a re y ‘vdieeal ati an: 4 MS whit - Pee a “ * me ora Paes at Pe a ett a Poa ary a nm $3 BSR StA ATER AEE SR2R bess 2A 56 8 WAS ALR SAS reaty Te Webs a ae rae h “ata SP aN D Lee 2 iatrasze oe! ee ee cle Spe eee ie ener ee eee ees E Coe pe ere are er eo as 9-H Sa oa nse det ATA SB atse eso Eo BREA Are 3.8 88d haa 5, 3 ek Siadpaitedsd.ncaceses See PUp Totor peri rae ete rte ee Aum Pte ras 6 TS ae Z sasicleeees ate Be a CcAsnsa 1 AGADR aac tehc te eaten cS, ati te cht a8 oe eatin 5 0 ROS be et hs toa ert ol a Soap air ietacale iene iemetiew were > kon es CA ed ehe he aoe TaN 6 gta 8 Rik Od AR thd, VeRO akan Beek 8K, ami Asics 4. Cera ne po eS ans aah Weems nee a ave eS k thet det see . PB 5a 2 oS, bata! ott... Part red SRA ae Ras eae -“ Sete= e382 pee h Sn a a ae lel ee . geese OS Pepe Pe Ree rg ti tates note estes . “Sg nig sie oN ay Tee elias | ene , rae Serene are= ee i ae ied Mineo ~~" pa BR ee aor < =e Se Pre) Spores cng ne Pte Me: : atresncets Ebert is atte ke a ee 39% ™ reece ty Seay a eee eb te br bade et agg sey ee Spee eA: Sos be es hab bh Sot kd by bebe be de bo ie “s ra yo Iain Soseteen ny det ee pte lees Bi Rick bak es a aS 5 : C oe be th pak Sans ies “ welch gece Ss : - L? \ } : PROCEEDINGS oF THE Bist OO IE TY or WASHINGTON SSIISAS ES SE Asser iieing SSS TES = aS ITER REGNELLIANUM SEPTIMUM. FLORA AEQUATORIENSIS. 23 APRIL 2003 VOLUME 116 NUMBER 1 fA THE BIOLOGICAL SOCIETY OF WASHINGTON | 2002-2003 Officers President: Roy W. McDiarmid Secretary: Carole C. Baldwin President-elect: Rafael Lemaitre Treasurer: T. Chad Walter Elected Council Michael D. Carleton G. David Johnson Clyde Roper Michael Vecchione Marilyn Schotte Don Wilson Custodian of Publications: Storrs L. Olson PROCEEDINGS Editor: Richard v. Sternberg Associate Editors Classical Languages: Frederick M. Bayer Invertebrates: Stephen L. Gardiner Plants: Carol Hotton Frank D. Ferrari Insects: Wayne N. Mathis Rafael Lemaitre Vertebrates: Gary R. Graves Invertebrate Paleontology: Gale A. Bishop Carole C. Baldwin Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Annual dues are payable on or before January 1 of each year. Renewals received after January | must include a penalty charge of $5.00 for reinstatement. Library subscriptions to the Proceedings are: $50.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an additional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued eee Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership, subscriptions, and/or back issues should be sent to: BIOLOGICAL SOCIETY OF WASHINGTON P.O. BOX 1897 LAWRENCE, KANSAS 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, and editorial questions should be sent to: EDITOR, RICHARD v. STERNBERG NATIONAL CENTER FOR BIOTECHNOLOGY INFORMATION—GENBANK BUILDING 45, ROOM 6An. 18D-30 NATIONAL INSTITUTES OF HEALTH BETHESDA, MD 20892-6510 Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Periodicals postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. © This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). (WPS) U.S. Forest Service, Pacific Northwest Research Station, Forestry Sciences Laboratory, Crania variation in Columbian white-tailed deer populations: implications for taxonomy and restoration Winston P. Smith, Leslie N. Carraway, and Thomas A. Gavin 2770 Sherwood Lane—Suite 2A, Juneau, Alaska 99801-8545, e-mail: Winston_Smith @fs.fed.us; (LNC) Department of Fisheries and Wildlife, Nash 104, Oregon State University, Corvallis, Oregon 97331-3803, U.S.A., e-mail: carver@proaxis.com; (TAG) Department of Natural Resources, Cornell University, Ithaca, New York 14853-2801, U.S.A., e-mail: tag] @cornell.edu UESTA] Abstract.—We examined variation in 18 cranial dimensions among three disjunct populations of white-tailed deer (Odocoileus virginianus) in the Pacific Northwest to test the hypothesis that they represent a single taxon. Previous allozyme analyses indicated considerable variation among the three popula- tions, but genetic divergences were less than conventional benchmarks used to distinguish subspecies. We observed substantial variation in cranial dimensions among the three populations that graphically sorted into three distinct morpho- logical groups and corresponded with east-west and north-south geographical gradients. Specimens of the northwestern white-tailed deer (O. v. ochrourus) from northern Idaho had longer and broader skulls than did Columbian white- tailed deer (O. v. leucurus) from the lower Columbia River or southwestern Oregon; specimens from southwestern Oregon had shorter rostra and narrower crania than those from the lower Columbia River. Even after controlling for differences in size related to age or sex, specimens from southwestern Oregon were relatively smaller animals with shorter faces and narrower posterior por- tions of the skulls than specimens in the other populations. These results do not support the hypothesis that the three groups represent a single taxon, nor do the results support the current taxonomy. Sample sizes were insufficient to fully evaluate if designating the three populations as distinct subspecies is war- ranted. Still, the three populations show considerable morphological and ge- netic variation, remain disjunct and isolated from each other, and likely are evolving along different trajectories because of geographical variation in hab- itat. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):1—15. 2003. The Columbian white-tailed deer (Odo- coileus virginianus leucurus {Douglas, 1829]) is one of three currently recognized subspecies of Odocoileus virginianus (Zim- mermann 1780) indigenous to the western United States (Smith 1991). Historically, Columbian white-tailed deer (CWTD) oc- curred throughout most of western Oregon and southwestern Washington lowlands, as- sociated with riparian vegetation of broad river valleys (Douglas 1829, Smith 1985). Extensive development of western Oregon following European settlement led to extir- pation of CWTD from most of its historic range, including the Willamette Valley of west-central Oregon (Smith 1985). Jewett (1914) and Bailey (1936) concluded that CWTD survived in the Willamette Valley until late in the 19" century. Today, its dis- tribution is limited to two isolated popula- i , rm A j 2 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON , Cotumbian Warre-14iLeo Deer ‘san, Nariomar Witouire Resuce bs Huntens ~~ Istanos Doweias county A 50 ku © Columbian White-tailed deer current distribution Fig. 1. Columbian White-tailed deer historic distribution é 29 om ptovet oy a co wd " \ Wiianere Rover F Uneeuca Rivers i OREGON {ff Idaho White-tailed deer current distribution Historic (stippled areas) and current (open circles and cross-hatching) distributions of Columbian white-tailed deer, Odocoileus virginianus leucurus Douglas (Smith 1985, 1987), and current distribution of Northwestern white-tailed deer, O. v. ochrourus Bailey, in Oregon and Washington (Johnson and Cassidy 1997, Washington Department of Fish and Wildlife 2000, Oregon Department of Fish and Wildlife, unpubl. data). Note that the Umpqua River branches into the North and South Umpqua rivers. tions: one along the lower Columbia River composed of several subpopulations that occur on several islands upriver from a Washington mainland subpopulation; and a second in the interior valleys of the Ump- qua River in Douglas Co., Oregon (Fig. 1). The CWTD remains allopatric with the oth- er two western subspecies; the nearest, northwestern white-tailed deer (O. v. och- rourus Bailey 1932), is about 300 km east of the current range of O. v. leucurus (see simauido Wess, NID). The limited distribution of CWTD and imminent threat to remaining habitat prompted the U.S. Department of the Inte- rior, Fish and Wildlife Service (FWS) to list O. v. leucurus as endangered in 1967 in the Federal Register (32 FR 4001). The Colum- bian White-tailed Deer National Wildlife Refuge (CWTDNWR) was established in 1972 and the Douglas Co. population was included in the listing in 1978 (Smith 1985). Since then, much effort has been ex- pended toward recovery of the endangered populations, but the process has been slow and arduous (Doremus and Pagel 2001). The FWS developed a recovery plan with specific goals and measurable objectives, including information needs, to help the CWTDNWR and Douglas Co. populations recover (Columbian White-tailed Deer Re- covery Team 1983). Numerous studies doc- umented the status and provide information on the population ecology of CWTD (Gav- in 1979, Suring & Vohs 1979, Dublin 1980, Gavin et al. 1984, Smith 1985, 1987; Ricca 2000, Whitney 2001), but little attention was given to the taxonomy or genetic in- VOLUME 116, NUMBER 1 tegrity of CWTD populations (Gavin & May 1988). The original taxonomic description of CWTD was based on specimens collected from near the mouth of the Columbia River and from the lower Willamette River [=falls at present-day Oregon City, Clack- amas Co., OR] (Douglas 1829). Douglas (1914) reported CWTD throughout the cen- tral river bottomlands of western Oregon, perhaps as far south as the Umpqua River valleys (in what is now Douglas Co.). Crews (1939) extended the range south to Grants Pass, Josephine Co., Oregon. To our knowledge, however, the relationship be- tween deer from Douglas Co. and deer from the region of the type locality was never rigorously examined. When Bailey (1932) described the northwestern white-tailed deer (O. v. ochrourus), he compared the type specimen to white-tailed deer collected by Jewett (1914) from Douglas Co. rather than to deer collected near the type locality of O. v. leucurus. Clearly, data supporting the original descriptions of these two taxa were limited. 5 Gavin & May (1988) evaluated the tax- onomic status of CWTD by comparing al- lozymes from 35 loci among multiple pop- ulations of white-tailed deer representing three subspecies, including O. v. ochrourus. They concluded that genetic distance be- tween the two CWTD populations and be- tween each of the CWTD populations and populations of O. v. ochrourus in Washing- ton and Oregon was less than the difference of two putative subspecies of widely sepa- rated geographic regions. Gavin & May (1988) did not observe a consistent pattern of differentiation at several loci; rather, their conclusions were based on variation at a single locus. Moreover, they recommended that an examination of additional evidence should occur before assigning subspecific status to any putative populations of CWTD. The purpose of this paper is to evaluate the taxonomy of O. v. leucurus by use of morphometric data. Our objectives were: 1) to quantitatively characterize cra- nia of white-tailed deer from Douglas Co., Oregon, the CWTDNWR, and the historic range of northwestern white-tailed deer; 2) to determine if significant variation in cra- nial features exists among the three groups; 3) to compare findings of this morphologi- cal investigation to earlier findings based on genetic distance among the populations (Gavin & May 1988); and 4) to use the re- sults of this study to test the working hy- pothesis that white-tailed deer in the three populations belong to a single taxon. Materials and Methods We examined crania of adult white-tailed deer from northern Idaho (n = 6 females, 12 males), the Columbian White-tailed Deer Na- tional Wildlife Refuge (CWTDNWR; Gavin & May 1988) in Washington and Oregon (1 = 65 females, 52 males), and from Douglas Co., Oregon (n = 80 females, 49 males; Smith 1982). Samples from northern Idaho are museum specimens; age was determined by toothwear (Severinghaus 1949, Larson & Tabor 1980, Gee et al. 2002). Tom Gavin col- lected samples from the CWTDNWR (Gavin & May 1988); age was determined by num- ber of tooth cementum annuli (Scheffer 1950). Samples from Douglas Co., Oregon, were collected by Winston Smith (1982); age was determined by either number of tooth ce- mentum annuli or by toothwear (Larson & Taber 1980:154, Gee et al. 2002). Eighteen measurements (Fig. 2, Table 1) were recorded for complete crania. Many specimens were recovered dead along roads, and had dam- aged crania because of collisions with vehi- cles, which resulted in incomplete datasets for these animals. Gavin recorded all measure- ments. Because growth in deer does not be- come asymptotic until about 4 and 6 years- of-age for females and males, respectively, missing measurements were not estimated. We used data only from complete crania in Statistical analyses. Females were sorted into three age clas- ses for each collection area: age class | con- tained 2—2.9 year olds, 2 contained 3—3.9 4 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ree it td Fig. 2. Cranium of female white-tailed deer (Odocoileus virginianus; OSUFW [Oregon State University, Department of Fisheries and Wildlife mammal collection] 1140) illustrating dimensions recorded. 1, basilar length, 2, palatilar length, 3, length of upper molar series at alveolus, 4, breadth between M3s, 5, postpalatal breadth, 6, maxillary breadth, 7, zygomatic breadth, 8, height of foramen magnum, 9, width of foramen magnum, 10, mastoid breadth, 11, length of external nares, 12, breadth of external nares, 13, nasal length, 14, least nasal breadth, 15, greatest nasal breadth, 16, least interorbital breadth, and 17, breadth of braincase. The last dimension recorded was 18, depth of rostrum (not illustrated), which was measured with the cranium resting on a flat surface. It is the distance from the dorsal side of the premaxillae to the flat surface. Scale bar equals 10 cm. VOLUME 116, NUMBER 1 year olds, and 3 contained 24 year olds. Males were sorted into four age classes for each collection area: age class | contained 2-2.9 year olds, 2 contained 3-3.9 year olds, 3 contained 4—5.9 year olds, and 4 contained 26 year olds. Data were analyzed in SPSS 10.0.7 for Windows by use of the General Linear Model within a Multivariate Multiple Anal- ysis of Variance (GLM MANOVA) and Ca- nonical Discriminant Function Analysis (CDFA) with jackknife classification of specimens (Hair et al. 1987, McLachlan 1992). Age classes were designated as co- variates because age was not a primary fac- tor in acquiring specimens (Hair et al. 1987). Sample location (n = 3) and sex (n = 2) were treated as factors. Significance levelewas =< 0:05: Initially, a GLM MANOVA was per- formed only with specimens having com- plete datasets (4 females and 2 males from Idaho, 14 females and 15 males from CWTDNWR, and 29 females and 10 males from Douglas Co., Oregon). The. GLM MANOVA was repeated after data for each specimen were standardized by dividing each measurement by the area of its fora- men magnum (A = 0.257-width-height) to remove effects of size (Radinsky 1967) and to examine differences in shape of crania among collection areas. A CDFA was per- formed on standardized data present for the 11 dimensions deemed significant in the second GLM MANOVA for distinguishing specimens among the samples (6 females and 3 males from Idaho, 20 females and 22 males from CWTDNWR, and 38 females and 11 males from Douglas Co., Oregon) to present a pictorial representation of sep- aration for specimens from the 3 localities. Results There was substantial variation among populations in cranial dimensions (Table 1). The initial GLM MANOVA of the original data indicated that significant differences (F = 3.673-123.501, df = 2) among speci- 5 mens from the 3 sample areas occurred for all variables (Fig. 3A). When the interac- tion of collection area and sex was consid- ered, however, only basilar length, least in- terorbital breadth, zygomatic breadth, and mastoid breadth were significantly different (f = 3.256—-9.487, df = 2). The second GLM MANOVA of the standardized data set indicated significant differences (F = 3.772—13.911, df = 2) in the shape of the skulls for specimens among the three sam- ples involving the following variables: bas- ilar length, nasal length, breadth of the braincase, greatest width of nasals, least width of nasals, mastoid breadth, length of upper molar row, maxillary length, palatilar length, depth of rostrum, and width of ex- ternal nares (Fig. 3B, Table 2). Values for these 11 standardized variables for speci- mens from the three samples were analyzed in CDFA (Fig. 4). The axis for Function 1 accounted for 71.4% of the variation in specimens among the areas and was related to skull shape. The axis for Function 2 in- corporated the remaining variation (28.6%) in cranial dimensions, which was associated with overall skull size. All specimens from area 1, 85.7% of specimens from area 2, and 93.9% of specimens from area 3 were correctly classified into their a priori groups. Furthermore, in the plot of axes 1 and 2, with the exception of four individ- uals, three distinct groups were formed (Fig. 4). Even after controlling for differ- ences in size related to sex and age, speci- mens from area 3 are distinguishable in the first axis from those in areas | and 2 by a combination of shorter basilar and nasal lengths, and narrower braincase and least width of the nasals (Table 1). On the second axis, Specimens from area | are distinguish- able from those in areas 2 and 3 by having longer basilar lengths and broader braincas- es. They also have narrower faces (as in- dicated by the narrower least width of the nasals) than specimens from area 2. Thus, it is apparent that even with size based on age and sex accounted for, specimens from area 3 (Douglas Co., Oregon) are still rel- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 69 *190°0 (S'9L-9'€S) cov 0 + 10°L9 8Z “€S0'0 (0'69-1'rS) 9LE 0 + LY'C9 LE €90'0 (SCH OW 9C1'0 + 7681 LL “S900 (6 €7V LI) ccO'O + 61°07 vL vVv0O'0 (F991 I-0r'06) vISO + L8 001 LL “GAO (y'98-€ OL) LLEO + €16L c9 “€L0'0 (8°L7-6'00) GENO) FE {SII WE OL “SS0'0 (9'Lv-T'9¢) V8CO + S8° CL 6S “CSO'0 (9°8L-€°LS) OLV'O + 60°0L 89 “Ly0'0 (SEAS LO) 0c9'0 + O£ 601 89 Pe0'0 (vr7—-O17) €l60 + PLCC so[Pulay $7 “6S0'0 (0'08-1'€9) Iv80 + 9LOL 6b “S00 (L°6L-€°79) Ico O + 8069 6r “OL0'0 (CEE-O AD) 9610 + LIT 6r “1L0°0 OCGA M)) L610 + 9S'6l ev 6c0°0 (y' 111-666) 09r'0 + L9'SOT br “Or0'0 (0'88-6'°€L) cov'0 + 1618 Irv “7900 (€'6C-V 17) CVT O + 88° 7T br “TS0'0 (CSL) 6S£0 + CISL I€ “9€0'0 (€LL-€'99) S9V' 0 + CCL Iv “Sr0'O (y'CCI-8 VOL) 8090 + 69'CII St “7VE0'0 USCS COS 1 + 96 SET Sole UOBIIQ “OD se[snog ce “Lv0'0 (9°LL-8'€9) 09S°0 + [8°69 v9 “8£0°0 (C'IL-L’°6S) Il€O + 9P'S9 79 *SS0'0 (CCHV EW velo + or 6l 79 “1L0°0 (9'€T-6'91) 9/10 + LOOI c9 “8700 (€'LOI—-6'S6) 8Se0 + Le lol 79 *LE0'0 (9° L8-L’CL) c9C 0 + O'8L cc “$900 WOe=e ©) 8170 + OL VC 09 “0S0'0 (€'6v-L6€) 6LC0 + 61 EV OS “pr0'0 (8'08-7 L9) 19V'0 + 68°¢L vs “Le0'0 (C L7I-€ 601) 16S 0 + SL8IT LE “S700 (9SC-IE7) SCOT + LS PVT so[euUlay ce “TSO (€'08-€°79) 8$90 + PrEL OS ‘OL0'0 (9'06-S'79) vVVLO + OI SL OS ~L90°0 (V'TC-S 91) C810 + 6c 6l IS *890'°0 (8: 17-¥'r1) O80 + 96°81 cr 6r0'0 (0'0ZI-S$ 96) 88Z£0 + €L 801 Ly “€v0'0 (CT 16-8°SL) SISO + 68°C8 8h “090'0 (L'0€-9'C7) 8S1O0 + 68°SC vr “rS0'0 (6'VS—-L Ch) 9800 + 6697 Ly “TE0'0 (8° 18-9'TL) 6Sc0 + LV OL Sr “SrO'0 (O'OrI-€ VI T) S730 += VL SCI Ee SEO (9L7-PV70) O6rl + 81 C9TC Sey IAN AMNGLMO 9 *Lr¥0'0 (L'9L-9'L9) eve | + OLOL 9 “L700 (6'9L-€'89) COV l + £O'EL 9 *9€0'0 (VCERHG OW) SOE 0 + £8°0C 9 “€v0'0 (1°€T-L 00) I8cO + £8 IC 9 “6700 (O'E1 1-6 POT) PLOT + 8C 801 9 ‘0¢€0'0 (C'S8-S'°8L) COO'T + 8€°C8 ¢ *6S0'0 (87-8 V7) v69'0 + CS'9T 9 *9¢€0'0 (16-1 SP) $890 + L89r 9 “6¢£0'0 (1'6L-S'0L) Coll + £9 PL 9 *6£0°0 (OEIC WC) 1661 + 8L°Scl 9 “C700 (097-tr7) LECT + LI IS sopeulay bv “Se0'0 (9'08-€ VL) CLE TL + 80 8L ZI “94070 (S'96-€ 78) PST T + €9°98 TI “C600 (0'C7-Z'91) 6£S 0 + 907 cI “7L0'0 (CERO SD) 6cV 0 + LINC [a tte) (S'PTI—-O'E IT) CSI I + 69°91T IT ‘rs0'0 (CT €6—-L'6L) 90r' I + VS'98 OI “cro (y'0€-¥'97) v8e 0 + S6'8C IT “S€0'0 (C’ES-P' Lt) LeSO + 196 6 900 (9'€8-€'EL) LOTT + 69° 9L 9 Lc0'0 (9°6€1-8 0ET) lov l + cOSEl v “0CC 0 (98C-€L7C) O10’ + ST 8LTE SIRI OYLp] UIOYION SolvU [BUIA}X9 JO yISUIT UIpeeig pilose] WINUSPU USUILIOJ JO UIPIAA WMNUSeL USWIPIOJ JO JYSIOH Ypeoig oNeWOsAZ Upeoiq AIv][IXepy ypeoig jeieyedisog SEW UseMIeq YIPRIg sotias 1ejour 1addn jo yysuay Yysug] Iwpneyed Ysug] Ieypiseg SUOTSUSUITCG ‘T “SI Ul UMOYS 212 SUOISUSUIIG] “UOBIIQ “OD se[snog pure ‘uOosdIM pu UOISUTYSeAA Ul (AMNGLMO) I8DJOY BUPA [PUONeN Jood psiey-s1yAA uPIquIN[OD 924) “oYyepy] UroY.10U WO} SNUDUIBAIA SNAP1OIOPE IeW PUL J[eUIOJ IOJ SUOISUDUTIP [[NYS JO syuoUToInsvau JO u Aq PaMOT[OJ SAD pur ‘(sosoyjuored url) sosuvl “WO F suvayj— |] IGP VOLUME 116, NUMBER 1 SO) WLIO €~ VOTO Se “Cr O 8c “VEL O ) LGN O b “SET 0 (COLAC LW (GS-0) 16) (S 6¢-6'VC) (r'9b-9'S7) SESE SO) (LEE Le) CEL O = 68 CE 9Ec I + 1C9E 6760 + £8 8E 1660 + 86° 8¢ OLOME == Gals 661'C + EO CE UMSOL JO UONVASTA OL “SSO'0 8h “be0'0 £9 “E00 cS “LVO'0 9 “8100 cI “SSO (9 Ly-T9¢) (8°9L-¥'L9) (8°9L-L'99) (9°08-S'19) (L°TL—0'69) (1°S8-7 TL) OST O + St 69 LS€O + 6€ CL voeO + Ec IL O8v 0 + OO PL 91S 0 + Ce IL 89c | + OC 6L oseouTesiq JO Yjpeoly 08 “OSO'0 6r 9F0'0 S9 “€r0'0 OS “9S0°0 9 “€70'0 cI “LEO'O (S°79-S'0S) (€'L9-8'SS) (9°€9-6'0S) (8°€L-6'PS) (€€9-1'09) (CT LL-0'89) cleO = 1098 coe 0 + $909 80¢ 0 + OI 8S 6670 + LOCO O8S'O0 + eS 19 vVSLO + 8S IL YIPeolqg [eIWGIOIOJUT ISvO"T OL “8600 8E “€80'°0 8h “S800 cr “L800 9 *SL0'0 6 “C600 (CAL SY) (O'T€-V'C7) CLES SE) (O'LE-9' VC) (SHEE (IRE San Sutz@) S870 + 8 7C cclO + VI9C SceO + £S'9C S6£ 0 + £7 672 89L0 + SOC yoo 0 + CLE OF Yjpeolg [eSseU jSoJeoIH 89 °860'0 8€ “€L0'0 8b “LLO0'0 Cr COLO 9 “7800 8 “LOLO (8'07-S'ET) (S'7C-8'91) (677-091) (6'ST-8'91) (€61-9'ST) (6° €C-V LI) SOc 0 + ELI CCC O + S88I cIcO + OO 6I 6c£ 0 + IV IC SLSO + STLI LSLO + S00 Yjpeelg [eseu 4seo'] 8Z “8L0'0 tr °880°0 OS “990'0 tr C900 9 “L700 oY ORO) (€'6L-G'0S) (1°6L-9°CS) (6'88-L'79) (0'S6-V'0L) (898-608) (y'00I-9'T8) 8LS°0 + 08'S9 L160 + S169 LILO + SE 9OL 88L0 + 8I'€8 8060 + CI'€8 987 c + 09°68 yisug] [eseN L9 *980'0 0c “€90'0 €€ “0800 Lc “¥90'0 9 ‘080'°0 bv “€S0'0 (y'vE-L 7) (Cl Sexo SC) (S'VE-L'€T) (6 VE-V LZ) (CT €€-G'9T) (9°SE-8' 1) 16c 0 + EL LC 9070 + 69° 87C 9TV 0 + 8S OE 06c 0 + L8' Ie 9960 + L967 C680 + 88 CE Soreu [PUIOJXO JO Yjpeoly so[euloy SOTRIA So[VUuldy SORIA So[PUuld,] So[eIN suOIsUOWIICG uosIIO “OD se[snog UMNGLMO OYLp] WIDYIION ‘ponunuoj— | a9R,, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 100°0 €00'0 “S9'E L100 ‘10'9 “8500 cO0'0 “T10'0 €00°0 “€00'°0 SorBU [PUIO}X9 JO YISUIT L000 900°0 “T€e€ $000 “69°¢ “SoO'0 cO0'0 O10'0 €00°0 ~900°0 UINSOI JO UONRAIA 1000°0 LOO LE 6r0'0 “FOr ‘L000 cO00'O “100°0 1000°0 “1000°0 ypeoiq peyeyedisog TOO 0 1000°0 CLV O10'0 “l6'r Ic0'0 8'L WEWO 800°0 “1S0'0 600°0 “810°0 600'0 “600°0 yisug] Iepneyed 1000 LOO LYE GND 100°0 “S00'0 SEW Us9MJoq YIpvol| 100°0 Ic0'0 60°C “8S0°0 7000 “£100 ypeoiq Arey[Ixey] 100°0 LO0'0 “6C'€ 700°O “10'°9 SNTOSATS JB SOLIIS ‘br0'0 Z00'0 “€10°0 7000 “8000 Jejout saddn jo yysuey 100°0 1000°0 CEE] 600°0 OI'S OSO'O VIE 1000'0 19° PE ‘TCO'O 900'0 “6€0'°0 €00°0 “S00'°0 cO0'0 “€00°0 L100 ‘LIO'O YIpeoig ployse|y 1000 £000 O8'€ 1€0'O “L9'E Cr0'0 “LI'L “€b0'0 cO00'O S100 c00'O “$000 £€00°0 “€00°0 ssBvoUuTeIg JO YJpeol| 100°0 O10'O 60°¢ cI1O'0 89°9 *C60'0 7000 “9700 600°0 600°0 ypesiq syeuUosKZ 1000°0 1000 Orv 1000 VL CI ‘ve0'O cO0'O “7100 L00'0 *L00°0 YIpes4lqg [BWQIOIOJUT ISCO" 10000°0 1000°0 ‘O07 L €00'0 979 €00'0 ‘OL6 1v0'0 VEY ‘C00'0 1000°0 “€00°0 1000 “100°0 100°0 “100°0 €000'0 “€000'0 Yjpeolq [Bseu ]Svo'T 10000 100'0 ~S'v 6100 “It r 0100 ‘OO'L “EMO 1000°0 “S00'0 100°0 ‘c00'O 100'0 “100'0 Yjpeolq [BSeCU JSO}BOIL) 100°0 10000 “StL 10000 “9S°TI veE0'0 “89°F *990'0 L00'0 ‘Orv0'0 7100 “$700 ¢$00'0 “S00'0 ysug] [eseN L000 1000°0 ‘78'S ¢$00'0 “99°¢ L00'0 “v9'L ‘LOV'O 1v0'0 “‘vr70 6£0°0 “6L0°0 €S0'0 “€S0'0 ysug] Ieyiseg Lg) = A72 9 = Te G= iv 6 = Hp La 1 it = th0 189} OJVLULATTNU IOUA [Pepoul ps}9911I0D xoag + AI][ROO] UOTIDITTOD AyBOOT UOTDIT[OD X9S§ Sselo o3V pue sajqeriea asuodsay “AWTEOOT UONIT][OO pue (a[eUl ‘aTBUIOJ) KOS SIOM SIOIOR] OY], “So[eUI IOJ p PUL SOTBUIOJ IOJ Sasse]O € PLY SSLIO BBY BCLIVAOD OY, “UUINTOS JONI oY} IOJ }d99x9 MOTEq d pur F YUM SIV ‘SS SB a][qulueA asuodsal yUROYIUSIS AT[VONSTIeIS YORd IO} pojuasaid st Japoyy IeoUTT [eIAUSyH sy (Z96] AYSuIpeyY) [eNpIArpur ie JO} (HM4ST'O = VY) Wnuseu USUILIOJ dy) JO vore oy) Aq JUSUOINSKOU YORE SuIpIAIp Aq saTqeieA ssuOodseal IOJ vIep PozIpsepuris 9A\ ‘OYLP] UJOYJIOU pue ‘UOSIIO “OD se[snod “UOsaIQ puL UOISUTYSeM “YAMNGLMO OU) WOI SNUDIUISA1A SNA]10IOPOE JO S][NYS WOI, Pop1Odal saypquiieA dsuodsal pozipslepuLIS OQ] SUTJSJJV SIOJOVJ puv aJVLIeAOD—Z I[qVL VOLUME 116, NUMBER 1 9 atively smaller animals with shorter faces and narrower skulls than those specimens from either area 1 (northern Idaho) or area 2 (CWTDNWR). Error = 67 af. 0.009, 0.001 Discussion Assumptions and limitations of analy- ses.—Although we collected a reasonably large number of skulls from each of the lo- calities, incomplete data from many speci- mens substantially reduced our sample sizes for statistical analysis, especially specimens assigned to O. v. ochrourus. Small sample size can be problematic, especially for MANOVA where statistical power is easily compromised (Johnson & Wichern 1998). In addition, departure from normality, an important assumption of MANOVA, occurs more frequently with small sample sizes. Fortunately, MANOVA is relatively robust to violations of assumptions in many cir- cumstances (Johnson & Wichern 1998). Also, because of the large effect size (dif- ferences among means of treatments) among populations with many cranial di- mensions, statistical power probably was not an issue in our analyses. Comparison- wise error rates ranged from 0.013 to 0.0001 (Table 2). Small sample size also contributes to classification bias in CDFA, a consequence of which is an overestimate of divergence among taxa (Lance et al. 2000). In this study, we used the results of CDFA strictly for illustrative rather than analytical pur- poses. Still, we used a less biased jackknife technique for subsequent classification of specimens (Hair et al. 1987, McLachlan 1992, Johnson & Wichern 1998, Lance et al. 2000). Cranial variation and taxonomy.-—The taxonomy of white-tailed deer, like that of most of the North American mammal fau- na, predates development of genetic tech- niques and consequently early descriptions of taxa were based on variation of morpho- logical attributes, especially cranial char- acteristics (e.g., Ovis canadensis, Cowan =6 3.71, 0.003 Corrected model af. 0.003, 0.001 2 df: Value = 0.449 f = 1.60 df. = 32, p = 0.040 Collection locality + Sex 2 0.001 0.008 af. = 5), 1), Value = 0.088 f=7.74 df, = 32, Collection locality p = 0.0001 0.001, Sex df: = | = 0.269 0.0001 f = 8.82 Value d.f. = 16, Dp Age class df. = | Value = 0.550 f = 2.65 df. p = 0.004 multivariate test Response variables and Table 2.—Continued. Breadth of external nares Lambda Wilkes’ 10 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 210 220 230 240 250 260 270 280 290 300 BASILAR LENGTH (in mm) lai 3. STANDARDIZED ZYGOMATIC 0.52 BREADTH (in mm) 0.22 0.58 STANDARDIZED BASILAR LENGTH (in mm) 0.78 0.98 1.18 1.38 A. Plot of basilar length and zygomatic breadth illustrating a decrease in size of female and male white-tailed deer (Odocoileus virginianus) from northern Idaho (females HJ, males (_]), the Columbian White- tailed Deer National Wildlife Refuge in Washington and Oregon (females A, males A), to Douglas Co., Oregon (females @, males ©). B. Plot of standardized basilar length and standardized zygomatic breadth illustrating the same relative sizes for female and male white-tailed deer (Odocoileus virginianus) from the same collection areas. 1940). Much of the historical taxonomy of species and subspecies lacks an adequate quantitative basis and reflects a typological view inconsistent with an evolutionary per- spective (Ball & Avise 1992, Wehausen & Ramey 2000). Recent developments in mo- lecular biology (e.g., Cook et al. 2001) and statistical analyses (e.g., Steppan & Sulli- van 2000) have changed the way mammal- ogists do systematics, which in many in- stances has resulted in revisions of existing taxonomy (Steppan & Sullivan 2000, We- hausen & Ramey 2000, Cook et al. 2001). Still, morphometry can be a useful tool in elucidating evolutionary and taxonomic re- lationships (Wehausen & Ramey 1993, Genov 1999, Molina & Molinari 1999), es- pecially when used in conjunction with ge- netic data (e.g., Wehausen & Ramey 2000). We used variation in cranial morphology to test the hypothesis that deer in the three populations belong to a single taxon. This hypothesis was proposed on the basis of al- lozyme variation among three white-tailed deer populations (Gavin & May 1988). The results of our analyses indicate significant variation among the three populations for several cranial dimensions (Table 2). Thus, our results do not support the current tax- onomy, which implies that white-tailed deer from the lower Columbia River and Doug- las Co. (O. v. leucurus) are similar, yet dis- tinguishable from white-tailed deer in east- ern Oregon, eastern Washington, and Idaho (O. v. ochrourus). Rather, our results clearly delineate three distinct morphological pop- ulations (Fig. 4, Table 2) rather than a sin- gle unified taxon. Similar geographical variation in cranial dimensions has been reported for bighorn sheep, Ovis canadensis Shaw (Wehausen & Ramey 1993, 2000), wild boar, Sus scrofa Linnaeus (Genov 1999), black bear, Ursus americanus Pallas (Kennedy et al. 2002), and other white-tailed deer (Molina & Mol- inari 1999). The key issue in interpreting cranial variation in the context of subspe- cific taxonomy is whether the morphologi- cal variation is indicative of corresponding genetic divergences; or, whether it is largely ecophenotypic variation that resulted from regional differences in habitat or other en- vironmental differences (Wehausen & Ra- mey 2000, Kennedy et al. 2002). Some taxa (e.g., black bear) show clinal variation, 1.e., significant correlations between skull mor- VOLUME 116, NUMBER 1 11 DISCRIMINANT FUNCTION II 3 iD 24 0 { 2 3 DISCRIMINANT FUNCTION | Fig. 4. Canonical-variates plot of specimens from | (MM), northern Idaho, 2 (A), the Columbian White-tailed Deer National Wildlife Refuge in Washington and Oregon, and 3 (@), Douglas Co., Oregon. Function 1 ac- counted for 71.4% and Function 2 28.6% of the variation among the areas. Group centroids are indicated by numbers. Specimens from the three areas sorted into three distinct morphological groups; straight lines, drawn by eye, within the graph delineate the groups. Differences in shape of the cranium are characterized as follows: specimens from area 3 have overall shorter and narrower skulls than those from areas 1 and 2; and specimens from area | have a longer rostrum (as indicated by significantly longer nasals) and narrower cranium than those from area 2. decreasing from north to south. The lower Columbia River population had features in- termediate between those of the Idaho and Douglas Co. populations. Unlike black phology and climatic or other environmen- tal gradients (Kennedy et al. 2002), and dis- play substantial genetic dissimilarity among regional populations (Miller 1995). In our study, the pattern of cranial variation was somewhat similar to that reported for black bears (Kennedy et al. 2002) with skull size varying along a west to east gradient and bears (Miller 1995), however, there was no clear evidence of corresponding genetic di- vergences at one locus among the disjunct regional populations (Gavin & May 1988). 12 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Gavin & May (1988) reported that white- tailed deer populations from the Pacific Northwest showed relatively low genetic divergence. In a large number of possible pair-wise comparisons Gavin & May (1988) found Nei’s (1971) genetic distances between O. v. borealis from New York and white-tailed deer populations from the Pa- cific Northwest (0.037) were an order of magnitude greater than genetic differences among white-tailed deer populations of Oregon and Washington. Moreover, white- tailed deer from Idaho showed less diver- gence from the Douglas Co. population than from the lower Columbia River pop- ulation. Genetic distances (Nei 1971) be- tween O. v. ochrourus from Oregon and Washington and O. v. leucurus in south- western Oregon, and between O. v. ochrou- rus and white-tailed deer from the lower Columbia River were 0.003 and 0.010, re- spectively. Also, they found that genetic di- vergence between O. v. leucurus popula- tions on the Oregon and Washington sides of the lower Columbia River (0.007) was greater than between sampled O. v. ochreu- rus populations in Oregon and Washington (Fig. 1), or between O. v. leucurus popu- lations in southwestern Oregon and O. v. ochrourus populations (0.002). Genetic and morphological data com- monly suggest different conclusions regard- ing taxonomy of mammals. Recent exam- ples include Ovis canadensis (Wehausen & Ramey 1993, 2000) and Sus scrofa (Genov 1999), where separation of subspecies based solely on morphology (Cowan 1940, Genov 1999) was not supported by more rigorous analysis in conjunction with ge- netic data (Wehausen & Ramey 2000). The tendency has been to rely on molecular data, which presumably provides less am- biguous evidence. Ball & Avise (1992) pro- posed that subspecies are major subdivi- sions of the gene pool diversity of species where such subunits can be corroborated by independent, genetically based traits. Ac- cording to this view, subspecies should have distinguishing attributes that have an evolutionary basis (Wehausen & Ramey 2000). We found white-tailed deer populations of Oregon and Washington distinguishable by cranial dimensions, but Gavin & May (1988) found no compelling evidence from an evolutionary basis for this variation. The putative historical ranges of O. virginianus populations in the Pacific Northwest (Bai- ley 1932, Grinnell 1933, Smith 1985, Wil- liams 1986, Gavin & May 1988) suggest that populations interbred freely. Before European settlement, white-tailed deer oc- cupied most of the riparian floodplains and other deciduous lowlands in western, cen- tral, and northeastern Oregon. The range of O. v. ochrourus extended from northeastern California (Grinnell 1933) north to west- central British Columbia and east to north- central Wyoming (Hall 1981, Smith 1991). In Oregon, O. v. ochrourus occurred in the Klamath Basin (Walsingham 1873), which is only about 100 km east of the southern- most range of O. v. leucurus in southwest- ern Oregon (Smith 1985). Throughout east- central Oregon, O. v. ochrourus occupied floodplain and riparian communities, fre- quenting deciduous woodlands and woody thickets associated with streams and marsh- es (Walsingham 1873, Cowan 1936). Sim- larly, O. v. leucurus occurred throughout the river valleys and other deciduous wood- lands of western Oregon (Smith 1985). The Cascade Range likely represented a barrier for free movement of white-tailed deer be- tween central and western Oregon; how- ever, opportunities for gene flow before Eu- ropean setthement presumably existed along the Columbia River and in south-central Oregon where river valleys cut through the Cascade Range at relatively low elevations. Without geographic isolation or strong se- lective pressures associated with markedly different environmental conditions (e.g., Wehausen & Ramey 1993, 2000), there is little reason to believe that historic popu- lations of white-tailed deer in Oregon (and the Pacific Northwest) were not a single, contiguous breeding population. VOLUME 116, NUMBER I Today, circumstances are very different; the populations clearly are isolated from one another (Fig. 1). White-tailed deer in northeastern Oregon apparently have ex- tended their range westward and southward in recent years (Oregon Department of Fish and Wildlife, unpubl. data). Still, land use and natural barriers throughout central Oregon represent significant impediments to dispersal and natural expansion. Efforts to translocate deer may establish isolated local populations, but much of the native habitat in central Oregon has been modified (Verts & Carraway 1998). Moreover, avail- ability and connectivity of habitat in west- ern Oregon and along the Columbia River is such that future opportunities for natural or facilitated expansion are unlikely. This, combined with the potential competition from black-tailed deer Odocoileus hemion- us (Smith 1985), renders the likelihood of O. v. leucurus reoccupying significant por- tions of its historic range extremely low. We believe it is prudent to consider the question of taxonomy in the context of cur- rent circumstances rather than belabor what might have been. Neither earlier genetic re- search nor our morphological study pro- vides compelling evidence to warrant an unambiguous resolution of this question. Consequently, the current taxonomy, al- though not directly supported by either line of evidence, cannot be refuted with certain- ty. Nonetheless, the three populations are morphologically distinct, geographically isolated, occupy different habitats (Gavin IGVD., simian IXsS, Wert 67 Carmenveyy 1998), and likely represent unique gene- pool subdivisions of O. virginianus (Ball & Avise 1992, Wehausen & Ramey 2000). With these populations isolated and gene flow interrupted, genetic divergence may become significant in time (Avise 1994). Implications for recovery and conserva- tion.—Nomenclature shapes the view of how nature is organized (Avise 1994) and taxonomic units have become the founda- tion of conservation efforts (Cook & Mac- Donald 2001). Current taxonomy views 13 white-tailed deer populations of the lower Columbia River and Douglas Co. as O. v. leucurus, which may allow translocation of individuals from either location for the pur- pose of restoring populations in portions of its historic range. Our results do not support current taxonomy, but indicate that deer from the lower Columbia River and Doug- las Co. are morphologically distinct. Be- cause of geographic isolation and differenc- es in habitat, we believe that in time the two populations will become sufficiently genet- ically divergent to warrant separation into two taxa. For that reason, we think it is pru- dent to choose a conservative approach to restoring white-tailed deer in western Oregon and refrain from translocating deer from Douglas Co. (or eastern Oregon) to supplement populations along the lower Columbia River or establish populations in the Willamette River valley. Acknowledgments We thank curators and collection man- agers at the American Museum of Natural History, Mammal Division, the University of Puget Sound, James R. Slater Museum of Natural History, and the University of Idaho, Bird and Mammal Museum for loan of or access to specimens in their care. N. Slade, Natural History Museum, University of Kansas and P. Sullivan, Department of Natural Resources, Cornell University, pro- vided statistical assistance. B. Albritton Coblentz assisted WPS with sectioning and staining of tooth sections. Portions of this research were supported by a grant from the U.S. Fish and Wildlife Service to WPS and LNC. A. Gardner, M. Kennedy, and J. We- hausen provided valuable comments on an earlier draft that improved the quality of this paper. Literature Cited Avise, J. 1994. Molecular markers, natural history, and evolution. Chapman and Hall, New York, 511 PP. Bailey, V. 1932. The northwestern white-tail deer.— 14 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Proceedings of the Biological Society of Wash- ington 45:43—44. 1936. The mammals and life zones of Oregon.—North American Fauna 55:1—416. Ball, M. R., & J. C. Avise. 1992. Mitochondrial DNA phylogeographic differentiation among avian populations and the significance of subspe- cies.—The Auk 109:626—636. Columbian White-tailed Deer Recovery Team. 1983. Revised Columbian white-tailed deer recovery plan. U.S. Fish and Wildlife Service, Portland, Oregon, 75 pp. Cook, J. A., & S. O. MacDonald. 2001. Should en- demism be a focus of conservation efforts along the north Pacific coast of North America?—Bi- ological Conservation 97:207—213. , A. L. Bidlack, C. J. Conroy, J. R. Demboski, M. A. Fleming, A. M. Runck, K. D. Stone, & S. O. MacDonald. 2001. A phylogeographic perspective on endemism in the Alexander Ar- chipelago of southeast Alaska.—Biological Conservation 97:215—227. Cowan, I. M. 1936. Distribution and variation in deer (Odocoileus) of the Pacific Coastal Region of North America.—California Fish and Game 22: 155-246. . 1940. Distribution and variation in the native sheep of North America.—The American Mid- land Naturalist 24:505—580. Crews, A. K. 1939. A study of the Oregon white-tailed deer, Odocoileus virginianus leucurus (Doug- las). M.S. thesis, Oregon State University, Cor- vallis, 45 pp. Doremus, H., & J. E. Pagel. 2001. Why listing may be forever: perspectives on delisting under the U.S. Endangered Species Act.—Conservation Biol- ogy 15:1258—-1268. Douglas, D. 1829. Observations on two undescribed species of North American mammals (Cervus leucurus et Ovis californicus).—Zoological Journal 4:330—332. . 1914. Journal kept by David Douglas during his travels in North America 1823-1827. W. Wesley and Son, London, U.K., 364 pp. Dublin, H. T. 1980. Relating deer diets to forage qual- ity and quantity—the Columbian white-tailed deer. M.S. thesis, University of Washington, Seattle, 135 pp. Gavin, T. A. 1979. Population ecology of the Colum- bian white-tailed deer. Ph.D. dissertation, Oregon State University, Corvallis, 149 pp. , & B. May. 1988. Taxonomic status and ge- netic purity of Columbian white-tailed deer.— The Journal of Wildlife Management 52:1—10. , L. H. Suring, P. A. Vohs, Jr., & E. C. Meslow. 1984. Population characteristics, spatial orga- nization, and natural mortality in the Columbian white-tailed deer.—Wildlife Monographs 91:1— 41. Gee, K. L., J. H. Holman, M. K. Causey, A. N. Rossi, & J. B. Armstrong. 2002. Aging white-tailed deer by tooth replacement and wear: a critical evaluation of a time-honored technique.—Wild- life Society Bulletin 30:387—393. Genov, P. V. 1999. A review of the cranial character- istics of the wild boar (Sus scrofa Linnaeus 1758), with systematic conclusions.—Mammal Review 29:205—238. Grinnell, J. 1933. Review of the Recent mammal fauna of California——University of California Publi- cations in Zoology 40:71—284. Hair, J. F, Jr, R. E. Anderson, & R. L. Tatham. 1987. Multivariate data analysis with readings, 2nd edition. Macmillan Publishing Company, New York, 449 pp. Hall, E. R. 1981. The mammals of North America. Wiley-Interscience, New York, 2:601—1181 + 90. Jewett, S. G. 1914. The white-tailed and other deer in Oregon.—The Oregon Sportsman 2:5—9. Johnson, R. A., & D. W. Wichern. 1998. Applied mul- tivariate statistical analysis, 4th edition. Pren- tice-Hall, Upper Saddle River, New Jersey, 816 Pp. Kennedy, M. L., P K. Kennedy, M. A. Bogan, & J. L. Waits. 2002. Geographic variation in the black bear (Ursus americanus) in the eastern United States and Canada.—Southwestern Naturalist 47:257—266. Lance, R. E., M. L. Kennedy, & P. L. Leberg. 2000. Classification bias in discriminant function analysis used to evaluate putatively different taxa.—Journal of Mammalogy 81:245—249 Larson, J. S., & R. D. Taber. 1980. Criteria of sex and age. Pp. 143—202 in S. D. Schemnitz, ed., Wild- life management techniques manual, 4th edi- tion, The Wildlife Society, Washington, D.C., 686 pp. McLachlan, G. J. 1992. Discriminant analysis and sta- tistical pattern recognition. John Wiley & Sons, Inc., New York, 526 pp. Miller, D. A. 1995. Systematic classification of black bears in the southeastern United States. M.S. thesis, Virginia Tech University, Blacksburg, 102 pp. Molina, M., & J. Molinari. 1999. Taxonomy of Ve- nezuelan white-tailed deer (Odocoileus, Cervi- dae, Mammalia) based on cranial and mandib- ular traits—Canadian Journal of Zoology 77: 632-645. Nei, M. 1971. Identity of genes and genetic distance between populations.—Genetics 68(Supple- ment):47. Radinsky, L. 1967. Relative brain size: a new mea- sure.—Science 155:836—837. VOLUME 116, NUMBER 1 Ricca, M. A. 2000. Movements, habitat associations, and survival of Columbian white-tailed deer in Western Oregon. M.S. thesis, Oregon State Uni- versity, Corvallis, 170 pp. Scheffer, V. B. 1950. Growth layers on the teeth of Pinnipedia as an indication of age.—Science 112:309-311. Severinghaus, C. W. 1949. Tooth development and wear as criteria of age in white-tailed deer.— Journal of Wildlife Management 13:195-216. Smith, W. P. 1982. Status and habitat use of Columbian white-tailed deer in Douglas County, Oregon. Ph.D. dissertation, Oregon State University, Corvallis, 273 pp. 1985. Current geographic distribution and abundance of Columbian white-tailed deer, Odocoileus virginianus leucurus (Douglas).— Northwest Science 59:243-251. . 1987. Dispersion and habitat use by sympatric Columbian white-tailed deer and Columbian black-tailed deer—Journal of Mammalogy 68: 337-347. . 1991. Odocoileus virginianus.—Mammalian Species 388:1—13. Steppan, S. J., & J. Sullivan. 2000. The emerging sta- tistical perspective in systematics: a comment on Mares and Braun.—Journal of Mammalogy 81:260—270. 15 Suring, L. H., & P. A. Vohs, Jr. 1979. Habitat use by Columbian white-tailed deer—Journal of Wild- life Management 43:610—619. Verts, B. J., & L. N. Carraway. 1998. Land mammals of Oregon. University of California Press, Berkeley, 668 pp. Wehausen, J. D., & R. R. Ramey. 1993. A morpho- metric reevaluation of the Peninsular subspe- cies.—Desert Bighorn Council Transactions 37: 1-10. , & R. R. Ramey, I. 2000. Cranial morpho- metric and evolutionary relationships in the northern range of Ovis canadensis.—Journal of Mammalogy 81:145-—161. Walsingham, T. L. 1873. On the distribution of the different species of deer and other ruminants in northern California and Oregon.—Proceedings of the Zoological Society of London 1873:561— DOs: Whitney, L. W. 2001. Ecological relationships between Columbian white-tailed and black-tailed deer in Southwest Oregon. M.S. thesis, Oregon State University, Corvallis, 106 pp. Williams, D. EF 1986. Mammalian species of special concern in California. Wildlife Management Division Administrative Report 86-1, Califor- nia Department of Fish and Game, Sacramen- tole Epp: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):16—22. 2003. New subgenus of leptarctine (Carnivora: Mustelidae) from the Late Miocene of Nebraska, U.S.A. Jong-Deock Lim and Larry D. Martin (J-DL) School of Earth and Environmental Sciences(BK21), Seoul National University, Seoul, Korea 151-742; (LDM) Natural History Museum and Department of Ecology & Evolutionary Biology, University of Kansas, Lawrence, Kansas 66045-2454, U.S.A. Abstract.—A new subgenus, Pseudoleptarctus, is described on the basis of a mandible from the late Clarendonian (late Miocene). The new leptarctine differs from other species in having a rounded trigonid and a well-defined hypoconulid on m1 and an enlarged accessory cusp on p3. Pseudoleptarctus genowaysi has the largest dentition known for leptarctines. The morphological similarities to procyonids indicate their shared aspects of diet. Mustelid carnivores occupy various hab- itats and their diet ranges from insects to vertebrates. Qiu & Schmidt-Kittler (1982) and Lim & Martin (2002) considered Lep- tarctinae to be a subfamily of mustelids in- cluding Craterogale (N. America, M. Mio- cene), Trocharion (Europe, M. Miocene), Hypsoparia (N. America, U. Miocene), and Leptarctus (N. America and Asia, L. to U. Miocene). Leptarctus is one of the least known fossil carnivores and it occurs in the middle and late Miocene of North America and the Miocene of the Tung Gur region, 13.5-13.8 MYA, Inner Mongolia, China (Lim 1997, Lim & Miao 2000, Lim et al. 2001, Lim & Martin 2002a, Zhai 1964). Leidy (1856) described a P4 from South Dakota (Fort Randall Formation, Barstovi- an) for Leptarctus primus, the type species. Wortman (1894) described a lower jaw with cl, p3, and p4 from Nebraska (Clarendon- ian) and referred the specimen to L. primus. However, Matthew (1924) erected a new species, Leptarctus wortmani Matthew, 1924 for the specimen (Runningwater For- mation, E. Hemingfordian) previously de- scribed by Wortman (1894). Simpson (1930) described a P4 from Florida (Lower Bone Valley Formation, L. Barstovian-E. Clarendonian) as Leptarctus progressus and concluded that it shows some resemblance to Nasua in its greater width and larger pro- tocone. Stock (1930) described a skull frag- ment with P4 and M1 from Oregon (Mas- call Formation, E. Barstovian) as L. ore- gonensis. Olsen (1957) placed Mephititaxus (Thomas Farm Local Fauna, E. Hemingfor- dian) as L. ancipidens. In 1959, a skull, mandible, atlas, and axis were discovered from the Tung Gur region, Nei Mongol, China and described as L. neimenguensis (Zhai 1964). Lim & Miao (2000) described L. martini based on a well-preserved skull from Nebraska (Valentine Formation, L. Barstovian) and Lim & Martin (2001a) de- scribed L. kansasensis based on fragments of a skull fragment and maxillary bone with left P3 and P4 from Kansas (Blick Quarry, Clarendonian). Lim et al. (2001) described an incomplete skull with right M1 and left P4-M1 from Texas (Hemphill Beds, E. Hemphillian) for L. supremus. Lim & Mar- tin (2001b) described an anterior portion of skull with left P2-M1 and right P3-M2 from Nebraska (Republican River Beds, E. Hem- phillian) for L. desuii. The mammalian diet is reflected in dental morphology, jaw shape, and masticatory muscles (Smith 1993). Leptarctus has den- tal characteristics indicative of diet different VOLUME 116, NUMBER 1 from those of other mustelids. The upper incisors of Leptarctus can be distinguished by the peg-like and greatly reduced first and second upper incisors (Lim 1999b). The re- duced upper canines suggest a non-carniv- orous diet. The lower canine is strongly curved and grooved on its lingual side (Lim 1999a). A survey of the lower canines in modern carnivorous mammals revealed that the coati, Nasua, has a similar groove. Na- sua 1S an Omnivorous procyonid and one of the most frugivorous members of the Car- nivora. The edge of the masseteric fossa of the mandible is an insertion site for the mas- seter muscle which elevates and deviates the mandible while chewing (Turnbull 1970). The deep masseteric fossa of Lep- tarctus indicates a large masseter and sug- gests that Leptarctus could deviate the low- er jaw laterally more than other mustelids. Olsen (1958) considered L. ancipidens to be a badger-like mustelid. However, the lower dentition of L. ancipidens (UF 5655) shows it differs more from Taxidea taxus than from Procyon lotor. The talonid on the m1 of L. ancipidens is long relative to the tri- gonid, a condition similar to the m1 of Pro- cyon lotor. The talonid and trigonid of m1 in Taxidea taxus are of similar length. The ml of Pseudoleptarctus genowaysi also has a long talonid (Fig. 1). This increased length of the talonid in Leptarctus and Pro- cyon indicates that the anterior part of M1 that occluded with the talonid of m1 has increased its surface area. This structure shows that these animals had similar diets and needed a strong chewing capacity. One distinction between Leptarctus and other mustelids is the great height of the zygomatic arches in Leptarctus (Lim 1996). The ventral edge of the zygomatic arch is an origin for the masseter, elevating the mandible and slightly deviating the man- dible from side to side. The heavy zygo- matic arch indicates a large masseter and suggests that Leptarctus had stronger chew- ing muscles than other mustelids. Abbreviations used are: AMNH—De- 17 partment of Vertebrate Paleontology, Amer- ican Museum of Natural History; F: AM— Frick American Mammals, American Mu- seum of Natural History; UF—Dyivision of Vertebrate Paleontology, Florida Museum of Natural History, Gainesville, University of Florida; KUVP—Division of Vertebrate Paleontology, Natural History Museum and Biodiversity Research Center, University of Kansas; UNSM—Division of Vertebrate Paleontology, University of Nebraska State Museum. Systematics Order Carnivora Bowdich, 1821 Family Mustelidae Fischer von Waldheim, 1817 Subfamily Leptarctinae Gazin, 1936 Genus Leptarctus Leidy, 1856 Pseudoleptarctus, new subgenus Diagnosis.—Large leptarctine with car- nassial trigonid sub-equal in size to talonid and rounded anteriorly. The ml crown is dominated by cross-lophs. The m2 is elon- gated. The p4 with primary and secondary cusps sub-equal in size. Thick cingula on premolars and molars. Pseudoleptarctus genowaysi, new species Figs. 1-3 Holotype.-—UNSM 25470, a right man- dible with p3-m1. Type locality.—Cr-111, 2% mi SW of Burge P.O. on west side of the Snake River, Cherry County, Nebraska. Referred specimens.—F:AM 25176, right mandible with canine, p3—4; F:AM Da eet le AME 2S hoy ariel nails) i: AM 49412, left mandible with canine, p2- m2; F:AM 49413, right mandible with p3- m2. Chronology.—Ash Hollow Formation, Late Clarendonian, Miocene. Age.—Pseudoleptarctus refers to similar- ity to Leptarctus. Pseudo means false. The trivial name honors professor Hugh H. Gen- oways, former director of the University of 18 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ileoaele waysi, holotype, UNSM 25470, m1 of L. wortmani F: AM 25351 (middle) and m1 of L. ancipidens UF 5655 (bottom). Scale bar = 1 cm. Occlusal view (right p3-m1) of P. geno- Nebraska State Museum, Lincoln, Nebras- ka. Diagnosis.—Larger and more robust than L. primus and L. wortmani. The ml has a labial cingulum, well-defined hypo- conulid and the trigonid is rounded anteri- orly. The m2 has a labial cingulum. Pre- molars and molar are more robust and rounded than in other Leptarctus. The sec- ond cusp of p4 enlarged and almost as big as the main cusp. The p3 has a small sec- ondary cusp and a thick posterior cingulum. Description.—The mandible is missing anterior parts of p3 and the ramus above the angular process. The dentition, p3—ml, is well-preserved. The teeth are more rounded and wider than in other Leptarctus (Fig. 1). The lower canines of F:AM 25176 and F: AM 49412 have a groove on the antero- labial side, characteristics of the leptarcti- nes. The masseteric fossa is larger and deeper than in L. primus and L. wortmani. In p3, a cingulum is located on the pos- tero-lingual face of the tooth and a well- defined posterior accessory cusp is present postero-labially (Figs. 1—3). The p4 has a round ridge on the anterior margin of the main cusp. The second cusp of p4 is well-developed and proportionately larger than in other Leptarctus. The first molar has three large and high cusps on the trigonid. The metaconid is the highest cusp and a labial cingulum is pre- sent on the lateral surface of the connection between paraconid and protoconid. The tal- onid of ml is expanded posteriorly by a median hypoconulid. The hypoconulid is connected to the hypoconid and the ento- conid by a ridge. The second lower molar of F:AM 49412 has a labial cingulum, which differs from other species. The m2 is double-rooted and large and the mandibular foramen is deeper and more prominent compared to L. primus and L. ancipidens. Discussion The mandible (UNSM 25470) of Pseu- doleptarctus genowaysi is the largest known leptarctine mandible. The presence of a distinctive cingulum in the labial side of ml is unique among carnivores. The lower canines of all leptarctines in- cluding P. genowaysi are curved and grooved on the lingual side. We know of no similar groove among modern carnivorous mammals with the exception of the coati, Nasua (Lim 1999a). Nasua is an omnivo- rous procyonid and one of the most frugiv- orous carnivores (Gompper & Decker 1998). The groove in both leptarctines and Nasua stops at the base of crown. Soleno- don, a modern insectivore, has a deeper and broader groove on the antero-labial side of the lower canine as a conduit for toxic sa- liva, but the groove in Solenodon continues to the very edge of the lower jaw. The p3 and p4 of P. genowaysi are very similar to those of Nasua nasua in having VOLUME 116, NUMBER 1 19 Fige 2: bottom), <1.5. a postero-lingual cingulum. The presence of an accessory cusp on the postero-external face on p3 is similar to that in L. wortmani while other leptarctines do not have the ac- cessory cusp. The dentition and mandible of Pseudoleptarctus genowaysi are robust and bigger than those of L. wortmani (Fig. 3). The height of mandible (below m1) of P. genowaysi (UNSM 25470) is 45% great- er than in L. wortmani (F:AM 25351). The p4 shows some similarities to procyonids. The secondary cusp on p4 is well-defined and similar in size to the first cusp as in Lateral views of mandibles of P. genowaysi (F:AM 49412, above) and L. primus (AMNH 18270, Procyon lotor and Nasua nasua. Other lep- tarctines and living mustelids have a much smaller secondary cusp. One of the most distinctive characters of the new species is the presence of a cin- gulum on the labial side of ml. The cin- gulum begins laterally at the paraconid and stops below the protoconid. The highest cusp of ml is the metaconid while proto- conids in L. primus, L. wortmani, and L. ancipidens are higher than the metaconid. The metaconids in Procyon lotor and Na- sua nasua are also higher than the other 20 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Mandibles of leptarctines in lateral view. a, P. genowaysi UNSM 25470, holotype; b, L. primus KUVP 9153; L. ancipidens UF 5655; d, L. wortmani cast of F:AM 25351. Scale bar = 5 cm. VOLUME 116, NUMBER 1 Table 1.—Measurements (mm) of lower teeth and mandibles of P. genowaysi, L. primus, L. wortmani and L. ancipidens. P. genowaysi L. primus L. wortmani L. ancipidens (UNSM 25470) (AMNH 18270) (F: AM 25351) (UF 5655) Height of mandible (below m1) 14.5 8.5 10 Sei p3: Length Di SES 4.9 Width 4.0 35) 3 p4: Length 8.3 5.4 Toh 6.5 Width Dall 33 4.1 4.1 ml: Length ielleg/ 9.6 ie 10.7 Width 6.7 4.2 5.6 4.9 Length between p3—m1 26.1 23.4 2], cusps on ml. The width of ml in P. gen- owaysi is 1.6 times wider than in L. primus (Table 1). The talonid of ml in P. geno- waysi is enlarged with well-defined cusps (hypoconid, hypoconulid, and entoconid) as in Procyon lotor. The carnassial is a critical tooth in car- nivorous mammals and major changes in its structure usually indicate changes in dietary preference. In general, the more carnivo- rous the diet the greater the emphasis on antero-posterior lophing of the trigonid coupled with reduction of the talonid. Her- bivory is usually signaled by reduction of the trigonid and enlargement of the talonid. Leptarctus follows this pattern. Cross-loph- ing is also characteristic of the teeth of her- bivorous mammals. Pseudoleptarctus pre- sents an unusual mixture of these features. The trigonid is not reduced being about the same length as the talonid, but its anterior margin is squared so that the tooth pattern is dominated by cross-lophs. This creates a tooth even more modified towards herbiv- ory than that of the raccoon, Procyon lotor, and in some ways similar to that of the leaf- eating koala bear, Phascolarctos cinereus. We have previously pointed out other sim- ilarities between Leptarctus and koalas, and think that leaf-eating might have played a role in the leptarctine diet, especially in Pseudoleptarctus (Lim 1997, 1999). Ex- amination of the muscle insertions indicates a greater capacity to deviate the mandible When chewing and this probably contrib- utes to the rearrangement of cusps and the accentuation of cross-lophs. The lack of tri- gonid reduction (already present in Lep- tarctus) and the large double rooted m2 (as shown by the alveoli) are primitive char- acter states as compared to Leptarctus and show that Pseudoleptarctus is not simply a late occurring and advanced species of Lep- tarctus but a separate line that split off early in leptarctine evolution. The widening of the trigonid occurs in parallel with widen- ing of the p4 through enlargement of the secondary cusp and accentuation of the cin- gulum. Chewing force has shifted forward and this is reflected in a slight forward ex- tension of the masseteric fossa as compared to Leptarctus. The reduced peg-like central incisors and the outward divergence of the canines in leptarctines suggests some spe- cialized function and might have provided an exit for a long, slender, mobile tongue. Acknowledgments For access to the type specimen, we thank G. Corner and R. Hunt (UNSM). We thank R. Tedford (AMNH) and B. Mac- Fadden (UF) for access to collections. We thank D. Miao, K. Gobetz, and J. Chorn for review and help. The senior author ac- knowledges financial support provided by the Collection Study Grant of the American Museum of Natural History and the Pano- rama Society Small Grant from the Univer- sity of Kansas Natural History Museum. We specially thank anonymous reviews for helpful comments. This research was finan- N N cially supported by the BK 21 Project of the Korean Government. Literature Cited Gompper, M. E., & D. M. Decker. 1998. Nasua na- sua.—Mammalian Species 580:1—9. Leidy, J. 1856. Notices of extinct Vertebrata discov- ered by Dr. E V. Hayden, during the expedition to the Sioux County under the command of Lieut G. K. Warren.—Proceeding of the Acad- emy of Natural Sciences of Philadelphia 8:31 1— SWIPE, Lim, J.-D. 1996. Dental evidence of Leptarctus as a frugivorous mustelid—Journal of Vertebrate Paleontology 16(3):48A. 1997. Functional morphology of Leptarc- tus.—Journal of Vertebrate Paleontology 17(3): 60A. . 1999a. Systematics and functional morphol- ogy of Leptarctus (Mammalia: Mustelidae). Un- published Ph.D. dissertation, University of Kan- sas, Lawrence, 151 pp. . 1999b. Evidence for Leptarctus as an arboreal frugivorous mustelid.—Journal of Vertebrate Paleontology 19(3):59A. Lim, J.-D, & D. Miao. 2000. New species of Leptarc- tus (Carnivora: Mustelidae) From the Miocene of Nebraska, U.S.A.—Vertebra PalAsiatica 38(1):52—57. Lim, J.-D, & L. D. Martin. 2001la. New evidence for plant-eating in a Miocene mustelid.—Current Science 81(3):314-317. . 2001b. A new species of Leptarctus (Mustel- idae) from the Miocene of Kansas, USA.— Neues Jahrbuch fiir Geologie und Palaontologie Monatshefte 10:633—640. Lim, J.-D, L. D. Martin, & R. W. Wilson. 2001. A new species of Leptarctus (Carnivora, Mustelidae) PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON from the Late Miocene of Texas.—Journal of Paleontology 75:1043—1046. Lim, J.-D, & L. D. Martin. 2002. A new fossil mustelid from the Miocene of South Dakota, USA.—Na- turwissenschaften. DOI 10.1007/sO0114-002- 0315-1 (Published online: 20 March, 2002) Matthew, W. D. 1924. Third Contribution of the Snake Creek Fauna.—Bulletin of American Museum of Natural History 50(2):59—210. Olsen, S. J. 1957. Leptarctines from the Florida Mio- cene (Carnivora, Mustelidae).—American Mu- seum Novitates 1861:1—7. Olsen, S. J. 1958. The skull of Leptarctus ancipidens from the Florida Miocene.—Special Publication No. 2 Contribution to Florida Vertebrate Pale- ontology Paper 2:1—11. Qiu, Z. & N. Schmidt-Kittler. 1982. On the phylogeny and zoogeography of the leptarctines (Carniv- ora, Mammalia).—Palaeontol Zeitschrift 56: 131-145. Simpson, G. G. 1930. Tertiary Land Mammal of Flor- ida.—Bulletin of American Museum of Natural History 59(3):149—211. Smith, K. K. 1993. The form of the feeding apparatus in terrestrial vertebrates: Studies of adaptation and constraint. Pp. 150—196 in J. Hanken & B. K. Hall, eds., The skull, vol. 3. University of Chicago Press, Chicago, 460 pp. Stock, C. 1930. Carnivora New to the Mascal Miocene Fauna of Eastern Oregon.—Carnegie Institution of Washington Publication 404:43—48. Turnbull, W. D. 1970. Mammalian masticatory appa- ratus.—Field Museum of Natural History (Fiel- diana: Geology) 18:149—356. Wortman, J. 1894. On the Affinities of Leptarctus pri- mus of Leidy.—Bulletin of American Museum of Natural History 6:229—231. Zhai, R. J. 1964. Leptarctus and other Carnivora from the Tung Gur Formation, Inner Mongolia.— Vertebrata PalAsiatica 8(1):27—32. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):23—28. 2003. Species limits in the pygmy antwren (Myrmotherula brachyura) complex (Aves: Passeriformes: Thamnophilidae): 1. The taxonomic status of Myrmotherula brachyura ignota Morton L. Isler and Phyllis R. Isler Department of Systematic Biology—Birds, National Museum of Natural History, Washington, D.C. 20560-0116, U.S.A., e-mail: piantbird@aol.com Abstract.—The taxonomic position of a trans-Andean avian population de- scribed as Myrmotherula brachyura ignota Griscom, 1929, has existed under a cloud of uncertainty for the past fifty years. Recent advances in the use of vocalizations in species-level taxonomy in suboscines, generally, and tham- nophilids, specifically, provide a basis for a reexamination of the relationships of M. b. ignota to two closely related cis-Andean taxa, M. brachyura and M. obscura. Four independent vocal characters distinguish M. b. ignota from M. brachyura, whereas the vocalizations of M. b. ignota cannot be distinguished from those of M. obscura. Myrmotherula b. ignota and M. obscura also share plumage characters distinct from M. brachyura. Myrmotherula ignota is rec- ognized as a distinct species, Moustached Antwren, of which M. 1. obscura is a subspecies. Smallest of the typical antbirds (Tham- nophilidae), Myrmotherula brachyura was also one of the first in its family to be de- scribed (Hermann 1783). Unlike most thamnophilid species with widespread dis- tributions for which a plethora of subspe- cies were described in the following almost 150 years, M. brachyura remained a mono- typic species as late as 1924 (Cory and Hellmayr 1924) with a geographic range extending throughout Amazonia, the Guianas, and the region immediately north- west of the Andes. In 1929, the trans-An- dean population in Panama and Colombia (now known to extend to northwestern Ec- uador) was described as M. b. ignota (Gris- com 1929) based on the briefest of descrip- tions: “‘Similar to typical M. brachyura of Cayenne and Amazonia, but the light streaking on pileum, wings, and back great- ly reduced, resulting in more solidly black- ish areas; female with much paler rufous crown Stripes, and no fulvous on throat.”’ Zimmer (1932), after examining about 100 cis-Andean specimens of M. brachyu- ra, noticed that some were “‘sharply distin- guishable.”’ He stated that if these speci- mens were found in allopatry with M. brachyura, he would have considered them conspecific, but because they occurred at the same locations as M. brachyura, he de- scribed them as a new species, Myrmoth- erula obscura, Short-billed Antwren. Zim- mer’s diagnosis and descriptions were much more complete than those of Griscom, but the salient points can be summarized as fol- lows. Compared to M. brachyura, male up- perparts and wings much darker (blacker), black mystacal stripe and black postocular stripe much broader, and length of bill re- duced (ranges of measurements of exposed culmen abut but do not overlap); female with pale head markings narrower and more rufescent, back markings narrower and whiter, and mystacal streak, postocular area, and bill length differed like male. The geo- graphic range of M. obscura has since been found to extend through southeastern Co- lombia, eastern Ecuador, northeastern Peru, 24 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON and extreme west-central and southwest Amazonian Brazil. In his description Zimmer (1932) did not compare specimens of M. obscura and M. b. ignota, apparently because he only had one male specimen of M. b. ignota at hand. Myrmotherula b. ignota is an uncommon bird in the field and in collections, and the paucity of specimens has hampered under- standing of its taxonomic position. How- ever, the similarity of plumage characters distinguishing M. b. ignota and M. obscura from M. brachyura is apparent from the de- scriptions. Subsequently, the similarity caused Bond (1950), who had obtained two additional female-plumaged specimens of M. b. ignota, to consult Hellmayr on the possibility that they could be more closely related to M. obscura than to M. brachyura. Their conclusions were inconclusive, con- fused by the finding that bill lengths of M. b. ignota fell into the range of those of M. brachyura, and Zimmer had emphasized bill length as an important character sepa- rating M. brachyura and M. obscura. Their considerations, however, led Meyer de Schauensee (1966) to write “It is possible that what is currently called M. brachyura ignota from Panama and northwest Colom- bia may prove to be specifically distinct from M. brachyura and conspecific with M. obscura ...’ Most recently, the seventh edition of the Check-list of North American Birds (American Ornithologists’ Union 1998) noted the possibilities that M. b. ig- nota might be specifically distinct or more closely related to M. obscura than to M. brachyura. Today, bolstered by the growing under- standing that vocalizations in nearly all su- boscines (including the Thamnophilidae) are innate, vocal characters are increasingly employed in examining species level ques- tions in these families (Baptista & Kroods- ma 2001). In an effort to provide a point of reference for considering species limits in the Thamnophilidae using vocal characters, Isler et al. (1998) compared characteristics of vocalizations of eight pairs of closely re- lated but reproductively isolated, syntopic antbird species. Among these were M. ob- scura and M. brachyura, whose vocaliza- tions in the region of sympatry were ana- lyzed. Three independent vocal characters were found to distinguish loudsongs (sensu Willis 1967) of the two species, and in ad- dition, each species included a call in its repertoire not known to be delivered by the other. Given the diagnostic differences found between vocalizations of M. obscura and M. brachyura, an obvious question for em- pirical analysis is which of these species are the vocalizations of M. b. ignota most closely related? Earlier consideration of the question was hampered by the lack of vocal recordings of M. b. ignota, a situation sim- ilar to that facing earlier specimen compar- isons. Ten years ago only one recording of M. b. ignota, a loudsong, was available. Its initial examination suggested that M. b. ig- nota was more closely related to M. obscu- ra (M. & P. Isler, R. Ridgely in Ridgely & Tudor 1994), but the single example pre- cluded analysis. Since then, a number of re- cordings of M. b. ignota have been ob- tained, and they allow us to address this fundamental question. Methods Because current species designations confuse rather than clarify the discussion, henceforth in this paper we refer to the three taxa simply as brachyura, obscura, and ignota. Morphology.—Examination of speci- mens available at the National Museum of Natural History (USNM) indicated that de- scriptions of earlier authors (cited in the in- troduction) accurately reflected plumage coloration of the three taxa. Consequently, no additional plumage data were obtained. Because of the importance of bill measure- ments in the description of obscura, and be- cause the differences in bill length between ignota and obscura deterred earlier workers from establishing a close relationship be- VOLUME 116, NUMBER I tween the taxa, measurements of bill length were obtained, using MAX-CAL electronic digital calipers, from specimens of the Lou- isiana State University Museum of Natural Science, Baton Rouge, Museo de Historia Natural de la Universidad Nacional Mayor de San Marcos, Lima, Peru, and USNM. Vocalizations.—The sample of record- ings of brachyura and obscura used and referenced in an earlier study of vocal dif- ferences (Isler et al. 1998) was employed in this analysis. The sample was obtained in the vicinity of the lower Rio Napo in north- ern Peru. The reason for confining the sam- ple to this area was to obviate the possibil- ity of introducing geographic variation in the vocalizations of the wide-ranging brachyura into the comparison. Analysis of geographic variation in the vocalizations of brachyura was beyond the scope of this pa- per but will be addressed subsequently. Seven recordings of ignota from Panama and three from Ecuador were employed. The recordings were not yet archived in an institutional collection. They were obtained at the following locations (listed as country, department, locality) which are followed by the name of recordist and number of indi- viduals recorded. PANAMA: Colon: Canal Zone (R. Ridgely 1, B. Whitney 3). San Blas: Nusagandi (B. Whitney 1). Darién: Cerro Pirre (B. Whitney 1), Cana (B. Whit- ney |). ECUADOR: Esmeraldas: 30 km E San Lorenzo (D. Lane 1). Pichincha: 20 km N Pedro Vicente Maldonaldo (D. Lane, 2). Recordings were analyzed using “‘Ca- nary 2.1.4”’ software (Charif et al. 1995). Procedures for analyzing vocalizations of brachyura and obscura were described in the earlier paper (Isler et al. 1998). Vocal- izations of ignota were analyzed as follows. First, sound spectrograms were printed of each recording to identify vocalization types and to permit visual comparisons among recordings. In each recording, the first three measurable loudsongs were se- lected for measurement. Measurements were directed to vocal characteristics that had been found to differ diagnostically in DS the earlier comparisons of brachyura and obscura. Number of notes, duration, and pace (notes/sec) were obtained for the entire vocalization. Then, loudsongs were divided into five sections (Isler et al. 1998) to obtain the pace for each section and ratios of change of pace between sections. Measure- ments were also taken of the duration of the second, middle, and next-to-last notes and the spaces following them. Finally, we determined whether vocal characters differed diagnostically between taxa, employing guidelines developed ear- lier (Isler et al. 1998, 1999). As described in more detail in the previous papers, such differences must be unambiguously distinct character states, or, in the case of continu- ous variables, ranges may not overlap and the means (x) and standard deviations (SD) of the population with the smaller set of measurements (a) and the population with the larger set of measurements (b) must meet the requirement: Xa ale LSD, = Xp = i s/D), (1) where f, = the f-score at the 97.5 percentile of the ¢ distribution for n — 1 degrees of freedom (except for ratios where this statis- tical test is inappropriate). Results and Discussion Morphology.—Principal plumage char- acters distinguishing the taxa are as follows. Males of ignota and obscura are similar to brachyura except mystacal streak is broader and more distinct, black postocular streak more prominent, and black of crown and upperparts more extensive. Male obscura differs from ignota in having fewer pale streaks and consequently more extensive black in the plumage, especially on upper- parts. Female ignota and obscura similar to brachyura except malar and postocular streaks are more prominent (as in male) and light portions of head, throat, and breast are darker: tawny brown in /gnota, and buff in obscura (distinguishing these two. taxa) rather than white or white tinged buff as in brachyura. 26 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Bill measurements were taken from a larger number of specimens of ignota, in- cluding the first specimens from Colombia, than were available to earlier authors. The mean of bill length (measured from the na- res) for ignota in Panama is 8.4 + SD of 0.2 mm (range 8.2 to 8.8, n = 8); in Co- lombia the mean was 7.9 + SD of 0.2 mm (range 7.5 to 8.1, n = 7). Equivalent values for obscura in Peru are 7.5 = SD of 0.3 mm (range 6.8 to 8.1, n = 16), and those for brachyura (various countries) are 8.7 + SD of 0.3 mm (range 8.4 to 9.3, n = 8). These values confirm the difference in bill length between ignota in Panama and ob- scura found by earlier authors, but they also show that bill length of zgnota in Colombia is intermediate, suggesting the possibility of clinal variation between populations of ig- nota in Panama and obscura in Amazonia. Vocalizations.—The earlier study of vo- cal differences between brachyura and ob- scura (Isler et al. 1998) concluded that there are at least four independent vocal characters that distinguish the two taxa. These vocal differences, although primarily determined by measurements, are also evi- dent in visual comparisons of sound spec- trograms (Fig. 1). The examples shown in the figure are typical for each taxon, 1.e., they represent a composite of the mean val- ues of measurable vocal characters obtained for all spectrograms that were studied. No diagnostic differences have been found be- tween vocalizations of ignota recorded in Panama and in Ecuador, and the example shown in Fig. | and the data that follow represent combined values for recordings obtained in both countries. The only re- cording available from near the type local- ity of ignota, that of a loudsong, conforms closely to the example shown. Visual examination of the spectrograms (Fig. 1) shows that the vocalizations of ig- nota are similar to those of obscura and dif- fer from those of brachyura. Visually ap- parent in comparisons, the loudsong of brachyura 1s faster in pace (notes/sec) than those of obscura and ignota, and the central notes of brachyura loudsongs are short and sharply downslurred sounding unmusical to the human ear, whereas those of obscura and ignota are longer and less steeply downslurred, having more of a musical quality. Less obvious visually, the pace of brachyura speeds up more than that of ob- scura and ignota (erroneously verbally de- scribed as the reverse in Isler et al. 1998 although the quantitative data were correct). In addition to the loudsong, brachyura and obscura deliver an abrupt note (Fig. 1B, G) that does not differ diagnosably between them (not yet recorded for ignota). How- ever, a second type of call is clearly distinct. The second call of brachyura (n = 4) is a stereotyped trill (Fig. 1C), but those of ob- scura (n = 12) and ignota (n = 4) a down- slurred note (Fig. 1E, H). The latter varies somewhat among individuals, as exempli- fied in Fig. 1, but is consistent in its essen- tial characteristics. Visual distinctions between loudsongs of brachyura (Fig. 1A) and those of obscura (Fig. 1D) and ignota (Fig. 1F) are support- ed by comparative analysis of measure- ments (data presented include the mean fol- lowed by the standard deviation and range). Comparing the entire loudsong, there was no overlap in ranges of the number of notes between brachyura loudsongs (28.3 + 3.0, 23-33 notes; n = 11) and loudsongs of ig- nota (17.4 + 1.8, 14-21; n = 11) and ob- scura (16.4 + 3.2, 11-22; n = 19) although only the difference between brachyura and ignota meets our test of significance. On the other hand, brachyura loudsongs are sig- nificantly faster in pace than loudsongs of both ignota and obscura; (11.3 + 0.7, 10.0-12.4 notes/sec) versus (6.6 + 0.5, 6.0—7.5) and (6.1 + 0.6, 5.0—7.0) respec- tively. Differences noted visually between the central notes of brachyura loudsongs and those of ignota and obscura are reflect- ed in significant differences in the lengths of the middle notes; (28 + 3, 21—32 msec) versus (62 + 5, 54-71) and (62 + 8, 45— 77) respectively. Finally, change in pace in the initial half of the song is reflected in the VOLUME 116, NUMBER 1 Frequency (kHz) 2.0 Time (seconds) Jerse, lL. 2nd Sound spectrograms of loudsongs and calls of taxa in the Myrmotherula brachyura complex. (A) loudsong of brachyura (Peru, Loreto, Quebrada Sucusari; ISL BMW.23:10; Whitney); (B) abrupt note of brach- yura (Peru, Loreto, Quebrada Sucusari; ISL BMW.23:10; Whitney); (C) stereotyped trill of brachyura (Peru, Loreto, Quebrada Sucusari; ISL BMW.94:16; Whitney); (D) loudsong of ignota (Ecuador, Esmeraldas, 30 km SE of San Lorenzo; ISL DFL.5:01; Lane); (E) downslurred note of ignota (Ecuador, Esmeraldas, 30 km SE of San Lorenzo; ISL DFL.5:01; Lane); (F) loudsong of obscura (Peru, Loreto, Quebrada Sucusari; MLNS 30881; Parker); (G) abrupt note of obscura (Peru, Loreto, Quebrada Papaya; MLNS 29192; Parker); (H) downslurred note of obscura (Peru, Loreto, Quebrada Papaya; MLNS 29192; Parker). Identification data include location, archive number, and recordist. Acronyms for recording archives: MLNS = Macaulay Library of Natural Sounds, Cornell Laboratory of Ornithology, Ithaca, New York; ISL = recordings not yet archived in an institutional collection but copied into an inventory maintained by the authors. ratio between the duration of the second note and space and that of the middle note and space, with the higher ratios expressing a greater reduction in duration and hence a more rapid acceleration in pace; values of this ratio are greater for brachyura (1.73 + 0.11, 1.61—2.04) than for ignota (1.28 + 0.10, 1.08—1.44) and obscura (1.29 + 0.08, 1.18—1.47). In summary, four independent vocal characters differ diagnostically be- tween brachyura and ignota and three be- tween brachyura and obscura; vocal differ- ences between ignota and obscura are in- significant. Conclusions Earlier studies have shown that brach- yura and obscura are syntopic species that exhibit numerous diagnosable differences in vocalizations and morphology. Our com- parison of brachyura and ignota demon- strates that these allopatric taxa differ to a similar extent vocally and morphologically 28 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON as do the sympatric brachyura and obscura. We therefore conclude that brachyura and ignota are specifically distinct, and we rec- ommend that ignota be raised to species level. However, ignota and obscura cannot be separated vocally, and given the minimal morphological distinctions between them and recognizing the principle of priority, we recommend that obscura be considered a subspecies of Myrmotherula ignota under the Biological Species Concept (Johnson et al. 1999). Additional data and analysis, in- cluding genetic studies, are needed to con- firm whether differences between M. i. ig- nota and M. i. obscura warrant their main- tenance as subspecies, or conversely, that they deserve recognition as distinct species. We propose the English name Mousta- ched Antwren for M. ignota. The name re- flects the wide black mystacal stripe char- acteristic of both subspecies. The name Short-billed Antwren, previously given ob- scura, 1S no longer appropriate because bill lengths of the newly constituted species overlap the bill lengths of brachyura. Nor do we believe that Griscom’s Antwren, pvro- posed for ignota if it were to be considered a distinct species (Ridgely & Tudor 1994), is satisfactory. In our opinion, Zimmer’s de- tective work in discovering obscura 1s much more worthy of recognition than Griscom’s minimal description of ignota, and it seems best to use a plumage feature as a basis for the English name. Acknowledgments We greatly appreciate the contributions of recordings of M. i. ignota by Bret M. Whitney, who provided six recordings from Panama, and Daniel E Lane, who provided three recordings from Ecuador. Without their generous support, this paper would not have been possible. Robert S. Ridgely also provided a recording from Panama. Con- tributors of recordings of M. brachyura and M. i. obscura were acknowledged in an ear- lier paper (Isler et al. 1998). We thank Rich- ard C. Banks, Robb T. Brumfield, and Gary R. Graves for their helpful comments in re- view of this paper. Literature Cited Baptista, L. E, & D. E. Kroodsma. 2001. Avian bio- acoustics. Pp. 11-52 in J. del Hoyo, A. Elliot, & J. Sargatal, eds. Handbook of the birds of the world, vol. 6. Mousebirds to Hornbills. Lynx Edicions, Barcelona. Bond, J. 1950. Notes on Peruvian Formicariidae.— Proceedings of Academy of Natural Sciences of Philadelphia 102:1—26. Charif, R. A., S. Mitchell, & C. W. Clark. 1995. Ca- nary 1.2 User’s Manual. Cornell Laboratory of Ornithology, Ithaca, New York. Cory, C. B., & C. E. Hellmayr. 1924. Catalogue of birds of the Americas. Pteroptochidae—Cono- pophagidae—Formicariidae.—Field Museum of Natural History (Zoological Series) 13, Pt. 3:1— 3GQ. Griscom, L. 1929. A collection of birds from Cana, Darién.—Bulletin Museum of Comparative Zo- ology, Harvard University 69:149—190. Hermann, J. 1783. Tabula affinitatum animalium olim academico specimine edita, nunc uberiore com- mentario illustratacum annotationaibus ad his- toriam naturalem animalium augendam facien- tibus. J. G. Treuttel, Strasbourg, 370 p. Isler, M. L., P. R. Isler, & B. M. Whitney. 1998. Use of vocalizations to establish species limits in antbirds (Passeriformes; Thamnophilidae).— Auk 115:577—590. . 1999. Species limits in antbirds (Passerifor- mes; Thamnophilidae): the Myrmotherula suri- namensis complex.—Auk 116:83—96. Johnson, N. K., J. V. Remsen Jr., & C. Cicero. 1999. Resolution of the debate over species concepts in ornithology: a new comprehensive biological species concept. Pp. 1470-1482 in N. J. Adams and R. H. Slotow, eds., Proceedings 22nd In- ternational Ornithological Congress, Durban. BirdLife South Africa, Johannesburg. Meyer de Schauensee, R. 1966. The species of birds of South America and their distribution. Living- ston Publishing Company, Narbeth, Pennsylva- nia. Ridgely, R. S., & G. Tudor. 1994. The birds of South America, vol. 2. The suboscine passerines. Uni- versity of Texas Press, Austin. Willis, E. O. 1967. The behavior of Bicolored Ant- birds. University of California Publications in Zoology 79:1—132. Zimmer, J. T. 1932. Studies of Peruvian birds. III. The genus Myrmotherula in Peru, with notes on ex- tralimital forms, part 1.—American Museum Novitates 523:1—19. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):29-—37. 2003. New fossil material with a redescription of the extinct Condor Gymnogyps varonai (Arredondo, 1971) from the Quaternary of Cuba (Aves: Vulturidae) William Suarez and Steven D. Emslie (WS) Museo Nacional de Historia Natural, Obispo 61, Plaza de Armas, La Habana, CH 10100, Cuba, e-mail: geopal@mnhnc.inf.cu; (SDE) University of North Carolina, Department of Biological Sciences, Wilmington, North Carolina 28403, U.S.A., e-mail: emslies@uncwil.edu Abstract.—New material of the Cuban Condor, originally described at An- tillovultur varonai (Arredondo, 1971), indicates that it is definitely referable to the genus Gymnogyps, but possesses other unique features that support its status as an extinct endemic species. It is characterized by distinct cranial and pre- maxillary features that are associated with more powerful musculature for feed- ing. These features include dorso-ventral deepening of the skull, long and mas- sive occipital processes, a large occipital condyle, and a more vaulted and broad rostrum compared with the living species G. californianus. Postcranial bones differ little from the living California Condor except in being larger and more robust. The Cuban Condor may have evolved relatively rapidly after a mainland population of G. kofordi or G. californianus reached the island and became specialized on the limited large prey available there during the Pleistocene. Resumen.—Nuevo material del Condor Cubano, originalmente descrito como Antillovultur varonai (Arredondo, 1971), indica que éste es definitiva- mente referible al género Gymnogyps, pero posee caracteres Unicos que sus- tentan su estado como especie endémica extinta. Se caracteriza por rasgos cra- neales y premaxilares distintivos, asociados a una musculatura potente, dado su modo de alimentacion. El craneo presenta una mayor profundidad dorso- ventral, procesos occipitales largos y masivos, condilo occipital grande, y rostro mas ancho y arqueado, en comparacion con la especie viviente G. californi- anus. Los huesos postcraneales difieren poco del Condor de California, excepto por ser mayores y mas robustos. El] Co6ndor Cubano pudo haber evolucionado relativamente rapido a partir de la colonizacion de la isla por una poblacion continental de G. kofordi 0 G. Californianus, que se especializo en el consumo de las limitadas carronas derivadas de la megafauna disponible en Cuba durante el Pleistoceno. Condors are large members of the family Vulturidae (=Cathartidae auct.) that appear to have originated in North America and later colonized South America (Emslie 1988). The only condor known outside the continental mainland of the Americas is the Cuban Condor Gymnogyps (“‘Antillovul- tur’) varonai (Arredondo 1971, 1976; Sua- rez 2000), providing evidence for the ca- pability of these scavengers to cross large bodies of water (see Tambussi & Noriega 1999). The genus Gymnogyps Lesson was most diverse in the Pleistocene and only the California Condor (G. californianus) sur- vives today (Emslie 1988). The earliest re- cord of the genus is from late Pliocene (Blancan) deposits of Florida (Emslie 1988, 1998). In addition, the fossil species Gym- 30 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nogyps kofordi is known from the early Pleistocene (middle Irvingtonian) of Flori- da (Emslie 1988). The only other fossil spe- cies in this genus are G. howardae from the late Pleistocene asphalt deposits of Talara Tar Seeps, Peru (Campbell 1979) and G. varonai, originally described as Antillovul- tur varonai (Arredondo, 1971) from Qua- ternary cave deposits of Cuba. The sup- posed species, G. amplus, from late Pleis- tocene asphalt deposits of Rancho La Brea (Rancholabrean) described by Miller (1911), is now considered to be a junior synonym of G. californianus and was a large, Pleistocene temporal subspecies (Em- She 1988). Recently, Suarez (2000) reviewed newly recovered fossil material and the generic status of the Cuban Condor as well as its paleodistribution and possible origin. As a result of this review, the monotypic and en- demic genus “Antillovultur’’, in which the species varonai was originally described, was considered congeneric with Gymno- gyps, an assumption that had been made by other authors (Olson 1978, Emshe 1988). Certain cranial and postcranial characters of Gymnogyps varonai were described by Suarez (2000), but were limited to only the generic, rather than the specific, position of the Cuban bird. Although some cranial fea- tures agreed with those described for fossil specimens of G. californianus amplus and G. kofordi, no direct comparisons were made with these taxa and the specific status of the Cuban bird has remained unclear since that time. Here we provide more de- tailed comparisons of the Cuban Condor with other fossil and living species. Materials and Methods Material examined.—The same _ speci- mens of the Cuban Condor, Gymnogyps va- ronai, listed with localities and depositories by Suarez (2000). Proximal end of a right radius, Instituto de Ecologia y Sistematica, La Habana, Cuba (IZACC 400-813). Fos- sils from the paleontological collection of the Museo Nacional de Historia Natural, La Habana, Cuba (MNHNCu), recovered in as- phalt deposits Las Breas de San Felipe (San Felipe II), 5.5 km west of Marti, Munici- pality of Marti, Matanzas Province, Cuba (for description and discussion of the chro- nology of this deposit see Iturralde-Vinent et al. 2000): fragmentary premaxillae (MNHNCu P4594, MNHNCu P4595), right coracoid (MNHNCu P4596), proximal right carpometacarpus (MNHNCu P4597) and distal left tarsometatarsus (MNHNCu P4598). Other fossil material examined is housed at: Museo Polivalente de Sagua la Grande, Villa Clara (MPSG), Collection of Oscar Arredondo, La Habana (OA), and Collection of William Suarez, La Habana (WS). Comparative material of the living, post- Pleistocene subspecies of the California Condor, Gymnogyps c. californianus ex- amined at the National Museum of Natural History, Smithsonian Institution (USNM), include the following skeletons: 13823, 17033, 17946-50, 345225, 346582, 489359, 489406, 489755, 492447. Compar- isons with specimens of the fossil Califor- nia Condor, G. c. amplus from Rancho la Brea, were made with the extensive collec- tions of the George C. Page Museum, Los Angeles, California. Specimens of G. ko- fordi from Florida Museum of Natural His- tory, Gainesville (UF) include the following paratypes: distal right ulna UF 63516; left femur UF 63513; distal left tarsometatarsus UF 31904. Casts of the holotypical right tarsometatarsus UF 63512, and paratypical fragmentary cranium UF 63517, also were used for comparison with this species. Os- teological terminology herein follows that of Howard (1929), Fisher (1944), and Jollie (1976-1977). Measurements were taken with a vernier caliper to the nearest 0.1 mm, following the methods of Fisher (1944) and Emslie (1988) for the cranium. Measure- ments designated with a plus sign (+) are from specimens with wear and abrasion, and are approximate. VOLUME 116, NUMBER 1 Systematic Paleontology Class Aves Order Ciconiiformes Family Vulturidae (Illiger, 1811) Genus Gymnogyps Lesson, 1842 Generic characters of Gymnogyps (Em- slie 1988) found in the Cuban specimens (Suarez 2000) are: cranium in dorsal view with constriction at postorbital, supraorbital edge convex, postorbital pit deep, upper mandible short and robust with nasal bar near the horizontal position in lateral view, bony circle formed by medial septum. Car- pometacarpus with large proximal symphy- sis (see Howard 1974); tarsometatarsus with concave surface of posterior shaft and with well-developed posterior protrusion on external cotyla. Gymnogyps varonai (Arredondo, 1971) Amended diagnosis.—Gymnogyps va- ronai differs from G. californianus Say, G. kofordi Emslie, and G. howardae Campbell (cranial material unknown in this species), in having: premaxillary short, broad and deeply vaulted, positioned near the level of the external nares (premaxillary larger and less vaulted in G. californianus; shorter, but less vaulted in G. kofordi), nasal bar very short, broad and flat (nasal bar relatively more slender, less broad and more rounded dorsally, not as flat in G. californianus; more slender, less broad, and constricted at the midpoint in G. kofordi), alinasal large, wide and flat, occupying more space in a shorter narial opening (alinasal small and less wide and flat, with narial opening larg- er in G. californianus; narial opening very large in G. kofordi); interpalatal space thin, maxilla wide, and maxillopalatines with a thin space between them so that they nearly touch at the mid line (interpalatal space wide in both G. californianus and G. ko- fordi); cranium high, with strong protrusion of the supraoccipital, large and massive oc- cipital processes with lateral areas beside supraoccipital protrusion thin and high (su- 3h] praoccipital less protruding, with lateral ar- eas wide and not as high in G. c. califor- nianus; Supraoccipital protrusion similar in G. c. amplus and G. kofordi, but with lateral areas wide and not as high), space between both temporal fossae small and flat (space between both temporal fossae wide and less flattened in G. c. californianus and G. ko- fordi), nuchal crest rostrally placed causing the area between this crest and the postor- bital process to be very short (larger in specimens of G. c. amplus and G. kofordi), interorbital space wide and flat (narrower and more rounded in G. californianus and G. kofordi), postorbital processes short with caudal orientation (postorbital processes larger with rostral orientation in G. califor- nianus; larger with slightly more rostral po- sition in G. kofordi); temporal fossa short antero-posteriorly, but deep and in general high and wide. Distal end of ulna flat in external-distal surface, external condyle Short and wide, distal radial depression large and pneumatic, carpal tuberosity large (rounded external-distal surface, external condyle large with distal radial depression small, less pneumatic in G. californianus; large and pneumatic distal radial depression in G. kofordi). Tibiotarsus with internal and external cnemial crests weakly developed and projected with reduction of muscular insertions, anterior and posterior intercon- dylar sulcus very thin causing the external condyle in distal view to be shorter and more voluminous. Tasometatarsus relatively short and robust, distal foramen placed low on shaft though this feature is variable. Comparative Description The Cuban Condor is particularly distinct from from G. californianus and G. kofordi in having a deeper, more laterally-com- pressed cranium with exagerated occipital and opisthotic processes, and the more ros- tal placement of the nuchal crest (Fig. 1). The rostrum (Fig. 2) also is more robust and deeply vaulted, with the dorsal surface bulging slightly above the level of the nasal 32 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Bice in fossil (right) and living (left, USNM 492447) G. californianus, in dorsal (A) and lateral (B) views. Scale bar = 1 cm. bar, more so than in G. californianus and G. kofordi. The mandible is similar to Gym- nogyps californianus except in being gen- erally more robust, with larger and blunter articular processes. This element is not known in G. kofordi. Postcranial characters of the Cuban Con- dor that differ from G. californianus in- clude coracoid with higher and more mas- sive furcular facet and coraco-humeral sur- face, attachment for anterior articular liga- ment in distal humerus extends farther distal-externally on shaft, presence of small Partial skull (MPSG 21) of Gymnogyps varonai (middle) in comparison with the equivalent element pneumatic foramen on proximal shaft of ra- dius below capital tuberosity (similar to G. kofordi), distal ulna with prominent shelf on the carpal tuberosity with a large foramen located distal to the shelf (similar to G. ko- fordi) and external condyle that tapers and extends less proximally, tibiotarsus (Fig. 3) with relatively long fibular crest and distal end with narrow intercondylar fossa when viewed distally, and tarsometatarsus (Fig. 4) with long and narrow middle trochlea (similar to G. kofordi). Measurements (mm).—Cranium (MPSG VOLUME 116, NUMBER 1 33 Rige 2. 1cm Sa enka Lateral view of the rostrum (upper mandible; MNHNCu P4613) of Gymnogyps varonai (middle) in comparison with the equivalent element in living (top, USNM 492447) and fossil (bottom) G. californianus. Scale bar = 1 cm. 21): temporal breadth, 43.2; cranial height, 38.8; postorbital position 51.9; postorbital breadth, 43.0; occipital breadth, 33.7. Pre- maxilla (MNHNCu P4613, formerly P588): breadth, 26.5; narial length, 21.9; narial breadth, 8.1; least breadth of nasal bar, 10.4; breadth of interpalatal space, 8.7; MNHNCu P4594: least breadth of nasal bar “lel: Coracoid (MNHNCu P4596): total length, 98.9; least breadth at midpoint, 17.0; depth at level of midpoint of glenoid facet, 21.8. Humerus (MPSG 30 and 31): proximal breadth, 53.0 and 52.6, respec- 34 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pa ba Fig. 3. Distal left tibiotarsus (MPSG 43) of Gymnogyps varonai (left) in comparison with the equivalent element in the living G. californianus (right, USNM 492447). Scale bar = | cm. VOLUME 116, NUMBER I 35 Fig. 4. Left (middle, WS 125) and right (right, WS 978) tarsometartarsi of Gymnogyps varonai in comparison with fossil G. californianus from Rancho la Brea (left). Scale bar = 1 cm. tively; depth of head, 18.0 and 18.2; IZ- ACC P80: least breadth and depth of shaft, 20.0 and 16.0; distal breadth and depth, 47.4 and 26.4; MPSG 32 and 33: distal breadth, 49.0 and 48.7, respectively; MPSG 33-) distal idepthy 25, aUinar (VIR SG 34)); distal breadth and depth, 22.9 and 23.3. Ra- dius (MPSG 36 and 37, IZACC 400-813): greatest diameter at proximal end, 16.9, 15.7, and 15.8, respectively; MPSG 35: dis- tal breadth, 23.6. Femur (OA 3202): total length, 141.0; proximal breadth, 37.3; depth of head, 17.0; least breadth shaft, 17.4; dis- tal breadth, 35.6; MPSG 38: proximal breadth, 35.7; depth of head, 17.8. Tibi- otarsus (MPSG 42): breadth of proximal 36 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON end through articular surface, 28.3; MPSG 43: least breadth and depth of shaft, 14.2 and 11.6; depth of external condyle, 23.5; distal breadth, 26.2; MPSG 41: length of fibular crest, 53.2. Tarsometatarsus (WS 125): proximal breadth, 25.7+; least breadth and depth of shaft, 15.0 and 8.8; MPSG 45e: proximal depth, 22.4; least depth shaft, 7.1+; breadth and depth of trochlea for digit 4, 8.2 and 16.3+; WS 978: least breadth and depth of shaft, 14.7 and 8.8; distal breadth, 32.2+; breadth and depth of trochlea for digit 2, 9.6+ and 13.1+; breadth and depth of trochlea for digit 3, 11.4+ and 16.5; breadth and depth of trochlea for digit 4, 7.6 and 14.9+; MNHNCu P4598: distal breadth, 33.0; breadth and depth of trochlea for digit 2, 10.0 and 13.4; breadth and depth of troch- lea for digit 3 depth, 11.7 and 16.6; breadth and depth of trochlea for digit 4, 8.3 and 13.4; OA 847: breadth and depth of troch- lea for digit 4, 8.0 and 15.9. Discussion The Cuban Condor differs from other fossil and living Gymnogyps mainly in fea- tures of the skull. The long and massive occipital processes, the shape of the brain- case, and the relatively large occipital con- dyle indicate that this species had large neck vertebrae and greater musculature in the head and neck associated with feeding. In addition, the relatively robust bill indi- cates that the Cuban Condor was more powerful and presumably could tear thick skin and sinew more easily than other con- dors of this genus. The more rostral place- ment of the nuchal crest in the Cuban Con- dor compared to other fossil and living Gymnogyps also is probably related to this more powerful feeding ability (see also Hertel 1995 for a description of cranial fea- tures related to scavenging). Postcranially, the Cuban species shows few differences in characters with other condors except for be- ing relatively larger and more robust than G. californianus californianus or G. kofor- di, and more similar to G. c. amplus (Suarez 2000). The Cuban Condor may have descended from Gymnogyps kofordi or G. californi- anus, both of which occurred in Florida (Emslie 1998, Suarez 2000). We hypothe- size that a population of one of these spe- cies reached Cuba in the early to late Pleis- tocene and quickly diverged into the en- demic form. Although fossil evidence is lacking for the presence of condors in Cuba prior to the late Pleistocene, G. varonai shares more cranial and some postcranial similarities with G. kofordi than with G. californianus, although the rostrum is most similar to that of G. californianus, espe- cially in the configuration of the nasal bar. Once established in Cuba, rapid diver- gence from its Florida counterpart would be facilitated by its insular isolation. In addi- tion, we speculate that limited large prey available at that time also might facilitate rapid morphological change in the cranial and bill regions. Large vertebrates known from the late Pleistocene of Cuba such as sloths [Megalocnus, Parocnus (=Mesoc- nus), Acratocnus (=Miocnus)], large ro- dents, and tortoises (Geochelone) were the main food sources for Gymnogyps (Suarez 2000). It is likely that the condors began to specialize on carcasses of these species and the thick hides of the sloths, and the hard carapaces of the turtles, selected for a more powerful feeding apparatus in Cuban Con- dors than in mainland species with more varied food sources. Strong competitive interactions at car- casses with other scavenging birds identi- fied in the fossil record of Cuba (Suarez 2001), including many raptorial species that were quite diverse in the Quaternary here, also may have increased the selective pres- sure for the robust cranial characters of Gymnogyps varonai. A more powerful bill for holding onto prey remains during inter- actions at a carcass may have made the Cu- ban Condor more competitive if food was limiting. Additional research on the func- tional morphology of the unique cranial VOLUME 116, NUMBER 1 characteristics of G. varonai with further studies of the extensive undescribed mate- rial of raptors from the Quaternary of Cuba are needed to further address this issue. Acknowledgements W. Suarez’s travels to Washington, D.C. and Los Angeles, California, were made possible by the Alexander Wetmore En- dowment Fund of the Division of Birds, National Museum of Natural History, Smithsonian Institution. We are grateful to Storrs L. Olson for his assistance at the Di- vision of Birds, Smithsonian Institution. Kenneth Campbell, Jr., Los Angeles County Museum, and Richard Hulbert, Florida Mu- seum of Natural History, Gainesville, as- sisted with loans of fossil specimens. John Steiner, Smithsonian Office of Photographic Services, provided the specimen photos. S. L. Olson and FE Hertel provided valuable comments on an earlier version of this pa- per. Literature Cited Arredondo, O. 1971. Nuevo género y especie de ave fdsil (Accipitriformes: Vulturidae) del Pleisto- ceno de Cuba.—Memérias de la Sociedad de Ciencias Naturales La Salle 31(90):311—323. . 1976. The great predatory birds of the Pleis- tocene of Cuba. Pp. 169-187 in S. L. Olson, ed., Collected papers in avian paleontology honoring the 90th birthday of Alexander Wet- more.—Smithsonian Contributions to Paleobi- ology 27:1—211. Campbell, K. E., Jr. 1979. The non-passerine Pleisto- cene avifauna of the Talara Tar Seeps, north- western Peru.—Life Sciences Contribution Royal Ontario Museum 118:1—203. Emslie, S. D. 1988. The fossil history and phyloge- netic relationships of condors (Ciconiiformes: Vulturidae) in the New World.—Journal of Ver- tebrate Paleontology 8:212—228. 37 . 1998. Avian community, climate, and sea-lev- el changes in the Plio-Pleistocene of the Florida Peninsula.—Ornithological Monographs 50:1— 113. Fisher, H. I. 1944. The skulls of the cathartid vul- tures.—Condor 46:272-296. Hertel, E 1995. Ecomorphological indicators of feed- ing behavior in Recent and fossil raptors —Auk 112:890—903. Howard, H. 1929. The avifauna of Emeryville shell- mound.—University of California Publications in Zoology 32:301—394. . 1974. Postcranial elements of the extinct con- dor, Breagyps clarki (Miller). Contributions in Sciences, Natural History Museum Los Angeles County 256:1—24. Iturralde-Vinent, M., R. D. E. MacPhee, S. Diaz-Fran- co, R. Rojas-Consuegra, W. Suarez, & A. Lom- ba. 2000. Las Breas de San Felipe, a Quaternary asphalt seep near Marti (Matanzas Province, Cuba).—Caribbean Journal of Science 36(3—4): 300-313. Jollie, M. 1976-1977. A contribution to the morphol- ogy and phylogeny of the Falconiformes.— Evolutionary Theory 1:285—298, 2:115—300, 3: 1-142. Miller, L. 1911. Avifauna of the Pleistocene deposits of California.—University of California Publi- cations, Bulletin, Department of Geology 6: 385—400. Olson, S. L. 1978. A paleontological perspective of West Indian Birds and Mammals.—Academy of Natural Sciences of Philadelphia, Special Pub- lication 13:99-117. Suarez, W. 2000. Contribuci6n al conocimiento del es- tatus genérico del céndor extinto (Ciconiifor- mes: Vulturidae) del Cuaternario cubano.—Or- nitologia Neotropical 11:109—122. . 2001. A reevaluation of some fossils identi- fied as vultures (Aves: Vulturidae) from Qua- ternary cave deposits of Cuba.—Caribbean Journal of Science 37(1—2):110—111. Tambussi, P.C., & J. I. Noriega. 1999. The fossil record of condors (Ciconiiformes: Vulturidae) in Ar- gentina. Pp. 177—184 in S. L. Olson, ed., Avian Paleontology at the close of the 20th Century: Proceedings of the 4th International meeting of the Society of Avian Paleontology and Evolu- tion, Washington D.C., 4—7 June 1996.—Smith- sonian Contributions to Paleobiology 89:1—344. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):38—46. 2003. A new gecko (Reptilia: Squamata: Genus Lepidodactylus) from Tuvalu, South-central Pacific George R. Zug, Dick Watling, Tataua Alefaio, Semese Alefaio, and Claudia Ludescher (GRZ) Department of Systematic Biology—Vertebrate Zoology, National Museum of Natural History, Washington, D.C. 20560-0162, U.S.A., e-mail: zug.george@nmnh.si.edu; (DW) Environmental Consultants Fiji Ltd., 259 Prince’s Rd., Suva, Fiji, e-mail: watling @suva.is.com.fj; (TA, SA, CL) Funafuti Marine Conservation Area, Private Mail Bag, Funafuti, Tuvalu Abstract.—A Group I Lepidodactylus was discovered during a recent bio- diversity survey of Tuvalu. These geckos uniquely share a heavily pigmented oral cavity with the Rotuman L. gardineri and also are most similar to L. gardineri in scalation, size, and body proportions; however, differences in the pubic patch of enlarged scales and the thickness of the head support speciation of the Tuvaluan population. This population is described and characterized. Two families of lizards are widespread and common colonizers of the islands of Oceania. The skinks (Scincidae) are diurnal and predominantly terrestrial lizards; the geckos are nocturnal and mainly arboreal ones. The origins and dates of these lizards’ colonization of Oceania remain debatable, although recent man-assisted colonization seems unquestionably the mode of dispersal for a few species, e.g., moth skink (Lipinia noctua; Austin 1999) and house gecko (Hemidactylus frenatus; Case et al. 1994). For other taxa, such as the geckos of the genus Lepidodactylus, dispersal throughout western Oceania likely occurred in the dis- tant past and well before human coloniza- tion of this area, because several endemic species occur irregularly from Rotuma, Viti Levu, ’Eua and westward. These species (L. gardineri, L. manni, L. euaensis, respec- tively, and others) are morphologically well differentiated and largely forest residents, thereby suggesting long periods of isola- tion. The interrelationships of these three taxa to one another and to congeners of the more western island groups are unresolved. Phenetically, Lepidodactylus consists of three species groups (Brown & Parker 1977). Group Ill (LZ. lugubris and relatives) consists of bisexual and unisexual species and populations, and one or more Group II species occur on almost every island in Oceania. Group III members are considered to be the most derived taxa of Lepidodac- tylus (Ota et al. 1995), and their current dis- tribution likely derives from natural and hu- man-assisted dispersal. The other two phe- netic groups, Group I (L. pumilis and allies) and Group II (L. guppyi and allies) are less specialized in morphology and are irregu- larly distributed among the islands west of and including the Tongan arc. L. manni (Fiji) and L. euaensis (Tonga) are members of Group I that is characterized by undivid- ed digital lamellae, and L. gardineri (Ro- tuma) is a Group II species, characterized by a few subterminal divided lamellae. It was, thus, surprising when a recent bi- otic survey in Tuvalu discovered another Group II Lepidodactylus. Individuals of this Lepidodactylus appear similar to Lepido- dactylus gardineri; however, some subtle differences suggest that the Tuvaluan pop- ulation represents a more ancient dispersal than a man-assisted one and that this pop- ulation’s isolation has resulted in speciation. We recommend that the Tuvulan population be known as: VOLUME 116, NUMBER 1 39 THOTT TTT TATTLE ET Fist Lepidodactylus tepukapili, new species ions! Holotype.-—USNM 531712, an adult male from Fuakea [Fuagea] (8°34’S, 179°04'E), Funafuti Atoll, Tuvalu, collected by Dick Watling on 4 September 1998. Paratypes.—USNM 531713-16, a juve- nile male, an adult female and two adult males, respectively, from Tepuka (8°28’S, 179°05'E), Funafuti Atoll, Tuvalu, collected by Dick Watling on 3 September 1998. Diagnosis.—Lepidodactylus tepukapili is a Group II species with the division or deep notching of two or three of the subterminal digital lamellae of second through fifth dig- its of the fore- and hindfeet. It differs from other Group II members: by the possession of a continuous row of 36 or more femoral- precloacal pores in adult males, 35 or less in L. novaeguineae, L. paurolepis, L. pulch- er, and L. shebae; by moderately dilated digital pads of fore- and hindfeet, only slightly dilated in L. vanuatuensis; by a blu- ish gray chin and throat, creamy white in L. guppyi; and by a larger pubic or pre- cloacal patch of enlarged scales (median 18 vs. 13.5; Table 3) and a flatter head (median Headih/SVib- 103 -vs-123%: Dable: 2)inwk. gardineri. Etymology.—The specific name tepuka- The holotype of Lepidodactylus tepukapili, USNM 531712. pili derives from the Tuvaluan language and is used as a noun in apposition. Pili refers to any small lizard, (either gecko or skink), and Tepuka is the island on which the first specimens were discovered. Puka of tepuka is the root word for two culturally important trees on the island, i.e., pukavai, Pisonia grandis, and pukavaka, Hernandia nym- Dhaeifolia. Description of the holotype.—Snout— vent length 50.3 mm; head length 11.5; head width 7.6; head height 5.4; snout—eye length 4.6; naris-eye length 3.6; orbit di- ameter 3.2; eye-ear length 3.0; snout width 1.9: interorbital width 3.6: snout—forelimb length 17.0; trunk length 20.9; crus length 6.0; tail length 37 (regenerate). All mea- surement here and subsequently are in mil- limeters. Mensural and scalation characters defined in appendix. Snout tapered, rounded at tip: rostral en- tering nares, width about 2.5 times height: nares bordered by five scales, three nasals, one rostral, and first supralabial; five scales touching rostral between left and right na- res; 35 interorbital scales; ten left and nine right supralabials; eight left and nine right infralabials; mental scale distinct, its ante- rior width equals midline length; six post- mental and seven chin scales. 40 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Piss 2s Lepidodactylus tepukapili, USNM 531712. Body slightly depressed; 118 rows of scales around midbody; dorsal and lateral scales granular, without enlarged tubercles, and in juxtaposition; ventral scales almost flat, cycloid, 2—3 times larger than dorsal scales; limbs well developed; subdigital la- mellae 16/16 and 11/12 on left/right sides of on digits IV of fore- and hindfoot (Fig. 2), respectively; digital lamellae ventrally covering nearly all of forefoot digits and I— II digits of hindfoot, about % of digits II-— V of hindfoot; all digits of fore- and hind- foot clawed except the first; ultimate claw- bearing and penultimate phalanges of fore and hind digits raised above pad although only claw free and extending over distal edge of pad; fore- and hindfoot webbing modest (<1/5 digit length); precloacal and femoral pore rows continuous with 38 ex- creting pores, reaching about % length of thigh; scales adjacent to pore bearing ones enlarged, usually in two rows anterior to pore row and posteriorly forming pubic patch of enlarged scales (Fig. 2). Posterior third of tail recently regenerate and likely regenerated from hemipenial sheath distally; tail subcylindrical through- out length, gradually tapering to a blunt tip; lateral margins without spines or skin flang- es; scales on tail annulate, cycloid, larger Ventral view of the posterior trunk and limbs (left) and right hindfoot (right) of the holotype of ventrally than dorsally, and subcaudal scales about 1.5 times belly scales; base of tail distinctly swollen by hemipenes; single large, blunt cloacal spur on each side. Color of holotype.—In preservation, dor- sal ground color of head, body, limbs, and tail brown with faint and discontinuous mottling of darker brown; ventrally, chin to anterior throat dusky, thereafter white with slight ventrolateral dark flecking on belly, ventrally tail white except for dusky on re- cently regenerate portion. When first found, the gecko was a rich chocolate brown dor- sally with lighter brown patches or mottling on the sides; the venter from chin onto tail was a bright dark yellow. Scales around eye and along upper lip were light, and interior of the mouth and tongue were black. The brighter coloration faded within an hour to a grayish brown dorsally and laterally, and a less intense yellow venter. Variation.—The two adult male para- types (USNM 531715-716) are distinctly smaller (43.5, 43.1 mm SVL, respectively) than the holotype but not greatly different from the adult female paratype (USNM 531714, 41.1 mm). There appear to be no proportional differences either between the smaller males and holotype or the female. The small sample size prevents any test of VOLUME 116, NUMBER 1 size dimorphism between adult females and males. The absence of dimorphism also ap- pears to be the situation for most aspects of scalation. Comparing the scalation of the holotype with the four paratypes yield the following: Ros, width usually 2.5 height; RosC, absent in all; NaRos, no contact in all; NaInf, no contact in all; CircNa, in- variant 3; SnS, 5 scales in holotype and ei- ther 4 or 5 in paratypes; IntorbS, 35 and 29-34; Suplab, 9 and 7-9; Inflab, 9 and 8— 9; Men, width equals height in all; PosMen, 6 and 7-10; Chin, 7 and 8—14; Midb, 118 andmlOO Sl 162CloacSe£ and l=2- Subcaud: width 1.5X height and 1.0—2.0X; ForefL, 6 ang Wéheilse leiiavchil,, eesti Wise LamNkE invariant 2; LamNL, invariant 4"; PoreRS, 43 and 37—42; Web, invariant bas- al 44": PrecIP, 17 and 12—18; for the males PreclPor, 38 and 39—40. These scalation traits show little variation within the Tuvalu sample. Color notes for the individual L. tepu- kapili are not available and likely would display no greater variation among individ- uals than within an-individual as an indi- vidual’s coloration shifted owing to physi- ological and psychological state. In preser- vation, the paratypes share the dorsal ground color with the holotype, although the dark brown mottling is more extensive on all paratypes, and the mottling largely dominates the dorsal coloration of USNM 531714 and 531716. Similarly the para- types share the holotype’s ventral coloration with more ventrolateral flecking from the neck to the hindlimbs; their chins and throats are dusky but somewhat lighter than the holotype’s. Distribution.—L. tepukapili is known presently from two islands, Tepuka and Fu- akea, in the Funafuti Atoll, Tuvalu. Limited searches on the main atoll island of Fon- gafale did not reveal any specimens. Natural history.—McLean and Hosking (1992) described the habitats of Funafuti Atoll, and Tepuka’s vegetation is almost en- tirely a “Coconut and Broadleaf Woodland.’ This mixed forest results from gardening 41 and regeneration, which create a medium density coconut woodland harboring stands and scattered individuals of broadleaf trees, such as the wide ranging Pisonia, Cordia, Calophyllum, Guettarda, Hernandia, Mor- inda, Hibiscus, Terminalia, and Thespesia. Of these, Pisonia and Hernandia are the most common species. The understory in- cludes Ficus and Pipturus scrub, and a groundcover of ferns, e.g., Asplenium and Nephrolepis. The smaller motu of Fuakea contains only a few coconuts in a similar mixture of broadleaf trees as on Tepuka. L. tepukapili was found under loose bark and in crevices, at one and two metres from the ground on the trunks of living trees, specifically Calophyllum inophyllum and coconut. Search time was limited by other bioinventory task, and we believe that L. tepukapili probably occurs in a larger va- riety of microhabitats and tree types. Comparison to Other Group II Members As noted in the Introduction, the three species groups of Lepidodactylus are phe- netically delimited. No study has tested the monophyly of these groups or, for that mat- ter, tested the monophyly of the taxon Lep- idodactylus (Kluge [1968] provided a set of diagnostic traits for this genus but did not address monophyly.). These two tasks are beyond the goals of our study; however, we wish to examine briefly the phylogenetic re- lationships of L. tepukapili. Our assessment of relationships assumes the monophyly of the guppyi complex (=Group II Lepidodactylus). Tables 1 and 2 provide a summary of select mensural and scalation characteristics of this com- plex. Only three species (gardineri, guppyi, vanuatuensis) are represent by reasonable, yet statistically inadequate, samples of adult specimens. Sexual dimorphism is a com- mon attribute among geckos. All members of the guppyi complex show this dimor- phism in the presence of secreting precloa- cal-femoral pores in adult males and their absence in adult females. Otherwise there & iw) PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Summary of selected mensural traits of adults of Group II Lepidodactylus species. Abbreviations are defined in section I of Appendix. Medians and ranges are presented for adults; SVL is in millimeters, proportions in percent; and sample size is in parentheses below specific name, females and males, respectively. SVL Taxon Female Male Head/SVL HeadW/SVL OrbD/HeadL Intorb/HeadL TrunkL/SVL CrusL/TrunkL gardinert 49.7 49.7 23 1] 30 Soy 44 27 (4; 5) 47.5—50.0 43.1—50.1 22—24 9-12 29-31 29-36 4446 24-33 guppyi 45.6 40.5 24 10 30 36 48 24 (3, 4) 37.3—-54.4 36.1-47.9 22—24 8-11 28-35 31-37 45-51 22-28 intermedius! —— 39.0 — a — — -— — (2, il) 242 lombocensis* 38.0 375) — — — — — — Gea) novae guineae OL 38.6 24 10 32 By) 48 24 Cn) 35.8-39.0 38.3-38.9 23-25 8-12 30-33 29-36 44-51 21-24 paurolepis — Soll DS 9 29 31 46 ZS (53) 37.4—-38.4 22-23 9-10 28-33 29-33 46-48 25-26 pulcher — 39.0 DS 11 33) 29 45 ei} (OX) shebae* 36.0 — 25 L7/ 30 — — —— (1, 0) tepukapili Al1.1 43.5 24 2 31 32 43 29 Gl; 3) 43.1-50.3 23-25 11-12 28-33 30-33 42-49 26-29 vanuatuensis 44.2 3553 24 10 30 31 47 24 (6, 4) 40.0-46.5 33.0—-39.2 22-25 8-12 27-34 29-33 44-50 21-28 Data from literature: ' Darevsky 1964; * Mertens 1929; >? Brown & Tanner 1949. is little commonality in the traits displaying statistically significant (Student’s f test, p < 0.05) sexual dimorphism among these three samples. Only one other character, ForefL displays sexual dimorphism in L. guppyi, five characters (Suplab, CloacS, HeadL/ SVL, HeadW/SVL, OrbD/HeadL) in L. gardineri, and seven characters (SVL, TrunkL, HeadW, EyeEar, SnW, Inflab, HeadL/SVL) in L. vanuatuensis. The sam- ple sizes are simply too small to decide Whether these dimorphic differences are real or a sampling bias. We provide body size differences for both males and females in Table 1, but otherwise the data are me- dians and ranges for all adult specimens (Tables 1, 2). In overall size, L. gardineri averages larger than any other guppyi member (Table 1) and appears to have equal-sized females and males. L. guppyi and L. tepukapili are the next largest geckos of this group; fe- males average larger in L. guppyi and pos- sibly the reverse in L. tepukapili, but the small samples argue for caution for such an interpretation. Caution is re-enforced by the L. vanuatuensis sample with females nearly as large as female L. guppyi yet with males Table 2.—Comparison of the relative head dimension of adult Lepidodactylus gardineri and L. tepukapili. Abbreviations are defined in section I of Appendix. Medians and ranges are presented for adults; proportions in percent; and sample sizes are same as in Table 1. Taxon HeadL/SVL HeadH/SVL HeadH/HeadL EyeEar/HeadL NeckL/HeadL gardineri 23 103 45 34 1S)7/ 22-24 98-116 44-49 31-36 150-165 tepukapili 24 123 50 30 148 22-25 107-126 47-53 26-35 138-154 VOLUME 116, NUMBER 1 43 Table 3.—Summary of selected scalation traits of adult Group II Lepidodactylus. Abbreviations are defined in section I of Appendix. Median and ranges of the traits are presented when known; all values are for females and males, except males only for CloacS and PoreRS; sample sizes are same as in Table 1. Taxon IntorbS Suplab Midbody CloacS gardineri 33 9 111 3 31-35 8-10 103-118 2-3 guppyi 35) 9 US 2, 33-37 8-11 110-133 0-4 intermedius! — 10-11 — — (10)? (CAD @): lombocensis?! — 9-10 — — Gi) ee COSI D)>s (= 2); novaeguineae 35) 9 118 2 32-39 8-10 108-125 2—2 paurolepis 32 10 100 1 31-32 10-11 99-110 1-2 pulcher 39 10 143 1 shebae* — 10 — — tepukapili 33 8.5 110 D, 29-35 7-9 105-118 1—2 vanuatuensis 33) 9.5 100 1 31-35 9-10 91-118 1-2 PoreRS PreclPore PreclP ForefL HindfL LamNL Web 39 38 18 IS) 12 4 ] 46-41 38-41 13-22 12-16 12-17 344 1-1 42 41.5 14 13 10 3} 2 39-44 33-43 12-15 11-15 9-12 24 1-2 — 24 a 9-10 10-12 — ] (8)° — 20 — 10-11 12-14 — — (1)° 18 16 16 12 1] 2 2 15-19 13-19 14-19 10-12 10-13 2-3 2-3 Sill 26 wD 14 13 3) D, 30-33 25-29 11-13 11-14 12-16 3-3 1-1 13 13 14 16 20 ] l — — — — 11 23 2 (30 or 32)? 41.5 39 35) 14.5 12 -| 1 37-45 38-40 12-17 14-16 12-13 44 1-1 40 26 3.5) 225) WIS) 3 ] 30-43 10—40 8-21 10-14 10-14 24 1-2 Data from literature: 'Darevsky 1964; * Mertens 1929; 3 Ota et al. 2000; * Brown & Tanner 1949. averaging smaller than all other guppyi group males or females (Table 1). The standard head proportion traits (Ta- ble 1) of Lepidodactylus systematics show little difference among guppyi members; however, the shorter relative trunk length (TrunkL/SVL) and the longer relative crus length (CrusL/TrunkL) differentiate L. gar- dineri and L. tepukapili from the other gup- pyi members. Although these latter two spe- cies appear quite similar, several aspects of head shape (Table 2) are different. L. te- pukapili has a thicker head relative to both body (HeadH/SVL) and head length (HeadH/HeadL) than does L. gardineri. This difference appears associated with a somewhat shorter head (EyeEar/HeadL, NeckL/HeadL; Table 2) in L. tepukapili. These proportional differences in head shape are not evident to the authors’ eyes. Discriminant function analyses (step- wise, backward entry) of male and female morphometric data show a strong differen- tiation of L. gardineri and L. tepukapili in multivariate space (Fig. 3). In the female analysis, none of the thirteen characters were eliminated in the final step, and clas- sification attained 100% for the five taxa. For males, the final step retained six char- acters (SnEye, SnForel, SnW, NarEye, OrbD, CrusL) and attained 100% classifi- cation for all taxa except L. guppyi and L. vanuatuensis (75% each). Neither the rela- tive positioning of the taxa clusters nor the classification accuracy should be weighed too heavily in interpretation of relationships Owing to the small sample sizes of all taxa. We note only that these data offer confir- mation to our interpretation of speciation of the Tuvalu population. L. gardineri and L. tepukapili are similar in scalation (Table 3) with the exception of the pubic patch of enlarged scales, which is larger and has more scales in L. gardineri. In this trait, L. gardineri differs from all other Group II members; all other members are similar with the exception of L. novae- guineae and its intermediate-sized patch. Our impression is that L. gardineri and L. tepukapili are more similar to one another ++ PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Discriminant Score 2 5 -10 -5 0 5 10 Discriminant Score 1 Fig. 3. 15 -20 -10 0 10 Discriminant Score 1 Discriminant functional analyses of morphometric characters of Group II Lepidodactylus, males (left) and females (right). Each ellipse defines the 60% confidence limit. Species symbols are: gardineri, circle; guppyi, square; novaeguineae, diamond (no confidence ellipse shown for males); paurolepis, triangle; pulcher, pentagon (no confidence ellipse); tepukapili, star; vanuatuensis, star burst. than either is to any other guppyi member. This similarity and the uniquely shared in- tense melanism of the oral cavity indicate that these two taxa share a common ances- tor. They also presently represent the deep- est penetration of Oceania by the Group II species. Their discovery in Tuvalu and the persistence of large tracts of forest in Sa- moa suggest that one of these taxa or a close relative probably occurs there also. Acknowledgements This new species was discovered during a biodiversity survey supported by the South Pacific Regional Environment Pro- gram. DW thanks that program for its sup- port and the various team members and the Funafuti community for their friendly assis- tance. We also thank Ms. A. Tiraa and Ms. S. Marks for their assistance in confirming the coloration of living specimens. GZ’s re- search into the biology and evolution of the Pacific herpetofauna has been supported by the Smithsonian’s Scholarly Studies Pro- gram, the Research Opportunity Fund of National Museum of Natural History, and the NMNH Department of Systematic Bi- ology—Vertebrate Zoology section. Similar- ly, the curators and collections management staffs of many museums regularly and gen- erously assist GZ; for this study, we thank J. Vindum and A. Leviton of the Califor- nian Academy of Science, and G. Schneider of the Museum of Zoology, University of Michigan for the loan of Lepidodactylus specimens. The manuscript greatly benefit- ed from the critiques of Drs. I. Ineich and H. Ota. Literature Cited Austin, C. C. 1999. Lizards took express train to Pol- ynesia.—Nature 397:113—114. Brown, W. C., & FE Parker. 1977. Lizards of the genus Lepidodactylus (Gekkonidae) from the Indo- Australian Archipelago and the islands of the Pacific, with descriptions of new species.—Pro- ceedings of the California Academy of Scienc- es, 4" ser. 41:253-265. , & V. M. Tanner. 1949. Discovery of the genus Pseudogekko with description of a new species from the Solomon Islands.—The Great Basin Naturalist 9:41—45. Case, T. J., D. T. Bolger, & K. Petren. 1994. Invasions and competitive displacement among house geckos in the tropical Pacific.—Ecology 75: 464-477. Darevsky, I. S. 1964. Two new species of gekkonid lizards from the Komodo island [sic] in Lesser Sundas Archipelago.—Zoologischer Anzeiger 173:169-174. Kluge, A. G. 1967. Systematics, phylogeny, and zoo- VOLUME 116, NUMBER I geography of the lizard genus Diplodactylus Gray (Gekkonidae).—Australian Journal of Zo- ology 15:1007—1108, plates 1-19. . 1968. Phylogenetic relationships of the gek- konid lizard genera Lepidodactylus Fitzinger, Hemiphyllodactylus Bleeker, and Pseudogekko Taylor.—Philippine Journal of Science [1966] 95:331—352. , & M. J. Eckardt. 1969. Hemidactylus garnotii Dumeéril and Bibron, a triploid all-female spe- cies of gekkonid lizard.—Copeia 1969(4):65 1— 664. Lewin, AY, 1855 IRs Ist Gilbjos, Ui, 18, lalezil, @e (Cs Is; Dawson. 1985. Standards in herpetology and ichthyology: part I. Standard symbolic codes for institutional resource collections in herpetology and ichthyology.—Copeia 1985(3):802—832. McLean, R. F, & P. L. Hosking. 1992. Tuvalu Land Resources Survey—Island Report No. 7 Funa- futi. Prepared for the FAO/UNDP by the Dept. of Geography, Auckland University, New Zea- land. Mertens, R. 1929. Zwei neue Haftzeher aus dem Indo- Australischen Archipel (Rept.).—Senckenber- giana 11:237—241. Ota, H, R. I. S. Darevsky, I. Ineich, & S. Yamashiro. 2000. Reevaluation of the taxonomic status of two Lepidodactylus species (Squamata: Gek- konidae) from the Lesser Sunda Archipelago, Indonesia.—Copeia 2002(4):1109—1113. , R. N. Fisher, I. Ineich, & T. J. Case. 1995. Geckos of the genus Lepidodactylus (Squamata: Reptilia) in Micronesia: description of a new species and reevaluation of the status of Gecko moestus Peters, 1867.—Copeia 1995:183—195. , & T. Hikida. 1988. A new species of Lepi- dodactylus (Sauria: Gekkonidae) from Sabah.— Copeia 1988:616—621. Zug, G. R. 1998. Australian populations of the Nactus pelagicus complex (Reptilia: Gekkonidae).— Memoirs of the Queensland Museum 42: 613— 626. Appendix I. Characters and analysis Kluge (1967) defined a basic set of measurement and scale counts for geckos and subsequently (Kluge and Eckardt 1969) added additional characters and re- defined some of the earlier ones. These characters and their definitions have been largely adopted by other herpetologists (e.g., Ota and Hikida 1989). We use a subset of these characters and their definitions. Each character and its abbreviation follow; we include a def- inition only where we record the character differently than the preceding researchers. Abbreviations follow Zug (1998) for ease of recognition. All characters re- ported for the right side. 45 Mensural_ characters.—Crus length: CrusL— Length of tibia from knee to heel. Eye-ear length: EyeEar. Head height: HeadH—Dorsoventral distance from the top of head to the underside of the jaw at the transverse plane intersecting the angle of jaws. Head length: HeadL. Head width: HeadW —Straight-line distance from left to right outer edge of jaw angles; this distance does not measure the jaw musculature broadening of the head. Interorbital width: Interorb— Transverse distance between the anterodorsal corners of left and right orbits. Nares-eye length: NarEye. Or- bit diameter: OrbD—Eye diameter or length of other authors, although they measure anteroposterior axis length of orbit. Snout—eye length: SnEye. Snout—fore- limb length: SnForel. Snout—vent length: SVL. Snout width: SnW—Internasal distance of other authors. Trunk length: TrunkL—Body length or axilla—groin length of others; distance between the posterior edge of the forelimb insertion (axilla) to the anterior edge of the hindlimb insertion (inguen). Meristic characters.—Circumnasal scales: CircNa—Number of scales abutting naris, exclusive of rostral and first infralabial. Chin (secondary post- mentals) scales: Chin—Number of scales transected by straight line from left to right 3"*-4" infralabial su- tures. Cloacal spurs: CloacS. Femoral pores: Fem- Por—Number of pores perforating scales and secret- ing. Forefoot lamellae (scansors): Forefl—Number of 4 digit lamellae; lamella is wider than deep and con- tacts the marginal scales; fragmented proximal scales are excluded. Hindfoot lamellae (scansors): HindfL— As for ForefL. Infralabials: Inflab. Interorbital scales: IntorbS. Lamellar notching, first: LamNF—The num- ber of the first lamella divided or deeply notched on 4" digit of hindfoot counting from terminal or ultimate lamella. Lamellar notching, last: LamNL—tThe last di- vided or notched lamellae, as in LamNE Mental size: Men—Width to height proportion; scored as for Ros. Midbody scale rows: Midb. Naris-infralabial contact: NaInf—Naris abuts or separated from first infralabial. Naris-rostral contact: NaRos—Naris abuts or separated from rostral by scale. Precloacal and femoral pore- scales in contact: PoreC—Precloacal and femoral scales bearing pores, separate or continuous. Pore row scales: PoreRS—Number of enlarged scales in the precloacal-femoral pore-scale row, whether or not the scales contain pores. Postmental (primary) scales: PosMen—Number of scales touching mental and in- fralabials from left to right 3-4" infralabial sutures. Precloacal (preanal) pores: PreclPor—As for FemPor. Precloacal scale patch: PreclP—Number of scales as large or larger than the scales bearing precloacal pores and slightly larger than surrounding scales. Rostral size: Ros—Width to height proportion: 1, W = H; 1.5, W 1.5 times H; etc. in 0.5 intervals. Rostral cleft (crease): RosC—Absence or presence of midline cleft or crease. Snout scales: SnS—Number of scales be- tween left and right nares and touching rostral. Sub- 46 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON caudal scales: Subcaud—Size of the median subcau- dal scales relative to the dorsal caudal scales; score as for Ros. Supralabials: Suplab. Webbing: Web—Rel- ative amount of webbing, four states: 0, none between the 2 and 3™ digit of hindfoot; 1, slight, basal 4 of 2™ digit’s length; 2, moderate, %4 to %; 3, strong, more than ¥%. Sex and maturity.—Examination of the gonads re- vealed sex and maturity. Females were considered ma- ture when they possessed vitellogenic follicles, typi- cally >1.5 mm diameter, oviducal eggs, or stretched oviducts; males when the testes and epididymides were enlarged, supplemented by the presence of secreting precloacal or femoral pores. Comments on characters.—Several researchers have attempted to quantify digit shape and length, as well as other traits. Although we support quantification be- cause it permits statistical analysis and presumably re- moves a degree of bias or subjectivity, many voucher specimens are not carefully prepared resulting in bent or folded specimens or parts thereof. Thus, we believe that quantification of some characters implies a degree of accuracy, which does not exist. Our selection of mensural characters emphasizes those possessing ter- mini ending on bone and along axes that have rigorous bony struts reducing compression or bending. SnForel and TrunkL, for example, are two useful measurements but also two that can have significant variation result- ing from poor preparation. II. Specimens examined Museum abbreviations follow Leviton et al. (1985). Lepidodactylus gardineri Boulenger 1897 [type-local- ity: ““Rotuma, north of the Fiji Islands” ]. Rotuma: USNM 268142, 268145, 268147—-48, 268151, 268 153-54, 268156, 268161, 268169. Lepidodactylus guppyi Boulenger 1884 [*‘Faro Is- land’’|]. Solomon Islands: CAS 139650, 156114; UMMZ 99966; USNM 120346, 120877-079, 313866. Lepidodactylus novaeguineae Brown & Parker 1977 [““Lake Sentani area, West Irian’’]. Papua New Guinea: CAS 11028—029, 12182, 89684; UMMZ 122450; USNM 112824—27, 119248. Lepidodactylus paurolepis Ota, Fisher, Ineich & Case 1995 [‘‘Ngerukewid Group (7°11'N, 134°16’E), Be- lau islands”’]. Palau: USNM 284400, 284402—403. Lepidodactylus pulcher Boulenger 1885 [‘‘Admiralty Islands” ]. Papua New Guinea: CAS 139832. Lepidodactylus tepukapili new species. Tuvalu: USNM 531712-716. Lepidodactylus vanuatuensis Ota, Fisher, Ineich, Case, Radtkey & Zug 1998 [“... Espiritu Santo Island ... J]. Vanuatu: USNM 323264-268, 334163, 334184-189. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):47-S1. 2003. Ompok pinnatus, a new species of silurid catfish (Teleostei: Siluriformes: Siluridae) from mainland Southeast Asia Heok Hee Ng (HHN) Fish Division, Museum of Zoology, University of Michigan, 1109 Geddes Avenue, Ann Arbor, Michigan 48109-1079, U.S.A. Abstract.—A new species of silurid catfish, Ompok pinnatus, is described from the Chao Phraya and Mekong River drainages in mainland Southeast Asia (Indochina). It belongs to the O. eugeneiatus species group and closely resem- bles O. eugeneiatus. Together with O. eugeneiatus, it can be distinguished from all other congeners in having an extremely long dorsal fin (about one-fifth SL vs. about one-tenth SL), and from all except O. eugeneiatus in having the maxillary and mandibular barbels reaching beyond the base of the caudal fin vs. not reaching that point. It can be distinguished from O. eugeneiatus in having a longer dorsal fin, deeper caudal peduncle, shorter mandibular barbels and fewer anal-fin rays. The family Siluridae is one of the most species-rich of Asian catfishes (Bornbusch, 1995), and yet little is understood of the phylogenetic relationships within the fami- ly. One such problematic group demonstrat- ed to be paraphyletic by Bornbusch (1995) is Ompok Lacépéde, which are medium- sized silurids usually found in lakes and large rivers throughout South and Southeast Asia. According to Bornbusch (1995), there are four distinct clades within Ompok, viz. O. bimaculatus group, O. leiacanthus group, O. hypophthalmus group, and O. eu- geneiatus group. The O. eugeneiatus group currently in- cludes two nominal species: Ompok euge- neiatus (Vaillant, 1902) (described from western Borneo) and O. sabanus Inger & Chin, 1959 (described from northeastern Borneo). While examining material identi- fied as O. eugeneiatus from mainland Southeast Asia (Indochina), differences were observed between them and suppos- edly conspecific material from Sumatra and Borneo. These differences were found to be significant enough to warrant the recogni- tion of a separate Indochinese species, which is described herein. Materials and Methods Measurements were made with a dial cal- iper and data recorded to 0.1 mm. Counts and measurements were made on the left side of the specimens when possible. In ta- bles and text, subunits of the head are pre- sented as proportions of head length (HL). Head length and measurements of body parts are given as proportions of standard length (SL). The measurements and terminologies fol- low largely those of Bornbusch (1991), with the following exceptions: pelvic-fin length is measured from the base to the tip of the second (usually the longest) ray. Cau- dal-fin length is the length of the first prin- cipal (usually the longest) ray of the upper lobe measured from the posterior margin of the hypural complex. Head width is mea- sured at the opercle (across its widest point) but discounting any lateral projection of the branchiostegal membranes. Head depth is measured at the base of the occipital pro- cess. Interorbital distance is determined at the dorsalmost point (the narrowest dis- tance) between the orbital margins. Insti- tutional acronyms follow Leviton et al. 48 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lang, Il. Chhnang. (1985) with the addition of CMK (collec- tion of Maurice Kottelat, Cornol, Switzer- land). Comparative material.—Ompok euge- neiatus: CMK 16344, 1 ex., 72.5 mm SL; Borneo (Indonesia): Kalimantan Barat, Ka- puas River drainage, Sungai Letang near Kampung Kandung Suli. UMMZ 209881, 1 ex., 91.3 mm SL; Borneo (Indonesia): Kalimantan Barat, Danau Piam near Ketun- gau, 38 km NNE of Sintang. ZMA 113.097, Table 1.—Morphometric data for Ompok pinnatus. In %SL Head length 17.9-19.6 Head width 9.7—10.5 Head depth 11.7-13.4 Predorsal distance 26.1—28.5 Preanal length 31.9—35.3 Prepelvic length 28.9—33.2 Prepectoral length 17.9—20.0 Body depth at anus 19.7—23.4 Depth of caudal peduncle 5.0—6.0 Pectoral-spine length 11.2—-13.1 Pectoral-fin length 18.0—20.9 Length of dorsal fin 21.1—24.2 Pelvic-fin length 4.2-6.1 Length of anal-fin base 66.5-69.9 Caudal-fin length 18.6—23.9 In %HL Snout length 32.8—37.5 Interorbital distance 43.5—47.0 Eye diameter 21.7—25.8 Maxillary barbel length 425.2—505.3 Mandibular barbel length 495.4-613.3 Ompok pinnatus, holotype, UMMZ 232679,80.5 mm SL, Cambodia: Tonle Sap at Kompong 1 ex., 104.8 mm SL; Sumatra: Taluk. ZMA 120.537, 1 ex., 100.2 mm SL; Sumatra (In- donesia): Jambi, Batang Hari. ZRC 11819— 11820, 2 ex., 72.7—78.7 mm SL; Peninsular Malaysia: Pahang, Tasek Chini. ZRC 30458, 1 ex., 92.8 mm SL; Peninsular Ma- laysia: Pahang, Sungai Chini. ZRC 38803, 3 ex., 61.0—62.3 mm SL; Borneo (Indone- sia): Kalimantan Barat, Kapuas River drain- age, Danau Basuk, lake adjacent to Kapuas immediately downriver of Jongkong. ZRC 39036, 2 ex., 82.7—84.5 mm SL; Sumatra (Indonesia): Riau, Sungai Bengkwang, trib- utary of Batang Kuantan (Indragiri River), 4 hours downstream of Rengat. ZRC 41678, 5 ex., 37.7—86.7; Sumatra (Indone- sia): Jambi, from aquarium trade. Ompok sabanus: FMNH 44828, 1 ex., holotype, 131.6 mm SL; FMNH 44829, 11 ex., paratypes, 101.6—117.8 mm SL; Bor- neo (Malaysia): Sabah, Lahad Datu district, Segama River at Segama Estate. Ompok pinnatus, new species Fig. 1 Ompok eugeneiatus.—Bornbusch, 1995: 44 (in part). Ompok sp. cf. eugeneiatus.—Rainboth, 1996: 149; Lim et al., 1999: 383. Holotype.-—UMMZ 232679, 80.5 mm SL; Cambodia: Tonle Sap at Kompong Chhnang, fishing lot 9 in second channel E VOLUME 116, NUMBER 1 of town; W. J. Rainboth, C. Rotha & N. van Zaineas 27 eo 1995: Paratypes.—_UMMZ 186749, 1 ex., 67.1 mm SL; Thailand: Maharaj province, Koh Tong canal (tributary of Chao Phraya Riv- er) 17.5 km N of Ayutthaya; A. Witt & S. Tongsangah, 12 Aug—12 Sep 1964. UMMZ 232375" 2iex., 67 427 > mm! SL: Cambo- GiazmekKandals Preke Wa Pova, 11) kimS* of Phnom Penh; W. J. Rainboth, 2 Feb 1995. Diagnosis.—Ompok pinnatus can be dis- tinguished from congeners in uniquely hav- ing the following combination of charac- ters: greatly elongated dorsal fin (about 20%SL vs. 10—-15%SL) and barbels (ex- tending beyond caudal fin), head width 9.7— 10.5%SL, caudal peduncle depth 5.1- 6.0%SL, and 53-58 anal-fin rays. Description.—Body and head laterally compressed. Dorsal profile of body slightly convex, descending gently from dorsal-fin origin to snout tip, and again from the pos- teriormost dorsal-fin ray to the caudal pe- duncle. Anterior profile of snout rounded, dorsal profile of nuchal region concave. An- terior pair of nostrils tubular and located an- teromedial to maxillary barbel base. Pos- terior pair of nostrils bordered by fleshy dorsal and ventral membranes and situated posteromedial to maxillary barbel base. Mouth terminal; gape oblique. Rictal lobes narrowly continuous at rictus and deeply subtended by submandibular groove, with upper rictal lobe lacking skin fold. Teeth villiform. Dentary teeth in bands that narrow posteriorly, teeth extending from symphysis to near posterior end of jaw; premaxillary teeth in broader bands, teeth extending from symphysis to near posterior end of jaw. Vomerine teeth in a single crescentic band. Maxillary barbels reaching base of cau- dal peduncle when extended posteriorly. Single pair of mandibular barbels present, originating slightly anterolateral to gular fold; barbels flattened for most of length, reaching beyond tips of caudal fin when ex- tended posteriorly. 49 Eyes small, subcutaneous; located in middle of head; visible dorsally, and more so ventrally. Gill membranes separate and free from isthmus. Anterior third of left and right membranes overlapping. Branchiostegal rays 9. Gill rakers short, anteriormost rakers on lower first arch small and widely spaced; 4 on epibranchial and 16—18 on ceratobran- chial. Pectoral fin with convex distal margin and 9-10 branched rays. Proximal two- thirds of first pectoral-fin element ossified into a spine without anterior and distal ser- rae. Pectoral spine and articulated segments of first pectoral-fin element sexually dimor- phic in mature individuals. Pectoral spine of males broad and somewhat flattened dor- soventrally, dentated with 3—4 distinct pos- terior serrae that increase in size distally; proximal articulated segments with 4 well- developed posterior serrae, decreasing in size distally. Pectoral-fin spine in females and juveniles more slender than in adult males, without serrae on posterior edges of either spine or articulated segments. Pelvic fin with convex distal margin and 1,6 rays. Dorsal fin with pointed distal mar- gin (first dorsal-fin ray longest), with 1,3 rays; segments of first ray not ossified to form spine. Anal fin with straight distal margin and 53—58 rays; not confluent pos- teriorly with caudal fin. Integument over anal fin thickened proximally for slightly more than half of ray lengths; fin-ray erec- tor muscles extending along anterior edges of anal-fin rays, ventralmost extent of mus- cles same as that of thickened integument. Caudal fin strongly forked; principal rays i,7,7,1. Urogenital papillae of both sexes lo- cated immediately posterior to insertion of pelvic fins. Vertebrae 12 + 36 = 48, 11 + 38 = 49 or 12 + 37 = 49. Morphometric data as in Table |. Color.—Head and body dark yellow, with scattered melanophores on dorsal sur- faces, flanks and thickened integument over anal fin; ventral surfaces of head, breast and belly with lighter covering of scattered me- 50 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lanophores. Faint black midlateral line var- iably present, usually as series of scattered melanophores broadening at base of caudal peduncle to form dark roughly triangular spot. Maxillary and mandibular barbels dark yellow, with color gradually fading distally. Fins hyaline, with small dark- brown spots occasionally present. Distribution.—Known only from the Chao Phraya and Mekong River drainages (an Thailand and Cambodia, respectively). Etymology.—From the Latin pinnatus, meaning feathered or plumed. In reference to the very long dorsal fin and barbels of this species. Used as a noun. Remarks.—The O. eugeneiatus group 1s characterised by the following combination of characters (after Bornbusch 1995): (1) cartilaginous plates supporting the mandib- ular barbels enlarged, roughly circular in shape, and each with a dorsolateral process that contacts the dorsal edge of the anterior ceratohyal; (2) anterior process of the hyo- mandibular short and extending anterodor- sally to or below the level of the cranial facet; and (3) the presence of 4 dorsal-fin rays. Bornbusch (1995) examined cleared and stained material of O. pinnatus (which he identified as O. eugeneiatus), and found them to possess the characters above, thus placing O. pinnatus within the O. eugeneia- tus species group. In the following discus- sion, detailed comparison of O. pinnatus with congeners will be confined largely to within the O. eugeneiatus group, which contains only two other nominal species: O. eugeneiatus and O. sabanus. Ompok pinnatus differs from all conge- ners in having a greatly elongated dorsal fin (about 20%SL vs. 10—15%SL), and (except for O. eugeneiatus) barbels (extending be- yond caudal fin vs. not reaching that point). It differs further from O. sabanus in having a narrower head (9.7-10.5%SL vs. 10.8— 11.9). Ompok pinnatus further differs from O. eugeneiatus in having a deeper caudal pe- duncle (5.1-6.0%SL vs. 3.9—4.8), shorter mandibular barbels (495.4—613.3%HL vs. 638.5—849.4) and usually fewer anal-fin rays (53-58 vs. 58-62). Ompok eugeneiatus and O. pinnatus share extremely long barbels extending be- yond the caudal fin (vs. extending just be- yond midway along body length in O. sa- banus and other members of the Siluridae), which is apparently derived within the O. eugeneiatus species group. On the basis of this unique derived character, O. pinnatus and O. eugeneiatus are hypothesized to be sister species. Given that O. pinnatus is re- stricted to the Chao Phraya and Mekong River drainages in mainland Southeast Asia, whereas O. eugeneiatus is found only in river drainages in central Sumatra and western Borneo, this hypothesized relation- ship conforms to the general biogeographic pattern in Southeast Asian siluriforms in which one sister species is found in main- land Southeast Asia (Indochina) and the other in Sundaic Southeast Asia (chiefly Sumatra and Borneo). A model of the his- torical biogeography of species with this pattern of distribution has been proposed by Bornbusch & Lundberg (1989), who hy- pothesized that the post-Pleistocene isola- tion of the North Sunda River system re- sulted in speciation. Acknowledgments I thank the following for permission to examine material under their care: Maurice Kottelat (CMK), Barry Chernoff (FMNH), Douglas Nelson (UMMZ), Isadc Isbriicker (ZMA) and Kelvin Lim. This study has been supported by the Rackham School of Graduate Studies of the University of Mich- igan. Literature Cited Bornbusch, A. H. 1991. Redescription and reclassifi- cation of the silurid catfish Apodoglanis furnessi Fowler (Siluriformes: Siluridae), with diagnoses of three intrafamilial silurid subgroups.—Cop- eia 1991:1070—1084. . 1995. Phylogenetic relationships within the Eurasian catfish family Siluridae (Pisces: Sil- uriformes), with comments on generic validities VOLUME 116, NUMBER 1 and biogeography.—Zoological Journal of the Linnaean Society 115:1—46. , & J. G. Lundberg. 1989. A new species of Hemisilurus (Siluriformes, Siluridae) from the Mekong River, with comments on its relation- ships and historical biogeography.—Copeia 1989:434—444. Inger, R. EF & P. K. Chin. 1959. New species of fresh- water catfishes from North Borneo.—Fieldiana, Zoology 39:279—296. Kevatons Ace = Reo Gibbso Jree. Heal & iC. E. Dawson. 1985. Standards in herpetology and ichthyology, part I. Standard symbolic codes for institutional resource collections in herpe- 51 tology and ichthyology.—Copeia 1985:802— 832. Lim, P., S. Lek, S. T. Touch, S.-O. Mao, & B. Chhouk. 1999. Diversity and spatial distribution of fresh- water fish in Great Lake and Tonle Sap river (Cambodia, Southeast Asia).—Aquatic Living Resources 12:379—386. Rainboth, W. J. 1996. Fishes of the Cambodian Me- kong. FAO Species Identification Field Guide for Fishery Purposes. FAO, Rome, xi+265, 27 pls. Vaillant, L. 1902. Résultats zoologiques de l’expédition scientifique Néerlandaise au Bornéo central. Pois- sons.—Notes from the Leyden Museum 24:1— 166. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):52—60. 2003. Halidesmus socotraensis new species and Haliophis guttatus (Forsskal), new records of congrogadine fishes from the Socotra Archipelago (Perciformes: Pseudochromidae) Anthony C. Gill and Uwe Zajonz (ACG) Fish Research Group, Department of Zoology, The Natural History Museum, London SW7 5BD, U.K., e-mail: t.gill@nhm.ac.uk; (UZ) Ichthyology, Senckenberg Research Institute and Natural History Museum, Senckenberganlage 25, D-60325 Frankfurt am Main, Germany, e-mail: uzajonz @sng.uni-frankfurt.de. Abstract.—Halidesmus socotraensis is described from six specimens, 39.6— 69.5 mm SL, from the Socotra Archipelago, western Indian Ocean. It is dis- tinguished from all other congrogadine species in having the following com- bination of characters: three lateral lines (dorsal, middle and ventral com- plexes); ventral lateral line branched on abdomen, with unpaired median section on ventral midline; and upper lateral line without anterodorsal branch on to nape. Haliophis guttatus is also newly recorded from the Socotra Archipelago on the basis of 15 specimens. The Socotra Archipelago in the western Indian Ocean at the entrance to the Gulf of Aden consists of four main islands (Fig. 1). The largest and most easterly island, So- cotra, is located some 400 km south of the Arabian Peninsula and 200 km east of Cape Guardafui, the Horn of Africa. The other islands are Abd al-Kuri in the west and the two smaller southern islands of Samha and Darsah, also known as “‘The Brothers’’. The coastal fishes of the Socotra Archi- pelago were poorly known until recently. Steindachner (1902, 1903) published the first study of fishes from Socotra, based on material collected by the Austrian Expedi- tion to Socotra and South Arabia in 1898— 1899. He listed 57 species of marine and brackish-water fishes. The first detailed study of the fish fauna was a sight survey conducted by Kemp (1998), who provided an account of 215 fish species and a prelim- inary zoogeographical analysis. In 1999 and 2000 the Senckenberg Re- search Institute, Frankfurt, conducted the Marine Habitat, Biodiversity and Fisheries Surveys in the framework of the United Na- tions Development Programme (UNDP) Global Environment Facility funded project ‘“‘Conservation and Sustainable Use of Bio- diversity of Socotra Archipelago,’’ execut- ed under the auspices of the Environmental Protection Council of Yemen. During two expeditions the second author collected fishes at inter- and subtidal stations throughout the archipelago. Among the col- lections were several specimens of the pseudochromid subfamily Congrogadinae, a group of eel-like reef fishes commonly called eel blennies or snakelets. The sub- family was revised by Winterbottom (1986) who recognised 19 species in eight genera (one with two subgenera); four additional species (and one new subgenus) have been described subsequently (Winterbottom & Randall 1994, Winterbottom 1996, Gill et al. 2000). Among the collections from the Socotra Archipelago, an additional new species referable to the western-central In- dian Ocean endemic genus Halidesmus Giinther was discovered, and specimens of the widely distributed western Indian Ocean species Haliophis guttatus (Forsskal) were VOLUME 116, NUMBER 1 [ot Indian Ocean §2°30' 52°00' 53°00' 12°50' 12°40 120 Abd al-Kuri 2 Sta. 173 Samha 12°20' 1240+ 8 Sta. 709 12°00 Ricmle identified. The purposes of this paper are to describe the new species and to document the new record of H. guttatus from the So- cotra Archipelago. Materials and Methods The nomenclature of cephalic laterosen- sory pores follows Gill et al. (2000). Al- phabetical codes for lateral lines in the new Halidesmus generally follow Winterbottom (1982), except that the structure of the ven- tral lateral line necessitates the following modifications (Fig. 2): the apparent homo- log of line G in H. scapularis is represented in H. socotraensis (and H. polytretus) by a median anterior section (termed G1) and a bilaterally paired posterior section (termed Socotra Archipelago 53°30’ 54°00" 54°30", a Darsah Map of the Socotra Archipelago showing collection sites. G2). Vertebral counts are presented in the form precaudal + caudal; the latter are de- fined as vertebrae bearing a haemal spine, and include the terminal urostylar complex. Terminology of ribs and intermuscular bones follows Gill (1998). Other methods of counting and measuring follow Winter- bottom (1982). Institutional codes follow Leviton et al. (1985), except for NHCY-F which is for the fish section of the Natural History Collection of Yemen. Osteological details were determined from x-radio- graphs. In the description of the new spe- cies, counts are given as values or value ranges for all type specimens, followed, Where variation was noted, by values for the holotype in parentheses. Where counts 54 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON leer, 2, lettering scheme. were recorded bilaterally, both counts are presented for the holotype, separated by a Slash; the first count given is the left count. Ratios for morphometric characters and fre- quency distributions for selected meristic characters are given in Tables | and 2, re- spectively. Comparisons with other Hali- desmus species are based primarily on data provided by Winterbottom (1982, 1986) and Winterbottom & Randall (1994). Halidesmus socotraensis, new species Socotran Snakelet Figs. 2—4, Tables 1-3 Holotype.-—SMF 29223, 64.3 mm SL, Socotra Archipelago, Abd al-Kuri Island, SY COASt, CA, WUBIN S2040718,, GHIS ion, Diagram of Halidesmus socotraensis in lateral and ventral views showing lateral-line systems and rotenone, U. Zajonz, EN. Saeed, M. Apel and E. Zandri, 4 April 2000 (station 709). Paratypes.—BMNH 2002.1.19.3, 1, 63.3 mm SL, ROM 72697, 1, 60.0 mm SL, SMF 29293, 3, 39.6-—69.5 mm SL, all collected with holotype. Diagnosis.—The following combination of characters distinguishes H. socotraensis from all other congrogadines: three lateral lines (dorsal, middle and ventral complex- es); ventral lateral line branched on abdo- men, with unpaired median section on ven- tral midline; and upper lateral line without anterodorsal branch on to nape. Description.—Data for morphometric characters appear in Table 1. Dorsal fin I + 58-61 (1 + 61), all seg- Table 1.—Morphometric values for Halidesmus socotraensis expressed as percentages of standard length (SL). Holotype mm SL 64.3 39.6 Soft dorsal-fin base Ok 80.0 Anal-fin base 61.6 59.8 Snout to first dorsal 7/o3l DOD Snout to soft dorsal 19.8 Ded Snout to anal origin IA 39.1 Head length SD) 15.4 Depth at parietal 7.0 8.1 Depth at anal origin eo 8.8 Eye diameter 2.8 4.0 Snout length 2.8 323) Interorbital width 12 1.3 Upper jaw length 4.5 4.8 Lower jaw length 6.2 Tes Pectoral length 5.6 el Paratypes 48.7 60.0 63.3 69.5 US2 78.3 78.2 Suey 60.6 61.7 63.0 61.2 19.7 eg) 18.3 od OD, (QZ 19.6 20.1 39.4 38.0 32 37.4 14.4 (33) 13.1 W217) 7.6 8.0 7.4 7.5 8.8 8.7 8.5 VD Dodd 30) 3.0 28) 28) 228 3.0 2 1.4 1.3 1.3 3) 4.7 4.5 4.4 4.6 6.8 0.3) 6.3 6.5 4.9 4.7 Dall ae * fin damaged. VOLUME 116, NUMBER 1 PIOP AIOP. SOTP NASP PN AN Fig. 3. JD pTp LLP Halidesmus socotraensis, holotype, SMF 29223, 64.3 mm SL; detail of head showing laterosensory pores, scalation and preserved pigmentation details. Abbreviations: AIOP, anterior interorbital pore; AN, anterior nostril; ATP, anterior temporal pore; DENP, dentary pores; ITP, intertemporal pore; LLP, lateral-line scale pores: NASP, nasal pores; PARP, parietal pores; PIOP, posterior interorbital pore; PN, posterior nostril; POPP, preopercle pores; POTP, posterior otic pore; PTP, posttemporal pore; SCLP, supracleithral pore; SOBP, suborbital pores: SOTP, supraotic pore. Scale bar = 2 mm. mented rays branched; anal fin 45—48 (48), all rays branched; pectoral fins 9—10 (10/ 10), upper 1—2 (2/1) and lower O-2 (1/1) rays unbranched, other rays branched; pel- vic fin absent, though small rod-like pelvis present internally; caudal fin with five dor- sal and five ventral principal rays, and one dorsal and one ventral procurrent rays; branched caudal fin rays 5 + 4. Caudal fin fully connected by membrane to last ray of dorsal and anal fins. Vertebrae 17—18 + 46— 49 = 63-66 (17 + 49); ribs present on pre- caudal vertebrae 3 through 4—5 (3 through 5); epineural bones present on precaudal vertebrae | through 3—5? (1 through 3?; bones difficult to distinguish on x-radio- graphs); hypural 5 absent; hypurals 3—4 fused to one another and to urostylar com- plex; hypurals 1—2 fused to one another and, variably, to parhypural; epurals 1; su- praneurals 2. Cephalic sensory pore openings (Fig. 3; all pores bilateral unless otherwise indicat- ed): nasal two, one pore just posterior to upper lip, second pore just above posterior nostril; anterior interorbital usually one, small second pore present unilaterally in one paratype, about two-thirds distance from typical anterior interorbital pore to posterior nasal pore; median (unpaired) posterior interorbital one, positioned above or slightly behind vertical through posterior edge of eye, additional posterior interorbital pore present in one paratype, displaced lat- erally in line with right side anterior inter- orbital pore; supraotic one; posterior otic usually one, two pores present unilaterally in one paratype; suborbital usually eight, 56 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. pelago. nine unilaterally in one paratype; preoper- cular seven; dentary usually four, three uni- laterally in one paratype; intertemporal usu- ally one, two unilaterally in two paratypes; parietal two, canals from left and right side of head not communicating in midline; an- terior temporal one; posttemporal usually one, two unilaterally in one paratype; su- pracleithral one. Gill membranes fused to one another ventrally but free posteriorly from isthmus; branchiostegal rays 6; outer (anterior) gill rakers on first arch 2 + 6—7 (examined in three largest paratypes only); pseudobranch filaments 6—7 (examined in three largest paratypes only). Lateral-line system com- plex (Fig. 2), with pores on either dorsal or ventral margins of lateral-line scales (or in the case of median lateral lines, on either left or right margins of lateral-lines scales): line A absent; line B + C (bilateral) with 186-210 (210/206) pored scales, extending from shoulder to beneath vertical through base of segmented dorsal ray 49—54 (51/ 51); line D (bilateral) with 177—205 (202/ 197) pored scales, extending from just be- hind upper edge of pectoral-fin base along midside to or almost to caudal-fin base, not connecting anteriorly with line B + C; line E (median) with 6—12 (9) pored scales, ex- tending from just beneath branchiostegal membranes to vertical beneath pectoral-fin base; line F (bilateral) with 78—90 (90/89) pored scales, extending posterodorsally from junction of lines E and G1 almost to pectoral fin, then along lower part of ab- domen to junction with line G2; line G1 (median) with 59—63 (62) pored scales, ex- tending from junction with lines E and F along ventral edge of abdomen to just in front of anus; line G2 (bilateral) with 10— Halidesmus socotraensis, holotype, SMF 29223, 64.3 mm SL, Abd al-Kuri Island, Socotra Archi- 14 (11/14) pored scales, connecting poste- rior terminus of line GI with junction of lines F and H; line H (bilateral) with 125— 152 (152/147) pored scales, extending from junction of lines F and G2 along lower part of body to vertical through base of anal ray 36—42 (41/42). Olfactory capsule with two openings; an- terior opening a short tube, positioned about midway between posterior opening and edge of lip; posterior opening with slightly raised rim, positioned near anterodorsal rim of or- bit (Fig. 3). Cheek, upper edge of operculum (above horizontal through opercular spine) and body with small cycloid scales, those on upper part of operculum imbedded and in- conspicuous; predorsal scales extending an- teriorly to parietal commissure (Fig. 3). Teeth in outer row of premaxilla and den- tary conical, recurved anteriorly, decreasing in size and becoming laterally compressed posteriorly, numbering 11—12 in the pre- maxilla and 13-15 in the dentary; single row of small conical teeth present anteriorly on premaxilla and dentary immediately be- hind outer row teeth; vomer edentate; pal- atine edentate. Live coloration: Not recorded. Preserved coloration (Figs. 3, 4): Head and body dark brown, with indistinct to dis- tinct paler mottling, which may align to form diffuse bars or reticulations; lower part of head and abdomen pale brown; dark grey-brown stripe extending from midside of lower lip through midside of upper lip to mid-anterior edge of eye, then behind mid- posterior edge of eye to upper edge of oper- culum; dark grey-brown stripe edged ven- trally with creamy white stripe; large pale- edged, dark grey-brown spot on shoulder; creamy white median stripe or series of VOLUME 116, NUMBER 1 spots extending from tip of upper lip to or- igin of dorsal fin; dorsal fin dark greyish brown with series of alternating small dark grey-brown and pale brown spots on base of fin, spots encroaching slightly on to body; anal and caudal fins dark greyish brown, sometimes with several pale brown or dark grey-brown spots; pectoral fin pale brown, with small dark grey-brown spot on middle of fin base. Comparisons.—The presence of three lateral lines on the body (dorsal, middle and ventral complexes) places the new species in the genus Halidesmus. There are four other species currently recognised in the ge- nus (Winterbottom, 1982, 1986; Winterbot- tom & Randall, 1994): H. scapularis Giin- ther from South Africa; H. polytretus Win- terbottom from Kenya; H. coccus Winter- bottom & Randall from southern Oman; and H. thomaseni (Nielsen) from Masirah Island (southeastern Oman) to Karachi (Pakistan) and the Bay of Bengal. Charac- ters distinguishing the four species are sum- marized in Table 3. Halidesmus is divisible into two apparent monophyletic groups, which are diagnosed by characters that are uniquely derived among pseudochromids. The first, which consists of H. coccus and H. thomaseni, is diagnosed by a single synapomorphy (fleshy crest on the snout and interorbital area). The other, which consists of H. so- cotraensis, H. scapularis and H. polytretus, is diagnosed by two synapomorphies (ven- tral lateral line with complex branching on abdomen; lateral-line pores opening to ei- ther ventral or dorsal margins of scales). Halidesmus socotraensis differs from H. scapularis in the following features: pelvic fins absent (versus pelvic-fin rays usually 1,2—3, fins rarely absent); lateral line A ab- sent (versus present); lateral line D with 177—205 pores, not joining C (versus with 157-179 pores, joining C); lateral lines E and G median (versus bilateral); medial pa- rietal pores bilaterally paired (versus single median pore); caudal, anal and dorsal fins fully confluent (versus united near fin-ray 7 bases only); and total vertebrae 63—66 (ver- sus 66—69). Halidesmus socotraensis differs from H. polytretus in the following: lateral line A absent (versus present); pores in lateral line D 177-205 (versus 204—216); cheek, nape and opercular scales present (versus ab- sent); preopercular pores 7 (versus 9); sub- orbital pores 8—9, usually 8 (versus 11—13); posterior interorbital pores usually 1, rarely 2 (versus 2); principal caudal fin rays 5 + DE VERSUS) Ona ©) sand seillinakers 2) = 6-7 (Wersusells ©): Relationships among H. socotraensis, H. scapularis and H. polytretus are ambiguous. The presence of lateral-line A is unique among congrogadines to H. scapularis and H. polytretus and suggests a sister relation- ship between the two species. Conversely, the presence of a median section of the low- er lateral line is unique to H. polytretus and H. socotraensis, and suggests relationship between those two species. These relation- ships will be tested in a forthcoming study of the phylogeny and historical biogeogra- phy of Halidesmus by the second author. Habitat and distribution.—All known specimens were collected at a single station in a small enclosed bay on the SW coast of Abd al-Kuri (Fig. 1, Sta. 709). The site is characterised by highly diverse assemblag- es of large massive and encrusting corals, and an associated diverse fish community. Etymology.—The specific epithet refers to the known distribution of the species. Haliophis guttatus (Forsskal) Barred Snakelet ang, DS; Ialolie 2 Haliophis guttatus is known from throughout the Red Sea to the northern Gulf of Aden, southern Oman, the east coast of Africa between Kenya and Mozambique, and the west coast of Madagascar (Winter- bottom 1984, Winterbottom & Randall 1994). Fifteen specimens were collected by the second author and associates in the So- cotra Archipelago. The morphology of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. these specimens agrees well with published descriptions of the species (e.g., Winterbot- tom, 1984, 1986; Randall, 1995). Winter- bottom (1984) noted that the species exhib- its geographic variation in several meristic and morphometric characters, which he in- terpreted as a step cline, with the step oc- curring between about 27°N and 20°N in the northern Red Sea. Frequency distribu- tions of counts relevant to this phenomenon are provided for the Socotran specimens in Table 2. Unfortunately, the specimens are somewhat distorted and difficult to measure accurately; therefore, we do not provide rel- evant morphometric details. Generally, me- ristic values agree well with data presented by Winterbottom for specimens from the southern Arabian area (central and southern Red Sea and Gulf of Aden), except for counts of precaudal vertebrae, which more closely approached values for specimens from Mozambique and Madagascar. How- ever, sample sizes for many of the localities Haliophis guttatus, SMF 29198, 59.5 mm SL, Abd al-Kuri Island, Socotra Archipelago. reported by Winterbottom are small (often five or fewer specimens). Material.—SMEF 29224, 1: 64.8 mm SL, Socotra Archipelago, Abd al-Kuri Island, SW coast) cay 127138" N 52-0472 6— omar rotenone, U. Zajonz, EN. Saeed, M. Apel and E. Zandri, 4 April 2000 (station 709); SME 29198, 7: 38.0—79.5 mm SL, NHCY- F uncataloged, 7: 43.2—66.0 SL, Socotra Archipelago, Abd al-Kuri Island, E coast, Ras Anjara bay, ca. 12°10’N 52°22’E, 3-9 m, rotenone, U. Zajonz and M. Apel, 8 April 1999 (station 173). Discussion The fish surveys of 1999 and 2000 com- prised 18 large sampling stations scattered all around the Socotra Archipelago. Sam- ples of congrogadine species were obtained only at Abd al-Kuri, whereas pseudochrom- ids of the subfamilies Pseudochrominae (six species of Pseudochromis Rippell) and Table 2.—Frequency distributions for selected meristic characters of Halidesmus socotraensis and Haliophis guttatus from the Socotra Archipelago. Bilateral counts of pectoral rays are included. Segmented dorsal rays 44 // 58 59 socotraensis guttatus 13 socotraensis Di) Caudal vertebrae // 47 guttatus 46 socotraensis gultatus Total caudal rays Anal rays 35 36 45 ~ Oo Precaudal vertebrae 14 14 // 13 py Total vertebrae 63 Sy VOLUME 116, NUMBER 1 Sb ar Gell 89-79 = OS Sr + 8I-9I juonyUuos AT[Ny SS juasqe 6—-L OS-SY p9-8S + I DAROUOD ‘padojaAadp-][OM I posed Ay]jeroye]Iq 8 It yuosoid juosoid juosoid V/N Q utof 1,ussop :sorod g61—O9T YIM juosqe poyouriqun SUISICUL Q[VOS IOLID}sod 0} usdo AS arn ee €L-69 = SSIS + 8I-LI yuonyuos AT[NJ GC OS-E juasqe CI-6 CS—-CS 89-79 + I X9AUO0D ‘pado][aAap-[][OM I posed Ayye197e]1q 8 [L quasqe juasqe juasqe V/N > ulof },uss0p :Sorod poI-ELT WIM juasqe poyouriqun SUISICUL g[evos IOLIa3sod 0} uado O) ap SG 69-99 = ¢S-6r + 8I-9I Ajuo Ay -jJeseq poilun souriquioul ZS *¢ *T Aliens) yuos -qe Ajarer ‘¢—Z‘]T Ayyensn 6 Ajyensn “QT-8 IS-8YP €9-09 + I juasqe 10 yeom I Je[nsuls ‘ueIpoul 8 Ajjensn *g—/ g Ajyensn *g—/ juasqe quosoid juosoid posed Ayyesajeyiq OD sutol :sorod 6/I-LS] UitM juosoid poyour.iq SUISIVUI aTBOS ]esiop 10 [eNUdA 0} usdo Gar VIED = STL ar OI juonyuos Ay[ny IOS 2 1-0 juasqe OI-6 9V-ST ES sp II juasqe IO yeom C poired Ay]yesoje]1q cI-Il 6 juosqe juosqe juosqe uvIPOU > ulof },ussop "sarod 91 7—VOT UM quasoid poyourig SUISIVUI OTROS [esIop JO [eNUdA 0} Uusdo L2) ae AE) = OV D7 ae till yuonyuos AT[Ny eS Se I yuasqe OI-6 gp Ayjensn ‘gp—Cr IBS ar It juasqe IO yeam [ Ajjensn ‘7—] posted Ayye19}e]1q g Ajjensn ‘6-8 [L. quosoid quosoid quosoid uvIpoul J utofl },uss0p :Sorod ¢QT-LLI WM juosqe poyour.iq SUISIUL 3[BOS [esiop 10 [eNUdA 0} usdo Stoyed [ID IeIGOII A, suy [es1op pue [eure ‘jepned uy jepned uy SIAJOd uy [e101909g uy [euy uy [esiog ysoioa AYSO[4 soiod [PIIGIOIDUL IOLIDISOg saiod jejoued [eIpayy saiod [eiigioqns soiod repno1adoaig sgjeos repnosiedO soyeos aden sgjeos yaoyD a6nos pue 6gGlos Ssoul’] Gurl Veco OUI] [eIOIe] [eQUIA SOUT] [e1O}e] UL SaI0g 1UISDUOY] SHIIOI siipjndvos snjasjKjod SISUADAJOIOS Joovieyo ss ee ‘soroads sniusapypy Sulysinsunsip siojovsreyo Jo AreumuNg—¢ 2] Qe], 60 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Pseudoplesiopinae (two species of Chlid- ichthys Smith) were collected from other is- lands as well. It seems possible that Abd al- Kuri receives a higher influx of pelagic eggs and larvae from East Africa through the Somali current compared to the other islands, enriching the fish communities. In particular along the south and east coasts there are isolated biodiversity “hotspots,” which may act like traps for recruits by pro- viding shelter and suitable habitats (U. Za- jonz, unpubl. data). Both sites where con- grogadines were collected host particularly rich fish and coral assemblages as com- pared to other areas visited at Socotra. The fish species compositions were exception- ally rich, suggesting that there is limited faunal exchange between these patches and the wider archipelago, at least of closely reef-associated species. Acknowledgments We would like to thank the staff of the Project Implementation Unit (PIU), espe- cially C. Cheung and E. Zandri, the repre- sentatives of the Environmental Protection Council (EPC) of Yemen, and the United Nations Office for Project Services (UN- OPS YEM/96/G32) for their assistance and support. All team members that helped in the field are cordially thanked. J. Seigel checked details on the lateral-line system of the holotype of Halidesmus polytretus, for which we are grateful. P. Hurst photo- graphed the specimens in Figs. 4 and 5. R. Winterbottom kindly discussed details of congrogadine systematics, reviewed a draft of the manuscript and provided useful sug- gestions. Literature Cited Gill, A. C. 1998. Homology of the anterior vertebrae, ribs, and dorsal fin pterygiophores and rays in congrogadine fishes (Perciformes: Pseudochro- midae).—Copeia 1998:1041—1045. , R. D. Mooi, & J. B. Hutchins. 2000. Descrip- tion of a new subgenus and species of the fish genus Congrogadus Gtinther from Western Australia (Perciformes: Pseudochromidae).— Records of the Western Australian Museum 20: 69-79. Kemp, J. M. 1998. Zoogeography of coral reef fishes of the Socotra Archipelago.—Journal of Bio- geography 25:919—933. Leviton, A. E., R. H. Gibbs, Jr, E. Heal, & C. E. Daw- son. 1985. Standards in herpetology and ichthy- ology: Part 1. Standard symbolic codes for in- stitutional resource collections in herpetology and ichthyology.—Copeia 1985:802-—832. Randall, J. E. 1995. Coastal Fishes of Oman. Crawford House Publishing, Bathurst, 439 pp. Steindachner, FE 1902. Wissenschaftliche Ergebnisse der stidarabischen Expedition in den Jahren 1898 bis 1899. Fische von Siidarabien und So- cotra.—Anzeiger der Akademie der Wissen- schaften Wien 39(24):316-318. . 1903. Fische aus Siidarabien und Sok6tra.— Denkschriften der Kaiserlichen Akademie der Wissenschaften, Wien 71:123—168. [dated 1902, but actually published in 1903] Winterbottom, R. 1982. A revision of the congrogadid fish genus Halidesmus (Pisces: Perciformes), with the description of a new species from Ken- ya and a list of the species included in the fam- ily—Canadian Journal of Zoology 60:754—763. . 1984. Revision of the congrogadid Haliophis (Pisces: Perciformes), with the description of a new species from Indonesia, and comments on the endemic fish fauna of the northern Red Sea.—Canadian Journal of Zoology 63:209— ONT. . 1986. Revision and vicariance biogeography of the subfamily Congrogadinae (Pisces: Perci- formes: Pseudochromidae).—Indo-Pacific Fish- es 9:1—34, pl. I. [dated 1985, but actually pub- lished 1986] . 1996. A new species of the congrogadin ge- nus Rusichthys from southern Oman (Percifor- mes; Pseudochromidae), with notes on its os- teology.—Canadian Journal of Zoology 74: 581-584. , & J. E. Randall. 1994. Two new species of congrogadins (Teleostei; Pseudochromidae), with range extensions for four other species.— Canadian Journal of Zoology 72:750—756. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):61—81. 2003. Calocidaris micans (Cidaridae) and Pseudoboletia maculata (Toxopneustidae): additions to the sea urchin fauna (Echinodermata: Echinoidea) of the Gulf of Mexico Richard L. Turner and Bruce D. Graham (RLT) Department of Biological Sciences, Florida Institute of Technology, 150 West University Boulevard, Melbourne, Florida 32901-6975, U.S.A., e-mail: rturner @ fit.edu; (BDG) Marine Resources, Inc., 7897 S. W. Jack James Drive, Suite A, Stuart, Florida 34997, U.S.A., e-mail: bgrahammri @ aol.com Abstract.—Iwo sea urchins (Calocidaris micans, Pseudoboletia maculata) were collected and photographed in the Gulf of Mexico between 1978 and 1998. These constitute new records for their distributions. Calocidaris micans has been reported previously off the northwestern coast of Cuba, in the Yucatan Channel, and off Barbados. One of our specimens was photographed and col- lected in the northeastern Gulf of Mexico off Alabama at 100 m. Unpublished museum holdings (USNM) add a second record off Louisiana at 129-144 m. Pseudoboletia maculata is distributed in the Pacific Ocean from Ceylon to Japan, in the South Atlantic only at Ascension and St. Helena islands, and in the North Atlantic from Venezuela, Barbados, and off the Atlantic coast of the United States from Florida to North Carolina. An unpublished document of a federal agency reported this sea urchin from the western Gulf of Mexico. Our collections and photographs from off Pensacola, Florida, at ca. 40 m depth revealed that P. maculata occurs there in mixed-species aggregations with the echinoid Lytechinus variegatus. These records bring the echinoid fauna of the Gulf of Mexico to 61 species, only one of which is endemic and 39 of which have West Indian-Caribbean distributions. Although these records probably are not range extensions within recorded human history, they might represent a post-glacial (re)invasion of the Gulf of Mexico within the last 9000 yr as patchily distributed live-bottom biotopes at intermediate depths became avail- able. These depths and biotopes—poorly sampled in the past—have become more accessible for survey using current sampling technology. In his review of the echinoid fauna of the Gulf of Mexico, Serafy (1979) listed 89 Species, an imenease ‘of 23) ispecies) from Clark’s (1954) list. Serafy pointed out that many of Clark’s species are not part of the Gulf fauna proper but have been collected only on its periphery, belonging instead to faunas of the Caribbean, Bahamas, eastern Florida, and the Florida Keys; Serafy’s list is annotated to reflect the species restricted to adjacent waters. Apparent inconsisten- cies in Serafy’s two tabulated lists give 53— 57 species of echinoid in the Gulf of Mex- ico proper: Cidaris abyssicola (Cidaridae), Araeolampas atlantica (Echinothuridae), and Pourtalesia miranda (Pourtalesiidae) are not marked as peripheral species, al- though their distributions would make them so by his criteria; Echinometra viridis 1s listed only as peripheral in his table | per- haps by lapsus because Serafy recorded it later in table 2 from several regions within the Gulf. Hulings’s (1955) report of EF. vir- idis from the northwestern Gulf of Mexico confirmed the presence of this species, al- though Pomory (2002) does not list it from 62 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the Texas coast. If, then, some interpretation is allowed, the known echinoid fauna of the Gulf stands at 54 species based on Serafy (1979), although numerous misidentifica- tions reported by Turner & Norlund (1988) are cause to question Gulf records of Bris- sopsis elongata. Defenbaugh (1976) identified specimens of Echinocardium from the northern Gulf of Mexico as the European species E. cor- datum and E. flavescens rather than as E. laevigaster reported by Clark (1954) and Serafy (1979). Harry (1979) believed that Defenbaugh’s specimens were misidenti- fied. Serafy (1979) did not list the toxo- pneustid Pseudoboletia maculata, which was recorded by Pawson (1978) from southeastern Florida and listed and illus- trated from the western Gulf of Mexico (Texas A&M University 1981). The reanal- ysis of Mellita quinquiesperforata by Har- old & Telford (1990) resulted in its restric- tion as a western Gulf species and in the elevation of M. gquinquiesperforata tenuis to species in the eastern Gulf. Hopkins et al. (1991) added Tretocidaris bartletti from the northern Gulf off the coast of Alabama; this species was one of Serafy’s (1979) periph- eral species. Most recently, the record of Cidaris abyssicola off the northern coast of the Yucatan Peninsula by Barbosa-Ledesma et al. (2000) places this echinoid well with- in the boundaries of the Gulf; and these au- thors also found Cidaris rugosa off the western coast of the Yucatan Peninsula and Stereocidaris ingolfiana off the northern coast. In the present report, we confirm Pseu- doboletia maculata as a part of the Gulf echinoid fauna with the find of a dense ag- gregation off western Florida, and we add Calocidaris micans from one specimen off Alabama and a second off Louisiana. These additions bring to 61 the echinoid species recorded from the Gulf of Mexico. Materials and Methods Sea urchins were obtained for this study from Continental Shelf Associates, Inc., Ju- piter, Florida, U.S.A., under contract with the U.S. Department of the Interior, Min- erals Management Service (Calocidaris mi- cans) (Continental Shelf Associates, Inc., & Texas A&M University 2001) and with Chevron U.S.A., Inc., New Orleans, Loui- siana (Pseudoboletia maculata) (Continen- tal Shelf Associates, Inc. 1996a, 1996b); from the Texas Cooperative Wildlife Col- lection of Texas A&M University (TAMU); and from the National Museum of Natural History (USNM) of the Smithsonian Insti- tution. Details of collecting localities are given under descriptions of the material. Other records for the Western Atlantic Ocean are based on holdings of USNM, Museum of Comparative Zoology (MCZ) of Harvard University, Harbor Branch Oceanographic Museum (HBOM) of Har- bor Branch Oceanographic Institution, Cal- ifornia Academy of Sciences, and the Mu- seum of Natural History (BMNH) in Lon- don. In most cases, the specimens were ex- amined by us. Calocidaris micans was collected off Louisiana, U.S.A., with the manipulator arm of R/V Johnson-Sea-Link I submersible operated by HBOI in 1989. R/V Edwin Link served as the tender. The R/V Tommy Munro served as the tender during survey photography and specimen collection of C. micans off Ala- bama, U.S.A., in 1998. Photographic data and the voucher specimen were collected using a Benthos® open frame SeaROVER remotely operated vehicle (ROV) equipped with color-imaging scanning sonar, manip- ulator arm, video and 35-mm still cameras, quartz-halogen lamps, and a strobe. The Photosea® underwater still camera and strobe were triggered manually by the on- board scientist. The voucher specimen of C. micans was collected using the ROV ma- nipulator arm. Pseudoboletia maculata was photo- graphed and collected from the East Flower Garden Bank in the NW Gulf of Mexico by Texas A&M University in 1978. The sub- mersible DRV Diaphus was equipped with VOLUME 116, NUMBER 1 video and still cameras, a manipulator arm, sediment scoop, and suction sampler (Texas A&M University 1981). The M/V Mr. Offshore was used during survey operations by Continental Shelf As- sociates, Inc., off Pensacola, Florida, U.S.A., in 1992 and 1993. Pseudoboletia maculata was photographed from a towed multicamera system with both video and 35-mm still cameras. Still photographs were taken with a Benthos® 1000 under- water camera and strobe. Specimens were collected by a 7.5-m semi-balloon “*mon- goose”’ trawl. Results Order Cidaroida Claus, 1880 Family Cidaridae Gray, 1825 Genus Calocidaris H. L. Clark, 1907 Calocidaris micans (Mortensen, 1903) (Figs. 1—2) Porocidaris sharreri Agassiz, 1880:71 (part.). Dorocidaris micans Mortensen, (original description), 28 (list). Calocidaris micans: Mortensen, 1928:312— 314 (for complete synonymy, key, de- scription), text-fig. 19-2.—Clark, 1954: 374 (list).—Fell, 1966:U333 (diagno- sis)—Phelan, 1970:7-9 (key, descrip- tion), text-fig. 2, pls. 2, 3.—-Serafy, 1979: 10, 14, 109 (key, geographic and bathymetric distributions). Non-Calocidaris micans: Downey, 1968:62 (see note in Phelan 1970:16). 1903:23 Material examined.—Gulf of Mexico: USNM E47929, 1 spec., dry, 65/76 mm test height/test diameter (TH/TD), R/V Edwin Link, R/V Johnson-Sea-Link I, dive 2585, Green Canyon, off Louisiana, U.S.A., 27°44'37'N, 91°7'54”"W, 129-144 m, 5 Sep 1989; USNM 1002220, 1 spec., alcoholic, 58/73 mm TH/TD, R/V Tommy Munro and remotely operated vehicle, Continental Shelf Associates, Inc., job 1600, cruise 3, site 4, sample 335, off Alabama, U.S.A., 29°19'39"N, 87°46'7.8"W, 100 m, manipu- 63 lator arm, 28 Aug 1998.—Western Atlantic Ocean: MCZ 283, | spec., dry, 41/58 mm TH/TD, USCSS Blake, sta. 297, off Bar- bados, 123 fm [225 m], 1879; USNM 10705, 1 spec., alcoholic, 28 mm TD, R/V Albatross, sta. 2348, Straits of Florida, off Havana, Cuba, 23°10'39"N, 82°20'21’W, 211 fm [386 m], 20 Jan 1885; USNM 10717, 2 spec., dry and alcoholic, 48 mm TD (dry), 50 mm TD (alcoholic), R/V AI- batross, sta. 2354, Yucatan Channel, off Cozumel Island, Mexico, 20°59'30’N, 86°23'45"W, 130 fm [238 m], 22 Jan 1885; MCZ 7720, 1 spec., dry, 41/58 mm TH/TD, R/V Atlantis, sta. 3305, off Playa Baracoa, Havana Province, Cuba, 330 fm [604 m], 23 Mar 1939; USNM E13068, | spec., dry, 77 mm TD, R/V Gerda, cruise 6433 (30th biological cruise), sta. 388, NW corner of Little Bahama Bank, Bahamas, 27°18'0"N, 79°12'0"W, 320 m, 19 Sep 1964; USNM E13069, 2 of 3 spec., dry, 53 and 70 mm TD, R/V Gerda, cruise 6717, sta. 899, Ar- rowsmith Bank, Yucatan Channel, off Yu- catan Peninsula, Mexico, 20°57’'0’N, 86°34’0"W, 40-165 m [one tag reads “‘102 mien lOL Sep. 1967-2 USNMEEI3025 lotr spec., dry, 56 mm TD, R/V Pillsbury, cruise 6802, sta. 595, Arrowsmith Bank, Yucatan Channel, off Yucatan Peninsula, Mexico, 21°8’30”N, 86°27'0"W, 33-586 m, 15 Mar 1968; USNM E13067, 3+ spec., dry, 17, 34, and 70 mm TD, R/V Pillsbury, cruise 7001, sta. 1141, off S coast of Great Inagua Island, Bahamas [one tag _ reads “(Carilolocain, ive, Jelaniiie |, 20> a2 Ol 73° VA OW, 403-458 m, 13 Jan 1970; HBOM 72:307, 1 spec., dry, 37/46 mm TH/ TD, R/V Seward Johnson, cruise 118, R/V Johnson-Sea-Link IT, dive 579, W of Nas- sau Harbor, New Providence, Bahamas, 25°7.1'N, 77°26.1'W, 488 m, manipulator, 30 Mar 1981; HBOM 72:339, 1 spec., al- coholic, 46/58 mm TH/TD, R/V Johnson- Sea-Link I, dive 1306, off Wood Cay, N of Grand Bahama Island, Bahamas, 274 m, manipulator, 7 Dec 1982; USNM E32546, 1 spec., dry, 57 mm TD, R/V Seward John- son, cruise 157, R/V Johnson-Sea-Link I, 64 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON dive 1357, W of Wood Cay, Bahamas, 26°42'36’N, 79°1'42"W, 244-309 m, 14 Jun L983.U SNM E3255 1Reespec. digys O2s mam TD, R/V Seward Johnson, cruise 157, R/V Johnson-Sea-Link I, dive 1359, W of Wood Cay, Bahamas, 26°42’48"N, 79°9'30’W, 618—624 m, 15 Jun 1983; USNM E32547, 1 spec., dry, 37 mm TD, R/V Seward John- son, cruise 159, R/V Johnson-Sea-Link I, dive 1498, French Bay, San Salvador Is- land, Bahamas, 23°56'0"N, 74°32'54’"W, 436—468 m, 21 Oct 1983; USNM E32550, | spec., dry, 50 mm TD, R/V Seward John- son, cruise 159, R/V Johnson-Sea-Link T, dive 1500, off Cockburn Town, San Sal- vador Island, Bahamas, 24°2'48’N, 74°32'30"W, 1600 ft [488 m], 22 Oct 1983; USNM E32548, 1 spec., dry, 34 mm TD, R/V Seward Johnson, cruise 159, R/V Johnson-Sea-Link I, dive 1506, Bonefish Bay, San Salvador Island, Bahamas, 24°4'42"N, 74°33'6"W, 914 ft [279 m], 25 Oct 1983; USNM E32606, 1 spec., alco- holic, 16 mm TD, R/V Seward Johnson, cruise 161, R/V Johnson-Sea-Link IT, dive 805, off Sandy Point, Great Abaco Island, Bahamas, 25°36'12”"N, 76°44'30’"W, 473 m, 7 Apr 1984; USNM E32580, | spec., dry, 48 mm TD, R/V Seward Johnson, cruise 161, R/V Johnson-Sea-Link II, dive 807, S of Rock Point, Great Abaco Island, Baha- mas, 25°59'30"N, 77°24'6"W, 300 m, 8 Apr 1984; USNM E32511, 1 spec., dry, 73 mm TD, R/V Seward Johnson, cruise 161, R/V Johnson-Sea-Link IT, dive 816, Chub Cay, Berry Islands, Bahamas, 25°23'42'N, 77°54'30"W, 226 m, 14 Apr 1984; HBOM 72:505, 1 spec., dry, 24/34 mm TH/TD, R/V Seward Johnson, R/V Johnson-Sea- Link I, dive 2001, San Salvador Island, Ba- hamasee24~ lo sOuNew ADO We Si/2 me a= nipulator, 21 Apr 1987; HBOM 72:498, 1 spec., dry, 43/54 mm TH/TD, R/V Seward Johnson, R/V Johnson-Sea-Link I, dive 2007, San Salvador Island, Bahamas, 23°0.7'N, 74°33.0’W, 263 m, manipulator, 24 Apr 1987; HBOM 72:743, 1 spec., dry, 48/63 mm TH/TD, R/V Seward Johnson, R/V Johnson-Sea-Link IT, dive 1738, 2.46 nm off Needham’s Point, bearing 044°, Bar- bados, 200-207 m, manipulator, 20 Apr 1989. Other sources.—In addition to specimens examined by us, we have relied on station data for Calocidaris micans deposited in California Academy of Sciences, HBOM, and BMNH. Descriptions of stations and material.— The specimen of C. micans (USNM E47929) taken off the coast of Louisiana at 129-144 m depth was collected by sub- mersible. The following description of the site is based on notes written by S. D. Cairns (in litt.) during the dive. Bottom temperature was 17°C. The site was a ridge oriented from northwest to southeast with a slope of 45° at least on one side. The slope was partly composed of large, flat, imbri- cated, calcareous slabs that Cairns com- pared to plates of deep-reef Agaricia. The substratum consisted also of gravel and of rocks of various sizes and shapes. A variety of scleractinian and antipatharian corals were Observed or collected (Cairns et al. 1993), including a new species of black coral (Opresko & Cairns 1992). Other ma- terial collected were three slit shells, anoth- er unspecified gastropod, and a crinoid; col- lection of C. micans was not mentioned in Cairns’s field notes. The specimen at hand had a test diameter (TD) of 76 mm and a test height (TH) of 65 mm, giving the test a high globose shape (TH:TD = 0.86) typical of C. micans. All primary spines above the ambitus were bro- ken except one (137 mm long; = 1.8 TD); Phelan (1970) reported that primary spines of C. micans can be up to 3 TD, although more commonly they are up to 1.5 TD. In addition to morphology of the test and ped- icellariae (Mortensen 1928, Phelan 1970), all primary spines had the smooth porce- laneous surface (Fig. 1B, C) that is unique to this species (Phelan 1970). The spines were free of epizoics. Although cidaroid spines are known to harbor epizoics (Hy- man 1955), the spines of most C. micans VOLUME 116, NUMBER I 65 C=) te] = fe = & eo £ Fig. |. Calocidaris micans. A. USNM 1002220 on ledge at collection site off Alabama, with a colony of the hard coral Madrepora carolina to left, the hard coral Madracis myriaster to the upper left, two unidentified comatulid crinoids to the right, several individuals of the black ahermatypic coral Rhizopsammia manuelensis, and the seastar Chaetaster nodosus at the bottom; note epizoics on some of the primary spines of the echinoid; courtesy of Continental Shelf Associates, Jupiter, Florida. B. USNM E47929, showing long, terete, smooth primary spines on ambitus and near apex of test. C. USNM 1002220, oral surface, showing smooth oral primary spines (arrows). 66 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON examined by us were rarely covered with epiZzoIcs. The specimen of C. micans (USNM 1002220) was photographed and collected off Alabama, U.S.A., at 100 m depth on a high-relief (>2 m) hard-bottom feature de- scribed by Ludwick & Walton (1957) as a trend of discontinuous reef-like pinnacles located near the edge of the continental shelf between the Mississippi River delta and DeSoto Canyon. The sampling site was the apex of a relatively large ridge gradu- ally sloping to the west-southwest and to the north. Visually and numerically domi- nant taxa at the site were hard corals (Rhi- zopsammia manuelensis, Madracis myrias- ter, Madrepora carolina), soft corals [Cten- ocella (Ellisella) spp., Nicella spp.], anti- patharians (Antipathes spp.), crinoids, basketstars, and various unidentified aher- matypic hard corals. The in-situ photograph (Fig. 1A) of the specimen at hand showed it on a ledge that was lightly coated with sediment. Associ- ated fauna visible in the photograph were a variety of anthozoans, crinoids, a seastar, and a galatheid crab. Most primary spines at and above the ambitus were broken, and little additional damage was inflicted upon collection of the specimen. The animal in life had pale yellow interambulacra and pal- er ambulacra, scrobicular spines, and regen- erating primary spines. The primary spines otherwise were medium brown, much dark- er basally, and largely unbanded. Many pri- mary spines of the live specimen exhibited a fuzzy coating, presumably of epizoics; but the only obvious epizoics on the alcoholic specimen were occasional patches of the fo- raminiferan Homotrema rubra. Among oth- er C. micans examined, One specimen from Barbados (HBOM 72:743) had spines with barnacles and tubicolous polychaetes. The test of the specimen from Alabama was glo- bose (TH:TD = 0.79). Remarks.—Calocidaris micans was. first collected off Barbados (Fig. 2) at 225 m depth by USCSS Blake in 1879 and was one of three specimens in Agassiz’s (1880) type material of Porocidaris sharreri. {It was not collected again off Barbados for 110 yr when one specimen (HBOM 72:743) was collected by TI.’ Askew at a depth of 200—207 m.] In 1885, R/V Albatross took two specimens each from the Straits of Florida (337—386 m) and the Yucatan Channel (238 m). One spec- imen from the Straits of Florida (NHM 1898.5.3.583) is Mortensen’s (1903) holotype of Dorocidaris micans. It was in his report on echinoids of the Danish Ingolf Expedition that Mortensen (1903) pointed out Agassiz’s (1880) error. Clark (1907:211) erected Calo- cidaris for D. micans, naming it presumably because it was “the most beautiful echinoid I have ever seen,’ an evaluation with which Mortensen (1910) later concurred. Clark (1907) characterized his new genus partly by the smooth, highly polished, porcelain-like primary spines (Fig. 1B) and by the absence of serrations (spinules) even on the oral pri- maries (Fig. IC). Phelan (1970) viewed the morphology of the primary spines as diag- nostic of Calocidaris, pointing out that all other cidaroids have serrations on at least some primaries. Although Clark (1925) later synonymized his Calocidaris with Cidaris, Mortensen (1928) retained Calocidaris, as have echinoid systematists since then. The genus remained monotypic and Recent until Cutress (1980) described Calocidaris palmeri from the Middle Miocene of Cuba; Cutress (1980) considered C. micans to have de- scended directly from C. palmeri in a process paralleled in Cidaris and Tretocidaris. For decades, the few specimens of C. micans were known only from the Greater and Less- er Antilles. The cruises of R/V Gerda and Pillsbury in 1964-1970 collected more specimens from the Yucatan Channel (33-586 m) but added the Bahamas (320—458 m) to its dis- tribution (Fig. 2). Many Bahamian speci- mens were collected during cruises of R/V Seward Johnson and R/V Edwin Link using the Johnson-Sea-Link submersibles in 1981-1998 at depths of 226—624 m. Not until 1989, 110 yr after its original discov- ery off Barbados, was C. micans taken from VOLUME 116, NUMBER 1 67 Fig. 2. Distribution of Calocidaris micans based on museum records and literature. Each plot might represent multiple nearby records. the Gulf of Mexico, off Louisiana at a depth of 129-144 m. The second Gulf specimen was photographed in situ (Fig. 1A) and col- lected in 1998 off Alabama at 100 m. Both specimens were from the northern Gulf of Mexico, well within Serafy’s (1979) criteria for inclusion; they are among the largest (73 and 76 mm TD) known specimens of C. micans and from among the shallowest stations (<150 m). Order Temnopleuroida Mortensen, 1942 Family Toxopneustidae Troschel, 1872 Genus Pseudoboletia Troschel, 1869 Pseudoboletia maculata Troschel, 1869 (Figs. 3-5) Pseudoboletia maculata Troschel, 1869:96 (original description). Pseudoboletia atlantica Clark, 1912:344 (original description), 345 (key). Pseudoboletia occidentalis Clark, 1921: 115-118, pl. II (original description). Pseudoboletia maculata: Mortensen, 1943: 528 (key), 532-534 (for complete syn- onymy, description), pl. XLII, figs. 4, 5; pl. LV, figs. 2, 5, 6, 16, 17, 21.—Pawson, 1978:3—5 (listed), 7 (description), 17—20 (description).—Continental Shelf Asso- ciates, Inc., 1979 (station data).—Texas A&M University, 1981:99 (ecology, sta- tion data), fig. X-10 (in situ), table X-C- 29 (listed).—Duke University Marine Laboratory, 1982:tables 5.9, 5.13, Ap- pendix 20 (listed).—Rowe & Gates, 1995:258 (systematics, distribution).— Continental @ Shelf *Assocrates,, Ince 1996a:148 (ecology).—Continental Shelf 68 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Associates, Inc., 1996b:15, 17 (ecology), table 1 (listed), pl. B-12. Pseudoboletia atlantica: Mortensen, 1943: 528 (key), 534-538 (synonymy, descrip- tion), pl. XXXII, figs. 1-5; pl. XXXIX, fig. 6; pl. XL, fig. 6; pl. XLII, figs. 1-3; pl. XLIII, figs. 1, 2; pl. XLIV, fig. 1; pl. IEA, ines I, 33, UY). 2X0), Pseudoboletia occidentalis: Mortensen, 1943:528 (key); 538-540 (synonymy, de- scription). ?Lytechinus variegatus: Continental Shelf Associates, Inc., 1996a:table 5.4 (part., listed). Material examined.—Gulf of Mexico: TAMU 3-1809, 1 spec., alcoholic, 27/51 mm test height/test diameter (TH/TD), DRV_ Diaphus, cruise 78-G9-D5, dive 5 [night dive], submersible, East Flower Gar- den, 27°53’N, 93°38'W, 160-190’ [49-58 m], 30 Oct 1978 [one tag reads “*9/30/78”’ |; USNM 1002222, 1 spec., dry, 34/66 mm TH/TD, M/V Mr. Offshore, Continental Shelf Associates, Inc., job 1390.3 for Chev- ron, Destin Dome Unit, survey 2, trawl 22, off Pensacola, Florida, U.S.A., 29°59'N, 87°11'W, 112’ [34 m], 9 Oct 1992; USNM 1002298, 2 spec., dry, 29/58 & 32/65 mm TH/TD, M/V Mr. Offshore, Continental Shelf Associates, Inc., job 1390.3 for Chev- ron, Destin Dome Unit, survey 2, trawl 22, off Pensacola, Florida, U.S.A., 29°59’N, 87°11'W, 112’ [34 m], 9 Oct 1992; USNM 1002299, 1 spec., alcoholic, 42/81 mm TH/ TD, M/V Mr. Offshore, Continental Shelf Associates, Inc., job 1390.5 for Chevron, Destin Dome Unit, survey 4, trawl 43, sam- ple QSA 2, off Pensacola, Florida, U.S.A., 2975 O25 ENE SI lA a IZ Oa |SSyemni| yaks Apr 1993.—Other Atlantic sites: USNM E4531, 1 spec. [holotype of Pseudoboletia occidentalis Clark, 1921], dry, 28/54 mm TH/TD, State University of lowa Barbados- Antigua Expedition, 1918, off Barbados, 30—100 fm [55—183 m], 1918; MCZ 7583, 2 spec., dry, 32/70 & 43/94 mm TH/TD, coll. T. Mortensen, off St. Helena, ca. 15°58'12”"S, 5°46'36"W [our estimate from Mortensen (1933)], 50 m, Feb 1930; USNM E5953, 1 spec., dry, 42/89 mm TH/ TD, coll. T. Mortensen, off St. Helena, ca. 15°58'12”S, 5°46'36”W [our estimate from Mortensen (1933)], 20 m, Feb 1930; USNM E12353, 4 spec., dry, 31/63—39/79 mm TH/TD, R/V Pillsbury, cruise 6806, sta. 736, off Venezuela, 10°57’0’N, 65°52'0"W, 38-85 fm [69-155 m], 22 Jul 19683 USNM E16096, 1 specridia; [crushed], coll. by “EM. for A. Love- ridge,’ James Bay, St. Helena, 14 Aug 1968; USNM E11732, 1 spec., dry, 39/82 mm TH/TD, coll. by “EM. for A. Love- ridge,”’ wreck of Papanui, James Bay, St. Helena, | Feb 1969; USNM E20593, 1 spec., dry, 5.8/12 mm TH/TD, R/V Pills- bury, cruise 6907, sta. 878, east of St. Vin- cent, St. Vincent and the Grenadines [other tags read “off of Saint Lucia” and “off Windward Is’’], 13°11'18’N, 61°6'30’W, 37-40 m, 6 Jul 1969; USNM E16204, 1 spec., dry, 39/85 mm TH/TD, coll. Rick Guest, off Hillsboro Beach, Florida, U.S.A., 70’ [21 mJ], 7 Mar 1974; USNM E16202, 1 spec., dry, 30/60 mm TH/TD, coll. R. Guest, off Hollywood, Florida, U.S.A., 65’ [20 m], spring 1974; USNM E16203, 1 spec., dry, 33/74 mm TH/TD, coll. R. Guest, off Hollywood, Florida, U.S.A., 65’ [20 m], 6 May 1974; USNM E16246, 2 spec. [1 crushed], 33/75 mm TH/TD, coll. M. Telford, Carlisle Bay, Barbados, 40’ [12 m], 1976; HBOM 72:278, 1 spec., dry, 34/ 74 mm TH/TD, Continental Shelf Associ- ates, Inc., sta. James Island-380, sample 14A-a, Charleston, South Carolina, U.S.A., from 32°34.9'N, 78°34.8'’W to 32°35.2'N, 78°35.0'W, 53 m, biological dredge, 2 Oct 1978; USNM E29871, 1 spec., dry, 35/75 mm TH/TD, R/V Dan Moore, Living Ma- rine Resources Study, sta. OSOS5 (field no. 818118), off North Carolina, U.S.A., 33°48'42’N, 76°34'12”"W, 102 m, trawl, 14 May 1981; USNM E32267, 2 spec., alco- holic, 73 & 74 mm TD, R/V Dan Moore, Living Marine Resources Study, sta. OSOS (field no. 818169), off North Carolina, U.S.A., 33°49'0"N, 79°34'0"W, 69 m [one VOLUME 116, NUMBER 1 tag reads ““33°49.1' 76°34.0’ 66 m’’], trawl, 10 Aug 1981; USNM E29626, 1 spec., al- coholic, 68 mm TD, R/V Dan Moore, Liv- ing Marine Resources Study, sta. OSO5 Gicldenomsisi71). ‘oi North) Carolina: WESPAC 349 301N5 76234, 0IW, 63) m, tramie blAus. 198i WSNM E30528)) 7 spec. [1 crushed], alcoholic, 66-85 mm TD, R/V Dan Moore, Living Marine Resources Study, sta. OSO5 (field no. 818179), off North Carolina, U.S.A., 33°49'24'N, 76°33'24"W, 68 m, trawl [W. W. Kirby- Smith, in litt.; but museum records indicate ““Smith-MaclIntyre grab’’], 11 Aug 1981. Other sources.—In addition to specimens examined by us, we have relied on station data for other Pseudoboletia maculata de- posited in BMNH and HBOM. Descriptions of stations and material.— Pseudoboletia maculata were photographed on and collected from the NW rim of DeSoto Canyon off Pensacola, Florida, in 1992-1993 on a soft substratum that sup- ported a biotal assemblage referred to as a “sand bottom algal community” (Conti- nental Shelf Associates, Inc. 1996a, 1996b). The sand bottom algal community was vi- sually dominated by calcareous red algae (Corallinaceae and Peyssonnelia inamoena) and closely associated with coarse substra- tum consisting of coralline algal rubble, Shell debris, and coarse sand. The substra- tum supporting the sand bottom algal com- munity produced a relatively strong acous- tic side-scan sonar reflection, indicating a coarse grain size and relatively high shell content (Marine Technical Services, Inc. 1985; John E. Chance & Associates 1991a, Ob eOOD): The presence of abundant algae on the sand substratum in water depths of approx- imately 30-55 m supported relatively dense aggregations of grazing echinoids (Astro- pyga magnifica, Eucidaris tribuloides, Ly- techinus variegatus, P. maculata, Styloci- daris affinis). Pseudoboletia maculata was only observed in mixed-species aggrega- tions with the echinoids Lytechinus varie- gatus and Eucidaris tribuloides (Fig. 3A). 69 Other biota observed and collected with P. maculata included the free-living hard coral Oculina tenella, the scallop Pecten rave- nelli, and rock shrimp Sicyonia brevirostris. The algal density within the area exhibited high variability (Continental Shelf Associ- ates, Inc. 1996a) probably due to seasonal changes in light or temperature in combi- nation with other oceanographic conditions. Peyssonnelia inamoena, the dominant alga closely associated with the presence of P. maculata, has been observed to vary sea- sonally in density in other areas where it is found (Schneider 1976). Eight non-overlapping photographs showed 1—18 P. maculata per frame, with occasional Eucidaris tribuloides and possi- ble L. variegatus (Fig. 3A). Many P. ma- culata had debris on their tests. Maculae were not visible in the photographs, but the large size and low hemispherical shape of most of the urchins distinguished them as P. maculata rather than L. variegatus. Many of the P. maculata were clumped, sometimes with spines in contact with neighbor’s spines; but, elsewhere, animals were several test diameters apart. Four Pseudoboletia maculata, test di- ameters 58—81 mm, were collected at two stations 5 nm apart in a sand bottom algal community. The specimens were mixed with Lytechinus variegatus (USNM 1002300 and 1002301), which has a pro- portionately higher test than P. maculata (Fig. 4). Only one of the four specimens had maculae (Fig. 3B), which occurred in three cycles: one cycle of pigmented spots subapical in the interambulacra; two cycles of spots, one just above and one just below the ambitus, each cycle consisting of five large interambulacral and five small ambu- lacral maculae. Each macula was formed by a group of primary and secondary spines, most primary spines dark brown basally, fading to pale green medially and white dis- tally. Otherwise, the spines and test were white. Our observations of spine coloration agree with those of Koehler (1908), Mor- tensen (1943), and Pawson (1978). 70 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. gation of P. maculata with two possible Lytechinus variegatus (arrowheads; based on size and shape) and two Eucidaris tribuloides (arrows); courtesy of Continental Shelf Associates, Jupiter, Florida. B. USNM 1002299, specimen from off Pensacola, Florida, showing maculae. C. USNM 1002222, specimen from off Pensacola, Florida, showing pore pairs in arcs of four and spination of plates in the buccal membrane. In addition to the material off Florida, one specimen (TAMU 3-1809) of P. ma- culata was collected by T. J. Bright in 1978 on a night submersible dive at 49-58 m on the western side of East Flower Garden Bank, a salt dome in the western Gulf of Mexico. The cidaroid Stylocidaris affinis Pseudoboletia maculata. A. Photograph of collection site off Pensacola, Florida, showing an aggre- was the only other echinoid recorded from the dive (Texas A&M University 1981). Submersible observations revealed high densities of P. maculata, Arbacia punctu- lata, and the asteroid Linckia nodosa at 46— 76 m, and one photograph taken in 1979 of the sea floor showed 46-50 P. maculata on VOLUME 116, NUMBER I 100 A o Pseudoboletia maculata a Tripneustes ventricosus — |ytechinus variegatus 80 60 40 Test Height (mm) 20 Ore 2097 405602 80 100 120 140 Test Diameter (mm) Fig. 4. Test dimensions for three species of toxopneustid echinoid. Data for Pseudoboletia maculata are from our material examined, Clark (1925), and Koehler (1908). Data for Tripneustes ventricosus are from Mortensen (1943). The line plots Serafy’s (1979) equation for Lytechinus variegatus using test diameters of his smallest and largest specimens. a ““nodule-covered bottom’”’ at 59 m depth (Texas A&M University 1981). Many of the sea urchins in the photograph held de- bris on their tests, and spines of most were in contact with those of neighboring ur- chins. Based on test diameter (51 mm) of specimen TAMU 3-1809, the density of the aggregation was up to 300 m-’. Maculae were not visible on P. maculata in the pho- tograph nor on the specimen at hand. Although the presence of maculae is not a reliable character in Pseudoboletia ma- culata from the Atlantic Ocean (Mortensen 1943, Pawson 1978), other characters read- ily distinguish this species from other com- mon toxopneustids, viz., Lytechinus varie- gatus and Tripneustes ventricosus. The test has a low, hemispherical, subpentagonal shape, with the ratio of test height (TH) to test diameter (TD) generally <0.50 (Fig. 4) in contrast to the higher dome-shaped test of L. variegatus (Serafy 1979) and even more depressed than the test of the similar T. ventricosus, at least in the range TD = 50-90 mm. Pseudoboletia is the only toxo- pneustid genus bearing spinelets on the buccal plates and other ossicles of the buc- cal membrane (Fig. 3C); Mortensen (1943) described this condition as a “bearded ap- pearance.’ The compound plates of the am- bulacral series are trigeminate (three pore- pairs) in Lytechinus and Tripneustes and [2 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. polyporous (four or five pore pairs) in Pseudoboletia (Fig. 3C); and the compound plates bear one primary tubercle and spine each in Lytechinus and Pseudoboletia but One primary spine only every three or four compound plates in Tripneustes (Mortensen 1943, Serafy 1979). Remarks.—Pseudoboletia maculata (s.s.) has a far broader distribution (Fig. 5) than does Calocidaris micans (Fig. 2) and is al- most pantropical, absent only from the east- ern tropical Pacific Ocean. Mortensen (1943) recognized four species of Pseudo- boletia: P. atlantica Clark, 1912 (S Atlantic Ocean), P. indiana (Michelin, 1862) (Mad- agascar to Hawaii), P. maculata Troschel, 1869 (Ceylon to Philippines), P. occiden- talis Clark, 1921 (West Indies). Pawson (1978) included all Atlantic specimens in P. Distribution of Pseudoboletia maculata in the Atlantic Ocean and adjacent waters based on museum records and literature. Each plot might represent multiple nearby records. Inset shows records for Ascension Island and St Helena in the South Atlantic Ocean. maculata, designated P. occidentalis as a junior synonym, and recognized two sub- species based on the number of pore pairs in each ambulacral arc: P. maculata atlan- tica Clark, 1912 (five pore pairs, S Atlan- tic), P. maculata maculata Troschel, 1869 (four pore pairs, West Indies). More re- cently, Liao & Clark (1995) synonymized P. maculata with P. indiana, effectively rendering Pseudoboletia monotypic; but they did not address the taxonomy of the genus in the Atlantic basin, and others (Rowe & Gates 1995, Lane et al. 2000, Liao 2001) have retained the distinction be- tween species. For their lack of comprehen- sive treatment, we reject the synonymy of Liao & Clark (1995) and tentatively adopt that of Pawson (1978), assigning material from the Gulf of Mexico to P. maculata VOLUME 116, NUMBER 1 maculata because of the presence of four pore pairs per arc (Fig. 3C). The genus must, however, be re-evaluated, particularly with the application of molecular tech- niques, some of which have already been applied to animals from the western Pacific Ocean (Matsuoka 1985, 1986, 1987; Yosh- ino et al. 1989). Pseudoboletia maculata has long been known from the Indian and Pacific oceans (Mortensen 1943), sometimes in commu- nities with other grazing echinoids (Shi- mabukuro 1991). The earliest possible specimens from the Atlantic Ocean might have been taken prior to 1875 by J. C. Mel- liss off St. Helena (Mortensen 1933), but the first definite record was Koehler’s (1908) two specimens taken by R/V Scotia in 1904 off Ascension Island at 40 fm [73 m]. Koehler’s (1908) description of the two specimens formed the basis for Clark’s (1912) new species P. atlantica. [A speci- men collected by E. W. Alexander off St. Helena is designated by museum records as “holotype” of P. atlantica H. L. Clark, 1912 (BMNH 1949.1.18.3). It has a label signed by H. L. Clark and dated “1924,” but the meaning of the date is unclear. Clark (1925) described the specimen but did not list it among the 111 echinoid types in the British Museum (Natural History).] A sec- ond possible discovery at St. Helena was by J. T. Cunningham in 1909 (Mortensen 1933), mistaken for Tripneustes ventricosus (reported as T. esculentus) perhaps because of the low, hemispherical, subpentagonal test bearing fine, short spines. Mortensen (1933, 1943) collected at least 12 P. ma- culata atlantica off St. Helena in 1930 “in patches in great numbers” from a depth of 20 m (Mortensen 1933:467; tags written in Mortensen’s own hand read “20 m’”’ for USNM E5953 but “50 m”’ for MCZ 7583) on a bottom of small stones, Lithothamnion, and “‘loose Furcellaria-like algae,” with which the urchins covered their tests. In 1968-1969, A. Loveridge took additional specimens from St. Helena (USNM E11732; El6096). 73 Pseudoboletia was first collected in the West Indies in 1918 by C. C. Nutting during the Barbados-Antigua Expedition (Fig. 5). The single specimen (USNM E4531), taken off Barbados “in all probability” from 30— 100 fm [55-183 ml], is the holotype of Clark’s (1921) P. occidentalis [but not listed by Downey (1968)], which Pawson (1978) synonymized with P. maculata maculata. Clark (1921:117) hailed this specimen as “certainly one of the most noteworthy re- sults of the Barbados-Antigua Expedition,” and he expressed surprise that the species had not been taken in earlier expeditions. Additional records include two more speci- mens from Barbados collected at 12 m (the shallowest Atlantic record) by M. Telford in 1976 (USNM E16246) and one near St. Vin- cent in 1969 by R/V Pillsbury (USNM E20593). Four specimens from the Carib- bean Sea collected in 1968 off Venezuela by R/V Pillsbury (USNM E12353) were listed by Pawson (1978), along with records off the SE coast of Florida, U.S.A., (USNM E16202, E16203, E16204) in 1974 at 20-21 m. An additional unpublished specimen from Key Largo, Florida, was taken in 1979 on an algal bottom at 51 m (HBOM 72:2572). In 1978—the same year in which the first specimens were found in the Gulf of Mexi- co—three specimens were collected off the coast of South Carolina, U.S.A., at 53-55 m (HBOM 72:278, 72:867); and in 1981, at least 21 specimens were collected in 12 of 24 trawl samples off North Carolina, U.S.A., at depths of 63-102 m (USNM E29626, E29871, E30528, E32267; at least 10 in col- lection of W. W. Kirby-Smith, in litt.), rep- resenting the northernmost Atlantic records to date. The North Carolinian P. maculata inhabited “‘‘live bottom’ rock outcrops with scattered sand in between” (W. W. Kirby- Smith, in litt.; Duke University Marine Lab- oratory 1982). The community was domi- nated in biomass by echinoderms, which clustered at a high level (0.7) of constancy along with sponges and decapod crusta- ceans. Crustose corallines were the only al- gae reported from the station, and a ridge- 74 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON trough system at 100 m depth included rub- ble with large “‘lithothamnion balls” (Duke University Marine Laboratory 1982). It is surprising that Project SEAMAP (National Marine Fisheries Service & Florida Depart- ment of Natural Resources) took no P. ma- culata in 1983-1985 and 1987 from its 450 trawl and dredge stations at 11-549 m be- tween Stuart (27°10'N) and Fernandina Beach (30°42'N), Florida (R. L. Turner, un- publ.). Discussion These records of Calocidaris micans and Pseudoboletia maculata bring the echinoid fauna of the Gulf of Mexico to 61 species (Table 1), add 2 West Indian-Caribbean spe- cies to its fauna, and further reduce its level of echinoid endemism. Of the 47 species that inhabit the continental shelf (O—200 m), 39 species (83%) have West Indian-Carib- bean distributions. Serafy (1979) recorded C. micans, Cidaris rugosa, and Stereoci- daris ingolfiana as slope species, but Cal- ocidaris micans occurs at 100—144 m in the Gulf of Mexico and Cidaris rugosa at 46 m (Barbosa-Ledesma et al. 2000); although Barbosa-Ledesma et al. (2000) did not re- port the depth for their S. ingolfiana, their station coordinates plot well within the 100- m isobath on the continental shelf north of the Yucatan Peninsula. Serafy (1979) con- sidered Brissopsis alta to be the only echi- noid endemic to the Gulf of Mexico. Since his work was published, B. alta has been collected at 12 stations sampled by R/V Delaware II and R/V Chapman during Pro- ject SEAMAP in 1984-1987 between 27°49'N (off Sebastian Inlet, Florida) and 30°20'N (off Jacksonville, Florida) at depths of 177—411 m in the Atlantic Ocean (R. L. Turner, unpubl.). But the addition of Mellita tenuis, restricted to the eastern Gulf of Mexico (Harold & Telford 1990), retains the level of endemism at 1 species out of 61 (<2%). Cutress (1980) considered Cal- ocidaris to be the only cidaroid genus, Re- cent or fossil, endemic to the Caribbean Sea; but our records of C. micans from the Gulf of Mexico and the Bahamas eliminate the endemic standing of the genus. We believe that the new records do not represent range extensions but rather in- creased sampling effort using recent tech- nology (SCUBA, submersibles, ROVs, camera sleds), in some cases in high-relief live-bottom biotopes that were avoided by earlier naturalists, except for the occasional use of rock dredges. This thought parallels those of Hendler & Miller (1984) and Hen- dler & Turner (1987) for four new species of deep-reef ophiuroids from the Caribbean Sea and Gulf of Mexico. Clark (1921:118) in the description of his new Pseudoboletia occidentalis remarked, “It is strange that neither the ‘Blake,’ the ‘Hassler,’ nor the ‘Albatross,’ nor any other collector in the West Indies, has met with the genus but the reason may be that the vessels mentioned did nearly all their collecting outside the 100 fms. line while the other collectors have done very little dredging at any depth. The teeming area between 10 and 100 fms. has scarcely been touched as yet.” Cutress (1980) postulated the extinction of many genera of cidaroid from the Carib- bean after the Cretaceous due to lowered salinity, reduced temperature, and heavy sedimentation from terrigenous sources. These conditions might have prevailed also in the Gulf of Mexico more recently during the Wisconsin Glacial Epoch of 100,000— 14,000 B.P. (reduced temperature) and with the subsequent postglacial flow of meltwa- ter from the Laurentide ice sheet via Lake Agassiz down the Mississippi River basin in the last 14,000 yr (reduced temperature and salinity, heavy sedimentation, altered surface currents; Broeker et al. 1989, Pielou 1991, Gore 1992, Williams et al. 1998). Only in the last 9000 yr might Caribbean echinoids have invaded the Gulf of Mexico to inhabit new biotopes of the recently flooded continental shelf. Live-bottom echi- noids such as Calocidaris micans and Pseu- doboletia maculata would have found suit- able habitat only in patches of high relief VOLUME 116, NUMBER 1 MAH VO OvS—-06 DUDISIOS DIUAIVS oepliusles Ayre ePplolus[es IapiO (SQ61) IIS 2% AJesas “(Or~e]) UssudIO/I LM VS MH VO VV QOOE-OTZL WNAD]JIUY DWUWAPPYIPOIS A] oepnewmopeipopidsy Ajue 4 (CQ61) JON 2 UosMmed LM VS MH VO OLE-EE snpnsuizigns snuidsisuo] snunydajsosjuag LM MA VO Q2-T I DIYUSYU DSKdOAISY IM VS 4d VO Hd VV OOr—-0 UTADIPIJUD DUOPOICT oepnewoprigq Ajimey eplojewapeiq JapiO (CES) UOSUDIOPI VM VO 0081-007 120qg8381s DJuaIv]d “dq (S€6]) UssUaIIO// VV OOLE-0S piuaonjd pyuaonjd “q pyuav]d DUOSOWAOYd (CEG) Uosus}IO/ Md LA VO VV OOLE-O00T 1sdajad DWOSOSKF] (S861) I19Hd 2 AJesog (CEG) Uosua Oy LM Ad VO VV O81 1—O9T UNIDASAUAL DULOSOADAY MH VO OcOI—O¢€I Hjeq DUOSOQDAV, oeplinuyjourTyoY, Ayre BploMYIOUIYIY Japi1O (1661) ‘Te 19 surydoH “(OL61) UeleUd MH VO CC9-8V x IOYADG SIADPIIOIOL T LM LIN Ad VO Hd VV OOOI—€7c siuyfo s1app1g0] Ag (OOOZ) ‘[e 19 BUISaps T-vsoqieg (OL61) Ur[eUd Md VO WV OSLI—-001 4~DUDY]OSU] S1IDP1IOIAAIS LM VS Ad VO Hd 008-0 SQPLO[NGIAD SOPlOINGIAL SiAD PION Apnjs Juosoid VO Hd P79-001 ySUDIIU SIADPIIOIDI (QOOOT) ‘[B 19 BUISOpoT-vsogieg (OL61) urloud LM AA VO O6L-9V 4 DSOSNA S1ADPID (QOOT) ‘[B 19 BUISOpoT-vsogieg (S861) Ted 29 Ayestog “(OL6T) Ue[eUd Wd LA VO 008-9 4 PJOIISSAGD SIUIDPID ovpuepiy Ape, Bprorepr) IspiO sooinos [RUONIPpYy o1ydeis00H (ur) oINOUTAYIe g uOXxe |, uonnqmysiq ‘(yINOS seIONeH odeD) oe19duia} Wem “EAA SOHUeNVY MVM sonuepy Ss ‘VWs :[eordonued ‘7g ‘onuepy N “VN ‘uvouroyIpsyy “LIN SePloly AS 2 SsAoy epLoyy “YH ‘(yInos pod sdea) siesrodura} euoyAIne ‘Pq “Uvs[HUY pue uvoqqlvy ‘yD :seureyeg ‘HY :onueperydure ‘yy :oorxoyy JO J[NH oy} puoAaq uotnqiunsip o1ydersooH ‘popnyjoxe ore jeroydised ureulor yey) IST] (661) S Afeslag ul satoods (6161) Aseslog 0} suoTIppe ore (,.) YSOIse UL YJIM poyIeUr exe], (661) AJes1oS WO] poyIpouw ‘oosrxepy JO J[NH oy) Jo sprourysq—'] eqQey PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 76 LM VS AH VO OIC-S snssaidapqns sajspada&]) (C661) ‘Te 19 Jo[pusH LM Ad VO S8c0 SNAIDSOA Aaspada]D LM AH VO O€T—OS 1J@UIAADA Aajspada] yD LM Md VO CL=GIl snjpsjsoid sajspada] J Ad VO LOIT—07¢ 14aYSayo Aajspada]D sepliajysvodATD Ajiuue7 eplolaysvadATD IopiO (861) Uosmed VS Ld 4H VO OLS-S SNWOJSOJIKI SNAUOUIYIT oeplououryoY, Ajrue7 eplodAjs9[0H{ IOpsO (SS6I) ssulny MA VO Or-0 SIPLMA DAJaWOUlYyIA (C661) ‘Te 39 Jo]pueH LM VS MH VO VV Cr-0 AIJUNIN] AIJUNIN] VDAJQUOUIYIT sepljououryoy ATuUue 7 BploulyoY JOp1O (S661) ‘Je 19 JO[pugH LM VS Md VO VV SS-O SNSOD1AJUAA SaJSnaud.A J Apnjs juasoid ‘(g/6|[) uosmed LM VS Id MH VO SSI-ZI 4 DIDINIDU DIDINIDU D1IJa]OGopNnas q (C661) ‘Te 19 Jo[pusH WH VO c6-S ISUIDIJIM SNUIYIAIAT VS MH VO Hd OST—-O SNIDSAIDA SNIVDSAIIDA “JT LM Md HA Cl=0) SNUIJOADI SNIDEAIADA “TJ SNIDBAIIDA SNUIYIAIAT LM MA VO LIL pfL=SS SaDAANA SNUIYIAJAT sepnsnoudoxoy, AjTwej7 LIN Ad LA VO VV OCr—-Cl DIDINIDU SIADP1IIOUAH oeplinojdouway, Ape; eploimos[douwlay, 1apiO Md VO O€L—-OET DIdNIS s1pplo0pod (S861) 119d 29 Ayesog (816) UOsMEd VS WH LA VO O8Ec-S9 snuppiioyf snina}doja0y WH LA VO Hd ScéC—-O pipjniound piovqay oeprorgry Ayre eploloeqily JOpIO S9dINOS [BUOTIIPpY o1yde1s00H (W) oIMew Aye g uOX® |, uonnginsiq ‘penunuoy— | 2191 Td VOLUME 116, NUMBER 1 (8961) JeyseyD Md LH VO Hd 619-OSI saplosunjpds viuanjoyja)d VS Ad VO Hd OL SIPUDAS SNSSIAGOISD] (C661) ‘Te 19 Jo[pusH Ad VO Hd O0C—C DSOIDIAJUAA DSOD1AJUAA DWUOAPW VO (WUCHE ({ )vIDsUuo]a DIDsUo]a Sisdossi1g (8961) teyseyD Md VO Hd LV9S9C poyupyw sisdossiag (8961) JeYyseyD D fal OIe—-SV pip sisdossig (C661) ‘TB 19 JO[PUSH “(8L61) Uosmed VS LIN Ad VO VV OVc—O AOJODIUN SUSSIAF oepisstig A[IuWey (OS61) UssusLIO VS VN WH VO VV OccS—OSS psafipy]ag Asay oepisdoisy Ae; LM VS AA VO Svr—0 sodosjD DAIOF& VO OveE-O£ snjooluy sajspjasday (TS6[) UssusVIOJy VO OSLI—OSL snjpjadkyovaig saisvjasday MH VO OO6-EV DI1AJUIIXA DIZISSD8Y Ad LH VO Hd OOS—97C SNUDKUBIGAO 1A]SDZ1IYIS oepliayseziyog Ayre, (8261) UOSmET 29 JOLY VN 3A OCI E-0061 NSAID LISDISADS (OS61) Ussus}IO/ VN Ad VO VV CesVy-O8e snyisiadxa AaIspiway seplio}seruay A[IUUey eprosuryeds IopiQ Md VO 008—-0€1 Jaqs31s spdupjouo) MAH VO OTE-LE pssaidap spdwupjou1yoq oepipeduivyjouryo, Aye; BployNpIsseD JaploO LM MA 06—€ mMiyayoiu adoouq LM MA Hd 06-TI suvsiagnv adoouq LM Md VO 09-0 pIvAOfsadsaixas DIPOIT (0661) PIOJTPL 29 PlosreH ¢—-0 SNOT DEAN (0661) PLOFIOL 2 plosrey VS VO O8I-O pypsofsadsainbuinb vjijaw oepaiyla Ayre (Qp6|[) UssusLIO/y MH VO VV OOSZ-OSI snsodipuvias snupaoou1yoq ovpreyngry Ayre, sooinos [PUONIPpY orydeis0a0H (uw) o1LnowAuIeEg UOXe |, uonnqiysiq ‘penunuoj— | 29,1 78 Table 1.—Continued. Distribution Additional sources Geographic Bathymetric (m) Taxon Family Loveniidae FK WT CA FK 40-220 360-3550 Echinocardium laevigaster Homolampas fragilis Family Asterostomatidae Mortensen (1950) CA FK 38-1610 150-1025 Archaeopneustes hystrix CA FK Palaeobrissus hilgardi CA FK 76-805 Paleopneustes cristatus PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON and other locations less prone to sediment accumulation. These events might be fac- tors that explain the strong Caribbean influ- ence and low endemism of the Gulf of Mexico echinoid fauna. Acknowledgments We thank the following individuals for lending material in their care or for hosting us on visits: D. L. Pawson and C. A. Ahearn (Smithsonian Institution); R. M. Woollacott and EK Collier (Harvard Univer- sity); L. W. Siemon (Harbor Branch Ocean- ographic Institution); M. K. Wicksten (Tex- as A&M University). We thank C. A. Ahearn also for making available to us some of the echinoid systematic literature. Information on museum records and collec- tion data were provided by R. Mooi (Cali- fornia Academy of Sciences), S. H. Halsey (Museum of Natural History, London), W. W. Kirby-Smith (Duke University), and K. D. Spring (Continental Shelf Associates, Inc.). We are especially indebted to S. D. Cairns (Smithsonian Institution) for allow- ing us to publish field notes from his sub- mersible dive. Thanks are also due M. B. Bush, R. A. Tankersley and J. T. Tolbert (Florida Institute of Technology) for dis- cussions of North American paleoclimates, for advice on graphics and for translation of parts of Koehler (1908), respectively. Literature Cited Agassiz, A. 1880. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Caribbean Sea in 1878-79, and along the Atlantic coast of the United States during the summer of 1880, by the U.S. Coast Survey Steamer “‘Blake,’””» Commander J. R. Bartlett, U.S.N., commanding. IX. Preliminary report on the Echini.—Bulletin of the Museum of Com- parative Zoology at Harvard College, in Cam- bridge 8:69—84. Barbosa-Ledesma, I. E, E A. Solis-Marin, & A. La- guarda-Figueras. 2000. New records for cida- roid echinoids (Echinodermata: Echinoidea) of the Gulf of Mexico, Mexico.—Revista de Biol- ogia Tropical 48:721. Broecker, W. S., J. PR Kennett, B. P. Flower, J. T. Teller, VOLUME 116, NUMBER I S. Trumbore, G. Bonani, & W. Wolfli. 1989. Routing of meltwater from the Laurentide Ice Sheet during the Younger Dryas cold epi- sode.—Nature 341:318—321. Cairns, S. D., D. M. Opresko, T. S. Hopkins; & W. W. Schroeder. 1993. New records of deep-water Cnidaria (Scleractinia & Antipatharia) from the Gulf of Mexico.—Northeast Gulf Science 13: 1-11. Chesher, R. H. 1968. The systematics of sympatric species in West Indian spatangoids: a revision of the genera Brissopsis, Plethotaenia, Paleop- neustes, and Saviniaster.—Studies in Tropical Oceanography 7:1—168. Clark, A. H. 1954. Echinoderms (other than holothu- rians) of the Gulf of Mexico.—Fishery Bulletin of the Fish and Wildlife Service 55:373-—379. Clark, H. L. 1907. The Cidaridae.—Bulletin of the Museum of Comparative Zoology at Harvard College 51:165—230. . 1912. Hawaiian and other Pacific echini. The Pedinidae, Phymosomatidae, Stomopneustidae, Echinidae, Temnopleuridae, Strongylocentroti- dae, and Echinometridae.—Memoirs of the Mu- seum of Comparative Zoology at Harvard Col- lege 34(4):1-179. . 1921. Report on the Echinoidea collected by the Barbados-Antigua Expedition from the Uni- versity of Iowa in 1918.—University of Iowa Studies in Natural History 9(5):103—121. . 1925. A catalogue of the Recent sea-urchins (Echinoidea) in the collection of the British Mu- seum (Natural History). British Museum (Nat- ural History), London, 250 pp. Claus, C. KF W. 1880. Grundziige der Zoologie: zum wissenschaftlichen Gebrauche. Fourth edition, volume 1. N. G. Ellwert, Marburg, 822 pp. Continental Shelf Associates, Inc. 1979. South Atlantic hard bottom study. Report to U.S. Department of the Interior, Bureau of Land Management, Washington, D.C., Contract AAS551-CT8-25. Continental Shelf Associates, Inc., Jupiter, Flor- ida, 356 pp. . 1996a. Baseline environmental surveys for the Destin Dome Unit. Report to Chevron U.S.A., Inc., New Orleans, Louisiana. Continental Shelf Associates, Inc., Jupiter, Florida, 155 pp. & ap- pendices. . 1996b. Photodocumentation surveys of Destin Domerarcas blocks 2 lS IAS Seill6N 545 Sp" 56, 57, 99, and 100. Report to Chevron U.S.A., Inc., New Orleans, Louisiana. Continental Shelf Associates, Inc., Jupiter, Florida, 80 pp. & ap- pendices. , & Texas A&M University. 2001. Mississippi/ Alabama pinnacle trend ecosystem monitoring, final synthesis report. Report to U.S. Depart- ment of the Interior, Geological Survey, Biolog- 7S ical Resources Division, Kearneysville, West Virginia, Contract USGS/BSR—2001-0007 & U.S. Department of the Interior, Minerals Man- agement Service, Gulf of Mexico OCS Region, New Orleans, Louisiana, OCS Study MMS 2001-080. Continental Shelf Associates, Inc., Jupiter, Florida, & Texas A&M University, Geochemical Research Group, College Station, 415 pp. & appendices. Cutress, B. M. 1980. Cretaceous and Tertiary Cidaro- ida (Echinodermata: Echinoidea) of the Carib- bean area.—Bulletins of American Paleontolo- gy 77(309):1=215. Defenbaugh, R. E. 1976. A study of the benthic mac- roinvertebrates of the continental shelf of the northern Gulf of Mexico. Unpublished Ph.D. dissertation, Texas A&M University, College Station, 476 pp. Downey, M. E. 1968. Catalog of Recent echinoid type specimens in the U.S. National Museum Smith- sonian Institution and the Museum of Compar- ative Zoology Harvard University.—Smithson- ian Institution, United States National Museum Bulletin, 264:1—99. Duke University Marine Laboratory. 1982. Final re- and Environmental port. South Atlantic OCS area living marine re- sources study. Year II. Volume II. An investi- gation of live-bottom habitats off North Caro- lina. Report to U.S. Department of the Interior, Minerals Management Service, Contract AA551-CT1-18. Duke University Marine Lab- oratory, Beaufort, North Carolina, 143 pp. & appendices. Fell, H. B. 1966. Cidaroids. Pp. U312—U339 in R. C. Moore, ed., Treatise on invertebrate paleontol- ogy, part U, Echinodermata 3(1). University of Kansas Press & Geological Society of America, New York, 367 pp. Gore, R. H. 1992. The Gulf of Mexico. Pineapple Press, Sarasota, Florida, 384 pp. Gray, J. E. 1825. An attempt to divide the Echinida or sea-eggs into natural families.—Annals of Phi- losophy 10/26:423—431. Harold, A. S., & M. Telford. 1990. Systematics, phy- logeny and biogeography of the genus Mellita (Echinoidea: Clypeasteroida).—Journal of Nat- ural History 24:987—1026. Harry, H. W. 1979. Echinoderms recorded from the northwestern Gulf of Mexico. H. W. Harry, Bellaire, Texas, 58 pp. Hendler, G., & J. E. Miller. 1984. Ophioderma deva- neyi and Ophioderma ensiferum, new brittlestar species from the western Atlantic (Echinoder- mata: Ophiuroidea).—Proceedings of the Bio- logical Society of Washington 97:442—461. , D. L. Pawson, & P. M. Kier. 1995. Sea stars, sea urchins, and allies: echinoderms 80 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of Florida and the Caribbean. Smithsonian In- stitution Press, Washington, 391 pp. , & R. L. Turner. 1987. Two new species of Ophiolepis (Echinodermata: Ophiuroidea) from the Caribbean Sea and Gulf of Mexico, with notes on ecology, reproduction, and morpholo- gy.—wNatural History Museum of Los Angeles County Contributions in Science 395:1—14. Hopkins, T. S., J. E Valentine, J. B. McClintock, K. Marion, & S. A. Watts. 1991. Community pat- terns of echinoderms associated with substrate and depth in the northern Gulf of Mexico. Pp. 231-239 in T. Yanagisawa, I. Yasumasu, C. Oguro, N. Suzuki, & T. Motokawa, eds., Biol- ogy of Echinodermata: proceedings of the Sev- enth International Echinoderm Conference, Ata- mi, 9-14 September 1990. A. A. Balkema, Rot- terdam, 590 pp. Hulings, N. C. 1955. An investigation of the benthic invertebrate fauna from the shallow waters of the Texas coast. Unpublished M.A. thesis, Texas Christian University, Fort Worth, 87 pp. Hyman, L. H. 1955. The invertebrates, volume 4, Echi- nodermata: the coelomate Bilateria. McGraw- Hill Book Co., New York, 763 pp. John E. Chance & Associates, Inc. 199la. Hazard study of block 15 (OCS-G-8321) and block 16 (OCS-G-8322) Destin Dome area. Report to Chevron U.S.A., Inc., New Orleans, Louisiana. John E. Chance & Associates, Inc., Lafayette, Louisiana, 11 pp. & appendices. . 1991b. Archeological and hazard study of block 12 (OCS-G-8320), block 13 (OCS-G- 10418), block 14 (OCS-G-10419), and block 57 (OCS-G-6407), Destin Dome area. Report to Chevron U.S.A., Inc., New Orleans, Louisiana. John E. Chance & Associates, Inc., Lafayette, Louisiana, 29 pp. & appendices. . 1992. Hazard study of the eastern half of block 100 (OCS-G-6411), Destin Dome area. Report to Chevron U.S.A., Inc., New Orleans, Louisiana. John E. Chance & Associates, Inc., Lafayette, Louisiana, 12 pp. & appendices. Kier, P. M., & M. H. Lawson. 1978. Index of living and fossil echinoids 1924—1970.—Smithsonian Contributions to Paleobiology 34:1—182. Koehler, R. 1908. Astéries, ophiures et échinides de V’ Expédition Antarctique Nationale Ecossaise— Transactions of the Royal Society of Edinburgh 46(Pt. II, 22):529-649. Lane, D. J. W., L. M. Marsh, D. VandenSpiegel, & E W. E. Rowe. 2000. Echinoderm fauna of the South China Sea: an inventory and analysis of distribution patterns.—The Raffles Bulletin of Zoology 2000(Supplement 8):459—493. Liao, Y. 2001. Echinodermata. Pp. 386—399 in Z. Huang, ed., Marine species and their distribu- tion in China’s seas [J. Lin, translator]. Krieger Publishing Co., Malabar, Florida, 599 pp. , & A. M. Clark. 1995. The echinoderms of southern China. Science Press, Beijing, 614 pp. Ludwick, J. C., & W. R. Walton. 1957. Shelf edge calcareous prominences in the northeastern Gulf of Mexico. Bulletin of the American Associa- tion of Petroleum Geologists 41:2054—2101. Marine Technical Services, Inc. 1985. A high-resolu- tion geophysical report on blocks 55, 56, 57, 99, and 100, Destin Dome area. Report to Con- oco, Inc., Algiers, Louisiana. Marine Technical Services, Inc., Stafford, Texas, 13 pp. & appen- dices. Matsuoka, N. 1985. Biochemical phylogeny of the sea- urchins of the family Toxopneustidae.—Com- parative Biochemistry and Physiology 8OB: 767-771. . 1986. Further immunological study on the phylogenetic relationships among sea-urchins of the order Echinoida.—Comparative Biochem- istry and Physiology 84B:465—468. . 1987. Biochemical study on the taxonomic situation of the sea-urchin, Pseudocentrotus de- pressus.—Zoological Science 4:339—347. Michelin, M. H. 1862. Echinides et Stellérides. An- nexe A (7 pp.) in L.-C. Maillard, Notes sur I’Tle de la Réunion (Bourbon). Dentu, Paris. Mortensen, T. 1903. Echinoidea (Part I).—The Danish Ingolf-Expedition 4(1):1—193. 1910. On some West Indian echinoids.— Smithsonian Institution, United States National Museum Bulletin 74:1-31. . 1928. A monograph of the Echinoidea, vol- ume 1, Cidaroidea. C. A. Reitzel, Copenhagen, Sepp: . 1933. Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-16. 66. The echinoderms of St. Helena. (Other than crinoids.)—Videnska- belige Meddelelser fra Dansk naturhistorisk Forening i Kg@benhavn 93:401—473. . 1935. A monograph of the Echinoidea, vol- ume 2, Bothriocidaroida, Melonechinoida, Lep- idocentroida, and Stirodonta. C. A. Reitzel, Co- penhagen, 647 pp. . 1940. A monograph of the Echinoidea, vol- ume 3(1), Aulodonta. C. A. Reitzel, Copenha- oi, IO) joo . 1942. New Echinoidea (Camarodonta). Pre- liminary notice.—Videnskabelige Meddelelser fra Dansk naturhistorisk Forening i K@benhavn 106:225—232. . 1943. A monograph of the Echinoidea, vol- ume 3(2), Camarodonta 1: Orthopsidae, Gly- phocyphidae, Temnopleuridae and Toxopneusti- dae. C. A. Reitzel, Copenhagen, 533 pp. . 1948. A monograph of the Echinoidea, vol- ume 4(2), Clypeastroida: Clypeastridae, Arach- VOLUME 116, NUMBER 1 noididae, Fibulariidae, Laganidae and Scutelli- dae. C. A. Reitzel, Copenhagen, 471 pp. . 1950. A monograph of the Echinoidea, vol- ume 5(1), Spatangoida 1. C. A. Reitzel, Copen- hagen, 432 pp. . 1951. A monograph of the Echinoidea, vol- ume 5(2), Spatangoida 2. C. A. Reitzel, Copen- hagen, 593 pp. Opresko, D. M., & S. D. Cairns. 1992. New species of black coral (Cnidaria: Antipatharia) from the northern Gulf of Mexico.—Northeast Gulf Sci- ence 12:93-97. Pawson, D. L. 1978. The echinoderm fauna of Ascen- sion Island, south Atlantic Ocean.—Smithsoni- an Contributions to the Marine Sciences, No. 2: i-iv, 1-31. , & J. E. Miller. 1983. Systematics and ecology of the sea-urchin genus Centrostephanus (Echi- nodermata: Echinoidea) from the Atlantic and eastern Pacific Oceans.—Smithsonian Contri- butions to the Marine Sciences, 20:1—15. Phelan, T. 1970. A field guide to the cidaroid echinoids of the northwestern Atlantic Ocean, Gulf of Mexico, and the Caribbean Sea.—Smithsonian Contributions to Zoology 40:1—67. Pielou, E. C. 1991. After the ice age: the return of life to glaciated North America. University of Chi- cago Press, Chicago, 366 pp. Pomory, C. M. 2002. A guide to the shallow-water Echinodermata of the Texas coast. University of Texas Press, Austin (in press). Rowe, FE W.E., & J. Gates. 1995. Zoological catalogue of Australia, volume 33, Echinodermata. CSI- RO Australia, Melbourne, 510 pp. Schneider, C. W. 1976. Spatial and temporal distribu- tions of benthic marine algae on the continental shelf of the Carolinas.—Bulletin of Marine Sci- ence 26:133-151. Serafy, D. K. 1979. Echinoids (Echinodermata: Echi- noidea).—Memoirs of the Hourglass Cruises 5(3):1-120. , & E J. Fell. 1985. Marine flora and fauna of 81 the northeastern United States. Echinodermata: Echinoidea. NOAA Technical Report NMFS 33:i-11i, 1-27. Shimabukuro, S. 1991. Tripneustes gratilla (sea ur- chin). Pp. 313—328 in S. Shokita, K. Kakazu, A. Tomori, & T. Toma, eds., Aquaculture in tropical areas [M. Yamaguchi, translator]. Mi- dori Shobo Company, Ltd., Tokyo, 360 pp. Texas A&M University. 1981. Northern Gulf of Mex- ico topographic features study, final report, vol- ume 3. Report to U.S. Department of the Inte- rior, Bureau of Land Management, Outer Con- tinental Shelf Office, New Orleans, Louisiana, Contract AA551-CT8-35. Texas A&M Univer- sity, College of Geosciences, Department of Oceanography, College Station, 130 pp., C-1-— C-34. Troschel, E H. 1869. Uber einige neue Seeigel.—Sit- zungsberichte und Verhandlungen des naturhis- torischen Vereines der preussischen Rheinlande 26:96. . 1872. Die Familie der Echinocidariden.—Ar- chiv fiir Naturgeschichte. 38:293—356. Turner, R. L., & C. M. Norlund. 1988. Labral mor- phology in heart urchins of the genus Brissopsis (Echinodermata: Spatangoida), with an illus- trated revised key to western Atlantic species.— Proceedings of the Biological Society of Wash- ington 101:890—897. Williams, M., D. Dunkerley, P. De Deckker, P. Ker- shaw, & J. Chappell. 1998. Quaternary environ- ments, Second edition. Arnold Publishers, Lon- don, 329 pp. Yoshino, K.-I., H. Kajiura, K. Nomura, T. Takao, Y. Shimonishi, M. Kurita, M. Yamaguchi, & N. Suzuki. 1989. A halogenated amino acid-con- taining sperm activating peptide and its related peptides isolated from the egg jelly of sea ur- chins, Tripneustes gratilla, Pseudoboletia ma- culata, Strongylocentrotus nudus, Echinometra mathaei and Heterocentrotus mammillatus.— Comparative Biochemistry and Physiology 94B:739-751. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):82—95. 2003. The genus Chrysopetalum Ehlers, 1864 (Annelida: Polychaeta: Chrysopetalidae) in the Pacific coast of Panama Maria Teresa Aguado, Maria Capa, and Guillermo San Martin Laboratorio de Biologia Marina e Invertebrados, Departamento de Biologia (Zoologia), Facultad de Ciencias, Universidad Autonoma de Madrid, Cantoblanco, 28049 Madrid, Spain Abstract.—The family Chrysopetalidae is represented in the Pacific coast of Panama by three genera: Bhawania Schmarda, 1961, Paleanotus Schmarda, 1961 and Chrysopetalum Ehlers, 1864. Up to now, the latter genus was rep- resented by a single species, Chrysopetalum occidentale Johnson, 1897. During a study carried out at the National Park of Coiba (Pacific coast of Panama), three species of this genus were found: C. occidentale and two new species: C. elegantoides n. sp. and C. maculata n. sp. In this paper the two new species are described and a key for identification is provided. Chrysopetalum elegan- toides closely resembles the descriptions of C. elegans Bush, 1900 and C. ehlersi Gravier, 1901. The paleae and neurosetae are very similar to those of C. elegans, but specimens from Coiba lack interramal glands like C. ehlersi. Chrysopetalum elegantoides differs from C. ehlersi principally in the tips of paleae and the length of the blades of the neurosetae. Chrysopetalum maculata differs from all others within the genus in having a wide body and broad paleae and in the lack of spines. The most similar species is C. heteropalea Perkins, 1985; both species lack spines (sensu Perkins) and have several symmetrical paleae; but the paleae of the C. maculata are ornamented with knobs, lacking the transverse ridges that are typical of C. heteropalea. The presence of C. occidentale in the area is verified. Polychaetous annelids are poorly known in the Panamanian Pacific, and only few studies in this area have provided records of these marine animals (Monro 1928a, 1928b, 1933a, 1933b; Fauchald & Reimer 1975; Fauchald 1977a; Lopez et al. 1997). Investigators from the Laboratorio de In- vertebrados y Biologia Marina of the Univ- ersidad Aut6noma de Madrid have been working at the National Park of Coiba since 1996 in order to increase the knowledge of the marine fauna in general (San Martin et al. 1997), and the polychaetes in particular (Capa et al. 2001a, 2001b, 2001c, San Mar- tin et al. 1998). The diversity of genera and species of the family Chrysopetallidae in the Panamanian Pacific appears to be low; only three gen- era: Bhawania Schmarda, 1961, Paleanotus Schmarda, 1961 and Chrysopetalum Ehlers, 1864, and four species: B. goodei Webster, 1884, B. riveti (Gravier, 1908), P. chryso- lepis Schmarda, 1861 and C. occidentale Johnson, 1897, have been reported. The present paper increases the knowledge of this family in the area offering a description of two new species of Chrysopetalum, as well as verifies the presence of C. occiden- tale. The National Park of Coiba (7°10’ to 7°53'N and 81°32’ to 81°56’ W) is a marine- terrestrial area, protected since 1991. Itis a set of a large amount of islands and islets, the largest being the one which gives the name to the Park. The study of the fauna and flora of this Park is very important is- sue for several reasons: the eastern central Pacific has a special biogeographical inter- VOLUME 116, NUMBER 1 est due to its past connection with the pre- sent Caribbean (Laverde-Castillo 1986); the lack of human impact; and the few studies carried out in the area. Material and Methods The samples were collected during four expeditions carried out between Jun 1996 and Sep 1998. Different substrata were sampled: blocks of dead coral (Pocillopora spp.), epibionts from dead gorgonians (Pa- cifigorgia sp.) and living oysters (Hyotisa hyotis) and coarse sand. Samples were col- lected in three different ways: the blocks of dead coral Pocillopora spp. (approximately 4 kg) were sampled by SCUBA and im- mediately put into a bag, the blocks were then placed in sea water for 24 h in order to induce the organisms to leave the coral cavities due to short supply of oxygen; the remainder of the organisms were extracted by fragmenting the branched blocks; the polychaetes associated with dead gorgoni- ans and living oysters were sampled by SCUBA, by removing the substrata and scraping off their surfaces; and for sam- pling the polychaetes in coarse sand, a corer 9.5 cm diameter and | liter volume was used, and the sediment was washed through a 0.25 mm gauge mesh sieves. The samples were fixed in a 10% form- aldehyde-seawater solution and preserved in a 70% alcohol solution. For identification a Olympus SZ30 stereomicroscope and Olympus CH30 optical microscope were used. Parapodia from most of the specimens were removed and slide mounts were made in glycerine gel. The drawings were made to scale, with a drawing tube, in a Nikon Optiphot optical microscope equipped with interference contrast optics (Nomarsky). Scanning electron micrographs (SEM) were made using standard procedure. The speci- mens in 70% alcohol were introduced grad- ually in solutions more concentrated in ac- etone. The specimens were critical-point dried in carbon dioxide, coated with gold and viewed through a SEM Phillips XL-30 83 system, in the SEM unit at the Universidad Autonoma de Madrid. The specimens are deposited in the Museo Nacional de Cien- cias Naturales de Madrid (MNCNM), Spain. The sample sites, co-ordinates and habi- tats are provided in Table 1. In 14 samples studied, a total of 65 specimens of Chry- sopetalids were identified belonging to only one genus, Chrysopetalum, and three spe- cles. Results Genus Chrysopetalum Ehlers, 1864 Type species.—Chrysopetalum fragile Ehlers, 1864. Diagnosis.—Medium in size, up to 65 segments. Prostomium rounded, partially retracted into anterior segments (about pos- terior margin of the fourth segment), with two pairs of eyes generally in rectangular disposition, anterior pair larger than poste- rior ones. A pair of elongate ventral palps. Pair of long lateral antennae emerging ven- trally on anterior margin of dorsal surface, and short median antenna, fusiform, origi- nating behind anterior eyes. Caruncle at- tached on posterior dorsal margin of pro- stomium, smaller than prostomium, sphae- roidal, and reaching the end of fifth setiger. Ventral mouth opening with triangular or rounded cover, extending from anterior margin of fifth segment. Peristomium re- duced, only visible as lip covering the mouth. First two segments each with pair of dorsal and ventral cirri, similar to those following segments. First segment asetig- erous, second segment only with notosetae. From third segment, parapodia birramous with notopodial paleae (modified notosetae) and compound spinigerous neurosetae on anterior segments, changing to compound falcigerous neurosetae through posterior segments. Dorsal cirri with long cirrophores and long cirrostyles with wide basis and fi- liform distal part. Ventral cirri shorter, but similar in shape. Dorsum partially covered by flatened paleae, distally serrated, with 84 Table 1.—Collection sites, coordinates, habitats, depths and weight/volumes of the studied samples. Weight/volume Samples Coordinates Station Depth Sampled substrate 3) inn 6m Isla de Uvas N, 81°46'W 4000 g Isla del Canal de Afuera Ensenada Maria Playa Rosario 50"N, 81°38'25"W 20"N, 81°41'W 4200 g 4100 g 2.5 m 30"N, 81°41'40"W 20"N, 81°41'W N, 81°46'W 30"N, 81°42 4400 g 5.8 m 6m 2m 2m Ensenada Maria Isla de Uvas 4000 g 3000 g Isla de Granito de Oro Ensenada Maria Playa Rosario 30" W W 3600 g 3900 ¢ 1.2m 11 m 14 m 40”"W ~ ~ ~ 20"N, 81°41 30"N, 81°41 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Islote Santa Cruz = N, 41°47'10 40"N, 81°49 Islote San Martin Punta Cirilo 30"W 10”"W 4-6 m 10 m Isla del Canal de Afuera Bajo Mali Rock ~ ~ ~ 10"N, 81°50 1 liter WwW < N, 81°41'40' ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . 7°49 T4l 7°24 7°38 7°24 T°49 F395 7°24 T38 P33 736 7°36 T°41 739 CM2FEB97 es CM3FEB97 es CM4FEB97 es CM2NOV97 es CM4NOV97 es CMSNOV97 CM1SEP98 eS es CM4SEP98 es CMS5SEP98 es eee eS oe = = = SS rT TS Se Dead cora Dead cora Dead cora Dead cora Dead cora Dead cora Dead cora Dead cora Dead cora CUAL8JUN96 Gorgonian epibionts CUAL2FEB97 Hyotisa hiotis epibionts Hyotisa hiotis epibionts Below rocks CUALSFEB97 CUAL8FEB97 ARISEP98 Coarse sand internal longitudinal ribs and transverse chambers. Paleae of middle segments form- ing at least three groups arranged in semi- circular or straight row, covering dorsum, extending from bundles occupying 0.5 to 0.67 of segmental width, serrated on both margins. Some species with an additional row of anterior spines. Middle group paleae consisting of two or three irregular trans- versely arranged rows of long and slender paleae originated in up to eight develop- mental centers, median paleae symmetrical, each with up to 12 internal longitudinal ribs, lateral ones with tips asymmetrical, bent toward middle of group. Lateral group paleae consisting of up to 10 more slender, symmetrically tipped, laterally orientated paleae, also with internal ribs. Midline group paleae arranged in longitudinal to slightly oblique row of up to six; each bent medially and posteriorly, shorter, slender than middle group paleae, with almost sym- metrical tips. Compound falcigers with blades decreasing in size dorsally to ven- trally, unidentate, with hooded tips and ser- rated margins. Pygidium with pair of anal cirri similar to dorsal cirri (Perkins 1985, San Martin in press). Chrysopetalum elegantoides new species Jeng, Il, 2 Material examined.—Holotype: CM2NOV97 (MNCN 16.01/8531a), para- types: CM2FEB97(2) (MNCN © 16.01/ 8352), CM3FEB97(1) (MNCN _ 16.01/ 8533), CM4FEB97(21) (MNCN_ 16.01/ 8534), CM2NOV97(5) (MNCN _ 16.01/ 8531b), CM4NOV97(2) (MNCN_ 16.01/ 8535), CMSNOV97(1) (MNCN_ 16.01/ 8536), CMISEP98(6) (MNCN __ 16.01/ 8537), CMS5SEP98(1) (MNCN _— 16.01/ 8539), CUAL8JUN96(1) (MNCN _ 16.01/ 8541), CUAL2FEB97(1) (MNCN_ 16.01/ 8541), CUALSFEB97(8) (MNCN_ 16.01/ 8543). Additional material examined.— CM3FEB97(fragments), CM4FEB97(frag- ments), CUAL8FEB97(fragments). VOLUME 116, NUMBER I 85 Fig. 1. Chrysopetalum elegantoides n. sp. Paratype CM1SEP98. A: Right parapodium, setiger 33, anterior view; B: Same, posterior view; C: Tip of symmetrical palea, middle group; D: Tip of palea, lateral part, middle group; E: Palea, lateral group; F: Palea, midline group; G: Spine, anterior group; H: Upper compound falcigerous neuroseta, middle segments; I: Lower compound falciger; J: Middle compound falciger. Scale A, B: 68.25 wm; CJ: 20 wm. 86 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Other material examined.—Chrysopetal- um elegans MNCN16.01/4107, MNCN 16.01/4194, MNCN16.01/4201, MNCN 16.01/4195 from Cuba, loaned by de Museo Nacional de Ciencias Naturales de Madrid. Chrysopetalum elhersi 1941.4.4.238 and A39 from Red Sea, 1969.192 from Kuwait and 1961.8.14 from Mombassa; lounded by the Natural History Museum, London and the Museum National d’Histoire Naturelle de Paris. Description.—Only two complete speci- mens (the holotype and one paratype in sample CUAL3FEB97). Holotype 8 mm long, 0.6 mm wide with 58 segments. Com- plete paratype 6 mm long, 0.5 mm wide, with 37 segments. Prostomium oval, slight- ly longer than wide. Some specimens hav- ing three pairs of eyes, with two posterior pairs slightly superimposed, anterior pair sometimes very close to each other but also separated in other specimens. Caruncle smaller than prostomium. Notosetae con- sisting of paleae and spines. Paleae serrated on their margins, without hoods on middle and midline groups, dorsal surface moder- ately knobbed (Figs. 1C—K 2B). Middle group paleae of midbody numbering 20— 30; middle, symmetric paleae with eight or nine internal longitudinal ribs and conical tapered tips (Figs. 1C, 2C); lateral middle group paleae asymmetrical with tips direct- ed to middle ones, with eight internal ribs (Fig. 1D); central paleae shorter than lateral paleae in middle group. Middline group pa- leae about seven, each with denticulated surface midrib (Fig. 1F). Lateral group pa- leae gradually more slender and smaller from the principal group to the anteriome- dial spines (Fig. 1E). Three spines slender and denticulated in each parapodia on an- terior group (Fig. 1G). Neuropodia with characteristic setae (Fig. 1H—J). Parapodial gland in the base of dorsal and ventral cir- rostyles and in the inferior part of neuro- podia (Fig. 1A—B). Remarks.—Two very similar species with slender paleae have been previously described: C. elegans Bush, 1910 and C. ehlersi Gravier, 1901. Chrysopetalum ele- gans differs from C. ehlersi in having well developed interramal glands with spindles of fibers, as well as in having wider neu- rosetal blades (Perkins 1985). The geo- graphical distribution of these two species is also different: C. elegans has been re- ported from the Caribbean (Bermuda and Cuba), Gulf of Mexico, and Florida, where- as C. ehlersi has been reported from the In- dian Ocean (Gulf of Aden, Persian Gulf and Kenia). Chrysopetalum elegantoides n. sp. from Coiba has cirrophoral glands but lacks interramal glands, although it is very simi- lar to C. elegans in the shape of the paleae and the width of the neurosetae. The spec- imens from Coiba have been compared with specimens of C. elegans from Cuba (San Martin 1986). The specimens of C. elegans from Cuba have very conspicuous interramal glands, confirming that this char- acter has value for segregating species (Per- kins 1985). Comparisons have also been made with specimens of C. elhersi 1941. Specimens from Kuwait and Mombasa are not C. ehlersi, and perhaps represent an un- described species. They differ from C. eh- lersi in having conspicuously broader hood- ed paleae, whereas the specimens from the Red Sea perfectly agree with the original description (Gravier 1901). Chrysopetalum elegantoides shares with C. ehlersi the ab- sence of an interramal gland and the similar shape of acute paleae, but they differ in sev- eral features: the median antenna in C. ele- gantoides is shorter than those in the spec- imens of C. ehlersi; C. elegantoides has proportionally more slender paleae than does C. ehlersi, which is more evident in the middle group and the lateral group pa- leae (Figs: 1@) EE, 3@ JE) the tips omihe paleae are sharper in C. elegantoides (Fig. 1D) than in those of C. ehlersi (Fig. 3D); midline group paleae have the denticulated midrib displaced laterally in C. ehlersi (Fig. 3F) and centered in C. elegantoides (Fig. 1F); the bidentation of neurosetae appears to be more pronounced in C. elegantoides (Figs. 1H—J, 2E, F) than in C. ehlersi (Fig. VOLUME 116, NUMBER I ene, 2. view; B: Paleae, anterior setigers; C: Paleae, middle setigers; D: Middle parapodium, posterior view; E—F: Compound falciger neurosetae. 3G—I); the neurosetal blades are shorter but they have the same width (they measure 44, 35 and 15 pm long in C. elegantoides, in the superior, median and inferior setae of a midbody parapodium respectively, and 57, 41 and 20 pm long in C. ehlersi, respec- tively. The cirrophoral gland is relatively larger in C. elegantoides than in C. ehlersi (Figs. 1A, B, 2D, 3A, B). The presence of three pairs of eyes has Chrysopetalum elegantoides n. sp. Scanning electron micrographs (SEM). A: Anterior end, dorsal been reported before on other species: Bush Gan Verrill 1900) found some specimens with six eyes in the syntypes of C. elegans, although Perkins (1985) reported only two pairs. Etymology.—tThe specific name is given because of the similarity between this spe- cies and C. elegans (-oides = similar to, in greak) which lives on the other side of the Isthmus of Panama. 88 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Chrysopetalum ehlersi. NHML ZK 1941.4.4.238. A: Parapodium middle segment anterior view; B: Same, posterior view; C: Tip of palea, central part of middle group; D: Same, from lateral part; E: Palea, lateral group; F: Palea, midline group; G: Upper compound falcigerous neuroseta; H. Middle compound falciger; I: Lower compound falciger. Scale A, B: 0.195 mm; C—I: 20 wm. VOLUME 116, NUMBER 1 89 2 =~ pee. Re ok 7 [D»- RAG ns Fig. 4. Chrysopetalum maculata n. sp. ARISEP98. A: Anterior end, dorsal view; B: Anterior end without paleae, dorsal view; C: Right parapodium, middle segment, anterior view; D: Left parapodium, middle segment, posterior view. Scale A: 0,195 mm; B: 0,18 mm; C, D: 97,5 wm. Habitat.—Blocks of dead Pocillopora Additional material examined.— spp. ARISEP98(10). Geographical distribution.—Pacific coast Description.—Holotype anterior frag- of Panama. ment 4.12 mm long, | mm wide without setae, 30 segments; largest paratype anterior Chrysopetalum maculata, new species fragment, 2.76 mm long, 0.8 wide, 23 seg- Figs. 4—6 ments. Fragile, broad body, white to pale Material examined.—Holotype: ARI _ brown. Silver to gold-colored paleae fans, SEP98(1) (MNCN 16.01/8530a), paratypes: often flecked with brown spots, covering ARISEP98(3) (MNCN 16.01/8530b). worm completely, median paleae interlock- 90 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON y AN OO PY, i SO. LOF CoLPF PD \RBY? ) Z 0’ 5 fe) fof J OF} ONS 9) © ale Fao 15 OPA \e ‘ 2705.2/. a [0 oe o ; 0\0, Qf Gof, |e 15° Ob ol 9°42 \ Q\o oF Iola 4° 099 3 ol%)\ 0 ( AW 4 ofo [0 j20f 2) 26/9 8 Voe04 || of,o 0 0° [9 9% @ 2o)° | odoolo 0} VS of eo l@ | eo iad od) | () 0 Q Bie) HES eB ie LE ae or 6 coo. Jo Foo o 8 10°08 Onl5 Sa eetoal dO} 50] 0% a) | 0°} a5) 0} o\-4 0° {2416 BOIS | ofS) 5210 Go lonbel® | ao |9: 2 O 9} 8 os oY Pick a ao % ig | OM ° O12. BEANO) go ee. 0} 9 ) fo) qd | ° O01 ~9] 09} 0 0G ° Cone) | JO--0 Volo} O01 2ofo 20 99 04 0|}0,) 0 ABI BI Se]}oF [Lo Ro [elo lo kd E FUE, 5). middle segment; B: Tip of palea from central part of middle group, middle segment; D: Palea, midline group; E: Paleae, lateral group; F: Palea, lateral group G: Lower compound falciger neuroseta; H: Middle compound falciger; I: Upper compound falciger. Scale A—C: 20 pm; D: 48 wm; E: 20 wm; F: 48 pm; G—I: 20 pm. ing at mid-line, forming distinct convex median ridge (Fig. 4A). Prostomium wider than long, with two pairs of garnet eyes, some specimens with an additional central eye. Palps about twice longer than wide (Fig. 4B). Rounded caruncle with cilia (Fig. 6B). Mouth covering semioval (Fig. 6A), eversible proboscis with two stylets. Noto- Chrysopetalum maculata n. sp. ARISEP98. A, C: Tips of paleae from lateral part of middle group, setae consisting of only very broad paleae, anterior group spines absent. Middle group paleae of middle segments about 20-22, with dorsal surface strongly knobbed, with 9—12 internal longitudinal ribs (Fig. SA—C) and, sometimes, with hooded tips; various symmetrical paleae in the centre of the fan (Fig. 6D). Midline group bearing about five VOLUME 116, NUMBER 1 el can Bas oe 2 Fig. 6. Chrysopetalum maculata n. sp. ARISEP98. (SEM). A: Anterior end, ventral view; B: Ciliated ca- runcula; C: Left parapodium, middle segment, anterior view; D: Tips of paleae from middle group, middle segment; E: Upper compound falciger neuroseta; F: Middle compound falciger neuroseta. paleae with denticulate midrib and six lon- gitudinal ribs (Fig. 5D). Lateral group con- sisting of one long, broad and acutely tipped palea with five or six longitudinal ribs, originating near acicula (Fig. 5F) and two or three more slender, shorter paleae originating laterally (Fig. 5E, 4C). Com- pound falcigerous neurosetae numbering 20 to 30 with blades moderately long. Upper few blades distinctly longer than adjacent ones, gradually shorter ventrally. Dorsal blades 55 wm long, with short and coarse serration (Figs. 5I, 6E); middle blades 38 wm long, with long and stout serration (Figs. 5H, 6F); ventral blades 28 wm long, with fine, moderately long serration (Fig. 5G). Interramal region of middle parapodia ciliated on anterior side. Glands containing spindles of fibers in dorsal cirrophore (Fig. 4D). 92 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Remarks.—Chrysopetalum maculata n. sp. resembles the general aspect of the spe- cies of the genus Arichlidon Russell, 1998, because of considerably broad shape of the body and paleae, as well as the golden color of the paleae with brown scale spots. How- ever, the diagnostic characters place this species in the genus Chrysopetalum. Chry- sopetalum maculata differs from other spe- cies of the genus in having very broad knobbed paleae and in lacking spines on the notopodia of middle segments. Chrysope- talum remaneii Perkins, 1985 and C. her- nancortezae Perkins, 1985, have very nar- row paleae and notopodial spines. Chryso- petalum elegans Bush, 1900 also possesses very narrow paleae and their typical inter- ramal glands could not be observed in the studied specimens. Chrysopetalum debile (Grube, 1855), C. occidentale Johnson, 1897, C. floridanum Perkins, 1985, and C. euripalea Perkins, 1985, have also narrow paleae, and tips of the paleae are displaced and have notopodial spines. The definition of C. elongatum (Grube, 1856) is conflict- ing and not clear (Perkins 1985) since drawings and descriptions are incomplete. The most similar species is C. heteropalea Perkins, 1985, because both C. maculata and C. heteropalea lack notopodial spines, posses dorsal cirrophoral glands, the ante- rior side of middle parapodia in the inter- ramal region is ciliated, and the shapes of neurosetae and paleae are similar. The tips of the paleae of C. heteropalea are also dis- placed as in C. maculata and both present symmetrical paleae in the centre of the mid- dle group. Paleae of C. heteropalea are con- siderably narrower, with 8 longitudinal ribs and lack knobs in their dorsal surface; in- stead, there is ornamentation consisting of numerous and irregular transversals ribs that is not present in C. maculata. Chryso- petalum heteropalea has also visible glands in the neuropodia and ventral cirrophores, which have not currently been found in specimens of C. maculata. Neurosetae are similar in both species, although C. macu- lata does not present the typical lowest compound falciger that occurs in C. heter- opalea and middle compound falcigers are slightly different. The caruncle of C. ma- culata is ciliated, differing from that de- scribed by Perkins (1985) for C. heteropa- lea. Etymology.—The species name derives from the latin macula (=spot), in reference to the numerous dark spots on the surface of paleae. Habitat.—Intertidal coarse sand. Geographical distribution.—Pacific coast of Panama. Chrysopetalum occidentale Johnson, 1897 jae, 7 Chrysopetalum occidentale Johnson, 1897: 161, pl. 5, figs 15, 16, pl. 6, Figs. 17-19; Monro 1933a: 19; Hartman 1961: 56, 57; 1968: 185, 186, Figs. 1-5; Fauchald & Reimer 1975: 82; Fauchald 1977a: 71, Figs. 18 a—c; 1977b: 10, Perkins 1985: 869-871, Figs. 3, 4. Material examined.—CM1 SEP98 (1+ 1 fragment), CUAL8FEB97 (1 fragment). Remarks.—Only one anterior fragment of 15 setigers and two middle fragments were collected. The specimens agree gen- erally with the description of Perkins (1985). The anterior fragment lacks the me- dian and left antennae, and we can not com- pare these structures with Perkins’ descrip- tion (1985). Middle group paleae have ob- tuse tips and eight to ten internal ribs (Fig. 7C, D). Midline group paleae have the den- ticulated midrib slightly displaced laterally (Fig. 7F). Lateral group paleae are slender (Fig. 7E). Blades of compound falcigerous neurosetae are bidentate, gradually decreas- ing in size dorsally to ventrally (Fig. 7G, H). Perkins (1985) established some differ- ences between the specimens from Califor- nia and those from Mexico. The number of paleae in the middle group of specimens from the first locality is between 35 and 40 and in Mexico they have a maximum of 30 paleae. In Coiba, the specimens found have VOLUME 116, NUMBER 1 93 Fig. 7. Chrysopetalum occidentale. CM1SEP98. A: Left parapodium, setiger 14, anterior view; B: Right parapodium, same setiger, posterior view; C: Palea, middle group, central part; D: Palea, middle group, lateral part; E: Palea, lateral group; F: Palea, midline group; G: Upper compound falciger neuroseta; H: Lower com- pound falciger. Scale A, B: 0.195 mm; C—H: 20 um. between 15 and 24 paleae in the middle group (Fig. 7A, B). Midline group paleae number about six in California and about four in Coiba. Type locality.—California. Habitat.—Rocks (Hartman 1968), hold- fast (Perkins 1985), intertidal (Fauchald 1977a), dead coral (Pocillopora spp.). Geographical distribution.—According to Perkins (1985) the distribution of this species is localized in the East Pacific, from the southern California to Panama. 94 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Key to Chrysopetalum Species in the Panamanian Pacific la. Anterior group of paleae and spines (sensu Perkins 1985) absent. Body and paleae very broad. ....C. maculata n. sp. lb. Anterior group of paleae and spines DPEESEMUV re ers eee ye Leen me y 2a. Middle group paleae with acute tips (see Fig. 2C—F), and narrow subtermi- nal region, without hoods or remnants inetherdistalvend loli paleace ae eller Ret HAS pe C. elegantoides n. sp. 2b. Middle group paleae with obtuse tips (see Fig. 8C—F) and enlarged subter- minal region, hoods or remnants gen- erally present C. occidentale ao0o006 9 -0°0 6.0.0 0 Acknowledgements We wish to thank the Agencia Espanola de Cooperacion Internacional (AECI) for partial financial support of the survey and Instituto de Recursos Naturales Renova- bles INRENARE) for logistical support in Coiba, especially S. Castroviejo, I. Tunon, L. Jimenez, C. Pechio and N. Bastida (Mali-Mali). Special thanks are due to E. L6pez and A. Laborda, members of the scientific staff. We are grateful to the fol- lowing people for the loan of material: M. Villena from Museo Nacional de Ciencias Naturales de Madrid (MNCN) for the loan of specimens of Chrysopetalum elegans, M. Lowe from the Natural History Muse- um, London (NHM) for the loan of spec- imens of C. ehlersi (1941.4.4.237—238, 1961.8.14, 1969.192) and FE Pleijel for the loan of specimens of C. ehlersi (A 39) from the Museum National d’ Histoire Na- turelle de Paris (MNHN). We appreciate the collaboration of E. Salvador at the SEM unit and J. Marugan for the revision of the English version. Special thanks are extended to T. Perkins for his comments and advise. Our acknowledgements to the referees of the manuscript. Literature Cited Capa, M., G. San Martin, & E. Lopez. 2001a. Syllinae (Syllidae: Polychaeta) del Parque Nacional de Coiba, Panama.—Revista de Biologia Tropical 49(1):103-115. BSG —. 2001b. Autolytinae, Eu- syllinae y Exogoninae (Syllidae: Polychaeta) con comentarios sobre la ecologia y la biogeo- gratia de la familia Syllidae del Parque Nacio- nal de Coiba, Panama.—Revista de Biologia Tropical 49(2):621—628. 5 . 2001c. Description of a new species of Parasphaerosyllis (Polychaeta: Syllidae: Syllinae).—Proceedings of the Bio- logical Society of Washington 114(1):280—284. Ehlers, E. 1864. Die Borstenwiirmer (Annelida Chae- topoda) nach systematischen und anatomischen Unterschungen dargestelt. Leipzig, Wilhelm Engelmann, xx + 1—268, pls. I-11. Fauchald, K. 1977a. Polychaetes from intertidal areas in Panama, with a review of previous sallow- waters records.—Smithsonian Contributions to Zoology 221:1—81. . 1977b. The Polychaete worms. Definitions and keys to the orders, families and genera.— Natural History Museum of Los Angeles Coun- try, Sciences Series 28:1—188. , & A. A. Reimer. 1975. Clave de poliquetos panamenos con la inclusi6n de una clave para todas las familias del mundo.—Boletin del Ins- tituto Oceanografico Universidad de Oriente Cumana 14(1):71—94. Gravier, M. C. 1901.Contribution a l’etude des anne- lides polychetes de la Mer Rouge.—Nouvelles Archives du Museum Paris, ser. 4. 3:147—268, pls. 7-10. Grube, E. 1855. Beschreinbungen neuer oder wenig bekannter Anneliden.—Archib fiir Naturgeschi- chte, Berlin 21:81—136, pls. 3-5. . 1856. Annulata Orstediana. Enumeratio An- nulatorum, quae in itinere per Indiam occiden- talem et Americam centralem 1845—1848 sus- cepto legit cl. A. S. Orsted, adjectis speciebus nonnullis a cl. H. Kré6yero in itinero ad Amer- icam meridionalem collectis. Pt. 1.—Videnska- belige Meddeleser fra den naturhistoriske For- ening 1 Kjobenhavyn, vol. for 1856:44—62. Hartman, O. 1961. Polychaetes annelids from Califor- nia.—Allan Hancock Pacific Expeditions 25:1— 22 . 1968. Atlas of the Errantiate Polychaetous Annelids from California. University of South- ern California, Los Angeles. 828 p. Johnson, H. P. 1897. A preliminary account of the ma- rine annelids of the Pacific Coast, with descrip- tions of new species, part 1. The Euphosynidae, Amphinomidae, Palmyridae, Polynoidae, and Sigalionidae.—Proceedings of the California Academy of Sciences, ser. 3, Zoology 1:153-— 199, Laverde-Castillo J. 1986. Lista anotada de los polique- VOLUME 116, NUMBER 1 tos (Annelida) registrados para el Pacifico Col- ombiano con notas preliminares de su zoogeo- grafia.—Actualidades Bioldgicas 15(58):123- 130. Loépez, E, G. San Martin,P. Cladera, & M. Capa. 1997. Fauna de Anélidos Poliquetos del Parque Na- cional de Coiba (Panama). In: S. Castroviejo (ed.). Flora y Fauna del Parque Nacional de Coiba (Panama). p 57-73. Monro, C. C. A. 1928a. Polychaeta of the Families Polynoidae and Acoetidae from the vicinity of the Panama Canal, collected by Dr, C. Crosland and Dr. Th. Mortesen.—Journal of the Linnean Society of London 36:553—576. . 1928b. Papers from Dr. Mortensen’s Pacific Expedition 1914—16, 45.On the Polychaeta col- lected by Dr. Th. Mortensen off the coast of Panama.—Videnskabelige Meddelelser fra Dansk Naturhisurisk Forening 85:75—103. . 1933a. The Polychaeta Errantia collected by Dr. C. Crossland at Colon in the Panama region and the Galapagos Islands during the expedition of the “St. George”’.—Proceedings of the Zoo- logical Society of London: 1—96. . 1933b. The Polychaeta Sedentaria collected by Dr. C. Crossland at Col6n in the Panama region and the Galapagos Islands during the ex- pedition of the ““St. George’’.—Proceedings of the Zoological Society of London:1039—1092. Perkins, T. 1985. Chrysopetalum, Bhawania and two new genera of Chrysopetalidae (Polychaeta). Principally from Florida.—Proceedings of the Biological Society of Washington 98:856—915. San Martin, G. 1986. Anélidos poliquetos procedentes de la I expedici6n cubano-espanola a la isla de la Juventud y archipiélago de los Canarreos. DS Familias Chysopetalidae y Amphinomidae.— Revista de Investigaciones Marinas 7(1):17—30. . In press. Anélidos poliquetos. Chrysopetali- dae. Jn Fauna Ibérica. Ramos, M. A. et al., eds., Museo Nacional de Ciencias Naturales, CSIC, Madrid. , E. Lopez, M. S. Redondo, M. Capa, P. Clad- era, & A. Laborda. 1997. El bentos marino del Parque Nacional de Coiba (Panama). Pp. 33-55 in S. Castroviejo, ed., Flora y Fauna del Parque Nacional de Coiba (Panama), AECI, Madrid, 534 pp. , E. Lopez, & A. I. Camacho. 1998. First record of a freshwater Pisionidae (Polychaeta): de- scription of a new species from Panama with a key to the new species of Pisione.—Journal of Natural History 32:1115—1127. Schmarda, L. K. 1861. Neue wirbellose Thiere beo- bachtet und gesammelt auf einer Reise um die Erde 1853 bis 1857, vol. 1. Turbellarien, Rota- torien und Anneliden, Pt. 2.—Leipzig, Wilhelm Engelmann, Pp. 1—164. 22 pls. Verrill, A. E. 1900. Adittions to the Turbellaria, Nem- ertina, and Annelida of the Bermudas, with the revision of some new England genera and spe- cies.—Transactions of the Connecticut Acade- my of Arts and Sciences 10:595—671. Watson-Russell, C. 1998. Description of Arichlidon new genus and two new species from Australia; Bhawania reyssi redescribed and assigned to Arichlidon (Chrysopetalidea: Polychaeta).— Beagle 14:159-176. Webster, H. E. 1884. The Annelida from Bermuda, col- lected by Mr. G. Brown Goode.—Bulletin of United States National Museum 25:307-—327, pls. 7-12. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):96—157. 2003. Annotated checklist of decapod crustaceans of Atlantic coastal and continental shelf waters of the United States Martha S. Nizinski National Marine Fisheries Service National Systematics Laboratory, Smithsonian Institution, P.O. Box 37012, NHB, MRC-0153, Washington, D.C. 20013-7012, U.S.A., e-mail: nizinski.martha@nmnh.si.edu Abstract.—The decapod crustacean assemblage inhabiting estuarine, neritic and continental shelf waters (to 190 m) of the temperate eastern United States is diverse, with 391 species reported from Maine to Cape Canaveral, Florida. Three recognized biogeographic provinces (Boreal in part, Virginian and Car- olinian) are included in this region. The assemblage contains 122 shrimp spe- cies (28 penaeids, 2 stenopodids, and 92 carideans), 10 thalassinideans, 8 lob- sters, 61 anomurans and 190 brachyurans. Since previous compilation of this fauna, 12 additional species have been described, including four carideans, one callianassid, four anomurans, and three brachyurans. Range extensions into the region have been reported for another five species (Parapenaeus americanus, Scyllarides aequinoctialis, Petrolisthes armatus, Dromia erythropus, Clythro- cerus nitidus). One species, Hemigrapsus sanguineus, has been introduced and become established throughout intertidal environments from southern Maine to northern North Carolina. Six species previously recorded from this region are no longer considered to occur there. Two of these species occur south of the Carolinian biogeographic province, three others are now known to occur only in the Pacific Ocean, and one species previously considered as likely to occur in the region has never actually been recorded there. Scientific nomenclature for all species recorded from the region is updated and referenced. Geographic distributions are summarized for each species incorporating recent published information where available. The decapod crustacean assemblage (shrimps, lobsters, and crabs) inhabiting es- tuarine, neritic and continental shelf waters (to 190 m) of the temperate eastern United States is diverse and fairly well known. The most recent comprehensive review of this fauna (Williams 1984) recognized 342 spe- cies of decapods and listed 14 extralimital species (species thought to occur inciden- tally in the region). Williams’ (1984) mono- graph is widely used and continues to be the best available reference for decapod crustaceans of the east coast of the United States. However, in the nearly 20 years since publication of Williams (1984), knowledge regarding shallow water (=190 m) decapods in this region has advanced. Numerous changes in taxonomy and/or sys- tematic placement of species occurring in this region have also been published in pa- pers scattered throughout the literature. Objectives of this paper are to re-evaluate the taxonomic status and to update nomen- clature for the 342 nominal species of deca- pods treated by Williams (1984). The list of species he presented is updated by inclusion of all decapods now known to occur in this region based on recently published infor- mation. Emendations to Williams’ original list include (1) the addition of species in this region described after Williams’ pub- lication, (2) adding of species newly re- VOLUME 116, NUMBER 1 ported from the region as published range extensions and introductions, and (3) re- moval of species no longer considered to occur in the region based on published range revisions. The emended list also in- cludes 13 species listed in Williams (1984) as extralimital, as well as 16 other species that Williams (1984) did not treat, but which previously had been recorded from this region. Methods This compilation largely follows the clas- sification and arrangement of Martin & Da- vis (2001). Based on other recent system- atic studies (cited below), some species and genera have been moved to different fami- lies, and other species have been reassigned to different genera. Where new systematic information has necessitated changes, up- dated nomenclature with relevant references is provided, together with a cross-reference to names used in Williams (1984). Symbols preceding species listings highlight the fol- lowing changes between Williams (1984) and the present compilation: ® denotes new additions to the decapod assemblage; * de- notes changes in nomenclature and/or sys- tematic placement. Additionally, distribu- tional information is reported for each spe- cies that occurs in the region with recently published range revisions incorporated and referenced. Justification explaining changes in systematic placement of taxa, together with appropriate reference sources, are pro- vided in a remarks section when relevant. The region of coverage in the present work is the same as that in Williams (1984), spanning depths from shallow water to the 100-fathom (ca. 190 m) depth contour in three biogeographic provinces. These prov- inces, defined by marine climatic zones (Williams 1984, Engle & Summers 1999), are, from north to south, Boreal (represent- ed by the region from Maine to north of Cape Cod, Massachusetts), Virginian (Cape Cod, Massachusetts, to Cape Hatteras, North Carolina) and Carolinian (Cape Hat- 97 teras, North Carolina, to Cape Canaveral, Florida). Williams (1984) commented that inclusion or exclusion of some species from his list was subjective. For example, he re- garded some species as extralimital (i.e., their center of distribution was outside the region) and did not consider them to be part of the assemblage despite the fact that they had been recorded from the region. Wil- liams’ treatment of extralimital taxa has proved to be problematic and confusing es- pecially in attempts to summarize decapod crustacean diversity of the region. To alle- viate this confusion, the present work does not consider species as extralimital (sensu Williams). Based on available distributional information, the present compilation treats species as either occurring in the geograph- ic region and thus included in the list, or as not occurring in the region and thus ex- cluded. Systematic Account Order Decapoda Latreille, 1802 Suborder Dendrobranchiata Bate, 1888 Infraorder Penaeidea Rafinesque, 1815 Superfamily Penaeoidea Rafinesque, 1815 > Family Aristeidae Wood-Mason, 1891 Remarks.—Pérez Farfante & Kensley (1997) diagnosed the family and genera, presented a key to the genera based on mor- phological characters, and illustrated im- portant diagnostic characters for the genera included in their study. > Aristaeomorpha foliacea (Risso, 1827) Remarks.—Species diagnosis and figures are provided in Pérez Farfante (1988) and Rodriguez (1993). This species occurs be- tween 170—1300 m (Pérez Farfante 1988). Known range.—Western Atlantic: Mas- sachusetts to Florida, Gulf of Mexico, Ca- ribbean Sea to Venezuela (Pérez Farfante & Kensley 1997), and Brazil (D’Incao 1998); Eastern Atlantic: Bay of Biscay to Western Sahara; Azores; Madeira; Canary Islands; Mediterranean Sea; Indo-West Pacific: 98 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON South Africa to Mozambique; Madagascar; Réunion Island; Maldive Islands; Sri Lan- ka; Indonesia; Philippines; Taiwan; Japan; Australia; New Zealand; New Caledonia; Wallis and Futuna Islands; Fiji (Pérez Far- fante & Kensley 1997). Family Penaeidae Rafinesque, 1815 Remarks.—Dall et al. (1990) provided a comprehensive account of many aspects of penaeid systematics, biology and ecology. Pérez Farfante & Kensley (1997) diagnosed the family and genera, presented a key to the genera based on morphological charac- ters, and illustrated important diagnostic characters for the genera included in their study. Former subgenera (Farfantepenaeus and Litopenaeus) of the more inclusive Penaeus were raised to full generic status by Pérez Farfante & Kensley (1997). Re- sults of some molecular phylogenies (Bald- win et al. 1998, Gusmao et al. 2000) have failed to support elevation of these subgen- era to genera, however, an alternative hy- pothesis of relationships proposed in the molecular phylogeny of Maggioni et al. (2001) supported conclusions of Pérez Far- fante & Kensley (1997) derived from their analysis of morphological characters. Tra- chypenaeus, as previously defined and as listed in Williams (1984), was shown to be a complex of five genera (Pérez Farfante & Kensley 1997). Only one of these five gen- era, Rimapenaeus Pérez Farfante & Ken- sley, 1997, is represented in the region by a single species. * Farfantepenaeus aztecus (Ives, 1891) Penaeus (Farfantepenaeus) aztecus Ives.— Williams, 1984:24. Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts, to Florida; Gulf of Mexico to Yucatan (Williams 1984, Pérez Farfante & Kensley 1997). * Farfantepenaeus brasiliensis (Latreille, 1817) Penaeus (Farfantepenaeus) brasiliensis La- treille—Williams, 1984:28. Remarks.—This species occurs in shal- low water to 366 m (Caetano da Costa et al. 2000), but more frequently at depths of less than 60 m (D’Incao 1998). Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Florida Keys; southern Gulf of Mexico (off Campeche) and Yucatan; Caribbean Sea to Rio Grande do Sul, Brazil; Bermuda (Williams 1984, Pérez Farfante & Kensley 1997). * Farfantepenaeus duorarum (Burkenroad, 1939) Penaeus (Farfantepenaeus) duorarum Bur- kenroad.—Williams, 1984:28. Known range.—Western Atlantic: lower Chesapeake Bay to Florida; Gulf of Mexico to tip of Yucatan Peninsula; Bermuda (Wil- hams 1984, Pérez Farfante & Kensley IOO7)) * Litopenaeus setiferus (Linnaeus, 1767) Penaeus (Litopenaeus) setiferus (Linnae- us).—Williams, 1984:32. Known range.—Western Atlantic: New York to St. Lucie Inlet, Florida; Gulf of Mexico to Yucatan (Williams 1984, Pérez Farfante & Kensley 1997). Metapenaeopsis goodei (Smith, 1885) Known range.—Western Atlantic: north- east of Cape Lookout, North Carolina, through the Florida Straits to Alabama; off Cape Catoche, Yucatan; Bahamas through the Caribbean Sea and along the coasts of Central and South America to Espirito San- to, Brazil; Bermuda (Williams 1984, Pérez Farfante & Kensley 1997, D’Incao 1998). ~ Parapenaeus americanus Rathbun, 1901 Remarks.—Pérez Farfante & Kensley (1997) record this species from off New England. This species occurs at 37—412 m (Pérez Farfante 1977b, D’Incao 1998). Known range.—Western Atlantic: off VOLUME 116, NUMBER 1 New England; off Ponte Vedra, Florida; Gulf of Mexico; Bahamas; Cuba; Puerto Rico; St. Lucia; Belize; Brazil (Rio de Ja- neiro to Rio Grande do Sul); northern Uru- guay (Pérez Farfante 1977b, Pérez Farfante & Kensley 1997, D’Incao 1998). Parapenaeus politus Smith, 1881 Remarks.—Parapenaeus politus occurs over mud and sandy mud sediments at depths of 3—752 m, but is usually found be- tween 65—275 m (Rodriguez 1993). Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts, south through Gulf of Mexico, Caribbean Sea to French Guiana, and Para, Brazil (Williams 1984, Pérez Farfante & Kensley 1997, D’Incao 1998). P Penaeopsis serrata Bate, 1881 Remarks.—Species diagnosis, descrip- tion, illustrations, as well as, color and size information were provided in Pérez Farfan- te (1980). Penaeopsis serrata occurs at 120—750 m, with maximum concentrations occurring between 300—450 m (Pérez Far- fante 1980, Rodriguez 1993). Known range.—Western Atlantic: New Jersey to Gulf of Mexico; Caribbean Sea to French Guiana; southern Brazil; Eastern At- lantic: Portugal to northwest coast of Africa (Pérez Farfante 1980, Pérez Farfante & Kensley 1997). * Rimapenaeus constrictus (Stimpson, 1871) Trachypenaeus_ constrictus (Stimpson).— Williams, 1984:38. Remarks.—Rimapenaeus constrictus is a new combination proposed by Pérez Far- fante & Kensley (1997). This species oc- curs in shallow water to 127 m (Caetano da Costa et al. 2000). Known range.—Western Atlantic: Nova Scotia; Chesapeake Bay to Florida Keys, Gulf of Mexico; Bermuda; Caribbean Sea to Santa Catarina, Brazil (Williams 1984, 99 Pérez Farfante & Kensley 1997, D’Incao 1998). Xiphopenaeus kroyeri (Heller, 1862) Remarks.—This species occurs in shal- low water to 70 m (Caetano da Costa et al. 2000). Known range.—Western Atlantic: Vir- ginia (Maris 1986) to Rio Grande do Sul, Brazil, including the Gulf of Mexico and Caribbean Sea (Williams 1984, Pérez Far- fante & Kensley 1997, Caetano da Costa et al. 2000); Eastern Pacific: Punta Piaxtla, Sinaloa, Mexico to Paita, Peru (Pérez Far- fante & Kensley 1997). Family Sicyoniidae Ortmann, 1898 Remarks.—Pérez Farfante & Kensley (1997) provided family and generic diag- noses and illustrations of diagnostic char- acters. Sicyonia brevirostris Stimpson, 1871 Known range.—Western Atlantic: Nor- folk, Virginia, to Florida, through the Gulf of Mexico; Campeche and Yucatan banks; Cuba; Bahamas; Eastern Pacific: off south- ern Mexico (Williams 1984, Pérez Farfante & Kensley 1997). Sicyonia burkenroadi Cobb, 1971 Known range.—Western Atlantic: Cape Lookout, North Carolina, to Florida, through the Gulf of Mexico; West Indies; Caribbean coasts of Central and South America to Bahia, Brazil (Williams 1984, Pérez Farfante & Kensley 1997). Sicyonia dorsalis Kingsley, 1878 Known range.—Western Atlantic: Cape Hatteras, North Carolina, through the Gulf of Mexico to Texas; Caribbean coasts of Central and South America to southern Bra- zil (Williams 1984, Pérez Farfante & Ken- sley 1997, D’Incao 1998). 100 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Sicyonia laevigata Stimpson, 1871 Known range.—Western Atlantic: Cape Hatteras, North Carolina, to southern Flor- ida; Gulf of Mexico; West Indies; Carib- bean coasts of Mexico, Central America and South America to Rio Grande do Sul, Brazil (Williams 1984, Pérez Farfante & Kensley 1997, Caetano da Costa et al. 2000); Eastern Pacific: southern Gulf of California; Costa Rica; Panama (Pérez Far- fante & Kensley 1997). Sicyonia parri (Burkenroad, 1934) Known range.—Western Atlantic: Beau- fort, North Carolina, through the Gulf of Mexico; West Indies to Sao Paulo, Brazil (Williams 1984, Pérez Farfante & Kensley 1997, Caetano da Costa et al. 2000). Sicyonia stimpsoni Bouvier, 1905 Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Florida through the Gulf of Mexico; West Indies; Caribbean coasts of Mexico, Central and northern South America to Suriname (Williams 1984, Pérez Farfante & Kensley 1997). Sicyonia typica (Boeck, 1864) Known range.—Western Atlantic: Wrightsville Beach, North Carolina, through the Gulf of Mexico; Cuba through the West Indies; Caribbean coasts of Cen- tral and South America to Rio Grande do Sul, Brazil (Williams 1984, Pérez Farfante & Kensley 1997, D’Incao 1998, Caetano da Costa et al. 2000). Family Solenoceridae Wood-Mason, 1891 Remarks.—Pérez Farfante & Kensley (1997) diagnosed the family and genera, presented a key to the genera based on mor- phological characters, and illustrated im- portant diagnostic characters for the genera included in their study. > Hadropenaeus affinis (Bouvier, 1906) Remarks.—Description, affinities, com- parisons, and illustrations are included in Pérez Farfante (1977a). This species occurs between 165-570 m (Pérez Farfante 1977a). Known range.—Western Atlantic: North Carolina to Florida; Gulf of Mexico; Carib- bean; Eastern Atlantic: Cape Verde Islands (Pérez Farfante 1977a, Pérez Farfante & Kensley 1997). & Hadropenaeus modestus (Smith, 1885) Remarks.—Description, affinities, com- parisons, and illustrations are included in Pérez Farfante (1977a). This species occurs at depths of 150—550 m (Pérez Farfante 1977a). Known range.—Western Atlantic: Dela- ware to Florida, Gulf of Mexico, Bahamas, Caribbean, and northern Brazil (Pérez Far- fante 1977a, Pérez Farfante & Kensley 1997). Mesopenaeus tropicalis (Bouvier, 1905) Known range.—Western Atlantic: north- east of Cape Lookout, North Carolina, to southern Brazil, including Gulf of Mexico (Williams 1984, Pérez Farfante & Kensley 1997). > * Pleoticus robustus (Smith, 1885) Hymenopenaeus robustus Smith.—Wil- liams, 1984:484. Remarks.—Williams (1984) considered this species to be extralimital. Pérez Farfan- te (1977a) concluded that Hymenopenaeus comprised a complex of genera and con- sequently transferred H. robustus and two other species (H. muelleri and H. stein- dachneri) to the genus Pleoticus Bate, 1888. Description, affinities, comparisons, and illustrations of P. robustus were includ- ed in Pérez Farfante (1977a) and Squires (1990). This species occurs at 70—915 m, but is most abundant between 250—475 m (Pérez Farfante 1977a, Rodriguez 1993). VOLUME 116, NUMBER 1 Known range.—Western Atlantic: Mas- sachusetts to Gulf of Mexico; Caribbean Sea to French Guiana (Pérez Farfante 1977a, 1988; Squires 1990; Pérez Farfante & Kensley 1997). Solenocera atlantidis Burkenroad, 1939 Known range.—Western Atlantic: North Carolina to southern Brazil, including Gulf of Mexico and West Indies (Williams 1984, Pérez Farfante & Kensley 1997). Solenocera necopina Burkenroad, 1939 Known range.—Western Atlantic: off Oregon Inlet, North Carolina, to Uruguay, including Gulf of Mexico and Bahamas (Williams 1984, Pérez Farfante & Kensley IDO). Solenocera vioscai Burkenroad, 1934 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina, to Florida and Gulf of Mexico (Williams 1984, Pérez Farfante & Kensley 1997). Superfamily Sergestoidea Dana, 1852 Family Luciferidae De Haan, 1849 Remarks.—Pérez Farfante & Kensley (1997) provided a family diagnosis and il- lustrated important diagnostic characters. Lucifer faxoni Borradaile, 1915 Known range.—Western Atlantic: Long Island Sound, New York, to Rio Grande do Sul, Brazil, including Gulf of Mexico, Ca- ribbean Sea, and Bermuda (Williams 1984, Pérez Farfante & Kensley 1997, D’Incao 1998); Eastern Atlantic: Senegal; Congo (Williams 1984, Pérez Farfante & Kensley 1997). & Lucifer typus H. Milne Edwards, 1837 Remarks.—Williams (1984) did not in- clude this species in his study because depth of occurrence was centered beyond the limits of bathymetric range he consid- 101 ered. Lucifer typus is usually found offshore (>180 m bottom depth), but occasionally is collected in shallower waters (Bowman and McCain 1967). Bowman and McCain (1967) provided information on diagnostic characters. Known range.—Western Atlantic: New- foundland; east coast of United States (Maine to Florida); Sargasso Sea; Brazil; Eastern Atlantic: Mediterranean Sea; Cape of Good Hope; east coast of South Africa; East-Cental Pacific: Baja California; Gulf of California to north of 4°; Indo-West Pa- cific: Bay of Bengal; Philippines; Queens- land, Australia (Abele and Kim 1986, Pérez Farfante & Kensley 1997, D’Incao 1998). Family Sergestidae Dana, 1852 Remarks.—Pérez Farfante & Kensley (1997) provided family and generic diag- noses and illustrations of diagnostic char- acters. Acetes americanus carolinae Hansen, IQs Remarks.—Pérez Farfante and Kensley (1997) considered subspecific designation valid. Known range.—Western Atlantic: lower Chesapeake Bay through Gulf of Mexico to Panama; Suriname; French Guiana (Wil- liams 1984, Pérez Farfante and Kensley IQO7). Suborder Pleocyemata Burkenroad, 1963 Infraorder Stenopodidea Claus, 1872 Family Stenopodidae Claus, 1872 Stenopus hispidus (Olivier, 1811) Known range.—Western Atlantic: Ber- muda; North Carolina to southern Florida, through the Gulf of Mexico to Fernando de Noronha and Espirito Santo, Brazil (Wil- liams 1984, Coelho & Ramos-Porto 1998a); Central Pacific: Hawaii; Indo-West Pacific: Durban, South Africa; Red Sea; Japan; western Australia; eastern Australia through New Caledonia; New Hebrides; Lord Howe 102 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Island; northern New Zealand to Tuamotu Archipelago (Williams 1984). Stenopus scutellatus Rankin, 1898 Known range.—Western Atlantic: Ber- muda; South Carolina; Gulf of Mexico to Fernando de Noronha and Rio Grande do Norte, Brazil (Williams 1984, Coelho & Ramos-Porto 1998a). Infraorder Caridea Dana, 1852 Superfamily Pasiphaeoidea Dana, 1852 Family Pasiphaeidae Dana, 1852 Remarks.—Subgeneric designations within Leptochela are considered valid by Holthuis (1993). Leptochela (Leptochela) papulata Chace, 1976 Known range.—Western Atlantic: Georges Bank, off Massachusetts; North Carolina to Georgia; eastern Gulf of Mex- ico (Williams 1984). Leptochela (Leptochela) serratorbita Bate, 1888 Known range.—Western Atlantic: Beau- fort, North Carolina, to South Carolina; western Gulf of Mexico; Florida Keys to Leeward Islands; Amapa to Pernambuco, and Sao Paulo, Brazil (Williams 1984, Ra- mos-Porto & Coelho 1998). Leptochela (Proboloura) carinata Ortmann, 1893 Known range.—Western Atlantic: Georges Bank, off Massachusetts; South Carolina; Gulf of Mexico, through Baha- mas to Alagoas, Brazil (Williams 1984, Ra- mos-Porto & Coelho 1998). Pasiphaea multidentata Esmark, 1866 Known range.—Western Atlantic: south- east of Greenland to Cape Cod, Massachu- setts, including Gulf of St. Lawrence and Gulf of Maine; Eastern Atlantic: Iceland; Norway to British Isles; Bay of Biscay; Mediterranean to Adriatic (Williams 1984, Squires 1990). Superfamily Bresilioidea Calman, 1896 * Family Disciadidae Rathbun, 1902 Remarks.—Martin & Davis (2001) rec- ognized the family Disciadidae. Species of Discias were previously considered part of the Bresiliidae (Williams 1984). Discias atlanticus Gurney, 1939 Remarks.—Criales & Lemaitre (1997) reported the discovery of this species in- habiting tubes of the polychaete Chaetop- terus variopedatus. This represents the first reported occurrence of this symbiotic rela- tionship for these species. These authors also discussed morphological sexual dimor- phism. Known range.—Western Atlantic: Ber- muda; Savannah, Georgia; Fort Pierce, Florida; Gulf of Mexico; Guadeloupe; Co- lombia; Maranhao and Ceara, Brazil (Ken- Ssley 1983, Williams 1984, Criales & Le- maitre 1997, Ramos-Porto & Coelho 1998); Eastern Atlantic: Cape Verde Islands; Ga- bon (Williams 1984); Indian Ocean: northern Kenya (Williams 1984); Red Sea; Western Pacific: Great Barrier Reef, Aus- tralia (Kensley 1983). & Discias vernbergi Boothe & Heard, 1987 Remarks.—Boothe & Heard (1987:506) described this species and provided a di- agnosis, illustrations, and size and sexual maturity information. This species occurs at 54-74 m (Boothe & Heard 1987). Known range.—Western Atlantic: Geor- gia; eastern Gulf of Mexico (west Florida) (Boothe & Heard 1987). Superfamily Palaemonoidea Rafinesque, 1815 > Family Anchistioididae Borradaile, 1915 Anchistioides antiguensis (Schmitt, 1924) Remarks.—This species was listed under family Palaemonidae, subfamily Pontoni- VOLUME 116, NUMBER 1 inae, in Williams (1984). Chace (1992) pro- vided evidence for familial separation of Anchistioides; Holthuis (1993) also adopted this arrangement. Known range.—Western Atlantic: Ber- muda; South Carolina; west Florida through West Indies to Maranhao, Pernambuco, and Alagoas, Brazil (Williams 1984, Ramos- Porto & Coelho 1998). Family Gnathophyllidae Dana, 1852 Gnathophyllum modestum Hay, 1917 Known range.—Western Atlantic: Beau- fort, North Carolina; Biscayne Bay, Florida (Williams 1984). Family Palaemonidae Rafinesque, 1815 Subfamily Palaemoninae Rafinesque, 1815 Remarks.—Subgeneric designations within Palaemonetes are considered valid by Holthuis (1993). Brachycarpus biunguiculatus (Lucas, 1849) Remarks.—This species occurs from shallow water to 105 m (Ramos-Porto & Coelho 1998). Known range.—Western Atlantic: Ber- muda; Cape Fear, North Carolina; western Gulf of Mexico through West Indies to Cu- ragao and Old Providence Island (Williams 1984); Amapa to Espirito Santo and Fer- nando de Noronha Archipelago, Brazil (Ra- mos-Porto & Coelho 1998); Eastern Atlan- tic: Mediterranean; West Africa; Eastern Pacific: west American coast; Indo-Pacific (Williams 1984). Leander tenuicornis (Say, 1818) Known range.—Western Atlantic: New- foundland Banks to Falkland Islands; trop- ical and subtropical waters worldwide ex- cept for west coast of Americas (Williams 1984). 103 Macrobrachium acanthurus (Wiegmann, 1836) Known range.—Western Atlantic: Neuse River estuary, North Carolina, to Rio Grande do Sul, Brazil (Williams 1984, Ra- mos-Porto & Coelho 1998). Macrobrachium carcinus (Linnaeus, 1758) Known range.—Western Atlantic: St. Augustine, St. Johns County and Silver Glen Springs, Marion County, Florida, to Rio Grande do Sul, Brazil, including Gulf of Mexico and Caribbean Sea (Williams 1984, Ramos-Porto & Coelho 1998). Macrobrachium ohione (Smith, 1874) Known range.—Western Atlantic: James River, Hopewell, Virginia, to southern Georgia; coastal Alabama to Aransas Bay, Texas; Freshwater: Mississippi River and tributaries upstream to McCurtain County, Oklahoma; Fort Smith, Arkansas; St. Louis, Missouri; Washington County, Ohio (Wil- liams 1984). Macrobrachium olfersii (Wiegmann, 1836) Known range.—Western Atlantic: lower Cape Fear River, North Carolina; Florida; Louisiana; Texas; Veracruz, Mexico, to Rio Grande do Sul, Brazil (Williams 1984, Ra- mos-Porto & Coelho 1998). Palaemonetes (Palaemonetes) intermedius Holthuis, 1949 Known range.—Western Atlantic: Vine- yard Sound, Massachusetts, to Port Aransas Texas; Bahia de la Ascension, Quintana Roo, Mexico (Williams 1984). Palaemonetes (Palaemonetes) pugio Holthuis, 1949 Known range.—Western Atlantic: Verte River, west of St. Modeste, Quebec, through Nova Scotia to Corpus Christi, Texas; Matamoros, Tamaulipas, to Campe- 104 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON che, Mexico (Williams 1984, Squires 1990, Rodriguez-Almaraz et al. 2000). Palaemonetes (Palaemonetes) vulgaris (Say, 1818) Known range.—Western Atlantic: south- ern Gulf of St. Lawrence to Cameron Coun- ty, Texas; Rio Champoton and near Progre- so, Yucatan, Mexico (Williams 1984, Squires 1990). Subfamily Pontoniinae Kingsley, 1878 Remarks.—Subgeneric designations within Periclimenes are considered valid by Holthuis (1993). Neopontonides beaufortensis (Borradaile, 1920) Known range.—Western Atlantic: Beau- fort, North Carolina, to Grand Isle, Louisi- ana; Caledonia Bay, Panama; Antigua (Wil- liams 1984). Periclimenaeus schmitti Holthuis, 1951 Known range.—Western Atlantic: Bogue Sound, Black Rocks off New River and Lockwoods Folly River, North Carolina; Dry Tortugas, Florida (Williams 1984). Periclimenaeus wilsoni (Hay, 1917) Known range.—Western Atlantic: Beau- fort, North Carolina; Sapelo Island, Geor- gia; Loggerhead Key, near Dry Tortugas; Franklin County, Florida (Williams 1984). * Periclimenes (Harpilius) americanus (Kingsley, 1878) Periclimenes americanus (Kingsley ).—Wil- hams, 1984:83. Known range.—Western Atlantic: Beau- fort, North Carolina, to western Gulf of Mexico, through West Indies to Aruba; Amapa to Pernambuco, Espirito Santo and Sao Paulo, Brazil (Williams 1984, Ramos- Porto & Coelho 1998). * Periclimenes (Periclimenes) iridescens Lebour, 1949 Periclimenes iridescens Lebour.—Williams, 1984:85. Known range.—Western Atlantic: Ber- muda; off Cape Hatteras, North Carolina; southern and northwestern Florida; Tobago; Cubagua Island, Venezuela (Williams 1984). * Periclimenes (Periclimenes) longicaudatus (Stimpson, 1860) Periclimenes longicaudatus (Stimpson).— Williams, 1984:86. Known range.—Western Atlantic: Cape Hatteras, North Carolina, to southwestern Florida; West Indies to Sao Paulo, Brazil (Williams 1984). * Periclimenes (Periclimenes) pedersoni Chace, 1958 Periclimenes pedersoni Chace.—Williams, 1984:87. Known range.—Western Atlantic: off Cape Lookout, North Carolina; off north- west Florida (?); Bahamas through West In- dies to Bonaire; Belize (Williams 1984). Pontonia domestica Gibbes, 1850 Known range.—Western Atlantic: Atlan- tic Beach near Beaufort Inlet, North Caro- lina, through Gulf of Mexico to South Pa- dre Island, Texas; Bahamas; Caribbean coast of Costa Rica (Williams 1984, Strenth & Chace 1995, Vargas & Cortés 1999); Eastern Atlantic: Madeira (Williams 1984). > Pontonia manningi Fransen, 2000 Remarks.—Fransen (2000:101) de- scribed this species and provided illustra- tions, size, color, and host information. This species is found at shallow depths to 80 m (Fransen 2000). Known range.—Western Atlantic: North Carolina to Caribbean Sea; Gulf of Mexico; VOLUME 116, NUMBER 1 Eastern Atlantic: Canary Islands; Cape Verde Islands (Fransen 2000). Superfamily Alpheoidea Rafinesque, 1815 Family Alpheidae Rafinesque, 1815 & Alpheus angulosus McClure, 2002 Alpheus angulatus McClure, 1995. Remarks.—McClure (1995:85) described this species and provided a diagnosis, illus- trations, color, size, and habitat information. McClure (2002) provided a replacement name (A. angulosus) for this species be- cause angulatus had previously been used by Coutiére (1905) for a subspecies of A/- pheus (A. strenuus var. angulatus) from the Indo-Pacific. Comparisons of morphologi- cal variation among species, including A. angulosus, of the edwardsii group of AI- pheus occurring in the northern Gulf of Mexico and northwestern Atlantic were provided in McClure (1995) and McClure & Wicksten (1997). Known range.—Western Atlantic: Beau- fort, North Carolina, to Quintana Roo, Mexico, including northern Gulf of Mexi- co; Haiti (McClure 1995). Alpheus armillatus H. Milne Edwards, 1837 Known range.—Western Atlantic: Ber- muda; North Carolina through Gulf of Mexico and West Indies to Santa Catarina, Brazil (Williams 1984, Christoffersen 1998). P Alpheus estuariensis Christoffersen, 1984 Remarks.—Christoffersen (1984:191) described this species and provided illustra- tions, color description, and ecological in- formation. This species occurs from the in- tertidal region to depths of 22 m (Christof- fersen 1984). Known range.—Western Atlantic: east coast of Florida (near Jacksonville and In- dian River region); northern Gulf of Mex- ico; Cuba; Dominican Republic; Trinidad; 105 Curacao; Ceara to Parana, Brazil (Christof- fersen 1984, 1998; McClure & Wicksten 1997). Christoffersen (1998) reported that his earlier records (Christoffersen 1984) of A. estuariensis in the Gulf of Mexico are actually A. angulatus (= A. angulosus, see above). However, McClure & Wicksten (1997) have reported A. estuariensis occur- ring in the northern Gulf of Mexico. Alpheus formosus Gibbes, 1850 Known range.—Western Atlantic: Ber- muda; near Beaufort, North Carolina, through Gulf of Mexico and West Indies to Sao Paulo, Brazil (Williams 1984, Chris- toffersen 1998). Alpheus heterochaelis Say, 1818 Remarks.—McClure (1995) redescribed this species and designated a neotype to clarify taxonomic confusion surrounding this species. Comparisons of morphological variation among species, including A. het- erochaelis, of the edwardsii group of AI- pheus occurring in northern Gulf of Mexico and northwestern Atlantic were provided in McClure (1995) and McClure & Wicksten CIQO7). Known range.—Western Atlantic: lower Chesapeake Bay to Aransas County, Texas; Bermuda; Cuba; Curagao; Suriname; Para to Paraiba, Brazil (Williams 1984, McClure 1995, Christoffersen 1998). Alpheus normanni Kingsley, 1878 Remarks.—Based on material examined from both western Atlantic and eastern Pa- cific locations, Kim & Abele (1988) con- cluded that variation between Pacific and Atlantic forms of A. normanni was suffi- cient to recognize two species. This deci- sion has not gained universal acceptance since A. normanni continues to be used for the Atlantic form. If A. normanni is restrict- ed to the eastern Pacific as suggested by Kim & Abele (1988), then western Atlantic specimens would be known as A. packardii 106 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Kingsley, 1880 (Christoffersen 1998). Fur- ther research is needed to resolve this tax- onomic question. Known range.—Western Atlantic: around Cape Charles, Virginia, and lower Chesapeake Bay through Gulf of Mexico and West Indies to Sao Paulo, Brazil; Ber- muda; Eastern Pacific: Gulf of California; Panama (Williams 1984, Christoffersen 1998). * Automate dolichognatha De Man, 1888 Automate gardineri Coutiére.—Williams, 1984:100. Remarks.—Chace (1988) placed A. gardineri and three other nominal species in the synonymy of ‘the variable and wide- ranging’ A. dolichognatha. Known range.—Western Atlantic: Beau- fort Inlet, North Carolina; Yucatan; Virgin Islands; Barbados; Rio de Janeiro, Brazil; Indo-Pacific: Red Sea to Samoa (Williams 1984, Christoffersen 1998). Chace (1988) considered the range to be pantropical, ex- cept for the eastern Atlantic. Automate evermanni Rathbun, 1901 Known range.—Western Atlantic: Vir- ginia to Texas; Puerto Rico; Sao Paulo and Rio Grande do Sul, Brazil; Eastern Atlantic: Cape Verde Islands; Liberia to Nigeria (Williams 1984, Chace 1988, Christoffersen 1998). Leptalpheus forceps Williams, 1965 Known range.—Western Atlantic: North Carolina to Sergipe, Brazil (Williams 1984, Christoffersen 1998). Synalpheus fritzmuelleri Coutiere, 1909 Known range.—Western Atlantic: Ber- muda; Beaufort, North Carolina, to Santa Catarina, Brazil (Williams 1984, Christof- fersen 1998); South Atlantic: St. Helena Is- land; Eastern Pacific: Baja California (Wil- liams 1984). Synalpheus longicarpus (Herrick, 1891) Known range.—Western Atlantic: Beau- fort, North Carolina, to west Flower Garden Reef, southeast of Galveston, Texas; Yu- catan, Mexico through West Indies to Rio de Janeiro, Brazil (Williams 1984, Chris- toffersen 1998). Synalpheus minus (Say, 1818) Known range.—Western Atlantic: Ber- muda; Cape Hatteras, North Carolina, to Sao Paulo, Brazil (Williams 1984, Chris- toffersen 1998). Synalpheus townsendi Coutiére, 1909 Known range.—Western Atlantic: Ber- muda; Beaufort, North Carolina, to Rio de Janeiro, Brazil; Rocas Atoll, Brazil (Wil- liams 1984, Christoffersen 1998); Eastern Pacific: Gulf of California (Williams 1984). Family Hippolytidae Dana, 1852 & Bythocaris nana Smith, 1885 Remarks.—This species was considered extralimital by Williams (1984). Abele & Martin (1989) redescribed the species, and provided illustrations and developmental notes. Bythocaris nana occurs at 79-1175 m depth and is not common (Abele & Mar- tin 1989). Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts, to southern Florida; northeastern Gulf of Mexico (Abele & Martin 1989). © Caridion gordoni (Bate, 1858) Remarks.—Williams (1984) considered this species to be extralimital. Squires (1990) provided a description and illustra- tions. This species occurs at 5-421 m depth (Williams & Wigley 1977). Known range.—Western Atlantic: south- western Newfoundland to Chesapeake Bay; North Atlantic: Iceland; Eastern Atlantic: northern Europe to Bay of Biscay (Williams 1984, Squires 1990). VOLUME 116, NUMBER 1 Eualus fabricii (Kr@yer, 1841) Known range.—Western Atlantic: Hud- son Bay, Foxe Basin, and northwestern Greenland to Cape Cod; North Pacific: Chukchi Sea; Bering Sea to British Colum- bia; Western Pacific: Sea of Okhotsk to Sea of Japan (Williams 1984, Squires 1990). Eualus gaimardii (H. Milne Edwards, 1837) Remarks.—Squires (1990) recognized two subspecies, FE. g. gaimardii and E. g. belcheri, both of which show considerable variation in diagnostic features and overlap completely in distribution. Williams (1984), however, believed that observed morpho- logical variation was not sufficient to war- rant recognition of more than one species among this material. Known range.—Western Atlantic: Greenland and Baffin Island to Cape Cod; Eastern Atlantic: Spitsbergen to North Sea; Arctic Ocean: Point Barrow to Siberia; North Pacific: south to Sitka, Alaska (Wil- hams 1984, Squires 1990). Eualus pusiolus (Kr@yer, 1841) Known range.—Western Atlantic: Gulf of St. Lawrence to Cape Henry, Virginia; Northeastern Atlantic: Iceland; Murman Sea to Channel Islands, southward along Bay of Biscay to Spain; Catalonian coast of Spain; North Pacific: Chukchi and Bering seas to British Columbia and Washington; Western Pacific: Sea of Okhotsk and Sea of Japan (Williams 1984, Squires 1990). Exhippolysmata oplophoroides (Holthuis, 1948) Remarks.—This species occurs over a bathymetric range of 5—45 m (Rodriguez 1993, Caetano da Costa et al. 2000). Known range.—Western Atlantic: Cape Fear, North Carolina, to Port Aransas, Tex- as; Guyana to northern Uruguay (Williams 1984, Caetano da Costa et al. 2000). 107 Hippolyte coerulescens (Fabricius, 1775) Known range.—Tropical and subtropical Atlantic Ocean, including south of the Grand Banks in the Gulf Stream; and Sar- gasso Sea (Williams 1984, Squires 1990). * Hippolyte obliquimanus Dana, 1852 Hippolyte curacaoensis Schmitt.—Wil- liams, 1984:117. Remarks.—Udekem d’Acoz (1997) ex- amined topotypic specimens of Hippolyte obliquimanus Dana and H. exilirostratus Dana and determined that these two nomi- nal species are identical. He also concluded that H. curacaoensis Schmitt is conspecific with H. obliquimanus. Udekem d’ Acoz (1997) redescribed H. obliquimanus based on this new information. Known range.—Western Atlantic: Beau- fort and Sneads Ferry, North Carolina; Puerto Rico; West Indies from Cuba to Cu- ragao; Venezuela to Santa Catarina, Brazil (Williams 1984, Udekem d’Acoz 1997, Christoffersen 1998). Hippolyte pleuracanthus (Stimpson, 1871) Known range.—Western Atlantic: Con- necticut to North Carolina (Williams 1984). Hippolyte zostericola (Smith, 1873) Known range.—Western Atlantic: south- ern Massachusetts; Bermuda; North Caro- lina to Yucatan; Trinidad; Curacao; Ceara, Brazil (Williams 1984, Christoffersen 1998). Christoffersen (1998) noted that pre- viously reported occurrences of this species in Brazil may actually refer to H. obliqui- manus. Latreutes fucorum (Fabricius, 1798) Known range.—Western Atlantic: New- foundland to Brazil (Pernambuco to Bahia), including Gulf of Mexico south to Bahia de la Ascension, Quintana Roo, Mexico; East- ern Atlantic: Azores; Cape Verde Islands 108 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Williams 1984, Christoffersen 1998, Rod- riguez-Almaraz et al. 2000). Latreutes parvulus (Stimpson, 1866) Known range.—Western Atlantic: Beau- fort, North Carolina, to Buenos Aires, Ar- gentina; Eastern Atlantic: West Africa (Wil- liams 1984, Christoffersen 1998). Lebbeus groenlandicus (Fabricius, 1775) Known range.—Western Atlantic: Hud- son Bay to Greenland, southward to Rhode Island; North Pacific: Arctic Canada, south- ern Chukchi Sea through Bering Sea to Pu- get Sound; Western Pacific: Okhotsk Sea southward to Vladivostok (Williams 1984, Squires 1990). * Lebbeus microceros (Kr@yer, 1841) Lebbeus zebra (Leim).—Williams, 1984: 25). Remarks.—Holthuis (1947) recognized both L. microceros and L. zebra as valid species. Couture & Trudel (1968) noted great similarity between these nominal spe- cies and commented that they might be syn- onymous, but rarity of material prevented an adequate evaluation of these taxa (Wil- liams 1984). Squires (1990) placed L. zebra into the synonymy of L. microceros without comment or explanation for this action. Chace (1997) also listed L. zebra in the syn- onymy of L. microceros. Known range.—Western Atlantic: South- ern Greenland; Foxe Basin; Ungava Bay; Newfoundland to New Brunswick, includ- ing Gulf of St. Lawrence to southeast of Isles of Shoals; North Pacific: possibly from Bering Sea to Kamchatka; Eastern Pa- cific: Checleset Bay, Vancouver Island (Williams 1984, Squires 1990). Lebbeus polaris (Sabine, 1824) Known range.—Circumarctic; Western Atlantic: southward to Chesapeake Bay; North Pacific: Okhotsk Sea; Bering Sea to British Columbia; Eastern Atlantic: Arctic to the Hebrides (Williams 1984, Squires 1990). Lysmata rathbunae Chace, 1970 Known range.—Western Atlantic: Ber- muda; Cape Fear, North Carolina; east coast of Florida to Yucatan; Venezuela (Williams 1984). Lysmata wurdemanni (Gibbes, 1850) Known range.—Western Atlantic: New Jersey to Port Aransas, Texas; Suriname; French Guiana; Ceara and Bahia to Rio Grande do Sul, Brazil (Williams 1984, Christoffersen 1998). Spirontocaris liljeborgii (Danielssen, 1859) Known range.—Western Atlantic: Foxe Channel and Davis Strait; Greenland, to off Delaware Bay; Northeastern Atlantic: Ice- land; Spitsbergen; Murman coast to south coast of England; west and southwestern Ireland; Arctic: Alaska (Williams 1984, Squires 1990). Spirontocaris phippsii (Kr@yer, 1841) Known range.—Circumarctic; Western Atlantic: Cornwallis Island; Hudson Bay to Martha’s Vineyard; North Atlantic: Spits- bergen to southern Norway; Britain; Arctic: north of Siberia; Beaufort Sea; North Pa- cific: Bering Sea to Siberian east coast; Shumagin Islands, Alaska (Williams 1984, Squires 1990). Spirontocaris spinus (Sowerby, 1805) Known range.—Circumarctic; Western Atlantic: Hudson Bay; Foxe Basin; Green- land, southward to Massachusetts Bay; North Atlantic: Spitsbergen to northern North Sea and Irish Sea; North Pacific: Plo- ver Bay, Siberia; Bering Sea; Shumagin Is- lands, Alaska; Puget Sound, Washington; Western Pacific: Okhotsk Sea; Sea of Japan (Williams 1984, Squires 1990). VOLUME 116, NUMBER 1 Thor dobkini Chace, 1972 Known range.—Western Atlantic: off Shackleford Bank, North Carolina, to Yu- catan; Louisiana; north coast of Cuba (Wil- liams 1984). Thor floridanus Kingsley, 1878 Known range.—Western Atlantic: Black Rocks off New River, North Carolina (7), to Yucatan (Williams 1984). Thor manningi Chace, 1972 Known range.—Western Atlantic: Beau- fort, North Carolina, to Yucatan, through West Indies to Curacao; Brazil (Paraiba, Bahia, and Sao Paulo); Eastern Pacific: Is- las Tres Marias, Mexico (Williams 1984, Christoffersen 1998). Tozeuma carolinense Kingsley, 1878 Known range.—Western Atlantic: Vine- yard Sound, Massachusetts, to Col6n, Pan- ama, including Gulf of Mexico; through West Indies to Curacao; Paraiba to Alagoas and Sao Paulo, Brazil (Williams 1984, Christoffersen 1998). Tozeuma serratum A. Milne-Edwards, 1881 Known range.—Western Atlantic: Non- amesset Island, Massachusetts; off Cape Hatteras and Cape Lookout, North Caroli- na; Cape Canaveral; extreme southern and northwestern Florida; Barbados; Colombia; off Rio de Janeiro and Sao Paulo, Brazil (Williams 1984, Christoffersen 1998). > Trachycaris rugosa (Bate, 1888) Remarks.—Based on available informa- tion, Williams (1984) presumed this species to occur beyond the southern limits of the region. However, Criales (1992) recorded T. rugosa from southern areas within this re- gion. Criales (1992) provided a redescrip- tion of the species, illustrations, and com- 109 mented on color, interspecific comparisons, and sexual dimorphism. Known range.—Western Atlantic: South Carolina through the Gulf of Mexico; Ca- ribbean Sea to Santa Marta, Colombia (Cri- ales 1992). Family Ogyrididae Holthuis, 1955 Ogyrides alphaerostris (Kingsley, 1880) Known range.—Western Atlantic: Vir- ginia to Rio Grande do Sul, Brazil, includ- ing Gulf of Mexico (Williams 1984, Chris- toffersen 1998). Ogyrides hayi Williams, 1981 Known range.—Western Atlantic: Beau- fort, North Carolina, to Sebastian Inlet, Florida; northwestern Florida to Mississip- pi; Puerto Rico; Pernambuco and Sao Pau- lo, Brazil (Williams 1984, Christoffersen 1998). > Superfamily Processoidea Ortmann, 1890 Remarks.—Previously the family Proces- sidae was placed in the superfamily Al- pheoidea (Williams 1984). Chace (1992) erected the superfamily Processoidea, which consists of a single family (Proces- sidae). Holthuis (1993) also adopted this ar- rangement. Family Processidae Ortmann, 1890 Nikoides schmitti Manning & Chace, 1971 Known range.—Western Atlantic: east of Cape Lookout, North Carolina; Biscayne Bay and Dry Tortugas, Florida; Guade- loupe; the Guianas (Williams 1984). Processa bermudensis (Rankin, 1900) Known range.—Western Atlantic: Ber- muda; Cape Hatteras, North Carolina, to northwestern Florida; Veracruz, Mexico; Cuba; Puerto Rico; Guadeloupe; Venezuela; Bahia, Rio de Janeiro and Parana, Brazil (Williams 1984, Christoffersen 1998). 110 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Processa fimbriata Manning & Chace, 1971 Known range.—Western Atlantic: off New River, North Carolina, to Rio de Ja- neiro, Brazil (Williams 1984, Christoffersen 1998). Processa guyanae Holthuis, 1959 Known range.—Western Atlantic: Cape Hatteras, North Carolina, to eastern Gulf of Mexico, including northern coast of Cuba; Suriname to Uruguay (Williams 1984, Christoffersen 1998). Processa hemphilli Manning & Chace, 1971 Known range.—Western Atlantic: east of Cape Lookout and Bogue Sound, North Carolina; east coast of Florida; northwest Florida; Guadeloupe; Brazil to Buenos Ait- res (Williams 1984, Christoffersen 1998). Processa profunda Manning & Chace, 1971 Known range.—Western Atlantic: south- east of Cape Hatteras, North Carolina; off South Carolina; Gulf of Mexico, off south- ern and western Florida; Suriname to Uru- guay (Williams 1984, Christoffersen 1998). Processa vicina Manning & Chace, 1971 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; northwest Florida; off Isla Margarita, Ve- nezuela (Williams 1984). Superfamily Pandaloidea Haworth, 1825 Remarks.—Christoffersen (1989) pro- posed a phylogeny of this taxon. Family Pandalidae Haworth, 1825 * Atlantopandalus propinqvus (G.O. Sars, 1870) Pandalus propinquus G.O. Sars.—WiI- liams, 1984:156. Remarks.—In his revision of the genus Pandalus, Komai (1999) transferred Pan- dalus propinqvus to a monotypic genus At- lantopandalus; a description was provided. Additionally, Komai (1999) stated that G.O. Sars consistently spelled the name of this species as “‘propinqgvus’ in the original de- scription and that this spelling should be used. This species occurs at depths of 20— 2180 m (Komai 1999). Known range.—Western Atlantic: Greenland; Davis Strait to Delaware Bay, including Gulf of St. Lawrence and Gulf of Maine; North Atlantic: Iceland; Eastern At- lantic: Norway to the British Isles; Bay of Biscay (Komai 1999). Dichelopandalus leptocerus (Smith, 1881) Known range.—Western Atlantic: Gulf of St. Lawrence and St. Mary’s Bay, New- foundland, to off Oregon Inlet, North Car- olina; North Pacific: Shumagin Bank, Alas- ka (Williams 1984, Squires 1990). P Heterocarpus ensifer A. Milne-Edwards, 1881 Remarks.—Diagnostic characters of this species were reported in Crosnier & Forest (1973), Chace (1985), and Rodriguez (1993). This species occurs on muddy bot- toms usually at 200—885 m (Chace 1985), but may be found at shallower depths (140 m; Rodriguez 1993). Known range.—Western Atlantic: North Carolina to the Guianas, including the Gulf of Mexico and Caribbean Sea; Brazil (Cha- ce 1985, Rodriguez 1993, Ramos-Porto & Coelho 1998). Pandalus borealis Kr@yer, 1838 Remarks.—The geographic range of Pandalus borealis has been reported as Arctic boreal in both Atlantic and Pacific oceans (Williams 1984). Recent studies (Squires 1992, Komai 1999) have demon- strated that morphological differences be- tween Atlantic and Pacific ‘varieties’ were VOLUME 116, NUMBER 1 sufficient to warrant full species rank for both. Pandalus borealis is restricted to the Atlantic Ocean and considered a geminate species of P. eous Makarov from Pacific localities (Squires 1992). Known range.—Western Atlantic: west- ern Greenland to Gulf of Maine; North At- lantic: Barents Sea to the North Sea (Komai IDE). Pandalus montagui Leach, 1814 Known range.—Arctic-boreal; Western Atlantic: Greenland and Hudson Bay to Rhode Island; North Atlantic: Iceland; White Sea; Eastern Atlantic: Norway to the western Baltic; North Sea; British Isles (Williams 1984, Squires 1990). Pantomus parvulus A. Milne-Edwards, 1883 Known range.—Western Atlantic: Cape Lookout, North Carolina, to Yucatan; Puerto Rico; St. Croix; Suriname; Uruguay (Williams 1984, Christoffersen 1989). & Plesionika edwardsii (Brandt, 1851) Remarks.—Chan & Yu (1991) provided a diagnosis, color description, and interspe- cific comparisons. This species occurs on muddy bottoms at depths of 50-690 m (Rodriguez 1993), commonly at 200—400 m (Chan & Yu 1991). Known range.—Western Atlantic: Vir- ginia to Gulf of Mexico; Eastern Atlantic: Mediterranean to Angola; Indo-Pacific (Chan & Yu 1991). > Plesionika martia (A. Milne-Edwards, 1883) Remarks.—Williams (1984) considered this species to be extralimital. This species occurs at 165—2100 m depth (Williams & Wigley 1977). Known range.—Western Atlantic: south of Nantucket, Massachusetts, to Brazil; Bermuda; Eastern Atlantic: southwest Ire- land; Bay of Biscay to southern Africa; 11] Mediterranean Sea; Central Pacific: Hawaii; Indo-Pacific: East Africa to Japan, includ- ing Gulf of Aden; New Zealand; southeast- ern Australia (Williams & Wigley 1977). P Plesionika tenuipes (Smith, 1881) Remarks.—Williams (1984) considered this species to be extralimital. This species occurs at 159-476 m depth (Williams 1984). Known range.—Western Atlantic: Rhode Island to southern Florida; Gulf of Mexico (Williams 1984). > * Plesionika willisi (L. H. Pequegnat, 1970) Parapandalus willisi L. H. Pequegnat.— Williams, 1984:484. Remarks.—Williams (1984) considered this species to be extralimital. Chace (1985) placed Parapandalus into the synonymy of Plesionika. This species occurs at 150—500 m depth (Williams & Wigley 1977). Known range.—Western Atlantic: south of Martha’s Vineyard, Massachusetts; Gulf of Mexico to French Guiana (Williams & Wigley 1977). Superfamily Crangonoidea Haworth, 1825 Family Crangonidae Haworth, 1825 Argis dentata (Rathbun, 1902) Remarks.—Komai (1997) evaluated the systematics of this species and several con- geners to clarify their taxonomic status. In this study, he also provided a detailed de- scription, illustrations, color, size and com- parative information for A. dentata. Known range.—Arctic-boreal; Western North Atlantic: northwest Greenland; Hud- son Bay; Canadian Arctic islands to Nova Scotia; North Pacific: Bering Sea to Sitka, Alaska; southeast coast of Kamchatka; northern Okhotsk Sea (Komai 1997). * Crangon (Crangon) septemspinosa Say, 1818 Crangon septemspinosa Say.—Williams, hOS4 aso: 112 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Remarks.—Subgeneric designation con- sidered valid by Holthuis (1993). Known range.—Primarily subarctic-bo- real; Western Atlantic: northern Gulf of St. Lawrence to east Florida (Williams 1984, Squires 1990). Squires (1990) restricted the geographic range to the north Atlantic. * Philocheras gorei (Dardeau, 1980) Pontophilus gorei Dardeau.—Williams, 1984:161. Remarks.—Chace (1984) provided mor- phological evidence for recognizing Phil- ocheras as a valid genus and removed it from the synonymy of Pontophilus. Chris- toffersen’s (1988) phylogenetic analysis supported this arrangement. Bathymetric range was reported as 59—194 m (Christof- fersen 1998). Known range.—Western Atlantic: central Georgia; central eastern Florida; Gulf of Mexico (southwestern Florida, Cape San Blas and Padre Island, Texas); Rio de Ja- neiro, Brazil; Uruguay (Dardeau & Heard 1983, Williams 1984, Christoffersen 1988). Pontophilus brevirostris Smith, 1881 Known range.—Western Atlantic: Gulf of Maine to eastern Gulf of Mexico, off Dry Tortugas, Florida; Cuba (Williams 1984). Pontophilus norvegicus (M. Sars, 1861) Known range.—Western Atlantic: Greenland to Maryland; North Atlantic: Iceland; Spitsbergen; Murman coast; East- ern Atlantic: northwestern Europe, includ- ing British Isles to Bay of Biscay; and Bal- earic Islands (Williams 1984, Squires 1990). Sabinea sarsii Smith, 1879 Known range.—Western Atlantic: Davis Strait to southeast of Nantucket, Massachu- setts; North Atlantic: Iceland; Eastern At- lantic: northern Europe (Williams 1984, Squires 1990). Sabinea septemcarinata (Sabine, 1824) Known range.—Western Atlantic: Hud- son Bay and Greenland to Massachusetts Bay; Arctic-North Atlantic: Iceland (except south coast), Kara, White and Barents seas; Wrangel Island; Eastern Atlantic: north of Faroes; Norway (north of 67°N); British Isles; North Pacific: Arctic Canada and Alaska to Point Barrow and Chukchi Sea (Williams 1984, Squires 1990). Sclerocrangon boreas (Phipps, 1774) Known range.—Western Atlantic: Hud- son Bay; east and west Greenland south to Cape Cod; Arctic-North Atlantic: Iceland; Spitsbergen; Kara and White seas; Franz Joseph Land; Novaya Zemlya; Eastern At- lantic: Faroes; Norway (north of approxi- mately 67°N); Arctic-North Pacific: Arctic Canada, north coast of Alaska; Chukchi Sea south to British Columbia (Williams 1984, Squires 1990). Infraorder Astacidea Latreille, 1802 Superfamily Nephropoidea Dana, 1852 Family Nephropidae Dana, 1852 Subfamily Nephropinae Dana, 1852 Homarus americanus H. Milne Edwards, i333) 7 Known range.—Western Atlantic: New- foundland to Cape Hatteras, North Caroli- na; occasionally Wilmington, North Caro- lina and south Florida (Williams 1984, Cof- er-Shabica & Nielsen 1988, Holthuis 1991). Williams (1984) noted that this species is occasionally found as far south as Wil- mington, North Carolina. One specimen of H. americanus, however, was collected off southern Florida (Miami Beach, 252 m; Cofer-Shabica & Nielsen 1988), suggesting that the range of this species occasionally extends further south, especially in deeper waters. > Subfamily Thymopinae Holthuis, 1974 > Nephropsis aculeata Smith, 1881 Remarks.—Species diagnosis provided in Hothuis (1991). This species occurs on mud VOLUME 116, NUMBER 1 or fine sand sediments at 137—824 m, but usually between 200—600 m (Holthuis 1991). Known range.—Western Atlantic: Mas- sachusetts to French Guiana, including Gulf of Mexico and Caribbean Sea; Bermuda (Holthuis 1991). * Infraorder Thalassinidea Latreille, 1831 Remarks.—Thalassinidea was considered a Section under the infraorder Anomura in Williams (1984:180). Poore (1994) pro- posed a phylogeny of the infraorder and provided a new classification, diagnoses, and keys to families and currently recog- nized genera. Superfamily Callianassoidea Dana, 1852 Family Callianassidae Dana, 1852 Remarks.—Manning & Felder (1991) re- stricted this family as a result of their re- vision of the American Callianassidae. In a phylogenetic analysis of generic relation- ships within the family based on 93 adult morphological characters, Tudge et al. (2000) determined that the Callianassidae comprised a monophyletic group. Sakai (1999) presented very different conclusions regarding the composition of, and generic relationships within, the Callianassidae. Sa- kai’s work, however, was not conducted within a phylogenetic framework and is widely regarded as controversial. More re- search is needed to resolve the composition and relationships of genera in this and re- lated families and subfamilies. Subfamily Callianassinae Dana, 1852 Remarks.—Tudge et al. (2000) conduct- ed a phylogenetic analysis of relationships within the family Callianassidae based on adult morphological characters and they de- termined that the subfamily Callianassinae was a monophyletic group. * Biffarius biformis (Biffar, 1971) Callianassa biformis Biffar.—Williams, 1984:182. WS: Remarks.—In a recent revision of the American Callianassidae, Manning & Feld- er (1991) concluded that Callianassa was a composite of numerous genera. Manning & Felder (1991) described the genus Biffarius and provided a diagnosis, illustrations, and comparative information regarding the gen- era within the subfamily Callianassinae. Known range.—Western Atlantic: Bass River, Yarmouth, Nova Scotia; Nantucket Sound, Massachusetts; Chesapeake Bay (7); North Inlet, South Carolina, to McIntosh County, Georgia; Franklin County, north- west Florida (Williams 1984). * Gilvossius setimanus (De Kay, 1844) Callianassa atlantica Rathbun.—Williams, 1984:180. Remarks.—Manning (1987) addressed the status of Gonodactylus setimanus DeKay. He determined this species was val- id as Callianassa setimanus (DeKay) and that it was the senior synonym of Calli- anassa atlantica Rathbun. Members of Cal- lianassa sensu stricto, as restricted by Man- ning & Felder (1991), are not represented in the American fauna. The genus Gilvos- sius was described (Manning & Felder, 1992) with Gonodactylus setimanus DeKay (= Callianassa atlantica Rathbun) as the type species. Known range.—Western Atlantic: Bass River, Nova Scotia, to Georgia; Franklin County, Florida (Williams 1984). > Necallianassa berylae Heard & Manning, 1998 Remarks.—Heard & Manning (1998: 883-884) described this genus and species and provided illustrations and diagnostic comparisons with eastern Atlantic conge- ners. Necallianassa berylae occurs at depths of 35-75 m (Heard & Manning 1998). Known range.—Western Atlantic: South Carolina and Georgia (Heard & Manning 1998). 114 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Subfamily Callichirinae Manning & Felder, 1991 Remarks.—Tudge et al. (2000) conduct- ed a phylogenetic analysis of relationships within the family Callianassidae utilizing adult morphological characters. They deter- mined that the subfamily Callichirinae was a paraphyletic group and that more research is needed to resolve relationships within this group. * Callichirus major (Say, 1818) Callianassa major Say.—Williams, 1984: 183. Remarks.—Manning & Felder (1986) re- defined Callichirus to remove the ambigu- ity surrounding previous concepts of this genus. Known range.—Western Atlantic: Beau- fort Inlet, North Carolina, to Cape Canav- eral, Florida; Grand Terre Island to Tim- balier Island, Louisiana; Sergipe to Santa Catarina, Brazil (Williams 1984, Rodrigues & Shimizu 1998). Family Laomediidae Borradaile, 1903 Naushonia crangonoides Kingsley, 1897 Known range.—Western Atlantic: Bass River, Yarmouth, Nova Scotia; Vineyard Sound and Elizabeth Islands, Massachu- setts, to Bogue Sound, North Carolina (Wil- liams 1984). Family Upogebiidae Borradaile, 1903 Upogebia affinis (Say, 1818) Remarks.—Williams (1993) provided an improved diagnosis and detailed description of this species and noted that its southern limits of distribution are in Texas. Speci- mens identified as U. affinis from more southern locations (West Indies to Sao Pau- lo, Brazil) are actually Upogebia paraffinis Williams, 1993. Known range.—Western Atlantic: Mas- sachusetts to southern Texas (Williams 1993) Superfamily Axioidea Huxley, 1879 Family Axtidae Huxley, 1879 P Axius armatus Smith, 1881 Remarks.—Kensley (2001) redescribed this species, provided illustrations, and compared A. armatus to its sympatric con- gener, A. serratus. Although Sakai & de Saint Laurent (1989) questioned the generic placement of this species, Kensley (2001) provided morphological evidence support- ing placement of this species in the genus Axius. This apparently rare species occurs at depths of 108—260 m (Kensley 2001). Known range.—Western Atlantic: Mas- sachusetts to South Carolina (Kensley 2001). Axius serratus Stimpson, 1852 Remarks.—Kensley (2001) redescribed, illustrated, and compared this species to its Atlantic congeners. This species occurs at depths of 19-220 m (Kensley 2001), which represents an increase in the maximum re- ported depth of occurrence. Known range.—Western Atlantic: Nova Scotia to Maryland (Kensley 2001). * Calaxius jenneri (Williams, 1974) Axiopsis jenneri (Williams).—Williams, 1984:185. Remarks.—Sakai & de Saint Laurent (1989) described the genus Calaxius in their revision of the family Axudae. Known range.—Western Atlantic: Cape Lookout, North Carolina (Williams 1984). > Family Calocarididae Ortmann, 1891 Remarks.—Kensley (1989) reinstated the Calocarididae Ortmann, which contained only Calocaris, and then expanded the fam- ily with the addition of several genera. Calocaris templemani Squires, 1965 Known range.—Western Atlantic: New- foundland to the Gulf of Maine; Cape Lookout, North Carolina (Williams 1984, VOLUME 116, NUMBER I Kensley 1989, Sakai & de Saint Laurent 1989, Squires 1990). The southernmost lo- cality reported for this species is southeast of Cape Lookout, North Carolina (Williams 1984). This appears to be the only recorded occurrence of C. templemani this far south. Infraorder Palinura Latreille, 1802 Superfamily Palinuroidea Latreille, 1802 Family Palinuridae Latreille, 1802 Panulirus argus (Latreille, 1804) Remarks.—Based on mtDNA _ samples from individuals collected from nine loca- tions between Bermuda and Venezuela, Sil- berman et al. (1994) hypothesized that Pan- ulirus argus was genetically homogenous throughout the tropical western Atlantic and Caribbean. However, three individuals with distinctly different mtDNA haplotypes, col- lected off Miami, Florida, were identified (Silberman et al. 1994). Sarver et al. (1998) compared mtDNA sequences between western Atlantic-Caribbean populations and Brazilian lobsters and found sufficient dif- ferences to suggest that P. argus is a com- plex of two species or subspecies. In addi- tion to genetic differences, characteristic color patterns were also identified distin- guishing the Brazilian P. argus from the Caribbean form (Sarver et al. 1998). They recommended provisional recognition of two subspecies until formal taxonomic re- vision could be completed: Panulirus argus argus representing populations from Ber- muda to Venezuela and Panulirus argus westonii representing populations from Bra- zil. Sarver et al. (2000) re-examined the ge- netically distinct individuals of Silberman et al. (1994) and determined that these un- usual individuals were the provisionally recognized Brazilian form of P. argus (P. argus westonii). However, until formal re- vision is conducted, the taxonomic status of these subspecies remains uncertain. Known range.—Western Atlantic: Ber- muda; North Carolina to Rio de Janeiro, Brazil, including Gulf of Mexico and Ca- LES ribbean Sea (Williams 1984, Holthuis 1991). Family Scyllaridae Latreille, 1825 Subfamily Arctidinae Holthuis, 1985 P Scyllarides aequinoctialis (Lund, 1793) Remarks.—This species was originally considered to occur outside the region, and therefore. was not included in Williams (1984). Lyons (1970) provided a detailed description. Scyllarides aequinoctialis oc- curs on sandy or rocky bottoms at depths of 0-180 m, usually 1—64 m (Lyons 1970, Holthuis 1991). Known range.—Western Atlantic: Ber- muda; South Carolina to southern Brazil, including the Gulf of Mexico, Caribbean Sea, and West Indies (Holthuis 1991). Scyllarides nodifer (Stimpson, 1866) Known range.—Western Atlantic: Ber- muda; Cape Lookout, North Carolina, to Yucatan, including Gulf of Mexico (Wil- liams 1984, Holthuis 1991). Subfamily Scyllarinae Latreille, 1825 Scyllarus americanus (Smith, 1869) Known range.—Western Atlantic: Bogue Inlet, North Carolina, to Campeche Banks, Mexico; Venezuela (Williams 1984). Scyllarus chacei Holthuis, 1960 Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Bahia, Brazil, including Gulf of Mexico, West Indies, and Caribbean Sea (Williams 1984, Coelho & Ramos-Porto 1998b). Scyllarus depressus (Smith, 1881) Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts; Cape Hat- teras, North Carolina, to Sao Paulo, Brazil, including Gulf of Mexico and West Indies (Williams 1984). 116 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Infraorder Anomura MacLeay, 1838 Superfamily Galatheoidea Samouelle, 1819 Family Galatheidae Samouelle, 1819 Galathea rostrata A. Milne-Edwards, 1880 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to southern Florida; northwestern Florida to Mississippi River delta; off Cape Catoche, Yucatan (Williams 1984). > Munida forceps A. Milne-Edwards, 1880 Remarks.—Williams (1984) did not in- clude this species in his study because the known geographic range and depth of oc- currence were outside the region. This spe- cies is now known from off southern New England and Virginia within the region. Munida forceps occurs at 80—337 m (Wil- liams 1988, Melo-Filho & Melo 1992). Melo-Filho & Melo (1992) provided a de- scription, illustrations, and measurements; Williams (1988) provided a color descrip- tion and figure of male carapace. Known range.—Western Atlantic: Veatch and Lydonia Canyons, off southern New England (Williams 1988); Virginia; Florida; Gulf of Mexico; Antilles; Guianas; Brazil (Melo-Filho & Melo 1992). Munida iris iris A. Milne-Edwards, 1880 Known range.—Western Atlantic: off Nova Scotia through southeastern Gulf of Mexico to near Cozumel Island, Yucatan; through Caribbean islands to Rio Grande do Sul, Brazil (Williams 1984, Squires 1990, Melo-Filho 1998). Munida irrasa A. Milne-Edwards, 1880 Known range.—Western Atlantic: off Bermuda; off Cape Lookout, North Caro- lina, through eastern Gulf of Mexico; Ca- ribbean Sea to Uruguay (Williams 1984). Munida longipes A. Milne-Edwards, 1880 Known range.—Western Atlantic: Balti- more Canyon, off the coast of Maryland (Williams 1988) southward through Gulf of Mexico to Belize; through West Indies to Curacao (Williams 1984); Sao Paulo to Rio Grande do Sul, Brazil (Melo-Filho 1998). Munida pusilla Benedict, 1902 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to Straits of Florida, through eastern Gulf of Mexico to Yucatan; Colombia; Trinidad (Williams 1984); Amapa, Brazil (Melo-Filho 1998). Munida valida Smith, 1883 Known range.—Western Atlantic: off Nova Scotia to Rio Grande do Sul, Brazil, including Gulf of Mexico and Caribbean (Williams 1984, Squires 1990, Melo-Filho & Melo 1992, Melo-Filho 1998). Family Porcellanidae Haworth, 1825 Euceramus praelongus Stimpson, 1860 Known range.—Western Atlantic: Dela- ware Bay to Aransas area, Texas (Williams 1984). Megalobrachium soriatum (Say, 1818) Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Port Aransas, Texas; West Indies to Barbados; Contoy Island, Mexico; Bahia Caledonia and Galeta Island, Panama (Williams 1984); Ceara to Sao Paulo, Brazil (Veloso 1998). Pachycheles pilosus (H. Milne Edwards, 1837) Known range.—Western Atlantic: Charleston, South Carolina; Key West to Sarasota Bay, Florida; through West Indies to Tobago and Aruba (Williams 1984). VOLUME 116, NUMBER 1 Pachycheles rugimanus A. Milne-Edwards, 1880 Known range.—Western Atlantic: Cape Lookout, North Carolina, through Florida to St. Thomas, U.S. Virgin Islands; Contoy Island, Mexico (Williams 1984); Amapa to Para (Veloso 1998) and Pernambuco, Brazil (Williams 1984). © Petrolisthes armatus (Gibbes, 1850) Remarks.—This species was recently re- ported in the region (Knott et al. 2000). Al- though Knott et al. (2000) reported the presence of this crab as an introduction, the occurrence of P. armatus in Georgia and South Carolina possibly represents a north- ern range extension because the original de- scription by Gibbes (1850) listed the local- ity of specimens examined as ‘*‘Florida.” This species is the dominant decapod crus- tacean on rocky substrates in Georgia and is well established on rocky rubble, oyster reefs, and other shallow subtidal and inter- tidal habitats throughout Georgia and South Carolina (Knott et al. 2000). Known range.—Western Atlantic: Ber- muda; South Carolina southward through the Gulf of Mexico; Bahamas; West Indies and Caribbean; northern South America to Santa Catarina, Brazil; Central Atlantic: As- cension Island; Eastern Atlantic: tropical West Africa; Eastern Pacific: Gulf of Cali- fornia to Peru (Veloso 1998, Knott et al. 2000). Petrolisthes galathinus (Bosc, 1802) Known range.—Western Atlantic: Cape Hatteras, North Carolina, through Gulf of Mexico and Caribbean Sea to Rio Grande do Sul, Brazil; Eastern Pacific: Isla San Lu- cas, Costa Rica, to off La Libertad, Ecuador (Williams 1984, Veloso 1998). Polyonyx gibbesi Haig, 1956 Known range.—Western Atlantic: Woods Hole, Massachusetts, to Uruguay (Williams 1984). AG Porcellana sayana (Leach, 1820) Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Rio Grande do Sul, Brazil, including Gulf of Mexico and Caribbean Sea (Williams 1984, Veloso 1998). Porcellana sigsbeiana A. Milne-Edwards, 1880 Known range.—Western Atlantic: off Martha’s Vineyard, Massachusetts, to southwestern Caribbean Sea off Colombia; West Indies to Virgin Islands (Williams 1984); Para and Maranhao, Brazil (Veloso 1998). Superfamily Hippoidea Latreille, 1825 Family Albuneidae Stimpson, 1858 Albunea gibbesii Stimpson, 1859 Known range.—Western Atlantic: Ber- muda; east of Cape Lookout, North Caro- lina, to Texas, through West Indies to Sao Paulo, Brazil (Williams 1984, Calado 1998). Albunea paretii Guérin-Méneville, 1853 Known range.—Western Atlantic: Beau- fort Inlet, North Carolina, to Corpus Christi, Texas, through West Indies to Rio Grande do Sul, Brazil (Williams 1984, Calado 1998); Eastern Atlantic: Cape Verde Is- lands; Senegal to Ghana (Williams 1984). Lepidopa websteri Benedict, 1903 Known range.—Western Atlantic: mouth of Chesapeake Bay; Drum Inlet, North Car- olina, to east central Florida; Tampa Bay, Florida; Petit Bois Island, Mississippi (Wil- hams 1984, Manning 1988). Family Hippidae Latreille, 1825 Emerita benedicti Schmitt, 1935 Known range.—Western Atlantic: Charleston County, South Carolina, to Ve- racruz, Mexico (Williams 1984). 118 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Emerita talpoida (Say, 1817) Known range.—Western Atlantic: Har- wich, Massachusetts, to Horn Island, Mis- sissipp1; Progreso, Yucatan, Mexico (Wil- liams 1984). Superfamily Coenobitoidea Dana, 1851 Remarks.—Williams (1984) recognized two superfamilies (Coenobitoidea and Pa- guroidea) for hermit crabs. McLaughlin (1983) had recommended that the super- family Coenobitoidea be suppressed and that a single superfamily of hermit crabs should be recognized. Martin & Davis (2001) accepted the arrangement proposed by McLaughlin (1983). However, Mc- Laughlin & Lemaitre (2001a) recently pro- vided evidence to support reinstatement of the Coenobitoidea. Based on their results, two distinct superfamilies should be rec- ognized with respect to hermit crab higher classification. Family Diogenidae Ortmann, 1892 Cancellus ornatus Benedict, 1901 Known range.—Western Atlantic: Cape Fear, North Carolina, through eastern Gulf of Mexico; Greater and Lesser Antilles, to Bahia, Brazil (Williams 1984, Rieger 1998). Clibanarius vittatus (Bosc, 1802) Known range.—Western Atlantic: Poto- mac River, Gunston, Virginia, to Santa Ca- tarina, Brazil, including the Gulf of Mexico (Williams 1984). Dardanus fucosus Biffar & Provenzano, noyZ Known range.—Western Atlantic: near Cape Hatteras, North Carolina, to Para, Brazil (Williams 1984, Rieger 1998). Dardanus insignis (de Saussure, 1858) Known range.—Western Atlantic: off Oregon Inlet, North Carolina, to Port Aran- sas, Texas; through West Indies to Argen- tina (Williams 1984, Rieger 1998). Paguristes hummi Wass, 1955 Known range.—Western Atlantic: New- port River, North Carolina, to off Sapelo Island, Georgia; southwestern Florida, in- cluding Tampa Bay, to off Isles Dernieres, Louisiana; Caribbean coast of Colombia (Williams 1984, Campos & Sanchez 1995, Strasser & Price 1999). Paguristes lymani A. Milne-Edwards & Bouvier, 1893 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; Flor- ida Keys to Swan Island, off Honduras; through West Indies to Guyana (Williams 1984). Paguristes moorei Benedict, 1901 Known range.—Western Atlantic: off Cape Lookout, North Carolina; Florida Straits; Puerto Rico (Williams 1984). Paguristes sericeus A. Milne-Edwards, 1880 Known range.—Western Atlantic: off Cape Lookout, North Carolina; West Flow- er Garden Bank, northwestern Gulf of Mex- ico, to Virgin Islands (Williams 1984). Paguristes spinipes A. Milne-Edwards, 1880 Known range.—Western Atlantic: Gulf Stream south of Cape Lookout, North Car- olina; off Cape Canaveral to Florida Straits; Sarasota, Florida; Barbados to Pernambuco and Alagoas, Brazil (Williams 1984, Rieger 1998). Paguristes tortugae Schmitt, 1933 Known range.—Western Atlantic: reefs off Beaufort, North Carolina, to southern and southwestern Florida; Gulf of Mexico; VOLUME 116, NUMBER 1 through West Indies to Sao Paulo, Brazil (Williams 1984, Rieger 1998). Paguristes triangulatus A. Milne-Edwards & Bouvier, 1893 Known range.—Western Atlantic: off Oregon Inlet, North Carolina, to Tortugas, Florida; Barbados: Trinidad (Williams 1984). Petrochirus diogenes (Linnaeus, 1758) Known range.—Western Atlantic: off Cape Lookout, North Carolina, to Uruguay, including Gulf of Mexico and West Indies (Williams 1984, Rieger 1998). Superfamily Paguroidea Latreille, 1802 Family Lithodidae Samouelle, 1819 Lithodes maja (Linnaeus, 1758) Known range.—Western Atlantic: West Greenland to Baltimore Canyon, off the coast of Maryland; North Atlantic: East Greenland; Iceland; Northeastern Atlantic: Spitsbergen to the British Isles; the Neth- erlands (Williams 1984, 1988; Squires 1990). Family Paguridae Latreille, 1802 > Anisopagurus hopkinsi Lemaitre & McLaughlin, 1996 Remarks.—Lemaitre & McLaughlin (1996:101) described and illustrated this species and discussed congeneric affinities. This species occurs at depths of 91-165 m (Lemaitre & McLaughlin 1996). Known range.—Western Atlantic: off Georgia; Gulf of Mexico (Lemaitre & McLaughlin 1996). > Catapagurus sharreri A. Milne-Edwards, 1880 Remarks.—Williams (1984) considered this species to be extralimital. Asakura (2001) rediagnosed this genus and species. This species occurs at 60-882 m depth (Williams & Wigley 1977) on sediments 119 consisting of sand, mud and shell fragments (Rieger 1998). Known range.—Western Atlantic: 40°N, southward to Rio de Janeiro, Brazil, includ- ing the Caribbean Sea (Williams & Wigley LOWE) * Goreopagurus piercei (Wass, 1963) Pagurus piercei Wass.—Williams, 1984: 218. Remarks.—McLaughlin (1988) exam- ined a large sample of P. piercei and ob- served several characters that excluded this species from the genus Pagurus. A new ge- nus (Goreopagurus) was erected and de- scribed (McLaughlin 1988) and G. piercei was rediagnosed and illustrated. Discovery of an additional species belonging to this genus required McLaughlin & Haig (1995) to emend and expand the original generic diagnosis. Known range.—Western Atlantic: Mid- Atlantic Bight, off eastern United States, to southeastern Florida; northern Gulf of Mex- ico (McLaughlin 1988). > * Hemipagurus gracilis Smith, 1881 Catapagurus gracilis (Smith).—Williams, 1984:484. Remarks.—Asakura (2001) removed Hemipagurus from the synonymy of Cata- pagurus, rediagnosed Hemipagurus, and re- described H. gracilis, the type species. Wil- liams (1984) considered this species to be extralimital. This species occurs at 73-418 m depth (Williams & Wigley 1977). Known range.—Western Atlantic: Mas- sachusetts to Barbados (Williams & Wigley D772): ~ lridopagurus reticulatus Garcia-Gomez, 1983 Remarks.—Garcia-Gomez (1983:37) de- scribed this species and provided a species diagnosis, color description, and behavioral observations. This species occurs at depths of 1—38 m (Garcia-Gomez 1983). 120 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Known range.—Western Atlantic: North Carolina to Florida; Bahamas; Jamaica; Do- minican Republic; U.S. Virgin Islands; Lee- ward Islands; Bonaire; Curacao; Belize; Colombia; Venezuela; Suriname; French Guiana (Garcia-Gomez 1983). * Manucomplanus ungulatus (Studer, 1883) Manucomplanus corallinus (Benedict).— Williams, 1984:224. Remarks.—Lemaitre & McLaughlin (1996) determined that the range of varia- tion of specimens of the nominal species M. corallinus and M. ungulatus was such that the two taxa could not be separated. Con- sequently, they placed M. corallinus in the synonymy of M. ungulatus. The authors provided a species diagnosis and illustra- tions. Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Florida; Gulf of Mexico; Straits of Florida; Baha- mas; Eastern Atlantic: South and West Af- rica (Williams 1984, Lemaitre & Mc- Laughlin 1996). Pagurus acadianus Benedict, 1901 Known range.—Western Atlantic: Straits of Belle Island and Notre Dame Bay, New- foundland; Gulf of St. Lawrence to mouth of Chesapeake Bay (Williams 1984, Squires 1990). Pagurus annulipes (Stimpson, 1860) Remarks.—Lemaitre et al. (1982) report- ed the depth distribution for this species as subtidal to 90 m, which represented an in- crease in the maximum reported depth of occurrence. Known range.—Western Atlantic: Vine- yard Sound, Massachusetts, to south central Florida (Williams 1984, Lemaitre et al. 1982). Pagurus arcuatus Squires, 1964 Known range.—Western Atlantic: Greenland to off Virginia Capes (Williams 1984). > Pagurus brevidactylus (Stimpson, 1859) Remarks.—Lemaitre et al. (1982) provid- ed a species diagnosis and commented on interspecific comparisons. This species oc- curs from the intertidal to 50 m (Lemaitre et al. 1982). Known range.—Western Atlantic: Ber- muda; northeastern Florida; Bahamas to Brazil; Gulf of Mexico; Caribbean (Lemai- tre et al. 1982). Pagurus carolinensis McLaughlin, 1975 Remarks.—Lemaitre et al. (1982) provid- ed species diagnosis and interspecific com- parisons. Known range.—Western Atlantic: North Carolina to southeastern Florida (Lemaitre et al. 1982); Tampa Bay, Florida (Strasser & Price 1999). Pagurus defensus (Benedict, 1892) Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Georgia; Dry Tortugas, Florida, to Alabama (Williams 1984). Pagurus impressus (Benedict, 1892) Known range.—Western Atlantic: off Di- amond Shoals, North Carolina, to near Cape Canaveral, Florida; Florida Bay (near Flamingo), north to Pensacola, Florida; Port Aransas, Texas (Williams 1984). Pagurus longicarpus Say, 1817 Known range.—Western Atlantic: Minas Basin and Chignecto Bay to Hutchinson Is- land, Florida; southwestern Florida to Texas (Williams 1984, Squires 1990). VOLUME 116, NUMBER 1 & Pagurus maclaughlinae Garcia-Gomez, 1982 Remarks.—Garcia-Gomez (1982:647) recognized and described this species. Le- maitre et al. (1982) reported this species in the region and provided a species diagnosis. Pagurus maclaughlinae occurs from the subtidal to 5 m (Garcia-Gomez 1982). Known range.—Western Atlantic: Was- saw, Georgia, to Florida; Gulf of Mexico; Caribbean (Lemaitre et al. 1982). Pagurus politus (Smith, 1882) Known range.—Western Atlantic: Nova Scotia to off Dry Tortugas, Florida (Wil- hams 1984, Squires 1990). Pagurus pollicaris Say, 1817 Known range.—Western Atlantic: Grand Manan, New Brunswick, to northeastern Florida; Key West, Florida, to Texas (Wil- liams 1984). Pagurus pubescens Kr@yer, 1838 Known range.—Western Atlantic: West Greenland; Foxe Basin; Hudson Bay to Cape Hatteras, North Carolina; Arctic- North Atlantic: East Greenland; Iceland; Spitsbergen; Barents Sea; Novaya Zemlya; Eastern Atlantic: Faroes and the British Isles (Williams 1984, Squires 1990). * Pagurus stimpsoni (A. Milne-Edwards & Bouvier, 1893) Pagurus hendersoni Wass.—Williams, 1984:214. Remarks.—Lemaitre et al. (1982), in comparing type material of P. hendersoni with specimens of P. stimpsoni, determined that the two species were conspecific, and placed P. hendersoni in the synonymy of P. stimpsoni. Pagurus stimpsoni occurs from the subtidal purportedly to 512 m (Wass 1963), usually from the subtidal to 73 m. Depths of occurrence reported for types of P. hendersoni (228 and 347-512 m) by WL Wass are problematic according to Lemaitre et al. (1982) who were unable to verify the accuracy of these collection data. Known range.—Western Atlantic: North Carolina to Florida; Gulf of Mexico; Carib- bean coast of South America (Lemaitre et al. 11982). Phimochirus holthuisi (Provenzano, 1961) Remarks.—McLaughlin (1981) reported the bathymetric range for this species as 1— 210 m. Known range.—Western Atlantic: North Carolina to Florida; Straits of Florida and Bahama Islands; Gulf of Mexico from Flor- ida to Texas; Caribbean and northern South America from Colombia to Bahia, Brazil (McLaughlin 1981, Williams 1984, Rieger 1998). Pylopagurus discoidalis (A. Milne-Edwards, 1880) Remarks.—McLaughlin & Lemaitre (2001b) rediagnosed the genus Pylopagurus and diagnosed, illustrated, and discussed morphological variation of all species with- in this genus. Known range.—Western Atlantic: North Carolina to central Brazil, including Gulf of Mexico and Caribbean (Williams 1984, McLaughlin & Lemaitre 2001b). Rhodochirus rosaceus (A. Milne-Edwards & Bouvier, 1893) Known range.—Western Atlantic: south of Cape Lookout, North Carolina, to Key West, Florida; northwestern Gulf of Mexi- co; Grenada; Suriname; Sao Paulo and Rio Grande do Sul, Brazil (McLaughlin 1981, Williams 1984, Rieger 1998). Tomopaguropsis problematica (A. Milne-Edwards & Bouvier, 1893) Known range.—Western Atlantic: north- east of Cape Lookout, North Carolina; southern Florida and Bahamas; Barbados; off Honduras (Williams 1984). 122 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON & Tomopagurus cokeri (Hay, 1917) Remarks.—McLaughlin (1981) rede- scribed the genus and provided a species diagnosis and illustrations. This species oc- curs at 44-302 m (McLaughlin 1981). Known range.—Western Atlantic: south of Cape Lookout, North Carolina; off northeastern Florida; Gulf of Mexico; Ca- ribbean; off French Guiana (McLaughlin 1981). > Tomopagurus wassi McLaughlin, 1981 Remarks.—McLaughlin (1981) rede- scribed the genus, described this species, and provided illustrations and comparative information. This species occurs at 75—360 m (McLaughlin 1981). Known range.—Western Atlantic: south- eastern United States; Straits of Florida; Gulf of Mexico; Caribbean to northern Bra- Zil (McLaughlin 1981). > Family Parapaguridae Smith, 1882 & Parapagurus pilosimanus Smith, 1879 Remarks.—Lemaitre (1989) provided a description, illustrations, and information on symbiotic associations and affinities. This species occurs at 102—3864 m, but is most frequently found in depths of 400— 1400 m (Lemaitre 1989). Known range.—North Atlantic: south- west of Iceland and the Faroe Islands to west of Ireland; Western Atlantic: Nova Scotia to Guyana; Eastern Atlantic: Bay of Biscay to Gulf of Guinea; South Atlantic: Tristan da Cunha (Lemaitre 1989). > Sympagurus pictus Smith, 1883 Remarks.—Lemaitre (1989) provided a description, illustrations, and information on symbiotic associations and affinities. This species occurs at 180—2322 m, but is most frequently found at depths of 200 to 800 m (Lemaitre 1989). Known range.—Western Atlantic: off Long Island, New York to off French Gui- ana (Lemaitre 1989). Infraorder Brachyura Latreille, 1802 Superfamily Dromioidea De Haan, 1833 Family Dromiidae De Hann, 1833 * Cryptodromiopsis antillensis (Stimpson, 1858) Dromidia antillensis Stimpson.—Williams, 1984:255. Remarks.—McLay (1993) redefined the genera within the family Dromiidae. As a result, Dromidia antillensis was reassigned to the genus Cryptodromiopsis. Known range.—Western Atlantic: Ber- muda; off Cape Hatteras, North Carolina, to Rio Grande do Sul, Brazil, including Gulf of Mexico and Caribbean Sea; Central At- lantic: Saint Helena (Williams 1984, Melo et al. 1998). > Dromia erythropus (George Edwards, 1771) Remarks.—Previously, the geographic range of this species was considered to be outside the region, however, Williams (un- published data) records this species from off North Carolina. Dromia erythropus in- habits various types of sediments from the intertidal to 360 m (Laughlin et al. 1982, Melo et al. 1998) and is usually found with sponges and ascidians on its carapace (Melo et al. 1998). Known range.—Western Atlantic: Ber- muda; Florida to Sao Paulo, Brazil (Laugh- lin et al. 1982, Melo et al. 1998). Williams (unpublished) noted a specimen ““Taken off North Carolina (33°48'06"N, 76°34’'24”"W), 105 m, 14 May 1981 (USNM 202800).”’ Hypoconcha arcuata Stimpson, 1858 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to west Florida; St. Thomas, U.S. Virgin Islands; Suriname to Sao Paulo, Brazil (Williams 1984, Melo et al. 1998). * Hypoconcha parasitica (Linnaeus, 1763) Hypoconcha sabulosa (Herbst).—Williams, 1984:258. VOLUME 116, NUMBER 1 Remarks.—Holthuis and Manning (1987) concluded that Hypoconcha parasitica was the oldest available name for the species formerly known as H. sabulosa. Known range.—Western Atlantic: off Cape Hatteras, North Carolina, through Gulf of Mexico to Sao Paulo, Brazil (Wil- liams 1984, Melo et al. 1998). Hypoconcha spinosissima Rathbun, 1933 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Gulf of Mexico, off Mississippi River delta; Yuca- tan; Jamaica (Williams 1984). Superfamily Homoloidea De Haan, 1839 Family Homolidae De Haan, 1839 > Homola minima Guinot & Richer de Forges, 1995 Homola barbata (Fabricius).—Williams, 1984:261 (in part). Remarks.—Guinot & Richer de Forges (1995:326) described this species, provided photographs, and discussed congeneric comparisons. Homola minima occurs at depths of 55—690 m (Guinot & Richer de Forges 1995, Martin & Zimmerman 2001). Homola barbata, formerly considered to be a wide-ranging species (Williams 1984), is now known to occur only in the Mediter- ranean Sea and also possibly in the adjoin- ing eastern Atlantic (Guinot & Richer de Forges 1995). Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts, to Rio Grande do Sul, Brazil; Bahamas (Guinot & Richer de Forges 1995, Martin & Zimmer- man 2001). Family Latreilliidae Stimpson, 1858 Latreillia manningi Williams, 1982 Known range.—Western Atlantic: Nan- tucket Shoals, off Massachusetts, to off Ha- vana, Cuba; Venezuela; Central Atlantic: Ascension Island (Williams 1984). 123 Superfamily Raninoidea De Haan, 1841 Family Raninidae De Haan, 1841 Remarks.—Guinot (1993) subdivided the family Raninidae into six subfamilies: Ran- ininae De Haan, 1841, Notopodinae Seréne & Umali, 1972, Symethinae Goeke, 1981, Raninoidinae Loérenthey & Beurlen, 1929, Lyreidinae Guinot, 1993, and Cyrtorhinae Guinot, 1993. Three subfamilies (Ranini- nae, Notopodinae, and Symethinae) were recognized in Williams (1984). * Subfamily Lyreidinae Guinot, 1993 > Lysirude nitidus (A. Milne-Edwards, 1880) Remarks.—Diagnostic characteristics and congeneric comparisons are provided in Goeke (1980, 1985) and Tucker (1998). This species is found on soft mud bottoms at depths of 119-823 m (Powers 1977, as Lyreidus bairdii). Known range.—Western Atlantic: Mas- sachusetts; Gulf of Mexico; Greater Antil- les to Venezuela; Suriname (Goeke 1980). * Subfamily Notopodinae Serene & Umali, 1972 Remarks.—Placement of Ranilia in sub- family Notopodinae was supported by Tucker’s (1998) phylogenetic study. Ranilia constricta (A. Milne-Edwards, 1880) Known range.—Western Atlantic: south- east of Cape Fear, North Carolina; Palm Beach, Florida, to Florida Straits and Yu- catan Channel; Cuba; off Barbados (Wil- liams 1984); Amapa, Rio de Janeiro to Rio Grande do Sul, Brazil (Melo et al. 1998): Central Atlantic: Ascension Island; Eastern Atlantic: Senegal to Congo; Annobon Is- land (Williams 1984). Ranilia muricata H. Milne Edwards, 1837 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to north- 124 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON western Gulf of Mexico; Colombia (Wil- liams 1984); Pernambuco, Brazil (Melo et al. 1998). * Subfamily Raninoidinae Loérenthey & Beurlen, 1929 Remarks.—Placement of Raninoides in subfamily Raninoidinae was supported by Tucker’s (1998) phylogenetic study. Raninoides loevis (Latreille, 1825) Remarks.—This species was listed under subfamily Ranininae in Williams (1984). Known range.—Western Atlantic: south of Cape Hatteras, North Carolina, to Sao Paulo, Brazil, including Gulf of Mexico, southern Caribbean Sea and Leeward Is- lands (Williams 1984, Melo et al. 1998). * Family Symethidae Goeke, 1981 Remarks.—This taxon was_ previously considered a subfamily of the Raninidae (e.g., Williams 1984, Guinot 1993). Tucker (1998) considered the subfamily Symethi- nae sufficiently distinct to warrant elevation of the subfamily to the rank of family. Symethis variolosa (Fabricius, 1793) Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina, through western Gulf of Mexico to Sao Paulo, Brazil, and Fernando de Noronha (Williams 1984, Melo et al. 1998). * Superfamily Cyclodorippoidea Ortmann, 1892 Remarks.—Tavares (1991a) proposed the use of the superfamilial name Cyclodorip- poidea to replace Tymoloidea. Martin & Davis (2001) accepted this arrangement. * Family Cyclodorippidae Ortmann, 1892 Family Tymolidae.—Williams, 1984:259. Remarks.—Family Tymolidae is placed in the synonymy of Cyclodorippidae (Ta- vares 1991a, 1993, 1996). Tavares (1996) conducted a revision of the family Cyclo- dorippidae. Clythrocerus granulatus (Rathbun, 1898) Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; Honduras; southern Florida, through Antil- les, to Venezuela and Trinidad (Williams 1984), and southward to Rio Grande do Sul, Brazil (Melo et al. 1998). © Clythrocerus nitidus (A. Milne-Edwards, 1880) Remarks.—Tavares (1996) included a de- scription and illustrations, and discussed morphological variation for this species. This species occurs at 12—531 m (Tavares 1996). Known range.—Western Atlantic: South Carolina; Florida; Barbados; Grenada (Ta- vares 1996). * Deilocerus perpusillus (Rathbun, 1901) Clythrocerus perpusillus Rathbun.—Wil- liams, 1984:260. Remarks.—As a result of his review of the New World cyclodorippoid crabs, Ta- vares (1993) identified and described four new genera, including Deilocerus. Clythro- cerus perpusillus was designated the type species of this new genus. Known range.—Western Atlantic: North Carolina to Georgia; Bahama Banks; Gulf of Mexico; Puerto Rico; Barbados; Amapa to Rio Grande do Sul, Brazil (Williams 1984, Tavares 1996, Melo et al. 1998). Superfamily Dorippoidea MacLeay, 1838 Family Dorippidae MacLeay, 1838 Subfamily Ethusinae Guinot, 1977 * Ethusa americana A. Milne-Edwards, 1880 Ethusa mascarone americana A. Milne-Ed- wards.—Williams, 1984:269. Remarks.—Hendrickx (1989) elevated VOLUME 116, NUMBER 1 the subspecies E. m. americana to full spe- cies status. Ethusa americana occurs only in the western Atlantic; references of this species at eastern Pacific localities refer to E. panamensis Finnegan. Ethusa mascaro- ne (Herbst) occurs in the Mediterranean (Manning & Holthuis 1981). Known range.—Western Atlantic: south of Cape Lookout, North Carolina, to Gulf of Mexico and West Indies; Maranhao to Rio de Janeiro, Brazil (Williams 1984, Melo et al. 1998). Ethusa microphthalma Smith, 1881 Known range.—Western Atlantic: off Martha’s Vineyard, Massachusetts, to Cuba; throughout Gulf of Mexico (Williams 1984); Sao Paulo, Brazil (Melo et al. 1998). Ethusa tenuipes Rathbun, 1897 Known range.—Western Atlantic: off Cape Lookout, North Carolina; East Florida to Gulf of Mexico, east of Mississippi River delta; Cuba (Williams 1984); Rio de Janeiro to Sao Paulo, Brazil (Melo et al. 1998). Superfamily Calappoidea H. Milne Edwards, 1837 Family Calappidae H. Milne Edwards, 1837 Remarks.—Bellwood (1996) evaluated phylogenetic relationships of four subfam- ilies within the Calappidae (Calappinae, Matutinae, Orithyiinae, and Hepatinae) us- ing cladistic analysis. She rejected the monophyly of an expanded Calappidae, but demonstrated support for the monophyly of each component taxon and proposed ele- vating the four subfamilies to family status. Additionally, Bellwood (1996) reassigned these families to different superfamilies, with Calappidae and Hepatidae remaining in the superfamily Calappoidea, Matutidae placed in the superfamily Leucosioidea, and Orithyiidae placed in the superfamily Do- rippoidea. Based on fossil evidence (cara- pace morphology), Schweitzer and Feld- 125 mann (2000) supported the conclusions of Bellwood (1996). Acanthocarpus alexandri Stimpson, 1871 Known range.—Western Atlantic: Georges Bank, off Massachusetts, to west coast of Florida; Puerto Rico to Grenadines: Rio de Janeiro (Williams 1984) to Rio Grande do Sul, Brazil (Melo et al. 1998). Calappa flammea (Herbst, 1794) Known range.—Western Atlantic: Woods Hole region, Massachusetts, to Florida Keys; Gulf coast of United States and Mex- ico; Bahamas; Bermuda (Williams 1984). Calappa ocellata Holthuis, 1958 Known range.—Western Atlantic: Ber- muda; Cape Hatteras, North Carolina, to Rio de Janeiro, Brazil (Williams 1984). Calappa sulcata Rathbun, 1898 Remarks.—This species occurs on sand, mud, and calcareous algal bottoms from shallow depths to 200 m (Melo et al. 1998). Known range.—Western Atlantic: Cape Hatteras, North Carolina, through Gulf of Mexico to Parana, Brazil (Williams 1984, Melo et al. 1998). * Calappa tortugae Rathbun, 1933 Calappa angusta A. Milne-Edwards.—Wil- liams, 1984:273 (in part). Remarks.—Williams & Child (1989) de- termined that ““Calappa angusta,”’ as pre- viously understood, was poorly defined and actually comprised a complex of species. The next available name for species of Ca- lappa in the western Atlantic, Calappa saussurel tortugae Rathbun, was removed from synonymy and elevated to full species rank (Williams & Child 1989). Species di- agnosis, illustrations and measurements were provided in Williams & Child (1989). This species occurs at 13—238 m (Williams & Child 1989). 126 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Known range.—Western Atlantic: North Carolina to off Venezuela, including Gulf of Mexico, Caribbean, and Leeward Islands (Williams & Child 1989). * Cryptosoma balguerii (Desbonne, 1867) Cycloes bairdii (Stimpson).—Williams, 1984:278. Remarks.—Galhil & Clark (1996) report- ed that populations of C. bairdii occurred exclusively in the eastern Pacific (Baja Cal- ifornia to Costa Rica); previous records of C. bairdii from Costa Rica to Ecuador were actually those of a new Cryptosoma spe- cies, whereas records reported as C. bairdii from Atlantic localities pertained to C. bal- guerii (a nominal species previously con- sidered to be a subjective synonym of Cy- cloes bairdii). Cryptosoma balguerii occurs in shallow waters to 230 m (Galil & Clark 1996). Known range.—Western Atlantic: North Carolina and Bermuda to Espirito Santo, Brazil (Galil & Clark 1996). > * Cyclozodion angustum (A. Milne-Edwards, 1880) Calappa angusta A. Milne-Edwards.—Wil- liams, 1984:273 (in part). Remarks.—Williams & Child (1989) concluded that Calappa angusta, as de- scribed by A. Milne-Edwards, was generi- cally misplaced. They described the genus Cyclozodion, of which C. angustum (A. Milne-Edwards) is the type species, and provided illustrations, measurements, and comparative information. This species oc- curs at 95—421 m (Williams & Child 1989). Known range.—Western Atlantic: off Cape Canaveral, Florida, to Isla Providen- cia, Colombia; Guyana (Williams & Child 1989). © Cyclozodion tuberatum Williams & Child, 1989 Remarks.—Williams & Child (1989:112) described this species and provided illustra- tions, measurements and color description. This species occurs at 31—188 m depth, rarely to 640 m (Williams & Child 1989). Known range.—Western Atlantic: off Cape Lookout, North Carolina, through Ba- hamas; eastern Gulf of Mexico; and Suri- name (Williams & Child 1989). * Family Hepatidae Stimpson, 1871 Remarks.—Species in the genera Hepa- tus and Osachila were previously assigned to the subfamily Matutinae (Williams 1984). Based on results of a phylogenetic study by Bellwood (1996) and corroborat- ing fossil evidence (carapace morphology) of Schweitzer & Feldmann (2000), these genera are now placed in the family He- patidae. Hepatus epheliticus (Linnaeus, 1763) Known range.—Western Atlantic: Ches- apeake Bay to western Gulf of Campeche, Mexico; Cuba; Jamaica; Dominican Repub- lic (Williams 1984). Hepatus pudibundus (Herbst, 1785) Known range.—Western Atlantic: Geor- gia to Rio Grande do Sul, Brazil (Williams 1984, Melo et al. 1998). Osachila semilevis Rathbun, 1916 Known range.—Western Atlantic: off Beaufort, North Carolina, to northwest Florida (Williams 1984). Osachila tuberosa Stimpson, 1871 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to northwest Florida and Yucatan Channel (Williams 1984); Rio de Janeiro and Rio Grande do Sul, Brazil (Melo et al. 1998). Superfamily Leucosioidea Samouelle, resi Family Leucosiidae Samouelle, 1819 Remarks.—Leucosiidae was previously assigned to the superfamily Calappoidea VOLUME 116, NUMBER 1 (Williams 1984). Based on results of a phy- logenetic study of the Calappidae, Bell- wood (1996) placed the family Leucosiidae in the superfamily Leucosioidea. Subfamily Ebaliinae Stimpson, 1871 Ebalia cariosa (Stimpson, 1860) Known range.—Western Atlantic: Bogue Sound near Beaufort, North Carolina, to west Florida; western Gulf of Mexico; Ja- maica; northeastern South America to Sao Paulo, Brazil (Williams 1984). Ebalia stimpsonii A. Milne-Edwards, 1880 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; west Florida to Barbados (Williams 1984); Ama- pa to Sao Paulo, Brazil (Melo et al. 1998). Speloeophorus nodosus (Bell, 1855) Known range.—Western Atlantic: Flori- da; West Indies (Williams 1984); Maranhao to Rio de Janeiro, Brazil (Melo et al. 1998). Speloeophorus pontifer (Stimpson, 1871) Known range.—Western Atlantic: south- east of Cape Lookout and off Beaufort, North Carolina, to west Florida; West In- dies to Barbados (Williams 1984). Subfamily [liinae Stimpson, 1871 Myropsis quinquespinosa Stimpson, 1871 Known range.—Western Atlantic: south of Martha’s Vineyard, Massachusetts, to Suriname, including Gulf of Mexico and Caribbean Sea (Williams 1984). Persephona mediterranea (Herbst, 1794) Known range.—Western Atlantic: New Jersey to Rio Grande do Sul, Brazil, in- _ cluding Gulf of Mexico and Caribbean Sea (Williams 1984, Melo et al. 1998). Subfamily Leucosiinae Samouelle, 1819 * Acanthilia intermedia (Miers, 1886) Iliacantha intermedia Miers.—Williams, 1984:290. 127 Remarks.—Galil (2000) removed Jf. in- termedia from the genus J/liacantha and placed it in the newly erected genus Acan- thilia Galil; generic description and species redescription are provided. This species oc- curs at depths of 10—329 m. Known range.—Western Atlantic: North Carolina to Brazil (Galil 2000). Callidactylus asper Stimpson, 1871 Known range.—Western Atlantic: south of Cape Lookout, North Carolina, through southeastern Gulf of Mexico to Panama, and southeastward to Alagoas, Brazil (Wil- liams 1984). Iliacantha subglobosa Stimpson, 1871 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to northwest Florida; through eastern Gulf of Mexico and Caribbean Sea, south to Alagoas, Brazil (Williams 1984). Superfamily Majoidea Samouelle, 1819 Remarks.—Recent investigations of in- trarelationships within the Majoidea include those of Drach & Guinot (1983), Griffin & Tranter (1986), Clark & Webber (1991), Guinot & Richer de Forges (1997), Guinot & Bouchard (1998), and Pohle & Marques (2000). Although all subfamilies within the superfamily have not been considered in a phylogenetic framework, previously recog- nized subfamilies within the family Majidae have been elevated to the level of family (Hendrickx 1995). Martin & Davis (2001) also adopted this arrangement. However, re- sults of a recent phylogenetic analysis based on larval characters (Pohle & Marques 2000) supported monophyly only of the Oregoniidae, Majidae and Inachidae. * Family Epialtidae MacLeay, 1838 Remarks.—AIll species listed below were previously placed in subfamily Epialtinae of the Majidae in Williams (1984). 128 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Epialtus dilatatus A. Milne-Edwards, 1878 Known range.—Western Atlantic: off Beaufort Inlet and New River, North Car- olina; southwest Florida; Yucatan; Bahamas to St. Thomas (Williams 1984). Sphenocarcinus corrosus A. Milne-Edwards, 1875 Remarks.—Tavares (1991b) provided a generic revision and diagnosis. Known range.—Western Atlantic: off Cape Lookout, North Carolina; Gulf of Mexico to Barbados (Williams 1984). * Family Inachidae MacLeay, 1838 Remarks.—Elevation of subfamily Ina- chinae to family level was supported by the cladistic analysis of Clark & Webber (1991). Other morphological evidence (Guinot & Bouchard 1998) supported this conclusion. All species of the genera An- omalothir, Metoporhaphis, Podochela, and Stenorhynchus (listed below) were previ- ously placed in subfamily Inachinae of the Mayidae in Williams (1984). The genus Ro- chinia was previously placed in the subfam- ily Pisinae of the Majidae (Williams 1984, Griffin & Tranter 1986). Cladistic relation- ships hypothesized by Clark & Webber (1991) indicated that Rochinia should be placed in the Inachidae. Tavares (1991b) conducted a generic revision of Rochinia. Anomalothir furcillatus (Stimpson, 1871) Known range.—Western Atlantic: off Cape Lookout, North Carolina, through eastern Gulf of Mexico; West Indies to Gre- nada (Williams 1984). Metoporhaphis calcarata (Say, 1818) Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Rio de Janeiro, Brazil, including Gulf of Mexico and Caribbean Sea (Williams 1984, Melo 1998). Podochela gracilipes Stimpson, 1871 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to Rio Grande do Sul, including Gulf of Mexico and Caribbean Sea (Williams 1984, Melo 1998). Podochela riisei Stimpson, 1860 Known range.—Western Atlantic: Ber- muda; North Carolina to Campeche, Mex- ico; Antilles; Rio de Janeiro, Brazil (Wil- liams 1984, Melo 1998). Podochela sidneyi Rathbun, 1924 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Veracruz, Mexico; northwestern Cuba; Yucatan Chan- nel (Williams 1984). Rochinia crassa (A. Milne-Edwards, 1879) Known range.—Western Atlantic: Nan- tucket Shoals, Massachusetts, to Gulf of Mexico, off southern Texas; northern Cuba; west of Cabo de la Vela, Colombia; off French Guiana (Williams 1984). Recently, a male measuring 89.2 mm carapace width and 89.5 mm carapace length was caught off Lunenburg, Nova Scotia, at 243 m (Mo- riyasu et al. 2001). This first recorded oc- currence of this species in Canadian waters constitutes a northern range extension from the previous northernmost occurrence on Nantucket Shoals, Massachusetts (Williams 1984). Moriyasu et al. (2001) considered this as a stray occurrence of this species in Canadian waters. Rochinia tanneri (Smith, 1883) Known range.—Western Atlantic: off Martha’s Vineyard, Massachusetts, to Straits of Florida (Williams 1984). Rochinia umbonata (Stimpson, 1871) Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina, VOLUME 116. NUMBER 1 through eastern and northern Gulf of Mex- ico to northeast of Nicaragua; through West Indies to St. Vincent (Williams 1984). Stenorhynchus seticornis (Herbst, 1788) Remarks.—Goeke (1989) determined that two co-occurring species were con- fused under the name S. seticornis. He re- stricted the specific description of S. seti- cornis, redescribed the species, and selected a neotype. This species occurs at 1-366 m (Goeke 1989). Known range.—Western Atlantic: Ber- muda; Cape Fear, North Carolina, to Rio Grande do Sul, Brazil, including Gulf of Mexico, Antilles, and northern South America (Williams 1984, Goeke 1989, Melo 1998). & Stenorhynchus yangi Goeke, 1989 Remarks.—Goeke (1989:631) described this species and provided a diagnosis, illus- trations, color description, and discussed morphological variation. This species oc- curs at 31—365 m (Goeke 1989). Known range.—Western Atlantic: off Martha’s Vineyard, Massachusetts, to Suri- name, including Gulf of Mexico (Goeke 1989). %* Family Inachoididae Dana, 1851 Remarks.—Drach & Guinot (1983) pro- posed elevating the Inachoidinae to family level. This decision was corroborated with morphological evidence (Guinot & Richer de Forges 1997). The genera included with- in this family are based on the recommen- dation of Guinot & Richer de Forges (1997). All species listed below were pre- viously placed in the subfamily Inachinae of the Majidae in Williams (1984). Aepinus septemspinosus (A. Milne-Edwards, 1879) Known range.—Western Atlantic: south of Cape Lookout, North Carolina; south- west of Cape San Blas, Florida; Bahama 129 Banks to Sao Paulo, Brazil, and Fernando de Noronha Archipelago and Rocas Atoll, Brazil (Williams 1984, Melo 1998). Anasimus latus Rathbun, 1894 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to Amapa, Brazil, including Gulf of Mexico and An- tilles (Williams 1984, Melo 1998). Arachnopsis filipes Stimpson, 1871 Known range.—Western Atlantic: south- east of Cape Hatteras, North Carolina; Gulf of Mexico, off northwest Florida, through Antilles to Rio Grande do Norte, Brazil (Williams 1984, Melo 1998). Batrachonotus fragosus Stimpson, 1871 Known range.—Western Atlantic: Cape Hatteras, North Carolina, to southern and western Florida; West Indies to Barbados (Williams 1984). Collodes robustus Smith, 1883 Known range.—Western Atlantic: Cape Cod, Massachusetts, to southeast of Cape Lookout, North Carolina (Williams 1984). Collodes trispinosus Stimpson, 1871 Known range.—Western Atlantic: near Cape Hatteras, North Carolina, to southern and western Florida (Williams 1984); Ama- pa, Rio de Janeiro, and Sao Paulo, Brazil (Melo 1998). Euprognatha rastellifera Stimpson, 1871 Known range.—Western Atlantic: off Georges Bank to Uruguay, including Antil- les (Williams 1984, Melo 1998 as E. acuta). Inachoides forceps A. Milne-Edwards, 1879 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; west coast of Florida to Rio de Janeiro, Brazil, 130 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON including Antilles; the Guianas (Williams 1984, Melo 1998). ® Pyromaia arachna Rathbun, 1924 Remarks.—Lemaitre et al. (2001) provid- ed information on distinguishing character- istics and congeneric comparisons. This species occurs on mud, mud-sand, and mud-shell sediments at 183-324 m depth (Powers 1977). Known range.—Western Atlantic: South Carolina to Gulf of Mexico (Powers 1977, Lemaitre et al. 2001). Pyromaia cuspidata Stimpson, 1871 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to west Florida; Cuba; Yucatan Channel to Nicara- gua (Williams 1984). %* Family Mithracidae Balss, 1929 Remarks.—All species of Hemus, Ma- crocoeloma, Microphrys, Mithrax, Mithra- culus, and Stenocionops (listed below) were previously placed in the subfamily Mithra- cinae of the Mayidae in Williams (1984). Based on results of his investigation of western Atlantic Mithrax crabs, Wagner (1990) considered the morphological evi- dence sufficient to warrant recognition of distinct genera instead of two subgenera within Mithrax; Mithraculus was elevated to full generic status. Hemus cristulipes A. Milne-Edwards, 1875 Known range.—Western Atlantic: off Cape Lookout, North Carolina; South Car- olina; northwest Gulf of Mexico and Yu- catan, through West Indies to Rio de Janei- ro, Brazil, and Fernando de Noronha Ar- chipelago (Williams 1984, Melo 1998). Macrocoeloma camptocerum (Stimpson, 1871) Known range.—Western Atlantic: Beau- fort Harbor, North Carolina; around south- ern Florida to Alligator Harbor, Florida (Williams 1984); Amapa to Maranhao, Bra- Zil (Melo 1998). Macrocoeloma eutheca (Stimpson, 1871) Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; off northwest Florida through Bahama Banks and West Indies; Central America; Maran- hao to Espirito Santo, Brazil (Williams 1984, Melo 1998). Macrocoeloma trispinosum (Latreille, 1825) Macrocoeloma trispinosum, variety of Rathbun, 1925 Macrocoeloma trispinosum nodipes (Desbonne, 1867) Remarks.—Williams (1984:328) listed these nominal taxa together in one species account. No further investigation of these taxa has been conducted, and further revi- sion is needed. Until further results are available, these taxa are considered as they were in Williams (1984). Known range (for M. trispinosum (La- treille, 1825).—Western Atlantic: Beaufort, North Carolina, to Alligator Harbor, Flori- da; Yucatan; West Indies to Sao Paulo, Bra- zil, and Fernando de Noronha Archipelago (Williams 1984, Melo 1998). Microphrys antillensis Rathbun, 1920 Known range.—Western Atlantic: Cape Hatteras to Cape Fear, North Carolina; An- tilles; Paraiba to Rio de Janeiro, Brazil (Williams 1984, Melo 1998). Microphrys bicornutus (Latreille, 1825) Known range.—Western Atlantic: Ber- muda; near Beaufort, North Carolina, through Gulf of Mexico to Rio Grande do Sul, Brazil; Fernando de Noronha Archi- pelago, Brazil (Williams 1984, Melo 1998). VOLUME 116, NUMBER 1 * Mithraculus forceps (A. Milne-Edwards, 1875) Mithrax (Mithraculus) forceps (Milne Ed- wards).—Williams, 1984:337. Known range.—Western Atlantic: Ber- muda; Cape Hatteras, North Carolina, through Gulf of Mexico and Antilles to Sao Paulo, Brazil: Fernando de Noronha Archi- pelago and Rocas Atoll, Brazil (Williams 1984, Wagner 1990, Melo 1998). * Mithrax cornutus de Saussure, 1857 Mithrax (Mithrax) acuticornis Stimpson.— Williams, 1984:332. Remarks.—Wagner (1990) determined that M. cornutus and M. acuticornis were conspecific. Mithrax cornutus was original- ly regarded as extralimital by Williams (1984:484). This species occurs between 20—458 m (Wagner 1990). Known range.—Western Atlantic: Ber- muda; off Cape Lookout, North Carolina; east and west coasts of Florida, through Gulf of Mexico, Yucatan Channel, and West Indies to Rio de Janeiro, Brazil (Wil- lams 1984, Wagner 1990, Melo 1998). * Mithrax hispidus (Herbst, 1790) Mithrax (Mithrax) hispidus (Herbst).—Wil- liams, 1984:333. Mithrax (Mithrax) pleuracanthus Stimp- son.—Williams, 1984:334. Remarks.—Wagner (1990) placed M. pleuracanthus in the synonymy of M. his- pidus. Known range.—Western Atlantic: Dela- ware Bay; Bermuda; Beaufort, North Car- olina; Charleston, South Carolina; Georgia to Pensacola, Florida: northwestern Gulf of Mexico to Yucatan Channel: Bahamas and Florida Keys through West Indies; Colom- bia to Sao Paulo, Brazil (Williams 1984, Wagner 1990, Melo 1998). * Mithrax spinosissimus (Lamarck, 1818) Mithrax (Mithrax) spinosissimus (La- marck).—Williams, 1984:335. 131 Known range.—Western Atlantic: North (?) and South Carolina to Nicaragua, through West Indies to Barbados; Venezue- la (Williams 1984, Wagner 1990). *% Mithrax verrucosus H. Milne Edwards, 1832 Mithrax (Mithrax) verrucosus H. Milne Ed- wards.—Williams, 1984:336. Known range.—Western Atlantic: Charleston, South Carolina; Florida; Cam- peche Banks; Curacao; Venezuela; West In- dies to Fernando de Noronha Archipelago and Rocas Atoll, Brazil (Williams 1984, Wagner 1990, Melo 1998). Stenocionops furcata coelata (A. Milne-Edwards, 1878) Known range.—Western Atlantic: Beau- fort, North Carolina, to northwest Florida and Alabama; Yucatan Channel; West In- dies to Barbados (Williams 1984). P Stenocionops furcata furcata (Olivier, 1791) Remarks.—This subspecies was only ref- erenced in the remarks section of S. f coe- lata in Williams (1984). It is unclear why S. f- furcata was not given a full account since its geographic range and depth distri- bution (shallow water to 64 m) were clearly within the limits of coverage. Much con- fusion surrounds identification of members of this species group, and more research is needed (D. Felder, pers. comm.). Until such time, both subspecies are considered valid as in Williams (1984). Known range.—Western Atlantic: Geor- gia to Rio Grande do Sul, Brazil, including Antilles and Colombia (Williams 1984, Melo 1998). Stenocionops spinimana (Rathbun, 1892) Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Florida Straits; Gulf of Mexico, off Mobile Bay, 132 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Alabama, and east of Chandeleur Island, off Mississippi (Williams 1984); Sao Paulo, Brazil (Melo 1998). » Stenocionops spinosissima (de Saussure, 1857) Remarks.—This species was considered extralimital by Williams (unpublished), but reported occurrence off North Carolina (Powers 1977) indicated that this species should be included. This species occurs at depths of 46—480 m on mud and sand sed- iments (Powers 1977, Melo 1998), with its center of distribution in the Gulf of Mexico at 110-183 m (Powers 1977). Known range.—Western Atlantic: off North Carolina; south and southwest Flor- ida; off Texas and east coast of Mexico; north coast of Cuba; Haiti; Guadeloupe; Dominica; Rio Grande do Norte to Rio Grande do Sul, Brazil; Fernando de Noro- nha Archipelago, Brazil (Powers 1977, Melo 1998). * Family Oregoniidae Garth, 1958 Remarks.—Family level status was sup- ported by a cladistic analysis (Clark & Webber 1991) and morphological evidence (Guinot & Bouchard 1998). In their classi- fication, Martin & Davis (2001) recognized this taxon at the subfamily level. All species listed below were previously placed in the subfamily Oregoniinae of the Majidae in Williams (1984). Chionoecetes opilio opilio (O. Fabricius, 1788) Known range.—Western Atlantic: Green- land south to St. Lawrence estuary and Gulf of Maine; Arctic-North Pacific: Point Barrow, Alaska, and northeastern Siberia through Ber- ing Strait to Alaskan Peninsula and Aleutian chain; Eastern Pacific: Kamchatka; Okhotsk Sea southward to Japan (Williams 1984). Hyas araneus (Linnaeus, 1758) Known range.—Western Atlantic: West Greenland to Rhode Island; Arctic-North Atlantic: between Greenland and Iceland; Spitsbergen; Kara Sea; Eastern Atlantic: through British Isles and northwest France (Williams 1984, Squires 1990). Hyas coarctatus coarctatus Leach, 1815 Known range.—Western Atlantic: Hud- son Bay and Greenland to North Carolina; Arctic-Eastern Atlantic: Murman Sea to Iceland and the British Isles (Williams 1984, Squires 1990). * Family Pisidae Dana, 1851 Remarks.—The following six species (listed below) were previously placed in the subfamily Pisinae of the Mayjidae in Wil- liams (1984). Coelocerus spinosus A. Milne-Edwards, 1875 Known range.—Western Atlantic: off Cape Fear, North Carolina, to Cape Can- averal, Florida; west Florida to east of Mis- sissippi River delta (Williams 1984). Libinia dubia H. Milne Edwards, 1834 Known range.—Western Atlantic: Cape Cod, Massachusetts, to southern Texas; Ba- hamas; Cuba (Williams 1984). Libinia emarginata Leach, 1815 Known range.—Western Atlantic: Prince Edward Island and Nova Scotia to western Gulf of Mexico (Williams 1984, Squires 1990). Nibilia antilocapra (Stimpson, 1871) Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Gulf of Mexico just east of Mississippi River delta; Gulf of Campeche; Antilles; off Guyana; Rio Grande do Norte and Rio Grande do Sul, Brazil (Williams 1984, Melo 1998). VOLUME 116, NUMBER 1 Pelia mutica (Gibbes, 1850) Known range.—Western Atlantic: Buz- zards Bay and Vineyard Sound, Massachu- setts, to Texas; Cuba; Puerto Rico; St. Thomas, U.S. Virgin Islands (Williams 1984). * Family Tychidae Dana, 1851 Remarks.—AIll species listed below were previously placed in the subfamily Tychi- nae of the Majidae in Williams (1984). Pitho lherminieri (Schramm, 1867) Known range.—Western Atlantic: off Beaufort Inlet, North Carolina, to west Florida; Veracruz, Mexico; West Indies to Sao Paulo, Brazil (Williams 1984). Tyche emarginata White, 1847 Known range.—Western Atlantic: off Beaufort Inlet, North Carolina, through Ba- hamas to west coast of Florida (Williams 1984); Antilles; Rio Grande do Norte, Bra- zil (Melo 1998). Superfamily Parthenopoidea MacLeay, 1838 Family Parthenopidae MacLeay, 1838 Subfamily Parthenopinae MacLeay, 1838 Remarks.—Ng & Rodriguez (1986) did not recognize subgenera within the genus Parthenope; Platylambrus was assigned full generic status. * Celatopesia concava (Stimpson, 1871) Cryptopodia concava Stimpson.—Wil- liams, 1984:346. Remarks.—Chiong & Ng (1998) deter- mined that the American species of the ge- nus Cryptopodia, including C. concava, dif- fered markedly in carapace appearance from that of the Indo-West Pacific species. Celatopesia was described, and the Amer- ican species formerly included in Crypto- podia were referred to the new genus 135 (Chiong & Ng 1998). Generic comparisons, species redescription and illustrations were also provided (Chiong & Ng 1998). Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; east central Florida; south of Cape San Blas, Florida, to St. Thomas, U.S. Virgin Islands; through Antilles; Maranhao to Rio de Ja- neiro, Brazil (Williams 1984, Melo 1998). Heterocrypta granulata (Gibbes, 1850) Known range.—Western Atlantic: from Nantucket Sound, Massachusetts, around peninsular Florida to southern Texas; through West Indies to Trinidad; Ceara to Parana, Brazil (Williams 1984, Melo 1998). Mesorhoea sexspinosa Stimpson, 1871 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; off northwest Florida to Flanagan Passage, Vir- gin Islands (Williams 1984). * Parthenope agona (Stimpson, 1871) Parthenope (Parthenope) agona (Stimp- son).—Williams, 1984:342. Known range.—Western Atlantic: off Cape Hatteras and Cape Lookout, North Carolina; east central Florida; Gulf of Mex- ico (Pensacola, Florida to near Ft. Myers, Florida), through Florida Straits, West In- dies and Caribbean Sea to Parana, Brazil (Williams 1984, Melo 1998). * Platylambrus fraterculus (Stimpson, 1871) Parthenope (Platylambrus) fraterculus (Stimpson).—Wilhiams, 1984:343. Known range.—Western Atlantic: off Cape Fear, North Carolina; east central Florida southward; Gulf of Mexico (off Cape San Blas, Florida) to Florida Straits; oft), Cape, Catoches~ Yucatan, IvMiexico; through West Indies to Rio Grande do Sul, Brazil (Williams 1984, Melo 1998). 134 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON * Platylambrus granulata (Kingsley, 1879) Parthenope (Platylambrus) granulata (Kingsley).—Williams, 1984:344. Known range.—Western Atlantic: Ber- muda; off Cape Hatteras, North Carolina, southward around Florida into Gulf of Mexico to Louisiana; Bahia Honda, Cuba (?); St. Thomas, U.S. Virgin Islands (Wil- liams 1984). * Platylambrus pourtalesii (Stimpson, 1871) Parthenope (Platylambrus) pourtalesii (Stimpson).—Williams, 1984:345. Known range.—Western Atlantic: off Martha’s Vineyard, Massachusetts; New Jersey to southern Florida; Gulf of Mexico through West Indies to Grenada (Williams 1984, Abele & Kim 1986). Solenolambrus tenellus Stimpson, 1871 Known range.—Western Atlantic: off Cape Lookout, North Carolina; east central Florida southward, including the Florida Keys, into eastern Gulf of Mexico to near Cape St. George, Florida; Bahamas; Bar- bados (Williams 1984, Abele & Kim 1986). Solenolambrus typicus Stimpson, 1871 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; western Gulf of Mexico off Corpus Christi, Texas, and north of Yucatan; Nicaragua Shelf; southern Florida through West Indies to Suriname and Rio de Janeiro, Brazil (Williams 1984, Melo 1998). Superfamily Cancroidea Latreille, 1802 Family Cancridae Latreille, 1802 Remarks.—Based on fossil evidence (pri- marily characters of the carapace and che- lipeds) Nations (1975) recognized four sub- genera (Glebocarcinus Nations, 1975, Rom- aleon Gistl, 1848, Metacarcinus A. Milne- Edwards, 1862, Cancer, sensu stricto Linnaeus, 1758) within the genus Cancer. Williams (1984) adopted this classification. Schweitzer & Feldmann (2000) re-evaluat- ed the Cancridae and elevated the subgen- era of Nations (1975) to full generic status, again basing their conclusions only on fos- sil evidence. This classification has not gained acceptance with researchers working on extant species. Until more thorough in- vestigation, incorporating a broader range of characters, 1s conducted, the classifica- tion of Williams (1984) will be followed. Williams & Wahle (1992) summarized and illustrated the differences between juvenile Cancer borealis and C. trroratus. Cancer (Metacarcinus) borealis Stimpson, 1859 Known range.—Western Atlantic: Nova Scotia to south of Dry Tortugas, Florida (Williams 1984). Cancer (Cancer) irroratus Say, 1817 Known range.—Western Atlantic: Lab- rador to off Miami, Florida (Williams 1984). Superfamily Portunoidea Rafinesque, 1815 Family Geryonidae Colosi, 1923 > * Chaceon quinquedens (Smith, 1879) Geryon quinquedens Smith.—Williams, 1984:485. Remarks.—Williams (1984) considered this species to be extralimital. Manning & Holthuis (1989) described the new genus Chaceon for 21 species including C. quin- quedens. This species occurs at 40—2155 m depth, but is usually found at the shelf edge or on the continental slope (Wigley et al. 1975, Williams 1984). Known range.—Western Atlantic: Nova Scotia southward to Gulf of Mexico; Cuba; Brazil and Argentina (Wigley et al. 1975, Williams & Wigley 1977, Williams 1984). VOLUME 116, NUMBER 1 Family Portunidae Rafinesque, 1815 Subfamily Carcininae Alcock, 1899 Carcinus maenas (Linnaeus, 1758) Remarks.—Behrens Yamada & Hauck (2001) provided extensive information on distinguishing characteristics of this species and evaluated the usefulness of these char- acters in making field identifications. Known range.—Western Atlantic (intro- duced): Northumberland Strait and Cape Breton to Virginia; Eastern Atlantic: Ice- land; Norway, including southwestern and rarely southern Baltic Sea, through North Sea and British Isles to Mauritania; north- west Africa; also introduced to South Af- rica; Madagascar; Red Sea; Myanmar; In- dia; Ceylon; Japan; Australia; Tasmania (Williams 1984, Squires 1990, Behrens Ya- mada & Hauck 2001); multiple eastern Pa- cific localities, including sites in California, Oregon, Washington, and Vancouver Is- land, British Colombia (Cohen et al. 1995, Grosholz & Ruiz 1996, Behrens Yamada et al. 2000). Subfamily Polybiinae Ortmann, 1893 > Bathynectes longispina Stimpson, 1871 Remarks.—Williams (1984) considered this species to be extralimital. This species has been captured and observed recently at depths somewhat shallower (124—152 m; V. Guida, pers. comm.) than previously re- ported (100—1455 m, commonly >200 m; Williams & Wigley 1977). Manning & Hol- thuis (1981) reported that western Atlantic specimens identified as B. superbus (Costa, 1853) are actually B. longispina. Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts, to off Mis- sissippi River delta southward to Goajara Peninsula, Colombia; Bermuda (Williams 1984). Ovalipes ocellatus (Herbst, 1799) Known range.—Western Atlantic: North- umberland Strait, Prince Edward Island, to Georgia (Williams 1984). 135 Ovalipes stephensoni Williams, 1976 Known range.—Western Atlantic: south- ern New Jersey to Biscayne Bay, Florida (Williams 1984, Stehlik et al. 1991). Subfamily Portuninae Rafinesque, 1815 Arenaeus cribrarius (Lamarck, 1818) Remarks.—Juveniles of this species were collected at 6—10 m on sand and mud-sand sediments (Scelzo 2001). Known range.—Western Atlantic: Vine- yard Sound, Massachusetts, to Mar del Pla- ta, Argentina (Williams 1984, Scelzo 2001). Callinectes bocourti A. Milne-Edwards, 1879 Known range.—Western Atlantic: occa- sionally North Carolina, Florida, and Mis- sissippi; otherwise Jamaica; Belize to Santa Catarina, Brazil, including Antilles and northern coast of South America (Williams 1984, Melo 1998). Callinectes danae Smith, 1869 Known range.—Western Atlantic: Ber- muda; New Hanover County, North Caro- lina (near Cape Fear); southern Florida; eastern side of Yucatan Peninsula to Rio Grande do Sul, Brazil, including Antilles and northern coast of South America (Wil- liams 1984, Melo 1998). Callinectes exasperatus (Gerstaecker, 1856) Known range.—Western Atlantic: Ber- muda; Duval County, east of Jacksonville, Florida, to Santa Catarina, Brazil, including Antilles and Venezuela; extreme southern Texas; Veracruz, Mexico (Williams 1984, Melo 1998). Callinectes larvatus Ordway, 1863 Known range.—Western Atlantic: Ber- muda; Beaufort, North Carolina, through Caribbean Sea to Sao Paulo, Brazil, includ- ing Antilles (Williams 1984, Melo 1998). 136 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Callinectes ornatus Ordway, 1863 Known range.—Western Atlantic: Vir- ginia through southern Florida; Bermuda; northwestern Yucatan to Rio Grande do Sul, Brazil, including Antilles (Williams 1984, Melo 1998). Callinectes sapidus Rathbun, 1896 Known range.—Western Atlantic: Cape Cod, Massachusetts, to northern Argentina, including Bermuda, Antilles, Central Amer- ica, and Venezuela (occasionally north of Cape Cod to Maine and Nova Scotia, in favorably warm periods); Eastern Atlantic: @Mresund, Denmark; the Netherlands and ad- jacent North Sea; northwest and southwest France; Mediterranean Sea, including northern Adriatic, Aegean, and western Black seas; Eastern Pacific: Japan (Wil- lams 1984, Squires 1990, Melo 1998). Callinectes similis Williams, 1966 Known range.—Western Atlantic: off Delaware Bay to Key West, Florida; north- western Florida around Gulf of Mexico to off Campeche, Yucatan; Isla de Providen- cia, Colombia (Williams 1984). Cronius ruber (Lamarck, 1818) Known range.—Western Atlantic: Little Egg Inlet, New Jersey; Rehoboth Bay, Del- aware; Virginia; South Carolina to Rio Grande do Sul, Brazil, including Central America, Antilles, and northern South America; Eastern Atlantic: West Africa from Mauritania to Angola; Cape Verde, Principe, Sao Tome and Annobon islands; Eastern Pacific: Baja California to Peru; Clipperton Island; Galapagos Islands (Wil- liams 1984, Melo 1998). Portunus anceps (de Saussure, 1858) Known range.—Western Atlantic: Ber- muda; Cape Hatteras, North Carolina, to Rio de Janeiro, Brazil, including Antilles (Williams 1984, Melo 1998). Portunus depressifrons (Stimpson, 1859) Known range.—Western Atlantic: Ber- muda; Fort Macon, North Carolina, through northwest Florida to Gulf of Campeche and Caribbean Sea (Williams 1984). Portunus floridanus Rathbun, 1930 Known range.—Western Atlantic: east of Cape Lookout, North Carolina, to Nicara- gua; through West Indies and northern South America to Suriname (Williams 1984). Portunus gibbesii (Stimpson, 1859) Known range.—Western Atlantic: south- ern Massachusetts through Gulf of Mexico along coast to French Guiana (Williams 1984); Bahia, Brazil (Melo 1998). Portunus ordwayi (Stimpson, 1860) Known range.—Western Atlantic: Vine- yard Sound, Massachusetts; Bermuda; North Carolina through Gulf of Mexico, West Indies and Caribbean Sea to Rio Grande do Sul, Brazil; Fernando de Noro- nha, Brazil (Williams 1984, Melo 1998). Portunus sayi (Gibbes, 1850) Known range.—Western Atlantic: off the Grand Banks through Gulf of Mexico to the Guianas; Bermuda; Eastern Atlantic: Ca- nary Islands; Morocco (Williams 1984, Squires 1990). Portunus spinicarpus (Stimpson, 1871) Known range.—Western Atlantic: off Oregon Inlet, North Carolina, to Santa Ca- tarina, Brazil, including Antilles and north- ern South America (Williams 1984, Melo 1998). Portunus spinimanus Latreille, 1819 Known range.—Western Atlantic: New Jersey to Rio Grande do Sul, Brazil, in- cluding Gulf of Mexico, West Indies and VOLUME 116, NUMBER 1 northern South America: Bermuda (Wil- liams 1984, Melo 1998). Superfamily Xanthoidea MacLeay, 1838 Remarks.—Guinot (1978) concluded that Xanthidae actually represented eight differ- ent families (Carpiliidae Ortmann, 1893, Menippidae Ortmann, 1893, Platyxanthidae Guinot, 1977, Xanthidae McLeay, 1838, Panopeidae Ortmann, 1893, Pilumnidae Sa- mouelle, 1819, Trapeziidae Miers, 1886, and Geryonidae Colosi, 1924) which should be recognized under the superfamily © Xanthoidea. Williams (1984) elected not to adopt this classification. Instead, he placed the majority of xanthid taxa (exclusive of the Goneplacidae) under the single family Xanthidae. In the intervening years, Guin- ot’s classification, with minor modification, has gained acceptance by the majority of crustacean researchers (Schubart, Neigel, & Felder 2000, Martin & Davis 2001). Family Goneplacidae MacLeay, 1838 Remarks.—The Goneplacidae has long been recognized as containing heterogenous groups of genera (Hendrickx 1998, and ref- erences therein). Because revisionary stud- ies of the Goneplacidae are still in progress, formal subdivision into subfamilies is not attempted here. Furthermore, after further revision of this family, it is likely that new families will be added and subfamilies el- evated to full family status (Guinot 1978, Williams 1984, Sternberg, pers. comm.). Euryplax nitida Stimpson, 1859 Known range.—Western Atlantic: Ber- muda; off Beaufort, North Carolina, to Heald Bank, Texas; Antilles to Santa Ca- tarina, Brazil (Williams 1984, Melo 1998). Frevillea hirsuta (Borradaile, 1916) Known range.—Western Atlantic: North Carolina to Rio Grande do Sul, Brazil (Wil- liams 1984, Melo 1998). 137 Goneplax sigsbei (A. Milne-Edwards, 1880) Known range.—Western Atlantic: east of Cape Fear, North Carolina; Grenada (Wil- liams 1984). Speocarcinus carolinensis Stimpson, 1859 Known range.—Western Atlantic: south of Cape Hatteras, North Carolina, to Rio Grande do Sul, Brazil, including West In- dies (Williams 1984, Melo 1998). * Family Menippidae Ortmann, 1893 Remarks.—Williams (1984) listed these taxa under the family Xanthidae. Eriphia gonagra (Fabricius, 1781) Known range.—Western Atlantic: Ber- muda; North Carolina to Patagonia, includ- ing Central America, Antilles, and northern South America (Williams 1984, Melo 1998). Menippe mercenaria (Say, 1818) Remarks.—Williams & Felder (1986) recognized two morphologically distinct populations of stone crab in the Gulf of Mexico and determined that these popula- tions represented distinct species. Range of Menippe mercenaria was restricted; M. ad- ina Williams & Felder, which ranges from northwestern Florida around the Gulf of Mexico to Tamaulipas State, Mexico, was described (Williams & Felder 1986). Known range.—Western Atlantic: Cape Lookout, North Carolina, through peninsu- lar Florida; Bahamas and Greater Antilles to Yucatan Peninsula, Mexico; and Belize (Williams & Felder 1986). * Family Panopeidae Ortmann, 1893 Remarks.—Williams (1984) listed these taxa under the family Xanthidae. Genera in- cluded in this family were based on the conclusions of Guinot (1978) and subse- quent acceptance of generic placements by 138 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Martin & Abele (1986). Schubart, Neigel, & Felder (2000) presented a molecular phy- logeny of western Atlantic Panopeidae. * Dyspanopeus sayi (Smith, 1869) Neopanope sayi (Smith).—Williams, 1984: 409. Remarks.—Martin & Abele (1986) de- scribed the new genus Dyspanopeus for D. sayi (occurring in the western Atlantic) and D. texanus (occurring in the Gulf of Mex- 1CO). Known range.—Western Atlantic: south- ern Gulf of St. Lawrence to Florida Keys; Eastern Atlantic Gntroduced): Bristol Chan- nel, United Kingdom (Williams 1984, Squires 1990). Eurypanopeus abbreviatus (Stimpson, 1860) Known range.—Western Atlantic: South Carolina to Rio Grande do Sul, Brazil, in- cluding Gulf of Mexico, Antilles, and northern South America (Williams 1984, Melo 1998). Eurypanopeus depressus (Smith, 1869) Known range.—Western Atlantic: Mas- sachusetts Bay through Florida to southern Texas; Dutch West Indies; Uruguay; Ber- muda (Williams 1984). Eurytium limosum (Say, 1818) Known range.—Western Atlantic: Ber- muda; South Carolina; Louisiana to Santa Catarina, Brazil, including West Indies and Caribbean Sea (Williams 1984, Melo 1998). Glyptoplax smithii A. Milne-Edwards, 1880 Remarks.—This species previously was placed in the family Goneplacidae (Wil- hams 1984). Guinot (1978) transferred this species to the Panopeidae; Martin & Abele (1986) adopted this arrangement. Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Gulf of Mexico and Yucatan Channel (Williams 1984). Hexapanopeus angustifrons (Benedict & Rathbun, 1891) Known range.—Western Atlantic: Vine- yard Sound, Massachusetts, to Port Aran- sas, Texas; West Indies; Pernambuco to Santa Catarina, Brazil (Williams 1984, Melo 1998). Hexapanopeus paulensis Rathbun, 1930 Known range.—Western Atlantic: South Carolina to Uruguay, including Gulf of Mexico (Williams 1984). Panopeus herbstii H. Milne Edwards, 1834 Remarks.—Williams (1983) examined the Panopeus herbstii species complex and determined that ‘forms’ previously recog- nized by Rathbun (1930) were sufficiently different and represented distinct species. As a result, P. herbstii was better defined, and its range determined to be the shallow intertidal and subtidal waters of the eastern United States. Known range.—Western Atlantic: Bos- ton Harbor, Massachusetts, to Indian River County, Florida (Williams 1983). > Panopeus obesus Smith, 1869 Remarks.—This species was recognized previously as a form of Panopeus herbstii (Rathbun 1930). Williams (1983) rediag- nosed this form and elevated it to full spe- cies rank; diagnosis, measurements, and color description of this species were also provided. Panopeus obesus is found in marsh edge, and shallow intertidal and sub- tidal waters of the Carolinian province. Known range.—Western Atlantic: Beau- fort, North Carolina, to northeastern Flori- da; Sarasota County, Florida, to Louisiana; Texas; northern Mexico (Williams 1983). VOLUME 116, NUMBER 1 Panopeus occidentalis de Saussure, 1857 Known range.—Western Atlantic: Ber- muda; North Carolina to Santa Catarina, Brazil, including Central America, Antilles and northern South America (Williams 1984, Melo 1998). Panoplax depressa Stimpson, 1871 Remarks.—This species was placed pre- viously in the family Goneplacidae (Wil- liams 1984). Guinot (1978) transferred this species to the Panopeidae; Martin & Abele (1986) adopted this arrangement. Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; off Jacksonville and Cape San Blas, Florida; through Antilles; Amapa to Pernambuco, Brazil (Williams 1984, Melo 1998). Rhithropanopeus harrisii (Gould, 1841) Known range.—Western Atlantic: south- western Gulf of St. Lawrence to Veracruz, Mexico; Eastern Atlantic (introduced): parts of Europe; Eastern Pacific (intro- duced): west coast of United States (Wil- lams 1984, Squires 1990). * Family Pilumnidae Samouelle, 1819 Remarks.—Williams (1984) listed these taxa under the family Xanthidae. Lobopilumnus agassizii (Stimpson, 1871) Known range.—Western Atlantic: Ber- muda; North Carolina; eastern Gulf of Mexico; Yucatan; Cuba; Venezuela; Trini- dad (Williams 1984). Pilumnus dasypodus Kingsley, 1879 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to Santa Ca- tarina, Brazil, including Gulf of Mexico, Caribbean Sea, and West Indies (Williams 1984, Melo 1998). 139 Pilumnus floridanus Stimpson, 1871 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to Bahia, Brazil, including Gulf of Mexico, Yucatan Channel, Central America, Venezuela, and West Indies (Williams 1984, Melo 1998). Pilumnus lacteus Stimpson, 1871 Known range.—Western Atlantic: near Beaufort, North Carolina, to Florida; Cuba (Williams 1984). Pilumnus pannosus Rathbun, 1896 Known range.—Western Atlantic: Bogue Sound, off Beaufort, North Carolina, to Port Aransas, Texas; West Indies to Virgin Is- lands (Williams 1984). Pilumnus sayi Rathbun, 1897 Known range.—Western Atlantic: North Carolina to Curacao, including Gulf of Mexico and West Indies (Williams 1984). % Family Pseudorhombilidae Alcock, 1900 Remarks.—This taxon was previously considered a subfamily (Pseudorhombili- nae) within the Goneplacidae (Williams 1984). Hendrickx (1998) proposed the fam- ily designation to accommodate six genera. Nanoplax xanthiformis (A. Milne-Edwards, 1881) Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Rio de Janeiro, Brazil, including Gulf of Mexico, Yucatan, West Indies, and northern South America (Williams 1984, Melo 1998). * Family Trapeziidae Miers, 1886 Remarks.—Williams (1984) listed this taxon under the family Xanthidae. 140 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Domecia acanthophora acanthophora (Desbonne & Schramm, 1867) Known range.—Western Atlantic: Ber- muda; Cape Lookout Shoals, North Caro- lina; northwestern Gulf of Mexico to Ala- goas, Brazil, including West Indies and Ca- ribbean Sea (Williams 1984, Melo 1998). Family Xanthidae MacLeay, 1838 Allactaea lithostrota Williams, 1974 Known range.—Western Atlantic: Ber- muda; off Cape Lookout, North Carolina; Florida Straits; off Cape Catoche, Yucatan; off Venezuela and Suriname (Williams 1984); Antilles; off Rio de Janeiro to Rio Grande do Sul, Brazil (Melo 1998). Carpoporus papulosus Stimpson, 1871 Known range.—Western Atlantic: be- tween Cape Hatteras and Cape Lookout, North Carolina; Gulf of Mexico, off Mobile Bay southeastward; Cape Catoche, Yucatan (Williams 1984). Glyptoxanthus erosus (Stimpson, 1859) Known range.—Western Atlantic: Cape Lookout, North Carolina, southward around Florida into Gulf of Mexico to off Grand Isle, Louisiana; Yucatan; West Indies to Guadeloupe (Williams 1984). Melybia thalamita Stimpson, 1871 Known range.—Western Atlantic: off Cape Lookout, North Carolina, southwest of Mississippi River delta to Sao Paulo, Brazil, including West Indies and northern South America (Williams 1984, Melo 1998). Micropanope nuttingi (Rathbun, 1898) Known range.—Western Atlantic: Cape Hatteras, North Carolina, to Sao Paulo, Bra- zil, including Gulf of Mexico and West In- dies (Williams 1984, Melo 1998). Micropanope sculptipes Stimpson, 1871 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina, to Port Aransas, Texas; West Indies to Bar- bados; Amapa to Rio de Janeiro, Brazil (Williams 1984, Melo 1998). Micropanope urinator (A. Milne-Edwards, 1881) Known range.—Western Atlantic: off Cape Hatteras and Cape Lookout, North Carolina; Florida Keys to St. Croix (Wil- hams 1984); Antilles; Para to Maranhao, Brazil (Melo 1998). Paractaea rufopunctata nodosa (Stimpson, 1860) Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; off Mississippi River delta, through West In- dies to Uruguay; Central Atlantic: Ascen- sion Island (Williams 1984, Melo 1998). Pseudomedaeus agassizii (A. Milne-Edwards, 1880) Known range.—Western Atlantic: Cape Hatteras, North Carolina, to southern Texas (Williams 1984). Pseudomedaeus distinctus (Rathbun, 1898) Known range.—Western Atlantic: off Cape Hatteras, North Carolina, through Straits of Florida to northwest of Dry Tor- tugas; Puerto Rico; Barbados (Williams 1984). Tetraxanthus rathbunae Chace, 1939 Known range.—Western Atlantic: off Cape Lookout, North Carolina, to Rio Grande do Sul, Brazil, including Gulf of Mexico and Antilles (Williams 1984, Melo 1998). VOLUME 116, NUMBER 1 Superfamily Pinnotheroidea De Haan, 1833 Family Pinnotheridae De Haan, 1833 Remarks.—Marques & Pohle (1995) conducted a phylogenetic analysis of this family using larval characters and demon- strated that several taxa are paraphyletic taxa and that further analysis will be needed to resolve relationships within this family. Subfamily Pinnotherinae De Haan, 1833 Remarks.—Griffith (1987) presented a hypothesis of phylogenetic relationships within the genus Dissodactylus based on adult morphology. Marques & Pohle (1995) conducted a phylogenetic analysis of mem- bers of this genus using larval characters and produced results that only partially cor- roborated the relationships proposed by Griffith (1987). In a separate analysis, both data sets (adult morphology and larval mor- phology) were combined (Marques & Pohle 1995). Results of this analysis provided a more robust hypothesis of relationships than when either character set was analyzed independently. Dissodactylus crinitichelis Moreira, 1901 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; off northwest Florida; Caribbean Sea and South America to Rio de la Plata, Argentina (Wil- liams 1984). Dissodactylus mellitae (Rathbun, 1900) Known range.—Western Atlantic: west- ern Vineyard Sound, Massachusetts, to Charleston, South Carolina; Hutchinson Is- land, east Florida; western Florida; off Gal- veston, Texas (Williams 1984). * Gemmotheres chamae (Roberts, 1975) Pinnoiheres chamae Roberts.—Williams, 1984:440. Remarks.—Campos (1996) described the 141 new genus Gemmotheres for P. chamae; il- lustrations were also provided. Known range.—Western Atlantic: North Carolina coast (Williams 1984, Campos 1996). > Parapinnixa beaufortensis Rathbun, 1918 Remarks.—Williams (1984) considered this species to be extralimital. Known range.—Western Atlantic: off Beaufort, North Carolina (Williams 1984). Parapinnixa bouvieri Rathbun, 1918 Known range.—Western Atlantic: off Charleston, South Carolina; south of Dry Tortugas, Florida; off Cape Catoche, Yu- catan; Puerto Rico; Amapa, Brazil (Wil- liams 1984, Melo 1998). Parapinnixa hendersoni Rathbun, 1918 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; off Tampa Bay, Florida, through West Indies to Curacao; Venezuela; Maranhao to Espirito Santo, Brazil (Williams 1984, Melo 1998). Pinnaxodes floridensis Wells and Wells, 1961 Known range.—Western Atlantic: off North Carolina to Georgia; northwest Flor- ida (Williams 1984). * Tumidotheres maculatus (Say, 1818) Pinnotheres maculatus Say.—Williams, 1984:441. Remarks.—Campos (1989) described the new genus Tumidotheres and discussed ge- neric relationships and life history traits. Known range.—Western Atlantic: off Martha’s Vineyard, Massachusetts, to Golfo San Matias, Argentina (Williams 1984). * Zaops ostreum (Say, 1817) Pinnotheres ostreum Say.—Williams, 1984: 444. 142 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Remarks.—Manning (1993) concluded that Zaops Rathbun, 1900 should be re- moved from the synonymy of Pinnotheres and recognized as a distinct genus. Addi- tionally, other pinnotherid genera formerly placed in the synonymy of Pinnotheres were diagnosed, figured and differentiated from Pinnotheres sensu Manning (1993). Known range.—Western Atlantic: Salem, Massachusetts, to Santa Catarina, Brazil (Williams 1984). Subfamily Pinnothereliinae Alcock, 1900 * Austinixa cristata (Rathbun, 1900) Pinnixa cristata Rathbun.—Williams, 1984:453. Remarks.—Heard & Manning (1997) recognized and described the new genus Austinixa for Pinnixa cristata and six other species formerly assigned to Pinnixa. Known range.—Western Atlantic: Beau- fort, North Carolina, to Miami, Florida; northern and southwestern Gulf of Mexico (Manning & Felder 1989); Central Ameri- ca; Brazil (Melo 1998). Pinnixa chaetopterana Stimpson, 1860 Known range.—Western Atlantic: Well- fleet, Massachusetts, to Rio Grande do Sul, Brazil (Williams 1984). Pinnixa cylindrica (Say, 1818) Known range.—Western Atlantic: North Falmouth, Massachusetts, to Pensacola, Florida, including Dry Tortugas (Williams 1984). Pinnixa floridana Rathbun, 1918 Known range.—Western Atlantic: south- east of Cape Lookout, North Carolina; Hutchinson Island, east Florida; west coast of Florida (Williams 1984). Pinnixa lunzi Glassell, 1937 Known range.—Western Atlantic: off the eastern shore of Virginia; North and South Carolina; Georgia; off Mississippi River delta; Seven and One-Half Fathom Reef, off Texas (Williams 1984). Pinnixa retinens Rathbun, 1918 Known range.—Western Atlantic: Dela- ware Bay; Chesapeake Bay; Little River In- let, South Carolina; Alligator Harbor, Flor- ida; Aransas area of Texas coast (Williams 1984). Pinnixa sayana Stimpson, 1860 Known range.—Western Atlantic: Vine- yard Sound, Massachusetts, to Beaufort, North Carolina; Hutchinson Island, east central Florida; Sarasota Bay, Florida to Grand Isle, Louisiana; Amapa to Rio Grande do Sul, Brazil (Williams 1984, Melo 1998). Superfamily Ocypodoidea Rafinesque, 1815 Family Ocypodidae Rafinesque, 1815 Subfamily Ocypodinae Rafinesque, 1815 Remarks.—Rosenberg (2001) conducted a phylogenetic analysis of the genus Uca using 236 discrete morphological charac- ters. Although many scientists ignore sub- generic designations, Rosenberg (2001) considered subgenera within Uca to be val- id. The three species of Uca occurring in the region are considered to be members of the subgenus Minuca (Rosenberg 2001). Ocypode quadrata (Fabricius, 1787) Known range.—Western Atlantic: Block Island, Rhode Island, to Rio Grande do Sul, Brazil; Fernando de Noronha Archipelago, Brazil; Bermuda (Williams 1984, Melo 1998). Uca minax (LeConte, 1855) Remarks.—Felder & Staton (1994) ana- lyzed electrophoretic allozyme assays and observed slight differentiation between Gulf of Mexico and Atlantic populations. VOLUME 116, NUMBER 1 Known range.—Western Atlantic: Buz- zards Bay, Cape Cod, Massachusetts, to Daytona Beach, Florida; Yankeetown, northwest Florida, to Matagorda Bay, Texas (Williams 1984, Barnwell & Thurman 1984). Uca pugilator (Bosc, 1802) Known range.—Western Atlantic: Cape Cod, Massachusetts, southward around the tip of peninsular Florida and westward to Pensacola Beach, Florida; possible rare oc- currences in Bahamas and western Gulf of Mexico (Barnwell & Thurman 1984). Uca pugnax (Smith, 1870) Known range.—Western Atlantic: Prov- incetown, Massachusetts, to Daytona Beach, Florida (Williams 1984, Barnwell & Thurman 1984). Family Palicidae Bouvier, 1898 Palicus alternatus Rathbun, 1897 Known range.—Western Atlantic: Cape Hatteras to southeast -of Cape Fear, North Carolina; Gulf of Mexico along west coast of Florida from Cape San Blas to Key West (Williams 1984). Palicus faxoni Rathbun, 1897 Known range.—Western Atlantic: off Cape Hatteras, North Carolina, to near Cape Canaveral, Florida; off Yucatan, Mex- ico, near Quita Sueno Banks; southwest of St. Christopher; Rio Grande do Norte to Rio de Janeiro, Brazil (Williams 1984, Melo 1998). P Palicus gracilis (Smith, 1883) Remarks.—Williams (1984) considered this species to be extralimital. This species occurs at 183-512 m (Williams & Wigley 1977): Known range.—Western Atlantic: Mar- tha’s Vineyard, Massachusetts, to Curacao, including Gulf of Mexico (Williams & Wigley 1977). 143 Palicus sica (A. Milne-Edwards, 1880) Known range.—Western Atlantic: off Charleston, South Carolina, to Cape Can- averal, Florida; west coast of Florida to Rio Grande do Sul, Brazil, including West In- dies (Williams 1984, Melo 1998). Superfamily Grapsoidea MacLeay, 1838 Family Grapsidae MacLeay, 1838 Remarks.—Previously, family Grapsidae was considered to be comprised of four subfamilies. Based on a cladistic study by Sternberg & Cumberlidge (1998) the Grap- sidae were redefined and restricted to in- clude all genera previously placed in the subfamily Grapsinae. Molecular data also supported elevation of grapsid subfamilies to full family status (Schubart, Cuesta, Die- sel, & Felder 2000). Pachygrapsus transversus (Gibbes, 1850) Known range.—Western Atlantic: Ber- muda; Cape Lookout, North Carolina, to Montevideo, Uruguay; Eastern Atlantic: Mediterranean Sea to northern Angola; Eastern Pacific: California to Peru; Gala- pagos Islands (Williams 1984). Planes minutus (Linnaeus, 1758) Known range.—Western Atlantic: off the Grand Banks of Newfoundland south to 11°N, exclusive of Gulf of Mexico; Eastern Atlantic: southern North Sea south to 11°N, including Mediterranean Sea (Williams 1984, Squires 1990). * Family Plagusiidae Dana, 1851 Remarks.—This taxon was_ previously considered a subfamily of the Grapsidae (e.g., Williams 1984, Guinot & Bouchard 1998). Based on results of a cladistic study, Sternberg & Cumberlidge (1998) concluded that the two genera (Percnon and Plagusia) placed in the subfamily Plagusiinae formed a monophyletic taxon; subfamily Plagusi- inae was redefined and elevated to family 144 status (Plagusiidae). Based on results of a molecular phylogeny, however, Schubart, Cuesta, Diesel, & Felder (2000) questioned taxonomic placement of Percnon in this family. Sternberg & Cumberlidge (1998) had placed Euchirograpsus in the Varuni- nae. However, based on morphological and molecular evidence (Schubart, Cuesta, Die- sel, & Felder 2000), Euchirograpsus be- longs in the Plagusiidae. Euchirograpsus americanus A. Milne-Edwards, 1880 Remarks.—This species was previously considered a member of the Varuninae (Williams 1984). Based on molecular evi- dence and larval morphology, Schubart, Cuesta, Diesel, & Felder (2000) proposed inclusion of this species in the family Pla- gusiidae. Known range.—Western Atlantic: Oceanographer Canyon, edge of Georges Bank; off Oregon Inlet, North Carolina; Florida to Venezuela, including West In- dies; Rio Grande do Sul, Brazil (Williams 1984, 1988; Melo 1998). Percnon gibbesi (H. Milne Edwards, 1853) Remarks.—Based on molecular evi- dence, Schubart, Cuesta, Diesel, & Felder (2000) suggested that this species does not belong in the family Plagusiidae. Those au- thors reported that the taxonomic position of Percnon gibbesi was uncertain and re- quired further investigation. Until more de- finitive results are available, placement of this species will remain in the Plagusiidae. Known range.—Western Atlantic: Ber- muda; Fort Macon, North Carolina; south- ern Florida and Bahamas to Fernando de Noronha Archipelago, Brazil, including Antilles; Eastern Atlantic: Azores to An- gola; Mediterranean Sea (Pipitone et al. 2001); Eastern Pacific: Cape San Lucas, lower California, to Chile; Galapagos Is- lands (Williams 1984, Melo 1998). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Plagusia depressa (Fabricius, 1775) Known range.—Western Atlantic: Ber- muda; Beaufort, North Carolina, to Bahia, Brazil, including Gulf of Mexico and West Indies; Fernando de Noronha Archipelago, Rocas Atoll, Saint Paul Rocks, and Trin- dade Island, Brazil; Eastern Atlantic: Azores; Madeira; Morocco to northern An- gola; Central Atlantic: St. Helena Island (Williams 1984, Melo 1998). * Family Sesarmidae Dana, 1851 Remarks.—This taxon was_ previously considered a subfamily of the Grapsidae (e.g., Williams 1984, Guinot & Bouchard 1998). Based on results of a cladistic study, Sternberg & Cumberlidge (1998) concluded that taxa previously included in the subfam- ily Sesarminae formed a polyphyletic group. However, one clade comprising the majority of Sesarma-like genera, including Sesarma and Armases, was redefined as the family Sesarmidae (Sternberg & Cumber- lidge 1998). * Armases cinereum (Bosc, 1802) Sesarma (Chiromantes) cinereum (Bosc).— Williams, 1984:465. Remarks.—Abele (1992) described the new genus Armases for American species of Sesarma assigned to the subgenus Chi- romantes; generic diagnosis, species de- scription, and illustrations are provided. The phylogeny of Armases proposed by Niem (1996) supported establishment of Armases Abele, 1992. Known range.—Western Atlantic: Ma- gothy River, Chesapeake Bay, Maryland, to Palm Beach, east Florida; Collier County, west Florida, to Veracruz, Mexico (Wil- hams 1984). * Sesarma reticulatum (Say, 1817) Sesarma (Sesarma) reticulatum (Say).— Williams, 1984:466. Remarks.—Subgenera were not recog- VOLUME 116, NUMBER 1 nized within the genus Sesarma by Abele (1992). Based on electrophoretic results, Felder & Staton (1994) concluded that al- lozyme divergence between Gulf of Mexico and Atlantic populations was comparable to levels previously reported for speciated populations, suggesting that Sesarma reti- culatum represented a species complex. Known range.—Western Atlantic: Woods Hole, Massachusetts, to Volusia County, east Florida; Sarasota, west Florida, to Bar- ra del Tordo, Tamaulipas, Mexico (Williams 1984, Felder & Staton 1994). %* Family Varunidae H. Milne Edwards, 1853 Remarks.—This taxon was_ previously considered a subfamily of the Grapsidae (e.g., Williams 1984, Guinot & Bouchard 1998). Cladistic analysis revealed that the Varuninae is an artificial group in need of re-examination (Sternberg & Cumberlidge 1998). It is possible that this group, pres- ently recognized at the family level (Schu- bart, Cuesta, Diesel, & Felder 2000, Martin & Davis 2001) may prove to represent sev- eral different families (Sternberg & Cum- berlidge 1998). Molecular data also support this conclusion (Schubart, Cuesta, Diesel, & Felder 2000). > Hemigrapsus sanguineus (De Haan, 1853) Remarks.—The Asian shore crab was most likely introduced in the western At- lantic via ballast water discharged some- time in the early 1980’s (McDermott 1998). The crab was first discovered in 1988 in southern New Jersey (Williams & Mc- Dermott 1990). Hemigrapsus sanguineus was the most abundant brachyuran at the intertidal monitoring site in southern New Jersey, some areas of Long Island Sound (McDermott 1998) and Narragansett Bay, RI (pers. obs.). McDermott (1998) reported this crab occurring in the upper to middle intertidal zone of New Jersey, whereas oth- ers (Lohrer & Whitlatch 1997, Ledesma & 145 O’Connor 2001) observed higher abun- dances in the middle and lower intertidal zone of sampling locations in eastern Long Island Sound and southeastern New Eng- land. Crab abundance increased with in- creased rock cover (Ledesma & O’Connor 2001). Known range.—Western Atlantic: Ap- pledore Island, Isles of Shoals, Maine (J. Morin, pers. comm.) and New Hampshire (McDermott 2000) to Oregon Inlet, North Carolina (McDermott 1998): Eastern Atlan- tics Ke Havres France: <@osterschelde’’, Netherlands (Breton et al. 2002); Western Pacific: Sakhalin, Korea; north China to Hong Kong; all coasts of Japan from Hok- kaido to Okinawa (Williams & McDermott 1990). Discussion The decapod crustacean assemblage oc- curring in shallow waters (<190 m) of the temperate eastern United States from Maine to Cape Canaveral, Florida, totals 391 spe- cies. This assemblage includes 122 shrimp species (28 penaeids, 2 stenopodids, and 92 carideans), 10 thalassinideans, 8 lobsters, 61 anomurans, and 190 brachyurans. By comparison, the previous comprehensive review of this assemblage by Williams (1984) recognized 103 species of shrimps (21 penaeids, 2 stenopodids, and 80 cari- deans), 8 thalassinideans (including calli- anassids, upogebiids and axiids), 6 lobsters, 51 anomurans, and 174 brachyurans (total decapods = 342). Since publication of Williams (1984), 51 species are new to this checklist. Thirteen species (Table 1) previously considered ex- tralimital by Williams (1984:484) because their centers of distribution or abundance occur beyond the boundaries of the region are now incorporated into the updated checklist because their geographic and bathymetric ranges are within the limits set for the region under consideration. An ad- ditional 16 species (Table 1), most likely excluded (i.e., not even considered as ex- 146 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Western Atlantic decapod Crustacea occurring in regions that were considered ‘extralimital’ or excluded from consideration by Williams (1984) because their geographic or bathymetric range centered beyond the boundaries of regions of consideration. “Extralimital’ Pleoticus robustus Bythocaris nana Caridion gordoni Plesionika martia Plesionika tenuipes Plesionika willisi Catapagurus sharreri Hemipagurus gracilis Mithrax cornutus Bathynectes longispina Chaceon quinquedens Parapinnixa beaufortensis Palicus gracilis tralimital) from consideration by Williams because their bathymetric distributions were centered beyond 190 m or their geo- graphic distributions were centered south of Cape Canaveral, Florida, are included in the present checklist for this same reason. Twelve species, discovered and described during the past 20 years, were also added to the regional assemblage (Table 2). Four species (Trachycaris rugosa, Axius arma- tus, Pagurus brevidactylus, Panopeus obe- sus) were added through a refined under- standing of the systematics of species com- plexes with improved recognition and de- lineation of component species. One nonindigenous species (Hemigrapsus san- guineus) was introduced and has become established in the region. An additional five species (Parapenaeus americanus, Scyllar- ides aequinoctialis, Petrolisthes armatus, Dromia erythropus, Clythrocerus nitidus) have also become part of the regional as- semblage through northward extension of their geographic ranges. Six species have been removed from Williams (1984) checklist (Table 3). Four species (Pontonia margarita, Anisopagurus pygmaeus, Iridopagurus caribbensis, and Hyas coarctatus alutaceus) once thought to Excluded Aristaeomorpha foliacea Hadropenaeus affinis Hadropenaeus modestus Penaeopsis serrata Lucifer typus Heterocarpus ensifer Plesionika edwardsii Nephropsis aculeata Tomopagurus cokeri Parapagurus pilosimanus Sympagurus pictus Munida forceps Lysirude nitidus Cyclozodion angustum Pyromaia arachna Stenocionops furcata furcata be members of this assemblage (Williams 1984) are removed from the checklist be- cause recent revisions and investigations in- volving these species (Garcia-G6mez 1983, Lemaitre & McLaughlin 1996, Fransen 2000) clearly indicate that they do not occur in the region. One eastern Pacific species (Notolopas lamellatus) was erroneously re- ported from the region by Rathbun (Wil- hams 1984). Additionally, Leptochela ber- mudensis was previously included in the checklist because it was considered as like- ly to occur in the region by Williams (1984). Since this species has never actually been recorded from the region, it has been removed from the checklist. Better understanding of species concepts and interspecific variation has necessitated placement of some nominal species previ- ously considered part of this regional as- semblage into synonymy. For decapod crustaceans of the eastern United States, eleven nominal species (Table 4) have been re-evaluated and determined not to repre- sent distinct species. Additionally, one fam- ily and one genus have also been placed into synonymy (Table 4). A significant proportion of systematic re- search on western Atlantic decapod crusta- VOLUME 116, NUMBER 1 147 Table 2—Summary of more recently described western Atlantic decapod Crustacea not previously listed in Williams (1984). Year of descrip- Depth of Family Genus and species tion Geographic range occurrence Disciadidae Discias vernbergi 1987 Georgia—west Florida 54-74 m Palaemonidae Pontonia manningi 2000 North Carolina—Gulf of Mexico, Ca- shallow—80 m ribbean Sea, eastern Atlantic Alpheidae Alpheus angulosus 2002 North Carolina—Gulf of Mexico, intertidal, shallow Mexico, Haiti Alpheidae Alpheus estuariensis 1984 Florida, Gulf of Mexico, Caribbean __intertidal—22 m Sea, Brazil Callianassidae Necallianassa berylae 1998 South Carolina—Georgia 35-75 m Paguridae Anisopagurus hopkinsi 1996 Georgia—Gulf of Mexico 91-165 m Paguridae Tridopagurus reticulatus 1983 North Carolina—Florida, Bahamas, 1-38 m Greater and Lesser Antilles, northern South America Paguridae Pagurus maclaughlinae 1982 Georgia—Florida, Gulf of Mexico, 1-5 m Caribbean Paguridae Tomopagurus wassi 1981 Southeastern United States, Florida 75—360 m Straits, Gulf of Mexico, Caribbe- an—northern Brazil Homolidae Homola minima 1995 Massachussetts—Brazil 55-690 m Calappidae Cyclozodion tuberatum 1988 North Carolina—Bahamas—eastern 31-188 m Gulf of Mexico, Suriname Inachidae Stenorhynchus yangi 1989 Massachussetts—Gulf of Mexico—Su- 31-365 m riname Table 3—Summary of western Atlantic decapod crustacean species previously (Williams 1984) considered to be part of the fauna off the eastern United States but now known not to occur in the region. These species were removed from the checklist. Species Range Reason for removal from checklist Leptochela bermudensis Bermuda, Puerto Rico Previously considered likely to occur in Gurney, 1939 through the Lesser Antilles the region (Williams 1984), but has never been reported. Pontonia margarita Eastern Pacific Formerly thought to be a widespread spe- Smith, 1869 cies; with resolution of species complex now known to occur only in the Pacific Ocean (Fransen 2000). Anisopagurus pygmaeus Florida Keys, Cuba to Curagao Resolution of species systematics indicated (Bouvier, 1918) geographic range occurs beyond limits of region (Lemaitre & McLaughlin 1996). Tridopagurus caribbensis Miami, Florida to Colombia Resolution of species systematics indicated (A. Milne-Edwards & Bou- geographic range occurs beyond limits vier, 1893) of region (Garcia-Gomez 1983). Hyas coarctatus alutaceus Eastern Pacific Resolution of species systematics indicated Brandt, 1851 geographic range occurs beyond limits of region (Squires 1990). Notolopas lamellatus Eastern Pacific Erroneously reported from the region Stimpson, 1871 (Williams 1984). 148 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 4.—Summary of nominal taxa of western Atlantic decapod crustaceans placed in synonymy since publication of Williams (1984). Nominal taxa appearing in Williams (1984) Automate gardineri Hippolyte curacaoensis Lebbeus zebra Callianassa atlantica Manucomplanus corallinus Pagurus hendersoni Hypoconcha sabulosa Homola barbata (in part) Cycloes bairdii Mithrax pleuracanthus Mithrax acuticornis Family Tymolidae Genus Parapandalus ceans 1s being conducted on higher-level re- lationships. Better understanding of rela- tionships based on identification of mono- phyletic groups has required reorganization of taxa at higher levels (e.g., within super- families, families, and genera) as well as elevating subfamilies and subgenera to fam- ily and generic levels, respectively. In the nearly 20-year period since Williams (1984), one new superfamily (Processoi- dea) and one new family (Anchistioididae) are recognized in the infraorder Caridea, and two subgenera have been elevated to genera (Farfantepenaeus and Litopenaeus) in the family Penaeidae, for shrimps occur- ring in the region. Thalassinideans, previ- ously placed in the section Thalassinidea within infraorder Anomura, are now rec- ognized at the level of infraorder. Within brachyuran taxa, three new subfamilies within the Raninidae are now recognized, 17 former subfamilies (seven in Mayidae, four in Xanthidae, three in Grapsidae, one each in Raninidae, Calappidae, and Gone- placidae) have been elevated to family, two subgenera elevated to genus (Mithraculus in family Majidae and Platylambrus in fam- ily Parthenopidae), and one subspecies (of Ethusa) elevated to species. Considerable changes in generic concepts involving west- ern Atlantic decapod crustaceans have also necessitated the recognition and redefinition Senior synonym Automate dolichognatha Hippolyte obliquimanus Lebbeus microceros Gilvossius setimanus Manucomplanus ungulatus Pagurus stimpsoni Hypoconcha parasitica Homola minima Cryptosoma balguerii Mithrax hispidus Mithrax cornutus Family Cyclodorippidae Genus Plesionika of genera to accommodate western Atlantic decapod species. Fifteen new genera have been described since publication of Wil- liams’ (1984) monograph, including one genus each in the Penaeidae, Pandalidae, Axiidae, and Paguridae, two in Callianas- sidae, and nine brachyuran genera placed in seven different families (Table 5). Twenty- One species of decapods occurring in the region have been reassigned to genera other than those listed in Williams (1984). Knowledge concerning well-known fau- nas, such as that of the decapod crustaceans of the western Atlantic, is not static, and new discoveries, additional collecting, and better understanding of systematic relation- ships will continue to improve our under- standing of regional biodiversity. New ev- idence from adult morphology, fossils, lar- val development, and molecular genetics has led to the reinterpretation of classical views of decapod crustacean relationships. Hypotheses of phylogenetic relationships are being proposed at a relatively fast rate compared with previous time periods and the taxonomic status of species continues to be re-evaluated. Information gained from these systematic studies will undoubtedly result in better understanding of the species, provide refined hypotheses of relationships among these taxa, and subsequently will continue to improve our knowledge regard- VOLUME 116, NUMBER 1 Table 5.—Summary of western Atlantic decapod crustacean species assigned to new genera or reassigned to established genera since publication of Williams (1984). Species name as appears in Williams (1984) Hymenopenaeus robustus Trachypenaeus constrictus Pandalus propinquus Parapandalus willisi Pontophilus gorei Callianassa biformis Callianassa atlantica Callianassa major Axiopsis jenneri Pagurus piercei Dromidia antillensis Clythrocerus perpusillus Tliacantha intermedia Cryptopodia concava Geryon quinquedens Neopanope sayi Pinnotheres chamae Pinnotheres maculatus Pinnotheres ostreum Pinnixa cristata Sesarma cinereum Species name as appears in present compilation (Y/N) Pleoticus robustus N Rimapenaeus constrictus Atlantopandalus propinqvus Plesionika willisi Philocheras gorei Biffarius biformis Gilvossius setimanus Callichirus major Calaxius jenneri Goreopagurus piercei Cryptodromiopsis antillensis Deilocerus perpusillus Acanthilia intermedia Celatopesia concava Chaceon quinquedens Dyspanopeus sayi Gemmotheres chamae Tumidotheres maculatus Zaops ostreum Austinixa cristata mei I= IRS, Ie IS Ie IG I IG, ZS I, I I, 4, IKI Armases cinereum Newly-described genus? ing the marine decapod crustacean assem- blage of the eastern United States. Acknowledgments This paper is dedicated to A. B. Wil- liams, carcinologist extraordinaire and gen- tleman scientist. The present study was ini- tiated when A. B. Williams, shortly before his death, expressed his belief that the “green book’’ must be updated. L. Cooper and R. McMillan, Jr., organized files, pho- tocopied literature, and began literature searches while working as scientific assis- tants to A. B. Williams. W. Blow, D. Felder, V. Guida, B. Kensley, R. Lemaitre, R. Man- ning, J. Morin, T. Munroe, and R. von Sternberg provided literature, information and/or consultation. W. Blow, B. Kensley, R. Lemaitre, T. Munroe, M. Vecchione, and three anonymous reviewers critically re- viewed earlier drafts of the manuscript and provided helpful comments. Literature Cited Abele, L. G. 1992. A review of the grapsid crab genus Sesarma (Crustacea: Decapoda: Grapsidae) in America, with the description of a new ge- nus.—Smithsonian Contributions to Zoology 527:1—60. , & W. Kim. 1986. An illustrated guide to the marine decapod crustaceans of Florida.—Tech- nical Series of the State of Florida Department of Environmental Regulation, Vol. 8, Part 1 & 2:1—760. , & J. W. Martin. 1989. American species of the deep-sea shrimp genus Bythocaris (Crusta- cea, Decapoda, Hippolytidae).—Bulletin of Ma- rine Science 45:26—51. Asakura, A. 2001. A revision of the hermit crabs of the genera Catapagurus A. Milne-Edwards and Hemipagurus Smith from the Indo-West Pacific (Crustacea: Decapoda: Anomura: Paguridae).— Invertebrate Taxonomy 15:823—891. Baldwin, J. D., A. L. Bass, B. W. Bowen, & W. H. Clark, Jr. 1998. Molecular phylogeny and bio- geography of the marine shrimp Penaeus.— Molecular Phylogenetics and Evolution 10: 399—407. Barnwell, EF H., & C. L. Thurman, II. 1984. Taxonomy and biogeography of the fiddler crabs (Ocypod- idae: Genus Uca) of the Atlantic and Gulf 150 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON coasts of eastern North America.—Zoological Journal of the Linnean Society 81:23-—87. Behrens Yamada, S., & L. Hauck. 2001. Field identi- fication of the European green crab species: Carcinus maenas and Carcinus aestuarii.— Journal of Shellfish Research 20:905—912. , C. Hunt, & N. Richmond. 2000. The arrival of the European green crab, Carcinus maenas, in Oregon estuaries. Pp. 94—99 in J. Pederson, ed., Marine bioinvasions: Proceedings of the First National Conference, January 24-27, 1999. MIT, MIT Sea Grant College Program, Massachusetts. Bellwood, O. 1996. A phylogenetic study of the Ca- lappidae H. Milne Edwards 1837 (Crustacea: Brachyura) with a reappraisal of the status of the family.—Zoological Journal of the Linnean Society 118:165—193. Boothe, B. B., Jr., & R. W. Heard. 1987. Discias vern- bergi, new species, a caridean shrimp (Crusta- cea: Decapoda: Bresiliidae) from the north- western Atlantic.—Proceedings of the Biologi- cal Society of Washington 100:506—5 14. Bowman, T. E., & J. C. McCain. 1967. Distribution of the planktonic shrimp, Lucifer, in the western North Atlantic.—Bulletin of Marine Science 17: 660-671. Breton, G., M. Faasse, P. Noél, & T. Vincent. 2002. A new alien crab in Europe: Hemigrapsus san- guineus (Decapoda: Brachyura: Grapsidae).— Journal of Crustacean Biology 22:184—-189. Caetano da Costa, R., A. Fransozo, E L. M. Mante- latto, & R. H. Castro. 2000. Occurrence of shrimp species (Crustacea: Decapoda: Natantia: Penaeidea and Caridea) in Ubatuba Bay, Uba- tuba, SP, Brazil.—Proceedings of the Biological Society of Washington 113:776—781. Calado, T. C. dos S. 1998. Malacostraca—Eucarida. Hippoidea. Pp. 407-411 in P. S. Young, ed., Catalogue of Crustacea of Brazil. Museu Na- cional, Rio de Janeiro. (Série Livros n. 6). Campos C., N. H., & H. Sanchez M. 1995. Los can- grejos ermitanos del género Paguristes Dana (Anomura: Diogenidae) de la costa norte col- ombiana, con la descripci6n de dos nuevas es- pecies.—Caldasia 17:569-—585. Campos, E. 1989. Tumidotheres, a new genus for Pin- notheres margarita Smith, 1869, and Pinnoth- eres maculatus Say, 1818 (Brachyura: Pinnoth- eridae).—Journal of Crustacean Biology 9:672— 679. . 1996. Partial revision of pinnotherid crab gen- era with a two-segmented palp on the third maxilliped (Decapoda: Brachyura).—Journal of Crustacean Biology 16:556—563. Chace, FE A., Jr. 1984. The caridean shrimps (Crusta- cea: Decapoda) of the Albatross Philippine Ex- pedition, 1907—1910, Part 2: Families Glypho- crangonidae and Crangonidae.—Smithsonian Contributions to Zoology 397:1—63. . 1985. The caridean shrimps (Crustacea: De- capoda) of the A/batross Philippine Expedition, 1907-1910, Part 3: Families Thalassocarididae and Pandalidae.—Smithsonian Contributions to Zoology 411:1—143. . 1988. The caridean shrimps (Crustacea: De- capoda) of the Albatross Philippine Expedition, 1907-1910, Part 5: Family Alpheidae.—Smith- sonian Contributions to Zoology 466:1—99. . 1992. On the classification of the Caridea (De- capoda).—Crustaceana 63:70-—80. . 1997. The caridean shrimps (Crustacea: De- capoda) of the Albatross Philippine Expedition, 1907-1910, Part 7: Families Atyidae, Eugona- tonotidae, Rhynchocinetidae, Bathypalaemonel- lidae, Processidae, and Hippolytidae.—Smith- sonian Contributions to Zoology 587:1—106. Chan, T-Y, & H-P Yu. 1991. Two similar species: Ple- stonika edwardsii (Brandt, 1851) and Plesionika crosnieri, new species (Crustacea: Decapoda: Pandalidae).—Proceedings of the Biological Society of Washington 104:545—555. Chiong, W. L., & P. K. L. Ng. 1998. A revision of the buckler crabs of the genus Cryptopodia H. Milne Edwards, 1834 (Crustacea: Decapoda: Brachyura: Parthenopidae).—The Raffles Bul- letin of Zoology 46:157—216. Christoffersen, M. L. 1984. The western Atlantic snap- ping shrimps related to Alpheus heterochaelis Say (Crustacea, Caridea), with the description of a new species.—Papéis Avulsos de Zoologia 35:189—208. . 1988. Genealogy and phylogenetic classifi- cation of the world Crangonidae (Crustacea, Caridea), with a new species and new records for the south western Atlantic.—Revista Nor- destina de Biologia 6:43—59. . 1989. Phylogeny and classification of the Pan- daloidea (Crustacea, Caridea).—Cladistics 5: 259-274. . 1998. Malacostraca. Eucarida. Caridea. Cran- gonoidea and Alpheoidea (except Glyphocran- gonidae and Crangonidae). Pp. 351—372 in P. S. Young, ed., Catalogue of Crustacea of Brazil. Museu Nacional, Rio de Janeiro. (Série Livros n. 6). Clark, P. F, & W. R. Webber. 1991. A redescription of Macrocheira kaempferi (TYemminck, 1836) zo- eas with a discussion of the classification of the Majoidea Samouelle, 1819 (Crustacea: Brach- yura).—Journal of Natural History 25:1259— 27S? Coelho, P. A., & M. Ramos-Porto. 1998a. Malacostra- ca. Eucarida. Stenopodidea. Pp. 323-324 in P. S. Young, ed., Catalogue of Crustacea of Brazil. VOLUME 116, NUMBER 1 Museu Nacional, Rio de Janeiro. (Série Livros TsO): PSG 1998b. Malacostraca—Eucarida. Palinuridea. Pp. 387—392 in P. S. Young, ed., Catalogue of Crustacea of Brazil. Museu Na- cional, Rio de Janeiro. (Série Livros n. 6). Cofer-Shabica, S. V., & R. B. Nielsen, Jr. 1988. First record of an American lobster, Homarus amer- icanus, from off southern Florida.—Bulletin of Marine Science 43:315-—317. Cohen, A. N., J. T. Carlton, & M. C. Fountain. 1995. Introduction, dispersal and potential impacts of the green crab Carcinus maenas in San Fran- cisco Bay, California.—Marine Biology 122: 225-237. Coutiere, H. 1905. Les Alpheidae. Pp. 852-921, figs. 127-139, pls. 70—87 in J. S. Gardiner, ed., The fauna and geography of the Maldive and Lac- cadive Archipelagoes, vol. 2, part 4, Cambridge University Press, Cambridge, England. Couture, R., & P. Trudel. 1968. Les crevettes des eaux cotiéres du Québec, Taxonomie et distribu- tion.—Le Naturaliste Canadien 95:857—885. Criales, M. M. 1992. Redescription of the hippolytid shrimp Trachycaris rugosa (Bate) (Crustacea: Decapoda: Caridea) from the western Atlantic, with notes on sexual dimorphism.—Proceed- ings of the Biological Society of Washington 105:562—570. , & R. Lemaitre. 1997. Discovery and morpho- logical variation of the caridean shrimp Discias atlanticus Gurney, 1939 (Crustacea: Decapoda: Bresiliidae) associated with the polychaete worm Chaetopterus variopedatus (Renier, 1804).—Caribbean Journal of Science 33:246— 25%). Crosnier, A., & J. Forest. 1973. Les crevettes profon- des de |l’Atlantique oriental tropical—Faune Tropicale 19:1—409. Dall, W., B. J. Hill, PR. C. Rothlisberg, & D. J. Sharples. 1990. The biology of the Penaeidae.—Advanc- es in Marine Biology 27:1—489. Dardeau, M. R., & R. W. Heard, Jr. 1983. Crangonid shrimps (Crustacea: Caridea), with a description of a new species of Pontocaris.—Memoirs of the Hourglass Cruises 6(2):1—39. D’Incao, FE 1998. Malacostraca. Eucarida. Dendrobran- chiata. Pp. 311—321 in P. S. Young, ed., Cata- logue of Crustacea of Brazil. Museu Nacional, Rio de Janeiro. (Série Livros n. 6). Drach, P., & D. Guinot. 1983. Les Inachoididae Dana, famille de Majoidea caractérisée par des con- nexions morphologiques d’un type nouveau en- tre carapace, pleurites, sternites, pléon.—Comp- tes rendus hebdomadaires des Séances de l’ Académie des Sciences 297:37—42. Engle, V. D., & J. K. Summers. 1999. Latitudinal gra- dients in benthic community composition in ily Western Atlantic estuaries.—Journal of Bioge- ography 26:1007—1023. Felder, D. L., & J. L. Staton. 1994. Genetic differen- tiation in trans-Floridian species complexes of Sesarma and Uca (Decapoda: Brachyura).— Journal of Crustacean Biology 14:191—209. Fransen, C. H. J. M. 2000. Pontonia manningi, new species, a bivalve-associated shrimp from the tropical and subtropical Atlantic (Decapoda: Pontoniinae).—Journal of Crustacean Biology 20 (Special no. 2):101—108. Galil, B. S. 2000. Acanthilia, a new genus of leuco- sioid crabs (Crustacea: Brachyura) from the At- lantic coast of the Americas.—Proceedings of the Biological Society of Washington 113:426— 430. , & P. FE Clark. 1996. A revision of Cryptosoma Brullé, 1837 and Cycloes de Haan, 1837 (Crus- tacea: Brachyura: Calappidae).—Zoological Journal of the Linnean Society 117:175—204. Garcia-Gomez, J. 1982. The Provenzanoi group of her- mit crabs (Crustacea, Decapoda, Paguridae) in the western Atlantic. Part 1. Pagurus maclaugh- linae, a new species.—Bulletin of Marine Sci- ence 32:647—655. . 1983. Revision of Jridopagurus (Crustacea: Decapoda: Paguridae) with descriptions of new species from American waters.—Bulletin of Marine Science 33:10—54. Gibbes, L. R. 1850. On the carcinological collections of the United States, and an enumeration of the most remarkable, and descriptions of new spe- cies.—Proceedings of the American Associa- tion for Advancement of Science 3:167—201. Goeke, G. D. 1980. Range extensions of six western Atlantic frog crabs (Brachyura: Gymnopleura: Raninidae) with notes on the taxonomic status of Lyreidus bairdii.imProceedings of the Bio- logical Society of Washington 93:145—152. . 1985. Decapod Crustacea: Raninidae.—Mé- moire du Muséum National de |’ Histoire Na- turelle (Séries A, Zoologie) 133:205—228. . 1989. Stenorhynchus yangi, a new western Atlantic species of arrow crab (Crustacea: Brachyura, Majidae) and a redescription of S. seticornis (Herbst, 1788).—Proceedings of the Biological Society of Washington 102:620— 636. Griffin, D. J. G., & H. A. Tranter. 1986. The Decapoda Brachyura of the Siboga Expedition. Part VIII. Mayjidae. Siboga-Expedition Monograph 39 (C4), 148:1—335. Griffith, H. 1987. Phylogenetic relationships and evo- lution in the genus Dissodactylus Smith, 1870 (Crustacea: Brachyura: Pinnotheridae).—Cana- dian Journal of Zoology 65:2292—2310. Grosholz, E. D., & G. M. Ruiz. 1996. Predicting the impact of introduced marine species: lessons 152 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON from the multiple invasions of the European green crab Carcinus maenas.—Biological Con- servation 78:59—66. Guinot, D. 1978. Principes d’une classification évolu- tive des Crustacés Décapodes Brachyoures.— Bulletin Biologique de la France et de la Bel- gique 112:211—292. . 1993. Données nouvelles sur les Raninoidea de Haan, 1841 (Crustacea Decapoda Brachyura Podotremata).—Comptes-rendus hebdomadai- res des Séances de l’Académie des Sciences (Paris), Series HI 316:1324—1331. , & J.-M. Bouchard. 1998. Evolution of the ab- dominal holding systems of brachyuran crabs (Crustacea, Decapoda, Brachyura).—Zoosyste- ma 20:613—694. , & B. Richer de Forges. 1995. Crustacea De- capoda Brachyura: révision de la famille des Homolidae de Haan, 1839.—Mémoires du Mu- séum National d’Histoire Naturelle 163:283— ST, , & . 1997. Affinités entre les Hyme- nosomatidae MacLeay, 1838 et les Inachoididae Dana, 1851 (Crustacea, Decapoda, Brachyu- ra).—Zoosystema 19:453—502. Gusmao, J., C. Lazoski, & A. M. Solé-Cava. 2000. A new species of Penaeus (Crustacea: Penaeidae) revealed by allozyme and cytochrome oxidase 1 analysis.—Marine Biology 137:435—446. Heard, R. W., & R. B. Manning. 1997. Austinixa, a new genus of pinnotherid crab (Crustacea: De- capoda: Brachyura), with the description of A. hardyi, a new species from Tobago, West In- dies.—Proceedings of the Biological Society of Washington 110:393—-398. , and . 1998. A new genus and species of ghost shrimp (Crustacea: Decapoda: Calli- anassidae) from the Atlantic Ocean.—Proceed- ings of the Biological Society of Washington 111:883—-888. Hendrickx, M. E. 1989. Notes on the genus Ethusa Roux, 1828, and description of Ethusa steyaerti n. sp. (Crustacea, Decapoda, Dorippidae), from the continental shelf of the Gulf of California, Mexico.—Bulletin du Muséum National d’ Histoire Naturelle, Paris, 4° séries, 11, Section A, n° 2:407—423. . 1995. Checklist of brachyuran crabs (Crusta- cea: Decapoda) from the eastern tropical Pacif- ic.—Bulletin de l’Institut Royal des Sciences Naturelles de Belgique, Biologie 65:125—150. . 1998. A new genus and species of “gonepla- cid-like” brachyuran crab (Crustacea: Decapo- da) from the Gulf of California, Mexico, and a proposal for the use of the family Pseudorhom- bilidae Alcock, 1900.—Proceedings of the Bi- ological Society of Washington 111:634—644. Holthuis, L. B. 1947. The Decapoda of the Siboga-Ex- pedition. Part IX. The Hippolytidae and Rhyn- chocinetidae collected by the Siboga and Snellius Expeditions with remarks on other species.—Si- boga-Expeditie, Mongraphie 39a°, 100 p. 1991. FAO species catalogue. Volume 13. Marine lobsters of the world.—FAO Fisheries Synopsis No. 125, Volume 13:1—292. . 1993. The recent genera of the caridean and stenopodidean shrimps (Crustacea, Decapoda) with an appendix on the order Amphionidacea. Nationaal Natuurhistorisch Museum, Leiden. 328 pp. , & R. B. Manning. 1987. Hypoconcha paras- itica (Linnaeus, 1763), a senior synonym of Hy- poconcha sabulosa (Herbst, 1799) (Crustacea: Decapoda: Brachyura).—Proceedings of the Bi- ological Society of Washington 100:1018—1022. Kensley, B. 1983. New records of bresiliid shrimp from Australia, South Africa, Caribbean, and Gulf of Mexico (Decapoda: Natantia: Cari- dea).—Smithsonian Contributions to Zoology 394:1-31. . 1989. New genera in the thalassinidean fam- ilies Calocarididae and Axiidae (Crustacea: De- capoda).—Proceedings of the Biological Soci- ety of Washington 102:960—967. . 2001. Two sympatric species of Axius from the North-West Atlantic (Decapoda, Thalassin- idea, Axiidae).—Crustaceana 74:95 1—962. Kim, W., & L. Abele. 1988. The snapping shrimp ge- nus Alpheus from the eastern Pacific (Decapo- da: Caridea: Alpheidae).—Smithsonian Contri- butions to Zoology 454:1—119. Knott, D., C. Boyko, & A. Harvey. 2000. Introduction of the green porcelain crab, Petrolisthes arma- tus (Gibbes, 1850) into the South Atlantic Bight. P. 404 (Abstract only) in J. Pederson, ed., Marine bioinvasions: Proceedings of the First National Conference. MIT, Massachusetts. Komai, T. 1997. Revision of Argis dentata and related species (Decapoda: Caridea: Crangonidae), with description of a new species from the Okhotsk Sea.—Journal of Crustacean Biology 17:135— 161. 1999. A revision of the genus Pandalus (Crustacea: Decapoda: Caridea: Pandalidae).— Journal of Natural History 33:1265—1372. Laughlin, R. A., P. J. Rodriguez, & J. A. Marval. 1982. The complete larval development of the sponge crab Dromia erythropus (George Edwards, 1771) (Brachyura: Dromiidae) from the Archi- piélago de los Roques, Venezuela.—Journal of Crustacean Biology 2:342-359. Ledesma, M. E., & N. J. O'Connor. 2001. Habitat and diet of the non-native crab Hemigrapsus san- guineus in southeastern New England.—North- eastern Naturalist 8:63—78. Lemaitre, R. 1989. Revision of the genus Parapagurus VOLUME 116, NUMBER 1 (Anomura: Paguroidea: Parapaguridae), includ- ing redescriptions of the western Atlantic spe- cies.—Zoologische Verhandelingen, Leiden 253:1—106. , N. H. Campos, & A. Bermtidez. 2001. A new species of Pyromaia from the Caribbean Sea, with a redescription of P. propinqua Chace, 1940 (Decapoda: Brachyura: Majoidea: Ina- choididae).—Journal of Crustacean Biology 21: 760-773. , & P. A. McLaughlin. 1996. Revision of Py- lopagurus and Tomopagurus (Crustacea: De- capoda: Paguridae), with the descriptions of new genera and species. Part V. Anisopagurus McLaughlin, Manucomplanus McLaughlin, and Protoniopagurus new genus.—Bulletin of Ma- rine Science 59:89-141. , & J. Garcia-Gomez. 1982. The Prov- enzanoi group of hermit crabs (Crustacea, De- capoda, Paguridae) in the western Atlantic. Part IV. A review of the group with notes on varia- tions and abnormalities.—Bulletin of Marine Science 32:670—701. Lohrer, A. M., & R. B. Whitlatch. 1997. Ecological studies on the recently introduced Japanese shore crab (Hemigrapsus sanguineus), in east- ern Long Island Sound. Pp. 49—60 in N. C. Bal- com, ed., Proceedings of the Second Northeast Conference on Nonindigenous Aquatic Nui- sance Species. Connecticut Sea Grant College Program, Connecticut.— Lyons, W. 1970. Scyllarid lobsters (Crustacea, Deca- poda).—Memoirs of the Hourglass Cruises 1(4): 1-74. Maggioni, R., A. D. Rogers, N. Maclean, & E D’Incao. 2001. Molecular phylogeny of western Atlantic Farfantepenaeus and Litopenaeus shrimp based on mitochondrial 16S partial se- quences.—Molecular Phylogenetics and Evo- lution 18:66—73. Manning, R. B. 1987. Notes on western Atlantic Cal- lianassidae (Crustacea: Decapoda: Thalassini- dea).—Proceedings of the Biological Society of Washington 100:386—401. . 1988. Notes on albuneid crabs (Crustacea: Decapoda: Albuneidae) from the central east coast of Florida.—Proceedings of the Biological Society of Washington 101:626—632. . 1993. Three genera removed from the syn- onymy of Pinnotheres Bosc, 1802 (Brachyura: Pinnotheridae).—Proceedings of the Biological Society of Washington 106:523-531. , & D. L. Felder. 1986. The status of the cal- lianassid genus Callichirus Stimpson, 1866 (Crustacea: Decapoda: Thalassinidea).—Pro- ceedings of the Biological Society of Washing- ton 99:437—443. , & . 1989. The Pinnixa cristata com- is6) plex in the western Atlantic, with a description of two new species (Crustacea: Decapoda: Pin- notheridae).—Smithsonian Contributions to Zo- ology 473:1—26. (Se . 1991. Revision of the American Callianassidae (Crustacea: Decapoda: Thalas- sinidea).—Proceedings of the Biological Soci- ety of Washington 104:764—792. OG . 1992 [1991]. Gilvossius, a new genus of callianassid shrimp from the eastern United States (Crustacea: Decapoda: Thalassin- idea).—Bulletin of Marine Science 49:558—561. , & L. B. Holthuis. 1981. West African brach- yuran crabs (Crustacea: Decapoda).—Smithson- ian Contributions to Zoology 306:1—379. RUSE . 1989. Two new genera and nine new species of geryonid crabs (Crustacea: De- capoda: Geryonidae).—Proceedings of the Bi- ological Society of Washington 102:50-—77. Maris, R. C. 1986. Larvae of Xiphopenaeus kroyeri (Heller, 1862) (Crustacea: Decapoda: Penaei- dae) from offshore waters of Virginia, U.S.A.— Proceedings of the Biological Society of Wash- ington 99:602—603. Marques, F, & G. Pohle. 1995. Phylogenetic analysis of the Pinnotheridae (Crustacea, Brachyura) based on larval morphology, with emphasis on the Dissodactylus species complex.—Zoologica Scripta 24:347—364. Martin, J. W., & L. G. Abele. 1986. Notes on male pleopod morphology in the brachyuran crab family Panopeidae Ortmann, 1893, sensu Guin- ot (1978) (Decapoda).—Crustaceana 50:182— 198. , & G. E. Davis. 2001. An updated classifica- tion of the Recent Crustacea.—Natural History Museum of Los Angeles County, Science Series 39:1-124. , & T. L. Zimmerman. 2001. Notes on three western Atlantic deep-sea crabs (Decapoda, Brachyura).—Crustaceana 74:41 1—414. McClure, M. R. 1995. Alpheus angulatus, a new spe- cies of snapping shrimp from the Gulf of Mex- ico and northwestern Atlantic, with a redescrip- tion of A. heterochaelis Say, 1818 (Decapoda: Caridea: Alpheidae).—Proceedings of the Bio- logical Society of Washington 108:84—97. . 2002. Revised nomenclature of Alpheus an- gulatus McClure, 1995 (Decapoda: Caridea: Al- pheidae).—Proceedings of the Biological Soci- ety of Washington 115:368—370. , & M. K. Wicksten. 1997. Morphological var- iation of species of the edwardsii group of Al- pheus in the northern Gulf of Mexico and north- western Atlantic (Decapoda: Caridea: Alphei- dae).—Journal of Crustacean Biology 17:480-— 487. McDermott, J. J. 1998. The western Pacific brachyuran 154 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Hemigrapsus sanguineus: Grapsidae), in its new habitat along the Atlantic coast of the Unit- ed States: geographic distribution and ecolo- gy.—ICES Journal of Marine Science 55:289-— 298. . 2000. Natural history and biology of the Asian shore crab Hemigrapsus sanguineus in the western Atlantic: a review, with new infor- mation. Pp. 193—199 in J. Pederson, ed., Marine bioinvasions: Proceedings of the First National Conference, January 24—27, 1999. MIT, MIT Sea Grant College Program, Massachusetts. McLaughlin, P. A. 1981. Revision of Pylopagurus and Tomopagurus (Crustacea: Decapoda: Paguri- dae), with the descriptions of new genera and species. Part I. Ten new genera of the Paguridae and a redescription of Tomopagurus A. Milne Edwards and Bouvier.—Bulletin of Marine Sci- ence 31:1—30. . 1983. Hermit crabs—are they really polyphy- letic?—Journal of Crustacean Biology 3:608-— 621. . 1988. The rediscovery of Ceratopagurus Yo- koya and a new genus for Pagurus piercei Wass (Decapoda, Paguroidea, Paguridae).—Crusta- ceana 55:257—267. , & J. Haig. 1995. A new species of Goreo- pagurus McLaughlin (Decapoda: Anomura: Pa- guridae) from the Pacific, and a comparison with its Atlantic counterpart.—Proceedings of the Biological Society of Washington 108:68— WD- , & R. Lemaitre. 2001la. A new family for a new genus and new species of hermit crab of the superfamily Paguroidea (Decapoda: Ano- mura) and its phylogenetic implications.—Jour- nal of Crustacean Biology 21:1062—1076. , & . 2001b. Revision of Pylopagurus and Tomopagurus (Crustacea: Decapoda: Pa- guridae), with descriptions of new genera and species. Part VI. Pylopagurus A. Milne-Ed- wards & Bouvier, 1891, Haigia McLaughlin, 1981, and Pylopaguridium, a new genus.—Pro- ceedings of the Biological Society of Washing- ton 114:444—483. McLay, C. L. 1993. Crustacea Decapoda: The sponge crabs (Dromiidae) of New Caledonia and the Philippines with a review of the genera.—Mé- moires du Muséum National d’Histoire Natu- relle 156:111—251. Melo, G. A. S. 1998. Malacostraca-Eucarida. Brach- yura. Oxyrhyncha and Brachyrhyncha. Pp. 455— 515 in P. S. Young, ed., Catalogue of Crustacea of Brazil. Museu Nacional, Rio de Janeiro. (Sér- te Livros m6): , M. E A. Torres, & O. Campos, Jr. 1998. Ma- lacostraca-Eucarida. Brachyura. Dromiacea and Oxystomata. Pp. 439—454 in P. S. Young, ed., Catalogue of Crustacea of Brazil. Museu Na- cional, Rio de Janeiro. (Série Livros n. 6). Melo-Filho, G. A. S. 1998. Malacostraca-Eucarida. Chyrostilidae and Galatheidae. Pp. 393-397 in P. S. Young, ed., Catalogue of Crustacea of Bra- zil. Museu Nacional, Rio de Janeiro. (Série Liv- ros n. 6). , & G. A. S. Melo. 1992. Reexamination of the material of Munida Leach (Crustacea: Anomu- ra: Galatheidae) collected by the H.M.S. Chal- lenger (1872-1876) along the Brazilian coast.—Proceedings of the Biological Society of Washington 105:760—774. Moriyasu, M., H. J. Squires, R. Campbell, & K. Ben- halima. 2001. Northern range extensions for two decapod crustaceans, the inflated spiny crab Rochinia crassa (A. Milne-Edwards, 1879) and the purplehead gamba prawn Aristeus antillen- sis A. Milne-Edwards & Bouvier, 1909, in the northwestern Atlantic.—Crustaceana 74:255— 260. Nations, J. D. 1975. The genus Cancer (Crustacea: Brachyura): systematics, biogeography and fos- sil record.—Natural History Museum of Los Angeles County Science Bulletin 23:1—104. Ng, P. K. L., & G. Rodriguez. 1986. New records of Mimilambrus wileyi Williams, 1979 (Crustacea: Decapoda: Brachyura), with notes on the sys- tematics of the Mimilambridae Williams, 1979, and Parthenopidae MacLeay, 1838, sensu Guin- ot, 1978.—Proceedings of the Biological Soci- ety of Washington 99:88—99. Niem, V. H. 1996. Phylogenetic relationships among American species of Sesarma (subgenus Ar- mases) (Brachyura, Grapsidae).—Crustaceana 69:330-348. Farfante, I. 1977a. American solenocerid shrimps of the genera Hymenopenaeus, Hali- poroides, Pleoticus, Hadropenaeus new genus, and Mesopenaeus new genus.—Fishery Bulle- tin 75:261—346. . 1977b. Range extensions of the shrimps So- Pérez lenocera necopina Burkenroad and Parapen- aeus americanus Rathbun (Crustacea, Decapo- da, Penaeoidea).—Proceedings of the Biologi- cal Society of Washington 90:597—-599. . 1980. Revision of the penaeid shrimp genus Penaeopsis (Crustacea: Decapoda).—Fishery Bulletin 77:721—763. . 1988. Illustrated key to penaeoid shrimps of commerce in the Americas.—NOAA Technical Report, National Marine Fisheries Service, 64: 1-32. , & B. FE Kensley. 1997. Penaeoid and serges- toid shrimps and prawns of the world. Keys and diagnoses for the families and genera.—Mé- moires du Muséum National d’Histoire Natu- relle 175:1—233. VOLUME 116, NUMBER 1 Pipitone, C., EF Badalamenti, & A. Sparrow. 2001. Contribution to the knowledge of Percnon gib- besi (Decapoda, Grapsidae), an exotic species spreading rapidly in Sicilian waters.—Crusta- ceana 74:1009—1017. Pohle, G., & E Marques. 2000. Larval stages of Par- adasygyius depressus (Bell, 1835) (Crustacea: Decapoda: Brachyura: Majidae) and a phylo- genetic analysis for 21 genera of Majidae.— Proceedings of the Biological Society of Wash- ington 113:739—760. Poore, G. C. B. 1994. A phylogeny of the families of Thalassinidea (Crustacea: Decapoda) with keys to families and genera.—Memoirs of the Mu- seum of Victoria 54:79—120. Powers, L. W. 1977. A catalogue & bibliography to the crabs (Brachyura) of the Gulf of Mexico.— Contributions in Marine Science 20 (supple- ment):1—190. Ramos-Porto, M., & P. A. Coelho. 1998. Malacostraca. Eucarida. Caridea (Alpheoidea excluded). Pp. 325-350 in P. S. Young, ed., Catalogue of Crus- tacea in Brazil. Museu Nacional, Rio de Janeiro. (Série Livros n. 6). Rathbun, M. J. 1925. The spider crabs of America.— United States National Museum Bulletin 129: 1-613. . 1930. The cancroid crabs of America of the families Euryalidae, Portunidae, Atelecyclidae, Cancridae and Xanthidae.—United States Na- tional Museum Bulletin 152:1—609. Rieger, P. J. 1998. Malacostraca—Eucarida. Paguroidea. Pp. 413—429 in P. S. Young, ed., Catalogue of Crustacea in Brazil. Museu Nacional, Rio de Janeiro. (Série Livros n. 6). Rodrigues, S. de A., & R. M. Shimizu. 1998. Mala- costraca—Eucarida. Thalassinidea. Pp. 379-385 in P. S. Young, ed., Catalogue of Crustacea in Brazil. Museu Nacional, Rio de Janeiro. (Série Livros n. 6). Rodriguez, B. 1993. Shrimps. Pp. 107—131 in K. Car- penter, W. Fischer, and L. Garibaldi, eds., Field guide to the commercial marine and brackish- water resources of the northern coast of South America. FAO, Rome. Rodriguez-Almaraz, G., A. Leija-Tristan, & R. Men- doza. 2000. Records of caridean shrimps (Crus- tacea: Decapoda) from the coasts of the Mexi- can Pacific Ocean, Gulf of Mexico and Mexican Caribbean.—Bulletin of Marine Science 67: 857-867. Rosenberg, M. S. 2001. The systematics and taxonomy of fiddler crabs: A phylogeny of the genus Uca.—Journal of Crustacean Biology 21:839-— 869. Sakai, K. 1999. Synopsis of the family Callianassidae, with keys to subfamilies, genera and species, and the description of new taxa (Crustacea: De- 5)5) capoda: Thalassinidea).—Zoologische Verhan- delingen, Leiden 326:1—152. , & M. de Saint Laurent. 1989. A check list of Axiidae (Decapoda, Crustacea, Thalassinidea, Anomura), with remarks and in addition de- scriptions of one new subfamily, eleven new genera and two new species.—Naturalists 3:1— 104. Sarver, S. K., D. W. Freshwater, & P. J. Walsh. 2000. The occurrence of the provisional Brazilian sub- species of spiny lobster (Panulirus argus wes- tonit) in Florida waters.—Fishery Bulletin 98: 870-873. , J. D. Silberman, & P. J. Walsh. 1998. Mito- chondrial DNA sequence evidence supporting recognition of two subspecies or species of the Florida spiny lobster Panulirus argus.—Journal of Crustacean Biology 18:177—186. Scelzo, M. A. 2001. First record of the portunid crab Arenaeus cribrarius (Lamarck, 1818) (Crusta- cea: Brachyura: Portunidae) in marine waters of Argentina.—Proceedings of the Biological So- ciety of Washington 114:605—610. Schubart, C. D., J. A. Cuesta, R. Diesel, & D. L. Feld- er. 2000. Molecular phylogeny, taxonomy, and evolution of nonmarine lineages within the American grapsoid crabs (Crustacea: Brachyu- ra).—Molecular Phylogenetics and Evolution 15:179-190. , J. E. Neigel, & D. L. Felder. 2000. Molecular phylogeny of mud crabs (Brachyura: Panopei- dae) from the northwestern Atlantic and the role of morphological stasis and convergence.—Ma- rine Biology 137:11-18. Schweitzer, C. E., & R. M. Feldmann. 2000. New spe- cies of calappid crabs from western North America and reconsideration of the Calappidae sensu lato.—Journal of Paleontology 74:230— 246. Silberman, J. D., S. K. Sarver, & P. J. Walsh. 1994. Mitochondrial DNA variation and population structure in the spiny lobster Panulirus argus.— Marine Biology 120:601—608. Squires, H. J. 1990. Decapod Crustacea of the Atlantic coast of Canada.—Canadian Bulletin of Fish- eries and Aquatic Sciences 221:1—532. . 1992. Recognition of Pandalus eous Maka- rov, 1935, as a Pacific species not a variety of the Atlantic Pandalus borealis Kr@yer, 1838 (Decapoda, Caridea).—Crustaceana 63:257— 262. Stehlik, L. L., C. L. MacKenzie, Jr, & W. W. Morse. 1991. Distribution and abundance of four brach- yuran crabs on the northwest Atlantic shelf.— Fishery Bulletin 89:473—492. Sternberg, R. von, & N. Cumberlidge. 1998. Taxic re- lationships within the Grapsidae MacLeay, 156 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1838 (Crustacea: Decapoda: Eubrachyura).— Journal of Comparative Biology 3:115—136. Strasser, K. M., & W. W. Price. 1999. An annotated checklist and key to hermit crabs of Tampa Bay, Florida, and surrounding waters.—Gulf Re- search Reports 11:35—50. Strenth, No Ed EAS Chace; In 995. New mecords of natant decapods (Crustacea: Palaemonidae) from the south Texas coast.—Texas Journal of Science 47:315-—318. Tavares, M. 1991a [1990]. Especes nouvelles de Cy- clodorippoidea Ortmann et remarques sur les genres 7ymolus Stimpson et Cyclodorippe A. Milne Edwards (Crustacea, Decapoda, Brach- yura).—Bulletin du Muséum _ National d’ Histoire Naturelle, Paris, 4° séries, 12, Section A, n°> 3—4:623—648. . 1991b. Redéfinition des genres Rochinia A. Milne Edwards, Sphenocarcinus A. Milne Ed- wards et Oxypleurodon Miers, et établissement du genre Nasutocarcinus gen. nov. (Crustacea, Brachyura, Majidae).—Bulletin du Muséum National d’Histoire Naturelle, Paris, 4° séries, 13, Section A, n° 1—2:159-179. . 1993. Description préliminaire de quatre nou- veaux genres et trois nouvelles especes de Cy- clodorippoidea Améericains (Crustacea, Deca- poda, Brachyura).—Vie Milieu 43:137-144. . 1996. Révision systématique des Cyclodor- ippidae ameéricains (Crustacea, Decapoda, Brachyura).—Bulletin du Muséum National d’ Histoire Naturelle, Paris, 4° séries, 18, Section A, n° 1—2:233-—295. Tucker, A. B. 1998. Systematics of the Raninidae (Crustacea: Decapoda: Brachyura), with ac- counts of three new genera and two new spe- cies.—Proceedings of the Biological Society of Washington 111:320—371. Tudge, C. C., G. C. B. Poore, & R. Lemaitre. 2000. Preliminary phylogenetic analysis of generic re- lationships within the Callianassidae and Cten- ochelidae (Decapoda: Thalassinidea: Callianas- soidea).—Journal of Crustacean Biology 20 (Special no. 2):129—149. Udekem d’Acoz, C. d’. 1997. Redescription of Hip- polyte obliquimanus Dana, 1852, and compari- son with Hippolyte williamsi Schmitt, 1924 (Decapoda, Caridea).—Crustaceana 70:469— 479. Vargas, R., & J. Cortés. 1999. Biodiversidad marina de Costa Rica: Crustacea: Decapoda (Penaeo- idea, Sergestoidea, Stenopodidea, Caridea, Thalassinidea, Palinura) del Caribe.—Revista de Biologia Tropical 47:877—885. Veloso, V. G. 1998. Malacostraca—Eucarida. Porcellan- idae. Pp. 399—405 in P. S. Young, ed. Catalogue of Crustacea in Brazil. Museu Nacional, Rio de Janeiro. (Série Livros n. 6). Wagener, H. P. 1990. The genera Mithrax Latreille, 1818 and Mithraculus White, 1847 (Crustacea: Brachyura: Majidae) in the western Atlantic Ocean.—Zoologische Verhandelingen (Leiden) 264:1—65. Wass, M. L. 1963. New species of hermit crabs (De- capoda, Paguridae) from the western Atlan- tic.—Crustaceana 6:133—157. Wigley, R. L., R. B. Theroux, & H. E. Murray. 1975. Deep-sea red crab, Geryon quinquedens, survey off northeastern United States——Marine Fish- eries Review 37:1—21. Williams, A. B. 1965. Marine decapod crustaceans of the Carolinas.—Fishery Bulletin 65:1—298. . 1983. The mud crab, Panopeus herbstii, s.1. partition into six species (Decapoda: Xanthi- dae).—Fishery Bulletin 81:863—882. . 1984. Shrimps, lobsters, and crabs of the At- lantic coast of the eastern United States, Maine to Florida. Smithsonian Institution Press, Wash- ington, D.C., 550 pp. . 1988. Notes on decapod and euphausiid crus- taceans, continental margin, western Atlantic, Georges Bank to western Florida, USA.—Fish- ery Bulletin 86:67—76. . 1993. Mud shrimps, Upogebiidae, from the western Atlantic (Crustacea: Decapoda: Thal- assinidea).—Smithsonian Contributions to Zo- ology 544:1-77. , & C. A. Child. 1989. Comparison of some genera and species of box crabs (Brachyura: Calappidae), southwestern North Atlantic, with description of a new genus and species.—Fish- ery Bulletin 87:105—121. , & D.L. Felder. 1986. Analysis of stone crabs: Menippe mercenaria (Say), restricted, and a previously unrecognized species described (De- capoda: Xanthidae).—Proceedings of the Bio- logical Society of Washington 99:517—543. , & J. J. McDermott. 1990. An eastern United States record for the western Indo-Pacific crab, Hemigrapsus sanguineus (Crustacea: Decapo- da: Grapsidae).—Proceedings of the Biological Society of Washington 103:108—109. , & R. Wahle. 1992. Distinguishing juvenile stages of Jonah and Atlantic rock crabs, Cancer borealis and C. irroratus (Decapoda: Cancri- dae).—Journal of Crustacean Biology 12:464— 466. , & R. L. Wigley. 1977. Distribution of deca- pod Crustacea off northeastern United States based on specimens at the Northeast Fisheries Center, Woods Hole, Massachusetts—-NOAA Technical Report NMFS Circular 407:1—44. Note Added in Proof Recent revision of the Albuneidae (Boy- ko 2002) provided new information for three species included in the present check- VOLUME 116, NUMBER I list. Western Atlantic specimens identified as Albunea paretii Guérin-Méneville, 1853 in Williams (1984) included two species, A. paretii and A. catherinae Boyko, 2002. Of these, only A. catherinae occurs in the re- gion (see account below). Albunea paretii ranges from the Florida Keys southward through the Caribbean to Rio Grande do Sul, Brazil, and should be removed from the checklist. According to Boyko (2002), all eastern Atlantic references to A. paretii refer to A. elegans. Boyko (2002) included additional distributional information for Lepidopa websteri that extends this species range to Texas. Based on new data for A. paretii and A. catherinae, 52 species are new to the checklist. Thirteen of these have been dis- covered and described since publication of Williams (1984), and seven species previ- ously recorded from the region are no lon- oi ger considered part of the regional decapod assemblage. Family Albuneidae Stimpson, 1858 > Albunea catherinae Boyko, 2002 Remarks.—Boyko (2002:343) provided a description, diagnosis, illustrations, and size information for this species. Albunea cath- erinae occurs at depths less than 64 m. Known~ range.—Western Atlantic: Vir- ginia to Palm Beach County, Florida; Col- ler County, Florida, through the Gulf of Mexico, to southern Texas; absent from the Florida Keys (Boyko 2002). Literature Cited Boyko, C. B. 2002. A worldwide revision of the recent and fossil sand crabs of the Albuneidae Stimp- son and Blepharipodidae, new family (Crusta- cea: Decapoda: Anomura: Hippoidea).—Bulle- tin of the American Museum of Natural History 272:1—396. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):158—167. 2003. A new caridean shrimp of the family Alvinocarididae from thermal vents at Menez Gwen on the Mid-Atlantic Ridge Timothy M. Shank and Joel W. Martin (TMS) Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, U.S.A., e-mail: tshank @ whoi.edu; (JWM) Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007, U.S.A., e-mail: jmartin@nhm.org Abstract.—A new species of alvinocaridid shrimp, Alvinocaris williamsi, is described from the Menez Gwen hydrothermal vent field on the Mid-Atlantic Ridge. The new species is most similar to another relatively shallow water alvinocaridid, A. stactophila Williams, known only from a cold seep area off the coast of Louisiana, but differs in details of the rostrum and appendages. Alvinocaris williamsi is also morphologically similar to two species of the genus recently described from the Mid-Okinawa Trough, Japan. Molecular phy- logenetic studies of A. williamsi, the eighth described member of the genus, may provide significant insights into the role that deep-sea hydrothermal vents at mid-ocean ridges and hydrocarbon seeps on continental margins have played in the evolution of fauna endemic to these chemosynthetic habitats. The caridean shrimp family Alvinocari- didae was originally proposed by Christof- fersen (1986) to accommodate a number of morphologically similar shrimp known from hydrothermal vents and hydrocarbon seeps. The family was further diagnosed by Segonzac et al. (1993) in a footnote, and also by Vereshchaka (1996a, 1996b, 1997) (see Martin & Davis 2001). To date, five genera have been proposed for the family. In the order in which they were proposed, they are: Alvinocaris Williams & Chace, 1982; Rimicaris Williams & Rona, 1986; Chorocaris Martin & Hessler, 1990; Opae- pele Williams & Dobbs, 1995; and Jorania Vereshchaka, 1996b (see Shank et al. 1998 for the synonymy of /Jorania and Rimicar- is). A sixth genus of vent shrimp, Mirocaris Vereshchaka, 1997, was proposed by Ver- eshchaka (1997) for Chorocaris fortunata Martin & Christiansen (1995) and for a new species (M. keldyshi Vereshchaka, 1997); however, this genus was transferred to the newly created family Mirocarididae by Ver- eshchaka (1997). The type genus of the Alvinocarididae, Alvinocaris, 1s the most diverse genus of the family and contains five hydrothermal vent-endemic species: A. brevitelsonis Kik- uchi & Hashimoto, 2000, from the Minami- Ensei Knoll in the Mid-Okinawa Trough off Japan; A. leurokolos Kikuchi & Hashimoto, 2000, also from the Minami-Ensei Knoll; A. longirostris Kikuchi & Ohta, 1995, from the Iheya Ridge off Japan; A. lusca Wil- liams & Chace, 1982, from the Galapagos Rift in the eastern Pacific; and A. markensis Williams, 1988, from the Mid-Atlantic Ridge. Another two species of Alvinocaris are endemic to chemosynthetic habitats as- sociated with hydrocarbon and brine seeps: A. muricola Williams, 1988, from the West Florida Escarpment; and A. stactophila Wil- liams, 1988, from the Louisiana Slope in the northern Gulf of Mexico. The study of additional collections, including unde- scribed hydrothermal-associated shrimp specimens from the Edison Seamount of the Bismark Archipelago in Papua New Guinea (Shank et al. 1999; K. Baba, pers. comm.), VOLUME 116, NUMBER 1 from the Logatchev vent field of the Mid- Atlantic Ridge (A. Vereshchaka, pers. comm.), and several recently recovered from active seamounts north of the Bay of Plenty (northern North Island, New Zea- land) (Rick Webber and Neil Bruce, pers. comm., http://www.niwa.co.nz/pubs/bu/05/ blind) may yield additional new species of Alvinocaris. The pan-equatorial biogeo- graphic distribution of the genus is note- worthy. Alvinocaris species occur in almost every biogeographic vent province (except the Northeast Pacific vents; Shank et al. 1999). However, recent initial investiga- tions of hydrothermal vents in the Indian Ocean did not reveal the presence of Alvi- nocaris shrimp despite the common ap- pearance of the other dominant Atlantic and Pacific genera Rimicaris and Chorocaris (Hashimoto et al. 2001, Van Dover et al. 2001). The Alvinocaris species described herein occupies the northernmost extent of the generic range, as it inhabits the Menez Gwen vent site, the most shallow vent field (850 m) and the most northern known ac- tive deep-sea hydrothermal site along the Mid-Atlantic Ridge (37°50.5’N, 31°31.3’W) Gebrmuk et ale 1997,. Colaco et al. 1998; Comtet & Desbruyeéres 1998, Desbruyéres et al. 2001). Below, we describe a new spe- cies of Alvinocaris from the Menez Gwen hydrothermal vent field, and suggest evo- lutionary relationships with other alvino- carid species. Materials and Methods All specimens were collected using the human occupied submersible DSV Alvin at the Menez Gwen hydrothermal vent field, northern Atlantic Ocean. Shrimp were col- lected from a cluster of active individuals among clumps of mussels, Bathymodiolus azoricus Von Cosel, Comtet & Krylova, within vent flow with a 30 cm X 30 cm square “black net’’ (BN) operated using the manipulator arm by the pilots of DSV Alvin. A total of 14 specimens were collected (from a single net sample) from areas of most intense diffuse fluid within the central 159 portion of a hydrothermally-active mussel bed community that included a more abun- dant alvinocaridid species, Mirocaris keld- yshi, amphipods, Luckia striki Bellan-San- tint & Thurston, gastropods, e.g., Lepeto- drilus spp. and Protolira valvatoides Waren & Bouchet, polychaetes, e.g., Branchipo- lynoe seepensis Pettibone, and crabs, Se- gonzacia mesatlantica (Williams, 1988) and Chaceon affinis (A. Milne-Edwards & Bouvier, 1894). Specimens were brought to the surface in an insulated container and placed in chilled water on-board ship. Whole shrimp were sorted by morphotype and either preserved in 4% buffered for- malin in seawater and subsequently trans- ferred to 70% ethanol or frozen at —70°C and subsequently transported on dry ice and stored at —80°C. Carapace length (CL) of each individual was measured in millime- ters (mm) from the orbital margin to the posteriomedial margin of the carapace. We recognize that the description of new spe- cies within this family from relatively few individuals can be risky (see Shank et al. 1998); however, preliminary results of mi- tochondrial DNA sequence data (Shank, pers. obs.) complement morphological evi- dence that distinction of this species from other Alvinocaris species 1s warranted. The holotype and paratypes are deposited in the collections of the National Museum of Nat- ural History, Smithsonian Institution, Washington, D.C. (USNM). The remaining specimens are cryo-preserved at the Woods Hole Oceanographic Institution for ongoing molecular phylogenetic investigations by TMS. Alvinocaris williamsi, new species Figs. 1-3 Material examined.—14 females CL: 8.2, SOS DISeMOLOMSIO OBO O1Oy OS29:G) 9.6, 9.7, 10.1, DSV Alvin, Dive 3117, Menez Gwen hydrothermal vent field, North Atlan- tic Ocean, 37°50.5'N, 31°31.3'W, 850 m, 7 Jul 1997. Types.—All from DSV Alvin Dive 3117. 160 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON > >>> so > Ss35°505S5>S5 \ ’ \ ‘ \ ‘ \ \ \ ‘ \ Figs le appendages, and abdomen, lateral view; b, anterior region of carapace, including eyes and antennae, dorsal view; c, telson and uropods, dorsal view. Scale bars equal 1.0 mm (bar | for a, bar 2 for b, bar 3 for c). Holotype: adult female CL 9.5 mm, ovig- erous (specimen ID: BN 70), USNM 1009651. Paratypes: 2 adult females CL 9.3, 9.7 (specimen ID: BN 68 and BN 51, respectively), abdomen torn from body, tail fan missing, USNM 1009652, USNM 1009653. Description.—Integument thin, shining, minutely punctuate. Rostrum (Fig. la, b) al- most straight, imperceptibly elevated above horizontal in distal half; sharply pointed tip (broken or slightly blunted in all speci- mens) reaching at least to level of articu- lation between first and second peduncular segment of second antenna or sometimes to 2 = S AS Alvinocaris williamsi, new species, holotype ovigerous female, USNM 1009651. a, body, cephalic midlength of second segment; dorsal mar- gin raised into thin serrate crest containing 13 or 14 teeth of nearly equal strength in central sector of row, about 0.5 length of crest continued onto carapace, deflecting from dorsal line of carapace at approximate 45° angle; ventral margin less prominent, straight, and unarmed or with | minute sub- terminal tooth; lateral carina (not visible in Fig. la) broadened proximally and conflu- ent with orbital margin. Carapace (Fig. la) with buttressed and distinct antennal spine; pterygostomian spine prominent, acumi- nate. Anterior antennal carina curving pos- teroventrally from near base of antennal VOLUME 116, NUMBER 1 spine to intersect at about midlength of car- apace with carina extending posteriorly from pterygostomial spine, its associated groove (not visible in Fig. 1b) continuing indistinctly posteriorly. Abdomen of female broadly arched dor- sally, gradually tapering posteriorly, height of sixth somite about half that of first so- mite. Pleura of third somite broadly round- ed, that of fourth somite irregularly round- ed, flanked dorsally by single remote ob- solescent spine; fifth pleuron with strong acute triangular posteroventral tooth flanked dorsally by cluster of 2 or 3 remote small spines on margin; sixth somite with middorsal length about 1.4—1.9 that of fifth somite, broad-based posterolateral tooth overlapping base of telson, smaller postero- lateral spine acute. Telson (Fig. 1c) elon- gate, subrectangular, length (not including posterior teeth) about 2.0—2.7 anterior width, 3.2—3.7 posterior width, about 1.3— 1.5 length of sixth somite; armed with 5—7 dorsolateral spines of nearly uniform size, occasionally unequal in number on either side; posterior margin slightly convex, armed with | spine at each corner and 12 feathered strong setae on posterior margin between them. Eyes (Fig. la, b) with cornea imperfectly developed; unfaceted though diffusely pig- mented; ovate in outline though fused to each other mesially beneath rostrum, each with upturned spine on anterodorsal sur- face. Antennular peduncle (Fig. 1b) extending slightly short of end of antennal scale; basal segment 2.1 length of second segment and about 2.2—2.5 length of third, all measured on dorsal margin; stylocerite well separated from peduncle, tapering to slender elongate tip variably reaching as far as midlength of second segment; basal segment with disto- dorsal margin exceeded by rostral tip though extended laterally into strong lateral spine reaching level nearly equal to that of stylocerite and closely appressed to second segment, much smaller distomesial spine slightly divergent; shorter second segment 161 with stronger mesiodistal spine. Antennal scale (scaphocerite) broadly rounded distal- ly, with broad triangular tooth on distola- teral border extending to same level as rounded distal border. Mouthparts (Fig. 2) fairly typical for the genus. Mandible (Fig. 2a, b) with 6 blunt, uneven, terminal teeth plus one slightly sharper dorsal subterminal tooth along cut- ting border, and with long, blunt, posterior tooth (““molar process”’ of some authors) separated from cutting edge by a wide gap; mandible deeply excavate on internal (pos- terior) surface just above this blunt poste- rior tooth; mandibular palp 2-segmented, distal article bearing numerous plumose se- tae; basal article with 2 plumose setae on distal border. First maxilla (Fig. 2c) with 2 endites; coxal (distalmost) endite with row of evenly spaced and evenly sized blunt spines (arrow) in addition to dense plumose and simple setae; basial endite (proximal endite) with longer, curved setae but lack- ing spine row; palp unsegmented, with 2 short and | long terminal setae as illustrat- ed. Second maxilla (Fig. 2d) scaphognathite with dorsal lobe broad, distally almost trun- cate, and lined with evenly spaced plumose setae; posterior lobe of scaphognathite ta- pering posteriorly, almost triangular, with setae increasing in length posteriorly, each seta angled and minutely serrate (for grooming); coxal lobe simple, blunt, setose; basial endite bilobed, each lobe fringed with plumose setae; palp thin, narrow, strap-like. First maxilliped (Fig. 2e) phyl- lopodous; coxal endite distally blunt and slightly recurved; basial endite bilobed, with distal lobe approximately twice length of proximal lobe; palp thin, tapering to acute distal tip bearing short setae and vis- ible only in posterior view; exopod broad, expanded and rounded distally, lined on ei- ther side with evenly spaced plumose setae; epipod bilobed at base, fused distally, with dorsolateral lobe bearing weak posterior projection. Second maxilliped (Fig. 2f) ped- iform, 6-segmented; coxa expanded and rounded on medial surface, which bears nu- 162 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Alvinocaris williamsi, new species, paratype female, USNM 1009653. Mouthparts, right side. a, mandible, outer (external) view; b, mandible and palp, inner view; c, first maxilla, with row of evenly spaced spines magnified and indicated by arrow; d, second maxilla; e, first maxilliped; f, second maxilliped; g, third maxilliped, with tip enlarged (upper arrow) and inner surface with rows of setae displayed (lower arrow). Scale bar 1 equals 0.5 mm (a, b, c), and 1.0 mm (d, e, f, g). merous plumose setae; fused basi-ischium tae on proximal region. Epipod nearly tri- with evenly spaced plumose setae along angular, with weak lobe coming off poste- medial border; merus and carpus short; pro- rior surface. Third maxilliped (Fig. 2g) ped- podus trapezoidal; dactylus short, distally iform, elongate, 4-segmented; coxal seg- rounded, with brush like patch of dense se- ment short, with minute epipod bearing | VOLUME 116, NUMBER 1 or 2 setae; distal 3 segments elongate and pediform; basal of these with plumose setae on proximal medial border and row of plu- mose setae along dorsolateral border; dac- tylus tipped with acute sclerotized spines surrounding central claw-like tip (arrow), and with evenly spaced densely packed rows of stiff, minutely serrulate setae on ventro-medial surface (curved arrow). << SD SIO Bigs 3 163 First pereopods (Fig. 3a, b) chelate, sub- equal; fingers curved ventrally and slightly laterally; dactylus more slender than fixed finger, tips varying slightly in relative length, mesial surface of each finger con- cave; cutting margins uniformly offset, closing without gape, each armed with row of almost uniform teeth so closely set as to be almost contiguous, line of sensory setae ——— ASS a SS Sy Za Alvinocaris williamsi, new species, paratype female, USNM 1009652. a, right first pereopod (che- liped), inner view; b, right first pereopod, outer view; c, left second pereopod, tip of chela; d, left second pereopod; e, left third pereopod; f, left fourth pereopod; g, left fifth pereopod. Scale bars | and 2 equal 1.0 mm; scale bar 3 equals 0.5 mm (bar | for d, e, f g; bar 2 for a, b; bar 3 for c). 164 mesial to cutting edges, acute tip of dacty- lus slightly spooned by elongate teeth slant- ed posteriorly and curving around external edge; entire leg slightly shorter than third maxilliped. Palm inflated but not elongate, approximately equal in length to fingers. Carpus measured along dorsal border slightly longer than palm, bearing oblique ventral crest, ending in strong distolateral spine (Fig. 3b) and flanked mesially by patch of setae on polygonal raised area; notch above spine smoothly concave, op- posing low ridge ending in small rounded spine on heel of palm; shallowly concave anteromesial margin of carpus leading dor- sally to 2 low rounded lobes. Merus swol- len in distal half, distinct from ischium but fused to it, neither armed. Second pereopod (Fig. 3c, d) shorter and more slender than first, reaching to between midlength and end of antennal peduncle; finger slightly longer than palm, similar in size and shape, opposed margins without gape, each pectinate with single row of teeth in distal half directed obliquely dis- tally and increasing slightly in size to end in noticeably stronger tooth crossing op- posite member when closed, but spineless proximally; carpus slender, about |.2 times longer than chela; merus unarmed, ischium with stout spine at approximately 0.75 length. Third to fifth pereopods (Fig. 3e, f, g) similar in length and structure, third reach- ing beyond antennal scale by about 0.3 the length of the propodus. Segments of these pereopods composed of: short dactylus armed with about 6 corneous spines on flexor surface, grading from small proxi- mally to longest and strongest distally; ven- tral row of spines on propodus leading to base of dactylus; carpus of each leg with distodorsal extension variously projecting as a stop along proximal part of propodal extensor surface; third leg with ischium and merus stronger than on fourth and fifth leg, merus of third with ventral spine at 0.3 and 0.6 length, distal spine tending to be stron- gest, and ischium with 2 spines in line with PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON those on merus; fourth leg with similar spines on merus but ischium not always bearing spines; fifth leg without spines on merus or ischium but with ventral row of small spines on propodus preceding base of dactylus. Pleopods (not illustrated) well developed, first pair with endopods about- half length of exopods; endopods tapering to acute tip. Uropods (Fig. 1c) with rami subequal in length, slightly exceeding posterior margin of telson; lateral ramus of exopod with movable spine mesial to larger distolateral tooth, diaeresis sinuous. Etymology.—The species name com- memorates the late Austin B. Williams, Re- search Scientist of the Systematics Labo- ratory and National Marine Fisheries Ser- vice at the National Museum of Natural History, Smithsonian Institution (see Le- maitre & Collette 2000). We are all bene- ficiaries of his ceaseless pursuit of excel- lence in his contribution to the taxonomy, systematics, biogeography, and evolution of various decapod groups, including vent shrimp and crab species. Remarks.—Alvinocaris williamsi appears morphologically most similar to A. stacto- phila, another species from a relatively shallow site (530 m). However, A. stacto- Phila is known only from a distant hydro- carbon seep site (as opposed to a hot vent) off the coast of Louisiana, northern Gulf of Mexico. Characters shared by the two spe- cies include a rostrum that is unarmed (or occasionally armed with a single tooth) on the ventral margin (a feature also shared with A. brevitelsonis and A. leurokolos from the western Pacific). The dorsal and ventral margins of the rostrum are heavily toothed in all other described Alvinocaris species. Additionally, both A. williamsi and A. stactophila have a relatively short ros- trum that continues posteriorly as a toothed carina that meets the dorsum of the cara- pace at an angle; this angle is sharper in A. williamsit and A. stactophila than in any other species except for A. longirostris, the distinctive rostrum of which, because of its VOLUME 116, NUMBER 1 length and armature, could not be confused with that of either A. stactophila or A. wil- liamsi. However, A. williamsi can be easily separated from A. stactophila by the sharper angle of the rostrum and carina where they meet the dorsal line of the carapace (ap- proximately 45° in A. williamsi vs. approx- imately 30° in A. stactophila), by the rela- tive length and width of the telson (which is considerably longer and narrower in A. stactophila), and by the shorter rostrum rel- ative to the carapace in A. williamsi. Based on the absence of ventral teeth on the rostrum and the angle of the rostral ca- rina and carapace, A. williamsi is also mark- edly similar to two species, A. brevitelsonis and A. leurokolos Kikuchi & Hashimoto (2000), recently described from the rela- tively shallow (~700 m) Mid-Okinawa Trough. However, Kikuchi & Hashimoto (2000) point out that there is significant var- iability in rostral characters, especially in A. brevitelsonis where the ventral rostral mar- gin is usually toothed (there are seven spines on the ventral rostral margin of the holotype) but may be unarmed, especially in young individuals. Alvinocaris leuroko- los lacks teeth on the ventral rostral margin, and is thus more similar to the new species A. williamsi. Alvinocaris williamsi 1s read- ily distinguishable from both of those spe- cies in having a lightly pigmented eye; the eye is unpigmented in both A. brevitelsonis and A. leurokolos. The distribution of Alvinocaris species within the global biogeographic vent prov- inces is centered in lower-latitude regions (Shank et al. 1999), and the strong morpho- logical similarities among extant Alvinocar- is species living in relatively shallow water vent and seep environments (<800 m) in the Atlantic and Gulf of Mexico is striking. Gebruk (1997) and others have hypothe- sized that vent-endemic shrimp species are derived from shallow-water seep ancestors. A molecular phylogenetic approach to this hypothesis by Shank et al. (1999) suggested that the analyzed vent-endemic Alvinocaris (a.e., A. markensis, A. lusca, and unde- 165 scribed Edison Seamount specimens) were derived from the A. stactophila (seep) lin- eage. This suggests that the extant Alvino- caris lineages share a common seep ances- tor. However, other vent-endemic lineages are basal to Alvinocaris, suggesting that vent lineages gave rise to a seep lineage that in turn gave rise to other extant vent line- ages. Based on morphology, A. williamsi from the relatively shallow Menez Gwen site more closely resembles a seep-endemic species (1.e., A. stactophila) than other hot vent species, and therefore molecular ge- netic comparisons of A. williamsi with con- geners from deep-sea hydrothermal vents, seamounts, back-arc basins, and hydrocar- bon seeps would markedly improve our un- derstanding of the evolution and radiation of these shrimp among diverse chemosyn- thetic environments throughout the world’s oceans. Acknowledgments We owe an enormous debt of gratitude to Austin B. Williams, whose early work on shrimp from hydrothermal vents established the general framework within which all subsequent work on these shrimp is now viewed, and who unfortunately did not live to see the completion of this species de- scription. Special thanks to Ruth Gibbons for persevering to make sure that several manuscripts, including this one, begun by A. B. Williams before his death were brought to fruition, and to M. Oremland and K. Hiratsuka Moore for providing il- lustrations. We thank the crews and pilots of the R/V Atlantis and the DSV Alvin for their skillful collecting efforts, A. Gebruk for shipboard expertise, and the co-chief Scientists of the expedition, R. C. Vrijen- hoek and R. A. Lutz; some of the shared funding was via the National Science Foun- dation (OCE-96-33131). Finally, we thank T. Komai and A. L. Vereshchaka, and two anonymous reviewers, for suggestions that greatly improved the manuscript. This is contribution No. 10700 of the Woods Hole 166 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Oceanographic Institution. , TMS was sup- ported in part by the Postdoctoral Scholar Program at the Woods Hole Oceanographic Institution, with funding provided by the Devonshire Foundation. JWM was support- ed in part by NSF grants DEB 9972100, DEB 9978193 (a PEET grant from the Sys- tematic Biology program), and DEB 0120635. Literature Cited Christoffersen, M. L. 1986. Phylogenetic relationships between Oplophoridae, Atyidae, Pasiphaeidae, Alvinocarididae fam. n., Bresiliidae, Psalido- podidae and Disciadidae (Crustacea Caridea Atyoidea).—Bolletim de Zoologia 10:273—281. Colago, A., D. Desbruyéres, T. Comtet, & A.-M. Alayse. 1998. Ecology of the Menez Gwen hy- drothermal vent field (Mid-Atlantic Ridge/ Azores Triple Junction).—Cahiers de Biologie Marine 39:237—240. Comtet, T., & D. Desbruyéres. 1998. Population struc- ture and recruitment in mytilid bivalves from the Lucky Strike and Menez Gwen hydrother- mal vent fields (37°17'N and 37°50'N on the Mid-Atlantic Ridge).—Marine Ecology Pro- gress Series 163:165—177. Desbruyeéres, D., et al. 2001. Variations in deep-sea hydrothermal vent communities on the Mid-At- lantic Ridge near the Azores plateau.—Deep- Sea Research 48(5):1325—1346. Gebruk, A. V., S. V. Galkin, A. L. Vereshchaka, L. I. Moskalev, & A. J. Southward. 1997. Ecology and biogeography of the hydrothermal vent fau- na of the Mid-Atlantic Ridge.—Advances in Marine Biology 32:93—144. Hashimoto, J., O Suguru, T. Gamo, H. Chiba, T. Ya- maguchi, S. Tsuchida, T. Okudaira, H. Watabe, T. Yamanaka, & M. Kitazawa. 2001. First hy- drothermal vent communities from the Indian Ocean discovered.—Zoological Science 18(5): 717-721. Kikuchi, T., & J. Hashimoto. 2000. Two new caridean shrimps of the family Alvinocarididae (Crusta- cea, Decapoda) from a hydrothermal field at the Minami-Ensei Knoll in the Mid-Okinawa Trough, Japan.—Species Diversity 5:135—148. , & S. Ohta. 1995. Two caridean shrimps of the families Bresiliidae and Hippolytidae from a hydrothermal field on the Iheya Ridge, off the Ryukyu Islands, Japan.—Journal of Crustacean Biology 15(4):771—785. Lemaitre, R., & B. B. Collette. 2000. Austin Beatty Willliams (17 October 1919—27 October 1999). Biographical summary. Proceedings of the Bi- ological Society of Washington 113:1—12. Martin, J. W., & J. C. Christiansen. 1995. A new spe- cies of the shrimp genus Chorocaris Martin & Hessler, 1990 (Crustacea; Decapoda; Bresili- idae) from hydrothermal vent fields along the Mid-Atlantic Ridge.—Proceedings of the Bio- logical Society of Washington 108:220—227. , & G. E. Davis. 2001. An updated classifica- tion of the Recent Crustacea.—Natural History Museum of Los Angeles County, Science Series No. 39:1—124. , & R. R. Hessler. 1990. Chorocaris vandov- erdae, a new genus and species of hydrothermal vent shrimp from the Western Pacific.—Contri- butions in Science, Natural History Museum of Los Angeles County 417:1—11. Milne-Edwards, A., & E.-L. Bouvier. 1894. Brachy- oures et anomures. Crustacés décapodes prov- enant des campagnes du yacht Il Hirondelle (1886, 1887, 1888). Premiére partie.—Résultats des campagnes scientifiques accomplies sur son yacht par Albert Ier, Prince Souverain de Mon- aco 7:1—112. Segonzac, M., M. de Saint Laurent, & B. Casanova. 1993. L-énigme du comportement trophique des crevettes Alvinocarididae des sites hydrother- maux de la dorsale medio-atlantique.—Cahiers de Biologie Marine 34:535—571. Shank, T. M., R. A. Lutz, & R. C. Vrijenhoek. 1998. Molecular Systematics of shrimp (Decapoda: Bresiliidae) from Deep-Sea Hydrothermal Vents: enigmatic ““Small Orange” Shrimp from the Mid-Atlantic Ridge are juvenile Rimicaris exoculata.—Molecular Marine Biology and Biotechnology 7(2):88—96. , M. B. Black, K. Halanych, R. A. Lutz, & R. C. Vrijenhoek. 1999. Miocene radiation of deep-sea hydrothermal vent shrimp (Caridea: Bresiliidae): evidence from mitochondrial cy- tochrome oxidase subunit I.—Molecular Phy- logenetics and Evolution 13(2):244—254. Van Dover, C. L., S. E. Humphris, D. Fornari, C. M. Cavanaugh, R. Collier, S. K. Goffredi, J. Hash- imoto, M. D. Lilley, A. L. Reysenbach, T. M. Shank, K. L. Von Damm, A. Banta, R. M. Gal- lant, D. Gotz, D. Green, J. Hall, T. L. Harmer, L. A. Hurtado, P. Johnson, Z. P. McKiness, C. Meredith, E. Olson, I. L. Pan, M. Turnipseed, Y. Won, C. R. Young, III, & R. C. Vrijenhoek. 2001. Biogeography and ecological setting of Indian Ocean hydrothermal vents.—Science 294:818—-823. Vereshchaka, A. L. 1996a. Comparative analysis of taxonomic composition of shrimps as edifica- tors of hydrothermal communities in the Mid- Atlantic Ridge.—Doklady Biological Sciences 351(1):576-578. . 1996b. A new genus and species of caridean shrimp (Crustacea: Decapoda: Alvinocarididae) VOLUME 116, NUMBER 1 167 from North Atlantic hydrothermal vents.—Jour- shrimp of the family Bresiliidae from thermal nal of the Marine Biological Association of the vents of the Galapagos Rift—Journal of Crus- United Kingdom 76(4):951—961. tacean Biology 2:136—147. . 1997. A new family for a deep-sea caridean . & E C. Dobbs. 1995. A new genus and spe- shrimp from North Atlantic hydrothermal cies of caridean shrimp (Crustacea: Decapoda: vents.—Journal of the Marine Biological As- Bresiliidae) from hydrothermal vents on Loihi sociation of the United Kingdom 77:425—438. Seamount, Hawaii.—Proceedings of the Biolog- Williams, A. B. 1988. New marine decapod crusta- ical Society of Washington 108:228—237. ceans from waters influenced by hydrothermal , & P A. Rona. 1986. Two new caridean discharge, brine, and hydrocarbon seepage.— shrimps (Bresiliidae) from a hydrothermal field Fishery Bulletin 86:263—287. on the Mid-Atlantic Ridge.—Journal of Crus- , & E A. Chace, Jr 1982. A new caridean tacean Biology 6:446—462. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):168—-189. 2003. New brachyuran crabs (Crustacea: Decapoda) from the Upper Pliocene Yorktown Formation of southeastern Virginia Warren C. Blow Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20013-7012, U.S.A. Abstract.—Six new species of crabs, representing three families and one new genus, are described and illustrated from the Upper Pliocene Yorktown For- mation of southeastern Virginia. They are remarkable for their state of pres- ervation and represent the first fossil record for three of these living genera in Virginia’s rich Neogene marine deposits. The presence of Stenocionops along with the common occurrence of Persephona in these deposits suggests that warm temperate waters covered southeastern Virginia during the deposition of late Yorktown sediments. Introduction In 1935, Mary J. Rathbun published the first comprehensive paper on the fossil crustaceans, principally decapods, of the Atlantic Gulf Coastal Plain of Eastern North America. In this landmark paper she describes or mentions all of the fossil deca- pod, stomatopod and isopod crustaceans from this region made available to her at that time from a host of sources (Rathbun 1935:1). Of the 25 families, 67 genera, and 167 species covered, and for the most part illustrated, in her paper only eight species representing six genera and six families were listed and treated from Virginia from deposits of Miocene, Pliocene and Pleisto- cene age. All but one were listed as occur- ring in deposits of the Yorktown Formation which at that time was regarded as Mio- cene. However the Neogene stratigraphy of southeastern Virginia has evolved consid- erably since 1935 (Hazel 1971la, Ward & Blackwelder 1980, Ward & Gilinsky 1993), and consequently only Cancer borealis Stimpson, 1859, Panopeus herbstii Milne- Edwards, 1834, and Libinia dubia Milne- Edwards, 1834, are listed from localities that are now regarded as old as Miocene. With the exception of Cancer borealis Stimpson, 1859, the remaining species— Callianassa suffolkensis Rathbun, 1935; C. atlantica Rathbun, 1926; Persephona punc- tata (Linné, 1758); Callinectes sapidus Rathbun, 1896; Panopeus herbstii Milne- Edwards, 1834; and Libinia dubia Milne- Edwards, 1834—are all listed from locali- ties now regarded as Upper Pliocene. Cal- linectes sapidus Rathbun, 1896 is the only species listed from the Pleistocene. In the present paper six additional taxa, Hepatus. bottomsi, n. sp., Pterocarcinus baileyi, n. gen, n. sp., Persephona niemey- erl, n. sp., P. rodesae, n. sp., Stenocionops dyeri, n. sp., and Euprognatha ricei, n. sp. are described as new and illustrated from deposits of late Pliocene age in the same general geographic area as that treated by Rathbun 67 years ago. Unlike Rathbun’s material, which is very fragmentary, often consisting only of fingers, the taxa present- ed in this paper are remarkable for their state of preservation and completeness. All are represented by their original exoskele- ton, and many even exhibit a uniform color similar to that of their living relatives. The material studied in this paper was collected from the Moore House Member of the upper part of the Yorktown Forma- tion from three localities in southeastern VOLUME 116, NUMBER I Virginia. They are: The Lone Star Lakes, near Chuckatuck, Suffolk; Riddick Pit, bor- row pit east side Virginia Routes 10/32, 4.4 km southeast of Benns Church; and Rices Pit, a now flooded borrow pit in Hampton (Ward & Blackwelder 1980). Detailed lo- cality data for the these localities are pro- vided in Appendix 1. With the exception of one female para- type of Stenocionops, the gender of the re- maining specimens described in this paper could not be determined. Stratigraphy & Paleoenvironment Mansfield (1943) expanded on his earlier biostratigraphic division of the Yorktown Formation into Zones I and II and regarded both units as entirely Miocene. His zona- tion, though flawed, was very workable in the field for many years. Hazel (1971a) di- vided the Yorktown into three ostracode zones of which the uppermost or Puriana mesacostalis zone, he regarded as Pliocene. Ward & Blackwelder (1980) refined the stratigraphy of the Yorktown Formation by redefining and dividing it into two forma- tions: the Upper Miocene Eastover Forma- tion overlain by the Lower Pliocene York- town Formation. The latter was further di- vided into four members which in ascend- ing order were named: Sunken Meadow, Rushmere, Mogarts Beach, and Moore House. Ward & Blackwelder (1980:44), in defining the uppermost member state, “‘the Moore House Member, representing the re- gressive phase of the Yorktown Formation, reflects a renewal of higher current and wave energy conditions. The member con- sists of sandy shell beds and cross-bedded shell hash and locally is cemented to form a very indurated rock.’’ They further (p. 45, 47) state, ““The Moore House Member is found only east of the Surry Scarp,”’ and “The Moore House Member reflects a pro- gressively shallowing, regressive sea. Mol- luscan assemblages indicate normal salini- ties, but some of the highest beds in the Williamsburg area contain a few brackish- 169 water mollusks. Locally large offshore bars were the site of rapid, large-scale, cross- bedded sand deposition.” All of the speci- mens studied in this paper and the strata at the localities from which they were derived are now regarded by L. W. Ward as be- longing to the Moore House Member and of late Pliocene age (Ward, pers. comm., 26 November 2002). Kier (1972), in a study of the echinoids from two (Chuckatuck and Rices Pit) of the three localities covered in this paper, gives a late Miocene age for his material. This age, though now regarded as incorrect, was based on the ostracode zones of Hazel (1971la) and the work of Gibson (1967). A late Miocene age for strata found at these localities was also the consensus of Neogene molluscan workers at that time. Kier’s interpretation of the life habits of the Yorktown echinoids is a valuable resource for understanding the paleoenvironmental conditions that existed during the deposi- tion of the Yorktown at Rices Pit and par- ticularly at the Chuckatuck Bar. (See John- son & Coch 1969, Johnson 1969 and Campbell 1993 for a detailed description of the structure, extent and mollusks of the Chuckatuck Bar, which includes the depos- its at the Riddick Pit). Kier (1972:3) states, ““The Yorktown echinoids, like the ostracodes, indicate a past climate warmer than now” ... “Al- though some of these living species range into cooler waters, they all occur in sub- tropical regions, suggesting that the fossil echinoids they resemble lived in waters warmer than the mild-temperate waters now occurring off the coast of Virginia” Earlier, Hazel (1971b) in a detailed study of the Yorktown ostracodes coupled with his own detailed studies of Western Atlantic Recent ostracodes concluded that the climate of the Yorktown Sea was much warmer than Vir- ginia’s coastal waters are today. Kier (1972: 2) in summarizing Hazel (1971b) states, ‘“This equable thermal regime is markedly different from that of any province and con- comitant climate zone now extant along the Atlantic coast of the United States. The 170 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON temperatures varied from about 12°C to 15°C in the winter to about 17.5°C to 20°C and finally to 20°C to 25°C in summer.”’ Ward & Gilinsky (1993:26) state, “‘The Yorktown Sea supported a large, warm- temperate to subtropical molluscan assem- blage.’’ Western Atlantic species of the spi- der crab Stenocionops are not reported north of Cape Hatteras, North Carolina but inhabit warmer waters further south, some to Brazil (see Rathbun 1925; Williams 1965, 1984 for geographic ranges). The purse crab, Persephona mediterranea (Herbst, 1794) though reported from as far north as New Jersey and the southern Ches- apeake Bay, is not commonly found there but is commonly found in warmer more southern waters from off Cape Hatteras to Brazil. The crabs of the Yorktown Sea, like the mollusks and other invertebrates found at the localities cited here, are representa- tive of similar taxa that would be found to- day in warm-temperate to subtropical, open marine waters of normal salinity. Terms and Conventions The minute mushroom-shaped structures found on the surfaces of numerous brach- yurans, and referred to by French authors for more than 100 years as “‘champignons”’ (mushrooms) in reference to their ““mush- room-like”’ appearance, are here referred to as bolitimorphs after the Greek bolites (mushroom) in combination with the Greek morphe (form, shape). Bolitimorph as here defined is intended to replace the phrases ““mushroom-like structure” or ““mushroom- shaped structure”. At least one living spe- cies, Merocryptus boletifer Milne-Edwards & Bouvier, 1894, from the Azores was named for this common feature. Haj & Feldmann (2002) observed similar structures in members of the family Ran- inidae and introduced the term “‘fungi- form” for their mushroom shaped _ struc- tures that form a “pebbled surface’? and stated, ““This pebbled surface has not been recognized in any other decapod taxon, nor has its structure and function been de- scribed previously.’ These structures have, however been known for some time as not- ed above, and both their development and function studied in great detail (Seréne 1954, Guinot 1979). Given the vast range of shapes suggested by the term Fungi, the term bolitimorph is more specific and is preferred here to denote mushroom-shaped structures. Repositories.—Primary and secondary types are deposited in the National Museum of Natural History, Smithsonian Institution, Washington, and The Virginia Museum of Natural History, Martinsville, Virginia as indicated. Abbreviations. —NMNH, National Mu- seum of Natural History, Smithsonian In- stitution; S.I., Smithsonian Institution; USGS, U.S. Geological Survey (when used with number indicates a locality; see local- ity register at end of paper); USNM, abbre- viation for catalogue numbers of the former U.S. National Museum now the National Museum of Natural History; VMNH, Vir- ginia Museum of Natural History. Measurements Measurements are expressed in millime- ters. Abbreviations are as follows: cl, car- apace length, maximum longitudinal mea- surement; cw, maximum transverse mea- surement; prl, propodus length, length of palm or combined length of palm and fixed finger; prh, propodus height; prt, propodus thickness. Systematic Paleontology Family Calappidae Subfamily Matutinae Genus Hepatus, Latreille, 1802 Hepatus bottomsi, new species lense Il Diagnosis.—Carapace arcuate (shaped like a drawn bow), front prominent, strong- ly elevated, produced well beyond orbits and anterolateral margin; dorsal surface VOLUME 116, NUMBER 1 171 Fig. 1. with eight distinct, pustulate protuberances; suborbital flank deeply depressed above de- scending, weakly sinuous anterolateral mar- gin; surfaces everywhere pitted, particularly anteriorly where appearing eroded. Description.—Carapace arcuate, strongly narrowing posteriorly, broad, length about three-fourths width, widest just forward of anterior-lateral angle, strongly convex lon- gitudinally and highly arched transversely, appearing tripartite in frontal view, gastric and branchial regions distinctly swollen, re- gions moderately well defined, greatly ele- vated in younger individuals; dorsal surface with eight prominent, pustulate protuber- ances. Surfaces everywhere pitted, giving an eroded appearance; pits most abundant on anterior one-third of carapace particular- ly on front, eyes, orbits, post orbital and anterolateral margins, and adjacent surfac- es. Surfaces somewhat smoother in larger, more mature individuals. Fronto-orbital Hepatus bottomsi Holotype (USNM 520705) dorsal view. Scale = 10 mm. width between one-fourth and one-third carapace width. Front prominent, thick, bi- lobate, strongly projecting beyond orbits and anterolateral margin (and placed well above anterolateral continuation of borders of carapace); lobes truncate with lateral margins rounded anteriorly, dorsally divid- ed by deep sulcus, ventrally fused or closed along entire length. Antennules very obliquely set. Orbit small, subcircular, de- fined by slightly raised wide rim; rim com- posed of three lobes, base of eye in orbit level with front; orbits directed obliquely downward parallel to slope of thick, round- ed dorsal margin above suborbital depres- sion. Anterolateral margin comprised of about 40 blunt denticles situated on from 12 to 13 or more teeth; teeth separated by closed incisions, most of which begin with a deep pit: teeth bidentate anteriorly other- wise most often tridentate with median den- ticle strongest. Margin, where descending 172 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON anteriorly to buccal cavity, weakly sinuous, comprised of single row of denticles of sub- equal strength. Posterolateral margin straight, strongly convergent posteriorly; anterior portion of margin, just posterior to strong tooth defining anterolateral angle, with a long low-angle tooth directed obliquely forward, tooth with double or tri- ple close set rows of low tubercles; tuber- cles continue as a single or double row to distinct, though small, subacute tooth defin- ing posterolateral angle; single row of tu- bercles continue transversely across poste- rior margin of carapace where it eventually continues along the opposing lateral mar- gins until returning to the opposite side of the buccal cavity from which it began. Pos- terior margin high, narrow, width about one-seventh to one-fifth carapace width; margin defined above by slightly convex row of tubercles continuous with row of posterolateral margin, and submarginal row of tubercles below. Postfrontal surface with medial, broad, very shallow sulcus leading to base of frontal lobes. Epigastric region defined by minute pustules on either side of base of postfrontal sulcus. Hepatic region depressed, bordered anteriorly by _ thick, rounded carapace margin leading to orbit. Remaining dorsal regions defined by eight prominent, pustulate protuberances distrib- uted as follows: paired protogastric, one mesogastric, One cardiac, paired epibran- chial, and paired mesobranchial. Protogas- tric and mesogastric regions large, of sub- equal size, subcircular. Epibranchial regions developed into oblique ridges directed to- ward marginal tooth of anterolateral angle. Mesobranchial regions developed into ridg- es aligned roughly with anterior ridge of cardiac region to form a transverse line of protuberances aligned with first posterolat- eral tooth. Anterior face of ridges, smooth, steep. Intestinal region weakly inflated, in- distinct. Posterior gastric pits minute, close- ly spaced. Color.—Rust or yellowish brown to brownish gray when dry, gray or dark gray when wet. Recent species in alcohol are usually gray. Etymology.—This species is named for E. Edward Bottoms, paleoanthropologist of Portsmouth, Virginia, in recognition of his numerous writings on Paleo Indians of the Eastern United States, for his years of teaching regional history on all student lev- els in the Tidewater Area, and for stimulat- ing the author’s early interest and develop- ment in paleontology. Holotype.—The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. USGS 26891; (USNM 520705) cl 42.88 mm, cw 54.92 mm. Paratypes: (USNM 520706), cl 49.97 mm, cw 66.18 mm. USGS 26891 as above; (USNM 520707), cl 33.63 mm, cw 42.77 mm. USGS 26891 as above. (VMNH 13546), cl 47.24 mm, cw 64.19 mm. Same geographic locality as USGS 26891 as above. Occurrence.—The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. Remarks.—Recent comparisons: The greatly produced front of H. bottomsi im- mediately separates it from all living spe- cies of Hepatus but may confuse it with species of the genus Hepatella, Smith, in Verrill, 1869, type species H. amica, from which it differs as follows. Adult specimens of Hepatus bottomsi have eight low dorsal protuberances compared to the six greatly elevated dorsal protuberances found in adults of Hepatella, and as Rathbun, 1937 notes, Hepatella lacks the suborbital de- pression found in members of the genus Hepatus. This feature is distinct in H. bot- toms. Fossil comparisons: Hepatus bottomsi differs from the middle Miocene H. nodo- sus Collins & Morris, 1976, of Trinidad, the only fossil species with which it might be confused, in having a greatly produced front and obliquely directed epibranchial ridges. In H. nodosus the front is marginal and the epibranchial ridges are transverse. Individuals of H. bottomsi become smoother and their dorsal protuberances VOLUME 116, NUMBER 1 lower and much less pronounced as they grow larger and older. Leucosiidae Subfamily Ebaliinae Pterocarcinus, new genus Diagnosis.—Carapace suboctagonal, wing like; ““‘wing”’ or branchial region ex- tending laterally well beyond basis of legs; margins very thin, carinate; front narrow with two very weak rostral “‘horns’’; car- diac region most prominent, a raised, sub- circular platform or “‘mesa’’ surrounded by a deep, wide sulcus; posterior margin broad, bilobate. Type species.—Pterocarcinus baileyi, n. sp. by present designation and monotypy. Related species.—?Ebalia rotundata (A. Milne-Edwards, 1880); See A. Milne-Ed- wards & E. L. Bouvier, 1902 for figures. Generic status uncertain. Etymology.—Ptero from the Greek, pter- on, wing, in combination with the generic name Carcinus derived from the Greek, karkinos, or crab. Gender masculine. Remarks.—In outline Pterocarcinus re- sembles a number of species currently placed in Ebalia Leach, 1817; Lithadia Bell, 1855 and Speloeophorus A. Milne- Edwards, 1865, but it differs dramatically in overall shape and form from the type species of all of these genera. It does not possess the subhemispherical form of the type species of Ebalia, E. tuberosa (Pen- nant, 1777). Nor does it exhibit the outline, ““thick”’ wings, and caverns of the type spe- cies of Speloeophorus, S. nodosus (Bell, 1855), and it does not exhibit the highly elevated and excavated form represented by the type of Lithadia, L. cumingii Bell, 1855. (See Rathbun 1937, for figures of North and South American species placed in these genera). In overall form and outline P. bail- eyi most closely resembles Ebalia rotun- data (A. Milne-Edwards, 1880), a species clearly not in the genus Ebalia, but one which might be its closest living relative. Ebalia rotundata, though very similar to 173 this new species, differs in having very thick or rotund branchial “‘wings’’ unlike the thin carinate wings of Pterocarcinus and in having the cardiac and urogastric re- gions joined and not completely separated by a deep sulcus as in Pterocarcinus (Fig. 2A). In the absence of the first right pleo- pod of a male and other features present in living material, the true relationship of this new fossil genus to other leucosiids may never be fully understood, but its form is quite different from all known genera. Frag- ments of this new genus are fairly common in strata of Pliocene age along the Atlantic Coastal Plain. Given this situation it seems best to give this unusual leucosiid a distinct generic name. Pterocarcinus baileyi, new species Figs. 2—4 Diagnosis.—As for genus. Description.—Carapace suboctagonal, wing-like, length about eight-tenths width, broadest at anterolateral angle; regions moderately well defined; dorsal ridge from front along midline to and including inter- secting diagonal branchial ridges; hepatic and cardiac regions and posterior lobes dis- tinct to prominent, ridges greatly elevated; irregular inner margin of elevated hepatic region deeply excavated. Cardiac region most prominent (Figs. 2A, 3B, FP), a raised, subcircular or oval free-standing “‘mesa- like” platform completely surrounded by a deep, wide sulcus filled with scattered, tall bolitimorphs (Fig. 3B, D); ““mesa”’ slightly elevated above ridge of branchial region and strongly produced posteriorly. Most surfaces of carapace covered with pavement of low, closely spaced or abutting boliti- morphs (Fig. 2A); bolitimorphs separate or free standing in depressions adjacent to ridge leading to front, along and anterior to diagonal ridge of branchial region, and in sulcus completely surrounding cardiac ‘““mesa-like”’ region as well as ventral sur- faces of pterygostomian and branchial re- gions where they are highest and most dis- 174 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nos) A ~ ay Bige2: Pterocarcinus baileyi Holotype (VMNH 13547) A. dorsal view; 1 mm. B. frontal view, close up. Scale = VOLUME 116, NUMBER 1 ipia, 3). tinct. Front truncate, elevated, narrow, width slightly less than one-sixth width of carapace, lateral margins defined by very short, rounded, blade-like “horns” (Fig. 2A, B); front weakly projecting. Horns ap- pearing more distinct in younger individu- als. Fronto-orbital width about one-fourth carapace width. Orbit sub-circular, margin- al, with three sutures, slightly produced at ex-orbital margin. Orbit and antennule fos- sae openly connected; basal antennal seg- ment not present (not preserved); antennule fossae oblique. Bolitimorphs on fronto-or- bital surfaces closely spaced forming pave- ment (Fig. 2B). Pterocarcinus baileyi Holotype (VMNH 13547) A. anterior view; B. posterior view; C. left lateral view; D. oblique posterior view. Scale = | mm. E. left lateral view of ventral surface beneath branchial “‘wing”’. Scale = 0.5 mm. EF oblique frontal view to show height of cardiac ““mesa’’. Scale = 1 mm. Pterygostomian region developed into long ridge subparallel to anterolateral mar- gin (Fig. 3A); ridge composed of free standing rows of bolitimorphs of median length that cluster to form a downward, obliquely directed tooth before terminating posteriorly. Buccal cavity triangular, ante- rior margins of afferent channels with mi- nute pustules, margin curving downward laterally to a narrow notch or sulcus. Bran- chial “‘wings’’ extending laterally well be- yond basis of ambulatory legs and chela (Fig. 3A—C); legs probably not visible in dorsal view. “Wings” ventrally covered with distinctly separated tall bolitimorphs 176 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Pterocarcinus baileyi Holotype (VMNH 13547) A., B. stereo pair in dorsal view. Scale = 1 mm. Pterocarcinus baileyi Paratype (USNM 520708) C., D. close-up of bolitimorphs in post-hepatic depression. Scale = 200 wm. Prerocarcinus baileyi Holotype (VMNH 13547) E. close-up of stalk of bolitimorph to show dagger-like structures that might function to hold sediment in place. Scale = 10 wm. Pterocarcinus baileyi Paratype (USNM 520708) EF high magnification to show unique structure of crown of bolitimorph seen in figures C. and D. above. Scale = 20 wm. (Fig. 3E); bolitimorphs much shorter and abutting around margins at insertion of am- bulatory legs and chela. Anterolateral, lat- eral, and posterolateral margins very thin, carinate (Fig. 3A—C); leading edge of each curving downward like an airfoil; edge ap- pearing beaded below where bordered by a single row of minute bolitimorphs. Antero- lateral margin longest, convergent anteri- orly, set at 40 degrees to longitudinal axis; margin interrupted medially by a very dis- tinct fissure between hepatic and branchial VOLUME 116, NUMBER 1 regions; margin anterior to fissure thickened and posterior to fissure sinuous; anterolat- eral angle rounded to subacute. Lateral mar- gin about two-thirds length of anterolateral margin, distinctly concave in dorsal view and raised medially in lateral view; anterior half weakly divergent anteriorly; posterior half subparallel to longitudinal axis and ter- minating in a subacute angle. Posterolateral margin perpendicular to longitudinal axis, length less than one-half length of antero- lateral margin; margin separated from pos- terior lobes by deep sulcus. Posterior dis- tinctly bilobate, width slightly less than one-half width of carapace; lobes thick, separated by a deep fissure; dorsal surface of each, smooth; lobes moderately pro- duced posteriorly, margin of each terminat- ing in a thin, raised, beaded rim. Apparent posterior gastric pits distinct, very widely spaced, set on branchial ridges just anterior to gastro-cardiac sulcus (Fig. 2A). Color.—Brown, buff, or sometimes weathered bluish grey. Etymology.—This species is named for Richard H. Bailey, geologist and paleontol- ogist, Professor, Northeastern University, Boston, Massachusetts, for his research contributions to Atlantic Coastal Plain ge- ology and paleontology, for his many years of superb teaching, and especially for his years of encouragement and field assistance to the author. Holotype.—Riddick Pit, about 4.4 km southeast of Benns Church, Isle of Wright County, Virginia. Same geographic locality as USGS 26892 (VMNH [3547), cl 10.04 mm, cw 12.75 mm. Paratypes: (VMNH 1S543)5 cl O77) mm. cw 13336 4mm este mate, incomplete. Same geographic locality as USGS 26892 above. (USNM 520708), ero losmm ecw 1022 mm: USGS Z6s9l. The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia; (USNM 520709), cl 8.79 mm, cw 10.95 mm, USGS 26891 as above. Occurrence.—Riddick Pit, about 4.4 km southeast of Benns Church, Isle of Wright 77 Co., Virginia, and The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. Remarks.—The function of the boleti- morphs found on Prerocarcinus baileyi and other crabs probably differs with each of the taxa on which they occur. Given the dif- ficulty of cleaning specimens covered with these boletimorphs, it is obvious to this au- thor that their primary function is most probably one of concealment, in that they very effectively trap sand particles and hold them in the grooves and excavations of the carapace, thus obscuring the crab’s outline and allowing it to blend into the substrate. Such concealment would allow it to hide from predators and ambush prey as well. These sometime very ornate, even flower- like, surface structures might also aid in an- choring the crab in the substrate. They also undoubtedly have a structural advantage in their reinforcement of the carapace. Seréne (1954) elaborates on their development and Guinot (1979) discusses their development and presents the hypothesis that they serve to channel water to the bases of the legs and that they aid in concealment as well. My own observations have indicated that their individual form can be species specific and therefore of particular use to paleon- tologists in the identification of crab frag- ments. Subfamily [iinae Genus Persephona Leach, 1817 Persephona niemeyeri, new species Biss. SAe GE. FE Diagnosis.—Carapace hemispherical with three posterior spines; front bilobate, elevated, produced with orbits well beyond anterolateral margin; post frontal, hepatic and all other dorsal surfaces thickly covered with relatively high, subacute tubercles of various sizes; tubercles strongest along lat- eral and posterior margins, and highly con- centrated on front, orbits and adjacent areas where developed into bolitimorphs. Description.—Carapace hemispherical, width about nine-tenths of length, regions PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON VOLUME 116, NUMBER 1 very weakly defined, front and orbits ad- vanced well beyond anterolateral margins; posterior with three equally spaced re- curved spines, two marginal, and one me- dian above margin. Dorsal surfaces thickly covered with relatively high, subacute tu- bercles of various sizes; tubercles strongest along lateral and posterior margins, most concentrated below plural suture, where very low and closely spaced, and highly concentrated on front, orbits and adjacent areas where developed into bolitimorphs. Front very narrow, width about 0.09 width of carapace, bilobate, lobes separated by distinct v shaped sulcus, laterally defined by raised clusters of tubercles; anterior mar- gins of lobes blunt, denticulate. Orbit very small, divided by three deep, narrow, open sutures creating four lobes. Minute, elon- gate tubercles span sutures from both sides, creating a pectinate appearance. Supraorbit- al eave and median lobe meet at a steep angle to form a broad v as seen from above. Large distinct bolitimorphs concentrated on outer lobe. Suborbital lobe with afferent channel below; channel deeply notched at anterolateral angle, inner angle rounded dis- tally, margin pustulose and channel pro- duced beyond front and orbits as seen in dorsal view. Hepatic region slightly swol- len, weakly defined, thickly covered with tubercles. Superior margin of subhepatic re- gion with low ridge of concentrated tuber- cles; ridge not joining but rather descending just below tubercles of anterolateral margin. Anterior and posterior lateral margins of carapace continuous, defined by a row of larger tubercles beginning at orbit and con- tinuing along upper edge of plural suture to a point just above tooth of posterolateral an- gle; row poorly separated from dorsal tu- — Big 5: 179 bercles, sometimes indistinct or appearing as double row where intermixed with dorsal tubercles; area below plural suture covered with closely spaced low tubercles, surface appearing very rough, bordered below by distinct row of strong tubercles just above bases of ambulatory legs. Posterior margin slightly more than one-third carapace width, produced posteriorly, laterally de- fined by a broad, stout, triangular, acute re- curved spine at each corner; margin be- tween spines comprised of from seven to 10 large recurved tubercles; intestinal mar- gin above with a single medially placed, stout, recurved spine. Spines coarsely gran- ulate. Posterior gastric pits not evident. Color.—Reddish orange to yellow when dry. Etymology.—tThis species is named for Antonio B. Niemeyer, Jr., science teacher, retired, of Chesapeake, Virginia, in recog- nition of his contribution to science through his many years of teaching science in the public schools of tidewater Virginia and for the fundamentals of general science that he instilled in this author so many years ago. Holotype.—The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. USGS 26891, (USNM 520710) cl 52.30 mm, cw 48.13 mm; Paratypes: (USNM 520711) cl 43.69 mm, cw 38.85 mm, USGS 26891 as above; (USNM 520712) cl 30.06 mm, cw 25.91 mm, USGS 26891 as above; (USNM 520713) cl 23.02 mm, cw 18.07 mm. Frag- ment of anterior portion of carapace only. USGS 26891 as above; (VMNH [3549) cl 36.83 mm, cw 33.27 mm. Same geographic locality as USGS 26891 above. Occurrence.—The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. Remarks.—Recent comparisons: Perse- Persephona niemeyeri Paratype (USNM 520711) A. closeup of left anterior surface in dorsal view. Scale = 4 mm. Persephona rodesae Holotype (USNM 520714) B. closeup of right anterior surface in dorsal view. Scale = 5 mm. Persephona niemeyeri Paratype (USNM 520711) C. dorsal view. Scale = 10 mm. Per- sephona rodesae Holotype (USNM 520714) D. dorsal view. Scale = 10 mm. Persephona niemeyeri Holotype (USNM 520710) E. dorsal view of geronic specimen, bare areas are a result of wear and abrasion. Scale = 10 mm. Persephona niemeyeri Paratype (USNM 520712) E dorsal view. Scale = 10 mm. 180 phona niemeyeri is most similar to the Re- cent western Atlantic purse crab, P. medi- terranea (Herbst, 1794) (see: Williams 1984 for synonymy) from which it can be immediately separated on the basis of its dense tuberculate covering alone. Fossil comparisons: Persephona niemey- eri most closely resembles P. rodesae de- scribed below from which it can be sepa- rated as follows: 1.) Persephona niemeyeri has far more dorsal tubercles than P. ro- desae, especially anteriorly on the postfron- tal, frontal, and orbital surfaces; 2.) these tubercles are much higher and more acute than those of P. rodesae; 3.) the densely tuberculate hepatic region of P. niemeyeri differs considerably from that of P. rodesae which possesses only a few tubercles; and 4.) the lateral row of tubercles just above the plural suture of P. niemeyeri is poorly separated from the dorsal tubercles with which it inner grades, where as this row in P. rodesae is well separated from the dorsal tubercles. Specimens of isolated “‘arms”’ referred to P. punctata (Linné, 1758) by Rathbun (1935:106), from her Virginia localities will probably, in time, be shown to represent P. niemeyeri and/or P. rodesae. The right pos- terior spine of paratype (USNM 520712) was broken off prior to photography but has since been restored. The barren surfaces seen on the figure of the holotype (USNM 520710) of P. niemeyeri are the apparent result of abrasion on this gerontic specimen. Remnants of the dense covering of these tubercles and bolitimorphs can be seen on the actual specimen. Persephona rodesae, new species Fig. 5B, D Diagnosis.—Carapace hemispherical with three posterior spines; front bilobate, elevated, produced with orbits well beyond anterolateral margin; post frontal surface barren of tubercles, hepatic region and all other dorsal surfaces sparsely covered with relatively low, small, obtuse to subacute tu- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON bercles of subequal size; tubercles strongest along lateral and posterior margins, and very small to minute where highly concen- trated on distal surface of front, and around orbits. Description.—Carapace hemispherical, width about nine-tenths of length, regions very weakly defined, front elevated, pro- duced with orbits well beyond anterolateral margin; posterior with three equally spaced recurved spines, two marginal, and one me- dian above margin. Dorsal surfaces sparsely covered with relatively low, blunt to sub- acute tubercles of subequal size; tubercles strongest along lateral and posterior mar- gins, most concentrated below plural suture, where very low, rounded and closely spaced; very small to minute tubercles highly concentrated on front and orbits where some are developed into small boli- timorphs; post frontal surfaces and adjacent areas behind orbits nearly barren, void of tubercles. Front very narrow, width about 0.11 width of carapace, bilobate, lobes sep- arated by distinct v-shaped sulcus, laterally defined by low ridge formed by raised clus- ters of tubercles; anterior margins of lobes blunt, denticulate. Orbit very small, divided by three deep, narrow, open sutures creating four lobes. Minute, elongate tubercles span sutures from both sides, creating a pectinate appearance. Supraorbital eave and median lobe meet at a steep angle to form a broad v as seen from above. A few very small bolitimorphs are present along margin of outer lobe and just posterior to orbit along anterolateral margin. Hepatic region slight- ly swollen, weakly defined, surface with only a few tubercles. Subhepatic region not preserved in type series. Anterior and pos- terior lateral margins of carapace continu- ous, defined by a distinct row of larger tu- bercles beginning at orbit and continuing along upper edge of plural suture to a point just above tooth of posterolateral angle; row of tubercles usually well ordered and well separated from dorsal tubercles, particularly anteriorly; posteriorly row may inner grade with some dorsal tubercles and appear as a VOLUME 116, NUMBER 1 double row; area below plural suture cov- ered with pavement of closely spaced low tubercles, surface appearing very rough. Posterior margin slightly more than one- third carapace width, produced posteriorly, laterally defined by a broad, stout, triangu- lar, acute recurved spine at each corner; margin between spines comprised of from seven to 11 large recurved tubercles; intes- tinal margin above with a single medially placed, stout, recurved spine. Spines coarsely granulate. Posterior gastric pits not evident. Color.—Pale yellow when dry. Etymology.—This species is named for Mary Betty Rodes, science teacher, de- ceased, formerly of Portsmouth, Virginia, in recognition of her contribution to science through her many years of teaching science in the public schools of tidewater Virginia. In addition Miss Rodes provided the author with his first formal introduction to pale- ontology. Holotype.—Rices Pit, Hampton, Virgin- ia. USGS 26893, (USNM 520714) cl 27.05 mm, cw 23.86 mm. Paratypes: (USNM 520715) cl 32.05 mm, cw 28.39 mm, USGS 26893 as above; (USNM 520716) cl 32.60 mm, carapace incomplete, cw 35.50 mm, USGS 26893 as above; (USNM 520717) cl 15.47 mm, cw 16.41 mm, fragment of an- terior portion of carapace only, USGS 26893 as above; (VMNH I[3550) cl 27.88 mm, cw 24.95 mm, same geographic local- ity as USGS 26893 above. Occurrence.—Rices Pit, Hampton, Vir- ginia. Remarks.—Recent comparisons: Perse- phona rodesae appears most similar to the Recent Western Atlantic purse crab, P. mediterranea (Herbst, 1794) (see: Williams 1984 for synonymy) from which it differs in having: 1.) relatively high, obtuse to sub- acute tubercles, compared to the very low rounded tubercles of P. mediterranea; 2.) small to minute tubercles concentrated on its front and orbits unlike P. mediterranea where these areas are all but barren; 3.) fewer, though stronger tubercles on its he- 181 patic region; 4.) fewer, though stronger tu- bercles along its lateral margins; 5.) a much rougher surface below its lateral margins; and 6.) far fewer and much stronger tuber- cles along its posterior border. Fossil comparisons: Persephona rodesae most closely resembles P. niemeyeri de- scribed above. See remarks for P. niemeyeri for comparison. Persephona rodesae and P. niemeyeri have not been found at the same locality or in similar sediments. Persephona rodesae has been found in the shelly sands of Rices Pit, a much calmer, much less violent en- vironment during its deposition than that seen at the Chuckatuck Bar where P. nie- meyeri 1s found. The morphological differ- ences between these two species therefore appear directly related to these environmen- tal differences. Family Majidae Subfamily Inachinae Genus Euprognatha Stimpson, 1871 Euprognatha ricei, new species Fig. 6 Diagnosis.—Carapace pyriform in out- line, surface with five unusually long nar- row blunt spines, mesogastric and cardiac spines longest, surface between base of me- sogastric spine and protogastric regions ex- hibiting a distinct, pentagonal array of gran- ulations mimicking the shape of a royal crown; intestinal spine present, well devel- oped; interantennular spine absent. Description.—Carapace pyriform in out- line; fronto-orbital region strongly project- ing, narrow, occupying less than one-fourth width of carapace; regions well defined, moderately swollen. Front bidentate, pro- jecting; projections represent superior sur- face of ““shoe-shaped”’ antennual fossae be- low; projections broadly triangular or ob- tuse distally, divided by a shallow sulcus extending posteriorly where terminating at an inverted v-shaped row of eight granules. Orbit circular, margins vertical, subparallel; supraorbital eave covered with triangular 182 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. field of granules posteriorly; point of field distal where defined by a strong tubercle on orbital margin. Eyes apparently on long stalks, resting within depression of postor- bital spine. Postorbital spine very strong; broad basally, tapering to a point distally, obliquely directed forward. Basal antennal article with inner and outer rows of gran- ules of which several are developed into short, obtuse spines (tubercles); article di- rected only slightly obliquely forward, ter- minating in a granulated spine subequal in advancement to the front. Antennal sternum weak, not produced; interantennular spine absent. Outer margin of pterygostomian re- gion with rows of coarse granules; superior row with large blunt tooth posteriorly. Buc- cal cavity broadest anteriorly, lateral mar- gins divergent, width about one-third width of carapace. Hepatic region conical, sparse- ly covered with rounded granules or small tubercles, terminating obliquely in a strong blunt spine; spine produced laterally to a distance subequal to advancement of post orbital spine. Mesogastric, cardiac, intesti- Euprognatha ricei Holotype (USNM 520718) A. right lateral view. Scale = 2 mm. B. ventral closeup of antennule fossae, and long basal segment of antenna. Scale = 1 mm. C. dorsal view. Scale = 2 mm. nal, and branchial regions each surmounted with a single prominent cylindrical blunt spine; mesogastric spine longest, vertically directed; cardiac spine subequal to meso- gastric spine, slightly directed posteriorly; intestinal and branchial spines much shorter and of subequal length; the former oblique- ly directed posteriorly; latter obliquely di- rected laterally. Spines smooth laterally with their distal ends granulated. Surface of carapace noticeably punctate, otherwise rel- atively smooth except for concentrations of low tubercles or rounded granules along frontal margins, supraorbital eaves, postor- bital spines, hepatic regions, lateral mar- gins, and bases of prominent spines. Irreg- ular concentrations of tubercles and gran- ules appearing as a distinct anteriorly di- rected row on each branchial region; row beginning just posterior to prominent spine of branchial region and converging anteri- orly where tapering to a point at branchial- hepatic junction. Surface between base of mesogastric spine and protogastric regions defined by a very distinct pentagonal array VOLUME 116, NUMBER 1 of granulations arranged in the shape of a royal crown. Tubercles strongest on or near hepatic and anterolateral margins. Posterior gastric pits distinct, close set at base of me- sogastric region, each bounded posteriorly by cluster of three granules. Distinct broad, smooth, groove separating cardiac and in- testinal regions from branchial region. Branchial region posterolaterally pro- duced into a relatively wide, flattened, gran- ulated margin. Posterior margin moderately produced posteriorly, width about one-third carapace width. Color.—Ash white when dry, or gray when wet. Etymology.—This species is named in honor and appreciation of Mr. William M. Rice, deceased, of Hampton, Virginia, who allowed the author unrestricted access to his borrow pit and who, with his wife, Made- line, and family built a museum on his property for the purpose of educating stu- dents in the Hampton Roads area. This small museum, known as The Kenneth E. Rice Memorial Museum (after his youngest son), the borrow pit behind it, and the tu- torage of Mr. Rice served more than five thousand students of all ages each year. His positive impact on the paleontology of the area iS unsurpassed. Holotype.—Rices Pit, Hampton, Virgin- ia. USGS 26893, (USNM 520718) cl 10.31 mm, cw 9.01 mm. Paratypes: (USNM 520719) cl 4.41 mm, cw 3.46 mm, USGS 22209 same geographic locality as USGS 26893 above; (VMNH [3551) cl 7.75 mm, cw 6.83 mm, USGS 26893 as above. Occurrence.—Rices Pit, Hampton, Vir- ginia. Remarks.—Recent comparisons: Of the known living species of Euprognatha, E. riceél appears most similar in overall form, number and placement of prominent dorsal spines and distribution of tubercles and granules to E. gracilipes A. Milne-Ed- wards, 1878b (Florida Keys to Barbados, Rathbun 1925). Euprognatha ricei can however be easily separated from E. gra- cilipes as follows: 1.) the prominent spines 183 of E. ricei are much longer than those of E. gracilipes and E. ricei unlike E. gracilipes possesses an intestinal spine; and 2.) the dorsal surface of FE. ricei is much less gran- ulate than the evenly, coarsely granulated sutace of 22 racilipes. ‘The presence of long mesogastric, branchial, and cardiac spines, and in particular a long intestional spine in combination with its sparsely gran- ulated surface, the absence of an interanten- nular spine and the presence of a pentago- nal array of granules in the shape of a royal crown on its mesogastric region easily sep- arate FE. ricei from all other living species of Euprognatha. Fossil comparisons: Rathbun (1935) identified two left dactyls from the Miocene of Liberty County, Florida as Euprognatha sp., p. 112, pl. 24, figs. 16-19. Lacking as- sociated chela, the type specimens of E. ri- cei cannot be compared with Rathbun’s ma- terial. The distal end of the left basal antennal article was broken and lost by the author in handling the specimen after the photograph was completed. Subfamily Mithracinae Genus Stenocionops Desmarest, 1823 Stenocionops dyeri, new species Piss Diagnosis.—Carapace oblong-ovate, strongly arched anteriorly; superior lateral marginal spines three; frontal depression laterally bordered by 10 to 18 small spines; cardiac region greatly elevated, armed with an irregular pentagonal array of five short spines. Description.—Carapace oblong-ovate, strongly arched anteriorly, surface uneven, regions well defined. Length from rostral notch along middorsal line from 1.33 to 1.41 times greatest width, broadest across middle of branchial regions; greatest height approximately one-half that of width. Ros- tral horns basally divergent, laterally sub- parallel, short, flattened, slightly upturned distally. Frontal depression deep, ovate, lat- 184 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ice. 10 mm. erally bordered by 10 to 18 small spines decreasing in height anteriorly. Fronto-or- bital region broad, more than one-half times that of width; orbits strongly projecting. Eyes small, retractable within deep tubular orbits; preocular spine superior, a small dis- tinct spine or tubercle medial to its base; postocular spine smaller, acute. Orbit com- pleted below by a blade-like lateral expan- sion of basal antennal article. Basal anten- nal article broader than long, armed with 3 spines, One anterior-external, short, acute, barely visible dorsally; another, posterior- external, usually very weak; and last, an- terior-internal, strong, acute, extending for- ward and obliquely downward from basal insertion of second antennal article. Second antennal article not extending forward to or beyond rostral notch. Opening of green gland oval, noticeably raised, a prominent, obliquely downward-directed, acute spine near its external-lateral border. Pterygosto- Stenocionops dyeri Holotype (USNM 520720) A. dorsal view and B. left lateral view. Scale = mian region triangular, armed with two small outer marginal spines; anterior lon- gest. Inner margins raised, rim like, anteri- orly terminated by an anvil-like process. Buccal frame broader than long, widest an- teriorly; width about one-third width of car- apace width. Superior lateral marginal spines three; one hepatic and two branchial. Secondary lateral marginal spines common, may exceed ten, their number apparently in- creasing with age. Hepatic region swollen, very distinct, reaching beyond fronto-orbit- al width, armed laterally with a cluster of five or more spines; superior inner margin with two spines, posteriormost longest and strongest. Mid-dorsal line slightly raised, ridge-like. Gastric region greatly arched an- teriorly; armed with from eight to 10 short spines, of which four are usually mesogas- tric, two metagastric, and three urogastric. Posterior gastric pits faint, bordered by VOLUME 116, NUMBER 1 curved row of spines just distal to cervical groove. Cardiac region greatly elevated, usually armed with a characteristic irregular pen- tagonal array of five small spines, posteri- ormost two, often close set. Cardio-intesti- nal and metabranchial regions separated by deep furrows. Intestinal region armed with two distinct spines along midline, posteri- ormost largest; spines aligned perpendicular to a lower horizontal row of from two to five smaller spines and/or tubercles, their number apparently decreasing with age; row of spines bordered below by wide, shallow sulcus. Posterior margin rounded, narrow, slightly projecting, width less than one-half fronto-orbital width. Branchial re- gions uneven, moderately spinous. A dis- tinct, armed, broadly conical prominence on each epibranchial, mesobranchial and me- tabranchial region, which when roughly aligned with rostral notch, forms an invert- ed V of approximately 30°. Epibranchial prominence usually with two secondary spines or tubercles. A prominent ridge armed with three low spines, posterior lon- gest, occupying inter-lateral margin of me- tabranchial region. Lower margin of entire carapace termi- nating in a rounded rim. Entire surface of body (chelipeds—excepting distal half of fingers) closely covered with minute, cir- cular pores. Coxa, basis, and ischium of chelipeds and ambulatory legs, unarmed. Merus of cheliped quadrangular in section, armed with four distinct lateral rows of spines, the dorsal row being most prominent with four Superior spines and as many as five smaller spines or tubercles. Carpus smooth. Hand elliptical in section, highest proximally, smooth to finely granulate. Fingers less than one-half length of palm, not gaping. Am- bulatory legs circular in section; merus of first and second pair moderately tuberculate above; third and fourth pairs smooth. Legs decreasing in length posteriorly. Color.—Ash white with faint patches of pink when dry, gray when wet. Some living 185 species of Stenocionops are dark red, such as Stenocionops furcata coelata (A. Milne- Edwards, 1878a). See Williams (1984:339). Etymology.—This species is named for Brian J. Dyer, microbiologist, Old Domin- ion University, Norfolk, Virginia for his contribution to science through his years of research and teaching and especially for his years of encouragement and field assistance to the author. Holotype.—The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. USGS 26891, (USNM 520720) cl including ros- trum 58.38 mm, excluding rostrum 54.47 mm, cw including spines 44.65 mm, ex- cluding spines 40.03 mm. Paratypes: (USNM 520721) Female: cl including ros- trum 53.91 mm, excluding rostrum 49.96 mm, cw including spines 41.70 mm, ex- cluding spines 37.72 mm, right propodus: pri 21.63 mm, prh 6.09 mm, prl 3.15 mm, left propodus, prl 20.96 mm, prh 5.13 mm, prt 3.22 mm, USGS 26891 as above; (VMNH 13552) cl including rostrum 42.95 mm, excluding rostrum 39.21 mm, cw in- cluding spines 32.22 mm, excluding spines 27.75 mm, same geographic locality as USGS 26891 above. Occurrence.—The Lone Star Lakes, near Chuckatuck, Suffolk, Virginia. Remarks.—Recent comparisons: Steno- cionops dyeri appears most similar in shape and overall character to S. spinosissima (Saussure, 1857) from which it can be eas- ily separated as follows: 1.) Stenocionops dyeri has three strong lateral spines com- pared to the five found in S. spinosissima; 2.) S. dyeri has short, flattened rostral horns compared to the much longer, more acute horns of S. spinosissima; and 3.) S. dyeri has between 13 to 15 short median spines, of which those on the cardiac region are arranged in an irregular pentagonal array in contrast to S. spinosissima which exhibits 10 median spines of which only two are found on the cardiac region. Stenocionops dyeri also superficially resembles a much smaller species of Stenocionops, S. trian- gulata (Rathbun, 1892) which also possess- 186 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON es three lateral spines. These spines in S. dyeri, however, are shorter and much less acute. Stenocionops dyeri also differs from this species in having: 1.) much flatter, shorter and less divergent rostral horns; 2.) 13 to 15 median spines compared to about nine on S. triangulata; and 3.) an irregular pentagonal array of five spines on its car- diac region compared to only one or two on this region in S. triangulata. In reference to S. triangulata, Rathbun (1925:461) states: ‘“Known only from small and immature specimens’’. The pentagonal array of spines on the cardiac region of S. dyeri alone sets it apart from all other known species of Stenocionops. Fossil comparisons: Rathbun (1935) de- scribed Stenocionops primus from the Up- per Cretaceous Brownstown Formation, of Howard County, Arkansas and Stenocion- ops suwanneeana from the Eocene Ocala Limestone of Suwannee County, Florida. The chela of these two species may indeed resemble those of adult males of Stenocion- ops or other mature majid chela but in size and coarse granulation they are very differ- ent from the nearly smooth diminutive fe- male chela of S. dyeri. The chela in both adult male and female species of some liv- ing representatives of Stenocionops are rel- atively small, like those of S. dyeri, but in reaching maturity the chela in gerontic males can become quite large. The geologic time difference between S. dyeri and these geologically much older taxa is such that it would be very unlikely that they represent the same species. The barnacle, Balanus sp., is often found completely covering the dorsal surface of the carapace of S. dyeri where it apparently lived as a symbiont during the life of the crab. This barnacle has not yet been found covering the orbital openings of this crab, thus revealing that the crab was indeed alive and able to keep its orbits clear during this relationship. The massive numbers of this barnacle found on some specimens of S. dyeri greatly increased the crab’s surface area and undoubtedly contributed to its be- ing washed about and buried in the Chuck- atuck Bar deposits by strong currents dur- ing storms. These same barnacles also strengthened the dorsal aspect of the cara- pace allowing it to be preserved in this high energy environment of deposition. Acknowledgments Mr. H. B. Roberts, deceased, formerly of the Department of Invertebrate Zoology, NMNH, SI, correctly identified Hepatus bottomsi, Persephona rodesae, Euprogna- tha ricei and Stenocionops dyeri as new in a letter to the present author, dated 13 Oc- tober 1964. Given the incomplete and often fragmentary nature of the material present- ed to him at that time, his identifications are remarkable for their accuracy, a testament to his knowledge and ability. Due to his de- clining health, Mr. Roberts was unable to pursue descriptions of these and other new taxa (see: Manning & Blow 1980, Blow & Bailey 1992) and passed that task on to this author before his retirement in June 1973. He died 14 March 1979. I thank E. E. Bottoms for introducing me to a number of fossil decapod localities in the Hampton Roads area in the fall of 1962, and for years of cooperative fieldwork and mutual assistance. I particularly thank my parents, R. V. and C. D. Blow, for years of logistical support, field assistance, and for their own collecting efforts which have benefitted and continue to benefit my re- search these past 40 years. Mr. W. A. Stay- lor, Mr. H. EK Saunders and the other men and women of the Lone Star Cement Co. kindly allowed the author unrestricted ac- cess to their diggings. R. H. Bailey, B. A. Bedette, B. J. Dyer, L. W. Ward, and the late D. Wilson helped the author in the field on a number of occasions, adding their ex- pertise and collections to the research and writing of the present paper. L. D. Campbell and Svein Nielsen kindly made their col- lections available to the author for study. I thank S. Whittaker, of the SEM lab, Smithsonian Institution, for providing the VOLUME 116, NUMBER 1 photomicrographs of P. baileyi, M. G. Har- asewych for providing the digital images of the remaining taxa except for the dorsal and lateral views of E. ricei, which were taken by Mr. T. FE Phelan. My sincere thanks to J. A. Sanner for enhancing the images in Ado- be Photoshop and constructing the finished digital plates. In addition B. A. Bedette read the manuscript and offered numerous sug- gestions and corrections. T. R. Waller pro- vided space in his laboratory, and scientific and editorial advice throughout the prepa- ration of this manuscript. Literature Cited Bell, T. 1855. A Monograph of the Leucosiade, with observations on the relations, structure, habits, and distribution of the family; a revision of the generic characters; and descriptions of new gen- era and species, Hore Carcinologice, or Notic- es of Crustacea, I—Transactions of the Linnean Society 21:277—314 + pls. 30—34. Blow, W. C., & R. H. Bailey. 1992. Chasmocarcinus robertsi, a new crab species from the Miocene of Virginia, with notes on the genus Falcono- plax (Crustacea, Decapoda, Goneplacidae).— Tulane Studies in Geology and Paleontology 25(4):175—185. Campbell, L. D. 1993. Pliocene molluscs from the Yorktown and Chowan River Formations in Virginia.—Virginia Division of Mineral Re- sources Publication 127:1—259. Collins, J. S. A., & S. EF Morris. 1976. Tertiary and Pleistocene crabs from Barbados and_ Trini- dad.—Palaeontology 19(1):107—131. Desmarest, A. G. 1823. Malacostracés. Pp. 158-425 in Dictionnaire des Sciences Naturelles, vol. 28. Strasbourg. Gibson, T. G. 1967. Stratigraphy and paleoenviron- ment of the phosphatic Miocene strata of North Carolina.—Geological Society of America Bul- letin 78(5):631—650, pls. 1, 2. Guinot, D. 1979. Morphologie et phyogenése des Brachyoures.—Mémoires du Muséum national d’Histoire naturelle (Paris), new series A (zo- ology) 112:1—354. Haj, A. E.. & R. M. Feldmann. 2002. Functional mor- phology and taxonomic significance of a novel cuticular structure in Cretaceous raninid crabs (Decapoda: Brachyura: Raninidae).—Journal of Paleontology 76(3):472—485. Hazel, J. E. 1971la. Ostracode biostratigraphy of the Yorktown Formation (upper Miocene and lower Pliocene) of Virginia and North Carolina.— 187 Geological Survey Professional Paper 704:1— 13. 1971b. Paleoclimatology of the Yorktown Formation (upper Miocene and lower Pliocene) of Virginia and North Carolina: Pp. 361—375 in H. J. Oertli, ed., Paléoécologie Ostracodes, Pau, 1970.—Bulletin Centre Recherche Pau-SNPA, 953 pp., 3 tab. Herbst, J. EF W. 1794. Versuch einer Naturgeschichte der Krabben und Krebse, nebst einer systema- tischen Beschreibung ihrer verschiedenen Arten 2(5):147-162 + pls. 37—40. Johnson, G. H. 1969. Guidebook to the geology of the York-James Peninsula and south bank of the James River—College of William and Mary Department of Geology Guidebook No. 1. At- lantic Coastal Plain Geological Association, Tenth Annual Field Conference and First An- nual Geological Field Conference: 1—33. , & N. K. Coch. 1969. A Coquina facies in the Yorktown Formation near Chuckatuck, Virginia and its geological implications.—(Abs.) Geo- logical Society America Special Paper 121: 448. Kier, P. M. 1972. Upper Miocene echinoids from the Yorktown Formation of Virginia and their en- vironmental significance.—Smithsonian Contri- butions to Paleobiology 13:1—41. Smithsonian Institution Press, Washington. Latreille, P. A. 1802. Histoire naturelle générale et par- ticuliére des Crustacés et des Insects 3:13—467. Paris. Leach, W. E. 1817. Malacostraca Podophthalma Bri- tanniz; or descriptions of the British species of crabs, lobsters, prawns, and of other Malacos- traca with pedunculated eyes. James Sowerby, London, XIV, 5 p., unpaged + pls 16, 25 and 44. Linnaeus, C. 1758. Systema naturae per regna tria na- turae, secundum classes, ordines, genera, spe- cies, cum characteribus, differentii, synonymis, locis, 10th edition. Laurentius Salvius, Stock- holm 1:1—824. Manning, R. B., & W. C. Blow. 1980. Henry B. Rob- erts 1 September 1910—14 March 1979.—Crus- taceana 39(1):104—107, pl. 1. Mansfield, W. C. 1943 [1944]. Stratigraphy of the Mio- cene of Virginia and the Miocene and Pliocene of North Carolina. Pp. 1-19 in J. Gardner, ed.., Mollusca from the Miocene and Lower Pliocene of Virginia and North Carolina: Pt 1, Pelecy- poda.—U.S. Geological Professional Paper 199-A:1—178. Milne-Edwards, A. 1865. Description de quelques Crustacés nouveaux on peu connus de la famille Survey des Leucosiens.—Annales Société Entomolo- gique de France 4(5):148—159, pl. 6. . 1878a. Note sur quelques Crustacés nouveaux 188 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON appartenant au groupe des Oxyrhynques.—Bul- letin de la Société Philomathique de Paris 7(2): 222-225. . 1878b. Etudes sur les Xiphosures et les Crus- tacés de la Région Mexicaine.—In Mission Scientifique au Mexique et dans l’Amerique centrale part 5(4):121—184, pls. 21—27, 29-30. . 1880. Etudes préliminaires sur les Crustacés, l ere partie. In reports on the results of dredging under the supervision of Alexander Agassiz, in the Gulf of Mexico, and in the Caribbean Sea, 1877, °78, °79 by the U.S. Coast Survey Steam- er “Blake,” Lieut. Commander C. D. Sigsbee, U.S.N., and Commander J.R. Bartlett, U.S.N., Commanding.—Bulletin of the Museum of Comparative Zo6dlogy at Harvard College 8(1): 1—68, pls. 1, 2. , & E. L. Bouvier. 1894. Brachyures et An- omoures. /n Crustacés décapodes provenant des campagnes du yacht I’Hirondelle (1886, 1887, 1888), Premiere Partie. Résultats des Campag- nes Scientifiques accomplies sur son yacht par Albert 1 er, Prince Souverain de Monaco 7:1— 112, pls. 1-11. , & E. L. Bouvier. 1902. Les Dromiacés et Ox- ystomes. Jn reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-78), in the Caribbean Sea (1978-79), and along the Atlantic Coast of the United States (1880), by the U. S. Coast Survey Steamer “Blake”, Lieut. Com. C. D. Sigsbee, U. S. N., and Commander J. R. Bart- lett, U. S. N., Commanding, XXXIX.—Mem- oirs of the Museum of Comparative Zodlogy at Harvard College 27(1):1—127, pls. 1-25. Milne-Edwards, H. 1834. Histoire naturelle des Crus- tacés, comprenant |’anatomie, la physiologie et la classification de ces animaux 1:1—468. Paris. Pennant, T. 1777. Crustacea, Mollusca, Testacea. Brit- ish Zoology, edition 4, 4:London, 1—154, pls. 1-93. Rathbun, M. J. 1892. Catalogue of the crabs of the family Periceridae in the U.S. National Muse- um.—Proceedings of the United States National Museum 15(901):231—277, pls. 28—40. . 1896. The genus Callinectes.—Proceedings of the United States National Museum 18(1070): 349-375, pls. 12—28. . 1925. The spider crabs of America.—United States National Museum Bulletin 129:1—613, pls. 1-283. . 1926. The fossil stalk-eyed Crustacea of the Pacific slope of North America.—United States National Museum Bulletin 138:1—155, pls. 1— 39), . 1935. Fossil Crustacea of the Atlantic and Gulf Coastal Plain.—Geological Society of America, Special Papers 2:1—160, pls. 1—26. . 1937. The Oxystomatous and allied crabs of America.—United States National Museum Bulletin 166:1—278, pls. 1—86. Saussure, H. de. 1857. Diagnoses de quelques Crus- tacés nouveaux de |’Amérique tropicale.—Re- vue et Magzin de Zoologie Pure et Appliquée, 2(9):501—505. Serene, R. 1954. Sur quelques especes rares de Brach- yures (Leucosidae) de Il’ Indo-Pacifique.—Treu- bia 22(3):453—499, pls. 7—10. Stimpson, W. 1859. Notes on North American Crus- tacea, No. 1.—Annals of the Lyceum of Natural History of New York 7(1862)(2):49—93, 1 plate. . 1871. Preliminary report on the Crustacea dredged in the Gulf Stream in the Straits of Florida, by L. E de Pourtales, Assistant United States Coast Survey. Part 1. Brachyura.—Bul- letin of the Museum of Comparative Zoology at Harvard College 2(2):109—160. Verrill, A. E. 1869. On the parasitic habits of Crusta- cea.—American Naturalist 3:239—250. [Hepa- tella amica Smith described as footnote p. 250] Ward, L. W., & B. W. Blackwelder. 1980. Stratigraphic revision of Upper Miocene and Lower Pliocene beds of the Chesapeake Group, Middle Atlantic Coastal Plain.—Contributions to Stratigraphy, Geological Survey Bulletin 1482-D:1—61, pls. 1-5. , & N. L. Gilinsky. 1993. Molluscan assem- blages of the Chowan River Formation, Part A, Biostratigraphic analysis of the Chowan River Formation (Upper Pliocene) and adjoining units, The Moore House Member of the York- town Formation (Upper Pliocene) and the James City Formation (Lower Pleistocene).—Virginia Museum of Natural History Memoir 3, Part A: 1-32, plate 1. Tables 1, 2, & 4-6 & fig. 6 not paginated (in pocket). Williams, A. B. 1965. Marine decapod crustaceans of the Carolinas.—Fishery Bulletin 65(1):1—298. . 1984. Shrimps, lobsters, and crabs of the At- lantic Coast of the eastern United States, Maine to Florida. Smithsonian Institution Press, Wash- ington, D.C., 550 pp. Appendix | Locality register USGS 26891 The Lone Star Lakes, formerly Lone Star Cement Company open-pit mine, about 1 km north of Chuckatuck, Suffolk, Virginia. USGS Chuckatuck, 72-minute quadrangle map, 1965 pho- torevised 1979. USGS 26892 Open-pit mine, locally known as Rid- dick Pit, about 4.4 km southeast of Benns Church, on east side of Va. route 10/32, Isle of Wright Coun- ty, Virginia. USGS Benns Church 7%-minute quad- rangle map, 1965 photorevised 1986. VOLUME 116, NUMBER 1 189 USGS 26893 Rices Pit belonging to Mr. William M. USGS 22209 Same geographic locality as 26893 Rice, about 0.4 km north of intersection of Fox Hill above but collected by a large group of Smithsonian Road (Rte. 167) and Harris Creek Road, Hampton, scientists lead by G. A. Cooper in 1966. Virginia. USGS Hampton, 7%-minute quadrangle map, 1965 photorevised 1986. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):190—197. 2003. Columbasellus acheron, a new genus and species of subterranean isopod from Washington (Crustacea: Isopoda: Asellidae) Julian J. Lewis, Joel W. Martin, and Regina Wetzer (JJL) J. Lewis & Associates, Biological Consulting, 217 W. Carter Avenue, Clarksville, Indiana 47129, U.S.A. (JWM & RW) Research & Collections Branch, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007, U.S.A. Abstract.—Columbasellus acheron, new genus, new species, is described from a male specimen discovered during the cleaning of the municipal water well of Kalama, Washington. The genus Columbasellus is geographically clos- est to the genera Caecidotea, Calasellus and Salmasellus, but is morphologi- cally more similar to the Eurasian Asellus aquaticus/hilgendorfii complex of species, with which it shares a male pleopod 2 that possesses a relatively large labial spur and basal spur, and fully formed, ovate exopods in pleopods 4 and 5. Columbasellus acheron inhabits the hyporheic zone of the Kalama River. Evidence is presented that indicates widespread use of stream gravel interstices by subterranean isopods in the Pacific Northwest, including the first record of Salmasellus howarthi from Oregon. In the eastern United States there are about 65 described species of subterranean asellid isopods, mostly known from caves (Henry et al. 1986, Lewis & Bowman 1995), although a significant number have now been described that apparently live in soil or gravel interstices (Lewis 1982, 2000; Lewis & Bowman 1981; Lewis & Holsin- ger 1985). The situation is different west of the Rocky Mountains, where caves and karst are much less commonplace and only five subterranean asellids have been de- scribed: Caecidotea sequoiae Bowman, 1975, Calasellus californicus (Miller, 1933), C. longus Bowman, 1981, Salma- sellus steganothrix Bowman, 1975 and S. howarthi Lewis, 2001. It is becoming increasingly apparent that subterranean isopods are more widespread in western North America than the number of described species might indicate. We add to the growing body of evidence (Lewis 2001) that subterranean isopods are com- mon in the hyporheic zone of western streams with the description of Columba- sellus acheron, the first record of Salma- sellus howarthi from Oregon, and several other records of undetermined eyeless, un- pigmented isopods. It is with some chagrin that we make this addition by describing a new genus and species of subterranean isopod that was dis- covered during the process of trying to eradicate it. The city of Kalama, Washing- ton, currently obtains groundwater from sands and gravels below the Kalama River, which is located approximately 3.2 km east of the confluence of the Kalama and Co- lumbia rivers near the Modrow Bridge (Fig. 1). The water withdrawal is accomplished via a Ranney well collection system that consists of three horizontal pipes extending below the Kalama River at a depth of five meters. The isopods were flushed from the municipal well during the process of clean- ing it with superchlorination, scouring and vacuuming. Of dozens of isopods emerging from the well, only one was collected. This single specimen is recognized here as a unique subterranean isopod flushed from its hyporheic habitat. VOLUME 116, NUMBER 1 191 Columbasellus acheron n. sp. Location of other asellids (see text) Salmasellus howarthi e>D* Cities iS Rivers and water bodies | eamcenen| . _ State boundaries Sa ; a. \ \ a —< oA) SS a = as )} Be ¢ “~ &x~f~ Portland es AF PAL [ne —_ ¢ Dak s } TS, >) [ a } 3 ee A {ic (a a we cay er Sars 480 Miles Bis. 1. g Map of the Pacific northwest showing collection localities of subterranean asellid isopods. Family Asellidae G. O. Sars, 1897 Columbasellus, new genus der, palm without processes; dactyls 1-7 with accessory unguis. Male pleopod | with reti- Diagnosis.—Eyeless, unpigmented. Head without rostrum or lateral incisions. Mandible with 3-segmented palp, 4-cuspate incisors and lacinia mobilis. Maxilla 1 inner lobe with 5 apical setae. Pereopods, coxae visible in dorsal view; male pereopod | propodus slen- nacula present. Male pleopod 2 endopod with digitiform basal spur, endopodial groove orig- inating from within basal part of endopod, labial spur present, cannula tapering to a sub- triangular stylet; exopod with catch lobe. Ple- opod 3 exopod with transverse suture. Pleo- 192 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pods 4 and 5 with well formed ovate exopods and endopods. Morphology and setation pattern of the mandible, maxilla 2, pereopods, uropods, essentially identical to other previously de- scribed asellids (e.g., see figures by Wil- liams 1970, Bowman 1981). Type-species.—Columbasellus acheron. Etymology.—**Columb” from Columbia River + “‘Asellus’’. Relationships.—Columbasellus is geo- graphically closest to Caecidotea from U.S. Pacific coastal states (Williams 1970; Bow- man 1974, 1975), Calasellus two species from California (Bowman 1981), and Sal- masellus two species in Washington and Al- berta (Bowman 1975, Lewis 2001). How- ever, Columbasellus is morphologically more similar to the Eurasian Asellus aqua- ticus/hilgendorfil assemblage discussed by Henry and Magniez (1995). These authors analyzed the Pacific Rim asellids and found anatomical similarities that suggested a common ancestor for the following genera: Asellus, Calasellus, Mesoasellus, Nippona- sellus, Phreatoasellus, Sibirasellus, and Uenasellus. These genera have the follow- ing characteristic states, which may be syn- apomorphic for the group: a first maxilla with 4 or 5 setae on the inner lobe; no pro- cesses on the palmar margin of the pereo- pod 1 propodite; a male second pleopod possessing an exopodite catch lobe, endop- odite basal and labial spurs. Mesoasellus, Phreatoasellus, Uenasellus, Calasellus comprise a clade characterized by the fol- lowing shared states: first maxilla with 5 setae on the inner lobe; mandibular palp of three articles. We add Columbasellus to this clade. Within the clade, Columbasellus is distinguished by the following autapomor- phies: labial spur arising from within en- dopodial groove not from the rim of the en- dopodial groove as in Calasellus and Sibi- rasellus; cannula short and tapering, not elongate and seta-like as in Calasellus. Columbasellus is separated from Caeci- dotea and Salmasellus by the absence of basal and labial spurs in the male second pleopod of the latter genera. Basal and la- bial spurs are present in Columbasellus and Calasellus, but the structure of the labial spurs are quite dissimilar. Columbasellus has a labial spur (see Henry & Magniez 1995) arising from within the endopodial groove, whereas the labial spur of Calasel- lus consists of a tiny cylindrical structure on the rim of the endopodial groove similar to that of Sibirasellus. The cannula in Col- umbasellus is a relatively short, tapering subtriangular stylet, while in Calasellus it is a Seta-like filament that traverses most of the length of the endopod. Bowman (1981) believed that Calasellus was related to Asellus (Phreatoasellus) Matsumoto, 1962 (five species in Japan, now elevated to generic status), which agrees with the later synthesis of Henry and Magniez (1995). Columbasellus and Cala- sellus are separated from Phreatoasellus by the dissimilar male pleopod 2 and by the absence of an accessory unguis on the pe- reopod dactyls. The pleopod 4 exopod of Phreatoasellus is ovate in P. higoensis and P. kawamurai, while pyriform in P. uenoi, P. iriei and P. minatoi (Matsumoto 1960, 1962, 1978). The pleopod 5 exopod is well developed in all Phreatoasellus. Columba- sellus is further separated from Calasellus by the oval pleopod 4 exopod (pyriform in Calasellus) and by the presence of a well formed pleopod 5 exopod (rudimentary or absent in Calasellus). Columbasellus acheron, new species Figs. 2—4 Material examined.—Washington, Cow- litz Co., Kalama, Ranney Well (approxi- mately 46.00N, 122.82W), about 5 m below Kalama River, collected May 2000, Carl McCrary, | male. This 14.5 mm male is the holotype and has been deposited in the col- lection of the Natural History Museum of Los Angeles County (LACM CR 2000- ONBoIO), Etymology.—Refers to the River Acher- VOLUME 116, NUMBER 1 Jane PAs 193 wex d Columbasellus acheron, holotype male. a, habitus; b, antenna 1; c, left mandible, incisor and lacinia mobilis; d, right mandibular incisor; e, mandibular palp; f, maxilla 1, inner lobe; g, same, outer lobe; h, max- illiped. on of Greek mythology, one of the rivers flowing through the underworld. Description.—Body linear, about 4.8 X as long as wide. Head trapezoidal, about 2 as wide as long, anterior margin slightly concave, postmandibular lobes weakly de- veloped. Pleotelson ovate, about 1.3X as long as wide, false sutures faintly visible on anterior dorsal surface, caudomedial lobe not produced. Antenna | reaching to mid-length of last article of antenna 2 peduncle, flagellum of 194 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ange. Be 23 articles, esthete formula equivocal due to damage, apparently 3-O0-1-0-12. Antenna 2 about 0.8X body length, basal segments slender, linear, cylindrical, specialized mod- ifications absent, flagellum of about 67 ar- ticles. Left mandible with incisor 4-cuspate and 2 minute denticles, lacinia mobilis 4-cus- pate, 16 spine-like setae in adjacent row; right incisor 4-cuspate, 17 spine-like setae in row; molars present, unremarkable; palp with rows of plumose setae on articles 2 and 3. Maxilla 1, outer lobe with 13 robust apical spine-like setae and | seta, | subter- minal plumose seta and 1 plumose seta along distal shaft, lateral margins with set- ules and minute comb spines; inner lobe with 5 stout plumose setae. Maxilla 2 with rows of setae on apical margins, unremark- able. Maxilliped with rows of stout plumose setae and 6 retinacula. Pereopod 1, propodus about 2.8 X as long as wide, palmar margin straight, with 8 spine-like setae, dactyl flexor margin with accessory unguis and 9 spine-like setae. Some pereopods missing or detached from Columbasellus acheron, holotype male. a, pereopod 1; b, pereopod 2; c, pereopod 7. holotype, but dactyls apparently all with ac- cessory unguis. Pleopod | longer than pleopod 2; proto- pod with 3 retinacula surrounded by minute comb spines; endopod subrectangular, about 2 length of protopod, setae of de- creasing length along concave lateral mar- gin, with 8 plumose setae along distolateral margin. Pleopod 2 protopod slightly longer than wide, mesial surface with numerous rows of minute comb spines; exopod prox- imal segment with 3 lateral plumose setae, distal segment with 20 plumose setae along margins, catch lobe present; endopod, dig- itiform basal spur present, mostly obscured in dorsal view; endopod produced into a ta- pering cylindrical process, curving distola- terally; endopodial groove prominent, tra- versing over half of endopod; labial spur originating from endopodial groove, ex- tending along axis of groove; cannula ex- tending into a robust, tapering stylet. Pleo- pod 3 exopod with 8 small, non-plumose setae along distal margin. Pleopod 4 exopod with | proximolateral seta and brush border VOLUME 116, NUMBER I 195 d Fig. 4. 3; e, pleopod 4; f, pleopod 5. of setules, transverse suture present. Pleo- pod 5 endopod and exopod subequal. Uropods cylindrical, linear, about 0.6 body length, 2.3 length of pleotelson; pro- topod about 0.7 length of endopod, en- dopod 1.9X length of exopod. Vernacular name.—Columbia ground- water isopod. Range.—This species is known only from the Kalama River at Kalama, in south- western Washington, just above its conflu- ence with the Columbia River, approxi- mately 55 km NNW of Portland, Oregon Gers 1). Notes on ecology.—The well from which Columbasellus acheron emerged consists of a large diameter concrete caisson that ex- tends approximately 8 meters below grade, Columbasellus acheron, holotype male. a, pleopod 1; b, pleopod 2; c, same, endopod tip; d, pleopod with three perforated lateral pipes that ex- tend horizontally approximately 5 meters below the bottom of the Kalama River. The isopods surfaced following superchlorina- tion and high pressure scouring of the lat- eral pipes. Most of the isopods settled to the bottom of the caisson and were later removed during vacuuming of the well by the contractor hired to conduct the cleaning. Although no additional isopods have been noted, the well is likely a window into the hyporheic habitat. Techniques that penetrate this environment (Pospisil 1992), e.g., Kar- aman-Chappuis for shallow stream gravel interstices or Bou-Rouch sampling of deep- er gravels with pump wells, should allow access to the isopods. It is likely that other subterranean invertebrates (amphipods, 196 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON bathynellids, etc.) occur with Columbasel- lus acheron and await discovery. Salmasellus howarthi Lewis, 2001 Material examined.—Oregon: Tillamook Co., Kilchis River drainage, Tilden Creek, 27 Sep 1998, 1 juvenile male. This species was previously known only from two lava tube cave streams in south- western Washington. This record extends the range approximately 100 km to the west. Asellid species Material examined.—Oregon: Clatsop Co., Sheewash Creek, 24 Aug 1994; Curry Co., Quail Prairie Creek near mouth, Oct 1992; Josephine Co., Howard Creek at mouth, 12 Oct 1997; Jackson Co., Crooked Crea, 13 Oct OO7s ILane €CO., Ceci Creek, 25 Sep 1999; Tillamook Co., Clear Creek at mouth, 29 Sep 1998; East Foley Creek at river mile 2.5, 31 Aug 1994; Fall Creek, 29 Aug 1996; tributary to North Fork Kilchis River, 28 Sep 1997; Tiiden Creek, 27 Sep 1998; Union Co., Limber Jim Creek at USFS Road 100, 20 Sep 1994; Wallowa Co., Chesnimnus Creek, 28 Aug 1991; Grouse Creek, 25 Aug 1992; Lick Creek, Aug 1992.—Washington: King Co., Lower Rock Creek, 15 Aug 1997; same lo- Calliiy, I Qet ION. This material consists of unidentifiable juveniles and females of subterranean spe- cies collected from stream gravels and pre- sumably represents populations of Calasel- lus, Salmasellus or Columbasellus. The col- lection data are mapped to further illustrate the occurrence of subterranean asellids in the streams of Oregon and Washington (Fig. 1). Acknowledgments We sincerely thank Mr. Carl McCrary, Kalama Public Works Director, for collect- ing this unique isopod, who later gave it to Dr. John T. Longino, Evergreen State Col- lege, Olympia, Washington. Dr. Longino recognized the unusual nature of the spec- imen and transferred it to us for identifica- tion. We also thank Jude Van Buren (State of Washington Department of Health) and Richard Hoey (Regional Engineer, South- west Drinking Water Operations) for infor- mation on the collection of the isopod. The new records of other asellids from Oregon and Washington are courtesy of Dr. Robert Wisseman, President, Aquatic Biology As- sociates, Corvallis, Oregon. We thank Da- vid L. White (University of South Carolina) for his help in preparing the map. We would also like to thank Dr. John R. Holsinger and Dr. Brian Kensley for reading the manu- script and making suggestions for its im- provement. We also thank Dr. Guy Mag- niez, who not only read the manuscript, but worked with us on understanding the ho- mologies of the Pacific Rim asellids. This work was indirectly supported by a PEET grant (DEB 9978193) from the Sys- tematic Biology Program of the U.S. Na- tional Science Foundation to J. W. Martin and D. K. Jacobs. Literature Cited Bowman, T. E. 1974. The California freshwater iso- pod, Asellus tomalensis, rediscovered and com- pared with Asellus occidentalis.—Hydrobiolo- gica 44:431—441]. . 1975. Three new troglobitic asellids from western North America (Crustacea: Isopoda: Asellidae).—International Journal of Speleolo- gy 7:339-356. . 1981. Calasellus longus, a new genus and species of troglobitic asellid from Shaver Lake, California (Crustacea: Isopoda: Asellidae).— Proceedings of the Biological Society of Wash- ington 94:866—-872. Henry, J.-P, & G. Magniez. 1995. Novelles donnees sur les Asellidae epiges d’Extreme-Orient (Crustacea, Isopoda, Asellota).—Contributions to Zoology 65:101—122. Henry, J-P., J. J. Lewis, & G. Magniez. 1986. Isopoda: Asellota: Aselloidea, Gnathostenetroidoidea, Stenetriodea. Pp. 434—464 in L. Botosaneanu, ed., Stygofauna Mundi. A faunistic, distribu- tional, and ecological synthesis of the world fauna inhabiting subterranean waters (including VOLUME 116, NUMBER 1 the marine interstitial). E. J. Brill, Leiden, 740 Pp- Lewis, J. J. 1982. A diagnosis of the Hobbsi group, with descriptions of Caecidotea teresae, new species, and C. macropropoda Chase and Blair (Crustacea: Isopoda: Asellidae).—Proceedings of the Biological Society of Washington 95: 338-346. . 2000. Caecidotea cumberlandensis, a new species of troglobitic isopod from Virginia, with new records of other subterranean Caecidotea (Crustacea: Isopoda: Asellidae).—Proceedings of the Biological Society of Washington 113: 458-4064. . 2001. Three new species of subterranean asel- lids from western North America, with a syn- opsis of the species of the region (Crustacea: Isopoda: Asellidae)——Texas Memorial Muse- um, Speleological Monograph 5:1—15. , & T. E. Bowman. 1981. The subterranean asellids (Caecidotea) of Ilinois.—Smithsonian Contributions to Zoology 335:1—66. ,& . 1995. The subterranean asellids of Texas (Crustacea: Isopoda: Asellidae).—Pro- ceedings of the Biological Society of Washing- ton 109:482—500. , & J. R. Holsinger. 1985. Caecidotea phrea- 197 tica, a new phreatobitic crustacean (Asellidae) from southeastern Virginia.—Proceedings of the Biological Society of Washington 98:1004— 1011. Matsumoto, K. 1960. Subterranean isopods of the Kyushu District, with the descriptions of three new species.—Bulletin of the Biogeographical Society of Japan 22:26—44. . 1962. Two new genera and a new subgenus of the Family Asellidae of Japan.—Annota- tiones Zoologicae Japonenses 35:162—169. . 1978. Three new species of subterranean asel- lids from southern Kyushu and the Kii Penin- sula, Japan.—Journal of the Speleological So- ciety of Japan 3:20—34. Miller, M. A. 1933. A new blind isopod Asellus cali- fornicus, and a revision of the subterranean asellids—University of California Publications in Zoology 39:94—109. Pospisil, P. 1992. Sampling methods for groundwater animals of unconsolidated sediments. Pp. 107— 134 in A. I. Camacho, ed., The natural history of biospeleology. Monografias Museo Nacional de Ciencias Naturales, Madrid, 680 pp. Williams, W. D. 1970. A revision of the North Amer- ican epigean species of Asellus (Crustacean: Is- opoda).—Smithsonian Contributions to Zoolo- gy 49:1-80. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):198—205. 2003. Bahadzia patilarga, a new species of subterranean amphipod crustacean (Hadziidae) from Cuba Thomas R. Sawicki, John R. Holsinger, Manuel Ortiz, and Abel Pérez (TRS) Department of Biological Sciences, Old Dominion University, Norfolk, Virgina 23529, U.S.A.; (JRH) Department of Biological Sciences, Old Dominion University, Norfolk, Virgina 23529, U.S.A.; (MO) Center for Marine Investigations, University of Havana, Cuba; (AP) Institute of Ecology and Systematics, Ministry of Science, Technology and Environment, Cuba Abstract.—A new species of Bahadzia is described from an anchialine cave on the southern coast of Cuba. Bahadzia patilarga is the second species of the genus to be described from Cuba. Both Cuban species have eye spots and are found with remipeds and thermosbaenaceans at or near the halocline. Since the initial description of the genus Bahadzia by Holsinger (in Holsinger & Ya- ger 1985) on the basis of two stygobitic species from anchialine caves in the Baha- mas and Turks and Caicos, seven more spe- cies have been described. These species are found in the Bahamas, Turks and Caicos, Cuba, Hispaniola and on the Yucatan Pen- insula. To date, the only species of Bahad- zia described from Cuba is B. yagerae (Or- tiz & Pérez 1995). The present paper de- scribes a second species of Bahadzia from the southern coast of Cuba. Although both this new species and B. yagerae exist in caves approximately 17 km apart, and both have eye spots, they are otherwise morpho- logically distinct. The following description raises the number of species in the genus to ten. Systematics Family Hadziidae Karaman, 1943 Genus Bahadzia Holsinger, 1985 (in Holsinger & Yager, 1985) Bahadzia patilarga, n. sp. (Figs. 1-4) Material examined.—CUBA. Metanzas Province: Cueva de los Carboneros, Playa Giron, holotype 2 (4 mm), A. Pérez, 11 June 1998; 2 paratypes (2 and juvenile), J. Yager, 14 September, 1992, and 2 paratypes (5 and 2), J. Bozanic, 15 September 1992. The holotype is deposited in the crusta- cean collection of the Center of Marine Re- search, University of Havana, Cuba (No. 178). The paratypes are in the collection of John R. Holsinger (H-3242, H-3249). Diagnosis.—Small to medium sized sty- gobitic species easily distinguished from other members of the genus except Bahad- zla yagerae Ortiz and Pérez 1995 by having a tiny, round, pigmentless eye, but differing from B. yagerae by having proportionately longer pereopods 6 and 7. Further distin- guished from all other species within the genus by possessing a much shorter row of setae on the extreme inner margin of the inner plate of maxilla 2 and fewer setae on anterior margin of the propod of gnathopod 2 of the female. Largest male 6.0 mm; larg- est female 6.5 mm. Female.—Head with tiny round, pig- mentless eye or eye spot. Antenna | ap- proximately 40% longer than body and 2.25 times longer than antenna 2; primary fla- gellum with up to 39 segments, accessory flagellum 3-segmented, subequal in length VOLUME 116, NUMBER 1 199 Fig. 1. Bahadzia patilarga, n. sp., paratypes from Cueva de los Carboneros, Playa Giron, Cuba. Juvenile (2.5 mm): A, head region with eyespot. Female (6.5 mm): B, left mandible; C, incisor and lacinia mobilis of right mandible E maxilla 2; G, maxilliped. Male (6.0 mm): D, maxilla 1; E, palp of other maxilla; H, telson; I, lower lip. 200 to the first 3 primary flagellar segments; pe- duncular segments becoming progressively shorter distally. Antenna 2: flagellum with up to 14 segments; peduncular segment 4 approximately 15% longer than segment 5. Mandible: molar well developed; spine row with 2 modified tooth-like spines distally and about 10 weakly serrate spines; lacinia mobilis of left mandible 4 dentate, that of right 3 dentate and smaller; incisor of left mandible 5 dentate, that of right 4 dentate and narrower; palp segment 3 as long as combined lengths of | and 2, bearing | long A seta, long row of approximately 13 D setae and 3—4 apical E setae. Lower lip: inner lobes distinct; lateral processes short, rounded apically. Maxilla 1: inner plate with 15 short, lightly plumose setae; outer plate bearing 9 pectinate spines; left and right palps similar, expanded and rounded distally, broad apex with 11 bladespines and single short, naked setae. Maxilla 2: in- ner plate narrowing distally, with row of ap- proximately 45 naked submarginal facial setae and row of approximately 5 thicker setae located distally on extreme inner mar- gin. Maxilliped: apex of inner plate even, armed on inner half with 4—5 short spines and a few short setae, inner margin with row of 10 weakly plumose setae; outer plate broader than inner with row of naked setae on inner margin and distally; palp segment 3 pubescent distally, distral inner margin of segment 3 slightly lobate; palp segment 4 almost as long as segment 3, nail small and spine-like. Gnathopod 1: propod subrectangular about twice as long as broad, posterior mar- gin heavily setose distally, palm _ short, transverse but lobate at defining angle and bearing 3 spines on lobe; carpus approxi- mately 2 times as long as propod, bearing several clusters of long setae toward distal end; merus weakly lobiform and pubescent; basis with 7 long setae on posterior margin; coxa long and deep with about 9 short setae and 3 short spines. Gnathopod 2: propod subrectangular, palm short, oblique bearing short setae and 3 long setae at defining an- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON gle, posterior margin with 4 sets of long setae, anterior margin with a few setae, not in clusters; carpus approximately 28% lon- ger than propod, posterior margin with 9 clusters of long setae; basis with 5 long se- tae; coxa deeper than broad, margin with about 6 setae and 4 short spines. Pereopod 3: coxa relatively small, deeper than broad, margin with 3 short setae. Pereopod 4: coxa broadly expanded distally and excavate posteriorly, margin with 13 short setae. Pe- reopod 6 approximately 5% longer than body, approximately 7% longer than pereo- pod 7 and 90% longer than pereopod 5. Pe- reopods 5—7: basis relatively narrow, with rather large, bluntly rounded distoposterior lobes. Dactyl of pereopod 5 approximately 50% length of corresponding propod; dac- tyls of pereopods 6 and 7 respectively, ap- proximately 30% and 13% of correspond- ing propods. Coxal gills on 2—6 subovate, with short peduncle, very large on pereo- pods 2—4. Brood plates sublinear, small rel- ative to gills. Pleonal plates: posterior margins with | setule each, posterior corners small but dis- tinct; ventral margin of plate 1 without spines, plates 2 and 3 with | spine. Pleo- pods normal, coupling spines rather long. Uropod 1: inner ramus shorter than pedun- cle, longer than outer ramus, bearing about 5 spines; peduncle with 12 spines, 1 of which is basofacial in position. Uropod 2: inner ramus approximately 15% longer than peduncle, longer and broader than outer ra- mus, armed with 13 spines; outer ramus with about 6 spines; peduncle with 9 spines 4 of which form a comb row on dorsodistal end. Uropod 3 approximately 23% length of body; inner ramus slightly longer and broader than outer ramus, margins with short spines and plumose setae; outer ramus with short terminal segment, inner margin with plumose setae and a few spines, outer margin with spines only; peduncle without spines. Telson rather long and narrow, in two separate lobes; lateral margins with about 5 spines each, none in sets of two; medial margins with 3—4 small spines each; t VOLUME 116, NUMBER 1 201 Fig. 2. Bahadzia patilarga, n. sp., paratypes from Cueva de los Carboneros, Playa Giron, Cuba. Male (6.0 mm): A, B, C, pereopods 7, 5, 6. Female (6.5 mm): D, gnathopod 1; E, enlarged distal end of propod and dactyl of gnathopod 1. 202 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Bahadzia patilarga, n. sp., paratype from Cueva de los Carboneros, Playa Giron, Cuba. Female (6.5 mm): A, B antennae 1,2; C gnathopod 2; D, enlarged distal end of propod and dactyl of gnathopod 2. 203 VOLUME 116, NUMBER 1 ey = = ve on Ae F Fig. 4. Bahadzia patilarga, n. sp., paratypes from Cueva de los Carboneros, Playa Giron, Cuba. Female (6.5 mm): A, uropod 1; B, uropod 2 (distomedial spine row enlarged); C, uropod 3; D, pleopod | (coupling spines enlarged); E, pleonal plates. Male (6.0 mm): E G, pereopods 3, 4; H, propod and dactyl of gnathopod 2. apices with 1—2 short spines and 3 long, distally plumose setae. Male.—Differing from female as fol- lows: maxilla | inner plate with 7 short, weakly plumose setae; palp of maxilla 1 with 12 bladespines and without short na- ked setae. Gnathopod 2: dactyl and propod proportionately longer; propod palm long, oblique with double row of about 9 blunt tipped spines; defining angle with 3 long setae; posterior margin longer than palm, with 4 sets of setae. Distal margin of pe- duncle of uropod 3 with 2 spines. Etymology.—The word “‘patilarga’’ is Spanish vernacular commonly used in Latin American countries to describe legs that are larger than normal. It is used here as a noun in apposition alluding to the extremely long pereopods 6 and 7 of the species. Remarks.—Bahadzia patilarga was 1n- formally referred to as “‘cubensis”’ in an ar- ticle written for a general SCUBA diving magazine and was not intended to be a for- mal taxonomic publication. Therefore the name ““cubensis”’ is unavailable and invalid and should not be considered a synonym or nomen nudum. Type-locality.—This species is known only from the type-locality, Cueva de los Carboneros. Discussion Bahadzia patilarga is recorded to date from a single anchialine cave, Cueva de los Carboneros, which is located in Playa Giron on the southern coast of Cuba in Matanzas Province. Playa Giron is commonly known in America as the Bay of Pigs. Collection of B. patilarga was made at or near the hal- ocline, where it occurs sympatrically with remiped and thermosbaenacean crusta- ceans. This association with remipeds and thermosbaenaceans and its specific micro- habitat defined as being near or within the halocline of anchialine caves is very com- mon, although not ubiquitous, for this ge- nus. Based on research in caves of the Yu- catan Peninsula, Pohlman et al. (1997) not- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ed that these crustacean taxa might be uti- lizing a chemoautotrophic energy source existing near the halocline. They suggested that a similar phenomenon may occur in an- chialine caves throughout the Caribbean and southern Atlantic, where organisms such as Bahadzia exist. Pohlman et al. (1997) noted a dramatic drop in oxygen concentration precisely at the halocline, Where they surmised that chemoautotroph- ism was occurring. Although no field data are available for the oxygen concentration at the halocline in the type locality for B. patilarga, it is interesting to note that, as with many (but not all) species of Bahad- zia, B. patilarga has extremely large gills. Enlarged gills may be an adaptation for hv- ing in low oxygen environments. Based on a track synthesis, Holsinger (1989, 1992) predicted the occurrence of Bahadzia in Cuba. The description of Ba- hadzia patilarga above brings the number of species so far discovered on the island to two. Both species exist in single caves a mere 17 km apart. Recent explorations by one of us (TRS) resulted in the discovery of a remiped crustacean in a cave on the northern coast of Cuba, also in Matanzas Province. Based on this discovery and the fact that remipeds and Bahadzia are often found living sympatrically, it is predicted that additional populations of Bahadzia, quite possibly representing new species, will be found in caves on the northern coast of Cuba. Acknowledgments We are grateful to Jeff Bozanic, Jill Ya- ger and the Biokarst Association of Cuba for assistance with collecting specimens, and we thank Tamara Connolly for assis- tance with preparation of the figures. This study was supported in part by a PEET (Partnerships for Enhancing Expertise in Taxonomy) grant from the National Science Foundation to JRH (DEB-9521752). VOLUME 116, NUMBER 1 Literature Cited Holsinger, J. R. 1989. Preliminary zoogeographic anal- ysis of five groups of crustaceans from anchia- line caves in the West Indian region. Proceed- ings of 10th International Congress of Speleol- ogy, 1:25—26. Holsinger, J. R. 1992. Two new species of the subter- ranean amphipod genus Bahadzia (Hadziidae) from the Yucatan Peninsula region of southern Mexico, with an analysis of phylogeny and bio- geography of the genus.—Stygologia 7:85—105. Holsinger, J. R., & J. Yager. 1985. A new genus and two new species of subterranean amphipod 205 crustaceans (Hadziidae) from the Bahamas and Turks and Caicos Islands.—Bijdragen tot de Dierkunde 55:283—294. Karaman, S. 1943. Die unterirdischen Amphipoden Sudserbiens. Srpska Kral’evska Akademiia Po- sebna Izdan’a, CXXXV Prirodn’achki 1 Mate- matichki Spici, 34 (4) Okhridski Zbornik: 163— SDs Ontiz, Mi SA betez, 1995) UnasNueva especie de Anfipodo Cavernicola Hadzioideo (Amphipoda, Gammaridea) de Cuba.—Graellsia 51:165—168. Pohiman, J. W., T. M. Iliffe, & L. A. Cifuentes. 1997. A stable isotope study of organic cycling and the ecology of an anchialine cave ecosystem.— Marine Ecology Progress Series 155:17—27. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):206—214. 2003. A new species of Cymbasoma (Crustacea: Copepoda: Monstrilloida) from the Pacific coast of Costa Rica, Central America Eduardo Suarez-Morales and Alvaro Morales-Ramirez (ESM) EI Colegio de la Frontera Sur (ECOSUR). Unidad Chetumal, A.P. 424, Chetumal, Quintana Roo 77000, Mexico, e-mail: esuarez@ecosur-qroo.mx; (AMR) Centro de Investigacion en Ciencias del Mar y Limnologia (CIMAR), Escuela de Biologia, Universidad de Costa Rica, San José 2060, Costa Rica Abstract.—A new species of monstrilloid copepod belonging to the genus Cymbasoma Thompson is described from a female specimen collected in Bahia Culebra, an embayment on the Pacific coast of northwestern Costa Rica, in Central America. The new taxon, C. concepcionae, n.sp., is part of a group of species of Cymbasoma with a fifth leg represented by a single lobe armed with three setae. An overall comparison of this morphological group is presented. The new species is distinguished mainly by a combination of characters in- cluding a short inner seta on the fifth leg, a slightly globous genital double somite, and transverse cuticular striations nearly encircling the cephalic area. Moreover, the species can be easily recognized by a distinctive cluster of round- ed cuticular processes on the middle ventral margin of the fifth pedigerous somite and on the anterior half of the genital double somite. The intercoxal sclerite of the first swimming leg shows a peculiar ornamentation not previ- ously described within the group. This is the first record of a monstrilloid copepod species from marine waters of Costa Rica and of Central America, and the fourth species of Cymbasoma recorded from the Eastern Tropical Pa- cific. Monstrilloid copepods are parasitic crus- taceans known mainly by their planktonic, non-feeding adult stage. These usually rare copepods are collected mainly in zooplank- ton surveys of coastal-neritic systems at all latitudes. The order is represented currently by over 100 nominal species contained in three valid genera: Monstrilla Dana, Mon- strillopsis Sars, and Cymbasoma Thompson (Huys & Boxshall 1991, Grygier 1994). Cymbasoma 1s a diverse genus with more than 40 nominal species; its species are dis- tributed world-wide. Most of these species have been reported from either the Eastern Atlantic or the Far East (Razouls 1996). To date three species, C. californiense Suarez- Morales, 1999, C. rigidum Thompson, 1888, and C. tumorifrons Isaac, 1974, are known to occur in tropical waters of the Eastern Pacific (Suarez-Morales & Alvarez- Silva 2001). Suarez-Morales & Alvarez-Sil- va (2001) predicted that the regional list of Monstrilloida recorded for the Eastern Tropical Pacific would increase, most likely from the coastal, near-shore environments where monstrilloid copepods are more fre- quently collected. As part of a survey of the coastal areas along the northern Pacific coastline of Cos- ta Rica, in Central America, zooplankton samples were collected in Bahia Culebra, a deep embayment that is part of the Gulf of Papagayo. A taxonomic analysis of the zoo- plankton samples revealed the presence of an undescribed species of monstrilloid co- pepod belonging to the genus Cymbasoma, which is described and illustrated here, fol- lowing the current description standards set VOLUME 116, NUMBER 1 by Grygier & Ohtsuka (1995) for monstril- loid copepods. Systematics Order Monstrilloida Sars, 1903 Family Monstrillidae Giesbrecht, 1892 Genus Cymbasoma Thompson, 1888 Cymbasoma concepcionae, new species (Figs. 1-3) Material examined.—Holotype: adult fe- male from Bahia Culebra, Costa Rica, un- dissected, ethanol-preserved. Date of col- lection: 6 November 1999. Vial deposited in the Collection of Zooplankton at El Col- egio de la Frontera Sur (ECOSUR), in Che- tumal, Mexico (ECOCH-Z-01211). Etymology.—The new species honors Mrs. Concepci6n Morales, mother of the first author (ES-M). Type locality.—Bahia Culebra (10°36'17.0"N, 85°39'36.6"W), northern part of the Pacific coast of Costa Rica; depth of collection: near surface. Habitat.—Bahia Culebra is located with- in the Gulf of Papagayo, in the northwest- ern part of the Pacific coast of Costa Rica. It is a deep (35 m average depth), relatively small (20 km’), semi-enclosed embayment. Bahia Culebra is influenced by seasonal variations of the coastal current flowing northward off Costa Rica. Because of wind influence, relatively intense coastal upwell- ing has been reported in the area (Glynn et al. 1983). The bay has a sandy bottom and scattered patches of coral reef (Jiménez 1998). The mean sea surface temperature in thesarea ts 27.5) C:; Description.—Female: Total body length of holotype: 1.5 mm measured from ante- rior end of cephalothorax to posterior end of anal somite. Cephalothorax measuring 0.95 mm, representing almost 61% of total body length (Fig. 1A, B). Oral papilla pro- tuberant (Fig. 1E), located ventrally a quar- ter the length of cephalothorax (Fig. 1B). Pair of ocelli present, pigment cups joined medially, well developed, weakly pigment- ed in central portion, round in dorsal view 207 (Fig. 1A, F). Cephalothorax with irregularly arranged cuticular protuberances on “‘fore- head” (Fig. 1A, D); sensilla not observed in this area. Longitudinal and oblique cutic- ular ridges overlying part of region of ocelli on dorsal surface; ridges stretch dorsally and end in a transverse pattern (Fig. 1F). Two pairs of nipple-like processes on an- terior ventral surface; anterior pair larger, eccentrically conical in shape, small pair regularly concentric (Fig. 1C, D). Other ventral cuticular ornamentation including pair of mammiform processes on anterior surface between bases of antennules (Fig. 1D) and wide zone of transverse ridges cov- ering most of the surface between the oral papilla and small nipples (Fig. 1A, B, D, E); dorsal and ventral transverse ridge zones at same level, but discontinuous on lateral surfaces. Antennule length 0.25 mm, relatively short, less than 14% of total body length, and 22% as long as cephalothorax. Anten- nules four-segmented, armed with O-I; 1-V; 2-I; 8-VIII setae (Arabic numerals) and spines (Roman numerals) (Fig. 2D, E). In terms of pattern described by Grygier and Ohtsuka (1995) for female monstrilloid an- tennular armature, setae (Roman numerals) and spines (Arabic numerals), element | present on first segment; elements on sec- ond sesinente dis) 205-22 Na, ov, and IId. Third segment with elements 3, IIId, and IIIv. Segment four bearing element 4v,, particularly well developed, asymmetrical, longer on left antennule (Fig. 2D, E); ele- ments 4d,, and 4v,_; present, poorly devel- oped, setae IVd, IVv, Vd, Vv, and Vm pre- sent. Element 5 absent. Subterminal ele- ments b, , and 6aes, 6,, and 6, present; 6aes larger on left antennule. Aesthetasc 4aes well developed, on ventral surface (Fig. 2D, E). First pedigerous somite, fused to cepha- lothorax, and succeeding three free pedi- gerous somites each bearing a pair of bi- ramous swimming legs. Pedigerous somites 2—4, together accounting for 21% of total length in dorsal view. Swimming legs 1—4 208 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0.01 mm Fig. 1. Cymbasoma concepcionae, new species. adult female, holotype from Bahia Culebra, Costa Rica. A. habitus, dorsal; B. habitus, lateral; C. detail of ventral nipple-like processes; D. cephalic area showing ventral cuticular processes; E. cephalic area, lateral view; FE cuticular ridges on dorsal surface posterior to ocelli; G. distal ends of ovigerous spines, ventral. Table 1.—Armature formula of swimming legs 1-4 of Cymbasoma concepcionae, new species. Symbols to left of dash are lateral elements, symbols to right of dash are medial elements; semicolons represent arthrodial membranes of rami; commas separate lateral, terminal, and medial elements on the distal segment of a ramus; Roman numerals are thicker, rigid elements; Arabic numerals are thinner, flexible elements. Coxa Basis Endopodite Exopodite leg 1 O08) 1-0 O-1; O-1; 1, 2, 2 I-1; 0-1; I, I+1, 2 leg 2 0-0 1-0 O-1; O-1; 1, 2, 2 I-1; 0-1; I, I+1, 3 leges 0-0 1-0 O-1; O-1; 1, 2, 2 I-1; 0-1; I, I+ 1, 3 leg 4 0) 1-0 0-1; 0-1; 1, 2, 2 I-1; O-1; I, I+ 1, 3 VOLUME 116, NUMBER 1 SSS SS ie eg eS NS SS 4 y) VJ ia Ye ldlt¢7 VE eit Rica: 209 Cymbasoma concepcionae, new species, adult female, holotype from Bahia Culebra, Costa Rica. A. urosome and fifth legs, ventral view; B. genital double somite and anal somite plus caudal rami, dorsal; C. fifth pediger and anterior part of genital double somite, lateral; D. right antennule, dorsal; E. distal segment of left antennule, ventral; E third free pediger and urosome, lateral view, protoendopod of fourth legs shown, fifth legs omitted to show ventral cuticular processes, caudal rami cut short. slightly increasing in size posteriorly (Fig. 3A—D), but their intercoxal sclerites de- crease in size posteriorly. Intercoxal sclerite of leg | with pair of longitudinal semi-cy- lindrical processes covered with short setae and flanking hairy medial depression. Pos- terior margin of this intercoxal sclerite flat at middle, with rounded processes on each side (see Fig. 3E). Intercoxal sclerites of legs 2—4 rectangular, without ornamentation on surface or on posterior margin (Fig. 3F). Basis of legs articulating with large, rect- angular coxa along diagonal arthrodial membrane. Basis with hair-like lateral seta 210 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Comparison of selected taxonomic features in species of Cymbasoma with uriramous fifth leg armed with three setae. GDS = anterior half genital of double somite globose; IS = inner seta of 5th leg smallest; IS <50% = inner seta <50% as long as others; L = longest seta on leg 5; OP on CT = position of oral papilla on cephalothorax; Al/CT = relative length (as a percent) of antennule to cephalothorax; = relative length (as a percent) of position of oral papilla anteriorly on cephalothorax; T = transverse striations on “‘neck’’ area. Key to species: (A) C. concepcionae, new species; (B) C. reticulatum Giesbrecht, 1892; (C) C. bowmani Suarez- Morales, 1998; (D) C. quintanarooense (Suarez-Morales, 1994) (Suarez-Morales & Escamilla 2001); (E) C. boxshalli (Suarez-Morales, 1993) (Suarez-Morales 20001a); (F) C. bali Desai & Krishnaswamy, 1962; (G) C. frondipes Isaac, 1975; (H) C. striatum (Isaac, 1974) (Suarez-Morales 2000); (1) C. claparedi Giesbrecht, 1892; (J) C. tumorifrons Isaac, 1975. A B c D E F G H I J GDS yes no no yes yes no no yes no yes IS yes no yes no yes no yes yes no yes IS <50% no no no no yes no no yes no no IL, outer none none none outer middle none none none outer OP on CT 23 20-25 15 28 18-20 NY u 18 14 BS A1/CT 22 27-30 m3 32 m) 27 21 33 29 31 OP/TL 43 89 42 47 42 41 y 34 49 36 T yes no no no no ? & yes no yes on legs 1—4 (Fig. 3A—D); on leg 3, this seta about 4.5 times longer, lightly setulated from distal half, and slightly thicker than those on the other legs (see Fig. 3C). En- dopodites and exopodites of swimming legs 1—4 triarticulated. Ramal setae biserially plumose except spiniform outer seta on ex- opod | and 3, and inner seta of first exo- podal segment, these latter being short and sparsely setulated (Fig. 3A—D). Also, out- ermost apical exopodal setae of swimming legs 1—4 with inner margin naked, outer margin lightly spinulose to tip (Fig. 3A—D). Armature of swimming legs in Table 2. Fifth legs separated at base, unsegment- ed, each consisting of relatively large, lobe- like expod (Fig. 2A); 3 setae present, outer 25% longer than terminal; inner seta nar- rower and shorter, about 50% as long as outer seta. All 3 setae with sparse, biserial setules (Fig. 2A). Urosome of 3 segments: fifth pedigerous somite, genital double somite and free anal somite. Fifth pedigerous somite with a group of rounded protuberances along mid- dle ventral margin between fifth legs lobes (Fig. 2C, F). Genital double somite rela- tively large, ratio of its length and anal so- mite: 69.5:30.5 = 100; anterior half ex- panded laterally, expansion visible in both ventral (Fig. 2A) and dorsal (Fig. 2B) views, with rounded protuberances along lateral margins of expanded anterior half. Anterior half of ventral surface of genital double somite with similar protuberances as those on fifth pedigerous somite anterior to base of ovigerous spines. Ovigerous spines paired, relatively long (0.65 mm), unbro- ken, inserted in a protuberant base visible in lateral view (Fig. 2F), separated at base, equalling almost 43% of total body length; adhering eggs covered by gelatinous sheath. Tips of both spines separated from each other, with slight but distinct subterminal narrowing (Fig. 1G). Caudal rami subrectangular, widely di- vergent, approximately 1.5 times longer than wide, each bearing 3 well developed setae, all of almost the same length (0.23— 0.30 mm) and width. Male.—unknown. Remarks.—The absence of a reliable way to associate females with conspecific males is one of the main problems with the tax- onomy of the Monstrilloida, a result of the complex life cycle of these parasitic cope- pods (see Suarez-Morales 2001b). As is the case of many other monstrilloids (see Sua- rez-Morales 1993, Suaérez-Morales & Dias 2001), C. concepcionae is known only from VOLUME 116, NUMBER 1 ZaM Ih mre EF p pH STIPE PI PL ET Me ee SFITIFTIT PT 7 pie Op NY: CSE: NAGS Se Sh 7 aS “i777 ) 4 Fig. 3: Cymbasoma concepcionae, new species, adult female, holotype from Bahia Culebra, Costa Rica. A. first swimming leg, anterior view, most swimming setae cut short; B. second leg, anterior view, ditto; C. third swimming leg, anterior view, ditto; D. fourth swimming leg, anterior view, ditto; E. intercoxal sclerite, first swimming legs; E intercoxal sclerite, fourth swimming legs. the female, and from a single specimen, the common case with monstrilloids (Suarez- Morales and Palomares-Garcia 1995, Sua- rez-Morales 2000). The new species described here is placed in the genus Cymbasoma because of the presence of two urosomites, or only one free somite posterior to the genital double somite, in the female (Isaac 1975, Grygier 1994). Following the most widely used key to the species of this genus (Isaac 1975), and considering a non-striated cephalic seg- ment option in the third couplet, the new species would match (in part) the female of either C. tumorifrons (Isaac, 1975) or C. claparedi (Giesbrecht, 1892), both found in the Eastern Atlantic. These two species also are members of a group of Cymbasoma with a single-lobed fifth leg armed with three setae; the group includes C. striatum (Isaac, 1974), C. frondipes Scott, 1904, C. reticulatum (Giesbrecht, 1892), C. bali De- sai and Krishnaswamy, 1962, C. bowmani Suarez-Morales, 1998 (see Suarez-Morales & Gasca 1998), C. boxshalli (Suarez-Mo- rales, 1993), and C. guintanarooense (Sua- rez-Morales, 1994). The presence of rounded protuberances on the ventral surface of the fifth pediger- ous somite and genital double somite of Cymbasoma concepcionae separates it from all other members of this group. Cymba- soma concepcionae shares with C. striatum, C. quintanarooense, and C. tumorifrons ce- phalic cuticular striations and a similar structure and armature of the fifth legs. Al- though C. concepcionae and C. tumorifrons are Similar in the structure and armature of fifth legs and transverse striations of the ce- phalic area (see Table 2), they differ in the body proportions, the shape of the cepha- lothorax, the length of the ovigerous spines, and the shape of the genital double somite, which is strongly globose in C. tumorifrons (Table 2 and see Sudarez-Morales and AI- varez-Silva 2001). Other important differ- ences include the respective cuticular or- namentation patterns and the distinctive ventral rounded processes present in the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON urosome of the new species. The new spe- cies can be distinguished from C. qguintan- arooense by the ocelli which are very well- developed and intensely pigmented in the latter (see complementary description by Sudrez-Morales and Escamilla 2001) vs. normally developed in C. concepcionae. The antennules are relatively longer in C. quintanarooense than in C. concepcionae, and the oral papilla is in a more posterior position. The three setae of the fifth legs are equal in length and breadth in C. quintan- arooense Vs. a short innermost seta in the new species (see Table 2). Cymbasoma concepcionae differs from C. striatum by the latter’s peculiar cuticular Striation pattern covering almost half the length of its cephalothorax, as redescribed by Suarez-Morales (2000); a strongly glo- bose shaped genital double somite vs. a slightly expanded shape in the new species; an oral papilla 18% along cephalic somite in C. striatum vs. 23% in the new species; ‘‘forehead’’ with medial protuberance formed by several irregularly arranged bumps; with a pair of sensillae in C. stria- tum vs. lower cuticular protuberances and no sensillae in the new species; ovigerous spines closely set distally and with subter- minal rounded protuberances in C. striatum vs. distally separated spines with slight but distinct subterminal narrowing in C. con- cepcionae; and an outer, apical exopodal seta of swimming legs 1—4 with setules in C. striatum vs. naked in the new species (Suarez-Morales 2000). Cymbasoma frondipes has a very long cephalothorax (over 66% of the total body length) and characteristically massive, short antennules (see Isaac 1975). Cymbasoma concepcionae differs from the remaining species shown in Table 2 by a combination of: an anteriorly weakly globose genital double somite; the inner seta of fifth leg shortest and outer seta longest; and trans- verse striations around the “‘neck’’ area. The taxonomic value of the morphology of the intercoxal sclerites has not been com- pared among these monstrilloids, although VOLUME 116, NUMBER 1 recent descriptions suggest a certain degree of ornamentation (Suarez-Morales and Dias 2001). The contours and setation of the first intercoxal sclerite in C. concepcionae is def- initely the most complex pattern known for species in the group. Cymbasoma concepcionae is the fourth confirmed species of the genus from waters of the Eastern Tropical Pacific (Suarez-Mo- rales and Alvarez-Silva 2001). This also is the first record of a monstrilloid species in Costa Rica or along the Pacific coast of Central America. More records are to be ex- pected from further examination of the zoo- plankton samples available from the region. Acknowledgments The plankton material from Bahia Cule- bras, Costa Rica was made available to us by the Centro de Investigacion en Ciencias del Mar y Limnologia of the University of Costa Rica (CIMAR). This specimen was sorted by Ivan Castellanos-Osorio (ECO- SUR-Chetumal) during a research visit to the CIMAR in San José, Costa Rica. Literature Cited Desai, H. V., & S. Krishnaswamy. 1962. Cymbasoma bali, a new species of monstrilloid copepod from the Bombay plankton.—Proceedings of the Indian Academy of Sciences 55B:163—166. Giesbrecht, W. 1892. Systematik und Faunistik der pe- lagischen Copepoden des Golfes von Neapel und der angrezenden Meeres-Abschnitte. Fauna und Flora des Golfes von Neapel und der an- grezenden Meeres-Abschnitte herausgegeben von der Zoologischen Station zu Neapel. XIX. Monographie. Verlag von R. Friedlander & Sohn, Berlin, 1—831. Glynn, P., E. Druffel, & R. Dunbar. 1983. A dead Cen- tral America coral reef tract: possible link with the Little Ice Age.—Science 203:47—49. Grygier, M. J. 1994. Identity of Thaumatoessa (= Thaumaleus) typica Kr@yer, the first described monstrilloid copepod.—Sarsia 78:235—242. Grygier, M. J., & S. Ohtsuka. 1995. SEM observation of the nauplius of Monstrilla hamatapex, new species, from Japan and an example of upgrad- ed descriptive standards for monstrilloid cope- pods.—Journal of Crustacean Biology 15:703-— WS) 2S Huys, R., & G. A. Boxshall. 1991. Copepod evolution. The Ray Society, London, 468 pp. Isaac, M. J. 1974. Copepoda Monstrilloida from south- west Britain including six new species.—Jour- nal of the Marine Biological Associaton of the United Kingdom 54:127—140. Isaac, M. J. 1975. Copepoda, Suborder: Monstrilloi- da.—Fiches d’Identification du Zooplancton 144/145:1-10. Jiménez, C. 1998. Arrecifes y comunidades coralinas de Bahia Culebra, Pacifico Norte de Costa Rica (Golfo de Papagayo). Unpublished M.Sc. thesis, Universidad de Costa Rica, San José, 218 pp. Razouls, C. 1996. Diversité et répartition géographique chez les copépodes pélagiques. 2.—Platyco- pioida, Misophrioida, Mormonilloida, Cyclo- poida, Poecilostomatoida, Siphonostomatoida, Harpacticoida, Monstrilloida.—Annales de I’ Institut Océanographique Nouvelle Série 72:1— 149. Sars, G. O. 1903. An account of the Crustacea of Nor- way with short descriptions and figures of all species. Vol. IV. Copepoda Calanoida. The Ber- gen Museum, Bergen. 28 pp. Suarez Morales, E. 1993. A new species of Thauma- leus (Copepoda, Monstrilloida) from the eastern coast of the Yucatan Peninsula, Mexico.—Crus- taceana 64:85-89. . 1994. Thaumaleus quintanarooensis, a new monstrilloid copepod from the Mexican coasts of the Caribbean Sea.—Bulletin of Marine Sci- ence 54:381—384. . 2000. Redescription of two species of Cym- basoma from southwest Britain and from In- donesia (Copepoda: Monstrilloida), with notes on taxonomy.—Beaufortia 50:139—-149. . 2001a. Redescription and first record of Cym- basoma boxshalli and Monstrilla inserta (Co- pepoda: Monstrilloida) from Curacao, eastern Caribbean Sea.—Cahiers de Biologie Marine 42:243-254. . 2001b. An aggregation of monstrilloid cope- pods in a western Caribbean reef area: ecolog- ical and conceptual implications.—Crustaceana 74:689—696. , & C. Alvarez-Silva. 2001. Cymbasoma tu- morifrons (Copepoda: Monstrilloida): an ex- panded description based on a new collection from the Eastern Tropical Pacific.—Pacific Sci- ence 55:183-189. , & C. Dias. 2001. Taxonomic report of some monstrilloids (Copepoda: Monstrilloida) from Brazil with description of four new species.— Bulletin de |’ Institut Royal des Sciences Na- turelles de Belgique. Biologie 71:65—81. , & J. B. Escamilla. 2001. Taxonomic report on some monstrilloids (Copepoda, Monstrilloida) from southeast Mexico with the description of 214 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a new species of Monstrilla.—Journal of Nat- of Monstrilla (Copepoda: Monstrilloida) from a ural History 35:1433—1445. coastal system of the Baja California Peninsula, , & R. Gasca. 1998. Cymbasoma bowmani Mexico.—Journal of Plankton Research 17: n.sp., a new monstrilloid (Copepoda: Monstril- 745-752. loida) from a Caribbean reef, with notes on spe- Thompson, I. C. 1888. Copepoda of Madeira and the cies variation.—Journal of Marine Systems Canary Islands, with descriptions of new genera 15(1—4):433-—439. and species.—Journal of the Linnean Society of , & R. Palomares-Garcia. 1995. A new species London (Zoology) 20:145—156. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):215—221. 2003. Description of subadult Pallisentis (Pallisentis) rexus (Acanthocephala: Quadrigyridae) from the vertebrate intermediate host in Thailand with an examination of the species identity Omar M. Amin and Horst Taraschewski (OMA) Institute of Parasitic Diseases, PO. Box 28372, Tempe, Arizona 85285, USA; (HT) Zoologische Inst. Abt., Okologie, Univ. Karlsruhe, Kaiserstrasse 12, 76128 Karlsruhe, Germany Abstract.—Subadults (considerably developed immatures) of the quadrigyrid acanthocephalan Pallisentis rexus Wongkham & Whitfield, 1999 are described from the swamp eel, Monopterus albus (Zieuw, 1793) in Thailand. The species is placed in the subgenus Pallisentis. Our specimens are compared with adults of the same species collected from the freshwater snakehead Ophicephalus striata Bloch also from Thailand. Discrepancies from the original description are noted, certain structures, e.g., proboscis hook roots and brain (cepahalic ganglion), are re-interpreted, and additional information provided. The genus Pallisentis Van Cleave, 1928 was revised, three subgenera were erected, and a key to the valid species, and other taxonomic information were provided by Amin et al. (2000). About the same time, Wongkham & Whitfield (1999) described Pallisentis rexus from the freshwater snake- head, Ophicephalus striata Bloch, from the Chiang Mai Basin in Thailand. More re- cently, we have collected subadults of P. rexus from the swamp eel, Monopterus al- bus (Zieuw, 1793) from Bangkok, Thai- land. Our specimens and type material ex- amined by us shed considerably more light on the morphology and status of P. rexus which the present study reports in a com- parative context. Materials and Methods Ten 40—80 cm long swamp eels, M. al- bus, were examined for parasites on 20 March 2001. The fish, obtained from a fish market in Bangkok, Thailand, had been col- lected from local streams. The 10 fish were infected with a total of 257 (range 5—74 per fish) ovoid whitish nodules in the body cav- ity especially at the external surface of the hind gut. The encapsulated worms were mostly alive when liberated from their cysts in fish physiological solution. Forty-two worms (16 males, 26 females) were refrig- erated in tap water overnight then fixed and shipped in 5% formalin. They were subse- quently transferred to 70% ethanol, stained in Mayer’s acid carmine overnight then briefly destained, dehydrated in ascending concentrations of ethanol, cleared in grad- uated concentrations of terpineol in 100% ethanol, and whole mounted in Canada bal- sam. Measurements are in wm unless other- wise stated. The range is followed by mean values (in parentheses). Width measure- ments refer to maximum width. Trunk length does not include proboscis, neck, or male bursa. The term subadults is used to describe the highly developed immature forms. They are practically identical to adults: all males and females were distend- ed with sperm and ovarian balls, respec- tively. Specimens were deposited in the United States National Parasite Collection (USNPC) no. 91753. Beltsville, Maryland, U.S.A. Seven adult paratypes (3 males, 4 females) from the British Museum of Nat- 216 ural History (BMNH) (no. 1997.6.3.2-13) were also studied. Results and Discussion It is important to note that all character- istics of our P. (P.) rexus subadults listed in Table 1 and illustrated in Figures 1—3 match those observed in the adult paratypes borrowed from BMNH. The anatomy of the subadults is generally similar to that of the adults as described by Wongkham & Whit- field (1999). However, misinterpretations and omissions exist in the original descrip- tion which otherwise was adequate. The quality of staining of paratypes produced faded and often indistinguishable structures, e.g., proboscis hook roots and cephalic gan- glia, which readily explains Wongkham & Whitfield’s (1999) version of worm anato- my. We shall report these discrepancies in the following section in an attempt to pro- vide an accurate and complete description of the species. Pallisentis rexus is placed in the subge- nus Pallisentis according to Amin et al. (2000) because the proboscis hooks gradu- ally decline in size posteriorly and the ce- ment glands are long and with many giant nuclei. Qualitative and quantitive traits of P. (P.) rexus subadults are provided and compared with those of adults, when available (Table 1). Measurements of subadults largely fell within the range of those reported for adults by Wongkham & Whitfield (1999). On the average, subadults, especially females, were relatively smaller than adults in only one character, trunk length. However, mean size of proboscis, proboscis hooks, proboscis re- ceptacle, and lemnisci were somewhat greater in subadults than in adults. This ob- servation is attributed to two factors: the apparent precocious state of development of subadult P. (P.) rexus (with sperm and ovarian balls) in their vertebrate interme- diate host M. albus; this phenomenon has been observed in other acanthocephalans (Amin 1982); and the early development of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON attachment organs to ensure establishment of new recruits in the gut of the definitive host (see Amin 1986, 1987 for other ex- amples). Omissions in the original description in- clude accounts of the giant nuclei of the hypodermis, lemnisci, and proboscis/neck, as well as some details of the male repro- ductive system and hook and spine inser- tion. We noted that one dorsal and one ven- tral elongate giant hypodermal nuclei were normally present. They were rarely absent on either side. Two similarly elongate giant nuclei were also noted within the proboscis retractor muscles in the proboscis/neck area. These nuclei are rarely reported in the genus Pallisentis but have been recently ob- served in Pallisentis (Brevitritospinus) viet- namensis Amin, Heckmann, Ha, Luc & Doanh, 2000. Each lemniscus had one large oval giant nucleus located at the level of the posterior half of the proboscis receptacle. Wongkham & Whitfield (1999) made no reference to the two large nuclei in the wall of Saefftigen’s pouch which was properly described posterior to the cement reservoir (Fig. 1B) but not “‘at the posterior margin of cement gland”’ as stated in the Abstract (Wongkham & Whitfield 1999). No refer- ence was made to the prominent sperm duct which ran parallel to the cement gland or the large common sperm duct which ran parallel to and overlapped Saefftigen’s pouch and the posterior half of the cement reservoir. Proboscis hooks had variably el- evated cuticular rims that were often prom- inent and body spines consistently had dis- tinct cone-shaped cuticular sleeves that left only spine tips naked. Misinterpretation in the original descrip- tion includes accounts of the proboscis hook roots, brain (cephalic ganglion), collar and trunk spines, and position of the female gonopore. Each of these characteristics has important taxonomic significance that can not be overlooked. Three aspects of pro- boscis hook roots need to be addressed. All hook roots extend laterally for a short dis- tance then loop back posteriorly in a direc- VOLUME 116, NUMBER 1 Figs. 1-3. D7] Diagnostic features of Pallisentis (P.) rexus. 1. One row of proboscis hooks and roots of female. 2. Posterior part of female proboscis receptacle showing triangular brain (cephalic ganglion) and ventro-lateral nerve cord passage through indented part of proboscis receptacle. 3. Posterior end of female showing slightly subterminal position (near terminal on ventral site) of gonopore. Scale bars = 100 um. tion paralleling that of the blade thus form- ing an inverted U-shaped structure (Fig. 1). In the original description, the angle be- tween hook and root was shown to be about 120—140° inaccurately creating an inverted open V-shaped blade-root complex (Fig. 2A). Roots were only slightly shorter than blades in all circles with a ratio of 1:1.0- LZ, JENS) BI Sib Zo aynvel ee ee from anterior (Fig. 1). In the original de- Scription, roots were shown to be consid- erably shorter than blades in all circles with Qi) UMM ESTOS eNO) Oe Wey, sz, WE, and 1:3 from anterior (Fig. 2A). In the same figure, Wongkham & Whitfield (1999) show the roots to be simple and with round- ed corners. Actually, the roots do not have rounded corners and are not all simple. Hook roots in the second circle have con- spicuously long anteriorly directed manu- bria and those in the third circle have prom- inent but shorter anteriorly directed manu- bria (Fig. 1). The brain was not mentioned in the orig- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CO N (sod) (97) 87-07 :(1UB) (EZ) 8~—-O~ (L'E1) OC-LAS' HI) 91-11 (900) L0'0-70'0 49 (8ZL'1) 7 C-IV'I (90'0) LO'0-S0'0 49 (617) 09°T-SS I (Q0Z) 8O0Z-LLI 49 (799) 8EL-TLS syooy ZI ‘(1€) rE-L7 syooy ZI :(Zr) Ly-SE syooy ZI *(19) L9-vS syooy ZI (LL) T8-OL (€OT) OTZ—-SLI 49 (COET) SHI-CII (8€'0) 6¢'0-67'0 49 (L6'b) OT S-9G'E Ajyeastp yyds ‘repnsueiw Ury posny sdn ynq ‘Ajlowsjue poyesedas AyTeurpmirsuoy wWyYds ‘repnsueLn peorg SOADZIS IeTNONNS podeys-su0d YIM ‘aseq je Jsapeolg gjorjdao -a1 srosogoid jo wed Jo1sa}sod-onUsA payUspuT UT ‘Te[NSuULII], BLIQnUeL JIM SYOOY psy} pue PUOSES JO S}OOY SUITE IePNONNS poyeaojsa ATQULIeA YA podeys-—) poWoAul “C"] :] (SOTOSNU JOJORNOI UL) 7 SNOSIUWTOT YORs UT | yuosge Ajoresr {[eUoA | ‘Tesrop (Z Ajeser) | ayesuojo ‘UI, Ajjoriajsod pemusd pedo ‘A[JOLA]Ue pesie[us ‘TeoTIpuT[AD (soyeulay L] ‘sofeut ¢{) (4oded sty}) syupeqns (vd) OVI SI-9/LT CI (S00) L0'0-Z0'0 44 (ZO'1) L8°T-ZI'0 (S00) LO'0-ZO'0 44 (171) SO"E-ST'0 (O9T) 077-06 44 (OSS) OOS I-OSZ syooy ZI *(8Z) €€-v7~ syooy ZI ‘(€r) OS-9E syooy ZI ‘(LS) 99-6r syooy ZI :(89) 6L-09 (O0Z) O€Z—-06 44 (OTT) OSI-OOI x(6€'0) $S°0-07'0 49 (Z0'9) S'TI-6'1 (qd ‘OZ ‘3tq) wads io , mm long, ca. O17 mm wide, without thin margins, apices shortly acute. Florets 30—35 in a head; co- rollas white, ca. 3 mm long, basal tube ca. 1.2 mm long, throat ca. 1.7 mm long, lobes ca. 0.5 mm long, densely pilosulous out- 243 side; anther thecae ca. 0.7 mm long, whit- ish. Achenes ca. 1.7 mm long, nearly gla- brous, with few short setulae on distal ribs; carpopodia short-cylindrical; pappus whit- ish, ca. 3 mm long, of ca. 25 fragile capil- lary bristles, slightly broadened distally. Ageratina pseudogracilis is known only from the type collection. It resembles A. gracilis (H.B.K.) R.M.King & H.Rob. in habit, with its small size and the small, sub- sessile leaves. The species is evidently more erect, the stems and leaves are pubes- cent, and the achenes are glabrous or nearly glabrous. Ageratina (Andinia) regalis H.Rob., sp. Nov. jaggy, Y) Type: Ecuador. Azuay: 8-10 km S of Cumbe on road to Loja, 9050—9400 ft, 22 Jan 1979, King & Almeda 7754 (holotype US, isotypes CAS, MO). Ad A. pseudochilcam in laminis foliorum coriaceis et subquinquenervosis similis sed in axilis venosi abaxialibus piliferis distinc- ta. Shrubs or small trees 1—4 m tall, mod- erately branched, branches ascending at ca. a 45° angle or less, not or scarcely viscid on stems, leaves and branches of inflores- cence, with sparse minute puberulence; stems brownish, terete with slight costae; internodes mostly 1.0—5.5 cm long. Leaves opposite, somewhat persistent along distal branches, petioles 5—15 mm long; blades thinly coriaceous, narrowly ovate to ovate- elliptical, mostly 4.0—6.5 cm long, 1.2—3.0 cm wide, base rounded or obtuse to broadly acute, margins with usually 10—22 teeth distal to the basal fifth, apex obtuse to acute, adaxial surface dark green, nitid, with prominulous larger veinlets, slightly insculpate smaller veinlets, glandular dots scattered, abaxial surface subnitid, with close reticulum of prominulous veinlets, smaller veinlets usually dark, minute glan- dular dots scattered, not restricted to vein- lets, pilose with pale hairs along sides of 244 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ag eralin a pseudogracilis MBL, Lf VENEZUELA Kupat . Get H. Robinson, U.S. Metions Herbenum. 4470 | — UNITED STATES Sufriitex. Hoja verde oscura. Corolas blancas. J, CUATRECASAS 2 69 2 y TACHIRA: Paramo de El Coloredo, {continuacion de El Zumbador), cumbre on 59 del pdramo 3200-3300 m. . .oQ0 ’ J. Cuatrecasas, No. 20381 20-1-1973 NATIONAL HERBARIUM Ne (28381 L. Ruiz-Terin M. Lépez-Figueiras Fig. 8. Ageratina pseudogracilis H.Rob., holotype, Cuatrecasas, Ruiz-Terdn & Lépez-Figueiras 28381 (US). VOLUME 116, NUMBER 1! 245 PLANTS OF ECUADOR Cyrological and/or Biochemical Vouchers 2 ROBERT MERRILL KING 22 January 1979 AND FRANK ALMEDA, No. 7754 ag Agera ina preudochiles (Bentl.) Ker. Azuay: 8-10 km S of Cumbe on the road to Loja. Elevation 9050-9400 ft. HH ieee slender shrub 1.5m tall. Florets white but linged {J O1O'Y} with pink. Common. centimeters UNITED STATES Ageratina regalis H.Rob. i ES a eh ee a Se NT det. Harold Robinson (US), July 2001 Speemens collected for the United States National Herhartn. Smithsonian NATIONAL HERBARIUM Institution and the Herbarium of the California Academy of Sciences. Fig. 9. Ageratina regalis H.Rob., holotype, King & Almeda 7754 (US). 246 midvein and on bases of secondary veins, especially obvious in axils of large second- ary veins; two pairs of strongly ascending secondary veins near base of blade, pair nearest base weak, near margin, distal pair more ascending, spreading at a 30—40° an- gle, usually straight or with convex curve from base. Inflorescences terminal on leafy stems, fastigiate with ascending branches, forming rather dense slightly rounded cor- ymbiform clusters of heads; peduncles O—1 mm long, with heads mostly in sessile sub- groups of 2 or 3. Heads 7—8 mm long; in- volucral bracts weakly subimbricate, with ca. 7 subequal bracts, pale green, oblong to elliptical, mostly ca. 4 mm long, apices sub- acute, scarious, scarcely bicostate outside at base, with many glandular dots. Florets 5— 7; corollas white to pale pink or with pink lines, ca. 5.0 mm long, basal tube 1.5—1.8 mm long, with scattered minute glandular dots, throat ca. 2.5 mm long, glabrous, lobes ca. 0.7 mm long, glabrous; anther the- cae 1.2—1.5 mm long, reddish. Achenes ca. 2.5 mm long, with minute setulae or glan- dular dots along ribs, glands on sides near pappus; pappus rather persistent, whitish or base pale pink, 3—4 mm long, of 30-35 bristles, slightly flattened, not broadened distally. n = ca. 42, K689/ (King et al. 1977, as A. pseudochilca). Paratypes: Ecuador. Cafiar: Region of San Marcos, ca. 10 km NE of Azogues, 9500 ft, 25 Sep 1944, Prieto P-69 (NY, US). El Oro: Cerro de Chilla, 03°30'S, 79°38'W, 3595 m, 3 May 1997, Vargas & Canaday 1423 (MO, QCNE, US). Chim- borazo/Canar: Western escarpment, near, El Tambo, 10,000—11,500 ft, 6-9 Jul 1945, Camp E-4099 (NY, S). Canar: Between Biblian and Cafar, ca. 3200 m, 21 Sep 1955, Asplund 17683 (S); ca. 2900 m, As- plund 17710 (S). Paramo between Biblian and Canar, 3350 m, 5 Aug 1959, Harling 6285 (S). Partidero El Corte—San Miguel de Porotos—Parcialidad Jatupamba—Cerro La Comuna, zona himeda, 9400—10,000 ft, 26 May 1979, Jaramillo 985, 1032 (AAU, QCA). Azuay: Cruz Pamba region above PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Banos (ca. 15 km SW of Cuenca), 9000— 10,000 ft, 29-30 June 1945, Camp E-3921 (NY, S, US). Cumbe, 3000 m, 9 May 1947, Harling 824 (S). 10 km S of Cumbe, 3000 m, 8 Jul 1959, Harling 5660 (S). Sayausid, ca. 3000 m, 1 Apr 1968, Harling, Storm & Strom 7930 (GB). Road Cuenca—Giron (ca. 10 km N of Giron), ca. 2700 m, 2 Apr 1968, Harling, Storm & Str6m 7949 (GB, US). Cumbe, S of Cumbe, ca. 2700 m, 22— 24 Apr 1968, Harling, Storm & Strom 8703 (GB, US). Along road to Limon, ca. 17 km E of Gualaceo, ca. 8900 ft, 31 Jan 1974, King 6631 (MO, US). Along road to Loja, ca. 6 km generally SE of Cumbe, ca. 9600 ft, 4 Feb 1974, King 6695 (MO, US). Por- tete del Tarqui, Cuenca—Gir6én, 2700—2900 m, 5 Apr 1974, Harling & Andersson 13193 (GB, US). Road Cuenca—Loja, upper northern slopes of Valle del Rio Leon and southern parts of Paéramo de Tinajillas, ca. 2800 m, 9 May 1974, Harling & Andersson 14518 (GB, US). Bafios, SW of Cuenca, cultivated ground, ca. 2700 m, 12 May 1974, Harling & Andersson 14572 (GB). Along road to Loja, ca. 17 km S of Cumbe, ca. 9700 ft, 24 Jan 1976, King & Garvey 6891 (MO, US). Partidero Llantera—Chi- quintad—Saucay—Guandum, zona htmeda, 9200 ft, 27 May 1979, Jaramillo 1051 (AAU, QCA). Tarqui, near the monument, 2600 m, 5 Feb 1982, Harling, Bohlin, Lind- strom & Roth 20243 (GB, US). Loja: Jera 10 km NE of Saraguro, 03°24’S, 79°14'W, 2 Mar 1989, Ellemann 91651 (AAU, QCA, US). Common name: “urcu chilca’” (Que- chua). Use: for bath, a decoction of whole plant is used. Ageratina regalis seems endemic to the area of Azuay and nearby parts of Canar, El Oro, and Loja. The species occurs in pa- ramo or subpardmo vegetation. Specimens have been consistently identifed in the past as the more northern A. pseudochilca. The present species is instantly distinguishable by the abaxial leaf hairs along the midvein and bases of the secondary veins, which are particularly evident in the axils of the sec- VOLUME 116, NUMBER 1 ondary veins. The stems are also more te- rete, the leaf blades generally thinner, and the corollas often pink. The species is named after the collector of a number of the specimens and specialist in the Eupatorieae, Robert Merrill King. Ageratina rhodopappa f. glandularis (B.L.Rob.) H.Rob., comb. nov. Basionym: Eupatorium pichinchense f. glandulare B.L.Rob., Contrib. Gray Herb., n.s. 61:10. 1920. Type: Ecuador. Chimbo- razo: vicinity of Huigra, mostly on Haci- enda de Licay, Aug 1918, Rose & Rose 22210 (holotype GH, isotypes NY, US). Ageratina (Ageratina) rosei H.Rob., Sp. NOv. Rie. LO Type: Ecuador. Chimborazo: Vicinity of Huigra, mostly on Hacienda de Licay, 21 Aug 1918, Rose & Rose 24078 (holotype US). Ad A. rhodopappam f. glandularum in setis pappi roseis et pedunculis glanduliferis similis sed in laminis foliorum base tripli- nervatis differt. Herbs 5—10 dm tall, scarcely to moder- ately branched, often with many prolifera- tions from axils; stems slender, brownish to purplish, terete, usually puberulous to hir- tellous; internodes 1.5—6.5(—14.0) cm long. Leaves opposite, petioles 4-16 mm long; blades thinly papyraceous, ovate, mostly 1.2—5.0 cm long, 0.6—3.0 cm wide, widest near basal 5th, base subtruncate or shallow- ly cordate to rounded, margins crenate or serrate with 5—12 blunt teeth on each side, apex acute, rarely slightly acuminate; ad- axial surface sparsely and evenly pilosu- lous, abaxial surface slightly paler, pilosu- lous mostly on veins; triplinervate from base of blade at petiole, rarely just distal to base. Inflorescences terminal on leafy stems, broadly and usually rather flatly cor- ymbiform; branches ascending, usually with many minute stipitate glands; pedun- cles 5-17 mm long. Heads 6-8 mm high; 247 involucral bracts ca. 18, greenish or brown- ish-green with purplish tips, oblong-lance- olate, mostly subequal, 4-6 mm long, 1.0— 1.3 mm wide, apices acute, inner bracts with paler membranaceous often erose tips, scarcely ciliate, outside minutely puberu- lous with usually reddish hairs, narrowly bi- costate. Florets ca. 35—45 in a head; corol- las whitish, 3.5—4.0 mm long, basal tube slender, 1.0—1.5 mm long, throat 1.5—2.0 mm long, lobes 0.5—0.8 mm long, with few hairs outside; anther thecae 0.6-0.9 mm long, white. Achenes ca. 2 mm long, fusi- form, with many short setulae along ribs from carpopodium to pappus, very rarely glabrous; carpopodium cylindrical; pappus of ca. 20 whitish to mostly reddish bristles, fragile at base, 3.5—4.0 mm long, scarcely to distinctly broadened distally. n = 17 + [KOS 52- nu — ai, | 2 aitae.. KODo7,, (Kane et al. 1977, as A. azangaroensis). Paratypes: Ecuador. Locality uncertain: In locis herbidis subandinis, Sodiro 549 (US). And. Quitenorum, s.d., Jameson s.n. (S). El Oro: Railway station Cotopaxi, 3550 m, 26 May 1939, Asplund 6498 (S). Pi- chincha: Pichincha, a little above Quito, 3200-3500 m, 1952?, Fagerlind & Wibom 2035 (S). Antisanilla, 11,500 ft, Oct 1923, Anthony & Tate 337 (US). Cerca de la es- taci6n de Chiriaco, 2820 m, 2 Apr 1928, Firmin 421 (US). Quito, Panecillo, 2900 m, 16 May 1939, Asplund 6078 (S). Saloya (W of Quito), 2000 m, 28 June 1939, Asplund 7360 (S). Dauli (flag stop) on railroad SE of Corazon Volcano, 00°34'S, 78°36’W, 3300 m, 31 May 1947, Fosberg 27638 (US). Cordillera Occidental, los Vertientes de Pichincha, 11 Jul 1959, Barclay, Juaji- bioy & Tinajero 7819 (US). Cordillera Oc- cidental, Paramo de Trujillo, border of Prov. Cotopaxi, between peaks I[liniza to W and Cotopaxi to E, ca. 3450 m, 15 Jul 1959, Barclay & Juajibioy 7933 (US). Quebrada Yanahuaico, ca. 3 km W Conocoto, 2700— 2800 m, 3 Dec 1966, Sparre 13292 (S). Road St Domingo—Quito, Cornejo Astorga (Tandapi), ca. 1600 m, 7-10 May 1968, Harling, Storm & Str6m 9393 (GB). Cerro 248 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Hee oan Seem UNITED STATES | 7) Bi ioRa | NATIONAL HERBARIUM | | Holotype Ageratina rosei H.Rob. det. Harold Robinson (US), July 2001 UNITED STATES NATIONAL MUSEUM UNITED STATES DEPARTMENT OF AGRICULTURE GRAY HERBARIUM OF HARVARD UNIVERSITY UNITED STATES NATIONAL MUSEUM NEW YORK BOTANICAL GARDEN EXPLORATIONS IN SOUTH AMERICA Xo LY O TF Beara alos Nepal epee Yak a 5 Gia pera Hee re fee Disteit ASE. Soca Viclolty of Hulgrn, Ecunidoy, mostly on the Wachonda de Dieay J. N. ROSE, Collector cel A GEORGE ROSE, Asalstane AR eee Fig. 10. Ageratina rosei H.Rob., holotype, Rose & Rose 24078 (US). VOLUME 116, NUMBER 1 Pichincha, above Quito, 3500-5000 m, 2 Nov 1974, Gentry 12381 (MO). Along route from Quito to Santo Domingo de los Colorados between San Juan and La Palma via Chiriboga, 15 km NE of Chiriboga, PASO im: ZaSep 1976. Croat 335725) (MO): Along road between Quito and Nanegal, via Cotocollao and Nono, 18-21 km NW of Quito, 3100-3200 m, 4 Sep 1976, Croat 3S802 (MO). Paramo and shrub vegetation on the eastern slopes of Cerro Pichincha, 3600—4000 m, 26 Jan 1977, Harling, Elias- son & Andersson 14834 (GB). W slopes of Cordillera Occidental, W of Calicali, OOF ONS 7832 Weicas 27000ms 25: Mar 1979, Lojtnant & Molau 11358 (AAU, GB). W of Calicali, W slopes of Cordillera Occidental, km 12 on road from Calicalt, OO°O1'N, 78°33'W, 24 May 1979, Lojtnant & Molau 13619 (AAU, GB). Quito, Par- raoquia El Salvador—Toctiuco—Chorrera— Faldas del Pichincha, zona himeda, pen- diente, 11,000—11,400 ft, 24 Jun 1979, Jar- amillo & Jaramillo 1206 (AAU, QCA); Heda. Mi Cielo—Faldas del Pichincha, zona humeda, 11,000 ft, 20 Jul 1979, Jaramillo 1260 (AAU, QCA): via Toctiuco—Chor- rera—Piramide (Antena de Television) Fal- das del Ruccu—Pichincha, zona hutmeda, 11,600 ft, 11 Aug 1979, Jaramillo & Las- cano 1328 (AAU, QCA); Pasochoa—Local- idad Sur Oriente, colecciones en bosque de Coja Andina y pajonal, zona humeda, 3500—3650 m, 17 Nov 1979, Jaramillo, Coello, Narvaes, Elisabeth, Bravo & Rasa 1567 (AAU, QCA). Carretera Chillogallo— San Juan—Chiriboga—Empalme, colecciones en alrededores de San Juan, 00°18’S, 78°39'W, 3100-3260 m, 8 Sep 1985, Zak & Jaramillo 621 (MO, US). Carretera Qui- to—San Juan—San José de la Victoria, en la cercania de San José, 2900-3400 m, 24 Dec 1987, Zak & Jaramillo 3280 (MO, US). Cotopaxi: Volcan Cotopaxi, above Limpio Pamba, 00°37'S, 78°26'W, 3781 m, 10 Jul 1963, Ugent & Albornoz 5662 (US). Volcan Cotopaxi WNW slopes (NE Ami Grande) 4000 m, 26 Apr 1967, Sparre 15777 (S). Road Salcedo—Napo, km 21, 249 3800 m, 5 Sep 1967, Sparre 18420 (S). Road Quevedo—Latacunga, Zumbagua, W of Zumbagua, ca. 3200 m, 2 May 1968, Harling, Storm & Strém 8942 (GB, US). Tungurahua: Along road to Bafios, ca. 1 km SE of Ambato, ca. 8700 ft, 20 Jan 1974, King 6532 (MO, US). Mocha, alrededores del Pueblo, 01°27’S, 78°40’W, 2900 m, 30 Sep 1995, Villacres 257 (F, US). Bolivar: Railway station Cotopaxi, 3550 m, 26 May 1939, Asplund 6498 (S, US). Along road to Guaranda (route 7), ca. 28 km generally SW of Ambato, ca. 11,000 ft, 25 Jan 1974, King 6587 (MO, US). Chimborazo: Vicin- ity of Huigra, mostly on Hacienda de Licay, 28 Aug 1918, Rose & Rose 24078a (US); 6 Sep 1918, Rose & Rose 24078b (US). Canon of Rio Chanchan near Huigra, 4000—4500 ft, 7-14 May 1945, Camp E- 3148 (NY, S, US). Between San Andrés and Cuatro Esquinas, Oct 1952, Fagerlind & Wibom 881 (S). El Retén (20 km south of Cebadas, 31 Jan 1968, Harling, Storm & Strom 6793 (GB). Carretera Riobamba—Lic- to—Pungala—Alao—Huamboya, en la quebra- da Supaychuan, 01°50'S, 78°30'W, 3570-— 3620 m, 26 Dec 1987, Zak & Jaramillo 3304 (MO, US). Cafar: between Biblidan and Canar, edge of thicket, ca. 3200 m, 21 Sep 1955, Asplund 17682 (S). Azuay: Cerro Soldados (W of Sayausid), below Quinoas, ca. 3400 m, 25 Apr 1968, Harling, Storm & Strém 8748 (GB). Napo: Alrededores de la Cordillera de Los Llanganati, Chihuila Sacha o Ainchilibi, 25—29 Aug 1959, Bar- clay & Juajibioy 9004 (US). Morona-San- tiago: Parque Nacional Sangay, Peace Corps Sangay Mountaineering Expedition, between Plazapamba and /Yanayacu, 02°02'S, 78°25'W, 3200-3600 m, 23-31 Dec 1995, Clark, Harris, Sulser, Teizeira & Bano 1829 (MO, QCNE, US). Three additional paratypes are tripliner- vate slightly above base of the blade: Ecuador: Pichincha: Crescit prope Quito, 1864, Jameson s.n. (US). Puingasi (cul.) en um cerco, ca. 3000 m, 18 Apr 1927, Firmin 46 (US, achenes without setulae). Quito Canton, carretera Quito—Nono—Tandayapa, 2506 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON OO°O1'N, 78°40'W, 1500-1800 m, 5 Sep 1991, Rubio 2083 (MO, QCNE). Peru: Dept. Lambayeque: Prov. Ferrena- fe: Sinchigual—Laguna Tembladera, Distrito Incahuasi, 3100 m, 13 Sep 1985, Sagdste- gui, Skillman, Mostacero & Ramirez 12859 (EF HUT, US). Ageratina rosei occurs mostly in central Ecuador at middle and higher elevations. Habitats cited include paramo on dry rocks of high cliff above grassy slope, on ridges in montane forest area, bunch-grass asso- ciation, under shrubs along ditch between cultivated fields, and roadside. The combi- nation of often reddish pappus and glan- duliferous inflorescence branches tends to distinguish the species. It is named for the collector of some of the earlier specimens, Joseph Nelson Rose, and its pappus color was also taken into consideration in the choice of the name. Some of the early col- lections of the species were identified by B. L. Robinson as his Eupatorium pichinchen- sis f. glandulare, the type of which has the triplinervation of the leaf far distal to the base of the blade, and it is placed now in A. rhodopappa. Both A. rosei and A. rho- dopappa \ack the long stem hairs of A. pi- chinchensis, although some specimens of that species have glands on the inflores- cence branches and reddish pappus bristles. More recent collections of A. rosei have of- ten been determined as A. azangaroensis (Sch.Bip. ex Wedd.) R.M.King & H.Rob., an apparent synonym of A. glechonophylla (Less.) R.M.King & H.Rob. that has gen- erally larger heads, corolla throats much longer in comparison to their lobes, and achenes with coarser setulae on the ribs and distal lateral surfaces. The Peruvian speci- men has generally longer peduncles, 2.5 cm long. Ageratina (Ageratina) serrulata H.Rob., n.sp. ioe isl Type: Colombia. Antioquia: Ituango, camino entre El Retiro y Cerro Paramillo, 3200—3600 m, 25 Feb 1993, Cuadros 5003 (holotype HUA, isotypes MO, US). In habitus ad A. gracilem similis sed in foliis flaccide pilosis in foliis erecto-paten- titer multiserrulatis et in achentis glabris distincta. Subshrubs 15-20 cm high, sparsely branched below, distal parts densely branched; stems dark brown when dry, densely hirsute with flaccid hairs; inter- nodes mostly 4—8 mm long, to 25 mm long on young shoots. Leaves opposite, petioles 2—4 mm long; blades papyraceous, ovate, 10-22 mm long, 5-12 mm wide, base rounded to broadly obtuse, margins with 5— 11 small sharp teeth, apex acute, adaxial surface pilosulous with sparse hairs, abaxial surface with long flaccid hairs spreading from midvein and larger secondary veins, triplinervate at the base of blade, secondary veins spreading at a 30—35° angle. Inflores- cences terminal on leafy stems, often over- topped by lateral leafy shoots, branches slender, few, sparsely hirtellous, peduncles 4—17 mm long. Heads ca. 5 mm high; in- volucral bracts 15—17, lanceolate, subequal, ca. 4.5 mm long, 0.7 mm wide, narrowly acute, glabrous outside, inner bracts thinner, paler, bicostate. Florets ca. 30 in a head; corollas white, ca. 3.5 mm long, basal tube ca. 1.5 mm long, throat ca. 1.3 mm long, lobes ca. 0.6 mm long, with few or no hairs; anther thecae whitish, ca. 0.7 mm long. Achenes ca. 2 mm long, glabrous; carpo- podium narrowly cylindrical; pappus whit- ish, ca. 3 mm long, of ca. 20 fragile bristles, scarcely broadened distally. Ageratina serrulata is known only from the type collection. The species is one of four described in this paper that have at times been identified as A. gracilis (H.B.K.) R.M.King & H.Rob. All are from above 3000 m and differ from the Colombian A. gracilis by the more erect habits and the pubescence on the leaf blades. The Ecua- dorian species, A. parviceps, is usually a larger plant with much longer petioles, and it has definitely setuliferous achenes. The western Venezuelan A. pseudogracilis and VOLUME 116, NUMBER 1 5) Lu U} vd i slay o QlIe Z\5_ 5 pe Er 216" 3 E aio —=3 E. cs Sa COLOMBIA | ASTERACEAE ; = 3) 2 Eee : Agertina grealis (\161 2s Antioquia: Ituango Municipio A yds I D { Aoeratine ser rulala H Pod, Antioquia, Ituango, camino entre El = Retiro y Cerro Paramillo. pee 3200-3600 m UNITED STATES isolype Inflorescencia blanca. 2S febrero 1993 3328083 NATIONAL HERBARIUM Hermes Cuadros 5003 MISSOURI BOTANICAL GARDEN HERBARIUM (MO) Fig. 11. Ageratina serrulata H.Rob., isotype, Cuadros 5003 (US). N southern Colombian A. ewanii differ by the broader more membranaceous tips on the involucral bracts, the very short internodes of the stems, and nearly but usually not completely glabrous achenes. Also, A. ewa- nii has a series of outer involucral bracts usually distinctly shorter than the inner ones. The present A. serrulata is most dis- tinct in the many sharper more jutting small teeth of the leaves and by the lanceolate essentially glabrous bracts of the involucre. Ageratina (Ageratina) villonacoensis H.Rob., n.sp. Fig. 12 Type: Ecuador. Loja: Eastern slopes of Cerro Villonaco, grassland or dense shrub vegetation, 2500—2600 m, 12 Apr 1974, Harling & Andersson 13439 (holotype US, isotype GB). Ad A. pichinchensem in caulibus villosis et in inflorescentiis compactis similis sed in laminis foliorum base cordatis margine acute dentatis et apice valde acuminatis et in bracteis involucri longe acuminatis dis- tincta. Subshrubs to ca. 1 m tall, sparsely branched; stems scrambling, terete, densely villous with often violet septate hairs over 1 mm long; internodes 6—9 cm long. Leaves opposite, petioles 5-18 mm long; blades broadly ovate, 2.5—-5.0 cm long, 2—4 cm wide, papyraceous, base rounded to shal- lowly cordate, margins coarsely serrate, with 7—10 teeth on each side, apex distinct- ly acuminate, adaxial surface pilose, abaxial surface pilose on veins; triplinervate from base, often from basal acumination, with secondary veins spreading at ca. a 45° an- gle. Inflorescences terminal on leafy stems, with branches spreading at a 30—45° angle, densely villous; ending in rather dense cor- ymbiform clusters of heads; peduncles 3—6 mm long. Heads 7-8 mm long; involucral bracts 20—22, in ca. 2 series, papyraceous, subequal, lanceolate, 6-7 mm long, 0.8—1.0 mm wide, outer bracts purplish, inner bracts stramineous, tips narrowly acute to long- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON acuminate, outside pilose with violet hairs. Florets ca. 30—35 in a head; corollas white, 4.0—4.3 mm long, pilosulous on distal part of throat and on lobes, basal tube slender, ca. 2 mm long, throat ampliate, becoming cylindrical, ca. 1.7 mm long, lobes ca. 0.7 mm long, pilosulous with fine hairs; anther thecae ca. 1 mm long, whitish. Achenes ca. 2 mm long, fusiform, setuliferous on distal part of ribs and on sides near pappus; pap- pus whitish, of ca. 25 capillary bristles ca. 3.5 mm long, fragile at base, slightly broad- ened distally. Paratype: Peru. Cajamarca: Prov. Caja- marca, Cerro Chupacotoy, 2800 m, 2 June 1992, Ochoa & Salas 16139 (US). Ageratina villonacoensis 1s known only from the type collection in southern Ecua- dor and a single similar but smaller-headed and less pubescent specimen from Cayja- marca, Peru. It is similar to A. pichinchensis and A. sodiroti. It has the longer pubescence of the former, without any trace of stipitate glands sometimes found in that species. It has the larger more acute teeth of the leaf margins as in A. sodiroi, but the latter has shorter pubescence. The present species dif- fers from both the related species by the more strongly cordate bases of the leaf blades and by the narrow often long-acu- minate tips of the involucral bracts. Ageratina (Ageratina) websteri H.Rob., sp. nov. Jae, 1S Type: Ecuador. Pichincha: Secondary montane rain forest, trail to waterfall, Re- serva El Pahuma, weed with white flowers, OO°O1'N, 78°38'W, 1920 m, 18 June 1996, Webster 31619 (DAV, US). In habitis decumbentibus et in caulibus rubri-puberulis et in laminis foliorum 20— 32 mm longis et margine argute serratis et in capitulis laxe dispositis et in bracteis in- volucri e acheniis vix longioribus et in flo- ribus in capitulo 20—25 et in tubis corollar- um sparsim minute pilosulis distincta. Herbs with long decumbent rooting VOLUME 116, NUMBER 1 lal eS = == =e o = OS nie <3 ae = pee | | | L } Ageritina villonz coensts HReb. Holotype 2c0l Get H Robinson, U.S. National Herbarium, 19 FLORA OF ECUADOR DR. GUNNAR HARLING'S FOURTH ECUADOREAN EXPEDITION | 13439 Aneratine prchinchen sis (BR. | ova: Lastern slopes of Cerro Villonaco, grassland or dense shrub OMB ea ee | pen on alt, 2500 - 2600 s.s.a. ; Liana-like. Corolla white, 2848966 | NATIONAL HERBARIUM | 49.19.1974 G. HARLING & L. ANDERSSON | 12.1¥.19 Fig. 12. Ageratina villonacoensis H.Rob.., holotype, Harling & Andersson 13439 (US). 254 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ag ertina webs teri Kicab Polat pe det H. Robinson, U.S. Nations! Herbertum, OE | UNITED STATES 3347396 NATIONAL HERBARIUM Asteraceae ECUADOR VASCULAR FLORA OF MAQUIPUCUNA Ageratin™ psilodoram (BLM) KR. PICHINCHA : Secondary montane rain forest, trail to waterfall, Reserva El Pahuma; weed with white flowers. Lat/Long: 00° 01 30 .N 78° 38 W EL. 1920 meters dune 18 1996 Grady L. Webster 31619 Alfonso del Hierro Det: The UC Dayis Herbarium, Fundacién Maguipucuna & Herbirio QNCE Fig. 13. Ageratina websteri H.Rob., holotype, Webster & Hierro 31619 (US). VOLUME 116, NUMBER 1 stems and projecting leafy branches 4—5 dm long, branching at ca. a 45° angle; stems brownish to dark reddish, densely puberu- lous with reddish hairs; internodes mostly 1.5—4.0 cm long. Leaves opposite, petioles 4—7 mm long; blades membranaceous, ovate, 20-32 mm long, 8-14 mm wide, base obtuse, margins serrate with 4—6 sharp teeth, apex acute, adaxial surface dark green, nearly glabrous, with scattered mi- nute hairs, more puberulous on veins near base, abaxial surface slightly paler, with few hairs along veins; triplinervate 1.0—1.5 mm distal to base of blade, secondary veins spreading at ca. a 25—30° angle. Inflores- cences terminal on leafy branches, with 1— 7 laxly disposed heads; branches puberu- lous with reddish hairs; peduncles slender, 12—25 mm long. Heads 5—6 mm high; in- volucral bracts ca. 15, in 1—2 series, mem- branaceous, few outer bracts darker green, narrowly oblong, subequal, 2.0—2.5 mm long, ca. 0.5 mm wide, apices obtuse, densely ciliate with fine hairs, outside pu- berulous with sparse minute hairs, costae not obvious. Florets 20—25 in a head; co- rollas whitish, ca. 2.8 mm long, with fine pale hairs on basal tube, basal part of throat and on lobes, basal tube slender, ca. 1.5 mm long, throat ca. 1 mm long, lobes ca. 0.3 mm long; anther thecae ca. 0.5 mm long, whitish. Achenes ca. 2 mm long, with mi- nute setulae often biseriate along ribs, sparse distally on sides; carpopodium cylin- drical; pappus ca. 2.8 mm long, whitish or 255 pinkish, of ca. 25 slender bristles, not or scarcely broadened distally. Ageratina websteri is known only from the type collection. It was originally deter- mined as possibly A. psilodorum, a Col- ombian species, but it does not have the glabrous stems of that species. Relationship is regarded here as closest to A. parviceps, which has a more erect habit with more heads on shorter peduncles, over 30 florets in the heads, corolla tubes and throats with- out hairs, larger more overlapping involu- cral bracts distinctly longer than the mature achenes, an almost strictly whitish pappus, and which occurs at elevations usually at least 1000 m higher. Acknowledgments The extensive technical help of Marjorie Knowles is acknowledged. The scans of the U.S. type specimens were prepared by Su- san Hunter of the Department of Botany. Literature Cited King, R. M., & H. Robinson. 1970. Studies in the Eu- patorieae (Compositae). XIX. New combina- tions in Ageratina.—Phytologia 19:208—229. , & . 1990. Notes on Ageratina in Me- soamerica (Eupatorieae: Asteraceae).—Phyto- logia 69:61—86. , D. W. Kyhos, A. M. Powell, P. H. Raven, & H. Robinson. 1977. Chromosome numbers in Compositae, XIII. Eupatorieae —Ann. Missouri Bot. Gard. 63:862—888. Turner, B. L. 1997. The Comps of Mexico, a system- atic account of the family Asteraceae. I. Eupa- torieae.—Phytologia Mem. 2:i—iv, 1—272. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(1):256—261. 2003. ERRATUM Erratum regarding.—Marcus, J. and S. Hourdez. 2002. A new species of scale- worm (Polychaeta: Polynoidae) from Axial Volcano, Juan de Fuca Ridge, northeast Pa- cific. Proceedings of the Biological Society of Washington 115(2):341—-349. Statement.—Higher printing quality of the figures illustrating the new species Vampiropolynoe embleyi (Polychaeta; Po- lynoidae) are presented here. Due to print- ing problems the figures issued with the original publication did not satisfactorily il- lustrate the morphological detail described in the text. VOLUME 116, NUMBER 1 257 a 500 um Fig. 1. A. Drawing of paratype (R473-6096), dorsal view of whole specimen. Dorsal cirri have broken off. B-E, SEM views. B. Elytron from the mid-section, dorsal view. C. Detail of elytral surface. D. Dorsal view of segments 11 (bottom)—14 (top). White arrow points to cirrophore on segment 12, black arrow points to digitiform dorsal tubercle on the same segment. E. Detail of first dorsal tubercle on segment 6 (black arrow). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 500 um Fig. 2. A, C, and D, SEM views. A. Dorsal view of anterior part of worm. White arrows point to the strong aciculae protruding from the Ist segment. Black arrow points to the left hooked frontal filament. B. Drawing of the anterior part of paratype (R473-6096), pharynx partially everted. C. Frontal view of prostomium showing median antenna, tentacular cirri, aciculae (white arrows). D. Detail of the buccal cirrophore (black arrow), and second ventral cirrus (3rd segment). VOLUME 116, NUMBER 1 259 Fig. 3. A and C, SEM views. A. Ventral view of a partially everted pharynx showing upper, lateral and ventral lips with their papillae and ridges. Black arrowheads point to lip ridges, asterisk points to shared bulbous base of the two median papillae on the lateral lip. B. Drawing of paratype (R473-6096), ventral view of a more everted pharynx, with the buccal papillae forming a crown at the base of the pharynx. C. Keratinized teeth at the junction between the mouth opening and the pharynx. 260 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1 mm 1 mm Fig. 4. A—D, SEM views; E and FE drawings. A. Posterior view of elytrigerous parapodium (segment 15). Arrow points to the ligule of the notopodium. B. Anterior view of the parapodium from the same segment. Arrow points to the ligule of the neuropodium. C. Posterior view of cirrigerous parapodium (segment 14). Arrow points to the ligule of the notopodium. D. Anterior view of the parapodium from the same segment. Arrow points to the ligule of the neuropodium. E. Posterior view of cirrigerous parapodium of segment 24, note the protruding acicula. K Anterior view of the same parapodium depicted in E. VOLUME 116, NUMBER 1 261 Fig. 5. A-—E, SEM views. A. Tip of a notoseta. B. Feather-like neurosetae on the lower side of the neuro- podium. C. Upper neurosetae. D. Detail of neurosetae tips. E. Ventral view of segments 11—14 showing incon- spicuous nephridial papillae (white arrows). Leet is eviomld sicdel, <_ bom) aoe = ‘ ) ee = se “ ~ \ j > Tz y re . ae : - ‘ Nie AS ; _— : : - he y en . i. 4 La i Pe= =i - i im, on * ; el a i it i ~~. ‘ 7 a *s Si : ¥ = t ~ —_ 4 a nS or 1 PSK Py eee « Peay, Hi) ieee ot a4 Pa 2 Pe oe eee ee : on : a ; ee eee , : 7 : ee Wt | fe bun & — os = = 1D yan. | Ce. oy an x mM i an : -. ram , ? it! 1 Swng » we Lal Ae uel ih ai i: Melt yere eer Liu ed . ay PN Sa ~ > dion ine eo ie a od co - @ 7 .-s ™~ ey ; _ coo io ; poagametnes 4 - wna “ mer -“_£; 7 : oe Te tall tneriacs : " rontvan sling eid ind rut (igs Cee 2 = . . =, A i . € , 4 fet “4 Wi ake INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Richard v. Sternberg, NCBI—GenBank, Building 45, Room 6An. 18D-30, National Institutes of Health, Bethesda, MD 20892-6510. (Do not submit manu- scripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in “GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON,” a supplement to Volume 114, number 4, December 2001. Authors are encouraged to consult this article before manuscript preparation. Copies of the article are avail- able from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page charges. The Society is a non-profit organization with limited funds, depending on financial resources and upon author request, it may subsidize a limited number of pages per vol- ume at the discretion of the editor and treasurer. Payment of full costs is encouraged and will facilitate speedy publication. Please note: Page charges are not automatically waived with membership. Authors requesting a partial waiver of page charges must justify need by emailing the editor and treasurer before proofs are returned and on the reprint request form received with page proofs. Please include email address on all correspondence. Costs.—Printed pages @ $65.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. Front cover—from this issue, p. 238. ne ee SSS SS ee, ee CONTENTS Cranial variation in Columbian white-tailed deer populations: implications for taxonomy and restora- tion Winston P. Smith, Leslie N. Carraway, and Thomas A. Gavin 1 New subgenus of leptarctine (Carnivora: Mustelidae) from the Late Miocene of Nebraska, U.S.A. Jong-Deock Lim and Larry D. Martin 16 Species limits in the pygmy antwren (Myrmotherula brachyura) complex (Aves: Passeriformes: Thamnophilidae): 1. The taxonomic status of Myrmotherula brachyura ignota Morton L. Isler and Phyllis R. Isler ZS) New fossil material with a redescription of the extinct Condor Gymnogyps varonai (Arredondo, 1971) from the Quaternary of Cuba (Aves: Vulturidae) William Suarez and Steven D. Emslie 29 A new gecko (Reptilia: Squamata: Genus Lepidodactylus) from Tuvalu, South-central Pacific George R. Zug, Dick Watling, Tataua Alefaio, Semese Alefaio, and Claudia Ludescher 38 Ompok pinnatus, a new species of silurid catfish (Teleostei: Siluriformes: Siluridae) from mainland Southeast Asia Heok Hee Ng 47 Halidesmus socotraensis new species and Haliophis guttatus (Forsskal), new records of congrogadine fishes from the Socotra Archipelago (Perciformes: Pseudochromidae) Anthony C. Gill and Uwe Zajonz a2 Calocidaris micans (Cidaridae) and Pseudoboletia maculata (Toxopneustidae): additions to the sea urchin fauna (Echinodermata: Echinoidea) of the Gulf of Mexico Richard L. Turner and Bruce D. Graham 61 The genus Chrysopetalum Ehlers, 1864 (Annelida: Polychaeta: Chrysopetalidae) in the Pacific coast of Panama Maria Teresa Aguado, Maria Capa, and Guillermo San Martin 82 Annotated checklist of decapod crustaceans of Atlantic coastal and continental shelf waters of the United States Martha S. Nizinski 96 A new caridean shrimp of the family Alvinocarididae from thermal vents at Menez Gwen on the Mid- Atlantic Ridge Timothy M. Shank and Joel W. Martin 158 New brachyuran crabs (Crustacea: Decapoda) from the Upper Pliocene Yorktown Formation of south- eastern Virginia Warren C. Blow ——-168 Columbasellus acheron, a new genus and species of subterranean isopod from Washington (Crustacea: Isopoda: Asellidae) Julian J. Lewis, Joel W. Martin, and Regina Wetzer 190 Bahadzia patilarga, a new species of subterranean amphipod crustacean (Hadziidae) from Cuba Thomas R. Sawicki, John R. Holsinger, Manuel Ortiz, and Abel Pérez 198 A new species of Cymbasoma (Crustacea: Copepoda: Monstrilloida) from the Pacific coast of Costa Rica, Central America Eduardo Suarez-Morales and Alvaro Morales-Ramirez 206 Description of subadult Pallisentis (Pallisentis) rexus (Acanthocephala: Quadrigyridae) from the ver- tebrate intermediate host in Thailand with an examination of the species identity Omar M. Amin and Horst Taraschewski 215 A new genus of the scleraxonian family Coralliidae (Octocorallia: Gorgonacea) Frederick M. Bayer and Stephen D. Cairns 222 New taxa and combinations of Ageratina from Ecuador, Colombia, and Venezuela (Eupatorieae: Oxylobinae) Harold Robinson 229 Erratum 256 NSTITUTION LIBRARIES a 01205 2478 iu 3 9088 ISSN 0006-324X PROCEEDINGS of THE Sremeeuri AL SOCIETY or WASHINGTON 11 AUGUST 2003 VOLUME 116 NUMBER 2 THE BIOLOGICAL SOCIETY OF WASHINGTON 2002-2003 Officers President: Roy W. McDiarmid Secretary: Carole C. Baldwin President-elect: W. Ronald Heyer Treasurer: TY. Chad Walter Elected Council Michael D. Carleton G. David Johnson Clyde Roper Michael Vecchione Marilyn Schotte Don Wilson Custodian of Publications: Storrs L. Olson PROCEEDINGS Editor: Richard v. Sternberg Associate Editors Classical Languages: Frederick M. Bayer Invertebrates: Stephen L. Gardiner Plants: Carol Hotton Frank D. Ferrari Insects: Wayne N. Mathis Rafael Lemaitre Vertebrates: Gary R. Graves Invertebrate Paleontology: Gale A. Bishop Carole C. Baldwin Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Annual dues are payable on or before January 1 of each year. Renewals received after January 1 must include a penalty charge of $5.00 for reinstatement. Library subscriptions to the Proceedings are: $50.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an additional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership, subscriptions, and/or back issues should be sent to: BIOLOGICAL SOCIETY OF WASHINGTON P.O. BOX 1897 LAWRENCE, KANSAS 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, and editorial questions should be sent to: EDITOR, RICHARD v. STERNBERG NATIONAL CENTER FOR BIOTECHNOLOGY INFORMATION—GENBANK BUILDING 45, ROOM 6An. 18D-30 NATIONAL INSTITUTES OF HEALTH BETHESDA, MD 20892-6510 Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Periodicals postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):263—274. 2003. Notes on the genus Ruellia (Acanthaceae) in Bolivia, Peru and Brazil D. C. Wasshausen and J. R. I. Wood (DCW) Department of Systematic Biology—Botany, National Museum Smithsonian Institution, Washington, D.C. 20560-0166, U.S e-mail: wasshausen.dieter@nmnh.si.edu; (JRIW) 111 Forest Road, Liss, Hampshire GU33 7BP, ‘Steat Brita. an, e-mail: jriwood@hotmail.com “to, NSS Op Natuyal History, “4V/ ae " Op; iia Abstract.—Taxonomic notes on Ruellia are presented in preparation for the } authors’ forthcoming annotated checklist of Bolivian Acanthaceae. Four new species (R. antiquorum, R. beckii, R. dolichosiphon and R. exserta) are de- scribed and illustrated, two new combinations are made and variation in R. haenkeana is discussed leading to the recognition of two varieties and two forms in the species. The little-known R. pearcei is described, illustrated and discussed in relation to the widespread R. brevifolia. Ruellia is the second largest genus of Acanthaceae in the flora of Bolivia with 27 species in the annotated checklist of Boli- vian Acanthaceae (Wasshausen and Wood, in prep.). Most species are quite well de- fined and there are relatively few novelties, perhaps a reflection of the fact that Ruellia is not primarily Andean in distribution. The following notes update our taxonomic knowledge of the genus in Bolivia prelim- inary to the publication of our checklist. 1. Infra-specific Variation Two species are especially variable but differ in the nature of their variation. Ruel- lia puri (Nees) Mart. ex Jackson varies greatly in the size of its corolla and the form of its inflorescence but detailed study of this widespread species is needed before the nature of this variation can be deter- mined. However, R. haenkeana (Nees) Wassh. is restricted to a relatively small area of Bolivia and Peru and we have been able to recognize one very distinctive va- riety and two forms below, based mainly on striking color variations. Ruellia haenkeana (Nees) Wassh. Ruellia haenkeana (Nees) Wassh., Beitr. Biol. Pflanzen 63: 423 (1988). Type: Penn ilaenikems nas dsoby pes, PIR: GZU!). var. haenkeana This is the common morphotype of the species found throughout its range in Peru and Bolivia. It is characterized by its gla- brous or puberulous leaves and crimson co- rollas. var. pilosa Wasshausen & J. R. I. Wood, var. nov. Fig. 1A—C Ruellia haenkeana var. pilosa Wasshausen & J. R. I. Wood, var. nov. Type.—BO- LIVIA: La Paz, Sud Yungas, Rio Bopi, San Bartolomé (near Calisaya), 750—900 mil —22 ulslOSO be AnekG nko, LOSO2 (holotype US!; isotypes NY!, K!). Varietas nova corolla rosea et foltis pi- losis distincta. Additional specimens examined.—BO- LIVIA: La Paz, Sud Yungas, Cerro Pelado, Puerto Linares-Yacumo, 1050 m, 14 Jul 1979, Beck 1756 (US, LPB); Larecaja, San Carlos, Mapiri, 850 m, 11 May 1927, Buch- tien 1372 (US); ibid, 5 May 1927, Buchtien 1373) (SS): ibid) 16 Apr 19275 Buchtien 264 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON My p. Y] VAY SSS SSS / icm Fig. 1. A—C, Ruellia haenkeana var. pilosa (Krukoff 10502). A, Habit; B, Bracteole and calyx lobes; C, Corolla, stamens and stigma. f. lutea Wasshausen & J. R. I. Wood, forma nov. Fig. 2A—D 1378 (US); Caranavi, 19 km from Caranavi towards Bella Vista, 1240 m, 8 Jun 1996, Wasshausen et al. 2109 (LPB, US); ibid, Wasshausen et al. 2111 (CAS, LPB, US). Leonard had annotated a number of du- plicate Krukoff 10502 collections as “*Ruel- lia nobilis Leonard, type’’. Since he never published this epithet it simply remains a nomen nudum. Ruellia haenkeana var. haenkeana forma lutea Wasshausen & J. R. I. Wood, f. nov. Type.—BOLIVIA: La Paz, Murillo, Val- le de Zongo, 1300 m, 14 Jun 1980, Beck 3681 (holotype US!; isotype LPB!). VOLUME 116, NUMBER 2 265 Nap BY 4 \ | id YA 1 aa ra weed a Z \S A A.RTANGERIN 2002 Fig. 2. A—D, Ruellia haenkeana var. haenkeana f. lutea (Beck 3681); E-G, Ruellia haenkeana var. haenkeana f. purpurea (Wasshausen & Wood 2144). A, Habit; B, Bracts, bracteoles and calyx lobes; C, Bracteoles, calyx lobes, corolla, stamens, style and stigma; D, Corolla expanded; E, Habit; K Bract, bracteoles and calyx lobes; G, Bracteoles, calyx lobes, corolla, stamens, style and stigma. 266 Forma nova corolla lutea et foliis glabris distincta. Additional specimen examined.—BO- LIVIA: La Paz, Murillo, Valle de Zongo, 1400-1600 m, 22 Apr 1982, Solomon 7535 (MO, US). f. purpurea Wasshausen & J. R. I. Wood, forma nov. Fig. 2E—G Ruellia haenkeana var. haenkeana forma purpurea Wasshausen & J. R. I. Wood, f- nov. Type.—BOLIVIA: La Paz, Carana- vi, by road ascending serrania E of Car- anavi towards Bolinda, 1200 m, 27 Jul 1998, Wood & Wasshausen 13768 (ho- lotype US!; isotypes K!, LPB). Forma nova corolla purpurea et foliis puberulis distincta. Additional specimens examined.—BO- LIVIA: La Paz, Caranavi, Bolinda above Caranavi, 1300 m, 27 Jul 1998, Wasshau- sen & Wood 2144 (US, LPB); Caranavi, 5 km from Serrania de Bella Vista on road to Caranavi, 1100 m, 8 Jun 1996, Wasshausen et al. 2104 (CAS, GOET, LPB, K, US); Caranavi, Serrania de Bella Vista, 17.6 km N of the bridge at Carrasco, 1600 m, 11 Jun 1985, Solomon 13984 (MO, US); Caranavi, N of Caranavi towards Linares, 17 km from Carrasco, 1700 m, 30 Jun 1983, Beck 9242 (LPB, US); Caranavi, 10 km N of Bolinda, 1450 m, 2 Jun 1991, Rea & Rea 10 (LPB, US). The new variety and forms are restricted to moist hill forest in the area NE of La Paz, occurring very locally amongst typical R. haenkeana. Forma lutea is apparently re- stricted to the Zongo Valley whereas the other variety and form have their centers of distribution on the mountains north and east of Caranavi. Variety pilosa is the most dis- tinct, as it has pilose leaves in addition to the distinctive pink corolla. 2. Species Clusters Ruellia geminiflora Kunth and R. brevi- folia (Pohl) Ezcurra are the best known and PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON most widespread representatives of two species clusters, which require detailed study throughout South America. The group centered on R. geminiflora comprises blue-flowered perennial herbs of open sa- vannas and cerrados. Most species have some kind of thickened rootstock, which enables the plant to survive burning, some- thing that often stimulates the plants to flower. Apart from R. geminiflora there are only two representatives of this group in Bolivia, R. bulbifera Lindau and R. brachy- siphon (Nees) Benth. & Hook. f. The three species differ from each other in indumen- tum and corolla size but are probably good species, although no final decision about their status and relationships can be made without detailed studies of the whole spe- cies complex, which is centered in Brazil. R. brevifolia is part of a cluster of spe- cies, which includes R. pearcei Rusby, R. gracilis Rusby and R. ruiziana (Nees) Lin- dau. These are bushy perennials of forest regions and are more diverse in the Andes than in the lowland regions. The four spe- cies in Bolivia are reasonably well defined although a few specimens can be difficult to assign. One of them, R. pearcei, has been universally neglected since its description and so we include a description of its main characters below, together with the citations of a range of specimens: Ruellia pearcei Rusby Fig. 3A—D Ruellia pearcei Rusby, Bull. New York Bot. Gard. 4: 429 (1907). Type.—BOLIVIA: Cochabamba, Bang 2056 (lectotype NY!, here chosen; isolectotype US!) Ruellia pearcei is related to R. brevifolia (Pohl) Ezcurra but can be distinguished from that species and all of its allies by its much longer corolla, which is 3.6—4.0 cm in length (up to 3.0 cm only in R. brevifol- ia) and differently shaped. The corolla tube has a very short cylindrical base and is then ventricose and gradually widened to the mouth whereas in R. brevifolia the basal cy- VOLUME 116, NUMBER 2 267 WEN NQTRRRS WW 7 MSS WA il) Wy i QQ K mi i Wi AZ Dae SS PES SS mv j SPM Fig. 3. A—D, Ruellia pearcei (Wasshausen et al. 2103); E-H, Ruellia brevifolia (Nee & Atha 50090). A, Habit; B, Bracteoles and calyx lobes; C, Calyx lobes, corolla and stamens: D, Anther enlarged; E, Habit, E Bracteoles and calyx lobes; G, Bracteoles, calyx lobes, corolla and stamens; H, Anther enlarged. 268 lindrical part is longer, the tube is then abruptly ventricose and even slightly con- tracted at the mouth (Fig. 3C, G). In habit, it is a perennial herb similar in general ap- pearance to R. brevifolia but the leaves are always lanceolate or oblong-lanceolate, acuminate at the tip, glabrous and usually dark violet on the undersurface. Habitat and distribution: R. pearcei 1s a local species of moist, tropical lowland forest along the eastern foothills of the Andes from the Santa Cruz area in Santa Cruz north through the Yungas of La Paz to southern Peru. Its distribution appears to be somewhat discontinuous, perhaps indicating that it has somewhat precise ecological demands. It is found between 600 and 1200 meters. Additional specimens examined.—BO- LIVIA: Beni, Ballivian, bridge Rio Qu- ibquibey, 4 km towards San Borja, 700 m, 14 Jul 1979, Beck 1691 (LPB, US); Beni, Ballivian, E side of Serrania de Pilon Lajas, 21 km from Yacumo [15°17’S, 67°40’W], 1035 m, 17 May 1989, Smith, Quintana & Garcia 13200 (K, LPB, MO); La Paz, Itur- ralde, Rio San Antonio, 46 km from I[xia- mas to Alto Madidi, 300 m, 17 Aug 1997, Kessler et al. 11204 (LPB, US); La Paz, Tamayo, Haciend Ubito, Rio Ubito Valley, 800 m, 12 Jul 1993, Kessler 3SO3 (LPB, US); ibid, Kessler 3836 (LPB, US); La Paz, Larecaja, 5 km from Consata to Sorato, 1250 m, 31 May 1995, Kessler et al. 4402 (LPB, US); La Paz, Sud Yungas, 5 km from Chamaca to La Asunta, 850 m, 6 Oct 1995, Kessler et al. 5793 (LPB, US); Sud Yungas, ca. 1.5 km W of El Sillar on road from Sapecho to Yacumo, 600 m, 28 Jul 1998, Wood & Wasshausen 137853 (K, LPB); La Paz, Caranavi, outskirts of Caranavi, along streambank, 680 m, 27 Jul 1998, Wasshau- sen & Wood 2139 (CAS, GOET, LPB, US); Caranavi, 17 km from Caranavi along road to Coroico, 500 m, 7 Jun 1996, Wasshausen et al. 2103 (CAS, GOET, K, LPB, US); Nor Yungas, ca. 25 km below Yolosa on road to Caranavi, near Chala, 900 m, 19 Feb 1999, Wood & Mondaca 14530 (K, LPB); La Paz, without exact location, Rio San PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Juan, 1100 m, 7 Apr 1902, Williams 258 (K, NY); Cochabamba, Chapare, Espiritu Santo, 160 km NE of Cochabamba, 750 m, Jan 1909, Buchtien 2239 (US); ibid, Buch- tien 2309 (US); Cochabamba, Tiraque, on escarpment above El Palmar, 1200 m, 6 Jun 1998, Wood 13673 (K, LPB); Chaparé, Rio Espirito Santo, at beginning of ascent to Cochabamba, 1000 m, 19 Jun 1994, Wood 5537 (K, LPB, US); Rio Espirito Santo, ca. 20—25 km E of Villa Tunari on road to Co- chabamba, 700 m, 27 Jul 1995, Wood 10081 (K, LPB); Cochabamba area without precise department, Bang 1223 (K, NY, US); Santa Cruz, Ichilo, ca. 2 km below Campamiento Mataratu, Amboro Park, 300 m, 30 May 1997, Wood 12212 (K, US); Ichilo, near San Rafael de Amboro, S of Buenavista, 500 m, 21 May 1995, Wood 9843 (K, LPB, US); Ichilo, Campamiento Mucunucu. Amboro Park, 500 m, 28 May 1996 Wood et al. 11150 (LPB): Amboro, SW of Buena Vista, near Macunucu Camp, 450-500 m, 28 May 1996, Brummitt et al. 19320 (K, LPB, US); ibid, Wasshausen et al. 2060 (CAS, GOET, K, LPB, US); ca. 2 km below Campamiento Mataratu, Amboro Park, 300 m, 30 May 1997, Wood 1212 (K, LPB); Santa Cruz, Valle Grande, 10 km from Loma Larga to Masicuri, 1450 m, 26 May 1996, Kessler et al. 6104 (LPB, US).—PERU: Puno, Sandia, between San Juan de Oro and San Ignacio, 1200 m, 7 Jun 1982, Wasshausen & Salas 1205 (K, MO, NY, US, USM); ibid, 1100 m, Wass- hausen & Salas 1211 (G, K, MO, NY, US, USM); without precise locality, 1863, Pearce 276 (K), 596 (K); Huanuco, Pachi- tea, Honoria, Iparia National Forest, along Rio Pachitea, 11 km above Tournavista Vil- lage, 3-400 m 19 Jul 1967, Schunke 2128 (K, US); Cuzco, Quispicanchis, Inambari bridge, 530 m, 4 Mar 1965, Vargas 16185 (CUZ, US); ibid, 580, m, 8 Dec 1967, Var- gas 18455 (CUZ, US). 3. New Combinations The following is a widespread species in the SW Amazon region although hitherto unnamed in most herbaria: VOLUME 116, NUMBER 2 Ruellia nitida (Nees) Wassh. & J. R. I. Wood, comb. nov. Ruellia nitida (Nees) Wassh. & J. R. I. Wood, comb. nov. Arrhostoxylum nitidum Nees, in Mart., Fl. Bras. 9(7): 59 (1847). Type.—BRAZIL: Amazonas: Borba, Rie- del s.n. (holotype LE!; isotype GZU!). This species appears to be common in moist lowland forest and we have seen many collections from northern Bolivia in the departments of Pando, Beni and Santa Cruz, and in western Brazil in the territory of Rond6nia [e.g. Prance et al. 5603 (K, INNG WS), D67G CXS INANE US) ene! iin Hite states of Amazonia and Mato Grosso. Ruellia multisetosa Rusby is not, in fact, a Ruellia and should be transferred to the genus Suessenguthia as follows: Suessenguthia multisetosa (Rusby) Wassh. & J. R. Il. Wood, comb. nov. Suessenguthia multisetosa (Rusby) Wassh. & J. R. 1. Wood, comb. nov. Ruellia mul- tisetosa Rusby, Mem. New York Bot. Gard. 7: 362 (1927). Type.-—BOLIVIA: La Paz: Huachi (alto Beni), 13 Aug 1921, White 547 (holotype NY!). 4. New Species The following four species are new, with three being from Bolivia and one being from Brazil. The Brazilian species is in- cluded because of its obvious similarity to the Bolivian novelty, R. beckii. Ruellia antiquorum Wassh. & J. R. I. Wood, sp. nov. Fig. 4A—C Species nova speciosa R. patula Jacq. re- vocans sed corolla longipedunculata statim recedens et inter speciorum austro-ameri- canarum corolla coerulea, longipeduncu- lata, solitare, axilliare valde distincta. Low perennial herb from a woody root- stock; stems to 25 cm long, trailing, terete, pilose with spreading multicellular tri- 269 chomes, becoming slightly woody when old, occasionally sending down adventi- tious roots from the leaf nodes; leaves shortly petiolate, the petioles 1-4 mm long, the blades ovate or elliptic, acute at apex, abruptly narrowed to the base, 1.4—3.0 cm long, 0.5—2.2 cm wide, sparsely to densely pilose with white, multicellular trichomes which are sometimes bulbous at the base, the margin entire; inflorescence of solitary pedunculate flowers borne in the axils of the upper leaves, in each leaf pair, one axil fertile and one sterile; peduncles slender, glabrous, 1.5—5 cm long; bracteoles paired, borne 1—2 mm below flower, linear-oblong, 9-10 mm long, 0.75—-1.00 mm wide, sparsely ciliate; calyx 1.4—1.7 mm long, di- vided to ca. 1 mm above base, the lobes narrowly oblong-elliptic, acute, glabrous or with a few short, marginal trichomes; co- rolla funnel-shaped, 55—70 mm long, ca. 2 mm wide at base, cylindrical for 25—30 mm, then gradually widened to 18—28 mm at the mouth, the tube dirty whitish-brown, pubescent on the exterior, the lobes blue, glabrous, spreading, broadly ovate to sub- orbicular, rounded, 10—16 mm long, 15—20 mm wide; stamens 4, didynamous, inserted ca. 30 mm above base of corolla, immedi- ately above narrow cylindrical part of the tube; filaments glabrous, 5—6 mm long, the outer one in each pair ca. | mm longer than the other; anthers included in corolla tube, ca. 3 mm long, glabrous; style 40 mm long, pubescent, persistent; ovary glabrous; cap- sule 10-15 mm long, 4 mm wide, narrowly obovoid, glabrous, gland-dotted; seeds len- ticular, ca. 3 mm long and wide, flattened, glabrous, the margin white. Type.—BOLIVIA: Santa Cruz: Florida, Samaipata, on hill by radio transmitter, 1700 m, 17 Feb 1995, J. R. I. Wood 9419 (holotype K!; isotypes LPB, US!). Additional specimens examined.—BO- LIVIA: Santa Cruz: Florida, El Fuerte, Sa- maipata, 1800 m, 15 May 1994, Wood 8376 (K, LPB); ibid, 11 Sep 1994, Wood 8630 (K, LPB, US); ibid, 2 May 1994, Rojas & Vargas 1994 (BOL); ibid, 1550-1650 m, 270 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SSS Sy \e es ot i OS LES Fig. 4. A—C, Ruellia antiquorum, A, B, (Wood 9419) C, (Beck 6769); D, E, Ruellia dolichosiphon (Balcazar, Montero & Alvarez 2422). A, Habit; B, Bracteoles, calyx lobes and corolla; C, Corolla expanded; D, Habit; E, Bracteoles, calyx lobes and corolla. VOLUME 116, NUMBER 2 23 Mar 1981, Beck 6769 (LPB, US); Rio Pirai Gorge between Las Cuevas and Ber- mejo, 1300 m, Wood 10561 (K, LPB). Ruellia beckii Wassh. & J. R. I. Wood, sp. nov. Fig. 5A—D Inter species generis Ruellia habitu scan- denti, ramis lignosis, corolla luteo-viridi et antheris valde exsertis bene distincta. Woody liana reaching 10 m in height; stems woody, weakly quadrangular, sparse- ly scurfy-pubescent when young, glabres- cent; leaves shortly petiolate, the petioles 0.8—1.0 cm long, the blades 6—14 cm long, 3-6 cm wide, broadly oblong-elliptic, acute, tapering at both ends, slightly asym- metric at the base, the margin obscurely re- pand, glabrous except for a few trichomes on the veins, cystoliths common above; in- florescence a small, terminal subumbelli- form cyme, the branches arising in opposite pairs, the branches scurfy-pubescent; brac- teoles ovate-triangular, 4-8 mm long, 2-3 mm wide, glabrescent or with a few scurfy marginal trichomes; calyx 5-lobed to just above the base, lobes 12—14 mm long, 5—6 mm wide, oblanceolate to obovate, imbri- cate, glabrous or with a few scattered mar- ginal trichomes; corolla ca. 38 mm long, tube ca. 7 mm wide at base, bulbous, then gradually narrowed to 4 mm at 13 mm above base, then abruptly widened to ca. 14 mm, shortly glandular pubescent on the ex- terior, lobes broadly oblong, 7 mm long, 6.5 mm wide, rounded, yellow-green; stamens 4, didynamous, inserted ca. 10 mm above base of corolla tube, the outer of each pair inserted ca. 1 mm below the inner; fila- ments 22 mm long, glabrous; anthers ex- serted 10-15 mm, 6 mm long, glabrous, sagittate at base; ovary glabrous except for the puberulent tip; style 45 mm long, pu- bescent, stigma bifurcate; capsule not known. Type.—BOLIVIA: La Paz: Caranavi, 20 km along road from Caranavi to Carrasco, Dial 1200 m, 1 Jul 1983, S. G. Beck 9298 (ho- lotype US!; isotype LPB). Only known from the type locality where it was growing in hill forest. Ruellia exserta Wassh. & J. R. I. Wood, sp. nov. Fig. SE-H A R. beckii inflorescentia spicata unila- terale et lobis calycis angustioribus, distan- tibus, corolla parviori diagnoscenda. Woody liana to 10 m, “climbing over trees’’; stems woody, rounded, glabrous be- low, scurfy above; leaves petiolate, the pet- idles 0.6—2.0 cm long, the blades 6—23 cm long, 2.0—10.5 cm wide, lanceolate, ovate or oblong-elliptic, acuminate at apex, taper- ing to a sometimes asymmetric base, the margin obscurely crenate, glabrous except for a few scurfy trichomes on the veins, cystoliths common above; inflorescence of small, apparently one-sided spike-like cymes terminal on branches arising alter- nately from each leaf axil; pedicels 3—5 mm long with numerous cystoliths; bracteoles ovate-triangular, 2—6 mm long, 1.5—2.5 mm wide, pubescent on the margins; calyx 5- lobed to just above the base, lobes 8—9 mm long, 2 mm wide, oblong, obtuse, distant, glabrous above, puberulent or sometimes scurfy glandular-pubescent below; corolla ca. 28 mm long, tube at base 2.0—4.5 mm wide, cylindrical for ca. 10 mm, not or hardly narrowed, then gradually widened to 12 mm at mouth, the exterior pubescent and with scattered subsessile glands, the lobes ovate, 6 mm long and 5 mm wide, rounded, yellow-green or green; stamens 4, didyna- mous, inserted ca. 17 mm above base of corolla tube; filaments ca. 22 mm long, gla- brous; anthers 5 mm long, glabrous, sagit- tate at base, exserted 10—13 mm; ovary pu- bescent; style 32—35 mm long, pubescent with a few glands, stigma bifurcate; capsule oblanceolate, apiculate, ca. 20 mm long, 8 mm wide, shortly and densely pubescent, 4- seeded; seeds suborbicular, 3.5 mm long, 3 DP PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SD i ] WS, ‘< < 1 Fig. 5. A-—D, Ruellia beckii (Beck 9298); E-H, Ruellia exserta (de Santos et al. 1798). A, Habit; B, Calyx lobes; C, Calyx lobes, disc and ovary; D, Corolla, stamens, style and stigma; E, Habit; KF Calyx; G, Calyx lobes, disc and ovary; H, Corolla, stamens, style and stigma. VOLUME 116, NUMBER 2 mm wide with a pale margin and scurfy surface. Type.—BRAZIL: Mato Grosso: N_ of Xavantina on Xavantina-Sao Felix road, 40 km N of base camp at [12°54’S, 51°52’W], 14 Jun 1968, R. R. de Santos, R. Souza & A. Ferreira 1798 (holotype K!; isotype US!). Additional specimens examined.—BRA- ZIL: Para: ca. 6 km on road NW of Camp 3-Alfa towards Camp 4-Alfa [5°47’S, 50°34’W], 250 m, 9 Jun 1982, Sperling et al. 6013 (K, NY); Maraba, 24 May 1982, Secco et al. 300 (MB, NY); Mato Grosso: Cuiba-Porto Velho, 15 km from frontier with Rond6énia, Patronal District, Vila Bela de Santissima Trinidade [12°13’S, 60°59’W], 9 Jun 1984, Cid et al. 4387 (K, NY, US); 16 kine NGoin [12 515S; 51-52) Wi, ca, 270 kim N of Xavantina, 20 May 1968, de Santos et al. 1483 (K, NY, US); Territory RondoOnia: Mun. de Ariquemes, Mineracgao Mibrasa, Sector Alto Candeias, km 128, [10°35’S, 63°35'W], SE of Ariquemes, 14 May 1982, Texeira 440 (INPA, NY); Mun. de Costa Marques, Chapada dos Parecis, Distrito de Alta Floresta, estrada P-56, km 17 [11°12’S, 62°63'W], 16 Jun 1984, Cid 458] (INPA, NY). Habitat and distribution: Apparently an uncommon Brazilian endemic plant of Para, Mato Grosso and Territory Rond6énia, growing on roadsides, in disturbed primary forest and in swampy gallery forest around 250 m. Ruellia beckii and R. exserta are closely related species, being similar in habit, leaf shape and flower color. After some hesita- tion, we felt confident in treating them as different species, since the calyx is very dif- ferent in the two species and the corolla is somewhat so. Possibly more important is the inflorescence, + umbellate in R. beckii but in apparently one-sided spikes in R. exserta. Although we have only seen one collection of R. beckii, there are various collections of R. exserta, all of which main- tain the distinctive inflorescence, calyx and corolla character even though collected 273 from different states in Brazil. Additionally, it is highly improbable that an Andean spe- cies such as R. beckii should be widespread in the Brazilian lowlands without any in- termediate populations. Ruellia dolichosiphon Wassh. & J. R. I. Wood, sp. nov. Fig. 4D-E Ob inflorescencia terminale et calyce quam bracteis bracteolisque longiore Ruel- lia glischrocalyx Lindau tingit sed ramis albo-pilosis, foliis ovatis, corolla longituba, lobis roseis distincta. Stout herb to 50 cm; stem rounded, densely white-pilose; petioles 0.4—1.4 cm long, white pilose; blades 3.5—8.5 cm long, 2.0—3.5 cm wide, ovate, acute, base abrupt- ly narrowed and then attenuate onto the pet- iole, sparsely white-pilose, the trichomes mostly on the upper surface, margins and veins, cystoliths very small, obscure, mar- gin slightly undulate; inflorescence of short terminal racemes; pedicels 2—4 mm long, glandular-pilose; bracts at base of pedicel, 6—8 mm long, 1.0—1.5 mm wide, subpe- tiolate, linear-oblanceolate, rounded, pilose with large, brownish trichomes; bracteoles 5—6 mm long, 1.0—1.5 mm wide, linear-el- liptic or linear-oblanceolate, weakly acute, pilose with large brownish trichomes; calyx much longer than the bracts, subequally 5- lobed to just above the base, lobes 20—22 mm long, 4 mm wide, oblong-lanceolate, obtuse, one slightly larger than the others, ciliate and with a thick line of brownish tri- chomes along the midrib, a few trichomes noticeably longer than the others; corolla funnel-shaped, the tube pubescent below, ca. 47 mm long, 4 mm wide at base, then narrowed to 1.5 mm after ca. 4 mm, before widening gradually to 6 mm at mouth, pale, probably dirty white, the lobes ca. 8 mm long, ca. 15 mm wide, broadly ovate, acute, pink, subglabrous outside; anthers not seen; style sparsely pilose; ovary minutely and densely pilose; capsule not seen. Type.—BOLIVIA: Pando: Manuripi, 274 Communidad Lago, 17 km NE of Pursima on road to Chivé [05°0012’S, 57°3581'’W] 220rim 29 un ZOO ei Balcazarme aC: Montero & J. Alvarez 2422 (holotype USZ!). Habitat and distribution: Only known from the type collection in Amazonian rain- forest in northern Bolivia. Acknowledgments The authors would especially like to thank Alice Tangerini who skillfully pre- pared the line drawings. Also, we express our appreciation to the National Herbarium in La Paz (LPB), the Museo Noel Kempf Mercado (UCZ) in Santa Cruz, the Natural PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON History Museum in London (BM) and the New York Botanic Garden (NY) for the loan of specimens used in this study. John Wood would also like to thank all the in- stitutions mentioned as well as the Royal Botanic Gardens, Kew (K) and the Smith- sonian Institution (US) for facilities provid- ed during his visits and to the Smithsonian Institution Office of Fellowships and Grants for a short-term visitor travel grant, under which a substantial amount of this research was carried out. Literature Cited Ezcurra, C. 1993. Systematics of Ruellia (Acantha- ceae) from South America.—Annals of the Mis- souri Botanical Garden 80:747—845. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):275—292. 2003. Phylogeography and systematic notes on two species of gracile mouse opossums, genus Gracilinanus (Marsupialia: Didelphidae) from Brazil Leonora Pires Costa, Yuri L. R. Leite, and James L. Patton Museum of Vertebrate Zoology, University of California at Berkeley, 3101 Valley Life Sciences Building, Berkeley, California 94720-3160, U.S.A. Abstract.—Although they were described more than a century ago, Graci- linanus agilis and Gracilinanus microtarsus are still mistaken for each other and their status as valid species has been challenged. Morphological studies are rare and accounts of molecular characters are even scarcer. In this paper, we present the first phylogeographic analysis for these two species based on mitochondrial cytochrome b sequences and provide a morphological analysis and extended diagnosis for both species. We show that G. agilis and G. mi- crotarsus are valid species, geographically bounded and distinguishable by morphological and molecular characters. Gracilinanus agilis is geographically widespread and genetically more homogenous, with low levels of divergence between three clades in central, northeastern and eastern Brazil. Gracilinanus microtarsus Occurs in southeastern and southern Brazil, being comprised of two clades that show a considerable level of sequence divergence. Although at this time we regard these two clades as geographic units of G. microtarsus, it is possible that more samples will show that there is greater diversity in this group than the current taxonomy recognizes. The genus Gracilinanus (Gardner and Creighton, 1989) comprises delicately built opossums, generally smaller than individu- als of Marmosops and Marmosa with which they are often confused. The dorsal color- ation varies considerably, from bright red- dish-brown to pale grayish-brown; the ven- tral pelage is often cream or pale orange with gray-based hairs, but sometimes white or pure cream as in G. emiliae. The tail is moderately long to very long, with non-pet- iolate central hairs in each caudal-scale trip- let. Among diagnostic cranial characters, the more trenchant are the highly fenestrat- ed palate, with maxillary vacuities often present, and bullae with an anteromedian strut forming a secondary foramen ovale (R. S. Voss, in litt.). Other characters and detailed descriptions of the genus can be found in Tate (1933) and Creighton (1984) under the microtarsus section of the for- merly inclusive genus Marmosa, and in Gardner & Creighton (1989) and Hershkov- itz (1992). Costa (in press) discusses the phylogenetic affinities of Gracilinanus among the didelphids, based on cytochrome b (cyt b) sequences and morphological characters. The genus ranges from the Guiana re- gion, through Venezuela and Colombia, bordering the western limit of the Amazon basin with scattered localities in Peru, Bo- livia and Paraguay, to the mouth of the Pa- rana river in Argentina, then northeast along the coast and interior tablelands of Brazil, reaching the southeastern border of Amazonia. It is apparently absent from all, or at least the majority of the lowland Am- azon basin in Brazil as specimens recorded from this vast region were either misiden- tittedia(Sces Rattonmet alasZ0005 Voss et ale 2001) or are of questionable occurrence (Patton and Costa 2003). Tate (1933) provided the most compre- 276 hensive analysis of the taxa currently in- cluded in Gracilinanus under the “micro- tarsus section” of his monograph. Later, Gardner & Creighton (1989) established the genus Gracilinanus to include six species: G. aceramarcae (Tate, 1931), G. agilis (Burmeister, 1854), G. dryas (Thomas, 1898), G. emiliae (Thomas, 1909), G. mar- ica (Thomas, 1898) and G. microtarsus (Wagner, 1842). Yet, the number of species within the genus remains debatable since Hershkovitz (1992), in his recent revision, recognized the same six species listed above but described three new ones from the Andean slopes of Colombia (G. longi- caudus and G. perijae) and Peru (G. kali- nowskii), and recorded what he considered an undescribed additional species from Ec- uador. Voss et al. (2001) subsequently de- fined a new genus, Hyladelphys, for G. kal- inowiskil. In addition, these authors regard G. longicaudus as a junior synonym of G. emiliae, and they pointed out that Hersh- kovitz’s undescribed “‘Gracilinanus’’ from Ecuador is a Marmosa (sensu stricto). In the present report we address two spe- cies: G. agilis and G. microtarsus. The range of G. microtarsus includes the mesic habitats of the Atlantic Forest in southeast- ern Brazil south to Rio Grande do Sul, while G. agilis is more widespread in Bra- zil, occurring primarily in dry and gallery forests of the interior plateau, isolated areas in the Northeast, and through the wet and dry forests of northeastern Argentina, Par- aguay, and Bolivia (Emmons & Feer 1997, Eisenberg & Redford 1999). Although G. agilis and G. microtarsus were described nearly 150 years ago, the lack of adequate comparisons between them has raised doubts about their validity as species. Al- though both Hershkovitz (1992) and Gard- ner (1993) recognize G. agilis and G. mi- crotarsus, the first author suggests that G. microtarsus could be “‘no more than a sub- species of G. agilis”, while the second au- thor states that “‘the forms agilis and mi- crotarsus may prove to be conspecific’. Here we provide a phylogeographic analy- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON sis of geographic samples of both G. mi- crotarsus and G. agilis, and present mor- phological comparisons and systematic comments. We also include a third species (G. aceramarcae, a rare Opossum known only from the type locality in Bolivia [Tate 1931] and two localities in southern Peru [Eisenberg & Redford 1999; L. H. Em- mons, in litt.]) in the molecular analysis to provide a more representative sampling of the taxonomic diversity within the genus. Methods Molecular analyses.—Our samples con- sist of a single specimen from Vilcabamba, Peru, tentatively identified as G. aceramar- cae, and 34 individuals of G. agilis and G. microtarsus from 24 localities in Brazil (Fig. 1), including topotypes of G. micro- tarsus (from Ipanema, Sao Paulo; see Ap- pendix). We also have samples from a lo- cality in the same region where the type specimen of G. agilis was collected (Lagoa Santa, Minas Gerais). We extracted DNA from frozen or etha- nol-preserved liver tissue using the Chelex® method (Walsh et al. 1991) and amplified the cytochrome b (cyt b) gene by the Poly- merase Chain Reaction (PCR), (Saiki et al. 1988) using primer pairs MVZO5 in com- bination with MVZ04 and/or MVZ 16. The double-strand PCR products were cleaned using the DNeasy™ Tissue Kit (QIAGEN, Inc.) and submitted to cycle sequencing re- actions that utilized the dRhodamine Ter- minator Cycle Sequencing Kit and Protocol (PE Biosystems, Inc.), and primers MVZO5, MVZ04 and MVZ65. All sequencing was done in an ABI Prism 377 automated se- quencer. Sequences were aligned by eye us- ing Sequence Navigator (Version 1.0.1, Ap- plied Biosystems, Inc.). The data set varied from 518 bp to 801 bp of the cyt D se- quence, each sequence beginning with the start codon of the gene. A Nexus file of all sequences is available from the authors upon request. Relationships among different haplotypes Od VOLUME 116, NUMBER 2 (SU) [NS Op spuvrH Ory puv (Os) eulieyey) vyurg “(Yd) vuried “(fy) OLlouvf op OTY “(dS) O[Ned ORS “(SIA) [NS OP Osso1n owP ‘(LIA) ossorg owW (OD) SBIOD “(DOW) Stetoy seul “(CW a) PIyeg “(Id) Meld (AO) Plea :ose sojeys uvI[Izeig “¢ “SI UL pafoqry] SoNteRoo] sues dy) 07 puodsasiod pue xipuaddy oy} ut pooua “19J910IS 9IV SONITPLOO] posoquuNyy “(€ “SI 99S) satoads ULYIIM sope[d dIVOIpPUL Soul] poysep pur salIepuNog satoads yWuNIJap sour] pljos ‘asuvs uleyunoU eillonbonueyy vp elIag oy) Aq payesedas ore snsipjosIIU “FD JO Sape{d OMI OYJ, (S715 “DH Jo Ayypesoy odAy oy sf Aypeso[) OW ‘eiueg vosey] sou snsuvjosonu “.) syoe]UoOd S1]18p snupulpovsy) soiseds peaidsapiM oy], “Wodas sty} UL pozrusooes sopr[o orydeisoes pur exe) oy) pue sojdures ino Jo uoNNQLNsIp oy) Bumoys dep *] ‘SI epejo weyseoujnos SHSADJOAIIUL °£) epeRjo sieiey seuly OF eianbnue/ BD CLOG Semin epRjoO jeyu00 @ 278 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GSL M1-M4 Fig. 2. Thirteen cranial measurements recorded from G. agilis and G. microtarsus, as follows: GSL— Greatest skull length: from the most anterior margin of the rostrum to the posterior margin of the occiput. ZB— Zygomatic breadth: greatest breadth across the zygomatic arches. BB (measurement 7 of Tate 1933:236)— Braincase breadth: breadth taken above the squamosal root of the zygomatic arch. I[OC—Least interorbital constriction: minimal breadth across the roof of the skull above the orbits. RL—Rostral length: from the anterior margin of the orbit to the midline tip of the nasals. NL—Nasal length: midline distance from anterior tip to posterior margins of nasals. RW—Rostral width: width of the rostrum at the level of the canines. M1—M4— Molar toothrow length: taken on the labial margin of the toothrow from M1 to M4. LPB (measurement 1 of Tate 1933:236)—Least pterygoid breadth: least breadth across pterygoid bones. PB (measurement 3 of Tate 1933:236)—Petrosal breadth: breadth across tympanic process of pars petrosa. ABB (measurement 2 of Tate 1933:236)—Alisphenoid bulla breadth: breadth across tympanic processes of alisphenoid. PL—Palatal length: midline distance from the posterior margins of the first upper incisor to the posterior margin of the hard palate. CD—Cranial depth: vertical distance between ventral margins of bullae and top of cranium. were examined by maximum parsimony us- tion, with 10 replicates and random se- ing PAUP*, version 4.0b8 (Swofford quence addition of taxa. The support for in- 1999). Trees were constructed using the ternal branches was evaluated by decay in- heuristic search option via stepwise addi- dices (Bremer 1988) and bootstrap analyses VOLUME 16, NUMBER 2 95/11 14.94% G. microtarsus G. agilis G. aceramarcae Lagoa Santa (1) Peti (2) Belo Horizonte (3) 90/2 33% Cruzeiro (4) 64/0 Jurumirim (5) 6.36% Rio Doce (6) 94/7 Caraca (7) go Ipanema (8) 100/10 Intervales (9) tok llha Grande (10) Boracéia (11) Coronel Murta (12) Serra do Cipé (13) 90/3 Il Rio Preto (14) 0.37% Bocaiuva (15) Andarai (16) Nova Ponte (17) 100/19 Passo do Lontra (18) 3.93% Poconé (19) Caldas Novas (20) Barra do Gar¢as (21) 78/2 2.28%| ‘ Serra da Mesa (22) Crato (23) Urucui-Una (24) Vilcabamba, Peru (25) —— 10 changes C Minas Gerais clade mi Southeastern clade @Central clade > ATS) Fig. 3. Bootstrap consensus tree based on maximum parsimony analysis of cyt b gene haplotypes (initial 801bp). Gracilinanus agilis and G. microtarsus are reciprocally monophyletic and highly divergent species. The two clades within G. microtarsus are also highly divergent. Sequences of Marmosops incanus, Thylamys sp., and Marmosa murina were used as outgroups to root the tree. Bold numbers at internal nodes are bootstrap values/decay indices; percentages are average Kimura two-parameter (K2p) distances. Voucher catalog numbers and localities for each haplotype are given in the gazetteer and correspond to the same localities on the map in Fig. 1. Consistency index = 0.593; homoplasy index = 0.407; retention index = 0.761. 280 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Kimura 2-parameter distances among cyt-b haplotypes of Gracilinanus, with Marmosops incanus also included. LC49 MNRJ31445 MNRJ31447 MCNM299 LCl RM26 M. incanus G. microtarsus G. microtarsus G. microtarsus G. microtarsus G microtarsus LC49 M. incanus MNRJ31445 G. microtarsus 21.00% MNRJ31447 G. microtarsus 21.00% 0.00% MCNM299 G. microtarsus 19.01% 0.00% 0.00% LC1 G. microtarsus 17.91% 4.95% 4.95% 4.94% RM26 G. microtarsus 17.01% 1.98% 1.98% 2.00% 3.95% MCNM394 G. microtarsus 18.33% 2.82% 2.82% 2.80% CPA) 3.00% RM4 G. microtarsus 18.50% 6.32% 6.32% 6.37% 6.71% 5.76% LPC805 G. microtarsus 20.77% 10.01% 10.01% 9.44% 9.60% 8.05% MVZ182056 G. microtarsus 19.77% 10.26% 10.26% 10.21% 10.36% 9.50% LP40 G. microtarsus 19.23% 8.60% 8.60% 8.59% 9.60% 8.03% MAM427 G. microtarsus 20.67% 10.86% 10.86% 9.81% 10.59% 9.27% LC189 G. agilis 19.02% 14.75% 14.75% 14.00% 16.56% 14.15% MNRJ31396 G. agilis 18.56% 14.36% 14.36% 14.31% 15.44% 13.72% YL64 G. agilis 19.03% 14.75% 14.75% 14.00% 16.56% 14.14% LPC241 G. agilis 19.56% 14.76% 14.76% 14.02% 16.34% 13.93% AP22 G. agilis 18.85% 14.59% 14.59% 13.80% 16.33% 13.93% LPC304 G. agilis 20.53% 15.22% 15.22% 14.53% 17.16% 16.29% LPC599 G. agilis 19.87% IS .23% i523 % 14.54% 16.90% 16.03% LPCS581 G. agilis 19.40% 15.05% 15.05% 14.31% 16.72% 15.43% UHEC0O4722 G. agilis 19.23% 14.73% 14.73% 14.15% 16.53% 15.25% LPC476 G. agilis 18.66% i270 1327296 13.62% 14.99% 14.28% UHESM1759 G. agilis 18.69% 14.58% 14.58% 13.75% 15.99% 14.79% LPC250 G. agilis 19.72% 14.08% 14.08% 12.94% 14.98% 13.65% UUP1292 G. agilis 19.49% 14.51% 14.51% 13.20% 15.10% 13.65% LHE1342 G. aceramarcae 19.89% 15.71% 15.71% 14.28% 15.66% 13.85% (Felsenstein 1985), with 100 bootstrap rep- licates each consisting of a full heuristic search as described above. On the assump- tion that Gracilinanus is monophyletic (Gardner & Creighton 1989; Voss et al. 2001), all trees were rooted by including sequences of Marmosops incanus, Marmo- sa murina, and Thylamys sp. as outgroups. Sequence divergence was calculated using the Kimura two-parameter algorithm (K1i- mura 1980) as implemented in PAUP*. Morphological analyses.—Statistical analyses were performed using the com- puter package StatView (Version 5.0, SAS Institute Inc.). For analysis of external mor- phological data, we used 71 specimens of G. agilis from 10 localities, and 25 speci- mens of G. microtarsus from 9 localities in the states of Sao Paulo, Rio de Janeiro, and Minas Gerais (see Appendix and Fig. 1). External measurements (total length and lengths of tail, ear, and hindfoot including claws) and weight (mass) were taken from specimen labels. We subtracted tail length from total length to obtain the length of the head and body. For analyses of cranial mea- surement data, we examined 60 specimens of G. agilis and 16 of G. microtarsus. Each individual was classified as young or adult, adults being defined as those individuals possessing a complete set of teeth Ginclud- ing the permanent third premolar, and four molars). We recorded 13 cranial measure- ments (Fig. 2) taken with digital calipers from each individual. Results and Discussion Phylogenetic analysis.—The maximum parsimony analysis resulted in 80 most-par- simonious trees of 672 steps each, with 215 of the 801 characters parsimony-informa- VOLUME 116, NUMBER 2 281 Table 1.—Extended. MCNM394 RM4 LPC805 MVZ182056 LP40 MAM427 G. microtarsus G. microtarsus G. microtarsus G. microtarsus G. microtarsus G. microtarsus LC49 M. incanus MNRJ31445 G. microtarsus MNRJ31447 G. microtarsus MCNM299 G. microtarsus LC1 G. microtarsus RM26 G. microtarsus MCNM394 G. microtarsus RM4 G. microtarsus 6.71% LPC805 G. microtarsus 9.16% 9.52% MVZ182056 G. microtarsus 9.79% 10.46% 1.34% LP40 G. microtarsus 8.24% 8.56% 3.88% 4.39% MAM427 G. microtarsus 9.64% 9.52% 4.60% Sle 1.43% LC189 G. agilis 14.75% 14.10% 15.98% 14.98% 14.31% 14.80% MNRJ31396 G. agilis 13.88% 14.38% 14.57% 14.20% 13.50% 14.79% YL64 G. agilis 14.74% 14.09% 15.97% 14.97% 14.31% 14.80% LPC241 G. agilis 14.69% 13.88% 15.98% 14.97% 14.33% 15.14% AP22 G. agilis 14.54% 13.87% 15.80% 14.73% 14.10% 14.64% LPC304 G. agilis 15.98% 16.04% 17.85% 15.78% 14.85% 16.82% LPC599 G. agilis SG 15.80% 17.43% 15.76% 14.41% 16.41% LPC581 G. agilis ISIN 15.52% 16.60% 16.50% 14.58% 15.91% UHECO4722 G. agilis 15.56% [5233596 16.10% 15.82% 14.00% 15.10% LPC476 G. agilis 14.04% 14.94% 15.76% 15.83% 14.62% 16.04% UHESM1759 G. agilis 14.65% 14.69% 16.45% 15.84% 14.44% 15.76% LPC250 G. agilis 14.01% LS AMUG%o 16.27% 16.12% IS JiYe 15.92% UUP1292 G. agilis 14.43% 14-75% 16.36% 15.04% 15.09% 16.52% LHE1342 G. aceramarcae 14.88% 11.76% 15.24% 16.10% 13.86% 14.68% tive. We present only the bootstrap consen- sus tree (Fig. 3) since it identifies the well- supported nodes and its topology does not differ significantly from the strict consensus of the most-parsimonious trees. A matrix of Kimura 2-parameter (K2p) distances among the 25 unique haplotypes of Gracilinanus, plus Marmosops incanus, is given in Ta- ble I. The bootstrap tree (Fig. 3) shows a basal, unresolved trichotomy between the G. ac- eramarcae sample, a clade from central Brazil (Figs. 1 and 3; localities 1-11), and a third clade formed by two groups from eastern Brazil (Figs. 1 and 3; localities 12— 24). These three clades are highly divergent from each other, differing by an average K2p distance of about 15%. The two clades from eastern Brazil [Minas Gerais (locali- ties 1-7; Fig. 1) and Southeastern clades (localities 8—11)] are also quite divergent (average K2p value of 9.55%), remarkably so considering the small geographic dis- tances separating them. On the other hand, there is a surprisingly amount of homoge- neity among the samples of the Central clade since its representatives come from widely separated localities but are only 3.93% divergent on average. Yet, the Cen- tral clade is divided into sub-clades: a more divergent one encompassing the northern part of Minas Gerais and central Bahia (lo- calities 12—16; Fig. 1), another in the North- east (localities 23—24), and a wide-spread third in central Brazil in the states of Mato Grosso, Mato Grosso do Sul, Goids and western Minas Gerais (localities 17—22). The bootstrap tree and the strict consen- sus of the parsimony trees differ only in the placement of two samples. First, G. acera- marcae is sister to G. agilis and G. micro- tarsus in the strict consensus tree, but with Table 1.—Extended. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON LC189 MNRJ3 1396 YL64 LPC241 AP22 LPC304 G. agilis G. agilis G. agilis G. agilis G. agilis G. agilis LC49 M. incanus MNRJ31445 G. microtarsus MNRJ31447 G. microtarsus MCNM299 G. microtarsus LC1 G. microtarsus RM26 G. microtarsus MCNM394 G. microtarsus RM4 G. microtarsus LPC805 G. microtarsus MVZ182056 G. microtarsus LP40 G. microtarsus MAMA427 G. microtarsus LC189 G. agilis MNRJ31396 G. agilis 0.60% YL64 G. agilis 0.00% 0.60% LPC241 G. agilis 0.76% 1.00% 0.76% AP22 G. agilis 0.13% 0.84% 0.13% 0.88% LPC304 G. agilis 4.49% ZING 4.49% 4.83% 4.66% LPC599 G. agilis 4.01% 3.07% 4.01% 4.34% 4.18% 0.77% LPCS581 G. agilis 3.47% 3.19% 3.48% 3.74% 3.61% 1.22% UHECO4722 G. agilis 3.34% 3.28% 3.34% 3.88% 3.47% 0.93% LPC476 G. agilis 3.33% 3.13% 283% 3.30% 3.57% 2.30% UHESM1759 G. agilis 3.48% 3.06% 3.48% 315% 3.61% 2.01% LPC250 G. agilis 4.82% 4.94% 4.82% 5.38% 4.96% 4.95% UUP1292 G. agilis 4.30% 3.99% 4.31% 4.85% 4.44% 3.79% LHE1342 G. aceramarcae 13.26% 14.84% 13.26% 13.59% 13.11% 14.67% low support (bootstrap < 50%), while the three form an unresolved polytomy in the bootstrap tree (Fig. 3). Second, in the strict consensus tree, the Caraga sample (locality 7; Fig. 1) falls in a trichotomy with the Southeastern and Minas Gerais clades, while in the bootstrap tree it joins the Minas Gerais clade, but again with low support (bootstrap = 64%; decay index = Q) and high sequence divergence (6.4% on aver- age; Fig. 3). The ambiguous position of the Caraga sample and the degree of divergence separating it from the remaining samples highlight its uniqueness and indicates that populations from this mountain range might have been isolated for a considerable period of time. Taxon allocation of the phylogeographic clades.—We identify three phylogeographic clades of Gracilinanus in eastern and cen- tral Brazil: Southeastern, Minas Gerais, and Central clades (Fig. 1). The Southeastern clade includes a specimen from Ipanema, the type locality of G. microtarsus (locality 8; Fig. 1; Appendix) and for that reason, as well as by comparison to the description of the holotype given by Tate (1933:191), we refer it to this species. The allocation of names to the two other clades, however, is more complicated. The name G. agilis has been historically applied to populations in central Brazil, from the state of Ceara in northeastern Brazil southwest to Lagoa Santa, and then continuing southwest, but interior to the coastal regions of Sao Paulo, Parana, Santa Catarina, and Rio Grande do Sul, to the Chaco of Paraguay and Argen- tina (Tate 1933). This range matches the distribution of our samples from the Central clade. However, the Minas Gerais clade in- cludes a specimen from Lagoa Santa (lo- cality 1; Fig. 1), the type locality of G. agil- VOLUME 116, NUMBER 2 Table 1.—Extended. LPCS581 G. agilis LPC599 G. agilis UHECO4722 G. agilis LPC476 G. agilis UHES1759 G. agilis LPC250 G. agilis UUP1292 G. agilis LC49 M. incanus . MNRJ31445 G. microtarsus MNRJ31447 G. microtarsus MCNM299 G. microtarsus LC1 G. microtarsus RM26 G. microtarsus MCNM394 G. microtarsus RM4 G. microtarsus LPC805 G. microtarsus MVZ182056 G. microtarsus LP40 G. microtarsus MAMA427 G. microtarsus LC189 G. agilis MNRJ31396 G. agilis YL64 G. agilis LPC241 G. agilis AP22 G. agilis LPC304 G. agilis LPC599 G. agilis LPC581 G. agilis UHECO4722 G. agilis 0.45% 0.79% 0.88% 2.07% IED S70 4.47% 3.31% 14.68% 2.05% 1.26% 3.88% 3.10% 14.38% 2.16% 1.39% JIS 2.97% 13.74% LPC476 G. agilis UHESM1759 G. agilis LPC250 G. agilis UUP1292 G. agilis LHE1342 G. aceramarcae 1.10% 5.07% 3.24% 15.83% 3.62% 2.84% 14.39% 2.04% 14.38% HSIN is, which is situated on the border with the Central clade. To resolve this problem, we compared our specimens with the descrip- tions of each holotype. If our specimen from Lagoa Santa is the same as Burmeis- ter’s type of that taxon, then G. agilis would become the name available for the Minas Gerais clade, recognized either as a separate species, or as a Synonym of G. microtarsus. In either case, the Central clade, which we identify, would require another name, pre- sumably one of those currently listed as synonyms of G. agilis. On the other hand, if the type specimen of G. agilis matches the morphology of the specimens from the Central clade, then we are left with a de- cision to make about the Minas Gerais clade, which could be recognized as a more interior clade of G. microtarsus or a sepa- rate, undescribed species, by virtue of its high level of divergence from topotypic G. microtarsus. The type of G. agilis is a young adult from Lagoa Santa, collected by H. Bur- meister and deposited at the Zoologisch Museum in Halle, Germany. The skin is mounted and the skull is in fragments, with only the rostrum and palate intact (Tate 1933: Plate 24, photo 215). Since we have not seen the type material of G. agilis, the best we can do at present is to follow Tate’s (1933) diagnosis. The following is his de- scription (p. 195) of the type of G. agilis: “Skin of type very faded brown. The hair rather close and even in length. Chest and posterior parts gray-based; throat and neck with self-colored hairs, buff-white. Vibris- sae short; eye-rings elongate before and be- hind, narrowed above and below. Feet small. Tail thickly covered with fine hair.” 284 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Means (+SE) and ranges for selected external and cranial variables of G. agilis and G. microtarsus?. Weight in grams, measurements in millimeters. Character Head and body length Tail length Hind foot length Ear length Mass (g) Greatest skull length Zy gomatic breadth Braincase breadth Least interorbital constriction Rostral length Nasal length Rostral width Molar toothrow length Least pterygoid breadth Petrosal breadth Alisphenoid bulla breadth Palatal length Gracilinanus agilis Male 100.0 + 7.06 82—115 n= 44 IB7Ol se 9.30) 110-158 n = 43 16.86 + 0.98 15-19 n= 44 Dix EZ MNBz 20-25 n= 44 MBN) 22 OI 15-40 n= 44 D&A s= i OW 25.87—30.59 n = 36 15.49 + 0.66 14.32—17.00 n = 35 EZ OR== 1032 10.75—-11.91 n = 36 AL 7) 35 (2D 4.20—5.06 n = 36 IOVS 2 O53) 9.29-11.56 n = 36 IZM 2 OI 10.32—13.37 n = 36 4.49 + 0.23 4.03-4.98 n = 36 5.49 + 0.16 5.19—-5.83 n = 36 BZ 22 OD 2.36—3.44 n = 36 oreot 22 O25) 7.81—8.84 n = 36 OD == 028 8.56—9.82 n = 35 13.89 + 0.56 12.68—14.95 n = 36 7k KK 7 OK 2 7k 2K ok 7K OK 2K 7K OK OK TK KK 2k 2K ok ns ns ns Female 89.63 + 7.28 81—108 in = ZI 123.96 + 7.54 110-139 n = 27 L503) 22 ZI 13-18 n = 27 ANS 2 NOI 20-24 n = 27 OMS 2 ZLO 13-25 n = 27 KS )s) 22. Il {Oks 25.23—29.69 n = 24 14.63 + 0.70 13.68—16.65 n= 24 11045 == 0:33 10.23-11.59 n= 24 4.46 + 0.22 3.96-4.86 n= 24 10.17 + 0.49 9.28-11.03 n = 24 11.39 + 0.74 10.22—13.18 n = 24 4.21 + 0.26 3.74 4.75 n= 24 S54 se ONS 5.12-5.63 n= 24 AGB) SE (DAY 2.59—3.34 n = 24 8.17 + 0.20 7.719-8.45 n = 24 8.96 + 0.33 8.15—9.35 n = 24 13.06 + 0.67 12.08—-14.39 if — 25 7K KK ns Gracilinanus microtarsus Male 04450) ae 114ES)5) 86-129 n = 20 ISAO ae FoI3 139-167 n = 20 IPOS 2 iB 15—20 n = 20 20.60 + 1.23 19-23 n = 20 AVAGO 22 IOG2 17-52 n = 20 0) 727) 32 lies! 28.16—33.20 n= 12 LO.37 ae I.i3} 14.77-18.38 wy, 11.94 + 0.20 11.30—12.25 n= 12 5.24 + 0:34 4.73-5.93 n= 12 ila) 22 O79) 10.30—12.98 n= 12 WAS a2 1 4! 11.22—14.52 n= 12 4.69 + 0.36 4.32-5.46 n= 12 5.1/3 22 OI 5.50-—6.07 n= 12 So OM4: 3.11—3.59 n= 12 85/80 == 10182 8.0—9.16 n= 12 NGOS 9.12—-10.04 n= 11 14.70 + 0.83 13.11-16.08 n= 12 ns TKK ns ns ns ns ns ns ns ns ns Female 95.20 + 16.86 81-116 n=5 140.20 + 9.34 131-155 n=5 15.80 + 0.84 15-17 n=5 IQA se O34! 18—20 n=5 22.40 + 10.74 12-37 n=5 28.29 + 1.48 26-80 29.75 n=4 ISe5il se 1 24! 14.28-16 99 n=4 INOS) se OL20) 11.35—11.80 n=4 4.98 + 0.28 4.70-5.29 n=4 10.59 + 0.96 9.68—-11.45 n=4 WANS) 22 1-40 10.61—13.63 n=4 4.43 + 0.39 4.03-4.85 n=4 5.66 £27020 5.42-5.90 n=4 3y610) 22 Oe I. II 3.25—-3.48 n=4 8.6372 0938 8.12—9.01 n=4 9.31 + 0.31 9.04-9.72 n=4 ioe 7S) 32 O88) 12.75—-14.57 n=4 VOLUME 116, NUMBER 2 285 Table 2.—Continued. Gracilinanus agilis Gracilinanus microtarsus Character Male Female Male Female Cranial depth Doie== Oil soa 9139) == 0334 ae 9) O)7/ aE (0).333) “s 9.57 + 0.18 8.80—10.10 8.63—9.84 ns 9.41—10.52 9.34-9.74 n = 36 n = 24 n= 12 n= 4 «Mean and standard deviation, with range below, for adult specimens of both sexes. The significance levels between males and females of each species and between the two species for both males (above) and females (Gelow) iaresindicateds ins) —\p)>a0!055 5p) = 01055 + — p=. 0 01 +25 — p< 0:00: He then (p. 195) gives a more complete de- scription of a specimen from Sapucay, Par- aguay deposited at the British Museum (now Natural History Museum of London; BM 3.4.7.22): ““‘Dorsal color a dull, almost dusty shade, nearest to russet (R.) or Brus- sels brown (R.). Color ventrally between warm buff (R.) and cinnamon-buff (R.). Gray-based hairs throughout, except from upper chest to chin and around the scrotal region. Face not markedly paler than dor- sum; ears large, light gray-brown; vibrissae short; feet rather pale; tail very slightly bi- color, brown.”’ Finally, in comparison with G. microtarsus, Tate wrote (p. 191): “Mi- crotarsus 1s much more vividly colored than agilis, and the spreading of the gray- based hairs completely across the ventral area, except for the chin, affords another probably constant difference.” Tate’s description of G. agilis fits quite well the individuals from the Central clade, especially the appearance and color of the dorsal pelage, the lack of strong contrast be- tween face and dorsum, and in the extent to which the gray-based hairs cover the ventral region. All seven specimens examined from the Minas Gerais clade—including the in- dividual from near the type locality of G. agilis at Lagoa Santa (MNRJ 31445)—are warm-brown on the dorsum with a much paler face, and gray-based hairs cover most of the underparts, except the chin. These are features that characterize individuals of G. microtarsus. In contrast, all 62 specimens of the Central clade are grayish-brown and without a sharp distinction between face and dorsum. The gray-based hairs in the under- parts are restricted to the lower pectoral and abdominal area leaving upper chest, throat and chin covered with self-colored hairs (same color from base to tip). These and oth- er traits (discussed below) convince us that the specimens forming the Central clade cor- respond to G. agilis. Despite the high ge- netic divergence found between the South- eastern clade and the Minas Gerais clade (Fig. 3), and the clear geographic break sep- arating them (Serra da Mantiqueira; Fig. 1), at this time we allocate the Minas Gerais clade also to G. microtarsus, since we could find no diagnostic morphological characters that justify specific distinction. Nevertheless, we point out that our sample size of G. mi- crotarsus is small and it is possible that more individuals will provide the morpho- logical support for distinction at the species level. It is important to note that G. agilis and G. microtarsus occur in near-sympatry in the state of Minas Gerais, since both spe- cies are now known from the general vicin- ity of Lagoa Santa. This locality is in a con- tact area between two biomes: the Cerrado, which has savanna-like vegetation character- istic of central Brazil, and the Atlantic For- est, represented at Lagoa Santa by semi-de- ciduous forests. Based on current distribu- tional information, Gracilinanus agilis oc- curs primarily in the Cerrado domain, while G. microtarsus is an Atlantic Forest species. In the following section we compare the two species and give an extended descrip- tion of both G. agilis and G. microtarsus. Morphological comparisons.—External body measurements: Gracilinanus agilis is sexual dimorphic, as males are significantly 286 3.8 4— ie es ee ee ee ee ee eee microtarsus SHegh | LPB w ! 7 T eee Ss a ee 22 a Soll So2 So) SA! 5:5 SiS S./ SS SY © Goll M1-M4 1OC un (aS eeu SUT ava Un RUN EU ge Le LT 2 23 2% 2 2 s® si 32 33 & GSL PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 12.4 2 Se eee ee Ee ee ee ee ee ee 122 >| - Ia 5 [ Wilss r a D1 ~T74 4 pet agilis F ws 10.8 4 Jd | 10.6 a a an ae een nam 25 26) 277728) 297 930 3 Car GSL 170 | ee | 1 { 1 ! l L i 1 ! 1 i 1 [eee | [ 1 microtarsus TAIL 80 90 100 110 120 130 HB Fig. 4. Bivariate plots comparing selected external body and craniodental measurements of Gracilinanus agilis (open squares) and G. microtarsus (solid circles). Above left: molar toothrow length (M1—MA4) versus least pterygoid breadth (LPB) of males and females. Above right: greatest skull length (GSL) versus braincase breadth (BB) of males. Below left: GSL versus least interorbital constriction (IOC) of males. Below right: head and body length (HB) versus tail length (TAIL) of males. larger than females in every trait examined (see Table 2). Adult males range in head and body length from 82 to 115 mm and weigh from 15 to 40 g. Females range in head and body length from 81 to 108 mm and weigh from 13 to 25 g. The tail is lon- ger in males, reaching 158 mm, while in females the tail reaches a maximal length of 139 mm. The ears of G. agilis are large, averaging more than 21 mm in both sexes. In G. microtarsus the tail, foot, and ears are longer in males than in females, while head and body and weight do not differ signifi- cantly (Table 2). In comparison, G. agilis is generally smaller than G. microtarsus, with females of G. microtarsus almost as heavy as males of G. agilis, and males of G. mi- crotarsus approximately 17% heavier on average than males of G. agilis. The tail in G. microtarsus 1s also longer, by about 11% more in both sexes (Table 2; Fig. 4). The ears are the only trait that is larger in G. agilis than in G. microtarsus, which may be related to the warmer and drier habitats this species occupies. Pelage differences: In both species, the bases of the hairs on the upper parts are dark gray and the tips are orange to buffy. However, in G. agilis the terminal (orange or buffy) portion of each hair is shorter than in G. microtarsus, giving the dorsal pelage a more grizzled tone, while G. microtarsus has a more uniformly colored reddish- brown pelage (Fig. 5; Table 3). The general color of the dorsal pelage is also paler in G. agilis than in G. microtarsus (Fig. 5). The fur is slightly longer and more lax in G. microtarsus than in G. agilis, although the series of G. microtarsus from Ipanema have the shortest fur of all specimens ex- amined. Although Tate (1933) emphasized the presence of numerous over-hairs in G. VOLUME 116, NUMBER 2 287 JEN S)s nanus agilis: first (LPC 599) and third (LPC 581) from left; and G. microtarsus: second (LPC 805) and fourth (LC 1) from left. Note the grizzled and paler dorsal pelage of G. agilis, and the ventral gray-based hairs extending to the throat in G. microtarsus. Note also the sharp contrast between face and body in G. microtarsus, and the gradual change in G. agilis. microtarsus, making the pelage “‘rough or shaggy-looking’’, our specimens appear smooth-coated. The general color of the ventral fur is yellowish cream in both species, but as mentioned above, the gray-based hairs of G. agilis are restricted to the lower pectoral and abdominal area, leaving the upper chest, throat and chin self-colored (Fig. 5). Also, the gray tone on the venter is slightly paler than that of the dorsum, making the gray color less evident. Individuals from the two more interior clades of G. agilis have the underparts of the arms also self-colored. Gracilinanus microtarsus, in turn, has gray- Differences in dorsal and ventral color patterns of Gracilinanus agilis and G. microtarsus. Gracili- based hairs throughout the ventral parts, ex- cept on the chin (Fig. 5), and the gray tone of the venter is as dark as it is in the dor- sum, making it obvious. Apparently, the population of G. microtarsus in the type lo- cality is an exception to this generality. Among the series of six topotypes we col- lected, one individual has the self-colored area restricted to the chin; three individuals have the self-colored hairs reaching the up- per chest; and the last individual has the self-colored hairs reaching as far as the lower abdominal area. Tate’s description (1933:191) of the type specimen of micro- tarsus is ‘“‘Underparts cinnamon-buff, the 288 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 3.—Diagnostic characters of G. agilis and G. microtarsus. Character G. agilis G. microtarsus External appearance of dorsum pelage spreading of gray-based hairs across the ventral area ocular-mark nose and ears grizzly grayish-brown throughout, except from upper chest to chin (underparts of arms also often self-colored) thin, small and not extending to uniform reddish-brown variable, but usually throughout except for the chin broad, large, and extending to nose and ears face not markedly paler than dorsum; markedly paler than dorsum, pro- transition is gradual, not sharp ducing a sharp contrast with contrast with the rest of body the rest of body ears larger; >21 mm in average smaller; <21 in average tail shorter; <140 mm in average longer; >140 mm in average Cranial posterolateral vacuities on palate size larger or comparable to that of the posteromedial vacuities always smaller than posteromedi- al vacuities Side view of the heads of Gracilinanus Fig. 6. agilis, above (LPC 599), and G. microtarsus, below (LPC 805), showing the differences in size and shape of the ocular mask. hairs gray-based, and only those of the chin self-colored’’, although he also recognized some variation since on page 189 he wrote: “‘underparts, posterior to the throat, with the hairs entirely gray-based’’. Intraspecific variation is also found in G. agilis: the se- ries of 13 individuals from Crato (locality 23) are even paler than the remaining spec- imens of this species, the color of the dorsal region approximating a grayish tone and the self-colored parts in the underparts being whitish cream. The ocular ring in G. microtarsus 1s broad and very dark (Fig. 6); the face is distinctly paler than the body—approxi- mately the same color as the cheeks—re- sulting in a sharp contrast between the face and body (Fig. 5). Although the mask is also prominent in G. agilis, it is thinner and antero-posteriorly restricted (Fig. 6); the face is not as pale—the color on top of the head is darker than the cheeks—and the transition between the face and body is gradual (Fig. 5). Cranial differences: When comparing G. microtarsus and G. agilis Tate (1933:191) observed that ““The skulls are extremely alike, differing solely in finer detail’. The only differences he pointed out were the VOLUME 116, NUMBER 2 289 Fig. 7. Ventral and dorsal views of skulls of Gracilinanus agilis to the left (LPC 307, from Nova Ponte, Minas Gerais), G. microtarsus from the Minas Gerais clade (YL 1, from Santa Rita de Jacutinga, Minas Gerais) in the middle, and a topotype of G. microtarsus from the Southeastern clade to the right (MVZ 197436, from Ipanema, Sao Paulo). Scale bar = 5 mm. Note the larger posterolateral palatal foramina in G. agilis, when compared to the size of the posteromedial vacuities. 290 longer tooth rows of G. microtarsus, its somewhat broader and shorter pterygoids and proportionately greater breadth across the pars petrosa. While we confirmed Tate’s findings with confidence for both males and females (Table 2; Fig. 4), we also detected additional differences. Skulls of G. micro- tarsus are longer (Figs. 4 and 7), and they also have a wider braincase, wider zygo- matic arches, and broader interorbital re- gion (Table 2; Fig. 4). Although the small number of females of G. microtarsus com- promises our statistical analyses, there is a significant difference between males of both species in all remaining traits exam- ined (Table 2), despite a comparable range of age classes. In terms of qualitative characters, the size of the posterolateral foramina of the palate is comparable to that of the posteromedial vacuities in G. agilis, being proportionally larger than in G. microtarsus in which they are always smaller than the posteromedial vacuities (Fig. 7). Incipient postorbital pro- cesses can be present in both species, con- trary to Hershkovitz’s (1992) observation that both species lack them. Conclusion Gracilinanus agilis and G. microtarsus appear to represent valid species, distin- guishable by morphological and molecular characters, with levels of sequence diver- gence equivalent to those separating each of them from a third species, G. aceramarcae. In short, these two do not even form a well- supported sister-pair within the genus, even with the very limited number of taxa sam- pled. However, it is also possible that the samples we allocate to G. microtarsus in fact represent two separate species, as in- dicated by the cyt b data. Verification of this hypothesis will have to wait until more samples are available. Acknowledgments We thank each of the following col- leagues for providing specimens used in our PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON research and/or for aiding in our own field collections: Louise H. Emmons, Lena Gei- se, Rachel T. de Moura, M6nica T. da Fon- seca, Alexandre M. Fernandes, Daniela C. Bianchini, Leonardo G. Vieira, Francisco P. C. Santos, Rodrigo L. Dias and Marco Au- rélio L. Sabato. We thank the curators Mar- i0 de Vivo (MZUSP), Leandro Salles (MNRJ) and Leonardo G. Lessa for making specimens available. Leonardo G. Lessa also contributed with tissues for the molec- ular analyses. Margaret E Smith, Sheda Morshed, and Yair Chaver provided aid in the laboratory. Bruce D. Patterson, Alfred L. Gardner, and Robert S. Voss offered crit- ical comments on the manuscript. The Na- tional Geographic Society, World Wildlife Foundation, Museum of Vertebrate Zoolo- gy, and National Science Foundation pro- vided financial support for laboratory anal- yses and/or fieldwork. Collecting permits were provided by IBAMA (Instituto Bras- ileiro do Meio Ambiente). LPC was sup- ported by a fellowship from the Fundagao Coordenagao de Aperfeigoamento de Pes- soal de Nivel Superior (CAPES) and YLRL was supported by a fellowship from the Conselho Nacional de Desenvolvimento Cientifico e Tecnol6gico (CNPq). Literature Cited Bremer, K. 1988. The limits of amino-acid sequence data in angiosperm phylogenetic reconstruc- tion.—Evolution 42:795—803. Costa, L. P. 2002. Relacoes filogenéticas intergenéricas de marsupiais didelfideos. In N. C. Caceres and E. L. A. Monteiro-Filho, eds., Marsupiais Bras- ileiros Editora da Universidade Federal do Pa- rana, Parana (in press). Creighton, G. K. 1984. Systematic studies on opos- sums (Didelphidae) and rodents (Cricetidae). University of Michigan, Ann Arbor, 220 pp. Eisenberg, J. F, & K. H. Redford. 1999. Mammals of the Neotropics: the central Neotropics, vol. 3: Ecuador, Peru, Bolivia, Brazil. The University of Chicago Press, Chicago, 609 pp. Emmons, L. H., & E Feer. 1997. Neotropical rainforest mammals: a field guide, 2nd edition. University of Chicago Press, Chicago, 307 pp. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap.—Evolution 39:783-791. VOLUME 116, NUMBER 2 Gardner, A. L. 1993. Order Didelphimorphia. Pp. 15— 23 in D. E. Wilson and D. M. Reeder, eds., Mammal species of the world. Smithsonian In- stitution Press, Washington, D.C., 1206 pp. Gardner, A. L., & G. K. Creighton. 1989. A new ge- neric name for Tate’s (1933) microtarsus group of South American mouse opossums (Marsu- pialia: Didelphidae).—Proceedings of the Bio- logical Society of Washington 120:3-—7. Hershkovitz, P. 1992. The South American gracile mouse opossum genus Gracilinanus Gardner and Creighton, 1989 (Marmosidae, Marsupia- lia): a taxonomic review with notees on general morphology and relationships.—Fieldiana: Zo- ology, New Series 70:1—56. Kimura, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences.— Journal of Molecular Evolution 16:111—120. Papavero, N. 1971. Essays on the history of neotrop- ical dipterology, with special reference to col- lectors (1750—1905), vol. 1. Museu de Zoolo- gia, Universidade de Sao Paulo, Sao Paulo, Bra- VAN Patton, J. L., & L. P. Costa. 2003. Diversidade, limites geograficos e sistematicos de marsupiais brasi- lerios. In N. C. Caceres and E. L. A. Monteiro- Filho, eds., Marsupiais Brasileiros. Editora da Universidade Federal do Parana, Curitiba (in press). Patton, J. L., M. N. E da Silva, & J. R. Malcolm. 2000. Mammals of the Rio Jurua and the evolutionary and ecological diversification of Amazonia.— Bulletin of the American Museum of Natural History 244:1—306. Saiki, R. K., et al. 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase.—Science 239:487—491. Swofford, D. L. 1999. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods), ver- sion 4. Sinauer Associates, Sunderland, Mas- sachusetts. Tate, G. H. H. 1933. A systematic revision of the mar- supial genus Marmosa, with a discussion of the adaptive radiation of the murine opossums (Marmosa).—Bulletin of the American Muse- um of Natural History 66:1—250. Vanzolini, P. E. 1993. As viagems de Johann Natterer no Brasil, 1817—1835.—Papeis Avulsos de Zoologia 38(3):17—60. Voss, R. S., D. B. Lunde, and N. B. Simmons. 2001. The mammals of Paracou, French Guiana, a Neotropical lowland rainforest fauna, part 2. Nonvolant species.—Bulletin of the American Museum of Natural History 263:3—236. Walsh, P. S., D. A. Metzger, & R. Higuchi. 1991. Chel- ex 100 as a medium for simple extraction of 2S) DNA for PCR-based typing from forensic ma- terial—Biotechniques 10:506—513. Appendix Gazetteer.—Localities from which we examined material of Gracilinanus for this study are listed by species and numbered to correspond with the mapped points in Fig. | and the labeled terminals in Fig. 3. Voucher numbers for specimens examined are given parenthetically for each locality; those marked with an asterisk correspond to haplotypes used in the molecu- lar analysis, and boldface place names are used to label the tree. Specimens examined are deposited in the col- lections of the Museum of Vertebrate Zoology, Uni- versity of California, Berkeley (MVZ); Museu de Zoologia da Universidade de Sao Paulo (MZUSP); Museu Nacional, Rio de Janeiro (MNRJ); and Museu de Ciéncias Naturais da Pontificia Universidade Ca- tolica de Minas Gerais (MCN); and Museo de Historia Natural, Universidad Mayor de San Marcos, Lima, Peru (MUSM). Specimens identified by other prefixes correpond to collector’s field number: AP (Adriano Paglia), CEMIG (Luiz Fernando B. M. Silva), RM (Raquel Moura) to be deposited at the Departamento de Zoologia, Universidade Federal de Minas Gerais, Belo Horizonte; LC and LPC (Leonora Pires Costa), LP (Luciana Pereira), UUPI (Maria José de J. Silva) and YL (Yuri Leite) to be deposited at one of the above Brazilian institutions; and UHECO and UHESM (Nelson da Silva) to be deposited at the Universidade Cat6élica de Goias, Brazil. Gracilinanus microtarsus 1 Fazenda das Bicas, 7.8 km (by rd.) SSE Lagoa Santa, Minas Gerais, Brazil, 19°38'’S, 43°53’W (MNRJ 31445*). 2 Estacgao de Pesquisas de Peti (CEMIG), 10 km (by rd.) Sao Gongalo do Rio Abaixo, Minas Gerais, Brazil, 19°49’S, 43°21’W (MNRJ 31447*, CEMIG SIL). 3 Parque das Mangabeiras, Belo Horizonte, Minas Gerais, Brazil, 19°55'S, 43°56’W (MCN 299*). 4 Cruzeiro, 8 km NE Santa Rita de Jacutinga, Minas Gerais, Brazil, 22°5'S, 44°2'W 560 m (LC 1*, LC J Nols; MN). 5 Jurumirim, Minas Gerais, Brazil, 20°8’S, 42°41'W (MCN 394*). 6 Parque Estadual do Rio Doce, 13 km E Marliéria, Minas Gerais, Brazil, 19°43’S, 42°39'W 300 m (MVZ 197587*). 7 Parque do Caraga, 25 km SW Santa Barbara, Mi- nas Gerais, Brazil, 20°5'S, 43°30’W 1300 m (RM 4*). 8 Floresta Nacional de Ipanema, 20 km NW Soro- caba, Sado Paulo, Brazil, 23°26'7"S, 47°37'41"W 701 m (LPC 801*, LPC 820-822, MVZ 197463). This is the historical site of ““Ypanema’’, the type N \O i) 11 locality of G. microtarsus, where the collector, Jo- hann Natterer, lived and worked (see Papavero 1971, Vanzolini 1993), and renamed as a formal conservation unit by IBAMA on May 20, 1992. Fazenda Intervales, Base do Carmo, 5.5 km S Ca- pao Bonito, Sao Paulo, Brazil, 24°20'S, 48°25'W 700 m (MZUSP 29158-29161, MZUSP 29165, MVZ 182054, MVZ 182055, MVZ 182056*, MVZ 182057). Vila Dois Rios, Ilha Grande, Angra dos Reis, Rio de Janeiro, Brazil, 23°9'S, 44°14’W (LP 40*). Estagao Biolégica de Boracéia, Sao Paulo, Brazil, 23°39'S, 45°54’W 850 m (MZUSP 29162, MZUSP 29163*, MZUSP 29164). Gracilinanus agilis [2 13 14 1S) 16 7) Ponte do Colatino, margem esquerda do Rio Je- quitinhonha, Coronel Murta, Minas Gerais, Brazil, 16°36'S, 42°12’W (LC 189*). Vargem do Retiro, Ribeirao Mascates, Parque Na- cional da Serra do Cip6, Minas Gerais, Brazil, 19°14’S, 43°33’W 800 m (MNRJ 31396*). Parque Estadual do Rio Preto, 15 km S Sao Gon- calo do Rio Preto, Minas Gerais, Brazil, 18°9’S, 43°23'W 950 m (YL 64*, LC 71). Fazenda Corredor, Bocaitiva, Minas Gerais, Brazil, 17°22'14"S, 43°52'16"W (AP 22*). Fazenda Santa Rita, 8 km E Andarai, Bahia, Brazil, 12°48'6"S, 41°15'41"W 399 m (LPC 241*). Mata do Vasco, 12 km W Nova Ponte, Minas Ger- ais, Brazil, 19°10'15"S, 47°42’29W 878 m (LPC 296-299, LPC 304*, LPC 305-307, LPC 309-311, 18 19 20 Di D7 W3} 24 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON LPC 314-316, LPC 318, LPC 324, LPC 325, LPC 327-328, LPC 330-331, LPC 339-340, MVZ 197438—-197446, 197451—197453, 197473, 197649— 197652, 197654-197657). Rio Miranda, above Passo do Lontra, Mato Grosso do Sul, Brazil, 19°34'35”S, 56°55’44”"W 100 m (LPC 599*, LPC 602, MVZ 197455). Base de Pesquisa do Pantanal—CENAP/IBAMA, 110 km SSW Poconé, Mato Grosso, Brazil, WAALS SOS 6 AI NN 98) im (ERGs sian 197454). Usina Hidrelétrica de Corumba, 30 km SE Caldas Novas, Goids, Brazil, 18°0'’S, 48°30’W (UHECO AWDO. Fazenda Lagoa Bonita, 36 km N Barra do Gargas, Mato Grosso, Brazil, 15°34’50”S, 52°22'29"W 331 m (LPC 476*). Usina Hidrelétrica de Serra da Mesa, Serra da Mesa, Goids, Brazil, 13°50’S, 48°18’W (UHESM WSS). Chapada do Araripe, 7 km SW Crato, Ceara, Bra- Zed V6 3978S) 3927.23 WW 960m (RGM Asn 244, LPC 249, LPC 250*, LPC 251, LP€ 269= 271, LPG 276-277," EE© 294. MNA Mo y447—= 197450, MVZ 197647—-197648). Estagao Ecolé6gica de Urucui-Una, Piaui, Brazil, 8°50'S, 44°10’W (UUPI 292*). Gracilinanus aceramarcae Vis) Cordillera de Vilcabamba, La Convencion (Camp 1), Junin, Peru, 11°39'56"S, 73°38'31"W (MUSM 13002*). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):293-—300. 2003. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 10. Cyanomyia salvini Brewster, 1893, is an intergeneric hybrid of Amazilia violiceps and Cynanthus latirostris Gary R. Graves Department of Systematic Biology, MRC-116, National Museum of Natural History, P.O. Box 37012, Smithsonian Institution, Washington, D.C. 20013-7012, U.S.A. Abstract.—Cyanomyia salvini Brewster, 1893, collected in Sonora, Mexico, is shown to be a hybrid between Amazilia violiceps ellioti and Cynanthus latirostris magicus, whose breeding ranges overlap extensively in northwestern Mexico. This specimen represents the only known intergeneric hybrid between species currently placed in Amazilia and Cynanthus (Sibley and Monroe 1990). The unique type of Cyanomyia salvini Brewster, 1893, was collected by John C. Cahoon at Nichosari, Sonora, Mexico, on Search S87. Eanly reterences treat (GC salvini as a valid species (Boucard 1895, Ridgway 1911, Cory 1918, Simon 1921), although only Ridgway’s entry indicated a personal examination of the specimen. Griscom (1934:378) proposed a hybrid or- igin for C. salvini in his revision of Ama- zilia violiceps: “By inference I had always doubted the existence of another species of this genus [Amazilia] in So- nora. Geographically and faunally there is no basis for one, and the failure to duplicate the type in over 40 years has further significance, and in part at least strengthens this view. My late esteemed colleague Outram Bangs always supposed that salvini was of hybrid origin. A careful study of the color and struc- tural characters of the type convinces me that Cy- anomyia salvini Brewster is a hybrid between Ama- zilia violiceps conjuncta [= Amazilia violiceps el- lioti] and Cynanthus latirostris Swainson.” Griscom’s brief description was insufficient to make a convincing case for hybridiza- tion, but Peters (1945) and Phillips (1964) cited Griscom’s treatment without substan- tive comment. A second hypothesis was in- troduced in a succinct footnote by Fried- mann et al. (1950), who suggested that Cy- anomyia salvini is an aberrant example of Amazilia violiceps. This idea was later en- dorsed by Weller & Schuchmann (1997) and Weller (1999), but neither of these ref- erences provided corroborating evidence. As a consequence, the taxonomic status of Cyanomyia salvini is still in doubt. Here I provide a taxonomic assessment of Cyano- myia salvini employing the methods and as- sumptions outlined in Graves (1990) and Graves & Zusi (1990), as modified by the findings of Graves (1998, 1999b). Methods The type of Cyanomyia salvini, (J. C. Ca- hoon field number 505), originally part of the William Brewster Collection (No. 24,124), was eventually cataloged in the Museum of Comparative Zoology, Harvard University (received in 1918, No. 224,124). The type was sexed as do on the Brewster Collection label and appears to be in defin- itive plumage as judged by the absence of striations on the maxillary ramphotheca, the absence of distinctive buffy feather tips on the dorsal plumage, and the presence of a strongly iridescent coronal patch. Descrip- tions in this paper refer to definitive male plumage. I compared the type of Cyano- myia salvini with specimen series of Calyp- te annae, C. costa, Selasphorus platycercus, S. rufus, S. sasin, Stellula calliope, Archil- ochus alexandri, Calothorax luciferi, He- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Ranges (mean + standard deviation) of measurements (mm) of wing chord, bill length, and rectrix length (RI—R5) of adult males of Amazilia violiceps ellioti, Cynanthus latirostris magicus and a probable hybrid, Amazilia violiceps ellioti X Cynanthus latirostris magicus (= type of Cyanomyia salvini Brewster, 1893; MCZ 224, 124). Amazilia Cynanthus violiceps latirostris Cyanomyia Character N = 15-17 N = 14-15 salvini Wing 53.5—56.1 48.4-51.9 SDS) (54.6 + 0.7) (50.4 + 0.8) Bill 20.1—22.5 18.7—21.6 21.4 (AA: 2 OO) @03) 227038) R1 26.3—29.3 22.4—25.3 Di) (28.2 22 O) 303 2 0.7) R2 28.0—30.5 23.9-27.0 28.0 QZQA z= OY) (Zd23 22 O71) R3 28.3-31.4 26.3—29.4 2). 1\ GOs==mIe0)) (28.6 = 0.8) R4 28.3—32.5 28.7—32.4 30.6 GO.7 2 1.1) (BOS 2 1) R5 28.8—-32.6 31.0—33.8 SZ (SOs7 22 12) G25 2 0.7) liomaster constantii, Eugenes fulgens, Lam- pornis clemenciae, Amaczilia beryllina, A. violiceps (including the type of Amazilia violiceps conjuncta Griscom, 1934; MCZ No. 224,112), Hylocharis leucotis, and Cy- nanthus latirostris magicus, all of which occur in Sonora, Mexico (Friedmann et al. 1950, Howell & Webb 1995), in the collec- tions of the Museum of Comparative Zo- ology. Because the generic allocation of species traditionally placed in Amazilia by Peters (1945) is in flux, I use the species taxonomy of Sibley & Monroe (1990). De- tailed descriptions and photographs of the type of Cyanomyia salvini were compared with series of the aforementioned species in the National Museum of Natural History, Smithsonian Institution, and with a sup- posed immature specimen of Cyanomyia salvini collected at Palmerlee, Cochise County, Arizona, on 5 July 1905 (Bishop 1906). The latter specimen (Field Museum of Natural History 160,998; wing chord = 53.4 mm; bill length = 21.2; R4 = 29.7; RS = 29.9) appears to be a female Amazilia violiceps and will not be further discussed. Measurements were taken with digital calipers and rounded to the nearest 0.1 mm: wing chord; bill length (from anterior ex- tension of feathers); and rectrix length (from point of insertion of the central rec- trices to the tip of each rectrix) (Table 1). Pairs of rectrices are numbered from the in- nermost (R1) to the outermost (R5). Scatter plots of measurements and least squares re- gression lines were used to illustrate size differences among specimens. General color descriptions presented in Appendix | were made under natural light. I evaluated the color of the medial vane of the dorsal surface of RI (7 mm from tip) with a calibrated colorimeter (CR-221 Chroma Meter, Minolta Corporation) equipped with a 3.0 mm aperture. The mea- suring head of the CR-221 uses 45° circum- ferential illumination. Light from the pulsed xenon arc lamp is projected onto the spec- imen surface by optical fibers arranged in a circle around the measurement axis to pro- vide diffuse, even lighting over the mea- suring area. Only light reflected perpendic- ular to the specimen surface is collected for color analysis. Colorimetric data from iri- descent feathers are acutely dependent on the angle of measurement, the curvature of plumage surfaces in museum skins, and the degree of pressure applied to the plumage surface by the Chroma Meter aperture. In VOLUME 116, NUMBER 2 Figé 1. magicus (bottom), and a probable hybrid, Amazilia violiceps ellioti * Cynanthus latirostris magicus (= type of Cyanomyia salvini Brewster, 1893; MCZ 224,124). order to reduce measurement variation, I held the aperture flush with the rectrix sur- face without depressing it. The default set- ting for the CR-221 Chroma Meter displays mean values derived from three sequential, in situ measurements. I repeated this pro- cedure twice (five times for the type of Cy- anomyia salvini), removing the aperture be- tween trials. Thus, each datum summarized in Table 2 represents the mean of 6 (paren- tal species) or 15 (type of C. salvini) in- dependent colorimetric measurements. Colorimetric characters were described in terms of opponent-color coordinates (L, a, b) (Hunter & Harold 1987). This system is based on the hypothesis that signals from the cone receptors in the human eye are coded by the brain as dark-light (L), green- Lateral views of males in definitive plumage: Amazilia violiceps ellioti (top), Cynanthus latirostris red (a), and blue-yellow (b). The rationale is that a color cannot be perceived as red and green or yellow and blue at the same time. slUherekores aredmess;.) and ) .ereen- ness”’ can be expressed as a single value, a, which is coded as positive if the color is red and negative if the color is green. Like- wise, ““yellowness”’ or “‘blueness”’ is ex- pressed by b for yellows and —b for blues. The third coordinate, L, ranging from O to 100, describes the “‘lightness”’ of color; low values are dark, high values are light. The more light reflected from the plumage, the higher the LZ value will be. Visual systems in hummingbirds (e.g., Goldsmith & Gold- smith 1979) differ significantly from those of humans and the relevance of opponent color coordinates to colors perceived by 296 a0) fs = O ) a N 3 = O io) oO \ Ya wa = O o) ~ fo J oo) Cc oO — = 18 48 50 52 54 56 58 WING Fig. 2. Bivariate plots of measurements (see Table 1): Amazilia violiceps ellioti (hollow triangle), Cynan- thus latirostris magicus (A), and a probable hybrid, (sx) Amazilia violiceps ellioti X Cynanthus latirostris magicus (= type of Cyanomyia salvini Brewster, 1893; MCZ 224,124). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON hummingbirds is unknown. In any case, the L, a, b color system permits plumage color to be unambiguously characterized for tax- Onomic purposes. Results and Discussion I considered three hypotheses proposed by previous authors—Cyanomyia_ salvini represents (1) a hybrid, Amazilia violiceps x Cynanthus latirostris, (2) an aberrant ex- ample of Amazilia violiceps, or (3) a valid species. For brevity I use the epithet, sal- vini, in the remainder of the paper. I found no evidence that salvini repre- sents a subdefinitive plumage or geographic variant of any known taxon. The possibility that salvini represents an aberrant plumage of Amazilia violiceps can be rejected be- cause salvini has substantially shorter wings. All evidence is consistent with the hypothesis that salvini represents an inter- generic hybrid, Amazilia violiceps X Cy- nanthus latirostris. Several characters of salvini facilitate the identification of its pa- rental species (Appendix 1): (a) brilliant bluish-purple crown; (b) white chin, throat, midline of breast and abdomen white; (c) greenish-blue subterminal spots or bars on white feathers at the lateral margins of the throat and upper breast; and (d) absence of rufous or buff pigmentation on the second- aries or rectrices (Fig. 1). Here I present a synopsis of the critical steps of the hybrid diagnosis. The pool of potential parental species may be quickly narrowed by focusing on the white ventral plumage of salvini. Among the potential pa- rental species that occur in Sonora, Mexico, only Amazilia violiceps possesses white ventral plumage from chin to undertail co- verts. Bluish-green spotting on the lateral margins of the chin, throat, and breast feathers of salvini were inherited from the other parental species. Both Amazilia ber- yllina and Cynanthus latirostris have green or bluish-green plumage from chin to upper breast. Because brown or reddish-brown pigments appear to exhibit consistent pen- VOLUME 116, NUMBER 2 etrance in hummingbird hybrids (Banks & Johnson 1961, Graves & Newfield 1996), A. beryllina can be eliminated from further consideration because the secondaries and rectrices of salvini lack buff or rufous pig- ment. Only one pair of species (A. violiceps x C. latirostris) could have contributed the unique combination of characters exhibited by salvini (Appendix 1). Hybridization of Amazilia violiceps with any of the small gorgeted species (1.e., Ca- lypte annae, C. costa, Selasphorus platy- cercus, S. rufus, S. sasin, Stellula calliope, Archilochus alexandri, Calothorax lucifer) would likely produce offspring with one to many iridescent gorget feathers (Graves & Zusi 1990; Graves 1996, 1999a). In a sim- ilar fashion, Lampornis clemenciae (bril- liant gorget and large white tail spots), Eu- genes fulgens (brilliant gorget and black breast and abdomen), Hylocharis leucotis (white postocular stripe and black chin and auriculars), and Heliomaster constantii (brilliant gorget and semi-concealed white rump patch) are unlikely to be a parental species because they each possess a suite of plumage characters not observed, even as traces, 1n salvini. As a second step, I tested the restrictive hypothesis with an examination of size and external proportions (Fig. 2). Measurements of trochiline hybrids fall within the men- sural ranges exhibited by their parental spe- cies aS a consequence of a polygenic mode of inheritance (Banks & Johnson 1961). Measurements of Amazilia violiceps ellioti and Cynanthus latirostris overlap in four of the seven characters and the percent differ- ence in character means is modest (larger species divided by smaller): wing chord (8.3%), bill length (5.4%), R1 (18.5%), R2 (14.0%), R3 (5.2%), R4 (0.3%), and R5 (5.9%). Measurements of salvini fall at or between the character means (or 1.0 mm less than the mean value for R4 of A. viol- iceps) of the postulated parental species, and, in several cases, approximate the val- ues predicted by least squares regression. Rectrix color values in salvini fall between ilia violiceps ~ Table 2.—Minima, maxima, and means (+ standard deviation) of opponent color coordinates (L, a, b) of rectrix 1 (RI) of adult males of Ama ellioti, Cynanthus latirostris magicus and a probable hybrid, Amazilia violiceps ellioti X Cynanthus latirostris magicus (= type of Cyanomyia salvini Brewster, 1893; MCZ 224, 124). —b/b —ala Blue/yellow Green/red Max Lightness Mean + SD Min. Max Mean + SD Mean + SD Min Max Min. WO 28 lets 303) 1.6 ec a) Va) 0.7 + 1.4 BD 4.4 29.8 + 1.9 lie. =03 2 32 DP N= 19 ilia violiceps Aj Ama =I 25 A) OQ 2 17 AO) 2 13.6 Cynanthus latirostris Hall —1.6 DOD Cyanomyia salvini 29), Greenness (-) or Redness (+) Blueness (-) or Yellowness (+) 5 10°15 20°25-30 35 Lightness (L) Fig. 3. Bivariate relationships of L, a, b color co- ordinates: Amazilia violiceps ellioti (hollow triangle), Cynanthus latirostris magicus (&), and a probable hy- brid (sx), Amazilia violiceps ellioti X Cynanthus lati- rostris magicus (= type of Cyanomyia salvini Brew- ster, 1893; MCZ 224,124). the character means of the postulated pa- rental species (Table 2, Fig. 3). In summary, evidence obtained from plumage color and pattern, as well as from external size and shape, is consistent with the hypothesis that Cyanomyia salvini is an intrageneric hybrid between Amazilia viol- iceps ellioti and Cynanthus latirostris mag- icus, whose breeding ranges overlap exten- sively in northwestern Mexico. Cyanomyia salvini Brewster, 1893, is thus available in taxonomy only for the purposes of hom- onymy. Acknowledgments I thank Richard Banks and Richard Zusi for reviewing the manuscript, Doug Cau- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON sey, Alison Pirie, and Jeremiah Trimble (Museum of Comparative Zoology, Har- vard University) for permission to study the type of Cyanomyia salvini, David Hafner (Occidental College) for specimen loans, Tom Schulenberg for tracking down the Bishop Collection specimen, and John Bates (Field Museum of Natural History) for lending it. Travel was supported by the Alexander Wetmore Fund, Smithsonian In- stitution. Literature Cited Banks, R. C., & N. K. Johnson. 1961. A review of North American hybrid hummingbirds.—Con- dor 63:3-—28. Bishop, L. B. 1906. Uranomitra salvini in Arizona.— Auk 23:337-338. Boucard, A. 1893. Genera of humming birds, part 2. Published by the author, London. Brewster, W. 1893. Description of a new hummingbird from northern Mexico.—Auk 10:214—215. Cory, C. B. 1918. Catalogue of birds of the Americas, part 2, No.1.—Field Museum of Natural His- tory Zoological Series 13:1—315. Friedmann, H., L. Griscom, & R. T. Moore. 1950. Dis- tributional check-list of the birds of Mexico, part 1. Pacific Coast Avifauna No. 29, 202 pp. Goldsmith, T. H., & K. M. Goldsmith. 1979. Discrim- ination of colors by the black-chinned hum- mingbird, -Archilochus alexandri.—Journal of Comparative Physiology A 130:209—220. Graves, G. R. 1990. Systematics of the “‘green-throat- ed sunangels”’ (Aves: Trochilidae): valid taxa or hybrids?—Proceedings of the Biological Soci- ety of Washington 103:6—25. 1996. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 2. Hybrid origin of Erioc- nemis soderstromi Butler.—Proceedings of the Biological Society of Washington 109:764—769. 1998. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 6. An intergeneric hybrid, Aglaiocercus kingi X Metallura_ tyrianthina, from Venezuela.—Proceedings of the Biologi- cal Society of Washington 111:511—520. . 1999a. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 7. Probable parentage of Calliphlox iridescens Gould 1860.—Proceed- ings of the Biological Society of Washington 112:443—450. . 1999b. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 8. A provisional hypothesis for the hybrid origin of Zodalia glyceria (Gould, 1858).—Proceedings of the Biological Society of Washington 112:491—502. VOLUME 116, NUMBER 2 , & N. L. Newfield. 1996. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 1. Charac- terization of Calypte anna X Stellula calliope and the possible effects of egg volume on hy- bridization potential—Proceedings of the Bio- logical Society of Washington 109:755—763. , & R. L. Zusi. 1990. An intergeneric hybrid hummingbird (Heliodoxa leadbeateri X Helian- gelus amethysticollis) from northern Colom- bia.—Condor 92:754—760. Griscom, L. 1934. The ornithology of Guerrero, Mex- ico.—Bulletin of the Museum of Comparative Zoology 75:368—422. Howell, S. N. G. & S. Webb. 1995. The birds of Mex- ico and northern Central America. Oxford Uni- versity Press, Oxford, UK, 851 pp. Hunter, R. S., & R. W. Harold. 1987. The measurement of appearance, 2nd edition. Wiley, New York, 411 pp. Peters, J. 1945. Check-list of birds of the world, vol. 5. Museum of Comparative Zoology, Cam- bridge, Massachusetts, 306 pp. Phillips, A. R. 1964. Notas sistematicas sobre aves Mexicanas, III].—Revista de la Sociedad Mexi- cana de Historia Natural 25:217—242. Ridgway, R. 1911. Birds of North and Middle Amer- ica. Bulletin of the United States National Mu- seum 50, part 5. Sibley, C. G., & B. L. Monroe, Jr. 1990. Distribution and taxonomy of birds of the world. Yale Uni- versity Press, New Haven, Connecticut, 1111 Pp. Simon, E. 1921. Histoire naturelle des Trochilidae (synopsis et catalogue). Encyclopedia Roret, L. Mulo, Paris. Weller, A.-A. 1999. Violet-crowned Hummingbird, Agyrtria violiceps. P. 599 in J. del Hoyo, A. Elliott, & J. Sargatal, eds., Handbook of the Birds of the World, vol. 5. Barn-owls to Hum- mingbirds. Lynx Edicions, Barcelona, 759 pp. , & K.-L. Schuchmann. 1997. The hybrid ori- gin of a Venezuelan trochilid, Amazilia distans Wetmore & Phelps 1956.—Ornithologia Neo- tropical 8:107—112. Appendix | Comparative description of selected characters of adult male Amazilia violiceps ellioti, Cynanthus lati- rostris magicus, and a probable hybrid, A. violiceps ellioti X C. latirostris magicus (= Cyanomyia salvini Brewster; MCZ 224,125). Descriptions of structural colors are unusually subjective, as color seen by the observer varies according to the angle of inspection and direction of light. For this reason I use general color descriptions. The forecrown and crown of violiceps exhibit bril- liant bluish-purple iridescence when viewed head-on 299 in direct light. The remainder of the dorsum from hind- neck to the rectrices is grayish-olive; the transition be- tween the purple crown and olive hindneck is abrupt. The back, shoulders, upper tail coverts, and central rectrices are faintly glossed with silvery-green when viewed head-on. The rectrices are unmarked. Feathers of the forecrown and crown of Jlatirostris (to a point immediately posterior of the eyes) are dark green, margined with bronze giving the forehead and frontal part of the crown a dull appearance. The re- mainder of the dorsum from hindcrown to rump is glossy green (showing scattered bluish-green, brilliant- ly iridescent feather barbs when viewed head-on). Up- pertail coverts are darker, contrasting slightly with the green rump and the bluish-black rectrices. The outer two or three pairs of rectrices (R3, R4, R5) are nar- rowly tipped with gray. The crown of salvini is intermediate in appearance between that of violiceps and latirostris, showing a mixture of deep blue, purplish-blue iridescence, be- coming greener toward the posterior of the coronal area. The intensity of iridescence increases posteriorly from the forecrown to center of the crown (viewed head-on). This bluish-green iridescence at the posterior edge of the coronal area blends into deep bluish-green on the hindneck and back, changing to dull green on the lower back and rump. The back and scapulars of salvini appear bluer than those of either parental spe- cies. This seems to be another example of the “blu- ing’’ phenomenon observed in some hummingbird hy- brids (Graves 1998, 1999b). The intensity of irides- cence on the lower back of salvini is intermediate be- tween that observed in the postulated parental species. There is no appreciable contrast between uppertail co- verts and rectrices in salvini. The rectrices are inter- mediate in color between those of /atirostris and viol- iceps (Table 2). The outer rectrices (R4, R5) of salvini are worn and appear to be retained from a subdefinitive plumage. R1 and R2 are fresh and unworn, whereas the tip of R3 is slightly worn. Wing coverts and flight feathers of salvini are intermediate in color and degree of melanism to those of violiceps and latirostris. The ventral plumage of violiceps is snowy white from chin to undertail coverts. Feathers on the chin, throat, and upper breast are pure white to the base, with only a few pale gray barbs at the base of the rachises. The chin and upper throat of /atirostris are deep purple (extending laterally to the auriculars and eye- ring), blending into dark bluish-green on the lower throat; this latter color continuing posteriorly to the vent. A few feathers in the chin are fringed with white. Feather disks from chin to vent exhibit iridescent high- lights when viewed head-on. Vent feathers are white. Undertail coverts of latirostris are gray, paling to white at the distal margins (most coverts are gray along the rachis to the tip). Some shorter coverts have an oval gray subterminal spot. 300 The chin and throat of salvini are white but basal feather barbs are much grayer than in violiceps. The auriculars are iridescent purple and greenish-purple (posteriorly). Feathers at the sides of the throat have purple (anterior) or purplish-green (posteriorly) subter- minal bars, imparting a spotted appearance to the sides of the throat. This spotting coalesces on the sides of the breast to form an incomplete pectoral band formed of white feathers with large subterminal greenish-blue or greenish-purple disks. The sides of salvini are dark bluish-green (about the same color as in /atirostris). The tendency toward an incomplete pectoral band is PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON present in violiceps in the form of grayish-olive sides, with scattered weakly iridescent, pale bluish-green feather disks. However, the medial extension of darker feather tips observed in violiceps never reaches the ex- tent observed in salvini. The few undertail coverts re- maining on the specimen of salvini are pure white. The bill of violiceps is dull grayish-yellow (red in life) tipped with dark brown or blackish-brown (<15% of the bill length). Bill color in Jatirostris is similar but the dark tip is more extensive. The pattern of pig- mentation in salvini is intermediate of that observed in adult male specimens of violiceps and latirostris. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):301—307. 2003. A new species of caracara (Milvago) from Quaternary asphalt deposits in Cuba, with notes on new material of Caracara creightoni Brodkorb (Aves: Falconidae) William Suarez and Storrs L. Olson (WS) Museo Nacional de Historia Natural, Obispo 61, Plaza de Armas, Ciudad de La Habana, CP. 10100, Cuba, e-mail: geopal@mnhnc.inf.cu; (SLO) National Museum of Natural History, Smithsonian Institution, Washington D.C. 20560, U.S.A. e-mail: olson.storrs@nmnh.si.edu Abstract.—An extinct caracara, Milvago carbo, new species, is described from Quaternary asphalt deposits of Las Breas de San Felipe, northern Matan- zas Province, Cuba, from tarsometatarsi, tibiotarsi, and a notarium. This species was much larger than other species of the genus, living or extinct. New material of the extinct species Caracara creightoni provides additional information on its characters and distribution. Resumen.—Se describe una nueva especie de caraira, Milvago carbo, sobre la base de tarsometatarsos, tibiotarsos, y un notarium, procedentes de Las Breas de San Felipe (depésitos cuaternarios de asfalto), al norte de la provincia de Matanzas, Cuba. Esta especie era mayor que las demas conocidas del género, vivientes 0 extinguidas. Nuevo material de la especie extinta Caracara creigh- toni Suministra informacion adicional sobre sus caracteres y distribucion. The caracaras (Caracarinae = Polybori- nae auct.) comprise 10 species of mainly terrestrial Falconidae found throughout the Neotropics and in southern Florida and in temperate southern South America. The larger species are at least partly scavengers with somewhat vulture-like habits. The only living representative of the group in the West Indies is the widespread Crested Caracara, Caracara cheriway Jacquin, found in Cuba and the Isle of Pines (now Isla de la Juventud) (American Ornitholo- gists’ Union 1998, Raffaele et al. 1998). The Cuban populations cannot be distin- guished from those of the mainland and are thought to result from a relatively recent colonization of the island (Suarez and Ol- son 2001). Despite their modern paucity, the fossil record indicates that caracaras were much more diverse in the West Indies in the Qua- ternary. Caracara creightoni Brodkorb (1959b) is known from the Quaternary of New Providence Island, Bahamas (Olson and Hilgartner 1982), and of Cuba (Suarez and Olson 2001). Caracara latebrosus Wet- more (1920) from Puerto Rico is enigmatic as it is known so far only from two bones that are not particularly diagnostic (Olson 1976). A caracara from Grand Cayman re- ported as C. creightoni (Morgan 1994), is too large for that species but smaller than a very large, nearly flightless caracara discov- ered on Jamaica (Olson, unpublished data). Smaller caracaras of the genus Milvago have hitherto been known in the West In- dies only from Hispaniola, home to Milva- go alexandri Olson (1976), a species about the size of the living M. chimachima. The latter species, essentially South American in distribution, expanded into southern Costa Rica from Panama about 1973 (Stiles and Skutch 1989), but probably once occurred throughout Middle America and southern North America because it is known from the Pleistocene of Florida. The Florida fos- 302 sils were originally described as Falco readei Brodkorb (1959a), which was trans- ferred to Milvago by Campbell (1980), and then synonymized with M. chimachima by Emslie (1998). Fossils of Milvago of about this size have also been recovered from Cuba (Suarez and Arredondo 1997), but those known so far are too incomplete to be diagnostic at the specific level (Suarez and Olson, pers. obs.). During the processing of bird fossils from asphalt deposits of Las Breas de San Felipe, northern Matanzas Province, Cuba (see Iturralde-Vinent et al. 2000), new evi- dence of yet another species of caracara came to light. Although of relatively large size, this species is nevertheless referable to the genus Milvago. We also report previ- ously unknown skeletal elements of Cara- cara creightoni from additional localities in Cuba. Methods and comparative material ex- amined.—Measurements were taken with digital calipers to the nearest 0.1 mm. Os- teological terminology follows Howard (1929) and Baumel and Witmer (1993). Skeletons of Caracarinae examined at the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM), include the following species: Milvago chimango (7), M. chimachima (8), Daptrius ater (5), D. americanus (2), Phal- coboenus australis (5), P. carunculatus (1), P. megalopterus (1), Caracara plancus (7), C. cheriway (12), and C. lutosa (1). Fossil specimens examined included: Milvago al- exandri, holotypical right tarsometatarsus USNM 214573, Haiti, Hispaniola; Milvago sp., proximal end of left tarsometatarsus, collection of William Suarez, La Habana, Cuba (WS 977); Caracara creightoni, dis- tal end of right tibiotarsus lacking posterior rim of internal condyle, Museo Polivalente de Sagiia La Grande, Villa Clara, Cuba (MPSG 83), complete left tarsometatarsus MPSG 103, and distal half of left tarso- metatarsus MPSG 106, Cuba. The descrip- tion and illustrations of the extinct caracara PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Milvago brodkorbi' Campbell (1979), of Peru, were also used for comparison. Systematics Class Aves Family Falconidae Fossils were referred to the Falconidae instead of Accipitridae by the characteristic three openings on the distal portion of the tibiotarsus; tarsometatarsus with medial cal- caneal ridge of the hypotarsus longer and centrally placed rather than short and me- dial. Referral to the subfamily Caracarinae instead of Falconinae is indicated by the short hypotarsus, with an abruptly truncate distal margin, rather than being long and tapering gradually down the shaft as in the Falconinae. Genus Milvago Spix, 1826 The new species is referred to Milvago rather than to Caracara, Daptrius, or Phal- coboenus by having the tarsometatarsus very slender, trochlea for digit 2 broad at base, rotated slightly posteriad and with posterior wing straight, projecting posteri- orly perpendicular to the main axis of the shaft (See Campbell 1980). Milvago carbo, new species (Fig. 1) Holotype.—Nearly complete right tarso- metatarsus, Museo Nacional de Historia Natural, La Habana, Cuba (MNHNCu P4569), lacking most of the posterior edge of the medial calcaneal ridge and part of the outer cotyla. Collected by members of the Geology and Paleontology Group of the MNHNCu, during field expeditions in 1998. Type locality and age.—Cuba, Matanzas Province, Municipality of Marti, 5.5 km west of the town of Marti, asphalt deposit known as Las Breas de San Felipe, San Fe- lipe II site (Instituto Cubano de Geodesia y Cartografia 1986, 1:50,000 map, sheet Mar- ti 4084-IV, X502, Y347). Quaternary, prob- VOLUME 116, NUMBER 2 Fig. 1; Milvago carbo, new species, holotype, MNHNCu P4569 (specimen coated to enhance photography); C, Cara- cara creightoni MPSG 103 Gmage reversed to facilitate comparison). Scale = 2 cm. ably late Pleistocene or early Holocene; not directly dated (for description, associated fauna, and discussion of the age of the de- posit, see Iturralde-Vinent et al. 2000). Measurements (mm) of holotype.—Total length, 85.7; least width and depth of shaft, 45 X 3.6; distal breadth, 3.0: width and depth of trochlea for digit 3, 4.6 X 6.1. Topotypical paratypes.—Fragmentary notarium MNHNCu P4567; distal ends of left tibiotarst MNHNCu P4568, MNHNCu P4570, MNHNCu P4571; proximal half of left tarsometatarsus lacking part of the me- dial and lateral calcaneal ridges MNHNCu B4as/2> “shatt of left tarsometatarsus MNHNCu P4573; proximal right tarso- metatarsus MNHNCu P4574; distal halves of right tarsometatarsi MNHNCu P4575, MNHNCu P4576 (abraded); distal ends of left tarsometatarsi MNHNCu P4577, MNHNCu P4578. Tarsometatarsi in anterior, posterior, and medial views: A, Milvago chimango (USNM 18473); B, Etymology.—L. carbo, charcoal, coal, from the black color of the tar-impregnated fossils. Diagnosis.—Much larger than any other species of the genus Milvago living or ex- tinct (Table 1). Description.—Notarium large, slender and laterally compressed (smaller, relatively shorter but also compressed in Milvago chi- machima and M. chimango). Although the tibiotarsus is large, no qualitative characters could be discerned, as this element shows great intraspecific variation in Milvago (Emslie 1998). Tarsometatarsus large with the flattened medial (inner) surface angled towards the midline of the posterior face so that the posterior crest that it forms is near the midline and nearly continuous with the medial calcaneal ridge, and the posterior metatarsal groove is narrow, as in M. chi- mango, and as opposed to M. chimachima, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 304 6 (LE) TV VE 6 (Co) LEO 6 (88) TOI-8 6 VO LACE 6 (TE) EOC 6 (C8) LT OI-6'L 6 (L'6S) 8'€9-8°9S 6 CO SerL L WV) CLOW CC OSUDULIYS “W 3) SALE 8 (COE) SS-CE 8 (L'8) 0O0I-78 3 CO) LSE 8 (CE) LCE 8 (88) $6-S 8 8 (81S) T7S—-0'0S 8 (€8) 68-GL € (6 VC) 6SC-VVT él WE) VyAve vi (L'6) vOI-1 6 Sl Or) vr9e 6 (O'OT) 6 01-6 C (86S) 8 19-67LS ET (16) 66-78 - xx(Q661) OSU pure xx(6L61) Meqdwued ‘.(9L61) YOSTO Wor syuoUToIMseoT] € (9S) 1 9-TS € (Cv) 9PV-6'E SCC) OSI-CEl VVOVEVE E(Srn) SV COTO Ole LS8 SCO Gilr90! Este € USIP 1OF vapyoon jo uidoq € USIP 1OF vafyoon JO YIPIM YIpim [Ist yeys Jo ujdop seo] yeys JO YIPIM jsvo'T YIpPIM [VWTXOld yjsuo] SNSIEIVIOUOSIL J, UIPIM [eISI snsiejOIq! |, sus] [210 L UINLIION see DULIYIDULIYD “PAL x41 G404YPO1G “W yl/PUDxXaID “We OG4DI “We JUSUTOINSRITAI ‘u (uRoUT) asURI :sI BDUaNbag ‘~OsHATIW JO soroods SUIAT] pure [ISSOJ JO (WU) sJUSUMOINSvOU [RIOTOYS— 1 STqeL VOLUME 116, NUMBER 2 M. alexandri, and M. brodkorbi, in which the posterior crest is offset medially and the groove is wider. The roughly triangular ex- cavation lying medial and distal to the hy- potarsus is short in M. carbo and M. chi- mango, and longer, extending farther down the shaft in M. chimachima, M. alexandri, and M. brodkorbi. Trochlea for digit 2 very wide and very excavated medially in pos- terior view, in this respect resembling the species of Phalcoboenus. Distal foramen high, more proximal on the shaft (similar to M. chimachima and M. alexandri, as op- posed to the more distal placement in M. chimango, and M. brodkorbi). Remarks.—The much greater diversity of the family Falconidae in the Quaternary of the West Indies than at present is again in- creased with the addition of Milvago carbo. This is the second fossil species of Milvago known in the Greater Antilles, and the third overall. Milvago brodkorbi Campbell (1979), from the Talara Tar Seeps, Peru, is larger than M. alexandri Olson (1976) of Hispaniola, but both would have been dwarfed by M. carbo. The Cuban species Milvago carbo and Falco kurochkini Suarez and Olson (2001), the latter of the subfamily Falconinae, agree in the marked elongation of the tarsometa- tarsus. Of the two living species of Milva- go, M. chimango, which ranges from south- ern Brazil and Chile south to Tierra del Fuego, has a long and slender tarsometatar- sus and occurs in open country, whereas M. chimachima, which ranges from Panama and Costa Rica southward east of the Andes to northern Argentina, has a shorter tarso- metatarsus and is more arboreal in habits (Vuilleumier 1970). From its size and the proportions of its tarsometatarsus, M. brod- korbi appears to be a trans-Andean repre- sentative of M. chimango. Milvago carbo would probably have been at least as terrestrial in habits as M. chimango or M. brodkorbi but was much larger. The total length of its tarsometatar- sus 1s between that of Caracara creightoni (smaller) and C. cheriway (larger), but be- 305 cause it is much more gracile, M. carbo probably took smaller prey than either. Genus Caracara Merrem, 1826 Caracara creightoni Brodkorb, 1959 (Biee 2; Table 2) Referred material.—Las Breas de San Felipe II, municipality of Marti, Matanzas: fragmentary anterior half of notarium MNHNCu P4579, 4 distal ends of left MNHNCu P4580-—83 and two distal ends of right tibiotarsi MNHNCu P4584-85, 6 dis- tal ends of night MNHNCu P4586—91 and two distal ends of left tarsometatarsi MNHNCu P4592-93. Cueva de Paredones, about 3 km SE of Ceiba del Agua, munic- ipality of Caimito, La Habana: complete left femur WS 1933. Cueva de Sandoval, about 4 km south of Vereda Nueva, munic- ipality of Caimito, La Habana: proximal end of right humerus WS 1035 and proxi- mal end of right femur WS 587. Comparisons with other species of Ca- racara.—The fragment of notarium agrees with Caracara in the less laterally com- pressed anterior vertebrae, unlike Milvago, in which these vertebrae are greatly com- pressed and thin anteriorly. It differs from C. plancus, C. cheriway, and C. lutosa in its relatively small size and by the lower position of the foveae costales. The proxi- mal end of the humerus is smaller with a capital groove that is thin, rather than wide as in Milvago. The femur is similar to that in the species of Caracara, but differs spe- cifically from C. plancus, C. cheriway, or C. lutosa in being smaller, with a very thin shaft at the midpoint and relatively wide proximal and distal ends (Table 2). The head of the femur is not reflected proxi- mally; the pneumatic foramen is large; the external condyle is thin and more vertical; and the intercondylar sulcus is wider. Thus this specimen agrees with two portions of femora recorded by Suarez and Arredondo (1997) as C. creightoni, but not further de- scribed by Suarez and Olson (2001) be- cause of their fragmentary condition. The additional tarsometatarsi of C. 306 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Left femora (A, B) and right humeri (C, D) of Caracara: A,.C. creightoni WS 1933; B, C. cheriway USNM 19670; C, C. creightoni WS 1035; D, C. cheriway USNM 19670. Scale = 2 cm. creightoni from Las Breas de San Felipe and more posteriorly rotated in Caracara indicate points of distinction not previously plancus, C. cheriway, and C. lutosa. This mentioned by Suarez and Olson (2001): the character is present in all specimens avail- trochlea for digit 2 is reduced and not ro- able and does not represent intraspecific tated posteriad, as opposed to being larger variation as we first supposed. Table 2.—Measurements (mm) of the humerus and femur in fossil and living species of Caracara. Sequence is: range (mean) n. Measurement C. creightoni C. cheriway C. plancus C. lutosa Humerus Depth of the head Dail 5.9-7.2 (6.7) 13 6.0-8.8 (7.1) 7 7.0 Least width of shaft ial Teco) eI) 12 7.49.8 (8.3) 7 8.4 Femur Length 65.6 68.4—76.8 (72.0) 12 66.1—80.8 (73.6) 7 Mitel: Proximal breadth 13.7, 14.4 13.2—15.2 (14.3) 12 13.2-18.5 (15.4) 7 14.8 Least width of shaft 6.2 6.8-8.2 (7.5) 12 6.9-8.8 (7.6) 7 TS Least depth of shaft 5).05) 6.2-7.4 (6.9) 12 5.9-8.4 (7.0) 7 7.0 Distal breadth NBe7 13.4-16.4 (15.0) 12 13.7—18.4 (15.7) 7 16.0 VOLUME 116, NUMBER 2 Remarks.—Remains of Caracara creigh- toni occurred together with Milvago carbo at Las Breas de San Felipe in a small area of less than 2 m?* (Suarez, pers. obs.). Al- though fossils of all known extinct Cuban scavenging birds occur at Las Breas, no re- mains of Caracara cheriway have been found there (Suarez, in prep.). The distal end of a left tarsometatarsus from a cave deposit in La Habana Province that Jiménez (1997) recorded as Caracara plancus (=cheriway), was re-examined (WS) and found to agree with C. creightoni instead. Possibly C. cheriway reached Cuba after Europeans introduced domestic animals that provided similar ecological conditions to those that may have been lost following the extinction of most of the large endemic mammals of the West Indies at the end of the Pleistocene and early in the Holocene. Acknowledgments Suarez’s research at the Smithsonian In- stitution was made possible by the Alex- ander Wetmore Endowment Fund of the Di- vision of Birds, National Museum of Nat- ural History. Photographs are by John Steiner, Smithsonian Office of Photographic Services, and composition of the figures is by Brian Schmidt, Division of Birds, Smithsonian Institution. Literature Cited American Ornithologists’ Union. 1998. Check-list of North American birds, 7th edition. American Ornithologists’ Union, Washington, D.C., 829 Pp: Baumel, J. J.. & L. Witmer. 1993. Osteologia. Pp. 45— 132 in J. J. Baumel, ed., Handbook of avian anatomy. Nomina Anatomica Avium, 2nd edi- tion—Publications of the Nuttall Ornithological Club 23. Cambridge, Massachusetts. Brodkorb, P. 1959a. The Pleistocene avifauna of Ar- redondo, Florida.—Bulletin of the Florida State Museum, Biological Sciences 4:269-291. . 1959b. Pleistocene birds from New Provi- dence Island, Bahamas.—Bulletin of the Florida State Museum, Biological Sciences 4:349—371. Campbell, K. E., Jr. 1979. The non-passerine Pleisto- cene avifauna of the Talara Tar Seeps, north- 307 western Peru.—Royal Ontario Museum, Life Sciences Contributions 118:1—203. . 1980. A review of the Rancholabrean avifau- na of the Itchetucknee River, Florida.—Natural History Museum of Los Angeles County, Con- tributions in Sciences 330:119—129. Emslhe, S. D. 1998. Avian community, climate, and sea-level changes in the Pilo-Pleistocene of the Florida Peninsula.—Ornithological Mono- graphs 50:1—113. Howard, H. 1929. The avifauna of the Emeryville Shellmound.—University of California Publi- cations in Zoology 32:301—394. Iturralde-Vinent, M., R. D. E. MacPhee, S. Diaz-Fran- co, R. Rojas-Consuegra, W. Suarez, & A. Lom- ba. 2000. Las Breas de San Felipe, a Quaternary asphalt seep near Marti (Matanzas Province, Cuba).—Caribbean Journal of Science 36:300— 33) Jiménez, V. O. 1997. Seis nuevos registros de aves fdsiles en Cuba.—E] Pitirre 10(2):49. Morgan, G. S. 1994. Late Quaternary fossil vertebrates from the Cayman Islands. Pp. 465—508 in M. A. Brunt and J. E. Davies, eds., The Cayman Islands: natural history and biogeography. Klu- wer Academic Publishers, Dordrecht, Nether- lands, 604 pp. Olson, S. L. 1976. A new species of Milvago from Hispaniola, with notes on other fossil caracaras from the West Indies (Aves: Falconidae).—Pro- ceedings of the Biological Society of Washing- ton 88:355—366. , & W. B. Hilgartner. 1982. Fossil and subfossil birds from the Bahamas. Pp. 22—56 in S. L. Ol- son, ed., Fossil vertebrates from the Baha- mas.—Smithsonian Contributions to Paleobiol- ogy 48. Raffaele, H. A., J. Wiley, O. Garrido, A. Keith, & J. Raffaele. 1998. A guide to the birds of the West Indies. Princeton University Press, Princeton, New Jersey, 511 pp. Stiles, EF G., & A. FE Skutch. 1989. A guide to the birds of Costa Rica. Cornell Univ. Press, Ithaca, New WOnkKk es) slap: Suarez, W., & O. Arredondo. 1997. Nuevas adiciones a la paleornitologia cubana.—E] Pitirre 10: 100—102. , & S. L. Olson. 2001. Further characterization of Caracara creightoni Brodkorb based on fos- sils from the Quaternary of Cuba (Aves: Fal- conidae).—Proceedings of the Biological Soci- ety of Washington 114:501—508. Vuilleumier, E 1970. Generic relations and speciation patterns in the caracaras (Aves:Falconidae).— Breviora 355:1—29. Wetmore, A. 1920. Five new species of birds from cave deposits in Porto Rico.—Proceedings of the Biological Society of Washington 33:77—82. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):308—316. 2003. A new species of penguin (Spheniscidae: Spheniscus) and other birds from the late Pliocene of Chile Steven D. Emslie and Carlos Guerra Correa (SDE) Department of Biological Sciences, University of North Carolina, Wilmington, North Carolina 28403, U.S.A., email: emslies@uncwil.edu; (CGC) Instituto de Investigaciones Oceanologicas, Universidad de Antofagasta, Casilla 170, Antofagasta, Chile, email: cguerra@uantof.cl Abstract.—We describe a new species of penguin, Spheniscus chilensis, from Cuenca del Tiburoén, late Pliocene, northern Chile. This species was found in association with a small species of cormorant, Phalacrocorax sp., and a cara- cara, Milvago sp., and is the first Pliocene penguin to be described from South America. Other vertebrates at this site include fish, sharks, and cetaceans. An extensive invertebrate fauna, including the late Pliocene muricid gastropod Herminespina mirabilis, also is present. The avifauna suggests a low diversity of seabirds existed in northern Chile from the late Pliocene to the present, unlike the much higher diversity found in Patagonia in the late Oligocene/early Miocene. The coastal region of southern Peru and northern Chile is well known for its mas- sive layers of marine shell beds exposed in cliffs. Geologic studies of these beds have provided considerable information on ma- rine faunas in the Miocene through late Pleistocene of this region. Previous inves- tigations in Chile have concentrated on the age and structure of the invertebrate faunas, and eustatic changes in sea level in corre- lation with tectonic uplift (Herm 1969, 1970; Tsuchi et al. 1988; Padilla & Elgueta 1992). Deposition of these beds occurred over a period of millions of years in coastal forearc basins, the eastern sections of which were later uplifted during crustal deforma- tions of the continental margin (Dunbar et al. 1990). Extensive and diverse deposits of late Pleistocene marine mollusks also occur throughout this region (Ortleib et al. 1994). In 1994, we visited an exposure of ma- rine shell beds on the Peninsula de Mejil- lones and approximately 14 km northwest of the Antofagasta airport and 6 km from the coast (Fig. 1). Here, over 10 m of ma- rine sediments are exposed on _ eroded slopes of valleys and hills that contain abundant marine mollusks. The upper lay- ers of these exposed sediments also contain numerous sharks’ teeth, fish bones, and fragmentary and complete bones of marine mammals and birds. This fossil exposure lies within the upper beds of the Caleta Herradura de Meyjillones Formation (23°21.453"S; 70°32.061”W), dating to the late Pliocene (Tsuchi et al. 1988). In addi- tion, the fossils were found in association with the muricid gastropod Herminespina mirabilis, which is only known from the late Pliocene of Peru and Chile (DeVries & Verne) IOD7). The area of concentration of the verte- brate fossils, known locally as Cuenca del Tiburon, is particularly rich in birds. Doz- ens of disarticulated bones of a single spe- cies of penguin (Spheniscidae) and a small cormorant (Phalacrocoracidae) are scattered on the surface, having eroded from exposed layers of shell beds nearby. Fossils of this penguin originally were found by one of us (CGC) at the locality in 1980. We visited the site twice in 1994 and once in 1997 to VOLUME 116, NUMBER 2 Peninsula de Mejillones Cerro Bandurria \ Cerro Moreno Antofagasta Pacific Ocean Jamz. Ihe collect additional material of this and other birds that are presented here. All the disarticulated penguin material appears to represent a new species of pen- guin that is referable to Spheniscus by the relatively long and slender mandibular shaft, humerus with relatively deep proxi- mal end and large pneumatic fossa that is weakly bipartite, proximal shaft slightly an- gled internally and not straight, and tarso- metatarus relatively short and broad with external proximal foramen equal or larger and placed slightly higher on shaft than in- ternal foramen. The fossil material most closely approaches the living S. humboldti and S. magellanicus in size and proportions, but differs from all species in this genus by distinct characters of the humerus and other postcranial elements. These comparisons indicate that the fossils represent a new spe- cies of penguin that is described herein; fos- sils of two other taxa of birds associated with the penguin material also are identi- fied. Cuenta del Tiburon 309 Map showing the location of Cuenca del Tiburé6n on the Peninsula de Mejillones, northern Chile. Materials and Methods Measurements and comparisons of Re- cent skeletons of penguins were completed in the collections of the Florida Museum of Natural History (FLMNH), Gainesville, and the American Museum of Natural His- tory (AMNH), New York. Fossil material was examined at the Museo de La Plata (MLP), Argentina, the Universite Claude Bernard Lyon 1, France, and at AMNH. Measurements were taken with Vernier cal- ipers to the nearest 0.1 mm and are self- descriptive. Measurements of the humerus include greatest length (GL), proximal breadth and depth (PB, PD), least breadth and depth of shaft (LBS, LDS), and distal breadth and depth (DB, DD). Terminology follows that of Howard (1929) and Simpson (1946). Fossil specimens reported here are housed at the Museo Geoldgico Prof. Hu- merto Fuenzalida V., Universidad Cat6lica del Norte (UCN), Antofagasta, Chile, or at 310 FLMNH where they are catalogued with University of Florida (UF) numbers. Systematic Paleontology Class Aves Order Sphenisciformes Family Spheniscidae Spheniscus chilensis, new species jae, 2 Holotype.—complete left humerus, UCN-1-130697 (Fig. 2A). Collected by Carlos Guerra Correa, 1980. Cast of spec- imen housed at FLMNH, UF 143300. Paratypes.—2 left mandibles missing ends, UF 144101, 144153; cervical verte- bra, UF 144102; 2 right scapulae, UF 143296—-143297; right coracoid, 144154 (Fig. 2B); four right coracoids missing ster- nal ends, UF 144124-144125, 144155-— 144156; shaft of right coracoid, UF 144103; two left humeri, UF 144104— 144105; 6 left humeri missing proximal ends, UF 144106, 144126, 144157—144159, 144171; 4 right humeri, UF 143295, 144107—144109; 3 right humeri missing proximal ends, UF 144127-144128, 144160; left radius, UF 144129; 6 right ra- dii, UF 143299 (Fig. 2C), 144110, 144130— 144133; 3 left ulnae, UF 144134, 144161- 144162; right ulna, UF 144163; 3 left car- pometacarpi, UF 143298 (Fig. 2C), 144111, 144135; 5 right carpometacarpi, UF 144112, 144136—-144138, 144164; wing phalanx, UF 144139; synsacrum, UF 144113; proximal end of synsacrum, UF 144147; left femur with ends damaged, UF 144165; proximal left femur, UF 144114; left femur missing proximal end, UF 144140; 3 distal halves left femora, UF 144115, 144141, 144148; distal end left fe- mur, UF 144166; right femur, UF 144116 (Fig. 2D); two right femora missing proxi- mal ends, UF 144117, 144149; 2 right fem- ora missing distal ends, UF 144118-— 144119; distal end right femur, UF 144167; shaft of right femur, UF 144142; two distal left tibiotarsi, UF 144150, 144168 (Fig. 2D); proximal end right tibiotarsus, UF PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 144169; distal shaft of right tibiotarsus, UF 144143; shaft of right tibiotarsus, UF 144170; right fibula, UF 144144; 3 left tar- sometatarsi with damaged proximal ends, UF 143293—-143294, 144120. Type locality and horizon.—Cuenca del Tiburon Fossil Locality, Peninsula de Me- jillones, Chile, late Pliocene (upper beds, Caleta Herradura de Mejillones Formation, Tsuchi et al. 1988). This is the only locality in which the fossil species occurs. Diagnosis.—Humerus with deep fossa at proximal anconal surface below head (fossa is Shallow in Spheniscus humboldti, S. de- mersus, and S. mendiculus, shallow to mod- erately deep in S. magellanicus), relatively smaller and narrower entepicondylar pro- cess (broad and rounded in all living Sphen- iscus), relatively slender shaft similar to S. magellanicus and S. demerus (more robust in S. humboldti; Table 1), and distal end with or without pneumatic fossa in distal view (no fossa present in all living Sphen- iscus). Tibiotarsus with relatively larger dis- tal foramina and broader distal external shaft than in all living Spheniscus. Tarso- metatarsus with shallow anterior grooves below proximal foramina (grooves deep in all Recent Spheniscus). The ulna, radius, carpometacarpus and femur of the fossil species show minor differences with the living species. Etymology.—Named after Chile, the country in which the fossil site is located. Description.—The fossil material most closely approaches the living S. humboldti and S. magellanicus in size and proportions, but differs from all species in this genus by distinct characters of the humerus and tar- sometatarsus. The fossil species is slightly smaller than Spheniscus humboldti and sim- ilar in size to S. magellanicus (Table 1). Only one other fossil species is known, S. predemersus from the early Pliocene of South Africa (Simpson 1971). The humerus of that species, however, is much longer (length, 84.7 mm) and more slender in the shaft (medial breadth, 8.6 mm) than S. chi- lensis (Table 1, plus see measurements in VOLUME 116, NUMBER 2 311 Fig. 2. Fossil specimens of Spheniscus chilensis from Cuenca del Tiburon: (A) holotype left humerus, UCN- 1-130697, in palmar (left) and anconal (right) view; (B) dorsal view of right coracoid, UF 144154; (C) anconal view of right radius (left), UF 144110, and internal view of left carpometacarpus (right), UF 144111; (D) posterior view of right femur, UF 144116 (left), and anterior view of distal left tibiotarsus; UF 144168. Scale bar = 1 cm. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON —_—___—_— rs 18 0°02 PS ETI = = 7 8SIrrl dn L’9 ear rs C1 = = = LSIrrl An = oa r9 al oF1 SCE = rOlrrl AN Sl r6l 79 STI = €07 OTL SOlprhl AN €9 Ll Sr Vol Vel st 769 LOIPrl AN vL E07 GS 611 THI L'0@ 789 60Irrl AN VL 0°61 9°¢ STI CSI E07 COL L690€1-I-NON sisuaplyo “§ €9-19 €LI-8'b1 CPOE 101-76 Purl 891-99] 8°8S-0'9S (€ = N) lO) == 69 Cl = O9I Ge C0 + 96 CO = Gili IO = LOI Ol SOLS snjnoipuau “§ 91-89 MED 19-T'r L'€1-8'01 [91-671 CIES OrL-€'L9 (GaN) N= il 61 + 0°02 alee AS Lae Sel ONG ze IY OC Eye in =e OL snuawap °s 8°L-8°9 807-761 Seo SOO 9°SI-8'€l l'1Z-6'81 0'SL-8°79 (= N) YO a Gil L'0 = 661 ZO = (OS Si0l= Oil 30 2s Si Cl Fe OS US = 1 SnoUD]JasDU “s CS €€7-T 07 yo-s'¢ LvI-971 89I1-S' 41 CEE IC WOE CieL @_ ND) r0 = 08 Il = Bie €0 ¥ 8's SO = rei 8:0 = 6SI EO 2 SK Sl ESSE upjoquny “s add sd] dd dd TO satoadg 312 ———=S=S—_——— SSS “‘sumAUOIOR JUSUTOINSBOUT 0} ADY JOJ 7X9} dG “ASULI ST MOI UIO}I0G pue UoneIAsp psepurys ~ UU ST MOI doy oy} ‘ANUS Yoo IO “sisuazIyD “§ 0} UOSTIedWIOD UI snosiuaydy yus09y JO LouNY Jo (WU) syUOWIOINSeaP—'] 2G] VOLUME 116, NUMBER 2 Simpson 1971, table 1). A large unde- scribed species of penguin close to Sphen- iscus is known from the late Miocene/Pli- ocene Pisco Formation, Peru (Cheneval 1993). Bones of this species were examined at the Universite Claude Bernard Lyon 1, France, and differ by their greater size, hu- merus with more strongly bipartite pneu- matic fossa (see Simpson 1946 for descrip- tion of this character) and deeper medial fossa, and broad and rounded entepicon- dylar processes. Another fossil penguin from the late Miocene to middle Pliocene Bahia Inglesa Formation in north-central Chile was referred to cf. Spheniscus sp. and described as very similar to S. humboldti, but at least 25% larger (Walsh & Hume 2001) and could not be referable to S. chi- lensis. Other fossil penguins that have been de- scribed from South America are restricted to the late Oligocene/early Miocene of Pa- tagonia where nine species and four genera are known from this region (Simpson 1946, 1972; Tonni 1980). Species in the genera Arthrodytes and Paraptenodytes are rela- tively large penguins with the largest (A. grandis) near the size of the modern Em- peror Penguin (Aptenodytes forsteri). In these genera, the humerus has a relatively straight shaft and, in Paraptenodytes, the tarsometatarsus is relatively long and slen- der. A partial associated skeleton of P. an- tarcticus (AMNH 3338) and several isolat- ed bones of P. robustus were examined at AMNH and found to differ from Sphenis- cus chilensis in the characters above and by their greater size and robustness. Two proximal humeri (AMNH 3341 and 3346) of Chubutodyptes biloculata differ from Spheniscus chilensis by their larger, more robust size, presence of a strongly bi- _ partite pneumatic fossa, and relatively lon- ger and deeper bicipital furrow. Numerous specimens of Palaeospheniscus patagoni- cus and P. wimani at AMNH and MLP ap- proach the size of S. chilensis, but are more robust and have a distinctly bipartite pneu- matic fossa on the humerus. The tarsometa- 313 tarsus of this genus also is relatively long and slender with the medial proximal fo- ramen greatly reduced or absent. Order Pelecaniformes Family Phalacrocoracidae Genus Phalacrocorax Brisson, 1760 Referred material.—right coracoid miss- ing ends, UF 144122; distal left humerus, UF 144151; proximal half left ulna, UF 144123; proximal end right ulna, UF 144152; left carpometacarpus with ends damaged, UF 144146; right carpometacar- pus, UF 144145. Measurements.—Ulnae, UF 144123 and 144152: PB) 93 and’ 94 mm: PD, 8-1 and 8.5 mm, respectively. Carpometacarpus, UF t44a145- Gie->0)lmme PByand 2D» 5-3.and 10.6 mm; LBS and LDS, 3.7 and 2.8 mm. UF 144146: GL, 48.3 mm; LBS and LDS, 3.5 and 2.8 mm. Discussion.—This material is from a small species of cormorant, near the size of the living Phalacrocorax brasilianus. It is possible that it represents an undescribed species, but the material is too fragmentary for systematic analysis with other fossil and living species. Another small cormorant is known from the late Miocene/Pliocene Pis- co Formation, Peru (Cheneval 1993). An ulna and carpometacarpus (both catalogued as AGL PPI 139) of this species were ex- amined at the Universite Claude Bernard Lyon 1, France, and found to be slightly larger and more robust than the Cuenca del Tiburon material. In addition, the proximal ulna (PB and PD, 10.0 and 9.3 mm) has a relatively broader impression of brachialis anticus and the carpometacarpus (GL, 51.8; PB yon aininewL BS andl DS. S:5-and oS mm) has a proximal end with a relatively deeper intercondylar fossa in anterior view compared to the Cuenca del Tibur6n ma- terial. A distal humerus from a larger cor- morant, near the size of P. bougainvilli, also is known from the Bahia Inglesa For- mation in north-central Chile (Walsh & Hume 2001). It was not examined here, but 314 Fig. 3. Distal left tarsometatarsus, UF 144121, of Milvago sp. Scale bar = 1 cm. is considered too large to represent the fos- sil species at Cuenca del Tiburon. Order Accipitriformes Family Falconidae Genus Milvago Spix, 1824 ices Referred material.—Distal left tarso- metatarsus missing external trochlea, UF 144121. Description.—The specimen is referable to Falconidae by the presence of two open- ings of the distal anterior foramen and by the morphology of the distal trochlea. It dif- fers from Falco in lacking a long posterior ridge extending distally from the hypotarsus to nearly %4 the length of the shaft and is most similar to Milvago and Polyborus in this feature. Polyborus and M. chimachima have only a single opening for the distal anterior foramen. The specimen is most similar to M. chimango in this feature and in size, but the fossil differs in having more robust trochleae similar to M. chimachima. UF 144121 also is similar to M. chimachi- ma is having only a single opening of the anterior distal foramen. It may represent an PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON undescribed species of Milvago, but addi- tional specimens are needed. Discussion.—Three fossil species have been described in this genus: Milvago read- ei, from the late Pleistocene of Florida (Brodkorb 1959; Campbell 1980) which has been synonymized with the living ™. chimachima by Emslie (1998), and M. al- exandri and M. brodkorbi from the late Pleistocene of Haiti (Olson 1976) and Peru (Campbell 1980), respectively. These last two species were not examined here, but M. brodkorbi is much larger than M. chima- chima (Campbell 1980) and probably does not represent the fossil species from Cuenca del Tiburon. Discussion Only one other marine Pliocene fossil lo- cality is known in Chile, also on the north- ern Coast. Walsh & Hume (2001) reported a marine avifauna within the Bahia Inglesa Formation, near Copiapo dating to the late Miocene to middle Pliocene based on as- sociated shark teeth, diatoms, and forami- niferans. This deposit contained the earliest records of a cormorant (Phalacrocorax sp.) and penguin (cf. Spheniscus sp.) in Chile, as well as several other species of seabirds, sharks, fish, reptiles, and marine mammals. Cuenca del Tibur6én is now the second ma- rine deposit that also contains fossil sharks, fish, marine mammals, and birds. The ver- tebrate fauna other than birds remains largely unstudied. The sharks represented at this site include Carcharodon carcharias, Carcharias sp., and Carcharhinus sp., plus mouth plates from unidentified Mylhobati- dae. In addition, there are at least two large species of bony fish (Osteichthyes), two small species of dolphin (Odontoceti), one small baleen whale (Balaenopteridae), and one sea lion (Otariidae) (G. Morgan, pers. comm.). Based on this fauna, as well as inverte- brate taxa, the Cuenca del Tibur6én deposit represents a nearshore marine community, including both sandy subtidal (from the VOLUME 116, NUMBER 2 presence of Perna, Glycymeris, and Cho- rus) and rocky intertidal zones (from the presence of Trochita and several barnacle genera). Most of the shells appear to have been transported a short distance before de- position occurred, based on leaching and abrasion plus the lack of paired valves (R. Portell, pers. comm.). The penguin remains also exhibit moderate wear and abrasion in- dicating that they are in a secondary site of deposition. The fossil record of penguins has been reviewed by Simpson (1946) and Fordyce and Jones (1990). Spheniscus chilensis is the first fossil penguin to be described from Chile and from the Pliocene of South America. One other fossil species within this genus was described by Simpson (1971) from the early Pliocene of South Af- rica. Olson (1983), however, reviewed this and three other taxa named by Simpson from this locality and concluded that they all belong to either one extinct genus or represent a primitive form of Spheniscus. In South America, at least nine fossil species have been described from the rich late Ol- igocene/early Miocene Patagonia Forma- tion of Argentina (see reviews by Simpson 1946 and Tonni 1980), and one from the early Miocene of the Pisco Formation in southern Peru (Cheneval 1993). The material from Cuenca del Tibur6n suggests that a low diversity of penguins occurred on the west coast of South Amer- ica, similar to today and unlike the earlier fossil record in Argentina and South Africa. This relative low diversity of penguins and other seabirds appears to have persisted along the Chile and Peruvian coasts for mil- lions of years. In Patagonia, penguin diver- sity has declined considerably since the ear- _ly Miocene as only one species (S. magel- lanicus) breeds in that region today, while only two species are found along the Chi- lean and Peruvian coasts (S. magellanicus and §. humboldti). Future investigations of fossil deposits in northern Chile are needed to further understand the paleoceanographic 315) conditions that prevailed during the Plio- Cene. Acknowledgments This research was supported by NSF Grant INT 96-04813. We thank J. Chene- val, C. Mourer-Chauviré, R. Hulbert, J. No- riega, C. Tambussi for their assistance in museum collections. W. Allmon, N. Guter- ez, D. Jones, G. Morgan, L. Ortleib, R. Por- tell, M. Sewolt, M. Stiger, and T. Verry pro- vided valuable assistance in the field. R. Portell provided identification of inverte- brate fossils, and G. Morgan for vertebrates, associated with the Cuenca del Tiburon fos- sil birds. B. Harmon provided the illustra- tion for Fig. 3. The paper was improved with comments from S. Walsh and C. Tam- bussi. Literature Cited Brodkorb, P. 1959. The Pleistocene avifauna of Arre- dondo, Florida.—Bulletin of the Florida State Museum Biological Sciences 4:269-291. Campbell, K. E., Jr. 1980. A review of the Ranchola- brean avifauna of the Itchtucknee River, Flori- da.—Natural History Museum of Los Angeles County Contributions in Science 330:119-129. Cheneval, J. 1993. L’avifaune Mio-Pliocéne de la For- mation Pisco (Pérou) étude préliminaire.—Doc- uments des Laboratoires de Géologie Lyon 125: 85-95. DeVries, T. J.. & G. J. Vermeij. 1997. Herminespina: new genus of Neogene muricid gastropod from Peru and Chile.—Journal of Paleontology 71: 610-615. Diaz, H. EF, & V. Markgraf. 1992. El Nifio—historical and paleoclimatic aspects of the Southern Os- cillation. Cambridge University Press, England, 475 pp. Dunbar, R. B., R. C. Marty, & P. A. Baker. 1990. Ce- nozoic marine sedimentation in the Sechura and Pisco basins, Peru.—Palaeogeography, Palaeo- climatology, Palaeoecology 77:235—261. Emslie, S. D. 1998. Avian community, climate, and sea-level changes in the Plio-Pleistocene of the Florida Peninsula.—Ornithological Mono- graphs 50:1—-113. Fordyce, R. Ewan, & C. M. Jones. 1990. Penguin his- tory and new fossil material from New Zealand. Pp. 417-446 in L. S. Davis and J. T. Darby, eds., Penguin Biology. Academic Press, New York. 316 Herm, D. 1969. Marines Pliozin in Nord- und Mittel- Chile unter besonderer Beriicksichtigung der Entwicklung der Mollusken-Faunen.—Zitteli- ana 2:1—158. Herm, D. 1970. Bostryx variabilis n. sp., eine Land- schnecke aus dem Altpleistozan von Meyjillones, Nordchile.—Mitteilungen Bayerische Staatssa- mmlung fiir Paléontologie und Historische Geo- logie 10:189—-198. Howard, H. 1929. The avifauna of Emeryville shell- mound.—University of California Publications in Zoology 32:301—394. Olson, S. 1976. A new species of Milvago from His- panola, with notes on other fossil caracaras from the West Indies (Aves: Falconidae).—Pro- ceedings of the Biological Society of Washing- ton 88:355—366. . 1983. Fossil seabirds and changing marine en- vironments in the late Tertiary of South Afri- ca.—South African Journal of Science 79:399— 402. Ortlieb, L., N. Guzman, & M. Candia. 1994. Moluscos litorales del Pleistoceno superior en el area de Antofagasta, Chile: primeras determinaciones e indicaciones paleooceanograficas.—Estudios Oceanologicos 13:57—63. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Padilla, H., & S. Elgueta. 1992. Neogene marine de- posits of Caleta Patillos, northern Chile: their relationship with Neogene sediments of the Peninsula de Meyjillones.—Revista Geoldgica de Chile 19:83-89. Simpson, G. G. 1946. Fossil Penguins.—Bulletin of the American Museum of Natural History 87: 1-99. . 1971. Fossil penguin from the late Cenozoic of South Africa.—Science 171:1144—1145. . 1972. Conspectus of Patagonian fossil pen- guins.—American Museum Novitates 2488:1— 371, Tonni, E. P. 1980. The present state of knowledge of the Cenozoic birds of Argentina. Natural His- tory Museum of Los Angeles County Contri- butions in Science 330:105—114. Tsuchi, R., T. Shuto, T. Takayama. A. Koizumi, M. Ibaraki, & R. Martinez-Pardo. 1988. Fundamen- tal data on Cenozoic biostratigraphy of Chile. Pp. 71—91 in R. Tsuchi, ed., Reports of Andean Studies, Spec, vol. 2. Shizuoka University, Ja- pan. Walsh, S. A., & J. P Hume. 2001. A new Neogene marine avian assemblage from north-central Chile.—Journal of Vertebrate Paleontology 21: 484-491. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):317-—329. 2003, Leptodactylus caatingae, a new species of frog from eastern Brazil (Amphibia: Anura: Leptodactylidae) W. Ronald Heyer and Flora A. Junca (WRH) Amphibians and Reptiles, MRC 162, PO. Box 37012, Smithsonian Institution, Washington, D.C. 20013-7012, U.S.A., e-mail: heyer.ron@nmnh.si.edu; (FAJ) Departamento de Ciéncias Biol6gicas, Laboratorio de Animais Pegonhentos e Herpetologia, Universidade Estadual de Feira de Santana, Campus Universitario—km 03 (BR 116), 44031-460 Feira de Santana, Bahia, Brazil, e-mail: fjunca@uefs.br Abstract.—A new species of frog of the genus Leptodactylus is described from eastern Brazil. The new species differs little morphologically from L. latinasus, but the advertisement calls are very different. The geographic dis- tribution of L. /atinasus and the new species, together with the respective levels of differentiation of morphology and call, is mirrored in the species pair L. bufonius and L. troglodytes. Presumably the same historical event or events lead to differentiation of these species pairs. In a revision of the Leptodactylus fuscus species group (Heyer 1978), Leptodactylus latinasus was characterized as having a ma- jor distribution pattern in Argentina, Uru- guay, and the State of Rio Grande do Sul, Brazil and a few disjunct localities in east- ern Brazil. The late Dr. Adaéo J. Cardoso recorded the advertisement call of a male L. latinasus from the State of Bahia in Brazil and brought the recording to the attention of WRH, pointing out that its call was very different from the other known recordings of L. latinasus from Argentina and Rio Grande do Sul, Brazil. More recently, FAJ collected a series of specimens of this east- ern Brazil form from the fossil sand dune region of Bahia. The purpose of this paper is to describe the eastern Brazilian popula- tion currently identified as L. latinasus as a new species and provide new field obser- vations for the taxon. Materials and Methods All specimens of the new species in the MZUSP, UEFS, and ZUEC collections (museum abbreviations follow Leviton et al. 1985 with UEFS being the Universidade Estadual de Feira de Santana) were bor- rowed and data were taken from them. Data for Leptodactylus latinasus are those used in Heyer (1978) with some additional data for eye-nostril distance, tympanum diame- ter, and belly patterns (data for these char- acters were not evaluated in the study pub- lished as Heyer 1978). Measurement data include snout—vent length (SVL), head length, head width, eye- nostril distance, thigh length, shank length, and foot length, following Heyer et al. (1990) except for the eye-nostril distance being measured by calipers as the distance from the anterior corner of the eye to mid- nostril and the tympanum diameter being measured by calipers as the maximum di- ameter of the tympanum including the an- nulus. Measurement data were analyzed using the software program SYSTAT 10 for prin- cipal component analysis (Stenson and Wil- kinson 2000) and discriminant function analysis (Engelman 2000). The advertisement call analyzed of the new species is USNM recording 234, cut 1, from Joazeiro, Fazenda Mary, Bahia, Bra- 318 zil, 7 March 1990, air temperature 26°C, 1910 h, voucher ZUEC 8833, by Adao J. Cardoso. All calls were recorded from the single voucher specimen. Call rate was based on eight samplings of the recording for a total of 396 calls. Other call parame- ters are evaluated from analysis of 10 calls. For comparative purposes, a recording of Leptodactylus latinasus, USNM recording 19, cut 1, from Embarcaci6n, Salta, Argen- tina, 22 December 1971, air temperature 22°C, time not noted, unvouchered, by W. Ronald Heyer, was analyzed. Call rate was based on three samplings of the recording for a total of 57 calls. Other call parameters are evaluated from analysis of seven calls (non-overlapping calls in a chorus) with fil- ter bandpass settings of 600 and 11,205 Hz. Advertisement calls were analyzed using Canary 1.2 software (Charif et al. 1995). The calls were digitized at a sample rate of 22,050 Hz, sample size of 16 bits. Call component terminology follows Duellman and Trueb (1986) and Heyer et al. (1990). Call duration was measured from the wave- form. Dominant frequency was determined using the spectrum analysis in Canary with settings of analysis resolution filter band- width 349.70 Hz, frame length 256 points, grid resolution time 128 points, overlap 50%, frequency 43.07 Hz, FFT size 512 points, window function hamming, ampli- tude logarithmic, clipping level —80 dB. Minimum and maximum call frequencies were measured from the audiospectrogram display (spectrogram in Canary terminolo- gy) using settings of analysis filter band- width 174.85 Hz, frame length 512 points, grid resolution 256 points, overlap 50%, frequency 43.07 Hz, FFT size 512 points, window function hamming, amplitude log- arithmic, clipping level —80 dB, display style smooth. Call amplitude modulation was evaluated from visual inspection of ex- panded waveform displays. Harmonics were determined using both expanded wave form displays and power spectrum analysis displays. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Advertisement Calls The call of the new species is given at an average rate of 160/min. The call dura- tion ranges from 64 to 77 ms. Each call has 7—8 distinct pulses (Fig. 1), given at a rate between 102 to 109 pulses/sec. The domi- nant frequency for the entire call has two peaks (Fig. 2), the slightly less loud peak ranges from 1072 to 1121 Hz, the loudest peak ranges from 1543 to 1648 Hz. The call is frequency modulated as rising from the beginning of the call to a maximum fre- quency at about *% the call duration, then slightly falling to the end of the call; the lowest dominant frequency ranges from 934 to 963 Hz, the highest dominant frequency ranges from 1543 to 1648 Hz. There are no indications of harmonic structure in the call. (Fig. 3). The call of the new species is dramati- cally different from the call of Leptodac- tylus latinasus in terms of amplitude struc- ture (Fig. 1, distinct pulses present in the new species, absent in L. latinasus) and fre- quencies (Fig. 3, Table 1). The call of the new species from Ibiraba, Brazil sounds very similar to Cardoso’s recording from Joazeiro, Brazil, and very different from the calls of Argentinian recordings of L. lati- nasus and L. fuscus, another species in the same species group that occurs in some of the same localities as the new species (FAJ, pers. obs.) The differences between the calls of the new species and L. latinasus convincingly demonstrate that they repre- sent two distinct species. Morphology The following variation occurs in the new species. Dorsal patterns include well-defined to weakly defined single or double mid-dorsal dark chevrons (as in Heyer 1978, Fig. 1A, C, p. 3), chaotically placed small to medi- um sized dark markings, or almost uniform. A very interrupted light or dark mid-dorsal pin stripe is present or absent. Lip stripes range from a light stripe well- VOLUME 116, NUMBER 2 a 0 smmercAn NN Nw | lcm le Brazil (b); scale bars = 0.02 sec. defined both above and below from the tip of the snout passing under the eye and con- tinuing under the tympanum onto the com- missural gland to indistinct or indiscernible. All of these conditions occur in specimens from Ibiraba, Bahia; all but one of nine specimens from Itiuba, Bahia have well-de- fined stripes. All but one specimen were scored as hav- ing a distinct light stripe on the lower face of each posterior thigh surrounded by irreg- ularly shaped broad dark outlines with the rest of posterior thigh having a relatively uniform scattering of melanophores to a mottled pattern. One specimen from Ibira- ba, Bahia has a series of light spots, rather than a continuous light stripe. The upper shanks have irregularly shaped dark transverse bands. Bi9 Wave form of advertisement call of Leptodactylus latinasus (a) and the new species from eastern The specimens from Ibiraba, Bahia have a greater development of belly pattern than observed from other localities. The Ibiraba specimens range from having an almost uniform distribution of densely packed me- lanophores (pale to the eye, but melano- phores distinct under magnification), a var- iegated pattern of melanophores, to groups of melanophores scattered over the belly. Specimens from the other localities either lack a belly pattern, have a weakly varie- gated pattern of melanophores, or have only a few scattered melanophores lateralmost on the belly. Many individuals lack any indication of dorsolateral folds. Several individuals have very interrupted folds from the eye to the sacrum. One individual has a very inter- rupted fold from the eye to the groin. Most 320 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON -50 ee an | | -60 | | Intensity in dB 0 1 z 3 4 5 Frequency in kHz Fig. 2. Power spectrum of advertisement call of new species of Leptodactylus from eastern Brazil. Frequency in kHz 0) 1 gee NG 1 2 Time in seconds Fig. 3. Audiospectrograms of advertisement calls of Leptodactylus latinasus (a) and the new species from eastern Brazil (b). VOLUME 116, NUMBER 2 Table 1.—Advertisement call parameters for Leptodactylus latinasus and L. latinasus-like specimens from Argentina and Brazil. * indicates data from Barrio (1965). Call data are average values. Maximum dominant frequency, Hz Minimum dominant frequency, Hz Dominant Call Call frequency in celsius rate/min duration/s Hz, entire call Temperature, Country/state or province 3000 3500 3274 0.08! 0.09! 0.07 30 13 145 26.0 Argentina, Buenos Aires* Argentina, Chaco* Argentina, Salta 20.0 3782 3334 22.0 Brazil, Bahia (new spe- 1616 943 1423 0.07 160 26.0 cies) ' Barrio (1965) gave these values as 0.8 and 0.9 s, but his audiospectrograms indicate that the values should be hundredths, not tenths of seconds. a2) individuals have complete to broken lateral folds. The upper tibia has scattered to many distinct white tubercles. Twenty of 21 spec- imens have many distinct white tubercles on the outer tarsal surface and sole of foot. One individual has scattered distinct white tubercles on the outer tarsal surface. Males range from 32.1 to 36.9 mm SVL (n = 8), females 36.2 and 39.1 mm (n = J). The dorsal pattern, lip stripes, posterior thigh pattern, upper shank pattern, mid-dor- sal pin stripes, upper shank texture, outer tarsal texture, and sole of foot texture traits are indistinguishable between the new spe- cies and L. latinasus. Belly pattern data were not taken for the Heyer (1978) study. WRH examined bellies of all USNM specimens of L. latinasus from Argentina, Uruguay, and Rio Grande do Sul, Brazil. Forty-nine individuals lack any belly pattern. One specimen from Ar- gentina (USNM 319587) and two speci- mens from Uruguay (USNM 535968, 539970) have a few melanophores on the lateralmost venter and/or in the arm inser- tion portion only of the venter. One indi- vidual from Uruguay (USNM 535969) has a pair of small blotches just behind the arm on the sides of the chest. None of the Ar- gentina, Uruguay, or Rio Grande do Sul, Brazil specimens examined have as exten- sive venter patterns as found in some indi- viduals of the new species. Body fold development was not scored in the same way for the Heyer (1978) study and this study. A series of 20 well-pre- served adult Leptodactylus latinasus from Salta Province, Argentina examined for this study shows the same expressions of lateral folds as in the new species, but with dif- ferent frequencies of occurrences of indi- vidual states. Only one of the 20 L. lati- nasus has a very broken dorsolateral fold from the eye to % distance to the sacrum; the other 19 show no indications of dorso- lateral folds. Thus, the new species appar- ently has a greater frequency of any dor- 322 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON solateral fold expression than occurs in L. 5 latinasus. i a: Sota There are two sets of measurement data. 5 eS A more extensive data set (in terms of num- I oP ber of individuals) is available for all adults A geicp aaa a examined for the Heyer (1978) study for ic) Es SVL, head length, head width, thigh length, 3 S shank length, and foot length. Additional 3 a IS NEES ws data for eye-nostril distance and tympanum a diameter are available for a smaller number of adults for both the new species and Lep- Z as} eas fs todactylus latinasus. g oo oa Due to the small number of adults avail- a able for the new species, principal compo- Z Srey Ge nent analyses are the primary statistical tool = aloge se used to explore measurement variation in ‘ adults of the new species and Leptodactylus = alia we latinasus (Table 2). The principal compo- ao El ee See nent analyses for males show a greater dis- bes tinction between the two species using the 2 5 data for all variables versus the data set 2 3 alm a lacking eye-nostril distance and tympanum 6 j 3 as diameter data (Fig. 4) than observed in the 5, 3 j female data (Fig. 5). The modest differen- cae a z/4 © tiation based on measurement data is ex- A € =o pected to exceed researcher measurement 4 5 error, based on previous experience (Hayek fe = -|Ge2 @ et al. 2001). Discriminant function analyses a 158 © for male data including eye-nostril distance § i and tympanum diameter data result in ae ere he 100% correct posterior classification for EH fs 2 oe both the complete data model and backward x & stepwise model. The discriminant function S = analyses for the male data lacking eye-nos- 3 I 5 ean tril distance and tympanum diameter result E = a ae in 97-98% correct posterior classification 3 Il ‘ of cases for both the complete model and g@ £2 Men a backward stepwise models. Results of mul- Ee Za a tivariate techniques can not be assumed to z 8 be generalizable beyond the individual data 5 = sets analyzed because any changes in the a3 © : ‘ ‘ iS & ans sample will yield different results to at least 3 2 aa & some degree (James and McCulloch 1990). aS E as E 3 Given this caveat, we interpret our results 25 4 2 2 5 2 to mean that there is at least some morpho- & I AN cs a 2 logical differentiation between the new spe- t = < Z ie gs Z cies and L. latinasus that would be con- 29 ae So 6 firmed with increased sample sizes. ic E 22 E E E Overall, the new species and Leptodac- ss] SS ie VOLUME 116, NUMBER 2 323 a b Principal Component 2 Principal Component 2 sot se ta 0 1 2 3 4 -4 2 0 2 4 6 8 Principal Component 1 Principal Component 1 Fig. 4. Male measurement data principal component analyses. A. Reduced variable data set. B. Variable set including eye-nostril distance and tympanum diameter data. Circles = Leptodactylus latinasus, filled triangles = new species. Principal Component 2 Principal Component 2 -3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 Principal Component 1 Principal Component 1 Fig. 5. Female measurement data principal component analyses. A. Reduced variable data set. B. Variable set including eye-nostril distance and tympanum diameter data. Circles = Leptodactylus latinasus, triangles = new species. 324 tylus latinasus are very similar morpholog- ically. There are no characters that com- pletely diagnose the two species. There are a few character states that demonstrate some differentiation between the species. If the only data available were the morpho- logical data, we would interpret the varia- tion to be within-species and would recog- nize only one species for the two taxa, as was done in the Heyer (1978) study. Given that the advertisement call data in- dicate that the specimens from eastern Bra- zil represent a species distinct from Lepto- dactylus latinasus and there is no name available for it, we hereby describe the tax- on lacking a name as: Leptodactylus caatingae, new species Figs. 6—7 Holotype.—ZUEC 8833, an adult male from Brazil, Bahia, Joazeiro, 09°25’S, 40°30'W. Collected by Adao J. Cardoso, 7 March 1990. Referred specimens.—BRAZIL; BAHIA; Bom Jesus da Lapa, 13°15’S, 43°25'W, UMMZ 109991(2); Ibiraba, 10°48’S, 42°50'W, UEFS 686, 688-695; Ititba, 10°42’S, 39°51’W, MZUSP 38556-38559, 38561-38563, USNM 547844-547845; Sao José do Rio Grande, 11°49’S, 44°44’ W, UMMZ 109992; ESPIRITO SANTO; Sao Mateus, 18°44’S, 39°51’W, MCZ A-92142 (recatalogued from the previously lot cata- logued specimen MCZ A-1298 cited in Heyer 1978); PERNAMBUCO; Exu, T°31'S, 39°43'W, MZUSP 51858; Ouricuri, 7°53'S, 40°05’W, MZUSP 77749. Diagnosis.—The species with a combi- nation of a distinct light stripe on the pos- terior surface of the thigh and obvious white tubercles on the outer surface of the tarsus and sole of foot in some or all indi- viduals are Leptodactylus albilabris, caatin- gae, elenae, fragilis, latinasus, and mysta- ceus. Leptodactylus albilabris and mysta- ceus have distinct dorsolateral folds (indi- cated by color pattern in poorly preserved specimens); L. caatingae has interrupted, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON indistinct dorsolateral folds or lacks them. Leptodactylus caatingae, fragilis, and lati- nasus have considerable morphological and color pattern overlap and cannot be consis- tently diagnosed from each other with these characters. The advertisement call of L. la- tinasus is not pulsed and has a high broad- cast frequency (3000-3780 Hz); the call of L. caatingae is pulsed and has a lower broadcast frequency (940-1620 Hz). The advertisement call of L. fragilis is longer (0.19 sec) than the call of L. caatingae (0.07 sec). Leptodactylus caatingae, fragil- is, and latinasus have allopatric distribu- tions: L. fragilis from southernmost Texas, United States to north coastal Venezuela; L. caatingae in eastern Brazil; and L. latinasus in southern South America (Argentina, Bo- livia, Rio Grande do Sul, Brazil, Uruguay). Description of holotype.—Snout rounded from above, acutely rounded in profile; can- thus rostralis indistinct; lores obtusely con- vex in cross section; tympanum well de- fined, moderate size, horizontal diameter about % eye diameter; vocal slits elongate, parallel to lower jaw, starting about mid- tongue level; vocal sac single with distinct lateral expansions indicated by pronounced folds under lower jaws; vomerine teeth in almost straight line patches separated from each other by about % length of a single tooth row, well behind and between almost round choanae; finger lengths II ~ IV 120 mm) of Amblyotrypauchen can have the pelvic fins more deeply incised than shown in the figure. VOLUME 116, NUMBER 2 Table 1.—Proportional measurements of Amblyotrypauchen arctocephalus. n Mean Range Standard length/TL 6 0.845 0.833—0.858 Head length/SL it 0.201 0.179-0.214 Pelvic-fin length (PEL)/SL 11 0.085 0.062-0.111 Pelvic-fin length/HL 11 0.421 0.320—0.550 Pectoral-fin length/SL 19 0.064 0.042—0.098 Pectoral-fin length/HL 1S 0.314 0.206—0.457 Pectoral-fin length/PEL 19 0.764 0.431-1.137 Head width/SL it 0.094 0.025—0.113 Snout length/SL 11 0.058 0.041—0.115 Jaw length/SL 11 0.068 0.056—0.076 Interorbital width/SL 11 0.031 0.023—0.039 Nape width/SL 1] 0.074 0.065—0.083 Body depth/SL 11 0.141 0.124—0.152 Predorsal length/SL 11 0.241 0.183—-0.261 Prepelvic length/SL 11 OnI92 0.174—0.205 Preanal length/SL 1] 0.406 0.370-0.433 shaped with dorsal rays longer than ventral ones; all pectoral-fin rays segmented, some of the longer rays are branched distally. Pel- vic-fin rays I, 5; frenum present; basal membrane uniting fins present for approx- imately one-third to one-half the length of innermost rays; smaller specimens typically have less deeply divided pelvic fins than larger specimens. Caudal fin with 17 seg- mented rays including 8+7 branched rays and a dorsal and ventral simple ray; unseg- mented procurrent rays 4, dorsally and ven- trally. Scales cycloid, present on entire body with scale patches on head, cheek, and op- ercle, largest scales near caudal-fin base. Scales on head, cheek, and opercle present in patches of as few as two or three to as many as nine or 10 scales, the largest scale patch often posterior to the orbit. Scales on body extending from areas dorsal and ven- tral to opercle posteriorly to caudal-fin base. Longitudinal scale count 60-77, scales dif- ficult to count with accuracy. Typically, two lateral rows of teeth in each jaw, more than two rows anteriorly; outer-row teeth much larger and more pointed than those of inner rows; lower-jaw teeth longer than upper-jaw teeth; 2—5 fang- like teeth in outer row of upper jaw, typi- cally interlocking with those of lower jaw; numerous conical teeth in inner row(s) of upper jaw; 4—7 fang-like teeth in outer row of lower jaw; numerous conical teeth in in- ner row(s) of lower jaw. No palatine or vo- merine teeth present. No sexual dimor- phism with respect to teeth. Tongue thick, tip rounded, free from floor of mouth. Gape wide, mouth oblique (about 45°); maxilla extending posteriorly to vertical below anterior half of orbit; pos- teriorly, near tip of maxilla, upper lip ex- Table 2.—Selected counts of Amblyotrypauchen arctocephalus. Character n Mean Dorsal-fin rays (total elements) hs) 49.7 Anal-fin rays (total elements) II) 40.1 Pectoral-fin rays 2) 16.2 Longitudinal scale rows 16 69.5 Upper jaw teeth (outer row) 10 3.6 Lower jaw teeth (outer row) 10 4.5 Frequencies 48(1), 49(6), 50(4), 52(2) 39(4), 40(6), 41(2), 43(1) 14(2), 15(5), 16(5), 17(5), 18(4) 60(3), 67(2), 68(1), 70(3), 71(1), 72(1), 74(2), 76(2), 771) 2X2) 3.) (O) DL) AC) @), 7) 334 panded into large fold that joins similar fold of lower lip at rictus, fold completely cov- ering posterior part of jaws even when agape. No barbels on underside of head. Eye absent. Posterior naris large and lo- cated in cavity where eye normally would be located; anterior naris at tip of small tube-like flap that slightly overhangs upper jaw. Cephalic sensory canals and pores ab- sent. Head papillae numerous, found on dorsum, cheek, opercle, and along lower jaw line. Papillae grouped in short series (<1 mm) that appear as whitish lines. Shallow pouch present along the dorsal edge of the operculum. Gill rakers very short, pyramidal-shaped, and not ossified; six or fewer on lower limb of first gill arch. Gill opening narrow, extending only the length of pectoral-fin base or slightly more ventrally. Genital papilla large and bulbous in fe- males, bilobed in gravid specimens. Male papilla fleshy and triangular with distinct tip. Coloration.—No fresh specimens were available for this study. Based on Alcock’s original description, this species is mottled pink with hyaline fins. In preserved mate- rial, dorsum of head and body dark brown to gray, remainder of head and body uni- formly pale or pale brown; no spots or stripes on body or fins; fins translucent; fang-like teeth on outer rows of both jaws often reddish brown in larger specimens. Distribution.—Northeast coast of India eastward to Hong Kong and the Philip- pines, southward to Papua New Guinea and off the north coast of Australia. Reports of specimens from the Arabian Sea and Gulf of Martaban (Myanmar) by Koumans (1941) are unconfirmed. Ecology.—This species inhabits muddy bottoms from almost 17 m to over 183 m in depth; for both amblyopines and gobies in general, these are uncharacteristic depths. None of the material examined was col- lected in typical amblyopine habitats, which are shallow, mud-bottomed areas near river PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON mouths. The type material of Amblyopus arctocephalus Alcock (1890) was collected from 37—92 m, whereas the type of Ambly- otrypauchen fraseri was collected from about 37 m (Hora 1924). The fang-like teeth suggest that this species is carnivo- rous, and a radiograph of one specimen (AMS 1.16753-024) revealed the presence of an elongate fish in the stomach. Remarks.—The genus Amblyopus (Va- lenciennes in Cuvier & Valenciennes, 1837: 157) is regarded as an unneeded substitute for Taenioides Lacepéde (1800) and is thus considered an objective synonym of Tae- nioides (Eschmeyer, 1998). Taenioides is diagnosed among Gobiidae by its posses- sion of a Y-shaped, second anal-fin ptery- giophore (Birdsong et al. 1988); Amblyotry- pauchen lacks this feature. After examining the types of both spe- cies, Koumans (1941) synonymized A. fras- eri with A. arctocephalus. Similar judg- ments were rendered by Smith (1959) and Menon and Yazdani (1968). Although wanting to examine the types of A. arcto- cephalus and A. fraseri, or at least obtain radiographs of them, I was unable to do so. I also was unable to determine the current status of Alcock’s and Hora’s types at the Zoological Survey of India. Regardless, there appears to be little reason to dispute the synonymy of A. fraseri with A. arcto- cephalus based on comparison of the orig- inal descriptions. When Hora (1924) de- scribed his new genus and species, he made no mention of Alcock’s species, which had been collected in the same general vicinity as Hora’s species (NW corner of the Bay of Bengal); I assume Hora was unaware of the existence of Alcock’s description. Two items in Alcock (1890) need men- tion. First, Alcock described a “‘short broad barbel”’ on each side of the upper jaw; this structure is a narial tube rather than a bar- bel. Secondly, Alcock stated that A. arcto- cephalus possesses 11 abdominal and 17 caudal vertebrae. Whereas it is possible that Alcock was not mistaken, in this study, I radiographed 13 specimens and all but one VOLUME 116, NUMBER 2 have 10 precaudal and 19 caudal vertebrae. The atypical specimen has 9 precaudal and 23 caudal vertebrae. Several items in Hora (1924) also require discussion. First, Hora stated the dorsal-fin ray count of his single specimen as **7/40.” Fin rays of amblyopines are very difficult to count without radiography. As none of the radiographed specimens in this study possess seven dorsal-fin spines, I assume Hora’s count was in error. Second, Hora mentioned in the text that the dorsal and anal fins were “separated from the caudal by a short distance.”’ However, Alcock (1890) stated that in his specimens, the dor- sal and anal fins were confluent with the caudal fin; Koumans (1941) made a similar statement about the specimens he exam- ined. In amblyopines, the membrane uniting the dorsal and anal fins with the caudal fins is thin and easily torn in handling. Some of the specimens examined in this study had torn membranes, and I assume that this also occurred to Hora’s specimen. Lastly, Hora’s figure of A. fraseri (reproduced here as Fig. 1) not only depicts the dorsal and anal fins not connected with the caudal fin, but also shows a cheek more fully scaled than on any specimen examined in this study. Pos- sibly head scales slough off over time or with frequent handling such that older pre- served material, like that examined in this study, does not accurately reflect the degree of head scalation found in fresher speci- mens. Alcock (1890) mentioned that the head was naked in his specimens and Kou- mans (1941) described the specimens he examined as having “‘some scales on head behind eye and on cheek and opercle.”’ The degree of scalation of the head of A. arc- tocephalus can probably only be adequately assessed by examining fresh material. Comparison of Amblyotrypauchen with other “Trypauchen’ group members.— Based on their shared absence of an inter- neural gap, Birdsong et al. (1988) created the monophyletic unit called the ‘Trypauch- en’ group that comprised Amblyotrypauch- en, Caragobius, Trypauchen, and Try- 333) pauchenichthys. Murdy (2002) added Cten- otrypauchen and a new genus (Karsten) to the “Irypauchen’ group and provided a key to the “Irypauchen’ group genera. Cara- gobius and Karsten lack an opercular pouch whereas the other “Irypauchen’ group gen- era have one. Hora (1924) distinguished Amblyotrypauchen from Ctenotrypauchen, Trypauchen, and Trypauchenichthys by its possession of canine teeth, as well as by the shape of the pelvic fins (Fig. 2). As a result of this investigation, I can further differ- entiate Amblyotrypauchen from these three genera by its possession of head scales (lacking in the other genera) and 10+19 vertebrae (all other genera typically have 20 or more caudal vertebrae). Murdy (2002) compared Amblyotrypauchen to Karsten, the only other blind amblyopine. Karsten differs from Amblyotrypauchen in having a typical spinous dorsal-fin pterygiophore formula of 3-123 or 3-132 (vs. 3-1221 in Amblyotrypauchen), possessing nine pre- caudal vertebrae (vs. 10 in Amblyotry- pauchen), degree of scalation (scaled only posteriorly in Karsten vs. scaled on entire body including scale patches on the head in Amblyotrypauchen), in presence/absence of an opercular pouch (absent in Karsten vs. present in Amblyotrypauchen), and tooth size (slightly enlarged teeth in outer row of jaws vs. very large and fang-like in Ambly- otrypauchen). Acknowledgments Numerous individuals aided this study in the loan and exchange of specimens, col- lections, habitat data, or in other diverse ways: D. Catania, C. Ferraris, S. Jewett, S. Raredon, K. Shibukawa, and S. Smith. For their contributions I am grateful. Literature Cited Alcock, A. W. 1890. Natural history notes from H. M. Indian marine survey steamer /nvestigator, Commander R. E Hoskyn, R. N., command- ing.—No. 20. On some undescribed shore-fish- es from the Bay of Bengal.—Annals and Mag- azine of Natural History 6: 425—443. 336 Birdsong, R. S., E. O. Murdy, & FE L. Pezold. 1988. A study of the vertebral column and median fin osteology in gobioid fishes with comments on gobioid relationships.—Bulletin of Marine Sci- ence 42:174—214. Cuvier, G., & A. Valenciennes. 1837. Histoire naturelle des poissons. Tome douzieme. Suite du livre quatorzi¢me. Gobioides. Acanthoptérygiens a pectorales pédiculées. Chez E G. Lerrault, Paris, 507 pp. Eschmeyer, W. N. (ed.). 1998. Catalog of fishes. Spe- cial publication No. | of the Center for Biodi- versity Research and Information.—California Academy of Sciences 1—3:1—2905. Herre, A. W. C. T., & E. S. Herald. 1951. Noteworthy additions to the Philippine fish fauna with de- scriptions of a new genus and species.—Phil- ippine Journal of Science 79:309—340. Hora, S. L. 1924. Notes on fishes in the Indian Mu- seum. VI. On a new genus of gobioid fishes (subfamily Trypaucheninae) with notes on re- lated forms.—Records of the Indian Museum 26:155—163. Koumans, FE P. 1941. Gobioid fishes of India.>—Mem- oirs of the Indian Museum 13:205—329. Livre quinziéme. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Lacepede, B. G. E. 1800. Histoire naturelle des pois- sons, vol. 2. Chez Plassan, Paris, 632 pp. Vevaton;9Ay E. Ree Gibbs; Jin E> Healy cc @ aes Dawson. 1985. Standards in herpetology and ichthyology, part 1. Standard symbolic codes for institutional resource collections in herpe- tology and ichthyology.—Copeia 1985:802— 832. Menon, A. G. K., & G. M. Yazdani. 1968. Catalogue of type-specimens in the Zoological Survey of India, part 2.—Fishes.—Records of the Zoolog- ical Survey of India, 61 (pts 1-2) [1963]:91— 190. [Not published until 1968.] Murdy, E. O. 2002. Karsten, a new genus of eel goby (Gobiidae: Amblyopinae) with a key to ‘Try- pauchen’ group genera.—Copeia 2002(3):787— TAL. , & K. Shibukawa. 2001. A revision of the go- biid fish genus Odontamblyopus (Gobiidae: Amblyopinae).—Ichthyological Research, 48(2001)1:31—43. Smith, J. L. B. 1959. Gobioid fishes of the families Gobiidae, Periophthalmidae, Trypauchenidae, Taenioididae and Kraemeriidae of the western Indian Ocean,—Rhodes University, Department of Ichthyology, Ichthyological Bulletin, 13: 185-225, pls. 9-13. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):337—340. 2003. The hyopalatine arch of a 25 mm larva of Synbranchus and homology of the single pterygoid in the Synbranchidae (Teleostei: Synbranchiformes) Ralf Britz, Sandra Favorito, and G. David Johnson (RB) Division Of Fishes, National Museum of Natural History, Smithsonian Institution, Washington D.C. 20560, U.S.A., and Lehrstuhl fiir Spezielle Zoologie, Universitat Tiibingen, Auf der Morgenstelle 28, 72076 Tiibingen, Germany, e-mail: britz.ralf@nmnh.si.edu; (SF-A) Universidade de Sao Paulo, Museu de Zoologia, Caixa Postal 42594, Sao Paulo, SP, 04299-970, Brazil, e-mail: favorito@uniban.br; (GDJ) Division Of Fishes, National Museum of Natural History, Smithsonian Institution, Washington D.C. 20560, U.S.A., e-mail: johnson.dave @nmnh.si.edu Abstract.—ldentity of the single pterygoid bone in Synbranchidae is eval- uated based on a 25 mm larva of Synbranchus. Two hypotheses have been proposed as to its homology: it is 1) the endopterygoid or 2) the ectopterygoid. We show that the bone in question develops in the position of the ectopterygoid and therefore represents the homologue of this bone in other teleosts. Thus synbranchids lack the endopterygoid, an observation that invalidates a previ- ously proposed synapomorphy of this family and the channids (snakeheads). Synbranchidae, or swamp eels, are a family of highly derived eel-like acantho- morphs, comprising 17 species (Bailey & Gans 1998) from fresh and estuarine waters of Middle and South America, Cuba, West Africa, Asia, and the Indo-Australian Ar- chipelago (Nelson 1994). Several species are well known for their amphibious habits and the presence of accessory air breathing organs (see e.g., Rosen & Greenwood 1976, Liem 1987, Munshi et al. 1989) that enable them to undertake extensive overland ex- cursions. There are two hypotheses about the re- lationships of the Synbranchidae to other acanthomorph taxa: 1) synbranchids are the sister group of channids (Lauder & Liem 1983) and 2) synbranchids are the sister group of mastacembeloids (= mastacem- belids plus chaudhuriids) (Travers 1984a, Johnson & Patterson 1993, Britz & Kottelat 2003). One of the characters cited as support for Lauder & Liem’s (1983) hypothesis is the presence of an enlarged endopterygoid in both, channids and synbranchids. The sin- gle pterygoid of synbranchids was consid- ered an ectopterygoid by Regan (1912), Rastogi (1964), Rosen and Greenwood (1976), Gosline (1983), Travers (1984), and Britz (1996), and an endopterygoid by Lau- der & Liem (1983). However, none of these authors specifically addressed the problem of the homology of this bone with respect to the ectopterygoid or endopterygoid of other teleosts. In the present paper we de- scribe the hyopalatine arch of a 25 mm lar- val specimen of Synbranchus sp., to resolve the identity of the synbranchid pterygoid. Material and Methods A cleared and double stained larval Syn- branchus sp. (USNM 372713) of 25 mm total length was studied. A Zeiss Tessovar was used to photograph the specimen. Ad- ditional comparative material comprised: Mastacembelidae; all cleared and stained: Mastacembelus erythrotaenia: AMNH 42129 (1, 277 mm); Mastacembelus sp. (as 338 Macrognathus aculeatus) AMNH 097654 (1, 158 mm); Macrognathus pancalus: AMNH 217414 (8, 4.5—36 mm). Synbranchidae; cleared and stained: Ma- crotrema caligans: MCZ 47107 (2, 172- 178 mm), Ophisternon aenigmaticum: AMNH 31573 (1, 72 mm); Synbranchus marmoratus: AMNH 30213 (1, 142 mm), AMNH 74541 (1, 47 mm), MCZ 52376 (3, 65-140 mm; 1, disarticulated); Monopterus albus: AMNH 41579 (1, 167 mm); dry skeletons: Ophisternon aenigmaticum (as Synbranchus marmoratus): USNM 111347 (1, partial skeleton); Monopterus sp. (as Synbranchus bengalensis): AMNH 220023 (1, ca. 550 mm). Terminology for the cartilaginous parts of the hyopalatine arch follows Arratia & Schultze (1991). Results The 25 mm larva of Synbranchus sp. still has a large yolk sac and prominent pectoral fins. The hyopalatine arch is largely carti- laginous (Fig. 1A). The hyosymplectic car- tilage articulates with the otic capsule of the chondrocranium. In the area around the fo- ramen for the hyomandibular branch of the facialis, there is a perichondral ossification, the hyomandibular, which bears a conspic- uous process of membrane bone that ex- tends ventrally between the body of the car- tilage and the pars metapterygoidea of the palatoquadrate. A thin perichondral ossifi- cation, the symplectic, surrounds the anter- oventral process of the hyosymplectic car- tilage. The opercle, which articulates with a posterior process of the hyosymplectic cartilage, and the remaining three opercular bones are present as thin platelets of bone. The palatoquadrate comprises two uncon- nected parts, the posterior pars quadrata et metapterygoidea and the anterior pars au- topalatina (Fig. 1A). The former is a rough- ly triangular cartilage, the ventral tip of which articulates with the lower jaw. Around this articulation and the lower third of the pars quadrata et metapterygoidea a PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON perichondral ossification is present, the de- veloping quadrate, with the usual poster- oventral process of membrane bone. The developing metapterygpoid is present as a thin lamina of perichondral bone surround- ing the posterodorsal corner of the pars quadrata et metapterygoidea. The pterygoid extends anteriorly as an elongate thin lam- ina of bone ventral to the anterodorsal cor- ner of the pars quadrata (Fig. 1A, B). The elongate cartilage of the pars autopalatina sits more anteriorly in the roof of the mouth and bears a long anterolaterally directed process, the distal tip of which articulates with the lacrimal. Ventral to this cartilage is a small splint of bone, the developing dermopalatine (Fig. 1A). The lower jaw consists of the long Meckel’s cartilage, its anterior part covered laterally by the den- tary, which bears a few teeth, and its pos- terior part by the angular. The retroarticular is present as a small ossification at the most posterior tip of Meckel’s cartilage, but the articular is not yet developed. Discussion In most actinopterygians, two dermal bones, the endopterygoid and the ectopter- ygoid, cover the medial face of the devel- oping palatoquadrate between the pars quadrata and pars autopalatina (Arratia & Schultze 1991). Usually, the pars quadrata and the pars autopalatina are connected by a thin strip of cartilage during at least some period in early development. The endopter- ygoid ossifies dorsomedial to this cartilage and the ectopterygoid ventromedial to it (see e.g., Arratia & Schultze 1991:figs. 14, 15; Britz 1996:figs. 3-5; Britz & Johnson 2002:figs. 4, 5). Even when the cartilagi- nous connection between the pars quadrata and the pars autopalatina is resorbed during ontogeny, a small projecting tip on the an- terodorsal face of pars quadrata usually re- mains for some time and can be used as a landmark. Such a stage is shown for a mas- tacembelid species in Britz (1996: Fig. 5). This landmark is also useful for taxa in VOLUME 116, NUMBER 2 539 exoccipital _« frontal first vertebra hyomandibular / nasal q ectopterygoid mesethmoid (#*-:** metapterygoid vorpet parasphenoid opercle premaxilla pars autopalatina dermopalatine e Ef subopercle maxilla interopercle \ quadrate 0.5 mm A Meckel's cartilage retroarticular hyomandibular ectopterygoid quadrate OLS toataal Fig. 1. Synbranchus sp., 25 mm: A, Skull and anterior two vertebrae, lateral view, cartilage white, bone light grey; B, Photograph of the hyopalatine area, lateral view, arrow points to ectopterygoid; C, S$. marmoratus, 47 mm, hyopalatine arch and anterior part of neurocranium, lateral view, modified from Britz (1996), cartilage white, bone light grey. 340 which the pars quadrata and the pars auto- palatina are never connected by cartilage. The pterygoid bone of the 25 mm Syn- branchus is located ventromedial to the projecting anterodorsal corner of the pars quadrata. Thus, the position of the developing pter- ygoid in Synbranchus is identical to that of Macrognathus (Britz 1996: fig. 5) or other teleosts (see e.g., Arratia & Schultze 1991: figs. 14, 15; Britz & Johnson: figs. 4, 5) and clearly demonstrates its homology with this bone. During subsequent development the ectopterygoid of Synbranchus enlarges greatly and bears numerous strong teeth. It becomes the dominant element of the adult synbranchid palatoquadrate (Fig. IC, see also Regan 1912:plate IX, fig. 1; Rastogi 1964:figs. 1-3; Rosen & Greenwood 1976: figs. 60, 61; Gosline 1983:fig. 3B; Travers 1984:fig. 10; Britz 1996:fig. 9C). Although we have no developmental information on the pterygoid of other synbranchids, it is reasonable to assume that it also represents the ectopterygoid, given its identical ap- pearance and position to the other bones of the hyopalatine arch. This homology falsi- fies Lauder & Liem’s (1983) interpretation of this bone as the endopterygoid and thus invalidates one of their putative synapo- morphies uniting the Synbranchidae and the Channidae. The wider phylogenetic impli- cations of our finding are beyond the scope of this paper and will be discussed in a forthcoming publication reevaluating the additional evidence for both hypotheses of synbranchid relationships. Acknowledgments Financial support of RB by a ‘visiting scientist’ fellowship of the Division of Fishes, National Museum of Natural His- tory, Smithsonian Institution, Washington D.C. is greatly acknowledged. Literature Cited Arratia, G., & H.-P. Schultze. 1991. Palatoquadrate and its ossifications: development and homology PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON within osteichthyans.—Journal of Morphology 208:1-81. Bailey, R. M., & C. Gans. 1998. Two new synbranchid fishes, Monopterus roseni from Peninsular India and M. desilvai from Sri Lanka.—Occasional Papers of the Museum of Zoology University of Michigan 726:1—18. Britz, R. 1996. Ontogeny of the ethmoidal region and hyopalatine arch in Macrognathus pancalus (Percomorpha, Mastacembeloidei), with critical remarks on mastacembeloid inter- and intrare- lationships.—American Museum Novitates 3181:1-18. , & G. D. Johnson. 2002. “Paradox Lost’’: Skeletal ontogeny of Indostomus paradoxus and its significance for the phylogenetic relation- ships of Indostomidae (Teleostei, Gasterostei- formes).—American Museum Novitates 3383: 1-43. , & M. Kottelat. 2003. Descriptive osteology of the family Chaudhuriidae (Teleostei, Synbran- chiformes, Mastacembeloidei), with a discus- sion of its relationships.—American Museum Novitates (in press). Gosline, W. A. 1983. The relationships of the masta- cembelid and synbranchid fishes.—Japanese Journal of Ichthyology 29:323-328. Johnson, G. D., & C. Patterson. 1993. Percomorph phylogeny: a survey of acanthomorphs and a new proposal.—Bulletin of Marine Science 52: 554-626. Lauder, G. V., & K. E Liem. 1983. The evolution and interrelationships of the actinopterygian fish- es.—Bulletin of the Museum of Comparative Zoology 150:95—197. Liem, K. E 1987. Functional design of the air venti- lation apparatus and overland excursions by tel- eosts.—Fieldiana 1379:1—29. Munshi, J. S. D., G. M. Hughes, P Gehr, & E. R. Weibel. 1989. Structure of the air-breathing or- gans of a swamp mud eel, Monopterus cu- chia.—Japanese Journal of Ichthyology 35: 453—465. Nelson, J. S. 1994. Fishes of the world. John Wiley & Sons, New York, 600 pp. Rastogi, M. 1964. The head skeleton of Amphipnous cuchia (Ham.).—Acta Zoologica 46:283—292. Regan, C. T. 1912. The anatomy and classification of the synbranchoid eels.—Annals and Magazine of Natural History, series 8, 9:387—390. Rosen, D. E., & P H. Greenwood. 1976. A fourth neo- tropical species of synbranchid eel and the phy- logeny and systematics of synbranchiform fish- es.—Bulletin of the American Museum of Nat- ural History 157:1—69. Travers, R. 1984. A review of the Mastacembeloidei, a suborder of synbranchiform teleost fishes. Part II: Phylogenetic analysis.—Bulletin of the Brit- ish Museum of Natural History (Zoology) 47: 83-150. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):341—355. 2003. Skull osteology of the characid fish Astyanax mexicanus (Teleostei: Characidae) Martha E. Valdéz-Moreno and Salvador Contreras-Balderas (MEVM) EI Colegio de la Frontera Sur, Carretera Chetumal-Bacalar km 2, Zona Industrial No. 2 A.P. 424 C.P. 77000, Chetumal, Quintana Roo, México, e-mail: mvaldez@ecosur-qroo.mx; (SCB) Bioconservacion, A.C., A.P. 504, San Nicolas de los Garza, Nuevo Leon, México 66450, e-mail: saconbal @axtel.net Abstract.—The skull of the characid fish Astyanax mexicanus is described based on twenty alizarin-stained adult specimens from Rio Salado, Rio Con- chos, Rio Alamo, and Rio San Juan populations, all of them Rio Grande trib- utaries in Northeastern Mexico. The skull has a circular shape in lateral view. The second infraorbital is triangular and never overlaps the inferior margin of the third infraorbital. The third infraorbital never reaches the laterosensory canal of the preopercular bone. The supraoccipital is short. The palatines, ec- topterygoids and mesopterygoids lack teeth. We describe two features not re- ported in Astyanax before: teeth on the second suspensory pharyngeal and posterior gill rakers on the four gill arches. Differences in skull osteology between A. mexicanus and other described species of Astyanax are denoted. Species of the fish genus Astyanax Baird & Girard, 1854, are found from the Nueces River, Texas, to Patagonia, Argentina (Ban- arescu 1990). This genus is among the most dominant group of freshwater fishes in this extensive region. The number of valid spe- cies of Astyanax that inhabits Mexico is un- certain (Contreras-Balderas & Lozano-Vi- lano 1998, Schmitter-Soto 1998). Because of the overall morphological similarity of the species of this genus, there is consid- erable confusion concerning species dis- tinctness and distribution. Contributing to this problem is the fact that the original de- scriptions of many of them are incomplete, and data for collection localities were not clearly provided and are often unknown. These problems are reflected in the present nomenclatorial confusion (Valdez-Moreno 1997, Contreras-Balderas & Lozano-Vilano 1998). Despite these uncertainties, most re- searchers, including Contreras-Balderas & Lozano-Vilano (1998), agree that one of the valid species is the northern form, Astyanax mexicanus (Filippi, 1853). It is known from central Texas, Rio Bravo (named Rio Grande in the U.S.A.), along Gulf of Mex- ico drainages such as the lower Panuco Riv- er, and through the Tecolutla—Cucharas sys- tems (Miller 1978, Obregon-Barboza et al. 1994, Valdéz-Moreno 1997). However, some authors, such as Espinosa-Pérez et al. (1993), considered forms from the Rio Bal- sas and the Papaloapan basins extending to Petén, Guatemala, to be A. mexicanus s. |. Others (e.g., Paulo Maya 1994), reported the presence of two distinct forms of Asty- anax in the Rio Balsas samples, without se- lecting names or referring them to either of the two nominal species described from the basin, Astyanax nitidus Bocourt (1868) and A. fulgens Bocourt (1868). The identity of the Rio Balsas form is still unresolved. We concur with Miller (in litt.) and Contreras-Balderas & Lozano-V1- lano (1998) that Mexican forms of Astya- nax are not the same as the South American representatives assigned to A. “‘fasciatus’’, and none of the Rio Balsas forms are A. 342 mexicanus. Such forms probably represent a species of the A. aeneus group. Further research is needed to solve this problem. Many studies of A. mexicanus deal pri- marily with morphometry (Schuppa 1984, Paulo-Maya 1994). Except for minor ref- erences to certain bones by Lozano-Vilano & Contreras-Balderas (1990), none deals with osteology. This approach contrasts with other studies in characid fishes. For example, the osteology of the American characid, Brycon meeki, was described in detail by Weitzman (1962). Menezes (1969) considered osteology in his study of the phylogeny of the tribe Acestrorhynchini, and Weitzman & Fink (1983) used osteo- logical characters to elucidate relationships among neon tetras, Paracheirodon spp. Vari & Harold (1998) diagnosed the genus Creagrutus as a monophyletic group within the family Characidae on the basis of some modifications of skull bones. Malabarba (1998) provided a new diagnosis of the Cheirodontinae using features of the den- tary, tooth morphology and other skull bones. Vari (1989a) studied the structure of the skull in members of the family Curi- matidae and the genus Pseudocurimata. Vari (1983) also used osteological infor- mation to hypothesize relationships within Curimatidae, Prochilodontidae, Anostomi- dae and Chilodontidae, as well as relation- ships among the Ctenoluciidae (Vari 1995). A cladistic analysis based on osteology by Buckup (1998) proposed relationships of the Characidiinae with the Crenuchinae. Langeani (1998) hypothesized the mono- phyly of the family Hemiodontidae using characteristics of skull bones along with other data. The posterior region of skull and pectoral girdle of a female A. mexicanus from Guay- Figs. 1-8. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON alejo River, Tamaulipas, Mexico, was de- scribed and figured by Weitzman & Fink (1983, 1985), but they did not describe the entire skull of this species. The skull anat- omy of A. “‘fasciatus’’ has been described only for the form from Balsas River, Mex- ico, by Mejia-Méjica & Diaz-Pardo (1991). Problems with the identity of this form were discussed above. The purposes of this work are to describe in detail the skull of A. mexicanus and to provide a baseline to compare with other species and genera. The use of the cranial characters described herein may provide a means to better study the taxa assigned to the genus Astyanax and could be useful to make inferences about the phylogeny of members of the genus or the family. Methods Material examined.—Twenty adult spec- imens of Astyanax mexicanus (>40 mm standard length, SL), from the Fish Collec- tion of the Autonomous University of Nue- vo Leon (UANL), were selected from Rio Grande/Rio Bravo localities: Rio Salado at Socavon, 5.2 km SW from Muzquiz, State of Coahuila (UANL-7271: 3 2 and 2 3 SL = 57.47 to 74.67 mm); Rio Conchos: Rio Chuviscar at Aldama, State of Chihuahua (UANL-6944; 4 2 and 1 do SL = 46.69 to 54.74 mm); Rio Alamo in Paso de las An- acuas, 26 km W from Ciudad Mier, State of Tamaulipas (UANL-3932; 3 2 and 2 d SL = 53.71 to 60.11 mm), and Rio San Juan: Canon de la Boca, | km after the dam Presa de la Boca, State of Nuevo Leon (UANL-4924; 5 2 SL = 67.65—75.54 mm). The specimens were cleared and stained following the technique of Hollister (1934). Terminology follows Weitzman (1962), — Skull of Astyanax mexicanus, lateral view, left side, female, SL 74.29 mm. 1, Full view; 2, Infraorbital I, different shapes; 3, Antorbital, different shapes; 4 Infraorbital 4, malformation; 5, Palatine, ectop- terygoid and mesopterygoid; 6, Lower jaw, opercular series, mandibular arch and palatine arch; 7, Quadrate; 8, Preopercular. Abbreviations.—AA—anguloarticular; AT—antorbital; AR—retroarticular; DE—dentary; EC— VOLUME 116, NUMBER 2 343 ctopterygoid; EL—lateral ethmoid; ET—ethmoid; FR—frontal; HY—hyomandibular; I1— Sue 5mm SC 5 mm L SA | 5 mm Figs. 17-22. Cranium of A. mexicanus, female, SL 74.29 mm. 17, lateral view. 18, Orbitosphenoid, different shapes; 19, Pterosphenoid, different shape; 20, dorsal view; 21, ventral view; 22a, Otolith asteriscus, dorsal view; 22b, Otolith sagitta, dorsolateral view; 22c, Otolith lapillus, dorsal view. Abbreviations.—AS—asteriscus; BO—basioccipital; EL—lateral ethmoid; EO—exoccipital; EP—epioccipital; ET—ethmoid; FR—frontal; LA— lapillus; Ol—opisthotic; OR—orbitosphenoid; PA—parietal; PC—pterotic; PP—parasphenoid; PR—prootic; PT—pterosphenoid; RN—rhinosphenoid; SA—sagitta; SC—supraoccipital; SP—sphenotics; VO—vomer. VOLUME 116, NUMBER 2 ulating with the lateral ethmoids. Two fo- ramina providing passages for ramus buc- calis of facial nerve. Ventrally, vomer artic- ulating with parasphenoid and dorsally with ethmoid. Ventral surface of vomer without anatomical markings, few present on dor- sum. Frontals (Figs. 1, 17, 20, 21).—Laminar bones comprising a large part of dorsal re- gion of cranium, connected to each other by epiphyseal bar that bridges the cranial fon- tanels. Each frontal with a ventral keel in medial position. Frontal articulating dorso- posteriorly with parietal, dorsoanteriorly with ethmoid, nasal and lateral ethmoid, and ventrally with orbitosphenoid, ptero- sphenoid, sphenotics, and pterotic. Each frontal containing canals of accoustico la- teralis system. Anterior portions of these canals wide and continuous with nasal ca- nal. Frontal bone with medial canals split- ting in two main branches: an epiphysial branch extending medially over the epiphy- sial bar and opening above the frontal fon- tanel, and posterior branch connected pos- teriorly with the parietal and this canal, but not to extrascapular canal. Another branch, smaller than the former two, runs laterally continuing with the pterotic canal. Small side branches of these canals opening on dorsal surface of frontal. Parietals (Figs. 1, 17, 20).—Rectangular laminar paired bones, without anatomical markings on their ventral and dorsal surfac- es. Each one articulating dorsoanteriorly with frontal, posteriorly with epioccipital, dorsoposteriorly with supraoccipital and la- teroventrally with pterotic. Parietals sepa- rated by the cranial fontanel. Laterosensory canal parietal connecting posterolaterally with extrascapular canal and extending dor- sally to cranial fontanel. Supraoccipital (Figs. 1, 17, 20).—Medi- an bone constituting posterior roof of cra- nium, articulating dorsoanteriorly with pa- rietals and forming posterior border of me- dian dorsal fontanel, ventrolaterally with epioccipitals, and ventrally with pterotics. In dorsal view, supraoccipital ‘““V’’ shaped, 347 with a wide anterior base. Dorsal surface with a groove extending posteriorly to end of supraoccipital spine. Epioccipitals (Figs. 17, 20).—Smooth tu- bular paired bones, each containing a major portion of the posterior vertical semicircular canal of the auditory and equilibrium sys- tems. Epioccipital articulating dorsally with supraoccipital, posteroventrally with exoc- cipital and anteroventrally with pterotic. Lateral epioccipital process anteriorly con- tacting parietal and pterotic, dividing pos- temporal fossa in two parts. Exoccipitals (Figs. 17, 21).—Smooth paired bones forming ventral part of pos- terior cranium, dorsally articulating with epioccipital, ventrally with basioccipital, anteriorly with prootic and anterolaterally with pterotic and opisthotic. Each exoccip- ital with laminar part enclosing the foramen magnum and ventral spherical part consti- tuting the roof of the otic capsule for la- gena. Opisthotics (Fig. 21).—Small, thin bones, having triangular shape with a curved base, located external to articulation between pterotic, exoccipital and prootic; opisthotic attaching by ligament to ventral process of postemporal bone. Basioccipital (Figs. 17, 21).—Constitut- ing posterior base of cranium, with one globular part and other laminar. Dorsally ar- ticulating with exoccipitals, anteriorly with prootic, ventromedially with parasphenoid and posteriorly with first vertebra. Basioc- cipital also forming ventral portion of otic capsule for lagena and posterior region of saccular cavity for asteriscus. No anatomi- cal markings on external surface. Pterotics (Figs. 17, 20, 21).—Paired bones articulating anteriorly with sphenotic, ventromedially with prootic, dorsally with frontal and parietal, and posteriorly with epioccipital. Pterotic with rounded process that projects posteriorly and ventrally, pro- viding insertion for levator operculi muscle. Pterotic articulating with hyomandibular at a groove. Internally, pterotic enclosing a major part of the horizontal semicircular ca- 348 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nal. Each pterotic with a ““V’’-shaped canal of the accoustico lateralis system, canal continuing ventrally to preopercular canal and dorsally to frontal canal. A small fo- ramen present between pterotic and sphen- otic. Extrascapular overlapping posterodor- sal part of pterotic. Sphenotics (Figs. 17, 20 21).—Paired bones, quadrangular with rounded edges in lateral view and with a strong rounded pro- cess anterodorsally. Sphenotics articulating anteriorly with pterosphenoid, ventrally with prootic, posteriorly with pterotic and dorsally with frontal. Sphenotic not asso- ciated with lateral sensory system. Prootics (Figs. 17, 21).—Complex paired bones. Dorsolaterally articulating with sphenotics and pterotics, medially with prootic on the other side, anteriorly with pterosphenoid, ventrally with parasphenoid, and posteriorly with opisthotic, exoccipital and basioccipital. Prootic with three foram- ina of different sizes, the auditory foramen largest. Parasphenoid (Figs. 1, 17, 21).—Lon- gest unpaired bone of cranium, spanning ventral median region of cranium and artic- ulating anteriorly with vomer, dorsoposter- iorly with prootics and basioccipitals. Ven- trally with a medial quilla joining with sus- pensor pharyngeals by ligaments. Ventro- posteriorly ending in two laminar process. Pterosphenoids (Figs. 1, 17, 21).—Paired bones, rectangular in lateral view, with smooth edges, forming part of wall and floor of cranial cavity. Together with orbi- tosphenoid, pterosphenoids constituting medial aspect of ocular cavity. Pterosphe- noid articulating dorsally with frontal, lat- erally with sphenotic, ventrally with para- sphenoid, and anteriorly with orbitosphe- noid. A small foramen for trochlear nerve present at articulation of orbitosphenoid and pterosphenoid. Orbitosphenoid (Figs. 1, 17, 21).—Lo- cated ventral to middle region of frontal and posterior to ethmoid. Anterodorsally artic- ulating with frontal and posteriorly with pterosphenoid. Ventrally, orbitosphenoid with a process directed toward anterior part of cranium. Orbitosphenoid highly varible in shape, length, and width at intra- and in- terpopulation levels (Fig. 18), usually re- sembling a “J” in lateral view. In dorsal view, orbitosphenoid resembling two wings or a “bird in flight” (Mejia-Mojica & Diaz- Pardo 1991). Rhinosphenoid (Figs. 1, 17).—Small, al- most square to irregular medial bone locat- ed anteriorly to orbitosphenoid, between lateral ethmoids. Rhinosphenoid ossified in five of our specimens. Nasals (Fig. 1).—Tubular paired bones, slightly curved with smooth surface, locat- ed anterolateral to cranium. Nasal articulat- ing anteriorly with premaxilla, posteriorly with frontal. Each nasal bone containing a branch of the laterosensory system. Otoliths (Fig. 22a, b, c).—The ovoid la- pillus contained in an utricular sac on floor of prootic. The spine-like sagitta contained in saccular recess formed by prootic, basi- occipital and exoccipital. The lenticular as- teriscus lying in capsule formed by the ex- occipital and basioccipital. Asteriscus larger than the other otoliths and bearing small projections around its border. Antorbital and Infraorbital bones.—This series Comprising seven paired elements as follows: Antorbitals (Fig. 1).—Each bone located ventrolaterally to respective nasal orifice. Antorbital with an elongated triangular shape, the anterior part considerably nar- rower than the posterior part; however, shape of antorbital varying among speci- mens. This variation apparently indepen- dent of sex (Fig. 3). Infraorbitals I (Figs. 1, 2).—Shape sim- ilar to the blade of a scalpel, the ventral border variably prolonged in width and length. First infraorbital partially over- lapped by the maxilla. Infraorbitals 2 (Fig. 1).—Triangular bone with inferior margin usually smooth or sometimes jagged. Posterior portion of bone deeper than anterior portion. Second infraorbital located lateral to anguloarticular VOLUME 116, NUMBER 2 and quadrate bones and never overlapping inferior margin of infraorbital 3. Infraorbitals 3 (Figs. 1, 4).—Largest bone of the series. Shape semicircular. In- ferior margin always smooth, lateral surface occasionally with small foramina. Third in- fraorbital located lateral to quadrate and preopercle, but inferior margin never reach- ing laterosensory canal of preopercular bone. Infraorbitals 4 (Figs. 1, 4).—Rectangular bone, shape depending on shape of articu- lation with infraorbital 3. Anteroventral edge projecting ventrally. Fourth infraorbit- al located lateral to hyomandibular bone. Infraorbitals 5 (Figs. 1, 4).—Quadran- gular bone with rounded corners. Located dorsal to infraorbital 4 and lateral to hyo- mandibular. Infraorbital 6 (Fig. 1).—Thumb-nail shaped bone, posteroventral margin roughly straight at articulation with fifth infraorbital. Sixth infraorbital located lateral to sphen- otic and frontal. All the circumorbital bones smooth and bearing canals for the accoustico lateralis system. Canal of sixth infraorbital connect- ing with laterosensory canals of pterotic and frontal bones. Astyanax mexicanus lacking supraorbital bones. Upper Jaw Premaxillae (Figs. 1, 9).—Strong, paired bones articulating dorsally with ethmoid and nasal bones, laterally with maxillae. Premaxilla triangular in dorsal view. Each premaxilla with two teeth rows. In one Specimen a replacement row present. An- terior row consisting of four teeth, each one having three, sometimes four, cusps. Pos- terior row with four to five teeth, each one with three to six cusps. Dorsal surface of premaxilla smooth, ventral surface with a small foramen. One specimen with a tooth malformation in the posterior row (Fig. 11). Maxillae (Figs. 1, 10).—Thin, paired bones, each articulating anteriorly with pre- maxilla. Posterior ramus of maxilla over- 349 lapping dentary and bearing one or two teeth, each with five to eight cuspids. Lower Jaw Dentaries (Figs. 1, 6, 12).—Paired robust bones, each articulating dorsoposteriorly with anguloarticular and retroarticular, pos- teromedially with coromeckelian bone. Dentary with a single row of four frontal teeth and several posterior ones. First and third frontal teeth similar in size and ap- pearance; second one about the same size or slightly smaller than the first. The fourth smaller than the other three frontal teeth. All of them with five to seven, usually six, cusps. Fifth dentary tooth smaller than the first four, with five or six cusps, usually five. Remaining teeth smaller (five to ten), with one or two cusps. Replacement row of teeth always present but variously devel- oped and positioned, teeth of replacement row directed vertically. Lateral surface of each dentary with a ca- nal for accoustico lateralis system, associ- ated with three small foramina. Posterior portion of this canal continuous with the an- guloarticular canal. Symphyseal and lateral diastemmas absent. Anguloarticulars (Figs. 1, 6, 12).— Paired bones that form the joint between the dentary and skull, shape similar to a fan with three peaks, middle one longest and pointed, other two rounded. Anguloarticular articulating anteriorly with dentary, ventro- posteriorly with retroarticular, posteriorly with quadrate via a condyle, and medially with Coromeckelian bone. Anguloarticular containing a canal of the lateral sensory system that communicates posteriorly with canal of preopercle. Retroarticulars (Figs. 1, 6, 12).—Paired, triangular bones, equal to or slightly smaller than coromeckelian bone. Retroarticular ar- ticulating ventroposteriorly with anguloar- ticular. Shape and length of retroarticular variable. Coromeckelian bones (Fig. 12).—Small, paired, oval bones with smooth external 350 surface, located along medial face of an- guloarticular. Medial surface of coromeck- elian with a median keel. Opercular Series Opercles (Figs. 1, 6).—Thin, paired, nearly rectangular bones with smooth sur- faces and borders. Opercle articulating an- teriorly with hyomandibular via a condyle on the latter, opercle articulating ventrally with subopercle. Subopercles (Figs. 1, 6).—Thin, paired bones with smooth surfaces and borders ar- ticulating dorsally with opercle. Anterior part of subopercle with a small dorsal pro- cess. Interopercles (Figs. 1, 6).—Paired, flat, smooth bones with triangular shape, poste- rior portion deeper than anterior part. Inter- opercle articulating dorsally with preoper- Cle: Preopercles (Figs. 1, 8).—Large, flat, paired bones, the dorsal limb lying between hyomandibular and opercle. Preopercle roughly triangular in shape, with a broad base and thin dorsal limb. A canal of the laterosensory system extending the length of preopercle; anteriorly, the canal contin- uous with anguloarticular canal, dorsally the canal continuous with pterotic canal. Mandibular Arch Hyomandibulars (Fig. 6).—Strong, paired bones with smooth surfaces. Dorsal portion of bone broad with rounded mar- gins, ventral part slender. Hyomandibular articulating dorsally with neurocranium through a fossa in the pterotic and another in the sphenotic. Dorsoposteriorly, hyoman- dibular articulating with opercle, ventrally with metapterygoid. Quadrates (Figs. 1, 7).—Paired bones ly- ing above preopercle and articulating ante- roventrally with anguloarticular by a con- dyle. Quadrate L-shaped, with vertical limb shorter than horizontal one. A posteriorly directed process of varying size sometimes present along dorsal margin of horizontal PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON limb. Variation in this process apparently independent of sex. Symplectics (Fig. 6).—Two smooth rod- like bones, slightly curved or straight, lo- cated medial to quadrate bone. Palatine Arch Metapterygoids, mesopterygoids, ectop- terygoids, palatines and the vertical part of quadrate forming the anterior suspensory mechanism of lower jaw. Metapterygoids (Fig. 6).—Paired bones of irregular shape with smooth surfaces. A small foramen present on posterior portion of metapterygoid. Posterior portion of bone deeper than anterior part. Two processes of variable size present on dorsal surface of posterior portion of metapterygoid. Metap- terygoid articulating dorsoanteriorly with mesopterygoid and dorsoposteriorly with hyomandibular. Mesopterygoids (Figs. 5, 6).—Paired tri- angular bones with smooth surfaces. Me- sopterygoid articulating anteriorly with pal- atine, anteroventrally with ectopterygoid, and ventrally with metapterygoid and quad- rate. Ectopterygoids (Figs. 5, 6).—Paired tri- angular bones adjacent to mesopterygoids. Ectopterygoid narrower than mesoptery- goid but of similar surface morphology and thickness. Palatines (Figs. 5, 6).—Paired bones, square shaped with slightly rounded angles. Palatines articulating anteriorly with max- illary and posteriorly with ectopterygoids and mesopterygoids. Mesopterygoid, ectopterygoid and pala- tine bones without teeth. Hyoid Arch Interhyals (Figs. 6, 15, 16).—Small, cy- lindrical, paired bones, lying ventral to hyo- mandibular and posterior to symplectic. In- terhyal located below interopercle, articu- lating with hyomandibular and symplectic. Posterior ceratohyals (Fig. 15).—Paired triangular bones with rounded posterior end VOLUME 116, NUMBER 2 and smooth borders and surfaces. A fora- men located centrally or somewhat closer to the posterior border, associated or not with a canal. Posterior ceratohyal articulat- ing anteriorly with anterior ceratohyal and posteriorly with interhyal. Ventrally poste- rior ceratohyal supporting fourth branchio- stegal. Anterior ceratohyals (Fig. 15).—Paired bones, rectangular in lateral view, with smooth surfaces. Anterior ceratohyal artic- ulating anteriorly with dorsal and ventral hypohyals, posteriorly with posterior cera- tohyal, and, ventrally, supporting first three branchiostegal rays. Third branchiostegal ray overlying and articulating with most posteroventral part of anterior ceratohyal. First and second branchiostegal rays insert- ed into small grooves on ventral margin of anterior ceratohyal. Dorsal hypohyals (Fig. 15).—Shallow paired bones overlying and articulating with ventral hypohyal and anterior ceratohyal. Ventral hypohyals (Fig. 15).—Paired quadrangular bones with rounded angles, articulating with anterior ceratohyals and dorsal hypohyals. Urohyal (Fig. 14b)—Triangular bone bearing projections on the ventral surface of varying length and number. Urohyal in- serting between and connected by ligaments to ventral hypohyals. Branchial Arches Basithyal (Fig. 13a).—This rectangular and slightly flattened median bone the an- teriormost element of the branchial skele- ton. Basihyal located between dorsal hy- pohyals. In one specimen, first basibranchi- al malformed (see Fig. 14a). Basibranchials (Fig. 13a).—Together with the basihyal, these four unpaired bones forming the median spine of the gill-arch skeleton. Anteriormost basibranchial flat- tened and attached by a ligament to basi- hyal; second and third basibranchials elon- gated; and fourth leaf-shaped and smaller 351 than preceding two. A shared basibranchial cartilage uniting all basibranchials. Hypobranchials (Fig. 13a).—Three paired bones comprising most proximal el- ements of first three gill arches. First hy- pobranchial tongue-shaped, articulating with first basibranchial and first ceratobran- chial. Second hypobranchial dome-shaped, articulating with second basibranchial and second ceratobranchial. Third hypobranchi- al saccular-shaped, articulating with third basibranchial and third ceratobranchial. All hypobranchials bearing gill rakers on ante- rior edge. Ceratobranchials (Fig. 13a).—Five paired bones that form most of the ventral part of each gill arch. First three cerato- branchials elongated and slightly flattened, each articulating proximally with a hypo- branchial and distally with an epibranchial. Fourth ceratobranchial elongated and slightly flattened, articulating proximally with a hypobranchial and distally with fourth epibranchial. First four ceratobran- chials bearing gill rakers along anterior and posterior edges. Fifth ceratobranchial mod- ified to form lower pharyngeal bone, which bears several small, scattered, unicuspid teeth on its dorsal surface and gill rakers along the anterior edge. Epibranchials (Fig. 13a).—Four paired bones forming most of dorsal portion of each gill arch. First two elongated, slightly flattened, and lacking uncinate processes. Third elongated and bearing posteriorly di- rected uncinate process near distal end of bone; uncinate process articulating with up- per pharyngeal teeth. Fourth epibranchial triangular. All epibranchials with small rak- ers along anterior edge, and first three with gill rakers along posterior edge. Third and fourth epibranchials support- ing upper pharyngeal plate. Upper pharyn- geal irregular in shape and with several small, scattered unicuspid teeth on ventral surface. Suspensory pharyngeals.—Three small, paired bones articulating with first three epibranchials and with parasphenoid. First S52 one triangular, without teeth (Fig. 13b). Second almost rectangular with few unicus- pid teeth (Fig. 13c). Third triangular with numerous unicuspid teeth (Fig. 13c). Gill Rakers.—Total number of anterior gill rakers on first arch 18 to 21, second and third arches 18 to 20, and fourth arch 13 to 18. Five posterior gill rakers on first arch, five to six on second, 14 to 16 on third, and five to six on fourth. Dorsal border of lower pharyngeal with six to ten structures similar to gill rakers. All gill rakers conical in shape and with several short conical spic- ules arranged irregularly. Discussion The skull of Astyanax mexicanus is made up of 58 bones, nine of them unpaired (eth- moid, vomer, parasphenoid, supraoccipital, orbitosphenoid, basioccipital, urohyal, ba- sihyal, rhinosphenoid), all others are paired. This number differs from that reported for Astyanax “‘fasciatus”’ (sensu Mejia-Mojica & Diaz-Pardo 1991), because they did not include the rhinosphenoid. A comparison of the skull between A. mexicanus and the descriptions of A. “‘fas- clatus”’ sensu Mejia-Mojica & Diaz-Pardo 1991, suggests that they are quite similar. However, differences are present in some bones. The opercle in A. mexicanus is al- most rectangular and wide, whereas in A. *‘fasciatus’’ it is more rhomboid-like and narrower. The supraoccipital (in dorsal view) 1s short with a wide base in A. mex- icanus, whereas in A. “‘fasciatus”’ it is lon- ger with a narrow base. Infraorbital 3 in A. mexicanus 1s semicircular, in in A. “‘fascia- tus’ it is angular. Infraorbital 2 has a wide margin articulating with infraorbital 3 in A. mexicanus, Whereas this margin is narrower in A. “fasciatus.” Also, Mejia-Mojica & Diaz-Pardo (1991) described the posterior ceratohyal in A. ‘fasciatus’ as having a foramen associated with a canal, whereas in A. mexicanus it may or may not be associated with a canal. The number of teeth cusps reported from PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the mandibular bones are different among the two species. All teeth are pentacuspid in A. “‘fasciatus,’’ whereas in A. mexicanus there are 3—4 cusps on teeth in the outer row, 3—6 cusps on teeth of the inner row of the premaxilla, 5-8 cusps on maxillary teeth, and 2—6 on dentary teeth. Finally, the number of gill rakers varies among the two species: in A. mexicanus 11 gill rakers are present on the first arch, whereas in A. “‘fas- ciatus’’ there are 13. The latter count is similar to that of A. aeneus, which has 12— 14 gill rakers (Schmitter-Soto 1997). Mejia-Mojica & Diaz-Pardo (1991) de- scribed as unique for A. “‘fasciatus” the presence of a small plate with teeth on the third suspensory pharyngeal; however, it is not exclusive to this species. We found this plate in almost all populations of A. mexi- canus studied. Here we provide the first re- port of teeth on the second suspensory pha- ryngeal for A. mexicanus. Most other osteological elements of A. mexicanus are very similar to those de- scribed for Astyanax “‘fasciatus”’ and are not discussed further. In general, osteology of A. “fasciatus” of Mejia-Mojica and Diaz-Pardo (1991) resembles that of other southern forms previously described by Valdéz-Moreno (1997), e.g., infraorbital 2 with a short base, infraorbital 3 with a dor- so-posterior projection and angulate shape, and supraoccipital long (in dorsal view) with a narrow base. Southern forms studied by Valdéz-Moreno (1997) appear to be A. aeneus, based on the work of Lozano and Contreras (1991). The main osteological difference found between the skull of A. mexicanus and Bry- con meeki, a well-described “representative of characid osteology”’ (Weitzman 1962), is the presence of supraorbital bones in the latter. The possible significance of the pres- ence or absence of the supraorbital in char- acid fishes was discussed by Weitzman & Malabarba (1998). The shapes of the in- fraorbitals, antorbital, quadrate and angu- loarticular are different in A. mexicanus and B. meeki. The rhinosphenoid is larger in B. VOLUME 116, NUMBER 2 meeki than in A. mexicanus. There are three rows of teeth on the premaxilla in B. meeki, two in A. mexicanus. There are 13—15 teeth on the maxilla is 13 to 15 in B. meeki, one or two in A. mexicanus. The pterotic pro- cess and sphenotic are longer in B. meeki than in A. mexicanus. The distribution of teeth on the superior pharyngeal (in five rows in B. meeki) is not the same in Asty- anax (irregular). The canals of the accous- tico lateralis system in A. mexicanus are similar to those of Brycon meeki, except the anterior portion of the canal in the frontal bone in the former is wider. B. meeki has more anterior gill rakers (31 to 33) than A. mexicanus (18 to 21). Finally, A. mexicanus has teeth on suspensory pharyngeals 2 and 3 whereas B. meeki does not. A comparison of A. mexicanus with other tetragonopterinae, like Paracheirodon ax- elrodi (Schulz, 1956), reveals even more differences: P. axelrodi lacks infraorbitals 4, 5 and 6, and major structural differences are evident in infraorbitals | and 2, the pter- otic, opercle, hyomandibular, sphenotic, ri- nosphenoid, and orbitosphenoid (Weitzman & Fink 1983). The figures of the posterior region of skull and pectoral girdle of female Astyanax mexicanus, that appeared with no explana- tions in Weitzman & Fink (1983), are con- sistent with our observations. The presence of two rows of teeth on the premaxillae in our specimens agrees with previous reports for the genus by Eigen- mann (1921), Géry (1977) and Mejia-Mo- jica & Diaz-Pardo (1991). We found the number of maxillary teeth to be only one or two; this agrees with data of Eigenmann (1921), Géry (1977) and Mejia-Mojica & Diaz-Pardo (1991) and is within the range (zero to three) given by Contreras & Rivera (1985) for Mexican forms. The last authors followed Gery (1977) in reporting up to seven maxillary teeth for Central American forms. None of the above citations referred specifically to A. mexicanus. Alvarez del Villar (1970) considered the maxillary teeth smooth (“‘no aserrados’’, sic), and Contrer- 353 as-Balderas & Rivera (1985) reported teeth tricuspidate for the genus Astyanax; we found these teeth with five to eight cusps in A. mexicanus. The number of dentary teeth in our spec- imens is the same as that reported by Géry (1977). Alvarez (1970) and Mejia & Diaz- Pardo (1991) mentioned that dentary teeth are pentacuspid in Astyanax, whereas Con- treras & Rivera (1985) indicated that they are tricuspid for the genus. Apparently, number of cusps is variable in this genus, and may be subject to abrasion, so this character should be considered with cau- tion. Malformations in the premaxillae (Fig. 12) and the fourth infraorbitals (Fig. 5) were detected in two specimens from Rio Salado and Rio Alamo. Cailliet et al. (1986) described deformations on the jaws and head of other species of fish, related to he- redity and/or enviromental factors. How- ever, this is the first time that malformations are reported for Astyanax, and the causes were not addressed in this study. Acknowledgments J. J. Schmitter-Soto, and M. Elias-Gu- tiérrez made valuable comments to the orig- inal manuscript. This study is part of a grant given to M. E. Valdéz-Moreno by Conacyt (Number 138485). Lourdes Lozano-Vilano allowed the use of material from the Fish Collection at Universidad Aut6noma de Nuevo Leon, Monterrey, México. H. Mejia- Moyjica kindly allowed examination of stained and cleared material from Balsas basin. Literature Cited Alvarez del Villar, J. 1970. Peces Mexicanos (claves). Secretaria de Industria y Comercio, México, 166 pp. Banarescu, P. 1990. Zoogeography of fresh waters. AULA, Verlag Wiesbaden, Germany, 541 pp. Bocourt, EM. 1868. Note sur les poissons du genre Tetragonoptére provenant du Mexique et du Guatemala-5 Ser. Sci. Nat. Zool., 9:62. Buckup, P.A. 1998. Relationships of the Characidiinae and phylogeny of characiform fishes (Teleostei: Ostariophysi). Pp. 123—144 in L. R. Malabarba, Re ES Reiss Re Pa VanikeZ. MeyS= Eucena é2(@sA- S. Lucena, eds., Phylogeny and classification of neotropical fishes. Porto Alegre, Brazil. Cailliet, G. M., M. S. Love, & A. W. Ebeling. 1986. Fishes: a field and laboratory manual on their structure, identification and natural history. Wadsworth Publishing Company, Belmont, Cal- ifornia, 194 pp. Contreras-Balderas, S., & M. L. Lozano-Vilano. 1998. Problemas nomenclaturales de las formas mex- icanas del género Astyanax (Pisces: Characi- dae).—Zoologia Informa 38:1—13. , & R. Rivera-Teillery. 1985. Bramocharax (Catemaco) caballeroi subgen. et. sp. nv., del Lago de Catemaco, Veracruz, México.—Publi- caciones Biolégicas del Instituto de Investiga- ciones Cientificas UANL 2(1):7—29. Eigenmann, H. C. 1921. The American Characidae, part 3.—Memoirs of the Museum of Compara- tive Zoology 43:209-—310. Espinosa Pérez, H., M. T. Gaspar Dillanes., & P. Fu- entes Mata. 1993. Listados faunisticos de Méx- ico. III. Los peces dulceacuicolas mexicanos. Instituto de Biologia, UNAM, México, D.E, 99 Pp. Filippi, EF 1853. Nouvelles espéces de poissons.—Re- vue Magazine Zoologie 5:164—171 Fink, S. V., and W. L. Fink. 1981. Interrelationships of the ostariophysan fishes (Teleoste1).—Zoologi- cal Journal of the Linnean Society 72:297—353. Géry, J.R. 1977. Characoids of the world. Tropical Fish Hobbyst Publications, Neptune City, 672 Pp. Hollister, G. 1934. Clearing and dyeing fish for bone study.—Zoologica 12(10):89-101. Langeani, EF 1998. Phylogenetic study of the Hemio- dontidae (Ostariophysi: Characiformes). Pp. 145-160 in L. R. Malabarba, R. E. Reis, R. P. Vari, Z. M. S. Lucena and C. A. S. Lucena, eds., Phylogeny and classification of neotropical fish- es. Porto Alegre, Brazil. Lozano-Vilano. M. L., & S. Contreras-Balderas. 1990. Astyanax armandoi, n. sp. from Chiapas, Méx- ico (Pisces, Ostariophysi: Characidae) with a comparison to the nominal species A. aeneus and A. mexicanus.—Universidad y Ciencia 7(14):95—107. Mejia-Mojica. H., & E. Diaz-Pardo. 1991. Descripcién del craneo de Astyanax “‘fasciatus” (Pisces: Characidae).—Anales Escuela Nacional de Ciencias Biolégicas 34:191—214. Malabarba, R. L. 1998. Monophyly of the Chaerodon- tinae, characters and major clades (Ostariophy- si: Characidae). Pp. 193-234 in L. R. Malabar- loz, IR. JB, INGIS. IR, Je Wersl, Z, IMI S. Lures anv C. A. S. Lucena, eds., Phylogeny and classifi- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON cation of neotropical fishes. Porto Alegre, Bra- zl. Menezes, A. N. 1969. Systematics and evolution of the tribe Acestrorhynchini (Pisces: Characidae).— Arquivos de Zoologia 18:1—150. Miller, R. R. 1978. Composition and derivation of the native fish fauna of the Chihuahuan desert re- gion.—Transactions of the Biological Research Chihuahuan Desert Region 4:773—802. Obreg6n-Barbosa, H., S. Contreras-Balderas, & M. L. Lozano-Vilano. 1994. The fishes of Northern and Central Veracruz, México.—Hydrobiologia 286:79-95. Paulo-Maya, J. 1994. Analisis morfométrico del gé- nero Astyanax (Pisces: Characidae) en México. Unpublished M. S. thesis, Escuela Nacional de Ciencias Bioldgicas, Instituto Politécnico Na- cional, México, 107 pp. Schmitter-Soto, J. J. 1998. Diagnosis of Astyanax al- tior (Characidae), with a morphometric analysis of Astyanax in the Yucatan Peninsula.—Ichthy- ological Exploration of Freshwaters 8(4):349— BOE Schulz, L. P. 1956. The amazing new fish called the scarlet characin.—New Jersey, Tropical Fish Hobbyst 4(4):41—43. Schuppa, M. 1984. Morphometrische und meristische Untersuchungen an verschiedenen Astyanax— populationen (Characidae) Mexikos. Unpub- lished thesis, Universitat Hamburg, Germany, 94 pp. - Valdéz-Moreno, M. E. 1997. Estudio comparativo os- teol6gico del género Astyanax en diversas cuen- cas de México. Unpublished M.S. thesis, Univ- ersidad Nacional Autonoma de México, 139 pp. Vari, R. P. 1983. Phylogenetic relationships of the fam- ilies Curimatidae, Prochilodontidae, Anostomi- dae and Chilodontidae (Pisces: Characifor- mes).—Smithsonian Contributions to Zoology 378:1—61. . 1989a. A phylogenetic study of the Neotrop- ical Characiform family Curimatidae (Pisces: Characiformes).—Smithsonian Contributions to Zoology 471:1-71. . 1989b. Systematics of the Neotropical Char- aciform genus Pseudocurimata Fernandez-Yé- pez (Pisces: Ostariophysi).—Smithsonian Con- tributions to Zoology 490:1-—71. . 1995. The neotropical fish family Ctenoluci- idae (Teleostei: Ostariophysi: Characiformes): Supra and Intrafamilial phylogenetic relation- ships, with a revisionary study).—Smithsonian Contributions to Zoology 564:1—90. , & A. S. Harold. 1998. The genus Creagrutus (Teleostei: Characiformes: Characidae): Mono- phyly, relationships, and undetected diversity. Phylogeny and classification of neotropical fish- es. Pp. 245-260 in L. R. Malabarba, R. E. Reis, VOLUME 116, NUMBER 2 RerevanilyZ Mes. lbucenaand CyAsS lbucena: eds., Phylogeny and classification of neotropical fishes. Porto Alegre, Brazil. Weitzman, S. H. 1962. The osteology of Brycon meeki, a generalized characid fish, with an osteological definition of the family.—Stanford Ichthyolog- ical Bulletin 8(1):1—77. , & W.L. Fink. 1983. Relationships of the neon tetras, a group of South American freshwater fishes (Teleostei, Characidae), with comments on the phylogeny of the New Word characi- forms.—Bulletin Museum of Comparative Zo- ology 150(6):339-395. 3)5)5) , & S. V. Fink. 1985. Xenurobryconin phylog- eny and putative pheromone pumps in Glan- dulocaudine fishes (Teleostei: Characidae).— Smithsonian Contributions to Zoology 421:1— Zits , & L. R. Malabarba. 1998. Perspectives about the phylogeny and classification of the Chara- cidae (Teleostei: Characiformes). Pp. 161—170 i ake WMialabarbas Re Es Reis) Re iP) Van, Z.- M. S. Lucena and C. A. S. Lucena, eds., Phy- logeny and classification of neotropical fishes. Porto Alegre, Brazil. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):356—379. 2003. Geographic variation in Poecilia Bloch and Schneider, 1801 (Teleostei: Poeciliidae), with descriptions of three new species and designation of lectotypes for P. dovii Gunther, 1866 and for P. vandepolli van Lidth de Jeude, 1887 Fred. N. Poeser Zoological Museum of Amsterdam, Department of Ichthyology, PO. Box 94766, 1090 GT Amsterdam, The Netherlands Abstract.—The South American cyprinodontiform fish species with the ver- nacular name ““mollies”’ are analyzed and three new species of the genus Poe- cilia are described and figured, viz., P. boesemani, n. sp. from Port of Spain, Trinidad, P. koperi, n. sp. from coastal areas of Venezuela and Colombia, and P. wandae, n. sp. from the Zulia district, Venezuela, west of Lake Maracaibo. Thirteen species of Poecilia are presently recognized from the northeastern part of South America and adjacent islands, viz., six mollies Gncluding P. vivipara and P. caucana), five micropoeciliids, P. heterandria and the guppy (P. retic- ulata). Two different populations in both P. koperi and P. wandae are recog- nized, based on variation in pigmentation. From the type series of P. dovii Ginther, 1866 and P. vandepolli Van Lidth de Jeude, 1887 lectotypes are se- lected. Four subspecies of P. gillii are recognized as morphologically and geo- graphically distinct populations. Poecilia reticulata, P. heterandria, P. bifurca, P. parae, P. picta, P. branneri and P. minima are not considered herein, whereas P. /aurae remains a species inquirendae. A geocline in characters of the species of Poecilia is recorded and an evolutionary scenario is discussed. This paper seeks to extend our biogeo- graphic knowledge of the genus Poecilia Bloch & Schneider, 1801 and describes three new species from the southern end of the range of this genus. The present study records seven species from the northeastern part of South America. Whereas the number of poeciliid species constitutes a dominant part of the Central American fish fauna, this is not the case in South America. Miller (1983) constructed a key and checklist for the Mexican species, but there are no com- prehensive lists for the species of Poecilia from Central and South America. Speci- mens of the species herein described as new, viz., P. koperi, P. boesemani and P. wandae, were previously identified as either P. vivipara Bloch & Schneider, 1801; P. sphenops Valenciennes, 1846 (Regan 1913: 1013; Price 1955:18; Boeseman 1960:122); P. vandepolli van Lidth de Jeude, 1887 (de Beaufort 1940:111; Schultz 1949:84, 97— 99; Feltkamp and Kristensen 1969); or mentioned as ‘unidentified species’ (Rosen & Bailey 1963:48). Regan (1906-08) examined variation within the species of Poecilia and extended our knowledge of the expression of mor- phology and pigmentation (Regan 1906—08, fig. 13), as well as the geographic range (from northern Mexico to the Leeward Is- lands) of the P. sphenops complex. With the discovery of the structures of the gonopodial tip as a taxonomic tool (Ei- genmann 1907), Regan (1913) reorganized species groups in the subfamily Poeciliinae. VOLUME 116, NUMBER 2 With respect to the P. sphenops complex, he allocated P. sphenops to the genus Mol- lienesia on the basis of identical gonopodia of M. latipinna and M. sphenops. Moreover, the number of dorsal fin rays of M. formosa was intermediate between the numbers found in M. latipinna and M. sphenops, which therefore rendered the character in- valid. Although Regan did not mention P. vandepolli, the geographic range of his M. sphenops still included the Leeward Is- lands, i.e., the Lesser Antilles. The importance of the gonopodial tip was also recognized by Hubbs (1924), who used the gonopodium as “‘the chief distinc- tive feature of the Poecillidae (as here de- limited).”” The M. sphenops group was de- fined by Hubbs (1926) as “‘the multitude of races allied to, or inseparable from M. sphenops.” This group was separated from M. latipinna and allies by the number of dorsal fin rays. For M. sphenops, Hubbs was “unable to delimit, in either distribu- tion or in characters, any multitude of ele- mentary species making up the sphenops complex of Middle America. The problem of determining the relationships of the di- verse types, many of which have received specific names, is in prospect a most fas- cinating study.”’ The situation remained promising, even after the revision of Rosen and Bailey (1963), who listed a total of 35 nominal species in their synonymy of P. sphenops. Schultz and Miller (1971) mentioned: “... a thorough study of the whole complex and the type specimens will be required before systematic units and the nomenclature can be convincingly coordinated.”’ Species rec- ognition was aided by dental analyses (Schultz and Miller 1971), i.e., unicuspid and tricuspid species were recognized. Oth- er alpha-taxonomic features (meristic and morphologic characters, body and fin color, even allozyme data) remained confusing. Rivas (1978) remarked that “proportional body measurements are [to be] omitted... , there is considerable variation in characters individually, ontogenetically, seasonally, 357 geographically, and environmentally and, therefore, they are of little or no value in distinguishing species (except in relative fin position).”’ Extensive and detailed investi- gations (Schultz and Miller 1971, Menzel and Darnell 1973, Miller 1975) resulted in a checklist for the Central American mollies (Miller 1983), which included eight species of Central American short-finned mollies. No species were mentioned from outside this range. Comparisons of the Mexican species to those of Central and South Amer- ica (Poeser 1992, 1995, 1998) have led to the present study. Methods Some 17 lots of Poecilia vandepolli, in- cluding the type series of P. vandepolli van- depolli and P. v. arubensis, have been re- examined. From the type series, a lectotype is selected. Meristic data and inner jaw den- tition (cf. Schultz and Miller 1971) received special attention. Vernier calipers were used to record distances to 0.1 mm. The average of the measurements and counts are given and compared to a similar study of Felt- kamp and Kristensen (1969) and to the data of P. vivipara and of P. gillii. In the description of the new species, proportional morphometric values (Table 2) are recorded in thousandths of the standard length (SL), following Miller (1975). The small size of P. wandae made measure- ments difficult to impossible and therefore some were omitted. Meristic characters fol- low Hubbs and Lagler (1947). In the type series of P. wandae, all specimens have damaged caudal fins; therefore, the caudal fin ray count is estimated. Terms concern- ing the shape of inner jaw dentition follow Garman (1895). Melanophore pigmentation of preserved specimens and gonopodial structures are also recorded (Figs. | to 4). The combined data are used to evaluate relationships between the newly described species. A key to the species is provided in Appendix 1. 358 The following abbreviations are used in the species diagnosis provided below. A = anal fin rays BS = scales around the body CPD = least depth of the caudal peduncle CPS = scales around the caudal peduncle D = number of dorsal fin rays G = gonopodial ray LLS = lateral scales P = pelvic fin rays PL = predorsal length PS = predorsal scales SL = standard length Systematics The genus Poecilia is diagnosed by the shape of gonopodial ray 4p, which has six to fourteen unserrated distal segments, fol- lowed by dorsally serrated subdistal seg- ments (cf. Miller 1975). The genus consists of several ill-defined subgenera. A more de- tailed analysis is in progress. In the subgenus Poecilia, ventral spine- like serrae are present on the third gono- podial ray. Gonopodial rays 4a and 4p are very similar, as are rays 5a and 5p. The P. sphenops species group has terminal seg- ments of gonopodial ray 4a and of gono- podial ray 4p close to each other, giving the gonopodium a sharp appearance. The mem- bers of this species group are moderately sized to large, with nuptial specimens usu- ally larger than 35 mm SL. The members of the P. caucana species group are defined by the relative thickness of gonopodial ray 4p versus 4a. The tips of these rays are split, making the gonopodium blunt. Nup- tial specimens are less than 35 mm SL. Poecilia vivipara Bloch & Schneider, 1801 Poecilia vivipara Bloch & Schneider, 1801: 452, pl. 86, Fig. 2 (type locality: Suri- nam). Poecilia surinamensis Humboldt & Valen- ciennes, 1821:158 (type locality: Suri- nam, French Guyana, Brazil). Poecilia unimaculata Humboldt & Valen- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ciennes, 1821:158 (type locality: Brazil, Rio de Janeiro). Poecilia schneideri Humboldt & Valenci- ennes, 1821:159 (type locality: Surinam). Neopoecilia holacanthus Hubbs, 1924:11 (type locality: Puerto Rico, introduced). Material.—West Indian Antilles: ZMA 120.384 (22), St. Lucia, freshwater pool, +1 km west from Fort Vieux, coll. J. H. Stock, 20-II-1974. Guyana: BMNH 1974. 10. 10. 527-619 (7 out 93), Georgetown, Seawall trench, no date; CAS 59364 (12), Georgetown trench- es, C. H. Eigenmann 1908; CAS (SU) 21784 (2), same data as CAS 59364; USNM 66278 (2), same data as CAS 59364; ZMA 100.629 (2), same data as CAS 59364; ZMA 119.912 (3), East Coast Demerara, Turkeyen, drainage canal, coll. M. Tamessar, 28-I-1987; ZMA 119.913 (6), East Coast Demerara, Bel Air, stagnant ditch, coll. M. Tamessar, 28-I-1987; ZMA 119.917 (4), East Coast Demerara, Indus- try, shallow pool, coll. M. Tamessar, 28-I- 1987; ZMA 121.005 (5), Guyana, coll. E Vermeulen, no date; ZMA 121.007 (15), Guyana, coll. E Vermeulen, 1992; ZMA 121.009 (5), Guyana, coll. EF Vermeulen, 1992. Surinam: MNHN B. 932 (5 syntypes of P. surinamensis), Surinam River, Levail- lant, no date; MNHN 4391 (48 syntypes of P. surinamensis), Cayenne, Rousseau, no date; MNHN B. 918 (60), without exact lo- cality, Duvernoy, no date; RMNH 18516 (2), Surinam, coll. J. Th. Noordijk, August 1887; ZMA 105.332 (2), Paramaribo-west, garden canals, coll. H. Nijssen, 12-XII- 1966; ZMA 106.757 (16), Marowijne Riv- er, coastal plain on Tijgerbank, west of mouth, coll. H. Nijssen, 1-IV-1966; ZMA 115.118 (45), Surinam district, Lagoon 7 & 8, 05°59’N, 54°49’W, coll. M. P. Panday, 22-IV-1974; ZMA 115.121 (20), same data as ZMA 115.118; ZMA 119.908 (6), Suri- nam River, pool near beach at Braamspunt, coll. I. Kristensen, 14-III-1960; ZMA 121.000 (4), Paramaribo, Mason Street, pol- VOLUME 116, NUMBER 2 luted trench, coll. EF Vermeulen & W. Suy- ker, 24-II-1991. French Guiana: USNM 121833 (9), Cay- enne, Carriera Guila, S. E Yolles 1945; USNM 121834 (2), Cayenne, Pont Magie, S. E Yolles 1945; USNM 121835 (9), Cay- enne, Ruisseau de l’institute-eau douce, S. E Yolles 1945; USNM 149938 (4), Cay- enne, S. FE Hildebrand 1945. Brazil: RMNH 2724 (2 syntypes of P. unimaculata), Rio de Janeiro, coll. Dela- lande, no date; ZMA 100.628 (1), Brazil, coll. Moesch, 1884; ZMA 116.210 (14), Rio Grande do Norte, Rio Acu (Rio Assu), downstream of Macau, coll. R. Boddeke, 10-I-1979. Diagnosis.—A = 8; D = 7; C = 16-18; LLS = 24-26; CPS = 16. Poecilia vivipara may have a midlateral spot or blotch in both sexes, a unique character in the subgenus. The gonopodium has no extruding hooks or spines, gonopodial ray 4a with serrae on dorsal surface. Description.—Medium sized species, mature males smaller than 50 mm and fe- males smaller than 60 mm. The body is truncate and displays dark stripes on the sides. The caudal and dorsal fins have black markings, with broad yellow margins. The caudal fin has black margins, at the base broader than at the terminal end. There is some variation in the presence or absence of the spot at the side of the body. No ex- tensive records are available of the extent of this variation. In the populations in which it is present, the blotch is also prom- inent in young specimens. The gonopodium is figured in Miller (1975), the presence of dorsal serrae on ray 4a are unique within the subgenus. Distribution.—Poecilia vivipara is found in coastal habitats from Venezuela to Ar- gentina. It is also found on some islands of the Lesser Antilles. Remarks.—Garman (1895) diagnosed this species and provided a list of syno- nyms. He also explained the obvious aber- rant figure in the original description. Hubbs (1926) synonymized Neopoecilia 359 holocanthus, correcting his misidentifica- tion. Poecilia mexicana Steindachner, 1863 Poecilia mexicana Steindachner, 1863:178, pl. 4, fig. 1, la (type locality: Mexico, Orizaba). Poecilia thermalis Steindachner, 1863:181, plate 4, fig. 3, 3a (type locality: Central America, warm springs). ?Gambusia (?) modesta Troschel, 1865:105 (type locality: Mexico). ?Gambusia (?) plumbea Troschel, 1865:106 (type locality: Mexico). Poecilia chisoyensis Gtinther, 1866:342 (type locality: River Chisoy, Vera Paz). Poecilia dovii Giinther, 1866:344 (in part; type locality: Mexico; lectotype is P. gil- lii). Poecilia limantouri Jordan & Snyder, 1900: 116-117, 129-131 (type locality: Tam- pica, Tamaulipas, Mexico). Mollienesia sphenops vantynei Hubbs, IOS seilil, jolawe A, tie. Ik Cajoe leecalliin Guatemala, Uaxactum, Rio Hondo). Mollienesia sphenops macrura Hubbs, 1935:12, plate 2, figs. 2—3 (type locality: Guatemala, Rio San Pedro de Martir). Mollienesia sphenops altissima Hubbs, 1936:239, plate 9, figs. 1-3 (type locality: Mexico, Yucatan Peninsula, Miramar Spring). Mollienesia sphenops melanistia Hubbs, 1937 (type locality: Mexico, Tamaulipas, Arroyo Marmoleyo). Material.—Barbados: BMNH 1970.1.29: 1 (1) Barbados, coll. R. Heath, no date; RMNH 24814 (55), rivulet near Three Mills CBO NS 59227 W)) collhale Keistensent 13.6.1961; RMNH 24804 (6), rivulet at Three Mills (13°10’N, 59°27’W), coll. I. Kristensen, 13.6.1961; RMNH 24809 (5), rivulet near Three Mills (13°10’N, 59°27'W), coll. I. Kristensen, 13.6.1961. Diagnosis.—A = 9; D = 9-10; C = 18- DD-MaeS =" 26-27 ACERS) = 18: Description.—Poecilia mexicana 1s rath- er variable in its morphology. This is a trun- 360 cate, torpedo shaped species in northern Mexico but ranges to high, laterally flat- tened specimens at the eastern end of the range in the Yucatan Peninsula. While the northern populations possess deeply blue pigmented bodies, alpha-males of the Yu- catan populations have a more copper-red body. The fin pigmentation is also varying, from totally black at the base of the fin with an orange outer area (in northern popula- tions) to a black and yellow reticulate pat- tern (on the Yucatan Peninsula). Between these extremes clinal variation is noted. In the Barbados material, both males and females exhibit considerable variation in their pigmentation, especially the speci- mens in RMNH 24804 have many spots on their body (perhaps a basis for why they were separated from the other lots). Fe- males have nine anal fin rays, some females have 10 dorsal fin rays. The membranous hook on gonopodial ray 3 is extremely small, without a bony spine. Several spec- imens have 20 caudal fin rays, whereas all specimens have 18 scales around the caudal peduncle. The diagnostic features of P. mexicana generally most closely agree with the nom- inal subspecies P. gillii gillii (see below), with the exception of the number of scales around the caudal peduncle (18 versus 16 in P. gillii). Distribution.—Poecilia mexicana occurs on the Atlantic coast of Central America, from the Texas border, through Yucatan into Guatemala and Costa Rica. The population found on Barbados is probably derived from escaped aquarium specimens, which is confirmed by their abnormal variability in body pigmentation. Poecilia gilli (Kner & Steindachner, 1864) Xiphophorus Gillii Kner and Steindachner, 1864: in Kner & Steindachner, 1865:25 (type locality: Panama, Rio Chagres); Poecilia gillii; Giinther, 1868:395; Poe- cilia sphenops gillii, Hubbs, 1953:145. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Poecilia mexicana {non Valenciennes, 1863]; Poeser, 1992:86 (misidentifica- tion). Poecilia thermalis [non Steindachner, 1863] Giinther, 1866:341 (type locality: San Salvador, warm springs). Poecilia dovii Gtinther, 1866:344 (in part: type locality: Lake of Nicaragua). Platypoecilus mentalis Gill, 1876:335 (type locality: Isthmus of Panama). Poecilia Boucardi Steindachner, 1876:386 (type locality: Colon, Panama). Poecilia cuneata Garman, 1895:179, plate V (type locality: Colombia, Gulf of Ur- aba, Turbo); Poecilia sphenops cuneata; Hubbs, 1926b:77. Poecilia salvatoris Regan, 1907:65, plate 14, figs. 2-3 (replacement name for P. thermalis Giinther, 1866). Platypoecilus tropicus Meek, 1907:146 (type locality: Costa Rica, Turrialba). Poecilia tenuis Meek, 1907:147 (type lo- cality: Costa Rica, Tiribi). Poecilia caudata Meek, 1909:209—210 (type locality: Costa Rica, Turrubares). Poecilia spilonota Regan, 1908:460 (type locality: San José, Costa Rica). Lembesseia parvianalis Fowler, 1949:267— 269 (type locality: Africa, Congo system, Oka). Mollienesia sphenops petersi Schindler, 1956:1—4, fig. 1 (type locality: Honduras, lake Yojoa). Material.—Nicaragua: Lectotype of P. dovit (BMNH 1863.12.16.77, Lake of Nic- aragua, coll. captain J. W. Dow, no date). 5 Paralectotypes of P. dovii, (BMNH 1863.12.16.78-92, same data as BMNH SOD l Ao Os 17). Costa Rica: 4 Syntypes of P. spilonota (BMNH 1907.2.11.44-50, San José, coll. P. Bidley, no date). Panama: NWM 21608 (1 syntype of Xiphophorus Gillii), Rio Chagres, Panama, no further data. MCZ 29433 (10), 1 mile south of Panama City, coll. USFC Steamer Albatross (Alex. Agassiz), 23-X-1904; MCZ 33847 (10), ca. 9°43'N, 79°43’W, be- VOLUME 116, NUMBER 2 tween Gorgona and Matachin, collected be- fore 1930; MCZ 54068 (10), small tributary of main stream on left about 0.75 km up- river of bridge, Bayano drainage, probably Rio Canita, Panama, coll. W. L. Fink & K. E. Hartel, 1-IV-1978; USNM 050368 (14), Panama, coll. C. H. Gilbert; USNM 64764 (17), Folks R. Swamp, Cristobal, coll. A. H. Jenning, 4-VIII-1909; USNM 65618 (15), 1 mile south of Panama city, coll. Str. Albatross, 23 X-1904; USNM 78837 (27), Upper Trinidad, coll. Meek & Hildebrand, 7-11-1911; USNM 247529, (24) canals in banana field, Bocas Province, California, coll. Loftin, 2-I[X-1962; USNM 247531, (44) Rio Gaurumo, Bocas Province, coll. Loftin, 18-IV-1963; USNM 247432 (10 of 50), Canal zone, behind Fort Clayton, res- idential area, coll. Loftin & Tyson, 19-IV- 1962; USNM 247436, (6) Bocas del Toro, Esendo de Verequas island, coll. J. Legler, 14-V-1962; USNM 247548, (35) San Blas, small river opposite to Mulatupo island, coll. Loftin & Evermann, 2-XII-1962; USNM 247550, (18) San Blas, Rio Acla, coll. Loftin, 16-XII-1962; USNM 293473, (10 of 51) 9°14’N, 78°58’W, Rio Tearbles, Bayano drainage, Panama province, coll. W. C. Sternes et al., 25-II-1985; USNM 293476, (23) 9°28'N, 79°3'’W, Comarca Kuna Yala, Rio Mandinga, coll. W. C. Ster- nes et al., 5-IHI-1985; USNM 293494 (42), Panama province, Rio Frijoles, above pipe- line rd., N. of Gamboa (Rio Chagres drain- ase. Atlantic side) colly W. ©) Stemies ‘et al., 26-II-1985. Colombia: 2 Syntypes of P. cuneata (MCZ 6458, Turbo, Gulf of Uraba, Colom- bia, coll. T. Barbour, no date); 3 Syntypes of P. cuneata (USNM 120285, same data as MCZ 6458); 5 Syntypes of P. Boucardi (MCZ 32959, San Pablo (Aspinwall (Co- lon), Quebrada San Pablo), coll. Hassler Expedition (Steindachner & Mr. Boucard), VII-1872); BMNH 99.3.15.27-29 (3), Mon- key Hill, Colon, coll. Dr. H. Festa, no date. Additional material from Nicaragua and Panama in Poeser (1992), from El Salvador in Poeser (1995). 361 Diagnosis.—A = 9; D = 9-10; CPS = 16; LLS = 26—29. In his checklist, Miller (1983) was uncertain if P. gillii was differ- ent from P. mexicana. Therefore, I have in- cluded a discussion of this species in the remarks section. Description.—This species is very simi- lar to P. mexicana, with the exception of the number of scales around the caudal pe- duncle. In P. mexicana, 18 scales around the caudal peduncle are common. Remarks.—All\ examined type material is consistent with the above meristic data. Over its extensive range, P. gillii shows considerable variation in body shape and in dorsal fin color, and moderate variation gonopodial features. These variations ap- pear to reflect intraspecific diversification, justifying taxonomic separation, 1.e., divi- sion in subspecies. Carr & Giovannoli (1950:17—18) reported the live colors of P. gillii and the sympatric P. marcellinoi Poe- ser, 1995 from Honduras: “‘“One male with yellow spots, a black blotch at the caudal base and spotted dorsal and caudal fin, and one female with black spotted sides (=P. marcellinoi). One female with yellow spot- ted sides, and a male with golden-orange blotches, with a black blotch on the dorsal base and rest of fin orange (=P. gillii).” Poecilia gillti salvatoris is reported from El Salvador, where males from most pop- ulations are reported to have red dorsal fins (Hildebrand 1925). This prompted Miller (1994) to redescribe P. salvatoris Regan, 1907, although all other characters are as found in the present study for P. gillii. Poe- ser (1995) in his redescription of Poecilia salvatoris agreed with the diagnosis of P. salvatoris by Regan (1908), with the excep- tion of the number of anal fin rays. Regan (1908:104) mentioned 8—9 anal fin rays; however, examination of 14 of his syntypes did not yield any specimen with eight anal fin rays, so Regan’s account is judged er- roneous. In the preserved material, no trace of red was found in the dorsal fins. The red finned western Central American popula- 362 tions are here considered as a subspecies, viz., P. gillii salvatoris. Villa (1982) provided a key to the genus Poecilia in Nicaragua, in which he recog- nized three species, viz., P. gillii, P. sphen- ops, and “‘an undescribed species with uni- cuspid teeth”. He mentioned (p. 134, trans- lated from Spanish): P. gillii: “‘Unicuspid inner teeth, incomplete supraorbital system, preorbital pores free, 26—28 (modally 27) lateral scales, 16 scales around caudal peduncle, 28-30 (modally 30) vertebrae. Guatemala, El Salvador, to Panama’’. P. spec.: ““Unicuspid inner teeth, com- plete supraorbital system, free preorbital pores, 28-30 (modally 29) lateral scales, 18 scales around caudal peduncle, 28—30 (modally 30) vertebrae. Rio Ulya (Hondu- ras), Lagunas Apoyeque and Xiloa, lake Managua and Nicaragua, and Rio Sapoa & Frio, Tilaran region, Costa Rica’’. P. sphenops: ““Tricuspid inner teeth, in- complete supraorbital pores, preorbital pores covered, 26—28 (modally 27) lateral scales, 16-18 scales around caudal pedun- cle, 28—30 (modally 29) vertebrae. Guate- mala and El Salvador, Atlantic side of Hon- duras, Nicaragua including the Great Lakes’’. Villa’s diagnosis of P. sphenops fits the description of P. marcellinoi (tricuspid in- ner teeth, 16 scales around the caudal pe- duncle). While his description of P. gillii is accurate, his figure of P. gillii shows 9 transverse scales on the caudal peduncle, which is a diagnostic character of P. mexi- cana. | suspect the figures were switched. His undescribed species might very well be P. mexicana. The apparently aberrant num- ber of lateral scales (also mentioned by Bussing [1987]) is explained by character displacement (cf. Poeser 1995). The only synonym of P. sphenops in Rosen & Bailey (1963) from Nicaragua is P. dovii Giinther, 1866. Poecilia dovii was considered syn- onymous with P. sphenops by Regan (1908). Since the type material of P. dovii contained specimens from P. mexicana PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (with 18 scales around the caudal pedun- cle), as well as from P. gillii (with 16 scales around the caudal peduncle), it is only par- tially alluded to the synonymy of P. gillii. In honor of captain J. W. Dow, the lectotype is selected from the lot he collected himself in Nicaragua, which is P. gillii. Giinther (1866) mentioned that P. dovii occurred only in Guatemala and Mexico. However, since the type locality includes Nicaragua, this country should be included in the range of distribution. The specimens from Lake Amatitlan (BMNH_ 1865.6.10.13-15) are not registered in the British Museum as syntypes, and these were not studied. The specific status of these types remains to be determined. Poecilia gillii appears to be nearly the only species of Poecilia present in Costa Rica. Meek (1914:116—117) placed all Cos- ta Rican taxa in synonymy with Platypoe- cilus tropicus (=P. gillii) as follows: “‘The inland or fresh-water forms of this species or variety found in Costa Rica are very var- iable, and as a result several species have from time to time described.... The fe- males and many males of these inland forms usually have a black spot at the base of the middle dorsal rays, .... On many specimens from salt and brackish water some of the scales have a dark spot which forms lines along the rows of scales... . On many of the males, especially from larger streams, the basal half of the caudal is black, or with black blotches; on some of these the basal half of the dorsal is also black.’ Specimens with a black basal half of the caudal fin, formerly described as P. caudata Meek, 1909, are herein recognized as a subspecies, viz., P. gillii caudata. Bussing (1987) identified most of the Costa Rican populations as P. gillii. He gave the following description (translated from Spanish): “‘Body with yellow spots, in some females these spots are black. Scales in a lateral series 26—28, mostly 27. Three orbital pores in one line. In the males some- times a large dorsal fin, with black spots or VOLUME 116, NUMBER 2 blotches at the base. Caudal fin with spots, blotches or solid black pigmentation. Some males with orange in their caudal fin. Other fins yellow, head and body bluish. Note: some males have a red dorsal. Large spe- cies, up to 105 mm.” Males possessing a red dorsal fin are P. gillii salvatoris. Bus- sing also recorded P. mexicana, in a much lower frequency, i.e., three populations from over fifty in total, widely apart from each other (Bussing 1987:144, map 20). His diagnosis for this species is: ““Very much like P. gillii. Yellow spots on body, black in some females. 28—30 (modally 29) Lat- eral scales, and three orbital pores, forming a triangle. Dorsal and caudal spotted, rest of fins yellow. It is also a large species, up to 110 mm.”’ Bussing (1987) illustrated P. gillii and P. mexicana in his paper, and his diagnoses of P. gillii and P. mexicana cor- respond with the account of Villa (1982) (see above). Surprisingly, Bussing did not record any populations of P. marcellinoi. Hubbs (1926, 1953) considered P. sphen- ops in Panama either P. sphenops cuneata (cf. Hubbs 1926), or P. s. gillii (cf. Hubbs 1953). Examination by me of Panamanian populations confirms earlier findings of profound morphometric differences, as well as constant meristic data, in all Panamanian populations. A conspicuous reduction of the gonopodial spine on ray 5 is noted. In pop- ulations near the Costa Rican border (Bocas del Toro district of Panama) the spine is distinct. In populations near Colombia, 1.e., the San Blas area, it is reduced or absent, as is found in P. koperi (Fig. 4a). This shift in character expression from west to east is not accompanied by meristic changes. Although some populations contain poor- ly pigmented, slender specimens, other pop- ulations manifest heavily pigmented, stout specimens. Nevertheless, all specimens have nine anal fin rays, nine (or rarely ten in the San Blas district) dorsal fin rays, 16 scales around the caudal peduncle, 18 scales around the body, and 26 to 29 scales in a lateral series. The consistency of these 363 counts suggests that all populations belong to the same species. Because the missing spine is of taxonom- ic significance, this form was named P. gil- lii cuneata Garman, 1895. The populations of P. gillii from Panama examined in the present study are partly sympatric with P. marcellinoi like in El Salvador (Poeser 1995). Poeser (1992) reported a male with an aberrant gonopodium from Nicaragua (GCRL 6697), identified then as P. mexi- cana mexicana. Re-examination of my notes proved that the identification and lo- cality was wrong. The sample containing this male was GCRL 8748, viz., P. gillii cuneata from Panama. Rosen & Bailey (1963) considered Lem- besseia parvianalis a synonym of P. sphen- ops. However, since L. parvianalis has uni- cuspid inner teeth, Miller (1983) placed it in the synonymy of either P. mexicana or P. gillii. Based on the original description, in which 8 transverse scales on the caudal peduncle are illustrated, I assign it to the synonymy of P. gillii. Fowler’s (1949) re- cord is important for its taxonomic value. The P. sphenops group, if raised to generic level, would become Lembesseia (with L. surinamensis (=P. sphenops) as type spe- cies). Distribution.—Poecilia gillii is recorded from the Pacific coast of Guatemala to the Atlantic coast of Colombia. In northern re- gions, from Guatemala to Costa Rica, it is represented by the subspecies P. gillii sal- vatoris. The subspecies P. gillii caudata oc- curs in Costa Rica, whereas P. gillii gillii 1s present in Panama. The South American populations, as well as adjacent populations in Panama, are considered to be P. gillii cuneata. The type locality of P. cuneata was designated “Turbo, Gulf of Darien” (Garman 1895). Rosen and Bailey (1963) added ‘Panama’ to this locality. However, study of several maps did not confirm this addition. The nearest village with that name is near the Gulf of Uraba, an extension of, and sometimes also so-called, the Gulf of 364 Darien, in Colombia. I presume that this is the correct type locality. Poecilia caucana (Steindachner, 1880) Girardinus caucanus Steindachner, 1880: 87, plate 4, figs. 4, 5 (type locality: Co- lombia, Caceres); Allopoecilia caucana Hubbs, 1924:11. Material.—Colombia: UMMZ 186317 (51), Depto. Cordoba, Tierra Alta, Rio Sinu, coll. W. Moberley & K. Adler, 1965. Venezuela: UMMZ 186931 (54), Rio Monay, coll. E E Bond, 1938; UMMZ 186934 (24), Quebrada Goayjira, EF E Bond, 1938; UMMZ 186937 (44), Rio Bucares, coll. E E Bond, 1938; USNM 86264 (4), Valeras, Rruyjililo;ncoly re Pitteim (9235 USNM 121677 (33), Rio Motatan, 4 km above Motatan, coll. L. P Schultz, 25-III- 1942. Panama: USNM 293444, (7), Rio Meteti, Darien Province, 40 km NNW of Yaviza, coll. W. C. Sterns et al., 24-II-1985; USNM 293574, (10), Rio Peresinico, Darien Prov- ince, coll. B. Chernoff, J. Lundberg, L. McDade, 23-II-1985. Diagnosis.—A = 8; D = 7-8; C = 18- 22 NAS; =" 26-27: (CS = 6; CRS aI Poecilia caucana is a small to medium sized species, characterized by a black band in its dorsal fin. The gonopodium has ray 4p with broader rays than ray 4a. Description.—This is one of the smaller species of the genus. The largest specimens examined are a female of 37.4 mm SL and a male of 27.2 mm SL. The dorsal fin has a black transverse band at the base. This fin is usually milky white or yellow, although also red fins are reported in aquarium lit- erature. The other fins are unmarked. The gonopodial characters figured in Ro- sen and Bailey (1963) are of considerable interest because gonopodial ray 4p is thick- er than ray 4a. In addition, the hook on gon- opodial ray 5 is weakly developed. Distribution.—Atlantic slopes of Pana- ma, Colombia and Venezuela. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Poecilia vandepolli van Lidth de Jeude, 1887 Poecilia Vandepolli van Lidth de Jeude, 1887:137, Pl. 2, Figs. 4 and 5 (type lo- cality: Curacao). Poecilia Vandepolli arubensis van Lidth de Jeude, 1887:138, Pl. 2, Figs. 6-10 (type locality: Aruba). Lectotype.—Curagao: RMNH 5155, adult male, Curacao, coll. Neervoort and v. d. Poll, no further data. Paralectotypes.—RMNH 33843 (4), same data as RMNH 5155. Material.—Aruba: RMNH 5156 (6 syn- types of P. Vandepolli arubensis), Aruba, coll. Neervoort and v. d. Poll, no further data. See also tables below, details in Poe- ser (1992). Diagnosis.—A = 8-9; D = 7-8; C = 16; LLS = 26-27; CPS = 16. Several speci- mens have humeral blotches anterior to the position of the lateral side spot as occurs in P. vivipara. The gonopodium lacks extru- sions on ray 3 and 5p. Description.—Poecilia vandepolli is re- ported to have specimens with orange on the ventral side of the body, yellow speci- mens with blue sides and grayish brown specimens in the same population (Felt- kamp and Kristensen 1969). Speckled spec- imens, allegedly only occurring in Vene- zuelan populations (Feltkamp and Kristen- sen 1969), also appear on Curacao. The pig- mentation at the base of the dorsal fin forms a blotch, in addition to dark spots. The body may have a humeral blotch, positioned more anteriorly than a similar such patch of color in P. vivipara. The females tend to be paler than males. The gonopodium, figured in Poeser (1992), is like that of P. vivipara, with the exception of the serrae on ray 4a. When fully developed, a little membranous bulge is found on gonopodial ray 3, which covers extruding serrae. Description of the types.—The lectotype is a mature male, 27.3 mm SL. It has 8 dorsal fin rays and 16 scales around the VOLUME 116, NUMBER 2 caudal peduncle. Several scales in the lat- eral series are missing on this specimen and the caudal fin is damaged, so no further me- ristic data can be given. Gonopodial ray 3 has a long terminal segment, the hood is short. An extremely short extrusion similar to a small hook is present. Gonopodial ray 4a is unmodified, ray 4p has eight unmod- ified distal segments followed by serrated segments. Gonopodial ray 5a is one seg- ment longer than 5p. Subdistal ventral mod- ifications occur on ray 5a. The largest fe- male paralectotype, 47.5 mm SL, has about 25 scales in a lateral series (this number is not accurate because some scales are miss- ing). It has nine dorsal fin rays and nine anal fin rays, the caudal fin is damaged. The specimens in the type series from Aruba are much smaller. The largest male, 22.0 mm SL, has a gonopodium like the lectotype, with seven distal unmodified seg- ments on gonopodial ray 4p and no trace of a hook on ray 3. It has 8 dorsal fin rays, 16 scales around the caudal peduncle, and about 24 scales in a lateral series (Some scales are missing). The caudal fin is dam- aged. This male has a pigmented band proximally on the dorsal fin and a pig- mented humeral blotch. The largest female (26.5 mm SL) has eight anal fin rays and nine dorsal fin rays. The caudal fin is dam- aged. Sixteen scales are found around the caudal peduncle and 27 in a lateral series. To establish morphometric and meristic variations a detailed examination of 15 pop- ulations was carried out. Some 12 lots were examined from Aruba, and three lots from Curacao (Table 1). Distribution.—Poecilia vandepolli oc- curs naturally in all kinds of waters of the Netherlands West Indies (Aruba, Curacao, and Bonaire). It is introduced on St. Maar- ten/St. Martin (Poeser 1992). Poecilia koperi, new species ene, Il, Walole 2 Poecilia vivipara (non Bloch & Schneider, 1801); De Beaufort, 1940:111. 365 Poecilia sphenops (non Valenciennes, 1846); Regan, 1913:1013 (in part). Poecilia sphenops vandepolli (non Van Lidth de Jeude, 1887); Schultz, 1949:84, 97-99 (in part). Poecilia sphenops cuneata (non Garman, ss lalulolos, WOAG37 7. Holotype.—Adult male, UMMZ 223343, Venezuela, Rio Curipe at Higuerote, coll. EF FE Bond, 2-V-1938. Allotype.—Adult female, UMMZ 223344, same data as UMMZ 223343. Material.—BMNH 1909.2.25: 53-56 (5), Venezuela, coll. Arnold, no date; UMMZ 200738 (15 of 123), 2 km N of Ocumare, lagoon on flats near mouth of Rio Cumboto, coll. EF EF Bond, 5-I-1938; UMMZ 200740 (15 of 57), Rio Guaiguaza, 3 km W of Por- to Cabello, 2 km from mouth of river, coll. FE E Bond, 15-I-1938; UMMZ 200744 (12), Rio Sanchon, 5 km W of Tavorda, 10 km W of Porto Cabello, coll. EF E Bond, 26-I- 1938; UMMZ 200753 (15 of 430), Lagun- ita, 5 km W of Coro, Estado Falcon, coll. E E Bond, 19-III-1938: UMMZ 200755 (15 of 115), Falcon, Laguna del Rio Capatarida, at mouth, 5 km N of Capatarida, coll. EF FE Bond, 2-III-1938; UMMZ 200760, (15 of 152), Falcon, Coastal lagoons, 15 km N of Maracaibo, coll. KE E Bond, 6-IV-1938; UMMZ 200761 (13 of 202), same data as UMMZ 223343; UMMZ 200762 (15 of 80), Estado de Miranda, Lagunita de Taca- rigua, at Tacarigua, 85 km E of Caracas on the coast, near the boca, coll. E E Bond, 3- IIl-1939; UMMZ 200764 (28), Boca del Rio Cumboto, 2 km NW of Ocumare; ZMA 109.206 (6), La Goajira, Rio Calancala, San Antonio, coll. P. Wagenaar-Hummelinck, 17-I-1937; ZMA 119.909 (30), Paraguana Estangue de Moruy, coll. P. Wagenaar- Hummelinck, 18-II-1937; ZMA 119.910 (15), Paraguana, Estangue de Santa Ana, coll. P. Wagenaar-Hummelinck, 16-I-1937; ZMA 120.885 (4), 2 km N of Barcelona, Rio Guanta, coll. P. Wagenaar-Humme- linck, 1936. 366 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Overview of standard lengths, proportional measurements and meristic data of Poecilia vandepolli. SL Pred CPd A D C LLS CPS Aruba ZMA 100.600. The fish were caught in a small freshwater stream, Rooi Prins at “Plantage Fontein’. The lot contains small specimens, of which the largest male (20.7 mm SL) has an unclear humeral spot. The gonopodium of this male is fully developed. The bodies are olive brown, without any additional pigmentation. Dorsal fin pigmentation lightly spotted, darker in males than in females. (Three additional small males are without a humeral blotch.) Eight females (22-31 mm SL), eight subadults and 30 juveniles complete this lot. Males 19.4 S75 WS — 8 16 De IS Females 26.0 610 156 8.8 VES 15.8 26.1 5) ZMA 100.606. This lot was collected in a freshwater pool at ‘Plantage Fontein’. It contains medium sized specimens, largest males about 35 mm SL with a clear humeral blotch (at scale 5 of LLS) and a large dorsal fin that reaches the base of the caudal fin. The dorsal fin has a large anterior blotch at the base. In this lot there is a total of six males with humeral blotch, all adults, of which two males do not possess a blotch, 14 females (30-48 mm SL), eight subadult and 30 juveniles. Males 34.6 551 XOW — 8.3 18 26.3 16 Females 43.3 613 173 9 8 16.4 26.5 16 ZMA 100.607 is a sample from a small pool of freshwater at “Fontein Plantage’. Only one fully grown male (without blotch) is present, with one fully grown female. It further contains three subadults and five juveniles. Male 33,6 533 162 — 8 7) 25 16 Female 41.9 616 159 9 8 7) 27 16 ZMA 100.608, again from a small freshwater pool at ‘Fontein Plantage’. Two males in this sample have a blotch, viz., the largest and the smallest males. One male is without blotch. One large female is present, together with one small female. 12 Juveniles. Males 30.9 581 200 — 8 17 DOS 16 Females 34.0 611 170 9 8 16 DI) 15 ZMA 102.212. Pool at the well “Rooi Prins’. Males in this lot have faded humeral blotches. The largest males have the dorsal fins to caudal base, with an anterior blotch at the base. Males 30.6 588 180 — 8 16 26 16 Females IBoll 653 171 9 7.8 18 26 So 7/ ZMA 120.413. ‘Spaans Legioen’. 54 Small, dark specimens, of which 22 are males, 25 are females, six subadults and one juvenile. No characteristic coloration can be observed. Dorsal fins and anal fins are damaged. Males Mol) 602 184 — — UT aT 26.3 16 Females 26.4 607 Le — — 16 BSI S)57/ ZMA 120.414. This lot is from ‘Rooi Awa Marga’, which is a slow stream. One small male with a completely developed gonopodium is present, with five small females and five juveniles. Male 16.0 563 163 — 8 16 BS 16 Females 21.6 602 D3 8 7 16.5 Des) 16 ZMA 120.415. Salinja master, caught at an abandoned saltpan (24 g Cl/l). The sample contains five males, three females, three subadults and 3 juveniles. The scales of these specimens are white and hard, probably caused by the high salinity. Males 24.7 633 167 —- 8 16 26 16 Females DES 627 149 9 VES Sod) 26 16 ZMA 120.422. Salinja master, east. Again the abandoned saltpan (6 g Cl/l). Nine males, of which the two largest were very dark. These two specimens are not recorded in the table. 19 Females are found, of which one has many irregularly positioned spots. Eight subadults and four juveniles complete this lot. Males 22.8 598 WZ — 8 16 26 16 Females Mee) 638 149 8.7 8 IS)o7/ Does 15.8 ZMA 120.423. ‘Bron Rooi Prins’ pool. It contains sixteen males, twenty-three females and 4 subadults. Males 31.0 606 187 — 8 16 M53 16 Females BOE 627 172 9 8 18 25.8 16 VOLUME 116, NUMBER 2 367 Table 1.—Continued. SIE Pred CPd A D € LLS CPS ZMA 120.425. ‘Salinja Balashi’. 9 Males, fifty-eight females, eighteen subadults and 2 juveniles. This population has some extra-ordinary features. The population consists of more or less colorless specimens. The males are few in number, and one specimen seems of intermediary sex, 1.e., does not have his gonopodium fully developed while being large enough. The other eight males are normally developed. Many females only have eight anal fins. Males 22.0 562 187 — 8 16 25 16 Females 293 616 169 8.3 7.8 16.6 26.3 16 ZMA 120.437. “Fontein pond’. Seven males, six females, five subadults and one juvenile. Males SS 577 190 — Tel) 16.7 Dei 16 Females Died 626 144 9 8 16.7 Dales 16 Curacao ZMA 100.603. Zaquito. A note in bottle reads: males with orange-red dorsal fins, orange-red gill-areas, ventral sides and caudal fins. On the sides 3 to 6 rows of orange-red spots. Males 3\0)7/ 510 168 —= 8 16 2S 14.5 Females 2m 622 163 8.2 7.8 16 Jee) 15 ZMA 123.465. 5 km SWS of Willemsstad, from a tidal pool. The complete population was poisoned, and contains more males than females. 48 Males, of which some have more than one humeral blotch. All males are heavily pigmented, i.e., are speckled. Some have rows of black spots on the caudal peduncle. 36 Females, 3 subadults, | juvenile. Males 38.4 506 171 — 8 16 26.3 LES Females 39.4 581 155 8.9 V2 16 26.6 16.9 ZMA 123.466. Hato. 127 mainly black or dark brown specimens. Male SileS 474 180 — 8 16 26 17 Females 30.2 674 164 8.5 8 16 DED 16 Averages of Morphometric and Meristic Data Aruba Males 26.8 580.8 181.2 — 8.0 16.5 DT, So) Females 30.7 620.5 161.8 8.8 7.8 16.5 26.3 15.8 Curacao Males 33:5 496.7 173.0 — 8.0 16.0 25.8 16.3 Females 3)3). Il 625.7 160.7 8.5 Ve 16.0 26.0 16.0 Colombia: BMNH 1899.3.15: 24-26, (2), Colombia, coll. H. Festa, no date. Diagnosis.—A = 8 (rarely 9); D = 8 (rarely 7); LLS = 25-26; BS = 18; CPS = 16; PS = 11-13 Descriptions.—Poecilia koperi has uni- cuspid inner teeth. The largest specimens examined are a female of 56.2 mm SL and a male of 47.0 mm SL. Holotype.—The largest male (38.5 mm SL) has a deep body, with depressed sides. The upper part of body exhibits two dark horizontal rows of dark spots, which are less conspicuous ventrally. The body is brown, dorsally darker than ventrally. The caudal fin has black spots on the membrane between the rays. The dorsal fin is pig- mented like caudal fin, reaching caudal base. The inner jaw dentition is unicuspid. Proportional body measurements in Table 2. Meristic data include 26 scales in the lat- eral series, 12 predorsal scales, 16 scales around caudal peduncle, and 18 scales around body. Pectoral fins exhibit 16 rays, the dorsal 8 fin rays, and the caudal 17 fin rays. The last segment of gonopodial ray 3 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Poecilia koperi, (a) holotype, (b) allotype. has an extruding spine, two subsequent seg- _trally on ray 3, a hood is present, not reach- ments without serrae, nine segments with ing the tip. Ray 4a of the gonopodium is ventral processes, and the other segments without serrae, ray 4p has ten distal unser- with serrae on both sides (Fig. 4a). Ven- rated segments, followed by ten segments 369 VOLUME 116, NUMBER 2 CLI = O¢E a O€T Gog 6£7 vIT ILE GUC yous] uly [epneD LI = -9T 6CC ETI O81 SOI 991 C81 96 YsuaT] UL SIAjad oan oss == == C6l CNC 781 091 CST 091 yysuaT Uly [e10}90q = = = = L9 — LS €¢ = a yjsuay] [eseg uly [euy HLT 6LI SOE LOT Gi 177 pri 971 ZOT 007 yysuaT] possaideq uly jeuy = = = —= Sel IS] 96 08 C9 Ie] yjsuayT [eseg uly [esiog Lvl 99] O77 Ye SET C67C == 9S I TSP 19¢ ysuay] possoidad uly [esiog = and = — 9¢ 79 OS SP 6S 09 yysuay [eiqiO == = = a OS Iv 6£ 67 SP SE ysuay ynoug 16 88 09 pL ian OI 8 OL OLl (45) WIPIM YIN 9TI LTI OL vil TEI LEI rel a orl TEI MIPIM Auog [e}IqQ1O19}U] C9] So] L8I L81 vLI Z6I pO SEI EST 861 yydaq 3svay] afounpod [epneD TIE TOE LES OES 9LT 98r GRE 997 ECS 89r yisuay] gjounped j[epneD ELI ISI COI OLI GG O17 ILI vr 161 LOI YIPIM PeeH 661 SLT 987 CVT CVT a6 617 861 CST THT ysuay] peoH CLT OLT ECE 697 OTE OTE CCT VIT Sr [87 yidaq Apog I8€ BoE TTS 6rS 89¢ bes Ore COL 79S C6r aseg [epneD 0} UISUO [euy (G49 ILE 06€ 68€ ZOV LO¢ IS¢ OT cor 96£ aseg [epneyD 0} UISIO [esiog 9¢9 909 GRE LLS 98¢ ELS 6LS 76S CES Crs yysuayT [esiopaid ee 661 ISI v0 SLY SEP 9° Or I've CSE SVE (uu) YJsUaT piepuris (adAjo][ Vv) So[Buld{ (adA}o[OH) Soe (odAjo0]V) (adAjo[oH) (edAo] Vv) So[euldy (adAjo;oH) SORIA AIDPUDM D1P1I00g 1UDWASa0g VILIIA0g l4adoy DIp1Ia0g ‘yISUS] Plepurs dy) JO syJpuvsnoy) UL sjyusWaInsvou [eUONIOdOAd ‘s1d}DUUITT [TU UT dIe SYISUS] PrepUe)s BY], ‘~AVPUDM ‘q pu 1UDUWIASaOg ‘q ‘14adoy ‘gq JO SyUusWIdINSKvoW [ePUOTIOdOIgG—'Z I1qR 370 with dorsal serrae. Rays 5a and 5p do not have spines, and are seven segments shorter than 4p. Allotype.—The largest female (40.6 mm SL) has spots on the body like those of the holotype, and has fins with indistinct pig- mentation. The dorsal fin is short, reaching halfway from the base of the first ray to the caudal fin base. The caudal fin has 16 rays, the anal fin has eight rays. The other me- ristic counts are like those in the holotype. Distribution.—Poecilia koperi occurs in coastal areas from the Peninsula de Araya, Venezuela to the eastern part of Colombia. Etymology.—This species is named after my friend Michel Koper, with whom dis- cussions have helped to keep my thinking flexible. Remarks.—Poecilia koperi is a medium sized species of the Poecilia sphenops com- plex (cf. Miller 1983), distinguished from P. gillii cuneata only by the reduced num- ber of fin rays in the dorsal and anal fin. From the other subspecies of P. gillii it is distinguished by gonopodial characteristics. Two geographically separated color morphs exist in this species. The pigmen- tation of the body of specimens east of Coro consists of two or three horizontal rows of spots, which are less conspicuous ventrally. The body becomes darker dorsal- ly, as opposed to the lighter ventral side. The unpaired fins are moderately pigment- ed, i.e., they only have spots on the mem- brane between rays. The dorsal fin reaches to the base of the caudal fin. The popula- tions west of Coro, 1.e., in the Maracaibo basin, are more intensely pigmented. The males have about 10 distinct vertical stripes from halfway the body to the caudal base. The scales on the body have pigmented margins. The base of the caudal fin is cov- ered with scales, which are irregularly pig- mented. The caudal fin has longitudinal spots, the dorsal fin has a median blotch from the base to halfway the fin, distally replaced by dark spots. The females are pigmented like the males, but with fainter stripes on the body. The fins in the females PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON are without spots and the snout is less pointed. Meristic data and the gonopodium are identical in the two populations, al- though Maracaibo populations never have seven dorsal fin rays, nor nine anal fin rays. The gonopodium has a fleshy hood that nearly reaches the tip. Gonopodial ray 3 has two or three weakly serrated terminal seg- ments. Other segments of ray 3 have ventral serrae and irregular dorsal extrusions. A membranous ventral hook is found on the edge of the last segment. Gonopodial ray 4a is with long unserrated segments. Gon- opodial ray 4a possesses seven to eleven unserrated distal segments, others segments have dorsal serrae. This ray is somewhat shorter than ray 4a. Gonopodial rays 5a and 5p have long unserrated segments ending at fourth to seventh segment of ray 4p count- ing from the tip (Fig. 4a). Schultz’ (1949:97) record of P. vivipara, partly based on de Beaufort’s (1940) ac- count, is erroneous; re-examination of the specimens mentioned by de Beaufort (1940: 111) in ZMA revealed that this material is P. koperi. Poecilia boesemani, new species jens 2, Iolle Z Poecilia sphenops (non Valenciennes, 1846); Regan, 1913:1013 Gn part); Price, 1955:18; Boeseman, 1960:122. Holotype.-—RMNH 21543, adult male, Trinidad, Port of Spain, Maraval River, coll. M. Boeseman, 27-VII-1954. Allotype.-—RMNH 32428, adult female, same data as RMNH 21543. Diagnosis.—A = 8; D = 8-9; LLS = 27; BS = 20; CPS = 16; PS = 13-16. Description.—Holotype: male, 43.8 mm SL. The body does not have a distinct pat- tern of pigmentation. The dorsal fin is an- gular, with dark spots on the membranes be- tween the rays. These spots form two reg- ular horizontal lines. In the caudal fin, a vertical line of similar spots is found on the posterior part. Proportional body measure- ments in Table 2. VOLUME 116, NUMBER 2 Fig. 2. Poecilia boesemani, (a) holotype, (b) allotype. The dorsal fin has 10 rays, the caudal fin has 18 fin rays. Pectoral fin with 16 rays, 20 scales around the body, 16 scales around the caudal peduncle, 27 scales in a lateral series. The holotype misses one or two pre- dorsal scales, so the exact number is un- known. All head pores are exposed. The gonopodium is like P. mexicana (Fig. 4b, see Miller 1975), i.e., with a spine in the membranous hook on gonopodial ray 3. In- B72 ner jaw teeth unicuspid or possibly subtri- cuspid (Garman 1895, Miller 1975). The female allotype is 47.8 mm SL. The morphology and pigmentation are like that noted in the holotype but without spots in the caudal fin. The anal fin has 8 rays, the dorsal fin contains 9 rays, the pectoral fin has 13 rays. All other counts are identical to those of the holotype. Distribution.—This species is known only from its type locality. Etymology.—This species is named after Dr. Marinus Boeseman, emeritus curator of the ichthyological collection of the National Museum of Natural History, Leiden (RMNH), who collected the specimens. Poecilia wandae, new species Figs. 3, 4, Table 2 Holotype—USNM 121683, adult male, Venezuela, Maracaibo basin, hot spring, creek tributary to Rio Mechango, 20 km above bridge, coll. L. P Schultz, 21-III- 1942. Allotype.—USNM 326142, adult female, same data as USNM 121683. Material.—USNM 121669 (4), Cano, % mile W of Sinamaica, coll. L. P Schultz, 11-III-1942; USNM 121670 (3), Rio Socuy, 3 km above mouth, coll. L. P Schultz, 24- 11-1942; USNM 121671 (42), Cienago del Guanavana, about 10 km N of Sinamaica, coll. L. PB Schultz, 11-lI-1942; USNM 121672 (12), Rio San Juan, 12 km S of Ro- sario, coll. L. P. Schultz, 26-II-1942; USNM 121673 (1), Rio San Ignacio, about 20 km S of Rosario, coll. L. P Schultz, 26-II-1942; USNM 121674 (1), Rio Negro, below mouth of Rio Yasa, coll. L. P. Schultz, 2- IlI-1942; USNM 121675 (27), Lago Tule, about 75 km W of Maracaibo, coll. L. P. Schultz, 1-I]I-1942; USNM 326143 (261), same data as USNM 121683. Diagnosis.—D = 5-7; A = 6-8; C = (probably) 14-16; LLS = 24-26; PS = 11- 13; BS = 16; CPS = 14. This species is very much like P. caucana, differing only PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON in regards to the reduced meristic and gon- opodial characteristics. Description.—Poecilia wandae is a small species, with mature males ranging from 13 to 22 mm SL. Most specimens have a brownish body with 7 to 10 vertical stripes extending from the caudal peduncle to half- way along the body. In some specimens a longitudinal stripe over the lateral line is present. The dorsal fin has a median basal blotch and a black margin, whereas the oth- er fins are unmarked. Poecilia wandae has an elongated body with a compressed head. Females have fewer vertical stripes than the males. Gonopodial ray 3 at the tip with 1 or 2 unserrated elements, all other segments of this ray are serrated on both sides (Fig. 4c). The fleshy hood reaches the tip. Gon- opodial ray 4a has long, unserrated seg- ments, with the last segment pointing downwards, while 4p has 7—8 square un- serrated elements at the tip and the distal segment pointing upwards. The tip is split. Gonopodial rays 5a and 5p are unserrated and do not reach the tip. Gonopodial ray 5p has a retrorse spine, which is more weakly developed at the end of ray 5a. Description of the types.—The holotype is a male, 20.6 mm SL. The body has seven vertical stripes that extend from the middle of the body to the middle of the caudal pe- duncle. This is combined with a reticulate pigmentation pattern that is dorsally darker than ventrally. The dorsal fin reaches half- way from the base of its last ray to the base of the caudal fin. The caudal fin is clear, whereas the dorsal fin has a median basal blotch and has a dark margin. The female allotype is 23.1 mm SL. Pigmentation and scale counts are like those noted on the ho- lotype. It has six dorsal fin rays and eight anal fin rays. The caudal fin rays are bro- ken. The meristic data are like those recorded for P. caucana. There are 25 scales in a lateral series, no more than 14 scales around the caudal peduncle and 16 scales around the body. Unfortunately, the caudal fins and the dorsal fins are damaged. Poecilia wan- VOLUME 116, NUMBER 2 Poecilia wandae, (a) holotype, (b) allotype. dae resembles P. caucana in gonopodial characters. Gonopodial ray 3 has the first two segments unserrated; all other segments are serrated on both sides. The fleshy palp covers the tip. Gonopodial ray 4a has un- serrated segments, with the last segment bending down. Gonopodial ray 4p has sev- en unserrated distal segments, the last seg- ment bends upwards, splitting the tip. Gon- opodial rays 5a and 5p are unserrated, both 374 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Tips of gonopodia of (a) Poecilia koperi, (b) Poecilia boesemani, (c) Poecilia wandae. do not reach the tip. Gonopodial ray 5p has a spinous retrorse extremity. Proportional body measurements in Table 2. Distribution.—This species is known from several river systems west of the Ma- racaibo basin. Etymology.—This species is named after Vanda Marisa Freitas de Leite, who wishes to be called Wanda. Remarks.—Some specimens collected from the population of the Rio San Juan/ Rio Negro drainage are pale, without any specific marking with the exception of two specimens. One female has a distinct black spot on the caudal base and another spot on the left side of the body under the dorsal fin. Another female had a similar spot on the left side at the upper margin of the cau- dal base. The number of specimens avail- able, however, was too limited to decide whether these latter populations are either specifically distinct, or only color morphs of P. wandae. Discussion The South American species of Poecilia inhabit mostly peripheral habitats, i.e., coastal areas or islands. Poecilia sphenops and P. vandepolli do not occur on the Ve- nezuelan mainland (Poeser 1992; present study), despite earlier reports (Schultz 1949, Feltkamp and Kristensen 1969). In the present paper, seven species from the northeastern part of South America are re- corded. One of these species, P. gillii (Kner and Steindachner, 1864), also occurs in Central America, and one species, P. vivi- para, extends from the Orinoco drainage and the Lesser Antilles south to Argentina. The re-examination of P. vandepolli con- firmed past observations concerning char- acter variation in this species, whereas ex- amination of the Colombian and Venezue- lan populations of mollies confirmed earlier conclusions (Poeser 1992) that the Vene- zuelan population are neither P. sphenops nor P. vandepolli. One of the new species described herein, P. wandae, is related to P. caucana based on its morphological and gonopodial characteristics. Venezuela is divisible into four distinct biogeographical regions, viz., the Maracai- bo basin, mainland Venezuela, the drainage of the Rio Orinoco, and the adjacent islands VOLUME 116, NUMBER 2 © NETHERLANDS ANTILLES LAKE MARACAIBO Rigeo- CARIBBEAN SE, TOBAGO ISLA DE MARGERITA A, TRINIDAD CARACAS ws A BARCELONA Map of Venezuela. The locations of UMMZ material (=P. koperi) are noted with black triangles, the USNM material (=P. wandae) are open circles, the RMNH material (=P. boesemani) is a black square. of the Lesser Antilles. Only one of the ex- amined species occurs in more than one dis- trict, viz., P. caucana (along the Atlantic coast from Panama to the delta of the Rio Orinoco). In the Rio Orinoco, P. vivipara finds its northernmost mainland extension. Poecilia vandepolli and P. boesemani are endemic species located on islands adjacent to Venezuela. Comparisons with Central American mollies reveals that the species of Poecilia manifest a clinal reduction in body size and a decrease in meristic characters from north to south. Poecilia mexicana, naturally occurring on the Atlantic side of Mexico, has unicuspid inner teeth, nine dor- sal fin rays, nine anal fin rays, 18 scales around the caudal peduncle, and two spines at the tip of the gonopodium (Table 3). Miller (1983) could not distinguish P. mex- icana from P. gillii, and genetic data of P. mexicana from Vera Cruz and from Costa Rican populations (labeled ‘*P. gillii’’) are virtually identical (Brett & Turner 1983, p. 136: cluster map of NEI identity values). Why then separate P. mexicana from P. gillii? Villa (1982) and Bussing (1987) both mentioned two species fitting the descrip- tions given in the present paper. Although 29 scales in a lateral series are not typical for P. mexicana (cf. Menzel and Darnell 1973), Bussing (1987) nevertheless consid- ered populations with this character as P. mexicana. The alternative hypothesis would be that P. gillii is a subspecies of P. mexi- cana with 16 scales around the caudal pe- duncle. The populations with 29 scales in a lateral series (Villa 1982; Bussing 1987) should then be considered another (possibly new) species. Poecilia gillii has been reported to ex- ceed 100 mm SL (Bussing 1987). The body size of the investigated specimens was al- ways much less than 100 mm SL. In con- trast to its variable body shape (Meek 1914, Meek and Hildebrand 1916), the meristic data are fairly constant. From El Salvador (Poeser 1995), through Nicaragua (Villa 1982) and Costa Rica (Bussing 1987) to Panama and Colombia (present study), P. gillii constantly displays nine anal fin rays, nine or ten dorsal fin rays, 16 scales around the caudal peduncle, and on average 27 scales in a lateral series. The only signifi- cant change in diagnostic characteristics is a reduction of the spine found on gonopo- dial ray 5p. The observed differences in gonopodial structures I regard as subspecif- ic variation within P. gillii. The next species in the Central to South American cline is P. koperi, found in Ve- nezuela. Poecilia koperi differs from P. gil- lii in the number of dorsal and anal fin rays (8 versus 9 in P. gillii). The reduction of the gonopodial spine on ray 5p is complete in 2. Koper lhe next Species south; PP: yi- vipara, continues the trend of reduction, ex- hibiting 7 dorsal fin rays and a lack of all gonopodial extremities. In addition, P. vi- 376 Table 3.—Clinal variation in meristic data and gonopodial characters in 5 examined taxa, from north to south. Gonopodial tip CPS Range Species spine on ray 3, spine on ray 5 18 16 16 16 16 9 or 10 9 or 10 9 or 10 Atlantic coast from Texas to Costa Rica Pacific coast from Guatemala to Panama P. mexicana P. gillii gillii P. g. cuneata P. koperi spine on ray 3, spine on ray 5 spine on ray 3 Panama and the Atlantic coast of Colombia Colombia and Venezuela Venezuela to Argentina spine on ray 3 no gonopodial spines PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON P. vivipara vipara has a median positioned humeral blotch and serrae on gonopodial ray 4a, unique characters in the subgenus, but not in the genus. In the subgenus Lebistes sensu Rosen and Bailey (1963), P. reticulata also has serrae on gonopodial ray 4a, P. bifurca exhibits the lateral body spot, and P. picta exhibits both the serrae on ray 4a and the lateral body spot. An exception to these clinal variations is seen in P. boesemani, which occurs on Trinidad. Poecilia boesemani has a reduced body size, less anal fin rays and a reduced number of scales around the caudal pedun- cle than seen in P. mexicana, whereas the number of dorsal fin rays and the gonopo- dial characters are similar in these two spe- cies. Based on the last two similarities, Price (1955:7, 18) and Boeseman (1960: 122) presumed that P. boesemani was merely an introduced aquarium strain with the vernacular name ‘Liberty molly’. Poe- cilia boesemani resembles western popula- tions of P. sphenops in fin pigmentation and general morphology (Schultz and Miller 1971). The inner jaw dentition, however, is unicuspid versus tricuspid in P. sphenops. Poecilia boesemani differs from most Cen- tral American species in the number of anal fin rays (8 versus 9 in the other species), in the number of scales around the caudal pe- duncle (mostly 18 in the other species) and in female pigmentation (rows of black spots in P. sphenops and in P. mexicana). Poe- cilia boesemani more closely resembles P. butleri, another widely distributed Mexican species, in the number of anal fin rays; it differs, however, in dorsal fin pigmentation (distinct basal blotch in P. butleri ([cf. Schultz and Miller, 1971]). Furthermore, P. boesemani and P. butleri are geographically the most separated species. Poecilia boe- semani differs from P. koperi and P. van- depolli in the presence of distal spines on both gonopodial rays 3 and 5. Boeseman’s (1960, p. 122) observation that “‘it lacks a distal retrorse segment on gonopodial ray 5”’, is erroneous (Fig. 4b). Poecilia vande- polli, occurring on the Dutch Lesser Antil- VOLUME 116, NUMBER 2 les, also has a reduced body size, less dorsal fin rays, 16 scales around the caudal pe- duncle, and gonopodial characteristics like those seen in P. vivipara. It also has a hu- meral blotch, although it is found more an- teriorly on the body than in P. vivipara. The clinal variation is evident in P. cau- cana and P. wandae. Poecilia wandae is superficially like P. caucana, resembling it in size, scale counts, pigmentation, and gonopodial structures. It differs, however, in the number of anal fin rays (6-8 versus always 8 in P. caucana), dorsal fin rays (5— 7 versus 7 or 8 in P. caucana), and in the absence of a membranous hook on gono- podial ray 3. The gonopodia of both species agree in several internal structures (Rosen & Bailey 1963:62, fig. 25B). Rosen & Bai- ley (1963) reported that this ‘undescribed species’, viz., P. wandae, was osteological- ly like P. caucana, as well as the Hispani- olan species of Poecilia, viz., P. elegans, P. dominicensis and P. hispaniolana, and “‘the smaller species of the P. sphenops group’’. Unfortunately, they did not specify to which species in the P. sphenops group they were referring. The occurrence of such similarities in widely separated species sug- gests that the osteological evidence could represent a primitive state. Furthermore, based on its gonopodial structures, it was suggested that P. caucana is related to the species of Limia of Hispaniola, and to the Central South American species of Pam- phorichthys (Rosen 1975, Rauchenberger 1989). This would than place the origin of P. caucana and P. wandae from an ances- tral species in South America at 80 to 70 Mya. (Pitman et al. 1990: fig. 2.5C). During this period, Hispaniola was situated be- tween North and South America, as part of the great Arc of Proto-Greater Antilles. The morphological differences, 1.e., the diver- sion from the ““molly-geocline’’, as well as the osteological and gonopodial character- istics linking P. caucana and P. wandae with Hispaniolan and South Brazilian spe- cies, suggest a separate origin for these two species. The overall similarities in gono- SILT) podial structures, however, warrant their present inclusion in Poecilia. Under a dispersalist scenario, one may conclude that members of the genus Poe- cilia entered South America after the com- pletion of the Andean uplift and the for- mation of the Panamanian landbridge, i.e., about 4 Mya. (Pitman et al. 1990: fig. 2.5 F). This rapid dispersal and speciation re- sulted in subsequent differentiation from the Central American species of Poecilia. This hypothesis is confirmed by a molecu- lar phylogeny presented by Breden et al. (1999), who concluded that P. vivipara was significally different from the species of the P. sphenops complex in Central America. Breden et al. (1999) suggested a separate taxonomic status for the P. sphenops com- plex and allocate these species, together with the broad finned mollies (cf. Miller 1983), in the subgenus Mollienesia (sensu Miller 1975). Rodriguez (1997) proposed a similar taxonomy based on a phylogenetic analysis of morphological characters. How- ever, all three studies (Miller 1975, Rodri- guez 1997, Breden et al. 1999) did not in- vestigate the clinal variation in characters found in the present study. The suggestion of a separate taxonomic status for P. vivi- para will be the subject of further investi- gations. Acknowledgements I like to thank Dr. H. Nijssen and Dr. I. J. H. Isbriicker (both ZMA, Amsterdam) for their support, Mr. J. Chambers (BMNH, London) for his influential moral support and for sending material under his care, Dr. O. Crimmen (BMNH, London) for continu- ing where Jim gave up, Dr. W. L. Fink (UMMZ, Michigan), Dr. S. Raredon (USNM, Washington) and Dr. M. J. P. van Oijen (RMNH, Leiden) for loaning material under their care. I also thank my supervisor Prof. Dr. EK R. Schram for his support. Literature Cited Beaufort, L. FE de. 1940. Freshwater fishes from the Leeward group, Venezuela and eastern Colom- 378 bia.—Studies of the Fauna Curacao, Aruba, Bo- naire, and the Venezuelan islands 2(7):109—114. Boeseman, M. 1960. The fresh-water fishes of the Island of Trinidad.—Studies of the Fauna of Curacao and other Caribbean Islands 10(48):72-153. Breden, E, M. B. Ptacek, M. Rashed, D. Taphorn, & C. A. Figueiredos. 1999. Molecular phylogeny of the life-bearing fish genus Poecilia (Cypri- nodontiformes: Poeciliidae).—Molecular Phy- logenetics and Evolution (12)2:95—104. Brett, B. L. H., & B. J. Turner. 1983. Genetic diver- gence in the Poecilia sphenops complex in Mid- dle America.—Biochemical Systematics and Ecology 11(2):127—137. Bussing, W. A. 1987. Peces de las aquas continentales de Costa Rica.—2. Edicion, San Juan, Costa Rica: Editorial de la Universidad de Costa Rica: 1-271. Carr Jr., A. EF, & L. Giovannoli. 1950. The fishes of the Choluteca drainage of Southern Hondu- ras.—Occasional Papers of the Museum of Zo- ology, University of Michigan (523):1—38, PI. I&II. Feltkamp, C. A., & I. Kristensen. 1969. Ecological and morphological characters of different popula- tions of Poecilia sphenops vandepolli (Cypri- nodontidae).—Studies of the Fauna of Curacao 31:102—130. Fowler, H. W. 1949. Results of the two Carpenter Af- rican Expedition, 1946-1948, part 2.—The fish- es.—Proceedings of the Academy of Natural Sciences of Philadelphia 101:233—275. Garman, S. 1895. The Cyprinodonts.—Memoires of the Museum of Comparative Zoology 19(1):1— 179, Plates I-XII. Ginther, A. 1866. A catalogue of the fishes in the Brit- ish museum.—London (6):1—368. Hildebrand, S. E 1925. Fishes of the republic of El Salvador, Central America.—U.S. Bulletin of the Bureau of Fisheries, Washington 41(985): 237-287. Hubbs, C. L. 1926. Studies of the fishes of the order Cyprinodontiformes VI.—Miscellaneous Publi- cations of the Museum of Zoology of Michigan, 16(1):1—87. . 1953. Geographic and systematic status of the fishes described by Kner and Steindachner in 1863 and 1865 in Panama and Ecuador.—Cop- eia 1953(3):141-148. Hubbs, C. L., & K. EK Lagler. 1947. Fishes of the Great Lakes region.—Cranbrook Institute of Science 26:i—xi, 1-186. Meek, S. E. 1914. An annotated list of fishes known to occur in the fresh waters of Costa Rica.— Publications of the Field Colombian Museum, zoological series 10(10):101—134. Meek, S. E., & S. E Hildebrand. 1916. The fishes of the freshwaters of Panama.—Publications of the Field Colombian Museum, zoological series 10(15):217—374. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Miller, R. R. 1975. Five new species of Mexicana poe- ciliid fishes of the genera Poecilia, Gambusia, and Poeciliopsis.—Occasional Papers of the Museum of Zoology, University of Michigan (672):1—44. . 1983. Checklist and key to the Mollies of Mexico (Pisces: Poeciliidae: Poecilia, subgenus Mollienesia).—Copeia 1983(3):817—-821. . 1994. Origin and classification of the Liberty Molly.—Tropical Fish Hobbyist (42)6:104, 106, 108. Pitman, W. C., S. Cande, J. La Brecque, & J. Pindell. 1990. Fragmentation of Gondwana: the separa- tion of Africa from South America. Pp. 15-34 in Peter Goldblatt, ed., Biological relationships between Africa and South America. Proceed- ings of the 37th Annual Systematics Sympo- sium.—Botanical Gardens 1990. Poeser, FE N. 1992. Re-establishment and redescription of Poecilia vandepolli Van Lidth de Jeude, 1887 (Pisces: Poeciliinae), with comments on related species.—Studies of the Natural History of the Caribbean Region 71:59-74. . Nonrandom variation in Poecilia marcellinoi n. sp. and P. salvatoris Regan, 1907 in EI Sal- vador (Pisces; Poeciliidae).—Bijdrage tot de Dierkunde 64(4):239—252. . 1998. The role of character displacement in the speciation of Central American members of the genus Poecilia (Poeciliidae).—Italian Jour- nal Zoology 65, Suppl.: 145-147. Price, J. L. 1955. A survey of the freshwater fishes of the Island of Trinidad.—Journal of the Agricul- tural Society of Trinidad & Tobago (863):1—28. Regan, C. T. 1908. Pisces. Jn EK Godman, DuC., and O. Salvin, Biologia Centrali-Americana. Lon- don, Ser. HI, pt. 193: 1-203, plates 1-24. . 1913. A revision of the cyprinodont fishes of the subfamily Poeciliinae.—Proceedings of the Zoological Society of London 1913(1):977— 1018, pl. 99-101. Rodriguez, C. M. 1997. Phylogenetic analysis of the tribe Poeciliini (Cyprinodontiformes: Poecili- idae).—Copeia (4):663—679. Rosen, D. E., & R. M. Bailey. 1963. The poeciliid fishes (Cyprinodontiformes), their structure, zoogeogra- phy and systematics.—Bulletin of the American Museum of Natural History 126(1):1—176. Schultz, L. P. 1949. A further contribution to the ich- thyology of WVenezuela.—Proceedings of the United States National Museum 99:1—211. Schultz, R. J.. & R. R. Miller. 1971. Species of the Poecilia sphenops complex (Pisces: Poeciliidae) in Mexico.—Copeia 1971(2):282—290. Villa, J. 1982. Peces Nicaragiienses de agua dulce.— Coleccion Cultural Banco Americana; Serie: Geografia y Naturaleza, 3:i—xiv, 1—253. VOLUME 116, NUMBER 2 379 Appendix 1.—Key to the species of the subgenus Poecilia (sensu Rosen and Bailey, 1963), south of Mexico. la. 1b. 2a. 2b. 3a. 3b. Aa. Ab. Sa. Sb. 6a. 6b. Wek 7b. 8a. 8b. 9a. Ob. Inner jaw dentition tricuspid Inner jaw dentition unicuspid Number of scales around the caudal peduncle 18 Number of scales around the caudal peduncle less than 18 Number of scales around the caudal peduncle 16 Number of scales around the caudal peduncle 14 Number of anal fin rays 9 Number of anal fin rays 8 Body never with humeral blotch, never with 8 anal fin rays Body often with humeral blotch, sometimes with 8 anal fin rays Gonopodial tip with spiny hooks on ray 3 and ray 5p Gonopodial tip with spiny hook only on ray 3 Gonopodial tip with spiny hooks on ray 3 and ray 5p Gonopodial tip without hooks or spines Gonopodial ray 4a never with dorsal serrae Gonopodial ray 4a with dorsal serrae Number of anal fin rays 8 Number of anal fin rays less than 8 P. marcellinoi p P. mexicana P. vandepolli P. gillii gillii* P. gillii cuneata P. boesemani> 8 P. koperi P. vivipara P. caucana P. wandae 4 Specimens with a red dorsal fin are diagnosed P. g. salvatoris, unicuspid specimens with a caudal blotch are P. g. caudata. > Miller (1983) reported P. butleri from El Salvador, fitting this description. However, he omitted the diagnostic characteristics for these populations. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):380—394. 2003. On the occurrence of the vestimentiferan tube worm Lamellibrachia luymesi van der Land and N@¢rrevang, 1975 (Annelida: Pogonophora) in hydrocarbon seep communities in the Gulf of Mexico Stephen L. Gardiner and Stéphane Hourdez (SLG) Department of Biology, Bryn Mawr College, 101 N. Merion Avenue, Bryn Mawr, Pennsylvania 19010, U.S.A.; (SH) The Pennsylvania State University, Department of Biology, 208 Mueller Laboratory, University Park, Pennsylvania 16802, U.S.A. Abstract.—Lamellibrachia luymesi van der Land & N@grrevang, 1975 is re- corded from shallow-water hydrocarbon seep communities along the Louisiana slope in the Gulf of Mexico. It is typically the most abundant tube worm species in these shallow-water communities, sometimes with individuals oc- curring in aggregations numbering in the thousands. Tubes extend deeply into the sediment, and apical portions may project above the sea floor a distance of one meter or greater. This study documents intraspecific variation for a number of important morphological features of L. /uymesi, such as the number of sheath lamellae, the number of branchial lamellae, the length and width of the obtu- raculum and vestimentum, the ratio of obturacular length to vestimental length, the ratio of vestimental diameter to vestimental length, and features of the tube. Comparisons of these features are made with other species of Lamellibrachia. The record of L. luymesi on the Louisiana slope extends the geographic range of this species from the southern North Atlantic Ocean into the northern Gulf of Mexico. Subsequent to the description of the first species of Lamellibrachia, L. barhami, by Webb (1969), van der Land & N@rrevang (1975) described a second species, L. luy- mesi, and discussed the possible relation- ships that vestimentiferans in general have with the Annelida, Polychaeta, and Pogo- nophora. Following their original descrip- tion, van der Land & N@rrevang (1977) provided an in-depth description of the anatomy and histology of L. luymesi. Both the original description and the subsequent anatomical and histological account were based on a single male specimen collected off the coast of Guyana in about 500 m of water. The record of L. luymesi off the coast of Guyana remained as the only documented vestimentiferan species from the Atlantic Ocean until Paul et al. (1984) reported the presence of vestimentiferan tube worms in 3000 m of water at the base of the Florida Escarpment in the Gulf of Mexico which were later named Escarpia laminata by Jones (1985). A second Atlantic species of Lamellibrachia, L. victori, was described by Mané-Garz6n & Montero (1985), based on two specimens collected in about 300 m of water off Uruguay. Since the description of L. victori, no additional confirmed re- cords of either Atlantic species of Lamelli- brachia have been published. However, be- ginning in 1985 (Kennicutt et al. 1985), nu- merous records of a Lamellibrachia sp., oc- curring along the Louisiana slope in the Gulf of Mexico, have been reported (see synonymy below). The purpose of the present study is to confirm the identity of the Lamellibrachia sp. that occurs along the Louisiana slope as VOLUME 116, NUMBER 2 L. luymesi and to provide an account of the intraspecific variation it displays with re- gard to certain important diagnostic mor- phological characters. Systematics As discussed in Gardiner et al. (2001), the systematic relationships of vestimenti- ferans remain unsettled. In order to main- tain consistency with other recent descrip- tions of vestimentiferans and broader dis- cussions of pogonophoran morphology and phylogeny (Southward et al. 2002, 2003), we consider vestimentiferans as comprising a subclass within the class Pogonophora of the phylum Annelida. Materials and Methods Specimens were collected by the manned submersible Johnson-Sea-Link I (JSL I) from two hydrocarbon seep sites on the Louisiana slope in the Gulf of Mexico and brought to the surface in a temperature-in- sulated container. On board ship, specimens were immediately fixed in their tubes in 10% buffered formalin in seawater. In the laboratory, tubes were sliced open with the aid of a Dremel Moto-Tool high speed drill, and specimens were extracted, rinsed in dis- tilled water and transferred to 70% ethanol. For scanning electron microscopy (SEM), materials were dissected from se- lected specimens, dehydrated in a graded ethanol series, critical-point dried, using carbon dioxide, and sputter coated with gold-palladium. Material was examined in either a Topcon ABT-60 or a JEOL JSM- 5200 scanning electron microscope. Nega- tives of figures were scanned at 600 dpi, and plates were constructed using Adobe Photoshop 7.0 (Adobe Systems, Inc., San Jose, California, USA). Subclass Vestimentifera Webb, 1969 Family Lamellibrachiidae Webb, 1969 Genus Lamellibrachia Webb, 1969 Type species.—Lamellibrachia barhami Webb, 1969, by monotypy. 381 Diagnosis.—Vestimentiferan worms with tapering tubes and bodies. Anterior obtu- racular region provided with branchial plume; orientation of branchial lamellae rel- ative to obturaculum axial and parallel; branchial filaments of obturacular plume of one type; plume covered by variable num- ber of peripheral sheath lamellae; anterior face of obturaculum bare, lacking crust or secreted structures; paired internal excreto- ry ducts opening by single, dorsal medial excretory pore. Anteroventral margin of vestimentum discontinuous, posteroventral margin of vestimentum broadly discontin- uous, lobes usually not overlapping. Opis- thosome with variable number of segments; chaetigerous segments with chaetae usually in single row; chaetae with variable number of teeth in anterior and posterior groupings. Tube hard, tapered, usually with variable number of collars anteriorly, becoming smooth posteriorly (Webb 1969; van der Land & N@rrevang 1975, 1977; Jones 1985; Southward 1991). Lamellibrachia luymesi van der Land & N@rrevang, 1975 Figs. 1-8 Lamellibrachia luymesi van der Land & N@rrevang, 1975:86—-101, Figs. 1-7; 1977:1—102, pls. 1—26.—Jones, 1985: 128.—Southward, 1991:872-874, ta- ble Lamellibrachia sp.—Kennicutt et al., 1985: 351, 352, table 1.—Childress et al., 1986: 1307.—Brooks et al., 1987:1139, table 1.—MacDonald et al., 1989:235-—245, Fig. 3A—D.—Fisher et al., 1990:1095, ta- ble 1.—MacDonald et al., 1990:246, Fig. 2.—Childress & Fisher, 1992:382.—Wil- liams et al., 1993:439, 440, table 1.— Carney, 1994:150.—Scott & Fisher, 1995:107, 109, table 2.—Young et al., 1996:514-516, Figs. la, 2b, d, e, m. Heldmanvret al, 1997-27 14272, 275, Fig. 2, tables 1, 2.—Fisher et al., 1997285, 88, 90, 91, table 1.—Miura et al., 1997: 455.—Julian et al., 1999:2245—2256.— 382 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON \ a ~ ' . In situ photograph of portion of large cluster of Lamellibrachia luymesi at Bush Hill hydrocarbon seep community. No scale. Riceee Southward, 1999:196, Fig. 5.2, 5.4, 5.5.—Tyler & Young, 1999:196-198, 200-204, table 3.—Schulze, 2001:1-9, Figs. 1E 2A, 4E, table 1.—Southward et al., 2002:1194. Lamellibrachia barhami [not Lamellibra- chia barhami Webb, 1969].—Fisher et al., 1988:232. Undescribed species of Lamellibrachia.— Southward, 1991:872. Lam.(GoM).—Williams et al., 443, Figs. 1—3, table 1. L. sp.—Fisher, 1995:307, table 4. Lamellibrachiid.—Fisher et al., Sie micsa lee: Lamellibrachia sp. nov. 2.—Nelson & Fisher, 2000:4, 8, 10, Figs. 2, 3. Lamellibrachia.—Salvini-Plawen, 2000: Ios 34! 1993:440, 1997:86, Lamellibrachia ct. luymesi.—Freytag et al., 2001:13408—13412.—Gardiner et al., 2001:705.—Bergquist et al., 2002:89—98, Fig. 1.—McMullin et al., 2003: 8-31, Figs) 2,4, 5, tables aly 2) Material examined.—Twenty specimens, Green Canyon hydrocarbon seep communi- ty, 27°44.1'N, 91°15.3'W, JSL I dive 3523, 25 Jun 1993, 540 m, coll. C. R. Fisher; 8 specimens, Green Canyon hydrocarbon seep community, 27°44.1’N, 91°15.3’W, JSL I dive 3525, 26 Jun 1993, 540 m, coll. E. Nix; 12 specimens, Bush Hill hydrocarbon seep community, 27°47'N, 91°31.5’W, JSL I dive 3530, 28 Jun 1993, 540 m, coll. J. J. Chil- dress. Selected specimens of this material are deposited in the National Museum of Natural History (USNM), Smithsonian In- stitution, Washington, D.C. VOLUME 116, NUMBER 2 383 Fig. 2. Lamellibrachia luymesi, light micrographs of selected specimens. A, dorsal view of obturacular region showing sheath lamellae not extending to anterior face of obturaculum, together with portion of vestimental region with folded vestimental wings. B, left lateral view of obturacular region showing sheath lamellae nearly covering obturaculum, together with portion of vestimental region. C, right lateral view of obturacular region with sheath lamellae removed to show branchial filaments. Note extent of bare surfaces on obturaculum. Arrow indicates vestimental sheath around base of obturacular region. D, ventral view of obturacular region and portion of vestimental region. E, ventral view of obturacular region with sheath lamellae removed from both sides, exposing branchial filaments. bf, branchial filaments; cf, ciliated field; ob, obturaculum; sl, sheath lamellae; v, vestimental region; vw, vestimental wing. Scale bars: A, B, D, E = 5 mm; C = 3 mm. Description.—Measurements of selected specimens (length by diameter, in mm; ob- turacular: vestimental: trunk: opisthosomal regions; + = incomplete; — = change in diameter; — = missing); (USNM 1007801): 9.4 by 4.5: 43 by 4: 142+ by 3.1: —; (USNM 1007803): 14.9 by 4.6: 56 by Aye WOO yy 23. 2 “CUSINIy 1007805): 16.3 by 5: 61.6 by 5.2: 110+ by 3.5: —; (USNM 1007807): 10.4 by 4.5: 51.5 by 4.2: 117+ by 3.4: —; (USNM 1007809): 7.8 by 4.4: 50.3 by 4.6: 115+ by 4.9: —. Measurements of selected tubes (length by diameter, in mm; + = incom- plete, — = change in diameter): (USNM 1007801): 1340+ by 6.4-1.3; (USNM 1007802): 1560+ by 7.6—2.5; (USNM 1007804): 1405+ by 8.9-0.9; (USNM 1007806): 1230+ by 10.2—0.7; (USNM 1007808): 1190+ by 6.4-1.2. Anterior face of obturaculum bare, with- out crust or secreted structures (Figs. 1, 2A-E). Obturaculum with 15—22 pairs of branchial lamellae (X = 17.8 + 1.8; n = 10), composed of single type of filament (Fig. 2C, E), fused for most of length; free ends of filaments with two rows of pinnules and two rows of ciliated cells (Fig. 3A, B), pinnules on some filaments of unequal size (Fig. 3B); branchial lamellae typically cov- ered by 4—8 sheath lamellae on each side Of branchialucrowninCi— 15.30: aiken — 30); number of sheath lamellae not directly correlated with size of specimen (Pearson’s correlation); filaments of sheath lamellae lacking pinnules and cilia, almost complete- ly fused, with very short free tips (Fig. 3C, D); sheath lamellae sometimes not extend- ing to apical region of obturaculum (Fig. 2A, D). Distal surfaces of obturaculum of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig.) preserved specimens usually not concealed by branchial lamellae (Fig. 2C—E); obtura- culum lenticular in cross-section (Fig. 4), lacking dorsal groove, with poorly devel- oped ventral ridge distally; paired internal excretory ducts opening by single pore on dorsal surface at base of obturaculum; length of obturaculum from 6.6 to 16.3 mm (X = 10.8 + 2.4 mm; n = 40); diameter of obturaculum from 3.4 to 5 mm; (X = 4.5 + 0.6 mm; n = 40); ratio of obturacular length to vestimental length variable, 1:2.6 to 1:6.7 (n = 40). Anterior margin of ves- timentum forming short sheath around base of obturaculum (Fig. 2C, arrow), with shal- low mid-ventral incision usually with over- lapping lobes; posteroventral margin of ves- Lamellibrachia luymesi, SEM. A, branchial filaments with pinnules of equal size. B, branchial fila- ments with pinnules of somewhat unequal size. C, inner face of fused filaments of sheath lamella. Note absence of pinnules and cilia. D, apical region of filaments of sheath lamella. Scale bars: A-C = 75 wm; D = 100 pm. timentum discontinuous, with posteroven- tral lobes of variable size and separated by large gap (Fig. 5A, B); length of vestimen- tum from 26.9 to 61.6 mm (X = 47.2 + 8.9 mm; n = 40); diameter of vestimentum from 3.7 to 5.2 mm (X = 4.3 + 0.6 mm; n = 40); ratio of vestimental diameter to vestimental length highly variable, 1:6.2 to 1:16.4 (n = 40); ventral surface of ves- timentum with ciliated field, wider in mid- dle region, bluntly to sharply tapered at an- terior and posterior ends (Figs. 2D, E, 5A, B); plaques associated with epidermis ven- trally and laterally, increasing in density in posterior one-half of vestimentum (Fig. 5A, B, arrowheads); males with paired dorsal ciliated grooves extending from gonopores VOLUME 116, NUMBER 2 Fig. 4. Lamellibrachia luymesi. Light micrograph of cross section through middle region of branchial plume. Arrowheads indicate obturacular blood vessels. bf, branchial filaments; ob, obturaculum. Scale bar = 1 mm. to near anterior end of vestimentum, di- verging anteriorly (Fig. 5C). Trunk very long in adult specimens, tapering to less than | mm posteriorly; epidermis with nu- merous associated plaques 36—45 wm in di- ameter (Fig. 6). Opisthosome of selected specimen with 26 segments (Fig. 7A), 17 anterior segments each with chaetae occur- ring in single row, separated middorsally and midventrally by narrow band of epi- dermis lacking chaetae (Fig. 7B, C), and 9 posterior segments without chaetae; chaetae with teeth in anterior and posterior group- ings; anterior grouping usually with 5 teeth in 2 rows, posterior grouping more variable, with 3—4 rows of 3-5 teeth (Fig. 7D). Tube hard, cream-colored to lightly yellowed tan with some dark banding anteriorly in living specimens (Fig. 1), becoming more darkly colored in preserved specimens; tube taper- ing posteriorly, diameter of aperture from By Ato OI) tatlad) (0.6 = eyo) ae sy aetna — 385 44), with variable number of weakly to moderately developed collars anteriorly (Fig. 8A, B), becoming smooth in middle and posterior regions. Distribution.—Presently known in the Atlantic Ocean from the type locality in 500 m of water off Guyana (8°01'N, 57°24’W) and in the Gulf of Mexico at several local- ities along the Louisiana slope in 500—650 m of water, including Green Canyon and Bush Hill hydrocarbon seep communities. Remarks.—Several external morphologi- cal features have been used to distinguish between species of Lamellibrachia, includ- ing features of the branchial crown, pro- portions of body regions, and features of the tube (see Table 1). One of the most con- spicuous features of the branchial crown is the presence of a number of sheaths com- posed of fused filaments, i.e., sheath lamel- lae, that enclose the inner branchial lamel- lae. Specimens of L. /uymesi examined for this study usually possess 4 to 8 sheath la- mellae on each side of the branchial crown, but most specimens have a different num- ber of sheath lamellae on each side. Typi- cally, one side will have 1 or 2 lamellae greater in number than the opposite side, although no pattern of preference is appar- ent in the specimens examined in this study. A few specimens have only 2 or 3 sheath lamellae on one side of the branchial crown. Close examination of the these spec- imens, however, always revealed damage to the branchial crown, and it is likely that some lamellae were torn away when the preserved specimens were removed from their tubes. The number of pairs of sheath lamellae reported for the holotype of L. /uymesi falls within the range recorded for specimens from the Gulf of Mexico and, therefore, supports their identification as L. luymesi. The majority of specimens with undamaged obturacula examined in the present study (23 of 30 specimens) exhibit a number of sheath lamellae intermediate between spe- cies with potentially fewer pairs (L. bar- hami and L. satsuma) and L. columna with 386 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Rigse: mm. its higher number (Table 1). Number of pairs of sheath lamellae, therefore, is a use- ful character to distinguish L. luymesi from these other species. However, the number of sheath lamellae of the type material of L. victori falls within the range of L. luy- mesi so this character does not distinguish these two species. Unlike the number of sheath lamellae, the number of branchial lamellae does not differ significantly so this feature is not a useful characteristic to distinguish between species (Table 1). Only L. barhami is re- ported to possess a number of branchial la- mellae outside the range exhibited by spec- imens of L. luymesi examined in this study (see Webb 1969). However, Jones (1985) states that L. barhami can possess up to 25 pairs of branchial lamellae so it is likely that this species exhibits a range in number Lamellibrachia luymesi, light micrographs of selected specimens. A, B, posterior end of vestimental region showing discontinuous posteroventral margin of vestimentum with variously developed lobes. Arrowheads indicate plaques on body wall. cf, ciliated field. C, dorsal view of vestimentum of male showing well-developed ciliated grooves (g) leading from gonopores to anterior end of vestimentum. Scale bars: A, B = 2 mm; C = 4 that is more consistent with other species of Lamellibrachia. Although lengths and proportions of body regions are affected by a certain de- gree of contraction at the time of preser- vation, some interesting trends can be ob- served between species of Lamellibrachia. Among the five species of Lamellibrachia currently described, L. columna is the larg- est species and L. satsuma is potentially the smallest species (see Table 1). Lengths of the obturacular and vestimental regions of L. barhami overlap somewhat with those of L. luymesi, but the latter species exhibits a broader range of lengths for both regions and is generally overall the larger of the two species. The obturacular length report- ed for L. victori falls within the range of L. luymesi examined for this study, but its ves- timental length is slightly greater than the VOLUME 116, NUMBER 2 Fig. 6. Lamellibrachia luymesi, SEM. Portion of body wall of trunk region showing numerous plaques (p) and openings of pyriform glands (arrows). Scale bar = 100 pm. maximum measured for L. luymesi (Table 1). Since the vestimental length of L. victori is based on a measurement taken from a single specimen, this difference is probably not significant. Three species of Lamellibrachia for Which multiple specimens have been ex- amined, i.e., L. barhami (Webb 1969, Jones 1985), L. luymesi (present study), and L. satsuma (Miura et al. 1997), demonstrate generally similar ranges for the ratio of ob- turacular length to vestimental length (Ta- ble 1). Ratios calculated for the type ma- terial of L. luymesi and L. victori fall within the range observed for specimens of L. luy- mesi from the Gulf of Mexico. When comparing ratios of vestimental di- ameter to vestimental length, L. columna and L. luymesi have generally similar val- 387 ues at the lower end of the range, but spec- imens of L. luymesi from the Gulf of Mex- ico exhibit a somewhat broader range over- all (Table 1). Although the value calculated for the holotype of L. luymesi is slightly outside the range of specimens from the Gulf of Mexico, the value for that of L. vic- tori falls within the range of L. luymesi from the Gulf of Mexico. Van der Land & N@rrevang (1975, 1977) provide illustrations of the dorsal surface of the anterior region of the trunk of the ho- lotype of L. Juymesi that show a number of transverse grooves in the epidermis (they mistakenly state that these grooves occur on the ventral surface in their 1977 study; see p. 12). Southward (1991), however, ques- tioned the taxonomic value of these grooves, suggesting that they may represent a contraction artifact. Specimens of L. luy- mesi from the Gulf of Mexico examined for this study vary greatly with regard to this feature. Some specimens possess numerous shallow grooves, whereas others have a few deep grooves, whereas still others lack grooves completely. We agree with South- ward (1991) that these grooves, when pre- sent, represent an artifact of contraction produced at the time of preservation. The opisthosome and opisthosomal chae- tae are known for three species of Lamel- librachia, L. luymesi (present study), L. co- lumna (Southward 1991) and L. satsuma (Miura et al. 1997). Although Webb (1969) discussed and illustrated what he thought was the opisthosome (= opisthomere in Webb) of four specimens of L. barhami, the absence of chaetae and external signs of segmentation cast doubt on his observations (also see Jones 1981). Southward (1991) suggested that the opisthosomal chaetae of vestimentiferans might display specific dif- ferences. Based on the limited chaetal ma- terial examined to date (one opisthosome each for L. luymesi and L. satsuma, and two for L. columna), significant specific differ- ences among these species are not apparent. Chaetae possess teeth in anterior and pos- terior groupings in the following arrange- 388 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 7. Lamellibrachia luymesi, SEM of opisthosomal region. A, overview of opisthosome of selected spec- imen. B, dorsolateral view of two chaetigerous segments showing chaetae occurring in single rows (arrows) and middorsal area lacking chaetae. C, ventrolateral view of two chaetigerous segments showing chaetae (arrows) and position of internal ventral nerve cord (vnc). D, enlargement of chaetae. Arrow indicates anterior direction. Scale bars: A = 300 pm; B = 50 pm; C = 75 wm; D = 2 pm. VOLUME 116, NUMBER 2 389 Fig. 8. Scale bars = 5 mm. ments for L. luymesi, L. columna, and L. satsuma, respectively: anterior groupings— 5 teeth in 2 rows, 2—7 teeth in 1—2 rows, 6—9 teeth in 2 rows; posterior groupings— 54 TOws) Of 5-5) teem 5—5) rows of 3=4 teeth, 3—4 rows of 3—4 teeth. Some differ- ences in the number of teeth in each group- ing and the number of rows in which they occur exist between these species. However, additional material should be examined to quantify better any possible differences in chaetal characteristics of these and other species of Lamellibrachia. With the possible exception of the max- imum tube length provided for L. satsuma, tube lengths of all species listed in Table 1 are of incomplete tubes so the full extent of possible tube length in most species of La- mellibrachia is uncertain at this time. For L. luymesi in the Gulf of Mexico, tube lengths may actually exceed 2 m (SLG, SH, pers. obs.). Based on the absence of discol- oration in the posterior region of the tube, van der Land & N@rrevang (1975, 1977) suggested that the tube of the holotype of L. luymesi was not buried in sediment. In contrast, specimens of L. luymesi in the Gulf of Mexico may have one-half or more Lamellibrachia luymesi. A, B, light micrographs of anterior ends of tubes showing growth collars. of their tube length buried in sediment (SLG, SH, pers. obs.). Van der Land & Ngrrevang (1975, 1977) acknowledged that the tube of the holotype of L. /uymesi was incomplete posteriorly. Based on our ob- servations of L. /uymesi in the Gulf of Mex- ico, we believe that the tube of the holotype was broken at the time of collection, leav- ing a posterior portion buried in sediment. The presence of collars on the anterior region of tubes is a common feature among vestimentiferans so it is not a particularly useful character to distinguish between spe- cies. It is interesting to note, however, that among the five described species of Lamel- librachia, L. columna is unique in lacking such collars on its tubes (Southward 1991). Status of Lamellibrachia victori.—In her comparison of species of Lamellibrachia known at the time, Southward (1991) con- cluded that L. columna and L. barhami were distinct from each other and that both species were different from L. luymesi and L. victori. She further noted, however, that L. luymesi and L. victori could not be clear- ly distinguished from each other, but the limited material available for each species at the time hindered comparisons. Although ‘suoumtoods aytuaanf ul 0197, , (1661) PreMmUINOS UI gE “BIJ Wo Joquinu ajeutxolddy . ‘(CQ6]) OIOIUOJ 2 UOZIeH-suL| Ul g “ST WOIJ Joquinu areullxoiddy p ‘(SQ6]) Souor WOT, sudtTDeds snosoUINU pue (6961) GI9M WOIF suourtoads adA} Ino] 5 ‘(LL6[) SUBASLION 29 pueyT Jop uvA UT QT “SI4 WoIF Joquinu ayeurlxo1ddy ‘suoulloads pasewepuy) v 8 0565656560000 See L661 “OoulYyseH =? vieyeyNs |, ‘VINIYA] L'8-S'7 0001 = Sl © PEL WC IL 8°6-8'1 61 01 dn -0 78 DUINSIDS "T 1661 ‘PreMYINOS OZ-+1 078 ET: 01 59:1 Sel 1 CEN 071-09 Tr-SI ITZ 91-8 rl puuinjoo “] S861 “OIC a UOZIUD-9UrIN, cl OVC 8:1 c-1 $9 el esl L G MOJIIA “T 6961 “G99 OFS Lh OVS I L9:) 86-1 8 L:1 0} SCT CS 9E-EC GOONS SC VC oS T< HUDYIDG “T CLO] “SURADLION 2 puery Jop ura PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ol L89 orl Sr] 69 ell a6] 9 I adAyojoy ‘saucny “7 suouttoeds OTXIW JO JIND LOVE 09ST 17 Otek CO) € Yelk L9:1 8 9CT Smo QI OO) CCS|| eB—V OV IsawUAn] “T (aimjiode suru) (xeul SUIUL) YJSud] WINJUSUINSOA YISUST WINjUSUINSoA (Uru) (uur) ov[[OWIR] feryoueiq se[[OUTe] Ywoys u IgjowIeIp sqny, yysus] oqn |, IOJOWUIVIP WIN}USUINSS A “WIZus] UNTNOVIMGOQ ~—«YSUST WINJUSLUTISS A, yysus] wnynovinigqO jo JoquInNy JO JoquUInN DIYIDAGIJIWUVT JO Sotseds IOF siajovIeYS NsouseIp ureyJa9 Jo uostIedwoy— | 214P.L 390 VOLUME 116, NUMBER 2 no publications to date indicate that new material of L. victori has been collected, abundant new material of L. luymesi from the Gulf of Mexico has been examined in the present study, allowing for a better un- derstanding of the morphological variation it exhibits for a number of external features (see Table 1). Results of this study corrob- orate Southward’s (1991) conclusion that L. luymesi is distinct from other species of La- mellibrachia, except for L. victori. For all features listed in Table 1, values for the type material of L. victori fall within the range of L. luymesi (holotype and Gulf of Mexico specimens), except for length of the vesti- mentum and aperture diameter of the tube. Because values for these features of L. vic- tori are taken from a very limited number of specimens, it is difficult to know their significance with regard to recognizing L. victori as a valid species of Lamellibrachia. It is likely, however, that the difference in vestimental length is not significant. To re- solve this issue, new material must be ob- tained from the area where the type material of L. victori was collected and compared with L. luymesi. Until such a study is com- pleted, L. victori remains questionably dis- tinct from L. luymesi. Discussion Presently, ten genera of vestimentiferan tube worms have been described, including Lamellibrachia Webb, 1969, Riftia Jones, 1981, Escarpia Jones, 1985, Oasisia Jones, 1985, Ridgeia Jones, 1985, Tevnia Jones, 1985, Alaysia Southward, 1991, Arcovestia Southward & Galkin, 1997, Seepiophila Gardiner, McMullin & Fisher, 2001, and Paraescarpia Southward, Schulze & Tun- nicliffe, 2002. With five described species, Lamellibrachia contains the largest number of species of these genera. Escarpia con- tains two recognized species, whereas all other genera are monotypic. Among these genera, Lamellibrachia also demonstrates the broadest geographic range with two de- scribed species occurring in the western Pa- 39] cific Ocean (L. columna and L. satsuma), One species in the eastern Pacific Ocean (L. barhami), and two species in the western Atlantic Ocean, including the Gulf of Mex- ico (L. luymesi and L. victori). Dando et al. (1992) found tubes of vestimentiferan worms in a shipwreck in 1160 m of water in the eastern Atlantic Ocean off Vigo, Spain. They indicated that the appearance of the tubes resembled that of tubes of L. barhami, but they were unable to confirm the identity of the species because fully in- tact worms were not available. Williams et al. (1993), however, were able to obtain de- graded tissue from these tubes. In their analysis of the 28S rRNA nuclear gene, they demonstrated that this tissue consis- tently grouped with species of Lamellibra- chia also examined in their study. This find- ing supports the suggestion that the tubes belong to a species of Lamellibrachia, thereby confirming the presence of Lamel- librachia in the eastern Atlantic Ocean. Re- cently, Lamellibrachia was reported from the southeastern Mediterranean Sea off Crete and Turkey in about 1700 to 2000 m of water (Olu-Le Roy et al. 2001). Cur- rently, no vestimentiferan species are known to occur in the Indian Ocean (see Van Dover et al. 2001). The two Atlantic species of Lamellibra- chia are presently known only from cold seep regions on continental margins. It is of interest to note that two Pacific species, 1.e., L. barhami and L. satsuma, have been col- lected from vent and seep sites, whereas a third described Pacific species, L. columna, is recorded only from vent sites (Southward & Galkin 1997, Tunnicliffe et al. 1998, Ko- jima et al. 2001). In a study based on a partial nucleotide sequence of the cyto- chrome oxidase I mitochondrial gene (COI) of specimens of Lamellibrachia collected from ten sites in the western Pacific, Koji- ma et al. (2001) recognized four tentative undescribed species, in addition to speci- mens identified as L. satsuma. Of these four undescribed species, two were collected from vent sites only, one was found exclu- 3} 9/2 sively in seep sites, and one occurred in vent and seep sites. Among species of Lamellibrachia pres- ently described, morphological characters of the specimens examined in this study are consistent with those of L. luymesi and L. victori. If these two species are shown to be identical in the future, L. /uymesi has priority. However, the type locality of L. luymesi off Guyana and localities in the northern Gulf of Mexico where specimens were collected for the present study are sep- arated by a distance of approximately 4000 km with localities of intermediate popula- tions not currently known. Based on se- quence similarity of the COI mitochondrial gene, McMullin et al. (2003) report that populations of a species of Lamellibrachia, which occur over a distance of 6000 km along the west coast of North America, rep- resent a single species, 1.e., L. barhami. Us- ing similar evidence, McMullin et al. (2003) also indicate that Paraescarpia echi- nospica exhibits a geographic range of at least 4500 km in the western Pacific Ocean. Based on morphological similarity, we sug- gest that L. /uymesi is distributed over a wide geographic range in the western At- lantic Ocean. Lamellibrachia luymesi co-occurs in the Gulf of Mexico with two additional vesti- mentiferan worms, Seepiophila jonesi, which is sympatric with L. luymesi in cer- tain seep communities (see Gardiner et al. 2001), and Escarpia laminata which was originally recorded from deep water at the base of the Florida Escarpment (Paull et al. 1984, Jones 1985). In addition to the seep sites reported in this study, a species of La- mellibrachia occurs along the Louisiana slope in the Gulf of Mexico in a large num- ber of other seep sites (see references in synonymy above for coordinates), extend- ing along a distance of at least 480 km and depths of 550 to 650 m of water. An anal- ysis of the COI gene of specimens taken from populations of Lamellibrachia along this distance of the Louisiana slope, includ- ing the seep sites used for this study, shows PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON genetic distances indicative of intraspecific variation (McMullin et al. 2003). Morpho- logical similarity of specimens examined in this study, together with the evidence from the analysis of the COI gene, support the suggestion that a single species of Lamel- librachia, 1.e., L. luymesi, inhabits hydro- carbon seep communities along the Loui- siana slope in the Gulf of Mexico. Acknowledgments This work was supported by the NOAA National Undersea Research Program at the University of North Carolina, Wilmington, Harbor Branch Oceanographic Institution, National Science Foundation OCE 0117050 to C. R. Fisher, and the Minerals Manage- ment Service, Gulf of Mexico Regional OCS Office through the contract number 1435-10-96-CT30813 to C. R. Fisher. We thank the captain and crew of the RV Ed- win Link and the submersible crew and pi- lots of the Johnson-Sea-Link. We thank Er- ica Nix, Jim Childress and Chuck Fisher for collecting and providing specimens and Dr. James Sidie and Dr. A. C. Allen of Ursinus College for making their SEM facilities available for our use. We thank Laura Shar- key and Adam Dewan, undergraduate stu- dents of Bryn Mawr College and Haverford College, respectively, for their assistance in obtaining and analyzing morphometric data used in this study. The senior author ex- tends special appreciation to Chuck Fisher, The Pennsylvania State University, for pro- viding ship time and submersible time on several cruises in the Gulf of Mexico. Literature Cited Bergquist, D. C., I. A. Urcuyo, & C. R. Fisher. 2002. Establishment and persistence of seep vestimen- tiferan aggregations on the upper Louisiana slope of the Gulf of Mexico.—Marine Ecology Progress Series 241:89—98. Brooks, J. M., et al. 1987. Deep-sea hydrocarbon seep communities: evidence for energy and nutrition- al carbon sources.—Science 238:1138-—1142. Carney, R. S. 1994. Consideration of the oasis analogy for chemosynthetic communities at Gulf of VOLUME 116, NUMBER 2 Mexico hydrocarbon vents.—Geo-Marine Let- ters 14:149-159. Childress, J. J.. & C. R. Fisher. 1992. The biology of hydrothermal vent animals: physiology, bio- chemistry, and autotrophic symbioses. Pp. 337— 441 in M. Barnes, A. D. Ansell, & R. N. Gib- son, eds., Oceanography and Marine Biology An Annual Review, vol. 30. Biddles Ltd, King’s Lynn and Guilford, England, 627 pp. 5 , J. M. Brooks, M. C. Kennicutt, II, R. Bidigare, & A. E. Anderson. 1986. A methan- otrophic marine molluscan (Bivalvia: Mytili- dae) symbiosis: mussels fueled by gas.—Sci- ence 233:1306—1308. Dando, P. R., A. J. Southward, E. C. Southward, D. R. Dixon, A. Crawford, & M. Crawford. 1992. Shipwrecked tube worms.—Nature 356:667. Feldman, R. A., M. B. Black, C. S. Cary, R. A. Lutz, & R. C. Vrijenhoek. 1997. Molecular phyloge- netics of bacterial endosymbionts and their ves- timentiferan hosts—Molecular Marine Biology and Biotechnology 6:268—277. Fisher, C. R. 1995. Toward an appreciation of hydro- thermal-vent animals: their environment, phys- iological ecology, and tissue stable isotope val- ues. Pp. 297-316 in S. E. Humphris, R. A. Zier- enberg, L. S. Mullineaux, & R. E. Thomson, eds., Seafloor hydrothermal systems: physical, chemical, biological, and geochemical interac- tions. Geophysical Monographs Series 91:1— 466. , J. J. Childress, & N. K. Sanders. 1988. The role of vestimentiferan hemoglobin in providing an environment suitable for chemoautotrophic sulfide oxidizing endosymbionts.—Symbiosis 5:229-246. , M. C. Kennicutt, HW, & J. M. Brooks. 1990. Stable carbon isotopic evidence for carbon lim- itation in hydrothermal vent vestimentiferans.— Science 247:1094—1096. , I. A. Urcuyo, M. A. Simpkins, & E. Nix. 1997. Life in the slow lane: growth and longev- ity of cold-seep vestimentiferans.—Marine Ecology 18:83—94. Freytag, J. K., P R. Girguis, D. C. Bergquist, J. P. Andras, J. J. Childress, & C. R. Fisher. 2001. A paradox resolved: sulfide acquisition by roots of seep tubeworms sustains net chemoautotro- phy.—Proceedings of the National Academy of Sciences 98:13408—13413. Gardiner, S. L., E. McMullin, & C. R. Fisher. 2001. Seepiophila jonesi, a new genus and species of vestimentiferan tube worm (Annelida: Pogono- phora) from hydrocarbon seep communities in the Gulf of Mexico.—Proceedings of the Bio- logical Society of Washington 114:694—707. Jones, M. L. 1981. Riftia pachyptila, new genus, new species, the vestimentiferan worm from the Ga- 598 lapagos Rift geothermal vents (Pogonopho- ra).—Proceedings of the Biological Society of Washington 93:1295—1313. . 1985. On the Vestimentifera, new phylum: six new species, and other taxa, from hydrothermal vents and elsewhere.—Bulletin of the Biologi- cal Society of Washington 6:117—158. Julian, D., EF Gaill, E. Wood, A. J. Arp, & C. R. Fisher. 1999. Roots as a site of hydrogen sulfide uptake in the hydrocarbon seep vestimentiferan Lamel- librachia sp.—The Journal of Experimental Bi- ology 202:2245—2257. Kennicutt, M. C., II, J. M. Brooks, R. R. Bidigare, R. R. Fay, T. L. Wade, & T. J. McDonald. 1985. Vent-type taxa in a hydrocarbon seep region on the Louisiana slope.—Nature 317:351—353. Kojima, S., S. Ohta, T. Yamamoto, T. Miura, Y. Fuji- wara, & J. Hashimoto. 2001. Molecular taxon- omy of vestimentiferans of the western Pacific and their phylogenetic relationship to species of the eastern Pacific. I. Family Lamellibrachi- idae.—Marine Biology 139:211-219. Land, J. van der, & A. N@rrevang. 1975. The system- atic position of Lamellibrachia [sic] (Annelida, Vestimentifera). Pp. 86-101 in A. N@rrevang ed., The phylogeny and systematic position of Pogonophora. Zeitschrift fiir zoologische Sys- tematik und Evolutionsforshung, Sonderheft 1, 143 pp. OS . 1977. Structure and relationships of Lamellibrachia (Annelida, Vestimenti- fera)—Det Kongelige Danske Videnskabernes Selskab, Biologiske Skrifter 21(3):1—102. MacDonald, I. R., G. S. Boland, J. S. Baker, J. M. Brooks, M. C. Kennicutt, I], & R. R. Bidigare. 1989. Gulf of Mexico hydrocarbon seep com- munities II. Spatial distribution of seep organ- isms and hydrocarbons at Bush Hill.—Marine Biology 101:235—247. , N. L. Guinasso, Jr., J. E Reilly, J. M. Brooks, W. R. Callender, & S. G. Gabrielle. 1990. Gulf of Mexico hydrocarbon seep communities: VI. Patterns in community structure and habitat.— Geo-Marine Letters 10:244—252. Mané-Garzon, FE, & R. Montero. 1985. Sobre una nue- va forma de verme tubicola Lamellibrachia vic- tori [sic] n.sp. (Vestimentifera)—Proposicion de un nuevo phylum: Mesoneurophora.—Revista de Biologia del Uruguay 8:1—28. McMullin, E. R., S. Hourdez, S. W. Schaeffer, & C. R. Fisher. 2003. Phylogenetics and biogeogra- phy of deep-sea vestimentiferan tubeworms and their bacterial symbionts.—Symbiosis 34:1—41. Miura, T., J. Tsukahara, & J. Hashimoto. 1997. La- mellibrachia satsuma, a new species of vesti- mentiferan worms (Annelida: Pogonophora) from a shallow hydrothermal vent in Kagoshi- 394 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ma Bay, Japan.—Proceedings of the Biological Society of Washington 110:447—456. Nelson, K., & C. R. Fisher. 2000. Absence of cospe- ciation in deep-sea vestimentiferan tube worms and their bacterial endosymbionts.—Symbiosis 28:1-15. Olu-Le Roy, K., M. Sibuet, G. Levitre, S. Gofas, & A. Fiala-Médioni. 2001. Cold seep communities in the deep Mediterranean Sea (South of Crete and Turkey): faunistic composition and spatial dis- tribution.—Program and abstracts. Second In- ternational Symposium on Deep-sea Hydrother- mal Vent Biology. Brest, France. [Abstract, p. 208]. Paull, C. K., et al. 1984. Biological communities at the Florida Escarpment resemble hydrothermal vent taxa.—Science 226:965—967. Salvini-Plawen, L. V. 2000. What is convergent/ho- moplastic in Pogonophora?—Journal of Zoo- logical Systematics & Evolutionary Research 38:133-147. Schulze, A. 2001. Comparative anatomy of excretory organs in vestimentiferan tube worms (Pogo- nophora, Obturata).—Journal of Morphology 250:1-11. Scott, K. M., & C. R. Fisher. 1995. Physiological ecol- ogy of sulfide metabolism in hydrothermal vent and cold seep vesicomyid clams and vestimen- tiferan tube worms.—American Zoologist 35: 102-111. Southward, E. C. 1991. Three new species of Pogo- nophora, including two vestimentiferans, from hydrothermal sites in the Lau Back-arc Basin (Southwest Pacific).—Journal of Natural His- tory 25:859-881. . 1999. Development of Perviata and Vesti- mentifera (Pogonophora).—Hydrobiologia 402: 185-202. , & S. V. Galkin. 1997. A new vestimentiferan (Pogonophora: Obturata) from the hydrothermal vent fields in the Manus Back-arc Basin (Bis- marck Sea, Papua New Guinea, Southwest Pa- cific Ocean).—Journal of Natural History 31: 43-55. , A. Schulze, & S. L. Gardiner. 2003. Pogo- nophora: form and function (Annelida).—Hy- drobiologia (in press). , & V. Tunnicliffe. 2002. Vestimenti- ferans (Pogonophora) in the Pacific and Indian Oceans: a new genus from Lihir Island (Papua New Guinea) and the Java Trench, with the first report of Arcovestia ivanovi from the North Fiji Basin.—Journal of Natural History 36:1179— 1197. Tunnicliffe, V., A. G. McArthur, & D. McHugh. 1998. A biogeographical perspective of the deep-sea hydrothermal vent fauna.—Advances in Marine Biology 34:353—442. Tyler, PR A., & C. M. Young. 1999. Reproduction and dispersal at vents and cold seeps.—Journal of the Marine Biological Association, United Kingdom 79:193—208. Van Dover, C. L., et al. 2001. Biogeography and eco- logical setting of Indian Ocean hydrothermal vents.—Science 294:818—823. Webb, M. 1969. Lamellibrachia barhami, gen. nov., sp. nov. (Pogonophora), from the Northeast Pa- cific.—Bulletin of Marine Science 19:18—47. Williams, N. A., D. R. Dixon, E. C. Southward, & P. H. W. Holland. 1993. Molecular evolution and diversification of the vestimentiferan tube worms.—Journal of the Marine Biological As- sociation, United Kingdom 73:437—452. Young, C. M., E. Vazquez, A. Metaxas, & P. A. Tyler. 1996. Embryology of vestimentiferan tube worms from deep-sea methane/sulfide seeps.— Nature 381:514—516. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):395—422. 2003. Four new genera of leucosiid crabs (Crustacea: Brachyura: Leucosiidae) for three new species and nine species previously in the genus Randallia Stimpson, 1857, with a redescription of the type species, R. ornata (Randall, 1939) Bella S. Galil National Institute of Oceanography, Israel Oceanographic and Limnological Research, P.O.B. 8030, Haifa 31080, Israel, e-mail: bella@ocean.org.il Abstract.—A study of the leucosiid genus Randallia Stimpson, 1857, led to the description of four new genera: Tanaoa, for R. distincta Rathbun, 1893, R. pustulosa Wood-Mason, in Wood-Mason & Alcock, 1891, and a new species, T. nanus; Tokoyo for R. eburnea Alcock, 1896, and a new species, T. cirrata; Toru for R. granuloides Sakai, 1961, R. trituberculata Sakai, 1961, R. pila Tan, 1996, R. mesjatzevi Zarenkov, 1990, and a new species, 7. septimus; and Ur- ashima, for R. lamellidentata Wood-Mason, 1892, and R. pustuloides Sakai, 1961. Randallia is restricted to its type species, R. ornata (Randall, 1839), and provisionally 12 other species currently placed in this genus pending further revision. All new genera are diagnosed and species assigned to them described or redescribed and illustrated; extended synonymies are given, and a key for species identification is provided. The type species, R. ornata, is redescribed. The genus Randallia Stimpson 1857a was established for //ia ornata Randall, 1839, a leucosiid crab known from the Pa- cific and Gulf coasts of California (Stimp- son, 1857a). The chaotic leucosiid system- atics and the fact that Stimpson (1857a:85; 1857b:471) gave but a cursory description, allowed for a miscellaneous assortment of leucosiid crabs to be relegated to that ge- nus. Several authors (Doflein 1904, Seréne & Soh 1976, Tan 1996) regarded Randallia as a heterogenous genus in need of revision. Yaldwyn & Dawson (1976) sorted Randal- lia species into four ill-fitting “‘species groups” according to rugosity of the cara- pace and length of chelipeds, while disre- garding the variation in the segmentation of the male abdomen, structure of the first male pleopod, and other morphological fea- tures. Though 30 species have been hitherto assigned to Randallia, doubts remained as to their systematic position (Ovaere 1989). A study of the extensive collections of the National Museum of Natural History, Smithsonian Institution, Washington, D.C., together with other major collections has enabled re-examination of many type spec- imens and much of the published material, and led to a reevaluation of Randallia. As result, the genus is herein restricted to its type species, R. ornata (Randall 1839), known from the eastern Pacific, and 12 oth- er species provisionally retained in Randal- lia s. s. pending further revision (Table 1). Of the other 17 species hitherto assigned to Randallia s. \., R. angelica Garth, 1940 was synonymized with R. ornata (Randall, 1839), by Hendrickx (1997). Randallia ja- ponica Yokoya, 1933 was declared a junior synonym of R. eburnea Alcock, 1896, by Sakai (1934). Four species were transferred to other genera: R. coronata Alcock & An- derson, 1894 to Pariphiculus Alcock, 1896, by Alcock (1896); R. lanata Alcock, 1896 and R. villosa Chen, 1989 to Jhleus Ovaere, 1989, by Ovaere (1989); and R. mirabilis 396 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Generic assignment of species hitherto attributed to Randallia Stimpson, 1857a (* species herein provisionally retained in Randallia s. s., pending further revision). *R. agaricias Rathbun, 1898. *R. americana (Rathbun, 1893) R. angelica Garth, 1940: synonymized with R. ornata (Randall, 1839) (see Hendrickx 1997). *R. bulligera Rathbun, 1898 R. coronata Alcock & Anderson, 1894: reassigned to Pariphiculus Alcock, 1896 (see Alcock 1896) *R. curacaoensis Rathbun, 1922 R. distincta Rathbun, 1893: placed in Tanaoa, new genus *R. eburnea Alcock, 1896: placed in Tokoyo, new genus *R. glans Alcock, 1896 *R. gilberti Rathbun, 1906 *R. granulata Miers, 1886 R. granuloides Sakai, 1961: placed in Toru, new genus R. japonica Yokoya, 1933: synonymized with R. eburnea Alcock, 1896 (see Sakai 1934) *R. laevis (Borradaile, 1916) . lamellidentata Wood-Mason, 1892: placed in Urashima, new genus . lanata Alcock, 1896: reassigned to Ihleus Ovaere, 1989 (see Ovaere 1989) . minuta Rathbun, 1935 R R R. mesjatzevi Zarenkov, 1990: placed in Toru, new genus R R . mirabilis Zarenkov, 1969: reassigned to Raylilia Galil, 2001 (see Galil 2001) *R. nana Zarenkov, 1990 R. ornata (Randall, 1839): type species of Randallia Stimpson, 1857a R. pila Tan, 1996: placed in Toru, new genus *R. pustulilabris Alcock, 1896 . pustuloides Sakai, 1961: placed in Urashima, new genus . pustulosa Wood-Mason, in Wood-Mason & Alcock, 1891: placed in Tanaoa, new genus . serenei Richer de Forges, 1983: is placed in synonymy with Tanaoa distinctus (Rathburn, 1893) . trituberculata Sakai, 1961: placed in Toru, new genus . villosa Chen, 1989: reassigned to [hleus Ovaere, 1989 (see Ovaere 1989) . vitjazi Zarenkov, 1994: placed in synonymy with Tanaoa pustulosus (Wood-Mason, in Wood-Mason & R R R *R. speciosa Chen, 1989 R R R Alcock, 1891) Zarenkov, 1969 to Raylilia Galil, 2001, by Galil (2001). Four new genera are herein established for three new species, and nine species previously in Randallia s. |. Ran- dallia serenei Richer de Forges, 1983, and Randallia vitjazi Zarenkov, 1994, were rec- ognized as junior synonyms of previously described species. Randallia s. s. differs from the newly established genera in hav- ing the antennular operculum entirely seal- ing the antennular aperture, the anterior margin of efferent branchial channel trilo- bate, and the male abdominal segments 3— 5 fused. All species in the new genera are described or redescribed and illustrated, ex- tended synonymies given, and a key for their identification is provided. The type species, R. ornata, 1s also redescribed. Abbreviations used are: btw, between; coll., collector; CP, chalut a perche (beam trawl); CH, chalut (trawl); DW, Waren dredge; I., Island; Is., Islands; Lt., Light; Pt., Point; Stn, station. The French expedition BATHUS was named after the Greek word for deep, bathys. The other French expedi- tions are identified by acronyms: BORDAU, a contraction of “‘bordure d’ Australo-indi- enne plateau”; CHALCAL, “‘chalutage New Caledonia’; HALIPRO, “‘halieutique pro- fonde”; KARUBAR, a contraction of the names of Kai, Aru and Tanimber Islands; MUSORSTOM was organized jointly by the Muséum national d’Histoire naturelle and the Office de la Recherche Scientifique et Technique Outre-Mer (ORSTOM). The length of each specimen was mea- VOLUME 116, NUMBER 2 sured along the vertical median line of the carapace, excluding intestinal spine. The material used remains deposited in the following museums: National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM); The National Natuurhistorische Museum, Leiden (former- ly Ryksmuseum van Natuurliyke History) (NNM); Museum national d’ Histoire natu- relle, Paris (MNHN); The Natural History Museum, London (NHM); Australian Mu- seum, Sydney (AMS); National Institute of Water & Atmospheric Research, New Zea- land (NIWA); National Taiwan Ocean Uni- versity, Keelung (NTOU); Queensland Mu- seum, Brisbane (QM); South African Mu- seum, Cape Town (SAM); Senckenberg Mu- seum, Frankfurt (SMF); Western Australian Museum, Perth (WAM); Zoological Muse- um, Amsterdam (ZMA); and Zoological Museum, Moscow University (ZMMU). Randallia ornata (Randall, 1839) Figs. 1A, 3A, B Ilia ornata Randall, 1839:129. Guaia ornata Gibbes, 1850:186. Randallia ornata Stimpson, 1857a:85; 1857b:471, pl. 20 fig. 3; 1860:69.—Rath- bones 9Sol3- (S04 5170- 1937-2 pl: 49, figs 1-2.—Holmes, 1900:100.—Wey- mouth, 1910:18, pl. 1, fig. 3.—Baker, oe O22 —Schmmtt. SOQ S Sear ie: 116.—Seréne, 1954:491.—Garth, 1960: 111; 1966:10.—Richer de Forges, 1983: 634 (tab.).—Austin, 1985:646.—Bonfil & Carvacho, 1989:83, fig. 4a—Hen- drickx., 1990:452 1992-7. 1995-129: 1997-103, fie. 114. Randallia angelica Garth, 1940:54; pl. 11, figs 1—-2.—Seréne, 1954:492.—Richer de Forges, 1983:634 (tab.).—Rodriguez de lasGruz. (9873120; Not Randallia ornata Boone, 1930:59, pl. 12 [= R. bulligera Rathbun fide Garth 1966:10]. Type material.—Paratypes of Randallia angelica Garth, 1940: Puerto Refugio, An- gel de la Guardia I., R/V Velero, Stn 541- 397) 36, 10m, 4 Mar 1936.6 '9'smme 2 1s)5 mm (USNM 139772). Material examined.—United States. Cal- ifornia, San Francisco, 1880, coll. D. S. Jor- dane 2 3022338)3 mm (OSNM 3115). San Francisco Bay, R/V Albatross, Apr 1914, 3 44.0 mm (USNM 55532). Golden Gate, 21 June Oils 2 Voie. 59°38) mim CuUSINIM 66506). Monterey Bay, Santa Cruz Light- house, R/V Albatross, Stn 4560, 18 m, 11 Jun 1904, 5 30.2 mm (USNM 66505). San- tauBarbaras iSs0s coll DES] Jordans 4 6 48.8-54.4 mm, 4 @ ovig. 30.2—33.4 mm (USNM 3101). Mugu Bay, Ventura Co., poy NEVZ., Coll, 135 12 (Cees, G ZY a0 2 ovig. 27.4 mm (USNM 57284). Santa Cruz I., R/V Albatross, 7 Feb 1889, 2 40.0 mm (USNM 17394). San Pedro I., Mar 1931, 6 26.3 mm (USNM 21791). Long Beach, coll. H. N. Lowe, 2 ¢ 28.9, 41.9 orn, 2 2 325, Joos iden (USN Aes Newport Bay, 16 Jan 1939, coll. S. A. Gla- valli 4G NOL2335 Sin, LS iey tase (USNM 207834). Santa Catalina I., R/V Anton Dohrn, 30 Dec 1912, 2 ¢ 16.4, 26.2 mm, @ 18.2 mm (USNM 50115). Catalina harbour, 23 Jun 1916, 5 15.6 mm, §& ovig. 17.6 mm (USNM 66488). SW Catalina har- bour, 23 Jun 1916, 2 19.4 mm (USNM 66496). San Nicolas I., R/V Albatross, Stn 4422, 57 m, 13 Apr 1904, 2 21.4 mm (USNM 66504). San Diego Bay, R/V AI- batross, 1 Mar 1904, 2 33.4 mm (USNM 66507). Mexico. Lower California, Playa Maria Bay, 24—26 Aug 1896, coll. A.W. Anthony, 3 6 SiLS=5s53) imma, S) 2 Ones Ys s/o) mm (USNM 19521). Corona del Mar, 13 m, Jul 1935, coll. G. E. MacGinitie, d 48.8 mm (USNM 89739). Mar 1948, coll. G. E. MacGinitie, 2 ovig. 30.8 mm (USNM 89742). Balboa, 0O—27.5 m, coll. S. H. Glas- Selly4ne IRS =27/-3mme4 2 1ssl=3523 mam (USNM 207834). Redescription.—Dorsal surface of cara- pace smooth, minutely shagreened anteri- orly. Frontal lobes triangulate, anteriorly granulate. Anterolateral margin with sub- hepatic granulate tubercle, 3 or more pear- 398 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. A, Randallia ornata (Randall, 1839): 3 cl 30.2 mm, dorsal view, California, Monterey Bay, R/V Albatross, Stn 4560, USNM 66505; B, Tanaoa distinctus (Rathbun, 1893): ¢ cl 40.5 mm, dorsal view, Banc Tuscarora, MUSORSTOM 7, Stn DW 556, MNHN; C, Tanaoa nanus, new species: holotype, ¢ cl 12.7 mm, dorsal view, Vanuatu, MUSORSTOM 8, Stn CP 1053, MNHN; D, Tanaoa pustulosus (Wood-Mason, in Wood- Mason & Alcock, 1891): d cl 35.8 mm, dorsal view, Btw Negros, Siquijor, R/V Albatross Stn 5538, USNM; E, Tokoyo cirrata, new species: holotype, d cl 13.0 mm, dorsal view, Vanuatu, MUSORSTOM 8, Stn CP 1086, MNHN; E Tokoyo eburnea (Alcock, 1896): d cl 14.8 mm, dorsal view, Japan, Tosa Bay, SMF 22577. liform granules on epibranchial margin. ic region tumid, topped by | or more pear- Posterolateral margin with small, triangular liform granules. Intestinal region slightly denticle. Posterior margin bearing 2 dorso- inflated, bearing a granule (Fig. 1A). ventrally flattened triangular denticles lat- Anterior margin of efferent branchial erally, pearliform granules medially. Hepat- channel granulate, with 3 subequal lobes. VOLUME 116, NUMBER 2 399 Fig. 2. A, Toru granuloides (Sakai, 1961): 5 cl 23.0 mm, dorsal view, Loyalty Is, MUSORSTOM 6, Stn DW 456 (MNHN); B, Toru pilus (Tan, 1996): 3 cl 10.8 mm, dorsal view, Vanuatu, MUSORSTOM 8, Stn CP 1047 (MNHN); C, Toru septimus, new species: holotype. d cl 13.1 mm, dorsal view, Loyalty Is., Stn DW 421 (MNHN); D, Toru trituberculatus (Sakai, 1961): 3 cl 13.0 mm, dorsal view, Japan, Mimase (SMF); E, Urashima lamellidentatus (Wood-Mason, 1892): holotype. 6 cl 11.9 mm, dorsal view, Andaman Is. (NHM 1896.9.8.7); E Urashima pustuloides (Sakai, 1961): 5 cl 32.8 mm, dorsal view, Taiwan, I-Lan county (MNHN B26326). Third maxilliped anteriorly setose, granu- podus swollen, smooth but for minutely lose. granulate upper margin; fingers, longer than Cheliped merus 0.75 as long as carapace, palm, set with longitudinal granulate ridges. set with pearliform granules; carpus with Pereiopodal carpi 1—4 with upper margin few granules distally on upper margin; pro-_ distally granulose; upper margin of propodi 400 i A B C D ica: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A E F G H A, B, Randallia ornata (Randall, 1839), ¢ cl 30.2 mm, California, Monterey Bay, R/V Albatross, Stn 4560 (USNM 66505): A, first pleopod; B, apex of first pleopod. C, D, Tanaoa distinctus (Rathbun, 1893), 3d cl 40.5 mm, Banc Tuscarora, MUSORSTOM 7, Stn DW 556, (MNHN): C, first pleopod; D, apex of first pleopod. E, E Tanaoa nanus, new species, d cl 13.7 mm, Indonesia, Tanimbar I., KARUBAR, Stn CP 39 (MNHN): E, first pleopod; EK apex of male first pleopod. G, H, Tanaoa pustulosus (Wood-Mason, in Wood- Mason & Alcock, 1891), d cl 34.0 mm, Japan, Tosa Bay (NHM 1961.6.5.38/39): G, first pleopod; H, apex of first pleopod. Scales = | mm. 1—4 bearing medially granulate ridge; lower margin of fifth pereiopodal merus, propo- dus granulate. Thoracic sternites, abdomen minutely pitted. Male first pleopod slightly sinuous, dorso-ventrally flattened, tip thickened, with dense tuft of setae subapically on in- ternal margin (Figs. 3A, B). Color.—**Carapace variegated with san- guineous spots, confluent anteriorly; cheli- peds variegated with red” (Rathbun 1937: 172). ““Ground color of carapace olive buff almost covered anteriorly with dots of chrome yellow. Large, regular designs vi- naceous russet, smaller patches paler and more orange. Posterior spines white. Che- liped yellow to pale buff, merus covered with carrot red, carpus and manus with a coarse netting of the same color. Merus of ambulatory legs pale yellowish white at base blending into intense scarlet on distal portion. Remaining segments yellowish white; dactyl yellow tipped.”’ (Garth 1940: 5D)s Distribution.—Eastern Pacific: Pacific and Gulf coasts of California, 10O—185 m. Remarks.—Randall’s (1839) specimens are no longer extant (Rathbun 1937:172). According to Garth (1940) R. angelica dif- fers from R. ornata in degree of granula- tion, more prominent frontal teeth and more pronounced angle of front with hepatic margin. Two decades later Garth (1960) considered it a synonym of R. ornata, and yet six years later (Garth 1966) reversed himself again and considered R. angelica a “Gulf of California cognate” of R. ornata. Hendrickx (1997:163), who examined spec- imens collected off the Pacific and Gulf of California coasts, maintained that R. angel- ica is but a synonym of R. ornata. Examination of large series of speci- mens, including Garth’s paratypes, revealed intraspecific variability in granulation. Smaller specimens (cl < 25 mm) tend to bear coarser, denser granules on the cara- pace and legs. The differences in granula- tion enumerated by Garth (1940:56) as characteristic of R. angelica fall within the range of variation observed for R. ornata. The 12 species provisionally assigned to Randallia s. s. differ from R. ornata in a number of characters. Randallia americana (Rathbun, 1893), R. glans Alcock, 1896, R. VOLUME 116, NUMBER 2 agaricias Rathbun, 1898 and R. speciosa Chen, 1989 differ from R. ornata in having a bilobate margin to the efferent branchial channel. Randallia granulata Miers, 1886, R. gilberti Rathbun, 1906, and R. nana Zar- enkov, 1990 differs from R. ornata in hav- ing segments 3—6 of male abdomen fused in. addition to the bilobate margin to effer- ent branchial channel. Randallia pustulila- bris Alcock, 1896, R. laevis (Borradaile, 1916) and R. minuta Rathbun, 1935 differ from R. ornata in having the antennular operculum sealing only the bottom half of antennular aperture. Randallia bulligera Rathbun, 1898 differs from R. ornata in its male first pleopod having a petaloid tip. Randallia curacaoensis Rathbun, 1922 dif- fers from R. ornata in having the sixth ab- dominal segment in the male bearing prox- imally a triangular denticle. As previously mentioned, these species are herein retained in Randallia s. s. pending further revision, rather than leave them as incertae sedis. Tanaoa, new genus Diagnosis.—Carapace subcircular, glo- bose. Front narrow, uptilted, bilobed. Eyes small, retractible. Outer orbital margin tri- sutured, V-shaped gap proximally on ven- tral margin. Antennules obliquely folded, basal antennular segment squat, operculi- form, sealing lower antennular aperture. Antennae small, slender, basal antennal seg- ment inserted in orbital hiatus. Postorbital region concave. Branchial, intestinal re- gions swollen, demarcated by grooves, 2 pairs of pits along cardiobranchial grooves. Posterior margin of carapace narrow, biden- tate. Third maxilliped exopod not quite seal- ing efferent branchial channel. Third max- illiped exopod slightly shorter than endo- pod; merus of endopod subtriangular, short- er than subrectangular ischium. Anterior margin of efferent branchial channel pro- duced, bilobed, separated by narrow groove from lower orbital margin. Chelipeds long, slender, equal. Cheliped 401 with merus and palm subcylindrical; fingers nearly as long as upper margin of palm, in- ner margins denticulate. Pereiopods 1-5 slender, short; all but last dactyl shorter than propodi; upper surface of pereiopodal dac- tyls setose, tips corneous. Fourth thoracic sternite swollen laterally. Male abdominal sulcus deep, nearly reach- ing buccal cavity, anterior margin raised, prominently granulate. Male abdomen tri- angular, abdominal segments 3-6 fused, basio-lateral regions of fused segments in- flated, anterior margin bearing denticle. Tel- son slender, one-third as long as fused seg- ments, not reaching tip of abdominal sul- cus. Walls of abdominal sulcus carinate along sternal sutures 2—4. Female abdomi- nal segments 4—6 fused, swollen, shield- like. Telson ogival, fitting between third maxillipeds, posteriorly sinuous. Male first pleopod elongate, slightly sin- uous, attenuate; bearing minute preapical process perpendicular to tip (Fig. 4a, b); second pleopod short, distally scoop-like. Type species.—Randallia pustulosa Wood-Mason, in Wood-Mason & Alcock, 1891. Gender: feminine. Species.—Tanaoa distinctus (Rathbun, 1893), Tanaoa nanus, new species, Tanaoa pustulosus (Wood-Mason, in Wood-Mason & Alcock, 1891). Etymology.—In the myths of the Mar- quesas islanders, Tanaoa is the god of dark- ness, confined to the depths of the ocean. The name Tanaoa is to be considered as an arbitrary combination of letters and here- tofore takes the gender masculine. Remarks.—Tanaoa, new genus, differs from Randallia s. s. as follows: antennular operculum seals only the lower antennular aperture; the third maxilliped exopod and efferent branchial channel gape anteriorly; the anterior margin of the efferent branchial channel is bilobate; male abdominal seg- ments 3—6 are fused; and the male first ple- opod is distally attenuate, bearing a preap- ical process. In contrast, in Randallia s. s. the antennular operculum seals entirely the antennular aperture; the third maxilliped ex- 402 opod seals the efferent branchial channel; the anterior margin of the efferent branchial channel is trilobate; the sixth segment of the male abdomen is free; and the male first pleopod is distally club-shaped. Tanaoa distinctus (Rathbun, 1893), new combination Figs. 1B, 3C, D Randallia distincta Rathbun, 1893:257; 1906:890, pl. 16, fig. 2, 3, text-fig. 44. Ihle, 1918:312.—Seréne, 1954:492; 1968:45.—Yaldwyn & Dawson, 1976: 97.—Richer de Forges, 1983:634(tab), 636, 638. Randallia serenei Richer de Forges, 1983: 634, figs 1—4.—Poupin & Richer de Forces, 199-21 Type material.—Holotype of Randallia distincta Rathbun, 1893: Hawaiian Is., Oahu I., R/V Albatross Stn 3472, 21°12'N, 157°49’W, 4 Dec 1891, 540 m, 2 20.4 mm (USNM 17516). Holotype of Randallia ser- enei Richer de Forges, 1983: Tahiti. Port Phaeton, 400—500 m, 1978, coll. B. Richer de Forges, 2 6 40.3, 41.8 mm (MNHN B8735). Material examined.—Marquesas Is., Ta- huata I., 430 m, 13 Sep 1987, 3 32.5 mm, 2? 31.0 mm (MNHN); 9°54.5’S, 139°08.2'W, 350 ma, I Sao IO, coll Ie ou, 2 6 ZLO, 40.0 tom, 2 2 Sis, 43.7 iain (MNHN). Elao I., 7°58.5'S, 140°44.5’W, 415 m, 19 Jan 1991, coll. J. Poupin, 6 d 39.7— 44.3 mm, 10 @ 22.6—-42.9 mm (MNHN). MUSORSTOM 9, Stn CP 1169, 8°59’S, 140°0S5'’W, 391-408 m, 24 Aug 1997, 6 32.2 mm (MNHN). Stn CP 1191, 8°46’S, 140°07'W, 390-400 m, 26 Aug 1997, 6 41.7 mm (MNHN). Stn CP 1251, 9°47’S 139°38'W, 500-650 m, 2 Sep 1997, 3 40.0 mm (MNHN). Stn CP 1268, 7°56’S 140°43’'W, 285-320 m, 4 Sep 1997, 3 43.6 mm, 2 ovig. 39.1 mm (MNHN). Stn CP 1270, 7°56'S 140°43'W, 497-508 m, 4 Sep IMQo7, 6 208 m0, 2 BilS imam, yun (MNHN). Stn CP 1276, 7°52'’S 140°37'W, MOBS mm S Seo IO, S Als saan PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (MNHN). Stn CP 1281, 7°48’S 140°21’W, 450-455 m, 7 Sep 1997, 2 31.5 mm (MNHN). Sth DW 1287, 7°54'S, 140°40'W, 163-245 m, 7 Sep 1997, 36 42.6 mm (MNHN). Tuamotu Archipelago. 18°04.2’'S, 141°01.8’W, 500 m, 2 Jun 1990, coll. J. Poupin, ¢ 43.0 mm, ¢ ovig. 42.7 mm (MNHN). Makemo, 300—600 m, 4 Jun 1988, coll. J. Poupin, & 32.5 mm (MNHN). Society Is., 15°48’S, 154°32’W, 500-700 m, 21 Jul 1988, coll. J. Poupin, ¢ 40.6 mm (MNHN). Tahiti. Port Phaeton, 400—500 m, 6 Oct 1978, coll. B. Richer de Forges, ¢ 40.3 mm (MNHN B8734). Bora-Bora L., 400-700 m, 23 Jul 1988, coll. J. Poupin: 6 39.5 mm (MNHN). Taravao, 17°47’'S, 149°21'W, 500—600 m, 11 Dec 1988, coll. J. Poupin, 3 ¢ 30.8-39.5 mm, 2 40.7 mm (MNHN). Hawaiian Is. Maui I., W Puniawa Pt, R/V Albatross Stn 4079, 21°01.40'N, 156°22.50'W, 261-326 m, 21 Jul 1902 3 42 5 mm, 2 ovie. 39. 7 mimi (USINIVE 29883). R/V Albatross Stn 4082, 21°04.35 N, 156°21.10'W, 402—435 m, 21 Jul 1902 3 5d 20.9-30.8 mm, 2 2 32.3, 43.6 mm (USNM 29884). Oahu I., SW Diamond Head Lt, R/V Albatross Stn 3813, 483— 335 m, 28 March 1902, 6 30.4 mm (USNM 29872). SW Diamond Head Lt, R/V Albatross Stn 3818, 536-540 m, 31 Mar 1902, 6 20.5 mm (USNM 29873). Oahu, SW Kahuku Pt, R/V Albatross Stn A115; 21415 N, 1587085 Wi 357544 m, 25 Jul 1902, 2 29.8 mm (USNM 29885). Off Honolulu, 27—40 m, Feb—Mar 1962: 2 ovig. 41.6 mm (WAM c24429). Hawaii I., Kawaihae Lt., R/V Albatross Stn 4044, 20°03.15'N, 155°55.20'W, 426— 362 m, 11 Jul 1902, 3 43.0 mm, 2 32.7 mm (USNM 29882). Pailolo Channel, btw Maui, Molokai Is.; R/V Albatross Stn 3883, 21°09.15’N, 155°34.15'W, 507-520 m, 16 Apr 1902, 2 6 12.6, 18.6 mm, 2 19.8 mm (USNM 29878). Pailolo Chan- nel, btw Maui, Molokai Is., R/V Albatross Stn 3865, 468-518 m, 10 Apr 1902, 4 ¢ 1353= 2007 Vain, SS rome COSINIVIE VOLUME 116, NUMBER 2 29877). Molokai I., R/V Albatross Stn 3836, 21°00.05N, 157°08.20'W, 435—467 im, 3 Aye IDO 2 2 HOY, ikesos) i170 (USNM 29874). Western Samoa. Upolu I., Apia, 250—846 m, 5-16 Sep 1980, coll. D. Popper, d 43.5 mm, 2 30.8 mm (NNM 35234). Wallis Is. MUSORSTOM 7, Stn DW ae lle Semi7olo wwe 500-600Nm: 13 May 1992, 2 42.2 mm (MNHN). Banc Tuscarora. MUSORSTOM 7, Stn DW 556, 11°49’S, 178°18'W, 440 m, 19 May 1992, 3 40.5 mm (MNHN). INeweZealande 3472 4700S) 173 0:3: 472 m, 21 Mar 1982, 3 35.9 mm (NIWA). Guam. Agana Bay, 303 m, 28-9 Aug OV a collemleaaEidredge.\2 24,9 smi (MNHN B8737). Redescription.—Dorsal surface of cara- pace covered with pearliform granules, in- terspaced with smaller granules, granules more pronounced posteriorly. Frontal lobes triangulate, minutely and closely granulate. Third maxilliped prominently granulose. Subhepatic margins of carapace somewhat inflated, 3 low granulate tubercles on epi- branchial margin, more pronounced in ju- veniles. Hepatic, branchial, and intestinal regions demarcated by shallow grooves. In- testinal region bearing small tubercle pos- teriorly, tubercle reduced in larger speci- mens; conical, upcurved in juveniles. Pos- terior margin bearing 2 stubby protrusions laterally (Fig. 3C, D). Cheliped and pereiopods 1-5 closely granulate throughout. Cheliped merus in adult male nearly as long as carapace; fin- gers nearly as long as upper margin of palm. Pereiopodal dactyls tomentose ante- riorly. Thoracical sternites granulate. Fused ab- dominal segment in male triangular, bearing transverse ridge, with preapical median denticle. Telson slender, third as long as fused abdominal segments. Female abdo- men granulate, granules larger, closer prox- imally, low denticle medially on distal mar- gin. Male first pleopod with transverse dig- itate process preapically (Fig. 1B). 403 Distribution.—Pacific Ocean: Marquesas Is., Tuamotu Archipelago, Society Is., Ha- walian Is., Samoa, Banc Tuscarora, Wallis Is., New Zealand, Guam; 27—805 m. Remarks.—Rathbun (1893:257) de- scribed Randallia distincta from a juvenile female specimen, but on examining addi- tional material collected by the Albatross, observed that adult specimens differ from juveniles in lacking tubercles on the bran- chial margins, denticles on the posterior margin, and a spine on the intestinal region (Rathbun 1906:890). Richer de Forges (1983:634) based his description of R. ser- enei on adult specimens, but noted (1983: 638) that a juvenile from Guam differs from the adults in possessing “‘les gros granules du bord latéral et le granule de I’ aire intes- tinal’’. Richer de Forges (1983:638), distin- guished R. serene: from R. distincta in hav- ing more rounded tubercules on the poste- rior margin of the carapace and pronounced branchio-cardiac grooves, though admitting ‘*Pour mieux décrire chacune de ces espe- ces, il serait nécessaire une gamme de taille de chaque espéce’’. Examination of the type series of R. distincta and R. serenei, and numerous additional specimens, including the male first pleopod, has shown that the latter is a junior synonym of the former. The specimen collected off New Zealand has slimmer, longer chelipeds than the other specimens examined. Tanaoa nanus, new species Fics, I@BEy Randallia pustulosa: thle, 1918:246 (not Randallia pustulosa Wood-Mason, in Wood-Mason & Alcock, 1891). Type material.—Holotype: Vanuatu. MU- SORSMOM 8) 7Stn TEP) 10535) 16°29.23"S, 167°58.70'E, 536-519 m, 1 Oct 1994, coll. B. Richer de Forges, d 12.7 mm (MNHN B.28510). Paratype: Indonesia: 1°17.5’N, 118°53’E, N of Kaniungan, Siboga Stn 90, 281 m, 21 Jun 1899, 3 14.7 mm (ZMA 242432). Material examined.—Vanuatu. MUSOR- 404 SHOMeS: Stn GPa O27 au ilii<5 32051 S: 168°39.35’E, 550-571 m, 28 Sep 1994, coll. B. Richer de Forges, 2 13.5 mm GMENTSIND), Sim Cle 1Q47> ie 53.62 7S. 168°10.49'E, 486-494 m, 30 Sep 1994, Coll, 13. JNicher ce ikormyes, © Y.5 main GMUNJBIND). Stn (iP WOS2. IkG 3y.37 oS. 168°00.29'E, 561-564 m, 1 Oct 1994, coll. B. Richer de Forges, 6 12.3 mm (MNHN). Sim CP 054 le 27'S. Woy s/ 44 18, 522-527 m, | Oct 1994, coll. B. Richer de Forges, d 13.1 mm, @ ovig. 12.4 mm GMONTEUND), Sie CIP W@S55 ile 3O. IS. lo. 13 TES 572-580 m, Oct 19947 coll B. Richer de Forges, 2 ovig. 15.3 mm (MNHN). Stn DW _ 1072, 15°39.89'S, 167°19.61'E, 622—625 m, 4 Oct 1994, coll. B. Richer de Forges, 5 10.9 mm (MNHN). Stnec€P WlOs9) Ss-08:827S, 16717 23-53 494-516 m, 6 Oct 1994, coll. B. Richer de Forges, 6 10.9 mm 2 Pov. 12745) nom @VIINIEUND A Sin CR nine IA oe O094SF 167°14.00’E, 1210-1250 m, 8 Oct 1994, coll. B. Richer de Forges, 2 ovig. 14.1 mm GMONDEIND)s “Sty (Oe wa SOS, 166°56.51’E, 532-599 m, 9 Oct 1994, coll. B. Richer de Forges, 2 2 ovig. 16.2, 17.1 mm (MNHN). Wallis I. MUSORSTOM 7, Stn DW 523, 13°12’S, 176°16'W, 455-515 m, 13 May 1992, | juv. (MNHN). New Caledonia. HALIPRO 1, Stn CP 867, 21°26’'S, 166°18’E, 720-950 m, 22 Mar 1994, coll. B. Richer de Forges, 3 10.4 mm, 2 15.6 mm (MNHN). BATHUS 4, Stn DW 911, 18°57.80'S, 163°08.47'E, 566— 558 m, 5 Aug 1994, coll. B. Richer de Forges, 2 12.5 mm (MNHN). Stn DW 915, IS" SIL2A0S; 63167278, SSO—7/5 im, 5 Aug 1994, coll. B. Richer de Forges, 3 6 9.9-10.4 mm (MNHN). Stn DW 920, i AS.318) SO. IGS 7 M618, OlO—G20 ian, © Aug 1994, coll. B. Richer de Forges, 3 8.5 mm (MNHN). Indonesia: Tanimbar I., KARUBAR, Stn CP 39, FAVS, (S220 78, AVAGO tim, 2 Oret MDM, G M37 ioren, 2 W4eal ania (VMINEIN) S Stum CR 59. 8:20 S.132 ME: 405-399 m, 31 Oct 1991, 2 3 12.4, 12.5 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON mm, 2 2 ovig. 12.0, 14.6 mm (MNHN). Stn CP 70, 8°41’S, 131°47'E, 413-410 m, 2 Nov 1991, 6 10.6 mm (MNHN). Description.—Dorsal surface of carapace covered with pearliform granules, inter- spaced with smaller granules. Frontal lobes rounded, closely granulate. Subhepatic mar- gins of carapace somewhat swollen, median subhepatic tubercle followed, in young specimens, by smaller tubercle. Hepatic re- gion bearing granulate tubercle. Anterolat- eral margin posteriorly set with 3 granulate tubercles, posteriormost tubercle largest. Posterolateral margin bituberculate, poste- rior tubercle larger. Posterior margin bear- ing 2 conicalal tubercles laterally. Branchi- al, intestinal regions demarcated by deep grooves. Intestinal region swollen, bearing granulate tubercle anteriorly, long, up- curved spur posteriorly (Fig. 1C). Anterior margin of efferent branchial channel deeply sutured. Third maxilliped granulose. Cheliped and pereiopods 1—5 closely granulate throughout. Cheliped merus in adult male 0.75 as long as carapace, in fe- male 0.66 carapace length; fingers as long as upper margin of palm. Thoracical sternites in male boldly gran- ulate. Telson slender, third as long as fused abdominal segments. Male first pleopod with lamellate process preapically (Figs. 38), 16), Etymology.—From the Latin nanus, small, minute. Distribution.—Indo-Pacific Ocean: Wal- lis I., Vanuatu, New Caledonia, Indonesia; 281-1250 m. Remarks.—Tanaoa nanus differs from T. pustulosus in its much smaller size, its coarsely granulated carapace, and the la- mellate preapical process of the first male pleopod. Tanaoa pustulosus (Wood-Mason, in Wood-Mason & Alcock, 1891), new combination Figs. 1D, 3G, H Randallia pustulosa Wood-Mason, in Wood-Mason & Alcock, 1891:266; 1892: VOLUME 116, NUMBER 2 pl. 5, fig. 4.—Alcock, 1896:196; 1899: 27.—Doflein, 1904:42, pl. 14, fig. 1-6 (part).—Seréne, 1954:491.—Sakai, 1976: 99, pl. 30, fig. 2—Yaldwyn & Dawson, 1976:95, figs 2—5.—Serene & Vadon, 1981:119, 124.—Richer de Forges, 1983: 634 (tab.).—McLay, 1988:100, fig. 20.— Chen, 1989:217, fig. 15, pl. 4, fig. 1.— Tan, 1996:1054.—Ikeda, 1998:82, pl. 19, fig. la-d.—wNg et al., 2001:10.—Chen & Sune 0022340" fie tale ply sae Randallia vitjazi Zarenkov, 1994:104, pl. 5, pl. 8b. Not Randallia pustulosa: thle, 1918:246 [= T. nanus, new species]. Material examined.—Fiji. MUSOR- STOMP TOF Blich™ Water “Sm EE 13315 17°02.4'S, 178°01.8'E, 694-703 m, 8 Aug 1998, 7 3 21.5—-30.1 mm, | juv. (MNHN). Sth CC 1332, 16°56.2’S, 178°07.9'E, 640— Gs ith co Awe IOs. ZO 2eiS)5 Asoo) main, ©) juvs. (MNHN). Stn CC 1337, 17°03.4'S, WTAE 655-070) my 9 Aus, 1998.76 27.8 mm, | juv. (MNHN). Stn CP 1342, 16°46.0'S, 177°39.7'E, 650-701 m, 10 Aug 1998, 2 35 28.7, 21.4 mm (MNHN). Stn CP LBA6, WP VNOG Ss WAS S248. O7S=G8s) ir, eae 998 6 230)5) mm, 229.9 mm (MNHN). New Caledonia. CHALCAL 2, Stn DW 75, 24°39.31'S, 168°39.67'E, 600 m, 29 Oct 1986, 3d 19.6 mm, 2 30.5 mm (MNHN BZ1210)., HALIPRO 1. Stn CP 867, 2126'S, 166°18'E, 720—850 m, 22 Mar 1994, 1 juv. 15.8 mm (MNHN). BATHUS 4. Sittin Cle Oil ISSO SS. MSS A47/ 18, 566-558 m, 5 Aug 1994, 1 juv. (MNHN). Caroline Is. Palau I., Mutremdiv Pt., Jun 1981, coll, W.B. Saunders, ¢ 37.7 mm (USNM 354775). Japan. Shikoku I., Mimase, Tosa Bay, 250 m, Apr 1968: ¢ 33.7 mm (SMF 15104, ex. coll. T. Sakai). Tosa Bay, 4 2 33.9-—34.3 mm (SMF 22555, ex. coll. T. Sakai). Tosa Bay, 34.2 mm (SME ex. coll. T. Sakai); Tosa Bay, ¢ 34.0 mm, 2 35.0 mm (NHM IIOleOD 38/59. exenco) led Salsa) 405 Taiwan. Tashi fishing port, 22 Mar 1986, coll. T.Y. Chan, 2 ovig. 34.8 mm (NTOU). Indonesia. Kai Is., KARUBAR, Stn DW 3S 48) SB, USPSA bs. VON WS inn, BA (Over 1991, 5 39.6 mm (MNHN). Philippines. Mindanao, Iligan Bay, R/V Albatross Stn 5508, 8°17.24’N, 124°11.42’E, 494 m, 5 Aug 1909, ¢ 34.4 mm (USNM). Btw Negros, Siquior, R/V Albatross Stn S538, SUB MSING WEI. AS son, iO Aug 1909, 5 35.8 mm (USNM). MUSOR- STOM, Stn 43, 13°50.5'N, 120°28.0’E, 484— 448 m, 24 Mar 1976, 2 3 23.8, 23.0 mm, 2 24.4 mm, | juv. (MNHN B18055). Stn 44, 13°46.9'N, 120°29.5'E, 610—592 m, 24 Mar 1976, 3 broken, 2 15.1 mm (MNHN B18057). MUSORSTOM 3, Stn CP97, 14°00'N, 120°18’E, 189-194 m, 1 Jun 1985, 1 juv. (MNHN B17999). Stn CP122, 12-20'N; 121 -42’'E, 673-675 ms 4 Juni 1985, 6 18.1 mm (MNHN B18000). Stn CP128, 11°S0’N, 121°42’E, 815-821 m, 5 Jun 1985, 2 Owe. BS.3, 2 joMeAstinec! 2225) tania (MNHN B18001). Waccadive Seay 8isi Ni oro BOE? parasitized 26.2 mm (NHM 1899.8.26.4, ex. Indian Museum). 9°34.57'N, 75°36.30’'E, @ 12.6 mm, (NHM_ 1896.9.8.10, ex. Indian Museum). Seychelles. 4°34.2’S, 56°26.6’E, 650— 630 m, 22 Oct 1987, 3 31.3 mm (MNHN RI OIOO), Ces SI IN, Ts SO. 30 Be PVA mm (NHM 1896.9.8.10, ex. Indian Muse- um). Réunion. 350—S500 m, 2 Feb 1974, coll. P Guézé, 2 31.8 mm (MNHN B19135). RUN] iViornion Dujresne, sine {eee i22: 20°57,9'S, 55°14.5’E, 450-580 m, 1 Sept 1982, 2 31.7 mm (MNHN B 19134). Geyser INGE, Sim We4e | we wo as. 46°28.2'E, 300-600 m, 11 Apr 1977, coll. M. Faubert, 2 24.1 mm (MNHN B19044). Comoro Is., Stn 61, 12°46.0’S, 44°58’E, 475-510 m, 29 Mar 1977, coll. M. Faubert, 1 juv. (MNHN B19045). Madagascar. 12°43.5'S, 48°14.5’'E, 370 m, 14 Apr 1971, coll. A. Crosnier, ¢ ovig. 30.9 mm (MNHN B18583). Stn CH 24, DDO SVSse4 3.07.2 9430-460 imneeiS Jan 406 1986, 2 ovig. 34.7 mm (MNHN B18585). Stn CHD 7 22,210 S34 305-545 Oman iS Jan 1986, coll. R. von Cosel, 2 32.6 mm Gay (MINISIN JBIGSS82), Stina Clal 32, 22°25.8'S, 43°04.3'E, 450-475 m, 19 Jan ISO, Coll, Ik. Cleya, 2 Ovi, 34.7 save (MNHN B19726). Stn CH 37, 22°18.2’S, 43°04.8'E, 450—475 m, 21 Jan 1986, coll. R. von Cosel, 2 ovig. 31.6 mm (MNHN BIBS). Sia Ciel BS, 223). 1S, 2:3) 5.5 Je; 400—500 m, 21.01.1986, coll. R. von Cosel, 2 ovig. 33.5 mm (MNHN B18581). Stn Clal 58, ZS 30.255 43 SOs Je, SO aol, 27 Feb 1973, coll. R. von Cosel, 2 ovig. 33.2 mm (MNHN B18588). Stn CH 59, 23°36.0'S, 43°29.6'E, 600-610 m, 27 Feb 1973, gd 30.3 mm, (MNHN B18586). Stn CHEGO0% 2225-6) S143.06)27E5 (47/59 mi, 18.10.1986, coll. R. von Cosel, @ ovig. 33.3 mm (MNHN B19038). Stn CH 6l, 23°36.1'S, 43°31.0'E, 445-455 m, 27 Feb 1973, 6 31.3 mm, (MNHN B19736). Same daa, 2 32.5 mom, 2 overs Sil. mon (MNHN B18584). Stn CH 81, 22°22.8'S, 43°03.3/E, 525, 25 Oct 1986, coll. R. von Cosel, 2 31.9 mm (MNHN B19039). Stn CH 122, 22°16.8'S, 43°02.7’E, 600 m, 30 Nov 1986, 3 33.0 mm, 2 22.7 mm (MNHN B19041). Stn CH 127, 22°S, 43°E, 610 m, 1 Dec 1986, coll. R. von Cosel, 2 22.3 mm (MNHN B19040). Redescription.—Dorsal surface of cara- pace unevenly granulate, obtuse granulate tubercles laterally on branchial region. Frontal lobes triangular, closely granulate. Third maxilliped minutely granulose. Sub- hepatic margins of carapace inflated, me- dian subhepatic tubercle followed, in young specimens, by smaller tubercle. Anterolat- eral margin posteriorly set with 3 granulate tubercles, posteriormost largest. Posterolat- eral margin bituberculate, posterior tubercle larger. Posterior margin bearing 2 dorsoven- trally flattened denticles laterally. Branchi- al, intestinal regions demarcated by deep grooves; intestinal region swollen, bearing prominent tubercle anteriorly, long, up- curved spur posteriorly (Fig. 1D). Cheliped, pereiopods granulate through- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON out. Cheliped merus in adult male nearly as long as carapace, in female 0.85 carapace length; fingers as long as upper margin of palm. Male thoracic sternites minutely granu- late; margin of abdominal sulcus raised, granulate. Telson slender, third as long as fused abdominal segments. Male first ple- opod with digitate process preapically (Figs. 3G, H). Color.—“*‘(after three weeks in alcohol) ... dorsal surface of the carapace ... pink- ish-orange, with the tubercles ... red; ven- tral surfaces ... pale pinkish-white”’ (Yald- wyn & Dawson 1976:95). Color photo: Ike- da, 1998:82, pl. 19, fig. la—d. Distribution.—Indo-Pacific Ocean: Fiji, New Caledonia, New Zealand, Caroline Is., Japan, Taiwan, Indonesia, Philippines, An- daman Sea, Laccadive Sea, Seychelles, Agalega Is., Comoro Is., Geyser Reef, Mad- agascar, Réunion, Mozambique channel, East Africa; 85—977 m. Remarks.—tThe description and drawings of Randallia vitjazi (Zarenkov 1994:104, pl. 5, pl. 8b) are clearly that of 7. pustu- losus, including the filiform preapical pro- cess of the first male pleopod. Sakai (1976: 99, pl. 14, fig. 6) believed that Doflein’s specimen (R. pustulosa) from the Nicobars was “‘a different species, which seems to be related to R. pustuloides Sakai’, whereas Chen (1989:217) declared it simply ““Non Randallia pustulosa’’ [Wood-Mason]; the specimen was not available to me at the time of writing. Key to Species of Tanaoa, new genus 1. Anterolateral margins of carapace bear- ing low granulate tubercles; intestinal re- gion bearing low tubercle posteriorly .. T. distinctus, new combination — Anterolateral margins of carapace bear- ing prominent granulate tubercles; intes- tinal region bearing prominent tubercle anteriorly, upcurved spur posteriorly .. 2 2. Carapace length of adult >30 mm; preapical process of male first pleopod filiform ... 7. pustulosus, new combination VOLUME 116, NUMBER 2 — Carapace length of adult >12 mm; preapical process of male first pleopod lane lates eee tee ee T. nanus, new species Tokoyo, new genus Diagnosis.—Carapace circular, globose, regions indistinct. Front narrow, bilobed. Eyes small, retractible. Outer orbital margin trisutured, V-shaped gap proximally on ventral margin. Antennules obliquely fold- ed, basal antennular segment squat, oper- culiform, sealing lower antennular aperture. Antennae small, slender, basal antennal seg- ment inserted in orbital hiatus. Postorbital region concave. Lateral margins rounded, bearing median tubercle. Posterior margin of carapace narrow, tridentate. Third maxilliped exopod sealing efferent branchial channel, slightly shorter than en- dopod; merus of endopod subtriangular, shorter than subrectangular ischium; endo- pod of adult female with vertical line of se- tae medially. Anterior margin of efferent branchial channel produced, deeply cleft, separated by narrow groove from lower or- bital margin. Chelipeds long, slender, subequal. Che- liped merus and palm subcylindrical; fin- gers shorter than upper margin of palm, in- ner margins denticulate. Pereiopods slender, short; dactyls nearly as long as propodi; upper surface of dactyls distally setose, tips corneous. Male abdominal sulcus deep, nearly reaching buccal cavity; lateral walls of ab- dominal sulcus with elongate cavities an- teriorly. Male abdomen narrow, twice as long as wide at base. Abdominal segments 3—6 fused, basio-lateral regions of fused segments slightly inflated, bearing denticle preapically; lateral margin carinate, carina fitting into groove between thoracic seg- ments 4 and 5. Telson triangular, fifth as long as fused segments. Abdominal seg- ments 4—6 of female fused, swollen, shield- like; telson lingulate, posterior margin arched. Margin of abdominal fossa in fe- male prominent, anteriorly thickened. 407 Male first pleopod elongate, slightly sin- uous, attenuate, distally sharply bent inte- riorly; second pleopod short, distally scoop- like. Type species.—Randallia eburnea A\- cock, 1896. Species.—Tokoyo cirrata, new species, Tokoyo eburnea (Alcock, 1896). Etymology.—Tokoyo, in Japanese my- thology, was a girl who slew a sea-serpent that intimidated the fisherfolks. The name Tokoyo is to be considered as an arbitrary combination of letters, and heretofore takes the gender feminine. Remarks.—Tokoyo, new genus, differs from Randallia s. s. in having the anten- nular operculum sealing only the bottom half of the antennular aperture, a bilobate anterior margin of efferent branchial chan- nel, and fused segments 3—6 of the male abdomen. TJokoyo differs from the other three new genera described herein in its tri- dentate posterior margin of the carapace, the preapically positioned denticle on the fused segment of male abdomen, and the lingulate telson in the female. Tokoyo cirrata, new species Figs. 1E, 4A—C Type material.—Holotype: Vanuatu. MU- SORSTOM 8, Stn CP 1086, 15°36.58’S, l6P-A16332708, Ns2=215) ime SyOct 19945 coll: B. Richer de Forges, | 5d 13.0 mm (MNHN B.28511). Paratypes: Same data, 12 3 9.7— ISO tm, © Vo Wiw7ealwA 2 inn, 4! 2 own: 11.7-13.2 mm, 7 juv. (MNHN B.28512). Material examined.—Vanuatu. MUSOR- SMOME Ss) Stn Ck 976) W9l25.22)S- 169°26.73'E, 160-182 m, 22 Sep 1994, coll. B. Richer de Forges, 6 broken, @ ovig. 12.3 mm, 1 juv. (MNHN). Stn CP lOVOS T5e8659"S= N67 11642 EF 184190 m, 4 Oct 1994, coll. B. Richer de Forges, 6 Ol imi, 2 OWS, NS) iii th UNE (MNHN). Australia. Queensland, Moreton Bay, 36 m, Sep 1966, 2 ¢ 23.4, 24.3 mm (AMS P15383). East of Swains Reef, 22°26.75'S, 408 ios USI 13, W32 in, o Sep IGS, 6 23Y mm (AMS P56719). Description.—Dorsal surface of carapace minutely and evenly granulate. Frontal lobes squat, minutely granulate. Subhepatic margin of carapace somewhat swollen, with row of granules, separated from anterolat- eral margin by shallow concavity. Lateral margin bearing medially small tubercle. In- testinal region weakly swollen, demarcated laterally by indistinct grooves. Posterior margin bearing 3 rounded, dorso-ventrally flattened denticles, median denticle smallest Cag, Ie). Third maxillipeds bearing conical gran- ules, setae anteriorly, low granulation pos- teriorly. Thoracic sternites indistinctly gran- ulate; anterior sternite with diagonal gran- ulate ridge laterally. Cheliped with well-spaced minute gran- ules. Cheliped merus in adult male 1.60— 1.80 as long as carapace; palm thicker dis- tally; dactyl two-thirds as long as upper margin of palm, gap proximally between dactyl, pollex. Cheliped merus in female one-third longer than carapace, palm cylin- drical. Pereiopods punctate. Fused abdominal segmenis of male bear- ing flattened triangular denticle. Telson one- fifth as long as fused abdominal segments. Tip of first male pleopod vermiculate, coiled, curled anteriorly (Figs. 4A—C). Etymology.—From the Latin, cirratus, curly, and refers to the shape of first pleo- pod. Color.—Dorsal surface of carapace or- ange, margins paler; posterior denticles white. Chelipeds pale orange, distal mar- gins of merus, carpus, propodus stained with darker orange. Distribution.—Southwestern Pacific: Va- nuatu and Australia; 36—215 m. Remarks.—Tokoyo cirrata, new species, differs from 7. eburnea Alcock, 1896, in having an anteriorly coiled tip of the first male pleopod, and color pattern of the car- apace. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Tokoyo eburnea (Alcock, 1896), new combination Figs. 1K 4D—-F Randallia eburnea Alcock, 1896:197.—A\I- cock & Anderson, 1897: pl. 30, fig. 4.— Ihle, 1918:246.—Sakai, 1934:289, pl. 18, fie. 4: 1935-545 pl. 9) fics 3 NSS ialew? fig. 22: 1965:42, pl. 17, fig. 15 197698 pl. 29, fig. 1—Uchida, 1949:720, fig. 2082.—Seréne, 1954:491, 1968:45.— Utinomi, 1956:72, pl. 36, fig. 8.—Tyn- dale-Biscoe & George, 1962:87, fig. 7.7.—Chang, 1963:7, fig. 1.—Zarenkov, 1969:24, fig. 7.3.—Takeda & Miyake, 1970:225.—Campbell, 1971:41.—Take- das, 19732325 tie. Se, fe 1975 allo ore 238; 2001:230.—Seréne & Soh, 1976:12, pl. 3, fig. c —Yaldwyn & Dawson, 1976: 96.—Serene & Vadon, 1981:118, 124.— Richer de Forges, 1983:634.—Chen, 1989:212, figs 12, 13.—Huang, 1989: 309, 1994:579.—Ng et al., 2001:10.— Chen & Sun, 2002:338, fig. 149.—Davie, D002 Ai Randallia japonica Yokoya, 1933:130, text- fig. 46. Material examined.—Japan: Shikoku L., Tosa Bay, Nov 1958, colls T. & K. Sakai, 2 2 ovig. 18.6, 17.3 mm (USNM 120708). Tosa Bay, 110 m, 10 May 1990, ¢ 14.8 mm (SMF 22577). Tosa Bay, 2 19.1 mm (SME ex. coll. Sakai). Off Ashizuri-Misaki, 366 m, 24 Nov 1958, & 17.3 mm (SME ex. coll. Sakai). Honshu I., SW Seno Umi, R/V Albatross Stn 3703, 57 m, 7 May 1900, 1 juv. (USNM 134214). China. Off Dougliai, 28 Jun 1976, d 12.3 mm, 2 13.2 mm (SMF13206). Taiwan. Tashi, 24 Jan 1997, coll. TY. Chan, 2 @ ovig. 20.1, 19.1 mm (NTOU). Tashi, Dec 1997, coll. TY. Chan, 2 6 17.8, 17.3 mm (NTOU). — Indonesia. Off Borneo, 5°57’N, 109°34’E, 150 m, 1963, 2 ovig. 24.4 mm (NHM 1964. 9.9.3). Btw Wowoni, Buton Is., 4°20’S, 122°58'E, 75-94 m, ‘Siboga’ Stn 204, 20 Sepels995 7 2937. sans Seu sen CAIN IE DAV BO VOLUME 116, NUMBER 2 a A B C D 409 F G H Fig. 4. A, B, C, Tokoyo cirrata, new species, holotype. dg cl 13.0 mm, Vanuatu, MUSORSTOM 8, Stn CP 1086 (MNHN): A, first pleopod; B, apex of first pleopod; C, apex of first pleopod, dorsal view. D, E, K Tokoyo eburnea (Alcock, 1896), d cl 14.8 mm, Japan, Tosa Bay (SMF 22577): D, first pleopod; E, apex of first pleopod:; E apex of first pleopod, dorsal view. G, H, Toru granuloides (Sakai, 1961), 5 cl 25.8 mm, Loyalty Is., MU- SORSTOM 6, Stn DW 487 (MNHN): G, first pleopod; H, apex of first pleopod. Scales = 1 mm. Philippines. MUSORSTOM 1, Stn 16, iS 92NS 120012:3:E. 164=150) m, 20 Mar 1976, det. H: Chen, 6 27.4 mm, 2 26.9 mm (USNM 237656 ex. Paris Museum). Stay GP34) 14701 NV 120715:3 Es 1912888 m, 23 Mar 1976, 2 32.1 mm (MNHN B ESO) este P 5313-58 _00NI20Ms97E: 143-178 m, 26 Mar 1976, 9 3 23.4—35.0 mm, 2 2 19.4, 28.3 mm, 9 ¢ ovig. 26.1— 29.0 mm, 4 juvs. (MNHN B 18058). MU- SORSHOMeS = sStna CRs Ss.) li 4Oile N, 120°17’E, 183-187 m, 31 May 1985, 8 3 I2DO—wS.5) wn, 2-2 MWO.S5 vO tir (MNHN B 17989). Stn CP 96, 14°00'N, 120°18’E, 190-194 m, 1 Jun 1985, 10 ¢ 11.1—23.8 mm, 5 @ 14.5—21.4 mm (MNHN B 17988). Andaman Sea. Thailand. Similan I., R/V Te Vega, 08°46'N, 97°46/E, 75—81 m, 4 Novel9OG3t4 6 HM ASI5!5 mm see) 6.4— 12.9 mm (USNM 273786). Laccadive Sea. 11°05.45'N, 75°04.08’E: 6 9.6 mm (NHM 1896.9.8.20, ex. Indian Museum). Redescription.—Dorsal surface of cara- pace minutely and evenly granulate. Frontal lobes squat, minutely granulate. Subhepatic margin of carapace somewhat swollen, with row of granules, separated from anterolat- eral margin by shallow concavity. Lateral margin bearing medially small tubercle. In- testinal region slightly swollen, demarcated laterally by indistinct grooves. Posterior margin bearing 3 rounded, dorsoventrally flattened denticles, median denticle smallest ei. JUS). Third maxillipeds bearing conical gran- ules, setae anteriorly, low granulation pos- teriorly. Thoracic sternites indistinctly gran- ulate; anterior sternite with diagonal gran- ulate ridge laterally. Cheliped with well-spaced minute gran- ules. Cheliped merus in adult male almost twice as long as carapace; palm thicker dis- tally; dactyl half as long as upper margin of palm, gap proximally between dactyl, pollex. Cheliped merus in female one-third longer than carapace, palm cylindrical, dac- tyl two thirds as long as upper margin of palm. Pereiopods punctate. Fused abdominal segments of male bear- 410 ing flattened triangular denticle. Telson one- fifth as long as fused abdominal segments. First male pleopod distally deflexed in- teriorly (Figs. 4D—F). Color.—‘**Yellowish-pink, with deeper pink on anterior margins of merus and hand of chelipeds; anterior portion of carapace and distal joint of merus of walking legs yellow. Fingers white’? (Tyndale-Biscoe & George 1962:87). “[B]right brick red above and white below” (Chang 1963:7). Distribution.—Indo-Pacific Ocean: Aus- tralia, Japan, China, Taiwan, Vietnam, In- donesia, Philippines, Andaman Sea, Lac- cadive Sea; 35—366 m. Remarks.—A\l|cock (1896:198) errone- ously described the male abdominal seg- ments 3—5 as fused, when in fact segments 3—6 are fused, as remarked by Ihle (1918: 246). Chen (1989, fig. 13b) erroneously de- picted the sixth male abdominal segment as articulate. Examination of immature speci- mens showed that the pleopod drawn by Zarenkov (1969, fig. 7.3) is of a young male. The specimens from Borneo and the Philippines are much larger than the others examined, but no morphological differences were detected. Key to Species of Tokoyo, new genus 1. First male pleopod with anteriorly coiled apical process . IT. cirrata, new species — First male pleopod with interiorly de- Hlexcdeaplcaleprocessmearic ir ee T. eburnea, new combination 2 © © © © © © © © © «© Toru, new genus Diagnosis.—Carapace subcircular, glo- bose. Front narrow, uptilted, bilobed. Eyes small, retractible. Outer orbital margin tri- sutured, V-shaped gap proximally on ven- tral margin. Antennules obliquely folded, basal antennular segment squat, operculi- form, sealing lower antennular aperture. Antennae small, slender, basal antennal seg- ment inserted in orbital hiatus. Postorbital region concave. Intestinal region swollen, demarcated by grooves, 2 pairs of pits PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON along cardiobranchial grooves. Posterior margin of carapace narrow, bilobate. Third maxilliped exopod slightly shorter than endopod, not quite sealing efferent branchial channel; endopod of adult female medially with vertical line of setae; merus of endopod subtriangular, shorter than sub- rectangular ischium. Anterior margin of ef- ferent branchial channel produced, bilobed, separated by narrow groove from lower or- bital margin. Chelipeds long, slender, equal. Cheliped merus, palm subcylindrical; fingers as long aS upper margin of palm, inner margins denticulate. Pereiopods slender, short; dac- tyls shorter than propodi; upper surface of pereiopodal dactyls setose, tips corneous. Fourth thoracic sternite not swollen lat- erally. Abdominal sulcus of male deep, elongate, nearly reaching buccal cavity, an- terior margin raised; lateral walls of abdom- inal sulcus with elongate cavities anteriorly. Abdominal segments 3—6 of male fused, basio-lateral regions inflated, fused segment narrowing distally, bearing denticle at distal margin; lateral margin carinate, carina fit- ting into groove at suture between sternites 4 and 5. Telson lingulate, two-fifths as long as fused segment, not reaching tip of ab- dominal sulcus. Abdominal segments 4—6 of female fused, shield-like; margin of ab- dominal cavity prominent, rampart-like. Telson ogival, basal margin sinuous. Male first pleopod elongate, attenuate, bearing long apical process; second pleopod short, distally scoop-like. Type species.—Randallia granuloides Sakai, 1961. Species.—Toru granuloides (Sakai, 1961), 7. mesjatzevi (Zarenkov, 1990), Toru pilus (Yan, 1996), Toru septimus, new species. Etymology.—Totru is the Polynesian god of the chasms of the deep. The name Toru is to be considered as an arbitrary combi- nation of letters and heretofore takes the gender masculine. Remarks.—Toru, new genus, is differ- entiated from Tanaoa, new genus, in having VOLUME 116, NUMBER 2 a lingulate telson in the male, the lateral walls of male abdominal sulcus are exca- vate anteriorly, the male first pleopod bear- ing a long apical process, the margins of the female abdominal cavity prominent, rampart-like, and the fourth thoracic ster- nite even; whereas in TJanaoa the male tel- son tapers narrowly, the lateral walls of the male abdominal sulcus are entire, the preap- ical margin of the male first pleopod bear a minute process perpendicular with tip, the margins of the female abdominal cavity do not form a rampart-like edge, and the fourth thoracic sternite is greatly swollen laterally. Toru granuloides (Sakai, 1961), new combination Figs. 2A, 4G, H Randallia granuloides Sakai, 1961:136, pl. 3, fig. 3; 1976:96, text-fig. 52.—Sereéne, 1968:45.—Yaldwyn & Dawson, 1976: 96.—Richer de Forges, 1983:634. Material examined.—Wallis 1. MUSOR- STOM 7, Stn DW 583, 13°11’S, 176°14'’W, 330-365 m, 22 May 1992, 9 18.5 mm (MNHN). Stn DW 584, 13°11’S, 176°14'W, 360—400 m, 22 May 1992, | juv. (MNHN). Stn DW 605, 13°21'S, 176°08'W, 335-340 m, 26 May 1992, 5 16.6 mm (MNHN). Fiji. MUSORSTOM 10, Stn CP 1386, 18°18.5'S, 178°05.1’E, 230-344 m, 19 Aug 1998, 2 ovig. 20.1 mm (MNHN). BOR- IDA IL, Sita CIP Nav WIS, IST Ae 400—401 m, 2 Mar 1999, 3 17.8 mm (MNHN). Loyalty Is. MUSORSTOM 6, Stn DW 416, 20°42.15’S, 166°59.60’E, 343 m, 16 Feb 1989, 6 27.3 mm (MNHN). Stn DW AS. DY 2354S, WOO W271 18, 420 tan, i7 Feb 1989, 5 18.9 mm, 2 9.5 mm (MNHN). Stn DW 456, 21°00.71'S, 167°26.35'E, 240 m, 20 Feb 1989, ¢ 23.0 mm (MNHN). Stn DW 487, 21°23.30S, 167°46.40’E, 500 m, 23 Feb 1989, 3d 25.8 mm, 2 13.8 mm QVINEIN)D BATHUS = 22 Stn CRY 737, 23°03.42'S, 166°59.97'E, 350—400 m, 13 May 1993, 3d 23.2 mm (MNHN). New Caledonia. MUSORSTOM 4, Stn 411 OAS 522815, oS Zik7 Es 545 mm, 198Sep 1985, 6 25.8 mm (MNHN B18411). Stn W330, 22 NW Sy WOW NS) 18, A =s 50) im, 2 Orsi 1985, 2 juv. (MNHN B21245). Jaoam, © 19.3 mum, 2 owe, ZS wen (SME ex. coll. Sakai). 2 d 18.8, 19.0 mm (SME ex. coll. Sakai). Redescription.—Dorsal surface of cara- pace granulate, granules closer set posteri- orly. Frontal lobes rounded, minutely and closely granulate. Subhepatic margins of carapace very slightly swollen, Lateral mar- gin medially set with small tubercle. Pos- terior margin bearing 2 lamellate, granulate, triangular tubercles laterally. Intestinal re- gion swollen, topped by low tubercle, de- marcated by deep grooves (Fig. 2A). Third maxilliped granulose. Cheliped and pereiopods closely granu- late throughout. Cheliped merus in adult male as long as carapace; fingers as long as upper margin of palm. Pereiopodal dactyls anteriorly tomentose. Thoracic sternites and abdomen minutely granulate. Preapical denticle on fused ab- dominal segment in male triangular. Shaft of first male pleopod slightly curved, with sickle-shaped apical process (Figs. 4G, H). Distribution.—Western Pacific: Fiji, Wal- lis I., Vanuatu, Loyalty Is., New Caledonia, Japan; S5O—550 m. Remarks.—Sakai (1961:136) believed ‘““The nearest relative of this new species [Randallia granuloides| is R. granulata Miers (1886). Yaldwyn & Dawson (1976: 96), as well as Richer de Forges (1983: 634), relegated R. granuloides and R. gran- ulata to the same “‘species-group’’. How- ever, despite superficial similarity owing to the granulate carapace and elongate cheli- peds they belong in different genera: Toru granuloides, new combination, differs from R. granulata in the form of the male telson, ogival rather than rounded as in R. granu- lata; and in the form of the first male ple- opod, bent distad in R. granulata. 412 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TS | | \ \ Ss 4 ( A B C D E F G H Fig. 5. A, B, Toru pilus (Tan, 1996), 3 cl 11.5 mm, New Caledonia, BATHUS 4, Stn CP 946, (MNHN): A, first pleopod; B, apex of first pleopod. C, D, Toru septimus, new species, holotype. d cl 13.1 mm, Loyalty Is., Stn DW 421 (MNHN): C, first pleopod; D, apex of first pleopod. E, EK Toru trituberculatus (Sakai, 1961), ¢ cl 8.6 mm, Indonesia, Tanimbar I., KARUBAR, Stn CP 67 (MNHN): E, first pleopod; FE apex of first pleopod. G, H, Urashima pustuloides (Sakai, 1961): paratype, d cl 34.3 mm, Japan, Shikoku I., Mimase, Tosa Bay (NHM 1961.6.5.46—48): G, first pleopod; H, apex of first pleopod. Scales = 1 mm. Toru mesjatzevi (Zarenkov, 1990), new combination Randallia mesjatzevi Zarenkov, 1990:67, pl. 7. Redescription.—Dorsal surface of cara- pace granulate, granules conical anteriorly. Frontal lobes rounded, minutely and closely granulate. Subhepatic margins of carapace very slightly swollen, Lateral margin uni- formly rounded. Posterior margin bearing 2 lamellate, granulate, triangular tubercles lat- erally. Intestinal region somewhat swollen, demarcated by shallow grooves. Third max- illiped granulose, endopod medially with vertical ridge of conical granules. Cheliped and pereiopods closely granu- late throughout. Cheliped merus in adult male as long as carapace; fingers as long as upper margin of palm. Anterior margin of pereiopodal meri, carpi and propodi prom- inently spinose. Pereiopodal dactyls anteri- orly tomentose. Thoracic sternites and abdomen minutely granulate. Preapical denticle on fused ab- dominal segment in male triangular. Shaft of first male pleopod slightly sin- uous, with sickle-shaped apical process. Distribution.—Off Kenya; 130-150 m. Remarks.—Zarenkov’s (1990, pl. 7, fig. 13) illustration of the first male pleopod of Toru mesjatzevi, with its sickle-shaped api- cal process, clearly places the species with- in the genus Toru. The species is larger than its cogeners (CL 36.0, 33.0 mm), and is no- table for its prominently spinose pereio- pods. The specimens were not available to me at the time of writing. Toru pilus (Tan, 1996), new combination Figs. 2B, 5A, B Randallia pila Tan, 1996:1051, fig 7I—n, 8a, Cen Galle Type Material.—Holotype: Philippines. R/V > Albatross® Stn 54545 swans 123°50.30'E, 300 m, 7 Jun 1909: 3 13.0 mm (USNM). Paratype: Luzon, San Ber- nardino Straits, R/V Albatross Stn 5453, 13°12'N, 123°48.18'E, 271 m, 7 Jun 1909, 2 ovig. 10.0 mm (USNM). Material examined.—Fiji. MUSOR- STOM> 10, "Stn® CR) B25 a idliowass 177°49.8'E, 282-322 m, 7 Aug 1998, 12 3 9.5-11.4 mm, 2 10.4 mm, 6 2 ovig. 9.1— 11.0 mm (MNHN). Stn CP 1327, VOLUME 116, NUMBER 2 IPB SS, WW lyole, STO srs) sas, Teor ISOS, © 6 WeSlOe tain, 2 OO inven, 2b 2 ovig. 9.5-10.6 mm, | juv. (MNHN). Stn CP IBA Seel7S0: 31S. 17/8 39:648.)353—390 mi; 11 Aug 1998, ¢ 10.5 mm, 2 ovig. 10.9 mm CVENIEIND 2 @ Sin CRIB 49%5 17 -SIAlw’S, 178°38.8'E, 244-252 m, 11 Aug 1998, @ OLOR nama “OVINEIN) 2 Stier 1390; 18°18.6'S, 178°05.1'E, 234-361 m, 19 Aug 1998, 2 11.9 mm (MNHN). BORDAU 1, Stn CP 1406, 16°39’S, 179°37'E, 360-380 m, 25 Feb 1999, 2 3 10.8, 10.0 mm (MNHN). Stn CP 1448, 16°45’S, 179°59’E, 410-500 m, 4 Mar 1999, 3S 10.0 mm (MNHN). Vanuatu. MUSORSTOM 8, Stn CP 964, 20°19.60'S, 169°49.00'E, 360—408 m, 21 Sep 1994, coll. B. Richer de Forges, 2 ovig. 13.0 mm, 2 juv. (MNHN). Stn CP 1047, 16°53.62’S, 168°10.49'E, 486—494 m, 30 Sep 1994, coll. B. Richer de Forges, Son OM tO! Seman, 210.4» anal uy. (MNHIN). Stn CP 1088, 15°09.23’S, 167°15.13'E, 425-455 m, 6 Oct 1994, coll. 5s ikimasr Gs lkorues, o Iie men 2) 8) mm (MNHN). Stn CP 1106, 15°05.27'S, 167°11.88'E, 305-314 m, 7 Oct 1994, coll. B. Richer de Forges, d 9.3 mm (parasit- ied) eoMovie., 13-25 12.7 mina 2 broken (MENTIN 2 oSta’ CP 1137, 15°41.527S, 167°02.67'E, 360-371 m, 11 Oct 1994, Coles Richer deszorces,. 6) IOhmanm, 2 broken (MNHN). New Caledonia. BATHUS 1, Stn CP 695, 20°34.59'S, 164°57.88'E, 410-430 m, 17 Mar 1993, 5 8.6 mm (MNHN). BATHUS A SinpCP) 946) 20°33.8'S, Woles8.3542; 386-430 m, 10 Aug 1994, 3 11.5 mm (MNHN). Redescription.—Dorsal surface of cara- pace boldly granulate. Front prominently produced, frontal lobes anteriorly rounded, minutely granulate, pilose. Postorbital re- gion depressed, concave. Rounded ridge extending from outer orbital margin across hepatic region. Subhepatic margin of cara- pace mammiform, separated by shallow arc from trituberculate anterolateral margin. In- testinal region swollen, demarcated by in- 413 distinct grooves, bearing conical tubercle posterioly. Posterior margin bearing later- ally 2 lamellar denticles (Fig. 2B). Third maxillipeds bearing conicalal granules an- teriorly. Thoracic sternites granulate. Cheliped and pereiopods closely granu- late throughout. Cheliped merus in adult male 0.85 as long as carapace; fingers near- ly as long as upper margin of palm. Male abdomen minutely granulate, preapical denticle narrowly triangular. Shaft of first male pleopod nearly straight, apical process incurved, looped (Figs. 5A, B). Distribution.—Western Pacific: Fiji, Va- nuatu, New Caledonia, Philippines; 234— 500 m. Remarks.—Toru pilus differs from the other species described herein of Toru, in having a prominently produced front, pro- nounced postorbital concavity, ridged he- patic region, and distally looped first male pleopod. The male holotype of 7. pilus is badly broken; the detached chelipeds (Tan 1996, fig. 4H, 8b) do not belong with the cara- pace; the distal segment of the first male pleopod is looped, not as depicted by Tan (1996, fig. 8a, f). The female paratype is entire. Toru septimus, new species | SWOSs AC Ha, 1D) Type Material.—Holotype: Loyalty Is., Stn DW 421, 20°26.27'S, 166°40.17’E, 245 m, 16 Feb 1989, 1 3 13.1 mm (MNHN B.28513). Paratypes: Idem, 1 ¢ 12.2, mm, 2 2 13.4, 13.3 mm (MNHN B.28514). Material examined.—Fiji. MUSOR- SOM! OF eStn EP r1386; 18785" S: 178°05.1’E, 230-344 m, 19 Aug 1998, 1 juv. (MNHN). Vanuatu. MUSORSTOM 8, Stn CP 963, 20°20.10'S, 169°49.08’E, 400—440 m, 21 Sep 1994, coll. B. Richer de Forges, 4 ¢ 10.1-14.8 mm, 6 2 10.5-13.9 mm, 1 juv. GVENEIN) Stn “CP “986% “19°20:57' S; 169°31.48’E, 602-648 m, 23 Sep 1994, colle BY Richer oder Forses, “2 128 am 414 (MNHN). Stn CP 1006, 18°50.24’S, 168°56.87'E, 574-611 m, 25 Sep 1994, coll. B. Richer de Forges, | juv. (MNHN). Sim (Cl WON, WW S2Ek0'S, Wee 26.2018, 294-295 m, 27 Sep 1994, coll. B. Richer de lorves, G 12.3 wom, 2 12.1 icon (VIINIEIN) SeSitnt S@R OSs li 2288) Se 168°25.08’E, 300-301 m, 27 Sep 1994, coll. B. Richer de Forges, ¢ 11.7 mm, @ broken (MNHN). Stn CP 1092, 15°10.80’S, 167°12.33’E, 314—321 m, 6 Oct 1994, coll. B. Richer de Forges, 3 juvs. (MNHN). Stn CP 1094, 15°08.02'S, 167°11.99’E, 312- 314 m, 6 Oct 1994, coll. B. Richer de Forg- es, | juv. (MNHN). Loyalty Is. MUSORSTOM 6, Stn DW 417, 20°41.80'S, 167°03.65'E, 283 m, 16 Feb 1989, 56 12.7 mm (MNHN). Sth DW 423, 20°25.85'S, 166°40.50’E, 280 m, 16 Feb 1989, 1 juv. (MNHN). Stn DW 440, 20°48.80'S, 167°17.25'E, 288 m, 19 Feb 1989, 2 12.6 mm (MNHN). Stn DW 451, 20°59'S, 167°24.50’E, 330 m, 20 Feb 1989, 2 7.3 mm (MNHN). Stn DW 457, 21°00.42’S, 167°28.71'E, 353 m, 20 Feb 1989, 2 juvs. (MNHN). Stn DW 479, 21°09.13'S, 167°54.95'E, 310 m, 22 Feb 1989, ¢ 13.0 mm, | juv. (MNHN). Sth DW 481, 21°21.85'S, 167°50.30’E, 300 m, 23 Feb 1989, 2 12.0 mm (MNHN). BATHUS ll, Sta CP WOW, ZI427I2Q"S, IOG35.75 8, 347-375 m, 19 Mar 1993, 2 3 9.9, 12.6 mm, | juv. (MNHN). BATHUS 2, Stn CP 742, 22°33.45’S, 166°25.86’'E, 340—470 m, 14 May 1993, 2 ovig. 13.6 mm (MNHN). New Caledonia. Lagoon, 22°33.41'S, 166°25.74’E, 300 m, 12 Sep 1994, 2 9.5 m (MNHN). BIOCAL, Stn CP 105, 21°31’'S, 166°22’E, 330-335 m, 8 Sep 1985, 2 3 9.0, 12.5 mm, 2 9.0 (MNHN B18412). Stn CP NOs, 2202S, lor Cols, 335 mm, 9 Seo IMss, 6 IOS mi! 2 ome. 13.5 iar (MNHN B19177). Chesterfield Is.. MUSORSTOM 5, Stn 258, 25°32.8'S, 159°46.10’E, 300 m, 8 Oct 1986, 6 12.5 mm (MNHN B21212). Stn 261, 25°26.58'S, 159°45.88’E, 300 m, 8 Oct 1986, 6 13.7 mm (MNHN B18372). Stn 268, 24°44,70'S, 159°39.20’E, 280 m, 9 Oct PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1986, 3d 13.9 mm (MNHN B21211). Stn 268, 24°44,70'S, 159°39.20’E, 280 m, 9 Oct 1986, 3 3d 9.1-13.0 mm (MNHN B18378). Stn 275, 24°46.60'S, 159°40.30’E, 285 m 9 Oct 1986, 6 10.8 mm, 2 2 13.1, 14.1 mm, 1 juv. (MNHN_ B18376). Stn 276, 24°48.90'S, 159°40.90'E, 269-258 m, 9 Oct 1986, 4 3 10.7-13.1 mm, 2 7.8 mm, ovig. 11.8 mm (MNHN B18493). Stn 277, 24°10.60'S, 159°34.90'E, 270 m, 10 Oct 1986, 2 5 10.4, 13.5 mm, 2 juv. (MNHN B18368). Stn 280, 24°09.99’S, 159°35.75’E, 270 m, 10 Oct 1986, 4 5 10.3-10.5 mm, 2 2 10.3, 10.0 mm, 3 juvs. (MNHN B18377). Stn 281, 24°10.54’S, 159°34.32'E, 272 m, 10.10.1986, @2@ ovig. 11.1 mm (MNHN B18484). Stn 284, 24°09.96’S, 159°33.49’E, 10 Oct 1986, 2 ovig. 13.5 mm (MNHN B 18374). Stn 285, 24°09.35'S, 159°34.04’E, 245-255 m, 10 Oct 1986, 3 11.8 mm (MNHN B18495). Stn 287, 24°05.40'S, 159°36.30’E, 270 m, 10 Oct 1986, 2 ovig. 11.2 mm (MNHN B18490). Stn 288, 24°04.80'S, 159°36.80'’E, 270 m, 10 Oct 1986, 3 12.6 mm, 3 juv. (MNHN B18488). Stn 289, 24°01.50’S, 159°38.40’E, 273 m, 10 Oct 1986, 8 6 10.3-12.4 mm, 9 2 9.2-13.0 mm, | juv. (MNHN B18454). Stn 291, 23°07 70'S, 159°28:40'E) 300 em SaiO@ct 1986, 6 11.2 mm, 2 damaged (MNHN B18485). Stn 307, 22°11.07’S, 159°24.07’E, 350-345 m, 12 Oct 1986, ¢ 10.4 mm, @ ovig. 13.7 mm (MNHN B18489). Stn 309, 22°10.20'S, 159°22.80'E, 340 m, 12 Oct 1986, 6 10.6 mm (MNHN B18492). Stn 319) 22 24'S) 159 11650) Ey 320=32 >see Oct 1986, 5 9.7 mm (MNHN B21321). 22°40'S, 167°10'E, 200-350 m, 10 Oct 1986, ¢ 9.2 mm (MNHN B19163). BA- THUS 4, Stn DW 902, 19°00.84’S, 163°14.83’E, 341-351 m, 4 Aug 1994, o 14.3 mm, | juv. (MNHN). Stn CP 905, 19°02.45'S, 163°15.65’E, 294-296 m, 4 Aug 1994, 3d 13.8 mm (MNHN). Description.—Carapace with dorsal sur- face evenly, minutely granulate. Frontal lobes squat, rounded, pilose. Subhepatic margin of carapace mammiform, separated by shallow arc from trituberculate antero- VOLUME 116, NUMBER 2 lateral margin. Intestinal region swollen, demarcated by indistinct grooves, bearing conical tubercle posterioly. Posterior mar- gin bearing laterally 2 lamellar, rounded denticles. Third maxilliped pilose, bearing conicalal granules anteriorly (Fig. 2C). Thoracic sternites granulate. Cheliped and pereiopods closely granu- late throughout. Cheliped merus in adult male nearly as long as carapace; fingers as long as upper margin of palm. Male abdomen minutely granulate prox- imally, preapical denticle triangular. Shaft of first male pleopod sinuous, apical pro- cess bent at right angle, sigmoid (Figs. 5C, D). Etymology.—From Latin, septem, seven, for the shape of the apical process of the first male pleopod. Remarks.—Toru septimus, new species, differs from the closely allied T. trituber- culatus in its finer granulation on carapace and chelipeds, the rounded, rather than sub- quade, frontal lobes, and in the sigmoid shape of the apical process of the first male pleopod. Distribution.—Western Pacific: Fiji, New Caledonia, Loyalty Is., Vanuatu; 200— 648 m. Toru trituberculatus (Sakai, 1961), new combination Rigs: 2D or, EF Randallia trituberculata Sakai, 1961:134, plas nee 2: 1965-42) pl li Sinee2-976: 98, pl. 29, fig. 1, text-fig. 53.—Sertene, 1968:45 —Seréne’ S& WVadon, 19812119; 124.—Richer de Forges, 1983:634. Chen, 1989:215, fig. 14, pl. 3, fig. 2.— Takeda, 2001:230.—Chen & Sun, 2002: 340, fig. 150, pl. 14.3. Material examined.—Japan. Shikoku I., Tosa Bay, 19 Nov 1958, coll. T. & K. Sakai, 6 12.7 mm, 2 11.4 mm (USNM 120707). Tosa Bay, 3 ¢ 11.8-13.0 mm (SME ex. coll. Sakai). 6 13.2 mm (SME ex. coll. Sa- kai). Mimase, Dec 1961, 6 13.5 mm (SME 415 ex. coll. Sakai). Mimase, Mar 1963, ¢ 13.0 mm (SME ex. coll. Sakai). Indonesia. Tanimbar I., KARUBAR, Stn GPO Sr98KS; 132706 E59233=146 an Nov 1991, ¢ 8.6 mm (MNHN). Stn CP 86, 926) S313) 1G4Ey225=223 am: 4 Nov 1991) 5 6 8.6-11.1 mm, 6 2 9.0-11.9 mm (MNHN). Philippines. MUSORSTOM 1, Stn 27, 13°59.8'N, 120°18.6’E, 192-188 m, 22 Mar 1976, 2 11.2 C(MNHN B18084). Stn 30, 14°01.3'N, 120°18.7’E, 186-177 m, 22 Mar 1976, d 7.1 mm (MNHN B18086). Stn 34, 14°01.0'N, 120°15.8’E, 191-188 m, 23 Mar 1976, 2 10.6 mm (MNHN B18083). Stn 51, 13°49.4’N, 120°04.2’E, 200-170 m, 25 Mar 1976, 6 damaged, 2 ovig. 11.3 mm (MNHN B18085). Stn 64, 14°00.5’N, 120°16.3’E, 194-195 m, 27 Mar 1976, 8.3 mm (MNHN B18082). MUSORSTOM 3, Stn 88, 14°O1’N, 120°17’E, 183-187 m, 31 May 1985, @ ovig. 10.7 mm (MNHN B93) st Ce 100s 14-00 NE 120 187E: LSPS) ti th Yom Wks. @ WO) tava (MNHN B17983). Stn CP 108, 14°01'N, 120-1384. 188195 sim 2 Jun 1985376" 9:9 mm, ¢ ovig. 11.1 mm (MNHN B17984). Stn CP 120, 12°06'N, 121°16’E, 219-220 m, 3 Jun 1985, 2 2 7.9, 8.8 mm (MNHN Big 986). sim CP 139) fl 530N 2 14): 240-267 m, 6 Jun 1985, 10 3d 10.1-11.8 mm, 6 2 7.6-11.3 mm (MNHN B17985). Redescription.—Carapace with dorsal surface granulate, granules more prominent anteriorly. Frontal lobes squat, subquadrate, minutely granulate, pilose. Subhepatic mar- gin of carapace prominently granulate, mammiform; separated by shallow arc from trituberculate anterolateral margin. Intesti- nal region swollen, demarcated by distinct grooves, bearing conicalal tubercle poster- ioly. Posterior margin bearing laterally 2 la- mellar, rounded denticles (Fig. 2D). Third maxillipeds pilose, bearing conical granules anteriorly. Thoracic sternites gran- ulate. Cheliped, pereiopods closely granulate throughout. Cheliped merus in adult male 416 0.85 as long as carapace; fingers as long as upper margin of palm. Male abdomen minutely granulate prox- imally, preapical denticle triangular. Shaft of first male pleopod sinuous, apical pro- cess bent at right angle, tip upcurved (Figs. SJ8, 16). Color.—*‘*|T]he carapace ... is yellow- ish-red, but the surface near the postero-lat- eral and posterior margins is pale whitish” (Sakai 1965:42). Distribution.—Pacific Ocean: Japan, In- donesia, Philippines; 35-267 m. Remarks.—Toru_ trituberculatus is not ‘““most closely related to R. distincta Rath- bun (1906)” as claimed by Sakai (1961: 135), it belongs in a different genus. Toru trituberculatus differs from Tanaoa distinc- tus in having a lingulate telson in the male, anteriorly excavate lateral walls of the male abdominal sulcus, prominent margins to the female abdominal cavity, and bearing a dig- itate apical process on the male first pleo- pod. Key to Species of Toru, new genus 1. Anterolateral margin of carapace lacking WUberculest rh wee 2 ee hued oak epee eke 2 — Anterolateral margin of carapace tritu- benculaterse et = me er ee ee Aiaek BO ee 3 2. Lateral margin medially set with small tubercle; pereiopods closely granulate T. granuloides, new combination — Lateral margin uniformly rounded; an- terior margin of pereiopodal meri, carpi and propodi prominently spinose ee oe T. mesjatzevi, new combination 3. Front prominently produced, postorbital region concave; hepatic region ridged; first male pleopod distally looped ..... T. pilus, new combination — Frontal lobes squat, postorbital region not concave; hepatic region lacking ridge; first male pleopod distally bent at right angle 4. Carapace and chelipeds finely granulate; apical process of first male pleopod sig- moid, tip not upcurved 2 © © © © © © ON 5 0 Oo Fo O-D oO So Oo T. septimus, new species — Carapace and chelipeds prominently OOO 50 0 Oo 0 oD ooo oo PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON granulate; apical process of first male pleopod not sigmoid, tip upcurved ... T. trituberculatus, new combination Urashima, new genus Diagnosis.—Carapace subrhomboidal, globose. Front narrow, bilobed. Eyes small, retractible. Outer orbital margin trisutured, V-shaped gap proximally on ventral margin. Antennules obliquely folded, basal anten- nular segment squat, operculiform, sealing bottom half of antennular aperture. Anten- nae small, slender, basal antennal segment inserted in orbital hiatus. Postorbital region concave. Hepatic, branchial regions demar- cated by grooves, 2 pairs of pits along car- diobranchial grooves. Intestinal region swollen, demarcated by grooves. Posterior margin of carapace narrow, bilobate. Third maxilliped exopod slightly shorter than endopod, not quite sealing efferent branchial channel; endopod of adult female medially with vertical line of setae; merus of endopod subtriangular, shorter than sub- rectangular ischium. Anterior margin of ef- ferent branchial channel produced, bilobed, separated by narrow groove from lower or- bital margin. Chelipeds subequal. Cheliped merus sub- cylindrical; fingers laterally compressed, as long as upper margin of palm, inner mar- gins denticulate. Pere1opods slender, prom- inently granulate, short; all but last dactyl as long as propodi; upper surface setose, tips corneous. Fourth thoracic sternite swollen laterally. Abdominal sulcus of male deep, elongate, nearly reaching buccal cavity, anterior mar- gin raised. Abdominal segments 3-6 of male fused, basio-lateral regions inflated, fused segment narrowing distally, bearing denticle at distal margin; lateral margin car- inate, carina fitting into groove at suture be- tween sternites 4 and 5. Telson slender, nearly half as long as fused segment, not reaching tip of abdominal sulcus. Abdomi- nal segments 4—6 of female fused, shield- like; margin of abdominal cavity promi- VOLUME 116, NUMBER 2 nent, rampart-like. Telson ogival, basal margin sinuous. Male first pleopod elongate, stocky, dis- tally flattened; second pleopod short, re- curved, distally attenuate. Type species.—Randallia pustuloides Sa- kai, 1961. Species.—Urashima_ lamellidentatus (Wood-Mason, 1892), Urashima pustulo- ides (Sakai, 1961). Etymology.—Urashima, according to Japanese legend, was a handsome fisher- man who married a mermaid and dwelt with her undersea. The name Urashima is to be considered as an arbitrary combina- tion of letters, and heretofore takes the gen- der masculine. Remarks.—Urashima, new genus, differs from the other genera discussed herein in the laterally compressed cheliped fingers, the granulate pereiopodal carpi and propo- di, and the oar-shaped first male pleopod. Urashima lamellidentatus (Wood-Mason, 1892), new combination Fig. 2E Randallia lamellidentata Wood-Mason, 1892: pl. 5, figs 5, 5a, 5b.—Alcock, 1894:404; 1896:195; 1899:26.—Kemp & Sewell, 1912:29.—Ihle, 1918:312.—Se- rene, 1954:491; 1968:45.—Yaldwyn & Dawson, 1976:96.—Richer de Forges, 1983:634. Type Material.—Holotype: Andaman Is. 11°3.40’N, 92°46.40’E, 3 juv. cl 11.9 mm (NHM 1896.9.8.7 ex. Indian Museum). Redescription.—Dorsal surface of cara- pace unevenly tuberculate, tubercles more pronounced laterally, posteriorly. Frontal lobes squat, rounded, minutely granulate. Anterior margin of efferent branchial chan- nel deeply sutured (Fig. 2E). Third maxil- liped unevenly granulate, endopod merus with median rise proximally. Subhepatic margin of carapace with la- mellate crest, followed by pearliform gran- ule. Branchial margin of carapace with 2 lamellate crests separated by granulate den- 417 ticle. Posterolateral margin set with pearli- form granules. Posterior margin laterally with lamellate rounded denticles. Intestinal region inflated, bearing posteriorly small, upcurved conical tubercle. Cheliped stout, prominently granulate; merus two-thirds as long as carapace, gran- ules larger, pearliform distally. Upper mar- gin of palm prominently crested, fingers lat- erally compressed; upper, lower margins carinate. Pereiopodal carpi and propodi bearing rows of conical granules on upper margin, as well as merus of last pereiopods. Thoracic sternites granulate. Fused ab- dominal segments in male lacking horizon- tal ridge, distal tubercle spur-like, promi- nent. Margins of abdominal sulcus in fe- male lamellate, prominent. First male ple- opod stout, sinuous, distally flattened, oar-like. Color.—*‘|W )hite, with a pinkish blush” (Alcock 1894:404). Distribution.—Indian Ocean: Andamans, Maldives; 340—640 m. Remarks.—Urashima lamellidentatus differs from U. pustuloides in bearing la- mellate crests on the anterolateral margins of the carapace, and lamellate rounded den- ticles laterally on the posterior margin of the carapace. Urashima pustuloides (Sakai, 1961), new combination Figs. 2E 5G, H Randallia pustuloides Sakai, 1961:135, pl. Sy ee A OOo alOU np SU) tise I. text fig. 54.—Yaldwyn & Dawson, 1976: 96.—Richer de Forges, 1983:634 (tab.)—Chen, 1989:219, fig. 16.—Tan, 1996:1054.—Takeda, 1997:238.—Ikeda, 1998:83, pl. 20, figs la, b, 2, 3a—b.—_-Ng eral OOleanO: Type Material.—Paratypes: Japan. Shi- koku I., Mimase, Tosa Bay, coll. K. Sakai, Gs4i35 imme 2) ovigy S580, 38:9) mm (NHM 1961.6.5.46—48). Material examined.—Japan. Shikoku L., Mimase, Tosa Bay, coll. K. Sakai, 2 37.6 418 mm (SMF). 6 35.3 mm (SMF). ¢ 35.6 mm, 2 38.4 mm (SMF). 3 36.7 mm (SMF). 6 37.2 mm, 2 37.6 mm (SMF TIO). DEC IDOISZ SG ZO), 3O.8 worn (SMF). 250 m, Apr 1968, ¢ 37.1 mm, ovig. 2 38.6 mm (SMF 15103). Off Ashizuri, 366 m, 24 Nov 1958, 2 2 36.7, 37.7 mm, ° broken (SMF). Taiwan. I-Lan county, 1998, coll. TY. Chan, 6 32.8 mm (MNHN B26326). Tong- kuang, 25 Feb 1995, 300 m, coll. T.Y. Chan, 2 § 33.3, 336 wan, 2 2 YOO, ZS.S) saora (NTOU). Tashi, 5 Mar 1997, 200 m, coll. T.Y. Chan, 2 18.3 mm (NTOU). Philippines. Btw Negros, Siquijor, R/V Albatross Stn 5538, 9°08.15'N, 123°23.20’E, 468 m, 19 Aug 1909, id. C.G.S. Tan, 2 38.3 mm (USNM). MUSORSTOM 2, Stn CP 20, 14°00'N, 120°18’E, 185-192 m, 6 25.9 mm (MNHN B18081). Indonesia. KARUBAR, Tanimbar I., Stn CP69, 356-368 m, 8°42’'S, 131°53'E, 2 Nov 1991; 2 6 37.4, 38.2 mm, 3 2 25.0—31.6 mm (MNHN). Stn CP 77, 352-346 m, SoS, ISI27 18, 3 Noy IO, 3 6 253= 27.2 mm, 2 38.9 mm (MNHN). Australia. 18°05'S, 118°08’E, 440—442 m, 22 Aug 1983, 2 37.9 mm (WAM c14731). 14°51’S, 121°35’E, 300 m, 3 Aug 1989, 5 37.0 mm (WAM 577-92). Redescription.—Dorsal surface of cara- pace unevenly tuberculate, larger, pustule- like tubercles on branchial region. Frontal lobes squat, rounded, minutely granulate. Anterior margin of efferent branchial chan- nel deeply sutured (Fig. 2F). Third maxil- liped unevenly granulate, merus with me- dian ridge proximally. Subhepatic margin of carapace inflated, median tubercle followed by 1 or 2 smaller, pearliform granules. Lateral margin of car- apace medially with 3 flattened, triangular, upcurved denticles, decreasing in size pos- teriorly. Posterolateral margin set with pear- liform tubercles. Posterior margin with lat- eral triangular denticles, closely set with pearliform tubercles. Intestinal region in- flated, bearing posteriorly small, upcurved conical tubercle. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Cheliped prominently granulate; granules larger, pearliform distally; merus_ three- quarters as long as carapace. Fingers later- ally compressed; upper, lower margins car- inate. Pereiopodal meri distally granulate on upper margin, fifth merus bearing coni- cal granules on posterior surface. Thoracic sternites with low granulation. Fused abdominal segment in male lacking horizontal ridge, distal tubercle spur-like, prominent. Margins of abdominal sulcus in female lamellate, prominent. First male ple- opod stout, sinuous, distally oar-shaped (Figs. 5G, H). Distribution.—Pacific Ocean: Japan, Tai- wan, Philippines, Indonesia, Australia; 85— 468 m. Remarks.—Sakai (1961), and Chen (1989) considered Urashima pustuloides re- lated to Tanaoa pustulosus. However, U. pustuloides in addition to the characters cit- ed by Chen (1989), and Takeda (1997), is easily distinguished from 7. pustulosus in having laterally compressed, rather than rounded, cheliped fingers; granulate, rather than smooth pereiopodal carpi and propodi; and oar-shaped, rather than bearing preapi- cal process on the first male pleopod. Key to Species of Urashima, new genus 1. Anterolateral margins of carapace bear- ing lamellate crests; posterior margin of carapace with lamellate denticles; upper margin of cheliped palm prominently CHESTS .ca eC ee Pe ea ee U. lamellidentata, new combination — Anterolateral margins of carapace lack- ing lamellate crests; posterior margin of carapace with triangular denticles; che- liped palm subcylindrical U. pustuloides, new combination 000000 2 © © © © © © © «© Acknowledgements The late R. B. Manning invited me to study the leucosiids in the National Muse- um of Natural History, Smithsonian Insti- tution, Washington D.C. This work is ded- icated to his memory. The visit was sup- ported by that Museum’s Collection Im- VOLUME 116, NUMBER 2 PLOVenTe ni MUnGd Sa slender atenul eto). Berents, T.-Y. Chan, P Clark, A. Crosnier, P. Davie, D. Eibye-Jacobsen, C. H. J. M. Fransen, D. Guinot, M. Hewitt, L. B. Hol- thuis, M. van der Merwe, D. Platvoet, V. Spiridonov and M. Tiirkay for entrusting me with valuable material from their col- lections. I thank B. Richer de Forges for elucidating the names of the French expe- ditions. I am deeply indebted to Alain Cros- nier for his generous hospitality. I am grate- ful to R. Lemaitre for comments on an ear- lier version of the manuscript. Special thanks to the librarians of the American Museum of Natural History, New York. The photographs were taken by A. Shoob, and H. Bernard inked the drawings. Literature Cited Alcock, A. 1894. XLIV.—Natural history notes from H. M. Indian Marine Survey Steamer ‘Investi- gator’, Commander R. FE Hoskyn, R.N., late commanding. Series II., No. 1. On the results of the deep-sea dredging during the season 1890—91 (concluded).—The Annals and Mag- azine of Natural History 13:400—411. . 1896. Materials for Carcinological Fauna of India. N. 2. The Brachyura Oxystomata.—Jour- nal of the Asiatic Society of Bengal 65(2):134— 296. . 1899. An Account of the Deep-Sea Brachyura collected by the Royal Indian Marine Survey Ship “Investigator”. Calcutta, 85 pp. , & A. R. J. Anderson. 1894. Natural history notes from H. M. Indian Marine Survey Steam- er ‘Investigator’, Commander C. EK Oldham, R.N., commanding. Series II., No. 14. An ac- count of a recent collection of deep-sea Crus- tacea from the Bay of Bengal and Laccadive Sea.—Journal of the Asiatic Society of Bengal 63(2):141-185. 5, . 1897. Crustacea, part V. Illustra- tions of the Zoology of the Royal marine sur- veying Steamer investigator, Calcutta pl. 28—32. Austin, W. C. 1985. An annotated checklist of marine invertebrates in the cold temperate Northeast Pacific. 3:1—682. Khoyatan Marine Laboratory. Baker, C. EK 1912. Notes on the crustacea of Laguna Beach.—First annual Report of the Laguna Ma- rine Laboratory at Laguna Beach, Orange County, California 1912:100—117. Bonfil, R., & A. Carvacho. 1989. The crabs of Bahia de Todos Santos, Baja California, part I. Drom- iidae, Leucosiidae, Majidae and Parthenopidae 419 (Crustacea: Decapoda: Brachyura).—Ciencias Marinas 2:79—109. Boone, L. 1930. Crustacea: Stomatopoda and Brach- yura. Scientific Results of the cruises of the Yachts “‘Eagle’”’ and “‘Ara’’, 1921-1928, Wil- liam K. Vanderbilt, Commanding. Crustacea: Stomatopoda and Brachyura.—Bulletin of the Vanderbilt Marine Museum 2:1—228, 74 pls. Borradaile, L. A. 1916. Crustacea, I. Decapoda. British Antarctic (“Terra Nova’’) Expedition, 1910. Natural History Report, Zoology 3(2):75—110. Campbell, B. M. 1971. New records and new species of crabs (Crustacea: Brachyura) trawled off southern Queensland: Dromiacea, Homolidea, Gymnopleura, Corystoidea, and Oxystomata.— Memoires of the Queensland Museum 16(1): 27-48. Chang, C. 1963. A check-list of Taiwan crabs with descriptions of 19 new records.—Biological Bulletin, Tunghai University 14:1—23, 2 pls. Chen, H. 1989. Leucosiidae (Crustacea, Brachyura). Jn J. Forest, ed., Résultats des Campagnes MU- SORSTOM, vol. 5.—Mémoires du Muséum na- tional d’ Histoire naturelle (A) 144:181—263. Chen, H., & H. Sun. 2002. Brachyura: marine primi- tive crabs. /n Fauna Sinica Invertebrata: Ar- thropoda: Crustacea, 30: xiii + 597, figs 1-237, pls 1—6. Science Press, Beijing. Davie, P. J. EK 2002. Crustacea: Malacostraca: Eucarida (part 2): Decapoda—Anomura, Brachyura. In A. Wells & W. W. K. Houston, eds., Zoological Catalogue of Australia. Vol. 19.3B. CSIRO Publishing, Melbourne, 641 pp. Doflein, E 1904. Brachyura. Jn Wissenschaftliche Er- gebnisse der Deutschen tiefsee-Expedition Auf dem Dampfer ‘Valdivia’, 1898-1899, Jena, 6: i—xiv, 1-314, 58 pl. Galil, B. S. 2001. A new genus of leucosiid crabs (Crustacea, Brachyura).—Zoosystema 23(1): 65-75. Garth, J. S. 1940. Some new species of brachyuran crabs from Mexico and the Central and South American mainland.—Allan Hancock Pacific Expedition 5(3):53—96, pl. 11—26. . 1960. Distribution and affinities of the brach- yuran Crustacea. Jn The biogeography of Baja California and adjacent seas, part Il. Marine Bi- otas.—Systematic Zoology 9(3):105—123. . 1966. Eastern Pacific Expeditions of the New York Zoological Society. Oxystomatous and al- lied crabs from the west coast of tropical Amer- ica.—Zoologica 51:1—16. Gibbes, L. R. 1850. On the carcinological collections of the United States, and on enumeration of spe- cies contained in them, with notes on the most remarkable, and descriptions of new species.— Proceedings of the American Association for the Advancement of Science 3:167—201. 420 Hendrickx, M. E. 1990. The stomatopod and decapod crustaceans collected during the GUAYTEC II cruise in the central Gulf of California, Mexico, with the description of a new species of Ple- sionika Bate (Caridea: Pandalidae).—Revista de Biologia Tropical 38(1):35—53. . 1992. Distribution and zoogeographic affini- ties of decapod crustaceans of the Gulf of Cal- ifornia, Mexico.—Proceedings of the San Diego Society of Natural History 20:1—11. . 1995. Checklist of brachyuran crabs (Deca- poda: Brachyura) from the eastern tropical Pa- cific.—Bulletin de |’ Institute Royal des Scienc- es Naturelles de Belgique 65:125—150. . 1997. Los Cangrejos Braquiuros (Crustacea: Brachyura: Dromiidae, hasta Leucosiidae) del Pacifico Mexicano. Comision Nacional para el Conocimiento y Uso de la Biodiversidad, 178 Pp. Holmes, S. J. 1900. Synopsis of California stalk-eyed Crustacea.—Occasional Papers of the Califor- nia Academy of Sciences 7:1—256, 4 pls. Huang, Z. 1989. Brachyura (Dromiacea, Gymnopleura, Oxystomata, Oxyrhyncha, Corystoidea). In Wei Ch., ed., Fauna of Zhejiang. Crustacea. Zhe- jiang Science & Technology Publishing House, pp. 280-387. . 1994. Marine species and their distributions in China’s seas. China Ocean Press, Beijing, 764 pp + 134 index. Thle, J. E. W. 1918. Die Decapoda Brachyura der S1- boga-Expedition. III. Oxystomata: Calappidae, Leucosiidae, Raninidae. Siboga Expeditie 39b2: 1-322. Ikeda, H. 1998. The deep-sea crabs of Sagami Bay. Hayama Shiosai Museum, Inamoto Printing, Kanagawa, 180 pp., 73 pls. Kemp, S., & R. B. Sewell. 1912. Notes on the Deca- poda in the Indian Museum. III. The species obtained by R.I.M.S. S. ‘Investigator’ during the season 1910—11. R.B.—Records of the In- dian Museum Calcutta 7:15—32. McLay, C. L. 1988. Crabs of New Zealand.—Leigh Laboratory Bulletin 22:1—463, 85 figs. Miers, E. J. 1886. Report on the Brachyura collected by H.M.S. Challenger during the years 1873- 1876.—Report on the Scientific Results of the Voyage of H.M.S. Challenger during the years 1873-76, (Zool) 17:1—362. Ng, P. K. L., C. H. Wang, P. O. Ho, & H. T. Shih. 2001. An annotated checklist of brachyuran crabs from Taiwan (Crustacea: Decapoda). Na- tional Taiwan Museum Special Publication Se- ries 11:1—86. Ovaere, A. A. 1989. Ihleus gen. nov., a new leucosiid genus (Crustacea, Brachyura).—Bulletin de l'Institut Royal des Sciences Naturalles de Bel- gique, Biologie 59:95—100. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Poupin, J., & B. Richer de Forges. 1991. New or rare crustaceans from French Polynesia (Crustacea: Decapoda).—Memoirs of the Queensland Mu- seum 31:211. Randall, J. W. 1839. Catalogue of the Crustacea brought by Thomas Nuttal and J.K. Townsend, from the west coast of North America and the Sandwich Islands, with descriptions of such species as are apparently new, among wich are included several species of different localities, previously existing in the collection of the Academy.—Journal of the Academy of Natural Sciences of Philadelphia 8:106—147. Rathbun, M. J. 1893. Scientific results of explorations by the U.S. Fish Commission Steamer Alba- tross. XXIV. Descriptions of new genera and species of crabs from the west coast of North America and the Sandwich Islands.—Proceed- ings of the United States National Museum 16: 223-260. . 1898. The Brachyura collected by the U.S. fish Commission Steamer Albatross on the voy- age from Norfolk, Virginia, to San Francisco, California, 1887—1888.—Proceedings of the United States National Museum 21:567—616. .. 1904. Decapod crustaceans of the northwest coast of North America.—Harriman Alaska Ex- pedition 10:1—211, 10 pls. . 1906. The Brachyura and Macrura of the Ha- waiian Islands.—Bulletin of the United States Fish Commission 23(3):829—930. . 1922. New species of crabs from Curagao.— Proceedings of the Biological Society of Wash- ington 35:103—104. . 1935. Preliminary descriptions of seven new species of oxystomatous and allied crabs.—Pro- ceedings of the Biological Society of Washing- ton 48:1—4. . 1937. The Oxystomatous and allied crabs of America.—Bulletin of the United States Nation- al Museum 166:1—278. Richer de Forges, B. 1983. Randallia serenei (Crus- tacea, Decapoda, Brachyura), espéce nouvelle de Leucosiidae du Pacifique.—Bulletin du Mu- séum national d’Histoire naturelle, Paris 5(2): 633-640. Rodriguez de la Cruz, M. C. 1987. Crustaceos deca- podos del Golfo de California. Sria. Pesca (ed.). Mexico, D.F 306 pp. Sakai, T. 1934. Brachyura from the coast of Kyusyu, Japan.—Science Reports of the Tokyo Bunrika Daigaku B 1(25):281—330. . 1935. Crabs of Japan. 66 plates in life colours with descriptions.—Tokyo, Sanseido Co. . 1937. Studies on the crabs of Japan. II. Ox- ystomata.—Science Reports of the Tokyo Bun- rika Daigaku, 3, suppl. 2:67—192. . 1961. New species of Japanese crabs from the VOLUME 116, NUMBER 2 collection of His Majesty the Emperor of Ja- pan.—Crustaceana 3(2):131—150. . 1965. The crabs of Sagami Bay collected by His Majesty the Emperor of Japan. Tokyo, Ma- ruzen Co, I-xvi, 1—206 [English]; 1—92 [Japa- nese], pl. 1-100, 1 map. . 1976. Crabs of Japan and the adjacent seas (Tokyo: Kodansha). [In three volumes: 1. En- glish text, xxx1x+773 pp., figs. 1-379; 2. Plates volume, 16 pp., 1-251; 3. Japanese text, 461 pp. Figs 1—2.]. Schmitt, W. L. 1921. The marine decapod Crustacea of California with special reference to the deca- pod Crustacea collected by the United States Bureau of Fisheries Steamer Albatross in con- nection with the biological survey of San Fran- cisco Bay during the years 1912—1913.—Uni- versity of California Publications in Zoology DS Mooe emo: Serene, R. 1954. Sur quelques espéces rares de brach- yures (Leucosidae) de 1’ Indo-Pacifique.—Treu- bia 22(3):453—500, pls 7-10. . 1968. Prodromus for a check list of the non- planctonic marine fauna of South ast Asia.— Singapore National Academy of Science. Spe- cial Publication no. 1. 122 pp. , & C. L. Soh. 1976. Brachyura collected dur- ing the Thai-Danish Expedition (1966).—Phu- ket Marine Biological Center Research Bulletin 12:1—37, figs 1-28, 8 pls. , & C. Vadon. 1981. Crustacés Décapodes: Brachyures. Liste préliminaire, description de formes nouvelles et remarques taxonomiques. In Résultats des Campagnes MUSORSTOM, vol. 1. Philippines. Mémoires ORSTOM 91: 117-140, pl. 1-4. Stimpson, W. 1857a. Notices of new species of Crus- tacea from western North America, being an ab- stract from a paper to be published in the Jour- nal of the society.—Proceedings of the Boston Society of Natural History 6:84—89. . 1857b. On the Crustacea and Echinodermata of the Pacific Shores of North America, part I. Crustacea.—Boston Journal of Natural History 6:444-532. . 1860. XI. Notes on North American Crusta- cea. I—Annals of the Lyceum of Natural His- tory in New York 7:49—93. Takeda, M. 1973. Report on the crabs from the sea around the Tsushima Islands collected by the Research Vessel ““Genaki” for the Trustees of the National Science Museum, Tokyo.—Bulle- tin of the Liberal Arts & Science College, Ni- hon University 1:17—68. . 1975. Crabs from the East China Sea, VI. A collection from off the Danjo Islands made by the R/V Hakuhé Maru Cruise KH-73-3.—Bul- 421 letin of the National Science Museum, Zoology 1(3):137-156. . 1997. Deep-sea decapod crustacean fauna of Suruga Bay, Central Japan.—National Science Museum Monographs 12:229—255, pls 1-5. . 2001. Annotated list of crabs from Tosa Bay, Southwest Japan, collected by the R/V Kotaka Maru during the years 1997—2000.—National Science Museum Monographs 20:217—262. , & S. Miyake. 1970. Crabs from the East Chi- na Sea. IV. Gymnopleura, Dromiacea and Ox- ystomata.—Journal of the Faculty of Agricul- ture, Kyushu University 16(3):193—235, pl.1. Tan, C. G. S. 1996. Leucosiidae of the Albatross ex- pedition to the Philippines, 1907-1910 (Crus- tacea: Brachyura: Decapoda).—Journal of Nat- ural History 30:1021—1058. Tyndale-Biscoe, M., & R. W. George. 1962. The Ox- ystomata and Gymnopleura (Crustacea, Brach- yura) of Western Australia with descriptions of two new species from Western Australia and one from India.—Journal of the Royal Society of Western Australia 45(3):65—96, pl. 1-3. Uchida, S. 1949. Illustrated Encyclopedia of the Fauna of Japan (Exclusive of Insects). Hokuryukan, Tokyo, 1898 pp. Utinomi, H. 1956. Coloured illustrations of sea shore animals of Japan. Hoikusha, Osaka, 167 pp., 64 color pls. Weymouth, EF W. 1910. Synopsis of the true crabs (Brachyura) of Monterey Bay, California——Le- land Stanford Junior University Publications University Series 4:1—64, 14 pls. Wood-Mason, J. 1892. Crustaceans, part |. Illustra- tions of the zoology of H. M. Indian Marine Surveying Steamer Investigator, pl. 1—5. Wood-Mason, J., & A. Alcock. 1891. XX VII. Natural history notes from H.M. Indian Marine Survey Steamer “Investigator’’, Commander R.E Hos- kyn, R.N., commanding. No. 21. Note on the results of the last seasons’s Deep-sea Dredg- ing.—The Annals and Magazine of Natural His- tory 7:258—272. (Crustacea attributed to Wood- Mason). Yaldwyn, J. C., & E. W. Dawson. 1976. First records of the crab genera Homola, Randallia, and Ro- chinia from New Zealand (Crustacea: Decapo- da: Brachyura).—Records of the National Mu- seum of New Zealand 1(6):91—103. Yokoya, Y. 1933. On the distribution of decapod crus- taceans inhabiting the continental shelf around Japan, chiefly based upon materials collected by s. Ss. Soyo-Maru during the years 1923—1930.— Journal of the College of Agriculture, Tokyo Imperial University 12(1):1—226, 71 figs. Zarenkov, N. A. 1969. Crabs of the family Leucosiidae (subfamilies Ebaliinae and Iliinae) collected in tropical waters of Indian and Pacific oceans.— Nauchnye Doklady Vysshei Shkoly, biologi- 422 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON cheskie Nauki, SSSR 12(10):16—26 [in Rus- sian]. . 1990. Plankton and benthos from the Nazca and Sala-y-Gomez ridges.—Transactions of the P. P. Shirshov Institute of Oceanology, Academy of Sciences of the USSR, Nauka, Moscow 124: 218-244 [in Russian]. . 1994. Crabs from seamounts of the western part of the Indian Ocean. Jn A. P. Kuznetsov & A. N. Mironov, eds., Bottom fauna of Sea- mounts.—Transactions of the P. P. Shirshov In- stitute of Oceanology, Russian Academy of Sci- ences, Nauka, Moscow 129:97—151 [in Rus- sian]. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):423—437. 2003. Type catalogue of the Crustacea Decapoda in the collections of the Museu de Zoologia da Universidade de Sao Paulo, Brazil Gustavo Augusto Schmidt de Melo, Renata M. Vezzani, and Oswaldo Campos Jr. (GASM, RMV) Museu de Zoologia da Universidade de Sao Paulo, Av. Nazaré 481, Ipiranga, Sao Paulo, SP, 04263-000, Brazil, e-mails: gasmelo@usp.br and rvezzani@uol.com.br; (OCJr.) Centro Universitario Sao Camilo, Av. Nazaré 1501, Ipiranga, Sao Paulo, SP, 04263-200, Brazil, e-mail: oswaldocamposjr@aol.com Abstract.—Data and pertinent information on the type specimens of eighty- four species of Crustacea Decapoda held in the Museu de Zoologia da Univ- ersidade de Sao Paulo, Brazil, are provided. The taxa are arranged in alphabetic order. Included are references where all taxa were described, collection data, and changes in nomenclature. Resumo.—Sao apresentados dados e informag6es referentes aos espécimes- tipo de oitenta e quatro espécies de Crustacea Decapoda, depositados na co- leg¢ao carcinol6gica do Museu de Zoologia da Universidade de Sao Paulo, Bra- sil. O tratamento das espécies é feito em ordem alfabética dentro das familias, e uma lista de referéncias inclui todos os trabalhos nos quais as espécies foram descritas. Brief History of Museu de Zoologia da Universidade de Sao Paulo The origin of Museu de Zoologia da Universidade de Sao Paulo can be traced back to 1886, the year of foundation of the Comissao Geografica e Geoldgica (Geo- graphical and Geological Commission, CGG) of the Province of Sao Paulo. The aim of the Commission was to improve sci- entific knowledge of the province. In the last decade of the XIX century, Francisco de Paula Mairynk acquired pri- vate collections belonging to Colonel Joa- quim Sertorio and offered them to the pro- vincial government in order to create a State Museum. After several years, these collections were incorporated into collec- tions already held by the Comissao Geo- grafica e Geoldgica. This acquisition cre- ated the nucleus of a museum of natural history. By the end of 1893, the State Museum received the name Museu Paulista. In 1895, as a commemoration of Brazilian indepen- dence, the present building of the Museu Paulista was inaugurated. This building is located near the streamlet where Emperor Dom Pedro I proclaimed independence for Brazil. The Section of Zoology of the Museu Paulista was established in 1925. In 1939 the Section was transferred to the Depart- ment of Zoology of the Secretariat of Ag- riculture, Industry, and Commerce of the state of Sao Paulo. A new building was then constructed in order to accommodate the collections. In 1969, The University of Sao Paulo in- corporated the Department of Zoology into its bureaucratic structure. At that time the Department received its present name of Museu de Zoologia (MZUSP). The Crustacea Collection The carcinological collection of the MZUSP was established on 19 March 1962, with 76 dried specimens found in the basement of the museum. Through 1966, 424 the collection was augmented by collections from rivers, mangrove swamps, and beach- es. The acquisition in 1967 of the Ocean- ographic Vessel Prof. W. Besnard by the Oceanographic Institute of the University of Sao Paulo (IOUSP) then enabled significant quantitative and qualitative improvements in the collection. The partnership between MZUSP and IOUSP in projects Ilha Grande (1966-1969) and GEDIP (Grupo Executivo do Desenvolvimento da Industria da Pesca) I (1968-1969) and II (1972), in addition to other, less extensive surveys, was crucial in bringing the collection to its present state. The collection was originally intended to lodge only species of the order Decapoda, because of space limitations. Nevertheless, other excellent collections were eventually added, for instance that of Cirripedia, do- nated by Dr. Paulo S. Young of the Museu Nacional do Rio de Janeiro, and a collec- tion of Isopoda assembled by the late Dr. Plinio Soares Moreira of IOUSP. Presently, the collection includes approximately 15,000 lots, and about 230,000 specimens of Decapoda, which are chronologically registered and sequentially numbered. There are about 6000 uncatalogued lots. This collection is considered one of the best in Latin America. Organization of Catalogue The catalogue presented herein includes all type specimens of Decapoda presently held in the MZUSP collection. Data and in- formation on 84 species of Crustacea De- capoda are presented, including Brachyura (46), Anomura (17), Caridea (10), and Thalassinidea (11). All specimens are pre- served in 70% ethanol. The highest hierarchical levels (super- families and families) are ordered according to Martin & Davis (2001). The genera and species are arranged alphabetically within each family. Species are listed by the scientific name under which each was originally described. Information regarding any subsequent name PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON changes is provided in the Remarks with a reference to the source of the change. In- consistencies between the original descrip- tion and the type specimen, register number, label or any additional comment are also noted. Instances of type specimens which have been donated to or lodged in other museums are also noted in the Remarks. Only types as recognized by the Inter- national Code of Zoological Nomenclature (1999), hereafter referred to as the ICZN, are included in this list. The collection includes some specimens considered “‘possible-types”’. This material, being potentially useful for future research on the group, is included in this paper, and pertinent details explained in the Remarks. Other abbreviations used: coll (collec- tor); DOUFPE (Departamento de Oceano- grafia da Universidade Federal de Pernam- buco); EPA (Expedicao Permanente do Amazonas); INPA (Instituto Nacional de Pesquisas da Amazonia); ITA (Instituto das Tecnologias Ambientais); juv Guvenile or juveniles); ovig. (ovigerous); Proj (Project); RD (Project Rio Doce); sta (station); TAAF MD 55 (Terres Australes et Antarctiques Francaises); UFRGS (Universidade Federal do Rio Grande do Sul). Suborder Pleocyemata Burkenroad, 1963 Infraorder Caridea Dana, 1852 Superfamily Palaemonoidea Rafinesque, 1815 Family Palaemonidae Rafinesque, 1815 Macrobrachium birai, Lobao, Melo, & Fernandes, 1986:50 Holotype.—6, Rio Branco, tributary of Rio Itapitangui, Cananéia, State of Sao Pau- lo, 24°54'44"S, 47°58'30"W, coll. V. L. Lo- bao and W. M. Fernandes, Dec 1984, MZUSP 8027. Paratypes.—All same data as holotype: 1 ¢6, MZUSP 8028; 1 6, MZUSP 8029; 1 65, MZUSE 80302 176) MZUSE S0SIesIaer MZUSP 8032; 1 2, MZUSP 8033. Remarks.—This species was redescribed in a later paper by Melo et al. (1988:89). VOLUME 116, NUMBER 2 Macrobrachium holthuisi Genofre & Lobao, 1978:273, fig. 1 Holotype.—6, Rio Guaeca, Sao Sebas- tiao, State of Sao Paulo, coll. V. Lobao, 21 Sep 1978, MZUSP 5283. Paratypes.—4 2, same data as holotype, MZUSP 4830. Macrobrachium inpa Kensley & Walker, 1982:6, figs. 7-9, 12c Paratype.—1 3d, Igarapé in the forest re- serve of INPA, Campina, km 60 BR-174, State of Amazonas, 21 Jan 1981, MZUSP 7014. Macrobrachium petronioi Melo, Lobao, & Fernandes, 1986:51 Holotype.—é6, Rio Branco, tributary of Rio Itapitangui, Cananéia, State of Sao Pau- lo, 24°54'44"S, 47°58'30"W, coll. V. L. Lo- bao and W. M. Fernandes, Dec 1984, MZUSP 8034. Paratypes.—All same data as holotype: INO MVIZAWSE "S035 —1iGa MZUSP 8036; | SS MZUWSP. 8037-1 2 MIZUSP 8038. Remarks.—This species was redescribed in a later paper by Melo et al. (1988:92). Pseudopalaemon gouldingi Kensley & Walker, 1982:18, figs. 20, 21 Paratypes.—118 juv, near the mouth of Rio Urubaxi, State of Amazonas, coll. M. Golding, 11 Feb 1980, MZUSP 7013. Superfamily Alpheoidea Rafinesque, 1815 Family Alpheidae Rafinesque, 1815 Alpheus chacei Carvacho, 1979:455, figs. 4-6 Holotype.—& ovig, Balneario Atalaia, State of Sergipe, coll. M. A. Santos, 12 Jul 1972, MZUSP 4545. Alpheus estuariensis Christoffersen, 1984:191, figs. 1, 2 Holotype.—é, Estuary of Rio Potengi, Natal, State of Rio Grande do Norte, in 425 nursery, Of shmmps, mud.) cole Ss 4S. Bueno, 26 Jan 1979, MZUSP 4544. Alpheus pouang Christoffersen, 1979:324 Holotype.—6é, Proj Sol, sta 1019, 238 90S. 4367 WWE 27 May 1970. MZUSP. 4543. Lepthalpheus axianassae Dworschak & Coelho, 1999:477, figs. 1-30 Paratypes.—1 @, 1 3, Praia do Araca, Sao Sebastiao, State of Sao Paulo, coll. V. R. Coelho and S. de A. Rodrigues, 9 Aug 1998, MZUSP 13010. Superfamily Processoidea Ortmann, 1890 Family Processidae Ortmann, 1890 Processa brasiliensis Christoffersen, 1979:364 Paratypes.—2 3, R/V Calypso, sta 77, 18°00’S, 38°18'’W, 48 m, 28 Nov 1961, MZUSP 4554. Infraorder Thalassinidea Latreille, 1831 Superfamily Callianassoidea Dana, 1852 Family Callianassidae Dana, 1852 Biffarius delicatulus Rodrigues & Manning, 1992a:324, figs. la—w Holotype.—s, Praia do Araga, Sao Se- bastiao, State of Sao Paulo, near mangrove, coll. S. de A. Rodrigues, 18 May 1985, MZUSP 10582. Callianassa guara Rodrigues, 1971:210, figs. 61—76 Holotype.—6, Guaruja, State of Sao Paulo, coll. T. K. S. Bjérnberg, Oct 1961, MZUSP 2729. Remarks.—Transferred to Sergio Man- ning & Lemaitre, 1994 by Manning & Le- maitre (1994:39). Callianassa guassutinga Rodrigues, 1971:204, figs. 41—60 Holotype.—6, Praia do Araga, Sao Se- bastiao, State of Sao Paulo, coll. S. de A. Rodrigues, Sep 1965, MZUSP 2728. 426 Remarks.—Transferred to Sergio Man- ning & Lemaitre, 1994 by Manning & Le- maitre (1994:39). Callianassa mirim Rodrigues, 1971:214, figs. 77-98 Holotype.—6, Baia de Santos, State of Sao Paulo, coll. S. de A. Rodrigues, Mar 1966, MZUSP 2730. Remarks.—Transferred to Sergio Man- ning & Lemaitre, 1994, by Manning & Le- maitre (1994:39). Eucalliax cearensis Rodrigues & Manning, 1992a:327, figs. 2a—v Holotype.—6, Barra do Rio Ceara, For- taleza, State of Ceara, coll. S. de A. Rodri- gues, 21 Feb 1984, MZUSP 10583. Lepidophthalmus sinuensis Lemaitre & Rodrigues, 1991:625, figs. 1—4 Paratypes.—5 2, Fazenda Agrosoledad, Cordoba, Colombia, nursery of Penaeidae, coll. S. de A. Rodrigues & R. Lemaitre, 20 Jul 1990, MZUSP 10751. Lepidophthalmus siriboia Felder & Rodrigues, 1993:367, figs. 2e—h, 4a—-f, 5a—g, 6a—| Holotype.—6, Rio Anil, Sao Luiz, State of Maranhao, coll. D. Felder and S. de A. Rodrigues, 18 Feb 1984, MZUSP 11083. Paratypes.—8 2, same data as holotype, MZUSP 11084. Poti gaucho Rodrigues & Manning, 1992b:10, figs. la—v Holotype.—&, State of Rio Grande do Sul, 33°43’S, 51°13'W, coll. S. de A. Rodri- gues, MZUSP 10581. Family Laomediidae Borradaile, 1903 Axianassa australis Rodrigues & Shimizu, 1992:317, figs. 1-20 Holotype.—é , Fazenda Maricultura, Val- enca, State of Bahia, near mangrove PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON swamp, coll. S. L. S. Bueno, Mar 1983, MZUSP 10584. Superfamily Axioidea Huxley, 1879 Family Axitidae Huxley, 1879 Eutrichocheles pindatyba Rodrigues & Kensley, 1991:556, figs. 1-16 Holotype.—<6, Santa Cruz, State of Es- pirito Santo, 19°57'S, 40°08'W, coll. M. L. Christoffersen, Mar 1979, MZUSP 10580. Remarks.—Transferred to Paraxiopsis De Man, 1905 by Kensley (1996:709). Family Micheleidae Sakai, 1992 Meticonaxius capricorni Coelho, 1987:64, figs. la—d, 2a, b, 3 Holotype.—&, R/V Alm. Saldanha, Proj Sul lil, sta 10, 23°527S, 43°11 We 27 3Milar 1972, MZUSP 7113. Paratype.—1 6, R/V Alm. Saldanha, Cabo Sao Tomé, State of Rio de Janeiro, sta 9, 11 Feb 1969, MZUSP 7114. Infraorder Anomura MacLeay, 1838 Superfamily Galatheoidea Samouelle, 1819 Family Aeglidae Dana, 1852 Aegla franca Schmitt, 1942:476, fig. 51, pl. 26d Paratypes.—3 ¢&, Franca, State of Sao Paulo, coll. E. Garbe, Oct 1910, MZUSP 622. Aegla inermis Bond-Buckup & Buckup, 1994:221, figs. 35a—e, 76a Paratype.— 1 2, Arr6io Kil6metro, trib- utary of Arrédio Cara, Osorio-Borussia, State of Rio Grande do Sul, coll. A. Rossi, 23 Jan 1979, MZUSP 11278 (Formerly UFRGS 345P). Aegla itacolomiensis Bond-Buckup & Buckup, 1994:212, figs. 3la—e, 75d Paratype.—1 ¢, Arrdio Demétrio, Fa- zenda Fialho, Taquara, State of Rio Grande do Sul, coll. G. Bond-Buckup, Mar 1989, MZUSP 11280 (Formerly UFRGS 881P). VOLUME 116, NUMBER 2 Aegla jarai Bond-Buckup & Buckup, 1994:229, figs. 40a—e, 72c Paratypes.— 2 6, Arroio Bandeirinha, Lages, State of Santa Catarina, coll. A. Ros- si, 22 Apr 1978, MZUSP 11279 (Formerly UFRGS 249P). Aegla ligulata Bond-Buckup & Buckup, 1994:216, figs. 33a—e, 74a Paratype.— 1 2, Arr6io Carvalho, Terra de Areia, State of Rio Grande do Sul, coll. IL, Ipwc uo, QD INOy IOV, IMAI WZ) (Formerly UFRGS 252P). Aegla microphthalma Bond-Buckup & Buckup, 1994:262, figs. 6la—e, 70c Holotype.—§, Caverna Santana, Iporan- ga, State of Sao Paulo, coll. E. Trajano, MZUSP 7405. Paratypes.—1 2, same data as holotype, MZUSP 7408; 2 3, 2 2, same data as ho- lotype, MZUSP 11274. Aegla perobae Hebling & Rodrigues, 1977:290, figs. la—e Holotype.—6, Caverna Peroba, Sao Pe- dro, State of Sao Paulo, 18 Jan 1973, MZUSP 4005. Paratypes.—12 3,4 @, same data as ho- lotype, MZUSP 4006. Aegla rosanae Campos Jr., 1998:137, figs. 1-4 Holotype.—é, head-water of cérrego Benfica, Bacia do Rio Paraiba, Piquete, State of Sao Paulo, coll. R.S. Lima, O. Tak- eshi and O. Campos Jr., 26 Jun 1992, MZUSP Vile 2: Remarks.—Considered a junior synonym of Aegla paulensis Schmitt, 1942:490 by Bond-Buckup & Buckup (2000:385). Aegla rossiana Bond-Buckup & Buckup, 1994:208, figs. 29a—e, 73a Paratypes.— 1 9, 2 36, Rio Jordao, Cri- ciuma, State of Santa Catarina, coll. L. 427 Buckup, 17 Sep 1977, MZUSP 11274 (For- merly UFRGS 230P). Aegla spinosa Bond-Buckup & Buckup, 1994:224, figs. 37a—e, 72d Paratypes— 1 3, 1 &, Rio Sado Joao, Campos Novos, km 310, State of Santa Ca- tarina, coll. A. Rossi, 22 Apr 1978, MZUSP 11276 (Formerly UFRGS 246P). Aegla violacea Bond-Buckup & Buckup, 1994:210, figs. 30a—e, 76b Paratypes.— 1 3, 1 ¢, Arroio do Cerro Negro, Guaiba, Mariana Pimentel, State of Rio Grande do Sul, coll. L. Buckup, D. Schossler and N. Fontana, 10 May 1988, MZUSP 11281 (Formerly UFRGS 1123P). Family Galatheidae Samouelle, 1819 Munida atlantica Melo Filho & Melo, 1994:50, figs. 1-7 Holotype.—&, Aracati, State of Ceara, R/V Canopus, sta 45, 04°14'S, 37°22’'W, 58 m, Aug 1965, MZUSP 11387. Munida brasiliae Coelho, 1973f:344 Paratypes.—5 3,3 2, R/V Alm. Saldan- ha, sta 1708, 02°44’S, 39°O1'W, 66 m, 22 Oct 1967, MZUSP 6611. Remarks.—Considered a synonym of Munida angulata Benedict, 1902 by Melo- Filho & Melo (2001:1143). Munida heblingi Melo-Filho & Melo, 1994:53, figs. 8-14 Holotype.—6, R/V Alm. Saldanha, Proj Leste I, sta 1953A, State of Espirito Santo, MZUSP 11388. Munida petronioi Melo Filho & Melo, 1994:55, figs. 15-21 Holotype.—é, R/V Alm. Saldanha, Proj Norte-Nordeste I, sta 1684B, 03°59’S, Soro em Se Oct 967. MZU SE 11389. 428 Munida victoria Melo-Filho, 1996:272, figs. 1-7 Holotype.—é, R/V Marion Dufrésne, Proj TAAF-55 BR, sta 55, 19°40’S, 38°43’W, 900 m, 30 May 1987, MZUSP 12256. Family Porcellanidae Haworth, 1825 Porcellana paivacarvalhoi Rodrigues da Costa, 1968c:405r Paratype.—1 6, Bairro Sao Francisco, Sao Sebastiao, State of Sao Paulo, MZUSP 8951. Remarks.—Considered a synonym of Porcellana platycheles Pennant, 1777 by Veloso & Melo (1993:171). Infraorder Brachyura Latreille, 1802 Section Dromiacea De Haan, 1833 Superfamily Dromioidea De Haan, 1833 Family Dromiidae De Haan, 1833 Dromia gouveai Melo & Campos Jr., 1999:281, figs. 3a, b Holotype.—<¢, Salvador, Geribotuba, State of Bahia, MZUSP 10222. Paratype.—|1 3d, Saco de Mamangua, Santos, State of Sao Paulo, MZUSP 5526. Superfamily Homoloidea De Haan, 1839 Family Latreilliidae Stimpson, 1858 Latreillia williamsi Melo, 1990b:28, meas, Il, Z Holotype.—é, R/V Prof. W. Besnard, Proj GEDIP, State of Rio Grande do Sul, sta 396, 34°26’S, 51°47'W, 155 m, sand and gravel, 27 Aug 1968, MZUSP 3295. Paratypes.—All same data as holotype: i SMA SR S492 2122) MZAUSP 5493.1 3d, MZUSP 5494. Section Eubrachyura de Saint Laurent, 1980 Subsection Raninoida De Haan, 1839 Superfamily Raninoidea De Haan, 1839 Family Raninidae De Haan, 1839 Ranilia guinotae Melo & Campos Jr., 1994:69, figs. 16-22 Holotype.—é, Santos, State of Sao Pau- lo, stomach content of the fish Diplectrum formosum Linnaeus, MZUSP 10932. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Paratype.—1\ 3, same data as holotype, MZUSP 11106. Ranilia saldanhai Rodrigues da Costa, ISWOS3, ie. Il Holotype.—é, R/V Alm. Saldanha, dredging, 26-60 m, 07°55'S, 34°48'W, MZUSP 8223. Remarks.—Considered a synonym of Ranilia muricata H. Milne Edwards, 1837 by Melo & Campos Jr. (1994:69). The ho- lotype is a male and not a female as listed by Rodrigues da Costa (1970:33). Superfamily Cyclodorippoidea Ortmann, 1892 Family Cyclodorippidae Ortmann, 1892 Clythrocerus analogus Coelho, 1973d:343 Paratype.—1 2, R/V Alm. Saldanha, Proj Norte-Nordeste, sta 1743A, 75 m, 00°51'S, 43°41'24"W, MZUSP 4828. Remarks.—Transfetrred to Deilocerus Ta- vares, 1996 by Tavares (1996:275). Clythrocerus bidentatus Campos Jr. & Melo, 1999:7; fies. 12 Holotype.—&, R/V Prof. W. Besnard, Proj Monitoramento, sta 28, 22°24’S, 40°42'W, 60 m, 09 Aug 1991, MZUSP 11077. Clythrocerus carinatus Coelho, 1973c:343 Paratypes.—1 3,1 2, R/V Alm. Saldan- ha, Proj Norte-Nordeste, sta 1804B, 27 m, sand, 25 Nov 1967, MZUSP 4827; 1 2°, R/V Alm. Saldanha, Proj Geomar, sta 27, 00°02'30"S, 46°15'00"W, 5 Jun 1968, MZUSP 4825; 1 @, R/V Alm. Saldanha, Proj Norte-Nordeste, sta 1763, O0°25’S, 47°17'W, MZUSP 12044. Cyclodorippe angulata Tavares, 1991:633, figs. 6b, 8c, |la—c Holotype.—¢&, R/V Marion Dufresne, Proj TAAF MD55-BR, sta 16, 310—360 m, 20°26'S, 36°41'W, 2 Jun 1987, MZUSP 10266. VOLUME 116, NUMBER 2 Cyclodorippe longifrons Campos Jr. & MeIOMIO99 385 lesa lee2 Holotype.—é, Proj Monitoramento, sta OS, ZAP 2OTSS AO WO WG TIS) tea, sy eee MZUSP 12055: Paratypes.—All same data as holotype: eSIMAUSP 120567 122 MZUSP 12057. Deilocerus coelhoi Campos Jr. & Melo, ONS NO), mss Ny 2 Holotype.—s, Proj Monitoramento, sta 57, 23°36'S, 41°56'W, 170 m, 20 Aug 1991, MZUSP W523: Paratypes.—All same data as holotype: SMA SP Ma524 I SaMAZAUSP tl522. Family Cymonomidae Bouvier, 1897 Cymonomus guillei Tavares, 1991:639, figs. 7b, 8d, 9b, 11d Holotype.—°, R/V Marion Dufresne, Proj TAAF MD55-BR, sta 54, 19°36’S, 38°53'W, 640 m, 30 May 1987, MZUSP 10268. Cymonomus guinotae Tavares, 1991:640, figs. 7c, 8b, 9c, 10a—c Holotype.-—&, R/V Marion Dufrésne, Proj TAAF MD55-BR, sta 64, 23°46’S, 42°10'W, 610 m, 2 Jun 1987, MZUSP 10269. Remarks.—Transferred to Cymonomo- ides Tavares, 1993 by Tavares (1993:141). Cymonomus magnirostris Tavares, 1991:635, figs. 7a, 8e, 9a, 1O0d—f Holotype.-—¢&, R/V Marion Dufresne, Proj TAAF MD55-BR, sta 64, 23°46’S, 42°10'W, 610 m, 2 Jun 1987, MZUSP LOZGTE Cymonomus meloi Campos Jr., 1997:129, fig. 1 Holotype.—é, Proj Integrado, Ubatuba, State of Sdo Paulo, MZUSP 10771. 429 Cymonomus oyakawai Campos Jr., OST s isle ine. 7 Holotype.—é, R/V Prof. W. Besnard, Proj GEDIP, State of Rio Grande do Sul, sta 1646, 34°25'S, 51°49’W, 166 m, 18 Jan 1972, MZUSP 6161. Cymonomus tavaresi Campos Jr., WOOD B25 sess Ss) Holotype.—?, R/V Emilia, Proj Ilha Grande, State of Rio de Janeiro, sta C. (288), 2 m, 30 Jul 1966, MZUSP 9193. Subsection Heterotremata Guinot, 1977 Superfamily Leucosioidea Samouelle, 1819 Family Leucosiidae Samouelle, 1819 Ebalia conica Coelho, 1973a:342 Holotype.—6, R/V Alm. Saldanha, sta 1784, off Cabo Norte, State of Amapa, 03°08’S, 49°07'W, 85 m, 16 Nov 1967, MZUSP 6603. Paratypes.—1 3,1 2, R/V Alm. Saldan- To, SAMOS OLAS) Bay LAW sy tangs 19/ Oct 1967) MZUSE 6551e 1 ORV Alm. Saldanha, sta 1676A, 04°46’S, 35°24'W, 39 m, 16 Oct 1967, MZUSP 6561. Remarks.—Transferred to Lithadia Bell, 1855 by Coelho & Ramos-Porto (1986:67). Ebalia obliqua Coelho, 1973b:342 Holotype.—és, North of State of Pernam- IDUICO, lero} IIMAS 37/ mms 17 darn WLYO2)- MZUSP 6544. Paratypes.—1 6, 2 carapaces, R/V Alm. Saldanha, Proj Geomar, sta 34, 00°21’S, 46°58'W, 30 m, 7 Jun 1968, MZUSP 6554; 1 3d juv, Proj Pernambuco, sta 27, 03 Feb 1969, MZUSP 6558; 1 3, Proj Pernambu- co, sta 35B, 07 Feb 1969, MZUSP 6556. Remarks.—Transfetred to Lithadia Bell, 1855 by Coelho & Ramos-Porto (1986:67). Ebalia vertiginosa Coelho, 1973e:343 Holotype.-—é, R/V Canopus, sta 14, OMFS OLSSRS S43 W,, 247 ems. 12 Jule 965, MZUSP 6559. 430 Paratypes.—1 2, R/V Akaroa, sta 37, 09°27'50"S, 35°17'45"W, 32 m, 6 Sep 1965, MZUSP 6549; 1 3, R/V Canopus, sta 61, O2°51'S, 38°51’ W, 60 m, 25 Aug 1965, MZUSP 6552; 1 2, R/V Akaroa, sta 59, 09°46'10"S, 35°34'40"W, 31 m, 7 Sep 1965, MZUSP 6555. Remarks.—Transferred to Lithadia Bell, 1855 by Coelho & Ramos-Porto (1986:67). Speloeophoroides capixaba Melo & Torres, 1998b:130, figs. 1-6 Holotype.—6, State of Espirito Santo, Proj Rio Doce, sta RD-54, 18°54’08’S, 39°15'04"W, 41 m, 1973, MZUSP 9149. Speloeophorus brasiliensis Melo & Torres, 1998a:126, figs. 7-10 Holotype.—1 6, Praia do Francés, Ma- rechal Deodoro, State of Alagoas, MZUSP 12045. Superfamily Majoidea Samouelle, 1819 Family Inachidae MacLeay, 1838 Podochela atlantica Coelho, 1997:223, figs. la—d Holotype.—éd, R/V Prof. W. Besnard, Proj GEDIP, State of Rio Grande do Sul, sta 554, 32°12’S, 50°12’W, 149 m, 9 Mar 1969, MZUSP 3533. Paratypes.—®? ovig., R/V Prof. W. Bes- nard, Proj GEDIP, State of Rio Grande do SW, SA SIO, 34° 2SS; SIDS WG WSS ian, 14 Mar 1969, MZUSP 3522; 1 @ ovig., R/V Prof. W. Besnard, Proj GEDIP, State of Rio Grande do Sul, sta 547, 30°48’S, 49°18’ W, 160 m, Jul 1969, MZUSP 5975. Podochela meloi Sankarankutty, Ferreira & Cunha, 2001:552, figs. 1, 2 Paratype.— 1 ¢&, Estuary near Macau, State of Rio Grande do Norte, MZUSP IBZ, Podochela minuscula Coelho, 1972:119, figs. Ib, c Paratype.—1 & ovig., R/V Alm. Saldan- ha, Proj Norte-Nordeste I, sta 1722, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 02°13.5'S, 40°43.5'W, 53 m, 29 Oct 1967, MZUSP 7213. Family Mithracidae Balls, 1929 Mithrax (Mithrax) besnardi Melo, IGoUas20)., ines, Il, 2 Holotype.—&, R/V Prof. W. Besnard, Proj GEDIP, sta 1871, off Maldonado, Uru- guay, 35°02'S, 52°42'W, 2100 m, 12 Aug 1972, MZUSP 4565. Paratypes.—All same data as holotype: 1 6, MZUSP 6241; 1 2, MZUSP 6242; 1 3d, MZUSP 7519. Remarks.—The subgenus Mithrax was elevated to generic status in a revision of the genus Mithrax by Wagner (1990). How- ever, it appears that Wagner was unable to consider M. (M.) besnardi as his revision, and Melo’s (1990a) description, appeared in the same year. Family Pisidae Dana, 1851 Libinia rostrata var. bellicosa Oliveira, 1944:87, figs. 1-3 Remarks.—A female specimen has been found from Ilha do Pinheiro, State of Rio de Janeiro, Brazil (MZUSP 8581), with similar information to that mentioned by Oliveira (1944) in his description of this taxon. Superfamily Parthenopoidea MacLeay, 1838 Family Parthenopidae MacLeay, 1838 Heterocrypta aloysioi Rodrigues da Costa, 1968a:144, figs. 1, 2 Holotype.—&, Praia do Araca, Sao Se- bastido, State of Sao Paulo, 3 Sep 1967, MZUSP 8224. Paratype.—1 ¢, same data as holotype, MZUSP 8225. Heterocrypta tommasii Rodrigues da Costay 19592595, ieee Holotype.—& ovig., Cananéia, State of Sado Paulo, 25°06’00”S, 47°51'05”"W, 26 Jul 1959, MZUSP 8226. VOLUME 116, NUMBER 2 Paratypes.—All same data as holotype: ll $, 2 2, MMUSIP S2278 Ils 25 Wu 8580. Remarks.—The holotype was originally deposited in IOUSP. Lambrus aylthoni Righi, 1965:57, fig. 1 Holotype.—é, Ilha Vitoria, State of Sao Paulo, coll. A. Joly, Jul 1963, MZUSP 1820. Paratypes.—1 6, Uha Vit6ria, State of Sao Paulo, 5 Jul 1959, MZUSP 2048; 1 &, Ilha de Alcatrazes, State of Sao Paulo, coll. IOUSP, 5 Oct 1959, MZUSP 2049. Remarks.—Transferred to Parthenope Weber, 1795 by Righi (1967:114). Solenolambrus brasiliensis Rodrigues da Cosia, IPollsil, ings. I, Z Holotype.—6é, State of Sao Paulo, 23°58'40"S, 46°80'30"W, MZUSP 8220. Paratypes.—All same data as holotype: cee VIS S22 iIe 2 MZ SP 8222. Superfamily Cancroidea Latreille, 1802 Family Cancridae Latreille, 1802 Cancer luederwaldti Rathbun, 1930a:200, pls. 86—89 Holotype.—Sex undetermined, Santos, State of Sao Paulo, coll. H. Ltiderwaldt, MZUSP 948. Remarks.—Synonymized under Cancer pagurus Linnaeus, 1758 by Rathbun G8 0b2578): Superfamily Portunoidea Rafinesque, 1815 Family Trichodactylidae H. Milne Edwards, 1853 Goyazana rotundicauda Magalhaes & Wir >> DDD DD > > > DD Table 1.—List of valid names of ‘‘Pylopagurus-Tomopagurus’ > group of taxa. Pylopagurus A. Mine-Edwards & Bouvier, 1891 sensu stricto . discoidalis (A. Milne-Edwards, 1880) . holmesi Schmitt, 1921 . pattiae Lemaitre & Campos, 1993 . macgeorgei McLaughlin & Lemaitre, 2001 gorei McLaughlin & Lemaitre, 2001 Lophopagurus (Lophopagurus) McLaughlin, 1981la . (L.) foresti McLaughlin & Gunn, 1992 . (L.) lacertosus (Henderson, 1888) . (L.) nanus (Henderson, 1888) . (L.) nodulosus McLaughlin & Gunn, 1992 . (L.) pumilus de Saint Laurent & McLaughlin (2000) . (L.) thompsoni (Filhol, 1885) Lophopagurus (Australeremus) McLaughlin, 1981la . (A.) cookii (Filhol, 1883) . (A.) eltaninae McLaughlin & Gunn, 1992 . (A.) cristatus (H. Milne Edwards, 1836) . (A.) kirkii (Filhol, 1883) . (A.) laurentae (McLaughlin & Gunn, 1992) . (A.) stewarti (Filhol, 1883) . (A.) triserratus (Ortmann, 1892) Rhodochirus McLaughlin, 1981la . rosaceus (A. Milne-Edwards & Bouvier, 1893) . hirtimanus (Faxon, 1893) Phimochirus McLaughlin, 1981la . operculatus (Stimpson, 1859) . californiensis (Benedict, 1892) . venustus (Bouvier, 1898) roseus (Benedict, 1892) randalli (Provenzano, 1961) . holthuisi (Provenzano, 1961) . occlusus (Henderson, 1888) . leurocarpus McLaughlin, 1981b Haigia McLaughlin, 1981a diegensis (Scanland & Hopkins, 1969) Agaricochirus McLaughlin, 1981la . boletifer (A. Milne-Edwards & Bouvier, 1893) . alexandri (A. Milne-Edwards & Bouvier, 1893) cavimanus (Chace, 1939) erosus (A. Milne-Edwards, 1880) gibbosimanus (A. Milne-Edwards, 1880) . hispidus (Benedict, 1892) echinatus McLaughlin, 1982 acanthinus McLaughlin, 1982 Anisopagurus McLaughlin, 1981a . bartletti (A. Milne-Edwards, 1880) . pygmaeus (Bouvier, 1918) vossi Lemaitre & McLaughlin, 1996 . actinophorus Lemaitre & McLaughlin, 1996 . hopkinsi Lemaitre & McLaughlin, 1996 VOLUME 116, NUMBER 2 Table 1.—Continued. 467 Manucomplanus McLaughlin, 1981la cervicornis (Benedict, 1892) spinulosus (Holthuis, 1959) longimanus (Faxon, 1893) varians (Benedict, 1892) ungulatus (Studer, 1883) SSS558 Enallopagurus McLaughlin, 1981a spinicarpus (Glassell, 1938) affinis (Faxon, 1893) coronatus (Benedict, 1892) provenzanol, new species Enallopaguropsis McLaughlin, 1981la guatemoci (Glassell, 1937) jJanetae McLaughlin, 1982 williamsi, new species Tomopagurus A. Milne-Edwards & Bouvier, 1893 rubropunctatus A. Milne-Edwards & Bouvier, 1893 cokeri (Hay, 1917) maclaughlinae Haig, 1976 cubensis (Wass, 1963) wassi McLaughlin, 1981la purpuratus (Benedict, 1892) merimaculosus (Glassell, 1937) . chacei (Wass, 1963) Protoniopagurus Lemaitre & McLaughlin, 1996 P. bioperculatus Lemaitre & McLaughlin, 1996 Pylopaguridium McLaughlin & Lemaitre, 2001 P. markhami McLaughlin & Lemaitre, 2001 Right cheliped with subovate or subrect- angular chela. Left cheliped with chela tri- angular in cross-section, but dorsal surface not elevated into prominent keel or crest. Dorsodistal spine of carpi of second and third pereopods lacking. Sternite of third pereopods with subcircular, subovate or subquadrate anterior lobe. Sternites of third to fifth pereopods often with capsulate se- tae. Fourth pereopod with propodal rasp consisting of single row of corneous scales; dactyl with small preungual process usually present. Males with paired gonopores, no distinct sexual tubes but occasionally with vas de- ferens protruded to form short tubular pa- pilla, usually on right; without paired ple- opods, with 3 uniramous or weakly bira- mous left pleopods on somites 3—5. Fe- males with paired gonopores, paired first pleopods modified as gonopods, 4 unpaired left pleopods, second to fourth with both rami well-developed, fifth usually unira- mous. Uropods symmetrical or asymmetrical. Telson without indication of division into anterior and posterior portions; terminal margin entire or with inconspicuous median indentation, unarmed. Remarks.—Faxon (1893, 1895: pl. 12, fig. 2e) reported that his single male spec- imen of Pylopagurus affinis (= Enallopa- gurus affinis) had the vas deferentia extrud- ed as “‘small threads’’ from both gonopores. McLaughlin (1982) reexamined Faxon’s (1893) specimen and concluded that while there were slight protrusions, these were more probably an artifact of preservation 468 rather than actual, very small sexual tubes. Since McLaughlin’s (1982) study, our in- formation on and understanding of sexual tubes in the Paguridae has increased sub- stantially. Among the “‘Pylopagurus-To- mopagurus’ group, species of Enallopa- guropsis and Agaricochirus can also have the vas deferentia protruded as tubular pa- pillae, albeit not as a distinct sexual tube. In other Paguridae, similar protrusions are known to occur in genera such as Discor- sopagurus McLaughlin, 1974, Paguritta Melin, 1939, and Pagurus (see de Saint Laurent 1970, McLaughlin & Lemaitre 1993, McLaughlin 2003). However, as re- ported by McLaughlin & Lemaitre (2001), short but distinct sexual tubes are present in some species of Pylopagurus sensu stric- to. Recently, McLaughlin & Jensen (1996) found a very small sexual tube developed on the right coxa of males of Parapaguro- des hartae McLaughlin & Jensen, 1996, and Komai (1998) provisionally transferred two Japanese species formerly assigned to Pagurus (P. gracilipes Stimpson, 1858 and P. nipponensis Yokoya, 1933), to Parapa- gurodes McLaughlin & Haig, 1973, be- cause of the observed small sexual tubes in males of both species. McLaughlin (1981a, 1982) included Fax- on’s (1893) Pylopagurus affinis in Enallo- pagurus, despite the fact that Faxon de- scribed the propodal rasp of the fourth pe- reopod as “‘multiserial”’. The type specimen has been reexamined, and Faxon’s descrip- tion and subsequent illustration (Faxon S95 spl ie 2d) pound stor besineenor: the propodal rasp of the fourth pereopod ac- tually has a single row of scales. This spe- cies is retained in Enallopagurus, although this generic assignment is still provisional given that E. affinis is known only from the male holotype. Until the discovery of the new species of Enallopagurus described herein, the genus had been considered exclusively an eastern Pacific taxon. Although the right cheliped of this new species lacks the operculate-like conformation seen in the other three species PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON assigned to the genus, the unarmed carpi of the second and third pereopods, the single row of scales on the propodal rasp of the fourth pereopod, and the presence in fe- males of paired first gonopods, support its placement in the genus. The terminal mar- gin of the telson in this new species of En- allopagurus has a minute median indenta- tion and is not clearly entire as described in the original generic diagnosis (McLaughlin 198la, 1982); thus, the need for the emen- dation. An updated key to the species of Enallopagurus is provided following the description of this new species. Enallopagurus provenzanoi, new species Jens. Il, 2 Holotype.—Ovig. 2 (sl = 2.40 mm), R/V John Elliott Pillsbury, sta 581, 21°05'N, 86°23'W, Arrowsmith Bank, 146— 265 m, 22 May 1967, USNM 1007525. Paratype.—1 6 in poor condition (sl = 1.80 mm), same data as holotype, USNM 1007526. Description.—Shield (Fig. la) longer than broad; anterior margin between ros- trum and lateral projections slightly con- cave; anterolateral margins sloping; poste- rior margin truncate, dorsal surface gla- brous. Rostrum triangular, terminating acutely. Lateral projections rounded, un- armed. Ocular peduncle approximately 0.65 length of shield, moderately stout, with cor- nea slightly dilated; ocular acicles subtrian- gular, each terminating subacutely, with moderately well-developed submarginal spine (not always visible in dorsal view); separated basally by approximately basal width of | acicle. Antennular peduncle overreaching ocular peduncle by 0.35—0.50 length of ultimate segment; ultimate and penultimate seg- ments unarmed; basal segment with small spine on lateral face. Antennal peduncle slightly shorter to equaling length of ocular peduncle. Fifth and fourth segments with scattered short se- VOLUME 116, NUMBER 2 469 < 2003: Errata Fumio Iwata. 2001. Nipponnemertes fernaldi, a new species of swim- ming monostiliferous hoplonemertean from the San Juan Archipelago, Washington, U.S.A.—Proceedings of the Biological Society of Washing- ton 114 (4): 833-857. Figure 4, page 842, due to author oversight figure captions were misplaced: Image for Fig. 4f appears above the Fig. 4h legend. Image for Fig. 4g appears above the Fig. 4f legend. Image for Fig. 4h appears above the Fig. 4g legend. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(2):529—530. 2003. BIOLOGICAL SOCIETY OF WASHINGTON 129th Annual Meeting, 23 May 2002 President-elect Roy W. McDiarmid called the meeting to order at 11:05 a.m. in the Waldo Schmitt Room, National Museum of Natural History. Council members present: W. Duane Hope, Michael Carleton (Elected Council), Richard Sternberg (Editor), Stephen Cairns and Frank D. Ferrari (Finance Committee), Storrs L. Olson (Custodian of Publi- cations), Richard Banks, Bruce Collette, Kristian Fauchald, David Pawson, Janet Reid, Richard Vari (Past Presidents), T. Chad Walter (Treasurer), Carole C. Baldwin (Secretary). Council members absent: Susan L. Jewett, G. David Johnson, Rafael Lemaitre, Jon No- renburg. McDiarmid announced that due to illness, President Kensley would not be presiding over the meeting. McDiarmid noted that the Society had dealt with a number of difficult issues during the past year, including the replacement of long-time Proceedings Editor Brian Robbins and mail problems related to anthrax contamination of a postal facility in Washington, D.C. Minutes of the 128th Annual Meeting of the Society were approved, and McDiarmid then called on Chad Walter for the Treasurer’s Report (Table 1). Society income for the period 1 January 2001 to 31 December 2001 was $126,129.06; expenses for the same period were $178,033.71. Total assets for the Society as of 15 April 2002 were $107,599.72, the Society’s endowment account declining by $7,942.12 because of stock- market fluctuations. Additionally, because of disruption to mail service at the Smithsonian Institution following the closure of the Brentwood mail facility after anthrax contamina- tion, $24,953.66 was withdrawn from the Society’s endowment fund to cover costs of publishing the Proceedings. By the end of April, all of the borrowed funds had been returned to the endowment account. Stephen Cairns noted that the Audit Committee had found the Treasurer’s records to be in good order. The new Editor of the Proceedings, Richard Sternberg, reported that his tenure began 1 January 2002, and that outgoing Editor Brian Robbins had handled the publication of Volume 114 and the first issue of Volume 115. Sternberg reported that four issues of Volume 114 were published comprising 85 papers and 988 pages. In addition, 31 papers and 363 pages were published in Bulletin No. 10 (S. D. Cairns and C. G. Messing, Editors), and a new Guidelines for Manuscripts (C. B. Robbins and D. B. Lellinger) was published as a supplement to Volume 114(4). As of 22 May 2002, there were 42 sub- missions for the year, up from 33 in May 2001; however, Sternberg noted that many of those 42 submissions were postmarked in the Fall of 2001. There continues to be no backlog for papers accepted in the Proceedings. McDiarmid announced that the Council had voted to adopt four recommendations of the Finance Committee: (1) Increase cost of reprints by approximately 20%; (2) increase cost of library subscriptions from $40 to $50 (which still keeps the price of the Proceed- ings in the lowest five percentile of scientific journals); (3) re-invest $55,000 of the So- ciety’s endowment funds into the American Funds Investment Company of America; and (4) establish gift-fund categories and list benefactors on a front page of the Proceedings. These measures are designed to help the Society reach a financial goal of $200,000 in the endowment fund. McDiarmid further noted that the Council had agreed to a vote at 530 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON next year’s annual meeting on a change suggested by the Finance Committee in By-Law No. 8, which establishes and describes the Society’s endowment fund. The proposed change would remove “‘and surplus funds from operations” from the first sentence of By- Law 8, and the amended first sentence would read as follows: ‘“Endowment Fund. There shall be an Endowment Fund which shall consist of gifts from members and miscellaneous gifts.” The Finance Committee (Stephen Cairns, Frank Ferrari, Oliver Flint, and Chad Walter, with assistance from Richard Banks) was thanked for its extended efforts on behalf of the Society. In view of the very low attendance by members at the annual meeting of the Society, Bruce Collette suggested a single annual meeting in the future that combines the Council meeting and the annual meeting; a decision regarding a single meeting will be made after an Executive Committee examines the issue in more detail. Results of the 2002 Election of Officers were then announced by President-elect McDiarmid: Rafael Lemaitre (President-elect), Carole Baldwin (Secretary), Chad Walter (Treasurer), Michael Carleton, G. David Johnson, Clyde Roper, Marilyn Schotte, Michael Vecchione, and Don Wilson (Elected Council). Assuming his new role as President, McDiarmid then thanked outgoing President Brian Kensley and adjourned the meeting. Respectfully submitted, Carole C. Baldwin Secretary Summary Financial Statement for 2001 General Endowment Total Fund Fund Assets Assets: January 1, 2001 29,028.46 SIU 3Z.ID WU TOL 2S Total Receipts for 2001 112,675.16 13,453.90# 126,129.06 Total Disbursements for 2001 156,637.69 (21,396.02)° 178,033.71 Assets: December 31, 2001 18,099.66 75,790.67 93,890.33 Net Changes in Funds (10,928.80) (7,942.12) (18,870.92) ‘annual gain in value of Endowment >’ annual loss in value of Endowment INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Richard v. Sternberg, NCBI—GenBank, Building 45, Room 6An. 18D-30, National Institutes of Health, Bethesda, MD 20892-6510. (Do not submit manu- scripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in “GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON,” a supplement to Volume 114, number 4, December 2001. Authors are encouraged to consult this article before manuscript preparation. Copies of the article are avail- able from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page charges. The Society is a non-profit organization with limited funds, depending on financial resources and upon author request, it may subsidize a limited number of pages per vol- ume at the discretion of the editor and treasurer. Payment of full costs is encouraged and will facilitate speedy publication. Please note: Page charges are not automatically waived with membership. Authors requesting a partial waiver of page charges must justify need by emailing the editor and treasurer before proofs are returned and on the reprint request form received with page proofs. Please include email address on all correspondence. Costs.—Printed pages @ $65.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. Front cover—from this issue, p. 343. CONTENTS Notes on the genus Ruellia (Acanthaceae) in Bolivia, Peru and Brazil D. C. Wasshausen and J.R.I. Wood 263 Phylogeography and systematic notes on two species of gracile mouse opossums, genus Gracilinanus (Marsupialia: Didelphidae) from Brazil Leonora Pires Costa, Yuri L. R. Leite, and James L. Patton 275 Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 10. Cyanomyia salvini Brewster, 1893, is an intergeneric hybrid of Amazilia violiceps and Cynanthus latirostris Gary R. Graves 293 A new species of caracara (Milvago) from Quaternary asphalt deposits in Cuba, with notes on new material of Caracara creightoni Brodkorb (Aves: Falconidae) William Suarez and Storrs L. Olson 301 A new species of penguin (Spheniscidae: Spheniscus) and other birds from the late Pliocene of Chile Steven D. Emslie and Carlos Guerra Correa 308 Leptodactylus caatingae, a new species of frog from eastern Brazil (Amphibia: Anura: Leptodactylidae) W. Ronald Heyer and Flora A. Junca 317 A review of Amblyotrypauchen (Teleostei: Gobiidae), a genus of blind amblyopine gobies Edward O. Murdy 330 The hyopalatine arch of a 25 mm larva of Synbranchus and homology of the single pterygoid in the Synbranchidae (Teleostei: Synbranchiformes) Ralf Britz, Sandra Favorito, and G. David Johnson 33) Skull osteology of the characid fish Astyanax mexicanus (Teleostei: Characidae) Martha E. Valdéz-Moreno and Salvador Contreras-Balderas 341 Geographic variation in Poecilia Bloch and Schneider, 1801 (Teleostei: Poeciliidae), with descriptions of three new species and designation of lectotypes for P. dovii Giinther, 1866 and for P. vandepolli van Lidth de Jeude, 1887 Fred. N. Poeser 356 On the occurrence of the vestimentiferan tube worm Lamellibrachia luymesi van der Land and N@¢rrevang, 1975 (Annelida: Pogonophora) in hydrocarbon seep communities in the Gulf of Mexico Stephen L. Gardiner and Stéphane Hourdez 380 Four new genera of leucosiid crabs (Crustacea: Brachyura: Leucosiidae) for three new species and nine species previously in the genus Randallia Stimpson, 1857, with a redescription of the type species, R. ornuta (Randall, 1939) Bella S. Galil 395 Type catalogue of the Crustacea Decapoda in the collections of the Museu de Zoologia da Universidade de Sao Paulo, Brazil Gustavo Augusto Schmidt de Melo, Renata M. Vezzani, and Oswaldo Campos Jr. 423 Establishment of a new genus for Panopeus bermudensis Benedict & Rathbun, 1891 and several other xanthoid crabs from the Atlantic and Pacific oceans (Crustacea: Decapoda: Xanthoidea) Darryl L. Felder and Joel W. Martin 438 A new species of Pseudopaguristes McLaughlin, 2002 (Crustacea: Decapoda: Diogenidae) from Japan Akira Asakura and Patsy A. McLaughlin 453 Revision of Pylopagurus and Tomopagurus (Crustacea: Decapoda: Paguridae), with descriptions of new genera and species. Addendum and taxonomic summary Rafael Lemaitre and Patsy A. McLaughlin 464 A new species of the genus Periclimenes Costa, 1844 (Crustacea: Decapoda: Palaemonidae) from the Ryukyu Islands, southern Japan Junji Okuno and Masako Mitsuhashi 487 A new genus and species, Tumidochelia randyi, from the Gulf of Mexico (Crustacea: Peracarida: Tanaidacea) Julianne Knight, Kim Larsen, and Richard Heard 497 A new cavernicole species of the genus Anelpistina (Insecta: Zygentoma: Nicoletiidae) from the Guanica subtropical dry forest, Puerto Rico Luis Espinasa and Monika Baker Alpheis 503 A second species of the rare milliped family Apterouridae (Diplopoda: Chordeumatida: Striarioidea) William A. Shear 509 Sponges of the family Chondrillidae (Porifera: Demospongiae) from the Pacific coast of Mexico, with the description of three new species José Luis Carballo, Patricia G6mez, José Antonio Cruz-Barraza, and Dulce Ma. Flores-Sanchez 515 Errata 528 Biological Society of Washington, 129th Annual Meeting, 23 May 2002 529 SMITHSONIAN INSTITUTION LIBRARIES 3 9088 01205 2486 th PROCEEDINGS of TH: Bi OGICAL SOCIETY or WASHINGTON oO HOON f; : “aN 2 DECEMBER 2003 At AN 07 2004 | GRR IES VOLUME 116 NUMBER 3 THE BIOLOGICAL SOCIETY OF WASHINGTON 2002-2003 Officers President: Roy W. McDiarmid Secretary: Carole C. Baldwin President-elect: W. Ronald Heyer Treasurer: T. Chad Walter Elected Council Michael D. Carleton G. David Johnson Clyde Roper Michael Vecchione Marilyn Schotte Don Wilson Custodian of Publications: Storrs L. Olson PROCEEDINGS Editor: Richard v. Sternberg Associate Editors Classical Languages: Frederick M. Bayer Invertebrates: Stephen L. Gardiner Plants: Carol Hotton Christopher B. Boyko Insects: Wayne N. Mathis Janet W. Reid Vertebrates: Gary R. Graves Invertebrate Paleontology: Gale A. Bishop Carole C. Baldwin Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Annual dues are payable on or before January 1 of each year. Renewals received after January | must include a penalty charge of $5.00 for reinstatement. Library subscriptions to the Proceedings are: $50.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an additional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership, subscriptions, and/or back issues should be sent to: BIOLOGICAL SOCIETY OF WASHINGTON P.O. BOX 1897 LAWRENCE, KANSAS 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, and editorial questions should be sent to: EDITOR, RICHARD v. STERNBERG NATIONAL CENTER FOR BIOTECHNOLOGY INFORMATION—GENBANK BUILDING 45, ROOM 6An. 18D-30 NATIONAL INSTITUTES OF HEALTH BETHESDA, MD 20892-6510 Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Periodicals postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):531—541. 2003. Rediscovery, systematic position, and re-description of **Leptoxis”’? melanoides (Conrad, 1834) (Mollusca: Gastropoda: Cerithioidea: Pleuroceridae) from the Black Warrior River, Alabama, U.S.A. Russell L. Minton, Jeffrey T. Garner, and Charles Lydeard (RLM) Department of Zoology, Field Museum of Natural History, 1400 S. Lake Shore Drive, Chicago, Illinois 60605, U.S.A., e-mail: rminton @ fieldmuseum.org; (JG) Alabama Fish and Game Division, PO. Box 366, Decatur, Alabama 35602, U.S.A.; (CL) Department of Biological Sciences, University of Alabama, Box 870345, Tuscaloosa, Alabama 35487, U.S.A Abstract.—The rediscovery of Leptoxis melanoides, a pleurocerid snail from the Black Warrior River drainage of Alabama, previously considered to be extinct, 1s presented. The radula of L. melanoides more closely resembles that of Elimia, and an analysis of 16S rDNA sequences positions the species within a monophyletic clade of Elimia. Therefore, we re-describe the species as E. melanoides, designate a neotype, precise type locality, and illustrate the spec- imen and its radula. The Mobile River basin, which drains much of Alabama and parts of Georgia, Mississippi, and Tennessee, 1s not only one of the largest drainage systems east of the Mississippi River (70,252 km’; Lydeard et al. 1997), but was also home to more fresh- water gastropod taxa than anywhere else in the world, accounting for 118 species in nine families (Bogan et al. 1995). Unfor- tunately, based on recent surveys and lit- erature records, 38 (32%) of the gastropod species in the basin are now presumed ex- tinct and many others have exhibited a marked decline in their distribution. One family that makes up a large component of the aquatic gastropod fauna in the basin is the Pleuroceridae. Pleurocerids are freshwater, gill-breath- ing, operculate snails that reach their high- est diversity in the southeastern United States (Burch & Tottenham 1980, Lydeard & Mayden 1995, Neves et al. 1997). Un- fortunately, 25 pleurocerid species and one genus (Gyrotoma) are now presumed ex- tinct due to the inundation of shoal areas by impoundment and habitat degradation from poor land use practices (Stein 1976, Bogan et al. 1995, Lydeard & Mayden 1995, Ne- ves et al. 1997). Besides Gyrotoma, the ge- nus Leptoxis has fared equally poorly in re- cent times, as 11 of 15 Leptoxis species in the Mobile River system are presumed ex- tinct. Eight of the presumed 11 species were restricted to the main channel of the Coosa River of Alabama (Bogan et al. 1995), which is now a highly modified and regulated system (Hershler et al. 1990). Of the three other species, L. compacta was re- stricted to the middle portion of the Cahaba River, which, although lacking major dams, was and remains subject to high environ- mental stress from urban development and activity near the city of Birmingham. An- other presumed extinct species, L. melanot- des, was restricted to the Black Warrior River of Alabama. Recent surveys of the Coosa (Bogan & Pierson 1993a, J. Godwin, pers. comm.) and Cahaba (Bogan & Pierson 1993b) Rivers failed to find any of the pre- sumed extinct Leptoxis species. In 1996, the authors conducted a gastro- pod survey in one of the few remaining 87.2°W 34.3° N 33.5° N 87.2° W Isis, Il. map shows individual sites in Locust Fork portion. Closed triangles indicate those sites where Leptoxis melanoti- des was present. free-flowing rivers in Alabama, the Locust Fork of the Black Warrior River. The main objective of the study was to determine the distribution of the third species, L. plicata, now restricted to the Locust Fork. During this survey, we rediscovered putative L. melanoides. This paper reviews the taxo- nomic and conservation history of L. mela- noides, examines the historical and recent distribution of the species, and evaluates its phylogenetic position based on an analysis of mitochondrial 16S rDNA sequences. In addition, causes for the evident contraction of the distribution of the species are pro- vided, along with a complete re-description of the species. Methods During the survey, 57 sites were visited in the Locust Fork drainage of the Black PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 86.2° W Map of sites visited in this study. Upper map shows location of Black Warrior River drainage, lower Warrior River in Blount, Etowah, Jefferson, and Marshall Counties (Fig. | & Appendix I). Leptoxis melanoides was found living in a total of seven localities in Blount County, five in the Locust Fork proper, and two in the Little Warrior River. Five individuals of L. melanoides from the Little Warrior River in Blount County, Alabama were collected in 2002 and used in this study. Radulae were prepared and visualized using the methods of Holznagel (1998) and compared to other pleurocerid radulae as figured in Holznagel (2000). Genomic DNA was isolated from head tissues using standard phenol/chloroform methods. Sequences were obtained for an amplified segment of the mitochondrial large ribosomal subunit (16S) gene using the SR14231/SNLO0O2 and H3080/L2510 VOLUME 116; NUMBER 3 primer pairs of Holznagel & Lydeard (2000) and Palumbi et al. (1991) respec- tively. The primers amplify an approxi- mately | kb fragment of the gene and were used for amplification and sequencing. Double-stranded amplifications via PCR were generated using 50—500 ng of tem- plate genomic DNA in 25 wl volumes (10 mM Tris, 50 mM KCl, 2.5 mM MgCl, 1 uM each primer, 0.1 mM each dNTP, 1.5 units Taq DNA polymerase). The amplifi- cation regime consisted of 35 cycles of de- naturation (92°C for 40 sec), annealing (62°C for 60 sec), and extension (/2°C for 90 sec). Double-stranded products were pu- rified using the Qiagen PCR Cleanup kit and provided the template for cycle se- quencing using the ABI BigDye 2.0 kit fol- lowing manufacturer’s instructions. Reac- tions were purified using Qiagen DyeEx spin columns and sequenced on an ABI 3100 genetic analyzer. Sequences of L. melanoides were entered in the software program BioEdit (version 5.0.9; Hall 1999), along with the 35 pleu- rocerid taxa used in Holznagel & Lydeard (2000), who provided the first phylogenetic survey of the Pleuroceridae (Appendix II). Sequences were aligned by eye to the sec- ondary structure models suggested by Holz- nagel & Lydeard (2000) and Lydeard et al. (2002). Phylogenetic hypotheses were gen- erated in PAUP* 4.0b10 (Swofford 2002) under maximum parsimony using 50 repli- cates of heuristic search with random ad- dition. The following options were em- ployed: uninformative characters were ig- nored, branches with a minimum Zero length collapsed, and minimal length trees kept. To test the internal stability of the data, a jackknife analysis (Farris et al. 1996; 1000 replicates, 10 random additions per replicate) was performed using PAUP* and Bremer decay indices (Bremer 1994) were determined in SEPAL (Salisbury 2001). Holznagel & Lydeard (2000) showed sig- nificant phylogenetic signal and no evi- dence of base composition bias in their analysis. They also reported no TS/TV sat- 3533 uration up to near 20% genetic distance, so no weighting schemes were employed. Specimens from the following museums were examined in this study: FMNH—Field Museum of Natural History, Chicago; FLMNH—Florida Museum of Natural His- tory, Gainesville; NCSM—North Carolina State Museum of Natural Sciences, Ra- leigh; UMMZ—University of Michigan Museum of Zoology, Ann Arbor. Results Radulae taken from putative Leptoxis melanoides most resemble Elimia species, and not Leptoxis species (Fig. 2; see below and Fig. 4 for complete description). Lep- toxis melanoides radulae have more rect- angular laterals and more lateral cusps than other Leptoxis species, though the main lat- eral cusps are larger than those normally seen in Elimia. Also in L. melanoides, den- ticles are narrower and more numerous on the inner marginals. Aligned sequences resulted in a matrix of 900 characters, including indels, of which 318 were parsimony informative. Maxi- mum parsimony analysis yielded a single tree (1338 steps, CI = 0.52; Fig. 3). All five specimens of L. melanoides shared a single sequence and were resolved in a terminal clade of Elimia taxa and not in the poly- phyletic groups of Leptoxis at the base of the tree. Jackknife values and Bremer in- dices strongly supported the resolution. Systematics Family Pleuroceridae Fischer, 1885 Genus Elimia Adams & Adams, 1854 (see Burch 2001) Elimia melanoides (Conrad, 1834) Fig. 4 Anculosa melanoides Conrad 1834:64, fig. 19% Leptoxis melanoides Haldeman 1843-1853: 5, figs. 145-146. Nitocris melanoides Adams & Adams 1858:308. 534 Fig. 2. D. Elimia showalteri. Anculosa turgida Haldeman 1840:2. Leptoxis turgida Haldeman 1843—1853:5, ny, Sle Type designation and locality.—Neotype FMNH 301993, Little Warrior River, dirt road bridge crossing off County Highway 15, RIE, T13S, sec. 30, Blount County, Al- abama, U.S.A, here selected, in accordance with article 75.3.4 of the International Code of Zoological Nomenclature (1.C.Z.N. 1999). Neoparatypes FMNH 301889 (4 specimens). Representative radulae from pleurocerids. A. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Leptoxis ampla. B. Elimia sp. C. Elimia alabamensis. Material examined (all in Alabama).— FMNH 33749, Black Warrior River; FMNH 57344, Black Warrior River; FMNH 76034, Black Warrior River; FMNH 80200, Warrior River, Jefferson County; FMNH 80243, Warrior River, Jef- ferson County; FLMNH 18219, Black War- rior River; FLMNH 81398, Locust Fork of Warrior River, Trafford, Jefferson County; FLMNH 81399, Black Warrior River, the forks; FLMNH 81400, forks of Warrior River, Walker County; FLMNH 81401; VOLUME 116; NUMBER 3 335 Melanoides tuberculata Melanopsis praemorsa are J. bulbosa 25 1.99 J. silicula 8 J. plicifera : ~Le. picta 93 Le. plicata 10 Le. praerosa cf. Li. armigera Le. ampla Le. taeniata Le. virgata Le. crassa anthonyi lo fluvialis Li. duttoniana Li. geniculata fuliginosa Li. geniculata geniculata P. acuta acuta P. unciale hastatum P. prasinatum 1 2 {100 40 100 15 100] 13 1100 1 23 100 5 8 1 88 7 100 10 1 85 2 P. prasinatum 2 P. canaliculatum filum P. walkeri P. vestitum 86 P. annuliferum “ P. pyrenellum E. virginica 99 6 99 6 68 Bigs: E. hydei E. crenatella E. showalteri E. caelatura Leptoxis melanoides = a E. cylindracea 3174 E. alabamensis 1 [90 E. olivula z E. haysiana The single most parsimonious tree generated using 16S rDNA sequences (1338 steps, CI = 0.52). Taxa historically considered Leptoxis are shaded. Values above branches are jackknife support, below are Bremer decay indices. 536 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Elimia melanoides. A. Neotype FMNH 301933, 10.8 mm. B. Rachidian and laterals. C. Inner and outer marginals. Scale bars: B, C = 50 wm. FLMNH 82143, Locust Branch, Warrior River, Trafford; FLMNH 82144, Black Warrior River, Squaw Shoals, Jefferson County; FLMNH 82145, Warrior River; FLMNH 82146, Black Warrior River; FLMNH 82147, forks of Warrior River, Walker County; FLMNH 82148, Black Warrior River; FLMNH 82149, Warrior River, lock 15, Tuscaloosa County; FLMNH 82150, Black Warrior River, Tus- caloosa, Tuscaloosa County; FLMNH 82151, Valley Creek, Toadvine, Jefferson County; FLMNH 82152, Black Warrior River; FLMNH 82153, Black Warrior Riv- er, Squaw Shoals, Jefferson County; FLMNH 230087, Sipsey Fork, Bankhead National Forest N.E 234, Winston County; FLMNH 230131, Sipsey Fork, Bankhead National Forest N.E 234, Winston County; FLMNH 230766, Locust Fork, at Black Warrior River, Jefferson County; FLMNH 230803, Sipsey Fork, above Alabama High- VOLUME 116, NUMBER 3 way 33, Winston County; UMMZ 10169; UMMZ 37463, Alabama River; UMMZ 37464, Warrior River; UMMZ 37465, Lo- cust Fork, Black Warrior River, Jefferson County; UMMZ 37466, forks of Black Warrior River, Walker County; UMMZ 49356, Black Warrior River, near mine, Jef- ferson County; UMMZ 55796 Locust (?) bridge of the Black Warrior River, Blount County; UMMZ 66275, Black Warrior Riv- er, Squaw Shoals, Tuscaloosa County; UMMZ 133928, Upper Black Warrior Riv- er; UMMZ 185760, Warrior River. Diagnosis.—Shell small relative to sym- patric pleurocerids, ovately conic and un- sculptured, with even sutures. Shell more conic and aperture more ovate than typical Leptoxis. Columella white, frequently with pink to purple tinge. Distribution limited to Black Warrior River drainage, Blount County, Alabama. Description.—Shell to 13 mm length, thin, ovately conic and smooth. Some early whorls and apex eroded. Fine transverse growth lines visible. Sutures flat to slightly impressed. Aperture oval, about one-half the length of the shell. Columella smooth and white, tinged with pink, red, or violet. Shell color green to brown, occasionally with brownish transverse bands. Operculum thin, dark, and oval, impressed at occasion- al growth lines (see Goodrich 1922, pl. 3, fig. 28). Radula rachidian tooth with nine triangular cusps, central cusp largest, slight- ly longer than wide. Lateral tooth broad, top edge straight with five cusps. Central cusp rectangular, broader and longer than others. Inner marginals with six to eight denticles, all equal size except for outer denticle on each side. Outer marginals with 12—14 den- ticles, all equal size except for outer denti- cle on each side. Distribution.—Historically included the upper half of the Black Warrior River drain- age, from Tuscaloosa and above. Currently restricted to stretches of the Locust Fork of the Black Warrior River and the Little War- rior River, Blount County, Alabama. Remarks.—Conrad’s (1834) original de- 587 scription of Anculosa melanoides was “Shell conical, with three entire volutions; apex eroded; whorls flattened, rounded only at the sutures; lines of growth prominent; body-whorl abruptly rounded; epidermis blackish, obscurely banded; aperture ellip- tical, about half the length of the shell’’ and the type locality, rivers of North Alabama, was ambiguous. Tryon (1873) reproduced Conrad’s (1834) original description of A. melanoides and provided three figures of the species. Goodrich (1922) included an observation by H. H. Smith, who doubted that A. melanoides belonged in Anculosa. Goodrich (1941) later suggested that A. melanoides more closely resembled north- ern species in the genus and that its taxo- nomic position may change when its radula is obtained. A neotype designation was re- quired, as the type specimen, once “‘in the possession of Mr. Anthony” (Tryon 1873: 400), was not available to Goodrich (1922), could not be located at the Museum of Comparative Zoology (Graf 2001) or the Academy of Natural Sciences in Philadel- phia (Baker 1964), and has apparently been lost. The type locality has also been sub- sequently restricted to the drainage where the species is currently extant, following recommendation 76A of the code, as the original type locality could include the Ten- nessee River drainage, where the species does not occur. Discussion The rediscovery of putative Leptoxis melanoides was important and exciting, es- pecially given the overall status of both many North American pleurocerid species and especially those in the Mobile River ba- sin. This study helps to illustrate the need for thorough survey work combined with systematic analysis to rediscover and iden- tify potentially “‘lost” or extinct taxa. Stein (1976:31) reviewed the conservation status of Alabama gastropods and stated “*. . . it is doubtful whether living populations of this species still exist.”’ In a subsequent report 538 on rare and endangered invertebrates, Har- ris (1990) listed L. melanoides as endan- gered without evidence to suggest it still ex- isted. Several formal and informal surveys had been conducted on various portions of the Black Warrior River and Locust Fork by U.S. Fish and Wildlife personnel and other investigators, but none documented the presence of L. melanoides. Given that the species had not been documented in over 50 years, Bogan et al. (1995) listed it as presumed extinct, a decision maintained by Turgeon et al. (1998). Comparisons of our material with museum specimens and cited illustrations support our notion of re- discovering L. melanoides in the Black Warrior River. Of the 57 sites visited in the Locust Fork of the Black Warrior River drainage, L. melanoides was found at seven sites in two disjunct areas. Although it is possible the species may be more or less continuously distributed between the two areas, it still represents a small portion of the river sys- tem. It 1s evident, however, that based on historical distribution the species has de- clined drastically since the studies of Good- rich in the 1920’s and 1940’s. Blount Coun- ty is home to poultry farms and coal mines which undoubtedly impact the surrounding water quality. Analysis of mitochondrial 16S rDNA se- quences suggests that L. melanoides should be placed in the genus Elimia, as it is re- solved in a well-supported monophyletic clade of Elimia from the Mobile River ba- sin. Such proposed generic reassignments are not uncommon in light of phylogenetic studies of freshwater mollusks (e.g., Ly- deard et al. 2000, Minton & Lydeard 2003). The observed radulae are consistent with other species of Elimia (Holznagel 2000), and combined with shell morphology, make this species diagnosable in the upper reach- es of the Locust Fork drainage. A redescrip- tion of what we are now considering Elimia melanoides, along with redesignation of type material, was needed given the pre- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON sumed loss of the holotype and the phylo- genetic placement resolved in this study. Given its relative rarity and restricted dis- tribution, and the overall decline of North American pleurocerids, remaining popula- tions of E. melanoides should be monitored. Efforts are underway to list the species as endangered with the U.S. Fish and Wildlife Service. Until formally listed, the species should be treated as endangered and efforts made to preserve its remaining habitat. Acknowledgments We thank D. Palmer, D. Pylant, W. Rake- straw, and P. West for their help in the field, with special thanks to D. Palmer for boat use in surveying the Locust Fork. Assis- tance with micrograph preparation was pro- vided by J. Nunley at The University of Al- abama. A. Bogan and P. Johnson provided helpful comments. Molecular work was done in the Pritzker Laboratory for Molec- ular Systematics and Evolution at the Field Museum, and the Johnson Molecular Sys- tematics Laboratory at the University of Al- abama. Some collecting effort was done with students as part of a field course of- fered at the University of Alabama’s Gads- den Education and Research Center. Partial support was provided by the U.S. Fish and Wildlife Service and the National Science Foundation. Literature Cited Adams, H., & A. Adams. 1853—1858. Genera of recent mollusca arranged according to their organiza- tion. Three volumes. John van Voorst, Pater- noster Row, London, 484 pp., 661 pp., 138 pl. Baker, H. B. 1964. Type land snails in the Academy of Natural Sciences of Philadelphia. Part III. Limnophile and thalassophile Pulmonata, part IV. Land and fresh-water Prosobranchia.—Pro- ceedings of the Academy of Natural Sciences of Philadelphia 116:149—193. Bogan, A. E., & J. M. Pierson. 1993a. Survey of the aquatic gastropods of the Coosa River Basin, Alabama: 1992. Final Report. Submitted to Al- abama Natural Heritage Program, Montgomery, Alabama, 13 pp. , & . 1993b. Survey of the aquatic gas- VOLUME 116, NUMBER 3 tropods of the Cahaba River Basin, Alabama: 1992. Final Report. Submitted to Alabama Nat- ural Heritage Program, Montgomery, Alabama, 20 pp. , & P. Hartfield. 1995. Decline in the freshwater gastropod fauna in the Mobile Basin. Rpe249— 252 Ea akoerG-7S> Farms; GxE: Puckett, P D. Doran, & M. J. Mac, eds., Our living resources: a report to the Nation on the distribution, abundance, and health of USS. plants, animals, and ecosystems. U.S. Depart- ment of the Interior, National Biological Ser- vice, Washington D.C., 530 pp. Bremer, K. 1994. Branch support and tree stability. — Cladistics 10:295—304. Burch, J. B. 2001. On the genus name Goniobasis (Eli- mia: Gastropoda: Pleuroceridae) and other re- cent nomenclatural inconsistencies.—Walker- ana 12:97—105. , & J. L. Tottenham. 1980. North American freshwater snails. Species list, ranges and illus- trations.—Walkerana 1:81—215. Conrad, T. A. 1834. New freshwater shells of the Unit- ed States, with coloured illustrations, and a monograph of the genus Anculotus of Say; also a synopsis of the American naiades. Judah Dob- son, Philadelphia, Pennsylvania, 76 pp. Farris, J. S., V. A. Albert, M. Kallersj6, D. Lipscomb, & A. G. Kluge. 1996. Parsimony jackknifing outperforms neighbor-joining.—Cladistics 12: 99-124. Fischer, P. 1885. Manuel de conchyliologie et de pa- léontologie conchyliologique ou histoire natu- relle des mollusques vivants et fossiles (Suivi dun appendice sur les Brachiopodes par D. P. Oehlert). EK Savy, Paris, pp. 689-784. Goodrich, C. 1922. The Anculosae of the Alabama River drainage.—University of Michigan Mu- seum of Zoology Miscellaneous Publications 7: 1-57, 3 plates. . 1941. Pleuroceridae of the small streams of the Alabama River system.—Occasional Papers of the Museum of Zoology, University of Mich- igan 427:1—10. Graf, D. L. 2001. The cleansing of the Augean stables, or a lexicon of the nominal species of the Pleu- roceridae (Gastropoda: Prosobranchia) of recent North America, north of Mexico.—Walkerana 12:1—124. Haldeman, S. S. 1840. A monograph of the Limnaides and other freshwater univalve shells of North America. Supplement to No. 1. J. Dobson, Phil- adelphia. . 1843-1853. Monograph of Leptoxis. Pp. 1—6 in J. C. Chenu, Illustrations Conchyliologiques ou description et figues des toutes les coquilles connues vivantes et fossiles, classes suviant le systeme de Lamarck modifie d’apres les progres 539 de la science, et comprenant les genres nou- veaux et les especes recemment decouvertes. Four volumes. Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT.—Nucleic Acids Sym- posium Series 41:95—98. Harris, S. C. 1990. Preliminary considerations on rare and endangered invertebrates in Alabama.— Journal of the Alabama Academy of Science 61:64—92. Hershler, R., J. M. Pierson, & R. S. Krotzer. 1990. Rediscovery of Tulotoma magnifica (Conrad) (Gastropoda: Viviparidae).—Proceedings of the Biological Society of Washington 103:815—824. Holznagel, W. E. 1998. A nondestructive method for cleaning gastropod radulae from frozen, alco- hol-fixed, or dried material—American Mala- cological Bulletin 14:181—183. . 2000. A phylogenetic study of selected North American Pleuroceridae (Gastropoda: Cerithioi- dea) using molecular and radula data. Unpub- lished Ph.D. dissertation, University of Ala- bama, Tuscaloosa, 140 pp. , & C. Lydeard. 2000. A molecular phylogeny of North American Pleuroceridae (Gastropoda: Cerithioidea) based on mitochondrial 16S rDNA sequences.—Journal of Molluscan Stud- ies 66:233—257. International Commission on Zoological Nomencla- ture (I.C.Z.N.). 1999. International Code of Zoological Nomenclature, 4th edition. Interna- tional Trust for Zoological Nomenclature, Lon- don, 306 pp. Lydeard, C., & R. L. Mayden. 1995. A diverse and endangered aquatic ecosystem of the southeast- ern United States——Conservation Biology 9: 800-805. , W. E. Holznagel, J. Garner, P. Hartfield, & M. Pierson. 1997. A molecular phylogeny of Mo- bile River drainage pleurocerid snails (Caeno- gastropoda: Cerithioidea).—Molecular Phylo- genetics and Evolution 7:117—128. , R. L. Minton, & J. D. Williams. 2000. Pro- digious polyphyly in imperiled freshwater pearly-mussels (Bivalvia: Unionidae): an ex- amination of species and generic designations. Pp. 145-158 in E. Harper & J. Taylor, eds., The evolutionary biology of the Bivalvia, The Geo- logical Society of London, Special Publications 177, 494 pp. , W. E. Holznagel, M. Glaubrecht, & W. F Pon- der. 2002. Molecular phylogeny of a circum- global, diverse gastropod superfamily (Ceri- thioidea: Mollusca: Caenogastropoda): pushing the deepest phylogenetic limits of mitochondrial LSU rDNA _ sequences.—Molecular Phyloge- netics and Evolution 22:399—406. 540 Minton, R. L., & C. Lydeard. 2003. Phylogeny, tax- onomy, genetics and global heritage ranks of an imperiled, freshwater snail genus Lithasia (Pleuroceridae).—Molecular Ecology 12:75-87. Neves, R. J., A. E. Bogan, J. D. Williams, S. A. Ahls- tedt & P. W. Hartfield. 1997. Status of aquatic mollusks in the Southeastern United States: a downward spiral of diversity. Pp. 46—85 in G. W. Benz & D. E. Collins, eds., Aquatic fauna in peril: the southeastern perspective. Lenz De- sign and Communications, Decatur, Georgia, 600 pp. Palumbi, S., A. Martin, S. Romano, W. O. McMillan, L. Stice, & G. Grabowski. 1991. A simple fools guide to PCR. Privately distributed, Honolulu, Hawaii. Salisbury, B. A. 2001. SEPAL: strongest evidence and parsimony analyzer. Version 1.4. Department of Ecology and Evolutionary Biology, Yale Uni- versity, New Haven, Connecticut. Stein, C. B. 1976. Gastropods. Pp. 1-41 in H. Bos- chung, ed., Endangered and threatened plants and animals of Alabama. Bulletin of the Ala- bama Museum of Natural History 2. Swofford, D. 2002. PAUP*: Phylogenetic Analysis Using Parsimony, version 4.0b10. Sinauer As- sociates, Sunderland, Massachusetts. Tryon, G. W. 1873. Land and freshwater shells of North America, part IV. Strepomatidae.— Smithsonian Miscellaneous Collections 253:1— 435. Turgeon, D. D. et al. 1998. Common and scientific names of aquatic invertebrates from the United States and Canada: Mollusks, 2nd edition. American Fisheries Society Special Publication 26, 526 pp. Appendix I Localities surveyed in 1996. Data include topo- graphic coordinates; exact locality information is avail- able from the authors. Localities where E. melanoides was found are indicated by an asterisk (*). ALABAMA. Blount County: Berry Creek, T12S, RIE, sec. 33.—Calvert Creek, T13S, RIE, sec. 32.— Calvert Creek, T13S, RIE, sec. 6.—Calvert Prong of Little Warrior River, T12S, RIE, sec. 22.—Dry Creek, T12S, R1IW, sec. 23.—*Little Warrior River, T13S, RIE, sec. 30.—*Little Warrior River, T13S, RIE, sec. 32.—Little Warrior River, T13S, R1IW, sec. 13.—*Lo- cust Fork, AL 160, T12S, R1W, sec. 23.—Locust Fork, T1OS, R2E, sec. 25.—Locust Fork, T10S, R2E, sec. 27.—Locust Fork, T11S, R2E, sec. 1.—*Locust Fork, TELS» R2ZE> sec) 4.—sLocust Fork DIZS RVE. see: 6.—*Locust Fork, T12S, RIW, sec. 13.—*Locust Fork, T12S, R1IW, sec. 23.—Locust Fork, T12S, RIW, sec. 27.—Locust Fork, T12S, RIW, sec. 32.—Mill Creek; NEISS, RIES see, O——Sucari@reckes Hiss: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON RIW, sec. 6.—Tucker Creek, T12S, R1IW, sec. 28.— Whippoorwill Creek, T11S, R3E, sec. 3. Etowah County: Bristow Creek, 0.5 miles S of Pine Grove, TI1S, R4E, sec. 7. Jefferson County: Campbell Creek, T1I5S, R4W, sec. 9.—Cane Creek, T15S, R4W, sec. 32.—Five Mile Creek, T16S, R4W, sec. 8.—Gurley Creek, T15S, R2W, sec. 21.—Locust Fork, 1.5 miles downstream of I-65, T14S, R3W, sec. 27.—Locust Fork, 5.5 miles downstream of I-65, T15S, R3W, sec. 32.—Locust Fork, 7 miles downstream of I-65.—Lo- cust Fork, T14S, R2W, sec. 16.—Locust Fork, T14S, R2W, sec. 20.—Locust Fork, T14S, R2W, sec. 29.— Locust Fork, T14S, R2W, sec. 30.—Locust Fork, T14S, R3W, sec. 24.—Locust Fork, T14S, R3W, sec. 34.—Locust Fork, T14S, R3W, sec. 35.—Locust Fork, T14S, R3W, sec. 35.—Locust Fork, T15S, R2W, sec. 9.—Locust Fork, TI5S, R4W, sec. 11.—Locust Fork, T15S, R4W, sec. 15.—Locust Fork, T15S, R4W, sec. 22.—Locust Fork, T15S, R4W, sec. 29.—Locust Fork, TI5S, R4W, sec. 29.—Locust Fork, T15S, R5W, sec. 25.—Locust Fork, T15S, R5W, sec. 36.—Locust Fork, TI5S, R6W, sec. 20.—Locust Fork, T16S, RSW, sec. 11.—Turkey Creek, T15S, R3W, sec. 2.—Village Creek, T16S, RSW, sec. 22.—Village Creek, T16S, R5W, sec. 36.—Village Creek, T17S, R4W, sec. 18. Marshall County: Clear Creek, T10S, R3E, sec. 35.— Drift Creek, T10S, R3E, sec. 17.—Locust Fork, T10S, R3E, sec. 34.—Slab Creek, T10S, R3E, sec. 6.—Slab Creek, T9S, R3E, sec. 33. Appendix II Systematic list of taxa used in this study. Taxon names (Burch & Tottenham 1980) are followed by mu- seum voucher and GenBank accession numbers. FMNH—Field Museum of Natural History, Chicago. NCSM—North Carolina Museum of Natural Sciences, Raleigh. PLEUROCERIDAE. Elimia: E. alabamensis, NCSM P4658, U73761; E. caelatura, NCSM P4659, AF100988; E. crenatella, NCSM P4660, U73762; E. cylindracea, NCSM P4661, U73765; E. haysiana, NCSM P4662, U73763; E. hydei, NCSM P4663, U73764; E. interrupta, NCSM 6595, AY010521; E. melanoides, FMNH 301889, AF540003; E. olivula, NCSM P4664, U73766; E. showalteri, NCSM P4665, U73767; E. virginica, NCSM P4666, AF100989. Jo: To fluvialis, NCSM P4667, AF100999. Juga: J. bul- bosa, NCSM P4668, AFI01005; J. plicifera, NCSM P4669, AF101004; J. silicula, NCSM P4670, AF101003. Leptoxis: Le. ampla, NCSM P4671, U73768; Le. crassa anthonyi, NCSM_ P4672, AF101001; Le. picta, NCSM P4673, U73769; Le. pli- cata, NCSM P4674, U73770; Le. praerosa, NCSM P4675, AFI101002: Le. taeniata, NCSM P4676, U73771,; Le. virgata, NCSM P4677, AF101000. Lith- asia: cf. Li. armigera, NCSM P4678, AF100998; Li. duttoniana, NCSM P4679, AF100997; Li. geniculata VOLUME 116, NUMBER 3 fuliginosa, NCSM P4680, AF100995; Li. geniculata geniculata, NCSM P4681, AF100996. Pleurocera: P. acuta acuta, NCSM P4684, AF100994; P. annulifer- um, NCSM P4685, U73772; P. canaliculatum filum, NCSM P4686, AF100991; P. prasinatum 1, NCSM P4689, U73774; P. prasinatum 2, NCSM P4688, 541 U73773; P. pyrenellum, NCSM P4690, AF100990; P. unciale hastatum, NCSM P4687, AF100993; P. ves- titum, NCSM P4691, U73775; P. walkeri, NCSM P4692, AF100992. THIARIDAE. Melanoides tuber- culata, NCSM P4682, AF101006. MELANOPSIDAE. Melanopsis praemorsa, NCSM P4683, U73776. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):542—547. 2003. A new species of Magelona Miller, 1858 (Polychaeta: Magelonidae) Maria Teresa Aguado and Guillermo San Martin Laboratorio de Biologia Marina e Invertebrados, Departamento de Biologia—(Zoologia), Facultad de Ciencias, Universidad Aut6noma de Madrid, Canto Blanco, 28049 Madrid, Spain Abstract.—Magelona magnahamata, new species, is described based on specimens found during a study of interstitial polychaetes carried out at Coiba National Park (Panama). It differs from other species of Magelona with regard to features of chaetiger 9, hooded hooks on anterior abdominal segments, and coloration pattern. Specimens of this new species coincide widely with speci- mens from the Gulf of México reported and described previously as Magelona sp. C. The family Magelonidae Cunningham & Ramage, 1888 currently includes three gen- era: Magelona Miiller, 1858, with about 55 described species (Fiege et al. 2000) plus ten others identified and described, without specific names, by Uebelacker & Jones (1984); Meredithia Hernandez-Alcantara & Solis-Weiss, 2000, with two species; and Octomagelona Aguirrezabala, Ceberio & Fiege, 2001, composed of a single known species. The genus Meredithia is characterized by bearing large hooded curved spines in some abdominal chaetigers (Hernandez-Alcantara et al. 2000), and the genus Octomagelona is distinguished by possessing eight thorac- ic segments instead of nine (Aguirrezabala et al. 2001). Diagnostic characters of the genus Ma- gelona are: presence and features of frontal horns, dimensions of the prostomium, pres- ence of dorsal median lobes and lateral pouches, shape of chaetae on chaetiger 9, structure of the abdominal hooks and the various shapes of the lateral lamellae. These characters have been discussed by Jones (1963, 1977, 1978) and recently by Rouse (2001). The most recent review work for European magelonids is found in Fiege et al. (2000), who outlined the terminology frequently used for magelonids. Several au- thors agree that a high number of species are yet to be described (Uebelacker & Jones 1984, Hernandez-Alcantara et al. 2000, Wilson 2000). The new species belongs to the genus Magelona, and the specimens were found during a study carried out in the National Park of Coiba (Panama), these specimens coincide widely with those described as Magelona sp. C found in the Gulf of Méx- ico by Uebelacker & Jones (1984). Only two species of this family have been reported in the Atlantic coast of Pan- ama (Magelona pacifica Monro, 1933 and M. riojai Jones, 1963); a single species both from the Atlantic and Pacific of Panama (Magelona papillicornis Miller, 1858); and only one from the Pacific (Magelona sp.., L6épez et al. 2002, later identified as M. sac- culata Hartman, 1961 after re-examina- tion). Reports on Magelonidae from Pana- ma were made by Monro (1933), Fauchald & Reimer (1975) and Fauchald (1977). Uebelacker & Jones (1984) described 12 species of Magelona from the Gulf of Méx- ico, without giving specific names (Mage- lona sp. A to L) and found specimens of M. pettiboneae Jones, 1963. In Mexican Pa- cific waters, Hernandez-Alcantara & Solis- Weiss (2000) reported four species of Ma- gelona: M. pacifica, M. sacculata, M. pitel- kai Hartman, 1944a, and M. californica Hartman, 1944b. Other reports from nearby VOLUME 116, NUMBER 3 regions were made by Jones (1963) (Carib- bean Sea) and Hartman (1969), Jones (1978), and Blake (1996) (California). Scientific expeditions dedicated to the knowledge of the flora and fauna of Coiba National Park (Pacific, Panama) became regular from 1996. Characteristics of this area and previous results of the study of polychaetes from these expeditions are de- fanled sim Wwopez, ct ale (19975 2002), San Martin et al., (1997) and Capa et al. (2001a, 2001b, 2001c). Materials and Methods Material from the National Park of Coiba occurs in two samples from soft substrates taken at sampling stations of Mali Rock (7°39'N, 81°41'40”"W) and Granito de Oro (7°35'30"N, 81°42'30"W). Sediment is com- posed of coarse sand, coming from coral destruction, sampled at depths to 0.5 m and 10 m. Samples were obtained using 1-liter capacity PVC tubes (core). Samples were filtered through a 0.1-mm gauge mesh sieve and retained material was fixed in formalin and preserved in ethanol (70%). Material from the Gulf of México was loaned by the Smithsonian Institution and belongs to the collection examined by Ueb- elacker & Jones (1984). The samples were taken along the northeastern Gulf of Mex- ico (from Alabama to Florida); 10-117 m; coarse to fine-very fine sand, silty fine to very fine sand, and sandy silt containing clay. Specimens were examined using a com- pound microscope equipped with an inter- ference Nomarsky system, and drawings were made using a camera lucida drawing tube. Pictures from scanning electron mi- croscopy (SEM) were taken at the SIDI (Servicio Interdepartamental de Investiga- cion), Universidad Auton6ma de Madrid. Holotype, paratypes and additional ma- terial from Coiba National Park are depos- ited at the Museo Nacional de Ciencias Na- turales (MNCN) de Madrid. Material from the Gulf of México is deposited in the 543 Smithsonian Institution, National Museum of Natural History, Washington, D.C. Results Genus Magelona Miiller, 1858 Magelona magnahamata, new species Jae, I, 2 Material examined.—Coiba National Park, Pacific of Panama; Granito de Oro, holotype (MNCN 16.01/8736a), paratypes (4 spec. MNCN 16.01/8736b) and 3 speci- mens. Mali Rock (1 spec. MNCN 16.01/ 8737). Additional material: Gulf of México (1 spec., USNM 86713); (1 spec., USNM SOS) CE specs USINIM S676); (@ spec= USNM 86717); (2 spec., USNM 86718); (1 spec., USNM 867119); (2 spec., 86720); (1 spec., USNM 86722): (1 spec., USNM 86723); (1 spec., USNM 86725); (5 spec., USNM 86725); (3 spec., USNM 86721). Description.—Holotype complete speci- men with 82 chaetigers, 19 mm long and 0.39 mm wide at chaetiger 5, dark pigment on several specimens at anterior dorsal part of thorax and on each side of anterior ab- dominal chaetigers in some specimens. Pro- stomium triangular to semicircular, rounded anteriorly, almost as long as wide, frontal horns absent. Palps longer than thorax, reaching to about chaetiger 23 (paratype), with 2 rows of papillae, becoming smaller toward bases of palps. Dorsolateral margins of first segment covering posterior part of prostomium and ventrally surrounding ba- ses of palps (Figs. 1A, 2A). Thoracic chae- tigers bearing, both on notopodia and neu- ropodia, a pair of lanceolate, wide lateral lamellae; dorso-medial and neuropodial lobes absent (Figs. 1G, 2B). Abdominal chaetigers each with pair of lateral, folia- ceous lamellae, smaller than thoracic ones (Fig. 1H). Lamellae gradually decreasing in size posteriorly, becoming digitiform (Fig. 11). Dorsal medial and neuropodial lobes small, papilliform on anterior and posterior abdominal chaetigers (Fig. 1H, I). Thoracic capillary chaetae slender, long and limbate (Fig. 2B). Chaetiger 9 lacking specialized 544 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Magelona magnahamata, n. sp. A: Anterior end, dorsal view; B: lateral papillae, posterior chaetiger; C: Posterior end, dorsal view; D: Hooks, anterior abdominal chaetiger; E: Hooks, abdominal chaetiger, segment 13; F: Hooks, posterior abdominal chaetiger; G: Parapodium, posterior view, anterior chaetiger; H: Parapodium, posterior view, median chaetiger; I: Parapodium, posterior view, posterior chaetiger. Scale A: 375 wm; B: 35 pm; C: 200 wm; D—-F: 35 pm; G, H: 200 wm; I: 50 pm. VOLUME 116, NUMBER 3 545 re gt oS 4 Fig. 2. Magelona magnahamata, n. sp. Scanning electron micrographs (SEM). A: Anterior end, dorsal view; B: Lamellae and capillary chaetae, thoracic segment; C: Hooded hooks on anterior abdominal chaetigers; D: Hooded hooks on posterior abdominal chaetiger. chaetae. Abdominal hooded hooks biden- tate, arranged in two groups on each ramus facing vis-a-vis. Hooded hooks of anterior abdominal chaetigers in three abruptly dif- ferent sizes; each ramus with a very small hook originating at base of lateral lamella, a single large and strongly curved hook, and three intermediate-sized hooks (Figs. 1D, 2C). From abdominal chaetiger 2-5 on (depending upon specimens), intermediate hooks numbering two (Fig. IE). On noto- podium, smallest and largest hooks face dorsally, and intermediate sized ones face ventrally (vice versa at neuropodium) (Fig. 1H). On posterior chaetigers, hooded hooks in two sizes, a single small hook nearest to lamellae, and several larger hooks (Fig. 1F); on most posterior chaetigers all hooks of one size, all similar (Figs. II, 2D). The hooks, on posterior chaetigers, more nu- merous (5—6) than those on anterior abdom- inal chaetigers, oriented 4 to 2 in notopo- dium, 4 hooks face dorsally and 2 ventrally, and vice versa in neuropodium (Fig. 11); and in others combinations (3 to 2, 4 to 1, 5 to 1). No lateral pouches. Conical pygid- ium with two short digitiform cirri (Fig. 1C). Abdominal chaetigers of some speci- mens (from chaetiger 40 in holotype, from the first abdominal chaetiger in others) with pronounced, rounded, lateral, glandular ar- eas anterior to parapodia, also pronounced, dark to red, and covered with small papillae (ies tE, ©): Remarks.—Magelona magnahamata 1s characterized by having a large, strongly curved, hooded hook on each anterior ab- dominal parapodial ramus and by lacking modified chaetae on chaetiger 9. Some less- er differences have been found between the Atlantic and Pacific material. For instance, the apical tooth of the large hooks is very 546 small in some specimens from the Gulf of México, while in other specimens from this area and in those of Coiba it is long, but this condition probably can be attributed to intra-specific variation. In some specimens of the Gulf of México, the large hooks seem to be slightly more robust than those from Coiba. Finally, the size is also somewhat different: specimens from the Gulf of Méx- ico are smaller than those Coiba, the largest being 10 mm long and 0.14 mm wide (at level of chaetiger 5), whereas in Coiba, the holotype is 19 mm long and 0.39 mm wide. However, these differences are not suffi- cient to consider the populations as belong- ing to two different species. The closest species known from the Pa- cific Ocean is Magelona pettiboneae Jones, 1963. Both seem similar in the capillary chaetae, bidentate hooks, and the absence of modified chaetae in chaetiger 9; how- ever, the hooks of M. pettiboneae are not so curved in anterior abdominal chaetigers, its lateral lamellae are shorter and rounded, and the hooks are more numerous. Magelona magnahamata is also related to three other species from the Atlantic coast of North America (M. papillicornis Miller, 1858, M. californica Hartman, 1944b, and M. minuta Eliason, 1962) for which Jones (1977) established common characteristics. In this way, all four species would share a subtriangular prostomium with rounded anterolateral margins, absence of dorsal medial and neuropodial lobes at the anterior region, abdominal chaetigers with subtriangular or sublanceolate lateral lamellae, dorsal medial and neuropodial lobes papilliform, and bidentate hooded hooks, one of them, the nearest to lamellae, smaller than the remaining. However, the presence of very bidentate and strongly curved hooks separates M. magnahamata from the others. Some specimens of M. magnahamata also possess modified posterior abdominal chaetigers with dark colored areas covered with papillae. These papillae have never PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON been cited for any described species in the genus. Etymology.—The specific name comes from Latin magna = large and hamutlus, di- minutive of hamus = hook, referring to the large strongly curved hooded hooks. Acknowledgments We wish to thank M. Capa, E. Lopez and R. Gallego for their comments and advice, and special thanks to the Smithsonian In- stitution for material loaned from the Gulf of México and to the Agencia Espafiola de Cooperacion Internacional (AECI) for par- tial financial support of the survey and IN- RENARE for partial financial and logistical support during the expeditions to the Na- tional Park of Coiba. The comments of two anonymous referees greatly improved the quality of the paper. Literature Cited Aguirrezabala, EF, A. Caberio, & D. Fiege. 2001. Oc- tomagelona bizkaiensis (Polychaeta: Mageloni- dae) a new genus and species from the Cape Breton Canyon (Bay of Biscay, north-east At- lantic).—Journal of the Marine Biological As- sociation of the United Kingdom 81:221—224. Blake, J. A. 1996. Family Magelonidae Cunningham and Ramage, 1888. Pp. 253—260 in J. A. Blake, B. Hilbig, & P. H. Scott, eds., The Annelida, part 3. Polychaeta: Orbiniidae to Cossuridae, vol. 6. Santa Barbara Museum of Natural His- tory: Santa Barbara, California. Capa, M., G. San Martin, & E. Lopez. 2001a. Syllinae (Syllidae: Polychaeta) del Parque Nacional de Coiba (Panama).—Revista de Biologia Tropical 48:101-113. , & . 2001b. Description of a new species of Parasphaerosyllis (Polychaeta: Syllidae: Syllinae).—Proceedings of the Bio- logical Society of Washington 114:280—284. 5 Ce . 2001c. Autolitynae, Eu- syllinae y Exogoninae (Syllidae: Polychaeta) con comentarios sobre la ecologia y biogeogra- fia de la familia Syllidae del Parque Nacional de Coiba, Panama.—Revista de Biologia Tropi- cal 49:621—628. Cunningham, J. T. & Ramage, G. A. 1888. The Poly- chaeta sedentaria of the Firth of Forth.—Trans- actions Royal Society of Edinburg 33:635—684. Eliason, A. 1962. Undersokningar 6ver Oresund. XXXXI. Weitere Untersuchungen tiber die Po- VOLUME 116; NUMBER 3 lychaetenfauna des Oresunds.—Lunds Univer- sitets Arsskrift, N. EK Avd. 2, 58:1—98. Fauchald, K. 1977. Polychaetes from intertidal areas in Panama, with a review of previous shallow- waters records.—Smithsonian Contributions to Zoology 221:1-81. , & A. A. Reimer. 1975. Clave de poliquetos panamefnos con la inclusi6n de una clave para todas las familias del mundo.—Boletin del Ins- tituto Oceanografico Universidad Oriente 14: 71-94. Fiege, D., E Licher, & A. S. Y. Mackie. 2000. A partial review of the European Magelonidae (Annelida: Polychaeta): Magelona mirabilis redefined and M. johnstoni sp. nov. distinguished.—Journal of the Marine Biology Association of the United Kingdom 80:215—234. Hartman, O. 1944a. Polychaetous annelids from Cali- fornia, including the description of two new genera and nine new species.—Allan Hancock Pacific Expeditions 10:239—307. . 1944b. Polychaetous annelids, 6. Paraonidae, Magelonidae, Longosomidae, Ctenodrilidae and Sabellariidae.—Allan Hancock Pacific Expedi- tions 10:311—389. 1961. Polychaetous annelids from Califor- nia.—Allan Hancock Pacific Expeditions 25:1— 226. . 1969. Atlas of the Sedentariate Polychaetous Annelids from California. Allan Hancock Foun- dation, University of Southern California, Los Angeles, 494 pp. Hernandez-Alcantara, P., & V. Solis-Weiss. 2000. Ma- gelonidae from the Mexican Pacific and North- ern Gulf of Mexico, with the description of a new genus (Meredithia) and four new spe- cies.—Bulletin of Marine Science 67:625—644. Jones, M. L. 1963. Four New Species of Magelona (Annelida, Polychaeta) and a Redescription of Magelona longicornis.—American Museum Novitates 2164:1—31. . 1977. A Redescription of Magelona papilli- cornis F. Miiller. Pp. 247—266 in D. J. Reish, & K. Fauchald, eds., Essays on Polychaetous An- 547 nelids in memory of Dr. Olga Hartman. Allan Hancock Press, Los Angeles, 604 pp. . 1978. Three new species of Magelona (An- nelida, Polychaeta) and a redescription of Ma- gelona pitelkai Hartman.—Proceedings of the Biological Society of Washington 91:336—363. L6épez, E., G. San Martin, P Cladera, & M. Capa. 1997. La fauna de anélidos poliquetos del Parque Nacional de Coiba (Panama) Pp. 57—73 in S. Castroviejo, ed., Flora y Fauna del Parque Nacional de Coiba (Panama), 534 pp. , P. Cladera, G. San Martin, A. Laborda, & M. T. Aguado. 2002. Polychaete assemblages in- habiting intertidal soft bottons associates to mangrove systems from Coiba National Park (East Pacific, Panama).—Wetlands Ecology and Management 10:233-—242. Monro, C. C. A. 1933. The Polychaeta Errantia col- lected by Dr. C. Crossland at Col6n in the Pan- ama region and the Galapagos Islands during the expedition of the “St. George’’.—Proceed- ings of Zoological Society of London 1—96. Miller, EK 1858. Einiges tiber die Anneliden Fauna der Insel St. Catharina an der Brazilianischen Ktis- te.—Archiv fiir Naturgeschichte Berlin 24:21 1— 220. Rouse, G. W. 2001. Magelona Miller, 1858. Pp. 261— 263 in G. Rouse, & E Pleijel, eds., Polychaetes. Oxford University Press, New York, 354 pp. San Martin, G., E. Lopez, M. S. Redondo, M. Capa, P. Cladera, & A. J. Laborda. 1997. El bentos marino del Parque Nacional de Coiba (Panama). Pp. 45-67 in S. Castroviejo, ed., Flora y Fauna del Parque Nacional de Coiba (Panama), 534 PP. Uebelacker, J. M., & M. L. Jones. 1984. Family Ma- gelonidae Cunningham and Ramage, 1888. Pp. 7.1—7.29 in J. M. Uebelacker, & P. G. Johnson, eds., Taxonomic Guide to the Polychaetes of the Northern Gulf of Mexico, vol. 2. Barry A. Vit- tor & Associates, Inc., Mobile, Alabama. Wilson, R. S. 2000. Family Magelonidae Pp. 194—195 in P. L. Beesley, G. J. B. Ross, & C. J. Glasby, eds., Polychaetes & allies. The southern synthe- sis. Fauna of Australia, vol. 4. Polychaeta, My- zostomida, Pogonophora, Echiura, Sipuncula. CSIRO Publishing, Melbourne, 465 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):548—556. 2003. Allonais inaequalis (Annelida: Oligochaeta: Tubificidae) in North America R. Deedee Kathman and Mark J. Wetzel (RDK) Aquatic Resources Center, 5109 Rock Bridge Lane, Thompsons Station, Tennessee 37179, U.S.A., e-mail: aquatres @ix.netcom.com; (MJW) Illinois Natural History Survey, Center for Biodiversity, 607 E. Peabody Drive, Champaign, Illinois 61820, U.S.A., e-mail: mjwetzel @uiuc.edu Abstract.—The oligochaete Allonais inaequalis (Stephenson, 1911), previ- ously thought to be unknown in North America, was found in the effluent of a sewage treatment plant in New Jersey and recently in the Greater Miami River in Ohio. It is redescribed from this material and compared to the other two species, A. paraguayensis (Michaelsen, 1905) and A. pectinata (Stephen- son, 1910) reported from North America. Identifications of specimens collected from North America, previously thought to be A. paraguayensis, were deter- mined to be incorrect; thus, this species is not yet known to occur in North America. Nais magnaseta Harman, 1973, which closely resembles A. pectinata, also is discussed. The naidid oligochaete! Allonais inae- qualis (Stephenson, 1911) was first de- scribed as Nais pectinata var. inaequalis Stephenson, 1911. Sperber (1948) erected the genus Allonais, using A. inaequalis as the generic type. She distinguished Allonais from Nais by the absence of eyes, by the anterior and posterior ventral chaetae being of similar size and shape, by no discernable stomachal dilatation, by a vascular plexus in the anterior segments, by no prostate gland cells, and by no formation of budding zones. She also indicated that A//onais, un- like Nais, was confined to the tropics. She ' Based on sequences of 18S rDNA and other mo- lecular and morphological data, Erséus et al. (2002) concluded that the family Naididae is polyphyletic and that the species of naidids are more correctly placed within a subfamily of the Tubificidae. Therefore, Nai- didae has become a junior synonym of the family Tub- ificidae. As the name Naididae is older than Tubifici- dae, the authors are requesting that Tubificidae take precedence because of the large number of species in the family compared to those within the naidids and await a ruling by the International Commission on Zoological Nomenclature (C. Erséus, pers. comm.; Er- séus & Gustavsson 2002). included five species (all formerly found in Nais) in this new genus, including two, A. pectinata (Stephenson, 1910) and A. para- guayensis, (Michaelsen, 1905), subsequent- ly reported from North America (Brinkhurst & Jamieson 1971, Brinkhurst 1986). Brink- hurst (1986) included A. inaequalis in his North American key, suggesting that it probably occurred in North America due to its worldwide distribution and because drawings of the chaetae of A. paraguayen- sis closely resembled the chaetae of A. in- aequalis (e.g., Hiltunen & Klemm 1980). New material collected and identified by the authors, or sent to the authors for iden- tification or verification, encouraged us to reassess the status of this genus in North America. Materials and Methods Oligochaetes were observed swimming in the effluent of the Linden-Roselle Sew- age Authority wastewater treatment plant in New Jersey, U.S.A., in October 1999. The facility is a typical tertiary treatment plant, VOLUME 116, NUMBER 3 Fig. 1. B = 1 pm. treating approximately 80% residential and 20% municipal waste. Specimens were col- lected using a small bucket in the uptake valve in the aeration ponds immediately prior to chlorination. Half of the worms were fixed in 10% formalin; the remainder were fixed in 70% ethanol. One oligochaete specimen collected in July 2001 from the Greater Miami River, Ohio, U.S.A., tenta- tively identified as A. paraguayensis, was sent to the senior author for verification. The description of A. inaequalis below is based on 10 specimens from the New Jer- sey material, preserved in ethanol, and mounted on slides using CMCP mounting medium for observations using light mi- croscopy. Measurements were made using a calibrated ocular scale on a Leitz Labor- lux-12 compound microscope, and draw- ings were made using a camera lucida drawing tube with the same microscope. Oligochaetes fixed in formalin from New Jersey material were prepared for scanning electron microscopy (SEM). Specimens were mounted on aluminum stubs by plac- ing one to two specimens in a drop of water on each stub that had been prepared with transparent double-coated sticky tape (No. 666, 3M Corp.). Stubs were immersed in liquid nitrogen to instantly freeze the spec- imens, then placed in an Edwards-Pearse 549 Allonais inaequalis (SEM). A, dorsal chaetal bundle; B, same as A, enlarged. Scale bars: A = 5 ym, Model EPD3 tissue dryer (temperature — 64°C; vacuum 0.0002 torr) for subsequent sublimation of liquid associated with the specimens. After sublimation was complete (4—6 hr), stubs were placed in an SPI Mod- ule Sputter Coater, where a thin (10—30 nm) film of gold-palladium was evaporated onto the specimens. After sputter-coating, spec- imens were examined using an Amray 1830 scanning electron microscope and photo- graphed on positive-negative film and pa- per. Allonais inaequalis (Stephenson, 1911) Piess ie 2 Material examined.—Allonais paraguay- ensis: Louisiana, 1963, and Florida, 1975, W. J. Harman collection from the National Museum of Natural History-Smithsonian Institution (USNM), Washington, D.C.; Ne- vis, Lesser Antilles, 1997, Illinois Natural History Survey (INHS) Annelida Collec- tion, Champaign, identified by the authors; Sudan, Africa, 1975 and 1984, R. Grimm collection, Zoologisches Institiit und Zoolo- gisches Museum (ZMUH), Hamburg. Al- lonais inaequalis: Puerto Rico, 1994, INHS Annelida Collection, identified by the au- thors. Allonais pectinata: South Africa, 96S) douisianay Wi9745.— 1975, Georgia: 550 A B lower upper tooth tooth Riga: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON av av BY pv Chaetae of Allonais inaequalis. A, dorsals; B, needles; C, ventrals. av, anterior ventral; pv, posterior ventral. Scale bars: A = 40 pm, B = 6.25 wm, C = 16 wm. 1967, and Florida, 1978, W. J. Harman col- lection from USNM. Nais magnaseta (ho- lotype): Texas, 1973, USNM (45510). Some specimens of Allonais inaequalis from New Jersey were deposited in the An- nelida collection of Aquatic Resources Cen- ter, Thompsons Station, Tennessee, and the remainder were placed in the INHS Annel- ida Collection. The one specimen from Ohio was deposited in the collection at SoBran, Inc., Cincinnati, Ohio. Redescription.—Body 3—9 mm long, consisting of 37-85 segments, no pigmen- tation. Prostomium rounded, without pro- boscis. No eyes. Dorsal chaetae begin in segment VI: 1—2 smooth hairs per bundle; if 2 hairs, then generally one longer (212— 288 jm) and one shorter (75-160 wm); 1— 2 needles per bundle, each usually with 2 intermediate teeth (but some with 1, 3, or 4), intermediate tooth next to upper is short- est of all teeth and intermediate tooth next to lower is second longest; needle length 64—96 wm, lower tooth (8-14 wm) almost twice as long as upper (5—8 wm). Ventral chaetae usually 5—7 per bundle (but as few as 3 or as many as 9) in II-V, 55—75 wm long and 1-3 pm wide, nodulus slightly proximal, upper tooth slightly longer and thinner than the lower; from VI posteriorad usually 5—7 per bundle (but ranging from 3-9), 60—93 wm long and 1.5—3 wm wide (Puerto Rico specimens up to 5 wm wide), nodulus median to slightly distal, upper tooth thinner and subequal or slightly short- er than the lower; number of ventral chaetae decreases to 2 per bundle near the tail. No mature individuals were collected. Discussion It can be difficult to distinguish Allonais inaequalis from A. paraguayensis based on external morphological characters. Al- though Sperber (1948) stated that the sexual organs of these two species of Al/onais are similar, there appear to be several characters that can be used to identify them as being different, as shown in Tables | and 2. The most obvious character is the number of in- termediate teeth found in the needle chae- tae. While the needle chaetae of A. inae- VOLUME 116, NUMBER 3 Nn N = Table 1.—Descriptions of the North American species of Allonais Sperber (1948) and Brinkhurst & Jamieson (1971) and of Nais magnaseta by Harman (1973). All measurements in 41m unless otherwise indicated. Allonais Allonais Character paraguayensis Allonais inaequalis pectinata Nais magnaseta Anterior ventral chaetae: no./bundle 2-8; 2—6 4—8*; 4—-6° 3-5 3-4 length 55-123 60-105 56-65 54 width ed) 2, <<) teeth (upper vs. lower) longer longer longer slightly longer Posterior ventral chaetae: no./bundle 2-8; 2—6° 4—8*; 4-6° 2-7 2-5 length 55—123 60—105 51-58 57-67 width 5) 2 3 teeth (upper vs. lower) equal equal equal thinner; slightly shorter Dorsal chaetae: no. of hairs/bundle ]—2 1-2 |—2 1-2 length of hairs 200-500 100-332 70-220 114-180 no. of needles/bundle 1-2 1-2 1-2 1-2 length of needles 60-192 67-112 42-68 50—60 length, outer needle teeth 35 5—6 no. of intermediate teeth 1—2 14 1-5 1-3 Total length of worm, mm 4—60 8-18 1.5-—8 2. No. of segments 15-200 40-95 15-65 Silsr No. of penial chaetae 3-11 4-6 3-5 immature No. of specimens 4 Brinkhurst & Jamieson (1971); ° Sperber (1948). qualis generally have two intermediate teeth (range 1—4; Figs. 1A, B, 2B), none of the specimens of A. paraguayensis exam- ined had needle chaetae with intermediate teeth (Fig. 3). Sperber (1948) stated, how- ever, that the upper tooth of A. paraguay- ensis can be bifid in some specimens. The posterior ventral chaetae generally are shorter and thinner in A. inaequalis (Fig. 2C) compared with those of A. paraguay- ensis; the upper teeth are always distinctly shorter than the lower teeth in A. para- guayensis, whereas in A. inaequalis the up- per and lower teeth are generally subequal. The hair and needle chaetae are almost al- ways shorter in A. inaequalis compared with A. paraguayensis. The needle chaetae of A. pectinata are distinct from those of its congeners, having both the upper and lower teeth subequal in length, and the interme- diates recessed between the two. l Although most of the individuals exam- ined from the USNM were in very poor condition, all of the specimens examined appear to be A. inaequalis, not A. para- guayensis, So it is very likely that the re- maining specimens also are A. inaequalis. This means that A. inaequalis had already been found in North America (as early as 1963), but previously had been misidenti- fied, and that A. paraguayensis has not yet been found in North America. Thus A. in- aequalis can now be recorded from Florida, Louisiana, New Jersey, and Ohio, which suggests that it may be widespread but un- common, at least in the eastern USA. The specimens of A. inaequalis from North America fit the original description of Stephenson (1911) very well. All of his measurements fall within the ranges of ours, and he also noted the differences in length when two hair chaetae were present PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON N ‘SISUIADNSDADA SIDUO]]V Se PoyMUsp! A][CUISLIC » G I 9 v V 9 suouwitsads JO “ON SUISSTUI O9-CY pues 1o1a\sod C8-Lt 19-CP 8IL1T—09 O6-SEt SJUSUISOS JO “ON SUISSTUL UWIUW “WIOM Q-¢ “BO puso Jo11a}sod 6-€ “BO C-9'T “BD G'Q-G “BO p-E “BO JO Y8UZ] [RIO], y390} (2) E=~G CQ) () € (€) f (1) E36 0) 0 QVIPSuULIo}UL JO “OU 96-SL e8-SL 98-19 08-19 el=@llll ScI—08 So]poou JO WISUE| a]puNnq/so[poou Cul Gaal GG Gu (@) 1 Cl Jo OU 88C_VCC/08-SL COC-8EC/TII-SL SL7-€E1C/O9I-88 OSC-CIC/STI-S6 Sce-SOC/N0C-OS I eSe-SCC/86I1-Ol I suey JO YISUe| o]punq/s1ey (@) 1 Gaal Cual Al (ou | CA gO OU savyjovyo [esioqd Joy1oys ApYSI[s 10 Joy1oys APYSITS 10 Jayoys APYSsI[s 10 Jay10ys ATYSITS 10 Ja}10Ys J9}.10Ys (19MO] jenbogns ‘19uuTY) Jenboqns ‘1ouury) jenboqns ‘1ouury) jenboqns ‘1ouuTY} ‘TouUTYy ‘ToUUTY) ‘sA Jaddn) 99} Se iS eG SS | (9) S-€ V-E YIprM t6-SL CLAWS) OL—09 cL-V9 SOI—88 8l1—08 y1sue] 9-7 (€) L-S 6-S (UL) G=6 (L) 9-P 9-¢ (€) 2[pung/ou :OBJOEYO [VNUDA IOIDISOg Jasuo] Jenboqns 10 Josuoy] Byei0 (0) Jasuo] jenbaqns jenbogqns (19MO] Apysiys ‘1ouury) Apysiys ‘rouury} Apystys ‘rouury) Apystys ‘rouuryy ‘TOUUTU} ‘TOUUTY} ‘sA Joddn) y}900} eC SCC S°C-C S(t! (y) €-G'I cme YIPIM OOI-VL V9-9S SL-SS CL-S9 SOI-8L 96-9 ysus] SV V OE Sag (9) rE (Qe e[punq/ ou :OBJIVYO [PUDA IOLIOUY OorY ovlond (o}0(@) Aosiof MON eCUueIsIno | SIAON uepns Jajoereya ‘epHole syonbavul sipuo]]y sisuakonspivd sipuo]]V ‘VY SUOCT SJOIPoI JOS PuOdas ‘sey WOYS S]OOYOI SITeY JO YISUS| JOJ sIoquINU JO 19S ISIE “(OZ6I ‘1 161) Uosueydais Jo UONdIIOsep oy) YIM sooISe SIY], ‘OYS SI | puv Buoy st [ uo) “JUosoid ovjovyo sIeY Z JI ‘PoVOIPUT SSIMIOYIO ssajuN WT! UT s]jUsWTOINSKoW [Ty “Apmis sIy) SuLInp pourwexe suouttoeds wioAy syonbapul ‘y pure sisuadvnsv.nd sipuojjy JO uostredwioy— Z 2198, VOLUME 116, NUMBER 3 ff Fig. 3. Chaetae of Allonais paraguayensis. A, nee- dles; B, ventrals. Abbreviations as in Fig. 2. Scale bars = 6.25 pm. (“‘one of the hair-setae is much shorter than the other’; Fig. 2A). Sperber (1948), in erecting the genus Allonais and describing the species she ascribed to it, made no men- tion of the differences in length of the hair chaetae, so we do not know if it applies to all Allonais species or only to those that we examined. She stated, however, that she felt that A. pectinata belonged in a different, but new, genus. Allonais pectinata, the third species in this genus reported to occur in North Amer- ica, 1s reported to be widespread in the east- ern USA (Illinois, Ohio, New York, Penn- sylvania, and Georgia) and Ontario, Canada (Hiltunen & Klemm 1980, Brinkhurst 1986), yet neither of the present authors has ever seen a specimen in our many collec- tions, and neither Klemm nor Brinkhurst (pers. comm.) was able to provide any spec- imens to corroborate its published distri- & av ser 393 bution. Because of the difficulty in distin- guishing A. pectinata and Nais magnaseta Harman, 1973 from one another using the descriptions provided in the literature (Ta- ble 1), specimens of each species were ob- tained from the USNM for study. Examination of the specimens from the USNM caused additional confusion. Two specimens from Lake Louisa at Clermont, Florida, previously identified as A. pectin- ata, were in fact N. magnaseta (see Table 3; Fig. 5), making this a new state record for the latter species, since it had previously only been reported from Bee County, Texas (Harman 1973). Another specimen from the USNM, collected from west of the Peach County line in Georgia and initially iden- tified as A. pectinata, is actually Nais var- iabilis Piguet, 1906 with pectinate dorsal needle chaetae. The possible confusion of some specimens of N. variabilis (those with pectinate needle chaetae), and subsequent misidentification of them as A. pectinata, was noted previously by Brinkhurst (1986). The distinct difference in size and shape of the ventral chaetae in IH-V compared with the rest of the ventrals in N. variabilis should easily separate those with pectinate needles from specimens of A. pectinata. Five specimens from Louisiana and three specimens from South Africa were exam- ined and appear to be A. pectinata, although there are some differences between the two sets (see Fig. 4). It is now clear that A. pectinata and N. av pv Fig. 4. Chaetae of Allonais pectinata. A, Louisiana specimen; B, South African specimen. av, pv as in Fig. 2; n, needle. Scale bars = 6.25 wm. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 554 “Opeul oq JOU P[NOD UONROYHUSp!I oUYSp we dIOJOIOY) pue [elIoyeUI SuIsULI ay) Aq poinosqo Ajpensed sem puoseas oy} yng ‘UOT}DITJOO sIy] UI SUaUTOSdS Z DIOM JIU)» ‘YISUST sINSvOW JO JaquINU [eNj}Oe dy) JUTUTIAJOP 0} YNOYJIP AIOA , ‘IVY BUOT SJOOPJoI Jos puodsas “I1eY JOYS SJOOYo1 SUeCY JO YISUI| JOJ SIOQUUNU JO 9S ISI ‘JOYS SI [| pue SUOT SI | UdY) ‘WUosoId avJORYS Mey Z Jl g ‘VIVUIJIAad SIDUO]]Y SB Paytuept AT[eUIsUOC » pl +9€ Cl on} eUIUIT vV9-L Van eS-CYV Gaal OCC If JoVIOYS ‘1OUUTY} EAS |! 9S-8V Se jenbo ‘1auuTy} I PUG cl onyeUIUI SOS Ad) E=C 9S 8V I 09 1-8Tl I 19}.10Ys Apysiys ‘rouury} I CO II VO-ES SE @ Jasuo] ApYsI[s 10 jenba ‘1duuTy) APYstys S aL Ol se EC Ci Ga oa oIN}eUIUIT CV-E € (7) OL-SV Cal 9SC-89I/CII aT) I Jayioys APYstiys Jo yenba ‘1ouuTY) cmon C8-VS CE Jasuoy ApYsiI[s 10 jenbo ‘1auury) AYsITs ¢ OE OW ae LG LGV V Savane EW toe GA OLI-8CI/CI 1-6 (C) I Jayioys AYstys Jo yenbo ‘19uuTY} is ISG I V9-8V 9-V Jasuoy ApYsITs ‘rouutyy APYSIys suoultseds JO ‘ON SJUOUIBIS JO ‘ON WU ‘UIOM JO YSU] [10], avjovyo [eluad Jo ‘oN 499} B[PseuU J9}NO ‘“YISUIT Y}99} d}VIPSULIOJUI JO ‘Ou So[poou JO YSU] a]puNq/se[psou Jo ‘ou sirey JO YSU] sirey JO ‘ou ravjavyo [esloq (19MO]T “SA Joddn) y}90) qIpIM sue] o]punq/‘ou :OBJOVYO [BUDA JOLIOSOd (19MO] “SA Joddn) yI903 CEI Cl=GSC | Sil Sc I-I yyprmn 8t-SL 9S-8r p9-8r 6S-ES ysus] ane v-€ Van S-r (€) 9[punq/‘ou :OBJOBYO [PIUSA JOLIOIUY pepllo[y SUXOT, BOLIFY YINOS evUuvISINOT Jojovieyy DIASDUSDU SIDN vypuyjoad sipuo]]y ‘pa]LOIPUT ISIMIJOYIO ssoyuN WIT UT syUSsUTAINSvOU [Py “Apmis sty) SuLINp poururexs suouttoeds DjaspUuspU SIDN' pu VIDUIJIad sipuo]]y JO uOstIedWIOD—’'¢E 2IqUL VOLUME 116, NUMBER 3 A B / oT av Fig. 5. Chaetae of Nais magnaseta. A, needles from Texas (TX) and Florida (FL) specimens; B, an- terior ventral (av) and posterior ventral (pv), from Tex- as specimen. Scale bars = 6.25 wm. pv magnaseta are recognized as, two distinct species (Table 3; Figs. 4, 5). The main dif- ferences between the two species are the length of the needle chaetae teeth, with those of N. magnaseta almost twice as long as those of A. pectinata, and the distinct difference in the length and width of the anterior ventral chaetae compared to the posterior ventral chaetae in N. magnaseta, while there is only a slight difference be- tween the anterior and posterior ventral chaetae in A. pectinata. Apparently most of the confusion asso- ciated with the identification of A. inae- qualis and A. paraguayensis results from the original misidentification of the speci- mens from Louisiana and Florida. Although identified as A. paraguayensis, these obvi- ously are A. inaequalis (Table 2). Brink- hurst (1986), followed by Kathman & Brinkhurst (1998), hinted at this possibility when they stated that the needle chaetae of A. paraguayensis from Louisiana seemed to resemble the needle chaetae of A. inae- gqualis. Hiltunen & Klemm (1980), assum- ing the identifications of A. paraguayensis to be correct, illustrated the needle chaetae of A. inaequalis but listed the species as A. paraguayensis. Adding to the confusion, the drawings of the needle chaetae of A. inaequalis by Sperber (1948) do not agree with the original drawing of Stephenson (1911) but resemble those of A. pectinata. Sperber referred to this as a form of A. in- aequalis, but it is unclear if it really belongs with A. inaequalis. Kathman & Brinkhurst Nn Nn Nn (1998) added Florida, Illinois, and Colora- do to the distribution list for A. paraguay- ensis. Examination of the Florida worms in- dicated that they were A. inaequalis, while those from Hlinois and Colorado do not be- long in Allonais. Brinkhurst (1986) and Brinkhurst & Marchese (1989) suggested that A. inaequalis had been synonymized with A. paraguayensis and A. pectinata but gave no references. In contradiction, Har- man et al. (1988) stated that A. inaequalis clearly is distinct from A. paraguayensis. We do not know of any publications that have discussed possible synonymies of these two species, and Brinkhurst (pers. comm.) was unable to provide information about these comments. Summary Allonais inaequalis is herein reported for the first time in North America, although we believe that this species was found as early as 1963 but erroneously identified as A. paraguayensis. Its current distribution includes New Jersey, Ohio, Florida, and Louisiana. This means that A. paraguay- ensis has not yet been reported from North America. Although previously reported as fairly widespread, A. pectinata could only be confirmed as found in Louisiana. Nais magnaseta, known previously only from Bee County, Texas, has been identified from Florida. Based on external morphological char- acters, it appears that the three A//onais spe- cies previously reported from North Amer- ica can be easily differentiated, although A. inaequalis and A. paraguayensis seem to be more closely related to one another than to A. pectinata, as suggested by Sperber (1948). We know of no current phyloge- netic analysis using either morphological characters or gene sequencing for this ge- nus that would provide a more definitive approach to recognizing the species. Acknowledgments We would like to give our sincere thanks to W. Moser (National Museum of Natural 556 History) and to R. Grimm (Zoologisches Institiit und Zoologisches Museum) for the loan of specimens. Thanks also to C. Wat- son (SoBran, Inc., Cincinnati) for recogniz- ing the Allonais worm from Ohio and pro- viding the senior author with the specimen. Allonais specimens from Nevis, sent to the junior author for identification, were pro- vided by D. Bass (Univ. Central Oklaho- ma); Allonais specimens from Puerto Rico sent to the junior author for identification were provided by J. Kurtenbach (U.S. En- vironmental Protection Agency). L. Crane and C. Warwick (INHS) assisted with SEM photomicrography, and T. Askegaard (Aquatic Resources) assisted with figure formatting. D. Webb (INHS), S. Fend (U.S. Geological Survey, Menlo Park, Califor- nia), and S. Gelder (Univ. Maine—Presque Isle) provided valuable comments on this manuscript. A very special thanks goes to D. Cavalli (Brown and Caldwell, Mahwah, New Jersey) for finding the specimens of A. inaequalis and sending them to the senior author for identification, thus initiating this investigation into Allonais in North Amer- ica. Financial support for publication of this research was provided in part by the Illinois Natural History Survey. Literature Cited Brinkhurst, R. O. 1986. Guide to the freshwater aquat- ic microdrile oligochaetes of North America.— Canadian Special Publication of Fisheries and Aquatic Sciences 84, 259 pp. , & B. G. M. Jamieson. 1971. Aquatic Oligo- chaeta of the world. University of Toronto Press, Toronto, Ontario, Canada, 860 pp. , & M. R. Marchese. 1989. Guide to the fresh- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON water aquatic Oligochaeta of South and Central America.—Colecci6n CLIMAX 6:1-179. Erséus, C., & L. Gustavsson. 2002. A proposal to re- gard the former family Naididae as a subfamily within Tubificidae (Annelida, Clitellata)—Hy- drobiologia 485:253—256. , M. Kallersj6, M. Ekman, & R. Hovmédller. 2002. 18S rDNA phylogeny of the Tubificidae (Clitellata) and its constituent taxa: dismissal of the Naididae.—Molecular Phylogenetics and Evolution 22:414—422. Harman, W. J. 1973. New species of Oligochaeta (Nai- didae) with additional distributional records from Oklahoma and Texas.—Southwestern Nat- uralist 18:151—164. , R. O. Brinkhurst, & M. Marchese. 1988. A contribution to the taxonomy of the aquatic Oli- gochaeta (Naididae) of South America.—Ca- nadian Journal of Zoology 66:2233—2242. Hiltunen, J. K., & D. J. Klemm. 1980. A guide to the Naididae (Annelida: Clitellata: Oligochaeta) of North America.—U.S. Environmental Protec- tion Agency, Cincinnati, Ohio, EPA-600/4-80- 031, 48 pp. Kathman, R. D., & R. O. Brinkhurst. 1998. Guide to the freshwater oligochaetes of North America. Aquatic Resources Center, College Grove, Ten- nessee, 264 pp. Michaelsen, W. 1905. Zur Kenntnis der Naididen.— Zoologica 44:350—361. Piguet, E. 1906. Observations sur les Naididées et re- vision systématique de quelques espeéces de cet- te famille.—Revue Suisse de Zoologie 14:185— Bilsy Sperber, C. 1948. A taxonomical study of the Naidi- dae.—Zoologiska Bidrag fran Uppsala 28:1-— 296. Stephenson, J. 1910. On some aquatic oligochaete worms commensal in Sporgilla carteri.—Re- cords of the Indian Museum V:233—240. , 1911. On some aquatic Oligochaeta in the col- lection of the Indian Museum.—Records of the Indian Museum 6:203-—214. , 1920. On a collection of Oligochaeta from the lesser known parts of India and from eastern Persia.—Memoirs of the Indian Museum 7: 191-261. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):557—600. 2003. A systematic review of Planaltina (Teleostei: Characiformes: Characidae: Glandulocaudinae: Diapomini) with a description of two new species from the upper rio Parana, Brazil* Naércio A. Menezes, Stanley H. Weitzman, and John R. Burns (NAM) Museu de Zoologia, Universidade de Sao Paulo, Caixa Postal 42694; CEP 04299-970, Sao Paulo, S.P., Brazil, e-mail: naercio@usp.br; (SHW) Department of Systematic Biology, National Museum of Natural History, Division of Fishes, MRC 0159, Smithsonian Institution, PO Box 37012, Washington, D.C. 20013-7012, U.S.A., e-mail: weitzman.stan@nmnh.si.edu; (JRB) Department of Biological Sciences, The George Washington University, Washington, D.C. 20052, U.S.A., e-mail: jrburns@ gwu.edu Abstract.—Two new species, Planaltina glandipedis and Planaltina britskii, glandulocaudine fishes of the family Characidae, tribe Diapomini, are described herein. They were collected from tributaries of the rio Grande and or the rio Tieté, both tributaries of the upper rio Parana in the State of Sao Paulo, Brazil. The previously described species, Planaltina myersi, until now known almost exclusively from the holotype, is here redescribed from 47 specimens taken from tributaries of the rio Corumba, a tributary of the upper rio Parana. The two new species differ most prominently from the previously known species by possession of three somewhat elongate scales along the dorsal border of the male’s caudal pouch opening. There is only one elongate scale in P. myersi. Planaltina glandipedis differs from P. britskii in having six or seven horizontal scale rows between the dorsal-fin origin and the lateral line versus five in the later species. The possible phylogenetic relationships among the three insem- inating species within the tribe Diapomini, subfamily Glandulocaudinae, are discussed along with the putative phylogenetic relationships of the Diapomini to the newly recognized glandulocaudine outgroup containing the characid gen- era Knodus and a new genus related to Aftonitus. Brief comments are presented concerning the geographical distribution and ecology of the species of Plan- altina. Resumo.—Duas novas espécies, Planaltina glandipedis e Planaltina britskii, peixes glandulocaudineos da familia Characidae, tribo Diapomini, sao aqui des- critas. Elas foram coletadas em afluentes do rio Grande e/ou do rio Tieté, ambos tributarios do rio Parana no Estado de Sao Paulo, Brasil. A espécie apenas descrita, Planaltina myersi, conhecida até 0 presente apenas a partir do hol6- tipo, é aqui redescrita com base em 47 exemplares originarios do rio Corumba, também tributario do rio Parana. As duas espécies novas diferem marcadamente da espécie previamente conhecida pela presenga de trés escamas algo alongadas ao longo da margem dorsal da abertura da bolsa glandular dos machos. Ha apenas uma escama alongada em P. myersi. Planaltina glandipedis difere de P. britskii pela presenga de seis ou sete séries horizontais de escamas entre a * This paper is part of the project, Diversidade de Peixes de Riachos e Cabeceiras da Bacia do Alto Rio Parana no Estado de Sao Paulo/Brasil, supported by the State of SAo Paulo Research Foundation (FAPESP)— The Biodiversity Virtual Institute Program (www.biotasp.org.br); coordinated by Dr. Ricardo Macedo Corréa e Castro; see Anonymous (2000). 558 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON origem da nadadeira dorsal e a linha lateral ao invés de cinco na Ultima espécie. As possiveis relagoes filogenéticas entre as trés espécies, todas inseminadoras dentro da tribo Diapomini, subfamilia Glandulocaudinae, sao discutidas jun- tamente com as relagoes filogenéticas tentativas de Diapomini com o grupo externo recentemente reconhecido de Glandulocaudinae constituido pelos gé- neros Knodus, Attonitus e um novo género relacionado a Attonitus da familia Characidae. Breves comentarios sao feitos sobre a distribuicao geografica e ecologia das espécies de Planaltina. Relatively little is known about Planal- tina Bohlke (1954:265) except from its original description based on Planaltina myerst described from a single damaged specimen collected through the use of dy- namite. This specimen, the holotype, was collected from “‘Sarandi brook, Planaltina, Goyaz, Brazil,” 21 September 1923 by Dr. Carl Ternetz. Bohlke (1954:265—267) was uncertain of the relationships of Planaltina, but considered it to possibly be among the ** ‘borderline’ genera (between the Tetra- gonopterinae and Glandulocaudinae of Ei- genmann) ...” Bohlke referred to these genera as Landonia, Argopleura, Acrobry- con, and Phenacobrycon. Bohlke (1954) doubted the monophyly of the Glandulo- caudinae and considered that these genera and the divergent members of the Glandu- locaudinae may have arisen “‘at different times from different members of such gen- eralized tetragonopterines as Bryconameri- cus.’ No additional specimens were ex- amined until Weitzman & Fink (1985:106) listed additional specimens with locality data. Weitzman et al. in Weitzman & Fink (1985:113) published a critique of the con- cept of the monophyly of the Glandulocau- dinae and concluded that there was insuf- ficient evidence to confirm or reject the monophyly of the subfamily. Burns et al. (1995:135—143) provided information that P. myersi 1S an inseminating species with aquasperm. Weitzman & Menezes (1998: 184) assigned this species to the Diapomini, which also included species of the characid genera Acrobrycon Eigenmann and Diapo- ma Cope, and, based on additional gross anatomical, histological, and sperm ultra- structural evidence, concluded that the Glandulocaudinae are probably monophy- letic. Weitzman & Menezes (1998) also il- lustrated the osteology of the caudal region and associated caudal organ of P. myersi. We provide a new definition of Planal- tina, redescribe P. myersi based on 47 spec- imens and discuss its putative relationships with the two new species described herein. We also briefly discuss the possible phylo- genetic relationships of the Diapomini to the Glandulocaudinae and to some of the characid species currently in the genera At- tonitus Vari & Ortega, Knodus Eigenmann and a species of a new genus Weitzman et al. (2004). Species of Attonitus, the new ge- nus and Knodus lack caudal organs as de- scribed for adult glandulocaudine males, but some species of Knodus and the three species of Aftonitus, and the species of the new genus were recently found to be insem- inating, Weitzman et al. (2004), and at least some of these have one or more of the syn- apomorphies of the primary and/or second- ary sexual systems of glandulocaudines that were used by Weitzman & Menezes (1998) to diagnose that subfamily. Furthermore, sexually active males of one of the new species described here, P. glandipedis, has numerous club cells at or near the surface of the skin of the pelvic fins and the anterior part of the anal fin whereas the species of Attonitus and the new genus possibly relat- ed to Aftonitus are found to have club cells at or near the surface of the skin in the an- terior part of the anal fin. These cells have been shown to secrete in a holocrine man- VOLUME 116, NUMBER 3 ner in the species of the new genus. See Weitzman et al. (2004) for a more detailed discussion of the taxonomic distribution of these apparent glandulocaudine and glan- dulocaudine outgroup synapomorphies. Those authors also discuss the possible phylogenetic relationships of species of the glandulocaudine tribes with the outgroup characid species currently in Knodus, the new genus, and Aftonitus as well as their apparent relatives not belonging to the Glandulocaudinae. The putative monophy- letic relationships among the glandulocau- dine tribes recognized by Weitzman & Me- nezes (1998) must again come into question and be reinvestigated in light of the newly recognized outgroup species for the puta- tive glandulocaudine genera and tribes. Methods and Materials Counts and measurements and methods of taking and presenting data follow Me- nezes & Weitzman (1990:382—383) unless otherwise stated. For the most part, statis- tical comparisons were calculated using SigmaStat 2.0 for Windows 95. Meristic and morphometric data of males and fe- males were treated separately in order to probe possible differences between the sex- es, but secondary sexual differences were detected only in morphometric data. Mor- phometric data were transformed into ratios of standard length, or head length when they were subunits of that length, and are presented separately for males and females in the tables and regression graphs. Although we summarize the statistical procedures used herein, see Weitzman & Malabarba (1999:2—4) and Weitzman & Palmer (1997:213—214) for more complete comments about their use. In making pair- wise statistical comparisons of counts, all population samples except those few with non-variable counts were found to be at least somewhat skewed about the mean. In running simple f-tests, all paired samples failed to pass the normality test and some failed to pass an equal variance test. Thus Nn Nn \O all pairwise comparisons of counts were made using Mann-Whitney rank sum tests, but also in all cases the results of simple ¢- tests were compared and although not test- ing the same statistical parameters, both were used to evaluate differences among the species. In most cases both kinds of tests were found to indicate the same over- all statistical results regarding a statistical test of null hypotheses of character similar- ities. However, in a few cases the f-test and the Mann-Whitney rank sum test provided opposing results. In these cases the statis- tical results were of marginal significance for both kinds of tests, and we opted to dis- regard the resulting hypotheses of differ- ences, especially considering the population sample size limitations. We utilized Tukey box plots for visual comparison of count distributions of the three species. In these plots the mean is rep- resented by a thick vertical bar, the median by a thin vertical line that is often the same as one of the vertical lateral borders of the boxes. Therefore, the numerical value of the median is given in the legend. The 25th and 75th percentiles are represented by the lateral borders of the boxes and indicate the respective values at which 25% of the sam- ples fall below and 25% fall above the lat- eral borders of the boxes. Error bars rep- resent the 10th and 90th percentile points, and the circles represent the 5th and 95th percentiles. Extremes, not shown in the plots, are given in the text. Regarding analyses for differences in sexual dimorphism in body ratios we exe- cuted a series of tests and present a few graphs using regression analyses. We used linear regression analysis, even though the statistics derived from it may rarely be a fully accurate representation of the more or less curvilinear mean growth lines ex- pressed by the data. Nevertheless linear re- gression graphs provide useful comparative visual presentations of the relative morpho- logical parameters of the population sam- ples being compared as well as reasonable population growth patterns. In using the lin- 560 ear regression model for comparing body ratios of population samples, statistical tests of a null hypothesis of character similarities for ratios of body measurements follow Weitzman & Palmer (1997:213-—214) and employed linear multiple regression analy- ses using dummy (indicator) variables. See Glantz & Slinker (1990:69—72) for an ex- planation. Also, analyses of covariance (ANCOVA) using BIOM-PC, version 2, Exeter Publishing Ltd. were performed. Such tests were done using base 10 log transformed data. However, it was found that in the cases discussed here the results of such tests did not need publication be- cause in those cases used to search for pop- ulation sample differences, the regression graphs, the mean regression lines, and 95% confidence intervals demonstrated clear sta- tistical differences in ratios of body mea- surements. Even though statistical tests were done using base 10 log transformed data, neither the regression graphs, their ac- companying regression equations, nor the r7 values presented here are derived from base 10 log transformed data because presenta- tion of non-transformed data in graphs is more visually accessible to the reader. In regression analyses a series of diagnostic tests using SigmaStat Version 2.0 were run to determine the suitability of the data for the linear regression model. In all cases the normality tests and the constant variance tests were passed. For histological analyses, tissues were re- moved from specimens initially fixed in 10% formalin and subsequently transferred to 70% ethanol. Tissues containing bony el- ements were first decalcified. Some tissues were then dehydrated in an ethanol series, embedded in paraffin, sectioned at 5—10 yum, and stained with modified Masson’s trichrome (Schreibman 1964:219). Other tissues were dehydrated in an ethanol series to 95% ethanol and infiltrated with and em- bedded in glycol methacrylate. Sections were cut at 2.0—3.5 wm and stained with toluidine blue and periodic acid-Schiff re- agent (PAS)/hematoxylin (Quintero-Hunter PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON et al. 1991:170). Measurements of testis ar- eas were taken from mid sagittal sections as described by Burns et al. (1995:132). For counts recorded in the descriptions, those of the holotype are provided first fol- lowed in parentheses by the mean of the population sample, (or median when the data are nonparametric), range, and the total number of specimens counted. The terminations of the dorsal and anal fins are morphologically different and are treated as follows. The posterior branched ray of the dorsal fin has a single pterygiop- hore at its base, 1.e., the ray is not split to its base and is counted as |, but when it is rarely split to its base, then its last two rays are counted as | because there is only 1 pterygiophore at the base of these rays. In the anal fin the most posterior pterygiop- hore has its ray almost invariably split to its base and these two rays are counted as one. The word pouch is used for that sac-like area internal to the modified caudal-fin scale or scales, called pouch scales, of the usually adult, sexually active males of glan- dulocaudine characids. The pouch opening to the surrounding water occurs along the posteroventral border of the scale or scales. In almost all case in males the pouch open- ing is provided with modified secretory cells presumably producing a pheromone. In diapomins both sexes have pouch scales. See Weitzman & Fink (1985) and Weitz- man & Menezes (1998) for further expla- nation. Specimens examined for this research are deposited in the Museu de Zoologia, Univ- ersidade de Sao Paulo (MZUSP); Museu Nacional, Universidade Federal do Rio de Janeiro (MNRJ); Laboratoério de Ictiologia de Ribeirao Preto, Faculdade de Filosofia Ciéncias e Letras de Ribeirao Preto, Univ- ersidade de Sao Paulo (LIRP); Departamen- to de Zoologia, Universidade Estadual Pau- lista, Campus de Sao José do Rio Preto, Sao Paulo (DZSJRP); and the Smithsonian In- stitution, National Museum of Natural His- tory (USNM). Comparative glandulocaudi- VOLUME 116, NUMBER 3 ne and other characid specimens used in this report are listed in Weitzman & Fink (1985: Appendix 1), Weitzman et al. (1994), Burns et al. (1995, 1997), Weitz- man & Menezes (1998:176), and Weitzman et al. (2004). Planaltina Bohlke Planaltina Bohlke, 1954:265 (type species: P. myersi Bohlke, 1954:267 by monotypy and original designation). Definition.—1) There are fewer scales forming the dorsal border of the caudal pouch in Planaltina (1-2) than in Acrob- rycon (4-7 or 8) and Diapoma (3-6). 2) In Planaltina one of the scales forming the dorsal border of the pouch opening is elon- gate and forms most of that border, whereas in Acrobrycon and Diapoma, even when one of the dorsal border scales of the pouch opening is somewhat elongate or larger, its actual contribution to the length of the dor- sal pouch border is litthke more than that of the other dorsal border scales of the pouch opening. 3) The three species of Planaltina produce aquasperm, the nuclei of which vary in shape from spherical to slightly el- liptical, whereas both Acrobrycon and Dia- poma have elongate sperm cell nuclei. Key to the species of Planaltina The characters used in this key, except those in couplet la, are not necessarily syn- apomorphies and the key is not intended to reflect phylogenetic relationships, only in- formation for species identification. The key assumes that at least adult male speci- mens are available and have a caudal organ consisting of hypertrophied scales enclos- ing a pouch that is in part open to the sur- rounding water and that the specimens do not fit the diagnosis of the Cheirodontinae proposed by Malabarba (1998:199—205). la. Male and female caudal organs as in- dicated by caudal squamation nearly equivalent in size and morphological development; three or more scales im- mediately ventral to lateral-line series extend posteriorly to form dorsal border Ompouch opening |... ac. Diapomini (2) lb. Female caudal organs as expressed by caudal squamation never as hypertro- phied as that of males; arrangement of scales bordering pouch opening not as above other Glandulocaudinae 2a. Dorsal border of pouch opening formed by one or two enlarged and elongate dorsal scales; sperm cell nuclei spheri- cal to slightly elliptical (aquasperm) . . 5g Oe a Oe a Planaltina (3) 2b. Dorsal border of pouch opening with- out especially enlarged scales or one somewhat enlarged scale present, but its perimeter entering pouch border not especially large; sperm-cell nucleus somewhat elongate Chea a ene Acrobrycon and Diapoma 3a. One much enlarged and elongate scale along dorsal border of pouch opening; adipose fin present .... Planaltina myersi 3b. Three enlarged and somewhat elongate scales along dorsal border of pouch opening; adipose fin present or absent (4) 4a. Six or seven scale rows between dorsal- fin origin and lateral line; 5 scale rows between anal-fin origin and lateral line; 16—17 scale rows around caudal pedun- cle (X = 16.2); 17—20 predorsal scales (X = 18.0); adipose fin absent ee nS eee, a2 © a9 eae Planaltina glandipedis 4b. Five scale rows between dorsal-fin or- igin and lateral line; 4 scale rows be- tween anal-fin origin and lateral line; 12—14 scale rows around caudal pedun- cle (X = 13.1); 14-16 predorsal scales Cx — lS) = adipwse fimypresent iene ars Planaltina britskii Planaltina myersi Bohlke Figs. 1—3, Table 1 Planaltina myersi Bohlke, 1954:267, orig- inal description, Brazil, ““Goyaz, Sarandi brook, Planaltina,’’ now in Distrito Fed- eral.—Weitzman & Fink, 1985:106—107, listed specimens from several locali- ties.—Burns et al. (1995:135—143) pre- sented evidence that species is insemi- nating with aquasperm.—Weitzman & 562 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. |. Planaltina myersit, USNM 221202, adult male above, SL 35.9 mm and adult female, below, SL 32.0 mm; corrego Pipiripau near Planaltina, Distrito Federal, Brazil, about 15°40'S, 47°39'W. Table 1.—Morphometrics of Planaltina myersi. Standard length is expressed in mm; measurements through head length are percentages of standard length; the last four entries are percentages of head length. Minimum, maximum, n, X, and SD are based on the holotype, SU 18636 (now at CAS) and the following specimen lots: MNRJ 10634, MNRJ 10635, USNM 363088, USNM 2210202, USNM 236416, USNM 258458, USNM 278966, USNM 278986, USNM 278987, USNM 278989, and MUSP 63588. Solo. Males Females type n Range xX SD n Range xX SD Standard length (mm) SOK WAP B= 3N310) — 3)27/ 33. -19.0-43.0 30.7 Body depth at dorsal fin origin DAI VA 222-262 24:0 MO) 33) 2128 352 aes Snout to dorsal-fin origin 54.3 14 54.3-61.6 59.2 1.8 33.4653. 1-015) 5) SON amales) Snout to pectoral-fin origin 2555 14 245-26.2 25.3 0.6 330 22-223 One Ole Snout to pelvic-fin origin 46.2 14 45.1-47.1 45.8 0.6 33. 46.7-50.0 464 1.4 Snout to anal-fin origin Soa Il4b Des HsOl D2 iO 33° 57.1-64:9" GOIS220 Caudal peduncle depth IO 7 4 > OSS OO Os) 33°) OF84=10l6F* 09:80 Caudal peduncle length W203 WAS NOS WTO) 33) eS 1 els, SIS ieee Pectoral-fin length 209 V4 201-2422 22), WkOls 33a) U8:8—25 35ers Pelvic-fin length 14.4 14 12.5-16.1 15.1 1.1 33 12.0-15.6 14.1 0.7 Dorsal-fin base length IO 4 OSQSIZ5 Om Os 33" 01938=12,9) OieOES Dorsal-fin height OFS Seek es 20D OES B38) NS 2=2 1 See SSE Sees Anal-fin base length 299 PAR 7:6—333 eS UlOer Es 33) 2610=3210 eZ StO uae les Anal-fin lobe length Ws WA NOM I O),7/ 33) NG 19:6 2) TshiieleO Eye to dorsal-fin origin 465 14 44647.9 46.1 1.0 33 41.7-48.6 45.7 1.5 Dorsal-fin origin to caudal-fin base 43.55 14 40.1-43.9 42.2 1.1 33. (38: 7-47.7" 42 27 Io Bony head length 23.4 14 226-254 23.7 0.8 33° 9205=26.47 9 245 el Horizontal eye diameter BS Ar 14AS 35: 0=37 7 SO Oy, 83) 933.712. Or ON eneles Snout length DAMN NWA NI VS PN 33" (li. 8=23-4.5 22 le 2 ee Least interorbital width 31.4 14 31.4-36.1 33.8 14 33) Diloll=3K333) DDD ILD Upper jaw length 43.0 14 40.0-45.9 43.1 1.4 33 40.3-47.7 43.8 1.5 VOLUME 116, NUMBER 3 563 Eis 2. Planaltina myersi, USNM 221202, adult male, SL 35.9 mm; corrego Pipiripau near Planaltina, Distrito Federal, Brazil, about 15°40’S, 47°39'W. Displays pigment pattern of head and anterior part of body. Menezes (1998:184) discussed relation- ships and assigned it to Diapomini. Specimens examined.—All specimens are from Brazil, Distrito Federal, drainage basin of rio Corumba, tributary of upper rio Parana. Holotype.—SU 18636 (now at CAS), 1 male, 36.8 mm SL. Type locality: Brazil: “Goyaz, Sarandi brook, Planaltina”’ (ap- proximately 15°40'S, 47°45’W). Now in Distrito Federal, see discussion below re- garding type locality. Additional specimens.—Brazil: Distrito Federal. MNRJ 10634, 24 (2 mature males, from 30.5 to 31.5 mm SL; 22 immatures of either sex to mature females, from 22.3 to 36.6 mm SL), corrego Fumal, where cross- es road between Brasilia and Planaltina, near Planaltina, about 15°20’S, 47°50’W, 11 Apr. 1982, L.E. de Macedo Cordoso. MNRJ 10635, 13 Gncluding | c&s male, 33.3 mm SL; | c&s female, 36.4 mm SL), corrego Fumal, where it crosses road between Bra- silia and Planaltina, near Planaltina, about LS OCS Ave SOW eas Aue 98, le Me, Ede Macedo Cordoso. USNM 363088, 1 (ma- ture male, 46.0 mm SL), rio Pipiripau near Planaltina, about 15°40’S, 47°39'W, 30 Sept. 1977, N. Menezes & party. USNM 221202, 6 (1 mature male, 35.9 mm; 5 ma- ture females, 32.0—43.1 mm SL), rio Pipi- ripau near Planaltina, about 15°40’S, 47°39'W, 30 Sept. 1977. N. Menezes & par- ty. USNM 236416, 3 Gmmatures to | ma- ture female, 19.0—39.7 mm SL), corrego Vargem de Tras, rio Pipiripau, near Plan- altina, about 15°40’S, 47°39'W, 1 Jun. 1979, N. Menezes & E. Bastos. USNM 258458, 1 (mature female, c&s, 41.0 mm SL), rio Pipiripau near Planaltina, about 15°40’S, 47°39'W, 19 Jan. 1976, E.C. Calaf. USNM 278966, 8 (7 mature sexually active males, 35.5 6.5) mm Sl i imaturestemale, 37-8 mm SL), corrego Papuda, tributary of rio Sao Bartolomeu, about 40 km south east of Brasilia, 1 Feb. 1985, M. Ribeiro & party. USNM 278985, 1 (mature male, 36.4 mm SL), rio Pipiripau near Planaltina, about 4 Ons 47, SO We LOSS Ma Ribeiro: USNM 278986, 1 (mature female, 39.6 mm SL), coérrego Papuda, tributary to rio Sao Bartolomeu, 28 Aug. 1985, M. Ribeiro & 564 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON R.C. Mendonga. USNM 278987, | (mature male, 37.1 mm SL), cérrego Forquilha, trib- utary to rio Sao Bartolomeu, Sept. 1985, M. Ribeiro. USNM 278989, 3 (mature females, 28.7—-37.4 mm SL), rio Taboca, tributary to rio Sao Bartolomeu, Sept. 1985, M. Ribeiro & J. Dalmaco. MZUSP 63588, 3 (mature females, 34.3—41.3 mm SL), cérrego Var- gem de Tras, rio Pipiripau, near Planaltina, 1 Jun. 1979, N. Menezes & E. Bastos. Note.—Bohlke (1954:267) with the aid of other Dr. Carl Ternetz localities of about the same date discussed the type locality ‘as a small brook somewhere along the Chapadao do Sarandy, lying between the headwaters of the Rio Maranhao (which empties into the Tocantins) and those of the Rio Sao Bartholomeo [= rio Sao Bartolo- meu] (which empties into the [rio] Cor- umba and then into the [rio] Paranahyba [= e 8 ws. 4 Riga Or rio Paranaiba].’’> Bohlke continued: ““Thus the particular brook which Ternetz collected may drain into the Amazon system to the north or into the Parana to the south, but more likely the latter ...’’ Subsequent to the original description, several collections were made in this region as listed above and in Weitzman & Fink (1985:106). Plan- altina myersi so far has been found only in the upper rio Parana system. Definition.—Following features distin- guish Planaltina myersi from P. glandipe- dis and P. britskii: One much enlarged and elongate scale along dorsal border of pouch opening; adipose fin present. In addition, anterior teeth of inner tooth row of premax- illa of P. myersi pentacuspid, and dentary with its anterior large teeth pentacuspid or quadricuspid. Lower jaw not extending be- yond upper. Planaltina glandipedis with tri- . . a ee eee, . AO 4 . . oe y OO PO » oe dg < vy v o« ° & < ve . > » AS on sae a; ore te me tave hs Be eee : < vse Koes 5 ALO mae AO ag « ea 4 . FF Ya e pe . . Mae? %e Planaltina myersi, MNRJ 10635, c&s, sexually active male, SL 33.3 mm. Premaxilla, maxilla and mandible, left side, external view. Note relative positions of maxilla and premaxilla with dentary not normal so that all jaw teeth can be illustrated. VOLUME 116, NUMBER 3 a O04 O70 Ro 565 rat oN scar egtese teens ceaeetet ELSE se 7 . 2. FOP A ay tara safes Lo 20 2 46 BOY 3 BIS eee he Aon 2 .? 95 9.99.9 x 2 6 eee Dis Seon A 5 eigen . o*e 0: 0,045 Pe Sule Seer. wt th ~ ete Fig. 4. Planaltina glandipedis, new species, paratype, USNM 362135, sexually active male, c&s, SL 24.3 mm. Premaxilla, maxilla and mandible, left side, external view. Note relative positions and premaxilla with dentary not normal so that all jaw teeth can be illustrated. cuspid teeth in inner row of premaxilla; dentary with tricuspid anterior large teeth, lower jaw protrudes somewhat beyond up- per jaw. Planaltina britskii with teeth of in- ner row of premaxilla and large anterior teeth of dentary pentacuspid and lower jaw included by upper (Figs. 3—5, but note: in drawings upper and lower jaws reproduced from independent drawings and not in a natural position in relation to one another, thus some dentary teeth not hidden by up- per jaw teeth). Figures 6—13 illustrate Tukey Box Plot distributions of population sample data for species of Planaltina. Captions of these fig- ures provide results of Mann-Whitney rank sum tests regarding differences and similar- ities among P. myersi, P. britskii, and P. glandipedis in number of lateral-series scales, Fig. 6; number of predorsal scales, Fig. 7; number of branched anal-fin rays, Fig. 8; total number of vertebrae including those of Weberian apparatus, Fig. 9; least number of scale rows around caudal pedun- cle, Fig. 10; number of lower-limb gill rak- ers, Fig. 11; number of scale rows between dorsal-fin origin and lateral line, Fig. 12; and number of scale rows between anal-fin origin and lateral line, Fig. 13. Description.—Morphometrics of the ho- lotype and additional specimens presented in Table 1. Tables, graphs, and descriptions 566 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON RieaoE Premaxilla, maxilla and mandible, left side, external view. Note relative positions of maxilla and premaxilla with dentary not normal so that all jaw teeth can be illustrated. provided below based on combined lots of this species because no statistical differenc- es were found among examined same-sex population samples. All collection localities geographically close and in one drainage system, rio Corumba, upper rio Parana. Body large, compressed and moderately elongate; greatest body depth between snout tip and dorsal-fin origin, near origin of pelvic fin. Dorsal body profile somewhat convex between nape and just anterior to dorsal-fin origin, only slightly elevated at dorsal-fin origin. Head profile to snout tip slightly convex. Snout rounded. Body pro- file nearly straight along dorsal-fin base and slightly concave in region of caudal pedun- Planaltina britskii, new species, paratype, MZUSP 62758, c&s, sexually active male, SL 35.6 mm. cle. Dorsal-fin origin nearer to caudal-fin base than snout tip. Ventral body profile convex from tip of lower jaw to anal-fin origin, nearly straight along anal-fin base and slightly concave from end of anal fin to origin of procurrent caudal-fin rays. Lower jaw terminal or only slightly pro- truding anterior to upper jaw. Mouth gape somewhat inclined posteroventrally towards mandibular joint. Maxilla extending poste- riorly beyond vertical line passing through anterior border of orbit, but falling short of reaching a vertical line passing through an- terior border of pupil of eye (see Fig. 2). Dorsal-fin rays 11, 8 in 44 of 45 speci- mens including holotype, I, 7 in one spec- VOLUME 116, NUMBER 3 Number of lateral series scales S1/ 38 39 40 ne 66 © n=27 n= 44 Si 38 39 40 567 41 42 43 @ P. britskii O P. glandipedis P. myersi 41 42 43 Fig. 6. Tukey box plots showing the statistical distribution of number of lateral-series scales for species of Planaltina. No significant difference in number of lateral series scales was found using a Mann-Whitney rank sum test for P. myersi and P. glandipedis. However, using this test a statistically significant difference was found between P. myersi and P. britskii (T = 3021.00, P = <0.001) and between P. glandipedis and P. britskii (T = 1733.0, P = <0.001). The median for P. myersi and P. glandipedis is 41 and for P. britskii it is 40. imen. Posterior most ray unbranched in all but one specimen in which count appears as 11, 9, but is actually 11, 8, posteriormost ray being branched, n = 45. Adipose fin present. Anal-fin rays iv, 21 (iv or v, usually v, branched rays X = 20.8, range 19-23, n = 45). Moderately developed anterior anal- fin lobe including anterior unbranched rays and first 4—5 branched rays. Anal fin of sex- ually mature males with bilateral hooks on fifth unbranched and anterior 9 or 10 branched rays distributed as in Fig. 14. Pec- toral-fin rays 1, 11 (anterior unbranched ray iin all specimens, posterior ray actually un- branched, but counted as a branched ray), branched rays X = 10.3, range 9-11, n = 45. Posterior tips of longest pectoral-fin rays not reaching pelvic-fin origin; pectoral fin about same length in both sexes, without hooks. Pelvic-fin rays 1, 6, n = 45 (posterior most ray unbranched). Sexually mature males with hooks present on rays of pelvic fin, distributed as in Fig. 15. Number of hooks per ray varying among different males, but usually approximately as shown in Fig. 15; a mature male (30.5 mm SL) with O hooks on first ray, 8 on second, 10 on third, 13 on fourth, 15 on fifth, 16 on sixth, 3 on seventh ray, all on ventral sur- face of right side pelvic fin. Pelvic fins of adult males longer than those of adult fe- males (see Table 1 and Fig. 15); distal tips of longest pelvic-fin rays extending to, or slightly beyond, anal-fin origin in adult males, but pelvic fins falling short of ante- rior origin of anal fin in adult females. Compare Figs. la, b. See also discussion under “Sexual dimorphism.” Scales cy- cloid, with few radu, often 3—6 along pos- terior exposed field, more numerlous, to approximately 10, on enlarged scale bor- dering dorsal border of pouch opening. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Number of predorsal scales 568 As 14 15 16 17 n= © n= 41 n= 44 @ As 14 15 16 17 ae, I 18 19 20 21 P. britskii Se) P. glandipedis P. myersi 18 19 20 21 Tukey box plots of number of predorsal scales for species of Planaltina. A Mann-Whitney rank sum tests found no significant difference in the number of those scales for P. myersi and P. britskii, but a significant difference was found between P. myersi and P. glandipedis (T = 2633.000, P = <0.001) and between P. glandipedis and P. britskii (T = 3854.000, P = <0.001). The median for P. myersi = 15, for P. glandipedis = 18 (= the same as mean), and for P. britskii = 15. Lateral line complete, perforated scales 42 (X = 41.0, range 39—43, n = 44). Pre- dorsal scales 15 (X = 15.3, range 13-17, n = 44). Scale rows between dorsal-fin origin and lateral line 6 (X = 5.9, range 5-6, n = 45). Scale rows between anal-fin origin and lateral line 5 (X = 4.9, range 4—5, n = 45). Scale rows around caudal peduncle 16 (X = 15.5, range 15-16, n = 45). Premaxillary teeth in two distinct rows (see Fig. 3). Larger teeth pentacuspid; smaller teeth tricuspid or with less well-de- veloped and sometimes vestigial additional lateral cusps. Outer row teeth 3 (X = 3.2, range 2—4, n = 47). Inner row teeth 4 (X = 4.0, range 4—5, but only one specimen had 5 teeth, n = 47). Maxillary teeth 1 (number of teeth does not appear to in- crease ontogenetically, X = 2.1, range 1-3, 47). Maxillary teeth tricuspid, occa- sionally bicuspid, (anterior tooth larger than eS remaining teeth). Dentary with 4 large an- terior pentacuspid or sometimes quadricus- pid teeth, nm = 47 and 7 smaller posterior teeth (X = 7.0, range 5—9, n = 46). Pre- maxillary, maxillary, and dentary teeth compressed so that flattened surfaces face both externally and internally. Vertebrae 39 (X = 39.3, range 38—41, n = 54). Upper limb gill-rakers 5 (X = 5.7, range 4—7, n = 47); lower limb gill-rakers 10 (X = 11.1, range 10—13, n = 47). Bran- chiostegal rays 4 in two cleared and stained specimens, 3 rays originating from anterior ceratohyal and | ray from posterior cera- tohyal. Color in alcohol.—Males and females with approximately the same color pattern (Figs. la, b, 2). Body pale to light brown and a little darker dorsally than ventrally. Dorsal body surface dark due to scattered dark chromatophores and a clear reticulate VOLUME 116, NUMBER 3 569 Number of branched anal-fin rays 23 24 20 21 22 n= 90 n= 48 P. britskii P. glandipedis n= 45 20 21 22 Fig. 38: P. myersi 23 24 Tukey box plots of number of branched anal-fin rays for species of Planaltina. A significant difference in number of these fin rays was found among all three species using a Mann-Whitney rank sum test. Between P. myersi and P. britskii (T = 2449.000, P = 0.004), between P. myersi and P. glandipedis (T = 1376.000, P = <0.001) and between P. glandipedis and P. britskii (T = 4387.000, P. = <0.001). The median for P. myersi pattern of dark chromatophores associated with scale borders. Fewer dark chromato- phores occurring on ventral posterior part of body where scattered concentration of chromatophores presents on body dorsal to anterior half of anal fin, most obvious in males. Dark lateral body stripe present, but mostly obscured by guanine pigment in freshly caught specimens or in specimens fixed in ethyl alcohol. This darkly pig- mented stripe, when guanine pigment de- stroyed by formalin, extending from pos- terior part of dorsal opercular region to cau- dal-fin base and onto anterior part of me- dian caudal-fin rays. Dark pigmentation of lateral stripe mostly concentrated along posterior two thirds of body and about equally evident on males and females. Also extending onto anterior part of median cau- dal-fin rays. Other median fins relatively free of pigment, but anal fin with consid- erable amount of dark pigment on the mem- 21, for P. glandipedis = 22 and for P. britskii = 21. branes between rays. Pectoral and pelvic fins pale. Head dark dorsally and dark around mouth. Circumorbital bones and op- ercle silvery with very few scattered dark chromatophores. Sexual dimorphism, reproductive mode and gonad anatomy.—The females lack the pelvic-fin and anal-fin hooks present in males as described above (see Figs. 14, 15). The pelvic fins are longer in adult males than in adult females. Our data indicate that sexual maturity is reached by about or at least by 30 mm SL and by this length the male’s pelvic fins have also reached their proportionately greatest length relative to that of the females. Comparisons of Fig. 16 of P. myersi, Fig. 17 of P. glandipedis and Fig. 18 of P. britksii indicate that the rela- tive pelvic-fin length in males of P. myersi and P. glandipedis varies less than in P. britskii. Although our population samples are represented by males and females of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Number of vertebrae including Weberian apparatus 570 Si 38 39 n= 28 n= 40 n= 54 BY 38 39 40 41 42 P. britskii P. g;andipedis P. myersi 40 41 42 Fig. 9. Tukey box plots showing the distribution of number of vertebrae for species of Planaltina. A Mann- Whitney rank sum test found no significant difference between number of vertebrae in P. myersi and P. glan- dipedis, but a significant difference (T = 612.500, P = <0.001) was found between P. britskii and P. glandipedis as well as P. britskii and P. myersi (T = 740.00, P = <0.001). Median for P. myersi and P. glandipedis = 39, and for P. britskii = 38. about equal length ranges for each sex, rel- atively few are fully sexually mature judg- ing from visual inspection of the gonad state of development in some specimens. The scales bordering the dorsal margin of the mature male pouch opening (see Fig. 19) are more elongate than those of the ma- ture females, but it is impossible to predict if greater elongation of scales dorsal to the pouch opening might be attained at com- paratively larger sizes in females. A gill gland as reported present in Cor- ynopoma riisei Gill by Burns and Weitzman (1996) was reported as present by Bush- mann et al. (2002: table 1) in a sexually mature male specimen of P. myersi (USNM 278966, 38.3 mm SL), but absent in a sex- ually mature female specimen (USNM 278966, 43.0 mm SL). In maturing males the presence of such a gland is uncertain. One maturing male specimen (MNRJ 10635, 33.0 mm SL) appeared to have sev- eral fused distal tips of anterior gill fila- ments, but no obvious gill gland was pre- sent. Histological analysis revealed that the epithelium covering both the anal- and pel- vic-fin rays was thicker in males than in females. Figure 20 shows the thickened ep- ithelium along a pelvic-fin ray of a mature male (USNM 278966, 38.3 mm SL). Abun- dant club cells, some close to the surface, were present in the anal- and pelvic-fin ep- ithelia of both sexes. Histological analysis of ovaries of two mature females (MNRJ 10634, 30.9 mm SL; USNM 278989, 35.0 mm SL) revealed the presence of abundant spermatozoa (see Fig. 21) and large mature oocytes within the ovarian cavity (see also Burns et al., 1995: table 3 and fig. 1A). Longitudinal sections through the testes revealed a pos- VOLUME 116, NUMBER 3 571 Least number of horizontal scale rows around caudal peduncle 11 12 18 14 n=55 n = 26 n= 41 11 12 13 14 15 16 7 18 P. britskii P. glandipedis P. myersi 15 16 17 18 Fig. 10. Tukey box plots of least number of horizontal scale rows around caudal peduncle for species of Planaltina. A Mann-Whitney rank sum test indicated a statistically significant difference among all species. That between P. britskii and P. glandipedis and between P. britskii and P. myersi is obvious comparing the graphs, but that between P. glandipedis and P. myersi, although apparently statistically significant (T = 1186.00, P = <0.001) is not particularly useful for identification of the species. Median for P. myersi and P. glandipedis = 16, and for P. britskii =-13. terior sperm storage area occupying 44.2% of the total testis area in a mid-sagittal testis section as was reported and discussed by Burns et al. (1995:134, fig. 2). Burns et al. (1995: table 3) also reported that the sperm cells are typical aquasperm, but with the nucleus spherical to slightly “‘deformed.”’ There appear to be no sperm cell modifi- cations as found in typical glandulocaudi- nes, but ultrastructure examination was not possible because only formalin-fixed spec- imens were available. Planaltina glandipedis, new species lang, 22, Walolle 2 Specimens examined.—All specimens from Brazil, Estado de Sao Paulo, Munici- pio de Corumbatai (except where noted), and tributaries to the drainage basin of the rio Tieté, a tributary of the upper rio Para- na. Holotype.—MZUSP. 62752, male, 26.2 mm SL, Corumbatai, rio Corumbatai, ap- proximately 22°13’S, 47°38'W, 19 Jan. 1976; H. A. Britsk1. Paratypes.—MZUSP 62753, 11 (23.0— 28.7 mm SL) and USNM 362135, 11 (Gm- matures to adults, sex undetermined, both collected with holotype, 20.0—27.0 mm SL, 2 specimens, male 26.3 mm SL and female 27.0 mm SL used for histological informa- tion). Following 5 lots of immature to adult paratypes collected at Corumbatai, rio Cor- umbatai, approximately 22°13'S, 47°38'W. MZUSP 62754, 7 (19.7—23.0 mm SL). USNM 362136, 5 (21.7—25.3 mm SL), 25 Jan. 1976, H. A. Britski. MZUSP 62755, 1 (28.5 mm SL), Dec. 1962, H. A. Britsk1. USNM 362137, 2 (25.8 and 26.5 mm SL), 372 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Number of lower limb gill rakers S) 10 11 12 13 14 n= 48 Pe @ P. glandipedis n= 47 © P. myersi SS) 10 4 12 13 14 Fig. 11. Tukey box plots of number of lower limb gill rakers for species of Planaltina. The differences in the number of lower limb gill rakers among the species is relatively small and provide little if any useful information for identification of the species. However, a nonparametric Mann-Whitney rank sum test found a statistical difference between P. britskii and P. myersi (T = 2192.500, P = <0.001) and may indicate a genetic difference in these species. However, using the same type of test, differences between the other two species combinations suggest that resulting differences should be interpreted cautiously and considering the numbers of specimens used we reserve judgment about any possible genetic differences in these species. Differences for P. britskii and P. glandipedis are T = 1936.5, P = 0.017 and for P. glandipedis and P. myersi are T = 2752.0, P = 0.009. Median for P. myersi and P. glandipedis = 11, and for P. britskii = 12. 2 Nov. 1963, H. A. Britski & A. E. Gomes. MZUSP 62756, 3 (matures 25.5—26.5 mm SL), 2 Nov. 1963, H. A. Britski & A. E. Gomes. DZSJRP 671, 4 (22.0—25.0 mm SL) and MZUSP 63690, 3 (matures 25.5—26.5 mm SL), Brazil, Sao Paulo, Municipio de Brotas, basin of rio Jacaré-Pepira, approxi- mately 22°17'S, 48°09’W, Jan. 1988, W. Bar- rela. Definition.—Planaltina glandipedis is sympatric with P. britskii, but can be easily distinguished from that species and the al- lopatric P. myersi by lacking an adipose fin and by characters provided in the key. Fur- thermore, P. glandipedis has tricuspid jaw teeth (see Figs. 3—5) and the lower jaw pro- trudes somewhat beyond upper, whereas in P. britskii the teeth of the inner row of the premaxilla and the large anterior teeth on the dentary are pentacuspid, and the lower jaw is included by the upper jaw. Description.—Morphometrics of the ho- lotype and paratypes presented in Table 2. Description based on all lots of this species examined because no statistical differences found among lots from different localities. Body relatively small, compressed and moderately elongate; greatest body depth between snout tip and dorsal-fin origin, near distal tip of adpressed pelvic fin. Dorsal body profile elevated at dorsal-fin origin, slightly convex from this point to snout tip, nearly straight along dorsal-fin base and slightly concave dorsal to caudal peduncle. Dorsal-fin origin nearer to caudal-fin base than to snout tip. Ventral body profile con- VOLUME 116, NUMBER 3 3/8: Number of scale rows between dorsal-fin origin and lateral line a 5 6 7 8 n=78 P. britskii n= 41 & P. glandipedis n= 45 P. myersi 4 S 6 7 8 Fig. 12. Tukey box plots of number of scale rows between dorsal-fin origin and lateral line for species of Planaltina. A nonparametric Mann-Whitney rank sum test detected a statistically significant difference between P. britskii and both other species, P. britskii and P. glandipedis (T = 4059.000, P = <0.001) and for P. britskii and P. myersi (T = 4350.000, P = <0.001). The same test for differences between P. myersi and P. glandipedis indicated no significant difference. Median for P. myersi and P. glandipedis = 6, and for P. britskii = 5. vex from tip of lower jaw to anal-fin origin, nearly straight along anal-fin base and con- cave from end of anal fin to origin of pro- current caudal-fin rays. Lower jaw protruding, anterior to upper jaw. Mouth gape inclined posteroventrally towards mandibular joint. Maxilla extend- ing posteriorly beyond vertical line passing through anterior border of orbit, but not reaching vertical line passing through an- terior border of eye pupil. Dorsal-fin rays 1, 8, n = 47. Adipose fin absent. Anal-fin rays iv, 22 (iv or v, usually iv, branched rays X = 2.2, range 20-25, n = 48). Anterior anal-fin lobe moderately developed and includes fourth or fifth un- branched ray and first 5—6 branched rays. Anal fin of sexually mature males with bi- lateral hooks on fourth or fifth unbranched and anterior 10 branched rays (Fig. 23). Anal fin of sexually active males with glan- dular cells on fin’s anterior region (see “Sexual dimorphism, ...” below and Fig. 24). Pectoral-fin rays i, 9 (anterior un- branched ray 1 in all specimens) branched rays X = 9.7, range 9-11, n = 48. Distal tips of longest pectoral-fin rays extending posteriorly beyond pelvic-fin origin. Pec- toral fins of about equal length in both sex- es and without hooks. Pelvic-fin rays i, 6 (posteriormost ray unbranched in most specimens, branched in two adult males, but always considered as “‘branched” in counts, nm = 48). Sexually mature males with hooks present on rays of pelvic fin, distributed as in Fig. 25. Number of hooks per ray in males varies; mature male (26.0 mm SL) with 7 hooks on first, 18 on sec- ond, 16 on third, 10 on fourth, and 13 on fifth branched rays of right pelvic fin. Pel- vic fins of adult males somewhat longer than those of most adult females (see Fig. 574 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Number of scale rows between anal-fin origin and lateral line 3 4. 5 6 n= 78 P. britksii n= 41 P. glandipedis n= 45 P. myersi 3 4 5 6 Fig. 13. Tukey box plots of number of scale rows between anal-fin origin and lateral line for species of Planaltina. A nonparametric Mann-Whitney rank sum test found a statistically significant difference between P. britskii and P. myersi (T = 4311.00, P = <0.001). The same test for P. myersi and P. glandipedis indicated no significant difference. The difference between P. britskii (all 4) and P. glandipedis (all 5) was absolute. Median for P. myersi = 5. 17); distal tips of longest pelvic-fin rays ex- tending to anal-fin origin in adult males but not in adult females. Males and females with pads of glandular tissue on pelvic-fin rays (see Figs. 26 and 27 and discussion in “Sexual dimorphism, ...” below). Scales cycloid with few radii along pos- terior border, more numerous on enlarged scales bordering and covering caudal pouch of male and female (see Fig. 28a, b). Lateral line complete, perforated scales 41 (X = 40.7, range 40—44, n = 27). Pre- dorsal scales 18 (X = 18, range 17-20, n = 41). Scale rows between dorsal-fin origin and lateral line 6 (X = 6.1, range 6-7, n = 41). Scale rows between anal-fin origin and lateral line 5 in all specimens (n = 41). Scale rows around caudal peduncle 16 (X = 16.2, range 16-17, n = 26). Premaxillary teeth in two distinct rows. Larger teeth distinctly tricuspid, smaller teeth with less well-developed, sometimes vestigial, lateral cusps (see Fig. 4). Outer row teeth 3 (X = 2.5, range 1-4, n = 48). Inner row teeth 4 (X¥ = 4.9, range 4—6, n = 44). Maxillary teeth 3, (increasing in num- ber with increasing SL from X = 3.6 in 10 specimens between 20.0 and 24.7 mm SL to X = 4.2 in 7 specimens between 22.0 and 26.5 mm SL). Maxillary teeth tricuspid, first 2 or 3 larger than remaining teeth. Den- tary with 4 large anterior tricuspid teeth in all specimens, 7 = 48 and 8 (X = 7.2, range 5-11, n = 48) posterior small teeth. Pre- maxillary, maxillary, and dentary teeth compressed, having flat inner and outer sur- faces. Vertebrae 39 (X = 39.5, range 38—41, n = 39). Upper limb gill-rakers 6 (X = 5.6, range 5—6, n = 48); lower limb gill-rakers 12 (X = 11.5, range 10-13, n = 48). Bran- chiostegal rays 4 in two cleared and stained VOLUME 116, NUMBER 3 Fig. 14. Planaltina myersi, MNRJ 10635, c&s, sexually active male, SL 33.3 mm. Anterior portion of anal fin, left side. Figure shows anal-fin hooks positioned along posterolateral border of anal-fin rays, larger more posterior hooks bent dorsoanteriorly. Inset drawing at right shows anal-fin hook remnants of a sexually inactive adult male or perhaps developing hooks of a maturing male. specimens, 3 rays originating from anterior ceratohyal and | ray from posterior cera- tohyal. Color in alcohol.—Males and females with approximately same color pattern (Fig. 22) at least in specimens preserved in al- cohol for more than 20 years. Body pale to light brown and slightly darker dorsally than ventrally. Dorsal body surface dark due to scattered dark chromatophores. Few- er dark chromatophores on the ventral pos- terior part of body. Dark lateral body stripe mostly obscured by guanine pigment when such pigment not destroyed by formalin and extending from posterior part of dorsal opercular region to caudal-fin base. Visible dark pigmentation of lateral stripe mostly concentrated on caudal peduncle and more evident on males. This dark pigment con- tinuing on anterior part of median caudal- fin rays. Pectoral and pelvic fins pale, dorsal and anal fins dusky with scattered dark chromatophores along fin rays. Head darker dorsally and around mouth. Circumorbital bones and opercle silvery with few scat- tered dark chromatophores. Ecology.—Dr. Heraldo Britski (pers. comm.) reports that the rio Corumbatai, where most of the samples were collected, is a clear-water river with a sandy bottom, approximately 8 meters wide and | meter deep at the collecting site. The area was al- most entirely deforested and the marginal vegetation predominately represented by grasses. Sexual dimorphism, mode of reproduc- tion, and gonad anatomy.—Histological analysis revealed a thickened epithelium along the anal-fin rays of a mature male (26.3 mm SL; Fig. 24). Only a slight thick- ening was evident in the anal fin of a ma- ture female (27.0 mm SL). Club cells were abundant in some areas, particularly in the epithelium near the anal-fin base. Although some club cells were present near the epi- thelial surface, no definitive secretion of the 576 Keegan ie a Spseiite eee the pen SS TNE, ‘o ats ate ses = aT se I a a Seekers, FSIS * Q Pha eh Pn re - ues SaaS sags, 13). Ventral surface showing pelvic-fin hooks. cell contents was observed. The females lack the anal-fin and pelvic-fin hooks pre- sent in the males as described above (Fig. 25). The mean pelvic-fin length is longer in adult males than in adult females (Fig. 17) and although difficult to document from our limited data, this divergence in relative length may begin at approximately 24.0— 25.0 mm. The skin is thickened along the ventral surface of the pelvic-fin rays of re- productively active males and forms a lin- ear and turgid mound of tissue. Histological analysis of pelvic fins of a mature male (26.3 mm SL) and a mature female (27.0 mm SL) revealed that the marked thicken- ing of the skin in the male was due to both growth of the epithelium and underlying connective tissue (Figs. 26, 27b). Club cells were particularly abundant near the pelvic- fin base, some near or at the surface. Un- fortunately we were not able to observe cells at the surface undergoing active ho- te leg par toe ele ae as PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Planaltina myersi, MNRJ 10634, c&s, sexually active male, SL 30.5 mm. Pelvic-fin rays, left side. locrine secretion as was observed for sim- ilar cells of the anterior region of the anal fin of a species of Attonitus and the new genus. See Weitzman et al. (2004) for a more complete description of the gland cells and their mode of secretion in species of Attonitus. In the related undescribed ge- nus and species these glandular cells resem- ble the classic “‘club” cells known to be present in most ostariophysan fishes (Pfeif- fer 1977:660). There is a great amount of evidence that most ostariophysan club cells produce a pheromone that serves as an alarm signal (Pfeiffer 1967:389—390). Clas- sic alarm substance cells (ASC’s) are club cells that do not open onto the epithelial surface and are said to release their contents only if the epithelium is damaged (Pfeiffer 1977:660—662). The club cells seen along the anal fin of Attonitus and the possibly closely related new genus and species do reach the epithelial surface and undergo de- VOLUME 116, NUMBER 3 S/F A = 13 males plus juvenile males, Y = -0.628 + 0.174; r* = 0.951: adj. r* = 0.947 @ = 32 females plus juvenile females, Y = 0.098 + 0.137X; p= 0.946; adj, r?=0.944 Pelvic-fin length (mm) HOw te20) (22 B24) 826. 2 e265 Standard Length (mm) Planaltina myersi 30) S29 $3411.36, 38 3405 42 4°44 Fig. 16. Planaltina myersi, pelvic-fin length as a function of SL by sex. Males and females were identified by external features and sometimes by examination of gonads through a slit in right side of body. Below 30 mm SL, the data is essentially useless for indicating differences in male and female pelvic-fin length, but beyond 30 mm SL, where most mature males occur and primarily where sex of both could males and females could be identified by gonad examination, all males have longer pelvic fins than females. generation with the contents of the entire cell being secreted (holocrine secretion). These cells may therefore have a function other than of ASC’s as may be the case in several catfish species (Pfeiffer 1967:389— 390, 1977:663). Some club cells observed in the species of Planaltina also appeared to be located near the epithelial surface, al- though cell degeneration and release was not observed in our histological sections. More specimens need to be analyzed his- tologically to see 1f some club cells of Plan- altina resemble those present on the anal fins of species of Afttonitus and the new spe- cies and genus possibly closely related to Attonitus. Although our population samples of P. glandipedis are represented by males and females of about equal length ranges from 24 to about 28.5 mm SL, only a few are sexually mature judging from visual in- spection of the gonads. The scales border- ing the dorsal border of the mature male pouch opening are more elongate than those of the mature females (Fig. 28a, b), but it is impossible to predict that greater elon- gation of scales dorsal to the pouch opening might be attained at comparatively larger sizes in females based only on the few available adult specimens of both sexes. Histological analysis of ovaries from a mature female (27.0 mm SL) revealed the 578 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 4 = 27 males plus juvenile males, Y = -1.180 + 0.180X; r* = 0.859: adj. r* = 0.853 @ = 19 females plus juvenile females, Y= 0.186 + 0.117X: r* = 0.829; adj. r* = 0.819 € E = O) S o & ig oO = 0) oO 20 21 22 23 Standard Length (mm) Planaltina glandipedis Fig. 17. 24 29 26 2 28 2g Planaltina glandipedis, pelvic-fin length as a function of SL by sex. Males and females identified by external features and sometimes by examination of gonads by a slit in right side of body. Sexual maturity according to relative pelvic-fin length in males versus those of females appears to strikingly occur at least by standard lengths of about 25 mm SL where there appears a sudden increase in male pelvic-fin length. However, the data in other regions of the graph are insufficient to confirm this estimate. Also, at these lengths the male’s testes become active and gland cells appear present in the pelvic fins. presence of spermatozoa within the ovarian cavity (Fig. 29). Longitudinal sections through the testes revealed a posterior sperm storage area like that described for P. myersi by Burns et al. (1995:134, fig. 2). This posterior sperm storage region in a mature male P. glandipedis male (26.3 mm SL) occupied 23.3% of the area of a mid sagittal testis section. In P. myersi the stor- age area occupied 44.2% of the total testis area while equivalent regions in outgroup glandulocaudine species ranged from 4.3— 12.2% (Burns et al. 1995). As reported for P. myersi by Burns et al. (1995: table 3) the sperm cell nucleus is spherical to slightly deformed. There appears to be no sperm cell modifications as are found in typical glandulocaudines, but ultrastructure exami- nation was not possible. There were no gill glands as reported in Corynopoma riisei Gill by Burns and Weitzman (1996) and other glandulocaudines by Bushmann et al. (2002). Distribution.—Planaltina glandipedis 1s known from the rio Corumbatai and rio Ja- caré-Pepira, both tributaries of the rio Tieté, upper rio Parana basin, in the State of Sao Paulo. Etymology.—The name _ glandipedis 1s derived from Latin glandula, diminutive for VOLUME 116, NUMBER 3 7/9 & = 41 males plus juvenile males, Y = -1.809 + 0.207X; r? = 0.865: adj. r*=0.861 ® = 51 females plus juvenile females, Y = -0.591 + 0.150X: r* = 0.970; adj. r* = 0.969 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0 3.8 3.6 3.4 3.2 3.0 2.8 2.6 2.4 2.2 5 Pelvic-fin length (mm) 18 19) 20 2) 22 2 Bey 5 As if 2s} 749) 60) Sil SBS) SE ES) S15) S)// Standard Length (mm) Planaltina britskil Fig. 18. Planaltina britskii, pelvic-fin length as function of SL by sex and maturity. Males and females identified by external features and sometimes by examination of gonads by a slit in right side of body. Sexual maturity according to relative pelvic-fin length in males versus that of females appears to occur by lengths close to 28—29 mm SL. At these lengths male’s pelvic fins become longer relative to those of females. Also, at these lengths, male’s testes become active. acorn, hence gland, and pedis, Latin for foot, in this case referring to the pelvic fins. A noun in apposition. The name is meant to signify a fish with a pelvic-fin gland or organ. Planaltina britskti, new species ces Oy ables Specimens examined.—All type speci- mens are from Brazil, Estado de Sao Paulo, drainages, tributary to rio Grande or rio Tieté, both tributaries of the rio Parana or in small streams directly tributary to the rio Parana. Holotype.—MZUSP 62757, male, 35.0 mm SL, Municipio de Sao José do Rio Preto, co6rrego da Barra Funda, tributary of rio Preto and tributary of rio Turvo, tribu- tary of rio Grande, approximately, 20°37’S, 49°23'W, 7 Dec. 1998, V. Garutti. Paratypes.—MZUSP 62758, 5 (33.0- 36.5 mm SL) and DZSJRP 668, 7 (imma- ture to adults, 22.0—36.5 mm SL), collected with holotype. Following 4 lots of imma- ture to adult paratypes collected in Muni- cipio de Sao José do Rio Preto, corrego da Barra Funda, tributary of rio Preto and then rio Turvo, tributary of rio Grande, approx- imately 20°37'S, 49°23’ W. MZUSP 62759, 4 (17.0—31.0 mm SL), 30 Dec. 1980, V. Ga- 580 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON —————— ——— SI Wasser, mom eewatoreser Fig. 19. Planaltina myersi, MNRJ 10635, c&s, sexually active male, SL 33.3 mm. Caudal skeleton, lateral view, left side. Inset at left shows modified scales removed. - j oes * i sac » ms i sent Fig. 20. Planaltina myersi, USNM 278966, histological section of pelvic fin showing abundant club cells (arrows), adult male SL 38.3 mm. Bar = 100 pm. VOLUME 116, NUMBER 3 Je, Zit. 581 ri a ws Planaltina myersi, USNM 278989, histological section of ovary of adult female, 35.0 mm SL, arrow points to spermatozoa, 0 = oocyte cytoplasm. Bar = 20 wm. rutti & party; DZSJRP 674, 5 (ammatures, 26.0—27.0 mm SL), 30 Dec. 1980, V. Ga- rutti & party; MZUSP 26911, 10 (21.0- 26.2 mm SL), 4 Jul. 1980, V. Garutti; USNM 362134, 10 (immatures to adults, 25.5—31.2 mm SL), 4 Jul. 1980, V. Garutti. MZUSP 62760, 10 (immatures to adults, 19.0—32.5 mm SL); LIRP 2255, 10 Gm- matures to adults, 31.5—33.5 mm SL), Mun- icipio de Auriflama, Fazenda Itapua, cor- Fig. 22. Planaltina glandipedis, new species, holotype, adult male above, MZUSP 62752, SL 26.2 mm and paratype, adult female below, MZUSP 62753, SL 28.6 mm. Corumbatai, rio Corumbatai, approximately 22°13'S, 47°38'W, Estado de Sao Paulo, Brazil. 582 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Morphometrics of Planaltina glandipedis. Standard length is expressed in mm; measurements through head length are percentages of standard length; the last four entries are percentages of head length. Minimum, maximum, n, X, and SD are based on the holotype, MZUSP 62752, and the following paratype lots: MZUSP 62753, USNM 362135, MZUSP 62754, USNM 362136, MZUSP 62755, USNM 362137, MZUSP 62756, and DZSJRP 671. Males Females Holotype wm Ranse lx 0) SD) Sin Ranscitn ao Din Standard length (mm) ND, All 20-28.7 24.0 20> 19 =28:6.) 2a Body depth at dorsal fin origin 20:1 27. N8:6—-28:8 2355" 3:0) 20) Qiks 2S ee eee) Snout to dorsal-fin origin 61.5 27 544-640 58.9 2.6 20) S47/-6223 9 ae Snout to pectoral-fin origin M0) AY ASO) PZ NO 20) 230-276) = 24.9 Snout to pelvic-fin origin 45.8 27 42.4-50.0 45.2 2.0 20 42.1-49.1 45.7 1.7 Snout to anal-fin origin 58.8 (27 °53:0-64.:0 57.7 2.6. (20) ~55.8-61-on SS OReaee Caudal peduncle depth lO 27 OWILI20 CQ iz DY MORO OL) Ill Caudal peduncle length 12.6 27 106-144 124 0.9 20 07.6-14.5 12.0 1.6 Pectoral-fin length WX 2 ZOC209) BZA i 20 21.8-264 234 0.1 Pelvic-fin length 14.5 27 12.5-145 134 0.5 20 1124144 > 2 e0w) Dorsal-fin base length WD BZ OSsO=2O O85 Obl 20” (082-1257 Ss 0l0 Ren Dorsal-fin height 1923 27 ISS 24 IO Led 20, AT.222.1 Soe Anal-fin base length 2.3 Al. 25-3203 290 16 20) 26:0=33:2 23/0. Anal-fin lobe length 1729 27 N36 -195) 2 EO 1) 20) 1330-20 ev eoeeeles Eye to dorsal-fin origin 46.9 27 43.149.6 46.0 2.1 20 43.0-50.0 46.5 1.6 Dorsal-fin origin to caudal-fin base 33.2 2 303420 39.5 1d 20 36.440.7 38.7 1.1 Bony head length MS 2 BQiNOG-#ds3' 23D 2 20) 25 —25)4 eS one Horizontal eye diameter 35.4 27 35.4-42.6 389 1.8 20 33.8-42.6 38.7 2.1 Snout length AZoM Zt IPOD ZO 25 20 WH8=25.4 ZS ZG Least interorbital width 33 27 ZUO-8e9 BOs ZI 2O\ 20: —36-2 les eo Upper jaw length 43.1 27 37.0-44.8 40.7 2.1 20 38.644.0 41.3 1.4 rego do Limoeiro, tributary of rio Sao José DZSJRP 656, 7 (immatures to adults, 22.0— dos Dourados, tributary of rio Grande, 29.2 mm SL), José Bonifacio, km 88 of 20°33'32.9"S, 50°35'29"W, altitude 350 m, road BR-153, stream ultimately tributary to 28 Jul. 1999, Ricardo M. C. Castro & party. rio Tieté, approximately 21°03’S, 49°41’W, MZUSP 62761, 6 (29.5-35.0 mm SL). 12 Feb. 1988, D. Brandao & party. MZUSP Fig. 23. Planaltina glandipedis, USNM 362385, c&s, sexually active male, SL 24.3 mm. Anterior portion of anal fin, left side. Figure shows anal-fin hooks positioned along posterolateral border of anal-fin rays, larger more posterior hooks bent dorsoanteriorly. VOLUME 116, NUMBER 3 N ioe) ie’) Fig. 24. Planaltina glandipedis, USNM 362136, histological section through anterior anal fin of mature male, SL 26.3 mm. Arrow indicates thickened epithelium. Bar = 100 pm. 62762, 3 (immatures to adults, 30.0—33.5 mm SL); DZSJRP 670, 3 (adults, 27.5—30.0 mm SL), Municipio de Mirassolandia, Ma- catbas, approximately 20°34’S, 49°28'W, 30 Apr. 1988, Marcelo Garcia. MZUSP MINS Sn immaminess: SssanGg) 20s mim SL). USNM 367195, 1 (27.3 mm SL), Mig- uelopolis, represa de Volta Grande, rio Grande, approximately 20°12'S, 48°07’'W, 6-7 Nov. 1975, CETESB. MZUSP 62763, 4 (ammatures, 21.5—23.3 mm SL), Conchal, rio Mogi-Guagu, tributary of rio Tieté, ap- proximately 22°19’'S, 47°07’W, 20 Apr. 1997, Flavio C. T. Lima & party. MZUSP 62764, 2 (immatures, 27.2 and 29.5 mm SL), Corumbatai, rio Corumbatai, tributary of rio Piracicaba, tuibutary of rio Tieté, ap- proximately 22° 137Ssi47-38 W,, 25 Jan: 1976, Heraldo A. Britski. Non-type material. M©MZUSP 62808, 4 (adults, 27.5 and 31.5 mm SL), Ilha Soltei- ra, rio Parana, approximately 20°23’S, 51°167'W, Sept. 1965. LIRP 2256, 1 (adult, 32.0 mm SL), Municipio de Olimpia, Cor- fecZondas |) Waranyjecimasm2OA4 1eS7S, 49°02'50.3"W, 15 July 2001, Ricardo M. C. Castro & party. LIRP 2257, 2 (immatures, 22.3 and 27.0 mm SL), Municipio de Guap- iacu, Cérrego do Modesto, 20°43'52.4’S, 49°14'35.5"W, 14 July 2001, Ricardo M. C. Castro & party. Definition.—Planaltina britskii and P. glandipedis sympatric in rio Corumbatat drainage and both with a scale arrangement on dorsal border of pouch scale different from that of P. myersi (key above and com- pane ies el Ow Ses 3aneb) . Rheseiworspe- cies readily distinguished from each other by characters provided in key and in defi- nitions of P. myersi and P. glandipedis. Description.—Morphometric data pre- sented in Table 3. Description based on all 584 ey ea Sr re SG B Ce NGS O15 One AO Ce eet ners ISO) OF0 ae eS Oe ee s, . lente W)e left side. Ventral surface showing pelvic-fin hooks. lots since no significant differences found among them. Body compressed and slender; greatest body depth located slightly anterior of dor- sal-fin origin at vertical crossing pelvic-fin origin. Dorsal body profile slightly convex from tip of snout to dorsal-fin origin, almost straight from this point to anterior to caudal peduncle and slightly concave dorsal to caudal peduncle. Dorsal-fin origin nearer to caudal-fin base than to snout tip. Ventral body profile convex from tip of lower jaw to anal-fin origin, almost straight from this point to posterior end of anal-fin base and slightly concave ventral to caudal peduncle. Snout conical, shorter than orbital diam- eter. Mouth subterminal, lower jaw includ- ed in upper when mouth closed. Maxilla ventrally convex, extending posteriorly to vertical line crossing anterior border of pu- pil of eye. Dorsal-fin rays 11, 8 (with posteriormost ray unbranched in all but two specimens in which count appears as 11, 9 but is actually ii, 8), m = 90. Adipose fin present. Anal-fin PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON EEE: SOS) SATE RNS EE ACNE SD aero aranrorad Planaltina glandipedis, USNM 362135, c&s, sexually active male, SL 24.3 mm, pelvic-fin rays, rays iv, 21 (iv or v, usually iv, unbranched rays, branched rays X = 21.3, range 19—24, n = 90). Anterior anal-fin lobe weakly de- veloped and including fourth and/or fifth unbranched rays and first 6—7 branched rays. Anterior part of anal fin in sexually mature males with bilateral hooks on ante- rior fourth or fifth unbranched and anterior 9-10 branched rays (Fig. 31). Pectoral-fin rays i, 11 (anterior unbranched ray 1 in all specimens; branched rays X = 10.4, range 9-11, n = 90). Distal tips of longest pec- toral-fin rays not reaching pelvic-fin origin in juveniles, just extending to pelvic-fin or- igin in immatures and slightly beyond pel- vic-fin origin in adult specimens; pectoral fin about same relative length in both sexes and without hooks. Pelvic-fin rays 1, 6 (pos- teriormost ray unbranched in most speci- mens, branched in 2 adult males, n = 90). Sexually mature males with hooks present on rays of pelvic fin, distributed as in Fig. 32. Number of hooks per ray varying among males; in mature male (43.0 mm SL) no hooks present on first branched ray, VOLUME 116, NUMBER 3 585 Fig. 26. Planaltina glandipedis drawing of a section of the ventral external surface of pelvic-fin glandular pads of a sexually active male, USNM 362135, SL 24.3 mm. The fin ray section displays its two halves or lepidotrichs and the proximal portion of this ray extends to the left and the distal portion extends to the right. A hook, at the right, is shown extending ventrally from the surface of the ventral lepidotrich. The fold of tissue extending ventrally from the fin ray constitutes a section of this particular ray’s glandular pad. The ventral dashed lines extending to the right and left of the fold of tissue indicate the continuation of the ventral border of the fold or pad that has been cut away. The dorsal dashed lines are continuous with the area where the pad of tissue has been cut from the membrane that extends between the pelvic-fin rays adjacent to the illustrated ray. 25 present on second, 23 on third, 30 on fourth, and 15 on fifth branched ray of right pelvic fin. Pelvic fins of adult males longer than those of females (Fig. 18); distal tips of longest pelvic-fin rays of males extend- ing slightly beyond anal-fin origin, but fall- ing short of anal-fin origin in adult females. See discussion under “Sexual dimorphism” for explanation of Fig. 18 and pelvic-fin length differences between males and fe- males. Principal caudal-fin rays 10/9, n = 90. Scales cycloid with few radii (3—5) along posterior border, more numerous and con- spicuous on enlarged scales bordering dor- sal edge of pouch opening (Fig. 33a). Lateral line complete, perforated scales 38 (X = 39.9, range 38—42, n = 55). Pre- dorsal scales 14 (X = 15.2, range 14-16, n = 73). Scale rows between dorsal-fin origin and lateral line 5 (n = 75). Scale rows be- tween anal-fin origin and lateral line 4 (n = 75). Scale rows around caudal peduncle 13 (X = 13.1, range 12-14, n = 54). Premaxillary teeth in two distinct rows. Outer row of premaxillary teeth tricuspid, inner row pentacuspid. Outer row teeth 3 (X = 3.4, range 2-5, n = 90). Inner row teeth 4 (X = 4.0, range 4-5, n = 90). Max- illary teeth 2 (X = 1.7, range 1-3, n = 90). Maxillary teeth tricuspid, first usually larger than remaining. Dentary with 4 large ante- rior pentacuspid teeth in all specimens, 1 = 90 and 6 (X = 6.2, range 4-9, n = 90) slightly tricuspid to nearly unicuspid teeth posteriorly. Premaxillary, maxillary and dentary teeth compressed. Vertebrae 38 (X = 38.7, range 38-41, n = 41). Upper limb gill-rakers 6 (X = 5.8, range 4—7, n = 88); lower limb gill-rakers 12 (X = 11.8, range 11-13, n = 89). Bran- chiostegal rays 4 in 2 cleared and stained specimens, 3 rays originating from anterior ceratohyal and | ray from posterior cera- tohyal. Color in alcohol.—Males and females displaying an identical color pattern in pre- served specimens (Fig. 30). Body light brown, darker on dorsum and top of head. Lateral body stripe dark with silvery reflec- tions, especially along ventral region due to deposits of guanine. This stripe beginning anteriorly as thin line at dorsal region of opercle and widens progressively towards caudal-fin base, reaching a maximum width at caudal peduncle, where it narrows and extends dorsal to enlarged dorsal pouch scales and onto median caudal-fin rays. Dark chromatophores present on dorsal body surface, mostly concentrated along 586 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 27. Planaltina glandipedis, USNM 362136; A, histological cross sections of pelvic fins of female SL 27.0 mm and B, glandular pads of the male SL 26.3 mm. Arrows indicate epithelium covering pelvic-fin rays, b = bone of fin rays; c = expanded connective tissue of male. Bar = 100 pm. VOLUME 116, NUMBER 3 Od : | RES Fig. 28. Planaltina glandipedis, A, USNM 362135, c&s, sexually active male, SL 26.9 mm. Caudal skeleton, lateral view, left side; B, USNM 362135, c&s, sexually active female, SL 24.5 mm. Caudal skeleton, lateral view, left side. 588 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 29. Planaltina glandipedis, USNM 362136, histological section of ovary of adult female SL 27.0 mm. Arrow indicates spermatozoa; O = oocyte cytoplasm. Bar = 20 pm. Fig. 30. Planaltina britskii, new species, holotype, adult male above, MZUSP 62757, SL 35.0 mm, and paratype and adult female, MZUSP 62758, SL 36.5 mm; coérrego da Barra Funda, tributary of rio Preto and tributary of rio Turvo, tributaries to rio Grande, Municipio de Sao José do Rio Preto, approximately 20°37'S, 49°23'W, Estado de Sao Paulo, Brazil. VOLUME 116, NUMBER 3 589 Table 3.—Morphometrics of Planaltina britskii, new species. Standard length is expressed in mm; measure- ments through head length are percentages of standard length; the last four entries are percentages of head length. Minimum, maximum, n, X, and SD are based on the holotype, MZUSP 62757, and the following paratype lots: MZUSP 62758, DZSJRP 668, MZUSP 62759, DZSJRP 674, MZUSP 26911, USNM 362134, MZUSP 62758, LIRP XXX, MZUSP 62761, DZSJRP 656, MZUSP 62762, DZSJRP 670, USNM 367195, and MZUSP 62763. Males Females Holotype 1 Range xX SD n Range xX SD Standard length (mm) BOR Soe 20-3550)" 2977 0 920-3625) 28-5 Body depth at dorsal fin origin papas) Ss) MOA) Za Nee SOMO 2525, wi 2222) 2S Snout to dorsal-fin origin DPN BY Dots) = SVD ils 50) 55.3—-01F0 35873) 165 Snout to pectoral-fin origin 23 39) UO IO Sumo A 50) 2224-273 24:4 1:0 Snout to pelvic-fin origin 46:0" 399 429-909" AD:6 18 49 42.8-51.1 46.5 1.8 Snout to anal-fin origin GO0R 39 0) 5356-6375) Osa ames a0) 54-/—-69.7. 602 2A Caudal peduncle depth US 3) ONO.) OD2° —0),7/ 50 =06.5-10.0 08.8 0.8 Caudal peduncle length 1435 394 F1OI0=14 3) 1235S del DO O94 2 Al Pectoral-fin length DEF ee) eee on 2 One LOS Ome mons —2 2-9) 20.6.) EO Pelvic-fin length i433 Ale t6My Pia oe 3S SOP eA 2S) OM Dorsal-fin base length O99 39) 0851s a G:OU Ors SY) Ose 092) .)57/ Dorsal-fin height LON See OO Onue 92 OT eee 4 Seles —20)2, 18:9) 10l6 Anal-fin base length 2350 a ee SO She Ome S05 eG) 50) 244-3158) 9 27-5, 14 Anal-fin lobe length OMe Onto 2S Oe 16.9 O19 SOM AG 19OD Tie deal Eye to dorsal-fin origin 47.1 39 41.5494 45.2 1.6 50 41.3-48.6 45.6 1.7 Dorsal-fin origin to caudal-fin base 42.33 39 38.6443 41.1 14 50° *37.2-45°3 40:5. 17 Bony head length WLS 3S) ~ PVS=7 3 PLO Oy) DO 2ia-25.0) 2350) 09 Horizontal eye diameter BOLO SOP 3429-48) 397 16 50. 434.5-41:8 38:0 1.9 Snout length WOES os) ZI Bsy Zo2 925 0220-90! Ste Orme Least interorbital width BAe, SOP Il 34 7, se Sle) C8 05. 23.135. 4 Siro) 2-5 Upper jaw length 42.77 39 36.4-45.5 41.4 24 50 34.7-44.8 40.9 2.3 exposed borders of scales. Scattered dark chromatophores on snout, tip of lower jaw, and below eye. Inconspicuous patch of dark chromatophores present dorsal to anterior half of anal fin, extending ventrally onto most of anal fin. Fins hyaline except faint scattered dark chromatophores present on basal portion of dorsal fin and upper and lower lobes of cau- dal fin. Most of iris, as well as circumor- bital, and opercular bones silvery. Color in life-—Unknown. Sexual dimorphism, mode of reproduc- tion and gonad anatomy.—tThe typical hooks on the anal and pelvic fins of males ane mepresemeduingimigs. sland) 32. (hese Fig. 31. Planaltina britskii, USNM MZUSP 62758, c&s, sexually active male, SL 35.6 mm. Anal fin, left side. Figure shows anal-fin hooks positioned along posterolateral border of anal-fin rays, hooks curved dorsoan- teriorly. 590 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON eins, BY, Ventral surface showing elongate narrow pelvic-fin hooks. are absent on the pelvic and anal fins of females. Adult males have pelvic fins lon- ger than adult females (see Fig. 18). In our population samples males and fe- males occur in equal length ranges, but very few adult females show the typical elongated scales on the dorsal border of the pouch opening that is present on the caudal fin of males of comparable lengths. The elongated scale pattern on the dorsal border of the pouch opening of adult males (illus- trated in Fig. 33a) is present even in 1m- mature males of 30.0 mm in SL, but mature females in the size range between about 28.0 and 34.0 mm SL have the scales on the dorsal border of the pouch opening only shightly modified (see Fig. 33b). This sug- gests that the full elongation of scales on the dorsal border of the pouch opening in females is attained at comparatively longer sizes than in males. A description of the ultrastructure of the sperm cell is not available. Histological analysis of mature ovaries revealed the presence of spermatozoa within the ovarian cavity (Fig. 34). Upon dissection, the testes of the only male specimen available for his- tological analysis (Fig. 35) were relatively thin, suggesting that it was not fully mature (holotype, MZUSP 62757, 35.0 mm SL). Planaltina britskii, MZUSP62758, c&s, sexually active male, SL 35.6 mm. Pelvic-fin rays, left side. However, a distinct posterior sperm storage region was present in the testis, occupying 13.1% of the total testis area in a mid-sag- ittal section. This is slightly larger than the percent area range (4.3—12.7%) for the out- group glandulocaudine species in the study of Burns et al. (1995:134, fig. 2). Consid- ering that this male was not fully mature, a distinct posterior sperm storage region com- parable to that found in the other Planaltina species probably occurs in P. britskii. As reported for P. myersi by Burns et al. (1995: table 3) the sperm cells are typical aquasperm with the cell nuclei spherical to slightly deformed. There appears to be no sperm cell modifications as are found in typical glandulocaudines. A gill gland, comparable to that seen in P. myersi (Bush- mann et al. 2002) was present in the holo- type (Fig. 36). Data from Garutti (1983) indicate that in corrego da Barra Funda the main reproduc- tive period of the species is coincident with that area’s rainy season, September through March, incidentally also the longest photo period of the year. The majority of the males and females, 98.3%, collected during the dry season, April through August, had undeveloped testes and ovaries, but 82.2% of the specimens of both sexes collected VOLUME 116, NUMBER 3 591 Fig. 33. Planaltina britskiit, MZUSP62758, A, c&s, sexually active male, SL 35.1 mm. Caudal skeleton, lateral view, left side; B, sexually active female, SL 33.8 mm. Caudal skeleton, lateral view, left side. 392 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 34. Planaltina britskii, MZUSP 26911, histological section of ovary of an adult female SL 32.0 mm. Arrow indicates spermatozoa; O = oocyte cytoplasm. Bar = 20 wm. Fig. 35. Planaltina britskii, MZUSP 62757, holotype, SL 35.0 mm, mid-sagittal histology section through posterior testis showing more anterior spematogenic regions (SG) and posterior sperm storage region (PS). Bar = 200 pm. VOLUME 116, NUMBER 3 593 Fig. 36. Planaltina britskii, holotype MZUSP 62757, holotype, SL 35.0 mm, frontal histological section through anteriormost gills showing gill gland with its epithelial covering (arrowheads) and reduced secondary lamellae (arrows) within gill gland chambers which number nine. Unmodified gills (u) with secondary lamellae of usual length also shown. Bar = 100 pm. during the rainy season were mature, the males having well-developed testes and the females well-developed oocytes. Garutti also found that the males reach larger sizes than the females. Ecology.—The population sample from Corrego do Limoeiro, rio Sao José dos Dourados sub-basin (MZUSP 62760 and LIRP 2255) was collected in a fourth order stream running in gallery forest represented by old cerrado vegetation. The stream width varied from 2.3—3.2 meters and the water depth from 0.07—0.08 meters. Rapids and small pools were found along the stream and the bottom consisted mostly of sand. Marginal vegetation was mostly represent- ed by palm trees (Arecaceae), ferns (Pteri- dophyta) and lichens. At the time of col- lecting activities the following data were taken: air temperature 35.5°C, water tem- perature 21.5°C, pH = 7.24, dissolved ox- ygen = 9.6 mg/l, and conductivity = 125.5 S/cm. Other species taken at this site were: As- tyanax bimaculatus, Astyanax eigenman- niorum, Moenkhausia sanctaefilomenae, Piabina argentea, Characidium sp., Hypos- tomus ancistroides, Hypostomus sp., Ster- nopygus macrurus and Poecilia reticulata. The populations samples from corrego da Barra Funda, rio Preto sub-basin (MZUSP 62757627 583,027 597 26091 15 DZSIRP 647 and USNM 362134), were all collected near the mouth where this stream empties into a larger stream called ribeirao da Barra Grande at an altitude of about 440 meters. According to Garutti (1983) the collecting site included a circular pool approximately 8 meters in diameter with a maximum depth of 2 meters. The pool lacked marginal veg- etation. Rapids occurred downstream from the pool and ran through a gallery forest to 594 a pool of the ribeirao da Barra Grande. The maximum stream width along the rapids was two meters, the water depth varied from 0.4—1.5 meters and the marginal veg- etation was grass. During the period collections were made, from March 1980 through February 1982 the near surface water temperature varied between 19° and 30°C and the pH between 5.0 and 6.0. The water current in the gallery forest stream varied between 0.244 and 0.600 meters per second. Other species at this collecting site were: Serrapinnus piaba, Odontostilbe stenodon, ““Cheirodon” sp., Astyanax bimaculatus, Astyanax fasciatus, Hemigrammus margin- atus, Moenkhausia sanctaefilomenae, Pia- bina argentea, Serrasalmus spilopleura, Acestrorhynchus lacustris, Oligosarcus pin- toi, Salminus hilarii, Characidium sp., Ho- plias malabaricus, Pyrrhulina australis, Cyphocharax gilberti, Cyphocharax van- deri, Leporinus friderici, Leporinus octo- fasciatus, Leporinus striatus, Eigenmannia sp., Cetopsorhamdia theringi, Nannorham- dia schubarti, Pimelodella cf. gracilis, Rhamdella minuta, Rhamdia hilarii, Aspi- doras fuscoguttatus, Hypostomus ancistro- ides, Microlepidogaster francirochat, Poe- cilia reticulata, Cichlasoma portalegrense, Crenicichla britiskii, Tilapia sp. and Syn- branchus marmoratus. Distribution.—This species is known to occur in the State of Sao Paulo in tributaries to rio Grande and rio Tieté, both tributaries of the upper rio Parana basin. Etymology.—tThe specific name, britskii, is a patronym for Dr. Heraldo A. Britski, ichthyologist at the Museu de Zoologia da Universidade de Sao Paulo, in recognition of his many contributions to the ichthyol- ogy of Brazil. Dr. Britski collected the first two known specimens and recognized that they represented a new species of glandu- locaudine fish. Discussion of generic and subfamily phy- logenetic relationships.—Bohlke (1954: 265) placed Planaltina in the Glandulocau- dinae and noted a physical similarity be- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tween Planaltina and Bryconamericus. However, he set aside the question of glan- dulocaudine relationships to other characid genera, suggesting that the Glandulocaudi- nae are probably polyphyletic and that per- haps its various component taxa had origins among several tetragonopterine genera. Bohlke (1958:43) and Géry (1977:35) also thought it likely that the Glandulocaudinae are polyphyletic. Weitzman et al. in Weitz- man & Fink (1985:112—117) discussed the question of glandulocaudine monophyly versus polyphyly at some length and con- cluded that there was not enough informa- tion to settle the question. Weitzman & Me- nezes (1998:178—180) hypothesized the subfamily as monophyletic based on a phy- logenetic analysis of the seven tribes of the Glandulocaudinae and used four unequiv- ocal synapomorphies to unite the members of all of these tribes. Outgroup information relative to the Glandulocaudinae and its tribes, as well as the synapomorphies used to diagnose the subfamily, were based on what was then known about the absence or near absence of the distribution of these four features in species of the non-glandu- locaudine genera of the Tetragonopterinae. Recently discovered information that some species of the tetragonopterine genus Kno- dus and all of the species of Attonitus and the related new genus and species are in- seminating and that the anatomy of the pri- mary and secondary sexual systems of these species display some to several of these same features used as synapomor- phies for the Glandulocaudinae by Weitz- man & Menezes (1998) suggest that the phylogenetic relationships of the tribes of the Glandulocaudinae need re-examination. We review at least some of the implications of this new information in our discussion below concerning the phylogenetic position of Planaltina and the other Diapomini. See also Weitzman et al. (2004) for a more de- tailed discussion of Attonitus, Bryconamer- icus, Knodus and the new genus and species regarding their possible relationships to the VOLUME 116, NUMBER 3 Glandulocaudinae and other characids known to be inseminating. The four synapomorphies for the Glan- dulocaudinae, each discussed at length by Weitzman & Menezes (1998:178), are as follows: 1) the presence of insemination, 2) the presence of a sperm storage area in the testes, 3) the presence of elongate sperm- cell-nuclei, and 4) the presence of an elon- gate cytoplasmic collar binding the flagel- lum to the sperm-cell nucleus in at least some stage of spermiogenesis (see Burns et al. 1998). So far as we know these char- acters remain valid for all species of the Glandulocaudinae with the exception of Planaltina, which differs by having essen- tially spherical sperm-cell nuclei and ap- parently lacks an elongate cytoplasmic col- lar binding the flagellum to the sperm-cell nucleus. However, their status as synapo- morphies at the levels proposed by Weitz- man & Menezes (1998) for most if not all glandulocaudines must now be reconsid- ered because Weitzman et al. (2004) found that these features are present in tetragon- opterine species placed in Aftonitus, some of the species currently placed in Knodus, the new genus and species and the species of a few other possibly related genera, for example Brittanichthys Géry and Monoto- cheirodon. However, all the species of these genera appear to lack the caudal-fin pher- omone organ found in a variety of derived states, each characteristic for one of the var- ious tribes of the Glandulocaudinae. It should be noted at this point that the puta- tive homology of the caudal-gland cells among the glandulocaudine tribes remains to be investigated and therefore the homol- ogy of the caudal organs in the subfamily is a not a fully defended hypothesis. It was recently discovered (Weitzman et al. 2004), that the species of these apparent glandulocaudine outgroup genera as well as at least some species of the Glandulocau- dinae have pheromone organs and in some cases simple putative pheromone secretory cells on raised integumentary pads of the anal fin and sometimes the pelvic fins of 595 mature, sexually active males. These fea- tures, when present, were also described and discussed above for the species of Planaltina. Until further research reveals the distribution of these primary and sec- ondary sexual structures among the above discussed genera, and among other insem- inating characids such as Brittanichthys and Monotocheirodon, we decline to discuss the phylogenetic significance of these features regarding the Glandulocaudinae as a whole. We suggest that new hypotheses concerning the phylogenetic relationships of the Glan- dulocaudinae to apparent outgroup taxa be held in abeyance until the distribution of the synapomorphies used by Weitzman & Me- nezes (1998) to diagnose the Glandulocau- dinae can be delineated for the inseminating species currently placed in the non-glan- dulocaudine genera mentioned here. We further note that the inseminating characid genera of the cheirodontine tribe Compsur- ini, although inseminating, lack three of the four synapomorphies (numbers 2, 3 and 4 mentioned above) and have the sperm cell synapomorphies discussed for the Cheiro- dontinae by Malabarba (1998). Regarding synapomorphy number 3 of the glandulo- caudines, their elongate sperm cell nuclei, it is true that the sperm nuclei of the species of the Compsurini are somewhat elongate, but they are never as elongate as in nearly all the glandulocaudine species and appar- ently the centrioles are located posterior to the nucleus rather than anterior to it as in most of the glandulocaudines. Thus nuclear elongation appears to occur through for- ward growth in the cheirodontines rather than backward growth in the glandulocau- dines (Burns et al. 1998). However, these putative characteristics of the ultrastructure of glandulocaudine and compsurin cheiro- dontine sperm cells need confirmation in many species and genera of both groups. Planaltina was considered by Weitzman & Menezes (1998:184) to belong in the glandulocaudine tribe Diapomini because P. myersi shares with the species of the oth- er two genera of the tribe, Acrobrycon and 596 Diapoma, two unequivocal synapomor- phies. 1) The male and female caudal or- gans are nearly equivalent in size. 2) A multiple series of caudal scales, at least three or more, Occur immediately ventral to the lateral-line series and extend posteriorly to form the dorsal border of a pouch open- ing of the caudal gland. Planaltina myersi is distinguished from the species of Acrobrycon and Diapoma by the possession of separate dorsal and ven- tral enlarged pouch scales and only three scales across the dorsal opening of the pouch (Weitzman & Menezes 1998:184, figs. 13-16 and Weitzman & Fink 1985:17, fig. 15). The two diapomin synapomorphies are shared with the two new species herein described (compare Figs. 19, 28, 33). How- ever, the description of caudal-organ fea- tures separating Planaltina from the species of Acrobrycon and Diapoma need some modification based on the inclusion of the two new Planaltina species. 1) The species of Planaltina have between two and six scales forming the dorsal region of the pouch while the number of scales in the dorsal region of the pouch in Acrobrycon and Diapoma exceeds eight or nine. 2) The number of scales forming the dorsal border of the pouch is greater in Acrobrycon and Diapoma than in Planaltina. The species of Planaltina have one or two scales forming the dorsal border of the pouch while in the species of Diapoma the number varies be- tween three and six. In Acrobrycon this number varies from four to at least seven. In Planaltina one of the scales forming the dorsal border of the pouch is elongate and forms most of that border while in Acrob- rycon and Diapoma, even when one of the dorsal border scales is somewhat elongate or larger, its actual degree of entry into the pouch border is little longer than that of the other dorsal border scales. At this time we are not prepared to hypothesize with con- fidence which of the characters listed above to separate these diapomin genera are ple- siomorphic or apomorphic relative to these three genera. In our view, the position and PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON number of adnate scales on ventral lobe of the caudal fin of the inseminating or even non-inseminating species of Knodus (some of which may form an outgroup for the Dia- pomini) must be described and recorded be- fore such hypotheses can be proposed. However, in view of the apparently more plesiomorphic nature of the sperm-cell nu- clei of the species of Planaltina compared to those of Acrobrycon and Diapoma, as discussed here, we suggest that the species of Acrobrycon and Diapoma with a greater number of caudal scales may be the more derived and that a greater number of caudal scales may be a synapomorphy for Acrob- rycon and Diapoma. Histological examination and measure- ment of the sperm cell bodies of the three species of Planaltina recognized herein show that their sperm cells are aquasperm, but with sometimes a slight ellipsoid shape (see Appendix 1), while in Diapoma spe- culiferum and presumably the other species of Diapoma, the cell body is elongate and has the typical derived ultrastructure of glandulocaudine sperm cell with a cyto- plasmic collar surrounding the flagellum as it parallels the side of the elongate sperm cell body (Burns et al. 1995:133, table 3, Burns et al. 1998:237, fig. 2). Although we have no information on the ultrastructure of the sperm cell of Acrobrycon, it has a cell body somewhat more elongate than that of the species of Diapoma. Thus in this re- spect, the three species of Planaltina appear plesiomorphic relative to the species of Diapoma and Acrobrycon. Although this information is useful for suggesting a hy- pothesis of the phylogeny of the genera and species of the Diapomini, confirmation of such relationships among all diapomin spe- cies and their relationships to outgroup in- seminating species currently in Knodus must await detailed research on species of the latter genus in order to obtain a greater knowledge of the characids that apparently form an outgroup or outgroups to the glan- dulocaudine tribes as diagnosed by Weitz- man & Menezes (1998). See also Weitzman VOLUME 116, NUMBER 3 et al. (2004) for a more complete discussion of this problem. One interesting undescribed species needs to be mentioned in regard to the statement just given above. Burns et al. 99521835 table ys) areconded spherical sperm cell nuclei, aquasperm, in Planaltina myersi, USNM 278989, and in Planaltina sp. (this latter sample should have been list- ed as USNM 362836 but was also listed as USNM 278989 in the table by error). See Appendix 2 for field data and in Weitzman et al. (2004) for sexual characteristics. Both of these taxa have aquasperm and are in- seminating. At this time we would place the population sample called Planaltina sp. in Burns et al. (1995) in Knodus because it has adnate scales on the lower caudal-fin lobe and therefore no caudal pouch. However, it is an inseminating species, and the type species of Knodus, K. meridae Eigenmann, although it too has adnate caudal scales and aquasperm, is apparently not an inseminat- ing species according to our histological ex- amination based on the specimens listed in Appendix 2 of Weitzman et al. (2004). The inseminating species currently in Knodus may belong to a series of species of that form a sister group to the Diapomini. How- ever, the histological and ultrastructure characteristics of the primary sexual organs in those species up to the present referred to Bryconamericus and Knodus is complex and mostly unexplored. See Weitzman et al. (2004) for further discussion of these fishes. In our opinion the species traditionally re- ferred to Bryconamericus and Knodus need a thorough phylogenetic study in order to better hypothesize possible relationships among the genera and species of the Dia- pomini. Phylogenetic relationships among the three species of Planaltina.—The phyloge- netic relationships among the species of Planaltina cannot be solved without better outgroup information than is currently available. There are over a dozen characters that provide information for distinguishing species; see especially the distinguishing Se] characters under the title “Definition” for each of the three species above and Figs. 3—13, 19, 28, 33. However, reliable out- group information for many of these char- acters is unavailable and some of the char- acters that might be hypothesized to be de- rived compared to those found in characids usually considered members of the Tetra- gonopterinae support contradictory hypoth- eses of phylogenetic relationships within the species of Planaltina. For example, P. myersi and P. glandipedis share the number of scale rows between the dorsal-fin origin and the lateral line, the number of scale rows between the anal-fin origin and the lat- eral line, the least number of scale rows around the caudal peduncle, and the number of vertebrae. We cannot at this time polarize any of these characters without appropriate outgroup information. Species in Acrobry- con and Diapoma, some undescribed in both genera, are not well known enough to be a source of outgroup information and some of the undescribed inseminating spe- cies currently referred to Knodus may in fact be the closest outgroup of the species of Planaltina. The species of Acrobrycon and Diapoma may be more derived dia- pomins than the species of Planaltina based on their more elongate sperm nuclei (in those so far investigated). Alternatively, if the less elongate and more numerous caudal organ scales of the species of Acrobrycon and Diapoma are considered plesiomorphic for the Diapomini and more like the caudal Squamation in some of the inseminating species of Knodus, then we arrive at a dif- ferent hypothesis. Parsimony ultimately must be used to settle these problems, but the available data is still insufficient for that purpose. There is some evidence that P. glandi- pedis is a paedomorphic relative to P. myer- si and P. britskii. It is smaller in size, has no adipose fin, and has fewer cusps on its jaw teeth than the other two species, all re- ductions common in relatively small char- aciform fishes (see Weitzman & Vari 1988, and Buckup 1993). On the other hand P. 598 glandipedis has a slightly greater number of scale rows around the caudal peduncle and more predorsal scales. In our view an anal- ysis of the phylogenetic relationships among the species of Planaltina must await detailed descriptions of the characters of all diapomins so far described and yet to be described as well as a complete study of the inseminating species that are currently as- signed to Knodus. Acknowledgments Financial aid for this research and asso- ciated travel for museum study in the Unit- ed States and field work and museum study in Brazil by N.A.M. were provided by funds for the project Diversidade de Peixes de Riachos e Cabeceiras da Bacia do Alto rio Parana no Estado de Sao Paulo, Brasil supported by the State of Sao Paulo Re- search Foundation (FAPESP) within the BI- OTA/FAPESP—tThe Biodiversity Virtual Institute Program. This project is under the coordination of Dr. Ricardo M. C. Castro. Also, this research and travel to museums and the field in Brazil for S.H.W. was fi- nancially supported by the Neotropical Lowland Research Program of the Smith- sonian Institution. This program was under the consecutive direction of W. R. Heyer and R. P. Vari for the duration of this re- search. Funds for histological and associ- ated research were provided by the Univer- sity Facilitating Fund of The George Wash- ington University to J.R.B. Funds for travel, equipment, and technical services were also received by S.H.W. and J.R.B. from the Tropical Fish Hobbyist, Inc. New Jersey, through Dr. Herbert R. Axelrod, then pres- ident of TFH, Inc. José Lima de Figueiredo and Osvaldo Oyakawa (MZUSP) loaned specimens of Planaltina in their care and/ or provided cataloging services. Lisa Palm- er and Jeffrey Clayton (USNM) provided cataloging and/or other technical services. Drs. Heraldo Britski (MZUSP), Mauro C. L. B. Ribeiro (IBGE, Reserva Bioldgicade Brasilia) and Valdener Garutu (DZSJRP) PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON generously made available information and specimens collected by them. Lisa Palmer prepared Figs. 1, 22, and 30. Tamara Clark prepared igs, 455) 41523) 25,260) 28no 33 and Figs. 3, 14, and 19 were prepared by Sara V. Fink (University of Michigan). Marilyn Weitzman, Richard P. Vari, both (Smithsonian Institution) and Luiz R. Mal- abarba (PURCS) reviewed the manuscript and offered valuable advice. This paper also benefited from the comments of two anon- ymous reviewers and Dr. Carole Baldwin (USNM). Finally we are indebted to all those thanked in the publications cited be- low for the loan of specimens and catalog- ing services supplied for our glandulocau- dine studies since 1975. Literature Cited Anonymous (Corréa e Castro, R. M.). 2000. Os peixes escondidos do Alto Parana. FAPESP Pesquisa, Sao Paulo, Brazil, 58:44—47. Bohlke, J. E. 1954. Studies on fishes of the family Characidae. No. 7, A new genus and species of glandulocaudine characids from central Bra- zil.—Stanford Ichthyological Bulletin 4(4):265— 274. . 1958. Studies of fishes of the family Chara- cidae, No. 14: a report on several extensive re- cent collections from Ecuador.—Proceedings of the Academy of Natural Sciences of Philadel- phia 110:1—121. Buckup, P. A. 1993. Phylogenetic relationships and re- ductive evolution in neotropical characidiin fishes (Characiformes, Ostariophysi).—Cladis- tics 9:305—341. Burns, J. R., & S. H. Weitzman. 1996. Novel gill-de- rived gland in the male swordtail Characin, Corynopoma riisei (Teleostei: Characidae: Glandulocaudinae).—Copeia 1996:627—633. , 8. H. Weitzman, H. J. Grier, & N. A. Menez- es. 1995. Internal fertilization, testis and sperm morphology in glandulocaudine fishes (Teleos- tei: Characidae: Glandulocaudinae).—Journal of Morphology 224:131—145. , K. R. Lange, & L. R. Malabarba. 1998. Sperm ultrastructure in characid fishes (Teleostei: Ostariophysi). Pp. 235—244 in L. R. Malabarba, R. E. Reis, R. P. Vari, Z. M. S. Lu- cena and C. A. S. Lucena, eds., Phylogeny and Classification of Neotropical Fishes. Museu de Ciéncias e Tecnologia, PUCRS, Porto Alegre, Brazil, Edipucrs, 603 pp. , 9. H. Weitzman, & L. R. Malabarba. 1997. VOLUME 116, NUMBER 3 Insemination in eight species of cheirodontine fishes (Teleostei: Characidae: Cheirodonti- nae).—Copeia 1997:433—438. Bushmann, P. J., J. R. Burns, & S. H. Weitzman. 2002. Gill-derived glands in glandulocaudine fishes (Teleostei: Characidae: Glandulocaudinae).— Journal of Morphology 253:187—195. Garutti, V. 1983. Distribuigao longitudinal da ictiofau- na do Cérrego da Barra Funda, bacia do rio Pa- rana. Unpublished Ph.D. dissertation, Instituto de Biociéncias, Universidade de Sao Paulo, 172 PP. Glantz, S. A., & B. K. Slinker. 1990. Primer of applied regression and analysis of variance. McGraw- Hill, Health Professions Division, New York, Xvilit+777. Géry, J. 1977. Characoids of the World. T. EK H. Pub- lications, Neptune City, New Jersey, 1-672. Malabarba, L. R. 1998. Monophyly of the Cheirodon- tinae, characters and major clades (Ostariophy- si: Characidae). Pp. 193—233 in L. R. Malabar- Daekenb wINGISMRel EE Valine Mer Seleucenarand C. A. S. Lucena, eds., Phylogeny and Classifi- cation of Neotropical Fishes. Museu de Cién- cias e Tecnologia, PUCRS, Porto Alegre, Bra- zil, Edipucrs, 603 pp. Menezes, N. A., & S. H. Weitzman. 1990. Two new species of Mimagoniates (Teleostei: Characi- dae: Glandulocaudinae), their phylogeny and biogeography and a key to the glandulocaudine fishes of Brazil and Paraguay.—Proceedings of the Biological Society of Washington 103:380— 426. Pfeiffer, W. 1967. Schreckreaktion und Schrecksto- ffzellen bei Ostariophysi und Gonorhynchifor- mes. Zeitschrift fiir vergleichende Physiologie 56:380-—396. . 1977. The distribution of the fright reaction and alarm substance cells in fishes.—Copeia 1977:653—665. Quintero-Hunter, I., H. Grier, & M. Muscato. 1991. Enhancement of histological detail using Me- tanil yellow as a counterstain in periodic acid Schiff hematoxylin staining of glycol methac- rylate tissue sections. Biotechnology and His- tochemistry 66:169—172. Schriebman, M. P. 1964. Studies of the pituitary gland of Xiphophorus maculatus (the Platyfish).— Zoologica 49:217—234. Weitzman, S. H., & S. V. Fink. 1985. Xenurobryconin phylogeny and putative pheromone pumps in glandulocaudine fishes (Teleostei: Characi- dae).—Smithsonian Contributions to Zoology 421:1-121. , A. Machado-Allison, & R. Royero L. 1994. A new genus and species of Glandulo- caudinae (Teleostei: Characidae) from Southern Sy) Venezuela.—Ichthyological Freshwaters 5:45—64. , & L. R. Malabarba. 1999. Systematics of Spintherobolus (Teleostei: Characidae: Cheiro- dontinae) from eastern Brazil.—Ichthyological Exploration of Freshwaters 10(1):1—43. , & N. A. Menezes. 1998. Relationships of the tribes and genera of the Glandulocaudinae (Os- tariophysi: Characiformes: Characidae) with a description of a new genus, Chrysobrycon). Pp. 171-192 in L. R. Malabarba, R. E. Reis, R. P. Vari, Z. M. S. Lucena and C. A. S. Lucena, eds., Phylogeny and Classification of Neotropical Fishes. Museu de Ciéncias e Tecnologia, PUCRS, Porto Alegre, Brazil, Edipucrs, 603 pp. ; , J. R. Burns, & H.-G. Evers. 2004. A new genus and species of inseminating characid fish from the upper rio Xingu and rio Tapajés, Brazil, (Teleostei: Characiformes: Characidae) with comments on relationships among insem- inating characids. Neotropical Ichthyology, Por- to Alegre, Brasil. , & L. Palmer. 1979. A new species of Hy- Phessobrycon (Teleostei: Characidae) from the Neblina region of Venezuela and Brazil, with comments on the putative ‘rosy tetra clade.’— Ichthyological Exploration of Freshwaters 7(3): 209-242. , & R. Vari. 1988. Miniaturization in South American freshwater fishes; an overview and discussion.—Proceedings of the Biological So- ciety of Washington 101:444—465. Exploration of Appendix 1 Sperm-cell dimensions for three species of Planal- tina.—The following measurements were taken of sperm nuclei that exhibited a slightly non-spherical shape. These cells were clearly in the minority, but may give a more accurate assessment of the nuclear shape. Most views of the sperm nuclei were spherical. The cells measured varied from slightly ovoid, to pear shaped, to indented, to slightly elongate. The differ- ences in measurements among the three species may be real or, alternatively, be due to variable shrinkage of cells having been treated differently in the past. For example separate collections may have been fixed in different concentrations of formalin. They may also have been slowly transferred to 70% ethyl alcohol by intermediate stages causing little or no shrinkage or, on the other hand, transferred from formalin fixative directly to 70% ethyl alcohol preservative, causing cell shrinkage. Measurements were taken directly under oil immer- sion (100 lens) using an ocular micrometer or from photographic enlargements of negatives taken with a photomicroscope using an oil immersion (100%) lens. For each species, 20 cells were measured. Means + standard deviations (n — | weighting) are given. 600 l. Planaltina myerst: USNM 278989: Male 37.4 mm SL, female 35.0. mm SL. n = 20: Male SZ nuclear length: 1.73 = 0.12 wm Female SZ nuclear width: 1.16 + 0.11 pm Planaltina glandipedis: USNM 362136: Paratype, male 26.3 mm SL, para- type, female 27.0. mm SL. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON n = 20: Male SZ nuclear length: 1.97 + 0.14 pm Female SZ nuclear width: 1.51 + 0.12 wm . Planaltina britskit: MZUSP 62757: Holotype, male 35.0 mm SL, MZUSP 62757: paratype, female 36.5 mm SL. n = 20: Male SZ nuclear length: 1.89 + 0.11 um Female SZ nuclear width: 1.48 + 0.11 wm PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):601—616. 2003. On a new species of tree-climbing crab of the genus Labuanium (Crustacea: Decapoda: Brachyura: Sesarmidae) from Taiwan Petem Ke Ne vandoEt-@ Jiu (PKLN) Department of Biological Science, National University of Singapore, Kent Ridge, Singapore 119260, Republic of Singapore, e-mail: peterng@nus.edu.sg; (HCL) Institute of Zoology, Academia Sinica, Institute of Zoology, Academia Sinica, Nankang, Taipei 115, Taiwan, e-mail: liuhc @ gate.sinica.edu.tw Abstract.—Specimens of a tree-climbing sesarmid crab from Taiwan previ- ously referred to Labuanium rotundatum (Hess, 1865) are here shown to belong to a new species, L. scandens. The new species differs from L. rotundatum s. s. in having a less prominently granulated carapace, the outer surface of the chela covered with numerous small and closely arranged granules, and a dif- ferently structured male first pleopod. The genus Labuanium Serene & Soh, 1970, currently contains 10 species from the Indo-West Pacific (Seréne & Soh 1970). Of these, one of the more often reported species is L. rotundatum (Hess, 1865), a taxon that has been reported from many parts of the Indo-West Pacific, although ap- parently, it is not a very abundant species, and not many specimens are known. Tesch (1917:193—-198), in a detailed discussion on the taxonomy of this species provided a long list of junior synonyms for this spe- cies, viz. Sesarma dentifrons A. Milne-Ed- wards, 1869, Sesarma oceanica De Man, 1889, Sesarma gardineri Borradaile, 1900, Sesarma (Episesarma) rotundata papuom- alesiaca Nobili, 1899, and Sarmatium fax- oni Rathbun, 1906. Labuanium rotundatum was first reported from East Asia by Sakai (1939) on the basis of a male specimen collected from Tansui in northern Taiwan, and although cited by subsequent authors (Sakai 1940, 1976; Dai en alg 9Son Dance yane 1991) Eaiet al: 1994), no further material was forthcoming until Liu (1999) recorded this species from southern Taiwan. Ng et al. (2001) subse- quently added some notes on this species from Taiwan. In 1999, through the courtesy of Lu Eldredge of the Bernice P. Bishop Museum, the first author examined some specimens of L. rotundatum from some Pa- cific islands that appeared to differ in sev- eral respects from the Taiwanese material. But because there was a lack of material from Taiwan and other parts of the Pacific, not much else could be done. Between 2000 and 2001, we obtained an excellent series of specimens from Guam and Taiwan, and the study of this material confirmed our sus- picion of the existence of two separate spe- cies, one of which is undescribed. The de- scription of the new species forms the basis of the present paper. Material examined is deposited in the Bernice P. Bishop Museum, Honolulu, Ha- wai, U.S.A. (BPBM); National Museum of Natural History, Smithsonian Institution, Washington D.C. (USNM); Taiwan Nation- al Museum, Taipei, Taiwan (TMCD); Insti- tute of Zoology, Academia Sinica, Nan- kang, Taipei, Taiwan (ASIZ); National Mu- seum of Marine Biology and Aquarium, Pingtung, Taiwan (NMMBA); and Zoolog- ical Reference Collection of the Raffles Museum, National University of Singapore (ZRC). The abbreviations Gl and G2 are used for the male first and second pleopods respectively. The measurements are cited as: cw (carapace width) X cl (carapace 602 length). The terminology essentially fol- lows that used by Ng (1988). The length of the ambulatory legs were obtained by add- ing the maximum lengths of the merus, car- pus, propodus and dactylus. Labuanium scandens, new species Figs. 1—4 Sesarma (Sesarma) rotundatum.—Sakai IDSOOS7, jole WO ing, 4, IDsssz, CNG! Sesarma rotundatum Hess, 1865.) Labuanium rotundatum.—Sakai 1976:663, text fig. 362, Dai et al. 1986:486, Dai & Yang 1991:532, Cai et al. 1994:597, Liu 1999:88, Ng et al. 2001:42. (Not Sesarma rotundatum Hess, 1865.) Material examined.—All localities in Taiwan. Holotype male, 42.3 < 41.2 mm, TMCD 3290, Hsiang-Chiao-Wan, Hengch- un, Pingtung County, coll. H.-C. Liu, 24 Jan 2002. Paratypes: 3 males, 38.5 X 37.5 mm, 31.8 X 30.3 mm, 14.6 X 14.4 mm, 1 female, 23.2 K 21.8 mm, ZRC 2002.444, Hsiang-Chiao-Wan, Hengchun, Pingtung County, coll. H.-C. Liu, 24 Jan 2002; 1 male, 16.3 * 15.8 mm, | female, 38.4 X 37.2 mm, TMCD 3291, Hsiang-Chiao-Wan, Hengchun, Pingtung County, coll. H.-C. Liu, 15 Dec 1999; 1 female, 27.1 xX 27.0 mm, NMMBA 2002-01, Hsiang-Chiao- Wan, Hengchun, Pingtung County, coll. P- H. Ho, 3 Oct 2001; 1 male, 36.1 X 35.8 mm, 2 females, 36.8 * 36.3 mm, 32.3 X 31.6 mm, IZAS 72860, Hsiang-Chiao- Wan, Hengchun, Pingtung County, coll. H.-C. Lith, ZZ Dee BVOile ZY wrees, SDD ~ 347 mm, 19.1 X 18.7 mm, 3 females, 28.1 xX MSyre) 100, Ard Jt XK AIS ion, Ailes —« HOS mm, ZRC 2001.29, Hsiang-Chiao-Wan, Hengchun, Pingtung County, coll. P. K. L. Ng, 7 Nov 2000; 1 female, 40.2 x 39.7 mm, ZRC 1998.447, Hsiang-Chiao-Wan, Hengchun, Pingtung County, coll. H.-C. Liu, 20 May 1998. Diagnosis.—Carapace slightly wider than long, dorsal surfaces finely granular; posterolateral regions with finely granular oblique striae; outer surface of chela with PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON numerous small, uniformly arranged round- ed granules, dorsal margin with a distinct granulated but non-pectinated ridge on out- er edge with 26—32 relatively longitudinally elongate granules, dorsal surface of dacty- lus with numerous granules; outer surface of ambulatory meri rugose, dorsal margin gently serrated; lateral margins of male seg- ment 6 with distal part convex, proximal part almost straight; G1 relatively stout, al- most straight; distal part bent 90° from ver- tical, distal chitinous part gently upcurved. Description of male holotype.—Carapace rounded, slightly wider than long, broadest at median part of carapace, dorsal surfaces finely granular, regions well defined, gastric and branchial regions swollen; posterolat- eral regions with finely granular oblique striae (Fig. 1A, B). Frontal margin deflexed, sinuous from dorsal view, vaguely divided into 2 low lobes by broad U-shaped cleft, margin relatively entire; postfrontal cristae distinct, sharp, separated into 4 parts, me- dian parts larger, wider, sharper, positioned more anteriorly than lateral parts, separated from lateral parts by short fissure, lateral parts adjacent to low inner supraorbital an- gle; frontal region prominently concave (Fig. 1A, B). Supraorbital margin smooth, with small cleft before external orbital tooth (Fig. 1B, C). Anterolateral margin strongly convex, gently curving into posterolateral margin, junction not discernible; external orbital tooth well developed, triangular, out- er margin convex, directed anteriorly, outer margin distinctly convex, separated from rest of anterolateral margin by deep V- shaped cleft, rest of margin with 2 relatively low teeth; posterolateral margin gently con- vex, converging to almost straight posterior carapace margin (Fig. 1A, B). Merus of third maxilliped longitudinally ovate, anterior part widest, posterior part ta- pering sharply to ischium (Fig. 1C), dis- tinctly longer than ischium; ischium with shallow median sulcus; exopod slender, reaching to mid-length of merus, flagellum long, reaching across width of merus. Male chelipeds subequal (Fig. 1A, B); VOLUME 116, NUMBER 3 603 Fig. 1. Labuanium scandens, new species. Holotype male, 42.5 < 40.5 mm, TMCD 3290, Taiwan. A, overall view; B, carapace and chelipeds; C, frontal view. 604 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. outer surface with numerous small, uni- formly arranged rounded granules (Fig. 3A). Ventral surface of ischium with 2 short rows of tubercles, distal ones more promi- nent; ventral surface of basis granulated. Outer surface of merus with numerous rounded granules on outer surface, with low but discernible longitudinal median granu- lated ridge; inner margins prominently ser- Labuanium scandens, new species. Holotype male, 42.5 x 40.5 mm, TMCD 3290, Taiwan. A, anterior thoracic sternum and abdomen, ventral view; B, propodi and dactyli of first two ambulatory legs. rated, dorso-distal part dilated to form sub- lamelliform structure that is just visible from dorsal view. Outer surface of carpus finely granulated; slightly longer than broad; inner distal angle with long sharp tooth, distal and proximal margins serrated with some spines larger (Fig. 1B). Palm in- flated, outer, dorsal and inner surfaces with numerous small, uniformly arranged round- VOLUME 116, NUMBER 3 a ei ten “y * ee Pea bo Fig. 3. Labuanium scandens, new species. Holotype male, 42.5 40.5 mm, TMCD 3290, Taiwan chela. A, outer view; B, dorsal view; C, inner view. 606 ed granules (Fig. 3A); dorsal margin with a main, always distinct granulated but non- pectinated ridge on outer edge with 27 or 29 relatively more longitudinally elongate closely arranged granules, joining another less well defined (sometimes barely dis- cernible) granulated ridge which extends from proximal edge of palm and curving along inward surface and gradually disap- pearing as it reaches median part (Fig. 3B, C); ventral margin (including pollex) with small sharp granules. Fingers shorter than palm, forming basal proximal gape when closed (Fig. 3A); dorsal surface of dactylus with numerous granules, most of which have a corneous tip, proximal 8—11 gran- ules relatively smaller, scattered, subse- quent 8 or 9 granules relatively larger, ar- ranged in uneven row (Fig. 3B, C); cutting edges of dactylus and pollex with numerous prominent teeth (Fig. 3A). Third ambulatory legs longest (Fig. 1A). Carpus of all legs with 2 subparallel low ridges on outer surface (Fig. 1A). Outer sur- face of propodus with short curved low ridge on proximal part. Meri of all legs rel- atively slender; surfaces rugose, dorsal mar- gin gently serrated, subdistal tooth promi- nent (Fig. 1A). Dorsal and ventral margins of all dactyli with densely packed short se- tae forming brush-like structures; promi- nent brush-like setae present on distal half of ventral margins of first and second pro- podus, that on second propodus relatively less extensive; distoventral margins of pro- podi of other legs with tufts of short black setae, but not as dense or continuous, not distinctly brush-like (Fig. 2B). Surfaces of thoracic sternites gently ru- gose; sternites 3 and 4 separated by almost straight, setae-lined ridge; abdominal cavity reaching to median part of sternite 4 (Fig. 2A). Abdomen triangular (Fig. 2A); telson shorter than segment 6, lateral margins gently convex on distal part but almost straight proximally, tip rounded; segment 6 with distal part of lateral margins convex, proximal part almost straight; segments 3— 5 increasingly trapezoidal; lateral margins PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of segment 5 gently convex, that of seg- ment 4 gently concave, that of segment 3 gently convex (Fig. 2A). Segments 1 and 2 transversely narrow. G1 relatively stout, almost straight; distal surface strongly setose, obscuring margins, distal part sharply bent 90° from vertical, distal chitinous part gently upcurved from lateral view (Fig. 4). G2 short. Etymology.—The name is derived from the Latin “‘scansus’’ for climb, alluding to the habits of this species. The name is used aS a noun in apposition. Distribution.—Known for certain only from Taiwan thus far. Variation.—The holotype is the largest specimen of L. scandens available, and oth- er than having its left branchial region slightly depressed (probably because of in- jury shortly after it molted), it is in excel- lent condition. Differences between sexes in this species are not substantial, most ob- vious being the proportionately more slen- der chela of females. The general forms of the male and female chelae are similar. Smaller specimens of L. scandens tend to be more squarish, with the lateral carapace margins almost straight or only gently con- vex; and the dorsal surface is relatively flat- ter. In larger specimens, the lateral carapace margins become prominently more convex and the dorsal surface is relatively more swollen. This is also true of L. rotundatum. In the series of specimens of L. rotundatum examined, the frontal margin of females may be more denticulate, appearing weakly serrated, although this is not always the case, with even a few of the smaller males also having a denticulate margin. The fron- tal margin of L. scandens is usually entire, although in some specimens, appears un- even; with only three or four individuals having distinct denticles. Based on the specimens of L. rotundatum examined, the presence of denticles along the frontal mar- gin does not appear to be correlated with SexeOMsiZe: The strength of the anterolateral teeth of L. scandens varies. In smaller specimens, VOLUME 116, NUMBER 3 607 Fig. 4. Labuanium scandens, new species. Left Gls (denuded). A—-E, holotype male, 42.5 x 40.5 mm, TMCD 3290, Taiwan; F—I, paratype male, 35.5 X 34.2 mm, ZRC 2001.29, Taiwan. A, FE dorsal (sternal) views; B, marginal view; C, G, ventral (abdominal) views; D, H, distal part, dorsal (sternal) views; E, I, distal part, ventral (abdominal) views. Scales equal 1.0 mm (A-C, E G), and 5.0 mm (D, E, H, I). 608 the two teeth are usually more pronounced and sharper, and in larger specimens, the second tooth is invariably low, sometimes barely discernible. The cleft on the supra- orbital margin is usually absent, with the margin appearing entire being present only on the holotype and a few other specimens. The strength of the inner dorsal granulated ridge of the palm varies a great deal, and in some specimens it is almost absent. The outer ridge on the other hand, 1s always dis- tinct, although the number of granules varies from 26 to 32. The density and extent of the brush-like setae on the ambulatory propodi and dactyli does not differ substan- tially between the sexes, but smaller spec- imens invariably have less setae, with the ventral margins of the propodus being often almost glabrous. These observations are also valid for L. rotundatum. Remarks.—lt is rather surprising that none of the five junior synonyms of L. ro- tundatum are conspecific with the present material from Taiwan. Still, those taxa had all been described, often from limited ma- terial, from various parts of the south and southwest Pacific, and Hawaii, with one re- cord from Nias in the eastern Indian Ocean, viz. Sesarma dentifrons A. Milne-Edwards, 1869 (one female, type locality Samoa), Se- sarma oceanica De Man, 1889 (one male, one female, type locality Ponape Island), Sesarma (Episesarma) rotundata papuom- alesiaca Nobili, 1899 (two males, one fe- male, type localities Nias Islands [western Sumatra, Indonesia], New Guinea), Sesar- ma gardineri Borradaile, 1900 (six males, four females, type locality Funafuti, Rotu- ma Islands), and Sarmatium faxoni Rath- bun, 1906 (three males, four females, Mar- shall Islands and type locality Oahu [Ha- wail]). Sesarma rotundatum Hess, 1865, s. str. was described from one male, ostensibly from Sydney, Australia, but this data is likely to be incorrect (see “‘Remarks”’ for L. rotundatum). De Man (1891, 1896) and Tesch (1917) provide compelling argu- ments to show that S. dentifrons, S. ocean- icum, S. gardineri, S. rotundata papuoma- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lesiaca and S. faxoni are all junior syno- nyms of L. rotundatum. Although we have not examined the type material for most of these species, fortu- nately, Hess (1865), A. Milne-Edwards (1869), De Man (1889), Nobili (1899), Bor- radaile (1900), Rathbun (1906) all provided good figures and/or sufficiently detailed de- scriptions for their species (supplemented by the detailed comments by Man 1891, 1896; Tesch 1917), and we are confident their taxa are conspecific with the excellent series of specimens of L. rotundatum we have from Guam and elsewhere. All these specimens are characterized by their cara- paces being prominently striated on the lat- eral surfaces, the median surface is strongly granulated, the regions are not well indi- cated and the gastric and branchial regions are not prominently swollen (Fig. 5), the outer and inner surfaces of the chelipedal carpus and chela having relatively few but larger and well spaced large conical or rounded granules, with the areas between them smooth (Fig. 7); the dorsal margin of the male chela possessing a outer ridge which has 10 to 14 relatively large granules (Fig. 7B); and the lateral margins of adult male abdominal segment 6 gradually di- verging posteriorly (Fig. 6A). The carapac- es of males tend to have smoother frontal margins (often more denticulate in females, see “‘Variations’’) and there is a clear ten- dency for the carapace to become propor- tionately wider with size, especially in males. Comparisons of the excellent series of specimens from Guam and Taiwan reveal numerous differences that confirm that we are dealing with two different species. Compared to L. rotundatum, L. scandens has the carapace granules and striae dis- tinctly smaller and lower (but more numer- ous), with the regions more prominent, the gastric and branchial regions distinctly swollen and the groove separating them broader and deeper (Fig. 1); the outer and inner surfaces of the chela has smaller but far more granules, the entire surface ap- VOLUME 116, NUMBER 3 609 Fig. 5. Labuanium rotundatum (Hess, 1865). Male, 42.9 39.7 mm, ZRC 2002.454a, Faifai Beach, Guam. A, overall view; B, carapace and chelipeds; C, frontal view. 610 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Labuanium rotundatum (Hess, 1865). Male, 42.9 x 39.7 mm, ZRC 2002.454a, Faifai Beach, Guam. A, anterior thoracic sternum and abdomen, ventral view; B, propodi and dactyli of first two ambulatory legs. pearing granular (Fig. 3); the dorsal margin of the male chela has an outer ridge with more (26-32) relatively smaller granules (Fig. 3B); and the lateral margins of adult male abdominal segment 6 are subparallel (Fig. 2A). These differences are apparent even for smaller male specimens for both species. The Gls of the two species also differ, with that of L. scandens being rela- tively more slender with the chitinous distal part relatively longer and bent at an angle of 90° (vs. relatively shorter and bent slight- ly upwards) (Fig. 4 vs. Fig. 8A—D, H, ID. The GI structure is slightly variable within VOLUME 116, NUMBER 3 611 Fig. 7. Labuanium rotundatum (Hess, 1865). Male, 42.9 * 39.7 mm, ZRC 2002.454a, Faifai Beach, Guam. Left chela. A, outer view; B, dorsal view; C, inner view. 612 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 8. Labuanium rotundatum (Hess, 1865). Left Gls (denuded). A-—D, male, 42.9 X 39.7 mm, ZRC 2002.454a, Faifai Beach, Guam; E—G, male, 28.0 X 29.1 mm, BPBM 4191, Swain’s Island; H, I, male, 27.9 x 17.0 mm, ZRC, Haputo Beach, Guam. A, E, FE H, dorsal (sternal) views; B, G, I, ventral (abdominal) views; C, distal part, dorsal (sternal) view; D, distal part, ventral (abdominal) view. Scales equal 1.0 mm (A, B, E—I), and 0.5 mm (C, D). VOLUME 116, NUMBER 3 each species, but when specimens of equiv- alent or similar sizes are compared, the dif- ferences are marked. The specimens of L. rotundatum from Swain Island in Samoa (BPBM 4191) are rather problematic in that their carapace is somewhat less granular compared to those from Guam (but still more so than in L. scandens), and their Gls appear to be more slender (Fig. 8E—-G). In fact, the overall form of the GI of these specimens is more similar to that of L. scandens, although the distal chitinous part is gently bent upwards like in L. rotundatum rather than at 90°. Also notable is that the carapaces of both males from Samoa are longer than broad. No other specimens of L. rotundatum ex- amined so far from elsewhere have such proportions, their carapaces always being broader than long. The form of their chelae, however, is typical for L. rotundatum. Nev- ertheless, in lieu of more specimens, it seems best to refer these specimens to L. rotundatum for the time being. However, it is quite possible that there is more than one species belonging to what is here identified as L. rotundatum, and some of the syno- nyms may be shown to be valid in the fu- ture. But the Gls of more specimens from throughout the Pacific must be examined before more can be said. In any case, there is no doubt that the Taiwan specimens are very different from L. rotundatum and all the other taxa now synonymized with it. Labuanium scandens is known only from southern Taiwan thus far. Although Sakai (1939) recorded L. rotundatum from Tansui in northern Taiwan, we have not yet found it there. The coastal areas of northern Tai- wan have been heavily developed in recent years and pristine habitats are no longer ex- tant there. Sakai’s (1939: pl. 110 fig. 4, 1976: text fig. 362) figure of L. rotundatum leave little doubt that his specimen is con- specific with L. scandens. Color.—The color of L. scandens varies somewhat with age. Smaller specimens are usually darker brown and mottled with light brown to white blotches and dark specks. 613 Larger specimens are usually of a more uni- form darker brown. The chela of larger males and females are mainly dirty white, with the dactylus purplish. In sharp con- trast, adult L. rotundatum are usually a more uniform grey to dark greyish-brown, although smaller ones have a blotchy car- apace. In general, the color of L. rotunda- tum is darker than L. scandens. The chelae of L. rotundatum are also uniformly white to dirty-white. Notes on habits.—Labuanium scandens is a phytotelmic tree-climbing species. Its presence is closely associated with the pres- ence of closed forest canopy and the avail- ability of water-filled tree-holes. Tree spe- cies where the crabs have been recorded from are Hernandia sonora, Barringtonia asiatica and Aglaia elliptifolia. The tree cavities in which L. scandens live can sometimes be deep, occasionally over 50 cm. The tree holes collect rain-water, and the pools vary in depth from several cen- timeters to over 40 cm, depending on the season. The phytotelm water ranges from clear to dark-brown and turbid with a lot of sediment. The tree holes in which the crabs are found are always sheltered from direct sunlight. Labuanium scandens not only uses these holes as refuges, but also molts there, under the water. In the several cases when L. scandens was observed molting in the wild, the exuviae were not eaten soon after, disappearing only after a few days, although it is not certain if the crab itself is responsible. In three cases, the crab molted in captivity, and the exuviae were con- sumed by the crab shortly after it had hard- ened. Labuanium scandens has been found to inhabit holes between 0.5 to 4 m from the ground, but have been observed climb- ing over 5 m high in trees. It is primarily nocturnal, hiding in tree-holes during the day. Of the over 52 specimens of L. scan- dens observed (not all collected), the only time they were found on the ground is when they are ovigerous and are making their way to the sea. Ovigerous females can be found from July to October. The eggs are 614 small and hatch out into pelagic, free-swim- ming larvae. The release of larvae seems correlated with the lunar cycle, with fe- males making their way to the sea during the full moon period. Not much is known about their diet. Specimens have been observed to scrape al- gae or lichens off the tree bark, as well as feed on dried leaves and flowers. The che- lae of L. scandens are not particularly adapted to scrape algae off trees like in many other known tree climbing sesarmid crabs (Sivasothi et al. 1993, Lim et al. 1999), suggesting that this species has a more varied diet, probably including animal matter. Labuanium rotundatum (Hess, 1865) Figs. 5-8 Sesarma rotundata Hess, 1865:149, pl. 6 fig. 9 (type locality ostensibly Sydney, Australia, but see ““Remarks’’). Sesarma dentifrons A. Milne-Edwards, 1869:31 (type locality Samoa). Sesarma oceanica De Man, 1889:429, pl. 10 fig. 9 (type locality Ponape Island). Sesarma (Episesarma) rotundata var. pap- uo-malesiaca Nobili, 1899:268 (type lo- calities Nias Islands [western Sumatra, Indonesia], New Guinea). Sesarma gardineri Borradaile, 1900:593, pl. 42 fig. 8 (type locality Funafuti, Ro- tuma Islands). Sarmatium faxoni Rathbun, 1906:841, pl. 7 fig. 1 (type locality Oahu, Hawai1). Labuanium rotundatum—Seréne & Soh 1970:401; Davie, 2002:221. (For complete synonymy, see Tesch 1917: 193) Material examined.—7 males, 16.4 X 15.9 mm, 316 < 30:8 mm, 34.4. x 32:1 nn, Son < SSO ion, 39.3 xX BY.y isan, ANS) XK BXO) 7) seat, 413.7 OX GLY inovom., 3) itS= males, 15.1 X 14.8 mm, 28.1 X 27.3 mm, 42.5 X 39.7 mm, ZRC 2002.454, in forest, Faifai Beach, near Gun Beach, Tumon Bay, Guam, coll. H.-C. Liu & P. K. L. Ng, 28 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Jul-Ol Aug 2001; 2 males, 22.2 X* 21.1 pam, 22-5212 mimes ZRE 20027457 Rae tidian Point, northern Guam, coll. H.-C. Liu & PK. L. Ng, 31 Aug 2001; 2 males, 27.9 94560 and 94561 (SEM 894, 1712-1716) (reported by Grasshoff, 1982b). Lesser Antilles: BL-232, MCZ 4807, sev- eral colony and branch fragments now sep- arated into 8 lots, alcohol and dry, previ- ously reported as N. regularis by Deich- mann (1936). Lesser Antilles: BL-233, MCZ 4808, sev- eral branches now separated into two lots, alcohol, previously reported as N. regularis by Deichmann (1936). Photographs of syntypes (MOM). Types.—Four syntypes are deposited at the MOM, one in alcohol and three dry (Carpine & Grasshoff 1985), cataloged as 12 0011. Type Locality: Hirondelle station 105: 38°23'45"N, 30°51'30"W (off Azores), DDH) sia Description.—Colonies have a short, stout main stem, which bifurcates into 2 subsidiary branches, the angle between these two branches being 100°-110°, each of which give rise to a series of regularly spaced (every 12—15 mm) branches on their facing sides, altogether forming a uniplanar, . 13) Feb: 623 lyriform colony shape. The largest speci- men examined (O-2644) has a main stem 2 cm in height and 5 mm in axial diameter, a total colony height of 43 cm, and approxi- mately 14 branches that originate from each of the subsidiary branches. The branches originating from the two subsidiary branch- es are up to 28 cm in length, most of which bifurcate only once, some of which bifur- cate twice, and several of which are un- branched. The axis is strongly calcified, fairly stiff, and golden-yellow in color; it bears fine longitudinal grooves. Polyps are arranged in whorls of 3—8 (average 5), the lower number characteristic of whorls at the ends of branches, the higher number found on larger-diameter basal branches. Whorls and polyps are so closely spaced that the branch coenenchyme cannot be seen, 10— 13 whorls occurring in 3 cm branch length, the tips of the opercular scales almost touching the basal scales of the adjacent proximal polyp and the sides of each polyps touching their adjacent polyp in the same whorl. As is characteristic for the genus, the polyps point downward; the length of a pol- yp, as measured parallel to the axis from basal scale to the tip of the operculars, is 2.0—2.2 mm. Each polyp is protected by three pairs of large abaxial body scales and a pair of small adaxial buccal scales. The pair of basal scales are quite prominent, standing perpen- dicular to the axis up to 1.8 mm in height, the upper half of each of these scales pro- jecting above the medial scales as a tall, broad, rounded lobe. The sides of these bas- al scales curve around the base of the polyp and meet on the adaxial side in a ring struc- ture, but do not fuse. At the point of cur- vature from the abaxial to lateral edges, there is usually a low longitudinal ridge on the lower half of the sclerite. The pair of medials are much smaller sclerites (0.7—0.8 mm in length), roughly square, and slightly creased transversely, resulting in a slight distal flare; they do not meet on the adaxial side. The pair of abaxial buccals are quite large (up to 1.3 mm in length and 0.90 mm 624 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Narella bellissima: A, B, JS-47, USNM 57444, a complete polyp whorl and lateral view of a polyp; C, JSL-IT 815, USNM 1004866, adaxial view. Stereo views. All scale bars are 0.5 mm. VOLUME 116; NUMBER 3 in width), and, although they do not meet on the adaxial side, form a protective cone enveloping the distal polyp and opercular scales, such that only the tips of the oper- cular scales can be seen from side or ab- axial view. The distal edges of these buc- cals are quite thin and delicate, appearing translucent for the distal 0.1 mm because of a lack of tubercles on their inner surface of this region. The three pairs of large abaxial scales overlap each other slightly on the sagittal line and are covered by very small granules arranged in lines that radiate from a central point on each scale about % the distance up from its base; however, the granulation is so fine that these scales ap- pear smooth. A pair of quadrate-shaped ad- axial buccal scales also occurs just beneath the adaxial operculars, these scales measur- ing about 0.3 mm in greater width. Their inner surface is tuberculate, the outer gran- ular, and their distal edges finely serrate, the serrations being 2—6 pm in width. Rarely, one can observe in some specimens 10-12 pairs of vestigial adaxial scales that are el- liptical in shape and only about 0.1 mm in greater diameter. The adaxial operculars are elongate-tri- angular and symmetrical, up to 0.7 mm in height and 0.31 mm in basal width (H:W = 2.1-2.9). They are longitudinally con- cave, sometimes bearing a short medial ridge within this concavity, and also bear- ing a prominent medial keel on the distal half of its under surface. The inner- and outer-lateral and abaxial opercular scales are only slightly taller and wider (up to 0.75 mm tall and 0.47 mm wide: H:W about 1.3-16) and asymmetrical, such that there is a slight shoulder on the adaxial side, making the interior keel appear off-center. The abaxial operculars are only slightly taller but much wider than the laterals, sometimes wider than tall (e.g., 0.78 mm tall and 0.80 mm wide: H:W = 0.81-1.0) and symmetrical, bearing shoulders on each side at mid height, making the longitudinal interior keel centrally placed. The opercular scales have a progressively deeper longitu- 625 dinal groove exteriorly and a more promi- nent keel on their inner surfaces in the ad- to abaxial gradient. At first glance, in situ, it would appear that the operculars are flat with a medial ridge, but when disarticulat- ed, it can be seen that these sclerites are deeply longitudinally grooved with a keel on lower face. When closed, the opercular scales form a tight, somewhat overlapping crown around the polyp, which is also pro- tected by the abaxial buccals, all of which are directed toward the branch axis. The tentacles contain small (80 wm length), granulated, slightly curved spindles. The coenenchymal scales are elongate and irreg- ular in shape, arranged in one layer, up to 1.2 mm in length, and usually longitudinal- ly ridged. The inner surfaces of all sclerites are tuberculate, except for the distal edges of the body wall scales. The tubercles are roughly 10 wm in diameter and spiny, the individual spines being about 1 wm in di- ameter. Comparisons.—Narella bellissima is the only Atlantic species in the genus to have dichotomous branching that results in a lyr- iform colony. Based on Table 1, it is most similar to N. alvinae, both species having prominent lobate basal scales, but N. bellis- sima differs in having smaller polyps and consequently more whorls per cm, less prominent ridging on the body wall scales, proportionately longer buccal scales and smaller operculars resulting in the opercu- lars being almost completely hidden from view, and occurs at a shallower depth. Distribution.—Eastern Atlantic: Bay of Biscay off France and NW Spain (Grass- hoff 1982b, 1986); off Portugal (Grasshoff 1986); Azores (Studer 1901, Tixier-Duri- vault & d’Hondt 1974); Madeira (Thomson 1927, Grasshoff 1982a); Canary Islands (Grasshoff & Zibrowius 1983); Cape Verde Islands (Grasshoff 1986); ?Dakar (Grass- hoff & Zibrowius 1983); 225-1968 m. Western Atlantic: Lesser Antilles and Ba- hamas (Fig. 14); 161—792 m. Remarks.—Studer (1901) was the first to describe and beautifully illustrate a speci- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. A—D, Narella bellissima (A, B, D, O-2644, USNM 52776; C, P-881, USNM 52777): A, underside of abaxial, outer-lateral, inner-lateral, and adaxial opercular scales; B, three coenenchymal scales; C, a tentacular scale; D, an adaxial buccal scale. E-H, Narella regularis from Alb-2752, USNM 49385: E, underside of abaxial, outer-lateral, inner-lateral, and adaxial opercular scales; EK three coenenchymal scales; G, two tentacular scales; H, an adaxial buccal scale. Scale bars: A, B, E, F = 0.5 mm; C, G = 0.05 mm; D, H = 0.25 mm. VOLUME 116; NUMBER 3 men of this species, but incorrectly identi- fied it as Stachyodes trilepis (see Bayer, 2001). Versluys (1906) examined Studer’s specimens, recognized the mistake, but, having examined only a small fragment of the type, declined to name it. He did note, and we concur, that Studer’s illustration (Studer 1901:pl. 5) of the larger syntype, which was purportedly drawn at natural size, depicts the polyps at almost twice their normal size, and thus results in only half as many whorls per cm. Finally, Ktikenthal (1915), in one line and one additional cou- plet in a key, provided the name S. bellis- sima for this taxon. He (Kiikenthal 1919) later redescribed the species, but incorrectly reported the depth of capture as 1700 m, apparently misinterpreting the depth given by Studer in meters as fathoms. In the same paper, Kiikenthal (1919) also reported S. re- gularis from St. Vincent, but examination of this specimen shows it to be typical N. bellissima. Likewise, examination of all the specimens reported as N. regularis by Deichmann (1936), which includes Kiiken- thal’s specimen, shows them to be WN. bel- lissima. Although widespread in the eastern Atlantic (see Distribution), these are the first correctly identified records of N. bel- lissima from the western Atlantic. Histological sections prepared from sam- ples of USNM 52772 specially fixed in Bouin’s fixative confirm Studer’s observa- tion (1901:pl. 11, figs. 3, 6) on Stachyodes trilepis (=Narella bellissima) that the sulcal side of the polyps is abaxial. The longitu- dinal muscles of the mesenteries are very well developed, as would be expected in polyps capable of strong contraction and adaxial flexion, as illustrated by Studer GEO ONEpIRSIE): Narella regularis (Duchassaing & Michelotti, 1860) Figs. 1B, 3E—H, 4A-C, 7 Primnoa regularis Duchassaing & Michel- otti, 1860:17, pl. 1, fig. 12, 13.—Ko6lliker, 1865:135, pl. 17, fig. 13.—Duchassaing, 627 1870:13 (isted).—Wright & Studer, 1889:56. Not Stachyodes regularis Wright & Studer, 1889:55 Gunior secondary homonym, re- placement name: S. studeri Versluys, 1906).—Kitikenthal, 1919:466—467 (=N. bellissima). Stachyodes regularis.—Versluys, 1906: 96.—Kiikenthal, 1924:316. Not Narella regularis.—Deichmann, 1936: 169.—Tixier-Durivault & d’Hondt, 1974: 1412.—Grasshoff, 1982a:738, map 3; 1982b:947 (all = N. bellissima). Narella regularis.—Bayer, 1956:F222, fig. OK) Material examined.—Bahamas, off Chub Cay, Berry Islands: 25°23.40’'N, 77°55.25'W, 792 m, JSL-H-815, 13 Apr 1984, 10 branch- es cut from a large colony, USNM 1004861 (SEM 1728). Off northwestern Cuba: 23°10'N, 81°28'W, 366 m, AZl-3480, 11 May 1939, 1 branch 6 cm long, MCZ 3674. Off northwestern Cuba: 23°12’N, 81°22'W, 777 m, Atl-3469, 9 May 1939, 2 large colo- nies (one 45 cm tall and 11 mm in basal stem diameter), MCZ 3601. Off St. Vincent: 13°34’00"N, 61°04’00’W, 514 m, bottom temperature 8.4°C, Alb-2752, 4 Dec 1887, 4 nearly complete colonies and 4 detached branches, USNM 49385 (SEM SIL, SIG, CMOS). Holotype (see below). Types.—The holotype of P. regularis is deposited at the Museo Regionale di Scien- ze Naturali, Turin (Coel. 275), but all pol- yps had fallen from the axis, making it dif- ficult to characterize the species. Type Lo- cality: Guadeloupe, Lesser Antilles, depth unknown. Description.—Colonies have a short main stem, above which they are equally and dichotomously branched at approxi- mate intervals of 15—25 mm, resulting in a uniplanar (not lyrate) fan. The largest col- ony examined (A/b-2752) has a main stem of at least 3.5 cm (but is broken from the substrate) and 2.7 mm in axial diameter, a 628 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Narella regularis from Alb-2752, USNM 49385: A, a complete polyp whorl; B, polyp in lateral view; C, polyp in adaxial view. Stereo views. All scale bars are 0.5 mm. VOLUME 116, NUMBER 3 total colony height of 19 cm, and 37 ter- minal branchlets, the longest branchlet the result of 7 bifurcations. The axis is heavily calcified, fairly stiff and golden-yellow in color. Polyps are arranged in whorls of 4 or 5, the predominant number being 5. Whorls of polyps are fairly closely spaced, such that 11-13 whorls occur in 3 cm branch length, the tips of the opercular scales sep- arated by about 0.45 mm from the basal scale of the adjacent polyp. The sides of each polyp do not touch each other, but are separated by 0.1—0.2 mm. The downward pointed polyps measure 2.0—2.3 mm in length parallel to the axis. Each polyp is protected by 3 pairs of large abaxial body scales and a pair of smaller adaxial buccals. The sclerites of the basal pair project at a downward angle of about 60° from the axis and are about 1.25 mm in height, only the distal 0.1—0.2 mm extending beyond the junction with the me- dials. The basal scales curve around the body of the polyp but do not meet on the adaxial side. The pair of medials are much smaller (e.g., 0.50—0.65 mm long) and nar- rower, projecting at a downward angle of about 45° from the axis and having a distal outwardly flared margin. The abaxial buc- cals are slightly more elongate (e.g., 0.70— 0.80 mm) and much broader, oriented es- sentially parallel to the branch axis and forming a cone that encircles only the very base of the opercular scales. The outer sur- face of the sclerites is distinctly roughened by close-set, sharp granules which impart a rough, shagreen-like appearance, and the inner surface where in contact with the me- sogloea, by crowded, complex tubercles. The basal pair of body scales is ornamented by several sharply raised irregular ridges ra- diating from the depositional center or “‘nu- cleus,’’ and similar, but less conspicuous, ridges mark the outer surface of the medial and buccal scales as well. The pair of small adaxial buccals are rhomboidal in shape, about 0.40 mm in greater diameter, and have a finely serrate distal edge. The adaxial operculars are symmetrical 629 isosceles triangles, varying in height from 0.6 to 0.9 mm and up to 0.45 mm in basal width (H:W = 2.0—2.2). The inner- and out- er-lateral operculars are larger (up to 1.0 mm tall and 0.65 mm wide: H:W = 1.5-— 1.7) and asymmetrical, a bearing a slight shoulder on their adaxial side. The largest (i.e., abaxial) operculars are symmetrical, up to 1.1 mm in height and about 0.75 mm wide (H:W = 1.45—1.55). They are slightly notched basally, thus resembling certain In- dian arrowheads. Thus all operculars, al- though differing in size and symmetry, have roughly the same H:W ratio, and all bear a longitudinal groove on their exterior surface and a prominent keel on their inte- rior surface. All of the opercular scales are easily seen in side view, except for the very base, which is protected by the abaxial buc- cals. The tentacles contain small (65—75 wm long and about 20 pm wide), straight, gran- ular rodlets. Coenenchymal scales are elon- gate and somewhat irregular in shape, ar- ranged in one layer, up to 0.75 mm in length, and ridged, the ridges sometimes anastomosing on the surface. Comparisons.—In addition to those char- acters mentioned in Table 1, N. regularis differs from N. bellissima in having larger, consistently triangular opercular scales (H:W always >1.5), which are largely un- protected by the abaxial buccal scales; a curved polyp, instead of one that is parallel to the branch axis; a smaller ratio in length of their buccal scales in relation their me- dial scales; lacking a distal translucent re- gion; basal scales that do not meet on the adaxial side; and more highly ridged body wall and coenenchymal scales, the ridges of the latter sometimes anastomose. Distribution.—Lesser Antilles, Bahamas, off Cuba (Fig. 7); 366—792 m. Remarks.—Eliminating the misidentifi- cations and simple listings of this species, as indicated in the synonymy, these are the first legitimate records of N. regularis sub- sequent to its description. 630 Narella pauciflora Deichmann, 1936 Figs. 1D, 5A—C, 6A—D, 7 Narella pauciflora Deichmann, 1936:170, Ol Ads WSs By Olls ZA, itis, 2: Material examined.—Off Morro Light, Cuba: 1473 m, BL-2, date, 3 branches, MCZ 4807?. Off Campeche Bank, Mexico: 23°52'N, 88°58'W, 1473 m, BL-35, date, 1 complete colony and several branches, MCZ 4814, 48 14a. NW of Cardenas, Cuba: 23°54’'N, 81°27'W, 1153-1190 m, G-375, 17 Sep 1964, several badly damaged branches re- taining only a few polyps, USNM 52771. Off northwestern Cuba: 23°24'N, 81°00'W, 676-1106 m, Atl-2995, 16 Mar 1938, 3 colony fragments, MCZ 3872. Off northwestern Cuba: 23°22'N, 81°O5'W, 859-1216 m, Atl-2996, 16 Mar 1938, 3 colony fragments, MCZ 3831. South of Cay Sal Bank: 23°51.9'N, 80°42.7'W, 1079-1089 m, G-1111, 30 Apr 1969, 9 branches, the largest 19 cm tall, USNM 52772 (SEM 377, 377A, C1013). N of Cardenas, Cuba: 23°51’N, 81°02’W, 1107-1162 m, G-372, 16 Sep 1964, 2 branches, the larger 12 cm tall, with one bifurcation, the other unbranched and partly denuded, USNM 52773. NE of Havana, Cuba: 23°45'N, 81°49'W, 1400-1395 m, G-965, 1 Feb 1968, 5 bro- ken branches, the largest 21 cm tall, with ophiuroids attached, cortex and polyps rubbed off in several places, USNM 52774. Off St. Lucia: 14°17'N, 60°45.2’W, 1280 m, P-892, 7 Jul 1969, 3 branches, the larg- est 27 cm tall with several bifurcations, ap- parently about half of a colony, without holdfast, severely damaged, with attached ophiuroids, USNM 52775 (SEM 1729). Tongue of the Ocean, Great Bahama Bank: 24°02'N, 77°17'W, 1335 m, CI-19, 6 Jul 1972, one nearly complete colony with part of calcareous holdfast, 31 cm tall, ter- minal branchlets broken off, ophiuroid at- tached, dry, USNM 57571 (SEM 1704); 5 branches, the largest 22 cm tall, and smaller PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON fragments more or less damaged, with ophi- uroid attached, USNM 57572. Tongue of the Ocean, Great Bahama Bank: 23°29’N, 77°05'W, 1234 m, CI-46, 24 Feb 1973, several badly broken frag- ments, the largest about 18 cm tall, with ophiuroids attached, in alcohol, USNM 57573 (SEM 406, 1730). Tongue of the Ocean, Great Bahama Bank: 23°40’N, 77°08'W, 1372 m, CI-47, 24 Feb 1973, one badly damaged branch about 10 cm tall, most polyps lost, USNM 57574. Straits of Florida between Delray Beach, Florida, and Grand Bahama Island: 26°24'N, 79°36'W, 738 m, CI-140, 28 Sep 1973, 6 broken branches severely decorti- cated, the largest 23 cm tall, USNM 57575. NE of San Juan, Puerto Rico: 18°40'N, 65°58'W, 1446-1510 m, P-830, 4 Feb 1969, several broken branches, the largest 12 cm tall, and decorticated fragments, with ophi- uroids attached, USNM 57577. Locality unknown: unknown Blake sta- tion, 3 branches, MCZ 4813. Types (see below). Types.—The holotype (BL-124, MCZ 4809) consists of one branch 13 cm long and several smaller branches preserved in alcohol. The paratypes (BL-227, MCZ 4810 and 4810a) consist of one broken col- ony and several branches, also preserved in alcohol. Type Locality: 17°47'30'N, 64°53'45"W (off St. Croix), 1061 m. Description.—Colonies have a stout main stem that is entirely calcified (white) and which bifurcates repeatedly in a fairly regular fashion (dichotomous branching) resulting in a uniplanar colony. The largest specimen (C/-19) is 31 cm tall, having a calcified basal stem 13 mm in diameter and 24 mm in height. Branching is fairly sparse, branch segments between branching nodes are 10—5O0 mm in length, and the longest terminal branch is 14 cm, the distal most branch being the result of 6 successive bi- furcations. The angle between branches is only 25°-30°. Above the calcified base, branches are fairly stiff, golden-yellow in VOLUME 116, NUMBER 3 631 Fig. 5. Narella pauciflora from G-1111, USNM 52772: A, a complete polyp whorl; B, polyp in lateral view; C, polyp in adaxial view. Stereo views. All scale bars are 0.5 mm. 632 color, and longitudinally grooved. Polyps are fairly well separated, arranged in whorls of 2—5 (usually 3) and only 6—9 (usually 7) whorls occurring in 3 cm. Thus, the lateral edges of each polyp in a whorl are separat- ed from its adjacent polyps in that whorl by approximately the width of a polyp, and whorls are separated from one another by about 1.5 mm. As is characteristic for the genus, the polyps are pointed downward, measuring 2.6—2.8 mm in length. Each polyp is protected by 3 pairs of large abaxial body scales and 1-2 pairs of much smaller adaxial buccal scales. The sclerites of the basal pair project at a down- ward angle of about 70° from the axis, and are about |.1—1.3 mm in height, only the distal 0.2 mm extending beyond the junc- tion with the medials. The basal scales curve around the body of the polyp but do not meet on the adaxial side. The medials are smaller (about 1.0 mm), projecting at a downward angle of about 25° from the axis. The abaxial buccals are of intermediate size (about 1.15 mm) and much broader than the medials, oriented parallel to the axis or at a slight angle from it, forming a cone that encircles only the very bases of the eight opercular scales. Although somewhat vari- able, the surfaces of the 3 pairs of large abaxial sclerites are ornamented with prom- inent ridges, these ridges radiating from a center in the basal third of each sclerite. Otherwise, the outer surface of these scler- ites is fairly smooth. The pair of adaxial buccals are square to slightly rectangular in shape and about 0.7 mm in width. They are usually highly ridged, the ridges sometimes reticulate, and have a finely serrate upper edge. Often there is another pair of smaller scales about half the size of the adaxial buc- cals that are directly proximal to the adaxial buccals, which are also ridged. The adaxial operculars are tall symmet- rical triangles, and are by far the smallest opercular scales, about 1.15 mm in height and 0.35 mm in width (H:W of 2.9—3.3). The inner-lateral operculars have a broad base, which is asymmetrically elongated on PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the adaxial side to partially cover the ad- axial operculars. These sclerties are up to 1.5 mm in height and 1.0 mm broad at the base (H:W = 1.4—1.5). Outer-lateral and abaxial operculars are even taller (up to 2.0 mm) and about 0.75 mm in width, resulting in H:W of about 1.5—2.0, these sclerites sometimes having 2 or 3 pointed tips. All opercular sclerites are deeply creased me- dially, edges of adjacent sclerites almost parallel with one another, the crease corre- sponding to a thick medial keel on the in- terior surface of the sclerite. Opercular sclerites also bear low, radiating ridges on their outer surfaces, and each adaxial oper- cular often bears a small (0.15 mm in di- ameter), short tube on its lower adaxial margin (Fig. 5C) of unknown function. The tentacles contain small, straight granular rodlets 65—90 wm in length and about 15 wm in width. Coenenchymal scales are elongate (up to 1.9 mm), irregular in shape, and ridged, the ridges often radiating from a central point, sometimes longitudinal, and sometimes quite high and reticulate, the lat- ter case most common on those coenenchy- mal scales adjacent to the basal scales. Comparisons.—Narella pauciflora is perhaps closest morphologically to N. re- gularis (see Table 1), but can be distin- guished by its larger polyps, fewer polyps per whorl, fewer polyps per unit distance on a branch, and in often having two pairs of highly ridged adaxial buccal scales. Distribution.—Antilles and Bahamas (Fig. 7); 738-1473 m. Remarks.—This is the first report of this species since its original description. Narella alvinae, new species Figs. 6E-I, 7, 8A—D *‘delicate coral’’ Calder, 1993:1, 26—27 (color figures). Material examined/holotype.—25 km NW of Bermuda: 32°35’'N, 64°55’W, 3419 m, Alvin-2566, 17 Mar 1993: main colony now fragmented into 47 fragments, the larg- est 27 cm in height with 6 terminal branch- inner- >. ial buccal cal de of abaxial, outer-lateral, inner-lateral, and tel AX1 a ra l D, an ad al scale with a sol outer- 5 5 RIG E; > > SPRITE B 5 si. A underside of abaxial two tentacular scales , an isolated coenenchym D > A G. Ss cale bar i s FE branch with intact coenenchymal scale S - S ale E, unde scales. cal al buc N ™ ™ N va) Z Y =) = B, three coenenchymal sc two adaxi 5 al S scale . Narella pauciflora from G- 1 opercular scales 3 two tentacular A-D ° Fig. 6. lateral, and adaxia scale. E-I, holotype of Narella alvinae, USNM 100778 adaxial opercular scales H tall crest VOLUME 116, NUMBER 3 634 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a, es, ROMIZ B3521; 12 branches and hun- dreds of separate polyps, USNM_ 100778 (SEM C1023-26, 1030), all from the same colony and considered to be holotypic. Type Locality: as stated above. Description.—The holotypic colony is sparingly and dichotomously branched in one plane, resulting in 26 rather long ter- minal branches 10-13 cm long. Most of these branches have been broken from the main colony, but the entire colony is figured in situ by Calder (1993). The thickest branch axis is only 0.9 mm in diameter, fairly rigid, smooth, and golden-yellow in color. Polyps whorls are spaced 1.2—1.7 mm apart. There are 4 polyps within each whorl and about 7 whorls occur in 3 cm of branch length. Fully developed polyps are 2.7—3.1 mm in length. The polyps are covered by 3 pairs of large abaxial scales and 1 pair of adaxial Distribution of Narella pauciflora (circles), N. regularis (squares), and N. alvinae (triangle). marginals (buccals), sometimes with a few much smaller adaxials in a more proximal position on the polyp. The sclerites of the abaxial basal pair project perpendicular to the axis and are about 1.7 mm in height, the distal 0.6 mm projecting beyond the junction with the adjacent medials as a prominent lobe, the 2 of which together form a collar around the base of the polyp. Each basal sclerite bears a prominent lon- gitudinal crest up to 0.45 mm in height at the point of curvature from the abaxial to lateral edges of the scale; these scales do not form a complete ring and thus do not touch or fuse on the adaxial side. In fact, the lower proximal edge of each basal scler- ite is often slightly notched, the notch fitted to a slightly apically indented coenenchy- mal scale. Furthermore, the coenenchyme between the basals of 2 adjacent polyps consists of 2 characteristically long (up to VOLUME 116; NUMBER 3 1.8 mm) coenenchymal sclerites that bear unusually tall crests (up to 0.48 mm). The structure and position of the coenenchymal sclerites that surround the basal sclerites is consistent and would appear to “‘lock in” or stabilize the basal sclerites into the coen- enchymal sclerites, the interpolyp scales preventing lateral motion and those at the proximal ends of the basals preventing lon- gitudinal movement, which might explain why in this species it is common for most of the polyps to be abraded from the axis but the basals remaining intact on the axis. The medial sclerites are 1.4—1.6 mm in length, not ridged, and have a thin, trans- parent, slightly upward curved distal edge. Buccal sclerites are similar to the medials but wider, overlapping only the basal region of the opercular scales. All abaxial scales are coarsely granular exteriorly. A pair of well-developed (0.45 mm in width), flat, rectangular adaxials occur contiguous to the adaxial operculars. In some specimens smaller (0.15 mm), elliptical adaxial scler- ites occur lower on the polyp wall. The adaxial operculars are the smallest of the operculars, symmetrically triangular, up to 1.05 mm in height and about 0.38 mm in greatest (basal) width (H:W = 2.5-2.7). The inner- and outer-lateral operculars are slightly larger (up to 1.4 mm and 0.70 mm in width: H:W = 1.8—2.2), and asymmet- rical, the widest point being about a third of the distance from the base caused by a small lobe on the adaxial side of the scler- ite. The abaxial operculars are symmetrical, up to 1.4 mm in height and 1.0 mm in width (H:W = 1.23-—1.45), widest about a third up from the base cause by symmetrical lobes on either side of the sclerite, and bluntly tipped. All opercular sclerites are deeply longitudinally grooved externally, which corresponds to a prominent keel on their in- terior surface. The tentacular rodlets are flattened and granular, up to 97 wm in length and 24 wm in width. The coenen- chymal scales are thin, slender (0.25 mm), elongate (1.2—2.0 mm) plates, imbricating on their edges, each plate having one thin, 635 transparent, longitudinal ridge up to about 0.20 mm tall, which gives the branch axis a “frilly” aspect. As mentioned above, the coenenchymal sclerites adjacent to the bas- als have even taller ridges. Etymology.—Named in honor of the col- lecting vessel, the R/V Alvin. Comparisons.—Although similar to WN. laxa in branching pattern, N. alvinae is morphologically most similar to N. pauci- flora (see Table 1), both species having a similar number of polyps per whorl and whorls per cm, and polyp size; however, N. alvinae differs in having prominent lobes on its basal scales, sparse branching, and prominent crests only on the basal scales at the point of curvature from the abaxial to lateral edges. N. alvinae appears to be unique in having coenenchymal sclerites with prominent longitudinal crests. Distribution.—Known only from the type locality off Bermuda (Fig. 7), 3419 m. This is the deepest known record of a Na- rella and one of the deepest collected prim- noids on record. Narella versluysi (Hickson, 1909) Figs. 1K 9A—C, 1OA—D, 14 Calypterinus Allmanni (sic).—Roule, 1896: 303—304.—Versluys, 1906:93. Stachyodes sp.—Versluys, 1906:93—94. Stachyodes versluysi Hickson in Stephens, 1909:10—13.—Ktkenthal, 1919:456— 457; 1924:309.—Thomson, 1927:30—32, pl. 2, fig. 19, pl. 5, figs. 12—13 (in part: not station 970).—Stiasny, 1941:80-—81. Narella_ versluysi.mDeichmann, 1936: 171.—Grasshoff, 1982a:738, 747 (map 3); 1982b:946—947, figs. 13—14.—Car- pine & Grasshoff, 1985:33.—Grasshoff, IQSOZie Narella elegans Tixier-Durivault & La- fargue, 1968:622—626, figs. 3—4. Material examined.—Straits of Florida, OlleSt mlucic sinlety Hlondas 27-06) Ne 79°32'W, 677—659 m, G-170, 29 Jun 1963, one unbranched stem 35 cm long, lacking 636 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 8. Narella alvinae, holotype, USNM 100778: A, lateral view of polyp; B, polyp in adaxial view; C, lateral view of polyp; D, oral (opercular) view of polyp. A and B stereo views. All scale bars are 0.5 mm. VOLUME 116, NUMBER 3 base and apex, USNM 52900 (unnumbered SEM stub). Straits of Florida, between Palm Beach, Florida, and Settlement Point, Grand Ba- hama Island: 26°28’N, 79°33'W, 751 m, G- 808, 13 Sep 1966, 2 pieces 5.5 and 12 cm long, probably of a single unbranched stem, lacking base and apex, USNM 52902. Straits of Florida, off Delray Beach, Flor- ida: 26°24'N, 79°36'W, 738 m, C/-140, 28 Seppe se epieces. 7 and 37 smmelong: probably of a single unbranched stem, lack- ing base and apex, USNM 57570 (SEM 413, C1014—1015). Straits of Florida off Delray Beach, Flor- ida: 26°23'N, 79°37'W, 743-761 m, CI-246, 29 Oct 1974, 2 partly decorticated pieces 6 and 24 cm long, apparently part of one un- branched stem, lacking base and apex, USNM 57569. Straits of Florida west of Riding Rock, Bahama Islands: 25°26'N, 79°23'W, 796 m, G-304, 23 May 1964, 3 unbranched stems, the largest 30 cm long, USNM 52901 (SEM 3)//3))). Off northwestern Cuba: 23°18'N, 80°46'W, 896 m, AZl-3474, 10 May 1939, 2 branches the longest 17 cm, MCZ 3727 and 43647. Off Bermuda: 32°14.47'N, 64°47.07'W, 900 m, “‘tangled in crab pot’’, 15 Nov 1990, one dry branch fragment 9 cm long, USNM 1004792; parent lot 76 cm long, dry, Ber- muda Biological Research Station, uncata- loged. Types (see below). Types.—Five syntypes are deposited at The Natural History Museum, London (1962.07.20.172) and an SEM stub (404) of one of these syntypes is at the USNM. Type Locality: off Ireland; 698-914 m. Description (western Atlantic speci- mens).—Colonies evidently are un- branched, straight or nearly so, with a near- ly uniform diameter of 5—7 mm (including polyps) and attaining a length up to 34.5 cm (G-170), although eastern Atlantic spec- imens are known to be as long as 78 cm (Hickson 1909). The axis is heavily calci- 637 fied, brittle, and longitudinally grooved, with dull golden reflections, in the most ro- bust specimen only 2.2 mm in axial diam- eter. In one specimen, the proximal part of the axis shows calcareous thickening of the kind present just above the holdfast of many other species. Polyps are fairly close- ly spaced, the lateral edges of polyps within a whorl touching each other, and the oper- cular scales of the zooids of one whorl touching (or nearly touching) the basal sclerites of the polyps of the next whorl. Polyps are arranged in whorls of 4—7, and 8-10 whorls occur in 3 cm of branch length. Polyps) measure 3.2—3.7 mm in length, measured parallel to the axis from the basal part of the basal sclerites to the apex of the abaxial operculars. Each polyp is protected by 3 pairs of large abaxial body scales and a variable number of smaller adaxial scales. The scler- ites of the basal pair project perpendicular or at a downward angle of 60° from the axis and are about 2.1 mm in height, the distal 0.7—0.8 mm flared outward abaxially as a pair of rounded, smooth-edged, very thin, petal-like processes without marginal points or spines. The inner surface of one or both basal scales may have one or more promi- nent keels, which may be visible even in abaxial view and which articulate with a notch or groove on the lower side of the adjacent medial sclerite (Fig. 9A). The bas- al sclerites are smoothly curved around the base of the polyp but do not meet adaxially. The medials are the smallest of the large body wall scales, about 1.2 mm in length and fairly narrow, encircling only about half of the polyp. They also have a free, upturned distal edge that is very thin and delicate, oriented about 45° from the axis. The abaxial buccal scales are of interme- diate size (about 1.7 mm in length), orient- ed parallel to the axis, and also concave in shape, their distal margins thin and flared outward as the other 2 pairs. These buccal scales form a prominent, open collar or cone that encircles only the basal region of the 8 operculars. Except for the internal 638 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 9: Narella versluysi: A, G-304, USNM 52901, a complete polyp whorl; B, syntype, polyp in lateral view; C, Cl-140, USNM 57570, polyp in adaxial view. Stereo views. All scale bars are 0.5 mm. VOLUME 116; NUMBER 3 ridges of the basal scales, the exterior sur- faces of the body scales are not ridged or keeled, but covered with a prominent gran- ulation that seems to radiate from a central region of each scale. The adaxial side of each polyp is furnished with a pair of rather large marginal (buccal) scales, quadrate to tear-drop shaped, up to 0.40 mm in diam- eter, and having finely serrate edges. Ad- ditional pairs (up to 5 or 6) of smaller (0.14—0.30 mm width) adaxial scales occur proximal to these two marginal adaxial buc- cals, their distal margins becoming more coarsely serrate toward the base of the pol- yP. As is typical in the genus, the adaxial operculars are the smallest of the opercular scales and shaped as isosceles triangles; they are about 1.0 mm in height and 0.30— 0.50 mm in width (H:W = 2.1-3.3), often variable in width within the same polyp, one adaxial opercular being considerably wider than the other. The inner- and outer- lateral operculars are taller and broader, up to about 1.6 mm in height and 0.70—0.95 mm in width (H:W = 1.6—2.1). The abaxial operculars are of a similar height (1.5—1.6 mm), but are wider than the laterals at mid- height (resulting in a H:W of 1.2—1.3), and are blunt-tipped. All opercular scales bear a prominent ridge or keel on the distal half of their interior surface, the keel increasing in prominence in the ad- to adaxial direc- tion, that of the abaxial opercular as much as 0.45 mm in height. The inner- and outer- lateral operculars are asymmetric in shape, each scale having a slightly broader shoul- der on the adaxial side of the medial keel; the ad- and abaxial operculars are symmet- ric in shape. The tentacles are filled with thin, elongate (up to 80 wm long and 20 zm width), finely granular rodlets, some of which are curved to conform to the shape of the tentacle. Coenenchymal sclerites are thin and scale-like, up to 1.4 mm in length, and not closely fitted but commonly with one edge flared outward as a thin expansion projecting over neighboring scales and 639 sometimes with a thin crest on the outer surface. Comparisons.—Within the Atlantic, a branch fragment of N. versluysi could easily be confused with a terminal branch of N. bellissima, both species having prominent basal scales, body wall scales with thin dis- tal edges, and a similar number of polyps per whorl (Table 1). But, N. versluysi dif- fers in having larger polyps and larger-di- ameter branches; shorter buccal scales that expose most of the opercular scales; small, stellate adaxial scales; unridged basal scales; and fewer polyps per cm. Distribution.—Eastern Atlantic: off Ire- land, Bay of Biscay, Azores, off Portugal, off western Sahara (Grasshoff 1982a, 1982b); 550-3100 m. Western Atlantic: Bermuda; Straits of Florida (Fig. 14); 677— 900 m. The record off Spitsbergen Bank, Arctic Ocean at 48 m (Thomson 1927) is an unlikely locality and depth; Carpine & Grasshoff (1985) implied that there might have been a labelling error associated with this record. Remarks.—The present specimens agree in the main with Hickson’s description of Stachyodes versluysi from off the west coast of Ireland. The two obvious differ- ences may be superficial and of no taxo- nomic importance. First, in the present ma- terial, the smallest number of polyps in a whorl ranges from 4 to 7, whereas in the type material the smallest number reported is 9 and the largest “‘about 14.”’ Although this could be no more than a reflection of colonial size, the diameter of Hickson’s specimen at its distal end was not much larger than in the present material but there it had 9 polyps compared with 5 in the pre- sent specimens. In regard to number of pol- yps per whorl, the western Atlantic speci- mens are more similar to specimens report- ed by Thomson (1927) from the Azores, which have predominantly 7 or 8 (range 6— 10) polyps per whorl or those described by Tixier-Durivault & Lafargue (1968) from the Bay of Biscay as N. elegans, which have 4—6 polyps per whorl; however, in the 640 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 10. A—D, Narella versluysi (A, B, CI-140, USNM 57570; C, D, G170, USNM 52900): A, underside of abaxial, outer-lateral, inner-lateral, and adaxial opercular scales; B, two coenenchymal scales; C, a tentacular scale; D, two adaxial buccal scales. E~H, holotype of Narella laxa, USNM 11853: E, upperside of abaxial, outer-lateral, inner-lateral, and two adaxial opercular scales; B, two coenenchymal scales; C, a tentacular scale; D, three adaxial buccal scales. Scale bars: A, B, E, F = 0.5 mm; C, G = 0.05 mm; D (right), H = 0.25 mm; D (left) = 0.09 mm. VOLUME 116, NUMBER 3 latter case the polyps are also smaller, only 2.5 mm in length. But, Grasshoff (1982b) considered a polyp length range of 2.5—4.0 mm to be within the intraspecific range for this species. A second difference appears in the coenenchymal plates, which in S. ver- sluysi, according to Hickson (1909:11), are large ““compound”’ plates over 1 mm long, which break down into smaller “oblong, square, leaf-shaped, stellate and quite irreg- ular” scales upon prolonged treatment with caustic potash. Perhaps by this drastic treat- ment (not used in the present study) Hick- son was merely breaking up fully devel- oped scales that may have been weakened by rough handling during collection. J. A. Thomson’s description (1927:30) of 18 specimens from the Josephine Bank and off the Azores also agrees in general both with Hickson’s account and with the present western Atlantic material. However, the weak lateral crests on the basal scales de- scribed by Thomson but not evident in his beautifully drawn figure of an isolated pol- yp (1927:pl. 2, fig. 19), casts some doubt on the identification of his specimens as Narella versluysi. Two small pieces of Hickson’s type specimen, which one of us (FMB) has examined in The Natural His- tory Museum (London) through the kind- ness of Dr. P. E S. Cornelius, show no trace of crests on the basal scales, although ad- jacent coenenchymal scales definitely have crests. Previously known only from the eastern Atlantic, this is the first report of this spe- cies from the western Atlantic. Although generally assumed to be an unbranched spe- cies, some authors have qualified this de- scription to include “slightly branched’’. Indeed, very few, if any, colonies have been collected intact including a holdfast, and thus the nature of its branching is difficult to verify. Narella laxa Deichmann, 1936 Figs. 1E, 1OE—-H, 11A—C Narella laxa Deichmann, 1936:170, pl. 26, fig. 1.—Grasshoff, 1985:305. 641 Material examined types.—South of Georges Bank: 40°34'18"N, 66°09'00’W, 3186 m, bottom temperature 3°C, Alb- 2573, 2 Sep 1884: one dry colony 14 cm high and 6 cm wide, with a detached branch, USNM 11853 (holotype) (SEM 410, 411, C1027-28). Type locality: as stat- ed above. Same data: One fragment 1.5 cm long, with 4 whorls only, in alcohol. From the same haul as the holotype and probably a piece of it. The preservative has at some been slightly acid, as the sclerites are chalky in appearance and extremely fragile, USNM 49426. Same data: one fragment 2.8 cm long, with 7 whorls; dry, glued to a piece of black cardboard. From same haul as holotype and undoubtedly a piece of it, MCZ 4811. Description.—The large holotypic colo- ny is sparingly branched in one plane, the short (5.4 mm tall and 1.6 mm in diameter) main stem dichotomously dividing, the re- maining divisions all occurring within 15 mm of the first bifurcation, resulting in 9 undivided terminal branches 10—13 cm in length. The axis is stiff, brittle, smooth, light yellowish brown with scant indication of metallic luster in dry condition. Polyp whorls are fairly well spaced, a distance of about 2.5 mm separating each whorl, the polyps within a whorl touching laterally. Polyps are arranged in whorls of 3-5, of which 5-7 (or even 8 in places where whorls of young polyps are present between older ones) occur in 3 cm of branch length. Fully developed polyps are about 3 mm long, measured parallel to the axis; younger individuals are about 1.75 mm long. The polyps are covered by 4 pairs of large abaxial body scales (1 pair of basals, 2 pairs of medials, and 1 pair of buccals), 2—3 pairs of smaller adaxial scales, and oc- casionally one or more inner-lateral buccal scales. The sclerites of the basal pair project perpendicular to the axis, and are about 1.4 mm in height, the distal 0.4 mm extending as short, rounded lobes beyond the junction with the first medial pair. The basal sclerites 642 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 11. MNarella laxa, holotype, USNM 11853: A, abaxial polyp view; B, polyp in lateral view; C, polyp in adaxial view. Stereo views. All scale bars are 0.5 mm. VOLUME 116, NUMBER 3 are smoothly curved around the base of the polyp, are not ridged, and do not meet adaxially. The 2 pairs of medial scales and the pair of abaxial buccals are progressively longer and wider toward the distal end of the polyp, the buccals almost encircling the polyp; however, in some polyps there are one or 2 medium-sized (0.35—0.43 mm in width) inner-lateral buccals that intercalate with the abaxial buccals and the adaxial buccals. The extra pair of medial scales al- lows for a slightly more flexible polyp than in most other species of Narella, the oper- cular scales usually directed straight toward the axis in a preserved polyp. The buccal scales cover only the basal-most region of the opercular scales. None of the larger body scales are ridged or crested, but all bear a coarse granulation that is arranged in lines radiating outward from the nucleus. Two or three pairs of well-developed, rect- angular adaxial buccals lie below the ad- axial operculars, those closest to the oper- culars about 0.30 mm in width, those more proximal, decreasing to about 0.15 mm in width. The distal edges of these scales are finely serrate. The adaxial opercular scales are the smallest of the operculars, shaped as slen- der triangles 0.70—0.85 mm in height and 0.20—0.24 mm in basal diameter (H:W = 3.3—4.2). The inner- and outer-lateral oper- culars are larger, up to 1.1-1.2 mm in height and 0.51 mm in basal width (H:W = 2.1—2.5), the base asymmetrically ex- tended on the adaxial side. The large ab- axial operculars are symmetrical, about 1.0—1.3 mm in height and 0.65—0.70 mm in basal width (H:W = 1.8—1.9). The opercu- lar scales are fairly flat becoming only slightly concave on their distal regions, which corresponds to but a weak keel on the internal side. The tentacles contain curved, granular rodlets up 0.105 mm in length. The coenenchymal scales are elon- gate (up to 1.10 mm in length), more or less imbricating, commonly with a prominent longitudinal crest. Comparisons.—Only one other species 643 of Narella at present known has 4 pairs of body scales, N. spectabilis n. sp., described below. It differs in the presence of distinct longitudinal crests on all the abaxial body scales, prominently keeled operculars, finer and smoother external sculpturing, and more elaborately crested coenenchymal scales (see also Table 1). N. laxa is also distinguished from all other western Atlan- tic species by having virtually flat opercular scales with rudimentary ventral keels, the opercular scales of all other species being strongly concave above. Distribution.—Western Atlantic: known only from the type locality (see Material Examined) near Balanus Seamount, New England Seamount Chain, off Georges Bank; 3186 m. Eastern Atlantic: Shamrock Canyon, Bay of Biscay, 2980 m (Grasshoff, IDS): Remarks.—Verrill, in his original manu- script on the Alcyonaria of the Blake and in the captions of his unpublished plates (Verrill, unpub.), generically separated WN. laxa from the other species of WNarella, which have only 3 pairs of large abaxial body scales. The discovery of another spe- cies with 4 pairs of scales lends support to that view, but clarification of this matter must await the study of a great deal more material than is available at present. Narella spectabilis, new species Figs. 1C, 12A—C, 13A—D, 14 Material examined/types.—Holotype: Tongue of the Ocean, Great Bahama Bank: DABSOINET 1 22 WANA 85: mbCI-125, 25) Sep 1973, one unbranched stem partly decorti- cated and lacking distal tip and holdfast, USNM 57578 (SEM 407-409, 411, 416, C1029). Type Locality: as stated above. Description.—The holotypic colony is 25 cm long and unbranched, but, because it is a fragment, it cannot be certain if this branch represents an unbranched species, like N. versluysi, or is an end branch of a sparingly branched species, such as N. laxa. The axis is 0.9 mm in diameter, stiff, brittle, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 12. Narella spectabilis, holotype, USNM 57578: A, polyp in abaxial view; B, polyp in lateral view; C, polyp in adaxial view. Stereo views. All scale bars are 0.5 mm. VOLUME 116, NUMBER 3 645 PigeaiS: lateral, and adaxial opercular scales; B, a tentacular scale; C, three highly crested coenenchymal scales; D, three adaxial buccal scales. Scale bars: A, C = 0.5 mm; B, 0.05 mm; D = 0.25 mm. yellowish white with pale golden reflec- tions, and nearly smooth, with only the faintest suggestion of longitudinal groov- ing. Polyp whorls are not directly adjacent, but separated by about 0.5 mm from one another. Polyps are arranged in whorls of 4, of which 8 occur in 3 cm of branch length. Fully developed polyps are 3.5 mm in length. The polyps are covered by 4 pairs of large abaxial body scales (1 pair of basals, 2 pairs of medials, and 1 pair of buccals) and 2 pairs of smaller adaxial buccals. The sclerites of the basal pair project perpendic- ular to the axis, and are about 1.15 mm in height, the distal 0.20 mm extending slight- ly beyond the junction with the proximal ends of the first medial pair. Each basal sclerite bears a prominent longitudinal crest up to 0.25 mm in height corresponding to the right angle curve it makes as it wraps around the lateral side of the polyp. The A-—D, holotype of Narella spectabilis, USNM 57578: A, upperside of abaxial, outer-lateral, inner- basal sclerites decrease in width toward the adaxial side where they meet to form a complete ring but do not fuse. The oper- culum is prominent, about % the length of the contracted polyp. The 2 pairs of medials and the abaxial buccal pair are about the same length (about 0.75 mm) but progres- sive wider toward the polyp tip, the buccals almost encircling the polyp. The second pair of medial sclerites is largely over- lapped by the first (more proximal pair). Like the basals, these body wall sclerites are also longitudinally ridged along their midline, corresponding to the right angle curve around the lateral side of the polyp, but the ridges are much less prominent, and in the case of the buccals one or more short- er accessory ridges may be present. The buccal scales cover only the basal-most re- gion of the large opercular scales. The lon- gitudinal ridges and crests are aligned among the 4 pairs of body scales, all oc- Gulf of Mexico PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Caribbean Sea Fig. 14. Distribution of Narella bellissima (circles), N. versluysi (squares), and N. spectabilis (triangle). curring at the right angle curve leading to the lateral side of the polyp wall, giving the polyp a polygonal shape in cross section. One might expect a corresponding linear depression on the underside of the sclerite which would align and allow the ridge of the more proximal sclerite to slid over it, but such depressions were not observed. All body scales are externally ornamented with fine, smoothly rounded granules arranged in lines radiating outward from a nucleus near the center of each sclerite. Two pairs of well-developed square to rectangular adax- ial buccal scales are present, the marginal ones being quite large (up to 0.4 mm in width) the second pair about half this size, the distal edges of which are finely serrate. The adaxial opercular scales are the smallest of the operculars, shaped as slen- der triangles but deeply concave exteriorly (essentially spoon-shaped), up to 0.85 mm in length and 0.35 mm in basal width (H:W = 2.4—3.2). Inner- and outer-lateral oper- culars are slightly longer (up to 1.3 mm in height and 0.50 mm in greatest width, H:W about 2.35) and are fairly symmetrical in shape, also highly concave but not spoon- shaped. Abaxial operculars are the largest opercular scales, up to 1.5 mm in length and unusually narrow (e.g., 0.6 mm, result- ing in a high H:W of 2.1—2.6). The abaxial operculars sometimes have two additional apices in addition to the main apex (Fig. 13A). Thus, all opercular scales are roughly the same shape, but of increasing size in the ad- to abaxial direction. They all bear prominent longitudinal keels on their inte- rior surfaces. The tentacles contain rotund granular rodlets up to 0.15 mm in length and 45 wm in diameter. The coenenchymal scales are thin, elongate, more or less im- bricating, usually with a thin longitudinal VOLUME 116, NUMBER 3 crest, the crests of those coenenchymal scales adjacent to polyps being extremely tall (e.g., up to 0.45 mm). Etymology.—Latin spectabilis = remark- able, notable. Comparisons.—Only one other species of Narella, N. laxa, has four pairs of large body wall scales. N. spectabilis differs from that species in having longitudinal ridges on all body wall scales, a closed ring of basal scales, more slender and deeply concave opercular scales with a high H:W ratio, and would appear to be unbranched. Although N. spectabilis appears to be re- lated also to N. studeri (Versluys), N. parva (Versluys), and N. bowersi (Nutting), those species have only three pairs of body scales, with crests on only the basal pair. Distribution.—Known only from the type locality of The Tongue of the Ocean, Bahamas (Fig. 14), 1485 m. Acknowledgments We wish to thank Ardis Johnson for the loan of Narella specimens deposited at the MCZ, and Dale Calder for the collection and loan of the type specimen of WN. alvinae. We are also grateful to Manfred Grasshoff for valuable advice about the type deposi- tion of various species, and we thank Linda Cole for her technical support. Literature Cited Bayer, EF M., 1956. Octocorallia. Pp. F166—189, 192— 231 in R. C. Moore, ed., Treatise on invertebrate paleontology, University of Kansas Press, Lawrence. . 2001. New species of Calyptrophora (Coelen- terata: Octocorallia: Primnoidae) from the west- ern part of the Atlantic Ocean.—Proceedings of the Biological Society of Washington 114(2): 367-380. , M. Grasshoff, & J. Verseveldt (eds.). 1983. Illustrated trilingual glossary of morphological and anatomical terms applied to Octocorallia. E. J. Brill, Leiden, 75 pp. Calder, D. R. 1993. Exploring Bermuda’s underwater mountainside.—Rotunda, the Magazine of the Royal Ontario Museum 26(2):18—27. Carpine, C., & M. Grasshoff. 1985. Gorgonaires, cat- alogue, Musée oéanographique de Monaco— 647 Pannatulaires, catalogue, Musée océanogra- phique de Monaco.—Bulletin de 1’Institut océanographique, Monaco 73(1435):71 pp. Deichmann, E. 1936. The Alcyonaria of the western part of the Atlantic Ocean.—Memoris of the Museum of Comparative Zoology at Harvard College s373:7/ipp-5/a pls: Duchassaing, P. 1870. Revue des Zoophytes et des Spongiaires des Antilles. 52 pp., 2 pls., Victor Masson et Fils, Paris. , & J. Michelotti. 1860. Mémoire sur les coralli- aires des Antilles——Mémoires des |’ Accadémie des Sciences de Turin (2)19:279—365 [reprint paginated 1—88], 10 pls. Grasshoff, M. (1981)1982a. Die Gorgonaria, Penna- tularia und Antipatharia des Tiefwassers der Biskaya (Cnidaria, Anthozoa). I. Allgemeiner Teil. Bulletin du Muséum national d’histoire Naturelle, Paris, Section A (4)3(3):73 1-766. . (1981)1982b. Die Gorgonaria, Pennatularia und Antipatharia des Tiefwassers der Biskaya (Cnidaria, Anthozoa). II. Taxonomischer Teil.— Bulletin du Muséum National d’histoire Natu- relle, Paris, Section A (4)3(4):941—978. . 1985. Die Gorgonaria, Pennatularia und An- tipatharia des Tiefwassers der Biskaya (Cnidar- ia, Anthozoa). III. Erginzungen. Jn L. Laubier and C. Monniot, eds., Peuplements profonds du golfe de Gascogne. Brest, IFRREMER, 17:299— 310. . 1986. Die Gorgonaria der Expeditionen von “Travailleur’’? 1880-1882 und ‘‘Talisman”’ 1883 (Cnidaria, Anthozoa).—Bulletin du Mu- séum national d’ Histoire Naturelle, Paris, Sec- tion A (4)8(1):9—-38. . & H. Zibrowius. 1983. Kalkkrusten auf Ach- sen von Hornkorallen, rezent und fossil.— Senckenbergiana maritima 15(4/6):111—145. Gray, J. E. 1858. Synopsis of the families and genera of axiferous Zoophytes or barked corals.—Pro- ceedings of the Zoological Society of London 1857:278-294. . 1870. Catalogue of the lithophytes or stony corals in the collection of the British Museum. British Museum, London, 51 pp. Kolliker, R. A. von. 1865. Icones histiologicae, oder Atlas der vergleichenden Gewebelehre. Zweite Abtheilung. Der feinere Bau der hoheren Thi- ere. Erstes Heft. Die Bindesubstanz der Ceolen- teraten. Pp. 87-181, pls. 10-19, Verlag von Wilhelm Engelmann, Leipzig. Kiikenthal, W. 1915. System und Stammesgeschichte der Primnoidae.—Zoologischen Anzeiger 46(5):142—158. . 1919. Gorgonaria.—Wissenschaftliche Ergeb- nisse der deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’, 1898-1899 13(2):946 pp., pls. 30-89. 648 . 1924. Coelenterata: Gorgonaria. Das Tierreich 47. Walter de Gruyter & Co., Berlin, 478 pp. Pourtalés, L. EK 1868. Contributions to the fauna of the Gulf Stream at great depths (2nd series).—Bul- letin of the Museum of Comparative Zoology 1:121-142. Roule, L. 1896. Résultats scientifiques de la campagne du ‘“‘Caudan”’ dans le Golfe de Gascogne— Aout—Septembre 1895. Coelenterés.—Annales de l’Université de Lyon 26:299-323. Stephens, J. 1909. Alcyonarian and madreporarian cor- als of the Irish coasts, with a description of a new species of Stachyodes by Professor S. J. Hickson, EF R. S.—Department of Agriculture and Technical Instruction for Ireland; Fisheries Branch. Scientific Investigations 1907, 5:1—28, IL jolts Stiasny, G. 1941. Studien tiber Alcyonaria und Gor- gonaria. V. (Parerga und Paralipomena).—Zoo- logischen Anzeiger 135(3/4):75—88. Studer, T. 1887. Versuch eines Systemes der Alcyon- aria.—Archiv fiir Naturgeschichte 53(1):74 pp., 1 pl. . 1901. Alcyonaires provenant des campagnes de l’Hirondelle (1886—1888).—Résultats des PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Campagnes Scientifiques accomplies sur son yacht par Albert Ier, Monaco 20:64 pp., 11 pls. Thomson, J. A. 1927. Alcyonaires provenant des cam- pagnes scientifiques du Prince Albert Ier de Monaco.—Résultats des Campagnes Scienti- fiques accomplies sur son yacht par Albert Jer, Monaco 73:77 pp., 6 pls. Tixier-Durivault, A., & E Lafargue. 1968. Quelques Octocoralliaires des cétes Fran¢aises.—Bulletin du Muséum National d’Histoire Naturelle (2)40(3):62 1-629. , & M.-J. d Hondt. 1974. Les Octocoralliaires des la campagne Biacgores.—Bulletin du Musé- um National d’Histoire Naturelle, Zoologie (3)174(252):1361—1433. Verrill, A. E. unpublished. The Alcyonaria of the “Blake” Expeditions. A. E. Verrill’s unpub- lished plates.—140 pls. Versluys, J. 1906. Die Gorgoniden der Siboga-Expe- dition. IJ. Die Primnoidae.—Siboga-Expeditie 13a:187 pp., 10 pls., 1 map. Wright, E. P., & T. Studer. 1889. Report on the AIl- cyonaria collected by H.M.S. Challenger during the years 1873—76.—Report on the Scientific Results of the Voyage of H.M.S. Challenger during the years 1873-76, Zoology 31(64):314 pp.. 43 pls. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):649—660. 2003. A systematic review of the land snail Euglandina singleyana (Binney, 1892) (Mollusca: Gastropoda: Spiraxidae) Kathryn E. Perez and Ned E. Strenth (KEP) Department of Biological Sciences, Box 870345, University of Alabama, Tuscaloosa, Alabama 35487; (NES) Department of Biology, Angelo State University, San Angelo, Texas 76909 Abstract.—A systematic review of Euglandina singleyana was undertaken to thoroughly examine shell morphology, allozyme, and mtDNA variation in specimens of FE. singleyana endemic to central Texas. Allozyme similarity ranged from 95% in the most geographically proximal individuals of E. sin- gleyana (Val Verde and Real counties) to 82% between the most distant indi- viduals (Comal and Val Verde counties). DNA sequence similarity, based on a 397 bp partial 16S mtDNA sequence, ranged from 98% in eastern specimens (Comal and Kerr counties) to 95% in individuals from Kerr, Real, and Val Verde counties. Analysis of variation in shell morphology, allozyme similarity, and mtDNA sequences supports the existence of a single, highly variable, wide- spread species of Euglandina in central Texas. This study also examines the validity of E. immemorata, using morphometric and DNA sequence data and E. exesa, using morphometric data. The morphometric analysis showed that E. immemorata and E. singleyana differ significantly in shape. The current status of E. immemorata and E. exesa are also examined. Euglandina singleyana (Binney, 1892) is found in a wide variety of habitats along the southern margin of the Edwards Plateau in Texas, from Terrell County in the west to Fayette County in the east, and south to Refugio County (Fig. 1; Singley 1893, Pils- bry 1946, Fullington & Pratt 1974, Hu- bricht 1985). In the eastern part of its range it is found under rocks and logs in wooded stream valleys in the limestone of the Ed- wards Plateau. In the clay and sandy areas of the Balcones Escarpment, it is restricted to wooded lowlands. The range of this spe- cies extends to the western Stockton Pla- teau where it is found under fallen Yucca and in rock crevices in desert shrub habitat dominated by Lechuguilla cactus (Fulling- ton & Pratt 1974). A great deal of the historical taxonomic confusion regarding this species appears to be related to its relatively wide geographi- cal range of ~250 km, compared to the me- dian range of land snail distribution of 50 km reported by Solem (1984) and high lev- el of variation in shell morphology. Shells collected in the eastern part of the range can readily be distinguished from those collect- ed in the western part of the range. One purpose of this study was to examine vari- ation in shell morphology of specimens from throughout the range of the species in central Texas to determine if there are dis- tinct differences in western versus eastern shell morphology or if there is continuous (clinal) change in shell shape and size across the range of the species. The purpose of this study was also to examine allozyme and DNA sequence variation in individuals from across the range of the species (Co- mal, Kerr, Real, and Val Verde counties) to determine if specimens conforming to the description of E. singleyana formed a monophyletic group. This study addresses the taxonomy of this species from a phy- 650 Jaiiys Ie PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The amended distribution of Euglandina singleyana in Texas. Counties in the historical literature are shown in gray. Counties added in this study (Blanco, Chambers, Coryell, Kimble, and Uvalde) are shown in black. Collection localities for allozyme and DNA studies are represented by a black square. Museum localities listed in the material examined are represented by a black circle. The distribution of this species is not continuous throughout the highlighted area, but instead occurs in isolated patches of appropriate habitat. logenetic species concept approach using both monophyly (sequence analysis) and diagnosability (shell morphometrics) crite- ria (Minton & Lydeard 2003). The carnivorous land snail genus Eu- glandina has a problematic taxonomic his- tory. Von Martens (1901:47) noted that “many species have been described only from one or a few examples, and not fig- ured. In this genus ... it seems to be very difficult, or rather impossible, to draw a clear line of distinction between local var- lations and nearly allied species.’ In addi- tion to these problems, Thompson (1987) also noted that many original descriptions are scattered among nineteenth century journals in several languages and often lack critical details of sculpture of the adult and embryonic shells. Euglandina singleyana from central Tex- as appears typical of this group in having a history of confusion regarding its identity. Initially, Roemer (1849) identified speci- mens from New Braunfels as Glandina truncata (Gmelin, 1788). Binney & Bland (1869) later considered Texas specimens to be Glandina corneola Pfeiffer, 1857. Bin- ney (1892) described and named this Texas snail Glandina singleyana. Pilsbry later (1907) noted that in the Binney (1885) monograph, shells from Texas were listed with the name G. decussata (Deshayes, VOLUME 116, NUMBER 3 Fig. 2. 2 8'e 3 Landmarks used in the geometric morphometric analysis. Numbers | and 12 were used as baselines é in all analysis. Number | is the most distal point on the columella to Number 12 at the apex of the shell. 1850), the description and figure were of G. corneola, and the locality and anatomical descriptions were those of G. singleyana. Singley (1893) listed this species as G. de- cussata var. singleyana. Pilsbry (1907; p. 175) removed all Mexican and mainland members of Glandina to the genus Euglan- dina due to taxonomic confusion regarding the name and types of Glandina and sub- sequent workers (Pilsbry & Ferriss 1906, Pilsbry 1907, 1946, Fullington & Pratt 1974, Hubricht 1985) have retained this ge- neric name. Euglandina immemorata Pilsbry, 1907 is another species associated with some de- gree of taxonomic uncertainty. This species was described by Pilsbry (1907) on the ba- sis of two shells from ‘“Texas”’ (exact lo- cality unknown). Fullington & Pratt (1974) stated that the holotype of E. immemorata is not especially distinct from E. singleyana but could not be precisely matched by any material they had seen. They concluded that E. immemorata is probably E. singleyana or, if distinct (based on the improved knowledge of the Texas fauna since 1946) is probably not a Texas species. Euglandina exesa Cockerell, 1930 was described from a single shell found in a limestone deposit in a cinnabar mine at Ter- lingua, Brewster County, Texas. Fullington & Pratt (1974) state that the holotype does not differ from many western specimens of E. singleyana. They consider E. exesa to be simply a western range extension of E. sin- gleyana. A systematic review of E. singleyana was undertaken to thoroughly examine the observed morphological differences and add further evidence using genetic tech- niques. This study also tests the conclusions of Fullington & Pratt (1974) by examining the validity of E. immemorata using mor- phometric and DNA sequence data and E. exesa using morphometric data. Materials and Methods Morphometric analysis.—Programs used for morphometric analysis are part of the Integrated Morphometrics Package (www. canisius.edu/~sheets/morphsoft.html) made available by David Sheets, Miriam Zeld- itch, and Donald Swiderski. Specimens of Euglandina singleyana (64), E. immemor- ata (six—including both type specimens), and E. exesa (holotype) were examined (Appendix 1). Twelve landmarks on each shell were digitized from photographs using the prosramptpsDis version 131) (Ey a Rohlf. tpsDig32: Digitize coordinates of landmarks and capture outlines. http://life. bio.sunysb.edu/morph/index.html) (Fig. 2). The landmarks were chosen to be repeat- 652 able and homologous (Swiderski 1993) across all shells, and most landmarks rep- resent shell sutures (Stone 1998). For each specimen, digitized landmark coordinates were transformed to procrustes distances using partial procrustes superimposition methods. This was carried out in Coord- Gen6 (H. D. Sheets, Dept. of Physics, Can- isius College, 2001 Main St. Buffalo, New York 14208, sheets @gort.canisius.edu). A principal component analysis was per- formed using partial procrustes distances (PCAGen6e; H. D. Sheets). This program computes partial warp scores for each spec- imen, using a procrustes mean specimen based on all data for comparison. The prin- cipal components (eigenvectors of the co- variance matrix) are then calculated based on the covariance matrix derived from the partial warp scores. Landmarks 1 and 12 were designated end points and all speci- mens were standardized according to this baseline. This transformation to partial warp scores accomplishes standardization so that the principal component analysis ex- amines variation in shape, excluding vari- ation due to scale, rotation, and translation (Swiderski 1993, Stone 1998). A canonical variates analysis was per- formed using the program CVAGen6 (H. D. Sheets) to determine the set of axes that al- lows for the greatest possible ability to dis- criminate between two or more groups. This program computes partial warp scores with reference to a common mean specimen then performs a multivariate analysis of variance followed by a canonical variates analysis. It determines how many distinct axes there are in the data, (p = 0.05) and computes the canonical variates scores of all the specimens. It also uses Mahalanobis distances to assign specimens to one of the groups. The single specimen of E. exesa was coded as an unknown and assigned to a group based on the canonical variates axes formed in the prior analysis (CVA- Gen6; H. D. Sheets). Lastly, a comparison was performed to determine if there was a significant difference in shape between E. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON singleyana and E. immemorata (Two- Group6c; H. D. Sheets). Allozyme analysis.—Each collection lo- cality of Euglandina singleyana was rep- resented by one or two specimens. A total of seven specimens from five collection lo- calities at ~50 km intervals throughout the geographic range of this species were ex- amined (Appendix 1). After collection, individuals were held without feeding for 7-10 days, then frozen in cryotubes in liquid nitrogen and stored in an ultracold freezer (—80°C) until anal- ysis. Samples were homogenized in two volumes of distilled water using a glass rod and centrifuged to obtain an aqueous ex- tract. Procedures for cellulose acetate elec- trophoresis and staining followed those of Hebert & Beaton (1993). Gels were pur- chased from Helena Laboratories Inc. (Beaumont, Texas), and the buffer used was tris-glycine pH 8.5. To examine variation within E. singleyana, scorable data for 19 loci (Table 1) were obtained and analyzed using Tools for Population Genetic Analy- sis 1.3 (M. P. Miller. Tools for Population Genetic Analysis (TFPGA 1.3): A windows program for the analysis of allozyme and molecular population genetic data. Com- puter software distributed by the author). To determine genetic similarity, Nei’s un- biased genetic identity was calculated (Nei 1978). An unweighted pair group method using arithmetic averages (UPGMA) cluster analysis was then performed using the ge- netic identity matrix. DNA sequence analysis.—Twelve tissue samples were either preserved in 70% eth- anol or frozen in liquid nitrogen and then stored in an ultracold freezer at —80°C. The outgroup taxa chosen were the closest rel- atives of Euglandina with sequences avail- able on GenBank, relationships from Wade et al. (2001). Total genomic DNA was ex- tracted from several milligrams of foot tis- sue by digestion with lysis buffer and Pro- teinase K and then purified by phenol: chlo- roform extraction according to standard procedures (see Palumbi, S., A. Martin, S. VOLUME 116, NUMBER 3 Romano, W. O. McMillan, L. Stice, & G. Grabowski. 1991. The Simple Fool’s Guide to PCR. Privately distributed, Honolulu, Hawai, 40 pp.). Mitochondrial DNA sequences were ob- tained for an amplified segment of the 16S rDNA gene using 16sar and 16sbr primers (see Palumbi above). Approximately 10 ng of genomic DNA provided templates for double-stranded reactions via the polymer- ase chain reaction (PCR). PCR reactions were done in a 50 pL solution containing each dNTP at 0.22 uM, each primer at 0.1 uM, 1.5 mM MgCl, | unit Taq DNA poly- merase, and 1X PCR reaction buffer. Re- actions were amplified for 30 cycles of 92°C for 45 sec, 50°C for 45 sec, and 68°C for 2 min. Samples were purified and dou- ble-stranded DNA provided the template for cycle-sequencing using BigDye (ABI) chemistry followed by analysis on an ABI3100 automated sequencer. Contigs were assembled in Sequencher™ 4.0.5 (Gene Codes Corporation, Ann Arbor, Michigan and aligned by eye using BioEdit (Hall 1999) with reference to secondary structure models to refine the alignment and identify regions corresponding to loops and stems (Lydeard et al. 2000). Sequences were deposited with Genbank (Accession Num- bers: AF405235—-AF405241, AY 149279, AY 167887—AY 167889). Aligned sequences were analyzed using maximum parsimony with PAUP*4.0b10 (Swofford 2002) using a heuristic search (10 addition replicates). The following options were used: uninformative characters were ignored, only minimal trees were kept, gaps were treated as missing, and zero length branches were collapsed. A bootstrap analysis with 1000 iterations was conducted (Felsenstein 1985). Bremer sup- port values (Bremer 1994) were calculated using the Decay function of MacClade 4.03 (Maddison & Maddison 2000). Results and Discussion Morphometric analysis.—In the exami- nation of shell variation within Euglandina 653 Table 1.—Presumptive enzymatic loci resolved. Tris-glycine (pH 8.5) was used as the buffer system for all of the enzymes listed. All stain and buffer rec- ipes are from Hebert & Beaton (1993). Enzyme commission Enzyme system and abbreviation number Adenylate Kinase (ADK) 2.7.4.3 Alcohol Dehydrogenase (ADH) [ESS FS Alcohol Oxidase (AQX) elie bs, Aspartate Aminotransferase (AAT) DEG Glucokinase (GK) DLT Mo Glucose-6-phosphate Dehydrogenase (G6PDH) 1.1.1.49 Hexokinase (HK) Dela L-Iditol Dehydrogenase (IDDH) 1.1.1.14 Isocitrate Dehydrogenase (IDH) 1.1.1.42 L-lactate Dehydrogenase (LDH) Hoe 27 Malate Dehydrogenase (MDH-1 & 2) edhe We 357/ Malate Dehydrogenase (NADP+) (MDHP-1 & 2) 1.1.1.40 Nucleoside Phosphorylase (NSP) 23-2 Phosphoglucomutase (PGM) 5.4.2.2 Phosphogluconate Dehydrogenase (PGDH) 1.1.1.44 Superoxide Dismutase (SOD) HS IeI Triosephosphate Isomerase (TPI) Deobllcl! singleyana the first principal component (PC1) accounted for 67.42% of the varia- tion present in the measurements and PC2 accounted for 9.66% of the variation pres- ent. A scatterplot, with specimens grouped by county, comparing the first two principal component axes, does not show any distinct groups within E. singleyana (Fig. 3). Shell variation described by principal component 1 appears to be continuous and the results of this analysis do not allow E. singleyana to be separated into groups based on shell characteristics. The canonical variates analysis (Fig. 4) to determine if FE. immemorata could be distinguished from E. singleyana was sig- nificant, (df = 20, p = 3.70685 x 10>) with 100% correct reclassification of both species. A two-group comparison showed that the shapes of the two species are sig- nificantly different (Hotelling’s T? f = 6.52, df = 24, 45, p = 3.8316 X 10-8). This result indicates that E. immemorata is morpholog- ically distinct from E. singleyana. This re- 654 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lene, 3). 0.02 0.08 0.04 0.06 Scatterplot displaying principal component scores of Euglandina singleyana. X-axis is principal component one, Y-axis is principal component two. Texas counties are represented by the following symbols: Bexar (asterisk) type-locality, Chambers (circle), Comal (plus sign), Coryell & Travis (star), Hays (square), Kimble, Uvalde, & Real (hollow triangle), Val Verde (filled triangle), Victoria (X). sult, along with the discovery of specimens of E. immemorata in Nuevo Leon, Mexico (Correo-Sandoval 1993), supports the re- moval of E. immemorata from Texas faunal listings. The unknown assignment test to deter- mine the placement of FE. exesa placed the single specimen in a cluster formed by E. singleyana (Fig. 4); however, this result was not significant (p = 0.001). This species is represented by only one specimen. There- fore, this test is not replicable and has little Statistical power. It is interesting that the CVA placed the specimen of E. exesa with- in E. singleyana, supporting the statements of Fullington & Pratt (1974) about this spe- cies. However, without more evidence, nothing conclusive can be stated about the validity of this species. Allozyme analysis.—The results of this study reveal a moderate to high degree of genetic similarity in all specimens of Eu- glandina singleyana examined across its range. Seven of 19 loci (MDH-2, MDHP- 2, IDH, LDH, SOD, G6PDH, PGM) were monomorphic for all specimens. The Val Verde County specimens exhibited poly- morphisms at nine enzymatic loci (MDH-1, MDHP-1, PGD, AOX, AAT-1, ADK, IDDH, ADH, HK), the Real County speci- mens exhibited polymorphism at six (MDH-1, MDHP-1, AOX, HK, NSP, ADH). In the specimens examined, there were no fixed allelic differences among populations of E. singleyana across its geo- graphic range. Allozyme variation appeared to follow a general geographic pattern with specimens having the smallest genetic distance from their most proximal geographic neighbors (Fig. 5). Genetic divergence calculated us- ing Nei’s unbiased genetic identity resulted VOLUME 116, NUMBER 3 Euglandina singleyana ® 4 Euglandina immemoratas Euglandina exesa * 655 -10 8 6 -4 -2 0 2 4 Fig. 4. Graph of the canonical variates axes displaying the separation between the means of Euglandina singleyana and E. immemorata. Plotted as an unknown is E. exesa. Oversized symbols indicate the means. X- axis is CVA 1, Y-axis is-CVA 2. in the most western specimens from Val Verde County and the Real County speci- mens displaying a genetic similarity of 94.5%. The Val Verde+ Real cluster was ge- netically similar to the more centrally lo- cated Kerr County specimen at a level of 89.7%. This cluster is 85.2% similar to the most eastern specimen from Comal County. Perez & Strenth (2002) found that spec- imens of Euglandina texasiana (Pfeiffer, 1857) from collection localities 150 km dis- tant, in south Texas and northern Tamauli- pas, had a similarity of 94.5%. In compar- ison, this study observed 94.5% similarity between specimens of E. singleyana located 157 km apart, and 85.2% similarity be- tween the most geographically distant spec- imens (246 km) from Comal and Val Verde Counties. Euglandina singleyana was found to display enzyme polymorphism at 12 of 19 loci examined. This level of allo- zyme variability is more similar to the re- sults of studies on Helix aspersa (Selander & Kauffman 1975) and differs from results found in Liguus by Hillis et al. (1991) and Rumina decollata by Selander & Kaufman (1975), which were notably less variable. The levels of genetic distance among the widely separated populations of Euglandina singleyana are within the range found in other organisms for genetic divergence be- tween subspecies (Quicke 1993). However, there are no fixed differences among the populations and genetic distance is low be- tween geographically proximal populations. This analysis indicates that the specimens of E. singleyana examined represent a sin- gle species. DNA sequence analysis.—The region of 16S mtDNA that was sequenced resulted in 656 1.800 1.350 0.900 Fig. 5. produced using Nei’s unbiased genetic identity. an aligned data matrix of 397 base pairs of which 191 were phylogenetically informa- tive. Phylogenetic analysis of the data using maximum-parsimony analysis resulted in three equally parsimonious trees of 480 steps (CI = 0.8646, gl = —2.024336). These three trees differed only in the place- ment of the Comal County (eastern) speci- men relative to the Kerr County specimen. One topology places the Kerr County spec- imen (central) sister to the Real+Devil’s River clade (Central & Western). An alter- nate topology places the Comal County specimen in this position, and the final to- pology describes Comal and Kerr County specimens as each others’ closest relative. Bootstrap analysis (1000 pseudoreplicates) of the aligned data matrix using maximum- parsimony produced the tree shown in Fig. 6. Euglandina singleyana forms a mono- phyletic group. The sister taxon to E. sin- gleyana is the specimen of Euglandina from Northern Coahuila. Also outside this grouping is E. corneola from Tamaulipas, Mexico. Pairwise sequence identity was calculated for each clade. Within the Val Verde County cluster (Devil’s River and Comstock) there is 98—100% sequence sim- ilarity. The two Real County specimens had identical 16S sequences (=100% similari- ty). Sequence similarity between Kerr 0.450 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0.000 E. singleyana -Val Verde Co E. singleyana -Real Co E. singleyana -Kerr Co E. singleyana -Comal Co Rabdotus A dendrogram of allozyme similarity in Euglandina singleyana produced by UPGMA on a matrix County and Comal County was 97%. All possible combinations of pairwise compar- isons were performed with a minimum sim- larity of 95% among specimens from Kerr and Real County. The two specimens of E. immemorata formed a group apparently not closely related to E. singleyana. Both allozyme and sequence analyses show an interesting geographic pattern with populations most closely related to their geographically proximal neighbors. This pattern of strong geographic structuring is often seen in land snail species (Thomaz et al. 1996, Schilthuizen et al. 1999). Thomaz et al. (1996) examined geographic variation within Cepea nemoralis and Helix aspersa and found very high levels of sequence di- vergence (12%) within these species of land snails. The authors conclude that the most likely explanation for the observed levels of divergence is the population structure of land snails with low dispersal and large populations divided into infrequently inter- acting demes. Avise et al. (1987) presents several tests of this hypothesis, one of which is: phylo- genetic differentiation between long sepa- rated demes should be reflected in nuclear as well as mitochondrial assays. The con- gruence of the allozyme and mitochondrial data in the present study appear to fulfill VOLUME 116, NUMBER 3 90 68 100 82 02 95 Fig. 6. (Si) Phaeohelix Euglandina sp. Coahuila E. singleyana -Comal Co E. singleyana -Kerr Co E. singleyana -Devil's River 66 E. singleyana -Comstock E. singleyana -Devil's River E. singleyana -Real Co 95 3 E. singleyana -Real Co E. comeola SS) 4 E. rosea E. immemorata 100 14 E. immemorata Nesiohelix Euhadra Strict consensus of three most parsimonious trees of partial 16S mtDNA sequences of specimens of Euglandina singleyana from Comal, Kerr, Real, and Val Verde counties. Numbers above the branches are bootstrap support percentages (1000 pseudoreplicates). Numbers below the branches are Bremer’s support values. this requirement, lending credence to the idea that the specimens of E. singleyana ex- amined in this study represent a single spe- cies. The present study does not have enough samples to conclusively examine geographic partitioning within this species; however, there does appear to be some geo- graphic structure in the data. Although sample sizes were low, they appear to be adequate to address the ques- tions posed in this research. Euglandina singleyana is considered to be an uncom- mon species (Singley 1893; Neck 1984, 1988), and the area where they are found is rapidly being disturbed ecologically due to a growing human population. As a result of their very specialized feeding habits, rarity, habitat preferences, and human activities, 658 the live specimens examined during this study represent a significant collection of living specimens of Euglandina singleyana. Conclusions All of the analyses, including morpho- metric, allozyme, and mtDNA sequences, support the premise that Euglandina sin- gleyana is a single, widespread, highly var- iable species. Both allozyme and DNA se- quence results indicate that there are some detectable genetic geographic patterns with- in E. singleyana, as well as the observable morphological gradient across the range of this species. The geographically most dis- tant specimens are the most genetically dis- similar and geographically proximal speci- mens are more similar. The results of the morphometric and se- quence analyses indicate that E. immemor- ata is distinct from E. singleyana. This re- sult, combined with the recent discovery of specimens of FE. immemorata from Nuevo Leon, Mexico leads to the conclusion that E. immemorata should be removed from Texas faunal listings. The taxonomic place- ment of Euglandina exesa could not be de- finitively determined. Acknowledgments We thank the following for assistance with collections and research: Dr. Robert Dowler, Dr. J. Kelly McCoy, Joel Brant, Jim Campbell, Dr. Brad Henry, Jerry F Hu- sak, Shailaja Marion, Mark Kitson, Jochen Gerber, Gene Hall, Jeanne Serb, Dr. Russ Minton, Dr. Charles Lydeard, Lynn Mc- Cutchen, and Bob Howells. Dr. Alfonso Correo-Sandoval was invaluable in obtain- ing specimens and permits in Mexico, Per- mit #NOM-126-ECOL-2001. We also thank Texas Parks and Wildlife Department for permits and access to Devils River State Natural Area. An NSF equipment grant at the University of Alabama (DBI-0070351) contributed to part of this research. Support was provided by an NSF-IGERT fellowship in the Freshwater Sciences program (DGE- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 9972810) and University of Alabama, Ecol- ogy & Systematics Enhancement fellow- ship. Thanks are also due the Beta Beta Beta association for two research awards that contributed to part of this research. Literature Cited Avise, J. C. et al. 1987. Intraspecific phylogeography: The mitochondrial DNA bridge between popu- lation genetics and systematics.—Annual Re- view of Ecology and Systematics 18:489—522. Binney, W. G., 1885. A manual of american land shells. Government Printing Office, Washington D.C., 351 pp. , 1892. Fourth Supplement to Terrestrial Mol- luscs.—Bulletin of the Museum of Comparative Zoology 22:163. , & T. Bland. 1869. Land and fresh water shells of North America. Part 1. Pulmonata Geophi- la.—Smithsonian Miscellaneous Collections, 194:13-21. Bremer, K. 1994. Branch support and tree stability.— Cladistics 10:295—304. Cheatum, E. P., R. Fullington, & L. Pratt. 1972. Mol- luscan records from west Texas.—Sterkiana 46: 6—10. Chiba, S. 1999. Accelerated evolution of land snails Mandarina in the oceanic Bonin Islands: evi- dence from mitochondrial DNA sequences.— Evolution 53:460—471. Cockerell, T. D. A. 1930. An apparently extinct Eu- glandina from Texas.—Proceedings of the Col- orado Museum of Natural History 9:52—53. Correo-Sandoval, A. 1993. Caracoles terrestres (Mol- lusca: Gastropoda) de Santiago, Nuevo Leon, México.—Revista de Biologia Tropical 41:683— 687. Deshayes, G. P. 1850. in Ferussac. 1820—1852. His- toire naturelle générale et particuliere des Moll- usques terrestres et fluviatiles. II, p. 182. Felsenstein, J. E 1985. Confidence limits on phyloge- nies: an approach using the bootstrap.—Evolu- tion 39:783-791. Fullington, R. W., & W. L. Pratt. 1974. The aquatic and land mollusca of Texas.—Bulletin of the Dallas Museum of Natural History 1(3). 48 pp. Gmelin. 1788—1790.—Systema Naturae 13. p. 3434, number 49. Hall, T. A. 1999. Bioedit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT.—Nucleic Acids Sym- posium Series 41:95—98. Hebert, P, & M. J. Beaton. 1993. Methodologies for allozyme analysis using cellulose acetate elec- trophoresis. Helena Laboratories. Beaumont, Texas, 32 pp. VOLUME 116, NUMBER 3 Hillis, D. M., M. T. Dixon, & A. L. Jones. 1991. Min- imal genetic variation in a morphologically di- verse species (Florida tree snail, Liguus fascia- tus).—Journal of Heredity 82:282—286. Hubricht, L. 1985. The distribution of the native land molluscs of the eastern United States.—Fieldi- ana Zoology 24, 191 pp. Lydeard, C., W. E. Holznagel, M. N. Schnare, & R. R. Gutell. 2000. Phylogenetic analysis of mollus- can mitochondrial LSU rDNA sequences and secondary structure.—Molecular Phylogenetics and Evolution 15:83—102. Maddison, W. P., & D. R. Maddison. 2000. MacClade 4.03: analysis of phylogeny and character evo- lution. Sinauer Associates, Sunderland, Massa- chusetts, U.S.A. Martens, E. von. 1890-1901. Biologia Centrali-Amer- icana. Pp 46-79 in E D. Godman, ed., R. H. Porter, and Dulau and Co., Soho Square W, London. McGee, P. L. 1971. Distribution and ecology of the terrestrial mollusks of the Texas coastal coun- ties —Molluscana, pp. 25-29, 37-39, 49-52, 61-64, 73-76, 85-88, 13-16, 25-28, 39-42, 53-57. Minton, R. L., & C. Lydeard. 2003. Phylogeny, tax- onomy, genetics and global heritage ranks of an imperiled, freshwater snail genus Lithasia (Pleuroceridae).—Molecular Ecology (12) (Gn press). Neck, R. W. 1984. Restricted and declining nonmarine mollusks of Texas.—Texas Parks and Wildlife Department Technical Series No. 34, 17 pp. . 1988. Urban refugia which support dense con- centrations of Euglandina singleyana.—Texas Conchologist 24:78—82. Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals.—Genetics 890:583—590. Perez, K. E., & N. E. Strenth. 2002. Enzymatic vari- ation in the land snail Euglandina texasiana (Gastropoda: Pulmonata) from South Texas and Northeastern Mexico.—Texas Journal of Sci- ence 54:37—44. Pfeiffer, von L. 1857. Novitates conchologicae. 8. p. ox, ll You, Wie, WI, WA, Pilsbry, H. A. 1907. Oleacinidae, Ferussacidae, vol. XIX in G. W. Tryon’s Manual of Conchology. Academy of Natural Sciences of Philadelphia, S66) pp san O2 plates: . 1946. Land mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia. Monograph 3, vol. 2(1):188—-199. , & J. H. Ferriss. 1906. Mollusca of the south- western states, I]—Proceedings of the Acade- my of Natural Sciences of Philadelphia 58:123— IWS Quicke, D. L. J. 1993. Principles and techniques of 659 contemporary taxonomy. Blackie Academic and Professional, Glasgow, 311 pp. Roemer, EF 1849. Mit einem naturwissenschaftlichen Anhange in A. Marcus, ed., Molluscs, 456 pp. Schilthuizen, M., J. J. Vermuelen, G. W. H. Davidson, & E. Gittenberger. 1999. Population structure in a snail species from isolated Malaysian lime- stone hills, inferred from ribosomal DNA se- quences.—Malacologia 41:283—296. Selander, R. K., & D. W. Kaufman. 1975. Genetic structure of populations of the brown snail (He- lix aspersa). I. Microgeographic variation.— Evolution 29:385—401. Singley, J. A. 1893. Texas molluscs. Geological Sur- vey of Texas. Fourth Annual Report, part 2. 375 Pp. Solem, A. 1984. A world model of land snail diversity and abundance. Pp. 6—22 in A. Solem & A. C. van Bruggen, eds., World-wide snails: biogeo- graphical studies of non-marine mollusca. Brill, Leiden, 289 pp. Stone, J. R. 1998. Landmark-based thin-plate spline relative warp analysis of gastropod shells.— Systematic Biology 47:254—263. Strecker, J. K. 1935. Land and fresh-water snails of Texas.—Transactions of the Texas Academy of Science 17:4—50. Swiderski, D. L. 1993. Morphological evolution of the scapula in tree squirrels, chipmunks, and grand squirrels (Sciuridae): an analysis using thin- plate splines.—Evolution 47:1854—1873. Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Ver- sion 4.0b10. Sinauer Associates, Sunderland, Massachusetts. Thomaz, D., A. Guiller, & B. Clarke. 1996. Extreme divergence of mitochondrial DNA within spe- cies of pulmonate land snails.—Proceedings of the Royal Society of London, Serial B 263:363— 368. Thompson, FE G. 1987. Giant carnivorous land snails from Mexico and Central America.—Bulletin of the Florida State Museum: Biological Sciences 2D, SZ 50\0- Wade, C. M., P. B. Mordan, & B. Clarke. 2001. A phylogeny of the land snails (Gastropoda: Pul- monata).—Proceedings of the Royal Society of London, Serial B. 268:413—422. Appendix | Material Examined and Distribution Morphometric Analysis Euglandina singleyana (64): The number of speci- mens of each lot that were used follows the museum catalog number in parenthesis (all undamaged shells with 5 or more whorls were used). Field Museum of Natural History: 29 specimens: FMNH 22347 (1), 660 30850 (2), 36309 (5), 50011 (4), 58295 (3), 62416 (6), 74849 (1), 78560 (2), 98199 (2), 109097 (1), and 175899 (1). The Academy of Natural Sciences of Phil- adelphia: 22 specimens: ANSP 186729 (1), 186731 (1), 346372 (3), 84622 (2), 104753 (1), 4312 (1), 84637 (1), 158379 (1), 134180 (4), 76849 (3), 150798 (1), 76837 (1) topotype, 186733 (1), and 87425 (1). Angelo State Natural History Collection: 13 speci- mens, individually numbered: (ASNHC 0008-0014, 042, 044-048). Euglandina immemorata (6): FMNH 11777 (1), 4359 (1) locality unknown, types; Florida Museum of Natural History 189621 (1), Nuevo Leon, Mexico, Santiago, | km. N of Laguna de Sanchez. University of Alabama Gastropod Collection, 3 specimens, indi- vidually numbered: 632—634, Nuevo Leon, Mexico 25°23'00.9"N; 100°14'28.6"W, 1 km N of Laguna de Sanchez. Allozyme Analysis Euglandina singleyana (7): Specimens used in this analysis are deposited in the Strecker Museum of Nat- ural History, (SMNH) Baylor University (Accession Number 2001-A-1-1; Catalog Numbers SM32439- SM32446). (1) Landa Park, New Braunfels, Comal County, Texas; (1) Kerrville, Kerr County, Texas; (2) 9 miles N of Leakey, Real County, Texas; (2) Devil’s River State Natural Area (DRSNA), Val Verde County, Texas; (1) Comstock, Val Verde County, Texas. Rab- dotus alternatus from Val Verde County was selected as an outgroup. DNA Sequence Analysis The same specimens listed in the above section on allozyme analysis provided the tissue samples which PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON were used in the sequence analysis. Additionally in- cluded were a specimen of an unidentified Euglandina from La Cuesta in Northern Coahuila; Euglandina cor- neola from 4 miles SW of Mante, Tamaulipas, Mexico; 2 specimens of E. immemorata from Nuevo Leon, Mexico 25°23'00.9"N; 100°14'28.6’W, 1 km N of La- guna de Sanchez; and one specimen of E. rosea from Lake County, Florida 28°30'53”N; 81°44'15”"W. Euhadra amaliae (AF098712), Phaeohelix phaeo- gramma (AF098714), and Nesiohelix omphalina (AF098713) sequences from Chiba (1999) were used as outgroups. Distribution Strecker (1935) listed Euglandina singleyana as oc- curring in Bexar, Caldwell, Comal, Goliad, Hays, Travis, Victoria, Atascosa, Frio, Gonzales, Guadalupe, and Wilson counties. McGee (1971) added Bandera, Fayette, Kendall, Kerr, Kinney, Val Verde, Medina, and Real counties. Cheatum et al. (1972) added Terrell County to the distribution. Collections by the authors and examinations of museum specimens from the Academy of Natural Sciences in Philadelphia and the Field Museum of Natural History during this study add Uvalde, Kimble, Chambers, Coryell, and Blanco coun- ties to the known distribution (Fig. 1). County records.—FMNH 78560, Horse Creek, Cor- yell Co., Texas; FMNH 98199, Galveston, Chambers Co., Texas; FMNH 210, 103/1, River Drift, Pedernales Falls State Park, Blanco Co., Texas; ANSP 186731, Garner State Park, Uvalde Co., Texas; ANSP 186727, Roadside Park, Nueces River W of Uvalde, Uvalde Co., Texas; ANSP 186733, Llano River, Highway 29, 12 miles S of Junction, Kimble Co., Texas. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):661—691. 2003. Redescription of adults and description of copepodid development of Dermatomyzon nigripes (Brady & Robertson, 1876) and of Asterocheres lilljeborgi Boeck, 1859 (Copepoda: Siphonostomatoida: Asterocheridae) V. N. Ivanenko and Frank D. Ferrari (VNI) Department of Invertebrate Zoology, Biology Faculty, Moscow State University, Moscow 119 899, Russia; (FDF) Department of Systematic Biology, National Museum of Natural History, Smithsonian Institution, Museum Support Center, 4210 Silver Hill Rd., Suitland, Maryland 20746, U.S.A. Abstract.—Adult and immature copepodids of Dermatomyzon nigripes (Bra- dy & Robertson, 1876) and Asterocheres lilljeborgi Boeck, 1859 were collected by SCUBA from the White Sea. All copepodids of D. nigripes were found on the bryozoan Flustra foliacea (Linnaeus, 1758); adults of D. nigripes also were washed out from the sponge Halichondria panicea (Pallas, 1766), the ophiuroid Ophiopholis aculeata (Linnaeus, 1767), and were collected among hydrozoans and other invertebrates from dead shells swept by tidal currents. Copepodids of A. lilljeborgi were washed from the starfish Henricia sanguinolenta (O. E Miller, 1776). Comparative analysis of development of D. nigripes and A. lilljeborgi with the related Scottomyzon gibberum (Scott & Scott, 1894) shows that both genders of A. Jilljeborgi and S. gibberum suppress the formation of fourth abdominal somite. Females of A. Jilljeborgi and D. nigripes develop a simple segmental complex when the arthrodial membrane separating the genital somite from the second abdominal somite fails to form; this arthrodial mem- brane develops on females of S. gibberum so that there is no genital complex. The antennule of A. /illjeborgi with a single, proximal complex of three seg- ments appears most similar to the ancestral siphonostomatoid. The antennule of D. nigripes has a proximal complex of two segments and a distal complex of three segments; the antennule of S. gibberum has a proximal complex of three segments and a distal complex of four segments. Setation of the maxil- liped of D. nigripes and A. lilljeborgi are identical and appear similar to the ancestral siphonostomatoid; the maxilliped of S. gibberum differs in that it fails to add a seta to its syncoxa and loses a seta on the penultimate endopodal segment. Asterocheres lilljeborgi and S. gibberum share derived states of se- tation on the exopod of swimming legs 1—4, leg 5 and leg 6; D. nigripes and S. gibberum share derived states of setation on the endopod of swimming legs 3 and 4. Copepods belonging to the family Aster- ocheridae Giesbrecht 1899 have been col- lected as free-living from the benthopelagic zone, or from benthic samples taken in as- sociation with cnidarians, echinoderms, sponges, ascidians and bryozoans from all oceans of the world. Forty of 45 genera of asterocherids have been described from shallow marine waters, and five from deep waters. Four of the five deep water genera are found in galls of corals (Stock 1981, 1984); the monotypic genus Cheramomy- zon Humes, 1989 is found in deep water seeps off Florida. In addition, one species 662 of the shallow water genus Collocherides Canu, 1893 has been collected from a deep water hydrothermal vent in the northeast Pacific Ocean (Humes 1999). The family Asterocheridae and the deep water hydro- thermal vent family Dirivultidae Humes & Dojiri, 1980 share character states often considered ancestral for the order Siphon- ostomatoida in phylogenetic analyses of co- pepod orders (Ho 1990, Huys & Boxshall 1991, Martinez Arbizu 2003). Detailed de- scriptions of development of asterocherid species are essential in analyzing the ances- tral states, transformations, and homologies of characters of that order and of the sub- class Copepoda. Ivanenko et al. (2001) pro- vide the only description of development of a siphonostomatoid, Scottomyzon gibberum (Scott & Scott, 1894), associated with an invertebrate, the starfish Asterias rubens Linnaeus, 1758. Scottomyzon gibberum pre- viously was included in Asterocheridae, but was placed in Scottomyzontidae by Ivanen- ko et al. (2001). In the present paper, adults are redescribed and all juvenile copepodid stages are described of Dermatomyzon ni- gripes (Brady & Robertson, 1876) and As- terocheres lilljeborgi Boeck, 1859, both collected from the White Sea. Dermato- myzon nigripes is the only species of the genus. Asterocheres lilljeborgi 1s the oldest of 60 nominal species in its genus, and tax- onomic studies of the genus have been pub- lished continuously since its discovery. Methods All measurements and dissections were made for copepodids from the Karelian coast of the Gulf of Kandalaksha, White Sea, near the Marine Station of Moscow State University, 66°31’N, 33°07'W. Ben- thic invertebrates were collected individu- ally in plastic bags during SCUBA dives from 5—25 m. About five volumes of fresh water were added to one volume of the sea- water in the plastic bag with invertebrates for 30 min. The bag was agitated and the water passed through a 20 micron mesh net. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Copepodids conforming to stages I-VI were recovered from these net samples and fixed with 70% ethanol; no nauplii were collected. Copepodids were cleared in lactic acid following the method of Humes & Gooding (1964), and stained by adding a solution of chlorazol black E dissolved in 70% ethanol/30% freshwater (Ferrari 1995). Drawings were made with a camera lucida. The first to sixth copepodid stages are CI—CVI. Thoracic somites are abbreviated Th; abdominal somites are Abd. CR is the caudal ramus. Somites are numbered ac- cording to their relative developmental age following Hulsemann (1991); thoracic so- mites and abdominal somites, except the most posterior anal somite, increase in age and decrease in numeral designation ante- riorly. The anal somite bearing the caudal rami is designated as the first abdominal so- mite because developmentally it is the old- est abdominal somite. The first thoracic so- mite bears the maxilliped; the genital open- ings are found on the seventh thoracic so- mite. The number of segments of the antennule often are difficult to determine because the arthrodial membrane separating segments may be very thin and difficult to observe; armament of an antennular segment is giv- en as “‘setae + aesthetascs’’. Although pat- terning of copepod swimming legs during development are only incompletely known, the model of proximal patterning (Ferrari & Benforado 1998) is followed here. Ramal segments of swimming legs 1—4 (thoraco- pods 2—5) are referred to by their presen- tation during development (Ferrari 1995). The terms “‘seta’’ and “‘spine”’ are used for articulating cuticular elements connected by an arthrodial membrane to an appendage segment; setae appear to be less rigid than spines. In order to maintain continuity among descriptive publications, tables of setae and spines on swimming legs 1—4 in the descriptive section follow the formula introduced by Lang (1934). In the formula, Roman numerals indicate spines and Arabic VOLUME 116, NUMBER 3 numerals are setae. Numerals to the left of a comma or dash indicate lateral elements; numerals between two commas are terminal elements, and numerals to the right of a comma or dash are medial elements. A semicolon separates ramal segments and an asterisk indicates that the segment is absent. It should be noted, however, that this kind of formula is not derived from the way a swimming leg is patterned during devel- opment, in which the distal arthrodial mem- brane of a segment is formed one copepod- id stage later than the formation of the ini- tial seta of the segment (Ferrari & Benfor- ado 1998). Thus, setal and segmental homologies cannot be determined correctly from the formula. Setules are epicuticular extensions of a seta; denticles are epicutic- ular extensions of an appendage segment; spinules are epicuticular extensions of a so- mite. Only authors who have contributed descriptions and/or illustrations are cited in the synonymy section. Dermatomyzon nigripes (Brady & Robertson, 1876) Figs. 1-8 Dermatomyzon nigripes Brady and Robert- son, 1876.—Giesbrecht, 1899:77-—78, pl. 1, fig. 4, pl. 5, figs. 1-14.—Sars, 1914: 95-97, pls. 59, 60.—Lang, 1949:5, fig. 8.—Eiselt, 1965:155—158, fig. 3A—J.— Boxshall, 1990:537-—539, figs. 9—-11.— Gotto, 1993:166, figs. 40P—Q, 41A—D. Dermatomyzon giesbrechti Brady, 1910: 574-577, textfigure 60, pl. 58, fig. 11. Dermatomyzon nigripes giesbrechti Brady, 1910.—Eiselt, 1965:155. Dermatomyzon elegans Claus, 1889:351, Ole ©, Dermatomyzon herdmani Brady, 1910:575, fig. 61. Cyclopicera nigripes Brady & Robertson, 1876:197.—Brady, 1880:54—56, pl. 89, figs. 1-11.—Thompson, 1893:36, pl. 25, ne, W. Ascomyzon thorelli Sars, 1880:474-—75. CVI female (Fig. 1A, B).—Body length 663 range 1.20—1.36 mm (based on 3 speci- mens); average length of prosome 0.84 mm; maximum width of prosome 0.62 mm; length of urosome 0.46 mm; length of gen- ital complex 0.15 mm; width of genital complex 0.17; ratio of length to width of prosome 1.4; ratio of length of prosome to length of urosome 1.8. Prosome (Fig. 1A, B): 4 articulating sec- tions; Ist a complex of 5 cephalic somites plus Th1l, 2; Th3—5 articulating. Urosome (Fig. 1C, D): 5 articulating sec- tions; Th6, genital complex of Th7 fused to Abd2 [an unsclerotized area separates an- terior neck from remaining part of Th7], Abd3, 4, | articulating. On genital complex, paired copulatory pores ventral-lateral to oviducal openings; paired oviducal open- ings dorsal. Egg sacs spherical with up to 8 eggs (Fig. 1P). Rostrum (Fig. view. Oral cone (Fig. 1E): Beak-like. Antennule (Fig. 1G, H): 19 articulating seumenus witin ll, 2%, 4, 25 2% 42 2% O5 lls 2, ~% DO) ONO Ra eee relies A Antenna (Fig. II): Coxa and basis with- out setae; basis with denticle. Exopod 1- segmented with | terminal seta and | prox- imal seta. Endopod 2-segmented; Ist seg- ment with denticles; 2nd segment with den- ticles and 5 setae, largest thick and with curved tip. Mandible (Fig. 1J): Gnathobase 2X length of palp, tapering and slightly curved distally with denticles; l-segmented palp with denticles and 2 terminal setae. Maxillule (Fig. 1K): Inner lobe bearing several series of denticles and armed with 5 setae; outer lobe articulating proximally, with | short and 2 long setae terminally and 1 long subterminal seta. Maxilla (Fig. 2C): Apparently 2-seg- mented subchela; Ist segment with proxi- mal asthetasc; 2nd segment curved and pointed distally with indistinct arthrodial membranes at distal 3rd and distal 6th of 1B): Pointed in lateral 664 Bien Le PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Dermatomyzon nigripes (Brady & Robertson, 1876), CVI Female: A, habitus, dorsal; B, habitus, lateral; C, urosome, ventral; D, urosome, dorsal; E, labrum, anterior; EK eggs; G, antennule, segments 1—11; H, antennule, segments 12—19; I, antenna; J, mandible; K, maxillule. Scale line 1 is 0.1 mm for A, B; line 2 is 0.1 mm for C, D; line 3 is 0.1 mm for E, G—K; line 4 is 0.1 mm for F its length, with small, inner seta and several sets of denticles. Maxilliped (Fig. 2A, B): Short, syncoxa with | inner seta, long basis with 1 inner seta on medial margin; endopod of proxi- mal, short, apparently subdivided section with | medially and 2 distally polarized se- tae, and distal segment with 2 terminal se- tae, largest thick and claw-like. Swimming legs 1—4 (Fig. 2D—G): Bira- mal, with 3-segmented rami; all with inter- coxal sclerite and 2-segmented protopods. Formula for spines and setae (Table 1). Dis- tal exopodal seta adjacent to terminal spine of swimming legs 3—4 very small. Leg 5 (Fig. 21): Basis not articulating with somite and bearing medial denticle and lateral seta; exopod elongate with | thin VOLUME 116, NUMBER 3 Fig. 2. 665 Dermatomyzon nigripes (Brady & Robertson, 1876), CVI Female: A, maxilliped; B, proximal part of endopod of maxilliped, posterior; C, maxilla; D, swimming leg 1; E, swimming leg 2; EK swimming leg 3; G, swimming leg 4; I, leg 5. Dermatomyzon nigripes CVI male: H, swimming leg 3, distal segment of endopod. Scale line 1 is 0.1 mm for A; line 2 is 0.05 mm for B; line 3 is 0.1 mm for C—G; line 4 is 0.05 mm for H, I. Table 1.—Spines and setae on swimming legs 1-4 of Dermatomyzon nigripes CVI female. Leg 1 Beg Leg 3 Leg 4 Coxa Basis 1-1 1—O 1-0 1-0 Exopod 2nd; 3rd; Ist I-1; I-1; Il, 1, 4 I-1; I-1; Ill, I, 5 Jes Vests IONS eS) eel i eS Endopod 2nd; 3rd; Ist O=h0—23 12553 O—1; 0-2; 1, 2,3 0-1: 0-2: 1; 1, 3 0-1; O-2; 1, I, 2 666 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Dermatomyzon nigripes (Brady & Robertson, 1876), CVI Male: A, habitus, dorsal; B, urosome, dorsal; C, antennule; D, maxilliped. Dermatomyzon nigripes CV Male: E, habitus, dorsal. Dermatomyzon ni- gripes CV Female: F habitus, dorsal; G, urosome, ventral; H, antennule. Scale line | is 0.1 mm for A, E, F; line 2 is 0.1 mm for B; line 3 is 0.1 mm for H; line 4 is 0.1 mm for D, G; line 5 is 0.1 mm for C. VOLUME 116, NUMBER 3 and 2 thick unarmed setae laterally and 2 longer setae with setules medially. Leg 6 (Fig. 1D): 1 long, thin seta and! 2 very small setae near oviducal opening. CR (Fig. 1C, D): Subquadrate with den- ticles and 1 lateral, dorsal seta, | medial dorsal seta, and 4 terminal setae. CVI male (Fig. 3A).—Differs from CVI female as follows: Body length range 0.90— 1.08 mm (based on 3 specimens); average length of prosome 0.59 mm; maximum width of prosome 0.42 mm; length of uro- some 0.33 mm; ratio of length to width of prosome 1.4; ratio of length of prosome to length of urosome 1.8. Urosome (Fig. 3B): 6 somites; Th7 artic- ulating with Abd2; copulatory pore ventro- lateral. Antennule (Fig. 3C): 17 articulating seg- mes wuling sry Yaris 4a 25 War il, 5 Waa lke Dory (Qe Ne Mee Deed 2 Si wel a Pera es ie ea 10. Geniculation between articulating seg- ments 16 and 17; articulating segments 14— 17 arc-like with articulating segment 15 flexed ventrally toward segment 14 and ar- ticulating segment 16 flexed ventrally to- ward segment 15. Maxilliped (Fig. 3D): Basis with raised pad-like section proximally and medially. Swimming leg 3 (Fig. 2H): Tip of lateral seta of distal endopodal segment not reach- ing to apex of terminal spine; segmental at- tenuation proximal to lateral seta points me- dially, not distally. Leg 6 (Fig. 3B): 3 distolateral setae. CV female (Fig. 3F).—Differs from CVI female as follows: Body length range 0.98— 1.00 mm (based on 2 specimens); average length of prosome 0.67 mm; maximum width of prosome 0.47 mm; length of uro- some 0.32 mm; ratio of length to width of prosome 1.4; ratio of length of prosome to length of urosome 2.1. Urosome (Fig. 3G): 5 articulating sec- tions; Th6, 7, Abdl-—3 articulating. Th7 without copulatory pores or oviducal open- ings. Antennule (Fig. 3H): 11 articulating seg- 667 mnemnrs writin We 24, Jil, Bs ©, 5 3 3, 4b Baril. 2 Leg 6 (Fig. 3G): 3 distolateral setae. CV male (Fig. 3E).—Differs from CV fe- male as follows: Body length range 0.73-— 0.74 mm (based on 2 specimens); average length of prosome 0.49 mm; maximum width of prosome 0.33 mm; length of uro- some 0.25 mm; ratio of length to width of prosome 1.5; ratio of length of prosome to length of urosome 2.0. Antennule: 11 articulating segments with I 2 ORL Om ssS O42 ek CIV (Fig. 4A).—Differs from CV female as follows: Body length range 0.65—0.72 mm (based on 2 specimens); average length of prosome 0.48 mm; maximum width of prosome 0.32 mm; length of urosome 0.21 mm; ratio of length to width of prosome 1.5; ratio of length of prosome to length of urosome 2.3. Urosome (Fig. 4B): 4 articulating sec- tions; Th6, 7, Abd2, | articulating. Antennule (Fig. 4C): 9 articulating seg- MEMES wat Wy ©, Ile Sy 25 dy 4 Zar ll, IZ. Antenna (Fig. 4D), maxillule (Fig. 4E) and maxilliped (Fig. 4K G): As illustrated. Swimming legs 1—4 (Figs. 4H, I; 5A, B): Small, distal seta adjacent to terminal spine of the exopod of swimming legs 3—4. For- mula for spines and setae (Table 2). Leg 5 (Fig. 4B): Basis not articulating with somite and bearing lateral seta; exopod elongate with | thin, unarmed seta medial- ly, | thick, unarmed seta laterally and | ter- minal seta with setules. Leg 6 (Fig. 4B): 2 distal setae. C/IlI.—Differs from CIV female as fol- lows: Body length 0.51 mm (based on | specimen); length of prosome 0.36 mm; maximum width of prosome 0.26 mm; length of urosome 0.15 mm; ratio of length to width of prosome 1.4; ratio of length of prosome to length of urosome 2.4. Urosome (Fig. 5C, D): Th6, 7, Abd] ar- ticulating. Antennule (Fig. 5G): 8 articulating seg- imoveroues waht 2-93). 4b ate ae We 668 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A 2S S> Fig. 4. Dermatomyzon nigripes (Brady & Robertson, 1876), CIV Female: A, habitus, dorsal; B, urosome, ventral; C, antennule; D, antenna; E, maxillule; EK maxilliped; G, maxilliped, proximal part of endopod, posterior; H, swimming leg 1; I, swimming leg 2. Scale line 1 is 0.1 mm for A; line 2 is 0.05 mm for B, C, H, I; line 3 is 0.05 mm for D-G. Antenna (Fig. 6B) and mandible (Fig. Maxilliped (Fig. 6E): Proximally, endo- 6A): As illustrated. pod indistinctly segmented with | seta. Maxillule (Fig. 6C): Outer lobe with 3 Swimming legs 1—4 (Fig. 6F—D): Swim- setae; inner lobe with 4 setae. ming legs 1-3 with 2-segmented rami, Maxilla (Fig. 6D): 3-segmented with in- swimming leg 4 with 1l-segmented rami. distinct arthrodial membrane distally. Distal exopodal seta adjacent to terminal VOLUME 116, NUMBER 3 Table 2.—Spines and setae on swimming legs 1-4 of Dermatomyzon nigripes CIV female. Exopod Endopod Coxa Basis 2nd; 3rd; Ist 2nd; 3rd; Ist eo lt 2021) Pia) “Er *; Tir 5 OSes De5 wesw, OS) 1-0, SA eS Olt ts 125 es. 3) O=l7) 1-0" 1: *: iil, 5 OFS a4 Leg 4 0-1 1-0 I-0; *; Ul, I, 5 OR): 1513 spine of swimming legs 3, 4 tiny. Formula for setae and spines (Table 3). Leg 5 (Fig. 5E): A unilobe ventrolateral bud with 1 distal spine and | distal seta. CR (Fig. 5F): As illustrated. Cll.—Differs from CIII as follows: Body length 0.4 mm (based on 1 specimen); length of prosome 0.29 mm; maximum width of prosome 0.19 mm; length of uro- some 0.13 mm; ratio of length to width of prosome 1.5; ratio of length of prosome to length of urosome 2.2. Prosome (Fig. 7A, B): 3 articulating sec- tions; Ist complex of 5 cephalic somites plus Thl, 2; Th3, 4 articulating. Urosome (Fig. 7A, B): Th5, 6, Abd1 ar- ticulating. Antennule (Fig. 7D): 6 articulating seg- iments with. so less tele Antenna (Fig. 7E), mandible (Fig. 7H) and maxilla (Fig. 7F): As illustrated. Maxillule (Fig. 71): Inner lobe with 3 se- tae. Maxilliped (Fig. 7G): Syncoxa without seta, basis without seta; proximal section of endopod with 1 distally polarized seta and 1 medial seta; distal segment with 2 ter- minal setae, | claw-like. Swimming legs 1—3 (Fig. 7K—M): Swim- ming legs 1-2 with 2-segmented rami, swimming leg 3 with l-segmented rami. Tiny distal seta adjacent to terminal spine of the exopod of swimming leg 3. Formula for spines and setae (Table 4). Swimming leg 4 (Fig. 7C): A ventrolat- eral, bilobe bud; dorsal lobe with 1 distal spine and | distal seta; ventral lobe un- armed. 669 CR (Fig. 7J): As illustrated. Cl.—Differs from CII as follows based on | specimen: length 0.36 mm; maximum width 0.16 mm; length of prosome 0.22 mm, urosome 0.14 mm; ratio of length to width 2.25:1; ratio of length of prosome to that of urosome 1.65:1. Prosome (Fig. 8A, B): 2 articulating sec- tions; Ist, complex of 5 cephalic somites plus Thl, 2; Th3 articulating. Urosome (Fig. 8A, B): Th4, 5, Abd1 ar- ticulating. Antennule (Fig. 8E): 4 articulating seg- rmneinrs watt 2, 2% Il, ilar Ie Antenna (Fig. 8F), mandible (Fig. 8G), maxillule (Fig. 8H) and maxilla (Fig. 8]): As illustrated. Maxilliped (Fig. 8L): Syncoxa and basis unarmed. Endopod of 2 distinct segments; proximal segment with | medial seta and distal segment with 2 setae. Swimming legs 1-2 (Fig. 8J, K): l-seg- mented rami. Formula for spines and setae (Table 5). Swimming leg 3 (Fig. 8C): Ventrolateral, bilobe bud; dorsal lobe with | distal spine and 1 distal seta; ventral lobe unarmed. CR (Fig. 8D): Inner terminal seta lon- gest; remaining terminal setae decreasing in length from inner to outer. Remarks.—The monotypic genera Der- matomyzon Claus, 1889, Australomyzon Nicholls, 1944, and Cheramomyzon Hu- mes, 1989 share appendage segmentation and segmental armature with species of Rhynchomyzon Giesbrecht, 1895. Females of Dermatomyzon with a 19-segmented an- tennule differ from females of Australo- myzon with a 21-segmented antennule. Fe- males of Dermatomyzon differ from fe- males of the poorly-diagnosed Rhyncho- myzon with antennules of between 14 and 16 segments, by the well-developed ros- trum or by the posterolateral extensions of thoracomeres 3 and 4. The mandibular palp of Cheramomyzon is 2-segmented but 1|- segmented in Dermatomyzon. The fifth leg of Australomyzon is 1-segmented but 2-seg- mented in Dermatomyzon. Dermatomyzon 670 lene, Se PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Dermatomyzon nigripes (Brady & Robertson, 1876), CIV Female: A, swimming leg 3; B, swimming leg 4. Dermatomyzon nigripes CII: C, habitus, dorsal; D, habitus, lateral; E, leg 5; EK caudal ramus, dorsal; G, antennule. Scale line | is 0.1 mm for C, D; line 2 is 0.05 mm for A, B; line 3 is 0.05 mm for E-F; line 4 is 0.05 mm for G. nigripes initially was described from the North Sea coast of England. Subsequently it has been reported from areas adjacent to the North Sea (Hansen 1923, Stephensen 1929, Jespersen 1940, Lang 1949), the Mediterranean Sea (Giesbrecht 1899), the Arctic Ocean (Shih et al. 1971), the Indian Ocean (Sewell 1949, Ummerkutty 1966), the Pacific Ocean (Boxshall 1990), and the Southern Ocean (Brady 1910). In the White Sea, copepodids of D. nigripes were found on the bryozoan Flustra foliacea (Linnaeus, VOLUME 116, NUMBER 3 671 Fig. 6. Dermatomyzon nigripes (Brady & Robertson, 1876), CII: A, mandible; B, antenna; C, maxillule; D, maxilla; E, maxilliped; K swimming leg 1; G, swimming leg 2; H, swimming leg 3; I, swimming leg 4. Scale line is 0.05 mm. 672 Table 3.—Spines and setae on swimming legs 14 of Dermatomyzon nigripes CII. Exopod Endopod Coxa__ Basis 2nd; 3rd; Ist 2nd; 3rd; Ist Leg 1 O-1 I-1 I-1; *; Ill, 4 Oj 205 Lee 2 O=] 1-0 Fie iia sS O-1; *; 1, 2, 4 Leg 3 O-1 1-0 I-0; *; II, I, 4 Opis * ails Leg 4 0-0 1-0 caer UU Ne (ne "apm al bo Cah) 1758). A small number of adults of D. ni- gripes were washed from the sponge Hali- chondria panicea (Pallas, 1766) and the echinoderm Ophiopholis aculeata (Linnae- us, 1767). Adults also were collected around hydrozoans and other invertebrates from bottom covered by dead shells scoured by strong tidal currents. These find- ings suggest that adults of D. nigripes may be an unspecialized symbiont associated with different invertebrates of the White Sea sublittoral. However, immature cope- podids of D. nigripes, along with other co- pepods, were associated only with Flustra foliacea (see Ivanenko & Smurov 1997) suggesting that the bryozoan may be the preferred host of the immature copepodids. Asterocheres lilljeborgi Boeck, 1859 Figs. 9-18 Asterocheres lilljeborgi Boeck, 1859:176, pl. 2, figs. 1-11.—Brady, 1880:64—65. Canu, 1892:264, pl. 27, figs. 1-6.—Gies- brecht, 1899:70, 73, pl. 3, figs. 21-26.— Wilson, 1944:547, pl. 30, figs. 161-— 162.—Roettger et al. 1972:259, figs. 1- 9.—Gotto, 1993:153, fig. 36 B—J. Ascomyzon asterocheres Sars, 1914:85—87, pls. 51-52. CVI female.—Body length range 1.41-— 1.47 mm (based on 3 specimens); average length of prosome 0.99 mm; maximum width of prosome 1.09 mm; length of uro- some 0.44 mm; length of genital complex 0.16 mm; width of genital complex 0.21; ratio of length to width of prosome 0.9; ra- tio of length of prosome to length of uro- some 2.3. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 4.—Spines and setae on swimming legs 1-3 of Dermatomyzon nigripes CII. Exopod 2nd; 3rd; Ist Endopod Coxa Basis 2nd; 3rd; 1st Ress 0S loa sine O-1; *; 1, 2,4 Leg 2, O-1 1-0 E0O;*;1, 15,4 O=-l2 =i Leg 3 0-0 1-0 cero Ud is 8} eaeie | 2) Prosome (Fig. 9A): Flattened dorsoven- trally with tergites pointed posteriorly; 4 ar- ticulating sections; Ist complex of 5 ce- phalic somites plus Thl, 2 with spinules; Th3-—5 articulating. Urosome (Fig. 9B): 4 articulating sec- tions; Th6, genital complex of Th7 fused to Abd2, Abd3, and Abd1l. Th6 and genital complex with spinules; paired copulatory pores ventrolateral; paired oviducal open- ings dorsolateral. Rostral area flattened; rostrum absent. Oral siphon (Fig. 10D): Tip reaching be- yond base of maxilla. Labrum: With spinules on tip. Antennule (Fig. 9E, F): 21 articulating Sesments) with 2; 2) 252.22 Oe ae Di Dri DD OD Dale ae mor Antenna (Fig. 10A): Coxa and basis without setae. Exopod apparently 1l-seg- mented with 2 terminal setae and proximal medial seta (Fig. 10B). Endopod 3-seg- mented; Ist segment with denticles, 2nd with 1 seta, 3rd with denticles and 4 setae, including large terminal claw. Mandible (Fig. 10C): Gnathobase slight- ly longer than palp, tapering distally with denticles; 2-segmented palp, both segments with denticles, distal segment with 2 ter- minal setae. Maxillule (Fig. 10E): Inner lobe with several sets of denticles, 4 large setae and 1 small seta; outer articulating lobe with denticles and 4 terminal setae. Maxilla (Fig. 1OF): An apparently 2-seg- mented subchela; Ist segment unarmed; 2nd segment distally pointed, curved, with set of denticles but without setae. Maxilliped (Fig. 10G, H): Short syncoxa VOLUME 116, NUMBER 3 673 Fig. 7. Dermatomyzon nigripes (Brady & Robertson, 1876), CII: A, habitus, dorsal; B, habitus, lateral; C, swimming leg 3; D, antennule; E, antenna; K maxilla; G, maxilliped; H, mandible; I, maxillule; J, caudal ramus; K, swimming leg 1; L, swimming leg 2; M, swimming leg 3. Scale line | is 0.05 mm for A, B; line 2 is 0.05 mm for C—M. 674 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 8. Dermatomyzon nigripes (Brady & Robertson, 1876), CI: A, habitus, dorsal; B, habitus, lateral; C, swimming leg 3; D, caudal ramus; E, antennule; EK antenna; G, mandible; H, maxillule; I, maxilla; J, swimming leg 1; K, swimming leg 2; L, maxilliped. Scale line 1 is 0.05 mm for A, B; line 2 is 0.05 mm for C—E, J, K; line 3 is 0.05 mm for E H, L; line 4 is 0.05 mm for G, I. VOLUME 116, NUMBER 3 Table 5.—Spines and setae on swimming legs 1—2 of Dermatomyzon nigripes CI. Exopod Endopod Coxa Basis 2nd; 3rd; Ist 2nd; 3rd; Ist Leg @0 Ik 2 IN,ILS = i124 Reon "0-0 9l—O' *; Mill I,.3 Soe aD tS with 1 inner seta, long basis with | tiny, inner seta on medial margin; endopod of proximal, short, apparently subdivided sec- tion with 4 setae, 2 proximal, | middle, | distal; long distal segment with denticles and 2 terminal setae, largest thick and curved. Swimming legs 1—4 (Fig. 11A—E): Bir- amal, with 3-segmented rami. Formula for spines and setae (Table 6). Lateral seta of the distal endopodal segment of swimming leg 3 not reaching tip of lateral apical seta; lateral apical seta of the distal endopodal segment of swimming leg 4 half the length of apical spine. Leg 5 (Fig. 11F): Basis not articulating with somite, with lateral seta; articulating segment elongate with 2 unarmed terminal setae and | medial seta, plus medial and lateral denticles. Leg 6 (Fig. 9C): 1 long seta and | short seta near oviducal openings. CR (Fig. 9D): Subquadrate with 2 dorsal setae and 4 terminal setae. CVI male.—Differs from CVI female as follows: Body length range 1.01—1.06 mm (based on 3 specimens); average length of prosome 0.70 mm; maximum width of pro- some 0.70 mm; length of urosome 0.36 mm; ratio of length to width of prosome 1.0; ratio of length of prosome to length of urosome 1.9. Urosome (Fig. 12A): Th6, 7, Abd2, 3, 1 articulating. Th7 with copulatory pore ven- trolaterally. Antennule (Fig. 12B—D): 18 articulating SSAHSMUS WANN? A, Le WE) I LED (Os a Ae en a GeniGulaton:be- tween 15th and 16th articulating segments; segment 10 unsclerotized dorsally. Maxilla (Fig. 12G): Apparently 3-seg- 675 mented subchela; Ist segment with proxi- mal denticles; 2nd with 1 small seta; 3rd distally pointed and curved. Maxilliped (Fig. 12H): Basis with raised bump proximally and medially. Swimming legs 1—4 (Fig. 11G—J): Swim- ming leg | with denticles on distal segment of endopod. Distal attenuation of middle segment of endopod of swimming leg 2 rel- atively longer than corresponding structure on other swimming legs; denticles on distal segment of endopod. Lateral seta on distal endopodal segment of swimming leg 3 al- most reaching to tip of apical seta. Lateral apical seta on distal endopodal segment of swimming leg 4 reaching %4 length of apical spine. Leg 5 (Fig. 12F): As illustrated. Leg 6 (Fig. 12E): Several sets of denti- cles and 2 setae, distal thickest. CV female (Fig. 13A).—Differs from CVI female as follows: Body length range 0.84—1.05 mm (based on 7 specimens); av- erage length of prosome 0.64 mm; maxi- mum width of prosome 0.62 mm; length of urosome 0.30 mm; ratio of length to width of prosome 1.0; ratio of length of prosome to length of urosome 2.1. Urosome (Fig. 13A): Th6, 7, Abd2, 3,1 articulating; no copulatory pore or oviducal openings on Th7. Antennule (Fig. 13C): 18 articulating SEMMEMUS Wt 25 25 2, 2 25 5 A, 5 (Sy 2 ew DD Dales Leg 6 (Fig. 13A): Uniramal bud with 1 lateral seta and medial denticles. CV male (Fig. 13B).—Differs from CV female as follows: Body length range 0.85— 0.89 mm (based on 2 specimens); average length of prosome 0.59 mm; maximum width of prosome 0.54 mm; length of uro- some 0.29 mm; ratio of length to width of prosome 1.1; ratio of length of prosome to length of urosome 2.5. Leg 6: Unilobe bud with | lateral seta and | terminal seta. CIV.—Differs from CV female as fol- lows: Body length range 0.65—0.80 mm (based on 7 specimens); average length of 676 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Vi Vv vw Vv ww vv YY Vv y fi mo, (i Pe = t le Df ? 2) 4 (C Fig. 9. Asterocheres lilljeborgi Boeck, 1859, CVI: A, Female, habitus, dorsal; B, urosome, dorsal; C, leg 6; D, caudal ramus, dorsal; E, antennule, segments 1—15; E antennule, segments 16—21. Scale line 1 is 0.1 mm for A; line 2 is 0.1 mm for B; line 3 is 0.1 mm for C, D; line 4 is 0.1 mm for E, F VOLUME 116, NUMBER 3 pie Co ines ~S P= a UTTT 677 Fig. 10. Asterocheres lilljeborgi Boeck, 1859, CVI Female: A, antenna; B, exopod of antenna; C, mandible; D, oral siphon; E, maxillule; EK maxilla; G, maxilliped; H, proximal part of endopod of maxilliped, posterior. Scale line | is 0.1 mm for A, C—G; line 2 is 0.05 mm for B, H. prosome 0.48 mm; maximum width of pro- some 0.40 mm; length of urosome 0.22 mm; ratio of length to width of prosome 1.2; ratio of length of prosome to length of urosome 2.2 (Fig. 13E). Urosome (Fig. 13G): Th6, 7, Abd2, | ar- ticulating. Antennule (Fig. 13F): 17 articulating Sesimentsawitn, 2, le eles 11 Al i 2: 440252, 240,114. Maxilla as illustrated (Fig. 14A). Swimming legs 1—4 (Figs. 14B—F): 2- segmented rami. Formula for spines and se- tae (Table 7). Leg 5 (Fig. 13G): Basis not articulating with somite, with lateral seta; free segment with 1 small seta laterally and 2 thick, un- armed setae distally. Leg 6 (Fig. 13G): Unilobe bud with 1 seta. CilI.—Differs from CIV as follows: Body length range 0.50—0.54 mm (based on 3 specimens); average length of prosome 0.37 mm; maximum width of prosome 0.30 678 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON LO 5G S oe eZ | aN oS So \ \ ays Fig. 12. Asterocheres lilljeborgi Boeck, 1859, CVI Male: A, habitus, dorsal; B, antennule, segments 1—16; C, antennule, segments 17—18 anterior; D, antennule, segments 17—18 anterior; E, leg 6; EK leg 5; G, maxilla; H, maxilliped. Scale line | is 0.1 mm for A; line 2 is 0.05 mm for D; line 3 is 0.05 mm for E; line 4 is 0.05 Mii PLOLs Ca e-) ine s1s,Ohlemmy tor Gere mm; length of urosome 0.14 mm; ratio of Antennule (Fig. 15C): 12 articulating lenethatomwidthmomprosomen: 2 ratlonot —segments, with 25 15 1, I 3. i i, 252.2. length of prosome to length of urosome 2.6. 3+1, 14. Urosome (Fig. 15A): Th6, 7, Abd] artic- Antenna (Fig. 15D), mandible (Fig. 15E) ulating. and maxilla (Fig. 15G): As illustrated. 680 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 6.—Spines and setae on swimming legs 1-4 of Asterocheres lilljeborgi adult female. Exopod Endopod Coxa Basis 2nd; 3rd; 1st 2nd; 3rd; 1st ecw O-1 1-1 I-1; I-1; Ul, 1, 3 O—1; O-2; 1, 2, 3 Leg 2 O-1 1-0 I-1; 1-1; IIL, I, 4 0-1; 0-2; 1, 2, 3 Leg 3 O-1 1-0 I-1; I-1; Ul, I, 4 O=1:, 0=2-7 1 Eris Leg 4 O-1 1-0 I-1; I-1; Ul, 1, 4 O=); 02. Wiki 5 Fig. 13. Asterocheres lilljeborgi Boeck, 1859, CV Female: A, habitus, dorsal. Asterocheres lilljeborgi CV Male: B, male, habitus, dorsal; C, antennule; D, antennule, segment 9. Asterocheres lilljeborgi, CIV: E, habitus, dorsal; EK antennule; G, urosome, ventral. Scale line 1 is 0.1 mm for A, B, E; line 2 is 0.1 mm for C; line 3 is 0.1 mm for D; line 4 is 0.1 mm for F; line 5 is 0.1 mm for G. VOLUME 116, NUMBER 3 Fig. 14. Asterocheres lilljeborgi Boeck, 1859, CIV: A, maxilla; B, swimming leg 1; C, swimming leg 2; D, swimming leg 3; E, swimming leg 4. Scale line is 0.1 mm. Maxillule (Fig. 15F): Outer lobe with 3 setae. Maxilliped (Fig. 16A): Proximal section of endopod 3-segmented each with | dis- tally polarized seta; distal segment with 2 terminal setae. Swimming legs 1—4 (Fig. 16B—E): Bira- mal; swimming legs 1—3 with 2-segmented 681 682 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 7.—Spines and setae on swimming legs 1-4 of Asterocheres lilljeborgi CIV female. Coxa Basis Leg 1 0-1 1-1 Leg 2 O-1 1-0 Leg 3 0-1 1-0 Leg 4 O-1 1-0 rami, swimming leg 4 with Il-segmented rami. Formula for spines and setae (Table 8). Leg 5 (Fig. 15B): Unilobe bud with 1 distal spine and | distal seta. ClI.—Differs from CIII as follows based on 5 specimens: length range 0.41—0.44 mm; average maximum width 0.2—0.22 mm; greatest dorsoventral thickness 0.13-— 0.14 mm; length of prosome 0.26—0.29 mm, urosome 0.14—0.15 mm; ratio of length to width 2.02:1; ratio of length of prosome to that of urosome 1.95:1. Prosome (Fig. 17A): 3 articulating sec- tions; Ist a complex of 5 cephalic somites plus Thl, 2; Th3, 4 articulating; Th4 nar- row. Urosome (Fig. 17A): Th5, Th6, Abd1 ar- ticulating. Antennule (Fig. 17C): 9 articulating seg- ones writin, 2, Il, Ze Il, Jl, il, il, wari, Wea Antenna (Fig. 17D), mandible (Fig. 17E) and maxilla (Fig. 17F): As illustrated. Maxillule (Fig. 17H): Inner lobe with 3 setae. Maxilliped (Fig. 17G): Syncoxa without seta, basis without seta. Endopod 3-seg- mented; proximal and middle segment each with | seta, distal segment with 2 terminal setae. Swimming legs |—3 (Fig. 17I-M): Swim- ming legs 1l—2 with 2-segmented rami, swimming leg 3 with 1l-segmented rami. Formula for spines and setae (Table 9). Swimming leg 4 (Fig. 17B): A bilobed bud; dorsal lobe with | distal spine and 1 distal setae; ventral lobe unarmed. Cl.—Differs from CII as follows: Body length range 0.38 mm (based on | speci- men); length of prosome 0.26 mm; maxi- mum width of prosome 0.16 mm; length of Exopod 2nd; 3rd; Ist 2nd; 3rd; Ist I-1; *: Ill, 1, 4 O=teets 2S eet Tie SS QSlEI ee ORES Ted S20. es O-1; *; 1, 1+1, 4 1-02 il iS Oils oe I. irl. 3 urosome 0.12 mm; ratio of length to width of prosome 1.6; ratio of length of prosome to length of urosome 2.2. Prosome (Fig. 18A): 2 articulating sec- tions; Ist complex of 5 cephalic somites plus Thl, 2; Th3 articulating. Urosome (Fig. 18B): Th4, 5, Abd1 artic- ulating. Antennule (Fig. 18C): 4 articulating seg- onenus wwiltin 2, 2, jl, Waser Il. Antenna (Fig. 18D), mandible (Fig. 18E), maxillule (Fig. 18F) and maxilla (Fig. 18G): As illustrated. Maxilliped (Fig. 18H): Syncoxa and ba- sis unarmed; endopod 2-segmented; proxi- mal segment with | medial seta and distal segment with | thick and | thin seta. Swimming legs 1—2 (Fig. 18I, J): l-seg- mented rami. Formula for spines and setae (Table 10). Swimming leg 3 (Fig. 18B): Bilobed bud; dorsal lobe with | distal spine and 1 distal seta; ventral lobe unarmed. CR (Fig. 18B): As illustrated. Remarks.—Asterocheres lilljeborgi, the type species of the family, can be distin- guished from its congeners by exceptionally wide tergites of the dorsoventrally flattened prosome. It has been collected in the Davis Strait, off coasts of the British Isles, Swe- den, Norway and France, associated with the starfish Henricia sanguinolenta (O. FE Miller, 1776) and other echinoderms (Hu- mes 1986, Gotto 1993). Wilson (1944) found A. lilljeborgi associated with Henri- cia leviuscula (Stimpson, 1857) in the Aleutian Islands near Alaska. Marchenkov (1997) reported that A. lilljeborgi at the White Sea is associated with H. sanguino- lenta. Roettger et al. (1972) studied ecology VOLUME 116, NUMBER 3 Fig. 15. . Asterocheres lilljeborgi Boeck, 1859, CII: A, habitus, dorsal; B, urosome, ventral; C, antennule; D, antenna; E, mandible; K maxillule; G, maxilla. Scale line | is 0.1 mm for A; line 2 is 0.1 mm for B, D—F; imers 1s) Os immrton. ce of A. lilljeborgi from H. sanguinolenta col- lected near Sweden. Discussion One somite is added to the body of both genders of Dermatomyzon nigripes and As- terocheres lilljeborgi during the molts to copepodids I-IV, a pattern similar to the related siphonostomatoid Scottomyzon gib- berum (see Ivanenko et al. 2001). During the molt to copepodid V, both genders of D. nigripes add a fourth abdominal somite, but females and males of A. /illjeborgi ap- pear to suppress the formation of the fourth somite as do both genders of S. gibberum. During the terminal adult molt to CVI, fe- males of A. lilljeborgi and D. nigripes de- velop a simple genital complex when an ar- throdial membrane separating the genital somite from the second abdominal somite fails to form. This arthrodial membrane does form in females of S. gibberum and in males of all three species. The body of the ancestral copepod is assumed to have added 684 ee es PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON oo Tee Fig. 16. Asterocheres lilljeborgi Boeck, 1859, CII: A, maxilliped; B, swimming leg 1; C, swimming leg 2; D, swimming leg 3; E, swimming leg 4. Scale line | is 0.1 mm. a somite during the molt of each copepodid stage (Hulsemann 1991), and the arthrodial membrane separating the genital somite from the second abdominal somite re- mained during the molt to copepodid VI of females, as it does for a few podoplean neo- copepods (Martinez Arbizu 2003). The body of the ancestral siphonostomatoid is assumed to have added a somite during the molt of each copepodid stage but the ar- throdial membrane separating the female genital somite from the second abdominal VOLUME 116, NUMBER 3 685 Table 8.—Spines and setae on swimming legs 1—4 of Asterocheres lilljeborgi Cll. Leg | O-1 1-1 Lee 2 O-1 1—O Leg 3 0-1 1-O Leg 4 0-0 1-O somite failed to form during the molt to co- pepodid VI. The body of D. nigripes ap- pears to be identical to the ancestral si- phonostomatoid. The body of A. lilljeborgi is derived because the fourth abdominal so- mite fails to form during the molt to cope- podid V of both genders. The body of S. gibberum also is derived because a fourth abdominal somite fails to form and because an arthrodial membrane separates the gen- ital somite from the second abdominal so- mite of females, a character state reversal. The maxilliped of D. nigripes and A. lill- Jeborgi, like S. gibberum, begins develop- ment at copepodid I with an unarmed syn- coxa, an unarmed basis, and a 2-segmented endopod with one seta on the proximal seg- ment and two setae on the distal segment. Both D. nigripes and A. lilljeborgi add one seta each to the syncoxa and basis at co- pepodid III. One seta also is added to a dis- tinctly segmented endopod at copepodid II and at copepodid III; a third seta juxtaposed to the proximal seta of the endopod is add- ed at copepodid IV. Patterning of the en- dopod of the maxilliped of D. nigripes and A. lilljeborgi during development is as- sumed to follow the general model for co- pepods (Ferrari 1995, Ferrari & Dahms 1998, Ferrari & Ivanenko 2001), as it is ap- plicable to siphonostomatoids (Ivanenko et al. 2001). Based on the stage at which each seta is added and its location, the endopod of adult D. nigripes and A. lilljeborgi is in- terpreted as 4-segmented. The maxilliped of S. gibberum fails to add a seta to its syn- coxa during development and the seta on the second, or penultimate, endopodal seg- ment is lost at CIV; both of these states are interpreted as derived. Exopod Endopod 2nd; 3rd; Ist 2nd; 3rd; Ist ek; =: EEE 1,.4 OSI Shea) isl IND Ble S) O=fe*3 dl, 254 IO; $2 0, ¥..4 DES erie Srl 3 OE) HH Lae eal ae bale, Swimming leg 1 of copepodid I of D. nigripes and A. lilljeborgi bears eight exo- podal setae and seven endopodal setae, the most common number of elements for these rami at this stage of copepod development (Ferrari 2000). Development from cope- podid I is similar for both species with one exception. The proximal medial seta on the third exopodal segment (or the ventral seta on the presumptive fourth segment) of D. nigripes fails to form on A. Jilljeborgi dur- ing the molt to CV so that there are three inner setae, not four, on the adult of A. Jil/l- Jeborgi. Swimming leg 2 of copepodid I of D. nigripes and A. lilljeborgi bears seven ex- opodal setae and six endopodal setae, the most common number of elements for these rami at this stage of copepod development (Ferrari 2000). Development from cope- podid I is similar for both species with one exception. The proximal medial seta on the third exopodal segment, or ventral seta on the presumptive fourth segment, of D. ni- gripes fails to form on A. lilljeborgi during the molt to CV so that there are four medial setae, not five, on the adult of A. Jilljeborgi. Swimming leg 3 of copepodid I of both species is a bud with two setal elements on the dorsal lobe, the presumptive exopod, and none on the ventral lobe, the presump- tive endopod. The common situation for co- pepods is three setae on the presumptive ex- opod and two setae the presumptive endo- pod (Ferrari 2000). At copepodid II, the transformed limb of D. nigripes and A. /ill- jJeborgi bears seven exopodal setae and six endopodal setae, the most common number of elements for these rami at this stage of copepod development (Ferrari 2000). De- 686 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 17. Asterocheres lilljeborgi Boeck, 1859, CII: A, habitus, dorsal; B, swimming leg 4; C, antennule; D, antenna; E, mandible; K maxilla; G, maxilliped; H, maxillule; I, swimming leg 1, protopod, exopod and inter- coxal sclerite; J, endopod of swimming leg 1; K, swimming leg 2, protopod, exopod and intercoxal sclerite; L, endopod of swimming leg 2; M, swimming leg 3. Scale line | is 0.1 mm for A; line 2 is 0.1 mm for C; line 3 is O.1 mm for B, D—M. VOLUME 116, NUMBER 3 687 Table 9.—Spines and setae on swimming legs 1-3 of Asterocheres lilljeborgi Cll. Lee Il 0-1 1-1 Leg 2 0-1 1-0 eces 0-0 1-0 velopment from copepodid II is similar for both species with two exceptions. D. nigri- pes fails to add a second terminal element to the endopod during the molt to copepod- id III, while A. lilljeborgi adds this second terminal element. The proximal medial seta on the third exopodal segment, or ventral seta on the presumptive fourth segment, of D. nigripes fails to form on A. lilljeborgi during the molt to CV so that there are four medial setae, not five, on the adult of A. lilljeborgi. Swimming leg 4 of copepodid II of both species is a bud with two setal elements on the dorsal lobe, the presumptive exopod, and none on the ventral lobe, the presump- tive endopod. The common situation for co- pepods is three setae on the presumptive ex- opod and two setae the presumptive endo- pod (Ferrari 2000). At copepodid HI, the transformed limb of D. nigripes and A. Iill- Jeborgi bears seven exopodal setae and six endopodal setae, the most common number of elements for these rami at this stage of copepod development (Ferrari 2000). De- velopment from copepodid III is similar for both species with two exceptions. D. nigri- pes fails to add a second terminal element on the endopod during the molt to cope- podid III, while A. Jilljeborgi adds this sec- ond terminal element. The proximal medial seta on the third exopodal segment, or ven- tral seta on the presumptive fourth segment, of D. nigripes fails to form on A. lilljeborgi during the molt to CV so that there are four medial setae, not five, on the adult of A. lilljeborgi. Leg 5 of copepodid III of both species is a bud with two setal elements on the dorsal lobe, the presumptive exopod; there is no ventral lobe. At copepodid IV, the trans- Exopod 2nd; 3rd; Ist Endopod 2nd; 3rd; Ist Oss i 4 EOE 2 1001, Mt, a! pane Te eS formed limb of both species bears three ex- opodal setae. At copepodid V of D. nigripes a medial and lateral seta are added to the exopod; no setae are added to A. Jilljeborgi. Leg 6 of copepodid IV of D. nigripes is a bud with two setae on the dorsal lobe which is assumed to be the presumptive exopod; a third seta is added during the molt to co- pepodid V. Leg 6 of copepodid IV of A. lilljeborgi is a bud with one seta on the dor- sal lobe; a second seta is added to the male during the molt to copepodid V, and to the female during the molt to copepodid VI. Following the simplifying assumption of oligomerization (Dogiel 1954, Monchenko & Von Vaupel Klein 1999) to infer derived states of swimming legs which develop from serially repeated elements, adults of S. gibberum share with those of D. nigripes absence of a second terminal element on the third endopodal segment of swimming legs 3—4. Adults of S. gibberum share with those of A. lilljeborgi absence of the proximal medial seta on the third exopodal segment of swimming legs 1—4, three setae on the exopod of leg 5 and two setae the pre- sumptive exopod of leg 6. Derived states of adult S$. gibberum include absence of the medial basal seta on leg 1, absence of the medial coxal seta of leg 4, and absence of the proximal lateral seta on the third exo- podal segment, or dorsal seta of the pre- sumptive fourth segment, of swimming legs 1-4. The proximal segment of the antennule of the adult female and male of A. Jillje- borgi bears two setae. The proximal seg- ment of the antennule of the adult female of D. nigripes and S. gibberum bears only one seta. One of two setae present on the proximal segment of D. nigripes early in 688 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 18. Asterocheres lilljeborgi Boeck, 1859, CI: A, habitus, dorsal; B, urosome, ventral; C, antennule; D, antenna; E, mandible; K maxillule; G, maxilla; H, maxilliped; J, swimming leg 1; I, swimming leg 2. Scale line 1 is 0.1 mm for A; line 2 is 0.1 mm for B; line 3 is 0.1 mm for C; line 4 is 0.1 mm for D-I. VOLUME 116, NUMBER 3 Table 10.—Spines and setae on swimming legs 1|— 2 of Asterocheres lilljeborgi Cl. Exopod Endopod Coxa Basis 2nd; 3rd; 1st 2nd; 3rd; Ist Beco OOF eal Ore FEV 3 rena ole) a Leg 2 0-0 1-0 soo i) ff Da Lee eee AAS development fails to form during the molt to copepodid IV; in contrast the proximal segment of S. gibberum bears only one seta throughout the copepodid phase of devel- opment. The ninth articulating segment of the adult female of D. nigripes bears only one seta as does the sixth articulating segment of the adult female of S. gibberum; these segments are not homologous. The ninth ar- ticulating segment of the antennule of the adult female of A. /illjeborgi is considered a complex of three segments with six setae, two from each segment. The third articu- lating segment of the antennule of the adult female of D. nigripes is considered a com- plex of two segments with four setae, two from each segment. The eighth articulating segment of D. nigripes is considered a com- plex of three segments also with six setae, and which corresponds to the complex of A. lilljeborgi. The antennule of the adult fe- male of S. gibberum also has two complex- es, a proximal complex of three segments of six setae and a distal complex of four segments with eight setae. These two com- plexes correspond to the two of D. nigripes but with the following articulating segment of D. nigripes fused to the corresponding complex of S. gibberum. The antennule of A. Jilljeborgi appears to have the fewest derived states: only one segmental complex composed of three proximal segments. The antennule of D. ni- gripes has the following derived states: one seta on the proximal segment; a proximal segmental complex of two segments; and a distal complex of three segments. The an- tennule of S. gibberum has the following derived states: one seta on the proximal segment; a proximal complex of three seg- 689 ments; and a distal complex of four seg- ments. Acknowledgments Research of VNI at the National Museum of Natural History, Smithsonian Institution, Washington, D.C. was supported by a Short-Term Visitor Grant from the Smith- sonian Institution. Thanks to Dmitry Zhad- an and Natasha Cherviakova (Department of Invertebrate Zoology, Moscow State University), and Lena Markhaseva and Al- exander Plotkin (Department of Embryol- ogy, St. Petersburg State University) for collecting samples. Zbigniew Kabata, Pa- cific Biological Station, Ju-shey Ho, Cali- fornia State University at Long Beach, and an anonymous reviewer, improved the man- uscript immeasurably. Literature Cited Boeck, A. 1859. Tvende nye parasitiske Krebsdyr, Ar- totrogus orbicularis og Asterocheres Liljebor- gii.—Forhandlinger i Videnskabers Selskabet 1 Kristiania 1859:171—182. Boxshall, G. A. 1990. Siphonostome copepods asso- ciated with sponges from Hong Kong.—Pp. 523-547 in B. Morton, ed., The marine flora and fauna of Hong Kong and Southern China, vol. 2. Taxonomy and ecology. Hong Kong University Press, Hong Kong, 449-942. Brady, G. S. 1880. A monograph of the free and semi- parasitic Copepoda of the British Islands. The Ray Society, London, 3:1—83. . 1910. Die marinen Copepoden der deutschen Sudpolar-Expedition 1901—1903. 1. Uber die Copepoden der Stamme Harpacticoida, Cyclo- poida, Notodelphyoida und Caligoida.—Deut- sche Sudpolar Expedition 1901—1903, 11 (Zo- ology 3) (5):499—593. Brady, G. S., & D. Robertson. 1876. Report on dredg- ings off the coast of Durham and North-York- shire in 1874.—Report to the British Associa- tion for the Advancement of Science 45:185— IO). Canu, E. 1892. Les Copepodes du Boulonnais. Mor- phologie, embryologie, taxonomie.—Travaux de I’ Institut Zoologique de Lille, Travaux de la Station de Zoologie Maritime de Wimereux 6: 1-292, pls. 1-30. . 1893. Un Copepode ascomyzontide sur une algue pelagique.—Notes de biologie marine, fauniques ou ethologiques. I. Annales de la Sta- 690 tion Aquicole de Boulogne-sur-Mer 1:100—107, pls. 6—7. Claus, C. 1889. Ueber neue oder wenig bekannte halb- parasitische Copepoden insbesondere der Lich- omolgiden- und Ascomyzontiden-Gruppe.—Ar- beiten des Zoologischen Institut der Universitat Wien 8:327—370. Dogiel, V. A. 1954. Oligomerization of the homolo- gous organs as one of the main paths in an an- imal evolution.—Leningrad University Press, Leningrad [now St. Petersburg], 368 pp. Eiselt, J. 1965. Revision und Neubeschreibung weiter- er siphonostomer Cyclopoiden (Copepoda, Crust.) aus der Antarktis.—Sitzungsberichten der Osterreichischen Akademie der Wissen- schaften, Mathematisch-Naturwissenschaftliche Klasse, Abteilung I. 174:51—169. Ferrari, EF D. 1995. Six copepodid stages of Ridge- wayia klausruetzleri, a new species of calanoid copepod (Ridgewaylidae) from the barrier reef in Belize, with comments on appendage devel- opment.—Proceedings of the Biological Society of Washington 108:180—200. . 2000. Patterns of setal numbers conserved during early development of swimming legs of copepods (Crustacea).—Hydrobiologia 417:81— 90. , & A. Benforado. 1998. Relationships between arthrodial membrane formation and addition of setae to swimming legs 1—4 of Dioithona ocu- lata, Ridgewayia klausruetzleri, Plearomamma xiphias, and Temora longicornis (Crustacea: Copepoda).—Crustaceana 71:545—564. , & H.-E. Dahms. 1998. Segmental homologies of the maxilliped of some copepods as inferred by comparing setal numbers during copepodid development.—Journal of Crustacean Biology 18:298—307. , & V.N. Ivanenko. 2001. Interpreting segment homologies of the maxilliped of cyclopoid co- pepods by comparing stage-specific changes during development.—Organisms, Diversity and Evolution 1:113-131. Giesbrecht, W. 1895. The subfamilies, genera, and spe- cies of the copepod family Ascomyzontidae, Thorell: diagnosis, synonymy and distribu- tion.—Annals and Magazine of Natural History (6)16:173—186. 1899. Die Asterocheriden des Golfes von Neapel und der angrenzenden Meeres-Abschn- itte—Fauna und Flora des Golfes von Neapel und der Angrenzenden Meeres-Abschnitte, Her- ausgegeben von der Zoologischen Station zu Neapel 25, 217 pp., pls. 1-11. Gotto, R. V. 1993. Commensal and parasitic copepods associated with marine invertebrates (and whales): keys and notes for identification of the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON species.—Synopses of the British Fauna, New Series 46:1—264. Hansen, H. J. 1923. Crustacea Copepoda. II. Copepoda Parasita and Hemiparasita.—Danish Ingolf-Ex- pedition 3, 92 pp. Ho, J-s. 1990. A phylogenetic analysis of copepod or- ders.—Journal of Crustacean Biology 10:528— 536. Hulsemann, K. 1991. Tracing homologies in appendages during ontogenetic development of calanoid copepods.—Bulletin of the Plankton Society of Japan, Special Volume, 105-114. Hulsemann, K. 1991. Tracing homologies in append- ages during ontogenetic development of calan- oid copepods.—Bulletin of the Plankton Soci- ety of Japan, Special Volume, 105-114. Humes, A. G. 1986. Synopsis of copepods associated with asteroid echinoderms, including new spe- cies from the Moluccas.—Journal of Natural History 20:981—1020. . 1989. Copepoda from deep-sea hydrothermal vents at the East Pacific Rise.—Bulletin Musé- um national d’ Histoire naturelle, Paris, 4e série 11, section A. no. 4:829-849. . 1999. Collocherides brychius, a new species (Copepoda: Siphonostomatoida) from a deep- water hydrothermal site in the northeastern Pa- cific.—Proceedings of the Biological Society of Washington 112:181—188. , & R. U. Gooding. 1964. A method for study- ing the external anatomy of copepods.—Crus- taceana 6:238—240. Huys, R., & G. A. Boxshall. 1991. Copepod Evolu- tion.—The Ray Society vol. 159, 468 pp. Ivanenko, V. N., FE D. Ferrari, & A. V. Smurov. 2001. Naupli and copepodids of Scottomyzon gibber- um (Copepoda: Siphonostomatoida: Scottomy- zontidae, new family), a symbiont of Asterias rubens (Asteroidea).—Proceedings of the Bio- logical Society of Washington 114:237—261. , & A. V. Smurov. 1997. Asterocheres flustrae n. sp. (Copepoda; Siphonostomatoida; Aster- ocheridae), associated with Flustra foliacea L. (Bryozoa) from the White Sea.—Systematic Parasitology 38:111—130. Jespersen, P. 1940. Non-parasitic Copepoda.—The Zo- ology of Iceland 3 (33), 116 pp. Lang, K. 1934. Marine Harpacticiden von der Camp- bell-Insel und einigen anderen Suedlichen In- seln.—Kunglia Fysiografiska Saellskapets I Lund Handlingar n.f. 45:1—56. . 1949. On some Swedish marine semi-parasitic and parasitic copepods.—Arkiv for Zoologi (A) 42(22):1-10. Linnaeus, C. 1758. Systema Naturae, per regna tria naturae, secundum Classes, Ordines, Genera, Species, cum characteribus, differentiis, synon- ymis, locis. Ed. decima, reformata. I, Regnum animale: 1—823. Laurentius Salvius, Holmiae. VOLUME 116, NUMBER 3 . 1767. Systema Naturae per regna tria naturae, secundum classes, ordines, genera, species cum characteribus, differentiis, synonymis, locis. Ed. duodecima, reformata, 1(2):533—1327. Lauren- tius Salvius, Holmiae. Marchenkov, A. V. 1997. Fauna of symbiotic and par- asitic crustaceans (Copepoda, Isopoda, Cirripe- dia, Amphipoda) from the benthic invertebrates of the White Sea and Arctic seas.—Parasitolo- giya 31:24—44. Martinez Arbizu, P. 2003. The paraphyly of Cyclopin- idae Sars, 1913, and the phylogenetic position of poecilostome families within Cyclopoida Sars, 1913 (Copepoda: Crustacea) develop- ment.—Organisms, Diversity and Evolution 3 (in press). Monchenko, V. I., & J. C. von Vaupel Klein. 1999. Oligomerization in Copepoda Cyclopoida as a kind of orthogenetic evolution in the animal kingdom.—Crustaceana 72:241—264. Mueller, O. E 1776. Zoologiae Danicae Prodromus, seu Animalium Daniae et Norvegiae indigena- rum characteres, nomina et synonyma imprimis popularium. M. Hallager, Havniae, 274 pp. Nicholls, A. G. 1944. Littoral Copepoda from South Australia. If. Calanoida, Cyclopoida, Notodel- phyoida, Monstrilloida and Caligoida.—Re- cords of the Australian Museum 8:1—62. Pallas, P. S. 1766. Elenchus Zoophytorum, Sistens Ge- nerum Adumbrationes Generaliores et Speci- erum Cognitarum Succinctas Descriptiones, cum Selectis Auctorum Synonymis. Hagae- Comitum, 451 pp. Roettger, R., H. Astheimer, M. Spindler, & J. Stein- born. 1972. Okologie von Asterocheres lillje- borgi, eines auf Henricia sanguinolenta paras- itisch lebenden Copepoden.—Marine Biology (Berlin) 13:259—266. Sars, G. O. 1880. Crustacea et Pycnogonida nova in itinere 2do et 3tio expeditionis norvegicae anno 1877 & 78 collecta.—Archiv for Matematik og Naturvidenskab, Christiana 4:427—476 + 60 pls. 1914. Copepoda Cyclopoida. Parts VII & VIII. Cyclopidae (concluded), Ascomyzonti- 691 dae.—An Account of the Crustacea of Norway, with short descriptions and figures of all the species. Bergen Museum, Bergen 6:81—104, pls. 49-64. Scott, T. & A. Scott. 1894. On some new and rare Crustacea from Scotland.—Annals and Maga- zine of Natural History (series 6) 13:137—149. Sewell, R. B. S. 1949. The littoral and semi-parasitic Cyclopoida, the Monstrilloida and Notodel- phyoida.—Scientific Reports of the John Mur- ray Expedition 9:17—199. ShihwG wee Gakisvcitayccab ne Grameen 1O7levA synopsis of Canadian marine zooplankton.— Bulletin of the Fisheries Research Board of Canada 176:1—264. Stephensen, K. 1929. Marine parasitic, semiparasitic and non-pelagic Crustacea Copepoda.—The Zoology of the Faroes 2, part 1, section 30:1— 18. Stimpson, W. M. 1857. On the Crustacea and Echi- nodermata of the Pacific Shores of North Amer- ica, and is pp. 444—532.—Boston Journal of Natural History 6:444—532. Stock, J. 1981. Associations of Hydrocorallia stylas- terina with gall-inhabiting Copepoda Siphon- ostomatoidea from the south-west Pacific. Part II. On six species belonging to four new genera of the copepod family Asterocheridae.—Bijdr- agen Tot De Dierkunde 51:287—312. . 1984. On the presence of gall-inducing Co- pepoda on stylasterine corals. Studies on Co- pepoda II.—Crustaceana supplement 7:377— 380. Thompson, I. C. 1893. Revised report on the Copep- oda of Liverpool Bay.—Transactions of the Liv- erpool Biological Society 7:175—230. Ummerkutty, A. N. P. 1966. Studies on Indian cope- pods. 13. Brief notes on the asterocherid cope- pods obtained from the south east coast of India with description of Indomyzon qasimi n. gen., n. sp. and a discussion of the family Asteroch- eridae.—Crustaceana 11:17—32. Wilson, C. B. 1944. Parasitic copepods in the United States National Museum.—Proceedings of the United States National Museum 94:529—582. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):692—698. 2003. A new Pseudocetopsis species (Siluriformes: Cetopsidae) from Suriname and French Guiana Richard P. Vari, Carl J. Ferraris, Jr., and Philippe Keith (RPV) Department of Systematic Biology—Fishes, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560-0159, U.S.A.; (CJF) Research Associate, Department of Systematic Biology—Fishes, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560-0159, U.S.A.; (PK) Muséum National d’ Histoire Naturelle, Laboratoire d’ichtyologie, 43 rue Cuvier, 75231 Paris cedex 05, France Abstract.—Pseudocetopsis orientale, a new catfish species of small body size of the cetopsid subfamily Cetopsinae is described from several Atlantic-slope rivers of Suriname and French Guiana. The combination of the presence of a dorsal fin with an ossified spinelet, the possession of pectoral- and dorsal-fin spines, the length of the pelvic and pectoral fins, the relative depth of the body, and the alignment of the dorsal and ventral profiles of the portion of the body posterior of the dorsal-fin base serve to distinguish the new species from all other cetopsines. Résumé.—Pseudocetopsis orientale, une nouvelle espece de poisson chat Ce- topsinae de la famille des Cetopsidae, est décrit de plusieurs riviéres du Suri- name et de Guyane frangaise. La combinaison de plusieurs caractéres incluant la présence d’une nageoire dorsale avec un crochet ossifié, la présence d’ épines dorsales et pectorales, la longueur des nageoires pelviennes et pectorales, la hauteur relative du corps et l’alignement des profils dorsaux et ventraux a larriére de la nageoire dorsale permet de distinguer cette espece de toutes les autres Cetopsinae. Recent studies have demonstrated that the species-level diversity among catfishes of the South American subfamily Cetopsinae as defined by de Pinna & Vari (1995) is sig- nificantly higher than the 12 species recog- nized by Burgess (1989), with five addition- al species described by Ferraris & Brown (1991), Lundberg & Rapp Py-Daniel (1994), Ferraris (1996), and Oliveira et al. (2001). Ongoing studies indicate that these 17 spe- cies, nonetheless, represent a significant un- derestimate of the actual species-level diver- sity within the subfamily, and we herein de- scribe a new species from Suriname and French Guiana discovered independently during a revisionary study of the subfamily (RPV, CJF) and faunal studies of the ich- thyofauna of French Guiana (PK). Materials and Methods All measurements were taken as straight line distances between points with head length (HL) measured from the tip of the snout to the end of the fleshy gill cover. Interorbital width was taken as the shortest distance between the orbits, but is, however, difficult to measure unambiguously because of the fleshy tissues around the orbits. The soft bodies of most cetopsins make it dif- ficult to accurately measure most standard morphometric features. As a consequence we do not provide these values for the spe- cies. Size of examined specimens is report- ed as standard length (SL) rounded to the nearest whole millimeter. Median fin ray counts include all elements apparent in ra- diographs. The number of vertebrae was VOLUME 116, NUMBER 3 Pigely Pseudocetopsis orientale, holotype, MHNG 2621.040, 27 mm SL, Suriname, Brokopondo District, Mindrineti Kreek, close to mouth of Maykaboeka Kreek, on the Gros Rosevel Mining concession (5°07'08.8’N, 55°16'59.4"W). taken from radiographs and includes the four elements of the Weberian apparatus and one element for the ural complex. Ver- tebrae were separated into preanal, precau- dal, and caudal elements with total verte- brae the sum of precaudal and caudal ver- tebrae. The presence of the sexually dimor- phic features present in males of other cetopsin species was considered indicative of the sex for those specimens that could not be dissected but with these attributes. The range of values for meristic and mor- phometric features in the species is pre- sented first, followed by the values for the holotype in square brackets. Numbers of specimens is indicated by “‘ex.”’ Institution- al abbreviations are: AMNH, American Museum of Natural History, New York; ANSP, Academy of Natural Sciences, Phil- adelphia; FMNH, Field Museum of Natural History, Chicago; MNHG, Muséum d’ Histoire Naturelle, Geneva; MNHN, Mu- séum National d’ Histoire Naturelle, Paris; MZUSP, Museu de Zoologia da Universi- dade de Sao Paulo; NZCS, National Zoo- logical Collection of Suriname, Paramaribo; and USNM, National Museum of Natural History, Smithsonian Institution, Washing- tom, Dic Pseudocetopsis orientale, new species ere Hemicetopsis sp., Boujard et al., 1990:347 [French Guiana, Fleuve Arataye]. Ponton & Copp, 1997:241 [French Guiana, Fleuve Sinnamary]. Pseudocetopsis cf. minutus Le Bail et al., 2000:146, unnumbered figure [French Guiana, Fleuve Maroni, Fleuve Iracoubo, Fleuve Comté, Fleuve Oyapock]. Holotype.-—MHNG 2621-040, 27 mm SL, Suriname, Brokopondo District, Min- drineti Kreek, close to mouth of Mayka- boeka Kreek, Saramacca River basin, on Gros Rosevel Mining concession (5°07'08.8"N, 55°16'59.4"W), R. Commerg- nat, J. Mol, and C. Weber. Paratypes.—All Suriname, Brokopondo District, Saramacca River basin: MHNG 2626.013, 9 ex., 18-27 mm SL (collected with holotype); USNM 369732, 2 ex., 21- 24 mm SL (collected with holotype); MNHN 2002.1625, 1 ex., 26 mm SL (col- lected with holotype); NZCS F7048, 1 specimen, 23 mm SL (collected with holo- type). MHNG 2621.044, 2 ex., 22—22 mm SL and MHNG 2621.039, 4 ex., 21—26 mm SL, close to mouth of Maykaboeka Kreek, on Gros Rosebel Mining concession (col- lected close to type locality), R. Commerg- nat, J. Mol, and C. Weber. MZUSP 65404, 1 ex., 27 mm SL, Maykaboeka Kreek, Gros Rosebel Area, near Golden Star Concession (5°04'45"N, 55°16'09"W), C.A. Figueiredo, EF Breden, and H. Brook, Jan 2000. Non-type specimens examined.—31 specimens, 18—49 mm SL. FRENCH GUIANA. MNHN 2002-1103, 3 ex., 38-39 mm SL, no specific locality. MNHN 1994-0092, 1 ex., 30 mm SL, Fleuve Arataye, Saut Parare (4°03’N, 52°42'W). MNHN 2002-1099, 2 ex., 38-39 694 mm SL; Crique Boulanger of Fleuve Comté (4°36’N, 52°19'W). MNHN 2002-1100, 1 ex., 22 mm SL; MNHN 2002-1098, 2 ex., 43-49 mm SL, Crique Balatée of Fleuve Maroni (5°29’N, 54°03'’W). MNHN 2002- 1101, 8 ex., 24—44 mm SL, Degrad Florian of Fleuve Iracoubo (5°29'N, 53°33’W). MNHN 2002-1102, 11 ex., 40—49 mm SL, Fleuve Oyapock. SURINAME. All Nickerie District. FMNH 96268, 1 ex., 25 mm SL, Kaiser- berg River (approximately 3°03’N, 56°35'W). AMNH 54828, 1 ex., 22 mm SL, stream near Avanavero, approximately 3 miles (=4.8 km) downstream of DeVis Falls (~4°48'N, 57°26'W). AMNH 55001, 1 ex., 32 mm SL, small stream on road from Lucie River Camp to Paramaribo, 25 km N of Sisa Creek (~3°34'N, 57°37'W). USNM 226146, 2 ex., 22-23 mm SL, woodland stream, 0.5 km from Camp Ma- taway (4°48’N, 57°43’W). USNM 226147, 7 ex., 18-24 mm SL (2 ex. cleared and stained), stream near Camp Anjoemara (4°50'N, 57°26'W). USNM 226148, 1 ex., 30 mm SL, stream at km 212 of road from Amotopo to Camp Geology, at Machine Park—Camp 212 (3°50'N, 57°34'W). Diagnosis.—The combination of the presence of a dorsal fin with an ossified spi- nelet and the presence of pectoral and dor- sal spines distinguishes Pseudocetopsis or- ientale from all other members of the Ce- topsinae with the exception of P. minuta (Eigenmann, 1912) from the Essequibo River basin of Guyana. Pseudocetopsis or- tentale differs from P. minuta in the length of the pelvic fin (tip of fin reaches posteri- orly to anterior margin of vent versus to anal-fin origin, respectively), in the length of the pectoral fin (tip of fin falls distinctly short of vertical through pelvic-fin insertion versus reaches that line), the relative depth of the body (0.21—0.23 of SL versus 0.17— 0.19, respectively), and in the alignment of the dorsal and ventral profiles of the portion of the body posterior of the dorsal fin (con- verging slightly posteriorly versus parallel, respectively). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Description.—Body moderately robust, somewhat compressed laterally anteriorly and becoming progressively distinctly com- pressed posteriorly. Body depth at dorsal- fin origin approximately 0.21—0.23 of SL, and approximately equal to distance from anterior of eye to rear of head. Lateral line on body unbranched, midlateral, incom- plete, and extending from vertical through pectoral-fin base posteriorly to point within region delimited anteriorly by vertical through middle of anal-fin base and poste- riorly by vertical located proximate to an- terior terminus of caudal peduncle. Dorsal profile of body nearly straight to slightly convex from nape to dorsal-fin origin and straight from that point to caudal-fin base. Ventral profile of body slightly convex along abdomen, approximately straight along anal-fin base and converging towards dorsal profile of body posteriorly. Caudal- peduncle depth greater than caudal-pedun- cle length. Head in profile acutely triangular overall with bluntly pointed snout. Dorsal profile of head slightly convex from tip of snout to vertical through posterior margin of orbit and broadly convex from that point to nape. Ventral profile of head slightly convex. Margin of snout in dorsal view broadly rounded. Postorbital margins of head slight- ly convex on each side from dorsal view. Enlarged jaw musculature very evident on dorsal and lateral surfaces of postorbital portion of head. Opercular membrane attached to isthmus only anterior of vertical through pectoral- fin insertion. Opercular opening moderately elongate; opening extending anteroventral of pectoral-fin insertion by distance approx- imately equal to one-third of head length and extending dorsal of pectoral-fin inser- tion by distance equal to width of eye. Eye situated on lateral surface of head; located entirely dorsal to horizontal through pectoral-fin insertion; eye visible in dorsal view, but not in ventral view, of head. Mid- dle of orbit at approximately 0.30 of HL. Eye diameter approximately one-third of VOLUME 116, NUMBER 3 snout length. Interorbital width approxi- mately equal to distance from tip of snout to point within region between middle of eye and posterior margin of orbit. Anterior narial opening circular, surrounded by short, anteriorly directed, tubular rim of skin. Opening of anterior nares located slightly dorsal of horizontal that extends through maxillary-barbel origin and at, or slightly ventral of, horizontal through tip of snout. Distance between anterior nares ap- proximately equal to distance from tip of snout to middle of orbit. Posterior narial opening located on dorsal surface of head, situated dorsal to anterior one-third of orbit. Narial opening nearly round and with flap of skin that nearly encircles opening, but with small gap posteriorly; flap highest an- terolaterally. Mouth subterminal, very wide; its width approximately 0.60 of HL. Margin of lower jaw gently rounded, its posterior limit reaching to vertical through middle of orbit. Premaxillary tooth patch in form of gently arched band continuous across midline; an- terior margin convex, and posterior margin concave and parallel to anterior margin. Teeth on premaxilla small, conical, sharply pointed, and arranged in three regular rows of uniform-sized teeth across entire pre- maxilla. Vomerine teeth in single arched row, with distinct gap in series at midline. Vomerine teeth conical, all of approximate- ly uniform size, and with largest teeth in series approximately same size as largest teeth on premaxilla. Dentary teeth compa- rable in shape to, but slightly larger than, premaxillary teeth. Dentary with three ir- regular tooth rows medially that taper to one row laterally. Maxillary barbel slender, its length ap- proximately equal to length of orbit plus postorbital portion of head, and _ slightly greater than three-quarters of HL; barbel or- igin located ventral to anterior margin of orbit. Medial mental barbel slightly shorter than lateral mental barbel, with latter short- er than maxillary barbel. Medial mental- barbel origin located along vertical through 695 rictus. Lateral mental-barbel origin situated slightly posterior of vertical through medial mental-barbel origin. Tips of adpressed mental barbels extend to, or barely beyond, opercular margin. Dorsal-fin rays 6 [6]. Dorsal fin moder- ate, its base approximately 0.38—0.40 of HL. Longest branched dorsal-fin ray, not including distal filament present in mature males, equal in length to approximately two-thirds of HL. Dorsal-fin spinelet pre- sent, first dorsal-fin ray spinous for basal one-half of length and flexible more distal- ly, with distal filament present in mature males. Distal margin of dorsal fin slightly convex, with first branched ray longest. Dorsal-fin origin located at approximately 0.29—0.33 of SL and along vertical that ex- tends through middle of adpressed pectoral fin. Tip of adpressed dorsal fin, not includ- ing distal filament present in mature males, reaching to vertical through anterior margin of vent. Last dorsal-fin ray with slight basal posterior membranous attachment to body. Caudal-fin rays 1,7,8,1 [1,7,8,1]. Caudal fin deeply forked, lobes symmetrical; tips of lobes rounded. Length of longest caudal-fin ray approximately two times length of mid- dle rays. Anal-fin rays 23 to 27 [26], rarely 27. Anal-fin base moderately long. Anal-fin or- igin located distinctly posterior of middle of SL and anterior of middle of total length. Anal-fin margin nearly straight in most ex- amined specimens, but convex in presumed mature male as evidenced by presence of filamentous dorsal- and pectoral-fin rays. Last anal-fin ray with slight membranous attachment to body. Pelvic-fin rays 1,5 [1,5]. Pelvic fin small; distal margin slightly convex with middle rays longest. Pelvic-fin insertion located an- terior to middle of SL and along vertical through posterior limit of dorsal-fin base. Tip of adpressed pelvic fin extending past middle of SL and reaching anterior margin of vent. Last pelvic-fin ray with membra- nous attachment to body along basal one- half of its length. 696 Pectoral-fin rays 7 or 8 [7], rarely 8. Pec- toral fin moderately long, its length slightly more than 0.60 of HL. Pectoral-fin margin distinctly convex, with middle ray longest. First pectoral-fin ray spinous with smooth margins, spine short with length slightly more than one-half that of first branched ray; ray prolonged as filament in presumed mature male specimens. Preanal vertebrae 12 to 15 [15]; precau- dal vertebrae 10 to 13 [11]; caudal verte- brae 27 to 29 [28]; total vertebrae 39 to 42 [39], with 39 and 40 vertebrae most com- mon, and 42 vertebrae in only | of 54 ra- diographed specimens (total includes radio- graphed non-type specimens). Ribs 8 to 11 [10]. Coloration in alcohol.—Overall ground coloration of head and body pale and over- lain with rounded, large, brown chromato- phores. Dark pigmentation on head and body tends to be more concentrated dorsal- ly. Expanded chromatophores blend togeth- er to form uniform brown cast on some por- tions of body in most specimens. Ventral surface of abdomen and head pale except for scattered chromatophores on abdomen in some specimens and broad band of scat- tered, dark chromatophores that extend from symphysis of lower jaw to opercle. Dorsal fin pale with some dark pigmen- tation basally that forms somewhat diffuse spot with margin of spot in form of semi- circle. Anal fin pale with scattered dark chromatophores basally. Pelvic and pectoral fins pale. Caudal fin with few scattered dark chromatophores that extend distally to at least middle of fin rays. Maxillary barbel with scattered dark pig- mentation basally and pale distally. Mental barbels pale. Coloration in life.—Le Bail et al. (2000: 147) provided a photograph of a live spec- imen of the species (identified therein as Pseudocetopsis cf. minutus), photographed in an aquarium immediately after capture. The overall dark pigmentation visible in the photograph is comparable to that in the pre- served specimens, but the head, body, and PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON fins of the specimen have an overlying sil- very sheen that is absent in preserved ma- terial. Sexual dimorphism.—The presumed ma- ture males of P. orientale have filaments present on the dorsal and pectoral fins and have the anal-fin margin slightly convex. Juveniles and females of the species, in contrast, lack filaments on the fins and have a straight anal-fin margin. Distribution.—Pseudocetopsis orientale is known from the Atlantic coastal rivers of Suriname and French Guiana, in the region from the Corantijn River, that forms the border between Suriname and Guyana, to the Fleuve Oyapock-Rio Oiapoque along the French Guiana-Brazil border. Given the presence of this species in the Surinamese tributaries of the Corantijn River, it likely also occurs in the left-bank tributaries to that river in Guyana. Similarly it is likely that the species also occurs in the portions of the Rio Oiapoque basin within Brazil. Etymology.—The species name, orienta- le, from the Latin for eastern, refers to the distribution of this species in the eastern most portions of the known distribution of the Cetopsinae. Ecology.—The type locality is a rainfo- rest stream bordered by overhanging vege- tation that at the time of the collection had low (40 cm deep), clear, slowly-moving wa- ter over a sand bottom with mud along the banks. The holotype and paratypes captured with it came from holes and fissures in de- caying branches submerged along the side of the stream (C. Weber, MHNG, pers. comm.). Mol et al. (2000:430) characterize the Maykaboeka Kreek as a low-gradient, second order drainage running through un- disturbed rainforest. Non-type specimens from Degrad Florian in the Fleuve Iracoubo basin, French Guiana (MNHN 2002-1101), were captured in a stream approximately 3 m wide and 10—59 cm deep in clear, but slightly tea-colored waters in areas with slow current densely shaded by the gallery forest. Observations at that locality indicate that during the day Pseudocetopsis orien- VOLUME 116, NUMBER 3 tale is hidden in the sediment under leaves or roots (PK). Remarks.—The question of generic lim- its within the Cetopsinae has been dis- cussed by various authors in recent years including Ferraris & Brown (1991), Lund- berg & Rapp Py Daniel (1994), and Ferraris (1996). Ongoing phylogenetic studies indi- cate that a revamping of some generic def- initions within the Cetopsinae may be nec- essary, but in the interim we assign the new species to Pseudocetopsis in keeping with the definition of that genus proposed by Ferraris & Brown (1991). Pseudocetopsis minutus, which was orig- inally described by Eigenmann (1912:211) as Hemicetopsis minutus from the Essequi- bo River system, has been reported by sub- sequent authors from a series of localities distant from that drainage basin including French Guiana (see synonymy above), the Rio Trombetas in the Amazon basin in Bra- zil (Ferreira, 1995:52), the Rio Caroni in Venezuela (Taphorn & Garcia Tenia, 1991, fig. 3; Lasso et al., 1990:117) and other riv- ers in that country (Machado-Allison et al., 1993:65, as Hemicetopsis minimus). Ongo- ing studies indicate that P. minutus is rather apparently endemic to the Essequibo River basin of Guyana and our comparisons here- in of P. orientale are limited to material of P. minutus from that river basin. Comparative material examined.—Pseu- docetopsis minuta: British Guiana (=Guy- ana) Amatuk (5°18’N, 59°18’W), FMNH 53262, 1 (18 mm SL, holotype of Hemi- cetopsis minutus). Siparuni VIII-2; Esse- quibo River at Essequibo campsite (04°45'41"N, 58°45'53"W), ANSP 175839, IG@ tim SIE): Acknowledgments Research and museum visits associated with the preparation of this paper were made possible by funding from the Neo- tropical Lowlands Research Program of the Smithsonian Institution. We thank Scott A. Schaefer and Barbara Brown, AMNH; John 697 L. Lundberg and Mark Sabaj, ANSP; Mary Anne Rogers, FMNH; Jean-Claude Hureau and Patrice Pruvost, MNHN; Claude Weber, MHNG; Carlos A. Figueiredo and Mario C.C. de Pinna, MZUSP; for access to or the loan of specimens that made this paper pos- sible. Claude Weber also provided infor- mation on the type locality. Sandra J. Ra- redon (USNM) provided technical support for the project and Pierre-Yves Le Bail, In- stitut National de la Recherche Agrono- mique, provided critical assistance during collecting efforts in French Guiana. Figure 1 was prepared by T. Britt Griswold. This paper benefitted from comments and sug- gestions from Marcelo R. Britto, Ricardo M.C. Castro, and Angela Zanata. Literature Cited Boujard, T., EF J. Meunier, M. Pascal, & J. E Cosson. 1990. Les Téléostéens d’un haut bassin fluvial Guyanais, |’Arataye. 2—Inventaires des “‘non- Characoides.”” Cybium, 14(4):345-351. Burgess, W. E. 1989. An atlas of freshwater and ma- rine catfishes. A preliminary survey of the Sil- uriformes. T.-EH. Publications, Neptune City, New Jersey, U.S.A., 784 pp. Eigenmann, C. H. 1912. The freshwater fishes of Brit- ish Guiana, including a study of the ecological grouping of species and the relation of the fauna of the plateau to that of the lowlands.—Mem- oirs of the Carnegie Museum, 5:xii + 578 pag- es, 103 plates. Ferraris, C. J., Jr. 1996. Denticetopsis, a new genus of South American whale catfish (Siluriformes: Cetopsidae, Cetopsinae), with two new spe- cies.—Proceedings of the California Academy of Sciences 49(6):161—170. , & B. A. Brown. 1991. A new species of Pseu- docetopsis from the Rio Negro drainage of Ve- nezuela (Siluriformes: Cetopsidae).—Copeia 1991(1):161—165. Ferreira, E. 1995. Composicao, distribuigao e aspectos ecol6gicos da ictiofauna de um trecho do rio Trombetas, na area de influéncia da futura UHE Cachoeira Porteira, Estado do Para, Brasil_— Acta Amazonica 23(1—4, Suplemento; issue for 1993, published 1995):1—88. Lasso, C. A., A. Machado-Allison, & R. Pérez Her- nandez. 1990. Consideraciones zoogeograficas de los peces de la Gran Sabana (Alto Caroni) Venezuela, y sus relaciones con las cuencas ve- cinas.—Memoria Sociedad de Ciencias Natur- ales La Salle 50(133—134):109—129. 698 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Le Bail., P.-Y., P. Keith, & P. Planquette. 2000. Atlas des poissons d’eau douce de Guyane. Tome 2. Fascicule II. Siluriformes. Muséum National d’Histoire Naturelle, Service du Patronomie Naturel, Paris, and Institut d’ Ecologie et de Gestion de la Biodiversité, Paris. 307 pp. Lundberg, J. G., & L. Rapp-Py-Daniel. 1994. Bathy- cetopsis oliveirai, gen. et sp. nov., a blind and depigmented catfish (Siluriformes: Cetopsidae) from the Brazilian Amazon.—Copeia 1994(2): 381-390. Machado-Allison, A., C. Lasso, & R. Royero-Le6n. 1993. Inventario preliminar y aspectos ecolo- gicos de los peces de los Rios Aguaro y Guar- iquito (Parque Nacional), Estado Guarico, Ve- nezuela.—Memoria Sociedad de Ciencias Na- turales La Salle 53(139):55—80. Mol, J. H., D. Resida, J. C. Ramlal, & C. R. Becker. 2000. Effects of El Nino-related drought on freshwater and brackish-water fishes in Surina- me, South America.—Environmental Biology of Fishes 59(4):429—440. Oliveira, J. C. de, R. P. Vari, & C. J. Ferraris, Jr. 2001. A new species of ““Whale catfish’? (Silurifor- mes: Cetopsidae) from the western portions of the Amazon basin.—Proceedings of the Biolog- ical Society of Washington 114(3):574—578. de Pinna, M. C. C., & R. P. Vari. 1995. Monophyly and phylogenetic diagnosis of the family Cetop- sidae, with synonymization of the Helogenidae (Teleostei: Siluriformes).—Smithsonian Contri- butions to Zoology 51:1—26. Ponton, D., & G. H. Copp. 1997. Early dry-season community structure and habitat use of young fish in tributaries of the River Sinnamary (French Guiana, South America) before and af- ter Hydrodam operation.—Environmental Bi- ology of Fishes 50:235—256. Taphorn, D. C., & J. G. Garcia Tenia. 1991. El Rio Claro y sus peces, con consideraciones de los impactos ambientales de las presas sobre la ic- tiofauna del bajo Rio Caroni.—Biollania 8:23— 45. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):699—709. 2003. Chromosomes of Philippine mammals (Insectivora, Dermoptera, Primates, Rodentia, Carnivora) Eric A. Rickart Utah Museum of Natural History, 1390 E Presidents Circle, University of Utah, Salt Lake City, Utah 84112, email: rickart@umnh.utah.edu Abstract.—Karyotypes of nine species of Philippine mammals representing five orders are presented. Chromosomes of six species are described for the first time, including three endemic insectivores (Podogymnura truei, Crocidura beatus, C. grayi), the endemic dermopteran (Cynocephalus volans), an endemic squirrel (Sundasciurus philippinensis), and a widespread viverrid (Viverra tan- galunga). Some species endemic to the oceanic Philippines have unique kar- yotypes whereas other endemics and widespread Asian species have karyotypes that are similar, or identical, to those of related species or conspecifics occurring outside the archipelago. These data corroborate patterns of karyotype variation previously documented for Philippine bats and murid rodents. For more than a century, the terrestrial mammal fauna of the Philippines has been recognized as a unique assemblage (Thom- as 1898). All portions of this remarkable fauna, which includes 179 species repre- senting nine orders, are characterized by a high proportion of endemic species. More than 60% of the mammal fauna, as a whole, is endemic (Heaney et al. 1998). For bats and murid rodents, the two largest faunal components, cytogenetic studies have pro- vided insight into how these groups have diversified within the archipelago (Rickart et al. 1989, Rickart & Musser 1993, Rickart et al. 1999, Rickart & Heaney 2002). These studies suggest that chromosomal data on other Philippine taxa may help clarify their phylogenetic and biogeographic relation- ships. Accordingly, this report presents kar- yotypes of nine species representing five or- ders (Insectivora, Dermoptera, Primates, Rodentia, and Carnivora). Methods Specimens were collected during field studies conducted in 1987 (on Leyte, Bili- ran and Negros islands), 1988 (Luzon Is- land), and 1993 (Mindanao Island). Live- trapped animals were processed and killed with sodium pentobarbital within 24 h of capture, and karyotypes prepared from bone marrow and/or spleen cells following in vivo methodology (Patton 1967, Rickart et al. 1989). Material from freshly killed an- imals caught in snap traps was processed in vitro (Rickart et al. 1998). Cells were pro- cessed and fixed in the field, and standard (non-differentially stained) karyotypes were prepared from stored cell suspensions. Preparations of silver-stained nucleolus or- ganizer regions (Ag-NORs; Howell & Black 1980) and G-banded chromosomes (Seabright 1971) were made for some taxa. A minimum of 10 chromosomal spreads was examined from each preparation. Chro- mosome terminology follows Rickart and Musser (1993). As used herein, fundamen- tal number (FN) refers to the total number of chromosome arms in the female karyo- type (including those of sex chromosomes). Species nomenclature follows Heaney et al. (1998). Voucher specimens were prepared as skins with partial skeletons, complete skel- etons, or preserved in fluid, and are depos- 700 ited in the Field Museum of Natural His- tory, Chicago, IL (FMNH), the National Museum of Natural History, Smithsonian Institution, Washington, DC (USNM), and the Philippine National Museum, Manila (PNM). Microscope slides of chromosomal preparations and photomicrograph nega- tives cross-referenced to cataloged voucher specimens are housed at the Utah Museum of Natural History, University of Utah, Salt Lake City. Specimens Examined Podogymnura truei Mearns, 1905.— Mindanao Island, Bukidnon Province, Mount Kitanglad Range, 16.5 km S, 4 km E Camp Phillips, elev. 1900 m, 8°10'30’N, 124°51’E, 2 males (FMNH 147793, 147798). Crocidura beatus Miller, 1910.—Leyte Island, Leyte Province, Mount Pangasugan, 10 km N, 4.5 km E Baybay, elev. 950 m, 10°47'N, 124°50’E, 1 male (USNM 457984). Crocidura grayi Dobson, 1890.—Luzon Island, Camarines Sur Province, Mount Is- arog, 4 km N, 22 km E Naga, elev. 1750 m, 13°40'N, 123°22’E, 1 female, 1 male (USNM 573607, 573617). Suncus murinus (Linnaeus, 1766).—Ne- gros Island, Negros Oriental Province, Sil- liman Farm, Dumaguete, elev. 5 m, O9°18'’N, 123°18’E, 2 males (USNM 457996, 457997); Mount Guinsayawan, 3 km N, 17 km W Dumaguete, elev. 1470 m, 9°22'N, 123°9’E, 1 female, 1 male (USNM 458970, 458972). Cynocephalus volans (Linnaeus, 1758).—Leyte Island, Leyte Province, 7 km Nenbaybaya mclev O> my 045 0Ne 124°47'30"E, 1 female (USNM 458982). Tarsius syrichta (Linnaeus, 1758).— Leyte Island, Leyte Province, 7 km N, 1.5 km E Baybay, elev. 50 m, 10°45’'N, 124°48’E, 1 male (PNM specimen, EAR field number 1441); Mount Pangasugan, 10.2 km N, 2.2 km E Baybay, elev. 320 m, 10°46’N, 124°49'E, 1 male (USNM PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 459818); Mount Pangasugan, 8.5 km N, 2.5 km E Baybay, elev. 500 m, 10°45'30’N, 124°49'30’E, 1 male (USNM 458723). Sundasciurus philippinensis (Water- house, 1839).—Biliran Island, Leyte Prov- ince, 5 km N, 10 km E Naval, elev. 850 m, 11°36’N, 124°29’E, 1 male (USNM 459821). Paradoxurus hermaphroditus (Pallas, 1777).—Leyte Island, Leyte Province, 7 km N, 1.5 km E Baybay, elev. 50 m, 10°45’'N, 124°48’E, 1 male (USNM 458891); Mount Pangasugan, 10 km N, 2 km E Baybay, elev. 300 m, 10°46’N, 124°49’E, | male (USNM 459999). Viverra tangalunga Gray, 1832.—Leyte Island, Leyte Province, Mount Pangasugan, 10 km N, 2 km E Baybay, elev. 300 m, 10°46’N, 124°49’E, 1 male (USNM 460000). Results Order Insectivora Family Erinaceidae Podogymnura truei. 2N = 40, FN = 76, Fig. 1A.—The karyotype of the Mindanao gymnure includes 11 pairs of small to large- sized metacentric or submetacentric auto- somes, 6 pairs of medium-sized subtelocen- tric autosomes, and 2 pairs of small or me- dium-sized telocentric autosomes. The small X chromosome is submetacentric, and the minute Y chromosome appears to be telocentric. Family Soricidae Crocidura grayi. 2N = 38, FN = 58, Fig. 1B.—The karyotype of the Luzon shrew in- cludes 4 pairs of small to medium-sized submetacentric autosomes, 5 pairs of small to large-sized subtelocentric autosomes, and 9 pairs of small to large-sized telocentric autosomes. The X chromosome is medium- sized and subtelocentric, and the small Y chromosome is submetacentric. Crocidura beatus. 2N = 38(?), FN = ?, not figured.—A poor quality in-vivo prep- A. >) AR KG WK Aa MM RM KA SR KR ae AR OR AL OH na na BMS XK KK SK XK hil fi an 20 ste RN OH NH AD ANNO fh AHA A KR Xx AN na | AbD AA AAan , , Fig. 1. Karyotypes of: ymnura truei, male (FMNH 147793) from in vitro preparation, 2N = 40 B) Crocidura grayi, mal , 2N = 38; C) Suncus murinus, female (USNM 458970), 2N 40 702 aration yielded preliminary information on the chromosomes of this species which is endemic to the Mindanao faunal region. The modal chromosome count from multi- ple spreads was 38, indicating a karyotype similar to that of C. grayi (Fig. 1B). Suncus murinus. 2N = 40, FN = 60, Fig. 1C.—Specimens of the Asian house shrew from Negros Island have a karyotype that includes 4 pairs of small to large-sized sub- metacentric autosomes, 5 pairs of small to medium-sized subtelocentric autosomes, and 10 pairs of medium to large-sized telo- centric autosomes. Both the X and Y chro- mosomes are submetacentric. Order Dermoptera Family Cynocephalidae Cynocephalus volans. 2N = 38, FN = 40, Fig. 2.—The standard karyotype of a female Philippine flying lemur from Leyte Island includes 18 pairs of small to large- sized telocentric autosomes and a pair of medium-sized submetacentric X chromo- somes. A G-banded preparation reveals G- positive bands on the nine largest auto- somes. The X chromosome is largely G- positive. Ag-NORs are located terminally on the smallest telocentric autosomes. Order Primates Family Tarsiidae Tarsius syrichta. 2N = 80, FN = 94, Fig. 3A.—Karyotypes of male Philippine tarsi- ers from Leyte Island include 6 pairs of small to large-sized metacentric or sub- metacentric autosomes and 33 pairs of small to large-sized telocentric autosomes. The submetacentric X and minute Y chro- mosomes are, respectively, the largest and smallest elements in the karyotype. Order Rodentia Family Sciuridae Sundasciurus philippinensis. 2N = 38, FN = 72, Fig. 3B.—The karyotype of a male Philippine tree squirrel from Biliran PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Island includes 9 pairs of metacentric or submetacentric autosomes, 7 pairs of sub- telocentric autosomes, and 2 pairs of small telocentric autosomes. The medium-sized X chromosome is submetacentric and the mi- nute Y chromosome appears to be telocen- tric. Order Carnivora Family Viverridae Paradoxurus hermaphroditus. 2N = 42, FN = 72, Fig. 4A.—Specimens of the com- mon palm civet from Leyte Island have a karyotype that includes 8 pairs of small to large-sized metacentric or submetacentric autosomes, 6 pairs of small to large-sized subtelocentric autosomes, and 6 pairs of small to medium-sized telocentric auto- somes. The medium-sized X chromosome is submetacentric, and the small Y chro- mosome is telocentric. Viverra tangalunga. 2N = 36, FN = 64, Fig. 4B.—The karyotype of a male Malay civet from Leyte Island includes 7 pairs of small to large-sized metacentric or sub- metacentric autosomes, 6 pairs of small to large-sized subtelocentric autosomes, and 4 pairs of small telocentric autosomes. The X chromosome is large and submetacentric and the small Y chromosome is telocentric. Discussion The 2N = 40, FN = 76 karyotype of Podogymuura truei (Fig. 1A) is the first re- ported for the gymnure subfamily Hylo- myinae. It differs substantially from those of hedgehogs (subfamily Erinaceinae) which have relatively uniform karyotypes of 2N = 48, FN = 90-96 and X chromo- somes that are substantially larger than that of P. truei (Gropp 1969, Bhatnagar & El- Azawi 1978, Hiibner et al. 1991, Reumer & Meylan 1986). The widespread genus Crocidura dis- plays extensive interspecific chromosomal variation (Maddalena & Ruedi 1994, Zima ct Tales 1998), ilitere) are jseventspeciesson Crocidura in the Philippines (Heaney & VOLUME 116, NUMBER 3 A BA aban aan 6O AR AR ON AA an Fig. 2. Karyotypes of Cynocephalus volans, female (USNM 458982), 2N = 38: A) standard, inset of Ag- o<2ge>s> i > 5>D>s2zS A RA AA AA AA me AA An ae pe VOLUME 116, NUMBER 3 A 705 & KK AN AA Ax BR AAR Ad AN DN AA 44 Amnah nea B A3 MR AE “uk OG AA Oh AR AR AG OR KQ@ Xk RE bh hO aA Fig. 4. Karyotypes of: A) Paradoxurus hermaphroditus, male (USNM 458891), 2N = 42; B) Viverra tan- galunga male (USNM 460000), 2N = 36. Ruedi 1994) of which two have been kar- yotyped (this study). The 2N = 38, FN = 58 karyotype of C. grayi (Fig. 1B) resem- bles those of several species with 2N = 38, FN = 54-58 from islands of the Sunda Shelf and from Sulewesi (Ruedi & Vogel 1995). It also is similar to the presumed an- cestral arrangement for Crocidura (Mad- dalena & Ruedi 1994). The apparent simi- larity in the karyotypes of C. grayi and C. beatus is consistent with morphometric and allozyme data that place these species as sister-taxa (Heaney & Ruedi 1994). The limited data available suggest that the di- versification of Philippine Crocidura has not involved major chromosomal rearrange- ments of the sort seen for members of the genus in Sulewesi (Ruedi & Vogel 1995), or for Philippine rodents of the genus Apo- mys (Rickart & Musser 1993, Rickart & Heaney 2002). Suncus murinus is a chromosomally polymorphic species (2N = 30—40) found throughout much of southeast Asia (Yosh- ida 1985... Zima-et! al. 1998)>. It oceuts throughout the Philippines as a non-native 706 commensal (Heaney et al. 1998), and on Negros Island, it also is naturalized and abundant in primary forest habitat (Heaney et al. 1989). Both commensal and natural- ized animals from Negros have identical karyotypes of 2N = 40, FN = 60 (Fig. 1C). This same arrangement has been reported for specimens from much of southeast Asia (Zima et al. 1998). Medina and Leonard (1977) reported a karyotype of 2N = 40, FN = 54 for S. luzoniensis (a synonym of S. murinus) from Luzon Island. The differ- ence in FN compared to the Negros speci- mens most likely reflects variable assess- ment of minute secondary arms on subtel- ocentric autosomes. The 2N = 38, FN = 40 karyotype of Cynocephalus volans (Fig. 2), one of two members of the mammalian order Dermop- tera, is reported here for the first time. This also corrects an erroneous report by Hsu and Benirschke (1973). They published identical karyotypes of 2N = 56, FN = 72 for a specimen of Galeopterus variegatus from Malaysia and for an animal at the Lin- coln Park Zoo, Chicago, originally identi- fied as C. volans. After the latter died, it was sent to the Field Museum where it was accessioned and correctly identified as Gal- eopterus (FMNH 60308; W. Stanley, pers. comm.). The two dermopterans have sub- stantially different karyotypes. Cynocephal- us lacks nine pairs of autosomes present in Galeopterus, including seven pairs of small to large-sized biarmed chromosomes, and the X chromosome of the former is sub- metacentric whereas that of the latter is metacentric (Fig. 2; Hsu & Benirschke 1973). The Ag-NORs of Cynocephalus are located on the smallest autosomal pair (Fig. 2). In Galeopterus, this same pair bears sec- ondary constrictions (Hsu & Benirschke 1973). This chromosomal distinctiveness is in accord with the substantial morphologi- cal and ecological differences that suggest ancient divergence of the two species and supports their placement in separate genera (Stafford & Szalay 2000). The karyotype of Tarsius syrichta from PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Leyte Island (Fig. 3A) is identical to those previously reported for specimens from Mindanao (Boer & Boer-Van der Vlist 1973, Dutrillaux & Rumpler 1988). Tarsius bancanus also has 2N = 80, FN = 94 (Klinger 1963), but its karyotype differs from that of 7. syrichta in the size and rel- ative arm lengths of several of the biarmed autosomes. Tarsius dianae from central Su- lawesi is the only other tarsier that has been karyotyped. Its arrangement of 2N = 46, FN = 82 (Niemitz et al. 1991) differs from those of 7. syrichta and T. bancanus by multiple Robertsonian and non-Robertso- nian events. Chromosomal data support phylogenetic hypotheses based on compar- ative morphology and behavior that asso- ciate the Philippine and Sundaic species (T. syrichta and T. bancanus) as relatively spe- cialized forms separate from the three spe- cies of Sulawesian tarsiers (Niemitz 1977, Musser & Dagosto 1987, Dagosto et al. 2001). The 2N = 38, FN = 72 karyotype of Sundasciurus philippinensis (Fig. 3B) 1s re- ported here for the first time. It is identical to that of S. jentinki, the only other member of the genus that has been karyotyped (Har- ada & Kobayashi 1980). Among callosci- urine squirrels, Dremomys rufigenis and Callosciurus albescens also share this ar- rangement (Nadler & Hoffmann 1970, Har- ada & Kobayashi 1980), and several other species of Callosciurus have karyotypes of 2N = 40, FN = 70-74 that differ only slightly from that of S. philippinensis (Nad- ler et al. 1975, Yong Hoi-Sen et al. 1975, Oshida & Yoshida 1999). The available data suggest that callosciurines are chro- mosomally conservative. A karyotype of 2N = 42, FN = 78 was reported previously for Paradoxurus her- maphroditus from India (Ray-Chaudhuri et al. 1966). The karyotype of Philippine specimens (Fig. 4A) is similar, but appears to have fewer subtelocentric and more telo- centric elements (FN = 72). In other re- spects, the two karyotypes are identical. The 2N = 36, FN = 64 karyotype of VOLUME 116, NUMBER 3 Viverra tangalunga (Fig. 4B) is the first re- ported for this widespread southeast Asian species. Viverra zibetha, the only other member of this genus that has been exam- ined has a substantially different karyotype of 2N = 38 and FN = 68 (Pathak 1971). Among members of the subfamily Viverri- nae that have been examined, Viverricula indica is the only species with 2N = 36 (Wurster & Benirschke 1968). However, Viverra tangalunga has more telocentric and fewer subtelocentric autosomes, and a significantly smaller Y chromosome than does Viverricula. Interpretation of these chromosome data is limited, in some cases, by the lack of comparative information on related taxa. Nonetheless, some general patterns of var- iation are apparent. Members of the endem- ic genera Podogymnura and Cynocephalus have karyotypes that differ substantially from those of related taxa. In contrast, en- demic species of Tarsius, Crocidura, and Sundasciurus, and the widespread species Suncus murinus and Paradoxurus herma- phroditus have karyotypes that are similar, or identical, to those of closely related spe- cies or conspecifics occurring elsewhere in Asia. These results mirror patterns of chro- mosomal variation documented for Philip- pine bats and murid rodents (Rickart et al. 1999, Rickart & Heaney 2002). Most en- demic taxa have karyotypes that are unique compared to relatives occurring outside of the Philippines. Some widespread species exhibit chromosomal polymorphism, in which Philippine populations have unique karyotypes. The general pattern is one of chromosomal rearrangement occurring as a result of isolation. The degree to which this pattern is expressed appears to be largely a function of how long a particular group has been present within the archipelago. Acknowledgments I thank N. Antoque, R. Fernandez, S. Goodman, L. Heaney, P Heideman, J. Klompen, D. Samson, D. Schmidt, B. Ta- 707 baranza, L. Tag-at, and R. Utzurrum for as- sistance with field work. W. Stanley (FMNH) provided information on the iden- tity and history of the dermopteran speci- men formerly at Lincoln Park Zoo. Com- ments from L. Heaney and two anonymous reviewers helped improve the manuscript. Permits and logistical support were provid- ed by the Philippine Protected Areas and Wildlife Bureau and the Philippine Bureau of Forestry Development. Field work was supported by grants from the National Sci- ence Foundation (BSR 8514223), the John D. and Catherine T. MacArthur Foundation (90-09272A), and the Barbara Brown and Ellen Thorne Smith funds of the Field Mu- seum. Literature Cited Bhatnagar, A. N., & T. EK El-Azawi. 1978. A karyotype study of two species of hedgehogs, Hemiechin- us auritus and Paraechinus aethiopicus (Insec- tivora: Mammalia).—Cytologia 43:53—59. Boer, L. E. M. de, & J. de Boer-Van der Vlist. 1973. The somatic chromosomes and the idiogram of Tarsius syrichta carbonarius Huede, 1898 (Pri- mates: Tarsioidea)—Genen en Phaenen 16:65— HA. Dagosto, M., D. L. Gebo, & C. Dolino. 2001. Posi- tional behavior and social organization of the Philippine tarsier (Tarsius syrichta).—Primates 42:233-243. Dobson, G. E. 1890. Descriptions of new species of Crocidura.—Annals and Magazine of Natural History, series 6, 6:494—497. Dutrillaux, B., & Y. Rumpler. 1988. Absence of chro- mosomal similarities between tarsiers (Tarsius syrichta) and other primates.—Folia Primato- logica 50:130—133. Gray, J. E. 1832. On the family of Viverridae and its generic sub-divisions, with an enumeration of the species of several new ones.—Proceedings of the Committee of Science and Correspon- dence of the Zoological Society of London 1832(2):63-68. Gropp, A. 1969. Cytologic mechanisms of karyotype evolution in insectivores. Pp. 247—266 in Com- parative mammalian cytogenetics. Springer, Berlin, 473 pp. Harada, M., & T. Kobayashi. 1980. Studies on the small mammal fauna of Sabah, east Malaysia II. Karyological analysis of some Sabahan mammals (Primates, Rodentia, Chiroptera).— 708 Contributions from the Biological Laboratory, Kyoto University 26:83-95. Heaney, L. R., & M. Ruedi. 1994. A preliminary anal- ysis of biogeography and phylogeny of Croci- dura from the Philippines. Pp. 357-377 in J. Meritt, G. Kirkland, & R. Rose, eds., Advances in the biology of shrews. Special Publication 18, Carnegie Museum of Natural History, Pitts- burgh, 458 pp. , P. D. Heideman, E. A. Rickart, R. C. B. Ut- zurrum, & J. S. H. Klompen. 1989. Elevational zonation of mammals in the central Philip- pines.—Journal of Tropical Ecology 5:259— 280. , et al. 1998. A synopsis of the mammalian fauna of the Philippine Islands.—Fieldiana: Zo- ology, n.s. 88:1—61. Howell, W. M., & D. A. Black. 1980. Controlled silver staining of neucleolar organizer regions with a protective colloidal developer: a 1-step meth- od.—Experientia 36:1014—1015. Hsu, T. C. & K. Benirschke. 1973. Cynocephalus var- iegatus, C. volans (flying lemur) 2n = 56. An atlas of mammalian chromosomes. Vol. 7, Folio 303, Springer-Verlag, New York, unpaged. Hiibner, R., T. Maddalena, & W. Poduschka. 1991. The karyotype of the middle-African hedgehog Afe- lerix albiventris Wagner, 1841 and its cytotax- onomical relationships to other Erinaceinae (In- sectivora: Erinaceidae).—Genetica 83:243—246. Klinger, H. P. 1963. The somatic chromosomes of some primates (Tupaia glis, Tarsius bancanus, Cercocebus aterrimus, Symphalangus syndac- tylus).—Cytogenetics 2:140—151. Linnaeus, C. 1758. Systema naturae per regna tria na- turae, secundum classes, ordines, genera, spe- cies, cum characteribus, differentiis, synonymis, locis, 10th edition. Laurentius Salvius, Stock- holm 1:1—824. . 1766. Systema naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis synonymis, locis, 12th edition. Laurentius Salvius, Stockholm 1: 1-532. Maddalena, T., & M. Ruedi. 1994. Chromosomal evo- lution in the genus Crocidura (Insectivora: Sor- icidae). Pp. 335-344 in J. Merritt, G. Kirkland, & R. Rose, eds., Advances in the biology of shrews. Special Publication 18, Carnegie Mu- seum of Natural History, Pittsburgh, 458 pp. Mearns, E. A. 1905. Description of some new genera and species of mammals from the Philippine Is- lands.—Proceedings of the United States Na- tional Museum 28:425—460. Medina, EF I. S., & A. Leonard. 1977. Observations sur les chromosomes de Suncus luzoniensis Pe- ters.—Acta Zoologica et Pathologica Antver- piensia 69:183-188. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Miller, G. S., Jr. 1910. Descriptions of two new genera and sixteen new species of mammals from the Philippine Islands.—Proceedings of the United States National Museum 38:391—404. Musser, G. G., & M. Dagosto. 1987. The identity of Tarsius pumilus, a pygmy species endemic to the montane mossy forests of central Sulawe- siimAmerican Museum Novitates 2867:1—53. Nadler, C. EF, & R. S. Hoffmann. 1970. Chromosomes of some Asian and South American squirrels (Rodentia: Sciuridae).—Experientia 26:1383- 1386. : , & M. E. Hight. 1975. Chromosomes of three species of Asian tree squirrels Callos- ciurus (Rodentia: Sciuridae).—Experientia 31: 166—167. Niemitz, C. 1977. Zur Funktionsmorphologie und Biometrie der Gattung Tarsius Storr, 1780. Her- leitung von Evolutionsmechanismen bei einem Primaten.—Courier der Senckenbergischen Na- turforschenden Gesellschaft 25:1—161. , A. Nietsch, S. Warter, & Y. Rumpler. 1991. Tarsius dianae: a new primate species from central Sulawesi (Indonesia).—Folia Primato- logica 56:105—116. Oshida, T., & M. C. Yoshida. 1999. Chromosomal lo- calization of nucleolus organizer regions in eight Asian squirrel species.—Chromosome Science 3:55—58. Pallas, P. S. 1777. Pg. 426 in Schreber, J. C. D. von. 1774-1785. Die Saugethier in Abbildungen nach der Natur mit Beshreibungen. Erlangen Wolfgang Walden, 1112 pp. Pathak, S. 1971. Karyotype of the Oriental Civet, Viv- erra zibetha, Linn. (Viverridae-Carnivora).— Mammalian Chromosomes Newsletter 12(2):61. Patton, J. L. 1967. Chromosome studies of certain pocket mice, genus Perognathus (Rodentia: Heteromyidae).—Journal of Mammalogy 48: 27-37. Ray-Chaudhuri, S. P., P. V. Ranjini, & T. Sharma. 1966. Somatic chromosomes of the common palm civet, Paradoxurus hermaphroditus (Viverri- dae—Carnivora).—Experientia 22:740—741. Reumer, W. F, & A. Meylan. 1986. New developments in vertebrate cytotaxonomy IX: chromosome numbers in the order Insectivora (Mamma- lia).—Genetica 70:119—-151. Rickart, E. A., & L. R. Heaney. 2002. Further studies on the chromosomes of Philippine rodents (Mu- ridae: Murinae).—Proceedings of the Biological Society of Washington 115:473—487. , & G. G. Musser. 1993. Philippine rodents: chromosomal characteristics and their signifi- cance for phylogenetic inference among 13 spe- cies (Rodentia: Muridae: Murinae).—American Museum Novitates 3064: 1—34. , L. R. Heaney, & M. J. Rosenfeld. 1989. Chro- VOLUME 116, NUMBER 3 mosomes of ten species of Philippine fruit bats (Chiroptera: Pteropodidae).—Proceedings of the Biological Society of Washington 102:520— Soup , J. A. Mercier, & L. R. Heaney. 1999. Cyto- geography of Philippine bats——Proceedings of the Biological Society of Washington 112:453— 469. , L. R. Heaney, B. R. Tabaranza, & D. S. Bal- ete. 1998. A review of the genera Crunomys and Archboldomys (Rodentia: Muridae: Murinae), with descriptions of two new species from the Philippines.—Fieldiana: Zoology, n.s. 89:1—24. Ruedi, M., & P. Vogel. 1995. Chromosomal evolution and zoogeographic origin of southeast Asian shrews (genus Crocidura).—Experientia 51: 174-178. Seabright, M. 1971. A rapid banding technique for hu- man chromosomes.—The Lancet 1971, II: 971— ae Stafford, B. J.. & E S. Szalay. 2000. Craniodental functional morphology and taxonomy of der- mopterans.—Journal of Mammalogy 81:360— 385. 709 Thomas, O. 1898. On the mammals collected by Mr. John Whitehead during his recent expedition to the Philippines. With field-notes by the collec- tor.—Transactions of the Zoological Society of London 14:377—412. Waterhouse, G. R. 1839. Description of a new species of squirrel (Sciurus philippinensis) from the Philippine Islands.—Proceedings of the Zoolog- ical Society of London 1839:117-118. Wurster, D. H., & K. Benirschke. 1968. Comparative cytogenetic studies in the Order Carnivora.— Chromosoma 24:336—382. Yong Hoi-Sen, S. S. Dhaliwal, Lim Boo-Liat, I. Mull, & The Kok-Leng. 1975. Karyotypes of four species of Callosciurus (Mammalia, Rodentia) from peninsular Malaysia.—Malaysian Journal of Science 3:1-5. Yoshida, T. H. 1985. The evolution and geographic differentiation of the house shrew karyo- types.—Acta Zoologica Fennica 170:31—34. Zima, J., L. Lukacova, & M. Macholan. 1998. Chro- mosomal evolution in shrews. Pp. 175-218 in Evolution of Shrews. Mammal Research Insti- tute, Polish Academy of Sciences, Bialowieza, 458 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):710—718. 2003. Early zoeas of Athanas parvus De Man, 1910 (Decapoda: Caridea: Alpheidae) reared in the laboratory Hoi Jeong Yang and Chang Hyun Kim (HJY) Department of Life Sciences, Silla University, Busan 617-736, Republic of Korea, e-mail: alpheidae @ hotmail.com; (CHK) Department of Biology, Pusan National University, Busan 609-735, Republic of Korea Abstract.—Early zoeas of Athanas parvus De Man are described and illus- trated in detail, based on laboratory-hatched eggs from females collected in Sangju, Korea. This is the first record of the genus Athanas Leach from Korean waters. The zoeal characteristics of Athanas species are compared with those of species of the other four alpheid genera for which larvae are known. Based on morphological similarity, larval characters of Athanas species support a close relationship between the Alpheidae and Palaemonidae. The small shrimps of the genus Athanas Leach, 1814 are usually found in dead coral heads, pools under rocks, and sandy or muddy beaches. Some species are found in permanent association with sea urchins, liv- ing between the spines on the oral surface (Banner & Banner 1973). Ten species of the family Alpheidae, belonging to four genera have been recorded from Korean waters (Yang & Anker 2003). However, species of the genus Athanas have not been reported from Korean waters to date. Larvae of three species of Athanas have been described so far: A. djiboutensis Cou- tiere, 1897, first zoea obtained from Ghar- daga (Gurney 1938); A. dimorphus Ort- mann, 1894, first zoea based on plankton material from the Red Sea (Gurney 1927) and first three zoeal stages from India (Bhu- ti et al. 1977); and A. nitescens (Leach, 1814), first three zoeal stages from the U.K. (Lebour 1932), and other descriptions based on plankton material (Sars 1906, William- son 1915, Webb 1921, Kurian 1956, Bour- dillon-Casanova 1960, Williamson 1967). This study describes and illustrates in de- tail the early zoeal stages of A. parvus De Man, 1910, and compares morphological characteristics of the zoea of species of Athanas with those of the other four alpheid genera for which larvae are known. Material and Methods On 21 July 1997, ovigerous females of Athanas parvus were collected from oyster raft cultures in Sangju, Korea (34°43'00’N, 127°59'30°E). Ten newly hatched zoeas were removed and placed in each of six glass bowls containing 33.3%o filtered sea- water, and kept in a growth chamber at 25°C. Larvae were fed daily with the mi- croalga Dunaliella tertiolecta Butcher, and some specimens in each stage were pre- served in 7% neutral formalin. Drawings were made with the help of a camera lucida. Measurements and setal counts were based on the mean of ten specimens for each zoeal stage. Total length (TL) was mea- sured from the rostral tip (postorbital mar- gin for the first zoea) to the posteromedian margin of the telson, excluding posterior se- tae. Carapace length (CL) was measured from the postorbital margin to the postero- median margin of the carapace. The setal armature of appendages is described from proximal to distal segments. The chromato- phore pattern was determined by observa- tion of living zoeas. VOLUME 116, NUMBER 3 Results Four zoeal stages were obtained. When the fourth zoea molted to the fifth zoea, the larvae died because of their inability to ex- tricate themselves from the fourth zoeal exoskeleton. The first zoeal stage is de- scribed in detail, and for subsequent stages only main differences from previous stage are given. First zoea ie Duration. 18—20 hours. ee oe @le26—4 6) imme. 026 (0.24—0.32) mm. Carapace (Fig. 1A). Rostrum absent; an- terior dorsomedian papilla present; ptery- gostomian spine present; supraorbital and antennal spines absent; anteroventral and posteroventral denticles absent; eyes ses- sile. Antennule (Fig. 1C). Peduncle unseg- mented; inner flagellum with long plumose seta; outer flagellum with 3 aesthetascs, short plumose seta, and long simple seta. Antenna (Fig. 1D). Peduncle with basal spine; endopod spine-like, less than one- third length of scale; scale 4-segmented, with 11 plumose setae and _ distolateral spine. Mandibles. Rudimentary. Maxillule (Fig. 1E). Coxal endite with subterminal seta and 3 terminal setae; basial endite with 2 stout spines and short subter- minal seta; endopod segmented, with 2 ter- minal setae. Maxilla (Fig. 1F). Coxal endite with 2 setae; basial endite bilobed, with 3 + 4 se- tae; endopod with 3 (1 basal + 2 terminal) setae and fine marginal hairs; scaphognath- ite with 5 plumose setae. First maxilliped (Fig. 1G). Coxa with seta; basis unarmed; endopod segmented, with 3 terminal setae; exopod with 4 ter- minal natatory setae. Second maxilliped (Fig. 1H). Coxa un- armed; basis unarmed; endopod 4-segment- Wil ed, with O, O, 1, 3 setae; exopod with 4 terminal natatory setae. Third maxilliped (Fig. 11). Coxa un- armed; basis with simple seta; endopod 4- segmented, with O, 0, 2, 2 setae; exopod with 5 (1 subterminal + 4 terminal) nata- tory setae. Pereiopods (Fig. IJ). Pereiopod 1 bira- mous; pereiopods 2—4 absent; pereiopod 5 uniramous. Abdomen (Fig. 1A). Composed of 5 so- mites, sixth somite not differentiated; all so- mites without spines; third somite bent; third to fifth somites with pairs of dorso- median setae; pleopods absent. Telson and uropods (Fig. 1B). Telson subtriangular, with 7 + 7 marginal setae; outermost 2 pairs plumose only on inner side; bases of all setae except outermost with row of minute spinules. Uropods vis- ible. Chromatophores (Fig. 1A). Dark red chromatophores present on peduncles of an- tennule and antenna, superolateral margin of each eye, anterior dorsomedian papilla of carapace, ventral margin of carapace, dor- solaterally on carapace, bases and exopods of first, second, and third maxillipeds, en- dopod of third maxilliped, rudiments of pe- reiopods | and 5, dorsally on second to fifth abdominal somites, ventrally on first and fourth to fifth abdominal somites, and tel- son. Yellow chromatophores present on pe- duncle of antennule and laterally on second and third abdominal somites. Second zoea igee2 Duration. 1—2 days. Seles (Gs 2—l->4) mms Cl, 028 (0.26—0.32) mm. Carapace (Fig. 2A, B). Rostrum short, not extending beyond eyes; eyes stalked. Antennule (Fig. 2D). Peduncle 2-seg- mented, with 3 and 4 setae. Antenna (Fig. 2E). Unchanged. Mandibles (Fig. 2F). Palp absent; left and TW PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON G H Fig. 1. First zoea of Athanas parvus De Man, 1910. A, habitus, lateral view; B, telson and uropods, dorsal view; C, antennule; D, antenna; E, maxillule; EK maxilla; G, first maxilliped; H, second maxilliped; I, third maxilliped; J, peretopods 1 and 5. Exopod shown truncated in G—I. Scale bars = 0.1 mm. right mandibles without spine between mo- 2 small marginal spiniform setae; endopod lar and incisor processes. 5-segmented, with O, 0, O, 1, 3 setae. Maxillule (Fig. 2G). Coxal endite with 5 Third maxilliped (Fig. 2K). Unchanged. setae; basial endite with 2 spines and 2 se- Pereiopods (Fig. 2L). Pereiopods | and 5 tae. longer; pereiopod 2 biramous as bud; pe- Maxilla (Fig. 2H). Unchanged. reiopod 3 uniramous. First maxilliped (Fig. 21). Unchanged. Abdomen (Fig. 2A, B). Unchanged. Second maxilliped (Fig. 2J). Basis with Telson and uropods (Fig. 2C). Telson VOLUME 116, NUMBER 3 713 Fig. 2. Second zoea of Athanas parvus De Man, 1910. A, habitus, lateral view; B, habitus, dorsal view; C, telson and uropods, dorsal view; D, antennule; E, antenna; K mandibles; G, maxillule; H, maxilla; I, first maxilliped; J, second maxilliped; K, third maxilliped; L, pereiopods 1—3 and 5. Exopod shown truncated in I-K. Scale bars = 0.1 mm. 714 with 8 + 8 marginal setae; outermost pair plumose only on innerside. Uropods un- changed. Third zoea Janez, 3 Duration. 2—3 days. MED GO TGS 61262) Finns Cle OZ! (O.28—0.32) mm. Carapace (Fig. 3A, B). Rostrum longer. Antennule (Fig. 3D). Peduncle with 8 and 6 setae. Antenna (Fig. 3E). Scale 2-segmented. Mandibles (Fig. 3F). Left mandible with 2 spines (lacinia mobilis) between molar and incisor processes; right mandible with spine at corresponding site. Maxillule (Fig. 3G). Unchanged. Maxilla (Fig. 3H). Unchanged. First maxilliped (Fig. 31). Unchanged. Second maxilliped (Fig. 3J). Unchanged. Third maxilliped (Fig. 3K). Unchanged. Pereiopod | (Fig. 3L). Basis with simple seta; endopod 4-segmented, with 0, 0, 2, | setae; exopod with 4 terminal natatory se- tae. Pereiopods 2—3. Unchanged. Pereiopod 5 (Fig. 3L). Endopod 5-seg- mented, with 0, O, 0, 0, 1 setae; dactylus elongated, sharply pointed, with 4 distal denticles facing mouthparts. Abdomen (Fig. 3A, B). Sixth somite now differentiated from telson; posterolateral margin acute. Telson and uropod (Fig. 3C). Telson with 7 + 7 marginal setae; posterior width slightly greater than anterior width. Uropod free: endopod rudimentary; exopod with 6 plumose setae. Fourth zoea Fig. 4 Duration. 3—4 days. mie eos) 5 (E6267) immer Cle O34 (0.27—0.36) mm. Carapace (Fig. 4A, B). Unchanged. Antennule (Fig. 4D). Proximal segment of peduncle with 10 plumose setae; outer PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON flagellum with 3 aesthetascs, 2 plumose se- tae, and simple seta. Antenna (Fig. 4E). Unchanged. Mandibles (Fig. 4F). Unchanged. Maxillule (Fig. 4G). Unchanged. Maxilla (Fig. 4H). Basial endite with 4 tz) Setaes First maxilliped (Fig. 41). Unchanged. Second maxilliped (Fig. 4J). Unchanged. Third maxilliped (Fig. 4K). Basis with 2 setae; endopod 5-segmented, with 0, O, 0, 2, 2 setae. Pereiopod | (Fig. 4L). Basis with 2 sim- ple setae. Pereiopods 2—4. Unchanged. Pereiopod 5 (Fig. 4M). Unchanged. Abdomen (Fig. 4A, B). Unchanged. Telson and uropod (Fig. 4C). Telson nar- rower, with 4 + 4 marginal setae; posterior width slightly narrower than anterior width. Uropod developed: endopod with 8 plu- mose setae; exopod with 12 plumose setae. Discussion There are larval descriptions for 17 spe- cies in seven genera of Alpheidae from the Indo-West Pacific (Yang & Kim 2002). The larval descriptions of Alpheopsis garricki Yaldwyn, 1971 and Betaeopsis aequimanus (Dana, 1852) (see Packer 1985) are not suf- ficiently informative for comparisons. How- ever, useful comparison of early zoeas is possible between the remaining genera: Al- pheus Fabricius, 1798, Athanas, Automate De Man, 1888, Synalpheus Bate, 1888, and Vexillipar Chace, 1988. The zoeas of spe- cies of these latter five genera can be dis- tinguished from each other by the setation of the endopods of the maxillule, the max- illa, and the second maxilliped. In zoeas of species of Athanas, Alpheus, and Vexillipar, the endopod of the maxilla has | + 2 setae, while in those of Automate and Synalpheus the endopod is furnished with O + 4 and 2 + 3 setae, respectively. Zoeas of species of Athanas are readily distinguished from those of Alpheus and Vexillipar by having two setae on the endopod of the maxillule, VOLUME 116, NUMBER 3 TANS Fig. 3. Third zoea of Athanas parvus De Man, 1910. A, habitus, lateral view; B, habitus, dorsal view; C, telson and uropod, dorsal view; D, antennule; E, antenna; K mandibles; G, maxillule; H, maxilla; I, first max- illiped; J, second maxilliped; K, third maxilliped; L, pereiopod 1; M, pereiopod 5. Exopod shown truncated in I-L. Scale bars = 0.1 mm. 716 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ses SSS Ps DDS = kerr Z SSS PS Sore Se Gb WSS Fig. 4. Fourth zoea of Athanas parvus De Man, 1910. A, habitus, lateral view; B, habitus, dorsal view; C, telson and uropod, dorsal view; D, antennule; E, antenna; K mandibles; G, maxillule; H, maxilla; I, first max- illiped; J, second maxilliped; K, third maxilliped; L, pereiopod 1; M, pereiopod 5. Exopod shown truncated in I-L. Scale bars = 0.1 mm. VOLUME 116, NUMBER 3 and an unarmed proximal segment of the endopod of the second maxilliped. Lebour (1932) pointed out that the dif- ferences between the larvae of Athanas and Alpheus were the length of the dactylus of the endopod of the third maxilliped in the second zoea, and the number of the exo- pods of the pereiopods in later stage zoeas. The dactylus of the endopod of the third maxilliped in the second zoea of Athanas dimorphus Ortmann, 1894 (see Bhuti et al. 1977), A. parvus (present study), and Syn- alpheus tumidomanus (Paulson, 1875) (see Bhuti et al. 1977) is never elongated; in species of Alpheus the dactylus is long, al- though it becomes short in the third zoea (see Yang & Kim 2002). The second zoea of Betaeus truncatus Dana, 1852 and B. emarginatus (H. Milne Edwards, 1837) (see Albornoz & Wehrtmann 1997) has the dac- tylus of the endopod of the third maxilliped elongated. The length of the dactylus of the endopod of the third maxilliped is, there- fore, an important character of the second zoea distinguishing the larvae of Athanas and Synalpheus from those of Alpheus and Betaeus. However, it is not possible at this time to distinguish the second zoea of Athanas from that of Automate dolichog- natha De Man, 1888 (see Bhuti et al. 1977) and Vexillipar repandum Chace, 1988 (see Saito et al. 1998) because larval descrip- tions of both species are limited to the first stage. Except for Periclimenes pandionis Hol- thuis, 1951, larvae of the Alpheidae and the Palaemonidae are characterized by both the lack of a proximal endite on the maxilla, and the lack of an outer seta on the base of the maxillule, in all zoeal stages. Moreover, the flexure of the third abdominal somite in the first zoea of Athanas also occurs in lar- vae of palaemonid species of the subfamily Pontoniinae, e.g., Coralliocaris graminea (Dana, 1852) and Harpiliopsis beaupresii (Audouin, 1826) (see Gurney 1938, the lat- ter as Harpilius beaupresii), and P. pan- dionis (see Gore et al. 1981). As suggested by Gurney (1938), these larval characters Jila support the view that the Alpheidae are closely related to the Palaemonidae. Acknowledgments The authors sincerely thank Drs. D. I. Williamson of Port Erin Marine Laboratory, Isle of Man, U.K., for reading the manu- script; A. Anker of the Muséum national d’ Histoire naturelle, Paris, France, for con- firmation of species identification; and I. S. Wehrtmann of Alfred Wegener Institute for Polar and Marine Research, Germany, for providing valuable literature. The authors are indebted by Drs. G. W. Pohle of Atlan- tic Reference Centre, Canada, and A. Anker for the suggestions and comments to the manuscript. Literature Cited Albornoz, L., & I. S. Wehrtmann. 1997. Descripcié6n y clave de los primeros estadios larvales de ca- marones carideos (Decapoda: Hippolytidae, Al- pheidae, Rhynchocinetidae) de aguas costeras de Chile.—Investigaciones Marinas, Valparaiso 25:121-133. Audouin, J. V. 1826. Explication sommaire des planch- es de Crustaceés de l Egypte et de la Syrie, pub- liées par Jules-César Savingy, membre de l'Institut: offrant un expose des caractéres na- turelles des genres, avec la distinction des es- peéces. Pp. 77—98 in J.-C. Savingy, ed., Descrip- tion de l Egypte, Histoire Naturella, Paris | (4). Banner, D. M., & A. H. Banner. 1973. The alpheid shrimp of Australia Part I. The lower genera.— Records of the Australian Museum 28:291—385. Bate, C. S. 1888. Report on the Crustacea Macrura collected by the H.M.S. “Challenger” during the years 1873—76.—Report on the Scientific Results of the Voyage of H.M.S. “‘Challenger”’ during the Years 1873-76 24:1—942. Bhuti, G. S., S. Shenoy, & K. N. Sankolli. 1977. Lab- oratory reared alpheid larvae of the genera Au- tomate, Athanas, and Synalpheus (Crustacea, Decapoda, Alpheidae). Pp. 588—600 in Proceed- ings of the Symposium on Warm Water Zoo- plankton, Special Publication, National Institute of Oceanography, Goa, India. Bourdillon-Casanova, L. 1960. Le méroplancton du golfe de Marseille: les larves de Crustacés Dé- capodes.—Recueil des Travaux de la Statation Marine d’Endoume 30:1—286. Chace, FE A., Jr. 1988. The caridean shrimps (Crusta- cea, Decapoda) of the Albatross Philippine ex- 718 pedition, 1907-1910. Part V. Family Alphei- dae.—Smithsonian Contributions to Zoology 466:1—99. Coutiére, H. 1897. Note sur quelques alphéidés nou- veaux OU peu connus rapporteés de Dyibouti (Afrique Orientale).—Bulletin du Muséum d’ Histoire Naturelle 3:233—236. Dana, J. D. 1852. Conspectus Crustaceorum quae in Orbis Terrarum circumnavigatione, Carolo Wilkes e Classe Republicae Foederatae Duce, lexit et descripsit—Proceedings of the Acade- my of Natural Sciences of Philadelphia 1852: 10-28. Fabricius, J. C. 1798. Supplementum entomologiae systematicae. Hafniae, 572 pp. Gore, R. H., C. L. V. Dover, & J. R. Factor. 1981. Studies on decapod Crustacea from the Indian River region of Florida. XVIII. Rediscovery of Periclimenes (Periclimenes) pandionis Hol- thuis, 1951 (Caridea, Palaemonidae) with notes on the males and zoeal stages.—Crustaceana 40(3):253—265. Gurney, R. 1927. Zoological results of the Cambridge expedition to the Suez Canal. Report on the lar- vae of the Crustacea Decapoda.—Transactions of the Zoological Society of London 22:231- 286. . 1938. The larvae of the decapod Crustacea. Palaemonidae and Alpheidae.—Great Barrier Reef Expedition 1928-29, Scientific Reports 6: 1—60. Holthuis, L. B. 1951. A general revision of the Pala- monidae (Crustacea, Decapoda, Natantia) of the Americas. I. The subfamilies Euryrhynchinae and Pontoniinae.—Allan Handcock Foundation, Occasional Paper 11:1—332. Kurian, C. V. 1956. Larvae of decapod Crustacea from the Adriatic Sea.—Acta Adriatica 6:1—108. Leach, W. E. 1814. Crustaceology. Pp. 28-36 in D. Brewster, ed., Edinburgh Encyclopedia, 7. Lebour, M. V. 1932. The larval stages of the Plymouth Caridea. IV. The Alpheidae.—Proceedings of the Zoological Society of London 1932:463— 469. Milne Edwards, H. 1837. Histoire naturelle des Crus- tacés, Comprenant |’ Anatomie, La Physiologie et la Classification de ces Animaux, 2: 532 pp. Man, J. G. de. 1888. Berichat tiber die von Herrn Dr. J. Brock in indischen Archipel gesammelten Decapoden und Stomatopoden.—Archiv ftir Naturgeschichte 53(1):215—600. . 1910. Diagnoses of new species of macrurous decapod Crustacea from the “Siboga-Expedi- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tion”’.—Tijdschrift der Nederlandsche Dierkun- dige Vereeniging (2)11:287—319. Ortmann, A. 1894. Crustaceen. Pp. 3—80 in R. Semon, ed., Zoologische Forschungreisen in Australien und dem malayischen Archipel.—V. Denk- schriften Medizinisch Naturwissenschaftliche Gesellschaft zu Jena 8. Packer, H. A. 1985. A guide to the larvae of New Zealand shallow water Caridea (Crustacea, De- capoda, Natantia).—Zoology Publications from Victoria University of Wellington 78:1—16. Paulson, O. 1875. Podophtalmata i Edriophtalmata (Cumacea). Chast I in Izsledovaniya Rakoob- raznykh krasnago moray s zametkami otnosi- tel’no Rakoobraznykh drugikh morei. Kiev, 144 pp. [English translation: Por, EK D. 1961 Podo- phtalmata and Edriophtalmata (Cumacea). Part I in Studies on Crustacea of the Red sea with notes regarding other seas. Jerusalem, Israel, 164 pp.]. Saito, T., K. Nakajima, & K. Konishi. 1998. First zoea of a rare deep-sea shrimp Vexillipar repandum Chace, 1988 (Crustacea, Decapoda, Caridea, Alpheidae), with special reference to larval characters of the family.—Publications of the Seto Marine Biological Laboratory 38:147—153. Sars, G. O. 1906. Post-embryonal development of Athanas nitescens Leach.—Archiv fiir Mathe- matik og Naturvidenskab 27:1—29. Webb, G. E. 1921. The larvae of the Decapoda Ma- crura and Anomura of Plymouth.—Journal of the Marine Biological Association of the United Kingdom 12:385—425. Williamson, D. I. 1967. Crustacea Decapoda: Larvae IV. Caridea, Families: Pandalidae and Alphei- dae.—Fiches d’Identification du Zooplancton 109:1—5. Williamson, H. C. 1915. VI. Crustacea Decapoda: Lar- ven.—Nordisches Plankton (6)18:315—588. Yaldwyn, J. C. 1971. Preliminary descriptions of a new genus and twelve new species of natant decapod Crustacea from New Zealand.—Re- cords of the Dominion Museum 7(10):85—94. Yang, H. J., & A. Anker. 2003. New records of alpheid shrimps (Decapoda, Caridea, Alpheidae) from Korea.—Korean Journal of Systematic Zoology 19(1):1-9. , & C. H. Kim. 2002. Early zoeas of two snap- ping shrimps Alpheus digitalis De Haan, 1850 and Alpheus japonicus Miers, 1879 (Decapoda, Caridea, Alpheidae) with notes on the larval characters of the Alpheidae.—Korean Journal of Biological Sciences 6:95—105. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):719—731. 2003. Morphological variation in glochidia shells of six species of Elliptio from Gulf of Mexico and Atlantic Coast drainages in the southeastern United States Christine A. O’Brien, James D. Williams, and Michael A. Hoggarth (CAO’B, JDW) U.S. Geological Survey, Center for Aquatic Resources Studies, Gainesville, Florida 32653; (MAH) Department of Life and Earth Sciences, Otterbein College, Westerville, Ohio 43081 Abstract.—The genus Elliptio, with 36 currently recognized species, is the largest genus in the family Unionidae in North America. The genus is repre- sented by two species, Elliptio crassidens and E. dilatata, in the Interior Basin and 34 species in drainages of the eastern Gulf of Mexico and Atlantic Coast. The paucity and variation of conchological characters in the genus Elliptio makes it extremely difficult to define species and determine relationships. We examined glochidia from six species of Elliptio in an effort to determine if there are useful characteristics for species level identification and/or characters for identification of species groups. Elliptio species were selected to represent different morphological groups from four drainages in the southeastern United States. The glochidia from EF. crassidens, E. dariensis, E. hopetonensis, E. icterina, E. shepardiana, and E. mcmichaeli were qualitatively compared, using scanning electron microscopy, with each other and with descriptions of these and other Elliptio glochidia described in the literature. Two groups were iden- tified. The crassidens group, including E. crassidens, E. dariensis, and E. memichaeli, had subtriangular glochidia with a triangular styliform hook ex- tending from the ventral margin of the valve and rough exterior valve sculp- turing. Adults of this group had wrinkled or corrugated sculpturing on the posterior slope of the shell. The complanata group, including E. hopetonensis, E. icterina, and E. shepardiana, had subelliptical glochidia with a broad flange extending the entire ventral margin and loose-looped exterior valve sculpturing. Adults of this group lack sculpturing on the posterior slope of the shell. Dif- ferences in glochidial morphometrics were found, however, additional work is needed to determine if they are reliable for species level identification. The genus Elliptio, with 36 currently rec- ognized species, is the largest genus in the family Unionidae in North America (Tur- geon et al. 1998). In the Interior Basin, this genus is represented by two species, E. crassidens and E. dilatata. The remaining 34 species are found in eastern Gulf of Mexico and Atlantic Coast drainages. These 34 species are characterized by ex- treme morphological variation ranging from compressed lanceolate forms (e.g., E. she- pardiana) to those that are oval and highly inflated (e.g., E. hopetonensis) (Burch 1975). The paucity of conchological char- acters and the extreme morphological vari- ation exhibited by most species makes it difficult to identify species and determine relationships. Clench & Turner (1956) characterized the genus Elliptio as “‘more confusing than any other in the Unionidae of North Amer- ica’’—a statement which unfortunately re- mains true today. As the validity of some Elliptio species has been questioned, and 720 some will probably be found to be syno- nyms, it is also likely there are valid species that are not currently being recognized. This uncertainty has resulted in reluctance on the part of conservation agencies and or- ganizations to take appropriate actions to provide protection for some species. At least one species, the winged spike, E. ni- gella, is considered to be extinct (Brim Box & Williams 2000), and another 22 have been assigned a national conservation status of endangered, threatened, or special con- cern (Williams et al. 1993). While most rec- ognized species of Elliptio have declined, it is still possible to protect much of the di- versity represented in this genus if action is taken in the near future. Once discrete tax- onomic entities are identified and their dis- tribution accurately delineated only then can conservation and management plans be implemented. The shells of Elliptio are usually mod- erately to greatly elongated, dark olive brown to black, often rayed in juveniles, and lack sculpturing with the exception of wrinkles or corrugations on the posterior slope and umbos of some species. Nacre color varies from white and highly irides- cent to all shades of pink and red to deep purple and violet. Soft parts are whitish to flesh colored. The eggs are white. Branchial papillae are present and usually simple but dendritic papillae may be present; anal pa- pillae are also present. The mantle margin lacks papillae anterior to branchial aperture. The glochidia are expelled from the mar- supium in compacted masses called conglu- tinates. All species of Elliptio utilize only the outer gills as marsupia (Britton & Fuller 1980). The genus Elliptio has been diagnosed as having hookless or spineless glochidia (Ort- mann 1912, Britton & Fuller 1980). How- ever, we found hook-like structures on the ventral margin of the glochidia of some El- liptio glochidia. This hook-like structure, ‘““pseudohook,”’ is not like those described as a diagnostic character of the subfamily Anodontinae (Clarke 1981, 1985). The PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pseudohook is an expansion of the flange on the ventral margin and is completely covered with microstylets of a near uniform SIZE. The paucity of information on Elliptio glochidial morphology (Hoggarth 1999) and host fish determinations (Watters 1994) is surprising considering the widespread distribution of the genus, high diversity (36 taxa), and the number of species with pop- ulations large enough to be utilized for study. Most of the work has been done on the two Interior Basin species, E. crassidens and E. dilatata, and one Atlantic Coast spe- cies, E. complanata. The difficulty in the identification and uncertainty surrounding their distributional limits has likely contrib- uted to the lack of glochidial morphology and host fish research for the Atlantic Coast and eastern Gulf species. We examined glochidia from six species of Elliptio in an effort to determine if there are characters useful in species and/or group identification. The Elliptio species were selected to represent different mor- phological groups from four drainages in the southeastern United States. The glo- chidial valve shape, lateral view, micro- points, microstyle, valve sculpturing, and hinge ligament length of E. crassidens, E. dariensis, E. hopetonensis, E. icterina, E. shepardiana, and E. mcmichaeli were com- pared qualitatively, using scanning electron microscopy (SEM). These species were then compared with other Elliptio glochidia described in published literature. Shells of adult Elliptio were also evaluated using mu- seum specimens and published descriptions to determine if there were characteristics useful in identifying species groups within the genus Eiliptio. Materials and Methods Gravid females representing six species of Elliptio were collected by hand and SCUBA. The valves were gently pried open 1 cm and the soft parts were visually in- spected for swollen gills, which is usually VOLUME 116, NUMBER 3 indicative of eggs or glochidia. Three spe- cies, E. dariensis, E. hopetonensis, and E. shepardiana were collected from the Alta- maha River, Appling County, Georgia, on 21 Apr 1996. A single gravid E. crassidens was collected from Cooleewahee Creek, Baker County, Georgia, on 2 Apr 1997, and a single gravid E. icterina was collected from Worthingon Springs, Dixie County, Florida, on 21 May 1996. Fifteen E. mcmi- chaeli were collected from the Choctaw- hatchee River on 5 May 1986. The gravid mussels were transported in coolers with ambient temperature creek wa- ter to the U.S. Geological Survey, Center for Aquatic Resources Studies, Gainesville, Florida, where they were held in 4-liter jars until they released their glochidia. Female mussels released their glochidia within two weeks of their capture date. All of the glo- chidia were collected from females that re- leased their glochidia in captivity, except for E. mcmichaeli. Glochidia of Elliptio mcmichaeli were removed from specimens that had been fixed in a 10% buffered so- lution of formalin and stored in 70% etha- nol. Glochidia collected alive were deter- mined viable when a snapping response was observed after a few salt crystals were added to a small subsample of 50 to 100 individual glochidia (Zale & Neves 1982). The viable glochidia were stored in 70% ethanol and were later used to describe glo- chidial morphology using SEM. Tissue inside the glochidial shell was re- moved by soaking them in a 5% sodium hypochlorite solution for about 10 minutes. The shells were then rinsed several times with tap water and preserved in 70% etha- nol (Kennedy et al. 1991). Several hundred preserved shells were mounted on a double- sided sticky carbon tape, air dried for 15 minutes, coated with gold, and examined using SEM. Photos were taken of the valve (250-500), flange region (300—3,000), hinge ligaments (SO0—1,000X), and shell sculpturing (15,000—30,000) of the glo- chidia from the six Elliptio species. The flange is defined as a flattened area along 72) the ventral margin of glochidial valve with microstylets, often referred to as micro- points (small tooth-like projections located along the flange). Measurements of the glo- chidia were determined by averaging the height (dorsal to ventral edge), length (an- terior to posterior edge) and dorsal margin (ong edge) length measurements of the glochidia under a stereo microscope with an ocular micrometer (10). The number of glochidia used to determine average mea- surements varied because some of the glo- chidia samples collected were small. Shells of adult El/iptio in the Florida Mu- seum of Natural History collection were ex- amined to determine the presence or ab- sence of wrinkled sculpturing on the pos- terior slope. This character ranges in its de- velopment from the obvious to the obscure and may be absent in some individuals. It is often best developed on the upper portion of the slope nearest the umbo, which is of- ten eroded, obscuring or removing any trace of the sculpturing. We examined lots of E. congarea, E. crassidens, E. dilatata, E. ho- petonensis, E. dariensis, E. fraterna, E. memichaeli, and E. shepardiana to evaluate this character. Results Glochidial shell descriptions.—Elliptio crassidens (Lamarck, 1819). Glochidium subtriangular (Fig. la) with a length of 134 + 2.8 pm (130-138 pm), a height of 150 + 69 um (141-160 pm), and a hinge length of 85 + 4.5 wm (80—92 pm) (Table 1). A triangular, styliform flange (hook), covered with microstylets, extends from the ventral aspect of each valve (Fig. 1b, f). Mi- crostylets are arranged in complete vertical rows. The microstylets are longer toward the middle and ventral margins of the valve and smaller toward the distal edge and the lateral margins of the valve. Valve outline is asymmetric with the anterior margin slightly more expanded than the posterior margin (Fig. la). Ventral margin produced into a nipple-like extension giving the valve T22 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. its triangular shape (Fig. la). The hinge lig- ament is located nearer the posterior margin (Fig. lc, d) than the anterior margin and the valve is covered with rough exterior valve sculpturing (Fig. le). Elliptio dariensis (Lea, 1842). Glochidi- um subtriangular (Fig. 2a) with a length of 142 + 5.5 wm (135-151 pm), a height of 166 = 6.4 pm (157-172 wm), and a hinge length of 88 + 5.0 wm (81-95 pm) (Table 1). Anterior and posterior margins subequal with the anterior margin slightly more fully rounded than the posterior margin. Ventral margin produced into a nipple-like exten- sion (Fig. 2a, e) giving the valve its trian- gular shape. A triangular, styliform flange Elliptio crassidens. Scanning electron micrographs (SEM) of glochidium. a, exterior valve; b, interior valve showing styliform hook; c, d, hinge; e, exterior valve sculpture; f, flange region with triangular styliform hook. extends from the ventral aspect of each valve to form a hook (Fig. 2b, e). Microsty- lets are arranged in complete vertical rows on the flange (Fig. 2b). They are larger to- ward the center of the flange and toward the proximate edge of the flange, and smaller near the distal edge of the hook and toward the lateral margins of the valve. The hinge ligament is slightly posterior of center (Fig. 2c) and the valve is covered with rough ex- terior valve sculpturing (Fig. 2d). Elliptio mcmichaeli (Clench & ‘Turner, 1956). Glochidium subtriangular (Fig. 3a) with a length of 146 + 7.9 wm (130-157 wm), a height of 153 + 5.8 wm (149-161 wm), and a hinge length of 90 + 9.4 wm VOLUME 116, NUMBER 3 Table 1.—The number of glochidia used to determine the measurements and standard deviation for six species of Elliptio. Number of Number of Species glochidia females (Triangular shape) Elliptio crassidens 6 l Elliptio dariensis 6 3 Elliptio mcmichaeli 10 S) (Subelliptical shape) Elliptio hopetonensis 10 4 Elliptio icterina 10 1 Elliptio shepardiana 10 2 (80-110 wm) (Table 1). Anterior and pos- terior margins subequal with the anterior margin slightly more fully rounded than the posterior margin. The ventral margin is pro- duced into a nipple-like point (Fig. 3a—c) giving the valve a triangular appearance. Extending from the ventral margin of each valve is a triangular, styliform hook (col- lapsed in Fig. 3c). Microstylets cover the hook in complete vertical rows. The mi- crostylets are longer near the center of the hook and become smaller laterally where they intergrade into micropoints (Fig. 3c). The hinge ligament is located slightly pos- terior of center (Fig. 3d) and the valve is covered with rough exterior valve sculptur- as (Cries, SYe)). Elliptio hopetonensis (Lea, 1838). Glo- chidium depressed subelliptical (Fig. 4a) with a length of 206 + 5.8 wm (197-215 Him) anherchtole2o— 295 OM imnC22—2 si wm), and a hinge length of 137 + 5.0 pm (130—145 pm) (Table 1). Anterior and pos- terior margins equal, but only slightly rounded to a more fully rounded ventral margin (Fig. 4a). A broad flange extends from the ventral margin (Fig. 4b). This flange is covered with microstylets in com- plete vertical rows (Fig. 4d). These micro- stylets are larger near the midpoint of the Mean measurement + SD 723 Length Height Hinge 134 + 3 ISO) 2s. 7/ iS) 25S (130-138 pm) (141-160 pm) (80-92 pm) 142 +6 NOD 2 © eX) ZED) (135-151 wm) (157-172 pm) (81—95 pm) 146 + 8 53) 22 © wl) ze © (130-157 pm) (149-161 pm) (80-110 xm) ANS 22 © 226 + 4 NW S7/ eS) (197-215 pm) (222-231 pm) — (130-145 ym) AVS) 28 © PNG sae 4 ae © (195-215 pm) (195-234 pm) (125-148 pm) 241 +3 284 + 6 148 + 6 (238-245 xm) (279-295 jm) (140-160 pm) ventral margin and grade into micropoints laterally. The dorsal margin is straight, long, and the hinge ligament is located much nearer the posterior than the anterior margin (Fig. 4c). The surface of the valve is covered with loose-looped exterior valve sculpturing (Fig. 4e). Elliptio icterina (Conrad, 1834). Glo- chidium depressed subelliptical (Fig. 5a) with a length of 203 + 6.4 pm (195-215 wm), a height of 216 + 10.5 wm (195-234 wm), and a hinge length of 141 += 8.8 pm (125-148 ym) (Table 1). Anterior and pos- terior margins equally, but only slightly rounded to a more fully rounded ventral margin. A wide ventral flange extends from the margin (Fig. 5b). The flange is covered in microstylets, (Fig. 5b, d) which are lon- ger near the midpoint of the flange and grade into micropoints laterally. The dorsal Margin is straight with the hinge ligament located much nearer to the posterior margin of the valve (Fig. 5c). Loose-looped exte- rior valve sculpturing covers the exterior surface of each valve (Fig. 5e) and that sculpturing can even be seen extending onto the surface of the microstylets on the flange (Fig. 5d). Elliptio shepardiana (Lea, 1834). Glo- chidium subelliptical (Fig. 6a) with a length 724 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON eis, Ze c, hinge; d, surface sculpturing; e, internal view. of 241 + 3.3 pm (238-245 pm), a height of 284 + 5.8 wm (279-295 wm), and a hinge length of 148 + 6.3 wm (140-160 wm) (Table 1). Anterior and posterior mar- gins more or less equal and only slightly rounded. The ventral margin is strongly arched and supports a wide ventral flange (Fig. 6b). The ventral flange is widest near the middle of the ventral margin and tapers to both anterior and posterior margins. The flange supports numerous microstylets ar- ranged in complete vertical rows which grade into micropoints distally on the flange and laterally toward the anterior and pos- terior margins of the flange (Fig. 6c, d). The hinge is straight and the hinge ligament is positioned nearest the posterior margin of Elliptio dariensis. SEM of glochidium. a, exterior valve; b, interior valve showing styliform hook; the valve (Fig. 6c). Loose-looped valve sculpturing covers the exterior surface of each valve (Fig. 6e), and this sculpturing is seen extending onto the ventral flange and covering the microstylets (Fig. 6d). Adult shell morphology.—Elliptio cras- sidens is widespread in the Interior Basin and eastward along the Gulf Coast from the Amite River in Louisiana to the Ochlock- onee River drainage in Florida (Brim Box & Williams 2000). Populations in the Ap- alachicola and Ochlockonee have been re- ferred to as a distinct subspecies, E. incras- satus. However, this taxon is generally not recognized as valid. Elliptio crassidens is large (length of 150 mm) and has a mod- erate to thick shell with a prominent pos- VOLUME 116, NUMBER 3 NZS Fig. 3. Elliptio memichaeli. SEM of glochidium. a, exterior valve; b, side view of valve; c, flange; d, hinge; e, exterior valve sculpture. terior ridge. It typically has sculpturing on the posterior slope consisting of distinct wrinkles or corrugations. Elliptio dariensis is endemic to the Al- tamaha River drainage, including the Ohoo- pee and Ocmulgee rivers in Georgia. John- son (1970, 1972) considered the distribu- tion of this species to extend southward into the St. Johns River system in peninsular Florida. The shell is thin, large (length to 135 mm), and moderately inflated. The pos- terior ridge is well defined with a broad posterior slope. Similar to E. crassidens, the posterior slope typically has sculpturing in the form of numerous wrinkles. Elliptio mcmichaeli is endemic to the Choctawhatchee River system (Gulf of Mexico drainage) in Alabama and Florida. The shell is thin, moderately large (>110 mm in length), with a poorly developed biangulated posterior ridge. The posterior slope is broad, somewhat concave, with sculpturing in the form of wrinkles that ra- diate from the posterior ridge across the slope to the dorsal margin. Elliptio hopetonensis is endemic to the Altamaha River drainage in Georgia. The shell is moderately thin and large (155 mm in length). The posterior ridge is broad and rounded dorsally, but terminates in a bian- gulation near the ventral margin. The pos- terior slope is broad, somewhat concave, and lacks any wrinkled or corrugated sculp- turing. Elliptio icterina is found in coastal drain- ages from the Escambia River system in Al- abama and Florida, east to peninsular Flor- ida and north on the Atlantic Slope to 726 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON woul ict Hivtel Fig. 4. Elliptio hopetonensis. SEM of glochidium. a, exterior valve; b, interior valve showing expanded flange of ventral margin; c, hinge; d, flange region showing rows of microstylets; e, exterior valve sculpture. White Oak River in North Carolina (John- son 1970). The shell of E. icterina is mod- erately compressed and small (length to ap- proximately 85 mm). The posterior ridge is low, somewhat rounded, with a broad pos- terior slope. Wrinkled sculpturing is absent from the posterior slope. Elliptio shepardiana is endemic to the Altamaha River system in Georgia, includ- ing the Ocmulgee, Oconee, and Ohoopee rivers. The shell is long dength to 190 mm), moderately compressed, rounded anteriorly, and somewhat pointed posteriorly. The pos- terior ridge is broadly rounded with a sec- ondary ridge above it, which often results in a slight biangulation at the posterior end. The posterior slope is devoid of wrinkled sculpturing. Discussion Based on glochidial and adult shell mor- phology of the six species described above we have identified what appears to be two clades within the genus Elliptio. While the Six species represent less than 20% of the genus, the nature and consistency of the characters would appear to be indicative of two distinct groups. As glochidia of the re- maining 28 species of Elliptio are examined it should be easy to assign them to one of the two groups described above. It is also possible that one or more additional clades may be found in the remaining 26 species of Atlantic Coast and eastern Gulf of Mex- ico drainage Elliptio. Elliptio crassidens group.—The glochid- ium of Elliptio crassidens was first figured VOLUME 116, NUMBER 3 Fig. 5. Elliptio icterina. SEM of glochidium. a, exterior valve; b, interior valve; c, hinge; d, flange showing microstylets; e, exterior valve sculpture. (Ortmann 1911) and subsequently de- scribed (Ortmann 1912) illustrating its sub- triangular shape (Fig. 6, plate 89, page 346— 347). Ortmann (1912) and Utterback (1915-1916) described the glochidium as small, suboval, without hooks. Ortmann and Utterback gave identical dimensions of 0.13 by 0.15 mm in length and height, re- spectively. Likewise, other reports (Surber ISA, Wilse Colxee Gi all, IYZI)) gemeclllhy describe the glochidium of E. crassidens (or its synonyms) as small, suboval, and spine- less or hookless. Surber (1915) described the glochidium as subtriangular and some- what intermediate between that of Alasmi- donata calceola (= viridis) and Quadrula (= Fusconaia) ebena. He did not identify the glochidium of E. crassidens as having a hook on the ventral margin and he gave the size of this glochidium as larger, 0.15 by 0.16 mm. We found the glochidium of Elliptio crassidens to be subtriangular with a defi- nite hook-like structure at the terminus of the ventral margin. While our measure- ments generally agree with those of Ort- mann and Utterback, their description does not mention the presence of a ““hook.”’ It is possible that they did not see the hook due to lack of magnification. The glochidial shells of the crassidens group were the only Lampsiline-Amblemine species observed to possess subtriangular glochidia (Hoggarth O92) The three species examined during this study which are assigned to the “‘Elliptio crassidens” group are E. crassidens, E. dariensis, and E. mcmichaeli. The glochidia 728 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Elliptio shepardiana. SEM of glochidium. a, exterior valve; b, interior valve showing wide ventral flange; c, hinge; d, flange region showing microstylets; e, exterior valve sculpture. of these species share subtriangular valve shape, the presence of a small styliform hook at the terminus of the ventral margin, and rough exterior valve sculpturing. The shells of adults of this group also share the wrinkled or corrugated sculpturing on the posterior slope. Elliptio complanata group.—The glo- chidium of Elliptio dilatata was first de- scribed by Lea (1863) (as Unio gibbosus). He described the glochidium as _ pouch- shaped, without hooks, and very much like that of Ptychobranchus fasciolaris. In fact, this glochidium is not similar to fasciolaris, except that both are subelliptical (Hoggarth 1999). Lefevre & Curtis (1910, 1912) pro- vide the first measurements for this glo- chidium (0.22 by 0.19 mm in length and height) and Ortmann (1911, 1919) adds that the glochidium of dilatata is small, suboval, and without hooks. Hoggarth (1999) described this glochid- ium as subelliptical with equal lateral mar- gins and equal length and height (0.22 mm). He found that the glochidium was covered in loose-looped valve sculpturing and that lanceolate micropoints cover the narrow ventral flange. He noted that only Ortmann (1919) found the dimensions of this glochidium to be equal (0.20 mm), while Lefevre and Curtis (see above) de- scribed this glochidium as longer than high, and all other authors found this glochidium to have a length of 0.20 mm and a height of 0.22 mm (Ortmann 1912, Surber 1912, Utterback 1915-1916). The light microscope photographs of the glochidia of Elliptio complanata in Matte- VOLUME 116, NUMBER 3 son (1948) clearly show the shape to be su- belliptical and hookless. The low magnifi- cation of the glochidial photographs is such that details of the sculpturing of the shell surface and micropoints are not visible. However, based on the subelliptical outline it appears to be identical to the glochidia of E. dilatata, E. hopetonensis, E. icterina, and E. shepardiana. These glochidia are all higher than long with E. shepardiana the largest and E. icterina and E. hopetonensis having almost identical dimensions with E. complanata. Furthermore, glochidia of all three species possess similar ventral mar- gins (hookless and micropoint structure) and loose-looped external valve sculptur- ing, except for E. complanata, which is yet to be determined. These are characteristics they share with E. dilatata reported by Hoggarth (1999). Based on the subelliptical shape of the glochidia of E. buckleyi (D. S. Ruessler, pers. comm.), it also belongs to the E. complanata group. Adult shells of the species in the E. complanata group are characterized by the absence of sculpturing on the posterior slope. Host fish.—Host fish data for most spe- cies of Elliptio is incomplete. Of the 36 cur- rently recognized species in the genus, only five, E. buckleyi, E. crassidens, E. com- planata, E. dilatata and E. icterina, have any published information on host fish (Watters 1994, Keller & Ruessler 1997). This may be explained in part by the fact that most species of Elliptio occur in Atlan- tic and eastern Gulf Coast streams where there are no commercially important spe- cies. The lack of economic interest resulted in very little research into the reproductive biology and host fish utilization by early mussel biologists, who concentrated their efforts on the commercially valuable Mis- sissippi Basin species. Howard (1914) reported finding Elliptio crassidens encysted on a skipjack herring, Alosa chrysochloris. This is the only host fish information available for E. crassidens (Watters 1994). The host fish for E. darien- sis and E. mcmichaeli remain unknown. q29 However, E. dariensis failed to produce ju- veniles when exposed to bluegill, Lepomis macrochirus, during a laboratory host fish experiment (C. O’Brien, pers. obs.). During the same laboratory host fish experiment, eastern mosquitofish, Gambusia holbrooki, and L. macrochirus were identified as host fishes for E. hopetonensis, but the large- mouth bass, Micropterus salmoides, failed to produce juvenile mussels. While the host for Elliptio mcmichaeli is unknown, there are data that suggest the host is a migratory fish. There are historic records of E. mcmichaeli from the upper Pea River, a tributary of the Choctawhatch- ee River, in southeast Alabama (Blalock- Herod, pers. comm.). The middle portion of the Pea River was impounded in the early 1900s, totally eliminating upstream access to native migratory fishes, including two migratory shads, Alabama shad, Alosa al- abamae, and skipjack herring, A. chrysso- clorris. During a recent survey of the entire watershed, E. mcmichaeli was found at nu- merous stations below the dam on the Pea River and elsewhere in the Choctawhatchee watershed, except it was absent in the Pea River above the dam (Blalock-Herod, pers. comm.). Since the Pea River above the dam currently supports a mussel community, which formerly included E. mcmichaeli, its absence today may be due to the lack of suitable migratory host fish that are blocked by the dam. Although circumstantial, we think this evidence strongly suggests a mi- gratory host fish for E. mcmichaeli. Matteson (1948) found yellow perch, Perca flavescens, as a host fish for Elliptio complanata via laboratory infestations and Wiles (1975) found E. complanata glochid- ia encysted on banded killifish, Fundulus diaphanus. Other fishes reported as hosts for E. complanata include green sunfish, Lepomis cyanellus, orangespotted sunfish, L. humilis, largemouth bass, Micropterus salmoides, and white crappie, Pomoxis an- nularis (Young 1911). Keller & Ruessler (1997) reported L. machrochirus, M. sal- moides, and Florida gar, Lepisosteus platy- 730 rhincus as host fishes for E. buckleyi. They also identified L. macrochirus and M. sal- moides as the host fishes for E. icterina. These were the only fish species tested with E. icterina during their study. The host fish for E. shepardiana remains unknown. Host fishes for the spike, FE. dilatata, include six fishes representing four families: flathead catfish, Pylodictus olivaris, and sauger, Sti- zostedion canadense (Howard 1914); giz- zard shad, Dorosoma cepedianum, and white crappie, P. annularis (Wilson 1916); black crappie, P. nigromaculatus, and yel- low perch, Perca flavescens (Clarke 1981). Based on host fish information available, it is not possible to make any definitive statements regarding possible association between species groups of Elliptio and par- ticular host fish usage. However, based on glochidial shell morphology and host fish information presented above, this could be the case. It appears that species of the cras- sidens group, triangular-shaped glochidia, may only use the highly migratory fishes of the genus Alosa (family Clupeidae), while species of the complanata group, subellipt- ical-shaped glochidia, may use a variety of fishes, 12 species representing 6 families. None of the host fishes used by the E. com- planata group are highly migratory. If fu- ture host fish research confirms this pattern this would provide an additional character- istic to distinguish these two species groups of Elliptio. It would also suggest a very ear- ly evolutionary divergence within the genus Elliptio. Species level identification for many spe- cies of the genus Elliptio remains problem- atic and their evolutionary relationships re- main a mystery. The proper identification of mussel species is a very important part of any management plan. However, concho- logical differences may not be the only tool needed to identify species of the complex Elliptio group. More research is needed to explore the genetic and host fish usage dif- ferences between these two Elliptio groups identified and any other groups that might become apparent. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Brim Box, J., & J. D. Williams. 2000. Unionid mol- lusks of the Apalachicola Basin in Alabama, Florida, and Georgia.—Bulletin of the Alabama Museum of Natural History 21:1—143. Britton, J. C., & S. L. H. Fuller. 1980. The freshwater bivalve mollusca (Unionidae, Sphaeriidae, Cor- biculidae) of the Savannah River Plant, South Carolina. Savannah River Plant Publication SRO-NERP-3. Department of Energy, Aiken, South Carolina, 37 pp. + 5 maps. Burch, J. B. 1975. Freshwater Unionacean clams (Mol- lusca: Pelecypoda) of North America, Rev. edi- tion. Malacological Publication, Hamburg, Michigan, 204 pp. Clarke, A. H. 1981. The tribe Alasmidontini (Unioni- dae: Anodontinae), part I. Pegias, Alasmidonta, and Arcidens.—Smithsonian Contributions to Zoology 326:1—101. . 1985. The tribe Alasmidontini (Unionidae: Anodontinae), part II: Lasmigona and Simpson- aias.—Smithsonian Contributions to Zoology 399:1—75. Clench, W. J., & R. D. Turner. 1956. Freshwater mol- lusks of Alabama, Georgia, and Florida from the Escambia to the Suwannee River.—Bulletin of the Florida State Museum of Biological Sci- ences 1:97—239. Coker, R. E., A. E Shira, H. W. Clark, & A. D. How- ard. 1921. Natural history and propagation of fresh-water mussels.—Bulletin of the U.S. Bu- reau of Fisheries 37(1919—1920):77-181, 17 pls. Separately issued as Bureau of Fisheries Document 893. Conrad, T. A. 1834. Descriptions of some new species of fresh water shells from Alabama, Tennessee, etc.—American Journal of Science and Arts 25: 338-343, 1 pl. [Reprinted in Sterkiana 9(1963): 49-50.] Hoggarth, M. A. 1999. Descriptions of some of the glochidia of the Unionidae (Mollusca: Bival- via).—Malacologia 41:1—118. Howard, A. D. 1914. Experiments in propagation of freshwater mussels of the Quadrula group. Re- port of the U.S. Commissioner of Fisheries for 1913, Appendix IV, pp. 1—52. Issued separately as Bureau of Fisheries Document 801. Johnson, R. I. 1970. The systematics and zoogeogra- phy of the Unionidae (Mollusca: Bivalvia) of the southern Atlantic Slope region.—Bulletin of the Museum of Comparative Zoology 140:263— 449. . 1972. The Unionidae (Mollusca: Bivalvia) of peninsular Florida.—Bulletin of the Florida State Museum of Biological Science 16:181— 249. Keller, A. E., & D. S. Ruessler. 1997. Determination VOLUME 116, NUMBER 3 or verification of host fish for nine species of unionid mussels.—American Midland Natural- ist 138:402—407. Kennedy, V2 S:, Sa@=buller Sana bntze 1 99 Shell hinge development of young Corbicula flumi- nea (Muller) (Bivalvia: Corbiculoidea).— American Malacological Bulletin 8:107—111. Lamarck, J. B. P. A. 1815-1822. Histoire des Animaux sans Vertébres. 8 volumes. Lea, I. 1834. Observations on the naiades; and descrip- tions of new species of that, and other fami- lies.—Transactions of the American Philosoph- ical Society 5[New Series]:23—119, pls. 1-19. . 1838. Description of new freahwater and land shells.—Transactions of the American Philo- sophical Society 6[New Series]:1—154, pls. 1— 24. . 1842. Description of new freahwater and land shells.—Transactions of the American Philo- sophical Society 8[New Series]:163—250, pls. 5-27. . 1863. Descriptions of the soft parts of one hundred and forty-three species and some em- bryonic forms of Unionidae of the United States.—Journal of the Academy of Natural Sciences, Philadelphia 5(N.S.):401—456. Lefevre, G., & W. C. Curtis. 1910. Experiments in the artificial propagation of fresh-water mussels.— Bulletin of the U.S. Bureau of Fisheries 28(1908):615—626. Issued separately as Bureau of Fisheries Document No. 671. 5 CZ . 1912. Studies on the reproduction and artificial propagation of freshwater mus- sels.—Bulletin of the U.S. Bureau of Fisheries 30(1910):105—201. Issued separately as Bureau of Fisheries Document No. 756. Matteson, M. R. 1948. Life history of Elliptio com- planatus (Dillwyn, 1817).—American Midland Naturalist 40:690—723. Ortmann, A. E. 1911. A monograph of the najades of Pennsylvania, parts I and II. Anatomical Intro- duction. The Systems of the North American Najades.—Memoirs of the Carnegie Museum 4: 279-347. . 1912. Notes upon the families and genera of the najades.—Annuals of the Carnegie Museum 8:222—365. ve . 1919. A monograph of the naiades of Penn- sylvania, part III. Systematic account of the genera and species.—Memoirs of the Carnegie Museum 8:1—385. Surber, T. 1912. Identification of the glochidia of fresh- water mussels. Report of the U.S. Commission- er of Fisheries for 1912 and Special Papers, pp. 1—10. Issued separately as Bureau of Fisheries Document No. 771. . 1915. Identification of the glochidia of fresh- water mussels. Report of the U.S. Fisheries for 1914, Appendix V, pp. 1—9. Issued separately as Bureau of Fisheries Document No. 813. Turgeon, D. D. et al. 1998. Common and scientific names of aquatic invertebrates from the United States and Canada: mollusks, 2nd edition. American Fisheries Society Special Publication XS, SAS {0}. Utterback, W. I. 1915-1916. The naiades of Missou- riimAmerican Midland Naturalist 4(3):41—53, (4):97—152, (5):181—204, (6):244—-273, (7):311- 327, (8):339—354, (9):387—400, (10):432—464. Watters, G. T. 1994. An annotated bibliography of the reproduction and propagation of the Unionoidea (primarily of North America).—Ohio Biologi- cal Survey Miscellaneous Contributions No. 1, 158 pp. Wiles, M. 1975. Parasites of Fundulus diaphanus (LeSueur) (Pisces: Cyprinodontidae) in certain Nova Scotian freshwaters.—Canadian Journal of Zoology 53:1578—1580. Williams, J. D., M. L. Warren, Jr., K. S. Cummings, J. L. Harris, & R. J. Neves. 1993. Conservation status of freshwater mussels of the United States and Canada.—Fisheries 18:6—22. Wilson, C. B. 1916. Copepod parasites of fresh-water fishes and their economic relations to mussel glochidia.—Bulletin of the U.S. Bureau of Fish- eries 34:333—374. Issued separately as Bureau of Fisheries Document 824. Young, D. 1911. The implantation of the glochidium on the fish.—University of Missouri Bulletin, Sciences Series 2:1—16. Zale, A. V., & R. J. Neves. 1982. Fish hosts for four species of lampsiline mussels (Mollusca: Unionidae) in Big Moccasin Creek, Virginia.— Canadian Journal of Zoology 60:2535—2542. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):732-—736. 2003. First fossil record of a finfoot (Aves: Heliornithidae) and its biogeographical significance Storrs L. Olson Division of Birds, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A., e-mail: olson.storrs@nmnh.si.edu Abstract.—The distal end of a humerus from the Lee Creek Mine in North Carolina is identified with the modern Neotropical species as Heliornis aff. fulica and provides the first fossil record for the family Heliornithidae. The fossil was determined to come from the Middle Miocene (14 Ma) Pungo River Formation based on sedimentary, preservational, and other lines of evidence. This in turn implies that Heliornis did not spread into South America until after the isthmian land connection about 3 Ma and that the interchange of Heliornithidae between the Old and New World probably took place through the Northern Hemisphere in the Paleogene. The finfoots or sungrebes (Heliornithi- dae) comprise 3 species, each in a mono- typic genus, that are placed in the tradition- al avian order Gruiformes. On the basis of both morphology and molecular data, the family is probably most closely related to rails, Rallidae (Houde 1994, Livezey 1998). The species are sedentary, swimming and diving birds of forested streams in the trop- ics of the New World, Africa, and southeast Asia. Hitherto, there has been no fossil re- cord of the family (Brodkorb 1967, Houde 1994, Tommy Tyrberg, Kimstadt, Sweden, pers. comm.). This has changed with the discovery of the distal end of a humerus in spoil piles of the Lee Creek phosphate mine near Aurora, North Carolina, from which tens of thousands of fossil bird bones had been recovered previously (Olson and Ras- mussen 2001). Systematics Class Aves Order Gruiformes Family Heliornithidae The fossil humerus is characterized by its short, curved overall shape, which at first suggests a medium-sized species of one of the “higher”? orders of arboreal land birds (Olson 1985). It differs from all of these in the very rounded, hemispherical condylus ventralis (ulnar condyle), which contrasts markedly with the narrow, transversely ovoid, and often proximally flattened con- dyle in the traditional orders Coliiformes, Trogoniformes, Coraciiformes, Piciformes, and Passeriformes. Another highly distinc- tive character is the extension of the epi- condylus ventralis distally well past the margins of the distal condyles. These two features (Fig. 1) in combination with the overall size and shape are diagnostic of the Heliornithidae. Genus Heliornis Bonnaterre, 1791 There is little in the way of qualitative differences in the distal half of the humerus to distinguish the three genera of Heliorni- thidae. Podica and Heliopais differ from Heliornis and the fossil in their much larger size, so that on size and geographical grounds it is more reasonable to assign the fossil to the type-genus of the family, He- liornis. VOLUME 116, NUMBER 3 ag @ S)aol)ayt Fig. 1. Left humeri of Heliornis (top row anconal view; bottom row palmar view): A, fossil Heliornis aff. fulica from Lee Creek Mine, North Carolina; B, C, modern Heliornis fulica (UF 38828 = PB 35036; USNM 321493; UF 33412 = PB 29333). The distally projecting epicondylus ventralis (long arrow) and the very rounded, hemispherical condylus ventralis (short arrow) are diagnostic features of the Heliornithidae. 734 Table 1.—Measurements (mm) of the humerus of modern Heliornis fulica (Heliornithidae) compared with fossil humerus of Heliornis aff. fulica (USNM 518297) from the Miocene of North Carolina. Heliornis fulica (N = 13) Heliornis aff. fulica Range (mean) SD fossil Distal width 6.2—7.0 (6.8) 0.24 6.9 Shaft width 2.4—3.2 (2.8) 0.19 3oZ Shaft depth 2.4—2.8 (2.6) 0.12 2D) Heliornis aff. fulica (Boddaert, 1783) Referred specimen.—Distal half of left humerus, National Museum of Natural His- tory, Smithsonian Institution, USNM 518297, collected 13 June 2001 by Geoff Keel. Locality.—Lee Creek Mine, south side of Pamlico River, near Aurora, Beaufort County, North Carolina (35°23'22’N, 76°47'06"W). Horizon.—Pungo River Formation, Mid- dle Miocene; Langhian stage, ca. 14 Ma (Gibson 1983). Fossil birds at the Lee Creek Mine come from deposits of two ages separated by an unconformity—the Middle Miocene Pungo River Fm and the Lower Pliocene Yorktown Fm (Olson and Rasmussen 2001). Assignment of the pre- sent fossil 1s based on several lines of evi- dence. It was recovered from a spoil pile that was topped by sediments of the Pungo River Fm; the preservation is excellent, the bone having a hard, shiny surface, unlike the etched appearance of many of the bones from the Yorktown Fm; blackish phosphatic sand characteristic of the Pungo River Fm was adherent to the fossil; and no strictly tropical elements occur in the Yorktown Fm but are known from the Pungo River and equivalent formations. Comparisons.—In size, the fossil falls within the range of modern Heliornis fulica in two measurements, though near the high end, and is very slightly larger in shaft depth (Table 1). Although the shaft of USNM 518297 is thus generally somewhat more robust, this character varies consid- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON erably in the modern series. Qualitatively, the only apparent differences are the some- what more curved shaft, the relatively larg- er condylus ventralis, and the wider sulcus musculo humerotricipitalis in the fossil. These differences may indicate slight dif- ferences in flight capability, but given the time disparity they are very minor. There is nothing in this single fossil to suggest that it represents a divergence from the lineage leading to the modern species Heliornis fu- lica. This perception might be altered if more of the skeleton were available. Discussion The Heliornithidae are among the classic examples pantropical families of birds (Fig. 2), such as barbets (Capitonidae) and tro- gons (Trogonidae), which likewise occur in the Old and New World tropics, but not in Australasia. This pattern invites the ques- tion of whether their distribution is the re- sult of vicariant events or dispersal. Modern Heliornis fulica extends practi- cally throughout the Neotropics from Mex- ico to northern Argentina. The fossil record from North Carolina is more than 2600 air line km NE of the nearest occurrence of Heliornis in Veracruz, Mexico. The speci- men comes from deposits about 14 Ma, whereas a land connection was established between North and South America only about 3 Ma, with mammalian overland dis- persal having taken place in both directions before about 2.5—2.8 Ma (Marshall 1985). Finfoots, being weak fliers and with very specific aquatic habitat requirements, are al- most certainly incapable of dispersing over an oceanic barrier. Thus, the evidence now at hand suggests that Heliornis fulica must have spread into South America from North America after the isthmian land connection formed, and perhaps considerably after that, given that it has no recognized subspecies (Hellmayr and Conover 1942). Heliornis differs from other finfoots in having altricial young that are carried in ax- illary marsupia (Houde 1994), so that it is VOLUME 116, NUMBER 3 ry “= Rice 2 Map modified from Fisher and Peterson (1964). presumably the most derived of the three species. Houde (1994) found Heliornis to be closer genetically to the Asian genus He- liopais, than to the African genus Podica, although the quality of the DNA of Helio- pais may have been equivocal. He also found that the degree of genetic divergence between Heliornis and the African genus Podica was too small to be compatible with a trans-Atlantic origin of the former and subsequent separation via sea-floor spread- ing, which would have necessitated a com- mon distribution across what is now Africa and South America well back into the Cre- taceous (Smith et al. 1994). Assuming finfoots to require tropical en- vironments and to be incapable of crossing marine water barriers, interchange of fin- foots between the Old and New World must have taken place early in the Tertiary across the Northern Hemisphere, because Helior- nis was already in North America by the Middle Miocene. There is abundant floral and faunal evidence for interchange be- tween North America and both Europe and Asia in the Eocene (Manchester 1999) and in particular for invasions of mammals from Asia in the late Eocene (Woodburne Modern distribution of the three species of Heliornithidae (solid) and the only fossil record (arrow). and Swisher 1995). Thus, fossils of Helior- nithidae might be anticipated in Paleogene deposits of the northern continents. Acknowledgments I am most grateful to Geoff Keel for col- lecting and donating the specimen, and to Mark Florence and David Bohaska, Dept. of Paleobiology, National Museum of Nat- ural History, Smithsonian Institution (USNM), for processing the specimen and associated information. Photographs are by John Steiner, Smithsonian Photographic Services. Production of the figures was by Brian Schmidt, Division of Birds, Smith- sonian Institution. I thank Helen James for processing statistics and Scott Wing for cer- tain bibliographical references. For lending specimens of comparative material I am grateful to Mark Robbins, University of Kansas Museum of Natural History, Lawr- ence (KU); J. V. Remsen and S. W. Cardiff, Museum of Natural Science, Louisiana State University, Baton Rouge (LSU); Carla Cicero, Museum of Vertebrate Zoology, University of California, Berkeley (MVZ); David W. Steadman, Florida Museum of 736 Natural History, University of Florida, Gainesville (UF); and Kristof Zyskowski, Peabody Museum of Natural History, Yale University, New Haven (YPM). I thank Da- vid W. Steadman and Gareth Dyke for com- ments on the manuscript. Literature Cited Brodkorb, P. 1967. Catalogue of fossil birds. Part 3 (Ralliformes, Ichthyornithiformes, Charadrifor- mes).—Bulletin of the Florida State Museum, Biological Sciences 11:99—220. Fisher, J., and R. T. Peterson. 1964. The world of birds. Doubleday and Co., Garden City, New York, 288 pp. Gibson, T. G. 1983. Stratigraphy of Miocene through Lower Pleistocene strata of the United States Central Atlantic Coastal Plain; pp. 35—80 in C. E. Ray, ed., Geology and paleontology of the Lee Creek mine, North Carolina, I—Smithson- ian Contributions to Paleobiology 53. Hellmayr, C. E., and B. Conover. 1942. Catalogue of birds of the Americas, part 1, number 1. Zoo- logical Series Field Museum of Natural History 13, 636 pp. Houde, P. 1994. Evolution of the Heliornithidae: recip- rocal illumination by morphology, biogeogra- phy and DNA hybridization (Aves: Gruifor- mes).—Cladistics 10:1—19. Livezey, B. C. 1998. A phylogenetic analysis of the Gruiformes (Aves) based on morphological PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON characters, with an emphasis on the rails (Ral- lidae).—Philosophical Transactions of the Roy- al Society B 353:2077-2151. Manchester, S. R. 1999. Biogeographical relationships of North American Tertiary floras——Annals of the Missouri Botanical Garden 86:472—522. Marshall, L. G. 1985. Geochronology and land-mam- mal biochronology of the transamerican faunal interchange. Pp. 49-85 in FE G. Stehli and S. D. Webb, eds., The great American biotic inter- change. Plenum Press, New York, 532 pp. Olson, S. L. 1985. The fossil record of birds. Pp. 9— 238 in DSS, Farmer eR. Kanes and aiken Parkes, eds., Avian biology, vol. 8. Academic Press, New York, 256 pp. Olson, S. L., and P. C. Rasmussen. 2001. Miocene and Pliocene birds from the Lee Creek Mine, North Carolina. Pp. 233-365 in C. E. Ray and D. J. Bohaska, eds., Geology and paleontology of the Lee Creek mine, North Carolina, I1].—Smith- sonian Contributions to Paleobiology 90. Smith, A. G., D. G. Smith, and B. M. Fussell. 1994. Atlas of Mesozoic and Cenozoic coastlines. Cambridge University Press, 99 pp. Woodburne, M. O., and C. C. Swisher III. 1995. Land mammal high-resolution geochronology, inter- continental overland dispersals, sea level, cli- mate, and vicariance. Pp. 335-364 in W. A. Berggren, D. V. Kent, M.-P. Aubrey, and J. Har- denbol, eds., Geochronology, time scales and global stratigraphic correlations—SEPM (So- ciety for Sedimentary Geology) Special Publi- cation 54. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):737-741. 2003. The type specimen of Anoura geoffroyi lasiopyga (Chiroptera: Phyllostomidae) Joaquin Arroyo-Cabrales and Alfred L. Gardner (JA-C) Laboratorio de Arqueozoologia ““M. en C. Ticul Alvarez Solérzano,”’ INAH, Moneda # 16, Col. Centro, 06060 México, D.F, México; and Biodiversity Program Office, Smithsonian Institution, Washington, D.C. 20560, U.S.A.; (ALG) USGS Patuxent Wildlife Research Center, National Museum of Natural History, Washington, D.C. 20560-0111, U.S.A. Abstract.—In 1868, Wilhelm Peters described Glossonycteris lasiopyga, based on a specimen provided by Henri de Saussure and collected in México. The type specimen was presumed to be among those housed in the collections of the Zoologisches Museum of the Humboldt Universitat in Berlin, Germany. Our study of one of Saussure’s specimens from México, discovered in the collections of the Museum d’Histoire Naturelle, Geneva, Switzerland, dem- onstrates that it and not one of the Berlin specimens is the holotype. Henri de Saussure (Saussure 1993) wrote several reports on the animals he collected during his intensive explorations of central México at the middle of the 19th Century. Among Saussure’s specimens, which in- cluded several mammals and birds he de- scribed as new to science (Baud 1977), was a bat he identified as Anoura ecaudata Geoffroy (Saussure, 1860). He described the specimen, provided some measure- ments, and said it occurred in temperate and warm regions of México. The specimen was deposited in the mammal collections of the Museum d’Histoire Naturelle in Gene- ve, Switzerland. The same specimen was redescribed by Peters (1868) as the basis of a new genus and species he named Glossonycteris la- siopyga. A translation of Peters’ description follows: B. Medial upper incisors smaller than outer ones; lower incisors deciduous [p. 364]. a. Zygomatic arch developed. fegeeee'| b. Zygomatic arch absent. 6. Glossonycteris nov. gen. ect s-3/5-5 5 l/l 2-2 /2-2) be 3.3/5.5. molars having distinct W-shaped [p. 365] formation, the cingulum of the two next-to-last upper cheek teeth more developed. Interfemoral membrane very short, having hair on both sides, calcar very short. Tail? On the only dried specimen that I was able to examine, through the kind efforts of Mr. H. de Saus- sure, there was no trace of tail to be seen, and it is probably absent, however this can only be deter- mined with certainty from fresh or alcohol speci- mens. Both zygomatic arches were lacking on the skull and this appeared natural, but I am not entirely sure of this point since I have had no chance to examine an intact skull. I. Glossonycteris lasiopyga n. sp. (PI. fig. 2.) 21844. Choeronycteris peruana Tschudi, Fauna Peruana. I. P. 71. Pl. 3. Fig. 1. 2.—Archiv f. Naturgesch. I. p. 247. 1860. Anoura ecaudata Geoffroy, H. de Saus- sure, Guérin Rev. et Mag. Zool. p. 81. The specimen examined by Mr. de Saussure is the only one available to me for inspection, while Choeronycteris peruana Tschudi, which is perhaps identical with it, cannot be decided upon exactly, since Mr. Coulon of the Neuchatel Museum informs me that this species cannot be found among the Chi- roptera given to the museum by Mr. von Tschudi. The description and drawing given in Fauna Per- uana are not adequate for an exact opinion. I give here the measurements of the dried speci- men from Mexico, sent to me by Mr. de Saussure, as far as these could be determined. FORE AUDI ds stk 18S come te est am MAS Bw ve 41 Ee Of iotineer: mtcp. 3-0: "1st ph. 3.35. 2nd ph: 2.0 Ss SE Age ee ea em Pe aaa Ue ren ah 8.0 738 IL, Ol 3) invoyyers ionic, 2953 Wee joo 3.7/8 Ziel jola, 21.0; 3rd ph. 0.011; cartilage 3.7 L. of 4. fing. mtcp. 37.3; Ist ph. 1.0; 2nd ph. 13.3; cartilage 1.5 Lg Oi Dy iloyys waMiKEO, DSADS Wie oN, HA0§ Aavel jolm, Ids cartilage 1.5 Ta geeys OSs 24h a ee gore ae ae tan el 3.0 FPOOb eres Bis eek eee Oe er Tee aos 11.5 CalCalrein. oa caks ot oe Ee ee ere ee 3.0 Although there is no clear statement of the kind of specimen preservation, from Pe- ters’ (1868) text we know that he had at hand only one specimen. The specimen was a dried skin and nearly complete skull be- cause his description of Glossonycteris la- siopyga contains diagnostic characters from both the skin and the skull, and he illus- trated the skull as Fig. 2 in the accompa- nying Plate. The dental formula Peters (1868) gave for Glossonycteris includes three upper and three lower incisors. However, Anoura has only two upper incisors on each side and lacks lower incisors. We interpret this dis- crepancy as indicating that Peters was at- tempting to provide a more complete dental formula that included the deciduous teeth. We reached this conclusion because Peters stated that the lower incisors were decidu- ous (see beginning sentence of translation) and, as evident from his Fig. 2, the type of Glossonycteris lasiopyga \acked lower in- cisors. Peters also indicated three upper and lower incisors in the dental formulae for Lonchoglossa Peters, 1868 (another junior synonym of Anoura) and for Choeronycter- is Tschudi, 1844, a taxon that also has only two upper incisors on each side and lacks lower incisors. Clearly, the specimen described by Peters is the holotype. Thomas (1893) treated Glossonycteris lasiopyga as a junior syno- nym of Anura [= Anoura| geoffroyi. San- born (1933) was the first to use the current name combination Anoura geoffroyi lasiop- yga to refer to the subspecies whose range he gave as México, Guatemala, and El Sal- vador. He gave the type locality as southern México. Carter and Dolan (1978) discussed two PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON specimens of Anoura geoffroyi lasiopyga that Carter examined at the Zoologisches Museum der Humboldt-Universitat zu Ber- lin (ZMB), Berlin, Germany. These speci- mens, an adult male (ZMB 3565) and an adult whose sex is unknown (ZMB 3564), had been collected in Cuernavaca, Morelos, México, by Boucard in August 1866. Both specimens are labeled as “‘typus’’, and were considered by Carter and Dolan (1978) to be syntypes. However, a review of the mu- seum catalog (handwritten by Peters him- self), and of the specimens labels (second- ary ones), resulted in no confirmation of any kind of type status (R. Angermann, in litt., 28 November 2000). These specimens are not syntypes be- cause Peters (1868) mentioned only the specimen he borrowed from Saussure and which had been collected before 1860. Dr. Renate Angermann (in litt., 28 November 2000) was kind enough to examine copies of two letters in the archives of the Zoolo- gisches Museum der Humboldt-Universitat from Wilhelm Peters to H. de Saussure. One is dated 28 December 1865, in which Peters requested the loan of six specimens collected in México, including the speci- men Saussure (1860) had identified as An- oura ecaudata. The other five specimens were those described by Saussure as new: Vespertilio mexicanus, Molossus aztecus, Stenoderma tolteca, Tylostoma mexicana, and Macrotus mexicanus. The other letter is from March 1866, in which Peters said he was sending back the five specimens men- tioned above with his opinion about their taxonomic status, and which he indicated as: Vespertilio mexicanus Sauss. = V. lu- cifugus Leconte; Molossus aztecus Sauss. = Nyctinomus brasiliensis Geoffroy; Stenod- erma tolteca Sauss. = Dermanura ciner- eum Gervais; Tylostoma mexicana Sauss. = Trachyops (Vamp.) cirrhosus Spix; and Macrotus mexicanus Sauss. = Macr. wa- terhousii Gray. Apparently the specimen Saussure had identified as Anoura ecaudata was not returned at that time. Recently, a nearly complete skull of An- VOLUME 116, NUMBER 3 759 Fig. |. Photographs of the cranium and mandibles of MHNG 515.88, holotype of Anoura geoffroyi lasiop- yga: a) lateral view of cranium and mandible, b) ventral view of cranium, and c) dorsal view of mandible; horizontal bar = 10 mm. 740 oura geoffroyi was found in the mammal collections of the Museum d’Histoire Na- turelle, Genéve, Switzerland (Fig. 2). It is of an adult individual cataloged as MHNG 515.88 and bears a label linking it to Henri de Saussure. The skull is missing both bul- lae; part of the right temporal bone; and part of the exoccipital region including the basioccipital and basisphenoid. The dental formula is 2/0-1/1-3/3-3/3. The skull label contains the following information: Ob- verse—Museum Geneve/No. 515-88/Glos- sonycteris geoffroyi Gray/Localite Me- xique, ach. Sumichrast. Reverse—Saussure 1860 VII L.49?/monto M.L [written in pen- cil and difficult to read]. Cranial measurements (in mm) are as fol- lows: greatest length of skull, 24.8; inter- orbital breadth, 4.9; braincase breadth, 9.5; palate length, 13.2; maxillary toothrow length, 9.4; width across M2, 3.4; dentary length, 17.3; mandibular toothrow length, 9.7; condylocanine length, 16.6; condylo- molar length, 6.6; coronoid height, 3.8; dentary depth below M1, 1.4; mandibular condyle length, 1.6. Photographs (Fig. 1) of the skull and mandibles of this specimen essentially du- plicate the illustrations that Peters (1868: Fig. 2 and 2b) provided of the holotype of Glossonycteris lasiopyga; although the skull appears to have had additional dam- age since the original description. The sim- ilarities, augmented by the label informa- tion, support our conclusion that specimen MHNG 515.88 is the holotype of Glosso- nycteris lasiopyga Peters, 1868. Although this specimen was not documented by Baud (1977) in his catalogue of type specimens at the Geneve museum, it was recently identified as a potentially significant speci- men by Claude Weber (pers. comm., 1999). To date, the skin has not been located. We have been unable to determine the actual type locality of Glossonycteris la- siopyga. Saussure (1860:494) simply said the specimen came from “... les régions chaudes et tempérées... .”’ (warm and tem- perate regions) of México, and there is no PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON further information in the museum cata- logue. Saussure’s field collector was Fran- ¢ois Sumichrast, and there is no information in Saussure’s notes (Saussure 1993) sug- gesting that either he or Sumichrast col- lected in or near Cuernavaca, Morelos (the locality Carter and Dolan [1978] provided for the specimens they identified as syn- types in the Berlin museum), during Saus- sure’s expedition to México. Sumichrast collected in Orizaba, Puebla, México City, Tampico, and in a few other small villages in between (Boucard 1884, Sumichrast 1881). Of course, Sumichrast also collected extensively at Santa Efigenia, his finca and famous collecting locality located in Oa- xaca near the border with Chiapas. How- ever collecting at this site did not begin un- til after 1867 following the Mexican civil war (1855 to 1860; Panni 2001la) and the brief reign from 1861 to 1867 of the Aus- trian Archduke Ferdinand Maximilian as Emperor of México (Panni 2001b). In ad- dition, we know that the holotype of Glos- sonycteris lasiopyga is not the specimen from San Geré6nimo (Isthmus of Tehuante- pec) later recorded by Sumichrast (1881). The type specimen must have been collect- ed during Saussure’s trip to México be- tween 1855 and 1856. It definitely was not collected later than 1860 when Saussure (1860) reported the specimen Peters (1868) later described as a new taxon. Although not helpful to us for locating the type lo- cality of Glossonycteris lasiopyga, Alvarez (1963) provided a valuable discussion on the importance of Sumichrast’s (1881) re- port for determining some of the actual type localities for several of the species Saussure described from México. Based on the knowledge that the specimen probably came from somewhere along the Atlantic slope of central México where Saussure did most of his field work, we further restrict the type locality to the State of Veracruz. Acknowledgments We thank Prof. Fran¢ois J. Baud for per- mitting the study of the mammal specimens VOLUME 116, NUMBER 3 under his care, as well as providing the nec- essary support for undertaking the study of Mexican mammal specimens in the collec- tions of the Département de Mammalogie et Ornithologie, Museum d’Histoire Natu- relle, Geneve, Switzerland. We are also very grateful to him for arranging for the photography of some of the type specimens. Prof. Claude Weber provided us with many of his unpublished notes on the type spec- imens in the mammal collection. Mr. C. Ratton took the photographs for this study. Our friends and colleagues Albérico No- gueira de Queiroz and Olivia Alexandre de Carvalho hosted JA-C’s visit to Geneve. Dr. Don Wilson approved the necessary funds to travel to Europe for JA-C, as well as all of the support from the Smithsonian Insti- tution (then through the Office of Biodiver- sity Programs). Dr. Renate Angermann gra- ciously provided information from the cop- ies of the difficult-to-translate, hand-written letters sent from Peters to Saussure. Literature Cited Alvarez, T. 1963. The type locality of Nyctomys sum- ichrasti Saussure.—Journal of Mammalogy 44: 582-583. Baud, FE 1977. Catalogue des types de mammiféres et d’oiseaux du Muséum d'Histoire naturrelle de Genéve.—Revue suisse de Zoologie 84:201-— PDD: Boucard, A. 1884. Notice sur Frangois Sumichrast.— 741 Bulletin de la Société Zoologique de France 9: 305-312. Carter, D. C., & P. G. Dolan. 1978. Catalogue of type specimens of Neotropical bats in selected Eu- ropean museums.—Special Publication, The Museum, Texas Tech University 15:1—136. Panni, E. 2001a. La Guerra Civil, 1858-1860. Pp. 21— 40 in J. Garciadiego, coord., Gran Historia II- ustrada de México. IV. De la Reforma a la Re- volucion, 1857-1920. Editorial Planeta Mexi- cana, México, 400 pp. Panni, E. 2001b. La Intervencion y el Segundo Im- perio, 1861—1867. Pp. 41—60 in J. Garciadiego, coord., Gran Historia Ilustrada de México. IV. De la Reforma a la Revoluci6n, 1857—1920. Ed- itorial Planeta Mexicana, México, 400 pp. Peters, W. 1868. Uber die zu den Glossophagae ge- h6rigen Flederthiere und tiber eine neue Art der Gattung Coléura.—Monatsberichte der K6nig- lich-Preussischen, Akademie der Wissenschaf- ten zu Berlin 1869:361—368. Sanborn, C. C. 1933. Bats of the genera Anoura and Lonchoglossa.—Field Museum of Natural His- tory, Zoological Series 20:23—27. Saussure, H. de. 1860. Note sur quelques mammiféres du Mexique.—Revue et Magasin de Zoologie Pure et Appliquée, Série 2, 12:1—11, 53-57, 97— 110, 241-254, 281-293, 377-383, 425-431, 479-494, 4 pls. [Also printed as an indepen- dently-paginated, 82-page separate. ] Saussure, H. de. 1993. Voyage aux Antilles et au Me- xique 1854—1856. Présenté par Louis de Roguin et Claude Weber. Editios Olizane, Genéve, Suis- See iSapp: Sumichrast, EF 1881. Enumeracion de las especies de mamiferos, aves, reptiles y batracios observados en la parte central y meridional de la Republi- ca.—Mexicana. La Naturaleza 5:199-213. Thomas, O. 1893. On some mammals from central Peru.—Proceedings of the Zoological Society of London 1893:333-341, plates 28-29. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):742-753. 2003. Observations on the structure of the mandibular gnathobase in some American Mesocyclops (Copepoda: Cyclopidae) E. Suarez-Morales, M. A. Gutiérrez-Aguirre, and M. Elias-Gutiérrez El Colegio de la Frontera Sur (ECOSUR) Unidad Chetumal. A.P. 424. Chetumal, Quintana Roo 77000. Mexico, e-mail: esuarez@ecosur-qroo.mx Abstract.—The mandibular edge of males and females of several American species of the cyclopoid genus Mesocyclops was examined. The general struc- ture, arrangement, and size of teeth was assessed for each species and shown to be variable among them. We used a modified version of a previous index, the Gnathal Index (GI) plus the Mandibular Size Index (MSI), and the Man- dibular Power Index (MPI) which, together are expected to 1) provide a quan- titative measure of morphologic variation amongst species and 2) suggest the feeding habits of these species. All species showed different values in terms of teeth number, size, and arrangement, as indicated by results of these indices. Known predatory species (1.e., M. edax, M. longisetus, and M. aspericornis) showed high values of MSI and MPI. The MSI values were positively corre- lated to the total body length. Our data suggest a predatory capability for these two species. Mesocyclops yutsil, M. reidae, and M. chaci had the lowest MPI and MSI values; they represent a group that probably are detritivores. The remaining species, all with lower MPI values, are probably omnivores. For the species examined here, most males are smaller than females but their mandibles tend to be similar in relative size and structure. Experimental observations support our conclusions about the predatory forms. Feeding experiments are needed to determine the limits of these indices for most species in the second and third groups. Knowledge of the mandibular structure in the species of the freshwater copepods genus Mesocyclops is limited to the basic descriptions. There are no previous com- parative studies about the morphology and specific variation of the mandible structures in freshwater copepods. Mandibles, togeth- er with other cephalic appendages, are di- rectly involved in food handling and inges- tion; hence, different feeding habits may be expected to correlate with particular man- dible types. Most species of the freshwater cyclopid genus Mesocyclops are omnivorous. How- ever, the role of some Mesocyclops as pred- ators of mosquito larvae and consequently as biological controls has been studied for several years in the neotropical and nearctic regions (Marten 1989; Marten et al. 1989, 1994a, 1994b), and in several Asian coun- tries (Sinh Nam 2000) suggesting that these species may be carnivores. However, there are no surveys dealing with the morpholog- ical bases of their capacity for carnivory in terms of the structure and size of their mouthparts, particularly of their mandibles. The morphological structure of marine cal- anoid copepod feeding appendages (1.e., the cutting edges of the mandible or the struc- ture of the maxillar setae) is a too! useful to infer their feeding habits (Itoh 1970). This aspect has not been explored in fresh- water cyclopoid copepods, a group found in a wide range of continental environments and with different feeding habits. There are ten species and one subspecies VOLUME 116, NUMBER 3 743 Table 1.—Gnathal Index (GI), Mandibular Size Index (MSI), and Mandibular Power Index (MPI) obtained from the analysis of females and males of some American species of Mesocyclops. Species arranged alphabet- ically; numbers between parentheses indicate the rank of the MPI value. Species GI MSI MPI M. aspericornis (Daday, 1906) (2) 67.9 4.16 282.4 (2) M. brasilianus Kiefer, 1936 (2) 63.6 2.79 V2: CG) M. brasilianus (3d) 60.6 2.92 176.9 (11) M. chaci Fiers, 1996 (¢ ) Siz 2.05 104.5 (15) M. edax (Forbes, 1891) (2) 82.4 4.51 BIG (A) M. evadomingoi Gutiérrez-Aguirre, 2001la (2) 75.6 2.41 182.1 (8) M. longisetus s.str. (Thiébaud, 1912) (¢) O25 BAS) 2035) M. longisetus curvatus Dussart, 1987 (@) V3 2.54 186.4 (7) M. pescei Petkovski, 1986 (@) 85.0 256 DNS) (5) M. pescei (d) 63.1 2.79 NWS. ik (CD) M. residae Petkovski, 1986 (2) WSO DOP Silks) Cl) M. reidae (¢) 80.2 DMS 178.4 (9) M. thermocyclopoides (Harada, 1931) (2) TRY) 2.63 191.7 (6) M. thermocyclopoides (¢) 87.0 2.719 234.8 (3) M. yutsil Reid, 1996 (¢) VY 1.92 149.5 (14) of Mesocyclops known to be distributed in Mexico (Suarez-Morales & Reid 1998, Gu- tiérrez-Aguirre & Sudarez-Morales 2001a, 2001b; Fiers et al. 2000), representing close to 60% of the species currently known from the neotropics (Gutiérrez-Aguirre & Sua- rez-Morales 2001b). In this work, analyzes the structure of the mandibular edge of ten species and one subspecies of Mesocyclops, all of them known to be distributed in Mex- ico and Central America, and some found also in North America. The mandible edge structure and relative size are determined for the species of Mesocyclops examined, and variations among species are discussed in relation to their feeding habits. A quan- titative aspect of this question is analyzed through different indexes, one of them de- rived from a formula first proposed by Itoh (1970) for the evaluation of the mandible structure of marine calanoid copepods. Methods Specimens were obtained from field col- lections of zooplankton in different fresh- water environments of Mexico. The meth- ods of collection and sampling sites in Mexico are described in Suarez-Morales et al. (1996) and in Gutiérrez-Aguirre and Suarez-Morales (2001la). Additional type and non-type material was requested on loan to different museums harboring collec- tions of American Mesocyclops, the Nation- al Museum of Natural History, Smithsonian Institution at Washington, D.C. (USNM), the Museum National d’ Histoire Naturelle, Paris (MNHN-Cop), and EI Colegio de la Frontera Sur, Unidad Chetumal, Chetumal, Mexico (ECO-CHZ) (see Sudarez-Morales & Gutiérrez-Aguirre 2001). Female speci- mens of eleven species or subspecies plus males of four species of Mesocyclops rec- ognized in Mexico and Central America were analyzed (see Table 1). Examination followed dissection of the mandibles and included camera-lucida illustrations of the mandibular edge detailing the teeth width, height, number, and separation. Mandibular morphology.—In general, the morphological interpretation proposed by Huys & Boxshall (1991) was followed. The gnathobase is a ventral extension of the coxa; it has a variable number of non-artic- ulation elements, that are the teeth. The mandible edge bears a proximal seta and in some cases an inner, flexible setiform ex- tension (named “inner”? proximal tooth). Both elements, the proximal seta and the 744 10 um i \ SS cx inner dorsal tooth y iE q Ng x e accessory G\\ eA teeth \G ANS n Pteay i Pe | denticles outer dorsal seta 10 um Fig. 1. Hi ky " [ i A ed W PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 10 um OU 10 um Structure and nomenclature of the mandibular edge of males and females of species of Mesocyclops from Mexico: A) M. yutsil (2), showing factors used in index formulae; B) M. chaci (2); C) M. edax (@), showing nomenclature of mandibular structures. Outer dorsal seta and inner dorsal tooth are proximal; blade is distal; D) M. aspericornis (2); E) M. evadomingoi (@ ). inner tooth, can have epicuticular exten- sions (such as setules or as pinnate elements mostly uniserially arranged). In some spe- cies these extensions are noticeably strong, we named these structures ‘accessory teeth’? (see Fig. 1C). In Mesocyclops, the distal edge of the gnathobase has a notice- ably larger tooth; it can be simple (as in Fig. 1A) or formed by a cluster of two or more teeth (see Fig. 1C, E). This structure is known herein as the blade (see Fig. 1C), but it was counted as one tooth and also is used as a reference for measurements of the other teeth. There are setiform structures along the mandible edge, arising near the base of the teeth; these are not true setae, and are named denticles for the purposes of this work (Fig. 1C). The same nomencla- ture was used in the descriptions of the mandibular structure. Formulae.—The mandibular elements (teeth number, edge width, height) were VOLUME 116, NUMBER 3 evaluated with an index based on the Edge Index (EI) proposed by Itoh (1970) for ma- rine calanoid copepods. The original for- mula of this EI is as follows: EI = & (wi/W X hi/H X 10+) X 1/N where W is the width of the edge of the gnathobase, measured from distal tooth to proximal seta, and wi the width of the space between each pair of adjacent teeth, respec- tively. Factor H is the height of the main tooth (blade) and hi the height of each re- maining tooth, respectively (Itoh 1970), and N is the number of teeth. We modified this formula substituting the meaning of wi by the actual width of the teeth (not the space between each tooth as in the Itho’s formula) and rearranged the factors. The name of the resulting value was changed herein to gna- thal index (GI), the resulting formula as: GI = > ((wi/W X 1/N)(hi/H X 1/N)) x 1000 where wi is the width of each tooth, not the space between them and W is the width of the edge of the gnathobase; hi is the height of each tooth and H is the maximum height of the blade tooth (see Fig. 1A). The first factor indicates the width of each tooth compared to the total edge width, the sum of all individual tooth widths gives an idea of the how much of edge of the gnathobase is take up by teeth. A value of 1 would imply that all the available edge is covered by teeth. However, for some species the factor figure is over 1, if teeth are arranged in more than one row. Dividing the result of this factor by the number of teeth, we obtain the average tooth width. The second factor iss designed to indicate how high teeth are when compared to the highest one (the blade); dividing this product by the number of teeth provides an average tooth height. The overall product of multiplying the average height and width is an idea on how strong each tooth is in the mandible, the higher and wider, the stronger. Addi- tionally, we formulated a mandibular size index (MSI) to quantify the length of the edge of the gnathobase as it relates to the 7TA5 total body length of the specimen. This in- dex is obtained from the following formula: MSI = W/TL X 100, where W is the width of the mandible edge (in wm) and TL the length of the species (Gn pm) including the caudal rami. This index is expected to pro- vide a comparative estimation on how large are the mandibles with respect to the size of the copepod. The figure obtained is a percentage, scaled to 100. The relation be- tween TL and MSI was analyzed statisti- cally by determining the correlation factor (r). Finally, by multiplying GI by MSI, the mandibular power index (MPI) is obtained, which provides a measure of the strength- size combination of the mandibles of each species. That is, a species with a high GI will not necessarily have a high strength- size combination, if the width of the blade is small relative to the length of the speci- men (as indicated by the MSI). Results Material examined.—Mesocyclops asper- icornis: 3 adult 22 from small reservoir near km 90 of the highway Culiacan-Los Mochis, Sinaloa, Mexico (25°17'N, 107°47'W); 1 adult 2 from Anapoima, Cun- dinamarca, Colombia (04°33'N, 074°32'W), USNM-216634; M. brasilianus: 6 adult 2 2, 2 adult dd from Sayaxché, Guatemala (16°31'57’”N, 90°11'18”W). 1 adult 2, Man- tecal, Venezuela, Collection of B. Dussart, MNHN Cp 821, | adult ¢, Lago Valencia, Venezuela (10°10’N, 0.67°45'00’"W), USNM-204662; M. chaci: 1 adult 2, 1 adult 6, Gruta Tza-Ma, Yucatdn, Mexico (USNM-274244): M. edax, 2 adult ° 2 from Cranes Pond, North Carolina, USA, | adult 2, Cenote Viejo, central Quintana Roo, Yu- catan Peninsula, Mexico, USNM-259699; M. evadomingoi, holotype specimen, pond at km 45 of the Jonuta-Villahermosa federal onl, Walowireo, IMilevaete) (C7 sans ZM Ine 92°14'11”"W), slides ECO-CHZ 01157 (2), paratype ECO-CHZ 01159 (d6); M. longi- setus Ss. str.: 4 adult 2 2 from small temporal pond near Comitan city, Chiapas, Mexico 746 (iG309' S57 INGV9220523) Wee adult? Sen2 adult ¢¢, pond at Km 51 Jonuta-Villaher- mosa federal road, Tabasco, México (18°23'16"N, 92°47'0"W); M. longisetus cur- vatus: 3 adult 22 from small pond at km 45, Jonuta-Villahermosa federal road, Tabas- co, Mexico (17°58'44"N, 92°14’11"W), 3 adult 2 2, Cenote Catedrales, central Quin- tana Roo, Yucatan Peninsula, Mexico, USNM-259687; M. pescei: 11 22,5 66 from small pond km 45, Jonuta-Villaher- mosa federal road, Tabasco, Mexico (17°58'44"N, 92°14’11"W); M. reidae: | adult 2, 1 adult ¢ from Tabano, Cuba; M. thermocyclopoides: 3 adult 22, 1 adult 3 from Pulsar, Tabasco, Mexico (17°39'10’N, 91°33'23"W), 2 adult 2 2, 2 adult dd, from small pond at km 45 of the Jonuta-Villaher- mosa federal road, Tabasco, Mexico (17°58'44'N, 92°14’11"W); M. yutsil: 2 adult 2 2 (USNM-259843, USNM-259846)) from cenote Yuncu, Yucatan Peninsula, Mexico; 1 adult 2, Cenote Mucuyché, Yucatan, Mex- ico (USNM-259848). Structure of mandibles.—The mandibu- lar edge in this genus follows a general pat- tern, with a strong blade which is, in most cases, the largest and strongest one on the gnathal edge (Fig. 1). In some species such as M. evadomingoi (2), M. pescei (3, &), and M. thermocyclopoides (2), the blade is formed by a cluster of two or more teeth. A row or rows of smaller teeth follow to- ward the proximal end. Teeth may be bi- cuspidal or multicuspidal. Teeth have a var- iable basal width and may have groups or rows of small denticles at their base. In some species, the inner proximal tooth is quite strong and armed with two or more internal accessory teeth (i.e., M. edax, M. longisetus s.str.). The structure of the blade showed some range of variation in the spe- cies of Mesocyclops studied (Figs. 1, 2). A brief description of the mandibular mor- phology of each species and sex examined is given below together with the percent variability of the indexes estimated; this is provided only for those species in which we could evaluate these variation. Total num- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ber of teeth includes the blade and the inner proximal tooth. GI, MSI, MPI are given in Table 1. When material was available to de- termine the variability of the indices, the percentage range of variation from the av- erage is indicated in each case for the spe- cies. Mesocyclops aspericornis (¢): Gnatho- base with 10—11 wide-based teeth, all mon- ocuspidal. Distal blade of three teeth. A sin- gle row of 6 denticles inserted near base of teeth on central part of gnathal edge. Prox- imal seta spinulated; inner proximal tooth with row of four accessory teeth (Fig. 1D). Variability in percent with respect to aver- age: GSI (11.9%), MSI (8.8%), MPI (18.4%). Mesocyclops brasilianus (¢): Gnatho- base with 8 wide-based teeth, all teeth mon- ocuspidal. Two paired sets of denticles in- serted near base of teeth on central part of gnathobase. Proximal seta slender, long, with row of setules on inner margin (Fig. 2B). Inner proximal tooth naked. Variability in percent with respect to average: GSI (14.1%), MSI (9%), MPI (5.1%). Mesocyclops brasilianus (6d): Gnatho- base with 4 wide-based teeth, all teeth mon- ocuspidal. Blade a cluster with two distal teeth. Two groups of denticles inserted near base of teeth on central part of gnathobase. Proximal seta short, lightly setulated (Fig. 2A). Variability in percent with respect to average: GSI (12.3%), MSI (8.7%), MPI (19.3%). Mesocyclops chaci (¢): Gnathobase forming cluster of 3—4 wide-based teeth, proximal seta biserially pinnate. Inner prox- imal tooth slender, setiform, next proximal tooth tetracuspidal, next one monocuspidal, next one bicuspidal, distal monocuspidal, blade formed by a cluster of three teeth. Row of six denticles inserted near base of bicuspidal tooth on central part of gnatho- base (Fig. 1B). Mesocyclops evadomingoi (@): Gnatho- base wide, with 12 wide-based teeth, all monocuspidal. Distal cluster of three teeth partially fused to blade. Row of five den- VOLUME 116, NUMBER 3 TAT | 1oum J Fig. 2. Structure of the mandibular edge of males and females of species of Mesocyclops from Mexico; A) M. brasilianus (3); B) M. brasilianus (2); C) M. pescei (3); D) M. pescei (2); E) M. longisetus curvatus (&femlae;); F) M. longisetus s.str. (2); G) M. thermocyclopoides (3); H) M. thermocyclopoides (2); 1) M. reidae (2): J) M. reidae (cd). 10 um 748 ticles inserted near base of teeth on central part of gnathobase. Outermost proximal seta long, with uniserial row of spinules along inner margin; inner proximal tooth relatively short, with row of 3—4 secondary teeth on inner margin (Fig. 1E). Variability in percent with respect to average: GSI (4.4%), MSI (3.4%), MPI (0.3%). Mesocyclops edax (2): Gnathobase with 13 wide-based teeth, all monocuspidal. A single row of 6—7 denticles inserted near base of the teeth on central part of the gna- thal edge. Distal cluster of three strong teeth clustering partially with blade. Proximal seta heavily spinulated along inner margin only. Inner proximal tooth with 2—3 strong accessory teeth (Fig. 1C). Variability in per- cent with respect to average: GSI (4.7%), MSI (2.3%), MPI (4.5%). Mesocyclops longisetus s.str. (@): Gnathobase with 8—10 wide-based teeth, all teeth monocuspidal. Row of five denticles inserted near base of teeth on central part of gnathobase. Proximal seta relatively long, slender, with inner row of short set- ules. Inner proximal tooth with three strong accessory teeth (Fig. 2F). Variability in per- cent with respect to average: GSI (20.6%), MSI (2.9%), MPI (22.3%). Mesocyclops longisetus curvatus (2): Gnathobase with 10—11 wide-based teeth, all teeth monocuspidal. Blade with blunt tip. Proximal seta relatively long, slender, with inner row of short spinules. Inner proximal tooth naked, shorter than the proximal seta (Fig. 2E). Variability in per- cent with respect to average: GSI (14.4%), MSI (16.4%), MPI (27.4%). Mesocyclops pescei (¢): Gnathobase with 12 teeth, all monocuspidal. Base of teeth unornamented. Distal cluster of four teeth partially fused with blade. Proximal seta long, strongly spinulated along inner margin. Inner proximal tooth slender, na- ked, slightly longer than proximal seta (Fig. 2D). Variability in percent with respect to average: GSI (15%), MSI (6.9%), MPI (Ga): Mesocyclops pescei (3): Gnathobase PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON with 11 wide-based teeth, all monocuspidal. Base of teeth unornamented. Blade partially fused to two adjacent teeth. Proximal seta relatively long, with row of spinules along inner margin. Inner proximal tooth slender, slightly shorter, with row of small, weak ac- cessory teeth along inner margin (Fig. 2C). Variability in percent with respect to aver- age: GSI (37.4%), MSI (15.5%), MPI (ZD.D%)). Mesocyclops reidae (°): Gnathobase with 10 wide-based teeth, all monocuspidal. Base of teeth with single row of four den- ticles inserted on central surface of base. Distal cluster of 2—3 teeth with blade. Prox- imal seta relatively long, with row of short spinules along inner margin. Inner proximal tooth slender, slightly shorter than proximal seta, with three accessory teeth on inner margin (Fig. 11). Mesocyclops reidae (d): Gnathobase with 9 wide-based teeth, all monocuspidal. Base of teeth with row of 2-3 denticles. Distal cluster of three teeth with blade. Proximal seta with row of short setules along inner margin. Inner proximal tooth slightly shorter, naked (Fig. 1J). Mesocyclops thermocyclopoides (@): Gnathobase with 12—13 wide-based teeth, all teeth monocuspidal. Row of four denti- cles inserted near base of teeth on central part of gnathobase. Distal cluster of 3—4 teeth partially fused with blade. Proximal seta relatively long, slender, with inner row of strong spinules. Inner proximal tooth slender, naked (Fig. 2H). Variability in per- cent with respect to average: GSI (10.3%), MSI (17.9%), MPI (7.6%). Mesocyclops thermocyclopoides (<4): Gnathobase with 8 teeth, all teeth mono- cuspidal. Blade clustered with one tooth. Two or three denticles inserted near base of teeth on central part of gnathobase. Proxi- mal seta relatively long, slender, naked. In- ner proximal tooth absent or reduced (Fig. 2G). Mesocyclops yutsil (2): Gnathobase with 4 wide-based teeth, including large blade; all moncuspidal. Row of five denticles in- VOLUME 116, NUMBER 3 749 1.2 Mesocyclops 0.8 7) = 06 5 = we = Females O Estimated values A a5 Males 0 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7 Teal UP 13 Ln TL Fig. 3. MSI vs. TL logarithmic (In) regression. Ellipses indicate two major groups with distinct features (see text por explanation) including both males (triangles) and females (squares). Arrow indicate the species with highest TL and MSI values. Species abbreviations as: Ma = M. aspericonis; Mb = M. brasilianus; Mc = M. chaci; Med = M. edax; Mev = M. evadomingoi; Mis = M. longisetus sensu stricto; Mlc = M. longisetus curvatus; Mp = M. pescei; Mr = M. reidae; Mt = M. thermocyclopoides; My = M. yutsil. serted diagonally near base of teeth on cen- tral part of gnathobase. Proximal seta rela- tively short, as long as adjacent teeth, with inner row of short spinules (Fig. 1A). Discussion The Mandibular Power Index (MPI) of species of Mesocyclops comparable numer- ical data which are presented in Table 1. The highest MPI value was that of Meso- cyclops edax, followed by M. aspericornis, whereas the lowest figure was shown by the females of M. yutsil and M. chaci. Addi- tional differences were detected between males and females of the same species. Out of the four species with both sexes evalu- ated, males of two (M. reidae, M. thermo- cyclopoides) had higher MPI values than females; females of M. brasilianus and M. pescei had an MPI figure higher than their males (see Table 1). The Mandibular Size Index (MSI), size of the gnathal edge relative to body length of the specimen, yielded interesting differ- ences among the species. Females M. edax have the largest mandibular edge (MSI = 4.5), followed by females of M. aspericor- nis (4.16), and females of M. longisetus s.str. (MSI = 3.26) (see Table 1). Females of Mesocyclops yutsil, M. chaci, and M. rei- dae have the smallest mandibles, relative to body size, of all the species examined here- in (MSI = 1.92 and 2.05, respectively). The TL vs. MSI graphic analysis, which used the logarithm (In) of both factors, showed two distinct groups (see Fig. 3) with a positive correlation (r = 0.598, p = >0.05) which is slightly higher if only fe- 750 males are considered (r = 0.62, p = >0.05). Group A includes the females of most species and group B contains the four males examined and the females of M. cha- ci and M. yutsil. The forms with highest values in terms of MSI and TL were the females of M. aspericornis, M. edax, M. longisetus s.str., and M. longisetus curvatus (arrowed in Fig. 3). Overall, the morphol- ogy of the mandibular edge of the studied Mesocyclops shows a relatively wide range of variation in terms of tooth structure. Itoh’s (1970) Edge Index was modified due to the differences between the calanoid teeth/mandibular pattern and the cyclopoid pattern (see Huys & Boxshall 1991), mainly with respect to the distribution of teeth along the gnathal edge. The two main fac- tors (width and height), when combined, provide an idea of the strength of the the teeth. Then, when divided by the number of teeth (N), a quantitative measure is obtained about the strnngth of each tooth, which is the actual interpretation of this index for Mesocyclops. These figures also provide a quantitative estimation of the morphologi- cal differences found in the species exam- ined. Although the taxonomic value of the mandible edge remains unexplored in Me- socyclops or in any other freshwater genus, this work reveals the many different char- acters with potential use for identification purposes, and that most of them can be evaluated quantitatively. Mandibles are one of the most heavily chitinized structures in the copepods; therefore, these appendages could be used to identify these crustaceans at the species level from stomach contents in trophic or ecological surveys. The size of the body (LT) is a key factor for a copepod used in mosquito control, but one of the largest species of Cyclopidae known (Homocyclops ater (Herrick)) can not be used as a mosquito predator (see Marten et al. 1994b). Therefore, the use of the MSI together with the TL was consid- ered necessary to infer the trophic habits of the species of Mesocyclops. The MSI vs. TL graphic (Fig. 3) yielded some interest- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ing facts: the relative size of the mandibular edge tends to be proportional to the total length of the species; males are smaller but at least in two of the males evaluated (M. pescel, M. reidae), mandibles are at least as large as those of their females. The value of males as potential predators of mosquito larvae is undetermined, although in this study the male of M. thermocyclopoides ranked higher (MPI) than the female. The females of Mesocyclops longisetus, M. aspericornis, and M. edax are unique in showing a proximal seta with 2—4 strong, wide-based accessory teeth. They had the highest MSI values which, complemented with their body size (1.0—1.2 mm) suggests they are the most well-equipped species for predation within the examined group (see Fig. 3). Overall, these results agree with those of field and laboratory experiments testing the predation capabilities of M. as- pericornis and M. longisetus (see Marten et al. 1994b, Suarez 1992) and of M. edax (Marten 1989). Mesocyclops thermocyclo- poides, recorded in Honduras, Costa Rica, and Mexico, has been successfully used as a mosquito control (Marten et al. 1994a, 1994b). Only recently were the neotropical records of this species confirmed with Asian specimens (Gutiérrez-Aguirre et al. 2003). M. thermocyclopoides ranked sixth in the MSI; its size (comparable to that of M. aspericornis and M. longisetus curvatus) suggests that this species might be capable of attacking a mosquito larva. Our results on this species agree entirely with the ob- servations obtained from experimental works by Kumar & Rao (1999a, 1999b) es- tablishing this species as an omnivorous form feeding on rotifers, ciliates, and algae. A second group of probably omnivorous species was found based on MSI vs TL val- ues and MPI, 1.e., the females of M. eva- domingoi, M. thermocyclopoides, M. pes- cei, and M. brasilianus (see Fig. 3). MSI values range between 2.2 and 2.8; they are medium to large-sized forms (0.75 mm-— 1.00 mm), most with small to medium- sized mandibles (except for M. brasilianus, VOLUME 116, NUMBER 3 a species with a mandible size equal to that of predator species). These omnivorous species may be capable of predation upon small prey items when conditions are ap- propriate to such behavior, but they may also ingest other food sources. Males of these species, some even with MSI values higher than their females, may not be able to feed on large prey (i.e., mosquito larvae) because of their relatively small size. Fi- nally, the females of M. chaci and M. yutsil are slender, probably planktic forms (Fiers et al. 1996). In terms of size of the man- dibular edge and teeth structure, these two species represent the diametral opposite end with respect to the predator group. Their mandibles are built quite differently, they probably handle algae, suspended organic matter along the water column, or even the heavily chitinized edge with these short, very solid teeth (shown by the top GI val- ues of all the examined group) could be used to scrape off food from littoral surfac- es or vegetation (Gf they are epibenthic forms). According to our results, the examined species of American Mesocyclops exhibit a wide variation of the mandibular size and armament. We speculate that at least the morphological extremes (M. yutsil—M. as- pericornis) are on correspondence with dif- ferent feeding habits and prey sizes. There- fore, not all the species of Mesocyclops are suitable to be developed or cultured as po- tential biological controls of mosquitoes. It is clear that the evaluation of the teeth strength through the GI alone could be mis- leading, a species with very strong teeth will rank relatively high in this index (i.e., M. yutsil) but this does not necessarily im- ply a predatorial capacity. This is why the GI should be complemented with the MSI and the MPI. A MSI value over 3 would represent a species conveniently armed for predation. We recognize, however, that the relatively reduced number of observations (see the Material examined section) is a drawback of this analysis, but still, our re- sults seem to make sense in terms of 1) the 751 estimated high capacity of three well- known predator species and their consistent tendency to cluster together in the three in- dices used, 2) the distinction of different and even contrasting structural-morpholog- ical patterns within the genus, 3) the rela- tion of this pattern with experimentally test- ed feeding habits (for predators and one omnivorous species), and 4) the relative body and mandibular size differences among the species examined herein. Of course, these patterns and estimations should be complemented with additional observations in order to have a more robust numerical analysis. Using the detailed drawings provided in redescriptions by Dahms & Fernando (1993), we extended our results of mandib- ular indices to other species and genera for which feeding habits are known or inferred. Mesocyclops leuckarti (Claus, 1857) is a predator species (Marten 1994). The MSI estimated for this species is 4.06 and the MPI is 315.8; these values are within the range value of other predator species we have detected in this work. Although these indices were designed for Mesocyclops, we tested our methods on Acanthocyclops brevispinosus (Herrick, 1884) of the “‘ro- bustus-vernalis” group complex (Dodson 1994). Recently, it was redescribed by Dahms & Fernando (1997). It has a MSI of 3.82, a value which suggests predating hab- its. Eucyclops conrowae Reid, 1992), a sup- posedly herbivore, had a GI of 65.1, a MSI of 2.7, and a MPI of 175; these figures cat- egorize this species within the omnivorous forms. In these or in the other instances, experimental observations will provide valuable data to find a link between the mandible structure and the feeding habits of freshwater copepods. It is probable that the differences among the species categorized here as omnivorous forms, regardless of the genus, rely mainly on the size of the poten- tial prey they are able to capture and handle (either protozoans, planktonic algae, or in- sect larvae). Literature Cited Deday, E. von. 1906. Untersuchungen uber die Cope- podenfauna von Hinterindien, Sumatra und Java, nebst einem Beitrag zur Copepodenkennt- nis der Hawaii-Inseln. (Reise von Dr. Walter Volz). Zoologischer Jahrbuch, Abt. Syst., Geogr. Biol. Thiere 24(3):175—206. Dahms, H. U., & C. H. Fernando. 1993. Redescription of Mesocyclops leuckarti (Copepoda, Cyclopoi- da), including a study of its naupliar develop- ment.—Internationale Revue der gesamten Hy- drobiologie 78:589—609. . 1997. Redescription of Acanthocyclops brev- ispinosus (Herrick, 1884) (Copepoda, Cyclo- poida) from Ontario.—Crustaceana 70:129— 144. Dodson, S. 1994. Morphological analysis of Wisconsin (U.S.A.) species of the Acanthocyclops vernalis group (Copepoda: Cyclopoida).—Journal of Crustacean Biology 14:113—131. Dussart, B. H. 1987. Sur quelques Mesocyclops (Crus- tacea, Copepoda) d’ Amerique du Sud.—Ama- zoniana 10:149—-161. Fiers, E, J. W. Reid, T. M. Iliffe, & E. Suarez-Morales. 1996. New hypogean cyclopoid copepods (Crustacea) from the Yucatan Peninsula, Mexi- co.—Contributions to Zoology 66:65—102. , V. Ghenne, & E. Suarez-Morales. 2000. New species of continental cyclopoid copepods (Crustacea, Cyclopoida) from the Yucatan Pen- insula.—Studies of Neotropical Fauna & Envi- ronment 35:209-—251. Forbes, S. A. 1890 [1891]. On some Lake Superior Entomostraca. Annual Report for the Commis- sion of Fish and Fisheries 1887. Government Printing Office, Washington, D.C., 701-718, pls. I-IV. Gutiérrez-Aguirre, M., & E. Suarez-Morales. 2001la. A new species of Mesocyclops (Copepoda, Cyclo- poida, Cyclopidae) from southeastern Mexi- co.—Journal of Limnology 60:143-154. . 2001b. Distribution and taxonomy of the trop- ical American Mesocyclops Sars, 1914 (Copep- oda, Cyclopoida).—Crustaceana 74:477—487. , J. W. Reid & E. Suarez-Morales. 2003. An Afro-Asian species of Mesocyclops (Copepoda: Cyclopoida) in Central America and Mexico.— Journal of Crustacean Biology 23. (in press). Harada, I. 1931. Studien tiber Siisswasserfauna For- mosas.—Annotationes Zoologicae Japonenses 13:149-168. Itoh, K. A. 1970. A consideration on feeding habits of planktonic copepods in relation to the structure of their oral parts.—Bulletin of the Plankton So- ciety Japan 17:1—10. Kiefer, E 1936. Brasilianische Ruderfusskrebse (Crus- tacea, Copepoda), gesammelt von Herrn Dr. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Otto Schubart. V. Mitt—Zoologischer Anzeiger 116:31-35. Kumar, R., & T. R. Rao. 1999a. Effect of algal food on animal prey consumption rates in the omniv- orous copepod, Mesocyclops thermocyclopo- ides.—International Revue of Hydrobiology 84: 419-426. . 1999b. Demographic responses of adult Me- socyclops thermocyclopoides (Copepoda, Cy- clopoida) to different plant and animal diets.— Freshwater Biology 42:487—-501. Marten, G. G. 1984. Impact of the copepod Mesocy- clops leuckarti pilosa and the green alga Kir- chneriella irregularis upon larval Aedes albop- ictus (Diptera: Culicidae).—Bulletin of the So- ciety of Vector Ecologists 9:1—5. . 1989. A survey of cyclopoid copepods for control of Aedes albopictus larvae.—Bulletin of the Society of Vector Ecologists 14:232—236. , & G. Thompson. 1997. Copepod production and application for mosquito control.—New Or- leans Mosquito Control Board, New Orleans, U.S.A., 42 pp. , R. Astaiza, R. Monje, & J. W. Reid. 1989. Natural control of larval Anopheles albimanus (Diptera: Culicidae) by the predator Mesocy- clops (Copepoda: Cyclopoida).—Journal of Medical Entomology 26:624—627. , G. Borjas, M. Cush, E. Fernandez, & J. W. Reid. 1994a. Control of larval Aedes aegypti (Diptera: Culicidae) by cyclopoid copepods in peridomestic breeding containers.—Journal of Medical Entomology 31:36—44. , E. S. Bordes, & M. Nguyen. 1994b. Use of cyclopoid copepods for mosquito control.—Hy- drobiologia 292/293:49 1—496. Omori, M., & T. Ikeda. 1992. Methods in marine zoo- plankton ecology. Krieger Publishing Co., Flor- ida, 332 pp. Petkovski, T. K. 1986. Zur Taxonomie des genus Me- socyclops G.O. Sars 1914 (Crustacea, Copepoda Cyclopoida) in der Neotropis.—Acta Musei Macedonici Scientiarium Naturalium 18:47—49. Reid, J. W. 1992. Copepoda (Crustacea) from fresh waters of the Florida Everglades, U.S.A., with a description Eucyclops conrowae n. sp.— Transactions of the American Microscopical So- ciety 111:229-—254. Sinh Nam, V., N. Thi Yen, M. Holynska, J. W. Reid, & B. H. Kay. 2000. National progress in dengue vector control in Vietnam: survey for Mesocy- clops (Copepoda), Micronecta (Corixidae), and fish as biological control agents.—American Journal of Tropical Medicine and Hygiene 62: 5-10. Suarez, M. E 1992. Mesocyclops aspericornis for the control of Aedes aegypti in Puerto Rico and An- guilla. Pp. 151-157 in S. B. Halstead, & H. Go- VOLUME 116, NUMBER 3 mez-Dantes, eds., Dengue-a worldwide prob- lem, a common strategy. Proceedings of the In- ternational Conference on Dengue and Aedes aegypti community-based control. Rockefeller Foundation and Mexican Ministry of Health, Mexico City, 232 pp. Suarez-Morales, E., & M. A. Gutiérrez-Aguirre. 2001. Morfologia y taxonomia de los Mesocyclops (Crustacea: Copepoda: Cyclopoida) de México. CONACYT/ECOSUR, México, 207 pp. , & J. W. Reid. 1998. An updated list of the 753 free-living freshwater copepods (Crustacea) of Mexico.—Southwestern Naturalist 43(2):256— XODE , J. W. Reid, T. M. Iliffe, & F Fiers. 1996. Ca- talogo de los copépodos (Crustacea) continen- tales de la Peninsula de Yucatan México. CON- ABIO/ECOSUR, 298 pp. Thiébaud, M. 1912. Copépodes de Colombie et des Cordilleres de Mendoza.—Mémoires de la So- cieté Neuchateloise des Sciences Naturelles 5: 160-175. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):754—802. 2003. A review of the freshwater crabs of the genus Hypolobocera Ortmann, 1897 (Crustacea: Decapoda: Brachyura: Pseudothelphusidae), from Colombia Martha R. Campos Universidad Nacional de Colombia, Instituto de Ciencias Naturales, Apartado Aéreo 103698, Bogota, Colombia, S.A., e-mail: mhrochad @unal.edu.co Abstract.—A review of the species of Hypolobocera from Colombia is pre- sented. A total of 18 species and four subspecies occur in this country. Diag- noses, illustrations, and a key for the identification of species and subspecies based on the morphology of the first male gonopod, are included. The geograph- ical distribution of the genus, species and subspecies is updated based on new material. Two new species, Hypolobocera murindensis and H. velezi, are de- scribed and illustrated. Two more groups of Hypolobocera species, groups 7 and 8, are defined to accommodate H. alata Campos, and H. andagoensis (Pretz- mann). One subspecies, H. bouvieri rotundilobata, is elevated to specific rank. The genus Hypolobocera Ortmann, 1897, includes 34 species of freshwater crabs that can be found in Venezuela, Colombia, Ec- uador, and Peru. This makes Hypolobocera the most widely distributed genus of all pseudothelphusid genera. In Colombia, Hy- polobocera is now represented by 18 spe- cies and four subspecies. In reviewing the genus for Ecuador, Rodriguez & Sternberg (1998) listed 12 Ecuadorian species, three of which were new. Rodriguez (1982a) re- corded three species for Peru, the southern distributional limit for the family Pseudo- thelphusidae. The systematics and bioge- ography of the genus have been reviewed by Rodriguez (1982a, 1994), Rodriguez & Sternberg (1998), and Prahl (1988). The morphology of the first male gono- pod, a basic characteristic for the diagnoses of the species, displays considerable inter- specific variability in freshwater crabs. Rodriguez (1982a) divided the genus Hy- polobocera into six groups based on mor- phological and biogeographical features. Accordingly, the Colombian Hypolobocera can be placed as follows: in group 1, Hy- polobocera beieri Pretzmann, 1968, H. martelathani (Pretzmann, 1965), and H. noanamensis Rodriguez, Campos & Lopez, 2002; in group 2, H. bouvieri (Rathbun, 1898); in group 3, H. steindachneri Pretz- mann, 1968; in group 4, H. cajambrensis Prahl, 1988, H. chocoensis Rodriguez, 1980, H. dentata Prahl, 1987, H. embera- rum Campos & Rodriguez, 1995, H. Ilo- roensis Campos, 1989, H. malaguena, Prahl, 1988, H. rotundilobata Rodriguez, 1994, and AH. velezi, new species; and in group 5, H. gorgonensis Prahl, 1983, and H. mutisi Prahl, 1988; group 6, H. anda- goensis (Pretzmann, 1965), and H. murin- densis, new species. However, H. alata Campos, 1989, and H. kamsarum Campos & Rodriguez, 1995 can not be assigned to any of the groups proposed by Rodriguez (1982a) as the features of their gonopods do not match any of Rodriguez’s groups. Thus, it is necessary to add two new groups in order to accommodate these latter two species. Group 7 is proposed for H. alata, and is characterized by having the lateral lobe of the first male gonopod with a strong triangular process; the apex is oval and bent caudocephalically, the cephalic border is VOLUME 116, NUMBER 3 expanded into triangular projection which is folded downwards and has a rounded pa- pilla on the tip. The known distribution of group 7 comprises the San Juan River basin on the Western Cordillera. Group 8 is pro- posed for AH. kamsarum, and is distin- guished by the lateral lobe of the first male gonopod which is small, transverse and dis- placed towards the cephalic side; the apex is oval with a rounded expansion distally, and the mesial border is projected proxi- mally, forming a strong triangular mesial lobe. The distribution of group 8 comprises the eastern foothills of the southern Andes of Colombia. The terminology for the male first gon- opod is that of Smalley (1964), and Rod- riguez (1982a, 1994). The measurements are reported in the order “‘cl X cb” (cara- pace length times carapace breadth). The material was collected by the author except where otherwise indicated, or in some cases by an unknown collector (uk). The material of the Museo de Biologia Marina, Univer- sidad del Valle, was reported by Prahl with numbers that apparently corresponded to his collection lot numbers (not museum cat- alogue numbers), followed by the abbrevi- ation “ADT-CRBMUV” (=Agua Dulce y Terrestres Crustaceos, Biologfa Marina Universidad del Valle). However, Prahl’s material examined that is still extant at the Museo de Biologia Marina, Universidad del Valle, Cali now has only museum catalogue numbers. Prahl’s collection numbers are in- cluded herein in parenthesis only for the material reported as such in his publica- tions. Color nomenclature follows Smithe (1975). Two species, Hypolobocera buen- aventurensis (Rathbun, 1905), and H. stein- dachneri Pretzmann, 1968, are not illustrat- ed due to the lack of material. The materials remain deposited in: Co- leccion de Referencia, Museo de Historia Natural, Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogota (ICN-MHN); Museo de la Sociedad de Ciencias Naturales La Salle, Caracas (LS); Museo de La Salle, Bogota (MLS); Museo V5 de Biologia Marina, Universidad del Valle, Cah (CRBMUY); Instituto Venezolano de Investigaciones Cientificas, Caracas (IVIC); National Museum of Natural History, Smithsonian Institution, Washington D.C. (USNM); Natural History Museum, Lon- don (BM); Museum of Natural History of Tulane University, New Orleans (TU); Field Museum of Natural History, Chicago (FMNH); and Musée de Strasbourg (SM). Tribe Hypolobocerini Pretzmann, 1971 Hypolobocera Ortmann, 1897 Diagnosis.—First male gonopod with strong longitudinal caudal ridge. Lateral lobe well developed, sometimes reduced, varying according to species as triangular, rounded or subquadrate, and with or with- out crenulations over distal margin. Apex outline either rounded, oval, or elongated in distal view. Mesial lobe triangular, semicir- cular or reduced as strong fold. Third max- illiped with exognath 0.20—0.60 times length of ischium of endognath. Type species.—Potamia chilensis H. Milne Edwards & Lucas, 1844. Distribution.—Venezuela, Colombia, Ec- uador and Peru. Hypolobocera alata Campos, 1989 Fig. 1A—F Hypolobocera alata Campos, 1989:145, fig. 2a—g. Hypolobocera alata.—Rodriguez, 1992: 183. Material examined.—Colombia. Huila Department, Villavieja, La Batea stream, 400 m alt., 3 Apr 1982, leg. R. Restrepo, 3 holotype, 13.1 X 20.2 mm, | ¢ paratype, 13.4 X 21.4 mm, ICN-MHN-CR 0853.— Risaralda Department, Pueblo Rico, Corre- gimiento Santa Cecilia, Amurropa stream, A90rmralt, 26, Sep) 199ilmles.GyA. 2007, 23) XS NDS) waaay WON SWS 2 oven, MEINE MHN-CR 1309; Vereda La Granja, 700 m allt, 24 Orci IID We 5 WAM Xe MS) (o) weave ICN-MHN-CR 1306.—Choc6 Department, 756 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Hypolobocera alata Campos, 1989, male, ICN-MHN-CR 1309: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, apex, distal view; E, right side of carapace, dorsal view; FE left third maxilliped, external view. Tado, Corregimiento Guarat6, Guadralito strongly bent caudocephalically, distal cau- stream, 500 m alt., 28 Sep 1991. leg. G. dal margin with row of blunt teeth (Fig. Andrade, 1 ¢, 12.1 X 19.4 mm, ICN- 1B). Caudal ridge long, strong, straight; MHN-CR 1308. ending beyond lateral lobe. Lateral lobe Diagnosis.—First male gonopod narrow, with strong triangular process (Fig. 1A—C). VOLUME 116, NUMBER 3 Apex bent caudocephalically, apex outline oval in distal view, lateral border raised, with 4 acute spines near caudal border; ce- phalic border expanded into triangular pro- jection, folded downwards, with rounded papilla on tip; prominent rounded cephalic papilla and small papilla on top. Mesocau- dal projection of spermatic channel termi- nating in slightly bifid papilla. Mesial lobe subtriangular (Fig. 1B—D). Third maxilliped with exognath approximately 0.30 times length of ischium (Fig. IF). Remarks.—The type locality of Hypolo- bocera alata is Villavieja, Huila, located on the Magdalena River valley. Subsequently, other specimens were collected in Pueblo Rico, Risaralda, Western Cordillera, upper reaches of the San Juan River, and in Tad6o, Choc6, Pacific coastal plain, middle course of the San Juan River. The first male gon- opods of these specimens is identical to that of the holotype. Thus, it appears that the type locality may have been misreported, and Hypolobocera alata is actually found in the vicinity of Pueblo Rico, Risaralda, and Tad6, Choco in the San Juan River ba- sin. Hypolobocera andagoensis (Pretzmann, 1965) Fig. 2A—G Strengeria (Strengeria) andagoensis Pretz- mann, 1965:6. Hypolobocera (Hypolobocera) andagoen- sis.—Preztmann, 1971:17; 1972:51, figs. LORNA Hypolobocera andagoensis.—Rodriguez, IQDSZERO Vee lilGs ile, Seen all, Ws 183.—Rodriguez, 1994:296, fig. la—c. Material examined.—Colombia. Choco Department. Andagoya, May 1957, leg. M. itathane, 1 holotype, 1922) mana xX 33)1.6 mm, USNM 106405.—No data, May 1957, le cer Mip Te Ativan Sicilians) ec 27/24. to dS Ye Os maim 2 cl 2038) S49 mora 7-5 x 28.8 mm, USNM 106407.—No data, May l957leces MM. eathams 2256, 12.5 < 302 WO 32) 06 IS.2 mom, 2D 26 Wilko >< 37/46) 75d to 9.4 Kk 11.0 mm, USNM 106409.—An- dagoya, Condoto and San Juan Rivers, leg. eG ryspurels 12097) as 2, 0 man BM 1915.11.1.1.—110 km N of Palestina, Do- cord6 stream, affluent of San Juan River, 4°55'N, 76°55'W, 22 Jan 1971, leg. B. Mal- kin and P. Bouchard, | m, 20.6 X 32.0 mm, 22, Gil WAG XK ASO cman, sil X< WO.6 tern, FMNH 3676. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile fingers and fixed fingers (Fig. 2G). First male gon- opod with caudal ridge long, strong, slight- ly sinuous; almost reaching to apex (Fig. 2A). Lateral lobe prominent, subquadrate, narrower distally than proximally, external margin faintly crenulated; caudal surface partially excavated, (Fig. 2A, B). Apex out- line oval in distal view, slightly expanded caudally; shallow notch on caudal border, and prominent cephalic papilla. Mesocaudal projection of spermatic channel terminating in acute papilla. Mesial lobe subtriangular (Fig. 2C, D). Third maxilliped with exog- nath approximately 0.25 times length of is- chim (ie, 2E): Remarks.—In his original description, Pretzmann (1965) designated as holotype a male specimen from lot USNM_ 106405 (19.2 mm X 31.6 mm). However, Pretz- mann (1972) changed the holotype lot as USNM 106407. Pretzmann designated 14 males and two females from USNM 196405 as paratypes, and incorrectly stated that the two lots were from the same local- ity. Pretzmann (1972) illustrated the whole holotype specimen (figs. 170, 171), and a detached first male gonopod (figs. 311, 312). As indicated by Rodriguez (1994), this appendage could not have belonged to the holotype (USNM 106405) as it was found still attached. Hypolobocera beieri Pretzmann, 1968 Fig. 3A—H Hypolobocera (Hypolobocera) bouvieri beieri Pretzmann, 1968:9; 1971:17; 1972: 46, figs. 176-181, 308, 309. 758 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Hypolobocera andagoensis (Pretzmann, 1965), male holotype, USNM 106405: A, left first gonopod, caudal view; B, same, cephalic view; C, same, mesial view; D, same, apex, distal view; E, carapace, dorsal view; FE left third maxilliped, external view; G, right chela, external view. VOLUME 116, NUMBER 3 Tf5)s) YCPi4MZSuy Fig. 3. Hypolobocera beieri Pretzmann, 1968, male, ICN-MHN-CR 1749: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; FE right side of carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. 760 Hypolobocera (Hypolobocera) monticola steindachneri.—Pretzmann, 1972:46 (part.) Hypolobocera beieri.—Rodriguez, 1982a: 46, figs. 19a, 1, 20b, g, 24a—d.—Campos, 1985:275.—Prahl, 1985:43—47, figs. 1—- 7.—Prahl, 1988:172, fig. 2.—Rodriguez, 1994:297.—Rodriguez et al., 2002:3, 4, fig. 1J—K. Material examined.—In addition to the material reported by Rodriguez (1982a), Prahl (1985), and Rodriguez et al. (2002), the following has been examined: Colom- bia, Valle del Cauca Department. Alto An- chicaya, 2 Oct 1982, leg. H. von Prahl, 1 6, carapace broken, TU 6383.—Da4agua, Vereda El Carmen, Santa Clara Farm 2 Feb 1984 ee EMReCIOn 276.27 3945.5 amas NS X& ALA men, ZL, Cl ZLB XK BSS scien, 19.5 X 30.6 mm, CRBMUV 84059.—Tul- ua, Vereda Salénica, 16 Aug 1982, leg. uk, 1 6, 17.3 X 27.9 mm, CRBMUV 84002. La Victoria, San Miguel, 3 km from high- way to Tacoté, 4 Feb 1984, leg. uk, 2 6, 74 S BOO win, ls % Wess wari, CRBMUV 84003.—Palmira, Corregimien- to La Buitrera, 5 Sep 1986, leg. R. Neira, 5S 2, Die «— SAY Wo OAs NSO sien, CRBMUV 82046.—Palmira, Corregimien- to La Buitrera, 25 Jun 1982, leg. R. Neira, 1 46, 20.8 X 32.9 mm, CRBMUV 82048. Bugalagrande, Guadualillo, 6 Jan 1986, leg. Ry Wem, 2 6, 148 « 220 mem, 2 x< N67) sore, 22 WOW << BSS nmin, D2) S83 mm, CRBMUV 86035.—Old road to Bue- naventura, Vereda La Elsa, La Elsa stream, 840 m alt., 25 Aug 1984, leg. H. Restrepo, 1 2, 30.7 X 48.6 mm, CRBMUV 84061. Cali, Corregimiento Felidia, 6 Mar 1988, les, tix, 3 °2, 23.6 X 42023 © WO.2 XK BOO mm, CRBMUV 88019.—Dagua, Puerta del Diablo near to Tacota, 3 Feb 1984, leg. H. iene, IO, LOD < 277 mining IL 2, 1D < 31.4 mm, CRBMUV 84066 (054 ADT- CRBMUV).—Jumbo, Vereda Manga Vieja, San Marcos stream, 4 Feb 1984, leg. uk, 1 6. MAS XX ANS ion, I SS XO OX ANS. 11 taaen, CRBMUV 84068.—Jamundi, San Antonio, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1 Apr 1984, leg. uk, 2 6, 20.9 X 33.8 mm, 19.8 X 30.5 mm, CRBMUV 84067.—Flor- ida, Corregimiento Betania, 15 Dec 1984, lee Wik, 3 Gy Biles < 34.0 to 11.3 << 16.2 mm, 52.2 OnxX 32510) 10:3 < 4 savan, ICN-MHN-CR 0135.—Calima, Azul River, 1600 m alt., 8 Feb 1984, leg. R.R., 1 2, 20.0 X 31.5 mm, ICN-MHN-CR 0548. Cali, Vereda Penas Blancas, Pichindé River, 1000m alt, 29 Sepl977. 1) 2 lSis ee snet mm, ICN-MHN-CR 1264.—Yotoco, 19 km from Buga to Buenaventura, La Cabana Farm, 1450 m alt., 6 Nov 1998, leg. A. Sua- rez, 1 6, 26.7 X 44.5 mm, ICN-MHN-CR 1749.—Bolivar, Inspeccion Cerro Azul, 900) malt, 22 Feb) 1994) lee Es Fiorezaa! 6, 19.5 X 30.7 mm, ICN-MHN-CR 1881.—Choc6é Department, between Cu- currupi and Noanama, Docordo stream, 5 Jan 1969, leg. B. Malkin, 1 6, 23.5 X 35.1 mim. 2245.8 < D724 won, tf 2, IDD «K Bey icin, KONE MHN-CR 0462, 0480.—Guaduas, Vereda El] Raizal y Cajon, 1300 m alt., 22 Jun ID83, 2S, Gl Ss2 XK Oil-3 ini, V4 XX BS mm, 1 ¢, 48.8 X 78.8 mm, ICN-MHN-CR 0467—0469.—Guaduas, Chaguani stream, OWS iol sllti., B42 uray WOR 4S, WAS < 934.0 imma sala 27.2 mm, ICN-MHN-CR 1202.—Fusaga- suga, Vereda La Pampa, Fusa-Chinauta Highway, 2000 m alt., 8 Jan 1993, leg. M. Ramirez, 1 2, 47.4 X 77.5 mm, ICN-MHN- CR 1293.—Silvania, Vereda Panama Alta, San Antonio de los Bogas Farm, El Hato stream, 1830 m alt., 8 Aug 1995, leg. R. Restrepo, | 1992) 43.0 4773.0 ini I@INE MHN-CR 1458.—Yacopi, Inspecci6n Apo- sentos, El Hatico stream, 850 m alt., 26 Oct 1995, leg. G. Galvis, 1 2, 19.0 X 30.4 mm, ICN-MHN-CR 1512.—Yacopi, Inspecci6n Guadualito, Vereda La Laguna, 900—1050 m alt., 28-29 Oct 1995, 2 ¢, 41.7 X 67.0 mm.3502 847-0 mm) Ion BAD aaa mm, ICN-MHN-CR 1519—1521.—Yacopi, Inspecci6n Guadualito, Vereda Lamal, Agua Blanca stream, 700 m alt., 31 Oct—4 Nov 1995, 2 36, 56.2 X 90.1 mm, 11.6 X 17 -A mma 4s 2 545) X SO 210) LIS eae mm, ICN-MHN-CR 1524, 532, 1535.—Ya- copi, Inspeccion Guadualito, Vereda Gra- males, Barbascales, Salitrona stream, 740 m alt., 2 Nov 1995, 1 3, 17.0 X 26.0 mm, ICN-MHN-CR_ 1528.—Yacopi, Inspeccion Guadualito, Vereda Lamal, Albercas, 900 m alt., 3 Nov 1995, 1 2, 55.0 X 89.3 mm, ICN-MHN-CR 1529.—Yacopi, Vereda La Oscura, La Oscura stream, 350 m alt., 24 Sep 1996, 6 3, 40.5 X 64.8 to 12.6 X 15.6 mm, 8 ¢, 24.4 * 37.2 to 10.6 * 15.6 mm, ICN-MHN-CR 1621. (4) Norte de Santan- der Department. Pamplonita, stream near to town, 1600 m alt., 26 Mar 1987, 2 6, 20.0 x 30.2 mm, 20.6 Kk 31.0 mm, 1 2, 29.8 X 46.2 mm, ICN-MHN-CR 0691.—Pamplon- ita, Vereda San José de Cunuta, 1 100—1250 m alt., 26 Mar 1987, 2 ¢, 39.5 X 63.2 mm, 32.8 X 51.3 mm, ICN-MHN-CR 0692, 0693.—Chindacota, Vereda El Urengue, 1075 m alt.) 26 Mar 1987, 1 63352 K373 VOLUME 116, NUMBER 3 mm, ICN-MHN-CR 0694.—Chindcota, Vereda Sonival, Chinacota-Toledo High- way, 1600 m alt., 27 Mar 1987, 1 @, 42.0 xX 67.7 mm, ICN-MHN-CR 0699.—Chin- Acota, Vereda El Asilo, Cacua stream, Chin- acota-Ragonvalia Highway, 1600 m alt., 9 OcilOSs. eSB 4o X15) 2mm: ICN- MHN-CR 0928.—Bucarasica, Vereda Santa Rita, Sardinata-Ocana Highway, 400 m alt., D9 MarlOs7iise oa2 16) 35.2 to L422 Dey met YPN6.y < 255) mm, ICN MHN-CR 0703.—Bucarasica, Vereda For- tunas, Sardinata-Ocana Highway, 500 m alt 29) Marel98S7- 10 6,7 19:45—% 31-4 to 12x OM mms seo | OORX< 4454 to 1167 xX 26.0 mm, ICN-MHN-CR 0704.—Boch- alema, Corregimiento La Donjuana, Vereda Cachiri, Durania-Lamus Highway, 1050 m Dies OMManl 98a Ga 22-5. 635651 min, d Sealise0) <2 77, mut. KONEMAN-ECR 0705.—Bochalema, Corregimiento La Donjuana, Vereda Cachiri, 1125 m alt., 30 Manil9s7=> 22 26:8 xX 418 to 112-x 1756 mm, ICN-MHN-CR 0706.—Durania, Ver- eda La Palma, Lavapatas stream, 900 m alt., 31 Mar 1987, 6 3, 34.6 X 55.6 to 16.0 X PASS mime e635: 82 4G 95-9) mimi) ICN= MHN-CR 0708.—Cucutilla, Vereda Cuesta Rica, Limoncito stream, 1700 m alt., 8 Oct 1988, 1 2, 25.9 X 38.6 mm, ICN-MHN- CR 0926.—Cucutilla, Vereda Aguada Bajo, Pamplona-Cucutilla Highway, 1450 m alt., SROGHMOSS,) te li35 Ml ex NS aSik x 45.4 mm, and 15 juveniles, ICN-MHN-CR 0711.—Simacota, Vereda Nauno, Socorro- Simacota Highway, 1050 m alt., 22 Sep 1988, 3 ¢, 19.1 X 28.8 to 16.4 X 24.9 mm, 29) WS: 76x QS9hmmAts.6 << 2307 sam: and 2 juveniles, ICN-MHN-CR 0906.—Si- macota, Vereda Pedregales, San Miguel stream, Sogamoso-Simacota Highway, 950 765 Malt wZiSep TOSSA 2s 6 aS9S 6277 mm, 14.7 X 22.1 mm, ICN-MHN-CR 0907.— Chima, Vereda Tierra Amarilla, El Guamal stream, Simacota-Chima Highway, 1050 m alt 22 596p lO8s8nld, 26:3) < 41-4 mama, ICN-MHN-CR 0908.—Suaita, Vereda La Aguadita, La Aguadita stream, 1600 m alt., 5 Jan 1994, leg. A. Rodriguez, 3 ¢, 16.7 X 29.5 to 12.2 X 19.3 mm, ICN-MHN-CR 1452. (6) Tolima Department. 8 km South of Ibagué, Combeima River, 12 Jun 1977, locas Diazaic 2.0) 84-0) mm ston ss 54.3 X 88.4 to 45.2 X 71.5 mm, IVIC.— Icononzo, Las Lajas stream, 1070 m alt., 28 Mare tO7sssleos Cre scalloni 3) Gin 57-15 x 60.6 to 35.3 XX 57.2 mm, ICN-MHN-CR 0499, 0500, 0501.—Icononzo, Vereda Cha- parro, La Juanita Farm, 1200 m alt., 16 Nov 1986, 1 3, 25.5 X 40.7 mm, ICN-MHN- CR 0681. Diagnosis.—Chelae of male with small rounded or oblong tubercle on external base of mobile fingers, and prominent, rounded tubercle on external base of fixed fingers (Fig. 4H). First male gonopod with caudal ridge long, slightly sinuous; ending in nar- row ridge beyond lateral lobe (Fig. 4A). Lateral lobe semicircular, with or without crenulations over distal margin. Cephalic surface with transverse crest on distal half, and tuberculated ridge parallel to lateral lobe (Fig. 4A—D). Apex outline oval, bor- ders raised; prominent cephalic papilla and auxiliary rounded papilla near spermatic channel. Mesocaudal projection of sper- matic channel terminating in blunt papilla. Mesial lobe subtriangular (Fig. 4D, E). Third maxilliped with exognath 0.20—0.30 times length of ischium (Fig. 4G). Remarks.—This subspecies is widely dis- tributed in the slopes of the Central and Eastern Cordilleras where waters drain to the Magdalena River. In males, the main feature that distinguishes this subspecies from the others are the chelae, which fea- ture a prominent, rounded tubercle on the external base of the fixed fingers (Fig. 4H). 766 Hypolobocera bouvieri angulata (Rathbun, 1915) Fig. SA—H Pseudothelphusa angulata Rathbun, 1915: 98.—Coifmann, 1939:106.—Rodriguez, 1966:129, fig. 9—Rodriguez, 1967:10. Strengeria (Strengeria) angulata Pretz- mann, 1965:7. Hypolobocera (Hypolobocera) bouvieri an- gulata Pretzmann, 1971:17, pl. 17.— Pretzmann, 1972:45, figs. 156-158, 208— 210. Hypolobocera bouvieri angulata.—Rodri- guez, 1982a:56—57.—Rodriguez, 1994: 299.—Campos, 1985:275—276, Rodri- guez et al., 2002:6. Material examined.—In addition to the material reported by Rodriguez (1982a), and Rodriguez et al. (2002), the following has been examined: Colombia, Magdalena Department. Santa Marta, Minca, 880 m alt., M holotype, 40.0 65.0 mm, USNM 98398.—Sierra Nevada de Santa Marta, 10 Aug 1987, leg. uk, 1 6, 30.4 X 47.9 mm, CRBMUV 87014.—Santa Marta, Parque Nacional, Natural Tayrona, Los Cedros. 2— 6 Jul 1983, leg. G. Galvis. 4 3, 43.3 X 61.4 (0) OJ *K AS t8 inn, © LS, 2 < Go SOW to 16.8 X 24.6 mm, ICN-MHN-CR 0528— 0530.—Santa Marta, Minca, 850 m alt., 23 IMileyy WES). lee, IRS Semele, I 2, 3440 Xx 55.6 mm, ICN-MHN-CR 0961.—Santa Marta, Vereda Alto Guachaca, stream afflu- ent of Guachaca River, 700 m alt., 30 Sep 1992, leg. A. Ferrer, 1 2, 54.6 X 87.4 mm, ICN-MHN-CR 1296.—Cesar, Serrania de Perla, La Jagua de Ibirico, Corregimiento La Victoria de San Isidro. Vereda Alto de las Flores, stream near Escuela Nueva de las Flores, 1200 m alt., 8 Mar 1996, 2 6, AT.) < VOA nin, Dl <— 33.8 om, ily 2. 13.4 kK 19.9 mm, ICN-MHN-CR 1551, 1556.—Vereda Nueva Granada, El Indio stream, 590 m alt., 9 Mar 1996, 4 3, 57.3 x OS. IS. << 23.9 sori, 2 2, 4310) Xx 78.1 mm, 20.0 X 31.4 mm, ICN-MHN-CR 1554, 1555, 1573.—Vereda El Zumbador, El Zumbador stream, 400 m alt., 10 Mar PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ie, 2d, Wes X SOL aan, YOO « Zil5 mm, 6 &, 38.2 X 59.5 to 13.0 X 20.3 mm, ICN-MHN-CR 1557.—Vereda Alto de las Flores, Tucuy River, 915 m alt., 11 Mar UO 2, AIS) XX BAKO ni, ZO XK 32.0) mm, 1 ¢, 21.8 X 35.2 mm, ICN-MHN-CR 1558.—Vereda El Zumbador, Zumbador River, 1000 m alt., 14-17 Mar 1996, 8 6, 48:39) 73:9 1to15.0>< 2310 min lone xX 89.2 to 8.3 X 13.8 mm, 2 juveniles, ICN- MHN-CR 1563, 1564, 1566, 1567.—Vere- da Nueva Granada, Buenavista Farm, Zum- bador River, 500 m alt., 20 Mar 1996, leg. OO) Ve Castanow ll Se 24513 6y/arnmenlee 42.4 x 69.9 mm, ICN-MHN-CR 1572.— Venezuela. Estado Zulia, Serrania de Per- 1a, Socuy River, Cueva Los Laureles, 1300 m, alt., 20 Dec 1990, leg. A. L. Viloria and T R. Barros, 1 6, 3958) x 6221 mime IVIC.—Estado Zulia, Serrania de Pera, Socuy River, Los Encantos, 850 m, alt., 20 Dec 1995, leg. F Herrera, 1 5, 29.5 X 47.7 mm, IVIC. Diagnosis.—Chelae of male with small tubercle on external base of mobile fingers, and swelling on external base of fixed fin- gers (Fig. 5H). First male gonopod with caudal ridge long, either sinuous or straight; ending in narrow ridge beyond lateral lobe (Fig. 5A, B). Lateral lobe usually subtrian- gular with small crenulations on distal mar- gin. Cephalic surface with transverse crest on distal half and tuberculated ridge parallel to lateral lobe (Fig. SA—D). Apex outline oval in distal view; prominent cephalic pa- pilla, and auxiliary rounded papilla near spermatic channel. Mesocaudal projection of spermatic channel terminating in slightly acute papilla. Mesial lobe subtriangular (Fig. 5D, E). Third maxilliped with exog- nath 0.20—0.32 times length of ischium (Fig. 4G). Remarks.—This subspecies is broadly distributed in an area that reaches from the Sierra Nevada de Santa Marta to both slopes of the Sierra de Perija and the Cor- dillera de Mérida in Venezuela. The distri- bution includes two disjunct areas that cov- er two different basins: the Cesar, and the VOLUME 116, NUMBER 3 767 Fig. 5. Hypolobocera bouvieri angulata (Rathbun, 1915), male, Colombia, Santa Marta, ICN-MHN-CR 0528: A, left first gonopod, caudal view; C, same, lateral view; D, same cephalic view; E, same, apex, distal view; E right side of carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. Male, Venezuela, Estado Zulia, Serrania de Perija, Socuy River, IVIC: B, left first gonopod, caudal view. 768 Catatumbo Rivers. This type of distribution, possibly due to temperature differences, is considered a relict distribution. There are slight differences in the first male gonopods of the specimens found in the Sierra Ne- vada de Santa Marta in Colombia and in the Sierra de Perijé in Venezuela. The most sig- nificant difference is the shape of the lateral lobe, which is slightly semicircular (Fig. 5B). This difference is most likely caused by different environmental conditions. Hypolobocera bouvieri monticola (Zimmer, 1912) Fig. 6A—H Pseudothelphusa monticola Zimmer, 1912: 3, figs. 6-10, pl. 1.—Coifmann, 1939: 108. Strengeria (Strengeria) monticola Pretz- mann, 1965:7. Hypolobocera (Hypolobocera) monticola monticola Pretzmann. 1971:17.—Pretz- mann, 1972:46, figs. 197-199, 206, 207. Hypolobocera bouvieri monticola.—Rodri- SUCH sea Do. mes, 192 30: Material examined.—Colombia. (1) An- tioquia Department. 28 Jun 1966, leg. uk, | 6, 22.7 X 37.1 mm, CRBMUV 66001. Los Lagos, leg. uk, 1 6, carapace broken, CRBMUV 67001.—Caucasia, Rojo River, 5 Jan 1969, leg. M. Serna, | ¢, 23.8 X 38.7 mm, 1 &, 14.5 X 26.1 mm, CRBMUV 69002.—San Luis, Vereda Manizales, 1600 m alt., 12 Nov 1981, leg. P. Pinto, 1 2, 49.5 < 1951 maim 10!6n<@ 1 5sGraamael 21353 < 9204" im, VICNe Mine Cr 0780.—8 km from Pueblo Rico to Villa Claret, Tolda Seca stream, 1550 m alt., 21 Aug 9875 296), 17-8) K 2720 nome slioremx< 24.1 mm, 2 @, one with carapace broken, 12.7 X 18.6 mm, 2 juveniles, ICN-MHN- CR 0781.—Vereda Zabarraga Piunda, Piun- da stream, 730 m alt., 21 Aug 1987, 1 2°, 28.5 X 45.9 mm, ICN-MHN-CR 0786. Vereda Ciat6, Cristalina stream, 1500 m alt; 20-23) Sep 1991, 7.72107 &, VOLUME 116, NUMBER 3 769 MPinzen Fig. 6. Hypolobocera bouvieri monticola (Zimmer, 1912), male, ICN-MHN-CR O086: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; FE right side of carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. 770 ZO KV AGOmton 16.2) 25) 3 mms ENE MHN-CR 1260.—La Florida, Vereda La Suiza, stream by “Camino Ecoldgico’’, 1850 m alt., 12 Jun 1989, leg. M. Morales, 1 3d, 24.0 xX 39.0 mm, ICN-MHN-CR 0966.—Mistrat6. Corregimiento San Anto- nio del Chami. Vereda La Florida, San An- tonio del Chami-Geguadas Highway, 1100 m alt., 30 Mar 1992, 13 3, 25.7 X 39.0 to its < 7S tamn, D LY. ZO «K S22 wo 130 xX 19.8 mm, ICN-MHN-CR 1268.—Vereda La Palestina, San Antonio del Chami-Mis- trat6 Highway, 1150 m alt., 30 Mar 1992, 36, 20.0 < BIL to IO K 18. Ota, 4 2. ISell x B27 to W454) dX DIL canimn, MONE MHN-CR 1269.—Vereda Arcacay, 900 m allt, ll Ayoe 19925 3 Gs 140 xX Zilles wo MOY x Goll nam, 2 Sy i2 « 2O.S wen, 135 xX 20.2 mm, ICN-MHN-CR_ 1270.—Vereda Empalado, 1800 m alt., | Apr 1992, 1 &, cl 34.8 mm, cb 55.2 mm (ICN-MHN-CR 1271).—Vereda Mampay, Sutu stream, L750 im allt, 3 Avoe 1992, 7 S, 19.5 K 292 (© I25 « Iso mm, 3 2, 17.7 K 2S to 11.5 X 17.2 mm, ICN-MHN-CR 1272.— Vereda La Florida, 1350 m alt., 4 Apr 1992, lO 6, 24.5 X37 4) © W206 X< 19.5 mern, V3 2, 31.0 X 47.0 to 12.0 X 18.0 mm, ICN- MHN-CR 1273.—San Antonio stream, 800 m alt., 4 Apr 1992, 1 3, 33.5 X 51.6 mm, 1 2, 30.3 X 47.6 mm, ICN-MHN-CR 1274.—Vereda La Gabriela, Sutti stream, LVOO son alt... O Aor IGOY. I Sy WSs XK ZB. mm, | 2, carapace broken, ICN-MHN-CR 1275.—Vereda Puerto de Oro, Carbones stream, 1080 m alt., 22 May 1992, leg. R. Sanchez, 1 2°, 19.8 X 31.6 mm, ICN-MHN- CR 1778. (4) Tolima Department, Chapar- ral. Vereda El Linday, La Miel stream, 900 m alt., 26 Jul 1981, 6 3, 37.7 X 59.7 to DL XK S39 mim, 8 YS, S8.O « GOA to 12.11 x 18.6 mm, ICN-MHN-CR 0069, 0070, 0071, 0072, 0073, 0074, 0075, 0076.—Ver- eda El] Linday, Agua Dulce stream, 850— 900 m alt., 22 May 1984, 15 Jul 1995, 4 6, 45.7 < W336 t® 23.4) X 36.7 tim, 2S. 40.2 < 64.1 mm, 31.8 X 51.9 mm, ICN- MHN-CR 0570, 1457.—Tuluni stream, 920 m alt., 24 Aug 1982, leg. D. H. Campos, | PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 6, 51.4 X 86.0 mm, ICN-MHN-CR 0086.—Corregimiento El Limon, Lake in Camacho Angarita Farm, 900 m alt., 20 Mar ri9s3, Gy 4am 697/516) nim GIN= MHN-CR 0139.—Corregimiento El Limon, Vereda Betania, 900 m alt., 15 Jul 1983,11 Jan 1988, 11 6, 38.7 X 59.5 to 13.8 X 20.7 mm, 9 2, 59.5 X 100.9 to 16.8 X 26.9 mm, ICN-MHN-CR 0506, 0507, 0509, 0845. Corregimiento El Limon, Vereda Buenosai- res, Miraflores Farm, 920 m alt., 17 Jul 1983, 30 Mar 1994, 6 6, 27.8 X 43.0 to 053) X< Ses Tomer 5) 2) OS) X 407 to 21.0 < 22-3 to OS x 17-3) mm: CRBMUV 83084.—Cajambre River, 25 Aus W983. leg uk, | 2) 5.6 x 254 mim, CRBMUV 83085.—Buenaventura, Pogod6 stream, affluent of Cajambre River, 50 m alt., 29 Apr 1999, leg. E. Florez, 1 6, 13.6 x 23.7 mm, ICN-MHN-CR 1876. Cauca Department, Timbiqui, Saija Riv- er, 100 m alt., 2°50’N, 77°30’W, 3 Nov 1971, leg. B. Malkin and A. Granja, 1 6, 24.9 X 47.6 mm, FMNH 3675. Diagnosis.—Carapace with front ex- tremely wide, approximately half of cara- pace width (Fig. 8F). Chelae of male with irregularly shaped tubercle on external base of mobile fingers, and swelling on external base of fixed fingers (Fig. 8H). First male gonopod with caudal ridge long, fusiform; ending in narrow ridge beyond lateral lobe (Fig. 8A). Lateral lobe triangular, and aux- iliary lobe parallel to lateral lobe on lateral side (Fig. 8A—C). Apex outline slightly oval TIS in distal view; caudolateral border trans- verse and expanded. Mesocaudal projection of spermatic channel terminating in round- ed papilla. Mesial lobe projected on ce- phalic surface as a wide, semicircular lobe (Fig. 8B—D). Third maxilliped with exog- nath approximately 0.30 times length of is- chium (Fig. 8G). Remarks.—This species can be easily distinguished from others within the genus by the presence on the first male gonopod of an auxiliary lobe parallel to the lateral lobe. Hypolobocera chocoensis Rodriguez, 1980 Fig. 9A-I Hypolobocera Hypolobocera dubia.— Pretzmann, 1972:48, figs. 224—226, 230— MSI, Don LOM Hypolobocera chocoensis Rodriguez, 1980: 891.—Rodriguez, 1982a:59, figs. 19, 21, 31.—Prahl, 1988:177, fig. 9.—Rodri- guez, 1994:300. Material examined.—Colombia, Choco Department. Guntas, Tamana River, leg. uk, 6 Inol@nijoe, US x Bikes mim, lel 1910.3.4.3-4.—Same data, | @ paratype, 18.2 X 26.9 mm, BM 1910.3.4.3—4.—\-Con- doto, leg. H. G. EF Spurrel, 2 6, 20.4 X 32.4 mm, 19.9 K 32.2 mm, BM 1913.10.28.1-— 3.—Mountains of upper San Juan River, Choco jungle, nearest village Playa de Oro, 28 Mar 1962, leg. M. Latham, 10 6, 22.8 xX 36.7 to 23.8 X 39.7 mm, USNM 240102.—Condoto. 23 Jul 1985, leg. uk, 2 6, ZNO « 2ZEO iim, TSS Se 27/0) tan OS = 0, 045-1 ADT-CRBMUV.—Llor6, Granja Codechocé, left margin of Atrato River, 50 m alt., 23 Mar 1988, leg. O. Reyes, 1 6, 16.0 X 27.1 mm, ICN-MHN-CR 0851. Pizarro, Bajo Baudo, Torreid6 stream, 20 m alt., leg. R. Sanchez. 30 Aug 1990, 2 d, cl NOD Seay Tanna, WS 27 ee rant gle 23.2 X 40.0 mm, ICN-MHN-CR 1201. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile and fixed fingers (Fig. 91). First male gonopod with caudal ridge strong, long, straight; ending 774 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 8. Hypolobocera cajambrensis Prahl, 1988, male, ICN-MHN-CR 1876: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; FE carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. VOLUME 116, NUMBER 3 775 Fig. 9. Hypolobocera chocoensis Rodriguez, 1980, male, ICN-MHN-CR 1201: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; EK same, apex, distal view; G, right side of carapace, dorsal view; H, left third maxilliped, external view; I, right chela, external view. Male, ICN- MHN-CR 0851: E, left first gonopod, caudal view. 776 in narrow ridge beyond lateral lobe (Fig. 9A, C). Lateral lobe triangular, with proxi- mal margin either rounded, or straight (Fig. 9A, C, E). Apex outline slightly rounded in distal view, expanded mesocaudally into rounded projection. Mesocaudal projection of spermatic channel terminating in semi- acute papilla. Mesial lobe semicircular (Fig. 9B, F). Third maxilliped with exognath 0.20—0.25 times length of ischium (Fig. 9H). Remarks.—This species is most similar to Hypolobocera lloroensis Campos, 1989. The two can be differentiated by features of the carapace and first male gonopod. The anterolateral border of the carapace in H. lloroensis (Fig. 14F) has a deep depression behind the external orbital angle, followed by another deep depression at the level of the cervical groove. In contrast, H. cho- coensis has the anterolateral border of the carapace with a deep depression behind the external orbital angle, but the border is not continuous with the margin of the depres- sion (Fig. 9G). The caudal edge of the apex of the first male gonopod is transverse in caudal view in Hypolobocera lloroensis (Fig. ISA), whereas it is rounded and ex- panded in H. chocoensis. Hypolobocera dentata Prahl, 1987 Fig. l1OA—G Hypolobocera dentata 1987:93-95, fig. | A-E.—Prahl, 1988:180, fig. 12. Hypolobocera dentata.—Rodriguez, 1992: 183. Material examined.—Colombia, Valle del Cauca Department. Bolivar, El Manzano, near Betania, 1600 m alt., 28 Jan 1984, leg. E. Velasco, 5 holotype, 15.6 x 24.0 mm, 071 ADT-CRBMUV.—Same Gaim, I 2 joemernyjoe, 17/0 x 25.7 iim, O72 ADT-CRBMUV. Diagnosis.—First male gonopod with caudal ridge long, sinuous, recurved prox- imally; ending in narrow ridge beyond lat- eral lobe (Fig. 1O0A). Lateral lobe subtrian- gular, with tubercles on distal external mar- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON gin, and on cephalic surface (Fig. 1OA—D). Apex outline oval in distal view, expanded caudocephalically into rounded projection; laterocephalic border toothed. Mesocaudal projection of spermatic channel terminated in rounded papilla. Mesial lobe slightly semicircular (Fig. 1OE). Third maxilliped with exognath approximately 0.40 times length of ischium (Fig. 10G). Remarks.—This species differs from the others in the genus in having a toothed la- terocephalic border in the apex of the first male gonopod. Hypolobocera emberarum Campos & Rodriguez, 1995 Fig. 11 A—-H Hypolobocera emberarum Campos & Rod- riguez, 1995:652, Figs. 3, 4. Material examined.—Colombia. Choco Department. E] Carmen de Atrato, Vereda El] Veinte, 2500 m alt., 30 May 1994, leg. I. D. Vélez, 3 holotype, 14.2 X 23.8 mm, ICN-MHN-CR_ 1358.—Same data, 2 6 paratypes, 13.1 X 22.0 mm, 13.0 X 21.8 mm, 3 2 paratypes, 13.47 22.6 tome 21.4 mm, ICN-MHN-CR 1359.—Antioquia Department. Urrao, Valle de Pérdidas, 1800 m alt., 3 Sep 1994, leg. P Duque, 3 d, 14.1 xX 23.8 to 10.9 X 18.3 mm, 3 &, 14.0 X 23.5 to 12.7 X 21.4 mm 9 juveniles, [CN- MHN-CR 1383. Diagnosis.—Carapace lateral sides con- spicuously pubescent in larger specimens (Fig. 11F). Chelae of male lacking tubercles on external base of mobile and fixed fingers (Fig. 11H). First male gonopod with caudal ridge long, fusiform; ending in narrow ridge beyond lateral lobe (Fig. 11A, B). Lateral lobe prominent, subtriangular, hatchet shaped, wide proximally, extending near apex of gonopod and forming deep notch distally; caudal face excavated, cov- ered partially with spinules (Fig. 11A—C). Apex outline oblong in distal view, caudo- cephalic expanded into elongated projec- tion. Mesocaudal projection of spermatic channel terminated in rounded papilla. Me- VOLUME 116, NUMBER 3 ee MPinzen gh mm Fig. 10. Aypolobocera dentata Prahl, 1987, male, CRBMUV 84077: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, apex, distal view; E right side of carapace, dorsal view; G, left third maxilliped, external view. sial lobe subtriangular (Fig. 11E). Third maxilliped with exognath approximately 0.20 times length of ischium (Fig. 11G). Remarks.—The shape of the lateral lobe of the first male gonopod of this species resembles that of Hypolobocera chocoensis Rodriguez, 1980 (Fig. 9A). However, the lateral lobe of H. chocoensis does not ex- tend near the apex, and lacks a distal notch as seen in H. emberarum. 778 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SCP i Alten Fig. 11. Mypolobocera emberarum Campos & Rodriguez, 1995, male holotype, ICN-MHN-CR 1358: A, left first gonopod, caudal view; B, same, lateral lobe, caudal view; C, same, lateral view; D, same, cephalic view; E, same, apex, distal view; EF right side of carapace, dorsal view; G, left third maxilliped, external view; H, left chela, external view. VOLUME 116, NUMBER 3 Hypolobocera gorgonensis Prahl, 1983 Fig. 12A—H Hypolobocera gorgonensis Prahl, 1983: 106, fig. 1—Prahl, 1988:181, fig. 13. Hypolobocera gorgonensis.—Rodriguez, 1992:183.—Rodriguez et al., 2002:7. Material examined.—Colombia, Cauca Department, Gorgona Island. Marranera stream, 22 May 1979, leg. uk, d holotype, 49.3 X 73.4 mm, 013 ADT-CRBMUV.—EI Wuecal leash speranzay leg. suke gle 2753.2) >< 80.0 mm, 012 ADT-CRBMUV.— 20 May IGT, lee ul 3 2, 508 XK S21 Silke x 49.5 mm, CRBMUV 79055.—El Azufral, [2 Noy IOs, leg, wis, | 25 Sao x SO mm, CRBMUV_ 82055 (027 ADT- CRBMUV).— 3 Nov 1989, leg. H. von Branko 4420. 725 mm, CRBMIUN: 89023.—26 Aug 1984, leg. uk, 1 2, 8.3 X 13.1 mm, CRBMUV 84072.—26 Aug lOg84, lew, wik, 1 2s Carapace lnirolxeii, CRBMUV 84073.—22 Aug 1980, leg. H. von Jieicdl, Il GG, SOG x€ wSins) inten, IP) 6303.—Pacific, | Feb 1962, leg. EF Medem, ” 6, U2 X* SIZ sim, 43% << ODD inaven, || 2. el 4272 nim, En VO4 marin, IEMONIe! 3687.—29 Apr 1985, leg. J. M. Rengifo, 3 36, 47.9 X 79.1 to 43.9 XK 71.2 mm, ICN- MHN-CR 0612. Diagnosis.—Chelae of male with small tubercle on external base of mobile fingers, and swelling on external base of fixed fin- gers (Fig. 12H). First male gonopod with caudal ridge strong, straight; ending in nar- row ridge distally. Mesial margin with rows of conspicuous setae (Fig. 12A). Lateral lobe triangular, increasing in width distally; distal external margin rounded (Fig. 12A, C). Apex outline elongated along mesola- teral axis in distal view; caudocephalic bor- der rounded. Mesocaudal projection of spermatic channel terminating in rounded papilla. Mesial lobe elongated and subtrian- gular (Fig. 12E). Third maxilliped with ex- ognath approximately 0.35 times length of ischium (Fig. 12G). Remarks.—This species is considered en- 779 demic to Gorgona, an island with a surface of 2400 ha. Hypolobocera kamsarum Campos & Rodriguez, 1995 Fig. 13A—H Hypolobocera kamsarum Campos & Rod- riguez, 1995:649, Figs. 1, 2. Material examined.—Colombia. Putu- mayo Department, Mocoa, Vereda Alto Campucana, 1350 m alt., 2 Jun 1994, leg. Or Ve) Castanion @ holotype, 14:0) e237 mm, ICN-MHN-CR 1349.—Same data, | d paratype, 13.4 X 21.8 mm, | @ paratype, 14.6 X 24.8 mm, 2 juveniles, ICN-MHN- CR 1350. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile and fixed fingers (Fig. 13H). First male gonopod with caudal ridge long, concave; almost reaching apex (Fig. 13A). Lateral lobe small, trans- verse, displaced towards cephalic side, evenly rounded in lateral view, placed far from apex (Fig. 13A—D). Apex outline oval in distal view, with rounded expansion di- rected distally. Mesocaudal projection of spermatic channel terminating in rounded papilla. Mesial border projected proximally, forming strong triangular mesial lobe (Fig. 13C, E). Third maxilliped with exognath relatively long, approximately 0.60 times length of ischium (Fig. 13G). Remarks.—This species differs from the others in the genus in having the mesial border projected proximally, forming a strong triangular mesial lobe (Fig. 13C, E). Hypolobocera lloroensis Campos, 1989 Fig. 14A—G Hypolobocera lloroensis Campos, 1989: 143, fig. 1. Hypolobocera lloroensis.—Rodriguez, 1992:183.—Rodriguez et al., 2002:7. Hypolobocera chocoensis.—Prahl, 1988: NIU; Mas Material examined.—Colombia, Choco Department. Llor6, Vereda Penalosa, Gran- 780 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON N) CPinzey Fig. 12. Hypolobocera gorgonensis Prahl 1983, male, ICN-MHN-CR 0612: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; E right side of carapace, dorsal view; G, left third maxilliped, external view; H, left chela, external view. VOLUME 116, NUMBER 3 781 NCRINtS4 ‘ Fig. 13. Hypolobocera kamsarum Campos & Rodriguez, 1995, male holotype, ICN-MHN-CR 1349: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; E right side of carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. 782 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON UCPinzen Fig. 14. Hypolobocera lloroensis Campos 1989, male, ICN-MHN-CR 1386: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; FE right side of carapace, dorsal view; G, left third maxilliped, external view. VOLUME 116, NUMBER 3 ja Experimental CEMA, Universidad de Choco, La Lana, 50 m alt., 22 Mar 1988, leg. M. C. Ardila, M holotype, 17.8 X 30.3 mm, 1 6 paratype, 11.0 x 17.5 mm, ICN- MHN-CR 0850.—Condoto, 23 Jul 1985, les. Jal vom imal, IG, Wtoll X 210) imovesk, Ih Bo WNL OS ZBeil maven, (IIBMWMONY soils (045-0, 045-1 ADT-CRBMUV).—Condo- to, Inspeccion Santa Rita de Iro, Iro River, 14 Jun 1987, leg. H. Almendiger, 1 ¢, 23.6 x 40.6 mm, CRBMUV 87016.—Istmina, San Juan River, 8 Aug 1969, leg. Dale Lit- tle, 1 d, cl 24.4 X 40.0 mm, 2 2, 26.3 X AS Om imime oro oN Zio mm, do) 693) == Tutunend6, km 29 Quibd6-Carmen de Atra- to Highway, 170 m alt., 25 Aug 1987, leg. IR, Seincnez, iI SG; LY « 19 inna, il 2, 11.2 X 17.9 mm, | juvenile, ICN-MHN-CR 0783.—Tad6o, km 20.6 Santa Cecilia-Quib- dd, 100 m alt., 10 Abr 1991, leg. G. Susa- (ema, 26 2, 1S « BilS (© NAO) « 220 icoviny | juvenile, ICN-MHN-CR 1262.—Tutunen- do, La Barbuda stream, 100 m alt., 17 Oct LOST, les, C. IRomem, 2 2. 29 « 372 mia, 20.4 X 34.5 mm, ICN-MHN-CR 1263. Antioquia Department, Urrao, Andabu stream, 1800 m alt., 23 Sep 1994, leg. I. D. Vélez, SG, FBS. 403 tt WO. x Zi mm, 6 2, 25.1 X 42.1 to 17.4 X 29.0 mm, ICN-MHN-CR 1386. Diagnosis.—Anterolateral margin of car- apace with deep depression behind external orbital angle, followed by another deep de- pression at cervical groove level. First male gonopod with caudal ridge long, straight, recurved proximally; ending in narrow ridge distally (Fig. 14A). Lateral lobe sub- triangular, wide proximally, narrow distally (Fig. 14A, C). Apex outline oval in distal view, mesocephalic border transversely ex- panded; prominent cephalic papilla. Meso- caudal projection of spermatic channel ter- minating in semiacute papilla. Mesial lobe semicircular, with acute spine directed lat- erally (Fig. 14C, E). Third maxulliped with exognath approximately 0.20 times length of ischium (Fig. 14G). Remarks.—Prahl’s (1988) report of two males of Hypolobocera chocoensis from 783 Condoto, Chocd (045-0, 045-1 ADT- CRBMUV), actually are of H. lloroensis. Contrary to Prahl’s report, the lot does not contain two males, but one male (18.1 X 28.0) and one female (14.7 X 23.1 mm). Hypolobocera malaguena Prahl, 1988 Fig. 1SA—G Hypolobocera malaguena Prahl, 1988:180, os, JKC hike Hypolobocera malaguena.—Rodriguez, 1992:183.—Rodriguez, 1994:300. Material.—Colombia. Valle del Cauca Department, Malaga Bay, La Alegria stream, 23 Dec 1985, leg. N. Ospina, ¢ ho- lotype, 24.6 x 40.7 mm, 0750 ADT- CRBMUV. Diagnosis.—Anterolateral margin of car- apace with depression fringed with approx- imately 8 papillae, behind external orbital angle, followed by shallow depression at level of cervical groove (Fig. 15F). First male gonopod with caudal ridge long, sin- uous; almost reaching apex (Fig. 15A). Lat- eral lobe subtriangular with distal border rounded. Lateral lobe oblique in relation to axis of appendage (Fig. 15B). Apex outline oval in distal view, mesocephalic border transversely expanded; prominent cephalic papilla. Mesocaudal projection of spermatic channel terminating in blunt papilla. Mesial lobe semicircular, with acute spine directed laterally (Fig. ISE). Third maxilliped with exognath approximately 0.25 times length of ischium (Fig. 15G). Remarks.—The specific name malague- na was given by (Prahl 1988) to this spe- cies. However, the Code does not allow the use of the “‘h”’ in a latinized scientific name and requires correction to H. malaguena. This species closely resembles H. lloroensis in the shape of the first male gonopod (Fig. 14A), but the lateral lobe is oblique in re- lation to the axis of the appendage in H. malaguena (Fig. 15B), while it is parallel in H. lloroensis (Fig. 14B). 784 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 15. Hypolobocera malaguena Prahl 1988, male, CRBMUV 85133: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; E right side of carapace, dorsal view; G, left third maxilliped, external view. VOLUME 116, NUMBER 3 Hypolobocera martelathani (Pretzmann, 1965) Fig. 16A—I Strengeria Strengeria martelathani Pretz- mann, 1965:6. Hypolobocera (Hypolobocera) martela- thami.—Pretzmann, 1971:17.—Pretz- mann, 1972:50, figs. 159-161, 242-244. Hypolobocera matelathami.—Rodriguez, 1982a:52.—Campos & Rodriguez, 1984: 538, fig. 4c, f—Rodriguez, 1994:300, fig. 1D—-F Hypolobocera merenbergiensis Prahl & Giraldo, 1985:2, fig. 1. Hypolobocera merenbergiensis.—Rodri- guez, 1992:183.—Rodriguez et al., AN T Material examined.—Colombia, Huila Department. San Agustin, De Quintero stream, near Yalconia Hotel, 1500 m alt., 27 ENS NOS, WO Ss NO X< Zoi OWS Xx 13-9 jen, IA Qe NGG Ye eval Tro ET keh) mm, ICN-MHN-CR 0653.—San José de Is- nos, Vereda Primavera, 1600 m alt., 27 Aug 198695) 6, 1885322 to-43,2 < 20/0) mm, HO cleo 290) to, IO 175 mim: ICN-MHN-CR 654.—San Agustin, Inspec- cidn Obando, 1400 m alt., 28 Aug 1986, 5 6, I4LIL XK BBB oO N23 <~ IOs sin, (GL, LOW << 25,5 ro JN 26 IY seating MONE MHN-CR 0655.—Acevedo, Inspeccié6n San Adolfo, Vereda Changuayaco, 1300 m alt., 29 Aug 1986, 13 6, 14.8 XK 24.8 to 12.4 X IS. cron, MOS, lS < BID ito WO T/ mm, ICN-MHN-CR 0660.—La Plata, Mer- enberg stream, 2300 m alt., 9 Apr 1982, leg. H. von Prahl and J. Giraldo, M holotype, 13.1 X 20.1 mm, No. 42 (lote 006 ADT) MBMUV.—Same data, 1 6 paratype, No. 43 dote 006 ADT) MBMUV.—Same data, 1 2 paratype, 14.2 X 22.5 mm, ICN-MHN- CR 0541.—Same data, 3 2, 14.0 X 23:4 to 11.5 X 18.2 mm, CRBMUV 82051.—Same Gata Dao) clal223e ee oeormmn OLS X16: mm, TU 6369.—Cauca, Inza, Vereda Tier- ras Blancas, 2200 m alt., 25 Mar 1982, leg. R. Restrepo, 4 6, 14.1 X 22.9 to 12.9 X 785 AQ Pn TOOT; 3 S25 iteall XX AE0) to U4! il 2x wy mm, ICN-MHN-CR 0087. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile and fixed fingers (Fig. 161). First male gonopod with caudal fusiform; ending in narrow ridge distally (Fig. 16A). Lateral lobe large, reaching middle of gonopod, wide distally, narrow proximally; covered with minute spinules and scattered short setae in caudal view (Fig. 16A, B, D, E); slightly semicir- cular in cephalic view (Fig. 16C). Apex outline oblong in distal view, with caudo- cephalic border slightly rounded. Mesocau- dal projection of spermatic channel termi- nating in wide papilla. Mesial lobe elon- gated and subtriangular (Fig. 16D, F). Third maxilliped with exognath approximately 0.50 times length of ischium (Fig. 16H). Remarks.—The morphology of the first male gonopods of Hypolobocera martela- thani and H. merenbergiensis Prahl & Gir- aldo, 1985 are identical, and both species are found in the Central Cordillera, in the upper reaches of the Magdalena River. Thus, H. merenbergiensis is considered a junior synonym of H. martelathani. The original spelling of the species name is martelathani (Pretzmann, 1965), although Pretzmann (1971, 1972) later used the spelling martelathami. Hypolobocera meineli Prahl, 1988 Fig. 17A-I Hypolobocera meineli Prahl, 1988:173, figs. 3, 4. Hypolobocera meineli.—Rodriguez, 1992: SS) Material examined.—Colombia. Valle del Cauca Department, Naya River, 15 Feb 1984, leg. R. Rios, 2 6, 043-00 ADT- CRBMUV. Narino Department, Ricaurte, 3 km to Pasto, Apr 1982, leg. L. Gomez, | 6, 27.5 X 44.6 mm, CRBMUV 82058.— Ricaurte, 22 May 1984, leg. L. Gomez, 1 3d paratype, 043-1 ADT-CRBMUV. Cauca Department, Naya River, Feb 1984, leg. uk, DBE SS SO (eis) imine Stee looall, sonra 786 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON J (Pine 9 Fig. 16. Hypolobocera martelathani (Pretzmann, 1965), male holotype, No. Cat. 82051, CRBMUV: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, lateral lobe, caudolateral view; FE same, apex, distal view; G, right side of carapace, dorsal view; H, left third maxilliped, external view; I, right chela, external view. VOLUME 116, NUMBER 3 787 SCPin2en Fig. 17. Hypolobocera meineli Prahl 1988, male, CRBMUV 84075: A, left first gonopod, caudal view; C, same, lateral view; D, same, cephalic view; E, same, apex, distal view; G, right side of carapace, dorsal view; H, left third maxilliped, external view; I, right chela, external view. Male, ICN-MHN-CR 1866: B, left first gonopod, caudal view; KE apex, distal view. 788 CRBMUV 84075.—Guapi, Guapi River, El Naranjo, 50 m alt., 20 Sep 2000, leg. E. Guerra, 1 6, 29.8 X 46.9 mm, ICN-MHN- CR 1866. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile and fixed fingers (Fig. 171). First male gonopod with caudal ridge long, either fusiform thickened at proximal half or prominent and wide; ending in narrow ridge beyond lateral lobe. Mesial margin with or without rows of con- spicuous setae (Fig. 17A, B). Lateral lobe small, slightly rounded, with or without slightly middle notch (Fig. 17A, B). Apex outline oval in distal view; expanded into a triangular projection cephalically. Mesocau- dal projection of spermatic channel termi- nating in wide papilla. Mesial lobe reduced as strong fold along margin (Fig. 17E, F). Third maxilliped with exognath approxi- mately 0.40 times length of ischium (Fig. 17H). Remarks.—Prahl (1988) designated as holotype a male (043-00 ADT-CRBMUV) from Naya River, and as paratype a male (043-1 ADT-CRBMUV) from Ricaurte, without indicating their sizes. He only pro- vided the measurements of the largest male (43.2 X 60.3 mm). In a recent examination of Prahl’s material at the CRBMUV the la- bels with the catalogue numbers corre- sponding to the type material could not be found. Only two males from Naya River (CRBMUV 84075) were located, but with sizes that do not match those indicated by Prahl (1988). Hypolobocera murindensis, new species Figs. 18A—D, 19A—F Holotype.—Colombia, Antioquia Depart- ment, Murindo, 25 m alt., 23 Sep 1994, leg. Ie ID, Wellez @, hig) << Dil’ sone, ICING MHN-CR 1388. Paratypes.—Same locality data as holo- iyoes 2 6, Wiles < QS mira, IOS X Wess mm, 1 2,9.5 X 16.5 mm, 5 juveniles, ICN- MHN-CR 1389. Diagnosis.—Chelae of male lacking tu- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON bercles on external base of mobile and fixed fingers (Fig. 18A). First male gonopod with caudal ridge long, fusiform; almost reach- ing apex (Fig. 19A, E). Lateral lobe prom- inent, subquadrate, external margin smooth (Fig. LOA, C, E). Apex outline oval in distal view; caudolateral expanded into rounded projection; cephalic border with row of spi- nules; prominent cephalic papilla. Meso- caudal projection of spermatic channel ter- minating in semiacute papilla. Mesial lobe slightly semicircular (Fig. 19C—F). Third maxilliped with exognath approximately 0.20 times length of ischium (Fig. 18C). Description.—Carapace (Fig. 18A, B) with cervical groove straight, shallow, end- ing some distance from lateral margin. An- terolateral margin with shallow depression behind external orbital angle. Lateral mar- gin with approximately 10 irregular tuber- cles. Postfrontal lobes small, rounded; de- limited anteriorly by 2 depressions. Median groove shallow. Surface of carapace in front of postfrontal lobes regularly inclined anteriorly. Front low, upper border well de- marcated with row of coalescent papillae; lower margin visible in dorsal view, sinu- ous in frontal view. Dorsal surface of car- apace smooth, covered by small papillae, regions distinctly demarcated (Fig. 18A, B). Third maxilliped with rounded angle on distal half of external margin of merus; ex- ognath approximately 0.20 times length of ischium (Fig. 18B, C). Orifice of efferent branchial channel irregularly ovate (Fig. 18D). First pereiopods heterochelous (Fig. 18A), right chela larger than left. Merus with 3 longitudinal crests as follows: upper crest with rows of tubercles, internal lower crest with rows of teeth, and external lower crest with row of tubercles. Carpus with 3 tubercles on internal crest, and blunt distal spine. Palms of both chelae smooth, and moderately swollen, fingers of larger chela slightly gaping when closed, finger tips crossing and surface of palms and fingers with rows of minute dark tubercles, (Fig. 18A). Walking legs (pereiopods 2—5) (Fig. VOLUME 116, NUMBER 3 ( \ pelt { EG; esi 5 Fig. 18. 789 Imm Oo mm Hypolobocera murindensis, new species, male holotype, ICN-MHN-CR 1388: A, carapace and pereiopods, dorsal view; B, carapace, frontal view; C, left third maxilliped, external view; D, opening of left efferent branchial channel, external view. 18A) with dactyli elongated, 5 longitudinal rows, and 4 to 6 spines on each row. First male gonopod with caudal ridge long, fusiform; almost reaching apex (Fig. 19A, E). Lateral lobe prominent, subquad- rate, external margin smooth (Fig. 19A, C, E). Apex outline oval in distal view, cau- dolateral expanded into rounded projection; cephalic border with row of spinules; prom- inent cephalic papilla. Mesocaudal projec- tion of spermatic channel terminating in semiacute papilla. Mesial lobe slightly semicircular and projected on the caudal surface (Fig. 19A, C—F). Color.—In alcohol, dorsal surface of the carapace light brown (near 136, Raw Si- 790 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 19. Hypolobocera murindensis, new species, male holotype, ICN-MHN-CR 1388: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, lateral lobe, caudal view; E same, apex, distal view. enna) with dark brown (near Raw Umber, 223) speck on postfrontal portion of cara- pace. Walking legs and chelae brown (near True Cinnamon, 139) dorsally, and light brown (near Sayal Brown, 223C) ventrally. Ventral surface of the carapace brown (Ve- rona Brown, 223B). Etymology.—The specific name refers to Murind6, where the specimens were col- lected. Remarks.—This species belongs to group 6, and is most similar to Hypolobocera an- dagoensis. The two can be distinguished by differences in the third maxilliped, and in the first male gonopod. In the new species the merus of the third maxilliped has a rounded angle, whereas in H. andagoensis it is sharp (Fig. 2F). The lateral lobe of the gonopod in H. murindensis is thick, and subquadrate, with the external margin rounded, slightly convex in lateral view, and with a narrow constriction below the VOLUME 116, NUMBER 3 lateral lobe. In contrast, H. andagoensis has the lateral lobe thin, narrower distally than proximally, with a faintly crenulated exter- nal margin, and a wide constriction below the lateral lobe (Fig. 2A, B). The mesial lobe is projected as a rounded lobe on the caudal surface, slightly semicircular in dis- tal view in H. murindensis, whereas, in H. andagoensis it is subtriangular, and not pro- jected on the caudal surface (Fig. 2D). Hypolobocera mutisi Prahl, 1988 Fig. 20A—G Hypolobocera mutisi Prahl, 1988:175, figs. De Os Hypolobocera mutisi.—Rodriguez, 1992: 183. Material examined.—Colombia, Valle del Cauca Department. Cajambre River, 500 m alt., 16 Jul 1983, leg. R. Rios, M holotype, 042-0 ADT-CRBMUV.—Same data, | d paratype (USNM 210728).—El Pinuelal, Cajambre River, 16 Aug 1983, leg. R.R., 1 ¢d, 23.3 X 38.6 mm, CRBMUV 83086. Diagnosis.—Chelae of male with small tubercle on external base of mobile fingers, and swelling on external base of fixed fin- gers (Fig. 20G). First male gonopod with caudal ridge long, straight; ending in nar- row ridge beyond lateral lobe (Fig. 20A). Lateral lobe outline irregular, slightly trap- ezoidal; projected into narrow expansion distally (Fig. 20A, C). Apex outline elon- gated in distal view, caudolaterally expand- ed into rounded projection; prominent, acute cephalic papilla. Mesocaudal projec- tion of spermatic channel terminating in rounded papilla. Mesial lobe elongated, subtriangular (Fig. 20C, D). Third maxilli- ped with exognath approximately 0.30 times length of ischium (Fig. 20F). Remarks.—An exhaustive search of the CRBMUV collections failed to produce the holotype of this species. Only the male specimen from E] Pinuelal, Cajambre River (CRBMUV 83086) was found, and is illus- trated herein. 791 Hypolobocera noanamensis Rodriguez, Campos & Lopez, 2002 Fig. 21A—H Hypolobocera noanamensis Rodriguez, Campos & Lopez, 2002:4—6, fig. 1A—H. Material examined.—Colombia, Choc6 Department, Noanama, San Juan River, 50 m alt., 4°42’'N, 26°56’W, 8 Aug 1969, leg. Dale Little, ¢ holotype, 50.9 X 80.3 mm, TU 6191.—Same data, 1 @ paratype, 53.9 X 30.3 win, NY 3337. Diagnosis.—Chelae of male without tu- bercle on external base of mobile fingers, with a swelling on external base of fixed fingers (Fig. 21H). First male gonopod with caudal ridge strong, fusiform; ending in narrow ridge beyond lateral lobe (Fig. 21A). Lateral lobe small, subtriangular, dis- tal angle rounded, placed transversely in re- lation to axis of appendage; cephalic sur- face with tuberculated crest (Fig. 21A—C). Apex outline oval in distal view, mesoce- phalic border rounded expanded; flat papilla on caudolateral border. Mesocaudal projec- tion of spermatic channel terminating in rounded papilla with spiny ridge on ce- phalic side. Mesial lobe subtriangular (Fig. 21E). Third maxilliped with exognath 0.30 times length of ischium (Fig. 21G). Remarks.—This species closely resem- bles Hypolobocera beieri in the shape of the lateral lobe, and the outline of the apex of the first male gonopod. The species can be differenciated from each other by the spiny ridge on the cephalic side of the me- socaudal projection in H. noanamensis, which is lacking in H. beieri. Hypolobocera rotundilobata Rodriguez, 1994 Fig. 22A—H Hypolobocera bouvieri rotundilobata Rod- Weve ZIAD O7 hes 2. Hypolobocera bouvieri rotundilobata.— Rodriguez et al., 2002:6. Material examined.—Colombia. Choco Department, Mountains of upper San Juan WE2 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON UCPiMIS 4 Fig. 20. Hypolobocera mutisi Prahl 1988, male, CRBMUV 83086: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, apex, distal view; E, right side of carapace, dorsal view; E left third maxilliped, external view; G, left chela, external view. VOLUME 116, NUMBER 3 793 Fig. 21. Hypolobocera noanamensis Rodriguez, Campos & L6pez, 2002, male holotype, TU 6191: A, left first gonopod, caudal view; B, same, lateral view; C, same, laterocephalic view; D, same, cephalic view; E, same, apex, distal view; EF right side of carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. 794 Fig. 22. ICP PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Hypolobocera rotundilobata Rodriguez, 1994, male holotype, USNM 240104: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; FE right side of carapace, dorsal view; G, left third maxilliped, external view; H, left chela, external view. River, nearest village Playa de Oro, 28 Mar 1962, leg. M. Latham, d holotype, 46.2 x 68.4 mm, USNM 240103.—Same data, 5 6 jearminjoes, 154! < 23 t© IDI « BWC mm, | &, 18.0 X 31.6 mm, USNM 240104.—Borderline between Choco and Valle del Cauca Departments, Paso de Ga- lapagos, 15 May 1985, leg. M. Alberico, 2 6, 13.8) X 23.3 iin, IAA) X< Dill mmm, © ©, 13.1 X 22.6 to 10.2 X 17.2 mm, CRBMUV VOLUME 116, NUMBER 3 85147.—Choc6 Department, San José del Palmar, Ingara River, 2000 m alt., 22 Oct 2000, leg. E. Guerra, 1 6, 13.1 X 22.8 mm, 1 2, 13.3 X 22.9 mm, ICN-MHN-CR 1867, 1868. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile and fixed fingers (Fig. 22H). First male gonopod with caudal ridge long, straight; ending in nar- row ridge distally (Fig. 22A). Lateral lobe slightly semicircular (Fig. 22A, C). Apex outline subtriangular in distal view; lateral border straight; rounded cephalic papilla. Mesocaudal projection of spermatic chan- nel terminating in semiacute papilla. Mesial lobe subtriangular and projected mesially (Fig. 22A, C, E). Third maxilliped with ex- ognath 0.30—0.40 times length of ischium (Fig. 22G). Remarks.—Rodriguez (1994) described this taxon as Hypolobocera bouvieri rotun- dilobata. The subspecies of H. bouvieri are characterized by the male chelae, featuring a small, rounded or irregularly shaped tu- bercle on the external base on each of the the mobile fingers, and a swelling or prom- inent tubercle on external base on each of the fixed fingers. The apex outline of the first male gonopod is oval, with prominent cephalic papilla, and an auxiliary rounded papilla near the spermatic channel. How- ever, in H. rotundilobata the male chelae lack tubercles on the external base of the mobile and fixed fingers. The apex outline is subtriangular, and without an auxiliary papilla. In view of this, H. b. rotundilobata is elevated to specific rank. Hypolobocera steindachneri Pretzmann, 1968 Hypolobocera (Hypolobocera) monticola steindachneri Pretzmann, 1968:9.— Pretzmann, 1971-17 —Pretzmann, 1972: 46, figs. 204, 205, 227-229. Hypolobocera steindachneri.—Rodriguez, 1982a:58. Material examined.—Colombia, Valle del Cauca Department. 1897, leg. Stein- 795 dachnerie S) holotype, cli 27-7 mm, 1d paratype, cl 20.6 mm (Museum Wien Nr. 3779).—Bitaco, 1957, leg. M. Latham, 2 6, Cll Ws iS3) ian Se ell Dil Os US6 Sisk USNM. Remarks.—\ have been unable to exam- ine any material of this species. Pretzmann (1968:9) in his original description of Hy- polobocera (Hypolobocera) monticola steindachneri described as features “the first male gonopod is slender; the lateral lobe is smaller, the external margin is fold- ed back strongly, and it is placed far from apex. The exognath of third maxilliped is 0.12 the length of ischium’’. Pretzmann (1972) described again, including illustra- tions, the morphological features of H. (H.) m. steindachneri, partially contradicting his previous description (Pretzmann, 1968). He stated that “‘the lateral lobe is placed not far from apex”’ and that “‘the exognath of third maxilliped is shorter, only 0.17 the length of ischium’’. Hypolobocera velezi, new species Fig. 23A—H Holotype.—Colombia. Choco6 Depart- ment, Carmen de Atrato, Vereda El Veinte, 2400 m alt., 25 May 1994, leg. I. D. Vélez, 1 5, 16.1 X 29.2 mm, ICN-MHN-CR 1387. Paratypes.—Same locality data as holo- Oyoe: | 2. IO X< 23-2 mim, | Owierous 2. 15.3 X 27.4 mm, ICN-MHN-CR 1889. Non-paratypes.—Colombia. Risaralda Department, Pueblo Rico, Corregimiento Santa Cecilia, Vereda La Granja, 600 m alt., 23, OCH 199Ie Is, 1555.6 25.4 sam, 12. 16.5 X 27.1 mm, ICN-MHN-CR 1307. Choc6 Department, Carmen de Atrato, km 53 Carmen de Atrato-Quibd6 Highway, 420 mali 27 Aue 1987) lees Re Sanchez, 2 2% Oe AS emms e996, mm: 1EN-= MHN-CR 0784. Diagnosis.—Chelae of male lacking tu- bercles on external base of mobile and fixed fingers (Fig. 23H). First male gonopod with caudal ridge strong, straight, narrow distal- ly; ending some distance from apex (Fig. 796 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 23. Hypolobocera velezi, new species, male holotype, ICN-MHN-CR 1387: A, left first gonopod, caudal view; B, same, lateral view; C, same, cephalic view; D, same, mesial view; E, same, apex, distal view; E right side of carapace, dorsal view; G, left third maxilliped, external view; H, right chela, external view. 23A). Lateral lobe subtriangular; proximal external margin angled and curved upward (Fig. 23A, C). Apex outline slightly round- ed in distal view, mesocaudally expanded into rounded projection; rounded papilla near cephalic border. Mesocaudal projec- tion of spermatic channel terminating in semiacute papilla. Mesial lobe subtriangular VOLUME 116, NUMBER 3 (Fig. 23C, E). Third maxilliped with ex- ognath approximately 0.20 times length of ischium (Fig. 23G). Description.—Carapace (Fig. 23F) with cervical groove deep, straight, recurved posteriorly, ending some distance from lat- eral margin. Anterolateral margin with shal- low depression behind external orbital an- gle, anterolateral border not continuous with margin of depression, but runing dor- sally to it, forming rounded lobe. Lateral margin with approximately 12 blunt teeth. Postfrontal lobes oval, low; delimited an- teriorly by 2 transverse depressions. Medi- an groove wide, shallow. Surface of cara- pace in front of postfrontal lobes regularly inclined anteriorly. Front lacking distinct upper border, regularly curving downward; slightly bilobed in dorsal view; lower mar- gin slightly sinuous in frontal view. Dorsal surface of carapace smooth, covered by small papillae, regions distinctly demarcat- ed (Fig. 23F). Third maxilliped with sharp angle on distal half of external margin of merus; exognath approximately 0.20 times length of ischium (Fig. 23G). Orifice of ef- ferent branchial channel slightly ovate. First pereiopods heterochelous, right chela larger than left. Merus with 3 longitudinal crests as follows: upper crest with rows of tuber- cles, internal lower crest with rows of teeth, and external lower crest with row of tuber- cles. Carpus with 3 tubercles on internal crest, and blunt spine distally. Palm of male chelae swollen, fingers gaping when closed. Walking legs (pereiopods 2—5) with 5 lon- gitudinal rows, and 4 to 6 spines on each row. First male gonopod with caudal ridge strong, straight, narrow distally; ending at some distance from apex (Fig. 23A). Lat- eral lobe subtriangular; proximal external margin angled and curved upward (Fig. 23A, C). Apex outline slightly rounded in distal view, mesocaudally expanded into rounded projection; rounded papilla near cephalic border. Mesocaudal projection of spermatic channel terminated in semiacute 797 papilla. Mesial lobe subtriangular (Fig. 23 C7): Color.—In alcohol, dorsal surface of the carapace brown (near 37, Antique Brown) with light brown (near Cinnamon, 39) speck on posterior portion of carapace. Walking legs and chelae brown (near 37, Antique Brown) dorsally, and light brown (near Sayal Brown, 223C) ventrally. Ventral surface of the carapace light brown (near Cinnamon, 39). Etymology.—The species is named in honor of Colombian scientist Dr. Ivan Dario Vélez, professor of the Universidad de An- tioquia, who promotes research in Paragon- imosis and other tropical diseases, and who collected the specimens. Remarks.—This species belongs to group 4, and is most similar to Hypolobocera llo- roensis. The two can be differentiated by feature of the first male gonopod. The prox- imal external margin of the lateral lobe in H. velezi is angled, and curved upward (Fig. 23A, C), whereas that margin is rounded, and not caudally curved in H. I/lo- roensis (Fig. 14A). Key to Species and Subspecies of Hypolobocera from Colombia 1. Outline of apex of gonopod in distal view oval, rounded or subtriangular 2 — Outline of apex of gonopod in distal WIEW TECLONPatediamet. 1 a5 2 ae te 16 2. Lateral lobe rounded or subtriangular 9 — Lateral lobe regularly rounded ..... 3) 3. Lateral lobe with strong triangular pro- CESS. CRIA ae AL Pte aden ed ht Oe H. alata — Lateral lobe without strong triangular PROCESS Bierce dr NEMS ARE CEE EEL oh OM + 4. Latero-cephalic border of apex toothed PCE MOS ia SEY. reo Tan H. dentata — Latero-cephalic border of apex not CODEN CCU re ir aerek. HME neh ESO ae rats Jens 5) 5. Lateral lobe displaced towards cephalic sidev(Eie= 13 B=) seam oe oe H. kamsarum — Lateral lobe not displaced towards ce- plalicgsiGe sarin. Aarts 2h oe See 6 6. Mesial lobe well developed, subtrian- CUAGFOISCOMCIRCULARM es 5 alee 7 798 IY), DO: . Outline of lateral lobe irregular PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Mesial lobe reduced as strong fold along margin H. meineli OO 0S OO 0 6.16 D0-.0' 10 . Lateral border of apex straight in distal view (Fig. 22E) H. rotundilobata Lateral border of apex rounded in distal Wey, (siless Be, 48, OS) 6 occa 006 8 With auxiliary papilla near spermatic Channel geen o48 ¢ oh Oe Oe 19 Without auxiliary papilla near spermat- ic Channell At ee eee eee ee 9 Mesocaudal projection with spiny ridge on cephalic side H. noanamensis Mesocaudal projection without spiny ridge on cephalic side H. beieri Lateral lobe rounded or subtriangular 11 2 © © © © © e Lateral lobe subquadrate .......... U3 . Proximo-external margin of lateral lobe OUNC SOs Sees Pears acon gene 12 Proximo-external margin of lateral lobe angled and curved upwards .... H. velezi . An auxiliary lobe parallel to lateral lobe ee © © © 6 8 on lateral side H. cajambrensis No auxiliary lobe on lateral side .... 13 . External margin of lateral lobe faintly crenulated fn, Aan at H. andagoensis External margin of lateral lobe not Crenullated te Buys H. murindensis . Distal border of lateral lobe rounded (Fig. 15A, C) H. malaguena Distal border of lateral lobe not round- EL Bie AO B))) ere a en ee ae see 15 . Mesocephalic border of apex transverse expanded (Fig. 14E) H. lloroensis Mesocephalic border of apex not trans- verse expanded (Fig. 9F) H. chocoensis . Distal end of lateral lobe with a deep notch H. emberarum Distal end of lateral lobe lacking notch 17 oe © © © © © © © © ee ew ew ell A. mutisi Outline of lateral lobe regularly re- curved . Apex in distal view with bifid papilla near spermatic channel H. gorgonensis Apex in distal view with slightly round- ed papilla near spermatic channel .... H. martelathani stad Lene 20 ee © © © © © © © ee ee ell elle lll Lateral lobe subtriangular Lateral lobe semicircular Mesocaudal projection of spermatic channel with distal spinule H. bouvieri monticola Mesocaudal projection of spermatic 2 2 © © © © Coc0C0O oO OOo oO oO Oo Goo oO channel without distal spinule H. bouvieri angulata 21. Chelae of male with prominent tubercle on external base of fixed fingers ee eI eo ad H. bouvieri bouvieri — Chelae of male without prominent tu- bercle on external base of fixed fingers Me Milan Si sue chy H. bouvieri stenolobata 2 2 © © © © © © © © ew ew ele lll Summary of Distributions This review of the genus Hypolobocera from Colombia has made possible the study of the geographic distribution of this genus. However, some species are known from single or few localities, and additional sam- pling is needed. The distribution of Hypolobocera species comprises a vast portion of the Colombian territory (Fig. 24), and includes most of Co- lombia’s major basins: the Magdalena, Cau- ca, and Atrato Rivers basins, which drain to the Caribbean; the San Juan, Anchicaya, Dagua, Cajambre, Naya, and Guapi Rivers basins, which drain to the Pacific; and the Caqueta River basin, which drains to the Amazon River. Seven species and two subspecies have trans-basin distribution: Hypolobocera buenaventurensis, H. cajambrensis, H. cho- coensis, H. lloroensis, H. meineli, H. velezi, H. beieri, H. bouvieri angulata, and H. b. monticola. The distribution of the latter subspecies is particularly interesting, be- cause it encompasses rivers that flow into the Atlantic and the Pacific Oceans. Hypolobocera bouvieri bouvieri, H. b. stenolobata, and H. martelathani are found exclusively in the Magdalena River basin, while H. alata, H. andagoensis, and H. ro- tundilobata have only been found in the San Juan River basin; H. emberarum occurs only in the Atrato River basin. Hypolobocera bouvieri bouvieri has an extensive distribution on the slopes of the Central and Eastern Cordilleras, in systems that drain to the Magdalena River. Another subspecies with a broad trans-basin distri- bution is H. b. monticola. The known dis- tribution of this subspecies includes the VOLUME 116, NUMBER 3 799 182 ey? Og 6° 4o 2° caqueta Riy, tf ECUADOR os 74° Fig. 24. Geographical distribution of Hypolobocera in Colombia. a, H. andagoensis; al, H. alata; b, H. beieri; an, H. bouvieri angulata; bo, H. bouvieri bouvieri; m, H. bouvieri monticola; sten, H. bouvieri stenolo- bata, bu, H. buenaventurensis; c, H. cajambrensis; ch, H. chocoensis; d, H. dentata; e, H. emberarum; g, H. gorgonensis; k, H. kamsarum; \l, H. lloroensis; ma, H. malaguena; mar, H. martelathani; me, H. meineli; mu, H. murindensis; mut, H. mutisi; n, H. noanamensis; r, H. rotundilobata; stei, H. steindachneri; v, H. velezi. The bold, dashed lines indicate continuous distribution ranges, and the dotted line a disjunct distribution. 800 Central and Western Cordilleras in the mid- dle course of the Magdalena and Cauca River basins. This subspecies has a great potential to colonize new habitats due, in part, to the ability of females to hatch large number of eggs. The distribution of H. Db. angulata includes the Sierra Nevada de Santa Marta to both slopes of the Sierra de Perija, and also the Cordillera de Mérida in Venezuela. This disjunct distribution, cov- ering the Cesar and Catatumbo River ba- sins, is considered a relict distribution mod- ified by temperature changes (G. Rodri- guez, pers. comm.). Hypolobocera lloroensis and H. cho- coensis are known from the headwaters of the Atrato River, which flows northwards into the Gulf of Uraba to the Caribbean Sea. In addition, records are here included from the basin of the San Juan River which drains into the Pacific Sea, for H. chocoen- sis (upper San Juan River, Choc6é, USNM 240102; Condoto, 045-0, 045-1 ADT- CRBMUV); and HA. lloroensis (Condoto, CRBMUV 85132; CRBMUV 87016; Ist- mina, TU 6193; Tad6, ICN-MHN-CR 1262). This trans-basin distribution can be attributed to the intermittent communica- tion between both basins which can occur during flooding seasons in areas below 100 m of altitude. Hypolobocera dentata, H. kamsarum, H. malaguena, H. mutisi, H. murindensis, and H. noanamensis are species known from single localities. The vertical distribution of the species (Table 1) ranges from 20 to 2500 m. Ay- polobocera lloroensis extends from 50 to 2400 m along the San Juan River and head- water of the Atrato River. Hypolobocera andagoensis, H. buenaventurensis, H. ca- jJambrensis, H. chocoensis, H. gorgonensis, H. malaguena, H. meineli, H. murindensis, H. mutisi, and H. noanamensis, have been found along the coastal plain of the Pacific, between 20 and 700 m. The other species from the Pacific drainage are H. beieri, be- tween 500 and 1600 m; HA. rotundilobata, between 70 and 2000 m. The inland spe- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Altitudes (in meters above sea level) re- ported for Colombian Hypolobocera. Hypolobocera alata 490-700 H. andagoensis 20-50 H. beieri 500—1600 H. bouvieri angulata 400-1300 H. bouvieri bouvieri 270-2050 H. bouvieri monticola 300-1850 H. bouvieri stenolobata 1175-1200 H. buenaventurensis 20-200 H.. cajambrensis 50—700 H. chocoensis 20—500 H. dentata 1600 H. emberarum 1800—2500 H. gorgonensis 20-300 H. kamsarum 1350 H. lloroensis 50-2400 H. malaguena 0-50 H. martelathani 1300—2300 H. meineli 50-100 H. murindensis DS) H. mutisi 500 H. noanamensis 50 H. rotundilobata 70—2000 H. steindachneri 1500 H. velezi 420—2400 cies, which range between 270 and 2500 m are: H. dentata, H. emberarum, H. marte- lathani, H. steindachneri, H. velezi, and the subspecies H. b. bouvieri, H. b. monticola, and H. b. stenolobata. Acknowledgments This study was supported, in part, by Di- vision de Investigaci6n Sede Bogota (DIB), Universidad Nacional de Colombia. The au- thor greatly profited from the valuable dis- cussions with G. Rodriguez. I am very grate- ful to R. Lemaitre, who contributed valuable criticism to the manuscript, and facilitated access to material deposited in the National Museum of Natural History, Smithsonian In- stitution, Washington, D.C. Similarly, I would like to thank G. Bolivar for access to the Museo de Biologia Marina, Universidad del Valle, Cali (CRBMUV). I would also like to express my gratitude to H. Suarez for his valuable help during my visit to the In- stituto Venezolano de Investigaciones Cien- tificas, Caracas (IVIC), and to G. Stiles for VOLUME 116, NUMBER 3 improving the manuscript with valuable sug- gestions. The illustrations were prepared by J. C. Pinzon. Literature Cited Campos, M. R. 1985. Decapodos de agua dulce del Suborden Brachyura reportados para Colombia. Caldasia 14(67):265—284. . 1989. Nuevas especies de cangrejos de agua dulce del género Hypolobocera (Crustacea: De- capoda: Pseudothelphusidae) para Colombia.— Trianea 3:143—147. , & G. Rodriguez. 1984. New species of fresh- water crabs (Crustacea: Decapoda: Pseudothel- phusidae) from Colombia.—Proceedings of the Biological Society of Washington 97:538—543. ns . 1995. Two new species of fresh- water crabs of the genus Hypolobocera from Colombia (Crustacea: Decapoda: Pseudothel- phusidae).—Proceedings of the Biological So- ciety of Washington 108:649-—655. Coifmann, I. 1939. Potamonidi della Guiana Inglese raccolti dal Prof. Nello Beccari.—Archivio Zo- logico Italiano 27:93—116. Milne Edwards, H., & H. Lucas. 1842-1844. Crusta- cés. In A. d’Orbigny, Voyage dans |’ Amérique méridionale dans le cours des annés 1826-1833 6:1—39, Atlas 9: pl. 1-17. Ortmann, A. 1897. Carcinologische Studien.—Zoolo- gische Jahrbiicher, Abtheilung ftir Systematik, Geographie und Biologie der Tiere 10:258—372. Prahl, H von. 1983. Hypolobocera gorgonensis sp. nov. (Crustacea: Brachyura: Pseudothelphusi- dae) un nuevo cangrejo de agua dulce de la Isla de Gorgona, Colombia. Cespedesia XII(45/46): 105-110. . 1985. Distribucion del cangrejo de agua dulce Hypolobocera beieri Pretzmann, 1968 y anato- mia de su gonopodo.—Actualidades Bioldgicas 14:43—47. . 1987. Hypolobocera dentata sp. nov.: a new freshwater crab (Crustacea: Brachyura: Pseudo- thelphusidae) from the Cordillera Occidental, Colombia.—Revista de Biologia Tropical 35: 93-95. . 1988. Fresh-water crabs (Crustacea: Decapo- da: Pseudothelphusidae) of the Pacific Drainage of Colombia.—Zoologische Jahrbiicher fiir Sys- tematik 115:171—186. , & J. Giraldo. 1985. Un nuevo cangrejo de agua dulce de la Cordillera Central de Colom- bia.—Lozania 49:1—5. Pretzmann, G. 1965. Vorlaufiger Bericht tiber die Fam- ilie Pseudothelphusidae—Anzeiger der Oster- reichischen Akademie der Wissenschaften Mathematische Naturwissenschaftliche Klasse (1)1:1-10. 801 1968. Neue Siidamerikanische Siisswasser- krabben der Gattung Pseudothelphusa.—Ento- mologisches Nachrichtenblatt, Wien 15:1—15. . 1971. Fortschritte in der Klassifizierung der Pseudothelphusidae.—Anzeiger der Mathema- tisch Naturwissenschaftliche der Osterreichisch- en Akademie der Wissenschaften (1)179(1—4): 14-24. 1972. Die Pseudothelphusidae (Crustacea Brachyura).—Zoologica 42(120) pt. 1: 1-182. Rathbun, M. J. 1898. A contribution to a knowledge of the fresh-water crabs of America. The Pseu- dotheiphusidae.—Proceedings of the United States National Museum 21(1158):507—537. . 1905. Les crabes d’eau douce (Potamoni- dae).—Nouvelles Archives du d’ Histoire Naturelle, Paris 7:159—321. . 1915. New fresh-water crabs (Pseudothelphu- sa) from Colombia.—Proceedings of the Bio- logical Society of Washington 28:95—100. Rodriguez, G. 1966. The freshwater crabs of the genus Pseudothelphusa from northern Venezuela and Trinidad (Brachyura, Potamonidae).—Zoologis- che Mededelingen, Leiden 41(6):111—135, pl. 1-7. . 1967. New species of Pseudothelphusa from the Venezuelan Andes (Crustacea: Brachyura, Potamonidae).—Zoologische Mededelingen, Leiden 42(2):1—10, pl. 1-2. . 1980. Description préliminaire de quelques esp’eces et genres nouveaux de Crabes d’eau douce de l’Amérique tropicale (Crustacea: De- capoda: Pseudothelphusidae).—Bulletin du Mu- séum nationale d’ Histoire naturelle, Paris (4) 2 Section A (3):889-894. . 1982a. Les crabes d’eau douce d’ Amérique. Famille des Pseudothelphusidae.—Faune Trop- icale 22:1—223. . 1992. The freshwater crabs of America. Fam- ily Trichodactylidae and supplement to the fam- ily Pseudothelphusidae.—Faune Tropicale 31: 1-189. . 1994. A revision of the type material of some species of Hypolobocera and Ptychophallus (Crustacea: Decapoda: Pseudothelphusidae) in the National Museum of Natural History, Wash- ington D.C., with descriptions of a new species and a new subspecies.—Proceedings of the Bi- ological Society of Washington 107:296—307. , & R. von Sternberg. 1998. A revision of the freshwater crabs of the family Pseudothelphusi- dae (Decapoda: Brachyura) from Ecuador.— Proceedings of the Biological Society of Wash- ington 111:110—139. , M. R. Campos, & B. L6pez. 2002. New spe- cies and records of pseudothelphusid crabs Muséum (Crustacea, Brachyura) from Colombia in the 802 Tulane Natural History Museum.—Tulane Stud- ies in Zoology and Botany 31(2):1-17. Schmitt, W. L. 1969. Colombian freshwater crab notes.—Proceedings of the Biological Society of Washington 82:93-112. Smalley, A. 1964. A terminology for the gonopods of the American river crabs.—Systematic Zoology 13:28-31. Smithe, E B. 1975. Naturalist’s color guide. The Amer- ican Museum of Natural History, New York. Part 1: unnumbered pages. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Young, C. G. 1900. The stalk-eyed Crustacea of the British Guiana, West Indies and Bermuda. John M. Watkins, London, 514 pp. Zimmer, C. 1912. Beitrag zur kentniss der Siisswasser dekapoden Kolumbiens. Jn O. Fuhrmann et E. Mayor, Voyage d’exploration scientifique en Colombie. Mémoires de la Société neuchatelo- ise des Sciences naturelles 5:1—8. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):803—810. 2003. A new species of Xenorhina (Anura: Microhylidae) from western New Guinea Fred Kraus and Allen Allison Bernice P. Bishop Museum, Honolulu, Hawaii 96817, U.S.A., e-mail: (FK) fkraus@hawaii.edu, (AA) allison@hawaii.edu Abstract.—We describe a new species of fossorial frog, Xenorhina adisca, from the Sudirman Mountains of western New Guinea. The new species is distinguished from its congeners in lacking discs on fingers and toes, having moderately long legs, large eyes, and a long, pointed snout, and having a red venter with black “‘pants’”’ markings under the groin. It is currently known only from the type locality. The limited mobility of members of this genus make them highly susceptible to endemism, and it seems likely that additional species await discovery along the south versant of the Central Dividing Range of New Guinea. Xenorhina is a genus of seven endemic New Guinea species (Zweifel 1972, Blum & Menzies 1988, Allison & Kraus 2000) closely allied with the equally endemic Xe- nobatrachus, from which it differs in lack- ing odontoid spikes. Clear synapomorphies distinguishing these two genera are lacking (Burton 1986) and further phylogenetic study may result in the synonymization of the latter with the former. Of the seven spe- cies of Xenorhina currently recognized, six are terrestrial or fossorial, and one has sec- ondarily adopted an arboreal life style (Al- lison & Kraus 2000). All species are found in the western part of the island, ranging from western Papua New Guinea to the western end of Papua (formerly Irian Jaya). Most Xenorhina species occur at moderate to high altitudes (1000—3500 m), although X. oxycephala is a widespread lowland spe- CIES: During a biological survey in southwest- ern Papua in 1997 we discovered a new species of Xenorhina and describe it below. Materials and Methods Specimens were euthanized in the field by immersion in chlorotone, fixed in 10% buffered formalin, and then transferred to 70% ethanol for storage. All measurements were made to the nearest 0.1 mm with dig- ital calipers or an optical micrometer, ex- cept that disc widths were measured to the nearest 0.03 mm. We follow the method- ology and use the terminology of Zweifel (1972, 2000) and Kraus & Allison (2001): distance from anterior corner of eye to cen- ten Ofemnanisy (EIN) idiametenionyeye (EN): width of disc on third finger (FD); foot length from proximal edge of sole to tip of 4th toe (FT); hand length from proximal edge of palm to tip of 3rd finger (HD); head width at widest point, typically at the level of the tympana (HW); internarial distance, between centers of external nares (IN); dis- tance from anterior corner of eye to tip of snout (SN); body length from snout to vent (SV); width of disc on fourth toe (TD); tibia length from heel to skin fold of knee (TLy.ig); tibia length from heel to outer sur- face of flexed knee (TL,,...); tympanum di- ameter (TY). The two different measures of TL are provided because measurement technique varies in the literature: the latter measurement is generally the more reliable across a variety of microhylid genera and is the standard in more recent taxonomic 804 a iceele Scale bar is 2 mm. treatments, but the former measurement is published for a wider array of taxa (Zweifel 1972). Sex was determined by examination of gonads and vocal slits. We confirmed generic assignment of the frogs by the presence of a symphygnathine jaw, absence of vomeropalatine spikes, lack of a broad subdermal sheet of bone behind the eye, absence of cutaneous turbercles on the eyelids, and presence of a pointed snout and small eyes. Type specimens are depos- ited in the Museum Zoologi Bogor, Indo- nesia (MZB) and the Bernice P. Bishop Mu- seum, Honolulu (BPBM). Additional com- parative material is housed in the collec- tions of the University of Papua New Guinea (UPNG). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Right foot (a), and hand (b) of holotype (MZB 8403) of Xenorhina adisca, showing absence of discs. Xenorhina adisca, new species Figs. 1-3 Holotype.-—MZB 8403 (field no. AA 15395), adult female, collected by Allen Allison at Tembagapura, 4.14009°S, 137.09782°E, 2200 m elev., Sudirman Mountains, Papua (=Irian Jaya), Indonesia, on 17 March 1997. Paratypes.—MZB 8404 and BPBM 14915, juvenile and adult female, respec- tively, same data as holotype. Diagnosis.—A small (SV = 18.8—23.6 mm) species of Xenorhina lacking finger and toe discs (Fig. 1), and having moder- ately long legs (TL,.gSV = 0.35-—0.38, Thine/SV = 0.39-0.40), large eyes (EY/SV = 0.072—0.080), and a relatively long point- VOLUME 116, NUMBER 3 Bien 2: is 3 mm. ed snout with a high EN/IN ratio (1.1—1.2, ies 2): Description of holotype.—An adult fe- male. Head moderately wide (HW/SV = 0.32), merging with body with no constric- tion at neck; oblique loreal region, no can- thus rostralis; nostrils much closer to tip of snout than to eyes (Fig. 2); internarial dis- tance less than distance from external naris to eye (EN/IN = 1.1, IN/SV = 0.068, EN/ SV = 0.072); snout rounded when viewed from side and from above (Fig. 2); eyes of moderate size (EY/SV = 0.072), eyelid less than one-half width of interorbital distance; tympanic ring indistinct, horizontal diame- ter equal to width of eye. Dorsal and lateral surfaces with scattered low rounded tubercles, especially concen- trated on end of snout (Fig. 2). Supratym- panic fold slight. Ventral surfaces smooth. Fingers unwebbed, relative lengths 3 > 4 > 2 = 1, tips somewhat flattened but lacking discs and circummarginal grooves (Fig. 1); very low areas of thickened skin on inner metacarpal surfaces, but not de- veloped into actual tubercles. Toes unweb- bed> relative ‘lenoths 4° > 3 > 5 > 2 > 1; tips rounded and lacking discs and circum- marginal grooves; low skin thickenings pre- sent on inner metatarsal surfaces, but not developed into tubercles. Hind legs mod- 805 b Dorsal (a) and lateral (b) views of the head of holotype (MZB 8403) of Xenorhina adisca. Scale bar erately long (TL,,),/SV = 0.35, TL,,../SV = 0.40). The vomeropalatines lack enlarged odon- toid spikes. Ground color of dorsal surfaces of body and limbs brownish yellow in preservative, heavily and evenly suffused with dark brown, which is densest dorsally and lighter laterally. Obscure chocolate brown flecks are scattered on dorsum and sides. Tympa- num somewhat lighter in color and with an obscure chocolate brown supratympanic stripe that extends across the dorsal margin of the tympanum and ends just anterior to the forelimb insertion. A chocolate brown patch surrounds the anus. Ventral ground color of body and limbs brownish yellow with many tiny dark brown flecks scattered throughout and most densely concentrated on chin, throat, groin, and undersides of limbs, forming “‘pants”’ in the groin and up- per thighs (Fig. 3). Palms and soles mottled light gray and chocolate brown. Measurements (in mm).—SV = 23.5, Pe = oo, Lo — 9.5. AW — 7.6, EN = 1.6, EN = 1.7, SN = 2.4, EY = 1.7, TY = 1.7, HL = 4.9, FL = 10.1, FD = 0.58, TD = 0.50. Variation.—One of the paratypes is an adult female; the other is near adult size but of undetermined sex. Mensural variation in 806 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 7h be of bc thd, ges Seite ; WA at aX. SHA\Sar IRS ths , ae Fhe F Ate wef Fig. 3. Underside of holotype (MZB 8403) of Xenorhina adisca, showing strongly melanized gular region and “‘pants”’ in the groin and upper thighs. Scale bar is 10 mm. VOLUME 116, NUMBER 3 807 Xenorhina Collection Sites 45 O Xenorhina adisca 0) 50 100 km = Contour intervals are 0, 500, 1000 and 1500 m. Elevations above 1500 m are not demarcated. Fig. 4. Map of western New Guinea showing the type locality of Xenorhina adisca. the small sample at hand is minor (Table 1) and there are no noteworthy differences in color pattern among the three specimens. Color in life.-—Dorsum dark brown; ven- ter bright red. Call.—Unknown. All available speci- mens are female or immature. Comparisons to other species.—Xenorhi- na adisca may be distinguished from all other species of the genus except X. similis 808 Table 1.—Mensural measurements (in mm) and ra- tios of the holotype and two paratypes of Xenorhina adisca. Specimen Character MZB 8403 MZB 8404 BPBM 14915 Mean SV 235 18.8 23.6 Dig PL, 8.3 6.7 8.9 8 Tae 9.5 AD 9.5 8.8 EN hod! 153 1.8 1.6 IN 1.6 Ji 1.5 1.4 SN 2.4 1-9 Dinh Dio TY Jog! 1.5 1.5 1.6 EY Ite 7! AES 1.8 be HW 7.6 6.3 8.2 TA HL 4.9 4.1 5.5 4.8 FL 10.1 8.2 10 9.4 FD 0.58 0.48 0.53 0.53 TD 0.5 0.38 0.53 0.47 TE ip 0.35 0.36 0.38 0.36 TLV 0.4 0.39 0.4 0.4 EN/SV 0.072 0.069 0.076 0.073 IN/SV 0.068 0.064 0.064 0.065 SN/SV On 0.1 0.11 0.11 TY/SV 0.072 0.08 0.064 0.072 EY/SV 0.072 0.08 0.076 0.076 HW/SV 0.32 0.34 0.35 0.34 HL/SV 0.21 Q22 0.23 O22 FL/SV 0.43 0.44 0.42 0.43 FD/SV 0.025 0.026 0.022 0.024 TD/SV 0.021 0.02 0.022 0.021 EN/IN I 1.1 1.2 et and X. minima in lacking expanded discs on both the fingers and toes. From X. similis, X. adisca may be distin- guished by its considerably smaller size (SV = 18.8—23.6 in X. adisca vs. 48—51 mm in X. similis), larger eye (EY/SV = 0.072—0.080 in X. adisca vs. 0.058—0.072 in X. similis), and dark dorsal color pattern. From X. minima, X. adisca may be dis- tinguished by its longer legs (TL,,,, = 0.35-— 0.38 in X. adisca vs. 0.29—0.34 in X. min- ima), longer and more pointed snout (EN/ IN = 1.1—1.2 in X. adisca vs. 0.83—0.84 in X. minima), dark dorsal color pattern, and red (vs. cream) venter. Xenorhina adisca 1s further distinguished from all other members of the genus by its ventral color pattern, which includes a dark chin and throat and dark “‘pants” in the groin and upper thighs (Fig. 3). Other PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Xenorhina species having melanized gular regions are either entirely dark below (X. parkerorum) or are mottled in the groin/ thigh region (X. arboricola, X. similis). Ecological notes.—Xenorhina adisca is fossorial. We collected our specimens dur- ing the day from within the surface litter of a heavily mossed montane closed-canopy forest on a steep southwest-facing slope drained by a small stream. The dominant trees at the collection site included Elaeo- carpus nubigenus, Saurauia calyptrate and Polyosma integrifolia. Canopy height was 20-30 m. Treeferns, Cyathea sp., were common throughout the subcanopy, as was a species of Pittosporum. The ground flora included a dense array of tree saplings, gin- gers, Elatostema, ferns, and mosses (Daw- sonia and Sphagnum). Intermittent rainfall and temperature re- cords maintained by P.T. Freeport Indonesia suggest that the mean monthly temperature at the collection site is around 11—12°C with little annual variation. Annual rainfall likely exceeds 6000 mm. Xenorhina adisca occurred sympatrically with Xenobatrachus macrops, X. ocellatus, and an undescribed species of Xenorhina, all closely related frogs of similar habitus. Other members of the local herpetofauna included a treefrog, Litoria angiana, and a skink Papuascincus stanleyanus, both of which are widespread in montane New Guinea. Etymology.—The name is a feminine compound adjective derived from the Greek “a”, meaning ‘“‘without’’, and the Greek noun “‘disc’’, meaning “‘a flat plate’’, in its adjectival combining form ‘“-disca’’, and refers to the absence of digital discs in the species. Distribution.—Xenorhina adisca 1s known only from the type locality (Fig. 4) but is likely to prove more widely distrib- uted along the southern versant of the Cen- tral Dividing Range of western New Guin- ea. VOLUME 116, NUMBER 3 Remarks Thirteen of the 26 species now recog- nized in Xenorhina and Xenobatrachus have been described in the last quarter cen- tury (Menzies & Tyler 1977, Blum & Men- zies 1988, Allison & Kraus 2000, Kraus & Allison 2002, this paper) and we are aware of additional undescribed species. Most re- cently described taxa of Xenorhina and Xe- nobatrachus have been discovered in the central valleys of the Central Dividing Range or in the outlying north coast ranges. In contrast, Xenorhina adisca 1s the first member of these genera described from the southern versant of New Guinea since X. minima (Parker 1934). Considering the large tracts of western New Guinea that have not been adequately surveyed, more species seem likely to occur. Our surveys confirmed the presence of at least six spe- cies of Xenorhina and Xenobatrachus along an altitudinal transect ranging from sea lev- el to ca. 3000 m along the road from Timika to Tembagapura in southern Papua. We heard but were unable to collect what may be additional species. This appears to be the richest concentration of species yet reported for this clade; other sites have 3—4 species recorded (Blum & Menzies 1988, Allison & Kraus 2000, Kraus & Allison 2002). Blum & Menzies (1988) have remarked that the rugged terrain of New Guinea, combined with the limited mobility of these fossorial frogs, has likely led to high levels of speciation and endemism—a view with which we agree. However, the degree of en- demism varies in scale: several taxa may prove endemic to single valley or mountain complexes, but others inhabit broader, though geographically limited, regions en- compassing several distinct mountain rang- es. The taxa described from the Eipo Valley (Blum & Menzies 1988) may prove to fit the former pattern, whereas Xenorhina ar- boricola, Xenobatrachus zweifeli, and Xe- nobatrachus tumulus are already known to inhabit multiple discrete mountain ranges along the north coast of Papua New Guinea 809 (Allison & Kraus 2000, Kraus & Allison 2002). What pattern generally holds for species along the southern versant of the Central Dividing Range remains to be seen but most known species from that region have broad distributions (Zweifel 1972). Acknowledgments We thank A. A. Dwiyahreni, T. Purad- yatmika and A. K. Mandesy, G. Shea, T. Boivin and the staff of P.T. Freeport, partic- ularly H. Lewis and D. Martindale for all aspects of fieldwork. We also thank H. D. Cameron for advice on the etymological derivation of the trivial epithet, C. Kishin- ami for assistance with specimens and for critically reviewing the manuscript; R. Sin- gadan for specimen loans, and B. Evans for preparing Figure 4. Our field research was funded by P.T. Freeport via P.T. Hatfindo. Laboratory work was supported by Nation- al Science Foundation grant DEB 0103794. Literature Cited Allison, A., & E Kraus. 2000. A new species of frog of the genus Xenorhina (Anura: Microhylidae) from the North Coast Ranges of Papua New Guinea.—Herpetologica 56:285—294. Blum, J. P, & J. I. Menzies. 1988. Notes on Xenoba- trachus and Xenorhina (Amphibia: Microhyli- dae) from New Guinea with description of nine new species.—Alytes 7:125—163. Burton, T. C. 1986. A reassessment of the Papuan sub- family Asterophryinae (Anura: Microhyli- dae).—Records of the South Australian Muse- um 19:405—450. Kraus, F, & A. Allison. 2001. A review of the endemic New Guinea microhylid frog genus Choero- phryne.—Herpetologica 57:214—232. SG . 2002. A new species of Xenoba- trachus (Anura: Microhylidae) from northern Papua New Guinea.—Herpetologica 58:56—66. Menzies, J. I., & M. J. Tyler. 1977. The systematics and adaptations of some Papuan microhylid frogs which live underground.—Journal of Zo- ology, London 183:431—464. Parker, H. W. 1934. A monograph of the frogs of the family Microhylidae.—British Museum, Lon- don. Zweifel, R. G. 1972. Results of the Archbold expedi- tions. No. 97. A revision of the frog subfamily Asterophryinae family Microhylidae.—Bulletin 810 of the American Museum of Natural History 148:41 1-546. . 2000. Partition of the Austropapuan micro- hylid frog genus Sphenophryne with descrip- tions of new species.—Bulletin of the American Museum of Natural History 253:1—130. Appendix Specimens Examined Xenorhina arboricola: BPBM 13745, 1.3 km S, 2.3 km E summit of Mt. Hunstein, Hunstein Mts., 1000 m, East Sepik Prov., Papua New Guinea; BPBM 13746-77, S slopes of Mt. Menawa, 8.5 km N, 14 km PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON E Utai aerodrome, Bewani Mts., 1200 m, West Sepik Prov., Papua New Guinea. Xenorhina bouwensi: BPBM 1015, Sibil Valley, Star Mts., 1250 m, Irian Jaya [= Papua], Indonesia; BPBM 3686, 3.2 km N Dasiga, New Guinea. Xenorhina eiponis: UPNG 7406 (paratype), Mung- gona, 1800 m, Irian Jaya [= Papua], Indonesia. Xenorhina minima: UPNG 7409, Serabum, 2400 m, Irian Jaya [= Papua], Indonesia. Xenorhina oxycephala: BPBM 13756—57, 2 km W Utai aerodrome, 260 m, West Sepik Prov., Papua New Guinea; BPBM 14269, 8.7 km N, 9.8 km E Mt. Hun- stein, 75 m, East Sepik Prov., Papua New Guinea. Xenorhina parkerorum: UPNG 9358-59, Nogar Vil- lage, Keowagi, 2200 m, Simbu Prov., Papua New Guinea. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):811-819. 2003. Acanthochondria hoi, a new species of parasitic copepod (Poecilostomatoida: Chondracanthidae) on the California halibut, Paralichthys californicus, from Santa Monica Bay, California, with an amended key to the genus Acanthochondria Julianne E. Kalman Department of Organismic Biology, Ecology and Evolution, University of California, Los Angeles (UCLA), 621 Charles E. Young Drive South, Box 951606, Los Angeles, California 90095-1606 Abstract.—A new species of parasitic copepod, Acanthochondria hoi, is de- scribed from specimens collected within the gill cavity of the California halibut, Paralichthys californicus (Ayers), from Santa Monica Bay, California. Acan- thochondria hoi can be distinguished from its congeners by the combination of a Type B-V antennule and Type A leg 2, in addition to leg | ornamentation. A revision of the key of Acanthochondria prepared by previous authors is provided and includes three new species. Santa Monica Bay is located in the Southern California Bight and is an open coastal embayment bounded by Point Dume to the north and Palos Verdes Point to the south. The Hyperion Treatment Plant (Bureau of Sanitation, Department of Pub- lic Works, City of Los Angeles) provides secondary treatment and disposal of treated wastewater through a 5-mile effluent outfall located in Santa Monica Bay. The Environ- mental Monitoring Division conducts quar- terly otter trawls to monitor the effects of the effluent on the fishes and macroinver- tebrates living in the vicinity of the outfall (Dojiri & Brantley 1991). During the July/ August and November 1998, and February and May 1999 trawls, several specimens of California halibut, Paralichthys californi- cus (Ayers), were collected with parasitic copepods within the gill cavity. These par- asites represent a new species of Acantho- chondria, which is described below. Materials and methods.—The fishes were collected in Santa Monica Bay, Cali- fornia. Quarterly otter trawls were made aboard the R/V La Mer in association with the Environmental Monitoring Division, Bureau of Sanitation, Department of Public Works, City of Los Angeles. Immediately after the catch was brought on board, the fishes were placed in plastic bags and kept on ice in a cooler for a later examination in the laboratory. The copepods were removed and preserved in 70% isopropyl! alcohol, then cleared in 85% lactic acid. They were measured with an ocular micrometer and selected specimens were dissected. [lustra- tion were drawn with the aid of a camera lucida. Holotype and paratypes were depos- ited in the National Museum of Natural History, Smithsonian Institution, Washing- ton, D.C. (1001623—1001628). Additional specimens are in the collection of the au- thor. Systematic Account Order Poecilostomatoida Thorell, 1859 Family Chondracanthidae Milne-Edwards, 1840 Genus Acanthochondria Oakley, 1927 Acanthochondria hoi, new species Figs. 1-3 Material examined.—A total of seven nonovigerous and 22 ovigerous females 812 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Acanthochondria hoi n. sp., female. A, habitus, dorsal; B, habitus, lateral; C, genito-abdomen, lateral; D, caudal ramus; E, antennule; K antenna; G, mandible. Scale: 1.0 mm in A, B; 0.1 mm in C, F; 0.05 mm in D, G; 0.2 mm in E. VOLUME 116, NUMBER 3 (each with an attached male) was collected from within the gill cavity (inner side of operculum and floor of oral cavity at the base of gill arches) of the California hali- but, Paralichthys californicus (Ayers). Female.—The trunk (Fig. 1A, B) is long and slender. The cephalosome is slightly longer than wide. Neck region consisting of first and second pedigers. Trunk with a sin- gle mid-lateral indentation and bearing a pair of posterior processes, which are mod- erately long and slender. The genital seg- ment (Fig. 1C) is longer than wide; and the abdomen (Fig. IC) is shorter than the gen- ital segment and bears two dorsal setules. The caudal ramus (Fig. 1D) has three setae, a knob, and a large spinulated terminal pro- cess. The antennule (Fig. 1E) is of Type B- V (Ho & Kim 1995), consisting of a large unarmed basal portion and a small cylin- drical distal portion with an armature for- mula of 2-2-8. The antenna (Fig. IF) is 2- segmented; the basal segment is large and unarmed; the terminal claw possesses a small mid-lateral seta and a minute basal setule. The mandible (Fig. 1G) is 2-seg- mented; convex margin armed with 31—35 teeth, concave margin has 21—24 teeth. The maxillule (Fig. 2A) is a lobe bearing two processes produced at the distal margin of the appendage. The maxilla (Fig. 2B) is 2-segmented, with the basal segment un- armed. The terminal segment carries a small seta, a large seta, and a large process armed with 15 teeth along its posterior mar- gin. The maxilliped (Fig. 2C) is 3-seg- mented; the first segment is unarmed; the second segment bears a protruded portion on which one patch of spinules is located and a row of 12 teeth on the distal margin; the terminal segment is clawlike, bearing a proximal patch of small spinules and a sub- terminal accessory process. Leg 1 (Fig. 2D) is biramous with the exopod bearing an out- er seta. The anterior surface is covered with irregular patches of spinules. Leg 2 (Fig. 2E) is biramous with long rami. The exo- pod carries an outer seta. The rami bear spi- 813 nules at the distal tips of the anterior sur- faces in addition to several setules. Measurements.—Total length (tip of ce- phalosome to tip of posterior process) 6.84 mm; trunk width 0.44 mm; cephalosome 0.39 mm X 0.35 mm; genital segment 0.47 mm X 0.44 mm; abdomen 0.19 mm X 0.16 mm; posterior process 1.32 mm. Male.—The body (Fig. 3A), 0.67 mm X 0.34 mm, is ventrally flexed. The cephalo- some and the first pedigerous segment com- prise more than half the total length. The antennule (Fig. 3B) is slender and bears an armature formula of 1-1-2-2-8. The antenna (Fig. 3C) is 2-segmented. The basal seg- ment possesses a rounded knob near the ar- ticulation with the terminal claw. The ter- minal segment bears two setae on the basal portion. The mandible (Fig. 3D) is 2-seg- mented; terminal segment armed with 20 teeth along convex margin, 9 teeth along concave margin. The maxilla (Fig. 3E) is 2- segmented and exhibits the usual sexual di- morphism for this genus by possessing a naked terminal process. Leg 1 (Fig. 3F) is larger than leg 2 (Fig. 3G). Both legs are similarly armed, with the protopod bearing a long outer seta, the exopod with two small elements, and the endopod a smaller un- armed lobe. However, the two elements on the exopod of leg 2 are unequal in size. Etymology.—This species is named after Dr. Ju-Shey Ho, an expert in parasitic co- pepod research and my mentor and former advisor. Remarks.—Acanthochondria hoi, new species, was previously reported by Dojiri (1977) as Acanthochondria sp. C. However, a literature search revealed that a descrip- tion of this species was never published. Ho (1975) tentatively identified a badly dam- aged specimen of Acanthochondria from the California halibut as A. soleae (?). In addition, Haaker (1975) and Allen (1990) reported A. soleae to occur on the Califor- nia halibut, directly and indirectly citing Ho (1975), respectively. Ho’s (1975) specimen is probably identifiable with A. hoi. Kabata (1979) comments that the literature contains 814 ry 4 Hey, yuna y ‘Ue, 5h irae And AG 4 a “an Oe $ Sah Via 4 af “4 ie maT 1 Vy ERY AINE sees v wy pis NS PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Acanthochondria hoi n. sp., female. A, maxillule; B, maxilla; C, maxilliped; D, leg 1; E, leg 2. Scale: 0.05 mm in A, B, C; 0.3 mm in D, E. a number of erroneous host listings of A. soleae. A. soleae is parasitic mainly on the sole, Solea solea, and its distribution 1s re- stricted to the Atlantic Ocean (Kabata 1979). A. hoi differs from A. soleae in the relative lengths of the endopod and exopod, leg 1 ornamentation, and structure of the maxilliped. Acanthochondria hoi is distinguished from its congeners by the combination of a Type B-V antennule and Type A leg 2 (Ho & Kim 1995), in addition to leg 1 orna- mentation. Legs | and 2 have relatively long rami with the endopod noticeably lon- ger than the exopod. A check into the key of Acanthochondria prepared by Ho and Kim (1995) revealed that this specimen col- lected from the California halibut is new to science. This specimen keyed out to step 33a, which is equivalent to A. exilipes (Ho 1971). Table 1 lists differences between A. hoi and A. exilipes. Key to the Species of Acanthochondria The following revised key includes all accepted species of Acanthochondria (Ho & Kim 1995). Three new species were add- VOLUME 116, NUMBER 3 A Rigaes: 815 Acanthochondria hoi n. sp., male. A, habitus, lateral; B, antennule; C, antenna; D, mandible; E, maxilla: les i: (Glee 2-2 Scale: 0:1) mm in’ A; 0:02 mm im By; € DYE, EG: ed, A. kajika (Ho & Kim 1996), A. zebriae (Ho et al. 2000), and A. hoi (Kalman, this report), increasing the number of species to 46. In addition, all typographical errors have been corrected (most notable, from Ho and Kim (1995): step 16b should lead to step 29, not step 28 as previously noted; A. cyclopsetta, A. exilipes, A. galerita, and A. physidis should all be cited as Ho 1971, not Ho 1970 as previously noted). Poly and Mah (2001) deeply criticize some of the characters used in the key by Ho and Kim (1995). However, this revised key is still valid and useful until further morphological characters can be discovered for some nominal species where the host Table 1.—Differences between Acanthochondria exilipes and A. hoi. A. exilipes Antennule type B-Ul Teeth on mandible Maxillule Leg 1 ornamentation 2nd segment of maxilliped naked 37-41 on convex margin 32—34 on concave margin 2 patches of spinules 2 patches of spinules A. hoi B-V 31—35 on convex margin 21—24 on concave margin naked patches of spinules 1 patch of spinules and 12 teeth on outer margin 816 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON family is used as a “‘character’’. Thus, the publication containing the best information to aid in species identification is provided in parentheses after each species name. The males of Acanthochondria do not show species differences; therefore, the characters used in this key refer strictly to adult ovigerous females (Ho 1970). For types of antennule and leg 2 found in this key, refer to Ho and Kim (1995). la b 2a 3a 4a 5a 6a Neck region consisting of first pediger COLUSA. 0 Oe eres Le ee eres eS D; Neck region consisting of first and sec- ONGEDECIGEES) Fares a. oon Ghee Aco 4 Neck region consisting of second pe- diger only; first pediger incorporated UNO INSAVG! TROON 1 do05000000600 triglae (Herrera-Cubilla & Raibaut 1990:82—-87) Second pediger indistinguishably fused COM DIUTM Gt eye ed ah 8 oe ed 3 Second pediger distinctly separated from trunk and bearing a pair of large HOUMNGISG! SUEUMNNGS 50200000508 limandae (Kabata 1979:127—-128) Antenna of B-VII type laemonemae (Capart 1959:102—103) Antenna of B-III type lepidionis (Ho 1972a:147-149) Antenna of B-I type zebriae (Ho et al. 2000:711—713) Neck very long, at least 8 times longer than wide Neck moderately long, at most about 3 times as long as wide; leg 2 with ex- tremely long protopod (Type E) Neck short; at most slightly longer than wide; protopod of leg 2 not greatly elongated Posterolateral processes short and blunt; terminal process of maxilla bear- ing a short row of fine denticles ... diastema (Ho & Dojiri 1988:273—279) Posterolateral processes long and slen- der; terminal process of maxilla bear- ing a long row of large teeth .. uranoscopi (Ho & Kim 1995:48-—51) Endopod of leg 2 much reduced, rep- resented by a little knob tchangi (Shuno 1959:361) Endopod of leg 2 at least half as large as exopod © 0°) (8 «| e ©) « (ce 7a 8a 9a 10a lla wa ial Cephalosome about as long as wide; parasite of Platycephalidae .. platycephali (Ho 1973:127—130) Cephalosome distinctly longer than wide; parasite of fishes other than Pla- tycephalidae inimict (Dojiri & Ho 1988:47—53) Trunk cylindrical, long (at least twice longer than wide), and without lateral e © © © © © © © © © © © © © eo 8 indentations: 2. 72.2. 4... ee 9 Trunk appearance otherwise ....... 11 Posterolateral processes shorter than head; caudal ramus shorter than abdo- Tem 60.25 accu e. See ee 10 Posterolateral processes longer than head; caudal ramus distinctly longer thanvabdomen! 5 5-46.45) eee elongata (Pillai 1985:125—-127) Cephalosome with small rounded knob at each anterior corner; hook-like an- tenna curved in distal region .... fraseri (Ho 1972b:523—527) Cephalosome with large anterolateral swellings; hook-like antenna curved in basal region pingi (YU & Wu 1932:66—-68) Trunk trapezoidal; postoral region elongated dilatata (Shino 1955:107—110) Trunk shaped otherwise; leg | close to ©) je: (ese) fej, e! (e) to) (oe) (eo) -oMie! elmo) Kel Vellielrell le Of eo O80 SO OO DOfO OOO OF Do 0.0 Oral TESION: 24.48. «: eee WZ Cephalosome bearing a pair of lateral horn-like projections; trunk without laterallindentations meee bicornis (Shiino 1955:103—107) Cephalosome without such projections; trunk mostly with lateral indentations i cagiitel fogs «Seateel «tate aes eee aS Leg 2 long and slender (Type D)... 14 Bes? shaped othenwiSe an eae 15 Posterolateral processes long, as long as or longer than 4th pediger .. soleae (Kabata 1979:128—129) Posterolateral processes short, distinct- ly shorter than 4th pediger ... cyclopsetta (Ho 1971:3) Legs 1 and 2 indistinctly bilobated @iype dB) oie A oe eee physidis (Ho 1971:11—15) Legs 1 and 2 distinctly bilobated ... 16 Antennule without inflated basal part (Type A) VOLUME 116, NUMBER 3 25a 26a Antennule with inflated basal part GRP Es) IES SE en 30 Leg 2 slightly larger than leg 1 18 Leg 2 distinctly larger than leg 1... 24 Both legs 1 and 2 covered with spi- MMU S eemeeme itt om tee ah eee We ar a3 19 Both legs | and 2 without spinules or bearing at most only patches of spi- [NO ESS SA rat ot PUN RA SS Or le Oe ee 20 First pediger with lateral protuberance sixteni (Dojiri & Ho 1988:53-56) First pediger without such protuber- ance ... dojirii (Kabata 1984:1708—1910) Cephalosome distinctly longer (at least Le Sk SUTRSS)) METIS SSeS Loe 3 Sie Da Cephalosome about as long as wide .. 22 Distal part of leg rami covered with SUE Sy eee ether mn ee A vancouverensis (Kabata 1984:1710) Distal part of leg rami not covered with spinules glandiceps (Shiino 1955:93—96) Trunk about as long as wide 23 Trunk distinctly longer than wide . spirigera (Shiino 1955:100—103) iw) [ey ©) epnoctohsed toile! 0) 10) ©: 0,0] tee ye. 'o Soo OG CaO Rarasiic om SillasiniG@dac a5 4 shawi @quT935-7—9)) PakAsitiC Onn GODMGAC) = onan 2 yul (Shiino 1964:30—33) Terminal process of maxilla armed with a long row of teeth (about 15) .. Terminal process of maxilla armed with a short row of teeth (at most 11) DS 2 ©) © © © © « © « ©] ss © © (© © © © © © » « © © 9 6 «© © © «© Cephalosome round in dorsal view brevicorpa (Yamaguti 1939:535) Cephalosome pear-shaped in dorsal Re orig, Soe eS CT longifrons (Shiino 1955:86—89) Both legs bearing large, prominent patchesyolgspmules 7.44% se a: margolisi (Kabata 1984:1705) Both legs naked or with small patches of spinules Trunk distinctly longer than wide and with prominent lateral indentations .. . Trunk about as long as wide with slight lateral indentations us) Parasitic on Serranidae constricta (Shiino 1955:96—100) Parasitic on Pleuronectidae .. hippoglossi (Kabata 1987:215) Labrum with lateral protrusion, legs | ao) spe! felieite? efile 28 @) (0) \@) |e je) Ve! ey 6) 10) ay ee) 0) @ 9595 bso 4. 4 30a 3la 36a 817 and 2 tipped with spinules on both rami .. kajika (Ho & Kim 1996:276—279) Labrum without lateral protrusion, legs rand’? talked": ee ays es es fissicauda (Shino 1955:90—93) Cephalosome with two lateral round swellings on ventral surface of head; antennule with prominent ventral pro- tuberance clavata (Kabata 1979:126—127) Cephalosome and antennule without SUCH CALUINES 2 (es) ken oe nen Ney apee ene Both rami of leg 2 large, coniform (Type C); a pair of large protuberances lateral to labrum in oral area ... galerita (Ho 1971:8—-11) Leg 2 and oral area without such fea- tures Os Oo GeO OO CO OOOO 0-o Go OYOLD 32 38) reel lense, e) 0) (e) Ve) (ef e” 0) Ls) 0 je (se) ese ,enie (e © 1's) (ee ta Leg 2 distinctly larger than leg 1 ... Leg 2 only slightly larger than leg 1 Trunk wider than long; posterolateral processes short and blunt .... fasmaniae (Heegaard 1962:154—155) Trunk about as long as wide; postero- lateral processes long priacanthi (Ho & Kim 1995:53-56) Trunk distinctly longer than wide; pos- terolateral processes either long or Yoo oO Ako O O16) d Wate ements, Ai 34 Endopod of leg 2 about as long as pro- COPOU re eens are ayy eae cee 35) Endopod of leg 2 distinctly shorter fain PROLO POG = eee ee vacae Or eee 36 Antennule of B-III type; leg 1 naked exilipes (Ho 1971:3-7) Antennule of B-V type; leg 1 with patches of spinules joel eo: (0, -et.e) (oi(er ele, tele) lente: ne fy ree oer a eee hoi (Kalman this report) Terminal process of maxilla bearing less than 10 teeth So Oe So 5 Qs Oo 6 epachthes (Kabata 1968:339—344) Terminal process of maxilla armed with at least 15 teeth oralis (Yamaguti 1939:536—537) Trunk as long as wide or slightly lon- Sek thane wider Meee. See ee ees 38 Trunk distinctly longer than wide... 40 Cephalosome large, as wide as trunk and bearing a pair of anterolateral pro- tnberanGesig: Pah aee A Ae macrocephala (Ho & Kim 1995:46—48) b Cephalosome distinctly narrower than trunk, without protuberance ....... 3) 39a Cephalosome slightly longer than wide; endopod of leg 2 distinctly lon- cer thaniexOpod) Saha jae eee incisa (Shino 1955:83—86) b Cephalosome distinctly longer than wide; endopod of leg 2 about as long as exopod ... ophidii (Ho 1977:158—160) 40a Cephalosome about as long as wide; both legs covered with spinules i Jaane rectangularis (Kabata 1984:1705) b Cephalosome wider than long; both legs with spinules on rami only .... 41 41a Cephalosome with two prominent lat- eral protrusions; antennule of Type B- ye ey eae ae sicyasis (Ho 1977:160—164) b Cephalosome with swollen oral region; antennule of Type B-V ........ cornuta (Ho 1970:121—127) o00 00 Acknowledgements This paper represents a modified portion of a thesis completed at California State University, Long Beach (CSULB). I thank Dr. Ju-Shey Ho (CSULB) for his guidance; Dr. Masahiro Dojiri (Bureau of Sanitation, Department of Public Works, City of Los Angeles), Dr. Donald G. Buth (University of California, Los Angeles), and Dr. Alan Miller (CSULB) for their valuable sugges- tions; the Environmental Monitoring Divi- sion staff (City of Los Angeles) for their field assistance; Danny Tang for his input; and my parents, Marjatta and Marvin Kal- man, for their support. Literature Cited Allen, M. J. 1990. The biological environment of the California halibut, Paralichthys californicus. Pp. 7—29 in C. W. Haugen, ed., The California halibut, Paralichthys californicus, resource and fisheries. California Department of Fish and Game, Fish Bulletin 174. Capart, A. 1959. Copépodes parasites. Jn Results Ex- pedition oceanographique Belge dans les Ezux Cotieres Africines de 1’ Atlantique sud (1948— 1949).—Institut Royal des Sciences naturelles de Belgique 3(5):57—126. Dojiri, M. 1977. Copepod parasites of Orange County sewer outfall. M.A. thesis, California State Uni- versity, Long Beach, 295 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON , & J. S. Ho. 1988. Two species of Acantho- chondria (Copepoda: Poecilostomatoida) para- sitic on fishes of Japan.—Report of the Sado Marine Biological Station Niigata University 18:47—-S6. , & R.A. Brantley. 1991. Lepeophtheirus spa- tha, a new species of copepod (Siphonostoma- toida: Caligidae) parasitic on the California hal- ibut from Santa Monica Bay, California.—Pro- ceedings of the Biological Society of Washing- ton 104:727—735. Haaker, P. L. 1975. The biology of the California hal- ibut, Paralichthys californicus (Ayres), in An- aheim Bay, California. Pp. 137-151 in E. D. Lane and C. W. Hill, eds., The marine resources of Anaheim Bay. California Department of Fish and Game, Fish Bulletin 165. Heegaard, P. E. 1962. Parasitic Copepoda from Aus- tralian waters.—Records of Australian Museum 25(9):149—234. Herrera-Cubilla, A., & A. Raibaut. 1991. Acantho- chondria triglae n. sp., copépode parasite des fosses nasals de poisons Triglidae.—Crusta- ceana 59(1):82—88. Ho, J. S. 1970. Revision of the genera of the Chon- dracanthidae, a copepod family parasitic on ma- rine fishes.—Beaufortia 17(229):105—218. . 1971. Parasitic copepods of the family Chon- dracanthidae from fishes of Eastern North America.—Smithsonian Contributions to Zool- ogy 87:1-39. . 1972a. Copepods of the family Chondracan- thidae (Cyclopoida) parasitic on south African marine fishes.—Parasitology 65:147—158. . 1972b. Four new parasitic copepods of the Family Chondracanthidae from California in- shore fishes.—Proceedings of the Biological Society of Washington 85:523—540. . 1973. Chondracanthid copepods parasitic on platycephalid fishes of Australia, with discus- sion of known species occurring on flatheads.— Parasitology 67:123—-131. . 1975. Parasitic Crustacea. Pages 69-72 in E. D. Lane and C. W. Hill, eds., The marine re- sources of Anaheim Bay. California Department of Fish and Game, Fish Bulletin 165. . 1977. Parasitic copepods of the family Chon- dracanthidae from fishes of the south-eastern Pacific (Crustacea, Copepoda).—Steenstrupia 4(13):157—166. , & M. Dojiri. 1988. Copepods of the family Chondracanthidae parasitic on Australian ma- rine fishes.—Australian Journal of Zoology 36: 273-291. , & I. H. Kim. 1995. Acanthochondria (Copep- oda: Chondracanthidae) parasitic on fishes of Sado Island in the Sea of Japan, with a prelim- inary review of the genus.—Report of the Sado VOLUME 116, NUMBER 3 Marine Biological Station, Niigata University 25:45—-67. , I. H. Kim, & A. B. Kumar. 2000. Chondra- canthid copepods parasitic on flatfishes of Ker- ala, India.—Journal of Natural History 34:709— 185% Kabata, Z. 1968. Some Chondracanthidae (Copepoda) from fishes of British Columbia.—Journal Fish- eries Research Board of Canada 25(2):321—345. , 1979. Parasitic Copepoda of British Fishes. Ray Society, London, 468 pp. , 1984. A contribution to the knowledge of Chondracanthidae (Copepoda: Poecilostomatoi- da) parasitic on fishes of British Columbia.— Canadian Journal of Zoology 62:1703—-1713. , 1987. Acanthochondria hippoglossi sp. nov. (Copepoda, Chondracanthidae), a crustacean parasitic of some flatfishes off the Pacific coast of North America.—Canadian Journal of Zool- ogy 65:213-216. Pillai, N. K. 1985. The fauna of India—copepod par- asites of marine fishes.—Zoological Survey of India 900 pp. Poly, W. J., and C. L. Mah. 2001. New host distribu- tion records for parasitic copepods in the North- 819 east Pacific Ocean with a discussion of taxon- omy of the genus Acanthochondria.—Bulletin of Marine Science 69(3):1121—1127. Shiino, S. M. 1955. Copepods parasitic on Japanese fishes, 9. Family Chondracanthidae, subfamily Chondracanthinae.—Report of Faculty of Fish- eries, Prefectural University of Mie 2(1):70— iil . 1959. Sammlung der parasitischen Copepoden in der Prafekturuniversitat von Mie.—Report of Faculty of Fisheries, Prefectural University of Mie 3(2):334—-374. . 1964. On two new species of the genus Acan- thochondria Oakley (Crustacea Copepoda) found in Japan.—Zoologische Mededelingen 34:30-36. Yamaguti, S. 1939. Parasitic copepods from fishes of Japan. 6, Lernaeopodida I.—Volumen Jubilare Professor Sadao Yoshida 2:529—578. Yu, S. C. 1935. Studies on the parasitic copepods of China belonging to the family Chondracanthi- dae.—Bulletin of the Fan Memorial Institue of Biology 6(1):1—15. , & H. W. Wu. 1932. Parasitic copepods of the flatfishes from China.—Bulletin of the Fan Me- morial Institue of Biology 3(4):55—75. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):820—826. 2003. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 11. Documentation of an intergeneric woodstar hybrid, Calliphlox mitchellii X Chaetocercus mulsant Gary R. Graves Department of Systematic Biology, MRC-116, National Museum of Natural History, Smithsonian Institution, PO. Box 37012, Washington, D.C. 20013-7012, U.S.A. Abstract.—A specimen in the Museum Alexander Koenig collected in Co- lombia is shown to be a hybrid between Calliphlox mitchellii and Chaetocercus mulsant. This is the first known example of intergeneric hybridization between species in these presumably closely related genera. External measurements of the hybrids are consistent with the proposed parental hypothesis. As far as I am aware, the sole reference to intergeneric hybridization between min- iature woodstars currently placed in Chae- tocercus and Calliphlox (sensu Schuch- mann 1999) is buried in the published cat- alog of the Otto Kleinschmidt Collection, which is now deposited in the Museum A\]- exander Koenig, Bonn, Germany (Klein- schmidt 1943:226): ‘“‘Chaetocercus—? 9837 6 ohne Orig.-Etik., wohl CC, héchst interessanter Vogel, Zwischenglied oder Mischling bzw. Bastard zwischen Chaetocercus mulsanti und Calliphlox mitchelli. Schwerlich Ab- erration von mulsanti. C. harterti scheint ahnlich, ist aber kleiner.” Kleinschmidt’s brief entry was accompa- nied by an inked drawing of the rectrices of the specimen and those of its proposed parental species. This record was _ over- looked in subsequent catalogs of humming- bird hybrids (Meyer de Schauensee 1949, Gray 1958, Panov 1989, Schuchmann 1999). In any case, Kleinschmidt’s presen- tation was insufficient to determine the tax- onomic status of the specimen or to make a convincing case for a particular hybrid combination. Here I provide an assessment of the specimen employing the methods and assumptions outlined in Graves (1990) and Graves & Zusi (1990), as modified by the findings of Graves (1998, 1999). Methods The specimen (Museum Alexander Koe- nig 9837) was obtained by the Fass] broth- ers in Colombia, possibly from the Cordil- lera Central, but little else 1s known about its provenance except that Anton and Ed- uard Fass! collected natural history speci- mens in Colombia from 1908 to 1911. The specimen, which was sexed as d on one of the three attached labels (November 2001), appears to be in definitive plumage as judged by the absence of striations on the maxillary ramphotheca, the absence of dis- tinctive buffy feather tips on the dorsal plumage or white spots in the rectrices, and the presence of a strongly iridescent gorget. Descriptions in this paper refer to definitive male plumage. I compared the specimen with all species in the subfamily Trochilinae, the typical hummingbirds (Zusi & Bentz 1982, Sibley & Monroe 1990, Bleiweiss et al. 1997), de- posited in the Museum Alexander Koenig. Color photographs of the specimen were compared with all trochiline species in the National Museum of Natural History, Smithsonian Institution. The diminutive specimen was similar in size and general appearance to several of the small wood- stars. For assessing the possibility of hy- bridization, I considered all species cur- VOLUME 116, NUMBER 3 87] Table 1.—Ranges (mean + standard deviation) of measurements (mm) of wing chord, bill length, and rectrix length (R1I—RS5) of adult males of Calliphlox mitchellii, Chaetocercus amethystina, Chaetocercus mulsant, and a probable hybrid, Calliphlox mitchellii < Chaetocercus mulsant (Museum Alexander Koenig 9837). Calliphlox mitchellii Calliphlox amethystina* Chaetocercus mulsant Character (N = 13-15) (N = 10-11) (N = 18-19) Hybrid Wing 34.3—37.2 30.4—33.1 37.6—40.6 38.9 (DLO == 9) (S20) 28 O43) (B93) 22 OY) Bill 12.9-14.3 11.4—13.6 15.5-17.3 16.2 Bi z= O45) G@DFa ==) @iorss==70:4) R1 11.9-13.1 13.3-15.1 14.4—17.2 So (P25), 2211025) (14.2 + 0.6) Cs.) 26 O72) R2 16.0-17.5 16.1—17.4 17.3—20.4 18.8 GOe=3 05) (COS = OS) GIS ie==50)9)) R3 M3921) 20.8—25.0 22.8—27.8 28.3 @53-=s183) (2340) 22 Mes) Qaslez2ie3)) R4 30.0—34.3 28.0—-31.1 22.8—26.6 iy G2205== 185) O97 2 O2) (24.7 = 0.9) R5 29.9-33.1 32.2—36.9 19.3—22.2 DAES Gieo==7170) (33.8 += 1.4) (20.4 = 0.8) @ Bahia (n = 1), “‘Brazil’’ (n = 3), Minas Gerais (n = 3), Rio de Janeiro (n = 4). rently placed by Schuchmann (1999) in Calliphlox (mitchellii, amethystina) and Chaetocercus (mulsant, bombus, heliodor, astreans, jourdanii) that occur in Colombia (Hilty & Brown 1986) as potential parental species (see Graves 1997). Measurements were taken with digital calipers and rounded to the nearest 0.1 mm: wing chord; bill length (from anterior ex- tension of feathers); and rectrix length (from point of insertion of the central rec- trices to the tip of each rectrix) (Table 1). Pairs of rectrices are numbered from the in- nermost (R1) to the outermost (R5). Scatter plots of measurements were used to illus- trate size differences among specimens. General color descriptions presented in Appendix 1 were made under natural light. I evaluated crown color with a calibrated colorimeter (CR-221 Chroma Meter, Min- olta Corporation) equipped with a 3.0 mm aperture. The measuring head of the CR- 221 uses 45° circumferential illumination. Light from the pulsed xenon arc lamp is projected onto the specimen surface by op- tical fibers arranged in a circle around the measurement axis to provide diffuse, even lighting over the measuring area. Only light reflected perpendicular to the specimen sur- face is collected for color analysis. Colori- metric data from iridescent feathers are acutely dependent on the angle of measure- ment, the curvature of plumage surfaces in museum skins, and the degree of pressure applied to the plumage surface by the Chro- ma Meter aperture. In order to reduce mea- surement variation, I held the aperture flush with the rectrix surface without depressing it. The default setting for the CR-221 Chro- ma Meter displays mean values derived from three sequential, in situ measurements. Colorimetric characters were described in terms of opponent-color coordinates (L, a, b) (Hunter & Harold 1987). This system is based on the hypothesis that signals from the cone receptors in the human eye are coded by the brain as light-dark (L), red- green (a), and yellow-blue (b). The ratio- nale is that a color cannot be perceived as red and green or yellow and blue at the Sdinemcmes slhcrelore. » Tedness: sand ‘““sreenness”’ can be expressed as a single value a, which is coded as positive if the color is red and negative if the color is 822 green. Likewise, “‘yellowness” or “‘blue- ness”’ is expressed by b for yellows and —b for blues. The third coordinate, L, ranging from O to 100, describes the “lightness” of color; low values are dark, high values are light. The more light reflected from the plumage, the higher the L value will be. Vi- sual systems in hummingbirds (e.g., Gold- smith & Goldsmith 1979) differ signifi- cantly from those of humans and the rele- vance of opponent color coordinates to col- ors perceived by hummingbirds is unknown. In any case, the L, a, b color sys- tem permits plumage color to be unambig- uously characterized for taxonomic purpos- es. Results and Discussion I considered the hypotheses that the spec- imen represents (1) a geographic variant or genetic color morph of a valid species; (i) a hybrid; or (iii) an undescribed species. Hybrids lack formal standing in zoological nomenclature. Therefore, hybridity must be ruled out before species status is conferred. Because the specimen differed significantly in size and shape from all species of Cal- liphlox or Chaetocercus, it does not appear to represent a previously undiscovered col- or morph or geographic variant of a known species. All evidence is consistent with Kleinschmidt’s hypothesis that the speci- men represents an intergeneric hybrid, Cal- liphlox mitchellii * Chaetocercus mulsant. Several characters of the hybrid facilitate the identification of its parental species (Figs. 1, 2): (a) purplish-rose gorget with a few scattered white barbs on the chin; (b) white pectoral band; (c) broad white mid- line from pectoral band to vent; (d) absence of rufous or buff pigmentation on the rec- trices; (e) the presence of cinnamon or rufus tipping on lower flank feathers; and (f) the unusual shape of the tail (rectrix length: R4 SIRS) S IRS SS IRD S IRI). Here I present a synopsis of the critical steps of the hybrid diagnosis (see Appen- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON eel age (top to bottom): Chaetocercus mulsant, a probable hybrid, Calliphlox mitchellii * Chaetocercus mulsant (Museum Alexander Koenig 9837), and Calliphlox mitchellii. Lateral views of males in definitive plum- dix). The pool of potential parental species may first be narrowed by focusing on ab- sence of rufus or buff pigmentation in the rectrices in the hybrid. Because brown and reddish-brown pigments appear to exhibit consistent penetrance in hummingbird hy- brids (Banks & Johnson 1961, Graves & Newfield 1996), Chaetocercus bombus (buff pectoral band) and C. jourdanii (rufus pigmentation on rachii and vanes of rectri- ces) can be eliminated from further consid- eration as parental species. Chaetocercus astreans can also be eliminated because its geographic range in the Sierra Nevada de Santa Marta does not overlap that of other potential parental species (Graves 1986) and because it possesses a dark red gorget and bluish-green dorsal plumage, neither of which is expressed in the hybrid. Similarly, Chaetocercus heliodor can be eliminated because hybridization between it and any of the remaining species would likely produce offspring with a grayish-white pectoral band, dark green dorsal plumage toned with VOLUME 116, NUMBER 3 823 eres 2 Koenig 9837). bluish-green, and dark greenish flanks (Graves 1997). Based on plumage charac- ters, the hybrid could be the product of two possible combinations of species: Calli- Phlox mitchellii X Chaetocercus mulsant and Calliphlox amethystina X Chaetocer- cus mulsant. On geographical grounds the hybrid must represent the former combi- nation because Calliphlox amethystina and Calliphlox mitchellii have allopatric distri- butions east and west of the Cordillera Ori- ental (Hilty and Brown 1986). The hybrid was collected in the Cordillera Central (Kleinschmidt 1943). Colorimetric mea- surements of crown color (Table 2) are con- sistent with the hypothesis that the speci- men represents an intergeneric hybrid be- tween Calliphlox mitchellit and Chaetocer- cus mulsant. As a second step, I tested the restrictive hypothesis with an examination of size and external proportions (Fig. 3). Measurements of trochiline hybrids fall within the men- sural ranges exhibited by their parental spe- cies aS a consequence of the polygenic Ventral view of a probable hybrid, Calliphlox mitchellii X Chaetocercus mulsant (Museum Alexander mode of inheritance (Banks & Johnson 1961). Measurements of Calliphlox mitch- ellii and Chaetocercus mulsant overlap for only two of the seven characters but the percent difference in character means (larg- er species divided by smaller) is moderate except for the outermost rectrix (R5): wing chord (10.4%), bill length (18.9%), Rl (26.4%), R2 (14.4%), R3 (0.8%), R4 (29.6%), and R5 (54.9%). Measurements of the hybrid fall within the cumulative range of parental measurements for six of the sev- en measurements. The third rectrix (R3) of the hybrid was 0.5 mm (1.8%) longer than the largest value recorded for that character in the parental species. There have been no well-documented cases of morphological luxuriance (where the size of hybrid off- spring exceeds that of the parental species) among avian hybrids (Graves 1990, 1996). I suspect that the cumulative range of mea- surements for R3 in the parental species would overlap the hybrid value if the sam- ple size was increased. I note for comparative purposes that Cal- Table 2.—Minima, maxima, and means (+ standard deviation) of opponent color coordinates (L, a, b) of crown color of adult males of Calliphlox mitchellii, Chaetocercus mulsant, and a probable hybrid, Calliphlox mitchellii X Chaetocercus mulsant (Museum Alexander Koenig 9837). IL darkness Min. Max. Mean + SD Calliphlox mitchellii M09) OY SBVDOM 35 ST ae Oe Chaetocercus mulsant Boy 9 PAO MING) Orae a le7/ Hybrid W222) al—a b/—b red/green yellow/blue Min. Max. Mean + SD Min. Max. Mean + SD ONS eee le EOE tee 122 O28) 13 03.) 4:4 19 = 14 a 1G SG 22 LS 146) Tell PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 35 30 st Si $2 Ras fs x 28 5 30 8 3 g ¢ g 26 <= cS eS D 25 S ore o o 22 20 20 15m ZOke250 SON Sonr40 28% 32» , 36,2 40 744 28. 32 36° 40ne44 Length Rectrix 5 Wing Length Wing Length Fig. 3. Bivariate plots of selected measurements (see Table 1) of males in definitive plumage: Calliphlox mitchellii (@), Calliphlox amethystina (X), Chaetocercus mulsant (A), and a probable hybrid (+), Calliphlox mitchellii X Chaetocercus mulsant (Museum Alexander Koenig 9837). liphlox amethystina * Chaetocercus mul- sant was not excluded as a parental com- bination by mensural measurements (Table 1). However, inspection of bivariate plots of the wing and rectrix measurements suggest that Calliphlox mitchellii was the most like- ly Calliphlox parent (Fig. 3). In summary, evidence obtained from plumage color and pattern, as well as from external size and shape, is consistent with the hypothesis that Kleinschmidt’s specimen represents a hy- brid between Calliphlox mitchellii and Chaetocercus mulsant. Acknowledgments I thank Andre Weller, Renate van den El- zen, Karl Schuchmann, and Angela Schmitz for their hospitality in Bonn and for per- mission to examine the Kleinschmidt col- lection in the Museum Alexander Koenig. Richard C. Banks and Richard L. Zusi cri- tiqued the manuscript. Travel was support- ed by the Alexander Wetmore Fund, Smith- sonian Institution. Literature Cited Banks, R. C., & N. K. Johnson. 1961. A review of North American hybrid hummingbirds.—Con- dor 63:3—28. Bleiweiss, R., J. A. W. Kirsch, & J. C. Matheus. 1997. DNA hybridization evidence for the principal lineages of hummingbirds (Aves: Trochili- dae).—Molecular Biology and Evolution 14: 325-343. Goldsmith, T. H., & K. M. Goldsmith. 1979. Discrim- ination of colors by the black-chinned hum- mingbird, Archilochus alexandri.—Journal of Comparative Physiology A 130:209—220. Co- lombia.—Condor 92:754—760. Graves, G. R. 1986. Systematics of the Gorgeted Woodstars (Trochilidae: Acestrura). Proceed- ings of the Biological Society of Washington 99:218-224. . 1990. Systematics of the “‘green-throated sun- angels” (Aves: Trochilidae): valid taxa or hy- brids?—Proceedings of the Biological Society of Washington 103:6—25. . 1996. Hybrid wood warblers, Dendroica stri- ata X Dendroica castanea (Aves: Fringillidae: Tribe Parulini) and the diagnostic predictability of avian hybrid phenotypes.—Proceedings of the Biological Society of Washington 109:373-— 390. 1997. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 4. Hybrid origin of Calo- thorax decoratus Gould. Proceedings of the Bi- ological Society of Washington 110:320—325. 1998. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 6. An intergeneric hybrid, Aglaiocercus kingi X Metallura_ tyrianthina, from Venezuela.—Proceedings of the Biologi- cal Society of Washington 111:511—520. 1999. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 8. A provisional hypothesis for the hybrid origin of Zodalia glyceria (Gould, 1858).—Proceedings of the Biological Society of Washington 112:491—502. , & N. L. Newfield. 1996. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 1. Charac- terization of Calypte anna X Stellula calliope VOLUME 116, NUMBER 3 and the possible effects of egg volume on hy- bridization potential. Proceedings of the Biolog- ical Society of Washington 109:755-—763. , & R. L. Zusi. 1990. An intergeneric hybrid hummingbird (Heliodoxa leadbeateri < Helian- gelus amethysticollis) from northern Colom- bia.—Condor 92:754—760. Gray, A. P. 1958. Bird hybrids. Commonwealth Agri- cultural Bureaux, Bucks, England, 390 pp. Hilty, S. L., and W. L. Brown. 1986. A guide to the birds of Colombia. Princeton University Press, Princeton, New Jersey, 836 pp. Hunter, R. S., & R. W. Harold. 1987. The measurement of appearance, 2nd edition. Wiley, New York, 411 pp. Kleinschmidt, O. 1943. Katalog meiner ornitholo- gischen Sammlung. Gebauer-Schwetschke, Hal- le, Germany. I-XII + 236 + A-H pages. Meyer de Schauensee, R. M. 1949. Birds of the Re- public of Colombia, part 2.—Caldasia 5:381- 644. Panoy, E. N. 1989. Natural hybridisation and etholog- ical isolation in birds (in Russian). Nauka, Mos- cow, 510 pp. Schuchmann, K. L. 1999. Family Trochilidae. Pp. 468-680 in J. del Hoyo, A. Elliott, & J. Sar- gatal, eds., Handbook of the birds of the world, vol. 5. Barn-owls to hummingbirds. Lynx Edi- cions, Barcelona, 759 pp. Sibley, C. G., & B. L. Monroe, Jr. 1990. Distribution and taxonomy of birds of the world. Yale Uni- versity Press, New Haven, Connecticut, 1111 PP. Zusi, R. L., & G. D. Bentz. 1982. Variation of a muscle in hummingbirds and swifts and its systematic implications.—Proceedings of the Biological Society of Washington 95:412—420. Appendix Comparative description of selected characters of adult male Calliphlox mitchellii, Chaetocercus mul- sant, and a probable hybrid, C. mitchellii * C. mulsant (Museum Alexander Koenig 9837). Descriptions of structural colors are unusually subjective, as color seen by the observer varies according to the angle of in- spection and direction of light. For this reason I use general color descriptions. The dorsal plumage in mitchellii, from forecrown to uppertail coverts, is weakly iridescent and dark green in coloration. When viewed head-on, the dorsum is sooty green. The dorsal plumage of mulsant is signif- icantly more iridescent than in mitchellii, appearing bluish-green, particularly on the rump and uppertail coverts. When viewed head-on, the dorsum of mul/sant exhibits a moderate degree of green iridescence which becomes progressively bluer from crown to uppertail 825 coverts. The color and intensity of dorsal iridescence in the hybrid were intermediate in appearance to that of the parental species. Both parental species and the hybrid possess silky white tufts of flank feathers. In mitchellii the central rectrices (R1) are black, toned with dull bronze, the outer vanes glossed with dark green. The remainder of the rectrices (R2—R5) are black, toned with bronze or purplish-bronze at certain angles of inspection under intense light. In mulsant, the central rectrices are black, glossed with dull green on the proximal two-thirds of the inner and outer vanes; the outer rectrices (R2—R5) are black. Rectrices of the hybrid are intermediate in color and shape (width of R4 = ~1.3 mm, width of RS = ~1.1 mm; both measured 5 mm from the feather tip). In mulsant, dark green wing coverts contrast with dull black secondaries and primaries. The contrast be- tween wing coverts and flight feathers in mitchellii is negligible. Wing coverts and flight feathers in the hy- brid are intermediate in color between those of mulsant and mitchellii. In mitchellii, a white post-ocular spot elongates to form a tenuous connection to the side of the white pectoral band. The postocular facial stripe (obscured by lateral gorget feathers in Fig. 2) in mulsant and in the hybrid is more pronounced. The purplish-rose gorget of mitchellii extends lat- erally to the lower edge of the eyering and posteriorly to the lower throat. Gorget feathers are pale gray ba- sally (with a few dark gray lateral barbs near the base of the feather), banded distally by a narrow transitional band of dark gray, and tipped with a purplish-rose disk. The gorget is bordered posteriorly by a creamy-white pectoral band. The abdomen and sides are dull green. Feather tips along the midline are tipped with gray. Feathers of the lower flanks are tipped with cinnamon or rufus. Vent feathers are white, whereas undertail coverts are dull green narrowly fringed with gray or buff. The purplish-rose gorget of mulsant does not extend laterally to the eyering. A small white chin spot speck- led with iridescent disks is found in most specimens. The gorget is bordered posteriorly by a white pectoral band (whiter than in mitchellii). Gorget feathers are pale gray, grayish-white or creamy white, banded dis- tally by a narrow transitional band of dull bronzy- green, and tipped with a purplish-rose disk. A broad white stripe extends along the midline from the pec- toral band to the vent (also white). The flanks and sides of the breast below the pectoral band are bluish-green. Undertail coverts are white with an indistinct grayish or bluish-gray lanceolate subterminal spot. Lower flank feathers lack cinnamon or rufus tips. The ventral plumage of the hybrid is intermediate in appearance between that of mitchellii and mulsant. A few feathers on the chin have scattered white barbs. 826 The gorget extends laterally to the lower margin of the eyering as in mitchellii. The gray basal portions of gor- get feathers are intermediate in darkness between those of mitchellii and mulsant; the narrow subterminal band of gorget feathers exhibits dull bronze reflections. The remainder of the underparts are nearly intermediate in appearance between that of the parental species al- though the elongated preparation of the skin makes it difficult to compare homologous parts efficiently. The PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON hybrid possesses a distinctive pectoral band (darker than in mulsant) and a white midline stripe extending to the vent area (Fig. 2). Feathers on the lower flanks are tipped with cinnamon or rufus. The undertail co- verts are creamy white or buff with a subterminal gray- ish-green spot. Longer coverts are narrowly tipped with gray whereas shorter ones are tipped with buff or rufus. The maxillary and mandibular ramphotheca of mitchellii, mulsant, and the hybrid are black. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):827—838. 2003. A new burrowing crayfish of the genus Cambarus Erichson, 1846 (Decapoda: Cambaridae) from the lower Flint River basin in the Dougherty Plain of Georgia, with notes on C. (D.) harti Hobbs, 1981 John E. Cooper and Christopher E. Skelton (JEC) North Carolina State Museum of Natural Sciences, Research Lab, 4301 Reedy Creek Road, Raleigh, North Carolina 27607, U.S.A., e-mail: john.cooper@ncmail.net; (CES) Georgia Department of Natural Resources, Wildlife Resources Division, Nongame Wildlife & Natural Heritage Section, Georgia Natural Heritage Program, 2117 U.S. Highway 278 S.E., Social Circle, Georgia 30025-4714, U.S.A.; (Present address for CES) Department of Biological and Environmental Sciences, Georgia College & State University, CBX 081, Milledgeville, Georgia, 31061, U.S.A., e-mail: cskelton @ gcsu.edu Abstract.—Cambarus (Depressicambarus) doughertyensis 1s a new species of obligate burrowing crayfish known from a single locality in the lower Flint River basin in the Dougherty Plain of the East Gulf Coastal Plain, Dougherty County, Georgia. It appears to be most closely related to Cambarus (D.) harti Hobbs, which is known from two localities in the western Piedmont Plateau, Meriwether County. Although the two species are morphologically similar in many respects, C. (D.) doughertyensis differs from C. (D.) harti in having a longer areola; a plethora of tubercles on the carpus and ventral surface of the palm; more tubercles on the opposable surfaces of both fingers of the chela, and differences in the morphology of those surfaces; a lack of spines or tu- bercles on the proximal podomere of the uropod; a radically different color pattern; and in a number of other characters. Spines on the ventral keel of the rostrum of crayfishes other than certain Mexican crayfishes of the genus Pro- cambarus Ortmann are reported for the first time. Inadvertent errors in the description of C. (D.) harti are corrected. Five similar species of highly special- ized, obligate burrowing crayfishes of the genus Cambarus Erichson, 1846, subgenus Depressicambarus Hobbs, 1981, latimanus Group Bouchard 1978, are known to oc- cupy limited, widely separated ranges in Georgia (Hobbs 1981). Cambarus (D.) cy- matilis Hobbs, 1970, occurs in the Cona- sauga-Coosa River basin in the Ridge and Valley physiographic province; Cambarus (D.) harti Hobbs, 1981, is known from the Flint-Chattahoochee River basin in the western Piedmont Plateau; Cambarus (D.) reflexus Hobbs, 1981, occupies parts of the Savannah and Ogeechee River basins in the Atlantic Coastal Plain and eastern Fall Line Hills District; Cambarus (D.) strigosus Hobbs, 1981, is limited to the Savannah River basin in the eastern Piedmont Pla- teau; and Cambarus (D.) truncatus Hobbs, 1981, is known only from the Oconee River basin in the Fall Line Hills District. Of these five crayfishes, C. (D.) harti, which is cobalt blue, occupies the southwesternmost range, and is known only from wetlands at two localities in Meriwether County. Its type locality is within the Cold Spring Creek subdrainage of the upper Flint River, the other is within the Flat Shoal Creek sub- drainage of the Chattahoochee River, about 18.4 air km northwest of the type locality. On 20 May 1999, CES excavated a num- 828 ber of chimneyed burrows in a seasonally flooded swamp forest adjacent to the flood- plain of Kiokee Creek, a tributary of the lower Flint River in the Dougherty Plain of the East Gulf Coastal Plain in Dougherty County. This locality is about 147 air km south-southeast of the type locality of C. (D.) harti. The initial digging efforts pro- duced four specimens of a relatively drab, brownish crayfish that superficially resem- bled C. (D.) harti except in color pattern. Additional specimens have since been col- lected, and it is apparent that this animal is an undescribed species related to C. (D.) harti and the four other previously men- tioned species. Examination of these spec- imens also revealed the presence of spines on the ventral keel of the rostrum, a char- acter that has been reported only in some Mexican crayfishes of the genus Procam- barus, subgenera Paracambarus Ortmann, 1906, and Villalobosus Hobbs, 1972 (Vil- lalobos 1955, 1983). Confusing statements in the original description of C. (D.) harti are Clarified, based on a recent examination of the primary types. Measurements of crayfish structures were made to the nearest 0.1 mm with a Fowler precision dial caliper, following the meth- ods of Hobbs (1981:9-10) except where noted. Abbreviations used in the paper are: GMNH, University of Georgia Museum of Natural History, Athens; j, juvenile; NCSM, North Carolina State Museum of Natural Sciences, Raleigh; PCL, postorbital carapace length; TCL, total carapace length; USGS, United States Geological Survey; USNM, National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C.; UTM, Universal Transverse Mercator. Cambarus (Depressicambarus) doughertyensis, new species Fie. 1 fable i Diagnosis.—Body and eyes pigmented, eye small (x adult diam. 1.2 mm, n = 15). Rostrum with slightly thickened, elevated margins, subparallel or moderately con- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON verging to base of short acumen, which not delimited by tubercles or spines; margins constricted at base of acumen, strongly con- verging and concave from there to small, dorsally directed apical tubercle; acumen comprising 28.6—42.9% (x = 34.6%) of rostrum length, latter constituting 11.7— 13.3% (& = 12.5%) of TCL; floor (dorsal surface) of rostrum broadly concave, ce- phalic half cuplike; ventral keel of rostrum bearing 1—4 corneous spines. Areola oblit- erated or nearly so, length constituting 40.9—45.6% (x = 43.7%) of TCL and 46.2— 51.2% (X = 48.9%) of PCL. Thoracic sec- tion of carapace dorsally crowded with punctations, dorsolaterally and laterally granulate or with small tubercles; cephalic section laterally with many, usually small, tubercles. Cervical spines reduced to tuber- cles, one to several each side of carapace. Postorbital ridge moderate, cephalic margin rounded and without spine or tubercle. Sub- orbital angle obsolete to broadly obtuse, without tubercle or spine; branchiostegal spine vestigial or a small tubercle. Antennal scale 2.5—3.1 (x = 2.8) times as long as wide, greatest width at midlength; lateral margin thickened and terminating distally in long spine, lamella narrow, cephalic mar- gin moderately or strongly declivous, me- sial margin subparallel to lateral margin; antennal peduncle without tubercles or spines. Palm of chela inflated, 1.7—-1.9 (« = 1.8) times wider than deep, depth 89.5—98.6% (x = 95.2%) of length of mesial margin, width 1.6—1.9 (x = 1.7) times length of me- sial margin; latter constituting 26.3—31.3% (x = 28.9%) of total chela (propodus) length, bearing subserrate mesial row of 6 (rarely 5 or 7) large, semierect tubercles, row subtended dorsally by second row of 5 or 6 (rarely 3 or 4) smaller ones, and other smaller but produced tubercles dorsolateral to this row; | or 2 small, produced tubercles ventral to mesial row. Fixed finger of chela costate laterally, with strong median ridge dorsally and weak submedian ridge ven- trally; opposable surface of finger with row VOLUME 116, NUMBER 3 829 ° 10» ° ° 2 0: ° ° 2 o- ° ,e So mros Fig. 1. Cambarus (Depressicambarus) doughertyensis, new species; all from holotypic male, form I (NCSM 7997), except E, EK from morphotypic male, form II (NCSM 7999), and I from allotypic female (NCSM 7998): A, lateral aspect of carapace; B, E, mesial aspect of gonopod (first pleopod); C, E lateral aspect of gonopod; D, dorsal aspect of carapace; G, caudal aspect of in situ gonopods; H, epistome; I, annulus ventralis and postannular sclerite; J, antennal scale; K, ventral aspect of left palm; L, dorsal aspect of distal podomeres of left cheliped: M, basis and ischium of third pereiopod. Scale line = 2 mm. 830 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Measurements (mm) of types of Cambarus (Depressicambarus) doughertyensis, new species. Holotypic male, form I Carapace Total length 28.5 Postorbital length Sal Width 14.5 Depth 11.0 Length rostrum 3.4 Length acumen 2 Length areola WaT) Width areola 0.2 Antennal scale Length 32 Width lea Abdomen Length De Width 10.1 Cheliped Length propodus DA Length mesial margin palm 6.5 Width palm LoS) Depth palm 6.0 Length dactyl 14.6 Gonopod length 6.8 of 5—7 (rarely 8) tubercles, in addition to subconical tubercle ventral to denticles, third tubercle from base usually very large; dactyl length 2.2—2.6 (x = 2.3) times length of mesial margin of palm, with very strong median dorsal ridge and weak submedian ventral ridge; mesial surface of dactyl with strong, broad tubercles on proximal two- thirds or more; opposable surface with row of 9-12 (usually 9 or 10) tubercles, fourth from base usually very large. Carpus with weak dorsomesial tubercles; large, stout, conical spine at distal margin of mesial sur- face; 2 low, subconical distal mounds at distal margin of ventral surface; and mul- tiple strong tubercles on mesial and ventral surfaces; merus with 2—4 small subdistal dorsal tubercles, 4—6 (usually 5) tubercles on ventrolateral ridge, and 7—11 (usually 9 or 10) tubercles on ventromesial ridge. Abdomen shorter and considerably nar- rower than carapace; proximal podomere of uropod without tubercles or spines; mesial Morphotypic male, Allotypic female form II 29.0 29.4 D558 26.2 14.5 14.7 11.4 aD Be, 3.8 1.4 Wed WZe7 13.0 oblit oblit 32) 3.4 eS Il 26.4 26.0 10.7 10.3 Dl) 24.4 6.4 740 Lele 12.6 6.3 6.9 14.0 15.8 N/A Teal ramus with small caudolateral and caudo- median spines, latter submarginal. Hook on ischium of third pereiopod of males, that of holotypic male, form I (Fig. IM) uniramous, tapered, oblique, over- reaching basioischial articulation by most of length, opposed by very weak tubercle on basis; coxa of fourth pereiopod of males with vertically disposed, caudomesial boss. In situ gonopods (first pleopods) of form I male (based on holotypic male; Fig. 1G) symmetrical; proximomesial apophyses moderate, partly rounded; pair of bulbosi- ties on base below apophyses; central pro- jection directed caudally; mesial process in- flated at base, tip extruded, directed cau- dally and inclined slightly proximally; in lateral aspect (Fig. 1C), central projection curved over 90° to plane of shaft, slightly tapered, not recurved, with moderate, prox- imally directed subapical notch; tip extend- ing proximally nearly to distal margin of mesial process but not as far caudally as tip VOLUME 116, NUMBER 3 of latter; mesial process subglobose, distal margin arched; tip of process extruded, sub- acute, caudally directed; caudal process rep- resented by small, rounded protuberance; in mesial aspect (Fig. 1B), setae at midlength of shaft and some near proximocaudal mar- gin of caudal curvature. Annulus ventralis (based on allotypic fe- male; Fig. 11) about 1.5 times wider than long, subovate in ventral outline; cephalic half moderately depressed, with broad, un- even median trough flanked each side by strong ridge; cephalomedian margin domed, with median concavity; sinistral ridge of ce- phalic half descending obliquely to join up- per arm of reverse C-shaped caudosinistral wall, dextral ridge slightly curved, descend- ing to cephalic margin of caudodextral wall; caudosinistral wall inflated, rounded caudally and laterally, caudodextral wall long, nearly horizontal, less inflated but broad; transverse tongue originating at broad median end of caudodextral wall, proceeding horizontally to plunge into deep fossa beneath sinistral wall; sinus not dis- secting caudal margin. Mirror image of this configuration observed in 7 of 10 females. Measurements of type specimens provid- ed in Table 1. Description of holotypic male, form I,.— Body and eyes pigmented, eye 1.4 mm di- ameter. Cephalothorax (Fig. 1A, D) subcy- lindrical, thoracic section 1.3 times wider than deep. Areola nearly obliterated, 1 punctation in narrowest part, length consti- tuting 44.6% of TCL (49.4% of PCL). Ros- trum with relatively thick, strongly elevated margins slightly converging to base of short acumen, where sharply constricted, more convergent and concave to small, dorsally directed apical tubercle; latter reaching midlength of penultimate podomere of an- tennular peduncle; acumen comprising 35.3% of rostrum length, latter constituting 11.9% of TCL; floor of rostrum excavate, cephalically cuplike, sparsely punctate, with usual row of deep punctations along inner surface of dorsal ridge; subrostral ridge strong, visible to base of acumen in 831 dorsal aspect; ventral keel of rostrum with 2 corneous spines. Postorbital ridge fairly strong, poorly de- fined ventrally, inflated caudally; groove shallow, lateral; cephalic margin rounded. Suborbital angle obsolete, margin without tubercle or spine; branchiostegal spine re- duced to small tubercle. Thoracic section of carapace dorsally crowded with deep punc- tations, dorsolaterally and laterally granu- late; row of small tubercles along ventral margin of anterior section of cervical groove; cephalic section of carapace 1.2 times longer than areola, constituting 55.5% of TCL; laterally with many scattered, small to moderate, tubercles, gastric region with scattered punctations. Cervical spine region with | tubercle and several granules. Antennal peduncle devoid of spines or tubercles; tip of adpressed antennal flagel- lum reaching caudal margin of third tergite; antennular peduncle with very small, sub- distal median spine on ventral surface of basal podomere. Antennal scale (Fig. 1J) 2.9 times as long as wide, broadest near midlength; lateral margin slightly convex, thickened, terminating in long spine, tip of which reaching distal margin of penultimate podomere of antennular peduncle; lamella 1.2 times width of thickened lateral portion, distal margin strongly declivous from base of spine, mesial margin subparallel to lat- eral. Abdomen narrow, greatest width 69.7% of greatest carapace width; length 89.5% of TCL; abdominal pleura fairly short, round- ed ventrally and caudally; terga with many large punctations, except articular surfaces glabrous. Proximal podomere of uropod without spines or tubercles; mesial ramus of uropod with weak median ridge bearing very small, submarginal caudomedian spine; caudolateral spine of ramus very small; cephalic section of right lateral ra- mus of uropod with weak ridge, caudal margin of transverse flexure bearing total of 19 fixed spines and 1 long, articulated sub- lateral spine (15 fixed and | sublateral on left). Telson with 2 spines in right caudo- 832 lateral corner of cephalic section, innermost articulated, left corner with single fixed and 2 articulated spines; transverse flexure strong, caudal margin broadly rounded. Epistome (Fig. 1H) with subcordiform to subtriangular cephalic lobe bearing short cephalomedian projection; margins of lobe relatively narrow, moderately elevated, lat- eral corners produced as tuberclelike pro- tuberances; floor (ventral surface) broadly convex, very punctate; central depression shallow, with deep median fovea; lamellae very broad, moderately punctate, laterally truncate, with | small tubercle and | strong caudal tubercle each side; zygoma well arched, cephalolateral margins flanked by deep pit. Third maxilliped with tip reaching mid- length of penultimate podomere of antennal peduncle; exopodite very hirsute, tip reach- ing nearly to midlength of merus of endop- odite; distolateral corner of ischium sub- acute, lateral half apunctate, glabrous; me- sial section very broad, with clumps of long bristles, mesial margin of right ischium with 21 denticles, 20 on left; incisor ridge of right mandible with 8 denticles, 7 on left. Right cheliped regenerated; left chela (Fig. 1L) 1.9 times longer than wide, palm 1.9 times as wide as deep, width 1.8 times length of mesial margin; latter 29.4% of to- tal chela (propodus) length, 44.5% of dactyl length. Dorsal surface of palm covered with mostly deep punctations, those on mesial half and some on lateral half with small basal tubercles; longitudinal dorsomesial sulcus weak, with row of small, produced tubercles; dorsolateral margin of palm cos- tate for much of length (continuous onto fixed finger), with slight impression and large punctations, some with tubercles at proximal bases; articular ridge strong, lat- eral eminence with dense, short setae on distolateral margin, continuing onto base of opposable surface of fixed finger; lateral margin of palm with row of large puncta- tions. Ventral surface of palm (Fig. 1K) very punctate, mesial portion sharply set off from greatly inflated portion; distolaterally PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON with weak depression (continuous onto fixed finger); lateral eminence of strong ar- ticular ridge with large, depressed, subcon- ical subdistal tubercle; 3 strong and 3 weak- er tubercles proximal to ridge (total of 13 obvious, produced tubercles on ventral sur- face). Mesial margin of palm with subser- rate mesial row of 6 strong tubercles, row subtended dorsolaterally by row of 6 gen- erally smaller tubercles, and 4 others dorsal to this row; | strong distal tubercle ventral to mesial row. Fingers gaping except at tips, greatest width of gape slightly more than width of dactyl base, largest tubercles on opposable surfaces of fingers separated when fingers closed. Fixed finger costate laterally, inner margin of proximal fourth scalloped; dorsal surface with very strong, glabrous median ridge, flanked each side by deep punctate groove; lateral surface with row of large punctations; ventral surface with fairly strong ridge, closer to opposable margin than median in position, flanked mesially by row of coalescing punctations, laterally by slanted surface crowded with large punctations; opposable surface with rela- tively small subconical tubercle ventral to denticles at base of distal third or so of fin- ger, and row of 5 other tubercles dorsal to or interrupting denticles, second from base largest; denticles in | or 2 rows. Dorsal sur- face of dactyl with very strong, glabrous median ridge, flanked each side by deep, punctate groove; mesial surface with very strong, broad tubercles on proximal three- fourths, basal tubercles in 3 rows, strongly encroaching dorsally, weakly encroaching ventrally; ventral surface with moderate, glabrous ridge, flanked each side by row of punctations; opposable surface with 9 tu- bercles, third from base largest, and exci- sion in surface proximal to this tubercle; denticles in | row. Carpus (Fig. 1L) 1.4 times as long as wide, 1.4 times as long as mesial margin of palm; dorsal surface of carpus with narrow, deep sulcus, lateral to which surface punc- tate, mesial to which surface with 3 dor- VOLUME 116, NUMBER 3 somesial tubercles; mesial margin with short, stout distal spine, 1 moderate tubercle close to its proximal margin, and 9 other strong, often subacute, tubercles; ventral surface with 2 subconical distal mounds, and 1 strong and 1 smaller proximomesial tubercle. Merus 1.7 times longer than deep, distodorsal surface with 1 large, broad ad- pressed tubercle and 1 smaller tubercle; ventrolateral ridge with 6 small, rounded tu- bercles, none on articular condyl, which rounded, glabrous; ventromesial ridge with 8 subacute tubercles and short, stout distal spine; ischium with 4 small, subacute tu- bercles on ventral ridge. See “Diagnosis”’ for description of gon- opods. Description of allotypic female.—Except for secondary sexual characters, differing from holotypic male in following respects: Areola obliterated, constituting 43.8% of TCL (49.2% of PCL). Acumen comprising 37.8% of rostrum length, latter constituting 12.8% of TCL. Postorbital ridge fairly weak, nowhere sharply defined. Cephalic section of carapace 1.3 times longer than areola, constituting 56.2% of TCL. Cervical spine region with 5 small tubercles and sev- eral granules. Greatest width of abdomen 73.8% of greatest carapace width, length 91.0% of TCL. Telson with single fixed spine in each caudolateral corner of ce- phalic section, transverse flexure weak, cau- dal margin domed. Palm of chela 1.8 times broader than deep, width 45.7% of length of mesial mar- gin, latter 29.8% of total chela length, 45.7% of dactyl length; dorsomesial sulcus of palm obliterated; ventral surface with to- tal of 10 obvious, produced tubercles. Op- posable surface of right fixed finger with broken subconical tubercle ventral to den- ticles at base of distal fourth of finger, row of 8 other tubercles (6 on left finger), third from base massive. Opposable surface of right dactyl with 10 tubercles, fourth from base massive. Carpus 1.3 times as long as wide, mesial surface with 10 tubercles of varying sizes and shapes in addition to 833 short, stout distal spine. Right merus with 1 moderate and 4 weak, rounded distodorsal tubercles (3 moderate tubercles on left); ventromesial ridge with 10 tubercles (9 on left), in addition to distal spine; ischium with 2 small ventral tubercles each limb. First pleopods strong, distally hirsute; preannular sternite relatively narrow, walls steep (see ““Affinities’’); ventral surface of postannular sclerite (Fig. 11) with large, subconical papilla. See “‘Diagnosis’”’ for description of an- nulus ventralis. Description of morphotypic male, form /7.—Differing from holotype in following respects: Areola obliterated, constituting 44.2% of TCL (49.6% of PCL). Ventral keel of rostrum with single spine near base. Branchiostegal spine obsolete. Cephalic section of carapace 1.3 times longer than areola, constituting 55.8% of TCL. Greatest width of abdomen 69.7% of greatest width of carapace, abdomen length 88.4% of TCL. Telson with 1 very small fixed spine and | minuscule articulated spine in right caudolateral corner of cephalic section, all spines congenitally absent from left. Lateral corner of lamella of epistome devoid of tu- bercles; zygoma moderately arched. Right cheliped regenerated; palm of left chela 1.8 times broader than deep; length of mesial margin 28.7% of total chela length, 44.3% of dactyl length; ventral surface of palm covered with tubercles of various siz- es; subconical tubercle on lateral eminence of articular ridge originating at proximal margin of ridge, not adpressed; 3 large tu- bercles proximal to ridge. Largest tubercle on opposable surface of fixed finger second from base, largest on opposable surface of dactyl fourth from base. Carpus length 1.5 times length of mesial margin of palm; dor- sal sulcus short, flanked mesially by row of 3 tubercles; mesial surface with total of 11 tubercles in addition to usual stout subdistal spine; ventral surface with single proxi- momesial tubercle. Distodorsal surface of left merus with | moderate, subconical and 2 smaller, rounded tubercles; ventrolateral 834 ridge with 5 very small tubercles, none on distal condyl; ventromesial ridge with 11 small, conical tubercles. Hook on ischium of third pereiopod moderate, not overreaching articulation, op- posed by extremely weak tubercle on ven- tral surface of basis. In situ gonopods with moderate, nearly abutted proximomesial apophyses; in lateral aspect (Fig. 1F), cen- tral projection broad, slightly tapered, not reaching as far caudally as tip of mesial process; latter inflated nearly entire length, directed caudally, tip subtruncate and with small, spiniform protuberance directed cau- domesially; juvenile suture absent; in me- sial aspect (Fig. 1E), short setae at mid- length of shaft. Color notes.—Most specimens discol- ored by ferrous deposits; the following based on a live adult female. Ground color of carapace brown, fading ventrally to pale greenish-brown, ventral margin orangish; cephaloventral area of lat- eral surface of cephalic section of carapace creamy orange. Coxa, basis, and ischium of cheliped orangish; merus brownish, fading to orangish ventrally and proximally; car- pus brown dorsally, orangish ventrally, with orangish dorsomedian furrow. Large distal tubercle on mesial surface of carpus orange, all others orange-cream. Dorsal surface of chela greenish-brown, articular ridge or- ange; tips of both fingers orange, not sub- tended by black band; tubercles on mesial margin of palm and dactyl orangish; lateral surface of propodus orangish. Largest tu- bercle on opposable surface of dactyl and fixed finger cream; other tubercles on both fingers orange-cream. Proximodorsal part and merus of second through fifth pereio- pods orangish, fading to brownish on basis and ischium; ventral surfaces of these pe- reiopods orangish. Dorsal surface of abdo- men brown; cephalic part of all pleura pink- ish-orange, color most obvious on cephalic two-thirds of second abdominal segment; uropods and telson light brown. Type _ locality.—Georgia, Dougherty County, burrows in wetland just south of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON access road near western boundary Albany Nursery Wildlife Management Area, ca. 12.8 km W of Albany (Pretoria 7.5” USGS quadrangle, UTM Zone 16, coord. 751779E, 3496706N). Locality within sub- drainage of Kiokee Creek, Flint River ba- sin, Apalachicola River drainage. Disposition of types.—The holotypic male, form I, allotypic female, and morpho- typic male, form II, are in the NCSM crus- tacean collection (7997, 7998, and 7999, re- spectively). The following are designated paratopotypes: 1h jos 3 2253) ae2 iu 1999, coll. CES, S. E. Cammack, E. E. Van De Genachte; 4°29 25) 2) (@ 987) e2aekel 2001, coll. CES; 1 ¢ 1 (7988), 13 Aug 1999 (was ¢o II when collected), coll. CES, S. K. Berckman; | do I, 1 2 (USNM 1004610), 2 Jun 1999, coll. CES, M. C. Freeman (MCF); | 2 (GMNH 6239), 2 Jun 1999, coll. CES, MCF; 1 2 (23692); 14 May 2002, coll. CES. Variations.—Some variations other than those previously addressed have been ob- served. In dorsal aspect, the margins of the rostrum are usually rather straight, and sub- parallel or only moderately convergent to the base of the acumen. In several speci- mens, though, the margins are slightly con- cave between the orbits, and in two very small females the rostrum is_ triangular. These juveniles also have marginal spines on the rostrum. In all specimens, the mesial (lamellar) portion of the antennal scale is only slightly wider than the thickened lat- eral portion, and the distomesial margin is moderately or strongly declivous from the base of the spine to the widest point. In some specimens the distomesial margin is deeply incised, bearing decidedly spinelike protuberances. In ventral outline, the ce- phalic lobe of the epistome varies in shape from subcordiform or subtriangular to sub- pentagonal, and in some specimens the margins are erose, with small angular pro- trusions. The central depression of the epi- stome is obsolete in 10 specimens, very shallow in the others, and always displays a deep fovea. VOLUME 116, NUMBER 3 The greatest width of the abdomen varies from 69.1—74.8% (* = 72.2%) of carapace width, and is only slightly wider in females than in males. The length of the abdomen varies from 86.5—95.2% (x = 90.3%) of TCL. The number of spines in the caudo- lateral corner of the cephalic section of the telson is highly variable. Four specimens have a single fixed lateral spine and a single articulated inner spine in each corner. Four have two spines in one corner, and either a single fixed or articulated spine in the other. One specimen has no spines in one corner, two in the other; another has two spines in one corner, three in the other; and two spec- imens have a single fixed spine in each cor- ner. In most specimens the articulated spines are very small. In most specimens the punctations on at least the mesial half or third of the ventral surface of the palm have small basal tuber- cles, but the surface also bears from 6—13 prominent, produced tubercles, several of which may be very large. The largest tu- bercle on the opposable surface of the fixed finger is almost always the third (rarely the second or fourth) from the base; the largest tubercle on the corresponding surface of the dactyl is usually the fourth from the base, but often is the third, and occasionally the first and fourth tubercles are equally large. The total number of tubercles or spines on the carpus, excluding the stout distal spine of the mesial surface and the two stout dis- tal tubercles or mounds of the ventral sur- face, ranges from 7—13 (usually 10-13). The ventral surface of the ischium of the cheliped bears from 2—4 (usually 3 or 4) small, subacute tubercles. The total length of the chela of adult males averages about 85% of TCL (94% of PCL), while the av- erage for adult females is about 75% of TCL (84% of PCL). Size.—The largest specimen is a female measuring 34.5 mm TCL (31.0 mm PCL). The largest form I male measures 31.0 mm TCL (27.5 mm PCL), the smallest 28.5 mm TCL (25.7 mm PCL). The smallest speci- 835 mens yet collected are two females with TCLs of 7.0 and 7.7 mm. Life history notes.—Form I males were collected in May and July. A form II male, collected in August 1999, underwent two molts in the laboratory, became form I in January 2000. No ovigerous females or those with attached young have yet been found. The two very small females, possi- bly representative of recruitment size, were dug from the same burrow on 25 February 2001. Ecological notes.—The type locality of C. (D.) doughertyensis is in a swamp forest dominated by an overstory of Quercus spp.., Nissa biflora, and Acer rubrum, while the understory has abundant Crataegus aesti- valis, Sebastiana fruticosa, Rhus radicans, and a variety of graminoids. The area is seasonally flooded, and in very wet periods produces a shallow outflow to Kiokee Creek. Burrows are typically found just above the standing water mark among roots of small trees and Serenoa repens. The soils in the area are Grady clay loams and pro- vide easy digging for the first 25 cm. Below that, the soil becomes much more dense and is more difficult to excavate. When the wa- ter table is high, burrows are anastomosed and usually have four or five openings, of- ten marked by well developed chimneys ap- proximately 10 cm high and 15 cm across. In this situation, the animals are most often found in one of the horizontal passages about 30 cm below the surface. As the wa- ter table drops, burrows are marked by only two or three openings, with chimneys, that angle down to a single subvertical passage that penetrates the water table. The animals are then found in the vertical passage, 5 to 10 cm below the water level. When a bur- row is excavated, the crayfish can rarely be induced to come to the air-water interface, and when encountered they are unmoving and rarely try to escape. Crayfish associates.—Faxonella clypea- ta (Hay, 1899) was collected in open water in the swamp forest when flooded, and was dug from simple burrows when the water 836 table dropped. Two female Cambarus (De- pressicambarus) cf. C. (D.) striatus Hay, 1902 (5157, 23691) and a female Camba- rus (Lacunicambarus) diogenes Girard, 1852 (23690) were dug from separate bur- rows in the floodplain of Little Kiokee Creek, not far from the C. (D.) dougher- tyensis site. Affinities.—Cambarus (D.) doughertyen- sis Shares so many similarities with C. (D.) harti and several of the other small Georgia burrowers that, although some of them can undoubtedly be attributed to convergence and the channeling effect of adaptation to an obligate burrowing existence, it seems reasonable these species have descended from a common proximate ancestor. The annuli ventrales of some of the species are very similar, as are their narrow abdomens, many aspects of their chelipeds, rostra, an- tennal scales, suborbital angles, and bran- chiostegal spines, and other features. Some, including C. (D.) harti, also have spines on the ventral keel of the rostrum. They differ significantly, however, in many ways. In C. (D.) doughertyensis the areola con- stitutes 40.9—45.6% (x = 43.7%) of TCL and 46.2—51.2% (x = 48.9%) of PCL, while in C. (D.) harti it constitutes 38.2—40.3% (x = 39.4%) of TCL and 44.2—45.8% (x = 44.8%) of PCL. Correspondingly, the ce- phalic section of the carapace in the former species is 1.2—1.4 times as long as the are- ola, while in the latter it is 1.5—1.6 times as long. The proximal podomere of the uropod in C. (D.) harti usually bears a small spine on both the lateral and mesial lobes, but such spines are lacking in C. (D.) dough- ertyensis. In C. (D.) harti, the preannular sternite of the female is broader and flatter than that of C. (D.) doughertyensis, which is narrow and steep. The width of this ster- nite, 1.e., the distance between the ventral angles of the walls at their articulations with the mesial surfaces of the fourth cox- ae, 1s about 38% of carapace width in the former species, about 31% in the latter. The ventral surface of the palm of the cheliped of C. (D.) doughertyensis (Fig. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1K), proximal to the articular ridge, bears many prominent, produced tubercles, whereas only a few such tubercles are pre- sent in C. (D.) harti. The opposable surface of the fixed finger of C. (D.) doughertyensis bears a row of from 5-—8 tubercles in addi- tion to the usual subconical tubercle ventral to the denticles, and there is a wide space between the base of the finger and the prox- imalmost tubercle. In C. (D.) harti, the number of tubercles that lie dorsal to the denticles is 4 or 5 (6 in one specimen). The opposable surface of the dactyl of C. (D.) doughertyensis bears 9—12 (usually 9 or 10) tubercles, whereas that of C. (D.) harti bears 5—7 tubercles. When the fingers of the chela are closed in C. (D.) doughertyensis, the largest tubercles on the opposable sur- faces are separated from each other, but in C. (D.) harti the largest tubercles either abut or overlap. The carpus of C. (D.) doughertyensis usually bears from 7—13 spines or tubercles in addition to the large, stout distomesial spine and the two broad distoventral structures, whereas that of C. (D.) harti has 3 or 4 tubercles in addition to the usual. The former species has a short- er, deeper merus, in which the length av- erages 1.7 times the greatest depth, whereas in C. (D.) harti the merus averages twice as long as deep. The following species differ from C. (D.) doughertyensis as indicated: C. (D.) cyma- tilis has a strong, usually subacute subor- bital angle, the caudomedian spine on the mesial ramus of the uropod extends well beyond the caudal margin, and the color is blue; C. (D.) reflexus has a long, strongly recurved central projection and a well-de- fined caudal knob, both on the form I male gonopod, and the color is reddish or blue; C. (D.) strigosus has a row of plumose se- tae on the caudal border of the mesial pro- cess of the form I male gonopod, and a sin- gle spine in each cephalolateral corner of the telson; and C. (D.) truncatus usually lacks both caudomedian and caudolateral spines on the mesial ramus of the uropod and a transverse flexure in the telson, and VOLUME 116, NUMBER 3 the form I male gonopod has a very short central projection. The members of some Georgia popula- tions of C. (D.) striatus and Cambarus (De- pressicambarus) latimanus (LeConte, 1856) display multiple strong tubercles on the carpus and the ventral surface of the chela. In these species, however, there is no excision in the opposable surface of the dactyl; there is a tubercle or spine on the ischium of the antennal peduncle; both lobes of the proximal podomere of the uro- pod bear a spine; the rostrum is longer, and the abdomen is longer and much broader; the antennal scale is subquadrate, with a broader lamella; and the annulus is sub- rhomboidal rather than subovate in outline. Remarks.—As far as we can determine, the presence of spines or acute tubercles on the ventral keel of the rostrum, while a not uncommon feature in some Malacostraca, has never been reported in freshwater cray- fishes except, as mentioned, some Mexican Procambarus of the subgenera Paracam- barus and Villalobosus. An investigation of the presence or absence of this character in other U.S. species is just getting underway, but it has to date been found in several oth- er members of subgenus Depressicamba- rus, including C. (D.) harti, and in one spe- cies of subgenus Erebicambarus. Etymology.—This species is named for both the physiographic province and the county in which it appears to be endemic. Suggested vernacular name: Dougherty burrowing crayfish. Cambarus (Depressicambarus) harti Hobbs, 1981 Some inadvertent errors appeared in the description of this species (Hobbs 1981: 104-109). On page 104, column 2, para- graph 2, part of the first sentence has been omitted. It should read: Cephalic lobe of epistome (Figure 45g) subrhomboidal with short cephalomedian projection and scal- loped margins; main body with broad, shal- low central depression lacking distinct fo- 837 vea.... On page 107, column 1, paragraph 1, line 10 says “‘ventrolateral row of setae on carpus reduced to 2....” This should read: ventrolateral row of tubercles on mer- us reduced to 2. The next line says “‘mesial surface of ischium of cheliped devoid of tu- bercles,’’ which should read: mesial and ventral surface of ischium. ... These latter statements (p. 107) apply to the “‘morpho- typic male, form II,’ but the specimen ac- tually appears to be a juvenile rather than an adult male of the second form. Acknowledgments Our sincerest appreciation is expressed to S. K. Berckman, S. E. Cammack, E. E. Van De Genachte, and M. C. Freeman for as- sisting CES in the field, to T. S. Patrick for characterizing the vegetation at the type lo- cality, and to G. Henry for providing access to the type locality. The USNM types of C. (D.) harti were examined through the cour- tesy of K. J. Reed and R. Lemaitre. The manuscript was improved by the comments of R. Franz, G. Schuster, C. A. Taylor and R. Lemaitre, and publication was facilitated with funds provided by the Georgia De- partment of Natural Resources, Wildlife Resources Division. Literature Cited Bouchard, R. W. 1978. Taxonomy, ecology, and phy- logeny of the subgenus Depressicambarus, with the description of a new species from Florida and redescriptions of Cambarus graysoni, Cam- barus latimanus, and Cambarus striatus (De- capoda: Cambaridae).—Alabama Museum of Natural History Bulletin 3:26—60. Erichson, W. FE 1846. Uebersicht der Arten der Gattung Astacus.—Archiv fiir Naturgeschichte 12(1): 86—103. Girard, C. 1852. A revision of the North American Astaci, with observations on their habits and geographical distribution.—Proceedings of the Academy of Natural Sciences of Philadelphia 6: 87-91. Hay, W. P. 1899. Description of two new species of crayfish.—Proceedings of the United States Na- tional Museum 22(1187):121—123. . 1902. Observations on the crustacean fauna of Nickajack Cave, Tennessee, and vicinity.— 838 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Proceedings of the United States National Mu- LeConte, J. 1856. Descriptions of new species of As- seum 25(1291):417—439. tacus from Georgia.—Proceedings of the Acad- Hobbs, H. H., Jr. 1969. On the distribution and phy- emy of Natural Sciences of Philadelphia 7:400— logeny of the crayfish genus Cambarus. Pp. 93-— 402. 178 in P. C. Holt, R. L. Hoffman, and C. W. Ortmann, A. E. 1905. Procambarus, a new subgenus Hart, Jr., eds., The distributional history of the of the genus Cambarus.—Annals of the Car- biota of the southern Appalachians, Part I: In- negie Museum 3(3):435—442. vertebrates. Research Division Monograph 1, . 1906. Mexican, Central American and Cuban Virginia Polytechnic Institute, Blacksburg, 295 Cambari.—Proceedings of the Washington pages. Academy of Sciences 8:1—24. . 1970. New crayfishes of the genus Cambarus Villalobos, A. 1955. Cambarinos de la Fauna Mexi- from Tennessee and Georgia (Decapoda, Asta- cana: Crustacea Decapoda. Tesis, Facultad de cidae).—Proceedings of the Biological Society Ciencias, Universidad Nacional Autonoma de of Washington 83(23):241—259. Mexico, 290 pp. . 1972. The subgenera of the crayfish genus . 1983. Crayfishes of Mexico (Crustacea: De- Procambarus (Decapoda: Astacidae).—Smith- capoda). Translation of Villalobos 1955 by H. sonian Contributions to Zoology 117:1—22. H. Hobbs, Jr. Smithsonian Institution Libraries . 1981. The crayfishes of Georgia.—Smithson- and National Science Foundation, Washington, ian Contributions to Zoology 318:1—549. D.C., 276 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(3):839—845. 2003. The rare deep-sea shrimp Bitias brevis (Rathbun, 1906) (Crustacea: Decapoda: Pandalidae) from the western Pacific Shiz-Chang Chuang, Tin-Yam Chan,* and T. Komai (S-CC and T-YC) Institute of Marine Biology, National Taiwan Ocean University, Keelung, Taiwan R.O.C., e-mail (T-YC): tinyamchan @ yahoo.com.tw (TK) Natural History Museum and Institute, Chiba, 955-2 Aoba-cho, Chuo-ku, Chiba 260, Japan Abstract.—During recent deep-sea expeditions in Taiwan, Japan and New Caledonia seven specimens were collected of the rare pandalid shrimp Bitias brevis (Rathbun, 1906), a species previously known from only two specimens. The study of this new material showed that characters used in separating B. brevis from its sole congeneric species, B. stocki Fransen, 1990, are variable. This study provides additional information on this rare shrimp, including col- oration. The recently established pandalid genus Bitias Fransen, 1990 contains two species: B. stocki Fransen, 1990, type species of the genus, and B. brevis (Rathbun, 1906). Both are rare and small species that occur in deep waters from 550 to 1350 m. Bitias stocki is known based only on four specimens from the North Atlantic (Fransen 1990, Gonzalez 1995, Gonzalez & Santana 1996, Gonzalez et al. 2001). Bitias brevis is known from the holotype originally described as Pandalus brevis Rathbun, 1906, collected in Hawai, and another specimen from Madagascar (Crosnier & Fransen 1994). These two spe- cies are very similar, and only slight differ- ences in the shape of rostrum, and length of telson, have been proposed to separate them (see Crosnier & Fransen 1994). How- ever, the limited number of known speci- mens renders the evaluation of these distin- guishing characters difficult. During recent deep-water expeditions in Taiwan, Japan and New Caledonia, seven more specimens of this genus were ob- tained. Careful examination showed that they can be assigned to B. brevis. The pre- sent report provides additional information on this rare shrimp, including a color pho- * Corresponding author. tograph. The shape of the rostrum and the length of the telson proved to be variable in the Indo-West Pacific material examined of this species. It is possible that B. stocki may be synonymous with B. brevis, but more material from the Atlantic is needed for a full evaluation. The specimens are deposited in the Na- tional Taiwan Ocean University, Keelung (NTOU), Natural History Museum and In- stitute, Chiba (CBM), and the Muséum na- tional d’Histoire naturelle, Paris (MNHN). The following abbreviations are used: cl, postorbital carapace length, measured from the orbital margin to the posterodorsal mar- gin of the carapace; BT, bottom trawl; CP, beam trawl; DW, Warén dredge; stn, station. Expeditions names used are: TAIWAN 2000, joint Taiwanese-French cruise; BA- THUS, French expedition named after the Greek word for deep, bathys; and HALI- PRO, French expedition named from “hal- ieutique profonde”’. Bitias brevis (Rathbun, 1906) Figs. 1-3 Pandalus brevis Rathbun, 1906:916 (in part), fig. 65, pl. 21-fig. 3 (type-locality: Hawaii).—Chace, 1985:44. 840 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON i 77a peewee d f Fig. 1. Bitias brevis (Rathbun, 1906). a, b, e, Tatwan, TATWAN 2000 stn CP 55, female 7.2 mm cl, NTOU 2000-55; c, d, Taiwan, TAIWAN 2000 stn CP 55, 6.3 mm cl, NTOU 2000-55; f, New Caledonia, HALIPRO 2 stn BT 58, female 6.8 mm cl, MNHN-Na. 14970. a, carapace lateral view; b, left second pereiopod; c, left third pereiopod; d, dactylus of left third pereiopod; e, f, posterior abdomen and tail fan, lateral view. Scales equal 1 mm. VOLUME 116, NUMBER 3 Bitias brevis—Crosnier & Fransen, 1994: 46, figs. 1, 2a—f, 3a—b. Material examined.—Taiwan. TAIWAN 2000, R/V Fishery Researcher I, stn CP 55, 24°26.9'N, 122°18.1'E, 638 m, 4 Aug 2000, 1 female 7.2 mm cl, 1 carapace 6.3 mm cl, NTOU 2000-55. Japan. KT 02-03 cruise, R/V Tansei- maru, stn B4, SE of Amami-Oshima Island, Ryukyu Islands, 27°56.94’N, 129°31.40’E, 3 m beam trawl, 751-810 m, 16 Apr 2002, 1 male 5.5 mm cl, 2 females 5.3, 6.0 mm cl, CBM-ZC 6469. New Caledonia. BATHUS 3, stn DW 809, 23°39.39'S, 167°58.94’E, 650-730 m, 27 Nov 1993, 1 female (posterior part of ab- domen missing) 6.8 mm cl, MNHN-Na. 14969; HALIPRO 2, stn BT 58, 25°4.13'S, 168°45.21'E—25°4.99’S, 168°44.86’E, 1303-— 1500 m, 17 Nov 1996, 1 female 6.8 mm cl, MNHN-Na. 14970. Description.—Tegumental scales absent. Rostrum short, slightly shorter than half carapace length, nearly horizontal, extend- ing to second segment of antennular pedun- cle, dorsal margin densely serrated with 1 1— 16 teeth, including 4—6 on carapace poste- rior to level of orbital margin; dorsal rostral teeth generally movable but distal 1—6 teeth may lack complete basal suture; ventral margin of rostrum usually armed with 1 subdistal fixed tooth (rarely unarmed); ros- tral lateral carina well marked and contin- uous with orbital margin. Carapace with or- bital margin concave; suborbital lobe poor- ly developed; antennal spine well devel- oped, pterygostomian spine small. Eye pyriform, with well developed, dark- ly pigmented cornea, lacking ocellus. An- tennule with peduncle slightly overreaching midlength of scaphocerite; stylocerite ter- minating sharply, extending nearly to distal margin of basal segment of peduncle; basal segment with plumose setae and bristles on dorsodistal margin, ventromesial ridge without tooth; second segment with 3 or 4 spinules on dorsodistal margin. Antenna with scaphocerite about 0.7 as long as car- 841 apace, lateral margin nearly straight, disto- lateral tooth reaching distal margin of la- mella; basicerite with moderately large ven- trolateral distal spine extending to proximal end of outer margin of scaphocerite. Third maxilliped with endopod over- reaching scaphocerite by about half length of ultimate segment; penultimate segment slightly longer than ultimate segment; ulti- mate segment with several slender spinules on lateral and dorsal surfaces; exopod ab- sent. Pereiopods slender, but not extremely elongate. First pereiopod not chelate, ex- ceeding scaphocerite by 0.3—0.4 length of propodus; ischium not expanded ventrally, lacking spinules on ventral margin; carpus with few spinules on lateral surface; pro- podus shorter than carpus, with few spi- nules on lateral surface; dactylus minute. Second pereiopods subequal, overreaching scaphocerite by length of chela and 0.5—0.1 length of carpus; carpus composed of 10 or 11 articles; chela rather large, about half length of carpus; dactylus slightly shorter than palm, terminating in 2 unguis corre- sponding to simple unguis of fixed finger. Third to fifth pereiopods similar, but de- creasing in length posteriorly. Third pereio- pod overreaching scaphocerite by length of dactylus, propodus about 0.1 length of car- pus; ischium with 2 ventral spines; merus almost as long as carapace, armed with 5— 7 lateral and 4 ventral spines; carpus slight- ly shorter than propodus, with few minute spinules on lateral surface and 6—8 (rarely 0) slender spinules on ventral surface; dac- tylus about 0.2 as long as propodus, later- ally compressed, not notably curved ven- trally, bearing 4 or 5 accessory spinules on ventral margin. Fourth pereiopod over- reaching scaphocerite by length of dactylus and 0.8 length of propodus; ischium with 2 ventral spines; merus slightly shorter than carapace, armed with 6 lateral and 4 or 5 ventral spines; carpus with 3 or 4 ventral spinules. Fifth pereiopod overreaching sca- phocerite by length of dactylus and 0.4 length of propodus; ischium without ventral spine; merus about 0.7 of carapace length, 842 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ae et 9g Fig. 2. Bitias brevis (Rathbun, 1906). a, Taiwan, TAIWAN 2000 stn CP 55, female 7.2 mm cl, NTOU 2000- 55; b, Taiwan, TAIWAN 2000 stn CP 55, 6.3 mm cl, NTOU 2000-55; c, New Caledonia, BATHUS 3 stn DW 809, female 6.8 mm cl, MNHN-Na. 14969; d, New Caledonia, HALIPRO 2 stn BT 58, female 6.8 mm cl, MNHN-Na. 14970; e, Japan, KT 02-03 stn B4, female 6.0 mm cl, CBM-ZC 6469; f, g, Japan, KT 02-03 stn B4, male 5.5 mm cl, CBM-ZC 6469: a-e, lateral view of rostrum; f, endopod of first pleopod; g, appendix masculina and appendix interna of second pleopod. Scales a-d = 1 mm, f, g = 0.5 mm. VOLUME 116, NUMBER 3 843 Fig. 3. 55. armed only with 2—4 lateral, and 0 or | ven- tral spines; carpus with 2 ventral spinules. Strap-like epipods on third maxilliped through fourth pereiopods, and correspond- ing setobranchs on first to fifth pereiopods. Gill formula identical to that of B. stocki (see Fransen 1990: table 1). Abdomen with third somite dorsally rounded, without posterior spine. Fourth abdominal pleuron rounded, fifth abdomi- nal pleuron bearing distinct posteroventral denticle. Sixth somite 2.2—2.5 times as long as height, distinctly shorter than telson. Tel- son reaching or falling short of posterior end of uropods, armed with 6 pairs (some- times 7 on one side) of dorsolateral spines and 2 pairs of terminal spines (inner pair Shorter). Uropod with endopod slightly shorter than exopod; exopod with movable spinule just mesial to small posterolateral tooth. Coloration (Fig. 3).—Carapace reddish Bitias brevis (Rathbun, 1906), Taiwan, TAIWAN 2000 stn CP 55, female 7.2 mm cl, NTOU 2000- except translucent rostrum. Eyes black. An- tennules and antennae translucent except basal parts of latter somewhat reddish. Tho- racic appendages reddish. Abdomen includ- ing tail fan translucent. Size.—Males ranging 5.5—7.8 mm cl, and females 5.3—7.2 mm cl. Distribution.—Indo-West Pacific; known from Madagascar, Taiwan, Japan, Hawaii and New Caledonia, at depths of 580—1500 m. Remarks.—Crosnier & Fransen (1994) reported that B. brevis can be distinguished from B. stocki by the shape of the rostrum, and the length of the telson. The seven specimens of B. brevis reported here all have complete rostrum, but the telson and/ or uropods are broken off in four speci- mens. One specimen from Taiwan (Fig. 2a), and one from New Caledonia (Fig. 2d), have the rostrum moderately deep, and with the ventral margin convex, somewhat sim- 844 ilar to those of the holotype and allotype of B. stocki (Fransen 1990: fig. la, c; Crosnier & Fransen 1994: fig. 3c, d). The rostrum of the other specimens (Fig. 2b, c, e) are rather slender, and similar to those of the holotype of B. brevis, and the Madagascar specimen of this species reported by Crosnier & Fran- sen (1994: fig. 3a, b). The lateral carina of the rostrum is distinct in all the material reported in the present study. Moreover, the rostrum of the paratype of B. stocki is also rather slender (Crosnier & Fransen 1994: fig. 3e). Therefore, the shape of the rostrum can not be used to separate B. brevis from B. stocki as suggested by Crosnier & Fran- sen (1994). The other distinguishing character pro- posed by Crosnier & Fransen (1994) is the length of the telson, being shorter than the uropods in B. stocki, and reaching the pos- terior end of the uropods in B. brevis. The three specimens of B. brevis reported herein with complete tail fan have the telson (ex- cluding the terminal spines) falling short of the posterior end of the uropods (Fig. le, f). Crosnier & Fransen (1994) had only one specimen of B. brevis and one of B. stocki with a complete telson. It should be pointed out that the paratype of B. stocki mentioned by Crosnier & Fransen (1994) as having the telson broken, is illustrated with a complete telson in Holthuis (1993: fig. 261), and the telson does not reach the posterior end of the uropods. Although both characters proposed by Crosnier & Fransen (1994) for separating B. brevis from B. stocki are variable, careful comparison of our specimens with the orig- inal description of B. stocki shows. that there are still minor differences between them. The specimens of the type series of B. stocki have 13-18 dorsal teeth on the rostrum (Fransen 1990, but Crosnier & Fransen 1994: fig. 3c—e illustrated a maxi- mum of 17 dorsal teeth on the rostrum), which are slightly more numerous than the 11—16 dorsal rostral teeth in B. brevis. The carpi of the third to fifth pereiopods bear only a few minute spinules on the lateral PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON surfaces in our specimens, while the carpi are armed with numerous spinules in B. sto- cki (Fransen 1990: fig. 3a—c). The shape of the endopod of the male first pleopod seems to be quite different in our male specimen, and the male holotype of B. stocki. In our male specimen, the endopod is narrower, the distal part is distinctly bilobed with a well differentiated appendix interna, and a mesially directed rounded lobe (Fig. 2f). In the holotype of B. stocki, the endopod is noticeably broadened distally and not bi- lobed (Fransen 1990: fig. 3g). However, our male (5.5 mm cl) is considerably smaller than the holotype of B. stocki (7.9 mm cl), and the appendix masculina is distinctly shorter than the appendix interna (Fig. 2g), whereas the appendix masculina and the ap- pendix interna are subequal in length in the holotype of B. stocki (Fransen 1990: fig. 3h). Therefore, the observed differences in the shape of the endopod of the first pleo- pod may merely reflect the maturity of the specimens. More material from the Atlantic is needed to determine whether or not B. stocki 1s distinct from B. brevis. A specimen of B. stocki has been reported from the Ca- nary Island by Gonzalez (1995), Gonzalez & Santana (1996) and Gonzalez et al. (2001). However, only a photograph of a preserved specimen probably with a broken rostrum was provided by Gonzalez (1995: Foto 41), without any descriptive details. An attempt to borrow this Canary Island specimen has been unsuccessful. As mentioned by Crosnier & Fransen (1994), it is often difficult to determine whether the small anterior, dorsal teeth on the rostrum are movable or not. In our spec- imens, the anterior four to six dorsal rostral teeth appear to lack a complete basal suture, and are probably fixed (Fig. 2a—e). There- fore, it is necessary to modify the key char- acters of Bitias given in Holthuis (1993: 262) to ““Rostrum short, not reaching the end of antennular peduncle; dorsal teeth densely packed and more or less movable. Second pereiopods equal. Stylocerite point- VOLUME 116, NUMBER 3 ed. Proximal lobe of scaphognathite trun- Catea Acknowledgments Grateful acknowledgment is extended to A. Crosnier (MNHN) for entrusting us to report the New Caledonian material, and for his warmest hospitality to the second and third authors during their stays in Paris. We would also like to thank the MNHN for providing short term research grants to the second (T-YC) and third (TK) authors to work in the MNHN, and therefore make this work possible. The cruise “TAIWAN 2000”’ was supported by the Taiwan Fish- eries Research Institute, National Science Council, Taiwan, R.O.C., MNHN, and the Institut de Recherche pour le Développe- ment, France. The third author (TK) thanks all participants of the KT0O2-03 cruise of R/V Tansei-maru and the ship personnel for their generous help and assistance at sea. This work is a contribution from a research grant supported by the National Science Council, Taiwan, R.O.C. Literature Cited Chace, E A., Jr. 1985. The caridean shrimps (Crustacea Decapoda) of the Albatross Philippines expe- 845 dition, 1907-1910, part 3. Families Thalasso- carididae and Pandalidae.—Smithsonian Con- tributions to Zoology (411):1—143. Crosnier, A., & C. H. J. M. Fransen. 1994. Découverte de Bitias brevis (Rathbun, 1906) a Madagascar (Decapoda, Pandalidae).—Crustaceana 66(1): 45-52. Fransen, C. H. J. M. 1990. Bitias stocki, a new genus and new species of Pandalid shrimp (Crustacea, Decapoda, Caridea) in the Eastern Atlantic Ocean.—Beaufortia 41(10):67—73. Gonzalez, J. A. 1995. Catalogo de los Crustaceos De- capodos de las Islas Canarias. Publicaciones Turquesa, Santa Cruz de Tenerife, 282 pp. Gonzalez, J. A., J. A. Quiles, V. M. Tuset, M. M. Garcia-Diaz, & J. I. Santana. 2001. Data on the family Pandalidae around the Canary Islands, with first record of Plesionika antigai (Cari- dea).—Hydrobiologia 449:71-76. Gonzalez, J. A., & J. I. Santana. 1996. Shrimps of the family Pandalidae (Crustacea, Decapoda) off the Canary Islands, eastern central Atlantic.— South African Journal of Marine Science 17: 173-182. Holthuis, L. B. 1993. The recent genera of the caridean and stenopodidean shrimps (Crustacea, Deca- poda), with an appendix on the order Amphion- idacea. Nationaal Natuurhistorisch Museum, Leiden, 328 pp. Rathbun, M. J. 1906. The Brachyura and Macrura of the Hawaian Islands.—Bulletin of the United States Fish Commission (1903), 23(3):827—930, pls. 1-24. Det? ; INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Richard v. Sternberg, NCBI—GenBank, Building 45, Room 6An. 18D-30, National Institutes of Health, Bethesda, MD 20892-6510. (Do not submit manu- scripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in “GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON,” a supplement to Volume 114, number 4, December 2001. Authors are encouraged to consult this article before manuscript preparation. Copies of the article are avail- able from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page charges. The Society is a non-profit organization with limited funds, depending on financial resources and upon author request, it may subsidize a limited number of pages per vol- ume at the discretion of the editor and treasurer. Payment of full costs is encouraged and will facilitate speedy publication. Please note: Page charges are not automatically waived with membership. Authors requesting a partial waiver of page charges must justify need by emailing the editor and treasurer before proofs are returned and on the reprint request form received with page proofs. Please include email address on all correspondence. Costs.—Printed pages @ $65.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. Front cover—from this issue, p. 789. CONTENTS Rediscovery, systematic position, and re-description of “Leptoxis” melanoides (Conrad, 1834) (Mollusca: Gastropoda: Cerithioidea: Pleuroceridae) from the Black Warrior River, Alabama, U.S.A. Russell L. Minton, Jeffrey T. Garner, and Charles Lydeard 531 A new species of Magelona Miiller, 1858 (Polychaeta: Magelonidae) Maria Teresa Aguado and Guillermo San Martin 542 Allonais inaequalis (Annelida: Oligochaeta: Tubificidae) in North America R. Deedee Kathman and Mark J. Wetzel 548 A systematic review of Planaltina. (Teleostei: Characiformes: Characidae: Glandulocaudinae: Diapomini) with a descripton of two new species from the upper rio Parana, Brazil Naercio A. Menezes, Stanley H. Weitzman, and John R. Burns 557 On a new species of tree-climbing crab of the: genus Labuanium (Crustacea: Decapoda: Brachyura: Sesarmidae) from Taiwan Peter K. L. Ng and H.-C. Liu 601 Studies on western Atlantic Octocorallia (Coelenterata: Anthozoa). Part 3: The genus Narella Gray, 1870 Stephen D. Cairns and Frederick M. Bayer 617 A systematic review of the land snail Euglandina singleyana (Binney, 1892) (Mollusca: Gastropoda: Spiraxidae) Kathryn E. Perez and Ned E. Strenth 649 Redescription of adults and description of copepodid development of Dermatomyson nigripes (Brady & Robertson, 1876) and Asterocheres lilljeborgi Boeck, 1859 (Copepoda: Siphonostomatoida: Asterocheridae V. N. Ivanenko and Frank D. Ferrari 661 A new Pseudocetopsis species (Siluriformes: Cetopsidae) from Suriname and French Guiana Richard P. Vari, Carl J. Ferraris, Jr., and Philippe Keith 692 Chromosomes of Philippine mammals (insectivora, Dermoptera, Primates, Rodentia, Carnivora) Eric A. Rickart 699 Early zoeas of Athans parvus De Man, 1910 (Decapoda: Caridea: Alpheidae) reared in the laboratory Hoi Jeong Yang and Chang Hyun Kim = 710 Morphological variation in glochidia shells of six species of Elliptio from Gulf of Mexico and Atlantic Coast drainages in the southeastern United States Christine A. O’Brien, James D. Williams, and Michael A. Hoggarth 719 First fossil record of a finfoot (Aves: Heliornithidae) and its biogeographical significance Storrs L. Olson 732 The type specimen of Anoura geoffroyi lasiopyga (Chiroptera: Phyllostomidae) Joaquin Arroyo-Cabrales and Alfred L. Gardner 737 Observations on the structure of the mandibular gnathobase in some American Mesocyclops (Copepoda: Cyclopidae) E. Suarez-Morales, M. A. Gutiérrez-Aguirre, and M. Elias-Gutiérrez 742 A review of the freshwater crabs of the genus Hypolobocera Ortmann, 1897 (Crustacea: Decapoda: Brachyura: Pseudothelphusidae), from Colombia Martha R. Campos 754 A new species of Xenorhina (Anura: Microhylidae) from western New Guinea Fred Kraus and Allen Allison 803 Acanthochondria hoi, a new species of parasitic copepod (Poecilostomatoida: Chondracanthidae) on the California halibut, Paralichthys californicus, from Santa Monica Bay, California, with an amended key to the genus Acanthochondria Julianne E. Kalman 811 Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 11. Documentation of an intergeneric wood- star hybrid, Calliphlox mitchellii x Chaetocercus mulsant Gary R. Graves 820 A new burrowing crayfish of the genus Cambarus Erichson, 1846 (Decapoda: Cambaridae) from the lower Flint River basin in the Dougherty Plain of Georgia, with notes on C. (D.) harti Hobbs, 1981 John E. Cooper and Christopher E. Skelton 827 The rare deep-sea shrimp Bitias brevis (Rathbun, 1906) (Crustacea: Decapoda: Pandalidae) from the western Pacific Shiz-Chang Chuang, Tin-Yam Chan, and T. Komai 839 Table of Contents and Abstracts available online: www.apt.allenpress.com/aptonline SMITHSONIAN INSTITU iin N 0006-324X SS PROCEEDINGS oF THE Pee CAL SOCIETY aH a) A)H or WASHINGTON 31 DECEMBER 2003 VOLUME 116 NUMBER 4 THE BIOLOGICAL SOCIETY OF WASHINGTON 2003-2004 Officers President: Roy W. McDiarmid Secretary: Carole C. Baldwin President-elect: W. Ronald Heyer Treasurer: T. Chad Walter Elected Council Michael D. Carleton G. David Johnson Clyde Roper Michael Vecchione Marilyn Schotte Don Wilson Custodian of Publications: Storrs L. Olson PROCEEDINGS Editor: Richard v. Sternberg Associate Editors Classical Languages: Frederick M. Bayer Invertebrates: Stephen L. Gardiner Plants: Carol Hotton Christopher B. Boyko Insects: Wayne N. Mathis Janet W. Reid Vertebrates: Gary R. Graves Invertebrate Paleontology: Gale A. Bishop Carole C. Baldwin Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Annual dues are payable on or before January 1 of each year. Renewals received after January 1 must include a penalty charge of $5.00 for reinstatement. Library subscriptions to the Proceedings are: $50.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an additional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership, subscriptions, and/or back issues should be sent to: BIOLOGICAL SOCIETY OF WASHINGTON P.O. BOX 1897 LAWRENCE, KANSAS 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, and editorial questions should be sent to: EDITOR, RICHARD v. STERNBERG NATIONAL CENTER FOR BIOTECHNOLOGY INFORMATION—GENBANK BUILDING 45, ROOM 6An. 18D-30 NATIONAL INSTITUTES OF HEALTH BETHESDA, MD 20892-6510 Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Periodicals postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(4):847—852. 2003. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 12. Amazilia bangsi Ridgway, 1910, is an intrageneric hybrid, Amazilia tzacatl X Amazilia rutila Gary R. Graves Department of Systematic Biology, MRC-116, National Museum of Natural History, Smithsonian Institution, RO. Box 37012, Washington, D.C. 20013-7012, U.S.A. Abstract.—Amazilia bangsi Ridgway, 1910, collected at Volcan les, Costa Rica, is shown to be a hybrid between Amazilia tzacatl rutila. The unique type of Amazilia bangsi Ridgway, 1910, was collected by C. EK Un- derwood (1896) at Volcan de Miravalles, Costa Rica, on 7 September 1895. Early references treated A. bangsi as a valid spe- cies (Ridgway 1911, Cory 1918) until Si- mon (1921) suggested that it represents an aberrant specimen of A. rutila. Bangs (1930:218) later noted in a catalog of avian types in the Museum of Comparative Zo- ology: “Simon (1921, p. 106) gives it as his opinion, without, however, having seen the type, that bangsi is simply an example of rutila with an undue amount of green feathers on the sides of the breast. I do not think that this is so, but regard the type, which is unique, as a hybrid. The type in its col- oring is nearly intermediate between A. rutila and A. tzacatl, both of which are common birds in the region whence it came.” Although Bangs’ brief comment was insuf- ficient to verify the taxonomic status of A. bangsi, subsequent authors accepted this treatment without additional comment (Pe- ters 1945, Gray 1958, Panov 1989, Weller 1999). Here I provide a more comprehen- sive assessment of A. bangsi employing the methods and assumptions outlined in Graves (1990) and Graves & Zusi (1990), as modified by the findings of Graves U2, Ley). Methods The type of Amazilia bangsi, originally part of the E. A. and O. Bangs Collection UB (No. 16682), was eventually cataloged in the Museum of Comparative Zoology, Har- vard University (No. 116682). Sexed as d on the Bangs Collection label, the specimen appears to be adult as judged by the ab- sence of striations on the maxillary ram- photheca. I compared the type with all tro- chiline species (n = 26) that occur in the Cordillera de Guanacaste and adjacent low- lands of northwestern Costa Rica (see Un- derwood 1896, Carriker 1910, Slud 1964, Stiles & Skutch 1989) in the Museum of Comparative Zoology (Appendix 1). De- tailed descriptions and photographs of the type were compared with series of Costa Rican species in the National Museum of Natural History, Smithsonian Institution. Descriptions in this paper refer to definitive male plumage. Measurements of selected specimens (Ta- ble 1) were taken with digital calipers and rounded to the nearest 0.1 mm: wing chord; bill length (from anterior extension of feathers); and rectrix length (from point of insertion of the central rectrices to the tip of each rectrix). Rectrices (R1—R5) and pri- maries (PI1—P10) are numbered from the in- nermost to the outermost. General color descriptions presented in Appendix 2 were made under natural light. I evaluated the color of the lower back and abdominal plumage (3 mm lateral of the midline) with a calibrated colorimeter (CR- fms a TA R IE a = 848 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Range (mean + standard deviation) of measurements (mm) of wing chord, bill length, and rectrix length (R1—R5) of adult male Amazilia tzacatl, A. rutila, and probable hybrid, A. tzacatl < A. rutila (= type of Amazilia bangsi Ridgway, 1910; MCZ 116682). Amazilia tzacatl* (n = 13-15) Wing 54.6-58.7 (OS) 22 1123) Bill 18.5—22.7 COs iD) R1 30.7—32.9 Biles. 22 0.7) R2 32.3-35.1 (3x0. 22 O57) R3 33.3—35.7 (34.6 + 0.6) R4 33.1—36.2 G50 =7 038) R5 32.4—35.9 (yr. 22 le10) 4 Costa Rica. > Costa Rica (n = 8), Honduras (n = 6). 221 Chroma Meter, Minolta Corporation) equipped with a 3.0 mm aperture (Table 2). The measuring head of the CR-221 uses 45° circumferential illumination. Light from the pulsed xenon arc lamp is projected onto the specimen surface by optical fibers arranged in a circle around the measurement axis to provide diffuse, even lighting over the mea- suring area. Only light reflected perpendic- ularly from the specimen surface is collect- ed for color analysis. To reduce measure- ment variation, I held the aperture flush with the plumage without depressing the surface. The default setting for the CR-221 Chroma Meter displays mean values de- rived from three sequential, in situ mea- Amazilia rutila® ““Amazilia bangs’ (n = 10-14) MCZ 116682 54.9-57.0 58.8 (56.2 = 0:8) 18.2—20.7 AAD). 7/ GIES == 028)) 30.4—34.0 32.1 @ILY = 11.3) 32.1—35.9 34.2 G33 2 1.3) 33.2—36.9 3)5)- 1) (34.9 + 1.2) 33.1—37.2 36.7 (553) == 1k4) 32.7—36.4 35.7 (4n7/ == 1.3) surements. I repeated this procedure twice (five times for the type of A. bangsi), re- moving the aperture between trials. Thus, each datum summarized in Table 2 repre- sents the mean of 6 (parental species) or 15 (type of A. bangsi) independent colorimet- ric measurements. Colorimetric characters are described in terms of opponent-color coordinates (L, a, b) as per Hunter & Harold (1987). This system is based on the hypothesis that signals from the cone receptors in the human eye are cod- ed by the brain as light-dark (ZL), green-red (+a/—a), and blue-yellow (+b/—b). The ra- tionale is that a color cannot be perceived as red and green or as yellow and blue at the Table 2.—Minima, maxima, and means (+ standard deviation) of opponent color coordinates (L, a, b) of rectrix 1 (R1) of adult male Amazilia tzacatl, A. rutila, and probable hybrid, A. tzacatl < A. rutila (= type of Amazilia bangsi Ridgway, 1910; MCZ 116682). Amazilia tzacatl (n = 15) Min. Max. Mean (+ SD) Back IL DS) 35.4 32S) (2 IW) a = 110.3} ell = 34 (a= Sel) b 16.7 DOS i (E ZY) Abdomen L 3533 47.7 40.6 (+ 3.0) a a ID 330 eo) (Gen le) b Tell SES WAS Boll) Amaczilia rutila (n = 8) ‘B: ie “‘Amazilia bangsi Min. Max. Mean (+ SD) MCZ 116682 3278 37.4 B5)o0) (2 16) 85:0 —3-0 10 —= 10) G= i.5)) LY) 14.7 DAES) Sell (22 23) 24.7 42.5 49.1 46.8 (= 2.3) 42.6 S50) 14.6 I@4) (Ge 27) 10.5 16.5 BSoe 24.4 (= 3:5) P23) 1 VOLUME 116, NUMBER 4 same time. Therefore “‘redness”’ and “‘green- ness”’ can be expressed as a single value a, which is coded as positive if the color is red and negative if the color is green. Likewise, “‘yellowness”’ or “‘blueness”’ is expressed by +b for yellows and —b for blues. The third coordinate, L, ranging from O to 100, de- scribes the “lightness” of color; low values are dark, high values are light. The more light reflected from the plumage, the higher the L value will be. Visual systems in hum- mingbirds (e.g., Goldsmith & Goldsmith 1979) differ significantly from those of hu- mans, and the relevance of opponent color coordinates to colors perceived by hum- mingbirds is unknown. In any case, the L,a,b color system permits plumage color to be unambiguously characterized for taxonomic purposes. Results and Discussion I considered three hypotheses proposed by previous authors: Amazilia bangsi rep- resents (1) a subdefinitive plumage, color morph, or geographic variant of Amazilia rutila or some other species (Simon 1921); (2) a hybrid, Amazilia tzacatl X Amazilia rutila (Bangs 1930); or (3) a valid species (Ridgway 1910). For brevity I use the epi- thet, bangsi, in the remainder of the paper. Populations of Amazilia tzacatl and A. rutila from northwestern Costa Rica are ad- equately represented in museum collections (Carriker 1910). I found no evidence that bangsi represented a subdefinitive plumage, color morph, or geographic variant of either of these species or any other taxon. Rather, all evidence suggests that bangsi represents an intrageneric hybrid. The hybrid diagnosis focuses on the identification of apomorphic character states in putative hybrids (Graves 1990). However, complete dominance and poly- genic inheritance of plumage characters may preclude or obscure the expression of parental apomorphies in hybrids. When pa- rental apomorphies are not identifiable, the parentage of a hybrid may be indicated, al- 849 though less conclusively, by the presence or absence of a suite of plesiomorphic char- acters. Plumage and soft-part characters of bangsi that facilitated the identification of its parental species include: (a) ventral plumage from chin to undertail coverts pre- dominately cinnamon-buff (Fig. 1); (b) feathers along the sides of the throat and breast spangled with dull pale green sub- terminal spots; (c) rectrices (RI—R5) dark rufous tipped with bronze; (d) mandibular ramphotheca predominately yellowish- brown (probably reddish-orange or red in life); and (e) shallowly forked tail (fork depth = 5.6 mm). Perhaps as informative, bangsi lacks several conspicuous traits that are present in some potential parental spe- cies (Appendix 1): (a) contrasting rump band; (b) brilliant frontlet or coronal patch; (c) brilliant gorget; (d) pronounced blue or violet iridescence on body plumage; (e) white spots on rectrices; and (f) thickened rachises of primaries (P8—P10). The combination of plumage characters observed in bangsi can be derived from only a single pair of species—A. tzacatl X A. rutila (see Appendix 2 for comparative description of plumages). Other pairwise combinations of species can be eliminated from consideration because they either lack characters found in bangsi, or possess one or more distinctive characters that are not expressed, even subtly, in bangsi. Colori- metric values largely corroborate the visual impression that plumage color of bangsi is intermediate between that of the postulated parental species (Table 2, Fig. 2). I tested the parental hypothesis by ex- amining size and external proportions (Ta- ble 1). Measurements of trochiline hybrids invariably fall within the mensural ranges exhibited by their parental species as a con- sequence of a polygenic mode of inheri- tance (Banks & Johnson 1961; Graves 1990, 1996). Amazilia tzacatl and A. rutila are very similar in size and the percent dif- ference in character means is negligible (larger species divided by smaller): wing chonday @2272)- sbills Mleneth W@r7 75) a Ral 850 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Jae, Ih A. tzacatl X A. rutila (= type of Amazilia bangsi Ridgway, 1910; MCZ 116682). (0.3%), R2 (0.0%), R3 (0.9%), R4 (0.9%), and R5 (0.6%). Measurements of bangsi fall within the cumulative range of parental measurements for six of the seven mea- surements (the wing chord of bangsi was 0.1 mm longer than the largest value re- corded for that character in the parental spe- cies). There have been no well-documented cases of morphological luxuriance (where the size of hybrid offspring exceeds that of the parental species) among avian hybrids (Graves 1990, 1996). I suspect that the cu- mulative range of measurements for wing chord in the parental species would overlap the hybrid value if the sample size was in- creased. Had the measurements of bangsi occurred well outside (e.g., >3% larger) the Lateral views of adult males: Amazilia tzacatl (top), A. rutila (bottom), and probable hybrid (middle), range of those of A. tzacatl and A. rutila, this particular hybrid hypothesis would have been rejected. In summary, concordance of results from analyses of plumage color and external mea- surements provides strong support for the hypothesis that bangsi is an intrageneric hy- brid between A. tzacatl and A. rutila. Ama- zilia bangsi Ridgway, 1910, is thus available in taxonomy only for the purposes of hom- onymy. The parental species are sympatric from the Yucatan Peninsula (Howell & Webb 1995) south to northwestern Costa Rica, particularly in the transition zone be- tween dry thorn forest and semi-humid for- est (Underwood 1896, Carriker 1910, Slud 1964, Stiles & Skutch 1989). VOLUME 116, NUMBER 4 ice) oOo 25 19 yellowness (+b) of abdomen 10 15 20 2S 30 oS 15 ep) ® S 10 ® G GE = 28 5 po, oue) —~ CO WK + O n 0 ep) ® (Ss So o -5 a 3110 -5 0 5 redness (+a) or greenness (-a) of back Fig. 2. Bivariate relationships of L, a, b color co- ordinates: Amazilia tzacatl (*), A. rutila (A), and prob- able hybrid (@), A. tzacatl X A. rutila (= type of Ama- zilia bangsi Ridgway, 1910; MCZ 116682). Acknowledgments I thank Carole Baldwin and an anony- mous reviewer for comments, and Doug Causey, Alison Pirie, and Jeremiah Trimble (Museum of Comparative Zoology, Har- vard University) for permission to study the type of Amazilia bangsi. Travel was sup- ported by the Alexander Wetmore Fund, Smithsonian Institution. Literature Cited Bangs, O. 1930. Types of birds now in the Museum of Comparative Zoology.—Bulletin of the Mu- seum of Comparative Zoology 70:145—426. Banks, R. C., & N. K. Johnson. 1961. A review of North American hybrid hummingbirds.—Con- dor 63:3—28. Carriker, M. A., Jr. 1910. An annotated list of the birds of Costa Rica including Cocos Island.—Annals of the Carnegie Museum 6:314—915. 851 Cory, C. B. 1918. Catalogue of birds of the Americas. Part 2, No. 1.—Field Museum of Natural His- tory Zoological Series 13:1—315. Goldsmith, T. H., & K. M. Goldsmith. 1979. Discrim- ination of colors by the black-chinned hum- mingbird, Archilochus alexandri.—Journal of Comparative Physiology A 130:209-—220. Graves, G. R. 1990. Systematics of the ““green-throat- ed sunangels”’ (Aves: Trochilidae): valid taxa or hybrids ?—Proceedings of the Biological Soci- ety of Washington 103:6—25. . 1996. Hybrid wood warblers, Dendroica stri- ata * Dendroica castanea (Aves: Fringillidae: Tribe Parulini) and the diagnostic predictability of avian hybrid phenotypes.—Proceedings of the Biological Society of Washington 109:373-— 390. 1998. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 6. An intergeneric hybrid, Aglaiocercus kingi X Metallura_ tyrianthina, from Venezuela.—Proceedings of the Biologi- cal Society of Washington 111:511—520. 1999. Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 8. A provisional hypothesis for the hybrid origin of Zodalia glyceria (Gould, 1858).—Proceedings of the Biological Society of Washington 112:491—502. , & R. L. Zusi. 1990. An intergeneric hybrid hummingbird (Heliodoxa leadbeateri X Helian- gelus amethysticollis) from northern Colom- bia.—Condor 92:754—760. Gray, A. P. 1958. Bird hybrids. Commonwealth Agri- cultural Bureaux, Bucks, England, 390 pp. Howell, S. N. G., & S. Webb. 1995. The birds of Mex- ico and northern Central America. Oxford Uni- versity Press, Oxford, UK, 851 pp. Hunter, R. S., & R. W. Harold. 1987. The measurement of appearance, 2nd edition. Wiley, New York, 411 pp. Panov, E. N. 1989. Natural hybridisation and etholog- ical isolation in birds (in Russian). Nauka, Mos- cow, 510 pp. Peters, J. 1945. Check-list of birds of the world, vol. 5. Museum of Comparative Zoology, Cam- bridge, Massachusetts, 306 pp. Ridgway, R. 1910. Diagnosis of new forms of Micro- podidae and Trochilidae.—Proceedings of the Biological Society of Washington 23:53—55. Ridgway, R. 1911. Birds of North and Middle Amer- ica.—Bulletin of the United States National Museum 50, part 5. Sibley, C. G., & B. L. Monroe, Jr. 1990. Distribution and taxonomy of birds of the world. Yale Uni- versity Press, New Haven, Connecticut, 1111 pp. Simon, E. 1921. Histoire naturelle des Trochilidae (synopsis et catalogue). Encyclopedia Roret, L. Mulo, Paris. Slud, P. 1964. The birds of Costa Rica.—Bulletin of the American Museum of Natural History 128: 1-430. Stiles, E G., & A. E Skutch. 1989. A guide to the birds of Costa Rica. Christopher Helm, London, 511 pp. Underwood, C. E 1896. A list of birds collected on the lower, southern, and southwestern slopes of the Volcano of Miravalles and on the lower lands extending to Bagaces in Costa Rica, with a few observations on their habits.—Ibis (sev- enth series) 8:431—451. Weller, A.-A. 1999. Rufous-tailed Hummingbird, Ama- zilia tzacatl. P. 595 in J. del Hoyo, A. Elliott, & J. Sargatal, eds., Handbook of the birds of the world, vol. 5. Barn-owls to hummingbirds. Lynx Edicions, Barcelona, 759 pp. Appendix 1 Species of trochiline hummingbirds that occur reg- ularly in the Cordillera de Guanacaste and adjacent lowlands of northwestern Costa Rica (see Underwood 1896, Carriker 1910, Slud 1964, Stiles & Skutch 1989). Taxonomy follows Sibley & Monroe (1990). Parentheses enclose a representative list of characters or traits that would probably be expressed in hybrid progeny of these species, but that do not occur in the type of Amazilia bangsi Ridgway, 1910 (MCZ 116682): Phaeochroa cuvierii (white spots on R4—R5; thickened rachises of P8—P10); Campylopterus hemi- leucurus (purple head and breast; white spots on R3— R5; thickened rachises of P8—P10); Florisuga melli- vora (blue head and breast; white collar; RI—R5 white with black tips); Colibri delphinae (violet auricular patch); Anthracothorax prevostii (black throat; purple rectrices); Klais guimeti (violet crown and _ throat; white tips on R2—R5); Lophornis helenae (elongated black head plumes; white rump band); Chlorostilbon canivetii (black tail); Thalurania colombica (purplish- black rectrices; violet crown, lower breast, flanks, and abdomen); Panterpe insignis (brilliant coppery-orange throat; blue crown; bluish-black rectrices); Hylocharis eliciae (golden-green tail; blue throat); Amazilia ama- bilis (bluish-violet upper breast; black rectrices); Ama- zilia saucerrottei (steel-blue tail); Amazilia cyanura (blackish-violet rectrices); Amazilia rutila; Amazilia zacatl; Eupherusa eximia (white medial vanes of R2— R5); Elvira cupreiceps (R2—R5 white); Microchera al- bocoronata (white crown; white vanes on R2-—R5; deep maroon body plumage); Chalybura urochrysia (bronzy-black rectrices; pale feet in dried skin); Lam- pornis castaneoventris (brilliant green crown; purple gorget; bluish-black rectrices); Heliodoxa jacula (bril- liant green crown; small blue gorget spot; bluish-black rectrices); Heliothryx barroti (R3—R5 white; violet crown; white ventral plumage); Heliomaster constantii (red gorget; white facial stripe; white tips on R3—R5); PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Heliomaster longirostris (reddish-violet gorget; white facial stripe; brilliant bluish-green crown; white tips on R4—R5); Philodice bryantae (purple gorget; white pec- toral band; white flank tufts). Appendix 2 Abridged description of adult male Amazilia tzacatl, A. rutila, and a probable hybrid, A. tzacatl < A. rutila (= Amazilia bangsi Ridgway, 1910; MCZ 116682). The forecrown and crown of tzacatl are dark bronze becoming dark green on the hind neck, lesser and me- dian wing coverts, and back, changing to bronzy-green or bronze on the upper tail coverts. The dorsum of rutila is similar in color but slightly paler and less iridescent (particularly on the lower back and rump) owing to narrow buffy feather margins. The color and quality of iridescence of the dorsum in bangsi are in- termediate to those of tzacatl and rutila. The rectrices (R1—RS5) of tzacatl, rutila, and bangsi share a similar color pattern. Rectrices are chestnut broadly tipped with bronze in rutila, slightly darker in tzacatl. The rectrices of bangsi are intermediate in appearance be- tween those of rutila and tzacatl. Likewise, the greater wing coverts, secondaries, and primaries of bangsi are intermediate in appearance between those of the pos- tulated parental species. Underwing coverts are dark glossy green in tzacatl, dull green broadly tipped with buff or rufous in rutila, and intermediate in appearance in bangsi. Primaries and secondaries of tzacatl, rutila, and bangsi lack rufous or chestnut pigmentation. Ra- chises of tzacatl, rutila, and bangsi are unmodified. The chin, throat, breast, and flanks of tzacatl are dark green (the throat and upper breast are glowing golden-green when viewed head-on). Grayish feather margins on the abdomen become progressively wider from the sides to the ventral midline; the center of the abdomen is buffy-gray. Undertail coverts are chestnut or dark rufous. The ventral plumage of rutila from chin to undertail coverts is buffy-cinnamon, palest on the throat (feathers with narrow buffy margins in many individuals). Ventrally, bangsi is most similar to rutila but shows the influence of a ““green breasted”’ parental species, particularly along the sides of the throat and upper breast where feathers have dull, pale green, sub- terminal disks which are broadly margined with buff. Feather of the chin and upper throat of bangsi are cin- namon-buff margined with very pale buff. Scattered feathers across the center of the lower throat have small greenish-bronze subterminal disks. The midline of the abdomen of bangsi is cinnamomeous as in rutila but duller with grayish tones. Undertail coverts of bangsi are intermediate in color but closer to rutila than to tzacatl. The maxillary and mandibular ram- photheca of tzacatl, rutila, and bangsi are pale yellow- ish-brown (red in life) tipped with dark brown or blackish-brown (20—30% of the bill length). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(4):853-—872. 2003. A new small-eared shrew of the Cryptotis nigrescens-group from Colombia (Mammalia: Soricomorpha: Soricidae) Neal Woodman USGS Patuxent Wildlife Research Center, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20013 Abstract.—Cryptotis colombiana Woodman & Timm, 1993 previously was known from few specimens from two isolated regions in the Cordillera Central and Cordillera Oriental of Colombia. Recent collecting in the northern Cordil- lera Central and review of older collections from the central Cordillera Oriental in the vicinity of Bogota yielded additional specimens that permit reevaluation of the two geographic populations of these small-eared shrews. Morphological and morphometrical studies indicate that the population inhabiting the Cordil- lera Oriental represents a distinct, previously unrecognized species that I de- scribe herein as Cryptotis brachyonyx. Study of 54 specimens of shrews from the Cordillera Oriental in systematic collections in North America, South America, and Europe yielded only four specimens of the new species, all col- lected before 1926. The paucity of modern specimens suggests that C. brach- yonyx may be extremely restricted in distribution, or possibly extinct. Small-eared shrews of the genus Cryp- totis occur from eastern North America south through Central America to the An- des Mountains of South America. Recent systematic studies on these shrews gener- ally partition the species among four infor- mal, groupings originally defined by Choate (1970) and modified by Woodman & Timm (1993, 1999, 2000). Only two of these groups have members that occur in South America. The C. nigrescens-group is pri- marily a Central American group, but with two species occurring in South America: Cryptotis mera, along the Panama/Colom- bia border (Woodman & Timm 1993), and Cryptotis colombiana, reported from the Cordillera Central and Cordillera Oriental of Colombia (Woodman 1996). The C. tho- masi-group 1s comprised entirely of Andean species. The C. nigrescens-group and the C. thomasi-group can be distinguished using a combination of external, cranial, and post- cranial characters. Two of the more reliable external characters are the relative size of the forefeet and length of the foreclaws: members of the C. thomasi-group tend to have large forepaws with notably elongate foreclaws, whereas species in the C. ni- grescens-group have smaller forefeet and foreclaws. Cryptotis colombiana Woodman & Timm, 1993 originally was described from a single specimen collected in 1950 from Rio Negrito on the Cordillera Central of Colombia and now housed in the Field Mu- seum, Chicago. A second specimen, col- lected in 1925 from San Juan de Rioseco on the Cordillera Oriental of Colombia, subsequently was identified in the collec- tion of the American Museum of Natural History, New York. At that time, I (Wood- man 1996) noted differences between the two specimens, but indicated that the tax- onomic significance of this variation was difficult to interpret based on only two specimens. Field work during the past few years in the Cordillera Central by Colom- bian colleagues has added a number of im- portant new specimens of C. colombiana that permit more comprehensive evaluation 854 of the characteristics of this species (Wood- man et al. 2003). However, its presence in the Cordillera Oriental has continued to be based on the single American Museum specimen. Recently, I studied available col- lections of Colombian shrews from the Cor- dillera Oriental, among them a series of ten specimens in The Natural History Museum, London, and the National Museum of Nat- ural History, Washington, that were col- lected from April to November 1895 at or near La Selva. These specimens include at least five of the individuals Merriam (1897) used when he described C. thomasi, the first valid species of shrew identified from South America. The pelage of each of the skins is faded, and most are irregularly and some- what overly stuffed, obscuring external dif- ferences. However, among the ten are three specimens that possess small forepaws, short claws, and distinctive cranial features that mark them as members of the C. ni- grescens-group, rather than C. thomasi. This conclusion is supported by morpho- logical and morphometrical characters of the skull. Additional morphological and morphometrical analyses of the four speci- mens from the Cordillera Oriental show them to be distinct from the population of C. colombiana on the Cordillera Central. Herein, I describe the population from the Cordillera Oriental as a new species and re- port on variation within and between the two species based on the available speci- mens from the two Colombian cordilleras. Materials and Methods Regional names, place names, and map coordinates derive from original collector tags, field notes, and field catalogs supple- mented with additions, corrections, and al- ternative readings based on my review of published localities, maps, and gazetteers of Colombia (Paynter & Traylor 1981, USBGN 1988). Capitalized lifezone names follow Espinal and Montenegro (1963) and IGAC (1988). The species synonomy lists only the first use of published names. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Information from personal correspon- dence is based on my inspection of original letters from George O. Child to Oldfield Thomas [# Df 232/1/55—56] and photocop- ies of correspondence from Clinton Hart Merriam to Thomas [# Df 232/1/252—257] in the Official Archives of The Natural His- tory Museum, London, and from microfilm of correspondence from Thomas to Merri- am in The Bancroft Library, University of California, Berkeley [BANC Film 1958]. Terminology of dentition and dental characteristics follows Choate (1970). Cap- italized color names for pelage hues are from Ridgway (1912). Pelage coloration was determined only from museum speci- mens. All four specimens of the new spe- cies were collected more than 80 years ago, and their pelages are faded and do not ac- curately represent those of live or newly- captured specimens. Measurements used in my analyses fol- low Woodman & Timm (1993, 1999) and are in mm, weights in g. Abbreviations are explained in Table |. External measure- ments were recorded from skin tags or orig- inal field notes, except for head-and-body length, which was calculated by subtracting length of tail from total length. Measure- ments of the skull were taken to the nearest 0.1 mm using either an ocular micrometer in a binocular microscope or a hand-held dial caliper (for CBL and CB). Univariate statistics include mean += SD and _ total range. Ratios of measurements were mul- tiplied by 100 to calculate percentages (Ta- ble 2). In describing species of Cryptotis, I compare them primarily with other mem- bers of the genus, unless stated otherwise. Comparative terms used for measurements and ratios reflect this context. A feature of ‘“*moderate”’ length or width is one whose mean value for the species falls within the range of plus or minus one standard devi- ation (SD) of the mean value for the genus. A “long” or “‘wide”’ feature is one greater than the mean plus one SD; a “‘short’’ or “narrow” feature is less than the mean mi- nus one SD. VOLUME 116, NUMBER 4 Principal components analysis (PCA) of a correlation matrix of 16 log-transformed craniomandibular variables (ZP, IO, M2B, Pb, WR, (UID IMO R= IMUDWE: IMI Eee Is(@le HCV, HAC, AC3, TRM, Lm1, BAC—Ta- ble 3) was used to investigate relationships in overall shape of the skull among 6 C. colombiana from the Cordillera Central, 6 C. mera, 32 C. thomasi, and 4 individuals of the new species. I carried out a second PCA to determine morphometrical relation- ships between just C. colombiana and the new species using a correlation matrix of 8 log-transformed cranial variables (ZP, IO, U3B, PL, TR, UTR, MTR, M1W—Table 4). Values of the variable M2B do not over- lap between the two species (Table 1), and it was excluded from the second PCA to determine whether the two samples would continue to be separated using combina- tions of other variables. In my investigations of Cryptotis, I op- erate under a systematic framework for the genus outlined previously (Woodman 2002). My comprehension of the biological species is under the philosophical influence of the evolutionary species concept as re- defined by Wiley (1978) and characterized by Wiley & Mayden (2000). In application, a species represents the largest monophy- letic entity whose constituent parts interact, that maintains its own identity, and that has an independent evolutionary trajectory (Frost & Hillis 1990). In practice, I use unique distributional patterns of morpho- logical characters among populations to dis- tinguish individual, presumably genetically- cohesive groups from other genetically-co- hesive groups. This application provides testable hypotheses for additional studies. Specimens from the following institu- tions were used in this study: American Museum of Natural History, New York (AMNH); Natural History Museum [for- merly British Museum (Natural History)], London (BM); Field Museum, Chicago (FMNH); Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogota (ICN); University of Kansas Natural His- 855 tory Museum, Lawrence (KU); Museum of Comparative Zoology, Cambridge (MCZ); Museo de la Universidad de Antioquia, Me- dellin (MUA); Royal Ontario Museum, To- ronto (ROM); National Museum of Natural History, Washington (USNM). Systematic Biology Cryptotis brachyonyx, new species Figs. 1A, 1B, 2, 4A, 5A, 6A Blarina thomasi Merriam, 1897:227 (in part). [Blarina (Cryptotis)| thomasi: Trouessart, 1904:138 (in part; name combination). Cryptotis thomasi: O. Thomas, 1921:354 (in part; name combination). Cryptotis avia: Tate, 1932:226 (in part; not Cryptotis avia G. M. Allen, 1923). Cryptotis thomasi thomasi: Cabrera, 1958: 48 (in part; name combination). Cryptotis colombiana: Woodman, 1996:417 (in part; not Cryptotis colombiana Wood- man & Timm, 1993). Holotype.—Dried skin (Fig. 1) and skull (Fig. 2) of adult male, BM number 99.10.3.2, collected by an unnamed worker for George O. Child of Bogota, Colombia, on 19 October 1895; purchased from G. O. Child and presented to the British Museum (Natural History) by Oldfield Thomas. Skin in fair to good condition: all fur present, faded; 2 small holes near right cheek; left hind foot sewn on; other feet loosely at- tached. Skull in good condition: slight dam- age to orbital areas; small crack running dorsally from foramen magnum; mandibles separate; both angular processes intact. Of standard external field measurements, only length of hind foot (= 12 mm) was record- ed on the skin tag, and this measurement probably was taken from the dried skin. Type locality.—_COLOMBIA: Depart- ment of Cundinamarca: “‘La Selva, near Bogota.”’ La Selva was the name of George O. Child’s estate on the Plains of Bogota, the altitude of which he estimated as ap- proximately 8900 feet (G. O. Child, in Litt. 856 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Measurements of selected Cryptotis from Colombia. Statistics are mean + SD, and range. Mea- surements follow Woodman & Timm (1993). C. mera C. colombiana C. brachyonyx C. thomasi External measurements: he= i n=4 n=O n= 12 Head-and-body length (HB) (9) ae Os) 22 7/ — oy Ze 6 66—73 60-76 74-96 Tail length (TL) Dae aaa B10) 2 3} — WA se D) 24-31 27-34 20—27 Skull measurements: n = 63 n = 63 n = 4 n= 17? Condylobasal length (CBL) G53 2 OA! 19.6 22 O28 AQ AL 2 OI IG EZ OS 17.7-18.7 19.4—20.1 20.3—20.4 20.7—22.6 (n = 5) (n = 3) Cranial breadth (CB) 9.4 + 0.3 Qe az OA! 7 OS sO 9.1-9.8 9.2—10.0 10.2—10.9 (n = 1) (n = 13) Breath of zygomatic plate (ZP) AQ) 28 @, Il A) z= Oz Il Ouse Oil DQ) 22. QD 1.9-2.2 1.8—2.1 1.7—2.0 BA Interorbital breadth (IO) Al 5) 2& (2 Alf) ae (Il 49 + 0.2 5.0 = 0.2 4.3—-4.7 4.7-4.9 4.6—-5.0 4.6—5.2 Breadth across first unicuspids (UIB) Do, Opi Ded) = OOS A 22 sil Dj == Oil 2.4—2.6 2.6—2.7 2.5—2.6 2.6—2.9 (n = 3) Breadth across third unicuspids (U3B) 2.9 + 0.1 Bil se @. II 3.0 + 0.05 31 se 0). Jl 2.7—3.0 2.9—3.2 2.9-3.0 2.9-3.2 Breadth across second molars (M2B) 56 22 OI 6), 28 (OY Scr 22, OI GY), 22 OY 5.5—5.8 6.0-6.4 5.7-5.9 5.8—-6.5 Palatal length (PL) W825 Oz of Se O38) 8.6 + 0.2 9.4 + 0.2 7.6—-7.8 8.1—9.0 8.3-8.8 8.89.8 Length of upper toothrow (TR) Fell 2 0),II TO 2 OD TO = O33 2 22 (0),3) 6.9-7.2 7.3-7.8 7.3—7.9 7.7-8.7 Length of unicuspid toothrow (UTR) Dg, se (0), 1l D5. 22 Opil Ds) SE (0), il Ags 26 (02 2,.3—2.5 2.4—2.6 2.4—2.5 2.4—3.0 Length of molariform toothrow (MTR) 5) Jl 32 Opi Sy.) 22 OW 5,5) 28, 07 Syaey 2, (0) 5.1-5.2 5.3—5.8 5.3—-5.7 5.5—6.0 VOLUME 116, NUMBER 4 Table 1.—Continued. C. mera C. colombiana Posterior width of first upper molar (M1W) 1ESe==OLO5 AAO) SE Oil 1.8-1.9 1.9-2.0 Length of mandible (ML) 5.9 + 0.4 G37 = O2 5.2-6.2 6.3—7.0 Height of coronoid process (HCP) AnSee= zl AnOy == 02 4.34.6 4.34.7 Height of coronoid valley (HCV) Def 22 (II Mages c= Q)2 2.62.8 2.5—3.0 Height of articular condyle (HAC) 3a 22 Obl Sai) 2 OY 3.6-3.8 3.64.2 Breadth of articular condyle (BAC) 3) Ze O, It BP), = (07 2.9-3.1 3.0-3.4 Articular condyle to m3 (AC3) 44+ 0.1 ALS) ce (0) II 4.3-4.6 4.6—5.0 Length of lower toothrow (TRM) So) 22 Oil So) se O2 5.4—-5.7 5.6—-6.3 Length of lower toothrow (M13) 4.30201 46+ 0.1 4.14.4 A a8) (a) Length of first lower molar (Lm1) less 22 Ohl (Q 2 Oz! 1.7-1.8 1.8—-1.9 Weight (g): 4 Except as noted. 13 July 1895). Merriam (1897) gave the el- evation of La Selva as 9000 ft. This slightly higher elevation probably was based on in- formation in a letter to him from Thomas (quoted below). Paratypes (2).—COLOMBIA: Cundina- marca: Plains of Bogota (USNM 80905; ““topotype”’ [of C. thomasi] written on tag); La Selva, near Bogota (BMNH 99.10.3.3). C. brachyonyx C. thomasi ile) = OS) Ores Out 1.8—1.9 1.8—2.1 6:85 ==3053 7/40) 220)? 6.4-7.0 6.6-7.4 4.6 *_0.1 Ang =ta Oui 4.54.7 4.54.9 2ge) 25, O05) Sell EOI 2.8-2.9 2.9-3.3 3L) 22 ell AVS} ae (Ol 3.84.1 4.14.6 Pell 22 OY 3(0, 22 OA 3.0—3.4 3.44.1 30) 22 Oeil 57 08 4.9-5.1 5.0-6.0 G105==02 6:4 = 02 5.7-6.2 6.1—6.7 4.5 + 0.1 Ae =O, 4.3-4.5 4.2-5.1 (n = 17) lees BE OI EO Rees Oat 1.7-1.8 1.7—2.0 Referred specimen (1).—COLOMBIA: Cundinamarca: San Juan de Rrifoseco (AMNH 70597). Etymology.—Greek: brachys (short) + onyx (claw), hence “‘short claw;”’ a noun in apposition. Distribution.—Presumably Premontane Moist Forest, Premontane Wet Forest, Low- er Montane Moist Forest, Montane Moist 858 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Characters of selected Colombian Cryptotis. C. mera C. colombiana C. brachyonyx (n = 7)8 (n = 7) (n = 4) Foramen of sinus canal present O% O% 0% @ = 3) Foramen dorsal to dorsal articular facet present 71% two 43% two 25% two 14% one 14% one 50% one 14% none 43% none 25% none Two dorsal foramina present 71% 86% 100% Petromastoid foramen present one or both petromastoids 14% (med-lg) 100% 100% (n = 5) Postero-lingual cuspules on cingulae of U1-—3 17% absent 83% minute 100% minute 100% obvious (n = 6) (n = 4) Size of U4 (% of U3) 54 + 10.0% 42 + 13% 40 + 7% 40-66 32-61 31-48 (n = 6) (@ = 3) Emargination of posterior borders of P4, M1 v. slight to slight v. slight to slight v. slight to slight Complexity of M3 100% simple 86% simple 14% complex 100% complex Lower sigmoid notch v. shallow to shallow shallow v. shallow to mod. deep Entoconid of m3 present 29% minute 100% minute 71% absent 100% absent Mensural characters: n = 63 n = 63 n = 4 Relative tail length (TL/HB x 100) 39 + 4 47 x9 — 34-46 36-57 Gi 7) (n = 4) Relative length of rostrum (PL/CBL xX 100) AD atamileal ABTS a= 15) 42.9 + 0.6 41.3-44.1) 41.8—-46.2 42.2-43.4 Ca— >) (n = 3) Relative breadth of zygomatic plate (ZP/PL x 100) AAS AD) SE I De se Oss Die se Neil 24.7—28.2 22.1—24.1 20.5—23.3 C. thomasi (n = 39) 8% (minute) (n = 38) 3% two 15% one 82% none 719% 100% (n = 35) 16% absent 62% minute 22% obvious (n = 32) 48 + 9% 22-64 @ = 35) v. slight to mod. deep 50% simple 50% complex (n = 28) v. shallow to mod. deep 11% obvious 32% minute 57% absent (n = 28) n = 17? ys) ES 21-36 (m = 22) LB9) a2 OS 42.4 44.0 Pp \ epee VES) 18.7—24.7 VOLUME 116, NUMBER 4 Table 2.—Continued. 859 C. mera C. colombiana C. brachyonyx C. thomasi (n = 7)? (n = 7) (n = 4) (n = 39) Relative breadth of zygomatic plate (ZP/CBL x 100) LIL) s= O07 KO) SE (05) OS) a2 (03) OD, 10:6 10.3-12.1 9.3-10.5 9.3-9.8 7.8-10.6 (n = 5) (n = 3) Breadth of interorbital area (IO/CBL) 24.8 + 0.6 MS) se OA! 24.4 = 03 MZ jl 2 Oca 24.3-25.8 23.9-25.1 24.0—24.6 21.0—24.6 (n = 5) (n = 3) Relative length of unicuspid toothrow (UTR/CBL x 100) 13.3) 28 O24! 12 = 0x4 2 se OK8 1D.O) 22 O46 12.6-13.6 12.0—-13.2 11.8-12.3 11.6—14.0 (= >) C3) Relative palatal breadth (M2B/PL xX 100) VAR SDLP) TZ ED? O70) 2 20 66.6 + 2.6 70.5—76.3 69.0—75.3 64.8-68.7 60.8—70.5 Relative height of coronoid process (HCP/ML x 100) 76.3 + 4.8 GBS 2 1 O78 22 LO GO, IL se 18) 71.0—82.7 65.7—71.2 64.3—70.3 63.4-70.6 Relative length of posterior portion of the mandible (AC3/ML xX 100) We 22 Si! Wall se BAO Jak = DP S1l.O) se BA 68.4—-82.7 70.0-—75.8 71.4-76.6 71.4-86.8 Relative length of posterior portion of the mandible (AC3/HCP x 100) 98.9 + 2.3 LOO 22 27 1O2.3) 22 12 ID2 S547 95.7—102.2 104.3-111.4 108.5-111.1 111.1-128.9 “Except as noted. Forest, and Montane Wet Forest life zones on the Cordillera Oriental in central and eastern Cundinamarca Dept., Colombia (Fig. 3); known elevational distribution, ca. 1300-2715 m. Diagnosis.—A small- to medium-sized Cryptotis with a moderately long tail, long dorsal pelage, small forepaws, and short foreclaws. Typically two large dorsal for- amina; no lateral branch of the sinus canal or associated foramen; a large foramen on postero-medial edge of tympanic process of each petromastoid. Interorbital area broad; rostrum, zygomatic plate, and palate of moderate breadth. Dentition bulbous; U4 large, but not visible in lateral view; teeth moderately pigmented, with distinct color in protoconal basins (and occasionally pale pigmentation in hypoconal basin) of P4 and M1-—2; posterior borders of P4 and M1-2 only slightly recessed; anterior and poste- rior elements of ectoloph of M1 about equal; M3 complex. Coronoid process mod- erately high, joins horizontal ramus at a high angle; posterior mandible short; artic- ular process low and broad; inferior sig- moid notch variably shallow; p3 short and high; minute entoconid on m3. Description.—A small- to medium-sized Cryptotis. Tail moderately long (Fig. 1; there are no external measurements, except length of hind foot, recorded for any of the four specimens). Forepaws and foreclaws small (Fig. 4), similar in proportions to those of C. nigrescens and other members of the C. nigrescens-group. Dorsal fur typ- ically 5—6 mm long, individual hairs up to 7 mm. Dorsal pelage of the four available 860 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 3.—Factor loadings for the first three factor axes from principal components analysis (PCA) of 18 cranial measurements from Cryptotis colombiana, C. mera, C. brachyonyx, and C. thomasi. Variables are listed in de- scending order by their loadings on the second axis. Abbreviations of measurements are explained in Table 1. Correlations Variable PC1 PC2 PC3 ZP —0.060 0.210 —0.610 IO —0.238 0.202 —0.081 M2B —0.245 0.406 0.079 PL —0.298 —0.088 0.058 TR = 02298 —0.077/ 0.082 UTR —0.264 —0.133 0.070 MTR AN} ,S)3 0.128 0.073 M1W (O20 0.647 —0.107 LM =O.277/ —=(O(QUS) = 0)087/ HCP =O 197, —0.068 0.057 HCV O.2VG —0.174 =0NS6 HAC —0.265 —0.288 V.O2ZZ AC3 =O Zl —0.224 —0.014 TRM SU. —0.009 0.158 Lml —0.161 0.303 0.443 BAC = 0205 a Onli 0.106 Eigenvalue: 10.183 1.308 1.143 Proportion of variation explained: 0.636 0.082 0.071 specimens Chestnut Brown to Prout’s Brown to Mummy Brown; venter Olive Brown to Buffy Brown to Hair Brown; dor- sal and ventral hairs 2-banded, with gray base and mid-section and ca. 1 mm brown- ish distal tip. [Pelage of these skins is fad- ed; live C. brachyonyx probably have a darker pelage, possibly similar to that of C. colombiana (see Comparisons, below).] Two obvious dorsal foramina present along the suture between the frontals (100%; Table 2); generally about equal in size (75%). Typically lacking lateral branch of sinus canal and associated foramen (see Woodman & Timm 1999) posterior to the dorsal articular facet (100%). A minute fo- ramen often present dorsal to one (50%) or both (25%) dorsal articular facets. Large, obvious foramen on the postero-medial edge of the tympanic process of both petro- Table 4.—Factor loadings for the first three factor axes from principal components analysis (PCA) of 8 cranial measurements from Cryptotis colombiana and C. brachyonyx. Variables are listed in descending order by their loadings on the second axis. Abbreviations of measurements are explained in Table 1. Correlations Variable PCl PC2 PC3 U3B —0.246 OF583 0.075 UTR —0.140 0.506 —0.695 M1W —=O2Bil 0.478 0.526 LP —0.404 —0.006 0.347 Pie —0.449 —0.034 SOs OS MTR —0.429 —0.093 —0.059 TR —0.417 —0.242 —0.310 IO =0)379 —0.404 0.054 Eigenvalue: 4.1890 B37 0.9006 Proportion of variation explained: 0.524 0.169 OnGIS VOLUME 116, NUMBER 4 A 861 L. vical Explorations, U.S. Dept. 45 Ae Bi ea S\ 2 Blevins Lromea : an PL OS “Bua Sclombio: ahd, emo we 2 al ons, [ S. De 5 : = C ARemae Whe % & eet SME Mf Bogsta Calon Colombia, Bary. " ATs. icm Bice Ie USNM 80904; D: USNM 80903). The smaller body size and shorter length of tail in C. brachyonyx are per- ceptible despite the irregular preparation. mastoids (100%); foramen not as large as in C. thomasi or C. colombiana; positioned more medially than in either of those spe- cies. Rostrum of moderate length (PL/CBL = 42.9%). Interorbital area broad (IO/CBL 24.4%). Zygomatic plate of moderate breadth in proportion to CBL (9.5%) and PL (21.7%); anterior border of zygomatic plate at posterior mesostyle-metastyle val- ley to metastyle of M1; posterior border from parastyle to posterior half of M3, and from posterior one-half to posterior edge of maxillary process. Palate of moderate breadth for the genus (M2B/PL = 67.0%). Anterior process of petromastoid variable in SIZe: Dried skins of C. brachyonyx (A: BM 99.10.3 .2—holotype; B: USNM 80905) and C. thomasi (C: Dentition bulbous. Teeth moderately pig- mented: medium red to dark red on tips of cones, styles, and cristae; pale to medium pigment typically extends into protoconal basins of M1 and M2; very pale pigment only occasionally in hypoconal basins. Short, somewhat crowded unicuspid tooth- row (UTR/CBL = 12.1%); U4 mostly ob- scured by U3 and P4 in lateral view of skull (100%; Fig. 5). Ul-—3 broad in lateral view and straight to convex on posteroventral margin. Cuspules on posterolingual cingu- lum of U1-—3 typically absent to minute; this region of these unicuspids typically pig- mented. U4 large, averaging ca. 40% of the surface area of U3 (Fig. 6). Posterior borders 862 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Skull of the holotype of Cryptotis brachyonyx (BM 99.10.3.2): (A) dorsal view of cranium; (B) ventral view of cranium; (C) lateral view of cranium and mandible; and (D) posterior view of articular process of the mandible. VOLUME 116, NUMBER 4 CARIBBEAN PACIFIC © <5 c o ce) o e) O Figs 3: mera, and C. thomasi. The 1000 m contour is shown. of P4, M1, M2 only slightly recessed. An- terior element of ectoloph of M1 approxi- mately equal in length to posterior element; protoconal basin of M1 about equal in size to hypoconal basin. M3 appears moderately complex, partly resulting from the presence of a small, but distinct hypocone that squares off the labial border of the tooth and from the relatively extensive pigmentation; par- astyle, paracrista, paracone, and precentro- crista well-developed and pigmented; me- sostyle, postcentrocrista, and metacone re- duced, but pigmented and obvious; proto- 863 C. brachyonyx C. colombiana C. mera C. thomasi 100 km Map of the northern Andes showing the distributions of Cryptotis brachyonyx, C. colombiana, C. cone pigmented and obvious; hypocone present, but reduced and unpigmented. Anterior border of coronoid process of mandible joins horizontal ramus at a rela- tively high angle; coronoid process mod- erately high (HCP/LM = 67.8%). Inferior sigmoid notch variably shallow to moder- ately deep. Posterior mandible behind m3 short (AC3/ML = 74.1%). Articular pro- cess relatively low and broad. Third lower premolar nearly as high as it is long. Mi- nute, uncolored, but distinct entoconid pre- sent (100%) on talonid of m3. 864 ===: —— a 1mm PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON {~ | eS oe \| gay: < ail se 7 Fig. 4. Ventral views of left forepaws of (A) Cryptotis brachyonyx (BM 99.10.3.2) and (B) C. thomasi (BM 99.10.3.4), both from La Selva, Colombia. Postcranial skeleton unknown. Comparisons.—Cryptotis colombiana: Externally, C. brachyonyx and C. colombi- ana may be difficult to distinguish. Based on cranial dimensions, C. brachyonyx 1s generally close in size to C. colombiana (e.g., see Multivariate Analyses, below), both appear to have a moderately-long tail, and both may have dark pelage. The orig- inal description of the pelage of C. colom- biana (dorsum Olive Brown to Fuscous; ventrum Buffy Brown to Hair Brown Woodman & Timm 1993) was based solely on the faded holotype. Study of a more re- cently-collected specimen (MUA 62) indi- cates that the pelage is closer to Sooty Black to Chaetura Black dorsally, with a Dark Olive Gray to Chaetura Black venter. Dorsal and ventral hairs have a medium- gray base and middle, with a ca. 1 mm dark-brown to black tip. However, the two species are easily distinguished cranially: C. brachyonyx has a smaller and more medi- ally located foramen on the tympanic pro- cess of the petromastoid; relatively and ab- solutely narrower palate; darker, more ex- VOLUME 116, NUMBER 4 Figs s: cuspids: (A) C. brachyonyx (USNM 80905); (B) C. colombiana (MUA 62); (C, D) Cryptotis thomasi (KU 157765, FMNH 71030). Arrows point to (a) curvature of posteroventral margin of U1 and (b) visible U4. Lateral view of upper first incisor and uni- 865 tensively pigmented dentition, with pigment more completely coloring protoconal basins of M1-—2 and hypocone of M1; narrower upper dentition; larger and more complex M3, resulting from more strongly pigment- ed protocone, better developed (but unco- lored) hypocone, and better developed and pigmented postcentrocrista and metastyle; minute entoconid in talonid of m3. Cryptotis mera: Cryptotis brachyonyx 1s absolutely larger in many cranial dimen- sions, and it possesses a large, obvious fo- ramen on the tympanic process of the pe- tromastoid; relatively shorter zygomatic plate (Table 2); relatively narrower palate; more extensively pigmented dentition, with darker pigment more completely filling pro- toconal basins of M1-—2 and coloring hy- pocone of M1; relatively shorter unicuspid toothrow; smaller average U4; more com- plex M3 possessing a small hypocone, strongly pigmented protocone, and more distinct postcentrocrista and metastyle; rel- atively lower coronoid process; minute en- toconid more commonly present in talonid of m3. Cryptotis thomasi-group: Like other members of the C. nigrescens-group, C. brachyonyx has noticeably smaller forefeet and shorter foreclaws; Ul-—3 typically are relatively broad and conical in lateral view of the skull, with a straight or convex pos- teroventral margin (Fig. 5); the anterior el- ement of ectoloph of M1 is approximately equal in size to the posterior element; the anterior border of the coronoid process of the mandible joins the horizontal ramus at a relatively high angle; the articular process is relatively low and broad. Cryptotis thomasi: This is the only spe- cies of shrew with which C. brachyonyx may be sympatric. In addition to the char- acters that separate it from members of the C. thomasi-group in general, Cryptotis brachyonyx can be separated by its smaller body size and longer tail. Cranially, C. brachyonyx has a smaller and more medi- ally located foramen on the tympanic pro- cess of the petromastoid; smaller average 866 1mm Fig. 6. 62); and (C) Cryptotis thomasi (USNM 80906). U4; less emarginate posterior borders of P4, M1, and M2; relatively shorter posterior portion of the mandible (AC3); minute en- toconid more commonly present on talonid of m3. Results of multivariate analysis.—Prin- cipal components analyses (PCA) of cran- iomandibular variables measured from C. brachyonyx, C. colombiana, C. mera, and C. thomasi strongly support the distinctive- ness of C. brachyonyx. The first analysis, which incorporated all four species, showed greatest separation among groups on factor axes | and 2. When factor scores on these two axes are plotted, individuals of the four species form four distinct groups of points (Fig. 7). Along the first factor axis, inter- preted as size (Table 3), the species sepa- rated into three size groupings with mini- mal overlap. Specimens of C. thomasi com- prise the largest individuals, and those of C. mera the smallest. Individuals of C. brachyonyx and C. colombiana, which overlap nearly completely along this axis, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Occlusal view of upper unicuspids: (A) C. brachyonyx (USNM 80905); (B) C. colombiana (MUA form a single group of intermediate size. The largest C. colombiana and the smallest C. thomasi also overlap along this axis. Along the second factor axis, which is in- terpreted as representing width of palate (M2B) and width of dentition (WM1—see Table 3), the groupings of individuals of C. colombiana and C. brachyonyx separate without overlap. A second PCA incorporating cranial var- 1ables from just C. colombiana and C. brachyonyx (Fig. 8) again shows nearly complete overlap along the first factor axis, interpreted as overall size of the cranium (Table 4). However, the two species segre- gate completely along the second factor axis, which represents the combined effects of variables that include IO, U3B, UTR, and WMI (Table 4). The variable M2B was specifically excluded from this analysis to assess whether the two species could be separated using other variables. The lack of intersection between the scatters defined by the two samples is notable, because over- VOLUME 116, NUMBER 4 Inrias We 867 C. colombiana C. brachyonyx C. mera C. thomasi pc 1 Plot of scores-on factor axes | and 2 from PCA of 18 craniomandibular measurements (Table 3) from Cryptotis brachyonyx, C. colombiana, C. mera, and C. thomasi. laps in multivariate space are common among species within the Cryptotis nigres- cens-group, even those that are geographi- cal neighbors (Woodman & Timm 1993, Woodman 2000). Discussion The histories of the original specimens of C. brachyonyx and C. thomasi are closely associated. At least one of the specimens that Merriam (1897) used to name C. tho- masi is now known to be of C. brachyonyx, indicating that since its original description, C. thomasi has been polytypic in content. This seeming oversight partly reflects an understandable ignorance at that time of the quality and extent of variation within and among species of small-eared shrews. It also can be attributed to the poor quality of preparation of the specimens, a point noted by Thomas in a letter to Merriam (see be- low). In his paper naming C. thomasi, Merriam (1897:227) wrote: ““For the opportunity to describe this very interesting shrew I am indebted to Mr. Oldfield Thomas, Curator of Mammals in the British Museum, who sent me seven specimens from the type lo- cality. Heretofore the genus Blarina has not been recorded from any point south of Cos- ta Rica; hence the discovery of the present species in South America is of unusual in- terest.” The seven specimens sent to Mer- riam were from a collection of shrews that eventually numbered at least ten skins (eight with accompanying skulls) collected at or near La Selva for George O. Child in 1895 and purchased by Oldfield Thomas for the BM (Table 5). Six of these specimens, including the holotype of C. thomasi, cur- 868 -2 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Dea O_ C. colombiana @ CC. brachyonyx Fig. 8. Plot of scores on factor axes 1 and 2 from PCA of 8 craniomandibular measurements (Table 4) from Cryptotis brachyonyx and C. colombiana. rently are in BM and four are in the USNM, the result of an ongoing exchange of spec- imens between Thomas and Merriam. In his letter to Merriam regarding these speci- mens, Thomas (in litt. 25 December 1895) wrote: I am sending you 6 or 7 Blarinas from Bogota, a long way south of any locality that I expect you know of. We have had an odd specimen from there since about 1845 but it has never been named, & this lot have just come in to me personally. It is sad that they are such awful skins, but they are made by natives, & one must be thankful for the perfect Table 5.—Specimens of shrews collected for G. O. Child from near the type locality of C. brachyonyx and C. thomasi. Specimens are listed in order by date. All specimens with dates were collected in the year 1895. Abbreviations and symbols: SS—skin and skull; SO—skin only; *—holotype. Museum number USNM 80904 USNM 80905 USNM 80906 USNM 80903 BM 99.10.3.3 BIMU S/S) AILS BM 99.10.31 BM 99.10.3.2* BM 99.10.3.4 BM no number (GDC #311) Q AIA AAYAY Species . thomasi . brachyonyx . thomasi . thomasi brachyonyx . thomasi thomasti . brachyonyx . thomasi . thomasi Tag locality Plains of Bogota Plains of Bogota La Selva La Selva Le Selva, near Bogota G. Child’s Estate, Plains of Bogota Boqueron, near Bogota La Selva, near Bogota La Selva, near Bogota Bogota (Child) Date 5 April 5 April 6 Aug. 9 Aug. 10 Aug. 14 Sept. 24 Sept. 10 Oct. 1 Nov. no date Preparation SO SS SS SS SS SS SS SS SS SO VOLUME 116, NUMBER 4 skulls & exact localities. A dozen more are likely to come to me, so I will give you all these except one, the best, which you shall select for me, & re- turn with your determination upon it, of course if new, & you would kindly describe it I would ask you to make that one the type. “Plains of Bogota’ means an estate some few miles out of the town, at an altitude of about 9000 feet. The other places are all about the Bogota dis- trict. The material is so poor compared to what you are accustomed to that you may despise it, & not care to work it out. I have kept then back some here hoping to get your paper on the genus, but I am now going away for a long holiday on account of my health, & prefer to send them you before I go. There is a discrepancy between the num- bers of specimens noted by both Thomas and Merriam and the current number of specimens available. Although Thomas clearly noted sending Merriam up to seven specimens (with permission to keep all but one), and, based on Merriam’s (1897) report all seven arrived, only four specimens ever were catalogued in the collection of the USNM. The difference does not result from counting skins and skulls separately, be- cause the four USNM specimens and the holotype in the BM include five skins and four skulls, which would yield nine speci- mens. There is no evidence that any speci- mens from Child were exchanged by either BM or USNM with any other collections in North America or Europe. One possibility is that Merriam returned to Thomas two specimens in addition to the holotype, in which case, the remaining specimens may be among those in the BM. Unfortunately, I have yet to find correspondence that men- tions the return of the holotype to Thomas. Alternatively, some of specimens may have been lost or discarded prior to cataloging in either collection. That there may at one time have been even more specimens is suggested by Thomas in his letter to Merriam, when he noted, “‘a dozen more are likely to come to me....’ This statement probably was based on a letter from Child to Thomas (in litt. 7 Nov. 1895), in which Child wrote, “‘I have got already some 12 shrews ready for 869 you and some 20 bats.’’? What happened to all of these specimens is not known. I have been able to trace only the 10 specimens now in the BM and USNM (Table 5). The AMNH specimen of C. brachyonyx and one of the AMNH specimens of C. tho- masi at one time were identified as the only other known specimens of Cryptotis avia G. M. Allen, 1923 (Tate 1932). Cryptotis avia originally was described on the basis of a single specimen, a dried skin with skull (MCZ 20091), collected by Nicéforo Maria at El Verjon on the Cordillera Oriental in 1922. Although the holotype is a small in- dividual based on both the external mea- surements from the dried skin reported by G. M. Allen (1923) and my cranial mea- surements, in my opinion it is conspecific with specimens of C. thomasi (AMNH 62789, 62790; MCZ 19995) taken by the same collector from the type locality of C. avia in March and October 1922 and with specimens of C. thomasi from elsewhere (Woodman 1996). The name C. avid is therefore a junior synonym of C. thomasi and not applicable to C. brachyonyx. The previous confusion of the AMNH specimen of C. brachyonyx with C. avia is under- standable, because the crania of both spec- imens are incomplete (Woodman 1996) and because, for some time, the mandibles of the two specimens had been interchanged. Both conditions served to obscure some of the characteristics important for distin- guishing the two taxa. As noted by Merriam (1897), C. thomasi was the first species of that genus (at that time a subgenus of Blarina) described from south of Costa Rica. It also represents the first valid species of the family Soricidae or of the order (Insectivora or Soricomorpha) from South America. Several species pre- viously had been interpreted as South American shrews, although all were deter- mined eventually to be otherwise. These er- roneous soricids included Mus araneus Marggraf, 1648, and Musaraneus brasilien- sis Brisson, 1762, both based on specimens of the didelphid marsupial Monodelphis 870 from Brazil [In any case, Marggraf’s names are pre-Linnaean, and Brisson’s names gen- erally are considered unavailable (Hop- wood 1947)]. Two species from Surinam, Sorex surinamensis Gmelin, 1789 and Blar- ina pyrrhonota Jentinck, 1910 (first men- tioned, but not described, by Jentinck 1888), were based on specimens of Euro- pean Sorex araneus mistakenly purported to hail from South America (Husson 1963, Hutterer 1993). It is reasonable to presume that the spec- imens collected for George Child in 1895 may have been the first specimens of true shrews collected in South America. How- ever, as noted by Thomas in his correspon- dence to Merriam, there 1s one other spec- imen in the BM (number 54.1.11.4)—a skull of C. thomasi without skin from ‘“‘New Grenada” that had been “purchased of Mr. S. Stevens.’’ Based on the catalog number, this specimen must have been col- lected in 1854 or earlier, and Thomas (in litt. 25 Dec. 1895) noted that it had been in the possession of the BM “since about 1845,’ making it the first verifiable speci- men of a shrew from South America. My inspection of 54 specimens of shrews from the Cordillera Oriental in Cundina- marca Department, Colombia, that are available in systematic collections has yielded only four specimens of C. brachy- onyx, all of which were collected either in 1895 or in 1925. The remainder are all C. thomasi. Cryptotis brachyonyx represents 30% of the ten existing specimens collected from near La Selva in 1895. The four spec- imens of C. brachyonyx comprise 13% of 26 shrews collected in this region through 1931. Among 24 specimens collected since 1950, all are C. thomasi; there are no spec- imens of C. brachyonyx. The overall scar- city of C. brachyonyx and its absence in later collections may reflect a species re- stricted to habitats that have not been ade- quately sampled since the 1920s and that have shrunken dramatically in size or dis- appeared in the region surrounding the type locality as a result of agriculture and urban PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON development. Alternatively, it is possible that the species may be extinct, but previ- ous experience with the history of collect- ing C. colombiana (Woodman et al. 2003) suggests that this is unlikely and new spec- imens will begin to be collected. However, intensive surveys in the Cordillera Oriental using appropriate methods will be required to elucidate the modern geographic and ecological distributions of C. brachyonyx. Acknowledgments The following curators and collections managers provided loans or permission to examine specimens under their care: Guy G. Musser and Robert Voss (AMNH); Pau- la Jenkins, Richard Harbord, and Daphne Hulls (BM); Lawrence R. Heaney, Bruce D. Patterson, John Phelps, and William Stanley (FMNH); Alberto Cadena (ICN); Robert M. Timm and Thor Holmes (KU); Maria E. Rutzmoser (MCZ); Carlos A. Cuartos and Carlos A. Delgado (MUA); Mark D. Engs- trom (ROM). Thanks to Susan Snell and Polly Tucker of the Official Archives, The Natural History Museum, London, and to David Kessler of The Bancroft Library, University of California, Berkeley, for pro- viding access to, or copies of, original cor- respondence. Jerry R. Choate, Alfred L. Gardner, and Robert M. Timm _ provided helpful comments on earlier versions of this manuscript. Literature Cited Allen, G. M. 1923. A new shrew from Colombia.— Proceedings of the New England Zodlogical Club 8:37—-38. Brisson, M. J. 1762. Regnum animale in classes 1x. Distributum, sive synopsis methodica sistens generalem Animalium distributionem in classes ix, & duarem primarium classium, Quadruped- um scilicet & Cetaceorum, particularem divisi- Onem in ordines, sectiones, genera & species. Theodurom Haak, Lugduni Batavorum. Cabrera, A. 1958. Catalogo de los mamiferos de América del Sur.—Revista del Museo Argenti- no de Ciencias Naturales “‘Bernadino Rivada- via’ 4 (1):i-1v, 1-307. Choate, J. R. 1970. Systematics and zoogeography of VOLUME 116, NUMBER 4 Middle American shrews of the genus Crypto- tis.—University of Kansas Publications, Muse- um of Natural History 19:195—317. Espinal T., L. S., and E. Montenegro M. 1963. For- maciones vegetales de Colombia. Memoria ex- plicativa sobre el mapa ecoldgico. Instituto Geografico “Agustin Cocazzi,” Departamento Agrologico, Bogota, 201 pp. Frost, D. R., & D. M. Hillis. 1990. Species in concept and practice: herpetological applications.—Her- petologia 46:87—104. Gmelin, J. FE 1789. Systema naturae per regna tria na- turae, secundum classes, ordines, genera, spe- cies; cum characteribus, differentiis, synonymis, locis, 13th edition. Volume I. J. B. Delamol- lierre, Lugdunum. Hopwood, A. T. 1947. The generic names of the man- drill and baboons, with notes on some of the genera of Brisson, 1762.—Proceedings of the Zoological Society of London 117:533—536. Husson, A. M. 1963. On Blarina pyrrhonota and Echi- mys macrourus: two mammals incorrectly as- signed to the Suriname fauna.—Studies on the Fauna of Suriname and Other Guyanas, 13:34— 41, pl. 1-2. Hutterer, R. 1993. Order Insectivora. Pp. 69—130 in D. E. Wilson and D. M. Reeder, eds., Mammal spe- cies of the world. Smithsonian Institution Press, Washington, 1206 pp. IGAC. 1988. Suelos y bosques de Colombia. Instituto Geografico Agustin Codazzi, Santafé de Bogo- ta;.135 pp: Jentinck, EK A. 1888. Catalogue systematique des mam- miferes (rongeurs, insectivores, cheiropteres, edentes et marsupiaux). E. J. Brill, Leiden. . 1910. Description of a shrew from Surinam.— Notes from the Leyden Museum, 32:167—168. Marggraf, G. 1648. Historiae rerum naturalium, liber sextus, qui agit quadrupedibus, & serpentibus. Pp. 221—236, 243—244 in W. Piso and G. Marg- graf, Historia Naturalis Brasiliae. EK Hackium, Lugdunum Batavorum. Merriam, C. H. 1897. Descriptions of five new shrews from Mexico, Guatemala, and Colombia.—Pro- ceedings of the Biological Society of Washing- ton 11:227—230. Paynter, R. A., Jr., & M. A. Traylor. 1981. Ornitholog- ical gazetteer of Colombia. Bird Department, Museum of Comparative Zoology, Harvard University, Cambridge, 311 pp. Ridgway, R. 1912. Color standards and color nomen- clature. Published privately, Washington, 43 Pp oUpL: Tate, G. H. H. 1932. Distribution of the South Amer- ican shrews.—Journal of Mammalogy 13:223-— 228. Thomas, O. 1921. New Cryptotis, Thomasomys, and 871 Oryzomys from Colombia.—Annals and Mag- azine of Natural History, Series 9, 8:354—357. Trouessart, E.-L. 1904. Catalogus mammalium tam viventium quam fossilium. Quinquennale Sup- plementum Anno 1904. Fasciculus 1, pp. 1— 288. R. Friedlander and Sohn, Berolini. USBGN [U.S. Board on Geographical Names]. 1988. Gazetteer of Colombia, 3rd edition. Defense Mapping Agency, Washington, D.C., 859 pp. Wiley, E. O. 1978. The evolutionary species concept reconsidered.—Systematic Zoology 27:17—26. , & R. L. Mayden. 2000. The evolutionary spe- cies concept. Pp. 70—89 in Q. D. Wheeler and & R. Meier, eds., Species concepts and phylo- genetic theory. A debate. Columbia University Press, New York, xiv + 230 pp. Woodman, N. 1996. Taxonomic status of the enigmatic Cryptotis avia (Mammalia: Insectivora: Sorici- dae), with comments on the distribution of the Colombian small-eared shrew, Cryptotis col- ombiana.—Proceedings of the Biological Soci- ety of Washington 109:409—418. . 2000. Cryptotis merriami Choate in Costa Rica: syntopy with Cryptotis nigrescens (Allen) and possible character displacement.—Carib- bean Journal of Science 36:289—299. . 2002. A new species of small-eared shrew from Colombia and Venezuela (Mammalia: Sor- icomorpha: Soricidae: genus Cryptotis).—Pro- ceedings of the Biological Society of Washing- ton 115:249-272. , & R. M. Timm. 1993. Intraspecific and inter- specific variation in the Crypftotis nigrescens species complex of small-eared shrews (Insec- tivora: Soricidae), with the description of a new species from Colombia.—Fieldiana: Zoology, New Series 1452:1—30. , & . 1999. Geographic variation and evolutionary relationships among broad-clawed shrews of the Cryptotis goldmani-group (Mam- malia: Insectivora: Soricidae).—Fieldiana: Zo- ology, New Series 1497:1—35. 5 Oe . 2000. Taxonomy and evolutionary relationships of Phillips’ small-eared shrew, Cryptotis phillipsti (Schaldach, 1966), from Oa- xaca, Mexico (Mammalia: Insectivora: Sorici- dae).—Proceedings of the Biological Society of Washington 113:339-355. , C. A. Cuartas-Calle, and C. A. Delgado-V. 2003. The humerus of Cryptotis colombiana and its bearing on the phylogenetic relationships of the species (Soricomorpha: Soricidae).— Journal of Mammalogy 84:832-—839. Appendix: Specimens Examined Specimens marked with an asterisk (*) in the fol- lowing listings are cranial remains recovered from owl 872 pellets. Specimens marked with an asterisk and in brackets are mandibles from owl pellets. Cryptotis brachyonyx (4).—COLOMBIA: Cundi- namarca: San Juan de Rioseco (AMNH 70597); La Selva, near Bogota (BM 99.10.3.2—holotype, 99.10.3.3); Plains of Bogota (USNM 80905). Cryptotis colombiana (14)—COLOMBIA: Antio- quia: Corregimiento San Antonio de Prado, 2100-— 2800 m (MUA 060, 062); Finca Campino, Alto de San Miguel, 2000 m (KU 157761); Reserva Ecolégica Alto de San Miguel, 2150 m (MUA 12009, 12010*, 12011* [MUA 12012*, 12013*]); Vereda San Francisco, 2750 m (KU 157762*, MUA 12005* [MUA _ 12007*, 12008*]); Rio Negrito, 15 km E of Sonso6n, 1750 m (FMNH 69816—holotype); Finca Los Sauces, 2150 m (MUA 12001). Cryptotis mera (7).—PANAMA: Darien: Cerro Ta- carcuna, 4800 ft (USNM 337967—337969); Cerro Mali, 4700 ft (USNM 337966); Mount Pirri [Cerro Pir- re], E slope near head of Rio Limoén, 4500—5000 ft (USNM 178974—178976—including holotype). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Cryptotis thomasi (50).—COLOMBIA: [no specific locality] (BM 54.1.11.4; MCZ 27596). Cundinamarca: Represa del Neusa (ICN 9659); Bogota (AMNH 34605); Plains of Bogota (USNM 80904); Paramo de Bogota, 2900 m (AMNH 37381; MNHN 1962-1068); La Selva [G. O. Child’s Estate], near Bogota, (BM 97.5.21.2—holotype, 99.10.3.4, BM no number [GDC #311]; USNM 80903, 80906); Boqueron, near Bogota (BM 99.10.3.1); Paramo de Monserrate, 3200—3300 m (ICN 9649, 9650, 9652, 9658; ROM 51870); San Fran- cisco, 3000-3500 m (AMNH 71354, 71355; FMNH 7NO23,, 7102455710255 7102637 10277 LOZ SsaalOZ9» 71035); San Cristobal, 2800-2900 m (FMNH 71030, 71031, 71032, 71033, 71034, 71036, 71037); Reserva Biolégica Carpanta, 3000 m (ICN 10995, KU 157765); Paramo de Choachi, 3000 m, (AMNH 38405, MCZ 19885, 20090, 20092, 27597, 27598); Paramo el Ver- jon (AMNH 62789, 62790, MCZ 19995, 20091—ho- lotype of C. avia); Chipaque (USNM 251960); Fusa- gasuga (MCZ 27599); Paramo de Chisaca, 3100 m (ICN 5223). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(4):873-—882. 2003. A new species of the catfish genus [tuglanis from French Guyana (Osteichthyes: Siluriformes: Trichomycteridae) Mario de Pinna and Philippe Keith (MDP) Museu de Zoologia da Universidade de Sao Paulo, Caixa Postal 42694 Av. Nazaré 481, Sao Paulo-SP 04299-970 Brazil, e-mail: pinna@ib.usp.br; (PK) Muséum national d’histoire naturelle, Laboratoire d’ichtyologie, 43 rue Cuvier, 75231 Paris Cedex 05, France, e-mail: keith@mnhn.fr Abstract.—Ituglanis nebulosus is described from the Aprouague basin in French Guiana. It differs from all other species of /tuglanis by a color pattern composed of large, round spots irregularly distributed on body and not coa- lesced into longitudinal stripes. It is also diagnosable from most congeners by the round caudal fin and the reduced number of vertebrae (35 or 36). The new species shares all synapomorphies so far proposed for the genus /tuglanis, as well as other characteristics supporting the placement of the genus within tri- chomycterids. The neotropical catfish family Tricho- mycteridae is known mostly for the semi- parasitic (hematophagous and lepidopha- gous) habits of some of its members in the subfamilies Stegophilinae and Vandelliinae, popularly called “‘candirus’’. However, most species in the family are free-living generalized predators of aquatic inverte- brates, usually inhabiting fast-flowing, clear-water streams. The majority of those are included in the subfamily Trichomyc- terinae, a vast non-monophyletic assem- blage (de Pinna 1989) comprising over 100 described species (cf. Eigenmann 1918, Tchernavin 1944) plus many others unde- scribed (de Pinna 1998). One consequence of the non-monophyly of the Trichomycterinae is that some forms included in the group are expected to be more closely related to other subfamilies than to remaining trichomycterines (de Pin- na 1989, 1998). Costa & Bockmann (1993) recognized a small monophyletic group of species formerly included in Trichomycte- rus, which they removed to a separate ge- nus, /tuglanis. Costa & Bockmann pro- posed that /tuglanis, rather than forming a clade with other species of Trichomycterus, is instead the sister group to a large clade comprising the subfamilies Tridentinae, Stegophilinae, Vandelliinae, Sarcoglanidi- nae, and Glanapteryginae. That hypothesis implies that /tuglanis, though superficially similar to Trichomycterus and other mem- bers of the subfamily Trichomycterinae, is a key taxon in understanding the evolution of the pronounced morphological and be- havioral specializations found in more dis- tal trichomycterids, including the parasitic forms. The genus /tuglanis, as defined by Costa & Bockmann (1993), currently comprises ten species, namely /. amazonicus (Stein- dachner), /. eichorniarum (Miranda Ribei- ro), I. gracilior (Eigenmann), J. herberti (Miranda Ribeiro), /. laticeps (Kner), I. me- tae (Eigenmann), J. parahybae (eigen- mann), /. parkoi (Miranda Ribeiro), /. pas- sensis Fernandez & Bichuette, and J. proops (Miranda Ribeiro). The distribution of species assignable to the genus, several of which remain undescribed, covers much of cis-Andean South America, from the Guyanas in the north to Uruguay in the south. 874 In this paper, we describe a distinctive new species of /tuglanis, representing the first species of the genus described from French Guyana. Material and methods.—Meristics and morphometrics are taken according to de Pinna (1992). Skeletal observations are mostly based on a single paratype, cleared and counterstained for bone and cartilage according to a procedure combined and modified from the methods of Taylor & Van Dyke (1985) and Song & Parenti (1995). Data on vertebral number and pleural ribs were taken from the cleared and stained paratype and from radiographs of the ho- lotype and alcoholic paratype. Radiographs were made with a digital x-ray system at the National Museum of Natural History, Smithsonian Institution. Vertebral counts do not include those in the Weberian complex, and the compound caudal centrum (PU1+U1) is counted as one. Number of branchiostegal rays and odontodes are based on the single cleared and stained paratype. Comparative material examined.—Itug- lanis amazonicus, MZUSP 30449 (1 ex c&s); I. eichorniarum, MZUSP 40792 (13 Ox, 24 C&S), SS33I ©} ee ik eraculionr FMNH 53264, x-ray of holotype, MZUSP 65415 (1 ex); 1. herberti, MZUSP 22186 (1 ex), 2209 (1 ex); I. parahybae, MZUSP TO SMOK CREAN SS) MIA S Rag is 2 Grex): I, metae, MZUSP 26030 (2 ex); I. parkoi, MNRJ 3849 (holotype), MZUSP 23368 (1 ex); I. passensis, MZUSP 80097 (3 ex), 80098 (3 ex), 80099 (2 ex); I. proops, MZUSP uncat (5 ex c&s); Jtuglanis sp. 1, MCP 10420 (1 ex c&s, from Uruguay); Ttuglanis sp. 2 MZUSP 24128 (2 ex c&s, from the Amazon); /tuglanis sp. 3 MNRJ 11489 (5 ex c&s, from Tocantins basin). The unidentified species listed above (sp. 1, sp. 2, etc.) refer to undescribed species which are comparatively relevant for this paper. They will be described in separate publications. Material of /. laticeps was not available for examination, and data on that species were obtained from Eigenmann PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (1918). Additional comparative material of other trichomycterids and other catfishes is listed in de Pinna (1992). Ituglanis nebulosus, new species Figs. | and 2 Ttuglanis sp.; Le Bail, Keith & Planquette, 2000:164, 165 (brief description and col- or photograph) Holotype.—MNHN 2001-1128, 36.6 mm SL, French Guiana, Arataye river (tributary of Approuague river), near the natural pre- serve “Reserve naturelle des Nouragues’”’, (04°00'22"N, 52°36'34"W), col. P. Plan- quette & et al., Sep 1985. Paratypes.—Two specimens, both col- lected with holotype. MNHN 2001-1129, 31.1 mm SL; MZUSP 69574, 35.1 mm SL (specimen cleared and stained). Diagnosis.—Ituglanis nebulosus 1s dis- tinguished from all other species of /tug- lanis by a color pattern composed of large dark blotches (larger than twice eye diam- eter), which often contact each other at their borders but do not coalesce. Dark spots in all other /tuglanis species, when present, are either very fine and irregular or regu- larly spaced and round, no larger than eye, or irregular and coalescent forming rough- ly-defined longitudinal stripes. /tuglanis parahybae has the color pattern most sim- ilar to that of I. nebulosus, but its dark markings are mostly irregular (vs. round) and partly coalesce as two or three longi- tudinal series. The differences in color pat- tern between the two species are well dem- onstrated in color photographs of fresh or live specimens included in fig. | of Costa & Bockmann (1993) and page 165 of Le Bail et al. (2000). The round caudal fin dis- tinguishes /. nebulosus from all of its con- geners except /. metae. In all other species of /tuglanis, the caudal fin is either truncate or truncate with rounded dorsal and ventral corners and a gently convex posterior mar- gin (see Discussion). The new species is further diagnosable from /. gracilior, geo- graphically its closest congener, by the VOLUME 116, NUMBER 4 875 Table 1.—Morphometric data for /tuglanis nebulosus. A—holotype, B and C—paratypes, D—mean. Mea- surements 2—13 expressed as % of SL, 14—27 as proportions of HL. 1—Standard length (mm) 2—Total length 3—Body depth 4—Caudal peduncle length 5—Caudal peduncle depth 6—Predorsal length 7—Preanal length 8—Prepelvic length 9—Antevent distance 10—Dorsal-fin base length 11—Anal-fin base length 12—Pelvic-fin length 13—Head length 14—Head width 15—Head depth 16—Interorbital 17—Eye diameter 18—Snout length 19—Mouth width 20—Length of interopercular patch of odontodes 21—Length of opercular patch of odontodes 22—-Posterior internarial width 23—Anterior internarial width 24—Posterior internarial diameter 25—Anterior internarial diameter 26—Pectoral-fin length 27—Length of pectoral-fin filament shorter maxillary barbel (reaching posteri- orly slightly beyond posterior margin of pectoral-fin base; vs. minimally to mid- length of pectoral fin exclusive of filament). The presence of only three pleural ribs dis- tinguishes J. nebulosus from all congeners except J. amazonicus and I. gracilior. How- ever, rib number is not known in all species of Ituglanis. The reduced number of verte- brae in J. nebulosus, 35 or 36, is also un- usual in the genus, where vertebral counts normally range otherwise from 39 to 41 (41 in J. gracilior, 39 in I. eichorniarum and I. parahybae, 40—41 in I. proops). The only other species with 36 vertebrae is /. passen- sis, a Cave species easily distinguished from all other /tuglanis species by the reduction of eyes and of integumentary pigmentation. Description.—Morphometric data for the holotype and paratypes are provided in Ta- ble I. Cross-section of body round or slight- A B (e D 36.6 sla 3)5)s 34.3 120.8 129 118.2 120.3 16.1 14.5 Ss 152 18.0 17.4 18.5 18.0 es) 11.6 11.4 IL) 72.4 73.0 70.4 TVS) 75.4 74.6 VIL WaT 62.6 62.7 63.5 62.9 68.9 68.5 69.8 69.0 Well 7.4 Teil 726 6.3 6.4 6.3 Gs 8.7 8.4 8.3 8.5 18.6 I).3 18.8 18.9 88.2 90.0 89.4 89.2 DLS 45.0 39.4 45.3 26.5 30.0 Pa Tfe3) PY SS, 10.3 8.3 6.1 8.2 36.8 38.3 39.4 a2 50.0 48.3 40.9 48.4 DSO) 26.7 25.8 DES 13.22 1323 10.6 12.4 iQ), 1 20.0 18.2 Oe 22a PAT, LOE Pal Dy) 5.0 6.1 5.6 SL) 5.0 4.5 il 58.8 55.0 DLO SA 29.4 Sy, Sy) 8) 33.0 ly depressed at pectoral-fin insertion, be- coming increasingly compressed posterior to midlength of trunk. Caudal peduncle ta- pering gradually to caudal fin in dorsal view. Dorsal profile of body straight from head to origin of dorsal fin, sometimes slightly convex at posterior half of trunk. Dorsal profile of head continuous with that of dorsum (Fig. 1). Caudal peduncle evenly deep, nearly as much as remainder of body, expanded posteriorly by procurrent rays. Ventral profile of body gently convex along abdomen, then practically straight from or- igin of pelvic fins to origin of caudal fin. Myotomes obscured by thick integument but visible by relief on caudal peduncle and dorsal part of posterior half of trunk. Re- gion of longitudinal skeletogenous septum visible as a shallow wide depression, prob- ably in part due to post-mortem shrinkage. Axillary gland inconspicuous on surface of 876 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. body, apparenly located dorsoposteriorly to base of pectoral fin, as inferred from slight- ly different texture of integument. Its small slit-like pore located shortly dorsoposterior- ly to first pectoral-fin ray, slightly antero- ventral to first lateral-line pore. Head depressed, parabolic in dorsal view (nearly round excluding snout), slightly longer than wide. Snout differentiated from rest of head by deep concavity on each side in dorsal view (Fig. 2). Jaw muscles hyper- trophied laterally, resulting in swollen cheeks. Eyes small, without free orbital rim, located dorsally on head but directed dor- JAS De 2001-1128, dorsal view of head. Ituglanis nebulosus, holotype, MNHN Ttuglanis nebulosus, holotype, MNHN 2001-1128, 36.7 mm SL, lateral view. solaterally. Skin covering eyes thin and transparent. Eyes located entirely on ante- rior half of HL. Interorbital nearly four times eye diameter. Eye lens well-formed and visible. Anterior nostril located slightly closer to margin of snout than to anterior margin of eye, surrounded by membrane of integument forming short tube, continuous laterally or posterolaterally with nasal bar- bel. Anterior nares separated by space slightly smaller than interorbital. Posterior nostrils slightly smaller than eye, located slightly closer to anterior nares than to eye, partly closed by thin cup-like flap of integ- ument along their anterior margin (no cor- responding flap posteriorly). Posterior in- ternarial width slightly smaller than anterior one. Opercular patch of odontodes small, dor- solaterally located on head, its posterior end reaching vertical through posterior margin of base of first pectoral-fin ray or to middle of fin-base. Approximately 12 opercular odontodes underlain by small, well-differ- entiated oval fold of integument. Interoper- cular patch of odontodes located ventrolat- erally on head, anteroventral to and approx- imately twice size of opercular patch. Oper- cular and interopercular odontode patches close to each other, almost juxtaposed in some specimens (Fig. 1). Approximately 17 interopercular odontodes, underlain by short flap of integument. Posterior odon- todes largest. VOLUME 116, NUMBER 4 Mouth subterminal, its corners only slightly curved posteriorly. Upper lip al- most straight, covered with large papillae. Premaxillary teeth arranged in two well-de- fined regular rows. Outer row teeth fine and conical, inner row teeth more robust, some- what blunt at tip. Lower lip thin, poorly- differentiated from remainder of ventral surface of head, covered with papillae smaller than those on upper lip. Flat, thin integumentary outgrowth at corner of mouth, adjacent to posterior margin of rictal barbel base. Branchiostegal membranes thick, narrowly joined to isthmus medially and forming large free fold. Six branchios- tegal rays. Branchial openings broad, un- constricted. All barbels large (Fig. 2). Max- illary barbel longest and broadest, reaching posteriorly slightly beyond posterior margin of pectoral-fin base. Rictal barbel smaller than maxillary one, extending to posterior margin of interopercular patch of odontodes or to anterior margin of pectoral-fin base. Nasal barbel approximately as long and wide as rictal one, extending posteriorly to perpendicular through posterior margin of opercular patch of odontodes, or slightly beyond. Lateral line reduced to short tube im- mediately posterior to head, its terminal pore slightly posterior to vertical through posterior margin of pectoral-fin base. Single additional pore anterior to it, located close to posterior margin of opercular patch of odontodes. Pectoral fin narrow, distally convex, with i+5 rays. Rays progressively shorter posteriorly. First pectoral-fin ray preceded by broad anterior margin of integ- ument, forming anterior edge of fin. First ray longer and thicker than others, pro- longed into filament representing approxi- mately one-third the total length of ray. Length of pectoral fin (filament included) approximately equal to HL. Pelvic-fin 1+4; first unbranched ray shorter than second and third. Origin of fin well anterior to that of dorsal fin. Large pelvic splint present. Posterior margin of pelvic fin slightly con- vex, covering anal and urogenital openings 877 but not reaching origin of anal fin. Dorsal fin small, with broadly round distal margin, its origin anterior to vertical through origin of anal fin. Dorsal-fin rays 11+6 plus one accessory anteriorly. Anal fin similar to caudal in size but deeper in shape, with 1i+5 rays plus one accessory anteriorly. Caudal fin round, approximately twice as long as deep, continuous with caudal pe- duncle, with 6+6 principal rays. Procurrent caudal-fin rays difficult to discern in alco- hol-preserved specimens, 12 dorsally and 10 ventrally in cleared and stained paratype. Vertebrae 35 (MNHN paratype) or 36 (holotype and MZUSP paratype). Pleural ribs three. Branchiostegal rays seven. Pigmentation.—A color photograph of a freshly-collected specimen is provided in Le Bail et al. (2000:165). In preservative, whole of dorsum and sides of body covered with irregularly-distributed, roundish dark markings (Fig. 1). Most individual marks larger than twice eye diameter, often partly coalescent, with those on ventral part of sides smaller than rest and those on dorsum less conspicuous due to darker background. Thin, dendritic, white threads evident on close examination, probably representing superficial blood vessels, sometimes outlin- ing limits between myotomes. Hypural plate outlined by large, dark, triangular spot. Dorsal part of head as dark as dorsum, but marbled tan, with spots less clearly de- fined. Area around eyes and nares markedly darker than rest, especially on region im- mediately lateral to nares (Fig. 2). Cheeks with well-defined spots smaller than those on body. Area of opercular odotodes nearly or totally white, but surrounding integu- ment rim very darkly-pigmented, forming narrow round black edge around odontode patch (Fig. 1, 2). Interopercular patch of odontodes white, except for dark area at base. Ventral surface of head mostly white, except for faint fields of dark chromato- phores on anterior part of mental region, lower jaw and lower lip. Part of cheek chro- matophores also visible in ventral view. Na- sal barbel and dorsal surface of maxillary 878 barbels covered with irregular fields of dark chromatophores along their basal three- fourths. Rictal barbel lacking dark integu- ment pigmentation. All barbels with clearly visible reddish-tan cores (cf. nasal barbels in Fig. 2). Caudal fin, except for its hyaline distal third, covered with spots similar to those on body. Dorsal fin with elongate dark fields on its proximal half. Anal fin with sparse, dark fields along base. Pectoral fin with small, elongate dark field along ba- ses of one or two anterior rays. Pelvic fin lacking dark pigment. Abdomen white. Etymology.—From the Latin nebulosus, meaning cloudy, misty, in reference to the integumentary pigmentation pattern of this species. Ecological notes.—AlI|l material of the new species was collected at a slow-flowing section of the river, about 3 meters wide, 20-50 cm deep, and densely shaded by tropical rain forest (gallery forest). The bot- tom was sandy, covered with leaf litter. Wa- ter was clear, slightly tea-stained. The exact microhabitat of the specimens could not be determined, because collecting was done with rotenone. Eight species were found to- gether with /tuglanis nebulosus: Dysichthys coracoideus, Helogenes marmoratus, Bry- conops affinis, Hoplias aimara, Gastero- pelecus sternicla, Copella carsevennensis, Astyanax meunieri, and Poptella brevispi- na. Stomach contents seen by transparency in the cleared and stained paratype of I. ne- bulosus show numerous arthropod remains. Distribution.—So far known only from type locality, in Approuague basin, French Guyana (Fig. 3). Comments on the taxonomy of Ituglan- is.—The genus /tuglanis seems monophy- letic as presently constituted and of partic- ular importance in being the sister-group to a large subclade of trichomycterids, includ- ing highly specialized parasitic forms (see section below). However, the limits of most of its species remain poorly-known, and there are, additionally, several yet unde- scribed species assignable to the genus. The PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SURINAME i Py St Laygnt ATLANTIC OCEAN cy d Ss ¥ & 3} | Kourou fi é > £ f R ae oie Cayenne \ Saut fracas a 8 ‘ \ pe X ) Dan re ey \ G. \ ‘ Re, égina IG ON t ry a 7 as » & - \ N 8 ® St Georges ry @Sa Is R Cd tof As BRASIL qIMOTeIN, < N = Nas 7 0 50 Km Fig. 3. Type locality (square mark) of /tuglanis ne- bulosus in French Guiana. systematics of the genus, therefore, is clear- ly in need of detailed revisionary studies. In the course of this study, some com- parative data were encountered that may be useful in elucidating the taxonomy of Jtug- lanis. The shape of the caudal fin was found to be of systematic value. In most species of [tuglanis, as in most of Trichomycterus, the caudal fin is vertically truncate with round corners, with the posterior margin gently convex or straight. The shape of the fin seems to vary little intraspecifically, even in the largest samples examined (over 100 specimens of /. proops). In a few spe- cies, the caudal fin is round, as in the new species herein described. Miranda Ribeiro (1908) and Eigenmann (1918) reported the caudal fin as round in J. proops, but this seems to be mistaken, since it is truncate in all specimens examined of the species. I- lustrations in Miranda Ribeiro (1908, 1912) clearly show a specimen preserved with a collapsed caudal fin, which partly obscures its truncate shape. Although the description in Eigenmann (1918) repeats the informa- VOLUME 116, NUMBER 4 tion on the round caudal fin, the illustration in his pl. 51 fig. 2 more accurately repre- sents the actual truncate condition. The cau- dal fin has also been erroneously reported as round in J. eichorniarum by Eigenmann (1918), even though the original description (Miranda Ribeiro 1912) reports it as trun- cate, a shape confirmed in specimens ex- amined. The shape of the caudal fin indi- cates that the form described by Dahl (1960) as Pygidium metae guayaberensis (actually an /tuglanis), with a truncate fin, probably represents a separate species, and should be referred to as /. guayaberensis. The fin in /. metae is round. The species geographically closest to J. nebulosus is I. gracilior, from the lower Po- taro River in Guyana. The two species are evidently distinct, as demonstrated by the characters already discussed and listed in the diagnosis and by a few other, less ob- vious, traits which could not be confirmed in the holotype of /. gracilior, such as the more numerous procurrent caudal-fin rays (15/12). However, diagnosis of I. gracilior is problematic. It seems like the original de- scription in Eigenmann (1912), based on a single, very small specimen, may in part be misleading. The only statement about pig- mentation pattern is “‘all upper parts ob- scurely spotted’’. The subsequent illustra- tion in Eigenmann (1918) shows a uniform coloration. We suspect that J. gracilior is actually strongly spotted, and corresponds to the form illustrated in Le Bail et al. (2000) as I. amazonicus. The very long bar- bels reported in the holotype of /. gracilior are seen in some specimens from French Guyana here identified as that species, al- though that character is subject to some de- gree of intraspecific variation and the con- dition illustrated by Eigenmann is one ex- treme. The superficial similarity in color pat- terns in J. nebulosus and I. parahybae is not reflected in other morphological characters. The two species differ markedly in the number of vertebrae (cf. Diagnosis above), ribs (3 in J. nebulosus and 6 in I. parahy- 879 bae), and branchiostegal rays (7 vs. 8). Ituglanis parahybae also has the fewest pectoral- and pelvic-fin rays in the genus (5 and 4, respectively, with a vestigial fifth pelvic-fin ray sometimes present). The re- duction in pelvic fin-ray number may be part of a trend to reduce the pelvic fins, since some specimens of that species have been reported to lack those fins entirely (Costa & Bockmann, 1993). /tuglanis ne- bulosus has 6 pectoral- and 5 pelvic-fin rays, values that are plestomorphic for /tug- lanis. The vertebral count (35 or 36) in J. nebulosus is one of the lowest among /tug- lanis. The only other species with a similar value is /. passensis, reported to have 36 vertebrae by Fernandez & Bichuette (2002: 276). So far as known, all other /tuglanis have at least 39 vertebrae. Samples exam- ined of multiple cleared and stained speci- mens of /. proops and Ituglanis sp. 3 show that vertebral number is subject to only mi- nor variation (+2 vertebra), and that the value observed for /. nebulosus is therefore taxonomically significant, despite being based on only three specimens. Costa & Bockmann (1993:44) reported 34—38 total vertebrae in the generic diagnosis of /tug- lanis, but their method of counting verte- brae was not specified and comparisons are therefore difficult. /tuglanis gracilior is the species described from nearest to the type locality of J. nebulosus and it has 41 (value for holotype) to 43 vertebrae. Ttuglanis proops, from the Rio Ribeira basin, and /. eichorniarum, from the Para- guay basin, are similar in external aspect and pigmentation. However, there are ad- ditional differences that indicate that they are not conspecific. The most conspicuous differences are found in the interopercular patch of odontodes. /tuglanis proops has the largest interopercular patch of odon- todes in the genus. Most of that enlarge- ment occurs dorsolaterally, so that the dor- soposterior margin of the interopercular patch closely approaches the ventral margin of the opercular one. The integumentary folds of the opercular and interopercular 880 patches contact each other in preserved specimens. In all other species of /tuglanis, the two patches are clearly separated by a broad band of normal head integument. The condition in J. proops seems to be autapo- morphic, and clearly diagnoses it from all other species of /tuglanis, including J. ei- chorniarum. Ituglanis herberti may be a synonym of I. eichorniarum. The two come from the Paraguay basin. The original description of I. herberti by P. de Miranda Ribeiro (1940) considered the species as “evidently alied to Trichomycterus proops” |= I. proops}, but fails to mention /. eichorniarum, from the same basin as I. herberti. The original description reports the color pattern of J. herberti as a series of vertical stripes along the sides of the trunk. This characteristic is unique in /tuglanis and, if confirmed, would provide evidence of specific distinctiveness. However, the original description also notes that the vertical striped pattern was not ev- ident in the live fish. It seems likely that the dark vertical stripes were artifactual, result- ing from folds of integument following preservation and shrinkage. This is com- mon in large trichomycterids with thick in- tegument, where deep ridges of the folds forming after preservation may be mistaken for darkly-pigmented stripes. We have seen this effect clearly in one specimen refer- rable to I. herberti (MZUSP 2209), which is an old specimen where dark pigment is entirely faded. No additional taxonomically relevant data were found to justify separa- tion of J. herberti and I. eichorniarum. Phylogenetic relationships.—The place- ment of the new species in /tuglanis is un- problematical. The three synapomorphies for the genus provided in Costa & Bock- mann (1993) are present in J. nebulosus. The first is the vestigial condition of the posterior fontanel, reduced to a small round orifice on the posterior half of the supra- occipital (Fig. 4). The second is a deep notch on the mesial margin of the palatine. The condition in /. nebulosus is not as ex- treme as that illustrated for J. parahybae by PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON premaxilla bone sphenotic supraoccipital posterior pterotic fontanel L epioccipital supracleithrum Weberian complex and capsule Fig. 4. Jtuglanis nebulosus, paratype, MZUSP 69574, dorsal view of neurocranium. Scale bar = 1 mm. Costa & Bockmann (1993, fig. 6), but is still clearly present. The third synapomor- phy /. nebulosus shares with its congeners is an anteriorly-directed anterior portion of the sphenotic (Fig. 4). Elucidation of the relationships of J. ne- bulosus within Jtuglanis requires a phylo- genetic analysis, preferably associated with a species-level revision, and is beyond the scope of this paper. However, characters found in the course of this study support preliminary inferences about relationships. Among the species of Ituglanis available for skeletal examination, /tuglanis sp. 2 has the fewest pleural ribs, with only 2 pairs. The next lowest value, 3 pairs, is seen in J. nebulosus, I. amazonicus and I. gracilior. Remaining species for which data are avail- able, /. eichorniarum, I. parahybae, I. proops, I. passensis, Ituglanis sp. 1 and Itug- lanis sp. 3 have between 5 and 7 pairs. Re- duction in rib number is considered derived because basal trichomycterids have nine or more ribs. A reduction in rib number (seven or fewer) has been considered synapomorph- VOLUME 116, NUMBER 4 ic for /tuglanis plus a large subset of Tri- chomycteridae (see below). A more extreme reduction (three or fewer) is a state syna- pomorphic for a subset of the genus, includ- ing I. nebulosus, I. amazonicus, I. gracilior and /tuglanis sp. 2. Such rib reduction is ho- moplastically seen, in similar or more ex- treme degrees, also in the highly especilized subfamilies which constitute the sister group to Ituglanis (see below). A smaller clade including only /. nebu- losus, I. amazonicus, and Ituglanis sp. 2. further shares a peculiar widening of the mesethmoid shaft. In those species, the width of the mesethmoid shaft equals or surpasses the length of one of the meseth- moid cornua. This condition is markedly distinct from that in other species of /tug- lanis and remaining species of the subfam- ily Trichomycterinae, where the meseth- moid shaft is narrow, its width being mark- edly smaller than the length of each mes- ethmoid cornua. Another subgroup of Jtuglanis, including I. eichorniarum, I. parahybae, I. proops, Ttuglanis sp. 1 and Ituglanis sp. 3, share a peculiar lateral, or posterolateral, process on the anterior third of the antorbital (fron- to-lacrimal tendon bone of some authors). In /. proops, some specimens have the pro- cess on one side only. The presence of this process is a derived condition, because the antorbital is a simple rod in al! other rele- vant trichomycterine taxa. The characters discussed indicate that Ituglanis seems to comprise two main monophyletic groups. One includes north- ern South American species from the Am- azon and the Guyanas. The other includes southern forms from the Paranda/Paraguai, Ribeira de Iguape, Paraiba do Sul and Southeastern drainages in Brazil and Uru- guay, plus /. eichorniarum and Ituglanis sp. 3, which also occur in Southern Amazonian tributaries. This idea is just a first approxi- mation to an understanding of I/tuglanis phylogeny. Further study, on the basis of additional characters and broader taxonom- ic sampling, is necessary to formulate a de- 881 tailed hypothesis on the relationships among species of the genus. Costa & Bockmann (1993) hypothesized Ttuglanis as the sister group to a large clade including the subfamilies Tridentinae, Ste- gophilinae, Vandelliinae, Sarcoglanidinae and Glanapteryginae. This group shares a synapomorphic reduction in the number of pleural ribs (2 to 6 pairs). Jtuglanis nebu- losus, with 3 pleural ribs, conforms to the condition of this character in the clade. The same happens with a more inclusive group, including the multi-subfamilial clade and Ituglanis, plus the genus Scleronema. This clade relies on two characters: a reduction in size of the interopercular patch of odon- todes and the fine, elongated lateral arms of the urohyal. Again, both derived conditions are present in J. nebulosus. Thus, the dis- covery of the new species does not bring in any new character state conflicting with current understanding of the relationships of [tuglanis within Trichomycteridae. Acknowledgments We thank PY. Le Bail, EF Meunier and the late P. Planquette, for help during in- ventories in French Guiana. We are also grateful to Sandra Raredom (USNM) for preparing excellent digital radiographs of the type series of J. nebulosus, and to C. Moreira for help with digital photography. R. Vari was generous with various kinds of assistance during the editorial process. The manuscript benefitted from comments by Victor Springer and an anonymous review- er. Research funding for MP is provided by CNPq (Conselho Nacional de Desenvolvi- mento Cientifico e Tecnoldgico; Brazilian Federal Government) and FAPESP (Fun- dacgao de Amparo a Pesquisa do Estado de Sao Paulo, Sao Paulo State Government). Literature Cited Costa, W. J. E. M., & E A. Bockmann. 1993. Un no- veau genre néotropical de la famille des Tricho- mycteridae (Siluriformes: Loricarioidei).—Re- vue Francaise d’ Aquariologie 20:43—46. 882 Eigenmann, C. H. 1918. The Pygidiidae, a family of South American catfishes——Memoirs of the Carnegie Museum 7:259—-398. Fernandez, L., & M. E. Bichuette. 2002. A new cave dwelling species of Jtuglanis from the Sao Domingos karst, Central Brazil (Siluriformes: Trichomycteridae).—Ichthyological Explora- tions of Freshwaters 13:273—278. Le Bail, P. Y., P. Keith, & P. Planquette. 2000. Atlas des poissons d’eau douce de Guyane. Tome 2, fascicule IH: Siluriformes. Patrimoines Naturels (M.N.H.N./S.P.N.) 4311): 307 pp. de Pinna, M. C. C. 1989. A new sarcoglanidine catfish, phylogeny of its subfamily, and an appraisal of the phyletic status of the Trichomycterinae (Te- leostei, Trichomycteridae).—American Muse- um Novitates 2950:1—39. . 1992. A new subfamily of Trichomycteridae (Teleostei, Siluriformes), lower loricarioid rela- tionships and a discussion on the impact of ad- ditional taxa for phylogenetic analysis.—Zoo- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON logical Journal of the Linnean Society 106:175— WDD, . 1998. Phylogenetic relationships of neotropi- cal Siluriformes (Teleostei: Ostariophysi): his- torical overview and synthesis of hypotheses. Pp. 279-300 in L. Malabarba, R. E. Reis, R. P. Vari, Z. M. Lucena, and C.A.S. Lucena, eds., Phylogeny and classification of neotropical fish- es. Porto Alegre, Edipucrs. Song, J., & L. R. Parenti. 1995. Clearing and staining whole fish specimens for simultaneous demon- stration of bone, cartilage and nerves.—Copeia 1995:114—-118. Taylor, W. R., & G. C. Van Dyke. 1985. Revised pro- cedures for staining and clearing small fishes and other vertebrates for bone and cartilage study.—Cybium 9:107—109. Tchernavin, V. 1944. A revision of some Trichomyc- terinae based on material preserved in the Brit- ish Museum (Natural History).—Proceedings of the Zoological Society of London 114:234—275. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 116(4):883-—891. 2003. A new species of the hatchetfish genus Argyripnus (Stomiiformes: Sternoptychidae) from the Indo-Pacific Antony S. Harold and Kelly Lancaster Grice Marine Laboratory, College of Charleston, 205 Fort Johnson, Charleston, South Carolina 29412, U.S.A., email (ASH): harolda@cofc.edu Abstract.—The sternoptychid species Argyripnus brocki Struhsaker, origi- nally described from the Hawauan Islands, has been reported in the literature from various Indo-Pacific localities and to the west off east Africa. We have made univariate and multivariate comparisons, using sheared principal com- ponents analysis, of specimens from this broad area and found there to be evidence of two distinct species: Argyripnus brocki, endemic to the Hawaiian Islands and vicinity, and A. pharos n. sp., occurring in all other investigated areas. These two species are distinguished from the other four in the genus by their relatively low numbers of gill rakers, vertebrae, and certain photophores. Argyripnus pharos 1s distinguished from A. brocki by several features but most clearly by its VAV + ACA photophore cluster, which is dorsally inflected in an arch-like form and reaches a higher elevation on the flank. Fishes of the order Stomuformes are mainly meso- and bathypelagic fishes, found in all oceans but more diverse at tropical and subtropical latitudes. The re- lationships and composition of the order have been examined in a series of papers by Weitzman (1974), Ahlstrom et al. (1984), Harold and Weitzman (1996), and Harold (1998). Four families are currently recognized in the order: Gonostomatidae, Photichthyidae (also spelled Phosichthyi- dae), Stomiidae, and Sternoptychidae. The last comprise the hatchetfishes, to which Argyripnus belongs. Unlike the deep-bod- ied hatchetfish genera, Argyropelecus, Po- lyipnus and Sternoptyx (see Baird, 1971, and Harold, 1994), Argyripnus species are quite shallow-bodied and overall rather similar to gonostomatids (bristlemouthes). Unlike the many pelagic stomiiforms, Ar- gyripnus species are considered to be ben- thopelagic (Badcock and Merritt 1972), liv- ing in association with the benthic com- munity and geographically restricted to the continental slope and other features of pos- itive topographic relief. Grey (1961, 1964) recognized three Ar- gyripnus species: A. atlanticus Maul 1952, A. ephippiatus Gilbert and Cramer 1897, and A. iridescens McCulloch 1926. Argy- ripnus brocki Struhsaker 1973 and A. elec- tronus Parin 1992 were added subsequent- ly, bringing the total in the genus to five. Struhsaker (1973) reviewed species occur- ring in the Indo-Pacific region and de- scribed A. brocki from off the Hawaiian Is- lands. Low gill-raker counts and other fea- tures in combination were used to distin- guish it from the most similar congeneric species, A. ephippiatus and A. iridescens. Strusaker (1973:835) examined several non-Hawaiian specimens, some of which he tentatively ascribed to A. brocki, stating that the geographic range of the species ““may extend from Hawaii to the Indian Ocean.” We report on additional collections of “‘low count” Argyripnus from the Philippines and other areas of the Pacific and Indian oceans and describe the non-Hawaiian specimens aS a new species, Argyripnus pharos. In making comparisons with A. brocki we rely on standard morphometric and meristic tab- 884 ular and graphical presentations, as well as a sheared principal components analysis (PCA) (sensu Humphries et al. 1981, Book- stein et al. 1985) of a subset of the mor- phometric characters. Materials and Methods Specimen length is standard length (SL) in mm. Morphometric, meristic, photophore and other characters were determined fol- lowing the methods and terminology of Weitzman (1986) and Harold (1994). Ter- minology of the photophores is repeated here for convenience, and to describe and illustrate certain photophore characters not covered by previously published works. The abbreviation for each photophore or photophore series (‘‘cluster,’’ in cases where associated photophores joined in a common organ) and its anatomical location, as defined by Weitzman (1986), and with modifications of the AC series terminology according to Harold (1994), are as follows: ACB (cluster dorsal to central portion of anal fin), ACC (cluster along ventral sur- face of caudal peduncle), BR (cluster on branchiostegal membranes), IP (cluster on isthmus), OP 1 (single photophore associ- ated with anterior base of preopercle), OP 2 (single photophore near anterodorsal mar- gin of opercle, posterior to center of eye), OP 3 (single very large photophore asso- ciated with subopercle), ORB (single pho- tophore anteroventral to eye), OV (dorsal abdominal series/cluster posterior to pecto- ral fin base), PV (ventral abdominal cluster between bases of pectoral and pelvic fins), and VAV + ACA (combined VAV and ACA clusters of other sternoptychids in a long organ extending from above pelvic-fin base to vertical through about seventh to tenth anal-fin ray; VAV/AC, of Badcock and Merrett, 1972:fig.1). Counts of fin rays and numbers of photophores in clusters are the total number of elements in all cases. Elevation of the VAV + ACA photophore cluster is a character developed specifically for this study, determined by taking the dis- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tance from the dark pigment border of the dorsalmost VAV + ACA photophore to the base of the immediately ventral anal-fin ray. A sheared principal components analysis was performed on a subset of morphometric characters for which the greatest number of specimens had values for all characters. Humphries et al. (1981) discussed compu- tational aspects of the procedure. The pro- gram “‘Shear,”’ written by N. MacLeod, was used for the analysis (McLeod 1990); this software was published as part of the Pro- ceedings of the Michigan Morphometrics Workshop (Rohlf & Bookstein 1990). One of the difficulties with systematic study of deep-sea fishes is their delicate na- ture and the resultant poor condition of trawl-captured material. Specimens may lack structures, rendering them of little val- ue for analyses such as PCA, in which a complete set of data for each is specimen is desirable. Damaged specimens were ad- equate for making counts and some mea- surements included in the description but were excluded from multivariate analysis. A complete truss network was not utilized, but instead selected homologous point-to- point measurements were used that provide critical shape comparisons. Sources of material for this project are listed as institutional acronyms and associ- ated catalog numbers for A. pharos in the account and for other Argyripnus species in Appendix I. Institutional abbreviations are as listed in Leviton et al. (1985). Argyripnus pharos new species Fig. | Argyripnus sp.—Grey, 1961:470—474, ta- bles 1, 2, 3 (USNM 135402, Philip- pines).—Grey, 1964:207 (off northern Mindinao, Philippines; Key to Spe- cies).—Bourret, 1985:57-58 (USNM 135402). Argyripnus brocki.—Struhsaker, 1973:835 (part, USNM 207984, Indian Ocean).— Parin, 1992:136 (part, northwestern Aus- tralia, northwestern Madagascar).—Har- VOLUME 116, NUMBER 4 885 Bice: old, 1999:1901 (part, western Central Pa- cific). [not Argyripnus brocki Struhsaker, 1973] Holotype.—BMNH 1986.9.22:41, (1, 78.9 mm), Saleh Bay, South Java, approx. 8°30'S, 118°00’E, field no. TGT 1676, coll. T. Gloerfelt-Tarp (no other data available). Paratypes.—AMS 1.24338002 (2, 55.7— 79.3 mm), Saleh Bay, South Java, 8°30’S, 117°46’E, 150—280 m, coll. T. Gloerfelt- Tarp, Jul 1981. BMNH 1986.9.22:42—45 (4, 59.4—76.6 mm), collected with holotype. Non-type material.—AMS 1.31174005 (11, 58.6—72.3 mm), Shark Bay, Western urstraliae G47 3 —49 eS ll 384 SS ek (O—) 285 m, coll. J. Paxton, 30 Jan 1991. CSIRO H2169-01 (1, 57.8 mm), E of Fraser Island, Queensland, Australia, 24°44.2— Adal S» la4-12 5-14.88. (O—) 492 m, R/V. Soela, Sta. S00685/7, 17 Nov 1985. IOAN uncataloged (2, ~45—51.4 mm), western In- dian Ocean, 12°31’S, 48°17’E, (O—) 380 m, R/V Vityaz, Cr. 17, Sta. 2603, 12 Nov 1988. MNHN 1979-98 (1, 67.9 mm), Philippines, 13°49’N, 120°04’E, (O—) 200 m, R/V MU- SORSTOM I, Sta. 51CP4, Mar 1976. MNHN 1979-99 (1, 62.4 mm), Philippines, 14°00'N, 120°16’E, (O—) 195 m, R/V MU- SORSTOM I, Sta. 64CP4, Mar 1976. USNM 135402 (1, 78.4 mm), N of Mindi- nao, Philippines, 8°48.5’N, 123°35.5E, (O—) 366 m, R/V Albatross, Sta. D5542, 20 Aug 1909. USNM 207984 (1, 47.8 mm), Indian Ocean, 06°51’'S, 39°54’E, (O—) 100 m, R/V Anton Bruun, Cr. 9, Sta. 422, 19 Nov 1964. YPM 10015 (12, 52.3-80.8 mm), Albay Argyripnus pharos, holotype, BMNH 1986.9.22.41, 78.9 mm SL. Gulf, Philippines, 13°10'N, 124°00’E, 22 Sep 1995: Diagnosis.—Characters that in combina- tion serve to distinguish Argyripnus phar- os: gill rakers 15—16, usually 16, with typ- ically 12 on the ventral limb; VAV + ACA photophores 13—18; total number of verte- brae 42—44, usually 42 or 43; dorsal-fin rays 9-12; VAV + ACA photophore ele- vation 3.7—-6.4% SL. No uniquely derived characters known. Description.—Moderate-sized species, maximum observed adult body length 80.8 mm SL (Table 1; Philippines, YPM 10015) with shallow body profile, body depth 24.7-31.9% SL. Head large (up to 35.1% SL), lateral profile forming a slightly round- ed angle of about 90 degrees, with its apex at dentary symphysis. Mouth large, oblique. Scales deciduous, none remaining in any Specimens examined; based upon observa- tions of scale pockets, scales quite large and dorsoventrally elongate. Adipose fin pre- sent. Maximum body depth located at ver- tical through occiput, body nearly as deep at dorsal-fin origin. Dorsal-fin origin ap- proximately at vertical through base of most lateral pelvic-fin ray, and located at about mid-body (predorsal length 47.0— 55.2% SL, prepelvic length 44.1-53.5% SL). Body profile gently tapered posterior to dorsal-fin origin to shallowest point at about middle of caudal peduncle; caudal peduncle slightly flared posteriorly. Orbit circular to slightly dorsoventrally elongate, and large, up to 47.1% of head 886 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Morphometric characters expressed as percentages of head length (orbit diameter and snout length) or standard length (all others) and meristic character values for Argyripnus pharos holotype (BMNH 1986.9.22: 41), and other material, and Argyripnus brocki holotype (USNM 207653), and other material combined. Argyripnus pharos Holotype Other Standard length 78.9 47.8—80.8 Orbit diameter 37 37.7-47.1 Snout length DAL) 21.6—26.0 Head length 32.6 29.0—-35.1 Body depth 27.6 24.7-31.9 Caudal peduncle length 16.2 12.8—-18.1 Caudal peduncle depth 8.3 8.1-11.3 Dorsal fin length IB 12.1-18.3 Anal fin length Deo) 23.7—35.1 Preanal length GIES 52.2-62.4 Predorsal length 48.9 47.0—55.2 Prepelvic length SO) 44.1—53.5 Postdorsal length 54.5 52.8—62.5 Postanal length 42.9 39.1-47.6 Dorsal-pelvic length 28.2 24.4—30.6 OP 3 height 30) 1.8—3.9 VAV + ACA elevation SO 3.7-6.4 VAV + ACA — ACB length Boll 1.8—8.0 ACB — ACC length 6.1 6.4—-11.2 ACC length 9.6 4.7-8.7 Dorsal-fin rays ll 9-12 Anal-fin rays 23 20-24 Pectoral-fin rays 15 14-17 Pelvic-fin rays 7 6-7 Dorsal gill rakers 4 3—4 Ventral gill rakers 12 11-12 Total gill rakers 16 15-16 Total vertebrae 42 42-44 VAV + ACA photophores 15 13-18 ACB photophores 5 5 ACC photophores 11 9-12 length. No spines or serrae present in asso- ciation with elements of the head skeleton or elsewhere. Premaxillary teeth uniserial, differing in size and shape, ranging from, medially, three widely spaced conical to re- curved large teeth (about | mm in length in 78.9 mm SL holotype) interspersed with much shorter teeth; laterally, a short comb- like patch of minute teeth (each about one tenth the length of the largest premaxillary teeth) along extremity of alveolar process. Maxillary teeth uniserial, subequal, conical to recurved, up to about one half the length of the largest premaxillary teeth, variable in number: 17 in series in 59.4 mm SL para- Argyripnus brocki Mean N Holotype Other Mean N 65.8 35) ills) 48.8—-89.0 69.6 80 41.1 33 39.8 38.6—-46.8 42.2 64 DEM By AM) 21.1—29.0 24.2 65 32,1 33 30.8 27.9-33.8 30.8 68 28.9 35) Zi MS),2=3) lO 28.3 We 14.2 34 BO 12.2-17.4 14.7 67 9.4 34 oe 7.J-\1.6 2 TZ 14.1 35 13.5 12.0-17.1 14.3 val BQ) 26 DEO 25.1—33.3 28.4 65 SZ 3)1I Dies 53.6—62.8 58.4 65 49.4 32 49.4 44.4-52.4 48.7 65 48.2 30 49.5 45.2—54.6 49.4 58 IDs 33 S3o 7) 52.0—60.0 56.3 60 43.5 33 40.4 41.0-47.8 43.9 63 28.4 33 27.4 24.3—34.6 PS jee) 66 2.8 Sy) 4.3 2.9-6.0 4.7 33 Do 34 36 2.84.9 3.8 56 4.3 34 4.4 2.9-7.4 50) 56 8.8 34 9.0 7.8-14.3 10.0 67 7.0 33) OS 8.6—-12.7 10.6 64 10.9 35 11 10-13 Lit, Il 15 22.1 op Z| 18—24 Dee WS 526 39) 16 14-16 14.9 VY 6.9 3 7 6-7 6.9 66 4.0 ES) 4 3-4 39) 14 ibe BD 12 12-13 12, 1h 14 UDI 2S) 16 15-17 16.0 14 42.8 21 4] 41-43 41.9 40 Wee) 34 5) 13-18 15.4 54 5.0 35) S > 50 76 10.2 a2 10 10.1 66 type, and 29 in 78.9 mm SL holotype. Den- tary teeth biserial, inner row comprising nu- merous (more than 20) minute, recurved teeth increasing in length toward the jaw an- gle, outer row three or four widely-spaced larged recurved teeth similar in size and shape to largest teeth of premaxilla. One or two large, recurved vomerine teeth located laterally on extremity of each anterior wing- like process; no teeth present on posterior shaft. Four or five minute, conical palatine teeth in a longitudinal row. Mesopterygoid teeth absent. Pseudobranch present and well- developed. Photophore counts as for genus (Ahls- VOLUME 116, NUMBER 4 trom et al. 1984), except VAV + ACA and ACC: AC group combined with VAV, from anterior to posterior, VAV + ACA (13-18), forming a distinctive arch in central portion of cluster directly dorsal to third or fourth anal-fin ray. ACB (5). ACC (9-12), begin- ning adjacent to second or third last anal- fin ray and continuing posteriorly to near anteriormost ventral procurrent caudal-fin ray. BR (6). IP (6). OP 3. ORB 1. OV (6) ae IE IP (I): Color in preservative.—Dark brown chro- matophores on head in association with iris, photophores (BR, IP, OP, and ORB), poste- rior portion of cranium, areas immediately dorsal, ventral, and posterior to orbit includ- ing dorsal portion of opercle, symphysial portions of premaxilla and dentary, anterior margin of medial one half of maxilla, and entire medial one half of posterior supra- maxilla. Reflective guanine pigment associ- ated with photophores, iris, and most of gill cover. Dorsal portion of body above lateral septum pigmented with dark brown chro- matophores, with greatest concentration be- tween pectoral girdle and dorsal-fin origin, in a patch on caudal peduncle, delineating myosepta, and forming predorsal stripe be- tween occiput and dorsal-fin origin. Other areas of concentration include patches of dark pigment at bases of dorsal- and pro- current-caudal rays, and covering main body of all photophores. Reflective guanine pig- ment on body on lateral surface of pectoral- fin base associated with posterior IP photo- phore reflector; guanine pigment also asso- ciated with other photophores, especially ventrolateral area of the abdomen lining OV and PV organs and region of body wall be- tween these clusters. Fin rays largely unpig- mented. Pectoral fins with lines of small dark brown chromatophores along basal por- tions of fin rays and widely separated spots of dark pigment distally. Longest three or four rays of both lobes of caudal fin darkly pigmented along approximately one quarter to one third of their lengths. Etymology.—The specific name pharos is based on the feminine Greek noun phar- 887 os, meaning lighthouse or beacon, with ref- erence to the dorsally displaced elements of the VAV + ACA photophore cluster. Distribution.—West Pacific, from Phil- ippines southward through Indonesian Ar- chipelago and Coral Sea, and western In- dian Ocean off Africa. Ecology.—In almost all cases Argyripnus species, including A. pharos, have been captured at mesopelagic depths (100—492 maximum depth of capture for A. pharos) but near the sea floor. Captures have most typically been made by bottom trawl, al- though Badcock and Merrett (1972) report- ed collections of A. atlanticus made by a rectangular midwater trawl (RMT8/5) when it was towed near (within 40 m) or on the bottom. The specimens caught were juve- niles and subadults (26.0—53.0 mm SL), suggesting the existence of a population of A. atlanticus near the sea floor. Available collection data for A. pharos indicate that this species likely occurs typically in this near-bottom, benthopelagic habitat. Certain other sternoptychids, such as Sonoda and Polyipnus, are also reported to have such affinity with the sea floor. Remarks.—A multivariate statistical ap- proach, sheared principal components anal- ysis, was used to quantify aspects of shape variation in the complex of forms here rec- ognized as A. brocki and the new species A. pharos. Characters used in the analysis are listed in Table 2. The principal com- ponents analysis extracted three compo- nents, with their eigenvalue and total vari- ance, respectively, as follows: PC 1 (0.049, 62.262), PC 2 (0.020, 23.604), and PC 3 (0.005, 5.964). The cumulative percent of variance accounted for by these principal components is 91.831. The size vector, PC 1, is not important in addressing the issue of interspecific discrimination in this prob- lem. Variable loadings for PC | and two sheared principal components (PC 2 and PC 3), the shape components, for each speci- men are reported in Table 2, with the five most important variables in terms of vari- ance explained indicated. Analyzed speci- 888 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Sheared principal component loadings for eighteen morphometric characters from analysis of Ar- gyripnus pharos and A. brocki, 27 and 39 specimens, respectively. Five highest loadings for each principal component followed by *. Character Standard length Head length Orbit diameter Snout length Caudal-peduncle length Caudal-peduncle depth Dorsal-fin length Anal-fin length Preanal length 10 Predorsal length 11 Prepelvic length 12 Postdorsal length 13 Postanal length 14 Dorsal-fin origin to pelvic-fin origin 15 ACC photophore cluster length 16 Posterior ACB to anterior ACC photophore ONAN BWN No) 17 Posterior VAV + ACA to anterior ACB photophore 18 VAV + ACA photophore cluster elevation mens are plotted with respect to sheared PC 2 and PC 3 scores in Fig. 2. The character VAV + ACA photophore elevation was found to be an important discriminating feature in PC space and is therefore plotted separately as a bivariate with standard length (Fig. 3). Argyripnus pharos shares with the other five species in the genus a united series of VAV + ACA photophores (18—32 photo- phores overall for the genus), dorsal fin with 10 to 14 rays and originating near or anterior of vertical through anal-fin origin (Weitzman 1974, 1986). The six species are separated into two main groups of morpho- logically similar species (see key to species, Grey 1964:207): (1) those with the dorsal fin originating nearly directly above the or- igin of the anal fin and with numerous (28— 32) VAV + ACA photophores (A. atlanti- cus and A. electronus), and (2) those with the dorsal fin originating well anterior of a vertical through the anal-fin origin, and fewer (13-21) VAV + ACA photophores (A. brocki, A. ephippiatus, A. iridescens, and A. pharos). The new species, A. pharos, and A. brocki are clearly most similar, based PC 1 (size) PC 2 (sheared) PC 3 (sheared) OZaI959 —0.026222 0.064297 0.240692 0.015219 —0.027908 03382285 —0.045382 0.010296 O2SsS 0m —0.008259 0.010706 0.223821 0.017732 0.216135* 0.205857 —0.097824 —0.035673 0.253762 =O) JENA NO 0.033499 0.191146 —0.129438* —0.051035 0.293429* —0.020121 0.062156 0.26249 1* 0.009478 0.054128 0.293861* —0.000413 0.084425 0.232723 —0.058346 0.078760 0.215870 —0.088821 0.027782 0.260888* —0.0005 12 0.083734 0.242147 0.064465 0.219743* 0.183837 el 0 Sol It YP 0.267798* 0.112730 —0.782794* —0.509556* 0.224150 0.445263* —0.740109* upon the presence of relatively few gill rak- ers (15-17, usually 16, as compared with 18—22 overall in A. ephippiatus and A. iri- descens), fewer VAV + ACA photophores (13-18, as compared with 19-21), and low vertebral counts (41—44 total vertebrae, as compared with 44—53). Meristic features that, in part, distinguish A. pharos from A. brocki include the presence of slightly more vertebrae (42—44 in A. pharos, as compared with 41—43 in A. brocki), and fewer dorsal- fin rays (9-12, as compared with 10-13, respectively). In terms of morphometric features, A. pharos compared with A. brocki has more highly elevated VAV + ACA photophores (elevation 3.7—6.4% SL, as compared with 2.8—5.9, respectively, a larg- er head (head length 29.0—35.1% SL, as compared with 27.9—33.8), and the dorsal- fin origin located slightly posteriorly (pre- dorsal length 47.0—55.2% SL, as compared with 44.4—52.4). Grey (1961) reported on an Argyripnus specimen (USNM 135402) from off Min- - dinao, Philippines, that she was unable to place to species level, and consequently re- ferred to Argyripnus sp. The specimen, with VOLUME 116, NUMBER 4 0.900 0.800 = ical Op Fa 0.700 a = a of ad @ - = es @ © L “an O © 0.600 ae @) = = 68 -) = fe) © 0.500 - = O 7) O 0.400 0.300 889 a i O 29 oO oO io C OO oe AO 9 O @ brocki non-types © brocki holotype @ brocki paratypes O pharos non-types ¢@ pharos holotype C pharos paratypes | -1.300 -1.200 -1.100 -1.000 -0.900 -0.800 -0.700 sheared PC 2 Ris, 2. Plot of principal component scores for Argyripnus pharos and A. brocki with respect to the first two sheared (shape) principal components, PC 2 and PC 3. The size axis, PC 1, is not illustrated. that same designation, was later incorporat- ed into a Key to Species (Grey 1964); lack of additional material apparently prevented Grey from describing what was evidently a representative of an undescribed Argyrip- nus species. Neither was Struhsaker (1973) able to assign USNM 135402 to a species, although the single specimen was reported to be very similar to his new species, A. brocki, from the Hawaiian Islands. He did, however, assign another geographically re- moved specimen (USNM 207984, western Indian Ocean) to A. brocki. These speci- mens were examined and placed herein well within the ranges of variation of A. pharos. Other Indian Ocean material ex- amined, from off the east coast of Africa (IOAN uncataloged, 12°31S, 48°17E) is as- signed to A. pharos. In conclusion, A. brocki is an endemic of the Hawaiian Islands vicinity. The similar, and likely phylogenetically related species, A. pharos, occurs in the Indo-Australian re- gion, from the Coral Sea northwards to the Philippines, and in the western Indian Ocean off Africa. There are, however, no records of A. pharos from the central Indian Ocean. This situation could be due to a pau- city of slope habitat at appropriate latitude and, concomitantly, collections of bentho- pelagic fishes in the region, although it is possible the species has disjunct western and eastern populations. Acknowledgments We thank M. McGrouther and J. R. Pax- ton (AMS), J. Maclaine and A. C. Gill PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON O Oo} 0 Or 2 mpeg © oO a © 80.0 70.0 90.0 100.0 standard length (mm) 890 6.0 @ brocki non-types © brocki holotype oe @ brocki paratypes O pharos non-types : 40 @ pharos holotype = 0 pharos paratypes O ) © > YB 3.0 @® ee Fa i ite ap ed - ey Hi rhe Appeal V ety far sk 5 C : P s Oa? 4 rt aa te . 4 debt eco he i iat wekss Ge 2 a, ie: ( Hives Sih hy pues wa nee tie id if ia y ee os ——— a i. mye © vite **Index to New Taxa’’ for volumes 115 and 116 will appear in volume 117, number 1. INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Richard v. Sternberg, NCBI—-GenBank, Building 45, Room 6An. 18D-30, National Institutes of Health, Bethesda, MD 20892-6510. (Do not submit manu- scripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in “GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON,” a supplement to Volume 114, number 4, December 2001. Authors are encouraged to consult this article before manuscript preparation. Copies of the article are avail- able from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page charges. The Society is a non-profit organization with limited funds, depending on financial resources and upon author request, it may subsidize a limited number of pages per vol- ume at the discretion of the editor and treasurer. Payment of full costs is encouraged and will facilitate speedy publication. Please note: Page charges are not automatically waived with membership. Authors requesting a partial waiver of page charges must justify need by emailing the editor and treasurer before proofs are returned and on the reprint request form received with page proofs. Please include email address on all correspondence. Costs.—Printed pages @ $65.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. Front cover—from this issue, p. 885. CONTENTS Diagnoses of hybrid hummingbirds (Aves: Trochilidae). 12. Amazilia bangsi Ridgway, 1910, is an intrageneric hybrid, Amazilia tzacatl x Amaczilia rutila Gary R. Graves A new. small-eared shrew of the Crypftotis nigrescens-group from Colombia (Mammalia: Soricomorpha: Soricidae) Neal Woodman A new species of the catfish genus J/tuglanis from French Guyana (Osteichthyes: Siluriformes: Trichomycteridae) Mario de Pinna and Philippe Keith A new species of the hatchetfish genus Argyripnus (Stomiiformes: Sternoptychidae) from the Indo- Pacific Antony S. Harold and Kelly Lancaster New findings and an,overview of the oligochaetous Clitellata (Annelida) of the North Atlantic deep sea a Christer Erséus and Emilia Rota On two new genera of pea crabs parasitic in holothurians (Crustacea: Decapoda: Brachyura: Pinnotheridae) from the Indo-West Pacific, with notes on allied genera Peter K. L. Ng and Raymond B. Manning A new crayfish of ne genus Cambarus Erichson, 1846 (Decapoda: Cambaridae) from the Cape Fear River basin in the Sandhills of North Carolina John E. Cooper and David G. Cooper A new species of freshwater anomuran crab of the genus Aegla Leach, 1821 (Crustacea: Decapoda: Aeglidae) from the Nahuelbuta Coastal Range, Chile Carlos G. Jara, Marcos Pérez-Losada, and Keith A. Crandall A new genus and new species of the family Paguridae (Crustacea: Decapoda: Anomura), from Hachijo- jima Island, Japan, with a list of hermit crab species found in the same collection sites Masayuki Osawa and Junji Okuno A new species of the hermit crab genus Diogenes (Decapoda: Anomura: Paguroidea: Diogenidae) from Pakistan, with a comparative diagnosis of D. guttatus Henderson, 1888 Feroz A. Siddiqui and Patsy A. McLaughlin A new species of Stenetrium Haswell, 1881 (Crustacea: Peracarida: Isopoda: Asellota), from Navassa Island, Northern Caribbean Sea Joel W. Martin, Richard W. Heard, and Regina Wetzer Eocuma petrescui, a new species of bodotriid cumacean (Crustacea: Peracarida) from Malaysia Charmaine K. Patel, Pilar A. Haye, and Irv Kornfield Metandania tordi, a new stegocephalid (Crustacea: Peracarida: Amphipoda) species from the Southern Ocean Jgrgen Berge and Wim Vader A new species of Stygonitocrella (Crustacea: Copepoda: Ameiridae), the first report of the genus in North America Janet W. Reid, Gary W. Hunt, and Emily H. Stanley Four new solitary entoprocts (Entoprocta: Loxosomatidae) from Okinawa Island, The Ryukyu Archipelago, Japan Tohru Iseto Contents, Volume 116 Table of Contents and Abstracts available online: www.apt.allenpress.com/aptonline IAN INSTITUTION LIBRARIES 847 853 873 883 892 901 920 YS 943 502 i i { 7 a fa. 0% tae ens ; | Avy bi, “at AY yy , * i) ‘i we WA) Dor een SMITHSONIAN INSTITUTION LIBRARIES AYN 9088 01482 3579 Bese iy ae e Aer Ral shee cera a See Wea AUS BS fa PeP aha Re hed Wale Ace eha he st hx ben, Saar Patella he Bre RR Re Wake ia hat in Fei Rhy Gish ct ie BoD we an 2 om yy