Yah i Ava bee as pT el ee : wadatteadickin§ WePacaanag. seni Odi g yyw Healy een oe , ei iN h Sree aly ‘ gy PAREN) oot r VOL DNOre , yuan NYS aM aN tA ‘ Y Aygt desde ibs Y ‘ " I PAN AL ‘ ! hana ty wile fhe wt : nels ae } t 4 ‘ Aine c 4 Arar eee Me AY Ae me Hata fet ‘ in yoo af ot Srv » Pikes te Ved gy wv Nl h yee a 34 pigs Te at aur age Soar vo? 4 eS Pp ies 7 5% tae , EAS UID: < Wore. is : kate! Yow aee Apa brasryn i pegiase is" oy peste elt ant a BOE usnet aed Fe rita Bed Ae, poate aD Vaceth Fe es ea Orr as a i _ z s = hi 4117 LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSNI~ NVINOSHLINS ~S314¥ fo) = pes = ow = a =) hs > > a = = = aA = ces = : 2 = B 7 o 7 = ; 7) = o . Z TION NOILNLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES SMITHSONIAN _ INSTI ” Zz 127) = wes ” z= = i = < NS = oT eh Ss z = SEWN F z SN » 2 E 2 F Z = WA Se ai 2 ; a ee = ie uit {SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOSHLIWS S31 uv : = pup? = em 2 SR Ee e XQ : > Ye VF: D> = Se SQ E E des 8 = cid \S i= a = UY fe ai | 3 7) ow 2 Sm 2 mn Cues UIT LIBRARIES, SMITHSONIAN INSTITUTION NOLLMLILSNI_NVINOSHLINS S314¥¥ ee 2 ke = ue = : < = 4 Za : x ws Noi 6 NO? 225 : Vi 7. ~ 30 » Oo 2? ee SNS 6 > Se ro) = WY sg E 2 i: Z = . = = S Ss = = Sl we eae 3 : ITION NOILNLILSNI_ NVINOSHLINS saiuvyag I1_LIBRARIES SMITHSONIAN _ m1 Zz tu z he Z = g g g 2 a : = oa re a fe = oc = Pe Cc roa) = oO = mo.” = _ a Ydit LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYY! c = a = ~ = o = ‘oO = o ° a = a 5 ae 5 2 eS 5 E a E a E ° it = a - m 2 GA m 2 m Zz a= = w s — TION, NOLLALILSNI_NVINOSHLINS SaIuvu a7 SMITHSONIAN _ INSTITL “\ = ass eS = < = z x Ss = z = = =! z SX SoZ. 2 zy 2B a 5 Ws » 2 = zi e Zz =o » See agen 2 a _ 3 B Wd LIBRARI ES SMITHSONIAN _ INSTITUTION NOILALILSNI_NVINOSHIIWS Sa 1Yyvi = Ww | = uw 2 A w g : te : ZX ar =r GS WSS Ss < 5 ec < c- RR WN < mm” OS = 5 S 2 ro} ie 2 FF if Zz a 2 a a JTION _NOILALILSNI _NVINOSHLIWS 2°34 iyvag Whee LIBRARI ES_ SMITHSONIAN _INSTIFU "5 GSM) 2 5 GAM) 2 GM) 5 Gey 2 1 RE Ue ELE ww Snows ar aS . te Le! S J RNs ee ie _ INSTITUTION NOILNLILSNI NVINOSHLINS S31u¥vudit 2, NOI. NO! NOI. <1ES SMITHSONIAN LIBRAI na 2: J INSTITUTION INSTITUTION INSTITUTION Ye @: S S3INYVHUSIT LIB ILSNI_ NVINOSHLINS S3INWYSIT_LIBRARIES SMITHSONIAN NOILAL 20" & BN Y SMITHSONIAN NVINOSHLIN SMITHSONIAN NYINOSHLINS Saluvuaiq NYINOSHLINS S3lYvuai7_ LIB LIBRA’ 21ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYaIT z aS z (@) 0 | Ze Len = 7) wn = (ep) = (7p) > us a uw 7 w a < }; pa < Piss < va =” E E = 3 : = 3 = 5 mo 5 a z J z ay z S31uvVusiIT LIBRARIES NOILAL : z : 3 : : > \ NSE > i= 22> = a] SS) x — Bs) — Dit 0) a a Ww ee - es = w oe = wo = no = R1ES SMITHSONIAN INSTITUTION NVINOSHLINS S31uVvVudiT LIBRAI 4 z = ip, = | < = re) ac fé O a6 \ \ 2 2 3 Ye? : Ss 2 F 2 ae z \ ws ay gs We = = \ 4ISNI_NVINOSHLINS S3Z1YVYUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNL NOILNLILSNI LIBRARIES NOILNLILSNI RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Sa31uvUugi7 LIBRA INSTITUTION ILSNI_NVINOSHLINS S3IYVYEIT_ LIBRARIES INSTITUTION NOILNLILSNI saiuvyugi INSTITUTION NOILNL c) tt b RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Saluvagty py NVINOSHLINWS NVINOSHLINS S3!IYvVudItT LIBRARIES SMITHSONIAN NVINOSHLINS S3iuvVudIT LIBRARIES —~ SMITHSONIAN NVINOSHLIWS SMITHSONIAN SMITHSONIAN LIBRA ES 1LSNI NOILAL Uy, T% oe ie) a > Pa a Saluvuds SMITHSONIAN €%, LIBRARIES INSTITUTION TUTION NOILNLILSNI VUdSIT LIBRARIES PRD 2 2 oO nay TUTION NOILNLILSNI VU¥dil LIBRARIES TUTION NOILALILSN WUGITLIBRARI Anubis Mir Hi : we a's Fi \ <4 Ay Les THE BIOLOGICAL SOCIETY OF WASHINGTON 1983-1984 Officers President: Donald R. Davis Secretary: Gordon L. Hendler Vice President: Austin B. Williams Treasurer: Leslie W. Knapp Elected Council J. Laurens Barnard Maureen E. Downey Frederick M. Bayer Louis S. Kornicker Isabel C. Canet Storrs L. Olson Custodian of Publications: David L. Pawson PROCEEDINGS Editor: Brian Kensley Associate Editors Classical Languages: George C. Steyskal Invertebrates: Thomas E. Bowman Plants: David B. Lellinger Vertebrates: Richard P. Vari Insects: Robert D. Gordon Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $15.00 include subscription to the Proceedings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $18.00 within the U.S.A., $23.00 elsewhere. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership and subscriptions should be sent to The Treasurer, Biological Society of Washington, National Museum of Natural History, Smithsonian Instutution, Washington, D.C. 20560. Manuscripts, corrected proofs, editorial questions should be sent to the Editor, Biological Society of Washington, National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. 20560. Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Second class postage paid at Washington, D.C., and additional mailing office. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 473-512 A REVISION OF THE MILLIPED GENUS DICELLARIUS WITH A REVALIDATION OF THE GENUS THRINAXORIA (POLYDESMIDA: XYSTODESMIDAE) Rowland M. Shelley Abstract.— The milliped genus Dicellarius consists of the valid species okefeno- kensis (Chamberlin), atlanta (Chamberlin), sternolobus Loomis, bimaculatus (McNeill), and talapoosa (Chamberlin), the last two of which are divided into three and two subspecies, respectively. The genus 7hrinaxoria is revived to ac- commodate /ampra (Chamberlin) and bifida (Wood). The other two species, dela and /eiacantha, both authored by Chamberlin, are relegated to the synonymy of D. bimaculatus fictus (Chamberlin). The seven species are distinguished chiefly by details of the male gonopods, particularly the configurations of the distal elements, but the enlarged pregonopodal sternal lobes distinguish sternolobus. Females can also be determined to genus because the two species of Thrinaxoria have a receptacle or remnant thereof on the cyphopods, whereas Dicellarius lacks this structure. Dicellarius is centered in the Gulf Coastal Plain and spreads inland into the southern extremities of the Blue Ridge, Ridge and Valley, and Applachian Plateau Provinces. Thrinaxoria occupies a more northern and western area, rang- ing from northwestern Louisiana to southwestern North Carolina, and extending southward into the Coastal Plains of Georgia and Alabama. The two species of Thrinaxoria are allopatric and separated by over 100 miles. However, specific ranges overlap to a high degree in Dicellarius, and only okefenokensis is allopatric. Dicellarius consists of two lineages, one leading to taiapoosa and atlanta, and the other to bimaculatus, okefenokensis, and sternolobus. Relationships between the three genera in the tribe Pachydesmini are presently unresolved. A new diagnosis is presented for the tribe to reflect improved knowledge of Dicellarius and res- urrection of Thrinaxoria. The Gulf Coastal Plain of the southeastern United States is the center of abun- dance of the milliped tribe Pachydesmini. Two of the component genera, Pachy- desmus and Dicellarius, are the dominant xystodesmids in lowland areas of Mis- sissippi, Alabama, the Florida panhandle, and southern Georgia. Pachydesmus has traversed the Mississippi and Tennessee Rivers occurring in northern Loui- siana and central Tennessee, but these waterways are distributional barriers for Dicellarius. Pachydesmus contains the largest American polydesmoids, which grow to over four inches in length, but the forms of Dicellarius are only a couple of inches long and equivalent in size to most apheloriine diplopods, though much stiffer. The two taxa possess common anatomical features whose significance was first recognized by Hoffman (1958), and in 1979 he formally united them in the tribe Pachydesmini. The taxonomy of Pachydesmus is stable. It was revised by Hoffman (1958), and Shelley and Filka (1979) reported new distributional records and gonopodal 474 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON variants of P. crassicutis incursus Chamberlin. The same cannot be said of Di- cellarius, however. Its taxonomy is highly confused, and specific names cannot be assigned with any degree of confidence. As with most Nearctic diplopod genera, the existing literature on Dicellarius consists of short, vague descriptive accounts accompanied by meaningless illustrations. None of its species has been adequately characterized, and only one, sternolobus Loomis, was assigned to Dicellarius in its original proposal. The generic validation, moreover, was a single sentence in a paper on other millipeds, so Dicellarius has not received a word of diagnosis. Thus, a revision of this genus is needed to stabilize its nomenclature and redescribe the component species, so that the millipeds can be used in field and experimental research in other biological disciplines. One of the first observations engendered by this study is that Dicellarius, as envisioned by Hoffman (1979), is heterogeneous. The Louisiana species formerly in Thrinaxoria, D. lampra (Chamberlin), is unique in having a sharp caudal bend on the gonopodal acropodite such that the distal elements, widely separated in this species, are discontinuous with the axis. In contrast to the other species, females of D. /Jampra possess a receptable on the cyphopods, and there is an allopatric eastern form in which both the receptacle and the bend of the acropodite are reduced. The eastern form, occurring in Tennessee, North Carolina, and Geor- gia, also differs in having a generally longer prefemoral process that is parallel to the acropodite in medial view, as opposed to the Louisiana populations in which the structure angles across the acropodite stem. The magnitude of these and other differences suggests that the two forms are reproductively isolated and that Thri- naxoria should be resurrected to accommodate them. Their removal from Di- cellarius leaves it with a more homogeneous content and results in two generic segregates with comparable differences between their component species. As a consequence of this change and from improved anatomical knowledge of all the species, alterations are necessary in Hoffman’s diagnosis (1979:187) of the tribe Pachydesmini. The species of both Dicellarius and Thrinaxoria lack coxal apophy- ses, which are therefore an apomorphic feature of Pachydesmus. The transverse sternal ridges called “‘podosterna”’ for Pachydesmus (Hoffman 1958) are reduced and sometimes barely detectable in both Dicellarius and Thrinaxoria, so their usage as a primary tribal character should be qualified. Hence, I also include a revised diagnosis of this tribe, whose chief features are gonopodal. The realization that 7Thrinaxoria is a valid genus comprised of two species affords the opportunity to resolve an outstanding enigma in diplopodology and assign the name Polydesmus bifidus Wood, heretofore regarded unofficially as a nomen inquirendum. In transferring bifidus to Epeloria, Chamberlin and Hoffman (1958) predicted that it might be a form of Thrinaxoria, and this prophecy can now be confirmed. The types of this species are lost, and Wood’s original de- scription and illustration (1864, 1865) do not clearly indicate the identity of the form(s) under study. However, he had material from both Georgia and Texas, which corresponds closely to the known range of Thrinaxoria, since lampra prob- ably occurs in the northeastern corner of Texas (see species account). Wood’s illustration of the bifidus gonopod is also very similar to the in situ configuration of the two species of Thrinaxoria (compare with Fig. 25), so I think he had both forms on hand and assumed they were conspecific. Wood did not give precise VOLUME 97, NUMBER 3 475 localities for his material, only the states of Georgia and Texas. However, the Georgia sample was probably from the mountains of north Georgia since the eastern species of Thrinaxoria is more common in this section of the state (see Fig. 29). Wood did not indicate which sample he considered the types, so his name could be assigned to either species of Thrinaxoria. However, since Jampra is available for the western species, both names are conserved by assigning bifida to the eastern one and changing the gender of the suffix. This resolution is also compatible with the law of priority, which does not apply here, since Georgia precedes Texas in Wood’s accounts. In the absence of his material, I designate a neotype for bifida from Polk County, Tennessee, which is adjacent to Fannin and Murray counties, Georgia, and the only county where more than one male has been collected. As with P. bifidus, the type specimens of Fontaria lamellidens Chamberlin are also lost. However, Chamberlin (1931) reported a precise type locality, Biloxi, Harrison County, Mississippi, and the identity of this nominal species can be determined from near topotypical material. My analysis indicates that F. /amel- lidens is a race of D. bimaculatus (McNeill). In the Dicellarius section of this paper, I depart from my usual practice of describing the type-species and comparing and contrasting the others with it. The type-species of Dicellarius, D. okefenokensis (Chamberlin), is atypical in its small size and gonopodal torsion, and I have therefore chosen D. bimaculatus as the basis for anatomical comparisons in this genus. I also show only the medial aspect of the gonopods of most forms because the prefemoral process is masked in lateral view and the positions of the solenomerite and tibiotarsus are merely reversed. This study indicates that Dicellarius consists of five valid species, two of which have recognizable geographic races, and that 7hrinaxoria is comprised of two species. Information on variation and distribution is presented in the species or subspecies accounts along with appropriate synonymies. Acronyms of sources of preserved study material are as follows: AU—Department of Zoology and Entomology, Auburn University, Auburn, Alabama. CC—Biology Department, Columbus College, Columbus, Georgia. FSCA— Florida State Collection of Arthropods, Gainesville, Florida. MCZ—Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts. MEM -— Mississippi Entomological Museum, Mississippi State University, Starkville, Mississippi. NCSM—North Carolina State Museum of Natural History, Raleigh, North Carolina. The invertebrate catalog numbers of material in this institution are indicated in parentheses. NMNH—National Museum of Natural History, Smithsonian Institution, Washington, D.C. RLH—Private collection of Richard L. Hoffman, Radford, Virginia. RVC—Private collection of the late Ralph V. Chamberlin, now being acces- sioned by the NMNH. UMMZ— Museum of Zoology, University of Michigan, Ann Arbor, Michigan. WAS —Private collection of William A. Shear, Hampden-Sydney, Virginia. 476 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Review The genus Dicellarius was established incidentally by Chamberlin (1920) in the introduction to a paper describing a new species of Eurymerodesmus. It was mistakenly assigned to the family Leptodesmidae, now a synonym of Chelodes- midae. The single sentence proposing the taxon reads, ““The second (new North American leptodesmid genus) may bear the name Dicellarius, with Leptodesmus okefenokensis Chamberlin the genotype.’ According to the International Com- mission of Zoological Nomenclature, this pre-1931 statement validates Dicellarius because it was published in combination with an indication, an available specific name. Thus, Dicellarius is the oldest genus-group name for forms referrable to Epeloria and Spathoria, and must be retained in modern xystodesmid taxonomy even though no diagnostic statements have ever been published. Its type-species was described by Chamberlin (1918) from Billy’s Island in the Okefenokee Swamp, Georgia. Chamberlin (1931) proposed Fontaria lamellidens from Biloxi, Mississippi, and (1939) designated it the type of the new genus Spathoria. Polydesmus bimaculatus, described by McNeill (1887) from Pensacola, Florida, was transferred into Spa- thoria as the other component species. Chamberlin (1939) also erected Epeloria for the new species E. talapoosa, misspelling the type locality, Tallapoosa, Georgia, in both the locality citation and the specific name. Since the name was misspelled in both places, it cannot be considered an emendable lapsus, and the error must be retained in modern nomenclature. During the 1940’s, Chamberlin named five new species of Epeloria as follows: fictus (1943), dela and leiacantha (1946a), atlanta (1946b), and nannoides (1949). Chamberlin and Hoffman (1950) erected Thrinaxoria for Fontaria lampra Cham- berlin, listing Zinaria aberrans Chamberlin as a synonym. These two nominal species were described from Louisiana by the author in 1918 and 1942, respec- tively. During this period new locality records of E. leiacantha were reported by Chamberlin (1947, 1951), and Causey (1955) published the first Florida locality of E. fictus. Causey also transferred bimaculatus into Epeloria from Spathoria and reported new localities for this species from Alabama and Mississippi. Cham- berlin and Hoffman (1958) recognized three genera in their checklist—Dice/larius, Epeloria, and Thrinaxoria—and assigned Spathoria to synonymy under Dicellari- us. Thus Dicellarius consisted of okefenokensis, the type-species; bimaculatus, transferred from Epeloria; and lamellidens, referred there because of the generic synonymy. They listed all the species under Epeloria that were then assigned to it except nannoides, which was synonymized with D. okefenokensis, and also brought Polydesmus (Fontaria) bifidus Wood into the genus, predicting that it might prove to be a form of Thrinaxoria. Thus Epeloria included talapoosa, dela, leiacantha, atlanta, ficta and bifida. As in 1950, Chamberlin and Hoffman (1958) still considered Thrinaxoria to be monotypic, and they added Tuscaloosa, Ala- bama, as the second locality for /ampra. Since 1958 only two authors have dealt with Dicellarius. Loomis (1969) pro- posed D. sternolobus from Tallapoosa County, Alabama, and published a key distinguishing bimaculatus, lamellidens, okefenokensis, and sternolobus, the four species then in the genus. Hoffman (1979) placed Epeloria, Thrinaxoria, and Spathoria in synonymy under Dicellarius and formally proposed the tribe Pachy- desmini for Pachydesmus and Dicellarius. He also stated that Dicellarius consisted VOLUME 97, NUMBER 3 477 of four species in the southeastern states, but since these were not mentioned specifically, the previous taxonomy still stands. Thus at this writing, Dicellarius consists of the following 11 species listed chronologically below with their type localities and other reported states of occurrence. Suffix endings have been changed to masculine where necessary for agreement in gender. D. bifidus (Wood, 1864). Described from Georgia and Texas, no definite lo- calities known. D. bimaculatus (McNeill, 1887). Pensacola, Escambia Co., Florida; also re- ported from Mobile Co., Alabama, and Jackson Co., Mississippi. D. okefenokensis (Chamberlin, 1918). Billy’s Island, Okefenokee Swamp, Charl- ton Co., Georgia; also reported from Gainesville, Alachua Co., Florida. D. lamprus (Chamberlin, 1918). Creston, Natchitoches Par., Louisiana; also reported from Shreveport, Caddo Par., Louisiana; and Tuscaloosa, Tuscaloosa Co., Alabama. ; D. lamellidens (Chamberlin, 1931). Biloxi, Harrison Co., Mississippi. D. talapoosa (Chamberlin, 1939). Tallapoosa, Haralson Co., Georgia. D. fictus (Chamberlin, 1943). Thomasville, Thomas Co., Georgia; also reported from Tallahassee, Leon Co., Florida. D. delus (Chamberlin, 1946). Morgan, Calhoun Co., Georgia, incorrectly re- ported as being in Morgan Co. by Chamberlin and Hoffman (1958). D. leiacanthus (Chamberlin, 1946). Along Spring Cr., Decatur Co., Georgia; also reported from Ft. Benning, Muscogee-Chattahoochee cos., Georgia, and an unspecified county in Alabama. D. atlanta (Chamberlin, 1946). Atlanta, Fulton-Dekalb cos., Georgia. D. sternolobus Loomis, 1969. Alexander City, Tallapoosa Co., Alabama. Taxonomic Characters The genera Dicellarius and Thrinaxoria can be distinguished in either sex. In females of the latter, the cyphopods possess a receptacle or remnant thereof, whereas this structure is absent in Dicellarius. In males of Thrinaxoria the so- lenomerite and tibial process of the acropodites, collectively referred to as “‘distal elements,” arise near midlength, are widely separated, and never lie over one another in medial or lateral views. In Dicellarius the elements arise distal to the midlength, are moderately separated at best, and one is frequently obscured by the other in these perspectives. Pachydesmus differs from both these taxa in the large postgonopodal sternal elevations, the apophyses on the gonopodal coxae, and the proximal origin of the elements of the acropodite, which arise at the juncture with the prefemur. In Dicellarius, body size is useful in distinguishing D. okefenokensis, since it is much smaller than any other congener. It is the only species whose adults are less than 30 mm long and 5 mm wide, and female only samples of D. okefenokensis and D. b. fictus can therefore be separated on this basis. The only other somatic features of taxonomic importance in Dicellarius are the pregonopodal sternal projections in D. sternolobus. Although variable, these structures are distinctly visible on segments 4 and 5, and are at least subequal in length to the widths of the adjacent coxae (Figs. 19-20). In all the other species, the processes are small and have no taxonomic value (Figs. 1-2). The remaining taxonomic characters in Dicellarius are located on the gonopods. 478 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Aspects that are important include the in situ configuration, the orientation of the prefemoral process, the general curvature and configuration of the acropodite, the course of the prostatic groove, and the shape of the distal elements. In most forms the tips of the acropodites overlap in situ (Figs. 7, 13), and where they are far enough apart, the distal elements may crisscross (Fig. 15). In D. atlanta and D. bimaculatus fictus, however, the acropodites do not overlap but lie side by side in a subparallel arrangement (Figs. 5, 17), thus allowing distinction from the other forms. The prefemoral process is acicular to some degree in most forms of Dicellarius, but it is directed differently due largely to the configuration of the acropodite. The process lies parallel to the acropodite stem (above in medial view) in many forms, but it may angle across the stem with the tip either lying over the far side or protruding beyond the opposite margin (Figs. 8, 14, 22). In D. okefenokensis and D. bimaculatus fictus, the acropodite curves subanteriad so that the distal elements overhang the level of the prefemoral process in the former (Fig. 10) and overlap the tip of the process in the latter (Fig. 6). The prostatic groove is typically completely visible in medial view from its origin in the base of the prefemur to its termination at the tip of the solenomerite. In some forms, however, it meets the anterior margin at the base of the soleno- merite, where it continues to the terminal opening (Fig. 14, 18). The extreme condition obtains in D. okefenokensis, where the groove crosses to the lateral side because of torsion in the acropodite (Figs. 10-12). This situation is convergent with torsion in the tribe Apheloriini, and represents a derived, autapomorphic trait in D. okefenokensis. The configuration of the distal elements and their position relative to each other are very important in the identification of a species or subspecies of Dicellarius. They are at most only moderately separated, and either lie parallel to each other or diverge apically. Both structures may be visible in medial or lateral views, in which case the gap between them also shows (Fig. 18), or one may overlie the other, completely obscuring the gap and at least partly hiding the opposite member (Figs. 8, 10). The tibial process may be shorter or longer than the solenomerite, but it is never narrower and sometimes is considerably wider. The configuration of the solenomerite is the most reliable determinant of the highly variable species D. bimaculatus. Here the structure is parallel-sided except for the tip, where the outer edge angles abruptly toward the inner forming a subacuminate inner corner (Figs. 4, 6, 8). Another way of describing this condition would be to say that the inner corner is produced or elongated. By contrast, the solenomerite is acicular in D. talapoosa, D. atlanta, and D. sternolobus, and both sides taper smoothly and continuously throughout the length to a subacuminate tip located in the midline (Figs. 14, 16, 18,22). In Thrinaxoria the taxonomic characters involve the size of the receptacle on the cyphopods, the degree of bend on the acropodites, and the length and ori- entation of the prefemoral process. Thrinaxoria lampra has a moderate-size re- ceptacle, whereas it is a remnant or vestige in 7. bifida. The acropodites display a smoother curve in the latter species, and the distal elements are more or less continuous with the curvature. The prefemoral process in 7. /ampra is shorter than in T. bifida and always angles across the acropodite stem in medial view. In VOLUME 97, NUMBER 3 479 T. bifida, however, the structure usually lies parallel to the acropodite and extends linearly along its anterior margin (Figs. 26, 28). Tribe Pachydesmini Hoffman Pachydesmini Hoffman, 1979:158, 187. Components. — Pachydesmus Cook, 1895; Dicellarius Chamberlin, 1920; Thri- naxoria Chamberlin and Hoffman, 1950. Diagnosis.— Relatively stiff, inflexible Xystodesminae of small to large size; sterna broad, with or without variable, transversely oval ridges or elevations on midbody segments; gonopods small to large, with or without dorsal coxal apoph- ysis, telopodite usually linear proximad, curving or bending distal to midlength, with long slender accessory process paralleling acropodite or with slender sub- terminal branch; cyphopods with or without receptacle or remnant thereof. Range.—Southeastern United States, from the eastern edge of Texas to coastal Georgia, South Carolina, and the southcentral part of North Carolina (Piedmont Plateau) just north of the state line, ranging northward across Tennessee and southward into the central Florida peninsula. Remarks.—This revised diagnosis incorporates several modifications necessi- tated by validation of Thrinaxoria and by improved knowledge of Dicellarius. It allows for the small size of D. okefenokensis; the absence of the coxal apophysis in all species of Thrinaxoria and Dicellarius, the reduced sternal ridges in both genera; and the variable condition of the receptacle on the cyphopods. These changes diminish the tribe’s distinctiveness from the Apheloriini, over which it has taxonomic priority. An additional diagnostic phrase is therefore inserted on the generally linear telopodites, demonstrated by all species except D. okefeno- kensis, in which the broad curvature is clearly derived. This feature of the telopo- dites does distinguish the Pachydesmini from the Apheloriini, where as described by Hoffman (1979), the telopodites are either semicircular, sigmoidal, or are bent or curved strongly mesad. The date of proposal of Dicellarius is also corrected to 1920. Genus Dicellarius Chamberlin Dicellarius Chamberlin, 1920:97.—Chamberlin and Hoffman, 1958:30.—Hoff- man, 1979:158. Epeloria Chamberlin, 1939:3; 1949:101.—Causey, 1955:24.—Chamberlin and Hoffman, 1958:32. Spathoria Chamberlin, 1939:6. Type-species.—Of Dicellarius, Leptodesmus okefenokensis Chamberlin, 1918, by original designation; of Epeloria, E. talapoosa Chamberlin, 1939, by original designation; of Spathoria, Fontaria lamellidens Chamberlin, 1931, by original designation. Description.—A genus of small to moderately large xystodesmids with the fol- lowing characteristics: Body composed of head and 20 segments in both sexes; size varying from 4— 480 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 8.5 mm in width and 22-42 mm in length; W/L ratio similarly varying from about 17—22.5%. Color in life variable; most forms with drab olive-brown base color, lighter margins, and darker stripes along caudal edges of metaterga. Head of normal appearance, moderately smooth and polished, occasionally finely granulate. Epicranial suture distinct, terminating in interantennal region, not apically bifid; interantennal isthmus variable; genae not margined laterally, with variable central impressions, ends broadly rounded and projecting slightly beyond adjacent cranial margins. Antennae moderately slender, varying in length, becoming progressively more hirsute distally, with 4 conical sensory cones on ultimate article; no other sensory structures apparent. Facial setae with epicranial, frontal, genal, clypeal, and labral series; with or without interantennal and suban- tennal. Terga usually smooth and moderately polished; becoming moderately coria- ceous on paranota. Collum variable, ends subequal to or extending well beyond those of following tergite. Paranota moderately to strongly depressed, caudolateral comers rounded on anteriormost segments, becoming blunt in midbody regions and progressively more acute posteriorly. Peritremata distinct, strongly elevated above paranotal surface; ozopores located caudal to midlength, opening dorsad to dorsolaterad. Prozonites smaller than metazonites; strictures moderately dis- tinct, slightly costulate. Caudal segments normal for family. Sides of metazonites variable, usually granulate, with or without shallow grooves and impressions. Gonapophyses broad, apically expanded. Pregonopodal sterna of males variously modified; that of segment 4 with minute projection, much shorter than widths of adjacent coxae, or large apically divided process, longer than widths of adjacent coxae; sternum of segment 5 with low, transverse ridge or large, ventrally directed process between anterior legs, latter much longer than widths of adjacent coxae, and with variable impressions between posterior legs; 6th sternum with variable impressions between caudal legs to accommodate apices of acropodites. Postgonopodal sterna mildly hirsute, with or without variable low, rounded, longitudinal or transverse elevations in midlines of midbody segments. Coxae without projections; prefemoral spines relatively long and sharply pointed; tarsal claws slightly bisinuate. Hypoproct broadly rounded; paraprocts with mar- gins slightly thickened. Gonopodal apertures ovoid to elliptical, with or without slight anteriolateral indentations, sides flush with or slightly elevated above metazonal surfaces. Gon- opods in situ with apices overlapping, interlocking, or lying parallel to one another; usually with tips of prefemoral processes crossing. Coxae moderate in size, without apophyses, connected by membrane only, no sternal remnant. Prefemora mod- erate, with variable prefemoral processes arising on anterior or anteriomedial sides. Acropodites moderately thick and robust, well sclerotized; configurations variable, either extending sublinearly from prefemur or curving slightly anteriad; divided at around *4 length into solenomerite and tibial process, both variable in length, configuration, and position relative to the other. Prostatic grooves arising in pits in bases of prefemora, usually running entirely along medial sides of acropodite stems, crossing to anterior and lateral surfaces in some species; opening terminally on solenomerites. VOLUME 97, NUMBER 3 481 Cyphopodal apertures elliptical, encircling 2nd legs, sides slightly elevated above metazonal surfaces, without pleurotergal lobes on anteriolateral corners. Cypho- pods in situ located lateral to 2nd legs, variably positioned in apertures; without receptacles. Valves moderate to large, subequal, surfaces finely granulate. Opercula relatively large, located on lateral sides of valves. Distribution.— Coastal Plain of Georgia, Florida, Alabama, and Mississippi; southern Ridge and Valley and Appalachian Plateau Provinces of Alabama; Pied- mont Plateau and southern fringe of Blue Ridge Province of Georgia and North Carolina. The Tennessee River is a distributional barrier in northern Alabama, and the Savannah River forms the range limit along the Atlantic Coast. Species.— Five, two of which are divided into two and three subspecies. Key to species of Dicellarius (based primarily on adult males) 1. Body size small; acropodites with torsion, prostatic groove crossing from medial to lateral sides; tibial process overlying and largely obscuring so- lenomerite in medial view; southeastern Georgia to central Florida .... I race a i us EE SET See fs Preys shades .0 okefenokensis (Chamberlin) — Body size moderate; acropodites without torsion, prostatic groove either running entirely on medial side or angling to anteriomedial edge at mid- length; distal elements variable in position and configuration, but tibial process never overlying solenomerite in medial view ................. 2 2. Sterna of segments 4 and 5 with large, ventrally directed processes, sub- equal to or longer than widths of adjacent coxae; solenomerite overlying tibial process in medial view, sides tapering smoothly and continuously to subacuminate tip; tibial process with several grooves and ridges on lateral surface; central Alabama ...................... sternolobus Loomis — Sterna of segments 4 and 5 with at most minute projections, much smaller than widths of adjacent coxae; solenomerite variable, occasionally over- lying tibial process in medial view and with sides parallel except apically; tibial process without noticeable ridges and grooves .................. 3 3. Tibial process broad, spatulate; sides of solenomerite parallel except api- cally, then narrowing abruptly to subacuminate tip on inner corner; central and eastern Georgia to western Florida, southern Alabama, and southern andueastenne VEISSISSIP DI mews oe eee bimaculatus (McNeill) — Tibial process and solenomerite acicular, sides narrowing smoothly and continuously to subacuminate tips located in midlines ............... 4 4. Acropodite sublinear; distal elements subequal, moderately separated; pre- femoral process on anterior side of acropodite; western North Carolina to COM Tal MG COREA ey Ser PB es esi ince, Saske, os pes hacaig ewe! atlanta (Chamberlin) — Acropodite curved gently caudad; distal elements contiguous to narrowly separated, solenomerite slightly longer; prefemoral process angling across stem of acropodite to caudal side; north-central Alabama to western Geor- BA 5S ES ee ee eee een ie | talapoosa (Chamberlin) Dicellarius bimaculatus (McNeill) Diagnosis.—Acropodite highly variable, either curved subanteriad, subcaudad, or submediad. Distal elements narrowly separated, continuing general curvature 482 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of acropodite, in some forms overlying one another in medial and lateral views; solenomerite laminate, sides parallel except at apex, then narrowing abruptly to subacuminate tip on inner corner; tibial process laminate and apically rounded, subequal in length to solenomerite. Remarks. —Dicellarius bimaculatus is the widely-distributed species along the Gulf of Mexico occurring in the Coastal Plains of Georgia, Alabama, Mississippi, and the Florida panhandle. The type locality is Pensacola, Escambia County, Florida. Forms in the eastern 7 of the range demonstrate considerable gonopodal stability, but the incidence of variation rises around the Alabama and Escambia Rivers. Populations become increasingly more variable west of this point, and the gonopodal differences between forms at Mobile, Alabama, and Biloxi, Mis- sissippi, a distance of about 45 miles, are much greater than the differences between forms at Savannah, Georgia, and Montgomery, Alabama, a distance of over 300 miles. The situation west of these rivers approaches a clinal transition, particularly along the Gulf Coast, and presents difficulties in making taxonomic decisions. The widespread, anatomically stable eastern form clearly represents a single geo- graphic race, and three Chamberlinian names are available, the oldest being D. b. fictus. In the rest of the range, one could justifiably recognize one large, highly variable subspecies or a number of smaller, more stable ones. Two names have been proposed for forms from this area, D. bimaculatus the oldest one referable to the complex and hence the specific name, and D. /amellidens (Chamberlin). Thus, if subspecies are designated, the nominate will be from this area of ana- tomical instability. One alternative would be to include the form at Pensacola with the eastern one, enlarge its range slightly, and make it the nominate sub- species. However, I think that the stable eastern morphotype deserves recognition in its own right and that the nominate subspecies should therefore refer to another form. After analyzing males from every sample west of the Alabama and Escambia Rivers, it became evident that a single taxonomic unit could be described from all of this area except the coast of Mississippi. This entity becomes the nominate subspecies and is necessarily much more variable than D. b. fictus. However, the much greater variation along the coast of Mississippi clearly requires separate treatment. Since the name /amellidens is available, I propose a third subspecies with much smaller range west of the Pascagoula River. East of this boundary and in the city of Pascagoula the forms are intermediate between the “‘average”’ variant of D. b. bimaculatus and are thus considered intergrades. Other resolutions are possible, but this one maximizes the existing nomenclature and provides a balance between naming nearly every local population west of the Alabama and Escambia Rivers and ignoring subspecies altogether. Dicellarius bimaculatus bimaculatus (McNeill), new status Figs. 1-4 Polydesmus bimaculatus McNeill, 1887:323, figs. 3-5. Fontaria bimaculata.—Attems, 1938:166. Spathoria bimaculata.—Causey, 1955:24. Dicellarius bimaculatus.—Chamberlin and Hoffman, 1958:30. Type specimens.—Three male and 15 female syntypes (NMNH) collected by VOLUME 97, NUMBER 3 483 C. H. Bollman, Mar-Apr 1886, at Pensacola, Escambia Co., Florida. I have designated one male as the lectotype. Diagnosis.—Gonopods in situ with tips of acropodites curved slightly mediad and overlapping, prefemoral processes overlapping; prefemoral process narrow and acicular, tapering gradually distad, angling across acropodite stem and lying nearly entirely over latter in medial view; acropodite with curvature in medial plane and thus masked in medial view, extending sublinearly from prefemur; distal elements subequal in width and length, continuing linear configuration of acropodite; solenomerite with edge directed mediad. Lectotype.—Length 31.3 mm, maximum width 6.3 mm, W/L ratio 20.1%, depth/width ratio 65.1%. Segmental widths as follows: collum 4.9 mm Sth-15th 6.3 2nd 5.4 16th 5.8 3rd 5.8 Wa 52 4th 6.0 18th 4.1 Color in life unknown. McNeill (1887) described the dorsum as being “‘obscure olive or chestnut, the scuta generally marked with an indistinct transverse dark band, with lighter color towards the margins; a well defined oval spot of gray is frequently present on the lateral margins; lateral laminae with a narrow pink border.”’ This corresponds generally to my observations on living species of Di- cellarius. Head capsule smooth, polished; width across genal apices 3.3 mm, interantennal isthmus 1.4 mm; epicranial suture thin but distinct, terminating above interan- tennal region, not bifid. Antennae relatively short, reaching back only to middle of 3rd segment, becoming progressively more hirsute distally, first antennomere subglobose, 2-6 clavate, 7 short and truncate, relative lengths of antennomeres 2>6>3>4>5>1 > 7. Genae not margined laterally, with distinct central impressions, ends broadly rounded and projecting slightly beyond adjacent margin of cranium. Facial setae as follows: Epicranial 2-2, interantennal absent, frontal 1-1, genal 3-3, clypeal about 14-14, labral about 20-20, merging with clypeal series and continuing for short distance along genal borders, about 3 setae on each side. Terga smooth, polished, becoming moderately coriaceous on paranota. Collum broad, ends extending slightly beyond those of 2nd tergite. Paranota moderately depressed, continuing slope of dorsum, anterior corners blunt with distinct, ele- vated scapulorae, caudolateral corners rounded through segment 5, becoming blunt and progressively more acute posteriorly. Peritremata distinct, strongly elevated above paranotal surface. Ozopores located caudal to midlength of peritre- mata, opening dorsolaterad. Sides of metazonites dull and granulate but without noticeable grooves or impressions. Strictures sharp, distinct, prozona elevated slightly above metazona. Gonapophyses broad, apically expanded. Pregonopodal sterna (Figs. 1-2) gen- erally unmodified, without noticeable lobes or projections; 6th sternum recessed slightly between 7th legs to accommodate apices of acropodites. Postgonopodal sterna higher than prozonites, with transverse grooves between leg pairs merging 484 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON into low, rounded, longitudinal elevations in midlines, elevations noticeable only on segments 8—15; caudal margins of sterna indented medially. Pregonopodal legs densely hirsute, postgonopodal legs becoming progressively less hirsute caudally. Coxae without projections; prefemoral spines beginning on segment 5, becoming longer and sharper on midbody region; tarsal claws faintly bisinuate. Hypoproct rounded, paraprocts with margins slightly thickened. Gonopodal aperture 2.0 mm wide and 0.9 mm long at midpoint, indented slightly anteriolaterad, sides flush with metazonal surface. Gonopods in situ (Fig. 3, not this specimen) with acropodites projecting anteriad from aperture, ex- tending well beyond anterior margin of aperture with tips overlapping between 7th legs, prefemoral processes also overlapping and extending beyond aperture. Gonopod structure as follows (Fig. 4): Prefemur moderate, with long acicular prefemoral process on anterior side angling over acropodite stem, bent anteriad distally and terminating just beyond base of tibial process, tip directed toward midlength of solenomerite. Acropodite curving submediad and curvature thus masked in medial view, projecting linearly from prefemur, sides narrowing grad- ually, divided at *%4 length into anterior solenomerite and posterior tibial process. Latter laminate, subequal in length to solenomerite, apically blunt. Solenomerite with sides subparallel except at apex, then narrowing abruptly to subacuminate tip on inner corner, curving broadly mediad with edge visible in medial view. Prostatic groove arising in pit in prefemur extending along medial surface of acropodite onto solenomerite, opening apically. Female paralectotype.— Length 30.8 mm, maximum width 6.9 mm, W/L ratio 22.4%, depth/width ratio 63.8%. Agreeing closely with lectotype in most somatic features, except paranota more strongly depressed, giving appearance of more highly arched body, and sternal elevations greatly reduced, barely detectable. Cyphopodal aperture elliptical, encircling 2nd legs, sides barely elevated above metazonal surface, without pleurotergal lobe on anteriolateral corners. Cyphopods in situ with valvular openings visible in aperture; valves large, subequal, surface finely granulate. Receptacle absent. Operculum relatively large, located on lateral side of valves, surface finely granulate. Variation. — The length of the prefemoral process varies, and it does not extend to the level of the acropodite division in some males. It is always acicular, but does not always extend across the acropodite stem. In one male there is a small, subterminal spur. In all forms the acropodite curvature visible in situ (Fig. 3) is masked in medial view because it lies in the plane of vision and perpendicular to the plane of drawing. However, not all the acropodites are as straight as that of the lectotype. Some bend anteriad, as in D. b. fictus, while others tend to lean caudad. In the eastern part of the range the distal divisions appear as in the lectotype, but farther west in Mississippi they lean caudad and the anterior surface is seen in medial view. This variation befits the intermediate geographical position of D. b. bimaculatus between the other two subspecies. Distribution. — Western Florida, southwestern and west-central Alabama, and east-central Mississippi. The Conecuh-Escambia River forms the approximate eastern boundary, but the western and northern extremes do not correspond with any physiographic features. Most localities are west of the Mobile-Tombigbee River. The subspecies is sympatric and syntopic with 7. Jampra in Washington VOLUME 97, NUMBER 3 485 County, Alabama, along the Tombigbee River. Specimens were examined as follows: FLORIDA: Escambia Co., Pensacola, 3M, 15F, Mar-Apr 1886, C. H. Bollman (NMNH), F, Nov 1969, C. Seal (FSCA), and 3M, 7F, 15 Mar 1970, C. Seal (FSCA) TYPE LOCALITY; 3 mi. W Pensacola, 3M, F, 22 Jan 1965, N. B. Causey (FSCA); Atmore, 3M, F, 15 Mar 1962, C. R. Parker (FSCA); along US hwy. 90 E jct. FL hwy. 29, M, 1 Jan 1954, N. B. Causey (FSCA); Cantonment, M, 1 May 1954, N. B. Causey (FSCA); and 3 mi. SE Cantonment M, 25 Nov 1967, M. Tidwell (FSCA). ALABAMA: Monroe Co., 2.5 mi. E Chrysler, F, 18 Apr 1976, M. R. Cooper (NCSM A735); and 6 mi. S Monroeville, M, 10 Mar 1977, R. E. Woodruff (FSCA). Conecuh Co., 0.8 mi. E Castleberry, along AL hwy. 6 at Murder Cr., M, 24 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4056); Evergreen, 3M, 10 Apr 1960, L. Hubricht (RLH); and 15 mi. N Evergreen, M, F, 9 Apr 1960, L. Hubricht (RLH). Baldwin Co., Daphne, M, F, 14 Mar 1948, B. A. Maina (RLH); jct. US hwys. 90 & 98 W Loxley, 2F, 22 Jan 1955, N. B. Causey (FSCA). Mobile Co., Spring Hill College, several MM, FF, 2 May 1954, C. E. Valentine (FSCA); Grand Bay, 4M, F, 17 Jun 1953, N. B. Causey (FSCA); Mobile, F, Apr 1954, N. B. Causey (FSCA); Theodore, M, 4F, 21 Jun 1961, L. Hubricht (RLH); 2.5 mi. SW Bucks, 2F, 13 May 1962, L. Hubricht (RLH); and along Alligator Cr. nr. Dog R., 2M, 3F, 21 Sep 1952, collector unknown (UMMZ). Washington Co., 3 mi. E Leroy, 3 Jul 1960, L. Hubricht (RLH). Clarke Co., Jackson, M, F, 23 Sep 1910, R. V. Chamberlin (RVC); Walker Springs, 2M, 25 Jul 1965, S. B. Peck (FSCA); and Coffeyville, F, 16 Jun 1959, N. B. Causey (FSCA). Choctaw Co., E of Lavaca nr. Tombigbee R., M, F, 4 Jul 1960, L. Hubricht (RLH). MISSISSIPPI: Oktibbeha Co., State College, M, 19 Apr 1917, S. B. Pearce (MCZ); Craig Springs, 6M, F, 22 Apr 1980, and 3M, F, 15 May 1980, G. Snodgrass (MEM); and Starkeville, M, 14 May 1982, G. Sullivan (MEM). Winston Co., Noxapater, 3 juvs., 29 Dec 1965, J. B. Black (FSCA). Lauderdale Co., Meridian, 7M, F, 10 Dec 1961, L. Hubricht (RLH). Jasper Co., Louin, 2M, 24 Mar 1963, L. Hubricht (RLH). Wayne Co., 8 mi. E Waynesboro, 3M, 3F, 13 Mar 1963, L. Hubricht (RLH). Jones Co., nr. Laurel, M, 2F, 15 Jul 1980, J. Robbins (MEM). Lamar Co., Oak Grove, F, 12 Apr 1957, L. Smith (FSCA). Forrest Co., Rawls Springs Twp. nr. Hattiesburg, M, 15 Feb 1957, B. D. Valentine (FSCA). Dicellarius bimaculatus fictus (Chamberlin), new combination, new status Figs. 5—6 Epeloria ficta Chamberlin, 1943:37—38, fig. 11.—Chamberlin and Hoffman, 1958: 32-33. Epeloria fictus.—Causey, 1955:24. Epeloria dela Chamberlin, 1946a:139, figs. 1-2.—Chamberlin and Hoffman, 1958:32. Epeloria leiacantha Chamberlin, 1946a:139-140, fig. 3; 1947:29; 1951:28.— Chamberlin and Hoffman, 1958:33. Type specimens.—Male holotype and female allotype (RVC) collected by H. Field, 5-10 Apr 1940, from Thomasville, Thomas Co., Georgia. 486 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON : ANN Figs. 1-8. Dicellarius bimaculatus: 1-4, D. b. bimaculatus. 1, Sternum of segment 4 of lectotype, caudal view; 2, Sternum of segment 5 of the same, caudal view; 3, Gonopods in situ, ventral view of male from Lauderdale Co., MS; 4, Telopodite of left gonopod of lectotype, medial view; 5-6, D. b. fictus. 5, Gonopods in situ, ventral view of male from Grady Co., GA; 6, Telopodite on left gonopod VOLUME 97, NUMBER 3 487 Diagnosis. —Gonopods in situ in subparallel arrangement, neither acropodites nor prefemoral processes overlapping; prefemoral process stout, wide basally and tapering sharply distad to acute tip, tip overlapping solenomerite; acropodite in medial view curved anteriad and extending over level of prefemoral process, distal elements subequal in width and length; solenomerite with profile visible in medial view. Variation.— The acropodite configuration in this race is more stable than that of the nominate subspecies. All males display the diagnostic acropodal curvature, with the solenomerite, and also the tibial process in some individuals, overhanging the prefemoral process. The latter is the most variable gonopodal feature. It is usually a wide, straight spine as in the holotype, but it is bent distally in some individuals and apically bifurcate with a subterminal spur in others. The orien- tation of the solenomerite also varies so that the produced inner corner is visible medially in some males and not in others. Distribution.—This subspecies has the widest distribution of any race of D. bimaculatus, and it is sympatric with T. bifida in Dougherty County, Georgia. The area includes the Fall Zone and Coastal Plain of Alabama east of the Tom- bigbee and Conecuh Rivers, the panhandle of Florida between the Blackwater and Suwannee Rivers, and the outer Piedmont Plateau and the Coastal Plain of Georgia except for the Suwannee River drainage in the southeastern corner. It abuts the ranges of D. okefenokensis in this area and D. atlanta and D. talapoosa in the north. A peripheral population in Chatham County, Georgia, is isolated from the rest of the range, although some of this hiatus may reflect inadequate collecting. The inner Coastal Plain has received little attention, and the Savannah record may connect with that in Tift County. Specimens were examined as follows: GEORGIA: Chatham Co., Savannah, Beaulieu Ave., 3M, F, 1 Nov 1959, L. Hubricht (RLH). Chattahoochee Co., Ft. Benning, M, 16 Apr 1950, D. E. Beck (RVC); and along GA hwy. 137 at Bagley Cr., M, 5 Sep 1980, G. E. Stanton (CC). Muscogee Co., Columbus College, M, F, 19 Oct 1977, G. E. Stanton (CC). Lee Co., N of Albany, F, 18 Jun 1959, N. B. Causey (FSCA). Tift Co., many MM and FF, Sep 1967-—Oct 1968, J. A. Payne (RLH). Dougherty Co., Albany, 2M, 12F, 12 Jun 1959, N. B. Causey (FSCA). Calhoun Co., Morgan, M, 4 Apr 1946, P. W. Tratt (RVC); and Leary, M, 25 Mar 1961, L. Hubricht (RLH). Early Co., between Saffold and Jakin, 2M, F, 18 Mar 1961, L. Hubricht (RLH); and Ko- lomoki Mounds St. Pk., 3M, 2F, 1 May 1983, R. M. Shelley and P. B. Nader (NCSM A4024). Baker Co., Pineland Plantation nr. Newton, 3M, F, 30 Mar 1959, D. B. Jester and H. Wyatt (FSCA). Seminole Co., 4.5 mi. SE Iron City, M F, and Ray’s Lake, F, 21 Mar 1954, T. H. Hubbell (UMMZ). Decatur Co., Spring Cr. W of Brinson, 2M, 3 Apr and 20 Aug 1946, P. W. Fattig (RVC), 2M, 2F, 31 Mar 1964, H. W. Levi (MCZ), and several MM and FF, 26 Jan 1965, N. B. Causey (FSCA). Grady Co., 3 mi. W Cairo, 4M, 6F, 26 Jan 1965, N. B. Causey — of holotype, ventral view; 7-8, D. b. lamellidens. 7, Gonopods in situ, ventral view of male from Jackson Co., MS; 8, Telopodite of left gonopod of the same, medial view. Scale line for Figs. 3, 5, and 7 = 1.00 mm; line for other Figs. = 1.23 mm for 1 and 2; 1.00 mm for 4, 6, and 8. Setae are omitted from all dissected gonopod and sternal drawings in this paper. 488 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (FSCA); and 8 mi. NE Cairo, along GA hwy. 188, M, 3F, 15 Sep 1979, R. M. Shelley and P. T. Hertl (NCSM A2875). Thomas Co., Thomasville, M, F, 5-10 Apr 1940, H. Field (RVC) and 2M, 15 Jul 1973, W. A. Shear (WAS) TYPE LOCALITY; Bar-M-Ranch nr. Boston, 8M, 2F, Apr 1968, W. Sedgwick (WAS) and M, F, 15 Jul 1973, W. A. Shear (WAS). Brooks Co., 8 mi. W Quitman, along US hwy. 84, F, 7 Jul 1960, N. B. Causey (FSC). ALABAMA: Marengo Co., Chickasaw St. Pk., M, F, 15 Jun 1959, N. B. Causey (FSCA) and 4M, F, 8 Jul 1980, R. M. Shelley (NCSM A3350). Dallas Co., Selma, 2M, 2F, 15 June 1959, N. B. Causey (FSCA). Wilcox Co., nr. Camden, 4M, 2F, 11 Nov 1962, L. Hubricht (RLH); and 5 mi. W Snow Hill, M, 10 Apr 1960, L. Hubricht (RLH). Lowndes Co., 2.5 mi. NE Sandy Ridge, F, 9 Apr 1960, L, Hubricht (RLH); and 8.4 mi. NE Ft. Deposit, 3M, F, 19 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4029). Montgomery Co., 7 mi. SW Montgomery, 4M, 3F, 16 Apr 1960, L. Hubricht (RLH); and 13.1 mi SSW Montgomery, along US hwy. 31 nr. Pintlalla, M, F, 19 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4025). Butler Co., McKenzie, M, F, 9 Apr 1960, L. Hubricht (RLH); 15.9 mi. NW Greenville, along AL hwy. 11, 0.2 mi. N jct. AL hwy. 54, 3M, 21 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4038); and 6 mi NE Searcy, 3M, F, 9 Apr 1960, L. Hubricht (RLH). Macon Co., 11.9 mi. SW Tuskegee, M, 10 May 1973, R. B. Little (AU). Crenshaw Co., Brantley, M, 17 Jul 1960, L. Hubricht (RLH); and Luverne, 4M, 19 Mar 1961, L. Hubricht (RLH). Pike Co., 10.8 mi. NW Troy, M, 26 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4060). Russell Co., 10.8 mi. SW Phenix City, 2M, 26 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4063). Coffee Co., jct. US hwy. 84 and AL hwy. 87 nr. Elba, 2M, 8F, 26 Jan 1965, N. B. Causey (FSCA). Geneva Co., 1 mi. E Geneva, along Choctawhatchee R., 2M, 2 Jul 1960, L. Hubricht (RLH); and 5.8 mi. S Hartford, M, 30 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4078). Houston Co., 1 mi. SE Grangeburg, M, 26 Mar 1961, and 4.5 mi. W Avon, 3M, 18 Mar 1961, L. Hubricht (RLH); 4 mi. E Webb, along Cedar Spring Cr., 3M, 9 Jul 1967, D. R. Whitehead (RLH); and Chattahoochee St. Pk., 2M, 30 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4079). FLORIDA: Walton Co., Gaskin, F, 29 Feb 1960, R. Cordle (FSCA); DeFuniak Springs, M, 6F, 7 Oct 1958, collector unknown (FSCA); and 8.5 mi. W DeFuniak Springs, along US hwy. 90, 7M, 2F, 22 Jan 1965, N. B. Causey (FSCA). Wash- ington Co., Chipley, 2M, F, 23 Jan 1965, N. B. Causey (FSCA). Bay Co., Panama City, M, 2 Feb 1960, R. E. Woodruff (FSCA) and 2M, 5 Mar 1961, R. C. Hallman (FSCA). Jackson Co., Florida Caverns St. Pk., M, 3F, 27 May 1968, R. E. Wood- ruff, and M, 2F, 14 Apr 1960, H. V. Weems (FSCA); Marianna vic., M, 26 Mar 1961, L. Hubricht (RLH) and M, 3 Apr 1968, J. A. Beatty (WAS); Cottondale, M, 31 Jul 1960, L. Hubricht (RLH); Three Rivers St. Pk. and Sneads vic., M, 2F, 7 Sep 1959, and 2M, 2 Mar 1961, L. Hubricht (RLH), M, 15 May 1964, M. Muma (FSCA), and 2M, F, 19 Jul 1973, W. A. Shear (WAS). Calhoun Co., 2 mi. NW Chason, F, 4 Feb 1960, T. E. Kunkle (FSCA). Liberty Co., along Appalach- icola R. nr. Bristol, M, F, 24 Jan 1965, N. B. Causey (FSCA) and Torreya St. Pk., M, 15 Mar 1963, H. W. Levi (MCZ), and 5M, 3F, 24 Jan 1965, N. B. Causey, and M, 22 Apr 1967, P. C. Drummond (FSCA). Leon Co., Tallahassee, many MM and FF collected from 1953-1959 by H. A. Denmark, H. V. Weems, and others (FSCA); Tall Timbers Res. Sta, ca. 22 mi. N Tallahassee, M, 2F, 5 Aug VOLUME 97, NUMBER 3 489 1958, W. W. Baker (FSCA) and M, 30 Jul 1973, D. Harris (FSCA). Jefferson Co., Monticello, M, 4F, 29 Aug 1959, H. W. Collins (FSCA); and Big Bend Horti- cultural Lab., 8M, date and collector unknown (FSCA) and 2M, 1971, H. Whit- comb (FSCA). Hamilton Co., Suwannee River St. Pk., White Springs, 2F, 29 May 1958, N. B. Causey (FSCA). Dicellarius bimaculatus lamellidens (Chamberlin), new status Figs. 7-8 Fontaria lamellidens Chamberlin, 1931:78—79.—Attems, 1938:167. Spathoria lamellidens.—Chamberlin, 1939:6, pl. 1, fig. 9. Dicellarius lamellidens.—Chamberlin and Hoffman, 1958:30. Type specimen.— The male holotype is missing from the RVC collection where Chamberlin (1931) and Chamberlin and Hoffman (1958) report its deposition. The type locality is Biloxi, Harrison Co., Mississippi, and since there are no paratypes, the following diagnosis is prepared from males from Ocean Springs in adjacent Jackson County. Diagnosis.—Gonopods in situ with acropodites bent strongly mediad, crossing at about *%4 length and extending beyond each other; prefemoral process short, terminating below level of acropodite division, moderately wide basally and ta- pering to acuminate tip, directed across acropodite stem; acropodite in medial view curved subcaudad, solenomerite overlying tibial process, latter longer and wider than solenomerite, profile visible. Variation. — The degree of caudal curvature of the acropodite, as seen in medial view, varies and is stronger in the Stone County males. Distribution.— Known only from a small area in coastal Mississippi west of the Pascagoula River. It has not been taken west of Long Beach, a suburb of Gulfport- Biloxi, but it could conceivably extend to the Pearl River and possibly even into southeastern Louisiana. Specimens were examined as follows: MISSISSIPPI: Stone Co., between Harrison Co. line and MS hwy. 26, 3M, 21 Jan 1965, N. B. Causey (FSCA). Jackson Co., Ocean Springs, Gulf Hills, 3M, 2F, 2 Oct 1958, and 3M, 1 Jun 1961, N. B. Causey (FSCA); and Gulf Coast Research Lab. nr. Ocean Springs, M, 16 Mar 1957, D. Moore; 2M, F, 1 Oct 1958, collector unknown; M, F, 30 May 1959, N. B. Causey; 2M, F, 27 November 1964, C. Guise; and M, 21 Nov 1965, D. E. Hahn (all FSCA). Harrison Co., Long Beach, F, 20 Mar 1910, collector unknown (RVC). Remarks.—In medial view the acropodite of D. b. lamellidens curves in the opposite direction to that of D. b. fictus (compare Figs. 6 and 8). However, the two forms connect through a series of intermediate stages along the Gulf Coast and therefore represent geographic races of a single species rather than different species. Dicellarius bimaculatus intergrades Intergrade populations of D. bimaculatus occur east of the Pascagoula River in Jackson County, Mississippi. The acropodites curve less than those from Ocean Springs (Fig. 8) and the gap between the distal elements is slightly visible in medial view. This configuration is intermediate between the nominate subspecies and D. b. lamellidens. Specimens were examined as follows: 490 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON MISSISSIPPI: Jackson Co., Pascagoula, M, Sep 1953, E. Waitt (FSCA) and 3M, F, 1 Jan 1954, N. B. Causey (FSCA); and Spanish Fort, 2M, 4F, 15 Jun 1953, collector unknown (FSCA) and 2M, F, 21 Jan 1965, N. B. Causey (FSCA). Dicellarius okefenokensis (Chamberlin) Figs. 9-12 Leptodesmus okefenokensis Chamberlin, 1918:370-371. Dicellarius okefenokensis.—Chamberlin, 1920:97.—Chamberlin and Hoffman, 1958:31. Epeloria nannoides Chamberlin, 1949:101, fig. 24. Type specimens. — Male holotype (MCZ) taken by unknown collector, Dec 1913, on Billy’s Island in the Okefenokee Swamp, Charlton Co., Georgia. Diagnosis.—Prefemoral process not extending across stem of acropodite; latter bent broadly anteriad at midlength, overhanging and extending slightly beyond level of prefemoral process, with torsion, prostatic groove crossing from medial to lateral sides at anterior bend; tibial process overlying and masking most of solenomerite in medial view. Holotype.— Length 22.0 mm, maximum width 4.2 mm, W/L ratio 19.1%, depth/ width ratio 73.8%. Segmental widths as follows: collum 3.4 mm 11th-16th 4.0 2nd 3.6 17th 3.6 3rd 3.9 18th 2.9 4th-10th 4.2 Color in life unknown. Somatic features similar to those of D. b. bimaculatus with following exceptions: Width across genal apices 2.4 mm, interantennal isthmus 0.9 mm. Antennae reaching back nearly to caudal edge of 4th segment, relative lengths of antenno- meres 2>6>5 >4=3>1> 7. Genae with very slight medial impression. Facial setae as follows: Epicranial 3-3, interantennal 1-1, frontal 1-1, genal 3-3, clypeal about 12-12, larbral about 16-16. Collum moderately broad, ends extending slightly beyond those of following tergite. Peritremata extremely thick and conspicuous, enlarged caudad. Sternum of segment 5 with 2 moderately elevated areas between 4th legs and slight recession between 5th legs, 6th sternum with convex recession between both leg pairs, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with slight midline elevation on segments 1 1—17, elevation oriented transversely rather than longitudinally as in other species. Gonopodal aperture elliptical, 1.6 mm wide and 0.5 mm long at midpoint, indented anteriolaterally, sides flush with metazonal surface. Gonopods in situ (Fig. 9, not this specimen) with acropodites curving broadly mediad at midlength and extending beyond anterior margin of aperture, apices overlapping above 6th sternum; prefemoral processes directed anteriomediad, crossing at midlength in midline of aperture, apices extending slightly beyond anterior margin of aperture. Gonopod structure as follows (Figs. 10-11): Prefemoral process wide and long, VOLUME 97, NUMBER 3 491 10 Figs. 9-12. Dicellarius okefenokensis: 9, Gonopods in situ, ventral view of male from Columbia Co., FL; 10, Telopodite of left gonopod of holotype, medial view; 11, The same, lateral view; 12, Telopodite of left gonopod of male from Hernando Co., FL, medial view. Scale line for Fig. 9 = 1.00 mm; line for other Figs. = 1.00 mm for each. about *4 length of acropodite, spatulate basally with sinuous margins narrowing at midlength, apex with several minute teeth directed toward tip of acropodite. Acropodite in form of broadly curved arc with torsion, overhanging and extending slightly beyond level of prefemoral process, prostatic groove crossing to lateral surface proximad, distal division masked in medial and lateral views, located at about 73 length. Distal elements widely separated, diverging continuously with greatest separation apically, gap visible mostly in ventral or dorsal perspectives; solenomerite anterior to, and slightly shorter than, tibial process, sides narrowing smoothly to tip; tibial process overlying and largely obscuring solenomerite in medial view, broader than latter, sides parallel for most of length, apically acute. Prostatic groove arising in pit in prefemur, running along medial surface of ac- ropodite basally, crossing to lateral side at anterior bend, terminating apically on solenomerite. Description of females. —Since the type collection contains no females and none have been encountered in Georgia, the following description has been prepared from females from Baker County, Florida, the closest available to the type locality. Length 23.2 mm, maximum width 4.0 mm, W/L ratio 17.2%, depth/width ratio 72.5%. Agreeing essentially with males in somatic features except paranota more 492 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON strongly depressed, creating appearance of more highly arched body, and sternal elevations greatly reduced barely visible. Cyphopods in situ with opening of valves visible in aperture. Receptacle absent. Valves relatively large, subequal, surface finely granulate. Operculum relatively large, located under dorsolateral corner of valves, surface finely granulate. Variation.—Southern populations vary considerably from those north of Gainesville. The prefemoral process is longer but of the same length relative to the acropodite since the latter is also longer; its width also varies, as it becomes much narrower and more acicular with decreasing latitude. The acropodite bends more abruptly at midlength (Fig. 12), and the solenomerite is longer relative to the tibial process, projecting beyond the latter in medial view. Ecology.— According to the labels in the vials, D. okefenokensis is typically found under logs in moist, predominantly hardwood habitats. The sample from Waycross, Georgia, was discovered under trash in a yard. Distribution. —Southeastern Georgia and peninsular Florida nearly to the level of Tampa. Except for the Waycross sample, all specimens have come from south of the Suwannee River or its headwaters in the Okefenokee Swamp. The known Florida localities all lie north of highway I-4. Specimens were examined as follows: GEORGIA: Ware Co., Waycross, under trash in yard, 2M, 8 Feb 1962, M. Barnette (FSCA). Charlton Co., Billy’s Island, Okefenokee Swamp, M, Dec 1913, collector unknown (MCZ) TYPE LOCALITY. FLORIDA: Nassau Co., Austin Cary Forest, M, 30 Jun 1960, W. J. Platt (FSCA). Baker Co., Macclenny, M, 26 Dec 1960, E. W. Holder (FSCA); and Glen St. Mary, 2M, 3F, 18 Jan 1961, E. W. Holder (FSCA). Columbia Co., Osceola Nat. For., 6M, 15 Feb and 16 Mar 1977, J. R. Wiley (FSCA). Hamilton Co., White Springs, M, F, 1959, A. Williams (FSCA). Suwannee Co., along Rocky Cr. at rd. S-136, M, F, 16 Mar 1977, J. R. Wiley (FSCA). Clay Co., no further locality, 2F, 7 Jan 1961, H. A. Denmark (FSCA). Marion Co., Confederate Cave, M, 19 Feb 1959, R. Cumming (FSCA). Alachua Co., 1 mi. N Newberry, along US hwy. 27, F, 14 Jan 1967, J. E. Lloyd (FSCA); and Gainesville, many MM and FF collected mostly during Oct—Feb from 1933-1965, by about 10 collectors (RVC, MCZ, RLH, & FSCA). Putnam Co., Ocala Nat. For., Johnson Field, F, 21 Jan 1973, R. Skinner (AU). Seminole Co., Sanford, M, 14 Feb 1960, G. W. Desin (FSCA). Hernando Co., Brooksville, M, 25 Jan 1960, L. Hill (FSCA); and 10 mi. S Floral City, along US hwy. 41, 2M, 23 Dec 1973, W. A. Shear (RLH). Pasco Co., Dade City, M, 15 Feb 1972, K. C. Lowery (FSCA), and M, 14 Jun 1983, R. M. Shelley and J. L. Staton (NCSM A4160). Remarks. — Dicellarius okefenokensis is the only allopatric species in the genus. Its range does not overlap that of any other species, though it abuts D. b. fictus in the west. The acropodite of D. okefenokensis is unique to both the genus and tribe in displaying two features typically found and convergent in the tribe Apheloriini: the broadly curved acropodite that overhangs the prefemoral process in medial view, and torsion, which causes the prostatic groove to cross from medial to lateral sides at 3 length. Just as the acropodite in D. b. lamellidens is turned counterclockwise so that the solenomerite overlies the tibiotarsus in medial view, the acropodite of D. okefenokensis probably evolved from eastern populations of D. bimaculatus in which the reverse, or clockwise, twisting occurred. Such rotation VOLUME 97, NUMBER 3 493 would place the tibiotarsus over the solenomerite in medial view, and an early stage in this torsion is apparent in some males of D. b. fictus, where the edge rather than the side of the solenomerite is visible medially. Thus, I think D. okefenokensis represents an eastern population of D. bimaculatus in which the distal half of the acropodite has turned about 90° clockwise, just as D. b. lamel- lidens represents ones in which 90° counterclockwise rotation has occurred. The latter still links with other populations of D. bimaculatus by intermediate or intergrade forms, and consequently is only a geographic race. Connections between D. okefenokensis and D. bimaculatus have disappeared, however, and the former has achieved reproductive isolation as evidenced by the torsion, the broadened curvature of the acropodite, and reduced body size. Dicellarius talapoosa (Chamberlin) Diagnosis.—Acropodite with slight but continuous curve; distal elements nar- rowly separated or touching, continuing general curvature of acropodite, never overlying one another in medial or lateral views; solenomerite acicular, with sides tapering smoothly and continuously to subacuminate tip; tibial process acicular, shorter than solenomerite; prefemoral process angling across stem of acropodite, relatively short, extending at most to base of distal elements, tip overlying tibial process. Remarks. —Dicellarius talapoosa occurs inland from D. bimaculatus in Ala- bama and Georgia. Its range abuts that of D. b. fictus in the south, and it is sympatric in parts of the range with both D. atlanta and D. sternolobus. In the material available to me a consistent difference exists in the degree of separation of the distal elements in populations east and west of the Coosa River. In the eastern populations they are contiguous for most of their lengths, whereas in the western ones they are narrowly, but completely, separated. One sample from St. Clair County, an intermediate geographical position, displays an intermediate degree of separation that I consider an intergrade. This situation is a much clearer indication of geographic races than that of D. bimaculatus, and since Chamberlin’s type is the eastern form, it becomes the nominate subspecies and a new subspecific name is needed for the western form. I hereby propose D. t. separandus to signify the greater separation of the solenomerite and tibial process. Dicellarius talapoosa talapoosa (Chamberlin), new combination, new status Figs. 13-14 Epeloria talapoosa Chamberlin, 1939:3, pl. 1, fig. 1.—Chamberlin and Hoffman, 1958:33. Type specimens.— Male holotype and one male paratype (RVC) collected by R. V. Chamberlin, 29 Jul 1910, from Tallapoosa, Haralson Co., Georgia. Chamberlin (1939) reported that a female allotype was taken in addition to the holotype, which is what the sample label says, but the vial actually contains two males, one of which must be a paratype. Diagnosis.— Distal elements closely appressed together, touching for most of length except for slight basal gap. Holotype. —Body badly fragmented and unmeasurable, segmental widths taken 494 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON from male from Cleburne Co., Alabama, the closest whole specimen to the type locality, and listed under variation. Color in life unknown. Somatic features similar to those of D. b. bimaculatus, with following excep- tions: Width across genal apices 4.4 mm. Antennae extending to caudal edge of 3rd tergite, relative lengths of antennomeres 6 > 5 > 3=4=2 > 1 > 7. Facial setae as in D. b. bimaculatus except subantennal 1-1, genal 2-2, and clypeal about 12- 12% Collum moderately broad, ends extending well beyond those of following tergite. Fourth sternum without process; 5th sternum with two elevated, medially co- alesced flattened areas between 6th legs and deep convex recession between 7th legs. Postgonopodal sterna with midline elevations in form of narrow, longitudinal ridges, distinct on segments 10-15. Gonopodal aperture 2.4 mm wide and 0.8 mm long at midpoint, indented slightly anteriolaterally, sides flush with metazonal surface. Gonopods in situ (Fig. 13, not this specimen) with acropodites projecting anteriad from aperture, ex- tending well beyond anterior margin of aperture, with apices crossing between 7th legs, prefemoral processes also overlapping and extending beyond aperture. Gonopod structure as follows (Fig. 14): Prefemur moderate in size, with relatively short, acicular prefemoral process extending in medial view across stem of ac- ropodite approximately half the length of the latter, terminating near level of distal division of latter. Acropodite curving slightly caudad, sides narrowing gradually and divided at about 7% length into distal elements; latter subspiniform and apically acute, closely appressed together, touching for most of length, with slight basal separation visible in subanterior view; solenomerite slightly longer and more acute. Prostatic groove arising in pit in prefemur, running along medial side of acropodite stem to level of distal elements, curving toward anterior margin of acropodite and entering solenomerite, opening apically. Male paratype.—The male paratype agrees with the holotype in all particulars and is also fragmented and unmeasurable. Description of Females.—Since the type collection contains no females, the following description was prepared from specimens from Randolph County, Al- abama, the closest females available to the type locality. Length 42.1 mm, max- imum width 8.8 mm, W/L ratio 20.9%, depth/width ratio 72.7%. Agreeing es- sentially with holotype in most somatic feature, except paranota more strongly depressed, giving appearance of more highly arched body, and sternal elevations greatly reduced, barely detectable. Cyphopods in situ with valvular openings visible in aperture; valves large, subequal, surface finely granulate. Receptacle absent. Operculum small, located on lateral side of valves, surface finely granulate. Variation. — The sample from Cheaha State Park, Cleburne Co., Alabama (NCSM => Figs. 13-16. Dicellarius talapoosa: 13-14, D. t. talapoosa. 13, Gonopods in situ, ventral view of male from Cleburne Co., AL; 14, Telopodite of left gonopod of holotype, medial view; 15-16, D. t. separandus. 15, Gonopods in situ, ventral view of paratype; 16, Telopodite of left gonopod of holotype, VOLUME 97, NUMBER 3 495 4 Yf | 16 18 ee / \\ Ay oe =e = ~ \ aS o3 y ° 7 ee medial view. Figs. 17-18, Dicellarius atlanta: 17, Gonopods in situ, ventral view of male from Peach Co., GA; 18, Telopodite of left gonopod of holotype, medial view. Scale lines for Figs. 13, 15, and 17 = 1.00 mm; line for other figs. = 1.00 for Fig. 14, 1.14 mm for 16, and 0.80 mm for 18. 13 5K 496 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A3117), is the closest to the type locality with whole males, one of whose mea- surements are as follows: Length 34.8 mm, maximum width 7.7 mm, W/L ratio 22.1%, depth/width ratio 66.2%. Segmental widths as follows: collum 5.8 mm 6th-13th 7.7 2nd 6.3 14th 6.8 3rd 7.1 15th 6.6 4th-Sth 7.3 16th 6.3 17th 5.8 18th 5.1 The gonopods of this subspecies are quite uniform. The prefemoral processes tend to be longer and the acropodites more distinctly curved in western popu- lations. However, the distal elements are always closely appressed together, except for the slight gap at their bases, and their lengths relative to each other and to the overall acropodite do not change. Distribution.—The range of D. t. talapoosa lies in the Piedmont Plateau and Fall Zone of west-central Georgia and east-central Alabama, east of the Coosa River. It crosses the Coosa River in the north and extends onto Lookout Mountain along the northern Alabama-Georgia boundary. Specimens were examined as follows: GEORGIA: Dekalb Co., Oglethorpe Univ., M, date unknown, A. L. Cain (FSCA). Haralson Co., Tallapoosa, 2M, 29 Jul 1910, R. V. Chamberllin (RVC) TYPE LOCALITY; and Bremen, M, 29 Jul 1910, R. V. Chamberlin (RVC). ALABAMA: Dekalb Co., 4.4 mi. SE Mentone, along AL hwy. 117 at Anna Branch Cr., M, F. 15 Apr 1978, R. M. Shelley and R. E. Ashton (NCSM A18442). Etowah Co., 1.5 mi. W Hokes Bluff, along Coosa R., 2M, 28 Feb 1961, L. Hubricht (RLH). Cleburne Co., 3.4 mi. N Shoal Cr. Camping Area, M, 28 Oct 1960, L. Hubricht (RLH); 4 mi. W Heflin, M, 5 Mar 1961, L. Hubricht (RLH); and summit Cheaha Mtn., Cheaha St. Pk., M, 10 Jun 1953, L. Hubricht (RLH) and 13 M, F, 20 May 1980, R. M. Shelley (NCSM A3117). Talladega Co., 5 mi. W Sylacauga, M, 3 Mar 1973, W. Redmond (AU). Lee Co., 3 mi. N Auburn, along AL hwy. 143 at Saugahatchee Cr., M, 7 Mar 1973, T. French (AU). Randolph Co., 0.7 mi. W Wadley, F, 1 Oct 1960, L. Hubricht (RLH). Macon Co., Tuskegee Nat. For., M, 2F, 14 May 1973, R. B. Little (AU); and Tuskegee, Tuskegee Institute Campus, 2M, 14 Mar 1978, S. B. Ruth (NCSM A3891). Remarks.—I collected 13 males on the roadway near the summit of Cheaha Mountain early in the morning of 20 May 1980. The animals must have ventured forth in the cool weather of the preceding night, and six had been freshly decap- itated by an unknown predator. The head and first three to four segments had been removed, but the bite stopped short of segment 5 and the anteriormost defensive glands. I have seen this phenomenon twice before, with individuals of Croatania catawba Shelley from Chester Co., South Carolina, and Gyalostethus monticolens (Chamberlin) from the Great Smoky Mountains National Park, Swain Co., North Carolina. In all instances the segments seem to have been removed by a single, clean bite, as there is no evidence of continuous chewing. At Cheaha Mountain the millipeds were exposed on the pavement away from protective cover, but in the other sites they were in forests lying on top of the litter. VOLUME 97, NUMBER 3 497 Dicellarius talapoosa separandus, new subspecies Figs. 15-16 Type specimens.—Male holotype (NCSM A3132) and 4 male and 4 female paratypes collected by R. M. Shelley, 23 May 1980, at Rickwood Caverns State Park, Blount Co., Alabama. Male and female paratypes deposited in FSCA. Diagnosis. — Distal elements subparallel but not touching, separated for entire lengths, diverging apically. Variation. — The degree of separation of the distal elements varies, but otherwise all males are closely similar to the holotype. Distribution.— This subspecies occurs generally west of the Coosa River, in the Cumberland Mountains around Birmingham and the Appalachian Plateau. It is sympatric and syntopic with 7. /ampra in Tuscaloosa. Specimens were examined as follows: ALABAMA: Dekalb Co., 3.7 mi. NW Collinsville, along AL hwy. 68 at Little Cr., 2.5 mi. W ject. I-59, M, 15 Apr 1978, R. M. Shelley and R. E. Ashton (NCSM A1841). Morgan Co., 8 mi. S Decatur, along Flint Cr., M, 24 May 1980, R. M. Shelley (NCSM A3143). Blount Co., Rickwood Caverns St. Pk., 5M, 4F, 23 May 1980, R. M. Shelley (NCSM A3132) TYPE LOCALITY. Jefferson Co., Birming- ham, 3M, 16-25 Nov 1958, collector unknown (FSCA). Tuscaloosa Co., Tus- caloosa, M, 2 juvs., 29 Mar 1948, G. Ball (RLH). Remarks.—The in situ gonopodal configuration of this subspecies is the same as that of the nominate, except the wider gap between the distal elements allows them to crisscross (Fig. 15, of paratype). The acropodites simply overly each other in D. t. talapoosa (Fig. 13). The sample from Birmingham had the following note about the smell of the defensive secretion: “odor good, strong, pineapple, with dash of shoe polish; not too unpleasant, sweetish.”’ Dicellarius talapoosa intergrades In two males from St. Clair County, the basal gap between the distal elements is larger than in D. t. talapoosa, but they still touch distally. I regard these spec- imens as intergrades, and they are indicated by the northernmost “X” in Fig. 29. They were taken about 5 miles west of the Coosa River, which generally forms the boundary between the subspecies. The data are as follows: ALABAMA: St. Clair Co., Steele, 2M, 4 Apr 1948, H. E. Evans (RLH). Dicellarius atlanta (Chamberlin), new combination Figs. 17-18 Epeloria atlanta Chamberlin, 1946b:151-—152, fig. 7.—Chamberlin and Hoffman, 1958:32. Type specimens.— Male holotype and female allotype (RVC) collected by P. W. Fattig, 1 Dec 1945, from Atlanta, Georgia. The label in the vial states 1946, but this must be incorrect since it would not have allowed publication in that year. Neither the publication nor the label in the vial gives the county, but since Atlanta was much smaller in the 1940’s and its center is in Fulton Co., the type specimens were probably collected there rather than in Dekalb Co. 498 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Diagnosis.—Acropodite relatively straight, sublinear; distal elements widely separated, either continuous with sublinear acropodite or very slightly curved, never overlying one another in medial or lateral views; solenomerite acicular, sides tapering smoothly and continuously to subacuminate tip; tibial process acicular, subequal in length to solenomerite; prefemoral process straight, not angling across acropodite stem but lying over or near anterior margin, relatively long, extending well beyond base of distal elements with tip either over soleno- merite or on anterior side of latter. Holotype.—Length 30.3 mm, maximum width 6.0 mm, W/L ratio 19.8%, depth/ width ratio 66.4%. Segmental widths as follows: collum 5.0 mm 16th 5.8 2nd—4th 5.5 17th 5.5 5th 5.8 18th 4.6 6th—-15th 6.0 Color in life unknown. Somatic features similar to those of D. b. bimaculatus, with following excep- tions: Width across genal apices 4.1 mm, interantennal isthmus 1.3 mm. Antennae extending back to caudal edge of 3rd tergite, relative lengths of antennomeres 6 > 5 >4=3=2> 1 > 7. Genae with distinct, linear, diagonal impression. Facial setae as in D. b. bimaculatus except genal 2-2, clypeal about 11-11, labral about 18-18. Collum moderately broad, ends extending slightly beyond those of following tergite. Sternum of segment 5 with small elevated area between 4th legs and depression between Sth legs; 6th sternum with wide, flattened, elevated area between 6th legs and deep, convex recession between 7th legs. Postgonopodal sterna with midline elevations generally large and distinct, acute and subconical, largest on segments 11-14, lower and more rounded before and after. Gonopodal aperture ovoid, 2.4 mm wide and 1.0 mm long at midpoint, in- dented anteriolaterad, sides nearly flush with segmental surface. Gonopods in situ (Fig. 17, not this specimen) with acropodites projecting anteriad from aperture, subparallel, not overlapping or crossing, extending beyond anterior margin of aperture to between 7th legs. Gonopod structure as follows (Fig. 18): Prefemur moderate, with long, acicular prefemoral process extending well beyond level of distal division of acropodite on anterior side. Acropodite extending sublinearly from prefemur, bent slightly caudad at level of division, sides parallel, not tapering, divided roughly at % length. Distal elements well separated, diverging apically; tibial process apically blunt; solenomerite slightly longer, tapering distad to sub- acuminate tip. Prostatic groove arising in pit in prefemur, running along medial surface of acropodite to level of division, crossing to anterior side and opening terminally on solenomerite. Female allotype.— Length 31.3 mm, maximum width 5.4 mm, W/L ratio 17.4%, depth/width ratio 77.8%. Agreeing closely with male except paranota more strong- ly depressed, creating appearance of more highly arched body, and sternal ele- vations greatly reduced, barely detectable. VOLUME 97, NUMBER 3 499 Cyphopods in situ with openings of valves visible in aperture; valves large, subequal, surface finely granuate. Receptacle absent. Operculum relatively large, located under lateral end of valves, surface finely granulate. Variation.—The gonopods of D. atlanta are uniform. The acropodite always extends linearly from the prefemur, and the solenomerite is always slightly longer than the tibial process. The distal elements continue in the axis of the acropodite stem, although they lean slightly caudad in some males; they converge apically in specimens from Paulding County, and the solenomerite is bowed outward basally away from the tibiotarsus. In medial view the prefemoral processes lean anteriad basally away from the acropodite then curve back toward the latter at midlength. Distribution. — Dicellarius atlanta is the only species occurring in the southern Blue Ridge Province (Appalachian Mountains) and ranges from western North Carolina to central Georgia and eastern Alabama. Its range overlaps those of D. b. fictus and D. t. talapoosa in central Georgia, and it is sympatric and syntopic with T. bifida in Dawson and Houston Counties, Georgia. The range spans the Chattahoochee River and the headwaters of the Ocmulgee River, and may also traverse the Savannah River into Oconee County, South Carolina. NORTH CAROLINA: Graham Co., 2 mi. NW Milltown, nr. Snowbird Picnic Area, 2M, 10 May 1961, L. Hubricht (RLH). GEORGIA: Rabun Co., Tallulah Falls, M, 28 Apr 1943, W. Ivie (RVC). Ha- bersham Co., Clarkesville, F, 4 Apr 1943, collector unknown (RVC). Dawson Co., Amicalola Falls St. Pk., 2M, 2F, 16 Apr 1978, R. M. Shelley and R. E. Ashton (NCSM A1855). Jackson Co., NW of Pendergrass, M, 23 Apr 1943, W. Ivie (RVC). Clarke Co., Athens, Univ. of GA campus, F, 27 Feb 1944, E. P. Odum (RVC); and 9 mi. W Athens, M, 6 May 1961, L. Hubricht (RLH). Oconee Co., Watkinsville, M, 26 Mar 1959, M. Pitken (FSCA). Fulton Co., Atlanta, M, F, 1 Dec 1946, P. W. Fattig (RVC), and Ft. McPherson, 2F, 9 Dec 1943, D. E. Beck (RVC). Dekalb Co., Atlanta, 2M, 2F, 7 Oct 1960, E. Davidson (FSCA); and Stone Mountain, M, F, 20 Oct 1946, and M, 6 Apr 1947, P. W. Fattig (RVC). Douglas Co., Sweetwater Creek St. Pk., M, 21 Nov 1977, R. M. Shelley (NCSM A1796). Paulding Co., Dallas, M, 30 Mar 1947, P. W. Fattig (RVC). Clayton Co., 6.4 mi. S Corely, M, 4 Dec 1960, L. Hubricht (RLH). Butts Co., Indian Springs St. Pk., M, 4F, 20 Nov 1977, R. M. Shelley (NCSM A1789). Bibb Co., Macon, M, 4 Mar 1961, L. Hubricht (RLH). Peach Co., 2 mi. W Fort Valley, along GA hwy. 96, 4M, F, 12 Apr 1978, R. M. Shelley (NCSM A1834). Houston Co., Perry, 50M, OF, 26-28 Dec 1982, J. A. Payne (RLH). Meriwether Co., 3.5 mi. S Woodbury, M, 11 Mar 1961, L. Hubricht (RLH). Harris Co., Franklin D. Roosevelt St. Pk., M, 2 May 1983, R. M. Shelley and P. B. Nader (NCSM A4083). Muscogee Co., Columbus College, M, 2 Nov 1977, G. E. Stanton (CC). ALABAMA: Lee Co., Auburn, M, Mar 1966, H. B. Cunningham (AU); Prath- er’s Pond nr. Auburn, 8M, 2F, Mar 1973, T. French (AU); and 3 mi. N Auburn, along AL hwy. 143 at Saugahatchee Cr., 3M, Apr 1973, T. French (AU). Remarks.—As shown in Figs 16 and 18, the gonopod of D. atlanta is closely similar to that of D. t. separandus, the main differences being the orientation of the prefemoral process, the degree of separation of the distal elements, and the curvature of the acropodite. In D. t. separandus the prefemoral process angles across the stem of the acropodite, whereas in D. atlanta it is located entirely on 500 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the anterior side. The distal elements are subequal and more widely separated in D. atlanta, and the acropodite is more linear in this species. These similarities suggest that D. atlanta originated as a subspecies of D. talapoosa, and has since achieved reproductive isolation as evidenced by the range overlap with D. t. talapoosa. Dicellarius sternolobus Loomis Figs. 19-23 Dicellarius sternolobus Loomis, 1969:247-248, figs. 5—7. Type-specimen. — Male holotype (NMNH) taken by unknown collector, 27 Oct 1944, from Alexander City, Tallapoosa Co., Alabama. The fragmented, immature paratype in the FSCA does not appear to be this species. Diagnosis. —Sterna of segments 4 and 5 with large, ventrally directed processes, usually longer than widths of adjacent coxae; solenomerite overlying tibial process in medial view, much narrower than latter with sides tapering smoothly and continuously to subacuminate tip; lateral surface of tibial process irregular, with several grooves and ridges. Holotype. — Length 36.8 mm, maximum width 6.8 mm, W/L ratio 18.5%, depth/ width ratio 75.0%. Segmental widths as follows: collum 5.4 mm 7th-15th 6.8 2nd 6.0 16th 6.5 3rd—Sth 6.2 17th 5.9 6th 6.6 18th 5.4 Color in life unknown. Somatic features similar to those of D. b. bimaculatus, with following excep- tions: Width across genal apices 4.5 mm, interantennal isthmus 1.2 mm. Antennae reaching back to middle of 4th tergite, relative length of antennomeres 2 > 3 > 6=5=4> 1 > 7. Genae with distinct medial depression. Facial setae as in D. b. bimaculatus except subantennal 1-1, genal 2-2, clypeal about 14-14, labral about 20-20. Collum broad, ends extending slightly beyond those of following tergite. Dorsum smooth and polished, glossy. Sternum of segment 4 with large, apically divided process between 3rd legs, greater in length than widths of adjacent coxae (Fig. 19); Sth sternum with similar process between 4th legs, longer than widths of adjacent coxae, and with slight recession between Sth legs (Fig. 20); 6th sternum with two low, transverse ridges between 6th legs and depression between 7th, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with lowly rounded, faint midline elevations no- ticeable on segments 8-13, undetectable on remaining segments. Gonopodal aperture ovoid, 2.1 mm wide and 1.0 mm long at midpoint, in- dented anteriolaterad, sites flush with metazonal surface. Gonopods in situ (Fig. 21, not this specimen) with acropodites bending mediad distad, apices overlap- ping; prefemoral processes angling mediad but not touching. Gonopod structure VOLUME 97, NUMBER 3 501 Figs. 19-23. Dicellarius sternolobus: 19, Sternum of segment 4 of holotype, caudal view; 20, Sternum of segment 5 of the same, caudal view; 21, Gonopods in situ, ventral view of male from Shelby Co., AL; 22, Left gonopod of holotype, medial view; 23, Telopodite of the same, lateral view. Scale line for Fig. 21 = 1.00 mm; line for other Figs. = 0.81 mm for Fig. 19, 1.00 mm for Figs. 20 and 22-23. as follows (Fig. 22—23): Prefemoral process acicular, curved slightly at midlength and angling across stem of acropodite to level of distal division. Acropodite curved gently caudad at midlength, overhanging and extending well beyond level of prefemoral process; stem widening basally then narrowing gradually to level of division at '4 length. Distal elements subparallel and subequal in length, narrowly separated, gap obscured in medial and lateral views but visible from dorsal and ventral perspectives; solenomerite located medial to tibial process and overlying it in this perspective, narrow and acicular, continuing general curvature of ac- ropodite and tapering smoothly to acuminate tip; tibial process broad, spatulate in shape but with grooves and ridges forming several smooth edges, one irregularly scalloped, obscuring solenomerite in lateral view, continuing general curvature of acropodite, widening distad, anterior margin apically flattened and forming acu- minate tip with caudal margin. Prostatic groove arising in pit in prefemur, running along medial sides of acropodite and solenomerite, opening terminally. 502 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Description of female.—There are no samples of D. sternolobus with both males and females from the same locality, so there are no certifiable females of this species. The one described here was taken about 9 mi. NW of the locality in Talladega Co. where a male was collected and is identified as D. sternolobus instead of the proximal D. talapoosa on the basis of size, since D. sternolobus is much larger than the latter. Length 39.9 mm, maximum width 8.4 mm, W/L ratio 21.0%, depth/width ratio 69.0%. Agreeing essentially with holotype in somatic features with following exceptions: Paranota more strongly depressed, creating appearance of more highly arched body; without sternal projections comparable to pregonopodal processes of male; postgonopodal sternal elevations of male absent, female sterna flat. Cyphopods in situ with valvular openings visible in aperture. Receptacle absent. Valves relatively large, subequal, surfaces finely granuate. Operculum relatively large, located under lateral end of valves. Variation. — There is considerable variation in the size of the sternal projections, one of the prime diagnostic features of the species. The one on segment 4, between the 3rd leg pairs, is highly variable and largest on the holotype. In the Shelby County males, this lobe is reduced to two widely separated knobs that are subequal in length to the adjacent coxal widths. The process between the anterior legs of segment 5 (4th legs) is less variable but still smaller than that of the holotype in the Talladega County male and the one from near Montevallo. On the gonopods the prefemoral processes are relatively uniform, some curving less than others. The acropodite of the Talladega County male is more linear and massive, and the solenomerite is twisted so that the groove crosses to the lateral side about midlength of the structure. On the non-type males, the tibial process is produced apically into a subacute tip. Ecology.—The male I collected in Shelby County was taken in a valley in a mesic hardwood forest not far from a creek. Distribution.—Known only from six localities in the southern extremity of the Ridge and Valley Province in central Alabama. Most sites lie between US highways 280 and I-65, but the species probably occurs in adjacent counties in all directions. Specimens were examined as follows: ALABAMA: Shelby Co., Oak Mountain St. Pk., M, 22 May 1980, R. M. Shelley (NCSM A3130); and 7.3 mi. NW Montevallo, M, 29 Dec 1962, collector unknown (FSCA). Coosa Co., 10 mi. N Rockford, M, F, 20 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4034). Talladega Co., 2.8 mi. NW Sylacauga, 2F, 6 May 1954, L. Hubricht (RLH); and 6 mi. SE Sylacauga, M, 7 May 1975, L. Hubricht (RLH). Tallapoosa Co., Alexander City, M, 27 Nov 1944, collector unknown (NMNH) TYPE LOCALITY. Remarks.—Loomis (1969) was correct in stating that D. sternolobus was the largest species of Dicellarius. However, his remark about a sternal lobe between the 6th legs is somewhat misleading. This structure is more of a narrow transverse ridge, divided medially, than a ventrally produced, apically tapering lobe. The affinities of D. sternolobus are unclear, but as in D. b. lamellidens, the acropodite curves caudomediad in situ and the solenomerite overlies the tibio- tarsus in medial view. Perhaps D. sternolobus also evolved from a form of D. bimaculatus that became reproductively isolated in central Alabama, in similar fashion to the evolution of D. okefenokensis in southeastern Georgia and Florida. VOLUME 97, NUMBER 3 503 Genus Thrinaxoria Chamberlin and Hoffman Thrinaxoria Chamberlin and Hoffman, 1950:4; 1958:51. Type-species.— Fontaria lampra Chamberlin, 1918, by original designation. Description.—A genus of moderate-size xystodesmids with the following char- acteristics: Body composed of head and 20 segments in both sexes. Head of normal ap- pearance, moderately smooth and polished. Antennae moderately slender, be- coming progressively more hirsute distally, with 4 conical sensory cones on ul- timate article; no other sensory structures apparent. Facial setae with or without epicranial, frontal, and genal series, with clypeal, and labral series; without in- terantennal and subantennal series. Terga finely granulate, not polished; paranota moderately coriaceous. Collum moderately broad, ends subequal to those of following tergite. Paranota flattened to moderately depressed, interrupting slope of dorsum, caudolateral corners rounded on anteriormost segments, becoming blunt in midbody regions and pro- gressively more acute posteriorly. Peritremata distinct, strongly elevated above metazonal surface; ozopores located caudal to midlength, opening dorsolaterad. Prozonites smaller than metazonites; strictures moderately distinct, slightly cos- tulate. Caudal segments normal for family. Sides of metazonites granulate, without grooves or impressions. Gonapophyses broad, apically expanded. Pregonopodal sterna modified as follows: that of seg- ment 4 with minute projection, much shorter than widths of adjacent coxae; that of segment 5 with small variable lobes or low transverse ridges between 4th legs and slight impression between Sth legs; that of segment 6 with impressions between both leg pairs. Postgonopodal sterna mildly hirsute, with or without low midline elevations. Coxae without projections; prefemoral spines relatively long and sharp- ly pointed; tarsal claws slightly bisinuate. Hypoproct broadly rounded; paraprocts with margins slightly thickened. Gonopodal apertures ovoid to elliptical. Gonopods in situ with apices criss- crossing in midline over 6th sternum. Coxae moderate in size, without apophyses, connected by membrane only, no sternal remnant. Prefemora moderate, with variable prefemoral processes arising on anterior or anteriomedial sides, either angling across acropodite stem or extending linearly along anterior margin. Ac- ropodites moderately thick and robust, configurations variable, either bent strong- ly mediocaudad at midlength (level of distal division), slightly curved, or nearly linear; divided near '2 length into variable solenomerite and tibial process, con- tinuous or discontinuous with axis of acropodite. Prostatic grooves arising in pits in bases of prefemora, running entirely along medial sides of acropodites onto solenomerites, opening terminally. Cyphopodal apertures elliptical, encircling 2nd legs, sides slightly elevated above metazonal surfaces, without pleurotergal lobes on anteriolateral corners. Cypho- pods in situ located lateral to 2nd legs, variably positioned in apertures. Recep- tacles varying from remnant to moderate structure cupped around part of valves, ‘surfaces irregular, with slight lobes. Valves moderate and subequal, variously oriented in aperture, surfaces finely granulate. Opercula large, located under lateral end of valves. 504 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Distribution. — Northwestern Louisiana to central Georgia, ranging northward into southwestern North Carolina (Blue Ridge Province) and west-central and southeastern Tennessee, and southward into the Coastal Plains of Alabama and Georgia. The Tennessee River is a distributional barrier in Tennessee and Ala- bama, as T. Jampra has not been encountered in the Nashville Basin and Highland Rim Provinces on the “‘inside”’ of the loop as has Pachydesmus (Hoffman 1958). This barrier effect of the Tennessee River is interesting since the genus spans many large rivers, most notably the Mississippi. Species. —Two, distinguished by the characters set forth in the following di- agnoses. Thrinaxoria lampra (Chamberlin) Figs. 24-26 Fontaria lampra Chamberlin, 1918:371-372. Thrinaxoria lampra.—Chamberlin and Hoffman, 1950:4; 1958:52.—Causey, 1963:77. Zinaria aberrans Chamberlin, 1942:4, pl. 1, fig. 7. Type-specimens.— Male paratypes (RVC, RLH) taken by K. P. Schmidt, 9 Mar 1915, from Creston, Natchitoches Par., Louisiana. The male holotype, collected at this locality on 26 Mar 1915 (Chamberlin 1918) and reported to be in the MCZ (Chamberlin and Hoffman 1958), is lost, since it is not at Harvard nor in the Chamberlin collection. Two female paratypes from Creston, supposedly of T. lampra, are in the MCZ, but they were taken on 5 Mar 1915 the same date that the type of Pachydesmus clarus (Chamberlin) was collected there (Chamberlin 1918). Since the features of females of 7. Jampra and P. clarus are nearly identical, it is impossible to tell which species these females belong to. Diagnosis.—Prefemoral process short, angling across stem of acropodite; distal elements bent abruptly caudomediad, discontinuous with general curvature of acropodite; cyphopods with moderate receptacle. Paratype. — Length 35.5 mm, maximum width 7.2 mm, W/L ratio 20.3%, depth/ width ratio 61.1%. Segmental widths as follows: collum 6.2 mm 16th 6.9 2nd 6.8 17th 6.1 3rd 7.0 18th 4.6 4th-15th 7.2 Color in life unknown. Head capsule smooth, polished; width across genal apices 4.2 mm, interantennal isthmus 1.6 mm; epicranial suture thin but distinct, terminating above interan- tennal region, not bifid. Antennae moderately long, reaching back to middle of 4th segment, relative lengths of antennomeres 2 > 3 > 4=5=6> 1 > 7. Genae not margined laterally, with distinct, linear, central impressions, ends broadly rounded and projecting slightly beyond adjacent cranial margin. Facial setae as follows: Epicranial 2-2, interantennal absent, frontal 1-1, genal 2-2, clypeal about 12-12, labral about 22-22. VOLUME 97, NUMBER 3 505 Collum relatively broad, ends not extending beyond those of adjacent tergite. Dorsum finely granulate, not polished. Paranota relatively flattened, interrupting slope of dorsum and subparallel to substrate, anterior corners rounded through segment 5, becoming blunt and progressively more acute posteriorly. Peritremata distinct, strongly elevated above paranotal surface. Ozopores located caudal to midlength of peritremata, opening dorsolaterad. Sides of metazonites dull and granulate but without noticeable grooves or impressions. Strictures sharp, distinct, prozona elevated slightly above metazona. Sternum of segment 4 with minute projection; that of segment 5 with two low, widely separated, transverse ridges between 4th legs and with slight impression between Sth legs (Fig. 24); 6th sternum with distinct impressions between both leg pairs. Postgonopodal sterna flattened, without trace of midline elevations; caudal margins indented medially. Gonopodal aperture ovoid, 2.2 mm wide and 1.2 mm long at midpoint, without anteriolateral indentations, sides flush with metazonal surface. Gonopods in situ (Fig. 25, not this specimen) with acropodites projecting anteriad and bending mediad, apices crisscrossing in midline over 6th sternum; prefemoral processes extending mediad and overlapping at midlength, tips projecting just beyond an- terior margin of aperture. Gonopod structure as follows (Fig. 26): Prefemoral process short, terminating well below distal division of acropodite, angling across stem of acropodite and directed toward distal extremity of tibial process. Acro- podite bent sharply mediocaudad at level of distal division, distal elements dis- continuous with curvature of acropodite, overhanging and extending well beyond level of prefemur; stem relatively narrow and parallel-sided proximal to bend. Distal elements widely separated by subovoid space, converging apically, directed generally mediocaudad; tibial process slightly longer and more proximal, with slight basal curve, flattened distally and produced into blunt tip; solenomerite flattened basally, sides roughly parallel then tapering and curving apically to blunt tip, tip directed toward tibial process. Prostatic groove arising in pit in prefemur, running entirely along medial side of acropodite onto solenomerite, opening ter- minally. Female topotype.— Length 35.0 mm, maximum width 6.3 mm, W/L ratio 18.0%, depth/width ratio 76.2%. Agreeing essentially with males in somatic features except paranota much more strongly depressed, angling sharply ventrad, creating appearance of more highly arched body; without trace of sternal elevations. Cyphopods in situ with corner of valves and receptacle visible in aperture. Receptacle cupped around medial end of valves, surface moderately irregular and rugulose, with slight lobes. Valves moderate and subequal, oriented subperpen- dicularly to body axis in aperture, surface finely granulate. Operculum large, located under lateral end of valves. Variation.—The sternal ridges between the 4th legs of segment 5 vary and become lobe-like in Alabama, resembling their condition in T. bifida. The pre- femoral processes of 7. Jampra are all short as in the holotype, in contrast to the variation in the congener. Distribution.— Thrinaxoria lampra is known from two allopatric populations, a fairly continuous one in northwestern Louisiana and one with widely separated samples from west-central Tennessee to southwestern Alabama. In Alabama it occurs sympatrically and syntopically with D. t. separandus in Tuscaloosa and 506 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON D. b. bimaculatus along the Tombigbee River in Washington County. The Ten- nessee River constitutes a range boundary because the known localities are all on the ‘‘outside” of the loop formed by the river as it passes through Alabama and northward into western Tennessee. The species can be listed as probable for eastern Texas and southwestern Arkansas, since Miller County, Arkansas, and Cass, Mar- ion, Harrison, Panola, and Shelby counties, Texas, are only a few miles from known localities in Caddo, Desoto, and Sabine Parishes, Louisiana. Thus, T. lambra becomes the third species whose occurrence in northeastern Texas can be predicted from distributions in adjacent states, the others being Pleuroloma fla- vipes Rafinesque and Auturus I. louisianus (Chamberlin) (Shelley 1980, 1982). Specimens were examined as follows: LOUISIANA: Caddo Par., 5 mi. NW Shreveport, M, 13 Apr 1936, L. Hubricht (RVC). DeSoto Par., 11 mi. N Mansfield, 2M, 25 Aug 1966, R. E. Tandy (FSCA). Sabine Par., 8 mi. W Noble, M, 5 Apr 1969, D. C. Marizot and R. M. Blaney (FSCA). Red River Par., 4 mi. NW Coushatta, M, 10 Apr 1971, P. Kimmick (FSCA). Natchitoches Par., no further data, 2F, 22 Mar 1957 and 20 Apr 1961, collector unknown (FSCA); Moreland, M, 16 Oct 1954, collector unknown (FSCA); and Creston, 2M, F, unknown date in 1915, K. P. Schmidt (RVC), M, 9 Mar 1915, K. P. Schmidt (RLH), and 2M, 2F, 26 Mar 1915, K. P. Schmidt (MCZ) TYPE LOCALITY. TENNESSEE: Henderson Co., Natchez Trace St. Pk., 2M, 3F, 26 May 1980, R. M. Shelley (NCSM A3159). MISSISSIPPI: Tishomingo Co., Tishomingo St. Pk., M, 15 May 1964, L. Hu- bricht (RLH). ALABAMA: Walker Co., Jasper, 800 Airport Rd., 3M, 21 Jun 1975, B. R. Wall (NCSM A596), and M, 29 Apr 1978 (NCSM A2591). Tuscaloosa Co., Tuscaloosa, M, F, 29 Mar 1948, G. E. Ball (RLH) and 2M, 3F, tunneling in Zoyzia grass, 13 Apr 1982, E. A. Cross (NCSM A3933). Washington Co., 3 mi. E Leroy, near Tombigbee R., M, F, 3 Jul 1960, L. Hubricht (RLH) and M, 28 Apr 1983, R. M. Shelley and P. B. Nader (NCSM A4048). Remarks.— Because of the general gonopodal similarity with Pleuroloma fla- vipes, T. lampra was described as a species of Zinaria (now a synonym of Pleu- roloma) by Chamberlin (1942). This similarity is merely coincidental, however, and a relationship between 7hrinaxoria and Pleuroloma was discounted by Cham- berlin and Hoffman (1950) and Shelley (1980). Thrinaxoria bifida (Wood), new combination Figs. 27-28 Polydesmus (Fontaria) bifidus Wood, 1864:7; 1865:223, fig. 52. Epeloria bifida.—Chamberlin and Hoffman, 1958:32. Type-specimen.—As stated in the introduction, Wood’s types are lost. I there- fore designate a neotype as follows: Male neotype (NCSM A2339) collected by R. M. Shelley and W. B. Jones, 8 Jul 1978, along road to Chilowhee Campground, 1.9 miles NW of the junction with US highway 64, 0.2 miles NE Parksville, Polk Co., Tennessee. The other specimens from Polk Co. in the locality listing are designated neoparatypes. Diagnosis.—Prefemoral process variable, short to long, usually extending lin- VOLUME 97, NUMBER 3 507 early along anterior margin of acropodite stem, rarely angling across latter; distal elements continuous with and continuing overall curvature of acropodite; cypho- pods with remnant of receptacle. Neotype. —Length 35.8 mm, maximum width 7.8 mm, W/L ratio 21.2%, depth/ width ratio 64.1%. Segmental widths as follows: collum 6.7 mm 14th-15th 7.4 2nd 6.9 16th 7.2 3rd 7.1 17th 6.8 4th 7.6 18th 5.5 5th-13th 7.8 Color in life: Dorsum generally bright orange in color, with darker ae lighter hues blending across metaterga. Somatic features similar to those of 7. /Jampra, with following exceptions: Width across genal apices 4.5 mm, interantennal isthmus 1.1 mm. Antennae reaching back to middle of 4th tergite, relative lengths of antennomeres 2 > 3 > 4=5=6 > 1 > 7. Genae with slight central impressions. Facial setae as follows: Epicranial, interantennal, frontal, and genal not detected and presumed absent, clypeal about 10-10, labral about 15-15. Collum broad, ends extending well below those of following tergite. Peritremata relatively indistinct, only slightly elevated above paranotal surface. Sternum of segment 5 with two apically separated paramedial lobes between 4th legs, height subequal to widths of adjacent coxae, and with moderate recession between 5th legs (Fig. 27); 6th sternum convexly recessed between both legs, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with low, almost indetectable, longitudinal elevations on segments 8-15. Gonopodal aperture elliptical, 2.6 mm wide and 1.3 mm long at midpoint, indented anteriolaterally, sides elevated above metazonal surface. Gonopods in situ with acropodites extending anteriad from aperture, distal elements curving mediad and crossing those of opposite member in midline over 6th sternum. Gonopod structure as follows (Fig. 28): Prefemoral process moderately long but terminating well below distal division of acropodite, angling slightly across stem of acropodite but not crossing caudal margin. Acropodite curving gradually me- diocaudad at midlength proximal to distal division, distal elements continuous with curvature and acropodite stem, extending well beyond level of prefemur; stem relatively narrow and parallel-sided basally, sides diverging gradually at beginning of curve. Distal elements separated by moderate space, generally di- verging and directed mediocaudad; tibial process longer and more proximal, broad and laminate, apically subacute; solenomerite with sides parallel, continuing cur- vature of acropodite but bent abruptly dorsad distally, tip directed toward tibial process. Prostatic groove arising in pit in prefemur, running entirely along medial side of acropodite onto solenomerite, opening terminally. Description of females.—Since no females have been taken in Polk County, Tennessee, a composite description of their characters has been prepared from the three specimens in adjacent Cherokee County, North Carolina. Length 37.7 mm, maximum width 7.1 mm, W/L ratio 18.8%, depth/width ratio 73.2%. Agreeing closely with males in somatic features except paranota more 508 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pinion i Aaa 28 y, \ pe oreis bess Figs. 24-28. Thrinaxoria spp. 24-26, T. lampra: 24, Sternum of segment 5 of paratype, caudal view; 25, Gonopods in situ, ventral view of male from Henderson Co., TN; 26, Telopodite of left gonopod of paratype, medial view. 27-28, T. bifida: 27, Sternum of segment 5 of neotype, caudal view; 28, Telopodite of left gonopod of the same, medial view. Scale line for Fig. 25 = 1.00 mm; line for other Figs. = 1.33 mm for 24 and 27, 1.00 mm for 26, and 1.14 mm for 28. strongly depressed, creating appearance of more highly arched body, and without trace of sternal elevations. Cyphopods in situ with opening of valves directed ventrad. Receptacle much smaller than in 7. /ampra, a thin darkened remnant of tissue located mediad or anteromediad to the valves. Valves moderate and subequal, surface finely gran- ulate. Operculum large, located under lateral end of valves. Variation.—The prefemoral processes vary, and in males from Whitfield and Dougherty counties, Georgia, they are long and acicular, projecting beyond the division of the acropodite. In most males this structure is linear and directed VOLUME 97, NUMBER 3 509 Fig. 29. Distribution of Dicellarius and Thrinaxoria. rectangles, D. b. bimaculatus; dots, D. b. fictus; stars, D. b. lamellidens; lower X, D. bimaculatus intergrades; triangles, D. t. talapoosa; squares, D. t. separandus; upper X, D. talapoosa intergrades; circles, D. atlanta; half shaded circles, D. ster- nolobus; diamonds, D. okefenokensis; asterisks, T. lampra; ovals, T. bifida. The arrows indicate apparently isolated populations of the species of Thrinaxoria within the range of Dicellarius. along the anterior margin of the acropodite. The neotype is thus an exception in having the process angle across the stem. In other males from Polk County the distal elements are more widely separated, and the solenomerite is broadly curved toward the tibial process instead of being abruptly bent. The overall curvature of the acropodite is similar in all males, but the solenomerite in the one from Dougherty County is nearly upright, with only a slight apical bend. The elements are thus more widely separated so that the overall configuration resembles that of D. atlanta. Distribution.—Southern extremity of the Blue Ridge and Ridge and Valley Provinces to the Coastal Plain, from southeastern Tennessee and southwestern North Carolina to southern Georgia. Specimens were examined as follows: TENNESSEE: Polk Co., 6.5 mi. NE Benton, Chilowhee Rec. Area, Cherokee Nat. For., M, 20 Jun 1950, L. Hubricht (RLH); Oswald Dome, Bean’s Mtn., 2M, 12 Jun 1953, L. Hubricht (RLH); and 0.2 mi. NE Parksville, along rd. to Chilowhee Cpgd., 1.9 mi. NW ject. US hwy. 64, M, 8 Jul 1978, R. M. Shelley and W. B. Jones (NCSM A2339) NEOTYPE LOCALITY. NORTH CAROLINA: Cherokee Co., Wolf Creek vic., ca. 1 mi. E TN st. line, F, 16 May 1979, R. M. Shelley and R. K. Tardell (NCSM A2706); and 6 mi. 510 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON himaculatus okefenokensis sternolobus tallapoosa atlanta 30 Fig. 30. Relationships in Dicellarius. WNW Culberson, along co. rd. 1107, 0.2 mi. N jet. co. rd. 1108, F, 27 Jul 1974, R. M. Shelley (NCSM A2405) and F, 6 Jul 1978, R. M. Shelley and W. B. Jones (NCSM A2330). GEORGIA: Dade Co., Cloudland Canyon St. Pk., F, 27 Apr 1979, R. M. Shelley (NCSM A2607). Whitfield Co., Dalton, along creek off I-75 at jct. GA hwy. 52, M, 16 Apr 1978, R. M. Shelley and R. E. Ashton (NCSM A1847). Dawson Co., along Amicalola River, 1.5 mi. E Amicalola Falls, M, 15 Apr 1961, L. Hubricht (RLH). Lumpkin Co., 16 mi. NW Dahlonega, M, F, 17 Jun and 2 Jul 1973, A. LaVallee and R. Dufhield (RLH). Floyd Co., Rome vic., F, 13 Oct 1954, P. J. Darlington (RLH). Houston Co., Perry, M, 26-28 Dec 1982, J. A. Payne (RLH). Dougherty Co., 8.4 mi. S Albany, just inside Mitchell Co. line on GA hwy. 3, M, 15 Sep 1979, R. M. Shelley and P. T. Hert! (NCSM A2877). Relationships At present I cannot resolve the generic relationships in the Pachydesmini. Of the three other eastern xystodesmid tribes, the Apheloriini is chosen as the out- group because of the synapomorphic absence of a sternal remnant between the gonopodal coxae. The genera of the Rhysodesmini possess such a remnant, a plesiomorphic trait, and the Nannariini, with small size, modified tarsal claws anteriorly in males, and subcoxal sternal spines, is clearly a derived, specialized group. Since apheloriine diplopods have a receptacle on the cyphopods, Pachy- desmus, which possesses this structure, could be considered sister to Thrinaxoria, which has a vestigial to moderate receptacle, and Dicellarius, which lacks the feature. However, since apheloriines have generally unmodified midbody sterna, Thrinaxoria, which has the most poorly developed elevations, could be considered sister to Dicellarius and Pachydesmus. These seem the only useful characters to resolve the question because others are not shared significantly with the out-group. For example, very few apheloriine taxa have a divided acropodite, and the coxal apophyses are autapomorphous in both Pachydesmus and Deltotaria. Conse- quently, until other features are discerned that clarify this matter, the relationships in the Pachydesmini are best considered an unresolved trichotomy. VOLUME 97, NUMBER 3 511 Within Dicellarius there are two obvious branches based primarily on the con- figurations of the tibial processes. One line leads to D. talapoosa and D. atlanta, in which the process tapers continuously to a subacuminate tip, and the other leads to D. bimaculatus, D. okefenokensis, and D. sternolobus, in which it is broad for most of its length, narrowing only near the tip. These two lineages are geo- graphically plausible, since their members are situated proximally to each other (Fig. 29). The key species appear to be D. bimaculatus and D. talapoosa; the others appear to represent segregated populations of these that have achieved reproduc- tive isolation. Dicellarius atlanta is very similar to forms of D. talapoosa, but their overlapping ranges are proof of reproductive isolation. Similarly, the acro- podal curvature of D. okefenokensis is an augmentation of that exhibited by eastern populations of D. bimaculatus (b. fictus), and the torsion culminates the west to east, counterclockwise to clockwise, rotation cline of the latter. These observations suggest that D. okefenokensis was once connected to D. bimaculatus by intergrades that have disappeared, leaving it reproductively isolated. Less conclusive evidence is available for D. sternolobus, but its caudally curved acropodite and the overlying of the tibiotarsus by the narrow solenomerite in medial view are similar to the conditions in D. b. lamellidens. Since this pattern exists in known populations of D. bimaculatus, it could also have evolved a second time in interior populations that are now fully isolated. These ideas on relationships are depicted in Fig. 30. Acknowledgments I am grateful to Gary Hevel and Ralph E. Crabill, Jr., NMNH, for the privilege of examining the types of Polydesmus bimaculatus and Dicellarius sternolobus, and to Herbert W. Levi, MCZ, for loan of that of Leptodesmus okefenokensis. The remaining available types, all in the RVC collection, were examined through the courtesy of Richard L. Hoffman. Dr. Hoffman and William A. Shear kindly provided their personal collections as did Michael L. Williams, AU; George E. Stanton, CC; Howard V. Weems, Jr., FSCA; Richard L. Brown, MEM; and Mark F. O’Brien, UMMZ, for the institutional material under their care. Specimens from Cheaha, Chickasaw, Oak Mountain, and Rickwood Caverns State Parks, Alabama, were collected with permission of the State Parks Division, Alabama Department of Conservation and Natural Resources; those from Amicalola Falls, Cloudland Canyon, Indian Springs, and Sweetwater Creek State Parks, Georgia, were taken through the courtesy of the Parks and Historic Sites Division, Georgia Department of Natural Resources; and that from Natchez Trace State Park, Ten- nessee, was secured through the consent of the Division of Parks and Recreation, Tennessee Department of Conservation. Renaldo G. Kuhler, NCSM scientific illustrator, prepared figures 3, 5, 7, 9, 11, 13, 15, 17, 20, and 26. I thank Donald R. Whitehead for his astute and highly beneficial review comments. This study was supported in part by National Science Foundation Grants Nos. DEB 7702596 and 8200556. Literature Cited Attems, Carl G. 1938. Polydesmoidea II].—Das Tierreich No. 69:1-487. Causey, Nell B. 1955. New records and descriptions of polydesmoid millipeds (Order Polydesmida) from the eastern United States.— Proceedings of the Biological Society of Washington 68:21- 30. S12 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1963. Additional records of Louisiana Millipeds.— Proceedings of the Louisiana Academy of Science 26:76-79. Chamberlin, Ralph V. 1918. Myriopods from Okefenokee Swamp, Georgia, and from Natchitoches Parish, Louisiana.— Annals of the Entomological Society of America 11:369-380. . 1920. A new leptodesmoid diplopod from Louisiana. — Proceedings of the Biological Society of Washington 33:97-100. 1931. A new milliped of the genus Fontaria from Mississippi (Chilognatha: Xystodesmi- dae). — Entomological News 42:78-79. 1939. On some diplopods of the family Fontariidae.— Bulletin of the University of Utah 30(2)[Biological Series 5(3)]:1-19. —. 1942. New southern millipeds.— Bulletin University of Utah 32(8) [Biological Series 6(8)]: 1-19. 1943. On nine North American polydesmoid millipeds.— Proceedings of the Biological Society of Washington 56:35—40. —. 1946a. On four millipeds from Georgia and Mississippi.— Proceedings of the Biological Society of Washington 59:139-142. ——. 1946b. On some millipeds of Georgia.— Entomological News 57:149-152. . 1947. Some records and descriptions of diplopods chiefly in the collection of the Academy. — Proceedings of the Academy of Natural Sciences, Philadelphia 99:21-58. 1949. Some millipeds of the families Polydesmidae and Xystodesmidae.— Journal of the Washington Academy of Sciences 39:94-102. . 1951. Records of American millipeds and centipeds collected by D. Elden Beck in 1950.— Great Basin Naturalist 11:27-35. , and Richard L. Hoffman. 1950. Onsome genera and families of North American diplopods. — Chicago Academy of Science, Natural History Miscellanea Number 71, 7 pp. , and 1958. Checklist of the millipeds of North America.— Bulletin Number 212, United States National Museum, 236 pp. Hoffman, Richard L. 1958. Revision of the milliped genus Pachydesmus (Polydesmida: Xystodes- midae).— Proceedings of the United States National Museum 108:181-218. . 1979. Classification of the Diplopoda. Museum d’Histoire Naturelle, Geneva, Switzerland, 237 pp. Loomis, Harold F. 1969. New and known millipeds from four southern states. —The Florida Ento- mologist 52:245-251. McNeill, Jerome. 1887. List of the myriapods found in Escambia County, Florida, with descriptions of six new species.— Proceedings of the United States National Museum 10:323-327. Shelley, Rowland M. 1980. Revision of the milliped genus Pleuroloma (Polydesmida: Xystodes- midae).— Canadian Journal of Zoology 58:129-168. 1982. Revision of the milliped genus Auturus (Polydesmida: Platyrhacidae).—Canadian Journal of Zoology 60:3249-3267. , and Marianne Filka. 1979. Occurrence of the milliped Pachydesmus crassicutis incursus Chamberlin in the Kings Mountain region of North Carolina and the Coastal Plain of South Carolina (Polydesmida: Xystodesmidae).— Brimleyana 1:147-153. Wood, HoratioC. 1864. Descriptions of new species of North American Polydesmidae.— Proceedings of the Academy of Natural Sciences, Philadelphia 16:6—10. —. 1865. The Myriapoda of North America.—Transactions of the American Philosophical Society 13:137—248. North Carolina State Museum of Natural History, P.O. Box 27647, Raleigh, North Carolina 27611. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 513-521 NEW SPECIES OF MAMMALS FROM NORTHERN SOUTH AMERICA: A LONG-TONGUED BAT, GENUS ANOURA GRAY Charles O. Handley, Jr. Abstract.—A new species of long-tongued bat, Anoura latidens, related to and sympatric with Anoura geoffroyi Gray is described. It is widespread and abundant in Venezuela, but elsewhere it is known only by single specimens from Colombia and Peru. The species of Anoura are keyed. Mammals and their ectoparasites were collected in Venezuela between 1965 and 1968 by the Smithsonian Venezuela Project (SVP), supported in part by a contract (DA-49-MD-2788) of the Medical Research and Development Com- mand, Office of the Surgeon General, U.S. Army. Numerous papers have described the ectoparasites and mammals of the Project. Throughout these papers unde- scribed species of mammals have been referred to by alphabetical designations. Some of these have been named subsequently by Handley and Ferris (1972), and Handley and Gordon (1979). In this paper I deal with another, a long-tongued bat of the genus Anoura Gray. When I named Anoura cultrata (Handley, 1960) the genus Anoura was believed to include only two other species, A. caudifera E. Geoffroy St.-Hilaire and A. geoffroyi Gray. Then for several years the genus was thought to include five species (Jones and Carter 1976), but Nagorsen and Tamsitt (1981), with whom I agree, regarded A. brevirostrum Carter (1968) and A. werckleae Starrett (1969) as syn- onyms of A. cultrata. Now I describe another species, this one related to 4A. geoffroyi. It was discovered in a survey of the mammals of the Parque Nacional Henry Pittier (“Rancho Grande’’) in Venezuela in 1960. Later, many more spec- imens were collected in Venezuela by SVP personnel. The cranial measurements reported here were taken as outlined by Handley (1959:98). Hind foot, tibia, calcar, and forearm were measured in the museum on dry specimens or on specimens preserved in alcohol. Other external dimensions were measured on fresh specimens in the field. All measurements are in milli- meters. Coloration was determined under Examolites (Macbeth Corp., Newburg, New York 12553) with natural light excluded. Capitalized color terms are from Ridgway (1912). Anoura latidens, new species Holotype.—USNM No. 370119, adult female, skin and skull, collected 18 Aug 1965, by A. L. and M. D. Tuttle, on Pico Avila, 2150 m, 5 km NNE Caracas (=“‘Hotel Humbolt, 9.4 km N Caracas’’), Distrito Federal, Venezuela, in mist net across trail through open forest; original number, SVP 629. This bat was pregnant when captured. Etymology. —Latin, latus, broad, and dens, tooth, referring to the unusual breadth PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 514 (L) (S) (L) (L) (L) (L) (L) OS—8P 901-66 OSCT—C HC 6€V—C 6E LI—SI cCI—-II 6ZL—OL 900 + 6 970+ V7 OL 970 + OVC Oll + OCP 090 + 7 9I Ovo + OTT OKC se LIL oN :douffoas pinoup (ZZ) (02) (ZZ) (91) (L1) (91) (L1) CS—9P CII—COl 8SC—-V VT 6ry—-EC IV EU's €I—IlI OL—I19 900 + 6Pf plo + LOI 910+ CSC OS 0 + Str CSO + CSI 970+ 6 vol + 6S9 STI Wy douffoas vanoup (ST) (ST) (ST) (pI) (rT) (pI) (p1) CS—9P LOIl—66 9VC—-I CC €tp—t OV SI—Il cI—II [L/L 9) 800 + 67 C10 + Ol O£ 0 + 6 €7C 8P0 + € CV 8LO+ VET 8c0 +7 II VOC + 9OLO BIIAY Old pue sopeue, so :suapijp] Danoupy (LZ) (€Z) (LZ) (81) (81) (81) (81) €S—8Y CII—c Ol €ST—6 £7 LSv—8 Iv Ete ll €I—Il CL @& vO0 + 0¢ O10 + 90I vIO + SVC 9VO+ LED pS 0 + 84rI 970 + 6TI 9V I + € 99 G7[ WY :suapijy] Danoup YIpesiq [e11GQ101sog YIpesig 9eWI0sAZ yisug] 1s9}BoIH WUe910J ieq (Arp) 100} pulyy yisug] [v10 ‘sasoyuoied Ul OZISs o[dwies ¢ OUI, PUe ‘SOUI9I1x9 9) Z SUT] ‘SIOLIO PIVPUL]S OM] SNUTUT IO Sn[d ULOW 9} SOPNIOUT | SUT] JUSWIOINSeOU YRS 104 ‘ODOULIO OLY 9Y} JO (Ope1Od [q ASS WY 68-89 ‘STI UP) yINOs pue (OILY “P[IAY OdIg ‘sOpeusA SOT) YLIOU WOIZ SUOTIDTIOD TAS JO Mouffoas DAnoUup pue SuapljD] DANOUPp s[eWo} I[Npe JO syUSWIOINSevIJI —"T 91QBL 515 VOLUME 97, NUMBER 3 (Z) (ZL) (9) (9) (Z) (9) (ZL) CAGE vr—-Iyv C9-8S £6—-Vv8 C6—16 LL Ol 86—£6 870 + SEI 800+ Cr cl 0 + 09 970 + 88 Tl 0 + 6 OCO FTL TL'0 + 96 OTL, :7Aouffoas pinoup (91) (81) (ZZ) (17) (07) (ZZ) (ZZ) Cri—r el 9br—-—OP C9—-6S 96-18 001-76 O8—E€L COI—S 6 910 + Sel 800+ CP 800+ c9 810+ 68 800 + 96 800 +9L 80°0 + 86 STI Wy Moiffoas Danoupy (pT) (9) (ST) (ST) (S1) (S1) (ST) OSI—S EI TH-BE €9-LS €6—-08 v6—98 SL—-IL 001-16 970+ Vl 80:0 + OF C10 +09 910 + 88 ClO + C6 OlO0+7L O10 + 96 RIIAY OdId pue sopeus,A SOT :suapljp] Dinoupy (81) (97) (97) (97) (L@) (LZ) (L2) OSI—E€ EI pr—-se C9O-LS 66-98 96—L8 O8—-TL TOI—€6 070 + THI 90:0 + I'v 90'0 + 79 80'0 + 16 800 + £6 800 + OL 90'0 + L'6 CT[ Wy :suapijD] DAnoup BIQILL SOUIUBD 1B YIPIA SIe[OW 12 YIPIAA yisug] [e1e[edisog MOIUIOO] ArlelIxXey yidop sseoulelg YyiIpeoig sseouIeIg ‘(ponunuod)—"T 2198.1 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 516 (8) (L) (8) (8) (8) (8) (8) OS—-tvY CII—v Ol VST—E PT FEV—-E IV L1—91 €I—Il 91-89 vIO + 8Y 810 + 801 970 + OSC 990 +S Cr 970 + 691 0S 0 + STII CEC FOCL OWT :7douffoas Dinoup (Z1) (Z1) (Z1) (Oo) (Ol) (on) (01) CS—9P CIl—v Ol C9T—V VC VEV—-S IV LI—vI €I—II CL—8S C10 + 6P 910 + 8 Ol 9€0 + VST 9V0 + LCL 090+ € SI 970+ 6TI CET FOSO SZ] Wy wdouffoas Danoupy (9) (9) (9) (9) (9) (9) (9) OS—8YP 8 0I—c Ol VVC—S ETC OCv—C OF SI—-TI cI—II 89—6S 800+ 61 810+ SOI 800 + 8 EC PSO+ETIP cyl + Let peor CII OL + 849 BIIAY OdIg pue SOpeUSA SOT :suapizD] DAnouy (IT) (6) (11) (8) () (8) . (8) TSB Pb TIl—0O01 VS7T—O' be LS¢—0 CP 9I—vI Gln 89—-£9 800 + 0$ v70 + LOI 070 + HZ p80 + Lr 8€0 + OSI 00°0 + OI 9p 1 + 759 STI Ury :suapyy] Danoup Yi peelq [e11G101sOg YIpesig 91eUI03AZ, yidug] 1Ss9}v01H W1Ie3104 eq (Arp) 100} purty yisug] [e101 ‘sosoyjuoled Ul OZIS o[]dwes ¢ SUIT pu ‘soUIOINX9 OY) Z SUT] ‘SIOLIO PIePULIS OM} SNUIUI IO sn{d ULSUT 9} SOPNISUI | SUT] JUOWIOINSPOU YOeS 10.J “ONOULIO OY 9} JO (OpelOd [Aq ASS UL] $8-89 “SZ| UH) yNOS pue (QIWLT] “e[IAY Oolg ‘sOpeuaA SOT) YOU WOI sUOTIDITIOD CAS JO Mosffoas Danoup Puke suapiID] Dénouyp seul I[Npe JO sjUUIOINSeI|] — “7 9QUL 517 VOLUME 97, NUMBER 3 (8) (L) (L) (8) (L) (8) (8) OvI—9 El Ly—-vy ¢ 9-09 ¢6—S8 86—V 6 SL-SL €0I—-S 6 C10 + 8 El O10 + SP vI0 +79 970+ 06 O10 + 66 900+ LL 910 + 86 ONWILI, :7douffoas pinoup (01) (Z1) (11) (11) (Z1) (Z1) (11) SrI—-Vel SY—-CL L9O—-8S 96-88 COI—I16 C8—-CL TOI—S 6 9c 0 + OVI 800+ SP 910 F €9 910+ C6 SIO +L6 910+ 9L O10 + 66 STI Wy douffoas Dinoup (9) (9) (9) (9) (9) (9) (9) 9rI—-S TI VV-CY v9—-09 C6—S'8 96-68 SEE LL 101-96 890 + LEI 900+ EP rIO+ 79 vOO + 88 0c 0 + £6 O10 + CL 910+ L6 BIIAY Odd pue sopeua, soy :suapijy] DAnoup (8) (11) (11) (01) (11) (11) (01) TSI—-Cel Cr—-CHL v9—-09 C6—-L8 ¢ 6-88 6L—OL 86-6 0S0 + IPvl 900+ 7P 900+ c¢9 910 + 06 ClO + C6 9TO + VL 800 + 96 CCI WY :Suapijy] Dinoup PIQIL SOUTUBD 1 YIPIAA SIB[OUI 18 YIPIM yigugl [e1epedisog MOIYIOO} AILT[IXLP] yidep osvourvig yipesiq sseourvig ‘(ponunjuoD)— "7 919qeL 518 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of P4 and the other premolars. This taxon has been referred to as ““Anoura sp. A” in previous publications dealing with the collections of the SVP. Distribution. — Widely distributed in Venezuela—in the foothills of the Sierra de Perija and the Mérida Andes, at medium and high elevations in the Sierra de la Costa and in mountains in the Northeast, at three localities in Bolivar state, and in the Ventuari Basin. Outside Venezuela, known only by a single specimen each from Colombia and central Peru. Probably also extends to adjacent parts of Brazil, Guyana, and Ecuador. SVP collectors netted Anoura latidens (n = 105) most often in moist situations (93 percent); in evergreen forest (71 percent), but also in yards and orchards (28 percent), and once in arid thorn scrub (1 percent). Elevation range 50-2240 m (81 percent below 1500 m). Holdridge life zones (Ewel and Madriz 1968): Tropical dry forest (3 percent), Tropical humid forest (28 percent), Premontane thorny forest (1 percent), Premontane humid forest (2 percent), Premontane very humid forest (48 percent), and Lower Montane humid forest (18 percent). Description. —In size similar to Anoura geoffroyi: skull averages slightly smaller, but body averages slightly larger (Tables 1 and 2). Dorsal coloration resembles that of A. geoffroyi but averages paler and browner, less blackish (between Verona Brown and Warm Sepia); hair bases on nape near Cartidge Buff, on remainder of dorsum somewhat grayer; chin, throat, and chest paler than remainder of under- parts, often whitish; abdomen near Sepia; bases of abdominal hairs brown to gray- brown, not well differentiated from buffy tips; and ears and membranes average paler than in A. geoffroyi. Skull, mandible, and teeth similar to those of Anoura geoffroyi but rostrum relatively short, nearly parallel-sided, and with little bulge at canine bases; brain- case relatively inflated; zygomata always incomplete; posterolateral edges of palate (behind toothrows) strongly pointed; horizontal ramus of mandible shallow and usually depressed anteriorly; coronoid-articular distance of ascending ramus short; P3 and P4 relatively wide (thick) and usually shortened; anterobasal cusp of P3 usually reduced; and medial internal cusp of P4 enlarged, rendering basal outline of tooth approximately triangular; P3 and P4 wide and usually long. Measurements. — Holotype female, adult: Total length 66, tail vertebrae 0, hind foot (dry) 11, ear fron notch 13, forearm 42.3, tibia 14.7, calcar 4.5. Greatest length of skull 24.0, zygomatic breadth 10.7, postorbital breadth 4.8, braincase breadth 9.5, braincase depth 7.5, maxillary toothrow length 9.1, post- palatal length 8.5, palatal breadth outside of M3 5.8, rostral breadth at base of canines 4.0. Additional measurements: Tables | and 2. Comparisons. — Anoura latidens is sympatric with three other species of Anoura. From Anoura caudifera it is easily recognized by its larger size (e.g., forearm 40- 45 vs. 35-39, maxillary toothrow 8.6—9.7 vs. 7.8—-8.7), lack of tail, incomplete zygomata, presence of medial internal cusp on P4, presence of anteroexternal cusp and crest on M1, and comparatively long, low P4. Note that most authors, fol- lowing Husson (1962:139), have used the name Anoura caudifer on the grounds that this was the spelling of the specific name originally proposed by E. Geoffroy St.-Hilaire (1818:418, Glossophaga caudifer). However, E. Goffroy St.-Hilaire was in error and so have been all authors (including me) who have used the masculine form “‘caudifer’ in combination with the generic names Glossophaga, Loncho- VOLUME 97, NUMBER 3 519 glossa, or Anoura, all feminine nouns. In this instance, the specific name is an adjectival modifer, and as such it must agree in gender with the generic name. Thus, Anoura caudifera. Compared with Anoura cultrata, Anoura latidens has coloration paler through- out, tail absent, skull much less robust, zygomata incomplete, upper canine much smaller, postcanine teeth (except P1) more robust, and PI not enlarged or blade- like. Anoura latidens is most like A. geoffroyi, but it can be distinguished by numerous characteristics: P3 and P4 (both upper and lower) thicker and more robust; lingual cusp of P4 enclosed in more or less triangular basal outline of tooth (vs. promi- nently protruding from narrow basal outline of tooth); P4 nearly crescent-shaped, with both front and hind ends flexed inward (vs. nearly straight or with the hind end flexed outward); rostrum shorter; canine slightly smaller and less bulging at base; posterolateral edge of palate usually more strongly pointed (vs. usually more rounded and often with multiple points); zygoma always incomplete (vs. always complete in southern Venezuelan and variable in northern Venezuelan 4. geof- froyi); horizontal ramus of mandible shallower and usually depressed (not straight) anteriorly; dorsal coloration usually paler, grayish-brown rather than blackish- brown; chin, throat, and chest pale (vs. undifferentiated from remainder of un- derparts); bases of abdominal hairs not well differentiated from tips, which are seldom whitish and never impart a “frosted’’ appearance as they often do in A. geoffroyi. Both A. geoffroyi and A. latidens show a slight amount of geographical variation between northern and southern Venezuelan populations, isolated from one another by the Llanos. The degree of differentiation is a little greater in Anoura latidens, suggesting a different evolutionary history (Tables 1 and 2). Key to Species of Anoura 1. First lower premolar greatly enlarged and blade-like; upper canine unusu- ally large and with prominent longitudinal sulcus on anterior face ..... re ace i tA ie Sine SE Te eh ie a a ea Anoura cultrata — First lower premolar approximately same size and shape as other pre- molars; upper canine not enlarged and its anterior face flat ........... DD 2. P4 without medial internal cusp; M1 without anteroexternal cusp and CRESUMLAIMPRESENO So) esearch Re oe Anoura caudifera — P4 with medial internal cusp; M1 with anteroexternal cusp and crest; tail LSC MLM Sige. EL aie eR ANB, AE PMT BCT 1 SA RE SEE. DRA CEFR 3 3. Medial internal cusp of P4 enclosed in broad triangular base of tooth; upper and lower P3 and P4 thick ....................:.. Anoura latidens — Medial internal cusp of P4 prominently protrudes from narrow base of tooth; upper and lower P3 and P4 narrow ............... Anoura geoffroyi Ecology.—In the SVP collections Anoura latidens proved to be almost as nu- merous and widespread in Venezuela as A. geoffroyi, its closest relative. Surpris- ingly, A. /atidens is represented by only two specimens in the extensive collections of Anoura in the American Museum of Natural History and by none at all in the collections of the British Museum (Natural History) and the Field Museum of 520 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Natural History. Although I have seen hundreds of specimens of A. geoffroyi that were collected in the 19th and early 20th centuries, I have found only one 4A. latidens AMNH 69187, San Juan de Rio Seco, Colombia, December 1923, Br. Niceforo Maria) that was collected prior to 1960. Either Anoura geoffroyi, or A. latidens, or both inhabited 26 localities in forested portions of Venezuela sampled by SVP. Anoura geoffroyi proved to be more widespread. It was found at 22 of the 26 localities, while A. /atidens was taken at only 14. However, A. /atidens outnumbered 4A. geoffroyi at six of the ten localities where the two species were taken together. The SVP records do not reveal any ecological or geographical segregation of Anoura latidens and A. geoffroyi in Venezuela. Although taken frequently in lowlands, both species were found most often (60 percent) at montane localities: A. geoffroyi to 2550 m near Tabay in the Andes and 4A. /atidens to 2240 m on Pico Avila in the Coast Range. Both were most numerous in the 1000-1500 m interval. Both were taken most often in moist situations in evergreen forest. About 30 percent of the total catch of each species was in openings. Specimens examined (Anoura latidens).—Colombia: CUNDINAMARCA, San Juan, Rio Seco, (1 AMNH). Peru: JUNIN, Prov. Tarma, 2 km NW San Ramon, 884 m, (1 AMNH). Venezuela (all USNM unless otherwise noted): ARAGUA, El Limon, 4 km NW Maracay, 524 m, (3); Est. Biol. Rancho Grande, Paso El Portochuelo and Pico Guacamayo, 13 km NW Maracay, 1130 m, (6). BARINAS, Altamira, 794 m, (1). BOLIVAR, El Manaco, 59 km SE El Dorado, 150 m, (12); Hato la Florida, 47 km ESE Caicara, 50 m, (3); Km 125, 85 km SSE El Dorado, 1032-1165 m, (44). CARABOBO, La Copa, 4 km NW Montalban, 1537 m, (4); Montalban, 598 m, (2). DITO. FEDERAL, Los Venados, 4 km NNW Caracas, 1465-1524 m, (9); Pico Avila, nr. Hotel Humboldt, and Boca de Tigre, 5 km NNE and 6 km NNW Caracas, 2092-2240 m, (13). FALCON, nr. La Pastora, 16 km ENE Mirimire, 70 m, (1). LARA, La Concordia, 47 km NE El Tocuyo, 592 m, (1). MONAGAS, Caripe, 1000 m, (2); Pico de Cerro Negro, Dto. Caripe, 2200 m, (1 UCV); Finca San José, 10 km NE Caripe, (1); San Agustin, 3 km NW Caripe, 1345 m, (1). SUCRE, Manacal, 26 km ESE Cartpano, 366-380 m, (2). T. F. AMAZONAS, San Juan, Rio Manapiare, 163 km ESE Pto. Ayacucho, 155 m, (15). ZULIA, Novito, 19 km WSW Machiques, 1135 m, (2). Total 117. Acknowledgments I am grateful to the following persons who helped me with this paper. Linda Gordon prepared the tables of measurements and assisted in the specimen ex- aminations. George Steyskal reviewed the nomenclature of Anoura with me. Carolyn Gamble, A. L. Gardner, Darelyn Handley, Dan Nicholson, Pippa Van- derstar, and David Webster read early drafts of the manuscript and made many helpful suggestions. Curators of the following collections kindly permitted me to study specimens under their care in the preparation of this description: American Museum of Natural History (AMNH), British Museum (Natural History) (BM), Carnegie Museum of Natural History (CM), Field Museum of Natural History (FMNH), Museum of Comparative Zoology, Harvard University (MCZ), Uni- versidad Central de Venezuela (UCV), University of Michigan Museum of Zo- ology (UMMZ), and United States National Museum of Natural History (USNM). A portion of the SVP specimens have been returned to Venezuela. VOLUME 97, NUMBER 3 521 Literature Cited Carter, D. C. 1968. A new species of Anoura (Mammalia: Chiroptera: Phyllostomidae) from South America. — Proceedings of the Biological Society of Washington 81:427—430. Ewel, J. J.. and A. Madriz. 1968. Zonas de vida de Venezuela.— Ministerio de Agricultura y Cria. Caracas. 265pp., map. Geoffroy Saint-Hilaire, E. 1818. Sur de nouvelles chauve-souris, sous le nom de Glossophages. — Mémoires du Muséum d’Histoire Naturelle (Paris) 4:411-418, pls. 17-18. Handley, C. O., Jr. 1959. A revision of American bats of the genera Euderma and Plecotus.— Proceedings of the United States National Museum 1 10:95-—246. 1960. Descriptions of new bats from Panama.—Proceedings of the United States National Museum 112:459-—479. 1976. Mammals of the Smithsonian Venezuelan Project.—Brigham Young University Sci- ence Bulletin, Biological Series 20(5):1—91. ; , and K. C. Ferris. 1972. Descriptions of new bats of the genus Vampyrops.— Proceedings of the Biological Society of Washington 84:519-523. , and L. K. Gordon. 1979. New species of mammals from northern South America. Mouse possums, genus Marmosa Gray, pp. 65-72. In J. F. Eisenberg, ed. Vertebrate ecology in the northern neotropics.—Symposia of the National Zoological Park, Smithsonian Institution Press, Washington. Husson, A. M. 1962. The bats of Suriname.—Zoologische Verhandelingen, Rijksmuseum van Na- tuurlijke Historie, Leiden 58:1-282. Jones, J. K., Jr., and D. C. Carter. 1976. Annotated checklist, with keys to subfamilies and genera, pp. 7-38. In R. J. Baker, J. K. Jones, Jr., and D. C. Carter, eds. Biology of bats of the New World family Phyllostomatidae, pt. 1.—Special Publications, The Museum, Texas Tech Uni- versity 10. Nagorsen, D., and J. R. Tamsitt. 1981. Systematics of Anoura cultrata, A. brevirostrum, and A. werckleae.— Journal of Mammalogy 62:82-100. Ridgway, R. 1912. Color standards and color nomenclature. [iv + 44 pp., 53 pls., Washington. ] Starrett, A. 1969. A new species of Anoura (Chiroptera: Phyllostomatidae) from Costa Rica.—Los Angeles County Museum Contributions in Science 157:1-9. Division of Mammals, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 522-525 OPAEOPHACUS ACROGENEIUS, A NEW GENUS AND SPECIES OF ZOARCIDAE (PISCES: OSTEICHTHYES) FROM THE BERING SEA Carl E. Bond and David L. Stein Abstract. — Opaeophacus (with O. acrogeneius, a previously undescribed zoarcid from the Bering Sea) is a new genus close to Nalbantichthys Schultz, Andriashevia Fedorov and Neyelov, and Puzanovia Fedorov. It differs from all known fishes in having a slot-like, vertically-oriented cavity in the lens of the eye. Opaeophacus has firm skin, normal penultimate and ultimate vertebrae, lacks scales, lacks palatine and vomerine teeth, has lateral neuromast organs, pectoral fins, gill open- ing entirely above the pectoral fin, and six branchiostegal rays. Those characters in combination differentiate the genus. Four specimens of a new fish from near Seguam Island in the Bering Sea, and another collected north of Umnak Island were presented to Oregon State Uni- versity. Comparison of the new form with the known genera of Zoarcidae showed that it is clearly distinct and we therefore propose a new monotypic genus for it. Methods Specimens were fixed at sea in 10% formalin-seawater solution and later trans- ferred to 50% isopropanol. Counts and measurements follow Hubbs and Lagler (1958). All specimens were radiographed, and one (USNM 260321) was cleared and counterstained with alizarin and alcian blue by the method of Dingerkus and Uhler (1977). The ranges of counts and ratios are given first, followed by the values for the holotype in parentheses. Specimens are on deposit at the National Museum of Natural History, Washington, D.C. (USNM), the California Academy of Sciences, San Francisco (CAS), and the Department of Fisheries and Wildlife, Oregon State University, Corvallis (OS). Opaeophacus, new genus Type-species.— Opaeophacus acrogeneius, n. sp. Diagnosis.—Optic lens with vertical, slot-like cavity, filled with soft hyaline gelatinous material; no pelvic fins or lateral jaw lobes; ultimate and penultimate vertebrae normal; scales, palatine teeth and vomerine teeth absent; body lateral line present, of free lateralis organs; pectoral fin present, with 4—5 rays; 6 bran- chiostegal rays; no opening behind last gill arch; gill openings entirely dorsal to base of pectoral fin; small pseudobranchiae present; only unbranched soft rays in all fins, with those in caudal, pectoral and posterior one-third of dorsal and anal fins segmented; no pyloric caeca. Comparisons. — This genus is part of the “‘natural group” suggested by Fedorov and Neyelov (1978) to contain Nalbantichthys Schultz, 1967, Puzanovia Fedorov, 1975, and Andriashevia Fedorov and Neyelov, 1978. It shares the general shape, VOLUME 97, NUMBER 3 523 massive lower jaw, blunt head, reduced pectoral fin, reduced gill opening, and reduced lateral line of those genera. It differs from all three in the absence of scales; from Nalbantichthys in having lateral sensory neuromasts, six branchios- tegals, and firm skin; from Puzanovia in having the gill opening entirely dorsal to the base of the pectoral fin; and in the absence of vomerine teeth, scales, and an opening behind the fourth gill; and from Andriashevia in having pectoral fins, in the absence of palatine teeth and an opening behind the fourth gill, and in having segmented soft rays. It also differs from all related genera in having a slot-like, vertical pit extending halfway to the center of the lens of the eye. Etymology. — The generic name is taken from the Greek “‘opaeos,’ “with a hole”’ and ‘phakos,’ “lens.” . Opaeophacus acrogeneius, new species Fig. 1 Holotype.—USNM 260320, 145 mm SL, female collected off Seguam Island, Bering Sea at 52°42’N, 172°15'W, depth of capture 500-800 m, by G. Hewitt, aboard M/V Mito Maru 82, a long-liner, 12 Apr 1979. It was associated with an unidentified orange colonial coelenterate. Paratypes.—USNM 260321, 137 mm SL, male and 147 mm SL, male, cleared and stained; CAS 52802, 151 mm SL male; same collection data as holotype. OS 10000, 154 mm SL, female, Bering Sea, North of Umnak Island, 53°33’N, 169°18’W, at 600-700 m by R. McClure, M/V Shintoko Maru 37, 27 Sep 1982. It was associated with unidentified black coral and basket starfish snagged by a hook of a long-line. | Diagnosis.—Same as for genus. Counts.— Vertebrae 144-149 (148), 25-26 (26) precaudal; dorsal fin rays 141- 148 (146); anal fin rays 121-124 (123); caudal fin rays 8—9 (9); pectoral fin rays 4-5 (4); branchiostegals 6 (6); gill rakers 11 + O + 3 on first arch. Ratios as % of SL. Head 11.2-12.0 (11.2); depth of head 7.4-8.3 (7.4); snout 3.0-—3.5 (3.2); eye 1.5-1.7 (1.5); width of lower jaw 5.8-6.8 (5.9). Distance from pectoral fin base to gill opening 1.4—2.0 (1.4); snout to anus 24.4—25.4 (24.5). Description. — Body elongate, tapering from blunt head to pointed tail, greatest depth about 8% SL. Skin thick and firm in newly preserved specimens. Head about 12% SL, broader ventrally. Massively fleshed lower jaw slightly longer than upper jaw; rictus of mouth extends to below posterior margin of eye; maxillae completely covered by skin; no separate lips. Eye small, about 14% HL, covered by membrane continuous with skin of head; lens with elongate vertical pit filled with soft gelatinous material extending into lens about 25% of its diameter (Fig. 1). One pair of tubular nostrils, directed forward. Cephalic lateralis pores small; One interorbital pore; 3 pores in occipital commissure; postorbital pores 3, one over gill opening; suborbital pores 5; supraorbital pores 2; preoperculomandibular pores 6; no lateral lobes, ridges or cirri on head. Teeth on dentaries and premaxillae sharp and recurved at tip, those at jaw symphyses larger; dentary teeth nearly hidden by tissue of jaws; palatine and vomerine teeth absent. Both upper and lower oral valves large; glossohyal covered by thick “‘tongue.”’ No slit behind last gill arch; pseudobranchiae present; gill opening small, entirely dorsal to pectoral 524 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Above. Opaeophacus acrogeneius, holotype, USNM 260320, 145 mm SL, showing dis- position of lateral neuromasts. Arrow indicates position of anus. Below. Head of holotype showing slit in eye lens and the disposition of cephalic sensory pores. Drawings by Bonnie Hall. fin. Lateral line midlateral with about 65 tiny crater-like pits extending posteriorly for about 65 to 70% of SL. Scales and pelvic fins absent. Pectoral fins small and slender; dorsal and anal fins completely hidden in skin. Only soft rays present, unbranched; those in caudal, pectoral, and posterior third of dorsal and anal fins segmented. Pyloric caeca absent. Vertebrae inequiamphicoelous; penultimate ver- tebra normal. Color in life and soon after fixation bright tangerine orange, very closely re- sembling the color of the colonial coelenterate from which the specimens from Seguam Island were collected. Color faded in alcohol to light tan within a few months. Etymology.— From the Greek ‘akrogeneios’, meaning “with prominent chin.” Remarks.—The strange eye lens of Opaeophacus with its slot-like vertical cavity constitutes a puzzling adaptation. If the soft gelatinous material that fills the cavity has a refractive index different from that of the lens, light striking the cavity would not be focused on the fundus of the retina but would be scattered to the anterior and posterior of the retina. Unless some unknown specialization provides oth- erwise, that would seem to have the effect of destroying visual acuity, but would spread any available light over a greater number of visual cells. A possible ad- vantage could be the detection of silhouettes of organisms moving against a dimly lighted background. Acknowledgments We thank Gary Hewitt, National Marine Fisheries Service, for discovering and preserving the first-known specimens and donating them to OSU, and Robert VOLUME 97, NUMBER 3 525 McClure, also of NMFS, for the specimen from Umnak Island. We thank Eric Anderson for his many helpful comments about the new species and its relation- ships, and for reviewing the manuscript. This work was supported by Oregon Agricultural Experiment Station (CEB) and OSU College of Oceanography (DLS). This is OAES Technical Paper 6919. Literature Cited Dingerkus, G., and L. D. Uhler. 1977. Enzyme clearing of alcian blue stained small vertebrates for demonstrations of cartilage. —Stain Technology 52(4):229-232. Fedorov, V. V. 1975. Description of a new genus and species of a zoarcid fish, Puzanovia rubra Gen. et sp. nov. (Pisces, Zoarcidae) from the northern part of the Pacific Ocean.—Journal of Ichthyology 15(4):527-531. , and A. V. Neyelov. 1978. A new genus and species of eelpout, Andriashevia aptera (Perci- formes, Zoarcidae).—Journal of Ichthyology 18(5):846—850. Hubbs, C. L., and K. F. Lagler. 1958. Fishes of the Great Lakes Region. Revised Edition. — Cranbrook Institute of Science, Bulletin (26):1-185. Schultz, L. P. 1967. A new genus and new species of zoarcid fish from the North Pacific Ocean. — Proceedings of the United States National Museum 122 (3598):1-5. (CEB) Department of Fisheries and Wildlife, Oregon State University, Corvallis, Oregon 97331. (DLS) College of Oceanography, Oregon State University, Cor- vallis, Oregon 97331. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 526-528 BIOLUMINESCENCE IN THE FRESHWATER AMPHIPOD, HYALELLA AZTECA, CAUSED BY PATHOGENIC BACTERIA Thomas E. Bowman and Fanny Phillips Abstract.— Bioluminescence is reported from a freshwater amphipod for the first time. Individuals in a culture of Hyalella azteca emitted a continuous green- ish-yellow glow, usually at night. Luminescence was always followed by death of the amphipod and is attributed to pathogenic bacteria. “Tiving light”? is emitted by members of almost all major groups of animals. Monera, Protista, and virtually all metazoan phyla have marine or terrestrial representatives that are bioluminescent. However, luminescence in freshwater animals is uncommon, and to our knowledge has not been reported in a freshwater amphipod. However, bacteria-based luminescence is known in at least one fresh- water crustacean, the caridean shrimp Xiphocaridina compressa. In Japan, in Lake Suwa and in a stream in rice fields near Sawara, these shrimp become luminescent en masse on hot summer nights. The luminescence is so attractive that He shrimp are protected by the government (Hanada 1955). Observations We report herein observations on bioluminescence in the freshwater amphipod Hyalella azteca (Saussure) and our reasons for believing that the bioluminescence is produced by pathogenic bacteria. We are not certain of the source of the am- phipods or the bacteria. At least some, and probably most of the amphipods came from plants taken from a small spring-fed trout pond on U.S. route 15 about 3 miles north of Leesburg, Loudoun County, Virginia. The amphipods were placed in an aquarium containing guppies that was kept in the second author’s house in Silver Spring, Maryland, next to a window with southern exposure. Other possible sources of the amphipods are with plants purchased from local aquarium stores and with plants collected by the second author and her husband, Craig Phillips, from several hot, peaty, rather stagnant pools in Florida. Whatever the source, the amphipods did not luminesce while the water in the aquarium remained cool. In July and August, however, when the aquarium water reached temperatures as high as 35°C and became cloudy, 1—3 luminescent am- phipods were observed during an evening. Luminescence was a continuous emission of a greenish-yellow glow. It was rather dim, and a darkened room was necessary to observe it. Almost all lumi- nescence occurred at night, but a few daytime occurrences were noted. Some time after an instance of luminescence was noted, usually in the morning following a nighttime observation, a moribund luminescent amphipod or one or more dead amphipods were found at the surface. The continuous rather than periodic or flashing luminescence, together with the inevitable death of the luminescent amphipods, indicates that the light was VOLUME 97, NUMBER 3 527 time of day Fig. 1. Observations of luminescence in a gallon jar culture of Hyalella azteca. Open circles, luminescence seen: closed circles, no luminescence seen. Concentric circles, 2 or 3 specimens seen luminescing. Daytime (Sunrise to sunset) indicated by dashed lines; nighttime (sunset to sunrise), by solid lines. produced by bacteria that killed the host amphipods. No other explanation appears possible. The light produced by the bacteria was not evident until the latter reached a concentration on the host amphipods that led to the latter’s death. Why this concentration was almost always reached nocturnally is not known. Possibly multiplication of the bacteria is inhibited by light. Lacking facilities, we did not attempt to culture and identify the bacteria. As for the source of the pathogenic luminescent bacteria, the trout pond near Leesburg remains cool in the summer and successfully supports water-cress and introduced rainbow trout. We think it likely that the bacteria came from the Florida plants, or possibly, but less likely, from aquarium store plants. A series of observations was made on a number of amphipods removed from the aquarium to a gallon jar and monitored for 15 days, from 22 July to 5 August. The jar was kept next to the first author’s bed and observed whenever he woke up. Some observations were made in a darkened closet during the day. The results are summarized in Fig. 1. At least one and as many as three amphipods were seen to luminesce each night. During the 15 days, at least 12 amphipods lumi- nesced and died. Discussion Luminescence is widespread among marine pelagic amphipods (Herring 1982), but is produced by luminescent organs rather than by luminescent bacteria. Ac- 528 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON counts of luminescence in amphipods that must have been caused by pathogenic bacteria date from the early 18th century and are summarized by Harvey (1952). Most of the amphipods involved are beach hoppers of the superfamily Talitroidea, to which Hyalella belongs. A recent account is that of Bousfield and Klawe (1963) for the beach hopper Orchestoidea gracilis. Beach hoppers inhabit sandy beaches above the wave line and are terrestrial, in contrast to the completely aquatic Hyalella. We are unable to explain why, in a culture of dozens of amphipods, only one or two per day developed a growth of bacteria sufficient to produce visible lu- minescence. Whatever the proximate cause, this pattern of infestation has adaptive value for the bacteria in prolonging the availability of the host. Acknowledgments We are grateful to Louis S. Kornicker and David Lapota for helpful reviews of the manuscript. Literature Cited Bousfield, E. L., and W. L. Klawe. 1963. Orchestoidea gracilis, a new beach hopper (Amphipoda: Talitridae) from Lower California, Mexico, with remarks on its luminescence. — Bulletin of the Southern California Academy of Sciences 62:1-8. Haneda, Y. 1955. Luminous organisms of Japan and the Far East.— Jn Frank H. Johnson, ed., The luminescence of biological systems.—AAAS, Washington, pp. 335-385. Harvey, E. Newton. 1952. Bioluminescence.—Academic Press, New York, 649 pp. Herring, Peter J. 1981. Studies on bioluminescent marine amphipods.—Joumal of the Marine Bi- ological Association of the United Kingdom 61(1):161-176. (TEB) Department of Invertebrate Zoology, National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560; (FP) 10109 Grant Av- enue, Silver Spring, Maryland 20910. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 529-531 A REDESCRIPTION OF ICHTHYOTACES PTEROISICOLA SHITNO (CRUSTACEA: COPEPODA: PHILICHTHIIDAE) FROM THE LIZARDFISH SYNODUS VARIEGATUS LACEPEDE (SYNODONTIDAE) Roger F. Cressey Abstract.—In 1932 Shiino described a new parasitic copepod from the scor- pionfish Pterois lunulata. This gall-producing parasite has not been reported since. The author recently collected another specimen from Synodus variegatus during his recent studies of Indo-West Pacific lizardfishes. Parasitic copepods of the family Philichthiidae are characterized by their pres- ence in the canals of the lateral line system of their hosts. The single female and male described below were recovered from a gall located on the right upper surface of the head behind the right eye of the host. Ichthyotaces pteroisicola Shiino Figs. 1-9 Ichthyotaces pteroisicola Shiino, 1932:417. Material examined.—A single female and single male from a gall on the head of a specimen of Synodus variegatus (CAS 30617) from Viet Nam housed in the collections of the California Academy of Sciences (Fig. 1). Description.—Female: Body form as in Fig. 2, about as wide as long (4mm). Ventral aspect with series of bulbous processes arranged symmetrically as in figure. Fig. 1. Synodus variegatus Lacépéde (CAS 30617) infested with Ichthyotaces pteroisicola Shiino. 530 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 2-9. Ichthyotaces pteroisicola: 2, Female, ventral view; 3—9, Male: 3, Dorsal view; 4, Last abdominal segment and caudal rami, ventral; 5, First antenna; 6, Oral area; 7, Leg 1; 8, Leg 2; 9, Leg 3. Oral area indicated by arrow on Fig. 2. (I was unable to dissect out the mouthparts from the single specimen as they are extremely small and difficult to see). No thoracic appendages could be found unless the bulbous processes represent them and the caudal ramus. VOLUME 97, NUMBER 3 531 Male: Body form as in Fig. 3. Total length 896 um, greatest width 251 um, measured at widest part of free thoracic segments. Thoracic segments bearing legs 1 and 2 fused, bearing articulated process at each posterior corner. Thoracic segment bearing leg 3 free. Following 2 thoracic segments without appendages. Genital segment bilobed ventrally and about 3 times as wide as long. Abdomen 3- or 4-segmented (last segment partly divided). Caudal rami (Fig. 4) longer than wide (83 X 32 wm) bearing | outer lateral seta, 3 small terminal-subterminal setae, and | longer, subdivided seta, all setae naked. First antenna (Fig. 5) 5-segmented, segments bearing 1, 2, 3, 2, 7 setae respec- tively and aesthete on each of last 2 segments, all setae naked. Second antenna and oral area as in Fig. 6. Second antenna oriented towards oral area indicating possible feeding appendage rather than for attachment as in other groups of para- sitic copepods. Labrum bearing 2 sclerotized processes similar to single process recently described as representing the labrum in Colobomatus by Cressey and Schotte (1983). Mandible represented by heavily sclerotized triangular process. First maxilla not found but possibly represented by 2 medial patches of spinules between mandible and second maxillae. Second maxilla bearing 2 terminal pro- cesses, largest bearing row of stout spinules along posterior border. Leg 1 (Fig. 7) rami 1-segmented, exopod outer edge with 2 short rows of spinules as in figure, 5 outer to terminal, short, stubby fringed spines and inner naked spine/seta, endopod with 3 terminal fringed spines as in exopod. Leg 2 (Fig. 8) as in leg | except outer rows of spinules on exopod not seen, | less terminal spine, and no inner spine/seta. Leg 3 (Fig. 9) simple lobe bearing 2 terminal fringed spines. Remarks.—This parasite is apparently rare as this is only the second reporting of it in over 50 years and seems especially so since the parasite causes the host to produce an obvious gall, not easily overlooked. During the course of my ex- aminations of lizardfish (Synodus) for a revision of the Indo-West Pacific species I examined over 1000 fish. The infestation by this parasite on only one examined fish indicates its rarity. The previous record from the scorpionfish Pterois also indicates that it is not host-specific. Acknowledgments The line illustrations were done by Hillary B. Cressey. The halftone was done by Jack Schroeder. Literature Cited Cressey, R. F., and M. Schotte. 1983. Three new species of Colobomatus (Copepoda: Phylichthyidae) parasitic in the mandibular canals of haemulid fishes.— Proceedings of the Biological Society of Washington 96(2):189-201. Shiino, S. 1932. Ichthyotaces pteroisicola, n. gen. and n. sp. a copepod parasitic on the fish Pterois lunulata Temm. and Schl.—Annotationes Zoologicae Japonenses 13(4):417-433. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 532-537 TECTASQUILLA LUTZAE, NEW GENUS AND SPECIES (CRUSTACEA: STOMATOPODA: LYSIOSQUILLIDAE) FROM THE GULF OF MEXICO Daniel L. Adkison and Thomas S. Hopkins Abstract.—A new genus and species of stomatopod is described based on two specimens collected from the western North Atlantic and the eastern Gulf of Mexico. Tectasquilla lutzae can be distinguished from other lysiosquilloids by the combination of the following characters: 1) eyes covered by rostral plate; 2) mandibular palp and five epipods present; 3) dactylus of raptorial claw with four teeth; 4) proximal portion of outer margin of uropodal endopod without strong fold; and 5) telson with false eave. In a recent faunal survey of the eastern Gulf of Mexico, a unique stomatopod specimen was collected. On examination, it became apparent that the specimen represented a new species. While similar to Heterosquilloides, the new species was difficult to place in any described genus, therefore a new genus is diagnosed for it. Tectasquilla, new genus Diagnosis.—Size moderate, maximum length at least 70 mm. Eye cornea bi- lobed. Rostral plate elongate, with apical spine. Antennal protopod with | mesial and 2 ventral papillae. Exposed thoracic somites lacking longitudinal carinae; eighth thoracic somite with median ventral keel. Mandibular palp and 5 epipods present. Raptorial claw, dactylus with 4 teeth. Endopods of pereopods of 2 articles, distal article elongate. Abdomen depressed, compact, smooth, unarmed except on sixth somite; sixth abdominal somite with spines at posterolateral angles, with ventrally directed process anterior to each uropod. Telson broader than long, with false eave; marginal armature on either side of midline consisting of row of slender submedian denticles, movable submedian tooth, 2 intermediate denticles, 1 in- termediate tooth, 1 lateral denticle and 1 lateral tooth. Uropod, proximal article with dorsal spine; proximal article of exopod with slender movable spines on outer margin and setae on rounded lobe on inner distal margin; endopod without strong proximal fold on outer margin; basal prolongation produced into 2 spines, triangular in cross section, inner longer. Type-species.— Tectasquilla lutzae new species Remarks.— Tectasquilla contains only the type-species. Tectasquilla only superficially resembles the nannosquillid genera which have a false eave on the telson. Tectasquilla can be easily differentiated from these genera by having a bilobed cornea, the mandible with palp, papillae on the an- tennal protopod, the endopods of pereopods slender, and a uropodal endopod without strong proximal fold on the outer margin. Of the lysiosquillid genera, Tectasquilla most closely resembles Heterosquilla Manning, 1963, and Heterosquilloides Manning, 1966. Tectasquilla will key out VOLUME 97, NUMBER 3 533 as Heterosquilla in Manning’s (1969) key to American Lysiosquillidae. In 1980, Manning elevated the subgenus Heterosquilloides to generic rank. Tectasquilla can be easily distinguished from Heterosquilla and Heterosquilloides by having: 1) eyes covered by the rostral plate; 2) 5 epipods present; 3) one mesial antennal papilla; 4) dactylus of raptorial claw with 4 teeth; and 5) telson with false eave. Etymology.—The name is derived from the Latin tectus, meaning “‘covered, roof” alluding to the false eave on the telson, and the name Squilla. The gender is feminine. Tectasquilla lutzae, new species Figs. 1, 2 Material examined.—Eastern Gulf of Mexico, Bureau of Land Management station 2529, sample 2529191880208; 29°55'55”N, 86°06'29”W; capetown dredge; 38 m; 8 Feb 1978; T. S. Hopkins coll; 1 6 (holotype) USNM 204717.—Atlantic Ocean off Florida, R/V Delaware IT station 23; 31°55'N, 79°43'W; tumbler dredge; 36-38 m; 22 Jun 1982; W. G. Lyons collector, 1 (paratype, anterior fragment) USNM 204716. Description. — Eye elongate; cornea bilobed, set slightly obliquely on stalk; ocular scales fused into rectangular plate with median notch; eyes reaching beyond middle of antennular peduncle. Antennular peduncle short, about '4 length of carapace; antennular process with broad, blunt, anteriorly directed spines. Antennal scale slender, about '4 length of carapace; antennal protopod with 1 mesial and 2 ventral papillae. Rostral plate subrectangular, elongate, extending beyond eyes; apical spine present. Mandible palp 3 articles. Raptorial claw, dactylus with 4 teeth, outer margin with basal notch; propodus with 4 movable spines proximally, proximal by far the longest, second and third spines approximately same length and half length of first, fourth spine half length of third spine; dorsal ridge of carpus poorly defined, terminating in blunt projec- tion; merus much longer than ischium, both unarmed. Lateral process of fifth thoracic somite blunt lobe, directed ventrally; lateral processes of sixth and seventh somites rounded; eighth thoracic somite with median ventral keel as low conical projection; distal article of endopods of walking legs elongate, most slender on last leg; basal article of walking legs unarmed. Abdomen depressed, compact, unarmed except on posterolateral angles of sixth somite; sixth somite with broad, blunt projection anterior to articulation of uro- pods; sixth somite with pair of broad low carinae, parallel to lateral margin and terminating in blunt lobe. Telson thick, with false eave, nearly twice as broad as long; ventral surface with strong spine on midline posterior to anus; | pair of tubercles present on antero- lateral dorsal surface; false eave with median trilobed prominence bearing 2 ca- rinae, one carina terminating with 1 spine, the other carina with 2 spines; 2 prominences laterally, each terminating in broad acute lobes; 7 to 10 spines on each side of midline between false eave and true posterior margin and between middle 2 prominences of false eave and submedian and intermediate teeth of posterior margin of telson; marginal armature of telson consists of row of 5 submedian denticles, inner denticle much the smallest, outer 2 denticles movable; 534 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON e Se Sa ra GF Se, Bee »® cece TEE EeESSeeee 2 \ |Omm \ Fig. 1. Tectasquilla lutzae, holotype. a, Dorsal view; b, Anterior part of body; c, Eyestalk; d, Antennule, ventral view; e, Antenna, dorsal view; f, Claw, medial view. Scales as indicated. movable submedian tooth; 2 intermediate denticles; 1 intermediate tooth; 1 lateral denticle; and | lateral tooth. Uropod, basal article segment thick, outer dorsal carina broad, inner dorsal carina terminating in strong spine, basal prolongation produced into 2 spines, triangular in cross section, inner spine longer; proximal article of exopod swollen dorsally, with row of 7 slender spines, distal spine not reaching middle of distal segment of exopod, ventrally with 1 broad spine distoventrally, inner margin with broad lobe fringed with setae; distal article of exopod nearly twice length of proximal article; endopod elongate, inner margin slightly concave, outer margin convex, proximal portion of outer margin curled dorsally (not folded). VOLUME 97, NUMBER 3 535 @ 2.0mm =] (g=i) Fig. 2. Tectasquilla lutzae, holotype. a, Telson, dorsal view; b, Telson, ventral view; c, Telson, posterior view; d, Telson, lateral view; e, Uropod, dorsal view; f, Uropod, ventral view; g, Maxilliped 5, lateral view; h, Walking leg 1, medial view; i, Walking leg 3, medial view. Scales as indicated. Remarks.—The paratype is a fragment consisting of the anterior part of body from rostral plate to fifth thoracic somite plus a part of the sixth somite and right first walking leg. Etymology.—This species is named for Ms. Linda B. Lutz, who has illustrated several crustaceans for us. Measurements. — Male, holotype, total length 73 mm; carapace length 16 mm; cornea width 2.2 mm; rostral plate length 5.6 mm, width 4.0 mm; fifth abdominal somite width 15.8 mm; telson length 3.5 mm, width 6.1 mm. Paratype, carapace length 17.8 mm; cornea width 2.6 mm; rostral plate length 6.3 mm, width 4.6 mm. Color.—In life, body not heavily pigmented; overall color straw (light red- brown) with scattered yellow to tan chromatophores. Rostral plate heavily pig- mented, dark brown to black. Carapace with 2 darker bands, anterior margin of 536 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON brown broad V-shaped band, laterally indistinct. Posterior margin of carapace darker. Four pigmented areas on posterior half of carapace (Fig. 1a) lateral pair of spots less distinct than spots on midline. Lateral spots with unpigmented or light yellow semicircular areas centrally. Posterior 3 thoracic somites and anterior 5 abdominal somites with paired pigmented spots dorsally (Fig. 1a), and with more diffuse pigmented area laterally. Sixth abdominal somites and telson without pigment spots. Antennal scale with distal pigment spot. Propodus of maxillipeds darker pig- mented than body, dactylus with few chromatophores. Raptorial claw, dactylar margins and teeth pale yellow-orange, central region pigmented like body; propodus yellow-orange distally, opposable margin also yellow-orange; walking legs bases only slightly darker than body, distal article much lighter than body, setae yellow. Uropods pigmented like body. Discussion.—In T. lutzae, the number of intermediate denticles is interpreted to be two but might be three. A rounded projection occurs between and just dorsal to the intermediate denticles. Since the posterior margin of the telson is not sharply defined in this area, this projection may in fact be the third denticle. The number of intermediate denticles present is an important character since Heterosquilla has two intermediate denticles and Heterosquilloides has four intermediate den- ticles. With respect to the other characters listed by Manning (1969), Tectasquilla appears closer to Heterosquilloides than Heterosquilla. Tectasquilla lutzae appears most similar to Heterosquilloides mccullochae (Schmitt, 1940). It shares a similar color pattern, an ornate posterior telsonal margin, four teeth on the dactylus of the claw, and five epipods. In addition to the differences noted above, 7. /utzae can be distinguished from H. mccullochae by the presence of three not four antennal papillae. Acknowledgments We thank Michael M. Dardeau, Dauphin Island Sea Laboratory, for suggestions on the manuscript. We thank David K. Camp, Florida Department of Natural Resources, for suggestions on the manuscript and for informing us of the existence of the paratype fragment. We also thank Raymond B. Manning, Smithsonian Institution, for suggestions on the manuscript and loan of the paratype. The drawings are by Linda Lutz of the Dauphin Island Sea Laboratory. The type was collected with support of the Bureau of Land Management contract to Dames & Moore number AA-550-CT7-34 via a subcontract to T. S. Hopkins. Literature Cited Manning, R. B. 1963. Preliminary revision of the genera Pseudosquilla and Lysiosquilla with de- scription of six new genera (Crustacea: Stomatopoda).— Bulletin of Marine Science of the Gulf and Caribbean 13(2):308-328. . 1966. Notes on some Australian and New Zealand stomatopod Crustacea, with an account of the species collected by the Fisheries Investigation ship Endeavour.—Records of the Aus- tralian Museum 27(4):79-137. 1969. Stomatopod Crustacea of the western Atlantic.—Studies in Tropical Oceanography (Miami) No. 8, 380 pp. VOLUME 97, NUMBER 3 537 . 1980. The superfamilies, families, and genera of recent stomatopod Crustacea, with diagnoses of six new families.— Proceedings of the Biological Society of Washington 93(2):362—372. Schmitt, W. L. 1940. The stomatopods of the west coast of America based on collections made by the Allan Hancock Expeditions, 1933-1938.—Allan Hancock Foundation Pacific Expeditions 5(4):129-225. (DLA) Department of Biology, Tulane University, New Orleans, Louisiana 70118; (TSH) University of Alabama, Dauphin Island Sea Laboratory, P.O. Box 369, Dauphin Island, Alabama 36528. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 538-543 NEW SPECIES OF FRESHWATER CRABS (CRUSTACEA: DECAPODA: PSEUDOTHELPHUSIDAE) FROM COLOMBIA Martha R. Campos and Gilberto Rodriguez Abstract.— Two new species of Pseudothelphusidae are described from Colom- bia. Strengeriana chaparralensis, n.sp. adds a fifth species to a primitive group of crabs from the Central Cordillera; its gonopod resembles that of its nearest geographic relative, S. tolimensis Rodriguez and Diaz, 1981. Chaceus davidi, n.sp. also belongs to a small genus of primitive crabs from the Sierra de Santa Marta; the gonopod has characters of both Chaceus and Hypolobocera. A geographical record is given for Hypolobocera martelatani Pretzmann, 1965, which was orig- inally described from an unknown locality in “Columbia.” The northern Andes of Colombia is an area of high taxon density for the family Pseudothelphusidae (Rodriguez 1982). Of the three groups that occur sympatri- cally in the Andes proper and in the Sierra de Santa Marta, the Strengerianini is a primitive one that displays considerable variability in the structure of the male gonopod, while the other two groups, consisting of the genera Neostrengeria and Hypolobocera, exhibit a more homogeneous gonopod structure, but at the same time they appear very fragmented and are dispersed over a wider area. During examination of material in the Museo de Historia Natural, from the Instituto de Ciencias Naturales, Bogota (ICN-MNH), we found two new Strengerianini. One of them is of particular interest since it displays characters in gonopod structure that are intermediate between the genera Chaceus and Hypolobocera. Tribe Strengerianini Rodriguez, 1982 Genus Strengeriana Pretzmann, 1971 Strengeriana chaparralensis, new species Figs. 1, 4a, d Material.— Quebrada Piedras Blancas, Municipio Rioblanco, 900 m above sea level, Tolima Department, Colombia; 29 Sep 1983; M. R. Campos : | male holotype, cb. 35.3, cl. 21.2 mm (ICN-MHN N° CR 0525).—Vereda Betania, Municipio Chaparral, 850 m above sea level, Tolima Department, Colombia; 17 Jul 1983; M. R. Campos: 2 males, cb. 33.4 and 28.2 mm, cl. 25 and 17.5 mm. Description. — The cervical groove is straight and shallow, deeper in its posterior half; it does not reach the margin of the carapace. The anterolateral margin has a shallow depression behind the orbit, followed by approximately 8 papillae regularly spaced on the anterior half, and approximately 8 minute teeth on the posterior half. The frontal lobes are wide but ill defined. The median groove is absent. The surface of the carapace behind the front is moderately inclined an- teriorly and towards the midline. The upper border of the front is almost straight in dorsal view, well marked with a row of conspicuous, irregularly placed tubercles. The lower margin is strongly sinuous in frontal view. The surface of the front VOLUME 97, NUMBER 3 539 Fig. 1. Strengeriana chaparralensis, holotype, ICN-0525, left gonopod: a, Total view, caudal; b, Detail of apex, caudal; c, Lateral; d, Cephalic; e, Mesial. between the upper and the lower borders is very narrow. The surface of the carapace is smooth, sometimes covered by small papillae not visible to the naked eye. The palm of the larger chela is moderately inflated; the fingers do not gape. The exopod of the third maxilliped overreaches the lateral margin of the ischium. The orifice of the efferent branchial channel is almost closed by a spine at the jugal angle and by the production of the lateral lobe of the epistome. The male gonopod is short and stocky; the marginal lobe is simple, with a short ridge on its lateral surface; the lateral lobe is produced cephalically to form with the cephalic lobe a long slit where the genital pore is located; the cephalic lobe bears 2 strong cephalically directed conical spines on its lateral surface, and another long and bifid spine on its mesial surface which is recurved and directed laterad and bears at its base a small spine. In addition to the strong caudal setae and the 540 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON e Fig. 2. Chaceus davidi, holotype. ICN-0083, left gonopod: a, Total view, caudal; b, Detail of apex, caudal; c, Lateral; d, Cephalic; e, Mesial; f, Apex in distal view. small spines of the genital pore, the gonopod bears a large patch of small dark spines on the lateral side, small spinules over the distal border of the cephalic lobe, and a few tiny setae on the mesial side. Size.—This is a small species, cb. 35.3 mm, but even so, it is the largest yet recorded for the genus. Remarks.—The species is most closely related to Strengeriana tolimensis Rod- riguez and Diaz, 1981. In both species the gonopod has two spines on the lateral side and a recurved bifid spine on the mesial side, with a small spine at its base, but all of these spines are weaker in S. tolimensis. The main difference between the gonopods of the two is the presence of a strong spinous process that extends laterally beyond the marginal process in S. tolimensis, but is lacking in S. cha- parralensis. Chaceus davidi, new species Figs. 2, 4b, e Material.—Ciudad Perdida, Sierra de Santa Marta, Magdalena Department, Colombia; 25 Jan 1982; C. Mejia: 1 male holotype: cl. 16.5 mm, cb. 29.0 mm (ICN-MHN N° CR-0083). Description.—The cervical groove is shallow and almost straight, and reaches VOLUME 97, NUMBER 3 541 Fig. 3. Hypolobocera marthelatani (Pretzmann, 1965), from Inza, Cauca Department, Colombia, left gonopod: a, Caudal view; b, Lateral view; c, Cephalic view; d, Apex in distal view. the lateral margin of the carapace. The anterolateral margin has a shallow notch behind the orbit, and a sharp angle at the place where the cervical groove meets the margin; it is completely devoid of teeth on its anterior half; the posterior half bears 20 small, ill defined teeth. The post-frontal lobes are weakly marked, the median groove is shallow and wide. The surface of the carapace behind the front is moderately inclined anteriorly and towards the midline. The front has a defined upper margin with ill defined tubercles; in dorsal and frontal view it is slightly bilobed; the lower margin is visible in dorsal view; it is strongly sinuous in frontal view. The carapace is smooth and shiny, with small punctae scattered all over its surface, not visible to the naked eye. The chelipeds are very unequal in size; the palm of the larger chela is moderately swollen, without a tubercle at the base of the fingers; the fingers are moderately gaping towards midlength. The merus of the third maxilliped does not have a low angle on the distal half of its external margin; the exognath is 0.7 the length of the ischium. The orifice of the efferent branchial channel is closed by a spine on the jugal angle and by the production of the lateral lobe of the epistome. The male gonopod is stocky; the caudal lobe is strongly produced beyond the apex, with a strong ridge on its middle part; the spermatic channel is bordered by a finger-like projection and another triangular projection; the caudal margin behind these projections is typically rolled. Size.—This is a small species. The holotype and only specimen has a cb. of 29.0 mm. Remarks.—The species is most closely related to Chaceus pearsei (Rathbun, 542 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ee a Va O d Fig. 4. Third maxilliped and left opening of branchial channel: a, d, Strengeria chaparralensis; b, e, Chaceus davidi; c, £, Hypolobocera marthelatani. 1915). The main differences between the gonopod of this species and the other in the genus is in the caudal lobe, which in C. davidi is strongly produced beyond the apex, and in the finger-like projection near the spermatic channel, which is smaller in this species, approximately of the same length of the triangular pro- jection located below it. The gonopods of all species of Chaceus have the same basic elements. Rodriguez (1982) has theorized on the possible derivation of the genus Hypolobocera from an ancestral Chaceus based on the homology of the finger-like and triangular projections with the two papillae found near the sper- matic channel in Hypolobocera. The present new species offers new support to this theory since these projections are surrounded by a ridge that somewhat resembles the button-shaped apex of Hypolobocera (Fig. 2c, f), and in caudal view there is an indication of the caudal ridge of this same genus. Etymology.—The species is named in honor of David Campos. Hypolobocera marthelatani (Pretzmann, 1965) Fig. 3, 4c, f Strengeria (Strengeria) marthelatani Pretzmann, 1965:6. Hypolobocera (Hypolobocera) marthelatami Pretzmann, 1971:17; 1972:50. Hypolobocera marthelatami (Pretzmann).— Rodriguez, 1982:52. Material.— Municipio Inza, Cauca Department, Colombia; 12 Oct 1982: 1 male cl. 23.0 mm, cb. 14.2 mm (ICN-MHN N?° CR-0087), 1 female cl. 25.5 mm, cb. 15.5 mm. VOLUME 97, NUMBER 3 543 Descriptions.—Our specimens from the Cauca Department agree well with Pretzmann’s description. The cervical groove is ill defined, consisting of a shallow oval depression on the posterior half, obsolescent on the anterior half. The an- terolateral border of the carapace has approximately ten ill defined small papillae. The postfrontal lobes are small, round, almost obsolete. The median groove is absent. The carapace is regularly inclined towards the middle; this surface in frontal view forms a sinuous line which parallels the line, also sinuous, of the lower frontal margin. The upper margin of the front is strongly bilobed in dorsal view; there is not a defined ridge, but the carapace in this area curves regularly downwards. The front is low. The lower margin in frontal view is sinuous. The third maxilliped has a well developed external angle. The exognath is 0.9 the length of the ischium. The surface is smooth and shiny, with numerous pores not visible to the naked eye. The chelae in our specimens are moderately swollen. In Pretzmann’s illustrations (1972, figs. 243, 244) they are considerably more in- flated. Size.— This is a small species. A mature female has a cb. of 25.5 mm. Remarks.—The species is very similar to Hypolobocera orientalis Pretzmann, 1968, from which it can be distinguished only by the shape of the gonopod apex, which in H. orientalis is more evenly rounded (see Rodriguez 1982, fig. 26d), not elongate as in H. marthelatani. The original spelling of the species name was marthelatani (Pretzmann, 1965) although Pretzmann (1971, 1972) latter used the spelling marthelatami. Acknowledgments We thank Dr. Horton H. Hobbs, Jr. for his critical review of the manuscript. This research was supported in part by a Colciencias Grant to M. R. Campos, Number 10000-1-138-82. Literature Cited Pretzmann, G. 1965. Vorldufiger Bericht iiber die Familie Pseudothelphusidae.—Anzeiger der Os- terreichischen Akademie der Wissenschaften Mathematische Naturwissenschaftliche Klasse, Jahrgant 1965, 1:1-11. . 1968. Neue Siidamerikanische Siisswassekrabben.—Entomologisches Nachrichtenblatt, Wien 15(1):1-15. 1971. Fortschritte in der Klassifizierung der Pseudothelphusidae.—Sitzungsberichter der Osterreichischen Akademie der Wissenschaften Mathematische Naturwissenschaftliche Klasse (1)179:14—24. . 1972. Die Pseudothelphusidae (Crustacea Brachyura).— Zoologica, Stuttgart 42(120):1-182. Rodriguez, G. 1982. Les crabes d’eau douce d’Amérique. Famile des Pseudothelphusidae. —Faune Tropicale 22:1—223. , and H. Diaz. 1981. New species of freshwater crabs from the Andes (Crustacea:Decapoda: Pseudothelphusidae). —Senckenbergiana Biologica 61(3/4):305—312. (MRC) Universidad Nacional, Instituto de Ciencias Naturales Apartado aéreo 7495, Bogota, Colombia; (GR) Instituto Venezolano de Investigaciones Cienti- ficas, Apartado 1827, Caracas 1010-A Venezuela. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 544-549 A NEW SPECIES OF THE GENUS CAMBARINCOLA (CLITELLATA: BRANCHIOBDELLIDA) FROM CALIFORNIA Perry C. Holt Abstract. — Cambarincola pamelae, new species, is described and compared with its near relative, C. mesochoreus Hoffman, 1963. Specimens from California previously assigned to C. mesochoreus are reassigned to C. pamelae and the status of presumptively eastern species of the branchiobdellids on the Pacific versant is discussed. The branchiobdellids of the Pacific versant in the United States have been studied most recently in a series of papers (Hoffman 1963; Holt 1960, 1967, 1974a, b, a, b, 1981a, b) with no pretensions that these efforts describe the totality of the branchiobdellid fauna of the region. Recently specimens were received for identification from Dr. Pamela Roe of California State College, Stanislaus, Tur- lock, California. Among them were specimens at first identified as Cambarincola mesochoreus Hoffman, 1963. Further examination revealed a distinct difference between the structure of the male bursal complex of the western worms and their more easterly congeners of the Mississippi Valley. This opportunity is taken to describe the new species from California. Cambarincola pamelae, new species Figs. 1-2 Cambarincola mesochorea Hoffman, 1963:307—311 (in part). Cambarincola mesochoreus.— Holt, 1973:10; 1981:689 (in part). Type-specimens.— Holotype, USNM 080687, 4 paratypes, USNM 080688—- 080691, and 10 paratypes, PCH 4065, taken on Procambarus (Scapulicambarus) clarkii (Girard, 1852) from an irrigation canal that drains into the San Joaquin River in the western part of Stanislaus County, California, by J. A. Meeuwse, 2 Dec 1982. Diagnosis.— Medium-sized worms (holotype 2.3 mm in length); lips entire; no oral papillae; no dorsal ridges; jaws subequal in size, small, dental formula 5/4; bursa about '4 body diameter in length, elongate ovoid, atrial fold present, penial sheath more than 4 total length of bursa, retracted penis greater in length than penial sheath; spermiducal gland short, thick, with deferent lobes, reflexed; pros- tate greater in length than spermiducal gland, subequal to latter in diameter, composed of granular cells, without ental bulb; spermatheca with long ectal duct, ovate bulb, subequal to body diameter in total length. Etymology.—For the discoverer, Pamela Roe. Description.— Five specimens of Cambarincola pamelae, including the holo- type, have the following approximate dimensions: total length, 2.6 mm; greatest diameter, 0.5 mm; head length, 0.4 mm; head diameter, 0.3 mm; diameter, segment I, 0.3 mm; diameter, sucker, 0.3 mm. VOLUME 97, NUMBER 3 545 Fig. 1. Cambarincola pamelae: a, Lateral view of holotype; b, Lateral view of reproductive systems of holotype. The lips are without lobes; the margins of the mouth without oral papillae. Mid-ventrally there is a shallow sulcus of the head and internally there is a prominent pharyngeal one. There are no supernumerary muscles, hence the body outline is smooth. The jaws are small, about '4, that of the head in length, and subequal in size. The dental formula is 5/4. 546 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. a—c. Cambarincola pamelae: a, Everted bursa and penis of a paratype; b, Optical section through bursa and penis of holotype; c, Jaws; d, Cambarincola mesochoreus, optical section through bursa and penis of paratype. The spermiducal gland is kidney-shaped and often extends to the dorsal border of the gut. Deferent lobes are present. The ental end of the posterior deferent lobe is often expanded with a wider lumen than the ectal part of the gland. The prostate is about 7% that of the spermiducal gland in width and greater than the latter in length. The lumen of the ental end, as in the spermiducal gland, is in some specimens expanded and in the holotype a clear ental bulb is present. VOLUME 97, NUMBER 3 547 The prostate is of the type known as “‘non-differentiated’’; its glandular cells are granular instead of vacuolated. The bursa is elongate ovoid, approximately 3 times its diameter and % of the diameter of the body in length. Somewhat more than half of the organ constitutes the penial sheath. The penis, though similar to that of other members of the genus, is folded or looped and parts of its lumen are seen in optical sections as narrow transverse spaces within the penial sheath. In specimens with an everted bursa, the penis is protruded as a relatively slender tube with strands of tissue (? muscular) extending internally its length (Fig. 2A). The ejaculatory duct is a relatively wide tube of the usual type. The spermathecal ectal duct is approximately equal to the body diameter in length, the bulb is ovoid. Variations.— The posterior deferent lobe of the spermiducal gland is dilated in some specimens, as is the ental end of the prostate. In some animals the prostate has an ental bulb (this is so for the holotype), but others just as clearly do not. Such variations have not been recorded for other species of the genus and their significance here is not clear; perhaps the ental bulb of the prostate should be re- evaluated as a specific character. Affinities. —Cambarincola pamelae is closely related to Cambarincola meso- choreus Hoffman, 1963:307, and the specimens of the type-series were at first identified as C. mesochoreus. The expanded lumen of the ental end of the prostate or of the spermiducal gland (absent in C. mesochoreus) in some of these specimens occasioned a more careful study and the difference in the structure of the penis and its enclosing sheath was noted. That of C. mesochoreus is of the more common type; the bursa is subspherical and the penis short and muscular. Holt (198 la: 689) said in his modification of Hoffman’s (1963:308) original diagnosis of C. mesochoreus that the spermiducal gland is slender and without deferent lobes. It is slender only in the sense that it is usually less in diameter than the prostate and never greater in this dimension. The deferent lobes ascribed to the species by Hoffman (1963:308) are either absent or obscure in four paratypes from Spencer County, Indiana (PCH 817), but are not prominent in C. pamelae. The sper- matheca of C. mesochoreus was described in the original diagnosis as “‘subfusi- form” and as having a “blunt ental process.”’ The paratypes from Indiana fail to confirm these statements; the spermathecal bulb varies in shape from subfusiform to broadly oval and no ental process can be seen; there are no obvious differences between the spermathecae of these species. The jaws of C. mesochoreus are perhaps marginally larger than those of C. pamelae. Hoffman’s specimens ranged in length from 2.8—4.2 mm; the length of the holotype and four paratypes of C. pamelae extends from 2.5 to 3.3 mm, probably an insignificant difference. The two species are closely related and can be distinguished reliably only by the difference in the penes (Fig. 2b, d). Host.— Procambarus (Scapulicambarus) clarkii. Distribution.— The type-series of C. pamelae is from the western part of Stan- islaus County, California, in the irrigated portion of the San Joaquin Valley. The specimens assigned by Holt (1981a:689) to C. mesochoreus, and herewith re- assigned to C. pamelae, were all taken from the introduced crayfish, Procambarus (Scalpulicambarus) clarkii, in Santa Barbara, Merced and Sonoma counties, Cal- 548 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ifornia. The assumption that C. pamelae is likewise an introduction is easily made, but not necessarily valid; Cambarincola gracilis Robinson, 1954, is un- doubtedly indigenous to the Pacific versant, is known from Santa Barbara County northward to southern British Columbia and is associated with C. pamelae in Santa Barbara and Stanislaus counties on the same host. Cambarincola barbarae Holt, 1981, is also associated with C. pamelae in Santa Barbara and Sonoma counties (Holt 1981a:678-679) and is not known from the east. The report of (Holt 1981a:680) of Cambarincola fallax Hoffman, 1963, from Sonoma County, California, from these considerations, becomes suspect: the specimen was iden- tified on the basis of the tentaculated upper lip and the 5/5 dental formula, diagnostic of C. fallax in the east. Perhaps C. pamelae and C. mesochoreus form another case of the phenomenon exemplified by Cambarincola shoshone Hoffman, 1963, and Cambarincola branchiophilus Holt, 1954; Cambarincola macrocephelus Goodnight, 1940, and Cambarincola holti Hoffman, 1963: pairs of very similar species, one western and the other eastern. The eastern crayfishes introduced into western waters might be expected to carry their branchiobdellid symbionts with them. But great care is often required to bring branchiobdellids from the field to the laboratory for even short distances. Since it is likely that these introductions were made by people unaware of the worms, it is unlikely that they took the care (moving the animals at low temper- atures and avoiding overcrowding) necessary to assure the survival of the bran- chiobdellids. But the range of none of the species involved is well enough known to allow reasonable conjectures as to the origin of the symbionts of the eastern crayfish introduced into the west. Cambarincola pamelae may be, though it prob- ably is not, an eastern, introduced species. Material examined.—The types from Stanislaus County, California, and that previously assigned to Cambarincola mesochoreus (Holt, 1981a:689). Acknowledgments The effort of Dr. Roe in making available the material on which the new species is based is appreciated as is the kindness of Dr. Horton H. Hobbs, Jr. and Mr. C. Willard Hart, Jr. in reading the manuscript. Literature Cited Goodnight, Clarence J. 1940. The Branchiobdellidae (Oligochaeta) of North American crayfishes. — Illinois Biological Monographs 17(3):1—75, pls. I-III. Hoffman, Richard L. 1963. A revision of the North American annelid worms of the genus Cam- barincola (Oligochaeta: Branchiobdellidae).— Proceedings of the United States National Mu- seum 114 (3470):271-371, 79 figs. Holt, Perry C. 1954. A new branchiobdellid of the genus Cambarincola Ellis (Oligochaeta, Bran- chiobdellidae) from Virginia. — Virginia Journal of Science (new series) 5(3):168—-172, 5 figs. . 1960. The genus Ceratodrilus Hall (Branchiobdellidae, Oligochaeta), with the description of a new species.— Virginia Journal of Science (new series) 11(2):53-77, pls. I-IV. 1967. Status of the genera Branchiobdella and Stephanodrilus in North America with de- scription of a new genus (Clitellata: Branchiobdellida).— Proceedings of the United States Na- tional Museum 124 (3631):1-9, 4 figs. . 1974a. An emendation of the genus Triannulata Goodnight, 1940, with the assignment of Triannulata montana to Cambarincola Ellis, 1912.—Proceedings of the Biological Society of Washington 80(8):57-72, 5 figs. VOLUME 97, NUMBER 3 549 1974b. The genus Xironogiton Ellis, 1919 (Clitellata: Branchiobdellida).— Virginia Journal of Science (new series) 25(1):5—-19, 6 figs. 1977a. An emendation of the genus Sathodrilus Holt, 1968 (Annelida: Branchiobdellida), with the description of four new species from the Pacific drainage of North America.—Pro- ceedings of the Biological Society of Washington 90(1):116-131, 7 figs. 1977b. A gill-inhabiting new genus and species of the Branchiobdellida (Annelida: Clitel- lata). — Proceedings of the Biological Society of Washington 90(3):726—-734, 5 figs. . 198la. A resumé of the members of the genus Cambarincola (Annelida: Branchiobdellida) from the Pacific drainage of the United States. — Proceedings of the Biological Society of Wash- ington 94(3):675-695, 5 figs. . 1981b. New species of Sathodrilus Holt, 1968, (Clitellata: Branchiobdellida) from the Pacific drainage of the United States, with the synonymy of Sathodrilus virgiliae Holt, 1977.—Pro- ceedings of the Biological Society of Washington 94(3):848-862, 3 figs. Robinson, Dorothy A. 1954. Cambarincola gracilis sp. nov., a branchiobdellid oligochaete com- mensal on western crayfishes.— Journal of Parasitology 40:466—469, pl. I. Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 550-554 PROBOPYRINELLA HEARDIN. SP. (ISOPODA: BOPYRIDAE) A BRANCHIAL PARASITE OF THE HIPPOLYTID SHRIMP LATREUTES PARVULUS (DECAPODA: CARIDEA) Daniel L. Adkison Abstract.—The bopyrid isopod Probopyrinella heardi n. sp. occurs in the bran- chial chambers of the shrimp Latreutes parvulus in coastal waters off Georgia, Mississippi, and eastern Florida. Probopyrinella heardi can be distinguished from P. latreuticola, the only other described species in the genus, by the asymmetrical development of the posterior lamina and first oostegite of the female of P. heardi. Chopra (1923) first recognized from published descriptions that Bopyroides latreuticola Gissler, 1882, represented a distinct genus, but he did not name the genus since specimens of B. Jatreuticola were unavailable for study. Based on the study of material, Nierstrasz and Brender a Brandis (1929) followed Chopra’s suggestion and proposed the new genus Probopyrinella with B. latreuticola as the type and only species. Markham (1977) reviewed the taxonomic history of Bopyroides latreuticola and also clarified the differences between Probopyrinella and the four genera (Bopyrella Bonnier, 1900; Bopyrina Kossmann, 1881; Bopyroides Stimpson, 1884; and Pro- bopyrus Giard and Bonnier, 1888) to which P. latreuticola had previously been assigned. An amended generic diagnosis is presented for Probopyrinella based on study of P. heardi, P. latreuticola and previous generic diagnoses. Probopyrinella Nierstrasz and Brender a Brandis, 1929 Generic diagnosis.— Female: Anterolateral edge of head acutely pointed on re- duced side; frontal lamina only slightly developed; posterior ventral lamina, 1 pair of projections. Eyes present. Maxilliped with palp. Head fused with pereomere 1 in one species, other pereon segments distinct. Coxal plates and dorsal bosses absent. Lateral margin of expanded side of pereon distinct; margin on reduced side less distinct. Pereopods of reduced side visible in dorsal view. Oostegite 1, posterior plate more than half width of anterior plate; oostegite 5 on reduced side large, sickle-shaped; other oostegites small, only marginally enclosing brood cham- ber. Pleon of 6 segments, fused dorsally, well defined laterally by incisions only on expanded side. Pleopods 4 and ’5 biramous pairs, except last pair either uni- ramous or biramous. Uropods absent. Male: head slightly narrower than pereomere |. Pereon with 7 distinct segments. Pleon, 6 segments at least laterally indicated. Pleopods, 5 uniramous pairs, reduced to low protuberances. Uropods absent. Discussion. — The large fifth oostegite appears to be an important generic char- acter (Heard, pers. comm.) that has not previously been recognized. If in further VOLUME 97, NUMBER 3 551 study the enlarged fifth oostegite proves to be an important character, then many of the Bopyridae parasitic on caridean shrimps of the family Hippolytidae are more closely related than is now thought. Probopyrinella heardi, new species Rigseley Bopyridae sp. A, Camp, Whiting and Martin. 1977:7, 27. Material examined. —(All are paratypes except holotype.) All infesting Latreutes parvulus. Georgia: Pass Warsaw Sound, Cabbage Island, Chatham County; 81°55'’N 31°55'W; 4-8 m; 30 Aug 1972; R. W. Heard collected and determined host; 1 9 (gravid, no male), USNM 172452 (holotype, host present). Same location; 4—8 m; 13 Oct 1972; R. W. Heard collected and determined host; 2 2, 2 6, USNM 172453 (host present).— Florida: East of Florida Power and Light electrical gen- erating plant, Hutchison Island, St. Lucie County; Station 1, EJ-72-132; 27°21'06"N 80°13'08"W; 8.5 m; 5 Jul 1972; R. M. Gallagher, F. S. Kennedy collected; D. K. Camp determined host; 2 2, 2 6, FSBC I 20838 (host FSBC I 16102). Same location, EJ-72-190; 6.7 m; 2 Nov 1972; R. M. Gallagher, C. R. Futch, N. H. Whiting collected; D. K. Camp determined host; 1 2, 1 6, FSBC I 20839 (host FSBC I 16103). Marco Island, Collier County; Station 18, 2, 1 (station location not known; 81°42'N 25°56’W; Nov 1972; D. L. Adkison determined host; 2 2, 1 6, MNHN Ep. 112, 113; 2 2 (gravid), 1 6, ZMC. Near pass, Lemon Bay; 82°21'N 26°54’W; trawl; 29 Jan 1983; O. Hartman collected; L. B. Holthuis determined host; 1 2, 1 6, USNM 172416 (host present).— Alabama: Pass, west end of Dauphin Island, Mobile County; 30°14’N 88°23’W; A-frame dredge net; 4—7 m; 14 Feb 1977; R. W. Heard collected and determined host; 1 2 (no male), USNM 172454.—Mis- sissippi: Approximately 19 km south of Mississippi River; approximately 29°11'N 88°37'W; trawl 22-26 m; 10 Oct 1977; collected by shrimper; D. L. Adkison determined host; 1 2, 1 6, USNM 172455 (host present). Description. — Female (Figs. 1, 2a, b): Body asymmetrical, distortion angle 40- 65°; length 1.7—2.0 mm; width across pereomere 3, 1.2—1.5 mm. Head fused with first pereomere; anterolateral corner of head acute on reduced side, other side with corner broader. First antenna, 3 or 4 segments; basal segment, no setae; second segment, 2 apical setae; distal segment, 4 setae as terminal tuft. Second antenna, 2 segments; no setae; distal segment, 3 to 6 setae. Second antenna greater than ' length of first antenna. Maxilliped quadrate, anterior plate larger than posterior plate; small palp present. Posterior lamina, | pair of projections; projection on reduced side longer than that on expanded side. Pereon segmentation often medially indistinct between pereomeres 2 and 3 (2 specimens). Lateral margins of pereomeres of expanded side well defined, notch often present in posterior 3 of pereomeres 2 and 3. Lateral margin on reduced side curled dorsally creating shallow trough medial to margin and exposing per- opods to dorsal view. Dorsal bosses absent. Coxal plates absent (present only on pereomere 1 of expanded side in 2 specimens). Tergal area of pereomere 1 on reduced side with conical projection. Pereopods with basal carina; pereopods decreasing in size from pereopods 4 or 5; pereopods smaller on reduced side. First oostegite, posterior plate width nearly equal to width of anterior plate; internal 552 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Probopyrinella heardi, female: a, Dorsal view; b, Antennae and oral cone; c, Oostegite 5 from reduced side; d, Pereopod 4, expanded side; e, Head, viewed in dorsal plane of head; f, Holotype, dorsal view; g, Maxilliped; h, Maxilliped palp; i, Oostegite 1 from expanded side, internal view; j, Oostegite 1 from reduced side, internal view; k, Posterior lamina; 1, Pleon of a, ventral view; m, Pleon of f, ventral view. Figures from holotype, e-—k, m. Figures from paratype FSBC I-20839, a and 1. Figures from USNM 172453, b-d. Scale: A = 1.0 mm (Figs. a, d, f, i, j, m). ridge unornamented; first oostegites asymmetrically developed, oostegite on re- duced side enlarged. Fifth oostegite on reduced side enlarged; other oostegites reduced, only fringing brood chamber. Pleon deeply set in pereomere 7. Six segments laterally indicated. Segments 1- 5 on expanded side defined by deep notch, anterior edge of segments curled ventrally. Pleomeres on reduced side and the sixth pleomere less clearly defined than on expanded side. Pleopods, 4 or 5 pairs, anterior pairs biramous, last pair VOLUME 97, NUMBER 3 553 Fig. 2. Probopyrinella heardi, a, b, female, immature; c-f, male: a, Dorsal view; b, Pleon, ventral view; c, Dorsal view; d, Pleon, dorsal view; e, Pleon, ventral view; f, Antenna. Figures from USNM 172454, a and b. Figures from FSBC I-20838, c-f. Scale = 1.0 mm. uniramous; pleopods decreasing in size posteriorly until posterior pair only small tubercle or ridge. Uropods absent. Male (Fig. 2c—f): Length 0.7—-1.1 mm; width across pereomere 3 or 4, 0.2-0.4 mm. Head wider than long, distinct from first pereomere. Eyes present. First antenna, 3 segments; basal segment, 0 to 2 setae; second segment, 2—4 setae; distal segment, 3 to 5 setae. Second antenna, 2 segments; basal segment, | or 2 setae; distal segment, 2—4 setae. Second antenna '2 to % length of first antenna. Maxillipeds not seen. Pereon 7 segments, increasing in width until pereomere 3 or 4, thereafter de- creasing in width posteriorly; pigment spots on dorsal surface of most pereomeres. Pereopods decreasing in size both anteriorly and posteriorly from pereopod 4. Pleon 6 segments indicated laterally by notches, medially fused; pleomere 1 subequal in width to pereomere 7. Pleopods, 5 uniramous pairs, decreasing in size posteriorly. Uropods absent. Etymology.—This species is named for Richard W. Heard who collected many of the specimens and who first recognized this species as undescribed. Type series.—The holotype and some paratypes have been deposited in the National Museum of Natural History, Smithsonian Institution (USNM). The remaining paratypes have been deposited in Florida Department of Natural Re- sources, St. Petersburg (FSBC); Muséum National d’Histoire Naturelle, Paris (MNHN); and Universitetets Zoologiske Museum, Copenhagen (UZM). Type-locality. — Warsaw Sound (off Cabbage Island) Chatham County, Georgia, U.S.A. Distribution.—Probopyrinella heardi is known from the coastal waters of the southeastern United States from Georgia to Mississippi. Discussion. — While the two species of Probopyrinella are very similar in dorsal 554 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON view, Probopyrinella heardi can be distinguished from P. /atreuticola by the fol- lowing characters for the former: (1) female with head and first pereomere fused, (2) female with second antenna of 2 segments, (3) female with posterior lamina and first oostegite asymmetrically developed, (4) male with second antenna of two segments, and (5) male with segmentation of pleon less distinct. Probopyrinella heardi, P. latreuticola, and the two undescribed species men- tioned by Chopra (1923:537), all infest members of the hippolytid genus Latreutes Stimpson, 1860: P. heardi on L. parvulus, P. latreuticola on L. fucorum (Fabricius, 1798), and the undescribed species on L. mucronatus (Stimpson, 1860) and L. pygneaus Nobili, 1904. Acknowledgments This paper represents part of a thesis submitted to the Biology Department of the University of West Florida. I wish to thank Thomas S. Hopkins, Dauphin Island Sea Laboratory, Dauphin Island, Alabama for financial support, encour- agement, and working space while I was working at Dauphin Island. I wish to thank Thomas E. Bowman (USNM) and David K. Camp (FSBC) for loan of specimens in their care. I am deeply indebted to Richard W. Heard for the loan of material he collected and sharing of his knowledge of the Epicaridea over the past several years. Literature Cited Bonnier, J. 1900. Contribution a l’étude des épicarides. Les Bopyridae.—Travaux de la Station Zoologique de Wimereux 8:1—476. Camp, D. K., N. H. Whiting, and R. E. Martin. 1977. Nearshore marine ecology at Hutchinson Island, Florida: 1971-1974. V. Arthropods.— Florida Marine Research Publications No. 25: 1-63. Chopra, B. 1923. Bopyrid isopods parasitic on Indian Decapod Macrura.—Records of the Indian Museum 25:411-550. Giard, A., and J. Bonnier. 1888. Sur deux nouveaux genres d’Epicarides (Probopyrus et Palaegyge).— Bulletin Scientifique de la France et de la Belgique 19:53-77. Gissler, C. F. 1882. Bopyroides latreuticola, a new species of isopod crustacean parasitic on a gulf- weed shrimp.— American Naturalist 16:591—-594. Kossmann, R. 1881. Studien iiber Bopyriden: II. Bopyrina virbii, Beitrage zur Kenntniss der Ana- tomie und Metamorphose der Bopyriden. — Zeitschrift fiir Wissenschaftliche Zoologie 35:662— 680. Markham, J. C. 1977. Distribution and systematic review of the bopyrid isopod Probopyrinella latreuticola (Gissler, 1882).—Crustaceana 33:189-197. Nierstrasz, H. F., and G. A. Brender a Brandis. 1929. Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-1916. 48. Epicaridea. I.— Videnskabelige Meddelelser fra Dansk Naturhis- torisk Forening 87:1—44. Stimpson, W. 1860. Descriptions of new species of marine Invertebrata from Puget Sound, collected by the naturalists of the North-west Boundary Commission, A. H. Campbell, Esq., Commis- sioner. — Proceedings of the Academy of Natural Sciences of Philadelphia 16:153-161. Department of Biology, Tulane University, New Orleans, Louisiana 70118. Present address: Department of Veterinary Pathology, Texas A&M University, College Station, Texas 77843-4463. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 555-582 NEW SPECIES OF PHYLLODOCIDAE AND HESIONIDAE (POLYCHAETA), PRINCIPALLY FROM FLORIDA Thomas H. Perkins Abstract.— The following new species are described: Eumida (Eumida) parvi- cirrus, Eumida (Pirakia) hutchinsonensis, Paranaitis gardineri, Heteropodarke lyonsi, H. formalis, Kefersteinia haploseta, and Podarkeopsis levifuscina. Kefer- steinia sp. is informally described. Oxydromus arenicolus glabrus Hartman, pre- viously referred to Oxydromus brevipalpa Hartmann-Schréder, is a distinct species which, with O. brevipalpa, Oxydromus capensis Day, and Gyptis maraunibinae Gibbs, are species of Podarkeopsis Laubier, and new combinations. This report is one of several (Perkins 1979, 1980, 1981, 1984) based primarily on collections made between September 1971 and July 1973 in an environmental baseline study of marine biota near the Florida Power and Light Company nuclear power plant at Hutchinson Island, St. Lucie County, Florida. Additional speci- mens collected later at Hutchinson Island and specimens from North Carolina and the eastern Gulf of Mexico are included. A species of Hesionidae, Microph- thalmus hartmanae, has been described previously from these collections (Westh- eide 1977), and another Microphthalmus species awaits description (Wilfried Westheide, in litt.). The study area and methods were described by Gallagher and Hollinger (1977). Sediments were described by Gallagher (1977). Other aspects of the physical and chemical environment were reported by Worth and Hollinger (1977). Brief de- scriptions of benthic sampling stations and methods were also given by Perkins (1979). Types and other material available for study are deposited in the Allan Hancock Foundation, University of Southern California (AHF); British Museum (Natural History) (BMNH); Invertebrate Reference Collection of the Florida Department of Natural Resources Bureau of Marine Research (FSBC I); Marine Environmental Sciences Consortium, Dauphin Island, Alabama (MESC); Mote Marine Labora- tory, Sarasota, Florida (MML); U.S. National Museum of Natural History, Smith- sonian Institution (USNM); Universitetets Zoologiske Museum, Copenhagen (ZMC); and Zoologisches Institut und Zoologisches Museum, Hamburg Univer- sity (ZMH). Dr. Marian H. Pettibone (USNM), Dr. Thomas Hopkins (MESC), and Mr. Jay Leverone (MML) loaned specimens. Mr. Robert G. Ernest, Applied Biology, Inc., Jensen Beach, Florida, and Mr. Harvey Rudolph, Florida Department of Envi- ronmental Regulation, donated specimens. Specimens of Kefersteinia sp. loaned by Mr. Leverone were collected and identified by personnel of Mote Marine Laboratory for the Bureau of Land Management, contract no. AA815-CTO-50. James F. Quinn, Jr., of the Bureau of Marine Research helped with the Latin names. Kristian Fauchald (USNM) provided a copy of a description not available 556 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON to me. Kristian Fauchald, Karen A. Steidinger and William G. Lyons critically read the manuscript. Many other individuals from the Bureau of Marine Research and from Applied Biology, Inc., participated in the Hutchinson Island study for which Florida Power and Light Company provided partial funding. Family Phyllodocidae Williams, 1851 Genus Eumida Malmgren, 1865, emended Diagnosis.—Body usually long, with numerous segments. Prostomium cone- shaped, oval, pear-shaped, or heart-shaped, with 2 eyes and 5 antennae. First segment reduced dorsally, visible laterally and ventrally, other tentacular segments complete and distinct from one another; 4 tentacular cirri, all cirriform, or ventral cirri of segment 2 slightly flattened and oval in cross section. Tentacular formulae: (seg rat geen ces al an al an Parapodia uniramous, with only compound setae; dorsal cirri oval, lanceolate or heart-shaped, ventral cirri originating near ventral margin, and equai to or smaller than setal lobes in posterior view, with axis parallel to aciculum or diverging ventrally. Proboscis smooth or nearly so or diffusely covered with conical, round- ed, or cylindrical papillae, not divided into 2 regions when everted. (Diagnosis modified from Hartmann-Schréder 1971, incorporating characters of ventral cir- rus suggested by Banse 1973.) Remarks.—The systematics of Eumida Malmgren, 1865, s.s., and related genera (e.g., Pirakia Bergstrom, 1914; Sige Malmgren, 1865; and Pterocirrus Claparéde, 1868) is complicated and somewhat confused. A system based on principles proposed by Bergstr6m (1914), with minor emendations and additions by later authors, was given by Fauchald (1977:45-51). Fauchald retained the identities of most genera maintained or described by Bergstr6m. Other authorities (e.g., Day 1967; Hartmann-Schréder 1971; Uschakov 1972; Banse and Hobson 1974) in- cluded some related genera of Fauchald’s system within the genus Eumida, with or without subgeneric status, or included these, Eumida, and still other genera of Fauchald’s system as subgenera of Eulalia Savigny, 1817. Eumida and related genera differ from Eulalia in having a dorsally reduced first segment; however, see Banse (1973) for a discussion of reservations concerning the importance of this character. Of the four generic taxa I consider related, Pirakia differs from Eumida in the degree and type of papillation of the proboscis. This may be an important generic character; Hartmann-Schréder (1971) stated that the proboscis of the type-species of Eumida, Eulalia sanguinea Orsted, 1843, was smooth or wrinkled only from contraction, whereas the proboscis of the type-species of Pirakia, Phyllodoce (Eu- lalia) punctifera Grube, 1860, was diffusely covered with small, conical papillae. However, there may be intermediate conditions of this character; other authorities (e.g., Pettibone 1963; Day 1967; Uschakov 1972) have described oval bumps or scattered papillae of various shapes on specimens reported as Eumida sanguinea and other Eumida species. These may prove to be incorrectly identified, and the apparent intermediate conditions may not exist. Therefore, I am retaining Pirakia, but only as a subgenus of Eumida. Sige and Pterocirrus, the two remaining similar generic taxa, differ from Eumida VOLUME 97, NUMBER 3 557 in having broadly lamellate ventral tentacular cirri on segment 2; I consider this an important generic character. In the strict sense, these genera differ from each other in the setation of tentacular segments, Sige having setae on segments 2 and 3, and Pterocirrus lacking setae but apparently having acicula on those segments (Banse 1959; Uschakov 1972). However, a form described from Japan [Sige macroceros (Grube, 1860) variety orientalis Imajima and Hartman, 1964:70, pl. 14, figs. c-f] is apparently intermediate between these character states. (This var- ietal designation has no status according to the International Code of Zoological Nomenclature 1964, Article 45.) The form has setae on segment 3, suggesting that setation of anterior segments is not an important generic character for this group, and that Sige and Pterocirrus constitute but a single genus without sub- genera. This is in strict agreement with the system of Uschakov (1972). Lack of importance of setation of anterior segments as a generic character is further sup- ported by intraspecific variation in this character noted in the species of Eumida (Pirakia), below. Uschakov used Pterocirrus Clapareéde rather than the older name Sige Malmgren for this taxon. He believed, after examination of a specimen from the Oresund, near the type-locality, that Malmgren’s original description and figure of Sige fusigera (Malmgren 1865:100, pl. 14, fig. 27), the type-species, were correct and concluded that Sige fusigera was a species of Eumida Malmgren. However, Berg- strom (1914) stated earlier that he had examined the holotype of S. fusigera deposited in the Swedish State Natural History Museum and other specimens of the species from the type-locality north northwest of the Oresund along the south- ern Norwegian and Swedish west coast. The “holotype” he examined [which should only be considered part of Malmgren’s original material, according to Kristian Fauchald (pers. comm.)] was damaged. The proboscis was torn loose from the mouth, and the part visible was the inner coelomic lining. This is suggested by Malmgren’s figure and is a condition common on one of the species described below. Bergstrém also stated that ventral tentacular cirri of the second segment were lamelliform and that the proboscis was diffusely papillate. At this time, Bergstr6m’s arguments are more convincing than those of Uschakov, and I continue to maintain Bergstr6m’s concept of Sige Malmgren, but as modified by Uschakov under the name Pterocirrus Claparéde. However, it is certainly possible that Malmgren had both the species reported as Eumida fusigera by Uschakov (1972) [here considered a Eumida species] and also the species reported by Bergstrom (1914) as Sige macroceros [not Phyllodoce (Eulalia) macroceros Grube, 1860, fide Banse (1959) but here considered to be Sige fusigera Malmgren, 1865]. This matter warrants further investigation. Additionally, I include in Eumida species having slightly flattened ventral ten- tacular cirri on segment 2. These occur on the species of Eumida (Pirakia) de- scribed below, on the species reported as Eulalia (E umida) sanguinea [not Eulalia sanguinea Orsted, 1843] by Day (1967:155, fig. 5.5.A—C) from southern Africa, and probably on Eulalia (Sige) falsa Day (1960:303, fig. 6A—C; 1967:155, fig. 5.5.D-F). There is a great difference between the slightly flattened ventral ten- tacular cirri of these species and the broadly lamellate ones of Sige (see Claparéde 1868:560—-562; 1869: pl. 17, fig. 2; and Bergstr6ém 1914:70, text figure 23f—-g for descriptions and figures of ventral tentacular cirri of segment 2 of Sige). Finally, the intraspecific variation in setation of anterior segments noted in the 558 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON description of the species of Eumida (Pirakia), below, seems to require that the limits of Eumida be expanded. Accordingly, I have emended the generic diagnosis to include all specimens of this species by the addition of a second setal formula indicating that setae may be absent from segment 2. Subgenus Eumida Malmgren, 1865 Diagnosis.—Proboscis smooth or with scattered papillae, bumps, wrinkles, folds, or warts of various shapes. Eumida (Eumida) parvicirrus, new species Fig. 1 Material examined.—FLORIDA: Hutchinson Island Sta II, 27°21.6’N, 80°13.2’W, 11 m, coarse calcareous sand, holotype (USNM 80510). Same, Sta V, 27°22.9'N, 80°13.9’W, 11 m, coarse calcareous sand, paratype (USNM 80511). Description.—Dorsum of holotype brown on segments 2-6, posterior lobes of brain mass brown (Fig. 1A), internal pigmented areas ventromedial to ventral cirri (Fig. 1C). Holotype complete, 3 mm long excluding proboscis, 30 segments; complete paratype smaller, about 2 mm long, 20 segments. Prostomium oval- pentagonal (Fig. 1A), about twice as broad as long; posterior margin almost straight; frontal antennae moderately slender, subulate; median antenna shorter and more slender than frontal ones, originating slightly in front of level of eyes; eyes pos- terolateral. First segment reduced dorsally, visible only laterally and ventrally. Tentacular formula: IL =F qt Be gt al an Setal lobes of segment 2 small, conical, with several setae. Tentacular cirri slender, subulate, about 7 as long as prostomial width on segment |; upper cirri on segment 2 longest, extending to about segment 7; ventral cirri of segment 2 slightly longer than those of segment 1; dorsal tentacular cirri of segment 3 almost as long as upper cirri of segment 2. Dorsal cirri (Fig. 1B, C) similar throughout, only slightly smaller on anteriormost and far posterior setigers, on short cirrophores, pyriform, extending to tips of setal lobes or shorter, about 11 times longer than wide, with thickness equal to '2 width. Setigerous lobes conical; each with short filiform presetal lobe above aciculum, with obscure, rounded, postsetal lobe, with rela- tively long, slender aciculum extending into presetal lobe. Ventral cirri subulate or fusiform, about as long as dorsal cirri, about '2 as wide, not extending past tips of setigerous lobes. Setae (Fig. 1D-F) numbering about 20 per parapodium in middle segments, with slender shafts and slender, spinigerous blades; shafts with about 6 short, slender spines on tips; blades with short serrations on edge. Anal Cilri missing. Proboscis (Fig. 1A) relatively short and bulbous when everted, with smooth surface; distal opening, when fully everted, surrounded by 15-20 soft, semicircular lobes. Remarks. —Eumida parvicirrus is a small species for the genus. Setigerous lobes, ventral cirri, and possibly setae are similar to those of the specimen reported as VOLUME 97, NUMBER 3 559 0.015mm TI Fig. 1. Eumida (Eumida) parvicirrus, holotype: A, Anterior end, dorsal view; B, Parapodium, middle segment, posterior view, with about half of setae figured; C, Same, setae omitted; D, Seta, lateral view; E, Same, enlarged view of hinge region; F, Tip of shaft, face view. Eumida fusigera by Uschakov (1972) [not Sige fusigera Malmgren]. However, the proboscis of E. parvicirrus is smooth and dorsal cirri are not as slender. Eumida parvicirrus appears to be unique in the genus in having relatively large posterior lobes of the brain mass, the posterior parts of which are pigmented. Etymology.—The specific name is derived from the Latin parvus, small, and cirrus, curl, and refers to the small cirri of setigerous segments. 560 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Subgenus Pirakia Bergstrom, 1914 Diagnosis.—Proboscis diffusely covered with conical, rounded, or cylindrical papillae. Eumida (Pirakia) hutchinsonensis, new species Figs. 2, 3 Material examined.—FLORIDA: Hutchinson Island Sta V, 27°22.9’N, 80°13.9'W, 11 m, coarse calcareous sand, holotype (USNM 80515), 29 paratypes (USNM 80517, 80519, 80520; AHF Poly 1390; BMNH 1983.931-937; FSBC I 30418; ZMC; ZMH P-17602). Same, Sta II, 27°21.6’N, 80°13.2’W, 11 m, coarse calcareous sand, 17 paratypes (USNM 54278, 80516, 80518; FSBC I 30419, 30420). Same, Sta IV, 27°20.7’N, 80°12.8’'W, 11 m, coarse calcareous sand, 1 paratype (FSBC I 30421). Off Cape Canaveral, 28°35.9'N, 80°18.6’W, 26 m, on scallops, 9 specimens (FSBC I 17407). Off Palm Beach County-Broward County line S of Boca Raton, on Phragmatopoma lapidosa Kinberg, 1 specimen (USNM 80521). Description.—Color variable, tan to dark brown; tan specimens with few scat- tered pigment spots more prominent on antennae, tentacular cirri, dorsal cirri, and ventral cirri; dark-brown specimens with numerous pigment spots on pro- stomium and dorsum of body; spots prominent in transverse segmental bands; dorsal and ventral cirri of dark-brown specimens also strongly pigmented, darker than body; many specimens with color intermediate between these extremes. Complete type-specimens less than 10 mm long, with maximum of 75 segments. Prostomium (Fig. 2A, B) rounded-pentagonal, longer than wide, with posterior margin almost straight; frontal antennae moderately stout, similar; median an- tenna originating from near middle of prostomium well in front of eyes, much shorter and more slender than frontal ones; pair of moderately large, brown, lensed eyes near posterior margin; moderately large, triangular lobe behind eyes. Segment 1 reduced dorsally, visible only laterally and ventrally, with tentacular cirri lateral to eyes. Tentacular formulae: ji ge a ge or erates ee al an al an Segment 2 parapodia with 0-3 setae; some specimens with setae on both para- podia, others with setae on only one; about half of specimens without setae on this segment. Ventral tentacular cirri of segment 2 slightly flattened, oval in cross section; remaining tentacular cirri cirriform; dorsal cirri of segment 2 longest, reaching to about segment 8; setal lobes of segment 2 reduced or absent. Parapodia after segment 3 (Fig. 2C, D) with dorsal and ventral cirri, slightly bilobed presetal lobes and shorter, rounded postsetal lobes; single neuroaciculum and composite setae. Dorsal cirri short on anterior segments, gradually longer on anterior 3 of body, shorter near far posterior end, about twice longer than wide on middle segments, flattened, lanceolate to cordiform in outline, on cirrophores of moderate length; ventral cirri somewhat flattened, oval in outline, extending almost to tips of setal lobes, on short cirrophores; parapodia of middle segments with less than 10 setae. Setae (Fig. 3B—D) composite, with short, flattened, awl-shaped blades; blades with very fine, short serrations on edge, with pointed tips; shaft tips asym- metrical with several spines all of same size. Anal cirri (Fig. 3A) similar to dorsal VOLUME 97, NUMBER 3 561 Fig. 2. Eumida (Pirakia) hutchinsonensis: A, Anterior end, dorsal view; B, Same, with proximal part of proboscis torn from mouth; C, Parapodium, middle segment, posterior view; D, Same, anterior view, setae omitted (A, USNM 80517; B, USNM 80516; C, D, USNM 80519). cirri of middle segments, slightly thicker and longer, almost oval in outline on holotype. Proboscis (Fig. 2A, B) in about anterior 23 segments when inverted, formed of papillate region in anterior 13 segments and thickly muscled region with numerous circular muscle rings in last 10 or so segments; proboscis when everted showing numerous, irregularly arranged, short papillae on all except possibly short prox- imal part near mouth, with distal opening surrounded by 12 rounded, soft papillae attached at anterior end of muscular part. Proboscis missing on several specimens; papillate part broken away from mouth on some others (Fig. 2B). Many specimens with polygonal eggs in coelomic cavity (Fig. 2A). 562 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON »~ a . iM 0.01mm Fig. 3. Eumida (Pirakia) hutchinsonensis: A, Posterior end, dorsal view; B—D, Setae, various views (A, USNM 80516; B—D, FSBC I 30419). Remarks.— Ventral tentacular cirri of segment 2 on most type-specimens of Eumida (Pirakia) hutchinsonensis are slightly flattened and oval in cross section. They are definitely flattened on a few type-specimens, but on another specimen (USNM 80521) collected north of Boca Raton, Florida on Phragmatopoma, they are cylindrical in cross section. The specimen is larger than the type-specimens, 17 mm long with about 100 segments, but I am certain the specimen is the same species. The structure of the proboscis is the same. The origin of the median antenna is the same, although it is relatively longer than on the smaller type- specimens. The shape of the setae is identical, although there are about 15 per parapodium on middle segments rather than 10 or less as on the type-specimens. There are no notopodial lobes on segment 2 of the larger specimen, and only two setae were found on one side of that segment, then only after removal of the parapodium. Parapodial lobes are similar, but dorsal cirri are more pointed on the larger specimen. Finally, the larger specimen is a female whose eggs are iden- tical with those of the smaller specimens. Eumida (Pirakia) hutchinsonensis differs from E. (P.) punctifera (Grube, 1860) (Phyllodoce (Eulalia) punctifera.— Grube, 1861:142, pl. 3, fig. 5; Eumida (Pirakia) punctifera.—Hartmann-Schréder, 1971:113, 114, Fig. 36A—C) in having a pro- stomium which is much longer than wide, in having a median antenna which originates from the middle of the prostomium rather than from between the eyes, and in having parapodia with only slightly bilobed presetal lobes. Eumida (P.) hutchinsonensis differs from E..(P.) fuscescens (Saint-Joseph, 1888) (Eulalia fus- cescens Saint-Joseph, 1888:296, 297, pl. 12, figs. 163-165; Eumida (Pirakia) fucescens.—Hartmann-Schréder, 1971:113) in having subulate rather than fili- form antennae and tentacular cirri and in having none to very few, rather than several, setae on parapodia of the second segment. Eumida (P.) hutchinsonensis differs from E. (P.) lanceolata (Hartman and Fau- chald, 1971:44—46, pl. 7) in having eyes and broader dorsal cirri and in lacking setal lobes on segment 2. Eumida (P.) hutchinsonensis differs from E. (P.) brunnea (Fauchald, 1972:53, 54, pl. 4, figs. c, d) in having eyes, numerous proboscidal papillae and broader dorsal cirri, and in lacking filiform lobes on tips of parapodia. VOLUME 97, NUMBER 3 563 Eumida (P.) hutchinsonensis may have been previously reported from Florida by Rullier (1974:23, 24) as Eulalia punctifera, but I cannot confirm this. Etymology.—The specific name refers to the type-locality. Genus Paranaitis Southern, 1914 Paranaitis gardineri, new species Fig. 4 Paranaitis polynoides.—Gardiner, 1976:110, fig. 6M—P [not Anaitis polynoides Moore, 1909]. Material examined.—NORTH CAROLINA: Cape Lookout, intertidal, sand mixed with gravel and shell fragments, S. L. Gardiner & C. J. Jenner, coll., 6 Apr 1974, holotype (USNM 52876), paratype (USNM 80523). Same, E. Powell, coll., 14 Feb 1975, paratype (USNM 52878). Intracoastal Waterway, Wrightsville Beach, intertidal, muddy sand, T. Fox and S. L. Gardiner, coll., 9 Mar 1974, 2 paratypes (USNM 52877). FLORIDA: Hutchinson Island Sta II, 27°21.6’N, 80°13.2'W, 11 m, coarse calcareous sand, 5 paratypes (FSBC I 30422-30424; USNM 80524; ZMH P-17600). Same, Sta IV, 27°30.7'N, 80°12.8’W, 11 m, coarse calcareous sand, 3 paratypes (AHF POLY 1391; BMNH 1983.938; FSBC I 30425). Same, Sta V, 27°22.9'N, 80°13.9'W, 11 m, coarse calcareous sand, 2 paratypes (FSBC I 30426, 30427). Description.—Specimens from North Carolina (collected intertidally) highly colored (Fig. 4A, C) with diffuse purple spots on prostomium, tentacular segments, tentacular cirri, dorsum and ventrum of posttentacular segments, and dorsal cirri; color spots often on cirrophores of dorsal cirri and on proximal parts of ventral cirri. Segmental spots on dorsum often coalesced, forming almost uniform trap- ezoidal groups with anterior side narrower; medial and pair of lateral irregular lines of pigment spots on ventrum; dorsal cirri with up to 10 diffuse pigment spots on medial half of cirrostyles. Florida specimens (collected subtidally) almost all without color pattern; trace of color pattern noted above present on one specimen. Maximum length 95 mm, width 4 mm (Gardiner 1976); maximum length of preserved specimens 60 mm, width 3 mm, 180 segments; specimens from North Carolina more than twice as long as Florida specimens. Prostomium (Fig. 4a) about as long as wide; posterior half surrounded on dorsal side by fused segments 1 and 2, with nuchal papilla. Tentacular cirri of segment 3 longest, extending to about setiger 8; dorsal pair of segment 2 often about as long as ones on segment 3. Dorsal cirri (Fig. 4C—F) on distinct cirrophores; cirrostyles imbricated, thin, about as broad as or shorter than length of aciculum, small anteriorly, gradually longer toward middle, smaller near posterior end, with margins gradually curved or with obscure dorsolateral angle, with dorsal margin slightly longer than ven- trolateral margin. Imbricated cirri exposing most of dorsum of body anteriorly, ¥3;—Y) of body on middle segments, '4—'3 of body on posterior segments. Setigerous lobes conical, with distally notched presetal lobes; upper parts of presetal lobes longer and broader than lower parts; postsetal lobes shorter, rounded. Ventral cirri extending about to tips of presetal lobes or slightly shorter, flattened, elongate- oval, with rounded tips, with dorsal margins usually concave, with ventral margins convex, Originating on short, broad cirrophores or extensions of ventral parts of parapodia. Up to 30 compound spinigers per parapodium depending on body 564 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0.2mm ad 6 \ SVD yg OTEH ged” Vad v avi \ \ Via N Do a ae te (Sen ae | H 0.015mm I 0.4mm Fig. 4. Paranaitis gardineri: A, Anterior end, dorsal view; B, Ventrolateral view of same; C, Parapodium, posterior view; D, Same, anterior view; E, Same, posterior view; F, Same, anterior view, small specimen; G, Pygidium, dorsal view; H, Hinge region of seta, anterior view; I, Same, posterior view (A-C, G, holotype; D, E, BMNH 1983.938; H, I, USNM 80524). size (Fig. 4D, F, H, I); blades flattened, serrated; shafts enlarged near tips, with large round-tipped tooth on anterior side, with row of numerous long spines below large tooth, and short, triangular tooth on same side of shaft as serrated margin of blade. Acicula gradually tapered to slender tips; tips often emergent from setal lobes, if not broken, in preserved specimens. Pygidium (Fig. 4G) on specimens VOLUME 97, NUMBER 3 565 of greater than 10 mm length with pair of long, slender, cirriform or tapered anal cirri; cirri 6—8 times longer than wide, about as long as ventral tentacular cirri of segment 2, but not as stout; often shorter, almost cylindrical on some small specimens. | Proboscis (Fig. 4A, B), observed everted on 2 specimens from North Carolina, dissected on specimen from Florida; dorsal surface with scattered, chitinous pa- pillae, with region of small papillae proximally, then short, smooth region more distally, then papillate region extending toward tip; papillae of distal region smaller proximally, gradually larger distally; 2 pairs of small, subulate, soft papillae lat- erally at proximal end; lateral and ventral surface proximally smooth, with about 6 indistinct, muscular ribs, changing to 6 rows of rectangular pads medially, then to 4 rows of rectangular pads distally; shape of anterior margin of proboscis when fully everted not satisfactorily observed. Remarks.—Paranaitis gardineri differs from P. polynoides (Moore, 1909), to which specimens of P. gardineri from North Carolina had been referred, in having dorsal cirri which are not as broad and in having anal cirri which are much longer and more slender. Dorsal cirri of P. polynoides are much broader than the lengths of acicula and conceal the parapodia and entire dorsum posteriorly, whereas on P. gardineri dorsal cirri are about as broad as the lengths of acicula and leave the middle '4—'4 of the body exposed posteriorly. Anal cirri of P. polynoides are stout, cylindrical, and about four times longer than wide, whereas anal cirri of P. gar- dineri are slender, gradually tapered, and 6—8 times longer than wide. The proboscis of the holotype of P. polynoides (Moore) is not everted. I dissected a specimen of P. polynoides from Puget Sound, Washington, identified by Dr. M. H. Pettibone (USNM 26839), and examined the proboscis. Although I could not exactly determine the details, the proboscis of the dissected specimen appeared similar to that of P. gardineri. Paranaitis gardineri is also similar to P. kosteriensis (Malmgren, 1865), but the latter species has spheroidal anal cirri (Bergstrém 1914). Etymology.—The species is named in honor of Dr. Stephen L. Gardiner, who was instrumental in collecting the excellent specimens of this and many other species from North Carolina. Family Hesionidae Sars, 1862 Genus Heteropodarke Hartmann-Schréder, 1962 Heteropodarke lyonsi, new species Figs. 5, 6 Material examined.—FLORIDA EAST COAST: Hutchinson Island Sta II, 27°21.6'N, 80°13.2’W, 11 m, coarse calcareous sand, holotype (USNM 80525), 2 paratypes (USNM 80530; FSBC I 30429), 2 specimens (USNM 80532; FSBC I 30430). Same, Sta III, 27°22.0’N, 80°12.4’W, 7 m, medium calcareous sand, 2 paratypes (BMNH 1983.939; ZMH P-17599), 3 specimens (FSBC I 30431-30433). Same, Sta IV, 27°20.7’N, 80°12.8’W, 11 m, coarse calcareous sand, 6 paratypes (USNM 80527, 85029; AHF Poly 1398; ZMH P-17598), 1 specimen (FSBC I 30434). GULF OF MEXICO (U.S. Bureau of Land Management, Mississippi, Alabama, Florida Study, 1975-76, box core samples): off Florida, Sta 28, 29°55’N, 86°05'W, 38 m, coarse calcareous sand-rubble, 2 paratypes (BMNH 1983.940- 566 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 941), 4 specimens (MESC). Same, Sta 29, 29°56’N, 86°06’W, 38 m, coarse cal- careous sand-rubble, 6 paratypes (MESC; BMNH 1983.942; ZMH P-17605; FSBC I 30435), 2 specimens (MESC). Same, Sta 30, 29°51'N, 86°06.5’W, 41 m, coarse calcareous sand-rubble, 6 paratypes (MESC; USNM 80526; AHF Poly 1399), 2 specimens (MESC). Same, Sta 31, 29°48'N, 86°09.5’W, 45 m, coarse calcareous sand-rubble, 1 paratype (USNM 80528), 1 specimen (MESC). Same, Sta 32, 29°46'N, 86°12.5'W, 45 m, coarse sand, 3 paratypes (FSBC I 30436, 30437), 3 specimens (MESC). Same, off Alabama, Sta 40, 29°43.5'N, 87°54.5’W, 36 m, coarse sand, | specimen (MESC). Same, Sta 41, 29°45.5'’N, 87°36.5'W, 37 m, medium sand, | paratype (USNM 80531). NORTH CAROLINA: off Beaufort, 34°34'N, 76°25'W, 19 m, fine sand and shell, J. H. Day, coll., BST 88X, 1 paratype (USNM 51078). Description.—Body without color pattern. Holotype complete, about 8 mm long, 75 setigers; largest paratype (USNM 80530) incomplete, in 2 pieces, about 20 mm long, about 200 setigers; remaining paratypes and specimens incomplete; single large pygidial fragment (USNM 80532). Body almost cylindrical, slightly flattened ventrally. Prostomium slightly longer than wide, rectangular, with broad- ly rounded corners on specimens with proboscis everted (Fig. 5A), shorter, pyr- iform on specimens with proboscis inverted (Fig. 5B), 2—*3 width of body without parapodia. Antennae and palps tapered, not articulated, perhaps pseudoarticulated on few partially dried specimens; lateral antennae attached on anterior margin at about middle of each side, about as long as prostomium; median antenna attached slightly back from anterior margin, slightly shorter than lateral antennae; palps attached on anterior margin on short palpophores at about lateral sides of pro- stomium, similar to lateral antennae. One or 2 pairs of minute eyespots usually present laterally on posterior 13 of prostomium. First segment reduced dorsally, visible only laterally and ventrally; tentacular cirri 4-6 pairs on 2-3 laterally visible, indistinctly marked segments, upper ones on long cirrophores; longest upper ones on segment 2, with about 25 articles, extending to about setiger 6; upper ones on segment | about *%3 as long; upper ones on posterior tentacular segment about 14 as long; lower ones all about same length, 4-4 length of longest upper ones, with about 10 articles. Parapodia (Figs. 5C, D; 6A—C) lateral on anterior segments, gradually changing to dorsolateral orientation on posterior segments, sesquiramous; each with articulated dorsal and ventral cirri, with blunt neuropodial lobe, with distinct presetal lobe on anterior 3 and middle and posterior setigers, with composite, heterogomph setae consisting of spinigers and falcigers, with blunt-tipped neuroaciculum, with tapered notoaciculum; internal structure of shafts of setae uniform, unmarked, similar to that of Syllidae. Dorsal cirri all similar, on moderately long cirrophores, about twice as long as neuropodial lobes, with up to 15 articles; ventral cirri s—' as long as dorsal cirri, on short cirrophores, with about half their length extending past neuropodial lobes; notoacicula solitary, short, slender on anterior segments, gradually longer, stouter in middle segments, very stout, emergent, and crossing one another middorsally on posterior segments. Spinigers (Fig. 6F, G) solitary, beginning on setiger 3, continuing to middle of body or beyond, or up to about setiger 80, with blades slightly longer and more strongly serrated on middle segments, aligned above aciculum on posterior side. Falcigers 4-7, almost always 5 per parapodium. Falcigers on setigers 1—3 (Fig. 6D, E) moderately stout; blades with rounded tips, without serrations on margins; VOLUME 97, NUMBER 3 567 Fig. 5. Heteropodarke lyonsi: A, Anterior end, dorsal view, small paratype; B, Same, large paratype; C, Parapodium, setiger 3, posterior view; D, Same from region of enlarged falcigers, anterior view; E, Posterior end of foregut, cirri omitted (A, B, USNM 80529; C, D, AHF Poly 1399; E, holotype). shafts with bifid tips, with one bifurcation strongly hooked laterally. Very stout golden brown falcigers beginning on setiger 4 (Fig. 6H) and extending for variable number of segments depending on size, maximally to setiger 37 on largest spec- imen, to setiger 12—37 on all specimens, with rounded or concave-edged, smooth, 568 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0.015mm 0.015mm 0.015mm Fig. 6. Heteropodarke lyonsi: A, Parapodium, about setiger 50, posterior view; B, Same, setiger 75, posterior view; C, Same, posterior parapodium, posterior view; D, Falciger, setiger 3; E, Shaft of falciger, setiger 3; F, Spiniger, setiger 4; G, Same, middle segment; H, Enlarged falcigers; I, Falciger, setiger 75; J, Subemergent falciger with sickle-shaped hood, setiger 75; K, L, Falcigers, middle segment, face view; M, N, Tips of shafts of falcigers, middle segment, face view; O, Upper falciger, posterior parapodium; P, Subacicular falciger of same; Q, Lower falciger of same; R, Neuroaciculum, setiger 75 (A, AHF Poly 1399; B, C, USNM 80530; D, E, FSBC I 30431; F, G, K—N, FSBC I 30430; H, FSBC I 30433; I, J, O-R, holotype). VOLUME 97, NUMBER 3 569 triangular blades; blades on setiger 4 twice as large as those on setiger 3, gradually larger to about setiger 10; shafts of very stout falcigers with broadly rounded tips, with diameter greater nearer tips than proximally. Falcigers of middle segments (Fig. 6I-N) with moderately stout blades; blades with hooded tips, with tips straight in lateral view but cupped in face view; tips of shafts slightly bifid, with similar bifurcations; hood of emerging setal blades long, pointed (Fig. 6J). Falcigers of posterior segments (Fig. 60—-Q) with blades more strongly hooked than on middle segments and with smaller, blunt tips or hooded areas, with shafts bifid but with one bifurcation strongly nobbed or hooked as on shafts of anterior-most falcigers. Neuroacicula solitary, very stout on parapodia with very stout falcigers, slightly constricted near tips and nobbed (Fig. 5D); remaining ones more slender with similar tips (Fig. 6R). Foregut with 10 soft papillae surrounding anterior margin when everted, ex- tending posteriorly to a few segments past segments having very stout falcigers, maximally to about setiger 40, ringed with thick muscle bands (Fig. SE). Remarks. — Heteropodarke lyonsi is similar to H. heteromorpha Hartmann- Schréder, 1962, the type-species, and H. heteromorpha africana Hartmann-Schro- der, 1974, in having a variable number of tentacular cirri, single spinigers on some but not all parapodia, and greatly enlarged falcigers on anterior parapodia. Heteropodarke lyonsi and H. heteromorpha have up to six pairs, whereas H. heteromorpha africana has up to eight pairs of tentacular cirri. Heteropodarke heteromorpha and H. heteromorpha africana have spinigers distributed approx- imately in the region of enlarged falcigers (Hartmann-Schréder 1974; Dorsey 1978), whereas such setae on H. lyonsi have a much more extensive distribution, from setiger 3 to about the middle of the body. Heteropodarke lyonsi differs from both H. heteromorpha and H. heteromorpha africana in having enlarged, emer- gent notoacicula on about the posterior 5—% of the body. Heteropodarke heter- omorpha is known from Peru, New Caldonia, and California; H. Heteromorpha africana is known from Natal; and H. lyonsi is known from North Carolina, the east coast of Florida, and the northeastern Gulf of Mexico. Etymology.—The species is named in honor of William G. Lyons, Supervisor - of Invertebrate Research of the Florida Department of Natural Resources Bureau of Marine Research. Heteropodarke formalis, new species Fig. 7 Material examined.—FLORIDA EAST COAST: Hutchinson Island Sta II, 27°21.6'N, 80°13.2’W, 11 m, coarse calcareous sand, holotype (USNM 80533), 81 paratypes (USNM 80536, 80537; AHF Poly 1396; BMNH 1983.943-951; FSBCI 30440, 30441; MESC; ZMC; ZMH P-17606, P-17607), 5 specimens (FSBC I 30438, 30439). Same, Sta IV, 27°20.7’N, 80°12.8’W, 11 m, coarse calcareous sand, 35 paratypes, (AHF Poly 1397; BMNH 1983.952-956; FSBC I 30443- 30445; USNM 80535; ZMC), 4 specimens (FSBC I 30442). Same, Sta V, 27°22.9’N, 80°13.9’W, 11 m, coarse calcareous sand, 7 paratypes (FSBC I 30448), 10 spec- imens (FSBC I 30446, 30447). GULF OF MEXICO (U.S. Bureau of Land Man- agement Mississippi, Alabama, Florida Study, 1975-76, box core samples): Off 570 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Florida, Sta 28, 29°55'N, 86°05’W, 38 m, coarse calcareous sand-rubble, 1 para- type (USNM 80538). Same, Sta 29, 29°56'N, 86°06’W, 38 m, coarse sand-rubble, 1 specimen (MESC). Same, Sta 30, 29°51'N, 86°06.5'W, 41 m, coarse sand-rubble, 2 paratypes (MESC; FSBC I 30449), 2 specimens (MESC). Same, off Alabama, Sta 41, 29°45.5'N, 87°36.5'W, 37 m, medium sand, 1 paratype (MESC). NORTH CAROLINA: off Beaufort, 34°34’N, 76°25'W, 19 m, fine sand and shell, J. H. Day, coll., BST 88X, 1 paratype (USNM 80534). Description. — Body without color pattern; maximum length about 9 mm, about 0.3 mm wide excluding parapodia, about 0.8 mm wide including parapodia and setae, about 75 setigers; body almost cylindrical, rounded dorsally, flattened ven- trally. Prostomium (Fig. 7A) indistinctly bilobed, slightly wider than long, almost straight anteriorly and posteriorly, rounded laterally, almost as wide as body excluding parapodia. Antennae and palps subulate or tapered, irregularly wrinkled, but not articulated; lateral antennae attached near anterolateral margins of pro- stomium, about as long as prostomial width; median antenna attached on anterior margin, perhaps half as long and much more slender than lateral antennae. Palps attached on short palpophores ventrolateral to lateral antennae, about on line with lateral margins of prostomium, similar and perhaps slightly longer than lateral antennae. Two pairs of small lensed eyes usually present in trapeziform arrange- ment open anteriorly, near lateral margins of prostomium slightly posterior to midtransverse line. Tentacular cirri articulated, 6 pairs on 3 laterally visible seg- ments, with first not visible dorsally and second and third indistinctly separated from each other dorsally; upper ones all similar, slightly longer than segmental width excluding parapodia, with about 13 articles, attached on moderately long cirrophores; lower ones about half as long, on short cirrophores. Pair of transverse segmental ciliary bands visible dorsally anterior and posterior to parapodia. Para- podia (Fig. 7B, C) similar throughout body, sesquiramous; each with articulated dorsal and ventral cirri, with slender blunt-tipped neuropodium, with papilliform presetal and rounded postsetal lobes, with heterogomph, compound setae con- sisting of spinigers and falcigers, with blunt-tipped neuroaciculum, with slender, tapered notoaciculum extending into base of dorsal cirrophore; internal structure of shafts of setae uniform, unmarked. Dorsal cirri about as long as body width excluding parapodia, with 10—15 articles, attached on moderately long cirrophores; ventral cirri '2—'’3 as long as dorsal cirri, with about 5 articles, attached directly to about middle of lower side of parapodia. Spinigers (Fig. 7E) on all parapodia, usually solitary, rarely 2, aligned above neuroaciculum, with shafts much more slender but similar to those of falcigers. Falcigers (Fig. 7F—I) 6-8 per parapodium, usually 4 above and 2 below neuroaciculum, slightly stouter on setigers 4 to 12 or 14 (Fig. 7F), but similar to others; blades short, unidentate, with short, triangular serrations; shafts with tips almost equally bifurcate. Neuroacicula of setigers 4— 14 slightly stouter than others. Pygidium (Fig. 7D) with pair of anal cirri similar to dorsal cirri. Foregut not divided into anterior and posterior parts by constriction, extending to about setiger 14, ringed with thick muscle bands, with 10 soft papillae sur- rounding anterior margin when everted. Remarks. — Heteropodarke formalis differs from other members of the genus in having a constant six pairs of tentacular cirri, in having falcigers only slightly enlarged on some anterior segments, and in having spinigers on all parapodia. VOLUME 97, NUMBER 3 571 0.015mm E Fig. 7. Heteropodarke formalis: A, Anterior end, dorsal view; B, Parapodium, setiger 7, posterior view; C, Same, middle segment, anterior view; D, Posterior end, dorsal view; E, Spiniger, middle segment, blade turned; F, Falciger, setiger 7; G, Same, middle segment; H, Same, posterior segment; I, Tip of shaft of falciger, face view (A, C, D, holotype; E, I, FSBC I 30444; F-H, USNM 80536). Sy/2. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Etymology.—The specific name is derived from the Latin and refers to the lack of variation in the number of tentacular cirri. Genus Kefersteinia Quatrefages, 1865 Kefersteinia haploseta, new species Fig. 8 Material examined.—FLORIDA: Hutchinson Island Sta II, 27°21.6'N, 80°13.2’W, 11 m, coarse calcareous sand, Applied Biology, Inc., coll., holotype (USNM 80513), paratype (USNM 80514). Description.— Body without color pattern. Complete paratype about 4.5 mm long, 0.22 mm wide excluding parapodia, 0.64 mm wide including parapodia at setiger 5, 30 setigers; holotype incomplete posteriorly, 29 setigers, about as wide as paratype. Body slightly narrower posteriorly for first 9-10 setigers, widest in middle (from reproductive products), tapered posteriorly. Prostomium (Fig. 8A, B) wider than long, slightly convex anteriorly, anterolaterally straight, rounded posterolaterally on paratype, slightly incised posteriorly on holotype. Antennae originating dorsally near anterior margin at about middle of each side, slender, gradually tapered, about as long as prostomial width; palps originating slightly ventrolateral to antennae; palpophores about twice diameter of antennae, less than half as long; palpostyles oval, slightly longer than palpophores; two pairs of large, lensed eyes on posterior half of prostomium near lateral margins, with anterior and lateral pair larger; nuchal organs behind posterolateral margins of prostomium possible. First tentacular segment reduced, visible only laterally and ventrally (Fig. 8A); three indistinctly marked tentacular segments visible dorsally, with first dorsally visible segment half as long as following one (proboscis everted in both specimens). Tentacular cirri 8 pairs; upper ones all similar, longer than body width including parapodia; lower ones less than half as long; all articulated distally like dorsal cirri, with proximal articles 4 times longer than broad and smooth, with distal articles slightly longer and pseudoarticulated; diameter of dorsal ones 1'4 times that of dorsal cirri; diameter of ventral ones slightly less than that of dorsal ones. Parapodia (Fig. 8C) sesquiramous, all similar, shorter on anterior few setigers and on posterior end; each with distinctly articulated dorsal cirrus, with smooth ventral cirrus, with notopodial lobe reduced to no- toaciculum extending into anterior part of base of dorsal cirrophore, with slender, conical neuropodial lobe, without notosetae, with neurosetae consisting of com- pound falcigers and slender spines. Dorsal cirri slightly longer than parapodia, on moderately long cirrophores; proximal articles 3—4 times longer than broad, smooth; distal articles longer, indistinctly divided into 4 pseudoarticles. Ventral cirri orig- inating on ventral margins of neuropodia near neurosetae, slender, extending ventrolaterally almost to tips of neuropodia. Neurosetae consisting of long to short-bladed falcigers on first 5 setigers, falcigers and slender spines thereafter (Fig. 8D-G). Falcigers with slender blades; blades shorter above and below, grad- ually longer near middle, with serrated margins and unidentate, hooded tips; shafts slender, internally cross-barred, with bifid tips. Simple spines solitary, slender, smooth, round-tipped, with embedded part aligned above neuroaciculum on an- terior side and separated from fan of falcigers, internally cross-barred; neuroacicula numbering 1-2 in middle parapodia, stouter than notoacicula, pointed, ending VOLUME 97, NUMBER 3 > ‘ 0.4mm Sas = a ey ys » CZ ae E> O we O 0.2mm ee cary eae uae SS -—~ ae oo) In EON pe ee pv Cr i N TALE WLLL y I Wt ESS \ | | y ai See Fig. 8. Kefersteinia haploseta: A, Anterior end, dorsal view; B, Prostomium and everted pharynx; C, Parapodium, middle segment, anterior view; D, Compound seta from middle of bundle, middle segment; E, Upper compound seta of same; F, Tip of shaft of compound seta of same; G, Simple seta of same (A, holotype; B—G, paratype). 0.2mm near tips of presetal lobes, internally cross-barred. Pygidium with pair of long anal cirri similar but slightly stouter than upper tentacular cirri. Everted proboscis (Fig. 8A, B) with margin surrounded by circle of about 20 papillae on paratype, fewer papillae on holotype; thickly muscled with circular muscle bands, extending to setiger 4 or 5; distal and proximal parts not separated by distinct groove or constriction. 574 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Both specimens sexually mature, but sex undetermined, with reproductive prod- ucts beginning in setiger 9 or 10 and extending for 10 segments in paratype, to posterior end of incomplete holotype. Remarks.—Hesionid genera diagnosed by Fauchald (1977:73-77) as having two antennae, eight pairs of tentacular cirri, sesquiramous parapodia and an unarmored pharynx are as follows: Hesione Savigny, 1818; Kefersteinia Quatre- fages, 1865; Dalhousiella McIntosh, 1901; and Wesenbergia Hartman, 1955. He- sione has minute, conical antennae and lacks palps. Wesenbergia is similar to Hesione, but has a pair of minute, simple palps about the same size and shape as antennae (Hesionella problematica Wesenberg-Lund, 1950:14, pl. 3, fig. 15a). Kefersteinia and Dalhousiella have longer, subulate antennae and biarticulate palps. Dalhousiella, Hesioneand Wesenbergia all have an eversible pharynx which is distally smooth; only Kefersteinia has an eversible pharynx surrounded distally by a circlet of papillae. The type-species of Kefersteinia is Psamathe cirrata Keferstein, 1862. Descrip- tions of this species available to me [Psamathe cirrata.—Claparéde, 1863:55, 56, pl. 14, figs. 1-7; 1868:537; Kefersteinia cirrata. —Fauvel, 1923:238-240, Fig. 89a— e, in part (parts comprised of citations of Kefersteinia cirrata of Saint-Joseph, 1888, Castalia fusca var. of Southern, 1914, and Kefersteinia cirrata of Fauvel, 1913, are probably examples of Hesiospina Imajima and Hartman, 1964); Hart- mann-Schréder, 1971:131, fig. 42] show that K. cirrata has the first three segments reduced dorsally and visible only laterally and ventrally. Thus, setae are first present on the second dorsally visible segment rather than on the fourth such segment as diagnosed by Fauchald (1977:76). In contrast to K. cirrata, Kefersteinia fauveli Averincev (1972: 146, 147, pl. 20 figs. 1-4; not K. fauveli of Hartman, 1978:145-148, fig. 12), and K. haploseta have only the first segment dorsally reduced, and Kefersteinia sp., described briefly below, has the first two segments dorsally reduced. Although dorsal reduction of anterior segments is usually con- sidered to be an important generic character, the above may represent a continuum of closely related species, and I prefer not to erect another genus without additional evidence. Further, the distal margin of the pharynx of the type-species is surrounded by a circlet of fimbriae (Fauchald 1977:76), here defined as slender filiform papillae. Kefersteinia haploseta and Kefersteinia sp. have a circlet of at most about 20 relatively stouter marginal papillae. Kefersteinia fauveli has a circlet of 11-12 small marginal papillae and a pair of dorsolateral bunches of 4—5 papillae back from the margin. It is my opinion that differences in number and size of marginal papillae from those of the type-species should not preclude assignment of species to Kefersteinia. However, the bunches of papillae back from the margin of the pharynx of K. fauveli are apparently unique among Hesionidae, and a re-evalu- ation of the importance of this character may be required. Simple neurosetae occurring on most parapodia of Kefersteinia haploseta are absent on K. cirrata, K. fauveli, and Kefersteinia sp., below. Similarly, simple neurosetae are present on Kefersteinia similis Hessle, 1925, the type-species of Hesiospina \majima and Hartman, 1964, but Fauchald (1977:76) did not consider such setae to be important enough to include in the diagnosis he gave for that genus. Likewise, I do not consider them to be of generic importance in Kefersteinia. However, the position and alignment of these simple setae suggest that they may VOLUME 97, NUMBER 3 575 be homologous with acicula, and a re-evaluation of the importance of this char- acter also may be required. Etymology.—The specific name is formed of a combination of the Greek haplos, meaning simple, and the Latin seta, meaning bristle, and refers to the simple neurosetae of this species. Kefersteinia sp. Fig. 9 Material examined.—FLORIDA: Hutchinson Island Sta V, 27°22.9'N, 80°13.9'W, 11 m, coarse calcareous sand, 1 specimen (FSBC I 30450). Gulf of Mexico off S.W. Florida, 25°45.7'N. 83°11.1’W, 54 m, coarse calcareous sand, Mote Marine Laboratory personnel, coll., 3 specimens (USNM 80512; MML). Diagnosis.—Complete specimen 6 mm long, 1.1 mm wide, 40 segments. Pro- stomium (Fig. 9A) longer than wide, with 4 eyes; all antennae lost; nuchal organs present. All tentacular cirri lost; first 2 segments reduced dorsally. Parapodia (Fig. 9B) similar to those of K. cirrata (Keferstein); dorsal cirri indistinctly articulated proximally, medially with long distinct articles formed of four indistinct pseu- doarticles, distally distinctly articulated. Setae all compound falcigers (Fig. 9C— F), all hooded, almost all with bidentate blades, few ventral ones with unidentate blades; shafts with bifid tips. Remarks.—This species differs from K. cirrata and K. haploseta in having falcigers with bidentate blades and from K. fauveli in lacking dorsolateral bunches of papillae back from the margin of the pharynx. It is not named because material is inadequate. Genus Podarkeopsis Laubier, 1961 Podarkeopsis levifuscina, new species Fige lO Gyptis vittata.—Taylor, 1971:155-159.— Day, 1973:25.—Hall and Saloman, 1975: 11 [list] [not Webster and Benedict, 1887]. Gyptis brevipalpa.— Gardiner, 1976:119, 120, figs. 8q—t, 9a [pot Oxydromus brey- ipalpa Hartmann-Schréder, 1959]. Material examined.—NORTH CAROLINA: Cape Lookout, intertidal, sand mixed with gravel and shell fragments, S. L. Gardiner, coll., 10 Nov 1973, holotype (USNM 52897). Banks Channel, Wrightsville Beach, intertidal, in burrow of Glycera robusta, Gardiner, coll., Feb 1972, 1 paratype (USNM 52895). Intra- coastal Waterway, Wrightsville Beach, intertidal, in burrow of Leptosynapta ten- uis, T. Fox, coll., Mar 1973, 1 paratype (USNM 52896). FLORIDA EAST COAST: Hutchinson Island Sta II, 27°21.6'N, 80°13.2’W, 11 m, coarse calcareous sand, 1 paratype (ZMH P-17601). Same, Sta III, 27°22.0'N, 80°12.4’W, 7 m, medium calcareous sand, | paratype (FSBC I 30451). Same, Sta IV, 27°20.7'N, 80°12.8'W, 11 m, coarse calcareous sand, 1 paratype (AHF Poly 1392). Same, Sta V, 27°22.9'N, 80°13.9'W, 11 m, coarse calcareous sand, 9 paratypes (BMNH 1983.957; FSBC 1 30452-30454: USNM 55586, 55587; ZMC). GULF OF MEXICO: Florida, U.S. National Marine Fisheries Service Tampa Bay Area Study, 1963-64, C. H. Sal- oman and J. L. Taylor, coll., exact locality unknown, 9 paratypes (USNM 45531). 576 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON AY 0.2mm B a Selle Fig. 9. Kefersteinia sp.: A, Anterior end, dorsal view; B, Parapodium, middle segment, posterior view; C, Upper seta, middle segment; D, Lower seta of upper group, middle segment; E, Upper seta of lower group, middle segment; F, Lower seta, middle segment (A, B, USNM; C-E, FSBC I). Same, Hillsborough Bay, 5 paratypes (AHF Poly 1393; BMNH 1983.958—959). Same, Upper Tampa Bay, 10 paratypes (AHF Poly 1395; FSBC I 12054; ZMH P-17603). Same, Old Tampa Bay, 22 paratypes (AHF Poly 1394; BMNH 1983.960- 963; FSBC I 10375; ZMC; ZMH P-17604). Same, Boca Ciega Bay, | paratype (ZMC). Anclote Anchorage, Tarpon Springs, Pinellas County, 28°12.6'N, 82°47.6'W, 3.5 m, 4 paratypes (FSBC I 17397). Description.— Usually without color pattern; body of one paratype (USNM 52895) tinged with scattered brown pigment spots dorsally, more prominent lat- erally above parapodia. Maximum length greater than 18 mm, width 2 mm including parapodia, about 60 segments; body broadest anteriorly, gradually ta- pered posteriorly, almost cylindrical but flattened ventrally. Prostomium (Fig. 10A, B) slightly wider than long, concave anteriorly, rounded laterally, covered posteriorly by fold of second segment, 74-34 as wide as body excluding parapodia. VOLUME 97, NUMBER 3 577 Median antenna originating just below anterior margin, short, 14 as long as lateral antennae, about half as broad, fusiform, broader proximally and gradually tapered or acuminate distally; lateral antennae originating below anterior margin about ¥; distance from middle to lateral margins, about twice as thick as median antenna at base, about half as thick as palps, gradually tapered, about 74 as long as pro- stomium and equal in length to palps; antennae perhaps irregularly wrinkled, but not articulated. Palps attached below lateral antennae, each with broad palpophore 7/; of total length and narrow, rounded palpostyle. Two pairs of moderately large, lensed eyes near lateral margins on posterior half of prostomium, in trapezoidal arrangement open anteriorly, anterior ones about twice as large as posterior ones. Ciliated nuchal organs present laterally between prostomium and tentacular seg- ments. First tentacular segment (Fig. 1OA, B) reduced dorsally, visible only lat- erally and ventrally, 3 indistinctly marked tentacular segments visible dorsally; tentacular cirri 8 pairs, distinctly to very indistinctly articulated, with articles shorter than broad; dorsal tentacular cirri longest on segment 2, extending to setiger 8-11, about half as long on segment 1, *4—*4 as long on segment 4, those on segment 3 about as long as on segment 1; ventral tentacular cirri on segment 1 about half as long as dorsal ones on same, about same length on segment 2, perhaps slightly longer on segment 4, about *4 as long as ventral cirri of segment 1 on segment 3; upper tentacular cirrophores long, perhaps 7% length of anterior parapodia, each housing tips of 2-3 slender acicula on larger specimens, 1-2 on smaller ones. Lower tentacular cirrophores shorter, shortest on segment 3, slightly longer on segments 1, 2 and 4; acicula not observed except in segment 1. Parapodia sesquiramous on setigers 1—4, biramous and similar in all remaining setigers (Fig. 10C), each with dorsal and ventral cirri, with small rounded notopodial lobes anteroventrally on tips of dorsal cirrophores beginning on setiger 5, with conical neuropodial lobes, with triangular presetal and rounded postsetal lobes, with notosetae consisting of simple spines and forked setae (Fig. 10D-—H), with neu- rosetae consisting of compound falcigers (Fig. 10I—K), with slender, pointed no- toacicula extending into dorsal cirrophores on setigers 1—4 and into bases of notopodial lobes beginning on setiger 5, and with slender, pointed neuroacicula extending into neuropodial lobes. Dorsal cirri smooth to indistinctly articulated, on moderately long cirrophores; relative lengths of dorsal cirri of anterior setigers, often obscured by regeneration or loss, as follows: cirri of setiger 1 about '4 as long as upper tentacular cirri of segment 1; cirri of setiger 2 abruptly shorter than those of setiger 1; following cirri gradually longer to setiger 4; cirri of setiger 4 slightly shorter than cirri of setiger 1; cirri of setiger 5 slightly shorter than cirri of setiger 4; following cirri gradually longer to setiger 8; cirri of setiger 8 similar in length to cirri of setiger 4; cirri of setiger 9 slightly shorter than cirri of setiger 8 on some specimens; remaining cirri more or less similar to those of setiger 8, slightly exceeding or exceeding tips of neuropodial lobes by up to almost '2 their length on middle segments. Ventral cirri smooth, originating at about midlength of neuropodial lobes near posterior side, usually directed ventrally, not exceeding tips of neuropodial lobes on anterior and middle segments, similar in length to those before but exceeding tips of neuropodial lobes on posterior segments. No- topodial spines (Fig. 10G, H) numbering 1-3, extending from upper part of no- topodial lobe, moderately stout to slender, stiff, smooth with rounded tips; forked setae (Fig. 1OD-—F) below spines, numbering 4—8, slender, each with bifid-long 578 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON C £ a c fo) 5 K (\ J Eee Mesa D 0.015mm Fig. 10. Podarkeopsis levifuscina: A, Anterior end, dorsal view; B, Same, ventral view, proboscis inverted; C, Parapodium, middle segment; D, Notopodial forked seta, middle segment, slightly turned; E, F, Same, in lateral view; G, Stout simple notopodial spine, middle segment; H, Slender spine of same; I, Compound neuroseta, upper group, middle segment; J, K, Same, lower group (A, I, J, USNM 52895; B, holotype; C, F, K, FSBC I 12054; D, E, G, H, BMNH 1983.957). VOLUME 97, NUMBER 3 579 tine, with pointed short tine, without serrations on shaft proximal to tip of short tine. Neurosetae consisting of numerous compound heterogomph falcigers (Fig. 10I—K), separated into upper and lower groups by acicula; blades moderately short above and below, long in middle, longest in lower part of upper group, with hooded, hooked, unidentate tips and finely serrated borders; shafts long, smooth with notched tips; hoods on few falcigers long, pointed, completely covering and exceeding tips of blades. Notoacicula in parapodia of setigers 1-4 numbering 2— 3 on large specimens, 1—2 on small specimens, single thereafter; neuroacicula numbering 2 per parapodium in middle segments. Foregut moderately long, extending through setigers 1—10 in holotype (inverted), through segments 1-6 in small specimens, to setiger 7 or 8 in large specimens with proboscis completely everted (USNM 52895), ringed with thick muscle bands, divided into anterior and posterior parts when everted by more or less distinct transverse groove or constriction, with anterior border when everted surrounded by 10 moderately long, conical soft papillae. Remarks. —In defining the monotypic genus Podarkeopsis, Laubier emphasized the importance of notopodial forked setae as a generic character on the type- species, P. galangaui Laubier, 1961. Presence of such setae has not been consid- ered an important generic character in the Hesionidae, but, in conjunction with other characters, this character appears to establish the close phylogenetic rela- tionship of a widely distributed group of species and tends to support the cor- rectness of Laubier’s arguments regarding its importance. Previously described species herein referred to Podarkeopsis Laubier are Ox- ydromus brevipalpa Hartmann-Schréder (1959:105-197; Gyptis brevipalpa. —Banse and Hobson, 1968:12, 13, in part, fig. 3B-E), Oxydromus arenicolus glabrus Hartman (1961:68, 69; Gyptis brevipalpa.—Banse and Hobson, 1968:12, 13, in part, fig. 3F—H), Oxydromus capensis Day (1963:397, fig. 4E-J; Gyptis capensis. — Day 1967:231, 232, fig. 11.2. L-O.— Gibbs and Probert, 1973:397, 398, fig. 2A;?in part), and Gyptis maraunibinae Gibbs (1971:137, 138, fig. 5). Banse and Hobson (1968) examined types of Podarkeopsis glabrus (Hartman), new combination, and referred the species to P. brevipalpa (Hartmann-Schré6der). Forked setae of the two species, which were first described by Banse and Hobson for P. glabrus and by Hartmann-Schréder in Banse and Hobson (1968) for P. brevipalpa, are distinctly different, and in my opinion the two species are not the same. Gibbs and Probert (1973) examined the holotype of Podarkeopsis capensis (Day). The forked setae they described resemble those of P. galangaui Laubier, and thus the two species may be the same. However, Laubier’s figure of the anterior end of P. galangaui (Laubier 1961: fig. 1A) and Day’s figure of the anterior end of P. capensis (Day 1967: fig. 11.2.N) suggest differences that may involve two species. Day’s figure shows the first two segments to be dorsally reduced, whereas Laubier’s figure shows only the first tentacular segment to be reduced dorsally. The two figures also suggest differences in relative lengths of tentacular cirri and dorsal cirri between P. galangaui and P. capensis. Therefore, I tentatively continue to consider them separate species. Podarkeopsis maraunibinae (Gibbs) apparently differs from P. brevipalpa in having enlarged papillae on the anterior end of the proboscis and smooth notopodial spines. Podarkeopsis levifuscina differs from other members of the genus in having 580 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON unique notopodial forked setae on which short prongs are smooth and long prongs are bifid. Notopodial spines of P. /evifuscina are similar to those of P. capensis, P. galangaui and P. maraunibinae in being smooth. Arrangement of tentacular and dorsal cirri appears similar to that of P. capensis. Etymology.—The specific name is derived from the Latin /evis, meaning smooth, and fuscina, meaning a three-pronged spear or trident, and refers to the forked setae. Literature Cited Averincev, V. G. 1972. Benthic polychaetes Errantia from the Antarctic and Subantarctic collected by the Soviet Antarctic Expedition.— Explorations of the Fauna of the Seas 11(19). Biological Results of the Soviet Antarctic Expedition 5:88—93, 41 pls. Banse, K. 1959. Uber die Polychaeten-Besiedlung einiger submariner Héhlen. Ergebnisse der Os- terreichischen Tyrrhenia-Expedition 1952, Teil XIJ.—Pubblicazioni della Stazione Zoologica di Napoli 30 (Suppl.):417-—469. 1973. The ventral parapodial cirrus of the benthic Phyllodocidae (Polychaeta), with special reference to Clavadoce Hartman and Bergstroemia Banse.—Journal of Natural History 7:683- 689. ,and K. D. Hobson. 1968. Benthic polychaetes from Puget Sound, Washington, with remarks on four other species. — Proceedings of the United States National Museum 125(3667):1—53. , and 1974. Benthic errantiate polychaetes of British Columbia and Washington. — Bulletin of the Fisheries Research Board of Canada, No. 185, 111 pp. Bergstrom, E. 1914. Zur Systematik der Polychaetenfamilie der Phyllodociden.—Zoologiska Bidrag fran Uppsala 3:37-224. Claparéde, E. 1863. Beobachtungen tiber Anatomie und Entwicklungsgeschichte Wirbelloser Thiere an der Kiiste von Normandie angestellt. Leipzig, vii + 120 pp, 18 pls. —. 1868. Les Annélides Chétopods du Golfe de Naples.— Mémoires de la Société de Physique et d’Histoire Naturelle de Genéve 19(2):313-584, Pls. 1-16. 1869. Les Annélides Chétopodes du Golfe de Naples. Seconde Partie. — Mémoires de la Société de Physique et d’Histoire Naturelle de Genéve 20(1):1-225, Pls. 17-31. Day, J. H. 1960. The polychaete fauna of South Africa. Part 5. Errant species dredged off Cape coasts.— Annals of the South African Museum 45(3):261-373. 1963. The polychaete fauna of South Africa. Part 8. New species and records from grab samples and dredgings.— Bulletin of the British Museum (Natural History), Zoology 10:381- 445. . 1967. A monograph on the Polychaeta of southern Africa. Part 1. Errantia. British Museum (Natural History) Publication No. 656, xxix + 458 pp. 1973. New Polychaeta from Beaufort, with a key to all species recorded from North Car- olina.— NOAA (National Oceanic and Atmospheric Administration) Technical Report NMFS (National Marine Fisheries Service) CIRC-375:140 pp. Dorsey, J. H. 1978. A first report of Heteropodarke heteromorpha Hartmann-Schréder, 1962 (Poly- chaeta: Hesionidae) from California. — Bulletin of the Southern California Academy of Science 77(2):82-87. Fauchald, K. 1972. Benthic polychaetous annelids from deep water off western Mexico and adjacent areas in the eastern Pacific Ocean.— Allan Hancock Monographs in Marine Biology Number 7, 573 pp. 1977. The polychaete worms. Definitions and keys to the orders, families and genera. — Natural History Museum of Los Angeles County Science Series 28:188 pp. Fauvel, P. 1913. Quatriéme note préliminaire sur les Polychétes provenant des campagnées de PHirondelle et de la Princesse-Alice.— Bulletin de 1’Institut Océanographique de Monaco No. 270, 80 pp. 1923. Polychétes Errantes.—Faune de France 5, 488 pp. Gallagher, R. M. 1977. Nearshore marine ecology at Hutchinson Island, Florida, 1971-1974. II. Sediments.— Florida Marine Research Publications No. 23:6-24. VOLUME 97, NUMBER 3 581 , and M. L. Hollinger. 1977. Nearshore marine ecology at Hutchinson Island, Florida, 1971- 1974. I. Introduction and rationale.— Florida Marine Research Publications No. 23:1-5. Gardiner, S.L. 1976. Errant polychaete annelids from North Carolina. — Journal of the Elisha Mitchell Scientific Society [Fall 1975] 91(3):77-—230. Gibbs, P. E. 1971. The polychaete fauna of the Soloman Islands.— Bulletin of the British Museum (Natural History) Zoology 21(5):101—211. , and K. Probert. 1973. Notes on Gyptis capensis and Sosane sulcata [Annelida: Polychaeta] from the benthos off the south coast of Cornwall.— Journal of the Marine Biological Association of the United Kingdom 53:397-401. Grube, A. E. 1861. Ein Ausflug nach Triest und dem Quarnero. Beitrage zur Kenntniss der Thierwelt dieses Gebietes. Berlin, Nicolaische Verlagsbuchhandlung, 175 pp, 5 pls. Hall, J. R., and C. H. Saloman. 1975. Distribution and abundance of macroinvertebrate species of six phyla in Tampa Bay, Florida, 1963-64 and 1969.— National Marine Fisheries Service Data Report No. 100:505 pp. Hartman, O. 1961. Polychaetous annelids from California.— Allan Hancock Pacific Expeditions 25: 1-226. . 1978. Polychaeta from the Weddell Sea quadrant, Antarctica. — Biology of the Antarctic Seas 6(4). Antarctic Research Series 26 (4):125-223 [published posthumously]. , and K. Fauchald. 1971. Deep-water benthic polychaetous annelids off New England to Bermuda and other North Atlantic areas. Part IJ.—Allan Hancock Monographs in Marine Biology, Number 6, 327 pp. Hartmann-Schréder, G. 1959. Zur Okologie der Polychaeten des Mangrove-Estero-Gebietes von El Salvador.—Beitrage zur Neotropischen Fauna 1(2):69-183. . 1962. Zweiter Beitrag von Polychaetenfauna von Peru. —Kieler Meeresforschungen 18:109- 147. —. 1971. Annelida, Borstenwitirmer, Polychaeta.— Die Tierwelt Deutschlands 58:1-—594. . 1974. Zur Polychaetenfauna von Natal (Siidafrika).— Mitteilungen aus dem Hamburgischen Zoologischen Museum und Institiit 71:35-73. Imajima, M., and O. Hartman. 1964. The polychaetous annelids of Japan.—Allan Hancock Foun- dation Occasional Paper 26:1—452. International Code of Zoological Nomenclature Adopted by the XV International Congress of Zoology. 1961. Second Edition, 1964. London, International Trust for Zoological Nomenclature, XIX + 176 pp. Laubier, L. 1961. Podarkeopsis galangaui, n.g., n.sp., hésionide des vases cotiéres de Banyuls-Sur- Mer.— Vie et Milieu 12:211-—217. Malmgren, A.J. 1865. Nordiska Hafs-Annulater.—Ofversigt af K. Vetenskapsakademiens Forhand- lingar, Stockholm 1865(1):51-110. Moore, J.P. 1909. The polychaetous annelids dredged by the USS Albatross off the coast of southern California in 1904. I. Syllidae, Sphaerodoridae, Hesionidae and Phyllodocidae. — Proceedings of the Academy of Natural Sciences of Philadelphia 61:321-351, pls. 15, 16. Perkins, T. H. 1979. Lumbrineridae, Arabellidae, and Dorvilleidae (Polychaeta), principally from Florida, with descriptions of six new species.— Proceedings of the Biological Society of Wash- ington 92(3):415-465. 1980. Review of species previously referred to Ceratonereis mirabilis, and descriptions of new species of Ceratonereis, Nephtys, and Goniada.— Proceedings of the Biological Society of Washington 93(1):1—49. 1981. Syllidae (Polychaeta), principally from Florida, with descriptions of a new genus and twenty-one new species.— Proceedings of the Biological Society of Washington 93(4):1080- 1172. 1984. Revision of Demonax Kinberg, Hypsicomus Grube and Notaulax Tauber, with a review of Megalomma Johansson from Florida (Polychaeta: Sabellidae).— Proceedings of the Biological Society of Washington 97(2):285-368. Pettibone, M. H. 1963. Marine Polychaete worms of the New England region. Part I. Families Aphroditidae through Trochochaetidae. — Bulletin, United States National Museum 227:1-356. Rullier, F. 1974. Quelques annélides polychétes du Cuba recueillies dans des €ponges.— Travaux du Museum d’Histoire Naturelle ‘“Grigore Antipa’ 13:9-77. 582 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Saint-Joseph, Baron A. de. 1888. Annélides Polychétes des cétes de Dinard. Seconde Partie.— Annales des Sciences Naturelles, Paris, Ser. 7, 5:141—338, pls. 6-12. Southern, R. 1914. Clare Island Survey. Archiannelida and Polychaeta.— Proceedings of the Royal Irish Academy 31(47):1-160, pls. 1-15. Taylor, J. L. 1971. Polychaetous annelids and benthic environments in Tampa Bay, Florida. Ph.D. Dissertation, University of Florida, Gainesville, Florida, 1332 pp. Uschakov, P. V. 1972. Polychaetes. Volume 1. Polychaetes of the suborder Phyllodociformia of the Polar Basin and the northwestern part of the Pacific (Families Phyllodocidae, Alciopidae, Tomopteridae, Typhloscolecidae, and Lacydonidae). Academy of Sciences of the USSR —Zoo- logical Institute. Fauna of the USSR. New Series 102, 259 pp. (In Russian, translated by the Israel Program for Scientific Translation, 1974). Webster, H. E., and J. E. Benedict. 1887. The Annelida Chaetopoda from Eastport, Maine.— United States Commission of Fish and Fisheries. Report of the Commissioner for 1885:705—758, pls. 1-8. Wesenberg-Lund, E. 1950. Polychaeta.—The Danish Ingolf-Expedition 4(14):1—92, pls. 1-10. Westheide, W. 1977. Phylogenetic systematics of the genus Microphthalmus (Hesionidae) together with a description of MM. hartmanae nov. sp. Pp. 103-114.—Jn D. J. Reish and K. Fauchald, eds., Essays on Polychaetous Annelids in memory of Dr. Olga Hartman, Allan Hancock Foun- dation, Los Angeles. Worth, D. F., and M. L. Hollinger. 1977. Nearshore marine ecology at Hutchinson Island Florida: 1971-1974. III. Physical and chemical environment.— Florida Marine Research Publications No. 23:25-85. Florida Department of Natural Resources, Bureau of Marine Research, 100 Eighth Ave. S.E., St. Petersburg, Florida 33701-5095. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 583-588 NEW SCYPHOZOAN RECORDS FOR HAWAITI: ANOMALORHIZA SHAWI LIGHT, 1921, AND THYSANOSTOMA LORIFERUM (Ehrenberg, 1835); WITH NOTES ON SEVERAL OTHER RHIZOSTOMES William J. Cooke Abstract. — Several adult specimens of the scyphomedusa Anomalorhiza shawi, unknown since the original collections in the Philippines during the 1920’s, have recently been collected in Hawaii. An adult specimen of the wide ranging oceanic species, Thysanostoma loriferum, previously collected from the Red Sea to the western Pacific, has also been collected in Hawaii. Anomalorhiza shawi is believed to be the latest example of several neritic scyphozoans introduced into Hawaii from the western Pacific. An earlier introduction, a Mastigias close to M. papua, was misidentified as Thysanostoma flagellatum in a recent handbook. Recent collections in Hawaii of two rhizostome medusae, Anomalorhiza shawi Light, 1921, and Thysanostoma loriferum (Ehrenberg, 1835), have increased the number of rhizostome species known from the central Pacific. This note records their collection, suggests the mode of introduction for A. shawi, and clarifies some previously published errors in the literature. Anomalorhiza shawi was described by S. F. Light (1921) on the basis of two complete specimens collected from Manila Bay, Philippine Islands. There are only two subsequent mentions of the species: Stiasny (1924) referred it to Lychnorhiza; Kramp (1961) maintained it in Anomalorhiza. Apparently there are no published collection records other than from the original location. The oceanic genus 7Thysanostoma was known from Hawaii on the basis of several Nineteenth Century collections of T. flagellatum (Haeckel, 1880). A recent T. loriferum collection established the presence of two species in the Hawaiian fauna. The medusa figured in Devaney and Eldredge (1977, figs. 8, 9) as “‘(?)Thysa- nostoma flagellatum’’ is actually a Mastigias close to M. papua (as figured by Hamner 1982). Anomalorhiza shawi Light, 1921 Fig. | Material examined.—One adult specimen collected 27 Apr 1983 by J. Brock from beach at northeast side of Coconut (Moku O Loe) Island, Kaneohe Bay, Oahu (21°26'18.5’N, 157°47'23”W). Accession 830004HA/1 Processing Center, Naval Ocean Systems Center-Hawaii Laboratory (NOSC). Two adult specimens collected 6 Jan 1984 by D. Krupp from lagoon on south- west side of Coconut Island (21°26'10”N, 157°47'34”W). Accession 830004HA/ 2 Processing Center, NOSC. Description.—The specimens range from 50 cm to 87 cm in diameter. The 584 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Anomalorhiza shawi canal system (traced from a color transparency) showing canal system in subumbrellar view and the gonads. Cent. Canal, centripedal canal between the radial canals. Rho- paliar radial canal is indistinguishable within the anastomosing marginal canal; inter-rhopaliar radial canal constricted at its inner end. exumbrella of the clear, fragile bell is marked by numerous low, warty bumps, each ringed with a light chocolate brown set against a general light purple ground color. The mouth arms are a deep purple while the gonads are a lighter pink- purple. No zooxanthellae were seen in examination of the bell and mouth arms. The single blind centripedal canal between each radial canal is especially obvious and clearly separates this species from all other daktyliophorans. The very com- plete original description is amended and extended only slightly as follows. Light’s (1921, fig. 2) rhopaliar radial canals were distinguishable beyond the ring canal, but in one specimen some are indistinguishable from the general anastomosing marginal canal (Fig. 1). The present specimens have several thick brownish filaments on the subumbrellar surface between the mouth arms where Light (1921) mentions, “‘small, very slender threadlike filaments.” The mouth arms of the present specimen are unbranched for about 74 (66% + 2%, n = 9; 95% limits) of their total length (excluding terminal filament), while Light’s smaller and larger specimens were unbranched for 71% and 82% respectively. (Most terminal filaments on the present specimens had lost their club-shaped distal part.) VOLUME 97, NUMBER 3 585 The cnidome of the species is unexceptional with ovoid “a” atrichs, 4-6 wm by 3-4 wm; small heterotrichous microbasic euryteles, 7-9 um by 6—7 wm; and almost spherical holotrichous isorhizas 10-11 wm by 9-10 ym (live, undischarged measurements). No large “A” atrichs (Weill 1934), polyspira atrichs (Spangenberg 1964), or “a’’ atrichs (Calder 1971) were seen. The stinging potential of this species seems low as several test subjects (including the author) experienced no pain or redness when live mouth-arm fragments were rubbed on the skin of the inner wrist. Thysanostoma loriferum (Ehrenberg, 1835) Fig. 2 Rhizostoma lorifera Ehrenberg, 1835:260. Himanostoma lorifera.— Haeckel, 1880:628, (additional bibliography see Kramp, 1961). Lorifera arabica Haeckel, 1880:628. Lorifera lorifera.— Mayer, 1910:694 (additional bibliography see Kramp, 1961). Thysanostoma lorifera.—Stiasny, 1940:24. Thysanostoma loriferum.—Kramp, 1961:364. Material examined.— One adult specimen collected 5 Nov 1982 by M. Morioka, Waikiki Aquarium from Kuuloa Beach Park (21°30'43”N, 157°50’16’W) Oahu, Hawaii. Accession 830010HA, Processing Center, NOSC. Description.— The specimen (Fig. 2) is 18 cm in diameter with a quite stiff, low- hemispherical bell of uniform pink-purple color. The ex-umbrellar surface is smooth. The broadly rounded marginal lappets are united by a thin common membrane. Each lappet pair is a darker purple than the purple of the exumbrella of the bell, from which it is delineated by a clear translucent circumferential band. The rhopalia and rhophaliar lappets are transparent and colorless. The circumferential bell muscles are well developed, clearly visible in sub- umbrellar view. The canal system is very finely meshed, with radial and perradial rhopaliar canals clearly distinguishable. The subumbrellar surface is smooth and translucent white. The subgenital ostia are approximately twice as broad as the mouth-arm pillars which are also trans- lucent white. The barely developed gonads in the present specimen are light pink. Distal parts of mouth arms were lost in present specimen. Their central stem is dark purple with fringes very dark (almost black) purple in life, fading to white in preserved material. This species has a cnidome of ovoid “a” atrichs 4-5 wm by 3-4 um, small heterotrichous microbasic euryteles, 6-8 wm by 5—6 wm and almost spherical holotrichous isorhizas, 8-10 um by 8-9 um (formalin preserved, undischarged measurements). Thysanostoma loriferum is differentiated from T. flagellatum (Haeckel, 1880) by having marginal lappets united into a membrane, and a short terminal filament on each mouth arm. The original description of T. flagellatum by Haeckel (1880) was quite short and had no figure. Stiasny (1940) provided clear diagnoses of both based on comparative material as well as detailing their color differences. Figure 2 clearly shows the lash-like weakly frilled nature of the mouth arms which 586 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Thysanostoma loriferum showing the bell margin and the proximal portions of the mouth- arms. Note the characteristic sparse fringing along the edges of the mouth-arms. (Original photo a color transparency courtesy Mr. L. Matsuura, Waikiki Aquarium.) distinguishes the Thysanostomatidae from all other rhizostomes. The endings -ferum and -latum agree (after Kramp 1961) in neuter gender with the generic name. Discussion Including the present species, there are now eight species of rhizostome medusae recorded from Hawaiian waters (Devaney and Eldredge 1977). Of these, Cephea cephea (Forskal, 1775), T. flagellatum (Haeckel, 1880) and T. loriferum (Ehren- berg, 1835) are oceanic species capable of reaching Hawaii as adult medusae. Thysanostoma flagellatum has been collected on only two occasions from Ha- waii (Haeckel 1880), but rather often (Kramp 1961) from the western Pacific (Philippines, Malaysian Archipelago). Thysanostoma loriferum had previously been recorded from the Red Sea to the western Pacific so its presence in Hawaii is not unexpected. It is probable that these oceanic species occur infrequently in Hawaii under favorable conditions but are more commonly seen in the island groups to the west and south. A different situation exists with the remaining rhizostome species whose mor- phology and habits suggest ecological restriction to plankton-rich, quiet neritic waters with little likelihood of transoceanic migration by adult medusae. Clearly, VOLUME 97, NUMBER 3 587 the two Cassiopea species, C. medusa Light, 1914 and C. mertensii Brandt, 1835, with their pseudobenthic habits are the most improbable adult immigrants. Scy- phistomal introduction from the Philippines has been suggested (Cutress in Doty 1961) for C. medusa and the species identified as Phyllorhiza punctata von Len- denfeld, 1884, in Devaney and Eldredge (1977, fig. 7). This medusa has also been identified as Cotylorhizoides pacificus by Cutress (in Doty 1961), as Phyllorhiza pacifica by Cutress (in Mansueti 1963), as Mastigias ocellatus by Walsh (1967), and M. papua by Reed (1971; fig. 14). It is quite clearly a mastigid, as is the species identified as “‘(?) Thysanostoma flagellatum (Haeckel, 1880)” in Devaney and Eldredge (1977, figs. 8, 9). Quite similar mastigids are well figured in Faulkner (1970 p. 20) and Hamner (1982:264, 276, 277). Reexamination of the numerous species in the Mastigiidae on a world-wide basis is clearly imperative, thus any attempt at more precise identifications of these two forms would be unwise. Clearly these mastigids (as well as C. mertensii) are also likely candidates for scyphistomal introduction as all are rather fragile-bodied species commonly found in lagoons and shallows. All of these species are now much less abundant than they were several years ago, but they are still believed to be present in the fauna. Anomalorhiza shawi, with its large distinctive medusa is also a recent intro- duction. Personal observations and discussions with long-time fishermen and others with several decades of experience in Kaneohe Bay confirmed that this medusa was not seen before 1983. The medusa 1s quite fragile, tearing and breaking under even gentle tension, and would be unable physically to withstand the forces of the open ocean. This fact and the collections and numerous other observations all in the sheltered waters of Kaneohe Bay reinforce the conclusion of a neritic or estuarine habitat for the species. In this case, introduction by scyphistome in fouling communities transported from the Philippines to Hawaii is the most probable route of introduction. The presence of these adult specimens suggests the potential for permanent establishment should they be able to breed success- fully. Acknowledgments I am indebted to J. Brock (Zoology Department, University of Hawaii), D. Krupp and P. Jokiel (Hawaii Institute of Marine Biology) who provided the specimens of A. shawi, and to M. Morioka and C. Hopper (Waikiki Aquarium) who provided the 7. Joriferum specimen to the Naval Ocean Systems Center- Hawaii Laboratory. This work was supported by NOSC contract N66001-83-C- 0025 to the Research Corporation University of Hawaii. Literature Cited Calder, D. 1971. Nematocysts of polyps of Aurelia, Chrysaora, and Cyanea, and their utility in identification. — Transactions American Microscopical Society 90:264—274. Devaney, D. M., and L. G. Eldredge. 1977. Class Scyphozoa.—Jn Devaney, D. M., and L. G. Eldredge, eds., Reef and shore fauna of Hawaii, Section 1.—Bernice P. Bishop Museum Special Publication 64(1):108-118. Doty, M.S. 1961. Acanthophora, a possible invader of the marine flora of Hawaii.— Pacific Science 15:547-552. Ehrenberg, C. G. 1835. Uber die Acalephen des Rothen Meeres und den Organimus der Medusen 588 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON der Ostsee.—Abhandlungen der Preussischen Akademie der Wissenschaften (1835):181—260. (not seen). Faulkner, D. 1970. The hidden sea. Viking Press, New York, ii + 149 pp. Haeckel, E. 1880. System der Acraspeden. Zweite Halfte der System der Meduseen; pp. 361-672. Jena. Hamner, W. M. 1982. Strange world of Palau’s salt lakes.— National Geographic 161:264—282. Kramp, P. L. 1961. Synopsis of the medusae of the world.—Journal of the Marine Biological Association of the United Kingdom 40:1-469. Light, S. F. 1921. Further notes on Philippine scyphomedusen jelly-fishes.— Philippine Journal of Science 18:25-—45. Mansuetti, R. 1963. Symbiotic behavior between small fishes and jellyfishes.—Copeia 1963:40-80. Reed, S. A. 1971. Some common coelenterates in Kaneohe Bay, Oahu, Hawaii.—Jn Lenhoff, H. M., L. Muscatine, L. V. Davis, eds. Experimental Coelenterate Biology. University of Hawaii Press, Honolulu. Spangenberg, D. B. 1964. New observations on Aurelia.—Transactions of the American Micro- scopical Society 83:448—455. Stiasny, G. 1924. Rhizostomen von Manila.—Zoologische Mededelingen 8:39—54. (not seen) 1940. Die Scyphomedusen.— Dana Reports 18:1—28. Walsh, G. W. 1967. An ecological study of a Hawaiian mangrove swamp.—Jn Lauft, G. H., ed. Estuaries. American Association Advancement of Science Publ. 83:420-431. Weill, R. 1934. Contribution a l’étude des cnidaires et de leur nematocysts.—Travaux Station Zoologie Wimereux 10:1—347;11:351-701. Processing Center, Naval Ocean Systems Center, Hawaii Laboratory, P.O. Box 997, Kailua, Hawaii 96734 and Research Associate, Hawaii Institute of Marine Biology, University of Hawaii. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 589-600 OSTRINCOLA AND PSEUDOMYICOLA (CRUSTACEA: COPEPODA: POECILOSTOMATOIDA) ASSOCIATED WITH MARINE BIVALVE MOLLUSKS ON THE PACIFIC COAST OF PANAMA Arthur G. Humes Abstract.—Two myicolid copepods occur in the mantle cavity of Bivalvia on the Pacific coast of Panama. Ostrincola falcatus, new species, is described from Protothaca asperrima, Mytella guyanensis, and Anadara obesa. The species is distinguished by the falciform seta on the third segment of the maxilliped of the male. A key to the species of Ostrincola is given. The widespread Pseudomyicola spinosus is recorded from a new host, Anadara obesa. The genus Ostrincola Wilson, 1944, contains several species, all of which live in the mantle cavity of bivalve mollusks in intertidal or shallow-water areas of tropical and subtemperate regions. More than 20 species of bivalves serve as hosts. Pseudomyicola spinosus (Raffaele and Monticelli, 1885) is known from the Mediterranean, the North and South Atlantic, and the Indo-Pacific, where it is associated with numerous bivalve hosts (Humes 1968). The copepods reported here were found in bivalves purchased in the market in Balboa. The localities on the Pacific side of Panama from which the bivalves came are known, however, and are indicated below. The copepods were studied in lactic acid and the figures were drawn with the aid of a camera lucida. The letter after the explanation of each figure refers to the scale at which it was drawn. The abbreviations used are: A, = first antenna, A, = second antenna, L = labrum, MX, = second maxilla, and P, = leg 1. Myicolidae Yamaguti, 1936 Ostrincola C.B. Wilson, 1944 Ostrincola falcatus, new species Figs. la-j, 2a—h, 3a—e, 4a—h Type-material.— 193 2° (including 75 ovigerous), 146 46, from 127 Protothaca asperrima (Sowerby), Chiman, Darien, Panama (Pacific side), approximately 100 km east of canal entrance, 5 Nov 1981. Holotype 2, allotype, and 331 paratypes (188 22, 143 46) deposited in the National Museum of Natural History, Smith- sonian Institution, Washington, D.C.; the remaining paratypes (dissected) in the collection of the author. Other specimens.—40 22, 59 46, from 155 Protothaca asperrima, type locality, 26 Oct 1981; 22 92, 30 66, from 60 Mytella guyanensis (Lamarck), Vera Cruz, approximately 10 km west of Panama City, Panama, 26 Oct 1981; 22 992, 15 46, from 60 Mytella guyanensis, same locality, 5 Nov 1981; 2 2°, from Anadara obesa (Sowerby), Vera Cruz, Panama, 5 Nov 1981. 590 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Wy MWY Wy yy y WV VY WW Wy | Wy 0.3 mm VOLUME 97, NUMBER 3 591 Female.— Body (Fig. 1a, b) elongate, widest in cephalosome. Length (not in- cluding setae on caudal rami) 0.99 mm (0.91—1.11 mm) and greatest width 0.26 mm (0.24—0.29 mm), based on 10 specimens from Protothaca in lactic acid. Ratio of length to width of prosome 2.2:1. Ratio of length of prosome to that of urosome 1.21:1. Segment of leg 1 weakly separated dorsally from cephalosome. Epimera of segments bearing second and third legs broadly rounded and projecting slightly in dorsal view. Small median sclerotized bar dorsally on head behind eye. Segment ofleg 5 (Fig. 1c)44 < 86 um. Genital segment elongate, 140 x 96 um, with slightly expanded lateral margins. Genital areas located dorsally just anterior to middle of segment and bearing 2 minute setae about 3 um long (Fig. Ic, e). Dorsal surface of genital segment smooth, but ventral surface ornamented with anterior and posterior transverse rows of spines joining laterally (Fig. 1d, e), thus forming circlet. Three postgenital segments from anterior to posterior 68 < 65, 65 x 52, and 38 xX 42 um, each segment ventrally with posterior transverse row of spines. Caudal ramus (Fig. 1f) elongate, 94 wm long, 14 um wide proximally, 10 um wide medially, and 4.5 wm wide distally. Ratio of length to greatest width 6.7:1. Armature consisting of proximal outer seta 34um, distal outer seta 33 um, dorsal seta 14 um, and 3 terminal setae from outer to inner 11, 33, and 14 um. All setae smooth. Body surface unornamented except for 4 minute hairs (setules ?) on dorsal surface of segment bearing fifth legs. Egg sac (in females from Protothaca) with seriate eggs, 2 eggs in sac measuring 218 < 96 um (Fig. 1g), 3 eggs in sac 286 xX 110 um (Fig. lh), and 4 eggs in sac 330 x 110 um (Fig. 11). Most females with 3 eggs in each sac. Rostrum (Fig. 1j) broad and not well defined. First antenna (Fig. 2a) 151 wm long, 7-segmented. Lengths of segments (mea- sured along posterior margin): 16.5 (45 wm along anterior margin), 22, 11, 14, 26, 17, and 20 um, respectively. Formula for armature: 4, 14,5, 3,4 + 1 aesthete, 2 + 1 aesthete, and 7 + | aesthete. All setae smooth. Second antenna (Fig. 2b) 160 um long, 3-segmented. First segment with diagonal row of spinules. Small second segment with minute seta. Elongate third segment slightly recurved, 96 x 16 um, ratio 6:1, bearing 1 minute seta near middle of concave edge, 3 small subterminal setae, and terminal claw 37 um. Segment ornamented along proximal half of convex side with small spinules. Labrum (Fig. 2c) with 2 hyaline posteroventral lobes and bearing prominent spines on outer lateral angles. Mandible (Fig. 2d) with 4 elements. Convex side having small process followed by unilaterally haired seta. Another longer unilat- erally haired seta nearby on dorsal surface. Distally mandible extended as long setiform element haired along one side and having bilaterally haired seta near its base. Paragnath not seen. First maxilla (Fig. 2e) with 4 setae. Second maxilla (Fig. 2f) 2-segmented. First segment large, ornamented with 3 groups of spinules. Sec- ond segment small, with minute proximal seta, smooth subterminal seta, and 2 unequal unilaterally haired terminal setae. —_— Fig. 1. Ostrincola falcatus, new species, female. a, Dorsal (scale A); b, Lateral (A); c, Urosome, dorsal (B); d, Urosome, ventral (B); e, Genital segment, lateral (C); f, Caudal ramus, dorsal (D); g, Egg sac, dorsal (A); h, Egg sac, dorsal (A); i, Egg sac, dorsal (A); j, Rostral area, ventral (D). 592 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Maxillipeds absent. Arrangement of appendages of cephalosome as in Fig. 2g. Legs 1—4 (Figs. 2h, 3b, c, d) with 3-segmented rami. Armature as follows (Roman numerals indicating spines, Arabic numerals representing setae): |B coxa O-I basis 1-I exp I-0; I-1; IV, 4 enp O-1; O-1; Js) PB coxa O-I basis 1-0 exp I-0; I-1; | As) enp O-1; 0-2; III, 3 IPs coxa O-I basis 1-0 exp I-0; I-1; TRS enp 0-1; 0-2; IV, 2 IP. coxa O-I basis 1-0 exp I-0; I-1; III, 5 enp O-1; 0-2; PV Coxa in all 4 legs with finely barbed inner spine. Basis in leg | with stout barbed spine and adjacent outer row of spinules, but in legs 2—4 this area without spine or spinules. Contour of margin of basis medial to endopod slightly indented in legs 3 and 4. Outer margins of segments of both rami bearing rows of spines, spines on endopods more prominent than those on exopods. Row of hairs in addition to spines on segments of endopods. In one female third segment of right exopod of leg | with IV, 3 (Fig. 3a), but left exopod with usual IV, 4. Leg 5 (Fig. 3e) 2-segmented. First segment 28 xX 23 um, bearing | dorsal seta and ornamented with 3 small spines. Second segment broad, 65 X 39 um, ratio 1.67:1, bearing 2 smooth setae and 2 finely barbed spines, lengths of these elements from dorsal to ventral 54, 20, 22, and 43 wm. This segment ornamented only with few small spinules distally on ventral edge. Leg 6 represented by 2 minute setae on genital area (Fig. le). Color in living specimens in transmitted light opaque gray, eye dark blue, egg sacs dark gray. Male.— Body (Fig. 4a, b) resembling that of female. Length (excluding setae on caudal rami) 0.86 mm (0.83—0.89 mm) and greatest width 0.21 mm (0.19-0.22 mm), based on 10 specimens from Protothaca in lactic acid. Ratio of length to width of prosome 2.2:1. Ratio of length of prosome to that of urosome 1.08:1. Segment of leg 5 21 < 75 um (Fig. 4c). Genital segment 120 x 96 wm, smooth dorsally but ventrally with band of spines in anterior half and row of spines on = Fig. 2. Ostrincola falcatus, new species, female. a, First antenna, ventral (scale D); b, Second antenna, antero-outer (D); c, Labrum, ventral (D); d, Mandible, ventral (E); e, First maxilla, ventral (E); f, Second maxilla, ventral (F); g, Midregion of cephalosome showing arrangement of appendages, ventral (C); h, Leg 1 and intercoxal plate, anterior (D). : Fig. 3. Ostrincola falcatus, new species, female. A, Abnormal third segment of left exopod of leg 1, anterior (scale D); b, Leg 2 and intercoxal plate, anterior (D); c, Leg 3 and intercoxal plate, anterior (D); d, Leg 4 and intercoxal plate, anterior (D); e, Leg 5, lateral (D). Fig. 4. Ostrincola falcatus, new species, male. a, Dorsal (scale A); b, Lateral (A); c, Urosome, ventral (C); d, Genital segment, lateral (A); e, Maxilliped, anterior (F); f, Maxilliped, posterior (F); g, Seta on third segment of maxilliped, posterior (E); h, Leg 5, lateral (F). VOLUME 97, NUMBER 3 sae (ALES ts / Gap YE x We, \ ‘ | ; X I LS , \ hs B CIE eee vw’, EZ yg, f ee Wp = Sfp LZ i === ; ia ——_ ya PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON VOLUME 97, NUMBER 3 595 596 Table 1.—Bivalve hosts, localities, and sources for the species of Ostrincola. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Species Host Locality Source O. gracilis Wilson, 1944 O. gracilis Wilson, 1944 O. gracilis Wilson, 1944 O. gracilis Wilson, 1944 O. gracilis Wilson, 1944 O. gracilis Wilson, 1944 O. clavator Humes, 1959 O. simplex Humes, 1959 O. koe Tanaka, 1961 O. koe Tanaka, 1961 . koe Tanaka, 1961 . koe Tanaka, 1961 . koe Tanaka, 1961 . koe Tanaka, 1961 . koe Tanaka, 1961 . koe Tanaka, 1961 . koe Tanaka, 1961 O O O O O. koe Tanaka, 1961 O O O O. koe Tanaka, 1961 O . koe Tanaka, 1961 O. japonicus Tanaka, 1961 O. japonicus Tanaka, 1961 O. japonicus Tanaka, 1961 O. portonoviensis Reddiah, 1962 O. portonoviensis Reddiah, 1962 O. portonoviensis Reddiah, 1962 Ostrea virginica Gme- lin Crassostrea virginica (Gmelin) Modiolus demissus granosissimus Sow- erby Mytilus recurvus Rafi- nesque Venus mercenaria mercenaria Linnaeus Tagelus gibbus (Spengler) Ostrea sp. Ostrea sp. Paphia (Paratapes) un- dulata (Born) Paphia Meretrix lusoria (R6d- ing) Cyclina sinensis (Gme- lin) Tapes japonica De- shayes Claudiconcha japonica (Dunker) Caecella chinensis De- shayes Mactra veneriformis Reeve Arcopagia (Merisca) diaphana (Deshayes) Tapes japonica De- shayes Tapes japonica De- shayes Tapes (Amygdala) phi- lippinarum Adams and Reeve Ostrea echinata Quoy and Gaimard Saxostrea echinata (Quoy and Gaimard) Ostrea (Crassostrea) gigas Thunberg Meretrix meretrix (Linnaeus) Meretrix casta (Chem- nitz, nec Deshayes) Meretrix casta (Chem- nitz, nec Deshayes) Beaufort, North Caro- lina Barataria Bay, Louisi- ana Barataria Bay, Louisi- ana Barataria Bay, Louisi- ana Barataria Bay, Louisi- ana Cotuit, Massachusetts Nosy Bé, Madagascar Nosy Bé, Madagascar Sakibe, near Sasebo, Japan Tsuyazaki, near Fuku- oka, Japan Near Saga, Ariake Bay, Japan Hiroshima, Japan Sakibe, Sasebo Bay, Japan Sakibe, Sasebo Bay, Japan Sakibe, Sasebo Bay, Japan Yue, Ariake Bay, Ja- pan Hiroshima, Japan Saga, Nagoya, Tsuya- zaki, Osaki, Japan Sasebo Bay, Japan Japan Sasebo, Japan Sakibe, near Sasebo, Japan Hiroshima, Japan Puddupeta, Portonovo, South India Puddapeta, Portonovo, South India Adyar estuary, near Madras, India Wilson (1944) Humes (1953) Humes (1953) Humes (1953) Humes (1953) Humes and Cressey (1960) Humes (1959) Humes (1959) Tanaka (1961) Tanaka (1961) K6, Murakami, and Daiku (1962) K6, Murakami, and Daiku (1962) Ko, Murakami, and Daiku (1962) K6, Murakami, and Daiku (1962) K6, Murakami, and Daiku (1962) Ko, Murakami, and Daiku (1962) K6, Murakami, and Daiku (1962) K6o (1961) K6 (1969a, b, c) K6, Yoshikoshi, and Ito (1974) Tanaka (1961) K6, Murakami, and Daiku (1962) K6, Murakami, and Daiku (1962) Reddiah (1962) Reddiah (1962) Reddiah and Mam- men (1966) ——————————— eS VOLUME 97, NUMBER 3 597 Table 1.—(Continued). Species Host Locality Source O. portonoviensis Sanguinolaria (Soletel- Puddupeta, Portonovo, Reddiah (1962) Reddiah, 1962 lina) diphos (Gme- South India lin) O. (?) portonoviensis Mesodesma trigona Quilon, Kerala, India Pillai (1968) Reddiah, 1962 (Deshayes) O. falcatus, new Protothaca asperrima Chiman, Darien, Pana- _ present paper species (Sowerby) ma O. falcatus, new Mytella guyanensis Vera Cruz, Panama present paper species (Lamarck) O. falcatus, new Anadara obesa (Sower- ‘Vera Cruz, Panama present paper species by) flap of leg 6 (Fig. 4d). Four postgenital segments from anterior to posterior 44 x 58, 49 x 52, 42 x 41, and 26 x 35 um, each segment with posteroventral row of spines. Caudal ramus similar to that of female. Rostrum like that of female. First antenna resembling that of female but 1 aesthete added on fourth segment (at point indicated by dot in Fig. 2a). Second antenna like that of female. Labrum, mandible, first maxilla, and second maxilla similar to those of female. Maxilliped (Fig. 4e, f) with 3 segments and long terminal claw. First segment with distal inner process bearing prominent spines. Second segment relatively short, bearing on inner side 2 groups of spines and 2 unequal setae, and on outer side patch of small spines. Small third segment with prominent sickle-shaped seta (Fig. 4g) 30 um long. Claw 83 wm, with small spurlike sclerotization at inner base and slight prominence midway along concave margin. Tip of claw blunt with small hyaline cap. Legs 1—4 like those of female. Leg 5 (Fig. 4h) with first segment 13 x 15.5 wm. Second segment not as broad as in female, 33 < 17 um, ratio 1.94:1, armed as in female. Leg 6 (Fig. 4c, d) represented by 2 small setae on posteroventral flap of genital segment. Color as in female. Etymology.—The specific name falcatus, Latin meaning sickle-shaped, alludes to the shape of the seta on the third segment of the maxilliped of the male. Remarks.—Six species of Ostrincola have been described, all associated with bivalve mollusks. Records of these species are summarized in Table 1. (Pillai’s 1963 record of Ostrincola portonoviensis Reddiah, 1962, from Mesodesma trigona in India needs confirmation, since there are discrepancies between Reddiah’s original description and Pillai’s subsequent description. For example, contrary to the original description, Pillai shows spines on a process of the first segment of the maxilliped of the male, the third segment of the endopod of leg 3 with IV, 2, and the same segment of the endopod of leg 4 with IV, 2.) Although it is sometimes impossible to determine various details from the published descriptions of the several species of Ostrincola, it appears that the falciform seta on the third segment of the maxilliped of the male of O. falcatus 598 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Distribution of Ostrincola. C = clavator, F = falcatus, G = gracilis, J = japonicus, K = koe, P = portonoviensis, and S = simplex. is diagnostic. In other species of the genus this seta is straight or only slightly recurved (this seta not described in O. japonicus). The spurlike sclerotization at the base of the claw of the maxilliped of the male also seems characteristic of the new species. Specimens from Mytella guyanensis are often somewhat larger than those from Protothaca asperrima. Twenty females from Mytella had a length of 1.15 mm (1.01—1.35 mm) and greatest width of 0.32 mm (0.29-0.38 mm) and 20 males a length of 0.93 mm (0.75—1.01 mm) and greatest width of 0.24 mm (0.22-0.26 mm). Their size ranges overlap with specimens from Protothaca. A combination of the specimens from Protothaca with those from Mytella shows a length of 1.09 mm (0.91-1.35 mm) and greatest width of 0.30 mm (0.26—-0.38 mm) for females and a length of 0.91 mm (0.75-1.01 mm) and greatest width of 0.23 mm (0.19- 0.26 mm) for males. Only two females from Mytella were ovigerous, but both of these had elongate egg sacs with the eggs arranged in a cluster rather than seriate. In one female the sac contained 11 eggs and measured 485 x 215 um. The ranges of the bivalve hosts are of interest since presumably Ostrincola falcatus may occur over part or all of their ranges. Protothaca asperrima is found from the Gulf of California to Peru, Mytella guyanensis occurs from Lower Cal- ifornia to northern Peru and on the coast of Venezuela southward to Brazil, and Anadara obesa ranges from Mexico to northern Peru (Olsson 1961). The various species of Ostrincola, whose geographical distribution is shown on the accompanying map (Fig. 5), may be distinguished by the use of the following key: Key to the species of Ostrincola 1. Endopod of leg | with third segment armed with II,4 ................ 2 — Endopod of leg 1 with third segment armed with I,5 ................. 3 VOLUME 97, NUMBER 3 599 2. Endopod of leg 3 with third segment armed with IV,1 .......... O. clavator — Endopod of leg 3 with third segment armed with III,3 ....... -.. O. simplex 3. Endopod of leg 3 with third segment armed with II,4 and that of leg 4 wwarinL JTL 3) “as oooh da ish tes RCS ore cre ve De O. portonoviensis — Endopod of leg 3 with third segment armed with IV,2 and that of leg 4 SOviTiL La Vg ROT ey CNEL TN SMBH See cits Pertepe ol CUNO E agate OF UE fe 4 4. Leg 5 in female with second segment round, flattened, ratio 1:1 . O. gracilis Leg 5 in female with second segment longer than wide, at least 1.36:1 . 5) 5. Leg 5 in female with second segment having several spinules on both sides — Leg 5 in female with second segment with few spinules only on one side Nr ee rey braineh eA a) ARMPIT UAT TOU RONDE, ONT O. falcatus 6. Second antenna with third segment long and slender, ratio approximately ia PD Ree Wes catenin gag toh cia 4 s+, « lam so egentees mle O. japonicus — Second antenna with third segment relatively short, ratio approximately 215 nar oe reread A Leister see boop pela WN ilar menesiiae ns bi aera roe spy opined tha a O. koe Pseudomyicola Yamaguti, 1936 Pseudomyicola spinosus (Raffaele and Monticelli, 1885) Material studied.—1 2, 1 6, from 60 Anadara obesa (Sowerby), Vera Cruz, Panama, 5 Nov 1981. Remarks.— Humes (1968) listed 39 hosts for this copepod. Since then the co- pepod has been reported in Japan from Mytilus edulis (Linnaeus) and Septifer virgatus (Wiegmann) by Ho (1980), who regarded Pseudomyicola ostreae Ya- maguti, 1936, as a synonym. On the southern Californian coast P. spinosus is common in Mytilus edulis and Mytilus californianus Conrad (Ho 1980). In New Zealand it is abundant in Crassostrea glomerata Gould (Dinamani and Gordon 1974). Furthermore, I have seen specimens of P. spinosus from Chione (Austro- venus) stutchburyi (Gray) from Cox’s Creek, Westmere, Auckland, New Zealand (collected by Dr. Philippa M. Black in 1971 in the upper reaches of Waitemata Harbor, within the city). The addition of these four new hosts increases the number of bivalves serving as hosts for P. spinosus to 44. Acknowledgments I thank the staff of the Naos Marine Laboratory, Smithsonian Tropical Research Institute, Panama, for many courtesies during my visit of one month, and Dr. Joseph Rosewater, Smithsonian Institution, for the identification of the bivalves. The trip to Panama and subsequent study was made possible by a grant (DEB 80 16421) from the National Science Foundation. Literature Cited Dinamani, P., and D. B. Gordon. 1974. On the habits and nature of association of the copepod Pseudomyicola spinosus with the rock oyster Crassostrea glomerata in New Zealand.—Journal of Invertebrate Pathology 24:305-—310. Ho, J.-S. 1980. Origin and dispersal of Mytilus edulis in Japan deduced from its present status of copepod parasitism.— Publications of the Seto Marine Biological Laboratory 25:293-313. 600 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Humes, A. G. 1953. Ostrincola gracilis C. B. Wilson, a parasite of marine pelecypods in Louisiana (Copepoda, Cyclopoida).— Tulane Studies in Zoology 1:99-107. . 1959. Copépodes parasites de mollusques 4 Madagascar. — Mémoires de I’Institut Scientifique de Madagascar, 1958, sér. F, 2:285-342. . 1968. The cyclopoid copepod Pseudomyicola spinosus (Raffaele & Monticelli) from marine pelecypods, chiefly in Bermuda and the West Indies. —Beaufortia 14:203-226. , and R. F. Cressey. 1960. Seasonal population changes and host relationships of Myocheres major (Williams), a cyclopoid copepod from pelecypods.—Crustaceana 1:307-325. Ko, Y. 1961. The biology of the commensal copepods in the Japanese clam, Tapes japonica De- shayes.— Records of Oceanographic Works in Japan, special number 5:129-141. 1969a. On the reproduction and metamorphosis of a commensal copepod, Ostrincola koe, in the Japanese clam, Tapes japonica (Preliminary note).— Bulletin of the Faculty of Fisheries, Nagasaki University 27:1-7. 1969b. The life-history of a commensal copepod, Ostrincola koe, in Japanese marine bi- valves.— Bulletin of the Faculty of Fisheries, Nagasaki University 28:73-82. 1969c. On the occurrence of four commensal copepods in the Japanese clam, Tapes ja- ponica.—Bulletin of the Faculty of Fisheries, Nagasaki University 28:83-91. , Y. Murakami, and K. Daiku. 1962. The biology of the commensal copepods in Japanese marine bivalves.— Records of Oceanographic Works in Japan, special number 6, pp. 113-119. — . K. Yoshikoshi, and N. Ito. 1974. External anatomy of a commensal copepod, Ostrincola koe-II. Nauplius stage.— Bulletin of the Faculty of Fisheries, Nagasaki University 38:87—93. Olsson, A. A. 1961. Mollusks of the tropical eastern Pacific particularly from the southern half of the Panamic-Pacific faunal province (Panama to Peru).— Paleontological Research Institution, Ithaca, New York. Pp. 1-574. Pillai, N. K. 1963. Copepods associated with South Indian invertebrates. — Proceedings of the Indian Academy of Sciences 58:235—247. Raffaele, F., and F. S. Monticelli. 1885. Descrizione di un nuovo Lichomolgus parassita del Mytilus gallo-provincialis Lk.— Atti della R. Accademia Nazionale dei Lincei, (4), Memorie della Classe di Scienza Fisiche, Matematiche e Naturali 1:302-307. Reddiah, K. 1962. Copepods associated with Indian mollusks (C). Ostrincola portonoviensis n. sp. from commercial bivalves at Portonovo, S. India.—Crustaceana 4:1-6. , and M. A. Mammen. 1966. Copepods associated with Indian molluscs (D). Copepods in Meretrix casta (Chemnitz).— Journal of the Marine Biological Association of India 8:141-145. Tanaka, O. 1961. Oncopepods associated with marine pelecypods in Kyushu. —Journal of the Faculty of Agriculture, Kyushu University 11:249-273. Wilson, C. B. 1944. Parasitic copepods in the United States National Museum. — Proceedings of the United States National Museum 94:529-582. Yamaguti, S. 1936. Parasitic copepods from mollusks of Japan, 1.—Japanese Journal of Zoology 7: 113-127. Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 601-614 A REVIEW OF THE GENUS PLATYNEUROMUS (INSECTA: NEUROPTERA: CORYDALIDAE) Michael J. Glorioso and Oliver S. Flint, Jr. Abstract.—The genus Platyneuromus consists of three species: soror (Hagen), honduranus Navas, and reflexus n.sp. Doeringia christel Navas is synonymized with P. soror, and P. auritus Kimmins with P. honduranus for which a neotype is designated. The male genitalia and postocular flanges of all species are figured and described. The flange of P. soror and P. honduranus varies greatly in size, but analysis shows that the variation is in direct relationship to gross size of the head. In 1979 the authors met and discussed various problems and projects in the Megaloptera. At that time a single specimen of an undescribed species of Platy- neuromus in the collection of the junior author served as the focus of an agreement between us to review this genus. Michael Glorioso began his study of the species, investigating characters, making measurements, and writing preliminary descrip- tions, while I (OSF) borrowed and forwarded material to him. On 17 October 1980, Michael died very suddenly. Dr. C. A. Triplehorn invited me to visit Ohio State University to work with them in sorting-out material and notes. I obtained all Michael’s notes and recovered the material of Platyneuromus. Study of these materials revealed considerable progress with this project on the part of the senior author. He had segregated the species, recognized the genitalic differences in the males, prepared many inked drawings of the male genitalia, made many measurements of various head dimensions and worked out the table of relationship between head size and flange width, and prepared preliminary descriptions of the genus and species. I have taken these notes and illustrations and filled them out to produce a uniform treatment of all species. Systematic Account Three genera of dobsonflies, Chloronia, Corydalus, and Platyneuromus, occur in Mexico and Central America. The first is known from northern Mexico to southern Brazil including the Lesser Antilles, and the second is widespread from southern Canada to northern Argentina, but is not known from the West Indies. The genus Platyneuromus is, however, totally restricted to middle America, being known from northeastern Mexico as far south as northern Panama. The genus Platyneuromus is readily distinguished by the postocular flange, the flattened lateral margin of the head (Glorioso 1981). The flange is developed to varying degrees depending on the species, sex, and, not surprisingly, the size of the example. The relationship between gross size and development of a sexually correlated structure is paralleled in Corydalus by the relationship between body size and length of male mandible in certain species. The very large postocular flange of a large example of P. soror (Hagen) led Navas (1925) to describe the example as a new species, genus and tribe—Doeringia christel, Doeringiini! 602 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON We here describe the new species, and provide descriptions and figures of the male genitalia for all known species, and report on the size relationship of the postocular flange. Platyneuromus Weele Platyneuromus Weele, 1909:252; 1910:23.—Lestage, 1927:94.— Penny, 1977:8.— Glorioso, 1981:282 (type-species Corydalus soror Hagen, by monotypy). Doeringia Navas, 1925:207.—Lestage, 1927:94.—Penny, 1977:8.—Glorioso, 1981: 282 (type-species Doeringia christel Navas, by monotypy; synonymized Glo- rioso, 1981:282). Diagnosis.—This genus, restricted to Mexico and Central America, is the only Corydalidae with notable flattened lateral margins of the head. Description. —Size: length 20-55 mm; wingspan 50-140 mm. Coloration: luteus to fulvus; pronotum margined laterally with fuscous to piceus vitta, vitta often on postocular flange; wings fulvus with darker tessellations. Head: broad, extremely flattened; posterior tentorial pit arcuate; postocular flange at least width of eye; antenna filiform, about length of forebody with 40- 55 articles; clypeal margin feebly to moderately incised medially; labrum broadly truncate, rounded at corners; maxilla relatively long and narrow; stipes with indistinct setae; lacinia with three long apical setae; galeal sensory peg well de- veloped; maxillary palp five-segmented, with distinct palpifer; labium with mod- erately setose ligula, four-segmented labial palp; terminal labial and maxillary palpal segments broadly rounded, with two sensory areas. Prothorax: notum more or less quadrate, prosternum often with 1 pair anterior acuminate spines, more pronounced in male. Wings: radial sector with 6-9 branches, last bifurcate; M,,> and M,,, each with 2 branches; Cu, with 4-5 accessories, Cu, unbranched; 1A 2 branched; 3 crossveins between R, and Rs and between Rs and M; 2 medial crossveins, 4 m-cu’s; 1 cubital crossvein, 24—30 costal crossveins; first R, cell one to one and a half times length of third. Males: ninth sternite nearly quadrate, broadly rounded or moderately incised posteriorly, membrane behind sternite thickened, bilobate; genital papillae weakly developed or absent, ninth tergite sparsely setose, internal inflection inverted, V-shaped; ninth gonostyli densely setose, clavate or unguiform, with terminal, glabrous, chitinous point; tenth ter- gites densely setose, clavate, occasionally with crenulate inner margin; tenth ster- nite broad, glabrous, moderately sclerotized; gonostyli long, digitiform, sparsely setose. Female: semicircular sternal pouch present between sixth and seventh segments; lateral sclerite well sclerotized; gonostylus fused with gonocoxite; bursa saclike, separate from spermathecal duct; accessory glands absent. Flange width.—A bivariate regression analysis was run in which the flange width was regressed on the interocular distance (as an indication of general body size) for 25 specimens of P. soror. Figure 4 shows the scattergram, regression line, regression equation, coefficient of correlation (R), R squared, and significance values. The coefficient of correlation for flange width on interocular distance is very high (.895) and the probability that there is only a random relationship between the measurements is less than 0.00001. The R squared value indicates that 80% of the variance in flange width is explained by the variance in interocular distance. Conclusion: it is almost certain that flange width is merely a function VOLUME 97, NUMBER 3 603 z PITT] Figs. 1-2. Platyneuromus honduranus (the holotype of P. auritus): 1, Habitus; 2, Head and prono- tum. of body size as expressed in interocular distance. The analysis of the flange width in P. honduranus is not quite as clear cut (Fig. 5). We suggest that this may be explained by a smaller sample size, in which the measurements are clustered at two extremes. We expect that specimens intermediate in size exist, but they were not available to us. Nevertheless, the results do corroborate the findings with P. soror, although at a lower R squared value—75%. Platyneuromus soror (Hagen) Figs. 3—4, 6-15 Corydalus soror Hagen, 1861:193 Neuromus soror (Hagen): Davis, 1903:467. Neuromus (Chloronia) soror (Hagen): Banks, 1908:30. Platyneuromus soror (Hagen): Weele, 1909:252; 1910, 23.—Stitz, 1914:199.— Navas, 1928:62.—Penny, 1977:8.—Glorioso, 1981:282. Doeringia christel Navas, 1925:29 (new synonymy). Diagnosis. — Male tenth tergites divergent basally, parallel distally; usually larger than P. honduranus; postocular flange with dark vitta, especially well developed in females. Description. —Size: body length 28—55 mm; forewing length males 34-59 mm [ave. 20 Mexican males, 43.9 mm; ave. 6 Costa Rican and Panamanian males 53.3 mm], females 42-63 mm [ave. 17 Mexican females, 50.2 mm; ave. 4 Costa Rican and Panamian females 60.0 mm]. Coloration: luteus to fulvus, fuscous to piceus vitta on lateral margin of pronotum and postocular flange. Postocular flange: in female barely exceeding eye, weakly concave anteriorly, moderately convex laterally, distinct from postocular spine; small males similar to females; flange much broader in large males, obliterating postocular spine. Pretarsus: un- guitractor longer than wide; median lobe of pulvillus with 2—4 setae. Male geni- talia: ninth sternite broadly rounded posteriorly; genital papilli poorly developed, often apparently absent; ninth gonostyli short, arcuate; tenth tergites diverging proximally, parallel distally; tenth sternite broad, moderately sclerotized, more heavily sclerotized anteriorly, styli long, digitiform, sparsely setose. 604 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Platyneuromus soror, fore- and hindwings. Variation.— The development of the postocular flange is relatively constant in females, but varies in males. The flange of small males is similar to that of females, while that of large males may be four times greater than the width of the eyes, with intermediate flange width in intermediate sized males (Figs. 6—9). Occurrence of large males with extremely large postocular flanges is greatest in southern Central America, where the species averages almost 10 mm larger in forewing length, although large-flanged males are infrequently collected at more northern localities. The dark vitta on the postocular flange may be restricted to half the width of the flange, or may cover the entire flange. Rarely the vitta is faintly developed or obscured by melanized areas on the head. Distribution.— The species is found from northeastern Mexico through central Mexico south to Chiapas, generally on the eastern coast. Records are then available from central Costa Rica south to northern Panama. It is not known if the lack of records from intermediate localities is an artifact of poor collecting or a true hiatus in the distribution of the species. Adults have been found flying together with those of P. honduranus in Chiapas, Mexico. In Mexico the species has been taken from late March to early September, and in southern Central America from late April into July. However, the peak of abundance would seem to fall between May and early July. Material examined.—Costa Rica, [Pcia. Cartago], Turrialba, May, 900 m, | 6 [DEI, holotype D. christel, 58 mm). Pcia. Heredia, Finca La Selva, 21-30 Jul 1976, J. C. Solomon, 1 2 [USNM, 62 mm]. VOLUME 97, NUMBER 3 605 FLANGE WIDTH 4.0 3.5] + 3.0 K- 2 = Os Lu © Za ee 2.0 =!) LL Y= .79848X -3.69604 1.5 Correlation (R) =0.89500 R squared= 0.80102 1.0 Significance = 0.00001 0.5 5.0 ES) 6.0 6.5 7.0 4.9 8.0 8.5 9.0 9.5 10.0 INTEROCULAR DISTANCE Fig. 4. Platyneuromus soror, regression of flange width on interocular distance. 3.5 3.0 2.5 2.0 1.5 Y =.856993X -2.49599 Correlation (R) =.868637 1.0 R squared = .75453 Significance = 0.001 0.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 INTEROCULAR DISTANCE Fig. 5. Platyneuromus honduranus, regression of flange width on interocular distance. 606 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 8 9 Figs. 6-9. Platyneuromus soror: 6, Head of male from Cuesta de Misantla, Mexico (wing, 49 mm); 7, Head of male from Santa Clara, Panama (wing, 51 mm); 8, Head of female from El Salto Falls, Mexico (wing, 48 mm); 9, Head of male from Santa Clara, Panama (wing, 59 mm). Mexico, 1 6 [MCZ, holotype C. soror, 45 mm]. Distrito Federal, J. R. Inda, | probably 2 [USNM, 62 mm]. Edo. Chiapas, 6 mi. S Puebla Nueva, 20 Mar 1953, Bechtel & Schlinger, 1 6 [CAS, 48 mm]. Edo. Chiapas, Palenque, 8 Jun 1969, Campbell & Bright, 1 ¢ [CNC, 45 mm]. [Edo. Mexico], Temescaltepec, 11 Jul 1933, Hinton & Usinger, 1 6, 2 9-[CAS, 6 42, 2 53, 58 mm]. Edo. San Luis Potosi, El Salto Falls, west Antiguo Morelos, ca. 2000 ft., 8 Jun 1961, U. Kans. Mex. Exped., 1 4, 1 2? [UKAL, 6 43, 2 51 mm]; same, but 17 Jun 1965, U. Kans. Mex. Exped., 2 °? [UKAL, 48, 54 mm]; same, but 29 Jun 1965, P. J. Spangler, 2 6, 1 ° [USNM, ¢ 39, 40, 2 45 mm]; same, but 11-14 Jul 1963, Duckworth & Davis, 2 2 [USNM, 43, 48 mm]; same, but 15-16 Jul 1963, U. Kans. Mex. Exped., 1 6, 4 VOLUME 97, NUMBER 3 607 Piewnl tity, Ao SEPA O gd 0 11 Figs. 10-14. Platyneuromus soror: 10, Male genitalia; dorsal; 11, Same, ventral; 12, Male tenth sternite and styli; 13, Pretarsus, ventral; 14, Female genitalia, lateral. 2 [UKAL, ¢ 38, 2 42, 43, 44, 45 mm]; same, but 4 Sep 1962, Ordway & Marston, 2 6, 1 2 [UKAL, 6 34, 42, 2 42 mm]. Edo. San Luis Potosi, Palitla, 3 Aug 1966, O. S. Flint, Jr., 1 ¢ [USNM, 43 mm]. Edo. San Luis Potosi, La (? Candiela), camino a Xilitla, 1 ¢é [IBUNAM, 50 mm]. Edo. Veracruz, Teocelo, Sep, E. A. Smyth collection, 1 2 [USNM, 61 mm]. Edo. Veracruz, Arroyo Claro, “‘Los Tux- tlas,’ 15 May 1977, 2 6 [IBUNAM, 45, 47 mm]. Edo. Veracruz, Balzapote, ‘“‘Los Tuxtlas,” 8 Sep 1977, J. Bueno, 2 6 [IBUNAM, 38, 40 mm]. [Edo. Veracruz], Cuesta de Misantla, M. Trujillo, 2 6 [BMNH, 47, 49 mm]. [Edo. Veracruz], Jalapa, M. Trujillo, 1 ?[BMNH, 58 mm]. [Edo. Veracruz], Rinconada, Sharis, 1 é[BMNH, 55 mm]. Edo. Veracruz, Ciudad Mendoza, 24 Apr 1953, Bechtel & Schlinger, 1 2 [CAS, 57 mm]. Edo. Veracruz, 5 mi. N Huatusco, 29 Jun 1971, Clark et al., 1 6 [TAMU, 48 mm]. Panama, Pcia. Chiriqui, Fortuna, 8°44’N: 82°15'’W, 1050 m, 29 Apr 1978, H. Wolda, 1 é[USNM, 53 mm]. Pcia. Chiriqui, El Valle de la Sierpe, 8°45’N: 82°15'W, 608 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON @® sOror Fig. 15. Known distribution of Platyneuromus soror. 5 May 1980, J. Peterson, 1 6, 1 2 [INPA, ¢ 50, 2 56 mm]. Pcia. Chiriqui, Santa Clara, 18 May 1977, Thurman, 3 6, 1 2 [INPA, USNM, ¢ 49, 51, 59, 2 63 mm]. [Pcia. Chiriqui], Volcan de Chiriqui, below 4000 ft., Champion, 1 ? [BMNH, 59 mim]. Platyneuromus honduranus Navas, new status Fig. 1-2, 5, 16-22, 31 Platyneuromus soror hondurana Navas, 1928:63. Platyneuromus auritus Kimmins, 1928:369 (new synonymy). Diagnosis.— Male tenth tergites with crenulate inner margins; ninth gonostyli unguiform; generally smaller than R. soror; vitta across postocular flange obscure or absent especially in females. Description. —Size: body length 20-28 mm; forewing length males 26-36 mm fave. 14 6, 31.0 mm], females 32—40 mm [ave. 13 2, 35.8 mm]. Coloration: luteus to fulvous, fuscus to piceus vitta on lateral margin of pronotum; vitta on postocular flange obscure or absent, especially in females. Postocular flange: in females, barely exceeding eye, concave anteriorly, moderately convex laterally, distinct from post- ocular spine; small males similar to females, flange much broader in large males, obliterating postocular spine. Pretarsus: unguitractor as wide as long, median lobe of pulvillus with 2 setae. Male genitalia: ninth sternite broadly and weakly incised posteriorly; genital papilli poorly developed, often apparently absent; ninth gon- ostyli long, unguiform, directed medially; tenth tergites nearly parallel, with cren- ulate inner margins; tenth sternite broad, moderately sclerotized, more heavily sclerotized anteriorly, styli long, digitiform, sparsely setose. VOLUME 97, NUMBER 3 609 Figs. 16-19. Platyneuromus honduranus: 16, Male genitalia, lateral; 17, Same, ventral; 18, Male tenth sternite and styli; 19, Pretarsus, ventral. Variation.— Development of postocular flange varies with size as in P. soror, although the clinal pattern seen in P. soror is not evident. Distribution.—The species is only known from southern Mexico through Gua- temala to northwestern Honduras. In Chiapas, Mexico it has been taken flying together P. soror or P. reflexus, but not yet with both at the same time. Collection dates range from the end of February to late May. Material examined.—Neotype, male, Honduras [Depto. Cortes], La Lima, 23 Feb 1965, C. Evers Q. [USNM 101511, 32 mm]. Guatemala, 1915, Wm. Schaus, 1 6 [USNM, 28 mm]. [Depto. Alta Verapaz], Panzos, Apr 1918, 1 2? [USNM, 37 mm]. [Depto. Izabal], Quirigua, Mar 1915, 610 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 21 Figs. 20-24. Platyneuromus honduranus: 20, Head of male from Agua Azul, Mexico (wing, 33 mm); 21, Head of male from Bonampak, Mexico (wing, 27 mm); 22, Head of female from Rio Tulya, Mexico (wing, 35 mm). Platyneuromus reflexus: 23, Head of male from Cascada Misolja, Mexico (wing, 40 mm); 24, Head of female from same locality (wing, 42 mm). VOLUME 97, NUMBER 3 611 YY iy, My y \! IW AS AYES UNA HOTA D LAG URANO NATTA ANIA 4 WK I nl AN Mf ‘ UM WAVY ar ety ACLANIIHAVAVON AW TPA HIME A NIDA TEAMMATE AN AMAA ANWRE DAM lt h oe OTANI eT TARY Intly Hs a AlN EA “clei Pe UNA AN (eINT TL RN ee £19 ; , TNA ALaTA aL A IRCA Or AW? 4 AY A RAMA me if 1 Figs. 25-30. Platyneuromus reflexus: 25, Male abdomen, lateral; 26, Male genitalia, dorsal; 27, Same, ventral; 28, Male tenth sternite and styli; 29, Pretarsus, ventral; 30, Same, lateral. Wm. Schaus, 2 2 [USNM, 34, 36 mm]. [Depto. Izabal], Cayuga, Mar, Schaus & Barnes, 1 6 [USNM, 36 mm]. Honduras, 2 6, 1 2 [BMNH accesion number 1921-503, 6 holotype P. auritus, second ¢ with replacement abdomen from another order, 6 32, 36, 2 39 mm]. [Depto. Atlantida], Lombardia, Wm. Mann, 1 2? [USNM, 37 mm]. 612 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON @ honduranus @ reflexus Fig. 31. Known distributions for Platyneuromus honduranus and P. reflexus. Mexico, Edo. Chiapas, Palenque, 8 May 1969, Campbell & Bright, | 2 [CNC, 38 mm]. Edo. Chiapas, Rio Tuliya, 48 km S Palenque, 17 May 1981, C. M. & O. S. Flint, Jr., 1 2[USNM, 35 mm]. Edo. Chiapas, Agua Azul, [50 km S Palenque], 1 May 1978, C. R. Beutelspacher, 2 6, 1 2 [IBUNAM, 4 28, 29, 2 32 mm]; same, but 22 May 1979, L. Rivera, 1 6, 1 2 [IBUNAM, ¢ 33, 2 33 mm]. Edo. Chiapas, Bonampak, 2—4 May 1978, C. R. Beutelspacher, 5 6 [IBUNAM, USNM, 26, 27, 30, 31, 35 mm]; same, but 20—25 May 1980, J. Bueno S., 1 2? [IBUNAM, 32 mm]. Edo. Chiapas, Tuxtla [Gutierrez], 7 May 1947, 1 é[PAAC, 31 mm]. Edo. Chiapas, Chiapa [de Corzo], 13 Mar 1953, Bechtel & Schlinger, 2 2 [CAS, 39, 40 mm]. Edo. Chiapas, Ocozocautla, 853 m, 17 May 1972, D. E. Breedlove, 1 2 [CAS, 34 mm]. Platyneuromus reflexus, new species Figs. 23-31 Diagnosis.—Postocular flanges pointed anterolaterally and slightly to strongly reflexed; male abdomen bearing setose tubercles laterally on segments 5, 6, 7; male tenth tergites gradually divergent. Description. —Size: body length 27-38 mm; forewing length males 38-47 mm [ave. 4 6, 42.8 mm], females 41-48 mm [ave. 4 2, 45.0 mm]. Coloration: luteus to fulvous, fuscous to piceus vitta on lateral margin of head and pronotum. Postocular flange: in females width barely exceeding that of eye, anterolateral angle pointed and directed slightly anteriad, lateral margin almost straight; males with flange broad and more strongly reflexed dorsad. Pretarsus: unguitractor longer than wide; median lobe of pulvillus with 4 setae. Male genitalia: ninth sternite broad, posterior margin broadly and slightly rounded; genital papillae strongly developed, often projecting as large, rounded, membranous lobes; ninth gonostyli VOLUME 97, NUMBER 3 613 short, arcuate, ending in sharp spine directed mesally; tenth tergites elongate, slightly inflated and slightly divergent apicad; tenth sternite broad, more heavily sclerotized anteriad, with small, lightly sclerotized anteromesal projection, gon- ostyli long, digitiform, sparsely setose, bases with distinct, lateral enlargement more strongly setose. Variation.— The series is quite uniform, but the smaller males have a slightly narrower and less reflexed postocular lobe, as in the other species of the genus. Distribution.—This poorly known species appears to have the most limited distribution of any in the genus, being known only from:-Chiapas, Mexico, and adjacent Guatemala. Adults of the species have been found flying with P. hon- duranus in southern Mexico. The species has only been collected in late May and June. Material examined.— Holotype, male, Mexico, Edo. Chiapas, Cascada Misolja, 20 km S Palenque, 17-18 May 1981, C. M. & O.S. Flint, Jr. [USNM 101510, 38 mm]. Paratypes: Same, data, 1 6, 2 2[USNM, 6 40, 2 41, 47 mm]. Edo. Chiapas, Agua Azul [50 km S Palenque], 22 May 1979, L. Rivera, 1 6, 2 2 [IBUNAM, ¢ 47,244, 48 mm]. Guatemala, [Depto. Alta Verapaz], Trece Aguas, Jun 1907, O. F. Cook, 1 6 [USNM, 46 mm]. Acknowledgments We thank the following institutions and individuals that made material avail- able: BMNH, British Museum (Natural History), London, P. C. Barnard; CAS, California Academy of Sciences, San Francisco, D. H. Kavanaugh; CNC, Cana- dian National Collection, Ottawa, J. E. H. Martin; DEI, Institute fiir Pflanzen- schutzforschung, Eberswalde, L. Diekman; IBUNAM, Instituto de Biologia, Uni- versidad Nacional Autonoma de Mexico, J. Bueno Soria; INPA, Instituto Nacional de Pesquisas da Amazonia, Manaus, N. D. Penny; MCZ, Museum of Comparative Zoology, Cambridge, M. Pearce & M. Thayer; PAAC, Philip A. Adams, Fullerton, California; TAMU, Texas A&M University, College Station, S. J. Merritt; UKAL, University of Kansas, Lawrence, G. W. Byers; USNM, National Museum of Natural History, Washington. Dr. R. Blum searched the collection of the Zoologisches Institiit und Zoolo- gisches Museum, Hamburg, for the type of P. honduranus, and reported its ab- sence. Dr. Jerry A. Louton ran the statistical computations and analyzed their significance. The Entomology staff artist, Mrs. Elaine R. Hodges, completed or corrected many of the genitalic figures and the wings, and executed the exquisite illustrations of the head. Mr. Victor E. Krantz photographed the type of P. auritus. Literature Cited Banks, N. 1908. On the classification of the Corydalinae, with descriptions of new species. — Pro- ceedings of the Entomological Society of Washington 10:27-30. Davis, K. C. 1903. Sialidae of North and South America.—New York State Museum Bulletin 68: . 441-487. Glorioso, M. J. 1981. Systematics of the dobsonfly subfamily Corydalinae (Megaloptera: Corydal- idae).—Systematic Entomology 6:253-290. Hagen, H. A. 1861. Synopsis of the Neuroptera of North America.—Smithsonian Miscellaneous Collections 4(1):xx + 1-347. Kimmins, D. E. 1928. New and little known Neuroptera of Central America.—Eos 4:363-—370. 614 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Lestage, J. A. 1927. La faune entomologique indo-chinoise, 2; les Mégaloptéres, deuxiéme parte. — Bulletin et Annales de la Société Entomologique de Belgique 67:93-119. Navas, L. 1925. Neuropteren, Megalopteren, Plecopteren und Trichopteren aus dem Deutschen Entomologischen Institiit (Berlin-Dahlem). I serie. —Entomologische Mitteilungen 14:205-212. . 1928. Insectos del Museo de Hamburgo. Primera serie. — Boletin de la Sociedad Entomologica de Espana 11:59-67. Penny, N. D. 1977. Lista de Megaloptera, Neuroptera e Raphidioptera de Mexico, America Central, ilhas Caraibas e America do Sul.—Acta Amazonica 7(4, supplement): 1-61. Stitz, H. 1914. Sialiden der Sammlung des Berliner Museums. Sitzungsberichte der Gesellschaft Naturforschender Freunde zu Berlin 1914:191-—20S. Weele, H. W. van der. 1909. New genera and species of Megaloptera Latr.— Notes from the Leyden Museum 30:249-264. . 1910. Megaloptera (Latreille), monographic revision. —/n Collections zoologiques du Baron Edm. de Selys Longchamps 5(1):1-93. (OSF) Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(3), 1984, pp. 615-617 A NEW SPECIES OF POTAMOCYPODA (CRUSTACEA: BRACHYURA: OCYPODIDAE) FROM MALAYSIA Ai-yun Tai and Raymond B. Manning Abstract.—Potamocypoda parapugil, the second species of the genus, is de- scribed. The new species differs from P. pugil in having the chelae equal and similar in both sexes. The crab characterized below was found in the collections of the Smithsonian by one of us (A.T.) during a visit to the United States in 1982. The holotype and some paratypes are in the collections of the Smithsonian; paratypes also have been deposited in the Institute of Zoology, Academia Sinica, Beijing. Potamocypoda parapugil, new species Fig. 1 Material.— Kuching, Sarawak, Malaysia [1°33’N, 110°20’E], in ditch by bridge, J. Crane, leg., 8 Aug 1955: male holotype, USNM 137241; 3 male, 6 female, and 1 juvenile paratypes, USNM 195341; 1 male and 1| female paratypes, Academia Sinica. Description.—Carapace subrectangular, length 0.75 times width in holotype, greatest width at or slightly posterior to midlength. Carapace convex from front to back, almost straight from side to side on anterior half, sloping down to lateral margin on posterior half. Lateral margin a raised, tuberculate ridge lined with setae, forming a curved crest over pterygostomian region, extending posteriorly from outer orbital angle to form anterolateral margin, then sharply turning ven- trally toward base of fourth leg, there turning posteriorly to form sinuous posterior margin. Posterolateral part of carapace with carina extending laterally, then curv- ing anteriorly as sharp ridge falling short of lateral margin. Low, transverse carina present on each branchial region above this ridge. Short curved carina extending mesially from point where lateral carina turns ventrally and meets posterolateral carina. Central part of carapace with broad, H-shaped depression. Cardiac and intestinal regions marked on each side by irregular depression ornamented with deep pits. Posterior margin paralleled by transverse ridge on carapace. Front broad, about 4 carapace width, deflected ventrally, appearing almost straight in dorsal view, with 2 low, short postfrontal ridges, 1 on each side of midline. Front marked with pits and an irregular linear depression. Dorsal surface of carapace smooth medially, pitted laterally, pits increasing in size posterolaterally. Ptery- gostomian region smooth but covered with short scattered hairs. Interantennular septum very small, inconspicuous. Orbits well formed, orbital margin a raised ridge lined with short setae. Eyes slender, tapering distally, cornea terminal. Third maxillipeds broad, mouthfield gaping slightly when closed. Ischium and merus very broad, former with oblique hairy ridge across surface. Palp much 616 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Potamocypoda parapugil: a—e, Male holotype: a, Dorsal view; b, Third maxilliped; c, Chela; d, Abdomen; e, Gonopod. f, Female paratype, terminal two segments of abdomen. narrower than basal 2 segments, carpus oval with central depression, propodus and dactylus much more slender. Second maxilliped slender, segments longer. Chelipeds small, slender, equal in both sexes. Fingers almost 2 times as long as palm, longitudinally ridged, gaping, unarmed, tips spatulate. Palm quadran- gular, length and depth subequal, surface with few punctations, upper surface with curved ridge. Carpus about twice as long as palm, measured dorsally, with raised inner ridge dorsally. Merus rather stout and short. Walking legs stout and short, second pair longest, fourth pair shortest. Merus tuberculate dorsally. Carpus longer than propodus, both with dense tomentum dorsally, heavier in males. Dactylus longer than propodus, diamond-shaped in cross-section, with 4 longitudinal ridges, tips corneous, sharp. Male abdomen with first 2 segments fused, not extending laterally to bases of fifth legs, crossed by sharp ridge. Abdomen constricted between second and third segments and at fifth segment, exposing gonopods, third and fourth segments broader than fifth to seventh segments. Female abdomen broad, but not extending to bases of fifth legs. Gonopod as illustrated, short, not extending to base of sternum of second leg. Measurements. — Male holotype, 8.0 . 7) bavedtr ood . 9k eriaacs..a8e D. vestitum 10. Undersurfaces of leaves and outer involucral bracts incompletely covered by pubescence, mostly hispid or pilosulous; heads usually born on elongate leafless DEGUNClESe ea thee seas oats ee one ae D. asperatum The following eleven species are recognized, including three transferred at this time from Oyedaea and two described as new. Because of some previous mis- identifications of material in the genus, specimens in the U.S. National Herbarium are cited below to indicate their correct identity and at the same time give some indication of the distribution. Dimerostemma annuum (Hassler) H. Robinson, comb. nov. Oyedaea annua Hassler, Repert. Nov. Sp. 14:175. 1915. PARAGUAY: In campis rupestribus calcareis in regione collis ““Margarita.”” Hassler 11030 (G, Photo US). The species apparently is known only from the type. Appearance would suggest that relationship is closest to D. humboldtianum. Dimerostemma asperatum Blake, Contr. Gray Herb. n.s. 52:12. 1917 BOLIVIA: East Velasco, 200 m, Jul 1892, O. Kuntze (Holotype US). More recent collections: BRAZIL: Distrito Federal: Cerrado, immediately south of Bra- silia, elev. 975 m. Irwin, Souza & Reis dos Santos 11112; Campo, north end of Lagoa Paranoa, Brasilia, elev. 975 m. Irwin, Grear, Souza & Reis dos Santos 13950; 46 km N of the bridge at Asa Norte in Brasilia on the dirt road to Vila Buritis, elev. 3100 ft. King & Almeda 8224; Vicinity of the University of Brasilia campus, elev. 2700 ft. King & Almeda 8230; Ecological Reserve S of Brasilia, elev. 3300 ft. King & Bishop 8921, 8927; Chapada da Contagem, 13 km nordeste de balao em BR 020 por estrada, 20.5 km nordeste de torre de televisiao, Brasilia. Kirkbride & Kirkbride 3098; Goias: Luziania (Saida da cidade). Heringer 18198; Campo, ca. 2 km N of Cristalina, elev. 1250. Irwin, Grear, Souza & Reis dos Santos 13294; 2 km S of Cristalina, elev. 2400 ft. King, Almeda & Eiten 8256; 13 km N of Cristalina along road to Brasilia, elev. 2800 ft. King & Bishop 8951. 622 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The species seems most notable for the large usually solitary heads on elongate leafless peduncles. The Bolivian type seems superficially different in appearance, but this is apparently because of the poor condition of the specimen. After careful examination I have found no essential difference. Dimerostemma bishopii H. Robinson, Phytologia 49:275. 1981 BRAZIL: Goias: 68 km NW along road from Iaciara to Nova Roma, elev. 1400 ft. King & Bishop 8803 (Holotype UB, isotype US). The species can be distinguished by its broad sharply acute to slightly acuminate leaves and by the lack of wings on the achenes. The lack of wings was passed over too lightly at the time the species was described, being a feature unique in the entire related group of Dimerostemma and Angelphytum, and a character that might cause the species to be placed in Aspilia according to more traditional concepts. Nevertheless, the species has all other characters of Dimerostemma and lacks any trace of the apical narrowing of the achene or the fiber sheaths in the throat of the corolla that are found in most species of Aspilia including all those seen from Brazil. Dimerostemma brasilianum Cassini, Bull. Soc. Philom. 1818:58. 1818. Ser- paea ovata Gardn., Lond. J. Bot. 7:296. 1848. Oyedaea ovata (Gardn.) Benth. ex Baker in Mart., Fl. Bras. 6(3):207. 1884. Oyedaea lippioides Baker in Mart., Fl. Bras. 6(3):208. 1884. Dimerostemma lippioides (Baker) Blake, Contr. Gray Herb. n.s. 52:14. 1917. BRAZIL: Distrito Federal: Bacia do Rio Sao Bartolomeu, cerrado as margens da rodovia DF 15. Heringer, Filgueiras, Mendonca & Pereira 6369; Goias: Portelandia. Hatschbach 34239, 20 km N of Corumba de Goias on road to Niquelandia, in valley of Rio Corumba, elev. ca. 1150 m. Jrwin, Maxwell & Wasshausen 187 90; Cerrado ca. 35 km NE of Catalao, elev. 900 m. Irwin, Onishi, da Fonséca, Souza, Reis dos Santos & Ramos 25266; 12 km S of Alto Paraiso de Goias, along road to Sao Joao da Alianga, elev. 3400 ft. King & Bishop 8865; 20 km N of Cristalina along road to Brasilia, elev. 2200 ft. King & Bishop 8946; Minas Gerais: Serra do Espinhaco, about 8 km N of Gouveia on road to Dia- mantina, elev. 1220 m. Anderson, Stieber & Kirkbride 35273; Ca. 4 km W of Campos Altos along Highway 262 to Uberaba, elev. 1160 m. Davidse & Rama- moorthy 10860; Morro das Pedras, ca. 25 km NE of Patrocinio, elev. 1050 m. Irwin, Onishi, da Fonséca, Souza, Reis dos Santos & Ramos 25502; Serra do Espinhaco, 13 km E of Diamantina, elev. 1000 m. Jrwin, da Fonséca, Souza, Reis dos Santos & Ramos 27510; Serra do Espinhago, 3 km N of Sao Joao da Chapada, road to Inhai, elev. 1200 m. /rwin, da Fonséca, Souza, Reis dos Santos & Ramos 28463; 10 km S of Diamantina, elev. 3600-4000. King & Bishop 8541; 75 km SE along road from Cristalina to Paracatu, elev. 2400 ft. King & Bishop 8979; Araxa, Barreiro?. Macedo 3138; Caldas. Widgren s.n., Morro das Pedras, near Belo Horizonte. Williams & Assis 5999. Blake (1917) was convinced that D. brasilianum and D. lippioides were separate species, so much so that he had difficulty understanding where Cassini had ob- tained his type material ofa species that Blake considered to be restricted to Goias in the interior of Brazil. A careful review of the material now available shows that D. lippioides is at best a form with somewhat larger heads in eastern Minas Gerais and according to literature and photographs in eastern Sao Paulo. There VOLUME 97, NUMBER 3 623 seems to be no significant discontinuity between such eastern specimens and material with smaller heads occurring as far east as central Minas Gerais. Dimerostemma episcopale (H. Robinson) H. Robinson, comb. nov. Oyedaea episcopalis H. Robinson, Phytologia 49:276. 1981. BRAZIL: Bahia: 14 km NW from the town of Rio das Contas along road to Pico das Almas, elev. 3300 ft. King & Bishop 8633 (Holotype UB, isotype US). Additional specimens: BRAZIL: Bahia: 12-14 km N of town of Rio das Contas on the road to Mato Grosso, elev. ca. 1200 m. Harley, Renvoize, Erskine, Brighton & Pinheiro 151636>7 100) have been collected from Little River, Blount County, Tennessee, than any other system, with the largest single series (11) collected there in 1971. The headwaters of Little River are 704 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 3.—Counts of total dorsal fin elements in Etheostoma cinereum. Population 22 23 24 25 26 27 n x SD CV Rockcastle R. 2 14 3 1 20 23.15 0.67 2.90 Buck Cr. 2 3 2) 7 23.00 0.82 3.55 Big South Fork 1 29 7 2 39 23.26 0.59 2.56 Obey R. 1 6 6 4 17 23.76 0.90 3.80 Roaring R. 1 1 25.00 mete Beef Red R. 1 2 3 22.67 0.58 2.55 Upper Duck R. 4 3 7 24.43 0.53 2.19 Buffalo R. 1 6 7 14 24.43 0.65 2.65 Clinch R. 1 1 25.00 tag nes! Emory R. 1 1 2 24.50 0.71 2.87 Little R. 3 15 34 16 3 71 25.01 0.89 3.57 Elk R. 1 1 23.00 protected in the Great Smoky Mountains National Park, and E. cinereum faces no immediate threat from habitat modification in that area, although a substantial reach of lower Little River has been subject to recent development. The only record of the species in the Elk River, Lincoln County, Tennessee, is a specimen collected by C. F. Saylor of the TVA near Fayetteville in 1981. The presence of this species in the Elk is surprising in light of water levels that vary more than 1 m daily due to power generation at Tims Ford Dam. Because this dam had been in operation for more than five years when the specimen was collected, there must have been recruitment in the population. Geographic Variation While considerable overlap among major drainage populations existed in the ranges of nearly all characters examined, univariate analyses revealed some sig- nificant differences and distinctive geographic trends. The limited range of E. cinereum and similarity of the streams it inhabits, particularly in regard to tem- perature regimes, suggest that genetic rather than environmental factors are re- sponsible for most geographic variation found. Results of discriminant analyses Table 4.—Counts of total vertebrae in Etheostoma cinereum. Counts of two specimens from the Obey River are taken from Bailey and Gosline (1955). Population 39 40 41 42 43 44 n x SD CV Rockcastle R. 4 16 20 41.80 0.41 0.98 Big South Fork 1 D 32, 1 3 38 40.97 0.49 1.20 Obey R. 9 9 18 41.50 0.51 1.24 Roaring R. 1 1 41.00 Bs 2 Bes & Upper Duck R. 6 1 7 43.14 0.38 0.88 Buffalo R. 1 13 14 42.93 0.27 0.62 Clinch R. 1 1 43.00 ees pote Emory R. 1 1 2) 42.50 0.71 1.66 Little R. D 47 18 67 42.24 0.50 1.18 Elk R. 1 1 43.00 ‘VOLUME 97, NUMBER 4 705 Table 5.—Counts of anal rays in Etheostoma cinereum. Population 7 8 9 n x SD CV Rockcastle R. 1 16 3 20 8.10 0.45 5.52 Buck Cr. 5 2 7 8.29 0.49 5.89 Big South Fork 31 8 39 8.21 0.41 4.99 Obey R. 16 1 17 8.06 0.24 3.01 Roaring R. 1 1 8.00 ke Nae Red R. 1 2 3 7.67 0.58 7.53 Upper Duck R. 6 1 7 8.14 0.38 4.64 Buffalo R. 3 11 14 8.79 0.43 4.85 Clinch R. 1 1 8.00 Emory R. 2 y) 8.00 ve Little R. 4 59 8 71 8.10 0.47 5.86 Elk R. 1 1 8.00 provided further evidence of isolation and divergence of some drainage popula- tions, and gave an indication of multivariate similarity among them. Counts of dorsal spines, dorsal rays, total dorsal fin elements, and vertebrae were all correlated and displayed similar geographic trends (Tables 1, 3, and 4). Counts of vertebrae and dorsal spines were also highly correlated in a population of Etheostoma exile (Gosline 1947) and in E. zonale (Tsai and Raney 1974), suggesting that these characters may be linked in development, possibly involving the differentiation of somites (Gosline 1947). Specimens from the Cumberland drainage, with the exception of Rockcastle River, typically had 41 vertebrae, 11 dorsal spines, and 12 dorsal rays. Most specimens from the Rockcastle River had 42 vertebrae, but displayed the same modal dorsal spine and ray counts as other Cumberland drainage populations. Duck and upper Tennessee drainage specimens typically had 42 or 43 vertebrae, 12 dorsal spines, and 12 or 13 dorsal rays. Only 27% of specimens from the upper Tennessee drainage had 43 or more vertebrae, compared to 95% in the Duck drainage. The Buffalo River population was unique in having a modal anal ray count of 9 compared to 8 in all others (Table 5). Scale counts were generally correlated with each other, and to a lesser degree, with Table 6.—Counts of lateral-line scales in Etheostoma cinereum. Population 51 52 53 54 55 56 57 58 59 60 61 62 63 n x SD CV Rockcastle R. 1 DEAD SES Del 20M IS St 5 VEO 23339 Buck Cr. 1 eX Zy 2 7 56.86 1.07 1.88 Big South Fork 2 SH Bein Sa Seas el 39 55.85 2.27 4.06 Obey R. 1 Nie Shel Si MSs oD: 17 58.47 1.66 2.84 Roaring R. 1 es Equals weight of ovaries x 1000/adjusted body weight. lygamous species of the subgenus Catonotus, in which females outnumber males (Lake 1936; Page 1974, 1975; Page and Burr 1976; Page 1980). Spawning.—As judged by size, weight, and color of gonads, and period of breeding coloration, the spawning season of E. cinereum probably extends from late January to mid-April, with peak spawning probably in mid-March. Spawning habitat and spawning behavior are unknown, but the conical-shaped genital papilla of ripe females (Fig. 2) indicates that the species may be an egg burier or egg attacher (following the classification of Page, 1983). Diet.—Of the 83 stomachs examined, only 72% contained at least 1 food item. The most common food items were chironomid larvae, larvae of the burrowing mayfly Ephemera, and oligochaetes. The abundance of Ephemera larvae and oligochaetes in the diet correlates the observation that the principal habitat of E. cinereum is sandy-bottomed pools where these organisms most frequently occur (Needham et al. 1935; Pennak 1978). Although the feeding behavior of E. cine- reum was not observed, the elongate snout and papillose lips may be specializa- tions for feeding on burrowing organisms. Less common items in the diet were, for the most part, either terrestrial organisms which had fallen into the water (e.g., formicids and cercopids), or aquatic organisms which are uncommon in the prin- cipal habitat of E. cinereum (e.g., Ferrisia and Stenonema). Substantial differences were found among food items of specimens from the Cumberland (N = 39) and upper Tennessee (N = 35) drainages. Chironomids accounted for a larger proportion of the diet of upper Tennessee drainage speci- mens (84%) than of Cumberland drainage specimens (52%). Ephemera accounted for 25% of the diet and occurred in 46% of the stomachs examined in Cumberland drainage specimens but were only 3.3% of the diet of upper Tennessee drainage specimens. Oligochaetes also comprised a larger proportion of the diet of Cum- berland drainage specimens (15.3%) than of upper Tennessee drainage specimens (2.2%). Too few specimens (2) from the Duck drainage were examined for stomach contents to make comparisons with other populations. Food habits varied among size classes. Specimens less than 40 mm SL fed almost exclusively on chironomid larvae. Chironomids were less important in the diets of 41 to 60 mm SL specimens, with larger food items such as ephem- ie PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON eropterans and oligochaetes accounting for 30% of the diet. Only 50% of stomachs of specimens larger than 60 mm SL contained chironomids, with 40% of the diet being ephemeropterans, plecopterans, trichopterans, and other large food items. Microcrustaceans, important in the diets of smaller size classes of many species of darters, were not found in the stomachs of any specimens of EF. cinereum, possibly because few specimens smaller than 40 mm SL were examined. Seasonal variation in the importance of several food items in the diet was apparent. While chironomid larvae were eaten every season, Ephemera larvae were a major food item only from late spring through early fall, and plecopteran larvae were present in the diet only from April until June. Parasites.—The only external parasites found on E. cinereum were fluke meta- cercariae; 76% of individuals examined had at least 1 metacercarial cyst, and 27% had 10 or more. The most heavily infested individual had 75 cysts. Cysts were present on individuals from all months in which the species has been collected. Of the 83 specimens dissected, encysted nematodes were found on the livers and intestines of nine, with the most heavily infested individual containing five cysts. The intestines of three specimens each contained one acanthocephalan worm. Acknowledgments We are grateful to Kevin S. Cummings, James M. Grady, Michael A. Klutho, and Stephen J. Walsh for helpful suggestions and aid in collecting specimens during the course of the study. Kevin S. Cummings made the genital papillae drawings, and Karen Schmitt of the Scientific Photography and Illustration Fa- cility, Southern Illinois University at Carbondale (SIUC) assisted in the prepa- ration of figures. Doris Sublette, former librarian for the Illinois Natural History Survey, provided library assistance and Karsten E. Hartel provided copies of correspondence between Hentz and Storer and minutes of a meeting of the Boston Society of Natural History. Branley A. Branson, Noel M. Burkhead, Neil H. Douglas, Robert E. Jenkins, Leslie W. Knapp, Robert A. Kuehne, John S. Ramsey, and Wayne C. Starnes provided color transparencies, field notes, and other helpful information about EL. cinereum. Charles F. Saylor provided a list of localities he had compiled for his account of E. cinereum in the “Atlas of North American Freshwater Fishes.” We are grateful to the following individuals and their institutions for granting specimen loans or otherwise allowing the examination of specimens: John S. Ramsey, Auburn University (AU); Robert Schoknecht, Cornell University (CU); Branley A. Branson, Eastern Kentucky University (EKU); Lawrence M. Page, Illinois Natural History Survey (INHS); Melvin L. Warren, Jr., Kentucky Nature Preserves Commission (specimens now at SIUC); Robert E. Jenkins (REJ), Roa- noke College; Neil H. Douglas, Northeast Louisiana University (NLU); Rudolph J. Miller, Oklahoma State Univesity (OAM); Charles F. Saylor, Tennessee Valley Authority (TVA); Herbert T. Boschung and David L. Nieland, University of Alabama Ichthyological Collection (UAIC); Carter R. Gilbert and George H. Burgess, University of Florida (UF); Reeve M. Bailey and Robert R. Miller, University of Michigan Museum of Zoology (UMMZ); Robert H. Gibbs and Susan Jewett, National Museum of Natural History (USNM); David A. Etnier and Bruce VOLUME 97, NUMBER 4 713 H. Bauer, University of Tennessee (UT); Bruce J. Turner, Virginia Polytechnic Institute (VPI); and Wayne C. Starnes (WCS), National Museum of Natural His- tory. Literature Cited Anonymous. 1973. Tennessee valley streams: their fish, bottom fauna, and aquatic habitat. Buffalo River drainage basin.— Division of Forestry, Fisheries and Wildlife Development, Tennessee Valley Authority. 17 pp. ' Bailey, R. M., and W. A. Gosline. 1955. Variation and systematic significance of vertebral counts in the American fishes of the family Percidae.— Miscellaneous Publications of the Museum of Zoology, University of Michigan No. 603. 44 pp. Branson, B. A. 1977. Threatened fishes of Daniel Boone National Forest, Kentucky.— Transactions of the Kentucky Academy of Science 38:69-73. ——, D. F. Harker, Jr., J. M. Baskin, M. A. Medley, D. L. Batch, M. L. Warren, Jr., W. H. Davis, W.C. Houtcooper, B. Monroe, Jr., L. R. Phillippe, and P. Cupp. 1981. Endangered, threatened, and rare animals and plants of Kentucky.— Transactions of the Kentucky Academy of Science 42:77-89. Burr, B. M. 1978. Systematics of the percid fishes of the subgenus Microperca, genus Etheostoma.— Bulletin of the Alabama Museum of Natural History No. 4. 53 pp. 1979. Systematics and life history aspects of the percid fish Etheostoma blennius with description of a new subspecies from Sequatchie River, Tennessee.—Copeia 1979:19 1-203. , and L. M. Page. 1978. The life history of the cypress darter, Etheostoma proeliare, in Max Creek, Illinois.—Illinois Natural History Survey Biological Notes No. 106. 15 pp. , and . 1979. The life history of the least darter, Etheostoma microperca, in the Iroquois River, Illinois.—Illinois Natural History Survey Biological Notes No. 112. 15 pp. Carrithers, R. B. 1971. An ecological survey of the West Fork of the Obey River, Tennessee, with special emphasis on the effects of the acid mine drainage.—M.S. Thesis, Tennessee Techno- logical University, Cookeville. 80 pp. Collette, B. B. 1965. Systematic significance of breeding tubercles in fishes of the family Percidae. — Proceedings of the United States National Museum 117:567-614. , and P. Banarescu. 1977. Systematics and zoogeography of the fishes of the family Percidae. — Journal of the Fisheries Research Board of Canada 34:1450-1463. , and L. W. Knapp. 1966. Catalog of type specimens of the darters (Pisces, Percidae, Etheos- tomatini).— Proceedings of the United States National Museum 119:1-88. Comiskey, C. E., and D. A. Etnier. 1972. Fishes of the Big South Fork of the Cumberland River. — Journal of the Tennessee Academy of Science 47:140-145. Dahlberg, M. D., and D. C. Scott. 1971. The freshwater fishes of Georgia. — Bulletin of the Georgia Academy of Science 29:1—64. Deacon, J. E., G. Kobetich, J. D. Williams, and S. Contreras. 1979. Fishes of North America endangered, threatened, or of special concern: 1979.— Fisheries 4:29—44. Ehrlich, P. R., and P. H. Raven. 1969. Differentiation of populations.—Science 165:1228-1232. Etnier, D. A., B. H. Bauer, and A. G. Haines. 1981. Fishes of the Gulf Coastal drainages of north Georgia (Part I), and fishes of Chickamauga, Chattanooga, and Lookout creeks, north Georgia and Tennessee (Part II).—Report to the U.S. Fish & Wildlife Service, Office of Endangered Species, Asheville, North Carolina. 226 pp. Gosline, W. A. 1947. Some meristic characters in a population of the fish Poecilichthys exilis: their variation and correlation.— Occasional Papers of the Museum of Zoology, University of Mich- igan No. 500. 23 pp. Hubbs, C. L., and M. D. Cannon. 1935. The darters of the genera Hololepis and Villora.—Miscel- laneous Publications of the Museum of Zoology, University of Michigan No. 30. 93 pp. , and K. F. Lagler. 1964. Fishes of the Great Lakes region.— University of Michigan Press, Ann Arbor. 213 pp. Jenkins, R. E., and J. A. Musick. 1980. Freshwater and marine fishes. Jn D. W. Linzey, ed., En- dangered and threatened plants and animals of Virginia.— Virginia Polytechnic Institute and State University, Blacksburg, pp. 319-373. 714 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Kirsch, P. H. 1893. Notes on a collection of fishes from the southern tributaries of the Cumberland River in Kentucky and Tennessee.— Bulletin of the United States Fish Commission 11:259— 268. Lachner, E. A., E. F. Westlake, and P. S. Handwerk. 1950. Studies on the biology of some percid fishes from western Pennsylvania.— American Midland Naturalist 43:92-111. Lake, C. T. 1936. The life history of the fan-tailed darter Catonotus flabellaris flabellaris (Rafin- esque).—American Midland Naturalist 17:816—830. Needham, J. G., J. R. Traver, and Y. Hsu. 1935. The biology of mayfilies.—Comstock Publishers, Ithaca. 759 pp. Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner, and D. H. Bent. 1975. Statistical package for the social sciences. 2nd ed.— McGraw-Hill Book Company, New York. 675 pp. Page, L. M. 1974. The life history of the spottail darter, Etheostoma squamiceps, in Big Creek, Illinois, and Ferguson Creek, Kentucky.—Illinois Natural History Survey Biological Notes No. 89. 20 pp. . 1975. The life history of the stripetail darter, Etheostoma kennicotti, in Big Creek, Illinois. — Illinois Natural History Survey Biological Notes No. 93. 15 pp. 1977. The lateralis system of darters (Etheostomatini).—Copeia 1977:472-475. 1980. The life histories of Etheostoma olivaceum and Etheostoma striatulum, two species of darters in central Tennessee.—Illinois Natural History Survey Biological Notes No. 113. 14 pp. 1981. The genera and subgenera of darters (Percidae, Etheostomatini).— Occasional Papers of the Museum of Natural History, University of Kansas No. 90. 69 pp. 1983. Handbook of darters.—Tropical Fish Hobbyist Publications, Neptune City, New Jersey. 271 pp. , and B. M. Burr. 1976. The life history of the slabrock darter, Etheostoma smithi, in Ferguson Creek, Kentucky.—Illinois Natural History Survey Biological Notes No. 99. 12 pp. , and R. L. Mayden. 1981. The life history of the Tennessee snubnose darter, Etheostoma simoterum, in Brush Creek, Tennessee.—IIlinois Natural History Survey Biological Notes No. 117. 11 pp. , and P. W. Smith. 1970. The life history of the dusky darter, Percina sciera, in the Embarras River, Illinois.—Illinois Natural History Survey Biological Notes No. 69. 15 pp. , and G. S. Whitt. 1973. Lactate dehydrogenase isozymes of darters and the inclusiveness of the genus Percina.—Illinois Natural History Survey Biological Notes No. 82. 7 pp. Pennak, R. W. 1978. Fresh-water invertebrates of the United States. 2nd ed.—John Wiley & Sons, New York. 803 pp. Ramsey, J. S. 1976. Freshwater fishes. Jn H. Boschung, ed., Endangered and threatened plants and animals of Alabama.—Bulletin of the Alabama Museum of Natural History No. 2, pp. 53-65. Raney, E. C., and E. A. Lachner. 1943. Age and growth of Johnny darters, Boleosoma nigrum olmstedi (Storer) and Boleosoma longimanum (Jordan).—American Midland Naturalist 29: 229-238. , and R. D. Suttkus. 1964. Etheostoma moorei, a new darter of the subgenus Nothonotus from the White River system, Arkansas.—Copeia 1964:130-139. Shaklee, J. B., and C. S. Tamaru. 1981. Biochemical and morphological evolution of Hawaiian bonefishes (A/bula).—Systematic Zoology 30:125-146. Smith-Vaniz, W.F. 1968. Freshwater fishes of Alabama.— Agricultural Experiment Station, Auburn. 211 pp. Starnes, W. C., and D. A. Etnier. 1980. Fishes. Jn D. C. Eagar and R. M. Hatcher, eds., Tennessee’s rare wildlife, volume I: The vertebrates.— Tennessee Wildlife Resources Agency, Nashville, pp. B-1-B-134. Storer, D. H. 1845. [No title].—Proceedings of the Boston Society of Natural History 2:47-50. Tackett, J. H. 1963. Clinch River basin: biological assessment of water quality, December 1963.— Virginia State Water Control Board. 67 pp. Tsai, C.-F., and E. C. Raney. 1974. Systematics of the banded darter, Etheostoma zonale (Pisces: Percidae).—Copeia 1974:1-24. Warren, M. L., Jr., and R. R. Cicerello. 1984. Drainage records and conservation status evaluations for thirteen Kentucky fishes. Brimleyana [in press]. Weatherley, A. H. 1976. Factors affecting maximization of fish growth.—Journal of the Fisheries Research Board of Canada 33:1046-1058. VOLUME 97, NUMBER 4 715 Williams, J.D. 1981. Threatened warmwater stream fishes and the endangered species act: a review. InL. A. Krumholz, ed., The warmwater streams symposium. —Southern Division of the Amer- ican Fisheries Society, pp. 328-337. Department of Zoology, Southern Illinois University, Carbondale, Illinois 62901. Present address (TES): Apartado 132, Canas, Guanacaste, Costa Rica. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 716-723 SYSTEMATIC IMPLICATIONS OF CHROMOSOMAL BANDING ANALYSES OF POPULATIONS OF PEROMYSCUS TRUEIT (RODENTIA: MURIDAE) William S. Modi and M. Raymond Lee Abstract.—C-band, G-band, and Ag-NOR analyses were carried out on the southern (AN = 54) and northern (AN = 62) cytotypes of Peromyscus truei from New Mexico, Texas, and Oregon. Results support the retention of comanche as a subspecies of P. truei, but populations possessing the southern cytotype are regarded as specifically distinct and referable to Peromyscus gratus. Peromyscus truei, as currently recognized, contains 15 subspecies including Peromyscus truei comanche (Hall 1981; Schmidly 1973). This arrangement differs somewhat from the latest taxonomic revision of the species (Hoffmeister 1951), which recognized 12 subspecies and regarded comanche as a subspecies of Pero- myscus nasutus (=P. difficilis). Several karyological studies have addressed the systematics and evolution of P. truei. These analyses indicate that two distinct chromosomal forms exist: a northern, 2n = 48, AN = 62 cytotype (AN = number of autosomal arms, see Lee and Elder 1977) from the western and southwestern United States, and a southern, 2n = 48, AN = 54 cytotype from New Mexico and Mexico (Hsu and Arrighi 1968; Lee et al. 1972; Zimmerman eft al. 1975). Further, the standard karyotype of P. t. comanche was shown to be identical to that of P. truei and quite different from that of P. difficilis (Lee et al. 1972). We present chromosomal banding comparisons of P. t. comanche and of the two P. truei cytotypes. We think the evidence derived both from our work and from previous studies most appropriately maintains P. t. comanche as a subspecies of of P. truei, while we recognize populations having the southern, AN = 54 cytotype as a distinct species referable to P. gratus Merriam, 1898. Methods. — Mice were live-trapped and shipped to the laboratory for processing. All chromosomal preparations were derived from bone marrow following Lee and Elder (1980). The differential staining procedures we utilized are also cited therein. Voucher specimens are deposited in the Museum of Natural History, University of Illinois, Urbana-Champaign. Specimens examined: P. ¢. truei (12), New Mexico: 2 mi NE Hanover, Grant Co., 1 female; 2 mi N Hanover, Grant Co., 2 males, 1 female; 8 mi S Magdalena, Socorro Co., 2 males, 1 female; 1 mi N Magdalena, 2 males; 2 mi up from Monica Canyon Ranger Station, San Mateo Mts., Socorro Co., 2 males, 1 female. P. ¢. comanche (4), Texas: 15 mi E Canyon, Randall Co., 3 males, 1 female. P. t. preblei aon) Fig. 1. a, G-banded karyotype of P. t. truei. Autosomes and X chromosome are from a female, the Y chromosome is from a male; b, G-banded karyotype of a male P. t. comanche; c, G-banded karyotype of a male P. gratus. 717 “VOLUME 97, NUMBER 4 g2 - a ae 6 Bs 1 * - GS wr) B,, wy tet 2D cas aii (at ke ate. coe _ st tg Pe: 68, ‘eakr om £ PS = z= wiser Cn os mia @ Be, enim ~ «ne eter we, ev oe™ ie, — 16 19 21 10 3 10 17 718 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (3), Oregon, 10 mi NW Terrebonne, Jefferson Co., 3 females. P. gratus (4), New Mexico: 2 mi NE Hanover, Grant Co., 1 female; 1 mi N Luna, Catron Co., 1 male; 3 mi up from Monica Canyon Ranger Station, San Mateo Mts., Socorro Co., 2 males. Results. — Peromyscus truei truei (2n = 48, AN = 62). All autosomes are acro- centric except 1, 2, 3, 6, 9, 15, 22, and 23. The X is a large submetacentric; the Y isa small acrocentric. The short arm of chromosome | is longer than that found thus far in other Peromyscus and appears to be the result of a pericentric inversion not involving the telomere (Figs. 1a, 3d). Autosomal heterochromatin is restricted _ to centromeric regions. The centromeric area and short arms of the X and the proximal one-third to one-half of the Y are heterochromatic (Fig. 2a). Our G- and C-band results appear to agree with those of Robbins and Baker (1981) for this species, although they apparently misidentified pairs 16, 17, 18, and 20 (see Committee 1977, Fig. 1a). Nucleolus organizer regions (NORs) are located te- lometrically on the short arms of pairs 1 and 22 (Fig. 3c). Peromyscus truei comanche (2n = 48, AN = 62). The karyotype is similar to that of P. t. truei. The elements within the G-banded karyotype of P. t. comanche and P. t. truei are homologous with the exception of the short arms of chromosome 6 and the X (Figs. 1b, 3d). These differences were not observed by Robbins and Baker (1981) in their examination of P. t. comanche. The chromosomal locations of C-bands are the same as in P. t¢. truei; however, the amount of heterochromatin in the karyotype of P. t. comanche is somewhat less (Fig. 2b). The silver-stained karyotype is indistinguishable from that of P. ¢. truei (Fig. 3a). Peromyscus truei preblei (2n = 48, AN = 62). The elements within the G-band- ed karyotype are homologous with those of P. t. truei and P. t. comanche, again with the exception of the short arms of chromosome 6 and the X (Fig. 3d). Each of the three populations of P. truei thus possesses different G-banding patterns for these two elements. The C-banded karyotype (not shown) is similar to that of P. t. comanche, and the silver-stained karyotype (also not shown) is identical to those of P. t. truei and P. t. comanche. Peromyscus gratus (2n = 48, AN = 54). All autosomes are acrocentric except pairs 1, 2,22, and 23. The X isa large submetacentric; the Y is a small metacentric. The Y chromosome and the short arm of the X have unique G-band patterns compared with other species of Peromyscus (Figs. 1c, 3d). (See Greenbaum ef al. 1978: Robbins and Baker 1981; Yates et al. 1980). Heterochromatin is located centromerically on all autosomes and the sex chromosomes, and on the short arms of the Y (Fig. 2c). Three pairs of Ag-NORs are located telomerically on pairs 3, 7 or 8 or 9, and 22 (Fig. 3b). Discussion.— Peromyscus truei truei, P. t. comanche, and P. t. preblei, differ karyotypically in the inversion patterns found in chromosome 6, in the G-banding patterns of the heterochromatic-short arm of the X, and in the presence of slightly more heterochromatin in the karyotype of P. t. truei. These findings indicate that P. t. comanche has undergone only a slight amount of chromosomal change in the 10,000-year period that it has been geographically isolated from the main population of P. truei (Blair 1950), and is no more differentiated than are either of the other two populations that were examined. We think that P. t. comanche should be retained as a subspecies of P. truei. This same conclusion was reached by Schmidly (1973) using morphological criteria. Electrophoretic comparisons of VOLUME 97, NUMBER 4 TA ov yen Pon ad — ~~ 9 “ “= > 7 cd * * - a “ ea ; a &s | =) é : ' 4 ’ | = re q ° ** fie a b XY =e ow ye om °* ci » ae es te | ey i) = oe =" a0 $f AG GR da An an ae 54 an a@ 40 aa FS AR (, XY Fig. 2. a, C-banded karyotype ofa male P. t. truei; b, C-banded karyotype of a male P. t. comanche, c, C-banded karyotype of a male P. gratus. 720 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ~ A ' nn i wv ub a ¥ ” - n r ¥ ~~ re A () (2 at Aa “ az : or | a 0 ") § ‘> 4 A x Th. ‘ ' a ) a | 4 e¢ ae 4a f : i : 5 f ‘ *, : i “ os ” be ti , i os » B 16 : ‘ g A OA Se aa na a3 és | b X Y ¥; > f “A AN Ke ee ea tc pg $3 a4e . * 223 8 gyte Bag 8 "ates 0 oF 28 S228 FAS Sees FRy 1 3 6 ] 19 Wes 255i a” eggN sce X Y d Fig. 3. a, Silver-stained karyotype of a male P. t. comanche; b, Silver-stained karyotype of a male P. gratus; c, Partial silver-stained karyotype of a male P. t. truei showing two pairs of autosomes with Ag-NORs and the sex chromosomes; d, Partial, composite, G-banded karyotype (P. t. truei (t), P. t. comanche (c), P. t. preblei (p), P. gratus (g), from left to right in each set) showing elements for which G-band differences exist among the four taxa. The Y was not examined in P. t. preblei. VOLUME 97, NUMBER 4 ; 721 P. truei truei and P. t. comanche (Rogers’ similarity = 0.96, Johnson and Packard 1974) also suggest to us no more than subspecific differentiation of commanche; Johnson and Packard (1974), however, recommended specific status for coman- che. Unpublished data from breeding experiments in our laboratory also show that P. truei truei (New Mexico) and P. t. comanche are completely interfertile. The problem regarding P. truei and P. gratus is more complex. Karyotypic data from conventional and banded preparations are available for 122 specimens be- longing to these two taxa. Seventy-five individuals (AN = 62) have been examined from Utah and California (Hsu and Arrighi 1968); Arizona, Colorado, Utah, and Texas (Lee et al. 1972); New Mexico and Texas (Robbins and Baker 1981; present study); Oregon (present study); and Arizona, Utah, and New Mexico (Zimmerman et al. 1975). Forty-seven specimens (AN = 54) have been reported from New Mexico (Hsu and Arrighi 1968; present study), Coahuila (Lee et al. 1972), Chi- huahua and Durango (Zimmerman et al. 1975), and Michoacan and Mexico (Schmidly and Kilpatrick, pers. comm.). No karyotypic variation has yet been reported within either cytotype with the exception of the differences reported here between P. ¢. truei, P. t. comanche and P. t. preblei. Most importantly, near and at the contact zone of the AN = 62 and the AN = 54 cytotypes in New Mexico, 24 specimens have been collected which range from being fully sympatric to occurring within 20 miles of one another in Grant and Socorro Counties, and none was karyotypically intermediate (Hsu and Arrighi 1968; Robbins and Baker 1981; Zimmerman and Kilpatrick 1972; present study). When the cytotypes of P. truei and P. gratus are compared at least five major chromosomal differences can be demonstrated: (1) based on the primitive Pero- myscus karyotype proposed by Robbins and Baker (1981) P. truei has undergone pericentric inversions in chromosomes 1, 2, 3, 6, 9, and 15, whereas P. gratus has experienced an inversion only in chromosome 2 (although we tentatively follow Robbins and Baker, 1981, who designated euchromatic differences of this type as pericentric inversions, we think that latent centromeric activity rather than inversions may account for at least some of these differences in euchromatic arm morphology, e.g., chromosome 3); (2) significant differences are seen in the G-banded and C-banded patterns of the short arms of the X chromosomes; (3) major differences exist in the morphology and in the G- and C-band patterns found in the Y chromosomes; (4) a difference occurs in the number and location of Ag-NORs; and, (5) there is a difference in the total amount of autosomal heterochromatin; more C-band positive material is found in P. gratus. Since both of the cytotypes are distinctive, wide-ranging, relatively homogeneous, and occur sympatrically, in our opinion each warrants recognition as a distinct species. Electrophoretic comparisons of P. truei and P. gratus yielded values (S = 0.866, 0.872, Avise et al. 1979; S = 0.863, Zimmerman et al. 1975) that are represen- tative of those found between some species or semispecies of Peromyscus (Avise 1975; Zimmerman et al. 1978). The northern limits of the geographic distribution of P. gratus are not precisely known. In his revision of P. truei, Hoffmeister (1951:29) recognized a subspecific boundary between P. ¢. truei and P. t. gentilis that closely coincided with the United States-Mexican border. However, the presence of P. gratus in New Mexico, as well as in Mexico, indicates that the limits of P. gratus should be extended northward at least to central and southern New Mexico. The distribution of P. 122 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON gratus thus includes the present ranges of P. t. gratus, P. t. gentilis, P. t. erasmus, P. t. zapotecae, and part of P. t. truei. The remaining northern populations and subspecies represent P. truei (see Hall 1981:702). Systematics and Synonymy Peromyscus gratus Merriam Peromyscus gratus erasmus Finley Peromyscus truei erasmus Finley, 1952, University, Kansas Publications, Mu- seum Natural History, 5:625. Type from 8 mi NE Durango, 6200 ft., Durango. Peromyscus gratus gentilis Osgood Peromyscus gratus gentilis Osgood, 1904, Proceedings Biological Society Wash- ington, 17:61. Type from Lagos, Jalisco. Peromyscus truei gentilis, Osgood, 1909, North American Fauna, 28:175. Peromyscus truei truei, Hoffmeister, 1951, Illinois Biological Monographs, 21:30 (part). Peromyscus gratus gratus Merriam Peromyscus gratus Merriam, 1898, Proceedings Biological Society Washington, 12:123. Type from Tlalpan, Distrito Federal, Mexico. Peromyscus truei gratus, Osgood, 1909, North American Fauna, 28:173. Peromyscus sagax Elliot, 1903, Field Columbian Museum Publications 71, Zoo- logical Series 3(8):142. Type from La Palma, Michoacan. Peromyscus pavidus Elliot, 1903, Field Columbian Museum Publications 71, Zoo- logical Series 3(8):142. Type from Patzcuaro, Michoacan. Peromyscus zelotes Osgood, 1904, Proceedings Biological Society Washington, 17:67. Type from Querendero, Michoacan. Peromyscus gratus zapotecae Hooper Peromyscus truei zapotecae Hooper, 1957, Occasional Papers Museum Zoology, University Michigan, 586:6. Type from 1 mi E Tlacolula, 5700 ft., Oaxaca. Acknowledgments We thank J. D. Diersing, V. E. Diersing, and C. A. Rinaldo for making the live Peromyscus available to us. D. J. Schmidly and C. W. Kilpatrick provided information concerning specimens from southern Mexico. J. L. Patton and D. S. Rogers kindly reviewed a draft of the manuscript. The Texas Parks and Wildlife Department provided a permit allowing collection of P. t. comanche. Literature Cited Avise, J.C. 1975. Systematic value of electrophoretic data.—Systematic Zoology 23:465—481. , M. H. Smith, and R. K. Selander. 1979. Biochemical polymorphism and systematics in the genus Peromyscus VII. Geographic differentiation in members of the truei and maniculatus species groups.—Journal of Mammalogy 60:177-192. Blair, W. F. 1950. Ecological factors in speciation of Peromyscus.—Evolution 4:253-275. VOLUME 97, NUMBER 4 723 Committee for standardization of chromosomes of Peromyscus. 1977.—Cytogenetics and Cell Ge- netics 19:38—43. Greenbaum, I. R., R. J. Baker, and P.R. Ramsey. 1978. Chromosomal evolution and its implications concerning the mode of speciation in three species of deer mice of the genus Peromyscus. — Evolution 32:646-654. Hall, E.R. 1981. The mammals of North America. John Wiley and Sons, New York, 2:601-1181 + 90. Hoffmeister, D. F. 1951. A taxonomic and evolutionary study of the pinon mouse, Peromyscus truei.—Illinois Biological Monographs 21:1-104. Hsu, T. C., and F. E. Arrighi. 1968. Chromosomes of Peromyscus (Rodentia, Cricetidae) I. Evo- lutionary trends in 20 species.— Cytogenetics 7:417—466. Johnson, G. L., and R. L. Packard. 1974. Electrophoretic analysis of Peromyscus comanche Blair, with comments on its systematic status. — Occasional Papers, Museum, Texas Tech University 24:1-16. Lee, M. R., and F. F. B. Elder. 1977. Karyotypes of eight species of Mexican rodents (Muridae). — Journal of Mammalogy 58:479-487. , and . 1980. Yeast stimulation of bone marrow mitosis for cytogenetic investigations. — Cytogenetics and Cell Genetics 26:36—40. ——, D. J. Schmidly, and C. C. Huheey. 1972. Chromosomal variation in certain populations of Peromyscus boylii and its systematic implications.— Journal of Mammalogy 53:697—707. Robbins, L. W., and R. J. Baker. 1981. An assessment of the nature of chromosomal rearrangements in 18 species of Peromyscus (Rodentia: Cricetidae).— Cytogenetics and Cell Genetics 31:194— 202. Schmidly, D. J. 1973. The systematic status of Peromyscus comanche.—Southwestern Naturalist 18:269-278. Yates, T. L., R. J. Baker, and R. K. Barnett. 1979. Phylogenetic analysis of karyological variation in three genera of peromyscine rodents.—Systematic Zoology 28:40-48. Zimmerman, E. G., B. J. Hart, and C. W. Kilpatrick. 1975. Biochemical genetics of the truei and boylii groups of the genus Peromyscus (Rodentia).— Comparative Biochemistry and Physiology 52B:541-545. ——, C. W. Kilpatrick, and B. J. Hart. 1978. The genetics of speciation in the rodent genus Peromyscus. — Evolution 32:565-579. Department of Ecology, Ethology, and Evolution, University of Illinois, 515 Morrill Hall, 505 S. Goodwin Avenue, Urbana, Illinois 61801. Present address of WSM: Laboratory of Viral Carcinogenesis, National Cancer Institute, Fred- erick, Maryland 21701. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 724-735 A NEW SPECIES OF SKATE, NEORAJA CAROLINENSIS, FROM OFF THE SOUTHEASTERN UNITED STATES (ELASMOBRANCHII: RAJOIDED John D. McEachran and M. Stehmann Abstract.— Neoraja carolinensis, n. sp. is described from the continental slope of the southeastern United States. Neoraja carolinensis is very similar to its congeners, N. caerulea and N. stehmanni, and Breviraja africana recently de- scribed from off West Africa, but can be distinguished from them mainly by the extreme length of its anterior pelvic lobe and coloration. Neoraja carolinensis lacks cross-bars on the dorsal surface of the tail and possesses a uniformly dark abdominal region which is sharply marked off from the light colored interbranchial and cloacal regions. A key is given to the three congeners and B. africana, and their distribution is discussed. In their revision of Breviraja and descriptions of Neoraja, n. gen., N. (Neoraja), n. subgen. and N. (Fenestraja), n. subgen., McEachran and Compagno (1982) briefly described and illustrated the skeletal anatomy of a new species of Neoraja (Neoraja) from off the coast of North Carolina. This species was not formally described, however, because of lack of adequate material. McEachran (1984) removed UN. (Fenestraja) from Neoraja and placed it in Gurgesiella as a subgenus, thus restricting Neoraja to the two described species in the subgenus Neoraja. Since publication of the generic revision, four more specimens of the undescribed species were captured off the coast of the southeastern United States by the FRV Anton Dohrn which conducted a bottom trawl survey from George’s Bank to Florida during October and November 1979 (Stehmann 1980). Herein the species is described and compared with its two congeners, and the eastern central Atlantic Breviraja africana Stehmann and Seret, 1983. The latter species appears closely related to those of Neoraja in several aspects, and one of us (JDM) feels it should be assigned to this genus. Specimens of the new species were obtained from the National Marine Fisheries Service Systematics Laboratory, Smithsonian Institution, and the Aussenstelle Ichthyologie des Instituts fiir Seefischerei, Hamburg, (ISH). The former two spec- imens were donated to the National Museum of Natural History, Smithsonian Institution (USNM). External measurements were made according to Bigelow and Schroeder (1953) with the exception of the pelvic lobe measurements. The anterior lobe was measured from the proximolateral margin of the left prepelvic process (which can be felt externally) to the tip of the lobe. The posterior lobe was measured from the proximolateral margin of the left prepelvic process to the tip of the lobe. Two specimens were dissected to reveal the structure of the neurocranium and scapulocoracoids. The neurocranium of one specimen was stained with alcian blue to distinguish the distal section of the rostral shaft. Unfortunately, none of the specimens were mature males so clasper structure could not be determined. VOLUME 97, NUMBER 4 725 All specimens were radiographed to verify the anatomical observations based on dissections and to count vertebrae and pectoral radials. Methods for making skeletal measurements followed Hubbs and Ishiyama (1968) and McEachran and Compagno (1979, 1982). Neoraja carolinensis, new species Figs. 1-6, Tables 1, 2 Neoraja, sp., McEachran and Compagno, 1982. Holotype. —ISH 3650/79, 285 mm TL, adolescent male, collected off Cape Fear, North Carolina, 33°38'N, 76°04'W, 796 to 800 m, 4.56°C, 34.958%o, 8 Nov 1979, FRV Anton Dohrn station 6410/79, by M. Stehmann. Paratypes.—ISH 3559/79, 280 mm TL, adolescent male, collected off the east coast of Florida, 29°11’N, 77°07’W, 1000 to 1008 m, 6.09°C, 35.035%o, 3 Nov 1979, FRV Anton Dohrn station 6385/79, by M. Stehmann.—ISH 3640/79 a+b, 143 mm TL, juvenile male, 232 mm TL, adolescent female, off Cape Fear, North Carolina, 33°27.5'N, 76°07'W, 990 to 1010 m, 4.18°C, 34.929%o, 8 Nov 1979, FRV Anton Dohrn station 6408/79, by M. Stehmann.—USNM 265013 (formerly Texas Cooperative Wildlife Collection, TCWC 2725.1), 247 mm TL, adolescent male, collected off Cape Fear, North Carolina, 33°56'N, 75°54'’W, 695 m, 31 Jan 1972, RV Oregon IT station 11757.—USNM 265014 (formerly TCWC 2740.1), 225 mm TL, juvenile female, collected off Cape Lookout, North Carolina, 34°22’N, 75°43'W, 805 m, 1 Feb 1972, RV Oregon IT station 11765. Diagnosis.— Anterior lobe of pelvic fin nearly as long or longer than posterior lobe of fin; dorsal surface of tail without distinct cross-bars; abdominal area uniformly dark colored and sharply marked off from light interbranchial and cloacal regions. Description. —Individual morphometrics and meristics given in Tables | and 2. Values of paratypes follow in parentheses those of holotype. Disc heart-shaped (Figs. 1, 2) 1.2 times (1.1—1.2 in paratypes) as broad as long; snout short, maximum angle in front of spiracles 120° (125°—136°); tip of snout with a small, flat, triangular process; anterior margin concave on either side of process, convex to level of orbits and concave from level of orbits to level of first gill slits (paratypes with less undulated anterior margin and smallest juvenile with almost convex margin); outer corners of disc broadly rounded; posterior margins and especially inner margins strongly convex. Axis of greatest disc width 0.76 (0.70-0.77) of distance from tip of snout to axil of pectoral fins. Pelvic fins deeply incised; anterior lobe long, narrow and tapering to acute tip; posterior lobe relatively short with convex lateral margin. Tail slender, little depressed, its width at mid-length about three- fourths eye diameter. Tail with lateral fold along each ventrolateral surface, ex- tending from about mid-length of tail to near tip (originating from mid-length to posterior third of tail); lateral folds broaden posteriorly to width equal to one- half height of epichordal caudal lobe at level of dorsal fins. Length of tail from center of cloaca to tip 1.4 (1.3-1.5) times distance from tip of snout to center of cloaca. Preorbital length 2.7 (2.4—2.7) times as long as orbit, 3.1 (2.9—3.2) times inter- orbital width; preoral length 1.7 (1.6—2.0) times internarial distance. Interorbital distance 0.9 (0.8—1.0) times length of orbit, orbit length 2.1 (1.9—2.4) times as 726 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Neoraja carolinensis ISH 3650/79, 285 mm TL (Holotype) adolescent male: a, Dorsal view; b, Ventral view. VOLUME 97, NUMBER 4 727 Table 1.—Proportional measurements and meristic values for Neoraja carolinensis. Proportions expressed as percentage of total length. Measurements made to nearest 1 mm are given to nearest 1 mm and those made to nearest 0.1 mm are given to nearest 0.1 mm. Holotype Paratypes Ma ISHie aNISH ISH wSH USNM USNM 3650/79 3559/79 3640/79 3640/79 265013 265014 x Sex ) ) g ) 3 9 Total length (mm) 285 280 232 143 247 225 Disc width 55.0 56.0 57.0 52.0 55.0 52.0 55.0 Disc length 47.0 47.0 48.0 43.0 47.0 48.0 47.0 Snout length (preorbital) 11.5 10.0! 11.3 97 10.4 11.4 # «210.8 Snout length (preoral) EO RS 27 1259) 6) SOF 124 Snout to maximum width 30.0 29.0 31.0 27.0 27.0 29.0 29.0 Prenasal length 9.4 7.6! 9.1 9.6 9.0 9.0 8.9 Orbit diameter 4.2 4.3 4.5 3.8 4.9 4.2 4.3 Distance between orbits 3.7 3.5 3.8 3.7 3.4 3.6 3.6 Orbit and spiracle length 4.9 4.9 5.4 4.9 6.0 5.0 5.2 Spiracle length 2.0 2.0 1.9 2.0 Del] 1.9 2.1 Distance between spiracles 7.0 7.3 7.3 7.0 6.6 6.7 7.0 Mouth width 7.3 7.9 7.0 6.2 7.4 6.7 V2 Nare to mouth 4.1 3.7 4.1 36 — 3.9 3.9 Distance between nostrils 7.6 7.5 7.9 6.9 V2 6.5 V3 Width of first gill opening 1.3 1.2 1.3 0.7 1.5 0.9 12 Width of third gill opening 1.4 1.2 1.4 1.0 1.6 1.0 1.3 Width of fifth gill opening 1.0 0.9 1.0 0.6 1.3 0.8 0.9 Distance between first gill openings 12.6 13.5 W283 Wai 132 Wai 29 Distance between fifth gill openings VD 7.6 7.8 7.4 8.2 Tol 7.5 Length of anterior pelvic lobe 13.8 15.1 14.7 12.8 15.8 145 14.6 Length of posterior pelvic lobe 15.0 15.2 15.0 13.2 160 149 15.0 Tail width of axil of pelvic fins 4.2 3.0 4.4 3.4 3.7 3.8 3.8 Distance-snout to cloaca 42 41 43 40 42 41 42 Distance-cloaca to first dorsal fin origin 47 47 44 47 48 47 47 Distance cloaca to 2nd dorsal origin 50 53 49 53 52 51 51 Distance-cloaca to caudal fin origin 55 56 54 57 53 56 55 Distance-cloaca to caudal fin tip 58 59 57 60 59 59 58 First dorsal fin, height 2.6 3.4 2.8 1.9 2.8 3.3 2.8 First dorsal fin, base length 5.7 6.0 5.8 6.6 5.8 5.6 5.9 Second dorsal fin, height 2.0 Ded 3.1 1.7 2.6 Doi 2.4 Second dorsal fin, base length 4.2 5.0 5.5 5.2 3.7 4.7 4.7 Tail at pelvic tips, height 2.6 2.5 Del DoS) 2.6 2.6 2.6 Tail at pelvic tips, width 3.5 3.3 3.8 3.5 3.0 3.4 3.4 Tail at first dorsal origin, height 1.1 0.9 1.1 1.0 0.9 0.9 1.0 Tail at first dorsal origin, width 1.9 1.4 1.9 1.6 1.5 1.4 1.6 Head length, to 5th gill slits 25.4 24.3 25.4 24.7 — 24.9 24.9 (5) Nasal curtain, length 5.1 4.6 48 43 — 4.6 4.7 (5) Nasal curtain, width each lobe 32) 2.6 3.0 2.8 2.8 D7 2.9 Nasal curtain, distance between lobes 3.0 3.6 3.0 3.0 2.4 2.5 2.9 Clasper, postcloacal length 13.4 7.0 — 6.3 5 = 9.1 (4) Angle of snout 120° 136° WGP WP Was IDsP Nae Pseudobranchial folds 8 10 9 9 8 10 9 Number of tooth rows (upper jaw) 44 45 46 40 42 42 43 Number of trunk vertebrae 24 24 25 24 26 26 25 Number of predorsal caudal vertebrae 73 70 65 70 72 70 70 Number of pectoral fin radials 64 64 62 63 65 66 64 1 Snout slightly deformed. 728 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Neurocranial and scapulocoracoid proportional measurements of Neoraja carolinensis expressed in percent of nasobasal length and greatest length respectively. ISH 3640/79" USNM 265013 232 mm TL 247 mm TL Q 3 ee ————— eee Nasobasal length (mm) 25.1 24.6 Cranial length 178 183 Rostral cartilage length 77 80 Prefontanelle length 61 58 Cranial width 117 117 Interobital width 34 33 Rostral base 17 15 Anterior fontanelle length 49 47 Anterior fontanelle width 14 12 Posterior fontanelle length 42 41 Posterior fontanelle width 12 10 Rostral appendix length 60 48 Rostral appendix width 20 20 Rostral cleft length 53 41 Cranial height 24 23 Width across otic capsules 58 60 Least width of basal plate 30 30 Greatest width of nasal capsules 49 49 Internarial width 15 15 Greatest length (mm) 11.7 11.4 Greatest height 84 77 Premesocondyle 44 43 Postmesocondyle 56 56 Postdorsal fenestra length 16 13 Postdorsal fenestra height 10 10 Anterior fenestra length 21 14 Anterior fenestra height 26 26 Height of rear corner 62 66 eS long as spiracle. Spiracle with 8 (8-10) pseudobranchial folds. Anterior nasal flap (nasal curtain) strongly undulated laterally, with distinct, short triangular loblet at mid-length and finely fringed posterior margin, posterior nasal flap poorly developed and very weakly fringed (smooth to weakly fringed) (Fig. 3). Mouth nearly straight, upper and lower jaws slightly arched on either side of symphysis (nearly straight in juveniles) and upper jaw slightly indented at symphysis. Teeth with very short pointed cusps near symphysis but with rounded cusps towards corners of jaws (juveniles with rounded cusps throughout); teeth in quincunx arrangement. Distance between first gill slits 1.6 (1.6—1.8) times as great as distance between nares; distance between fifth gill slits 0.9 (1.0—1.1) times as great as between nares; length of first gill slits 1.3 (1.1-1.4) times length of fifth gill slits and 0.2 (0. 1-0.2) times mouth width. First dorsal fin about equal in size and shape to second, their bases confluent and second dorsal confluent with short and low epichordal caudal lobe; hypochordal caudal lobe poorly developed. Not fully formed claspers ex- tending distal to posterior pelvic lobe (not exceeding pelvic lobe in paratype males), inner dorsal lobe of glans with two proximal clefts or slits, ventral lobe with elongate shield. ‘VOLUME 97, NUMBER 4 729 Fig. 2. Neoraja carolinensis ISH 3640/79,, 143 mm TL (Paratype) juvenile male: a, Dorsal view; b, Ventral view. Upper surface of disc except narrow posterior margins densely and uniformly covered with denticles bearing slender spines with recurved tip; anterior pelvic lobes naked; posterior pelvic lobes sparsely covered with denticles from origin to center with broadly smooth outer margins; distomedial aspect of claspers sparsely 730 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Neoraja carolinensis, ISH 3650/79 285 mm TL (Holotype) adolescent male, mouth and nasal region. covered with denticles (naked in male paratypes). Ventral surface naked except for denticles along lateral aspects of tail to level of first dorsal fin (some paratypes with lateral denticles only on posterior half of tail). Orbital thorns small, set in an almost continuous inner half-ring consisting of: 5 preorbital, innermost two being largest, and 3 postorbital, plus 2 supraorbital, of which | on left and both on right side broken off; 1 supra- and 1 interspiracular thorn on each side, right interspiracular broken off (paratypes with 3-5 pre-, 0—2 supra-, and 1-4 postor- bital, 0-1 supra- and O-1 interspiracular thorns). An irregular median row of 6 thorns over nape/shoulder area, smaller posterior two in suprascapular position; 2 thorns on each shoulder, left outer one broken off (paratypes with 2-5 median nuchal, 1—2 suprascapular, and 1—2 scapular thorns). About 75-80 small thorns from about level of maximum disc width along midline onto mid-length of tail in irregular row, on origin of tail in narrow band of two or three irregular rows; remaining tail section to origin of first dorsal fin without median thorns, but with longitudinal groove also lacking denticles (paratypes with 40—50 thorns in irregular midrow originating at posterior third of body, or at level of pectoral or pelvic axils; smallest paratype with 40 thorns in almost regular midrow along body and tail to origin of first dorsal and lacking smooth longitudinal groove on tail (Fig. 2a). Thorns with roundish bases and recurved tips bearing small lateral keels. Dorsal and upper caudal fins sparsely covered with denticles. Color.—After preservation in formalin and storage in alcohol, dorsal surface (Figs. la, 2a) uniformly grayish-brown, except area on either side of rostrum, VOLUME 97, NUMBER 4 731 Fig. 4. Neurocranium of N. carolinensis, ISH 3640/79,, 232 mm TL (Paratype) adolescent female: a, Dorsal view; b, Lateral view; c, Posterior view; d, Ventral view. ac—anterior cerebral vein foramen, af—anterior fontanelie, antc—antorbital condyle, end—endolymphatic foramen, es—efferent spirac- ular artery foramen, hf—hyomandibular facet, ic—internal carotid artery foramen, into—interorbital vein foramen, ja—jugal arch, obf—otic branch of facial nerve foramen, oc—occipital condyle, of— orbital fissure, onc—orbitonasal nasal canal foramen, os—optic stalk, pc—posterior cerebral vein foramen, peri—perilymphatic foramen, pf—posterior fontanelle, poc—preorbital canal foramen, postp— postorbital process, prep—preorbital process, prof—profundus nerve foramen, ra—rostral appendix, rb—rostral base, m—rostral node, rs—rostral shaft, II—optic nerve foramen, III—oculomotor nerve foramen, IV—trochlear nerve foramen, VII—hyomandibular branch of facial nerve foramen, IX— glossopharyngeal nerve foramen, X— vagus nerve foramen. posterior pectoral and outer margins of posterior pelvic lobes and apices of dorsal fins semitransparent and lighter; anterolateral margin of anterior pelvic lobes and claspers yellowish-white, but distolateral margin of left ventral clasper lobe brown; lateral tail folds marbled brown and white, almost yellowish-white in distal sec- tion; tail colored like remainder of disc with several light spots on left side in its posterior third. Ventral surface yellowish-white, except for margins of disc pos- terior to first gill slits, area lateral to gill slits, transverse axis of disc, abdomen, area lateral to cloaca, and margin of posterior pelvic lobes, all of which more or less spotted, blotched, or clouded grayish-brown; abdomen uniformly grayish- brown and sharply marked off from predominantly whitish interbranchial and cloacal areas; claspers yellowish-white, except for brown distolateral margin and pale brown cross-bar distally on left; origin of tail whitish, remaining section pale brown with scattered darker spots, and tip of tail whitish except for two grayish- brown cross-bars at level of dorsal fins. Paratypes similarly colored, except that tail may have vague dark cross-bars in anterior third and at dorsal fins (ISH 3559/ 79, 3640/79b, USNM 265013) and some variation in extension of brownish spots and blotches ventrally; ventral surface varying from predominantly whitish in smallest paratype (Fig. 2b). with abdomen thus appearing distinctly darker than remainder of ventrum, or predominantly dark with only few pale areas as in USNM 265014; ventral side of tail varying from entirely whitish with few dark spots, to brown on anterior and white on posterior half, or to completely light brown with white tip, with cross-bars at dorsal fins less evident. Freshly caught four ISH specimens showed brown color more intensively on both surfaces. 732 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1cm Fig. 5. Lateral view of left scapulocoracoid of N. carolinensis, ISH 3640/79,, 232 mm TL (Paratype) adolescent female: af—anterior fenestra, msc—mesocondyle, mtc—metacondyle, pdf—postdorsal fe- nestra, prc—procondyle, pvf—postventral fenestra, rc—rear corner, scp—scapular process. Neurocranium (mainly based on paratypes ISH 3640/79b and USNM 265013) with very slender, unsegmented rostral shaft tapering to filament abutting against but free of rostral node and appendices (Fig. 4a); rostral node consisting of narrow bridge joining rostral appendices at tip of snout; rostral appendices plate-like anteriorly with large elliptical foramina, posteriorly extending nearly to anterior fontanelle and conical in cross section, 60-80% of rostral length; propterygia of pectoral girdle terminating lateral to anterior margin of appendices; rostral base narrow, 18—23% of cranial width; nasal capsules ovoid-shaped, greatly laterally expanded and set at about 33° angle to transverse axis and at 67°-75° angle to longitudinal axis of neurocranium; profundus nerve with two foramina on:‘leading edge of each nasal capsule; interorbital region moderately narrow 26-31% of cranial width (Table 2); preorbital processes poorly developed, separated from incised supraorbital crests by narrow groove; anterior fontanelle narrow, tapering anteriorly, with rounded anterior and posterior margins; posterior fontanelle nar- row and constricted at about mid-length; foramen for anterior cerebral vein on vertical with dorsal rim of optic nerve foramen and on line with foramina for preorbital and orbitonasal canals (Fig. 4b); jugal arches very slender (Fig. 4c); basal and nasal plates moderately narrow (Fig. 4d). Scapulocoracoids moderately expanded (Fig. 5; and Fig. 13c McEachran and Compagno 1982), with postdorsal and anterior fontanelles little expanded, those of female more expanded than those of male (Table 2); scapular process directed dorsomedially; anterodorsal margin (anterior to rear corner) straight to slightly concave, rear corner little elevated, if at all; posterodorsal margin concave and steeply sloping to metacondyle. Pelvic girdle with relatively slender puboischiadic bar, nearly straight anteriorly; strongly concave posteriorly, more deeply concave in males than in females (Fig. 6a, b); prepelvic processes short and oriented laterally; illiac processes long, strong- ly curved and directed anteromedially; two obturator foramina. VOLUME 97, NUMBER 4 WS i b Fig. 6. Neoraja carolinensis pelvic girdles: a, ISH 3650/79, 285 mm TL (Holotype) adolescent male; b, USNM 265014, 225 mm (Paratype) juvenile female. Drawn from radiographs. Etymology.—Named after the type-locality, the Carolinian Province of the western North Atlantic. Remarks. — Neoraja carolinensis is very similar to its two congeners, N. caerulea and N. stehmanni, and to Breviraja africana in external morphology, squamation, coloration, and skeletal structure but differs from them in the several characters. In N. caerulea, N. stehmanni and B. africana the anterior lobe of the pelvic fin is about 70 to 80% of the length of the posterior lobe (lobes are about equal in length in N. carolinensis). Neroraja caerulea has 47 to 55 (x = 50) and B. africana 45 to 49 (k = 47) tooth rows in the upper jaw (N. carolinensis has 40 to 46 (x = 43). Both N. caerulea and N. stehmanni have 6 to 9 distinct cross-bars on dorsal surface of tail (N. carolinensis either lacks or has indistinct cross-bars on tail). Abdominal region of N. caerulea, N. stehmanni and B. africana is light colored or light with dark blotches (N. carolinensis has a uniformly dark abdominal region). Neoraja carolinensis also differs from its two congeners in several aspects of its neurocranial and scapulocoracoid structure (McEachran and Compagno 1982). Neoraja caerulea and N. stehmanni neurocrania possess shorter rostral shafts, shorter rostral appendices and less laterally expanded nasal capsules. The scapulocoracoids of N. caerulea and N. stehmanni possess an elevated rear corner and a more concave dorsal margin than that of N. carolinensis. Material examined in the course of this study necessitates several changes in McEachran and Compagno’s description of NV. (Neoraja). McEachran and Com- pagno (1982) stated, that species of NV. (Neoraja) possess a single row of thorns along the proximal midline of the tail, that the rostral shaft of NV. sp. nov. (=N. carolinensis) is distally segmented and that the anterodorsal margin of the scap- ulocoracoids is concave with the rear corner elevated. However, the holotype and larger paratypes of N. carolinensis possess a narrow band of two or three irregular rows of thorns on the proximal aspect of the tail unlike the smaller specimens or 734 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the three other species. Staining one of the dissected neurocrania of N. carolinensis revealed that the rostral shaft is not distally segmented. Apparent segmentation of the rostral shaft in the previous study is an artifact of dissection. The rostral shafts of N. caerulea and N. stehmanni are probably also unsegmented. While the scapulocoracoids of N. caerulea and N. stehmanni possess a concave anterodorsal margin and an elevated rear corner, those of N. carolinensis and B. africana are straight to slightly concave, without or at most with a slightly elevated rear corner. Key to Species of Neoraja and Breviraja africana 1. Anterior and posterior pelvic lobes of about equal length, dark cross-bars on dorsal surface of tail indistinct to absent, abdomen uniformly grayish- brown and marked off from light interbranchial and cloacal regions cad lace sheet Goo Wlghatte: Gees 4 aE ee eA esis one N. carolinensis — Anterior pelvic lobe at most 70 to 80% of posterior pelvic lobe length, dark cross-bars on dorsal surface of tail usually distinct, abdomen uni- formly light colored to light with dark blotches but not uniformly dark colored nor marked off from interbranchial and cloacal regions ....... My, 2. Dorsal surface of disc and pelvic fins grayish with irregular pattern of dark blotches and light spots, 38 to 44 tooth rows in upper jaw ... N. stehmanni — Dorsal surface of disc and pelvic fins plain brown to bluish-violet, without blotches;/45° to SS: tooth rows insupperyaw 4). 420 eee 3 3. Dorsal surface of disc and pelvic fins bluish-violet, dorsal side of tail light colored, dark cross-bars on dorsal surface of tail distinct ...... N. caerulea — Dorsal surface of disc, pelvic fins and tail uniformly brown (faint bluish shade centrally when newly dead), dark cross-bars on dorsal surface of LATING ISTITICE cyeting® gerbe hb iAhW ue 5 mth a? AE pls ha Ws dae B. africana Distribution.— The three species of Neoraja and Breviraja africana are allopat- rically distributed but occur in similar habitats. Neoraja carolinensis was captured along the continental slope off the southeastern United States, 34° 22’N to 29°11’N, at 695 to 1010 m, at temperatures of 4.18° to 6.09°C. Neoraja caerulea was captured on the slopes of the Rockall Basin, west of Scotland and Ireland, 61°06’N to 55°44’N, at 600 to 1262 m, at temperatures of 6.41° to 9.10°C (Stehmann 1976). Neoraja stehmanni was captured along the continental slope off the west coast of South Africa, 33°53.7’S to ca. 34°S, at 292 to 640 m, at temperatures of 5.55° to 5.70°C (Hulley 1972). Another very small specimen of N. stehmanni reported from off northern Mauritania, 22°50'N, at 1490 to 1620 m, by Golovan (1978) probably represents another species (Stehmann, in press). Breviraja africana was captured on the continental slope off Gabon, 2°41'S to 3°25’S, at 900 to 1030 m, at temperatures of 4.35° to 4.66°C (Stehmann and Seret 1983). Acknowledgments Daniel M. Cohen kindly furnished two specimens of the new species. The remaining four specimens were collected on cruise 213-3 of the German FRV Anton Dohrn, Project MFE 0319.7, supported by the Ministry of Research and Technology of the Federal Republic of Germany. Gudrun Schulze (ISH) radio- graphed and photographed specimens and prepared Figs. 1, 2, 3, 6. Janice D. VOLUME 97, NUMBER 4 : 735 Fechhelm prepared Figs. 4 and 5. This study was supported in part by grants to JDM from the National Science Foundation, Grant nos. DEB 78-11217 and DEB 82-04661. Literature Cited Bigelow, H. B., and W. C. Schroeder. 1953. Sawfishes, guitarfishes, skates and rays. Pp. 1-514. In J. Tee Van, ed., Fishes of the Western North Atlantic, Memoirs Sears Foundation for Marine Research 1(2):1-588. Golovan, G. A. 1978. Composition and distribution of the ichthyofauna of the continental slope of North-Western Africa. Jn Taxonomy and ecology of the deep-sea fishes. — Trudy P. P. Shirshov Institut Okeanologii Moscow /11:195-258, figs. 1-7 [in Russian, English summary]. Hubbs, C. L., and R. Ishiyama. 1968. Methods for the taxonomic study and description of skates (Rajidae).—Copeia 1968 (3):483-491, 1 fig. Hulley, P. A. 1972. A new species of southern African brevirajid skate (Chondrichthyes, Batoidei, Rajidae).— Annals of the South African Museum 60:253-263. McEachran, J. D. 1984. Anatomical investigations of the New Zealand skates Bathyraja asperula and B. spinifera, with an evaluation of their classification within Rajoidei (Chondrichthyes). — Copeia 1984(1):45-58. , and L. J. V.Compagno. 1979. A further description of Gurgesiella furvescens with comments on the interrelationships of Gurgesiellidae and Pseudorajidae (Pisces, Rajoidei).—Bulletin of Marine Science 29:530-553. , and . 1982. Interrelationships of and within Breviraja based on anatomical structures (Pisces: Rajoidei).— Bulletin of Marine Science 32:399—425. Stehmann, M. 1976. Breviraja caerulea spec. nov. (Elasmobranchii, Batoidea, Rajidae); eine neue archibenthale Rochenart und zugleich ein Erstnachweis ihrer Gattung im Nordostatlantik. — Archiv fiir Fischereiwissenschaft 27:97-114. 1980. Tiefenfischerei am norwestatlantischen Kontinentalabhang.—Informationen fur die Fischwirtschaft 27 (1):4—10, 1 fig. , and B. Seret. 1983. A new species of deep-water skate, Breviraja africana, sp. n. (Pisces, Batoidea, Rajidae), from the Eastern Central Atlantic slope, and remarks on the taxonomic status of Breviraja Bigelow & Schroeder, 1948.—Bulletin du Muséum National d’Histoire Naturelle Paris (4° ser) 5 (sect. A. No.3):903-925. . [In press]. Rajidae. In J. C. Quero, A. Post, and L. Saldanha, eds., Check-list of the fishes of the eastern tropical Atlantic. (JDM) Department of Wildlife and Fisheries Sciences and Department of Oceanography, Texas A&M University, College Station, Texas 77843; (MS) Ich- thyologie, Institut fiir Seefischerei, Bundesforschungsanstalt fur Fischerei, Martin- Luther-King Platz 3, D-2000 Hamburg 13, Federal Republic of Germany. ‘ PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 736-740 A HAMERKOP FROM THE EARLY PLIOCENE OF SOUTH AFRICA (AVES: SCOPIDAE) Storrs L. Olson Abstract.—Two bones from early Pliocene deposits at Langebaanweg, south- western Cape Province, South Africa, constitute the first fossil record of the family Scopidae. These are described as a new species, Scopus xenopus, that was larger than the living Hamerkop, S. umbretta, and that had a foot structure possibly indicating that the fossil species was better adapted for swimming. The Hamerkop, or Hammerheaded Stork, Scopus umbretta, is the sole living member of the family Scopidae. The exact systematic position of this family has long been disputed (Kahl 1967), although it is generally conceded that Scopus is probably related to storks (Ciconiidae) and the Shoebill (Balaenicipitidae), with which it is associated in all modern classifications. Scopus umbretta is widely distributed through sub-Saharan Africa and also occurs in Madagascar and south- western Arabia (Snow 1978). Up to the present, there has been no fossil record for the Scopidae (Brodkorb 1963). This has now changed with my discovery at the South African Museum of two bones of a new species from the early Pliocene at Langebaanweg, in southwestern Cape Province, South Africa. The deposits at Langebaanweg are renowned for the number of vertebrate remains they have yielded, including a great diversity of birds (Rich 1980). A variety of ecological conditions existed in the vicinity of Langebaanweg in the Pliocene (Hendey 1981la, b; 1982), resulting in remains of freshwater, marine, and terrestrial organisms being deposited. The geology and chronology of the Langebaanweg succession have been detailed by Hendey (1981b). Order Ciconiformes Bonaparte, 1854 The content and relationships of this polyphyletic order are still being actively debated. One of the families traditionally included here, the flamingos (Phoeni- copteridae), has been shown conclusively to belong in the Charadriiformes near the Recurvirostridae (Olson and Feduccia 1980). The ibises and spoonbills (Pla- taleidae) seem to be more closely related to the Charadriiformes and Gruiformes than to storks (Olson 1979), a conclusion that has received strong support from the fossil record (Peters 1983). On the other hand, the vultures of the families Vulturidae (=Cathartidae) and Teratornithidae, currently placed with the hawks in the Accipitriformes, are probably best included in or near the Ciconiiformes (Ligon 1967; Olson, in press). Thus, at this point it is neither possible nor prudent to provide a diagnosis of the order Ciconiiformes. Nevertheless, Scopus would surely be included in the same ordinal level taxon as Ciconia, and as the ordinal name Ciconiiformes has priority over that of any of the groups that might possibly be included with the storks and allies (e.g., Pelecaniformes), the Scopidae would belong in the Ciconiiformes regardless of its ultimate composition. VOLUME 97, NUMBER 4 U3 Table 1.—Measurements (mm) of fossil and living species of Scopus. S. umbretta S. xenopus i= on... Ranee (Mean TARSOMETATARSUS Distal width through trochleae 9.2 6 8.2-8.7 8.4 Shaft width at proximal end of distal foramen 6.6 6 6.1-6.3 6.2 Width of middle trochlea 3.3 6 2.9-3.3 3.1 Shaft width 25 mm proximal to middle trochlea 3.6 Om 3.0-3.4 33.3) CORACOID Distance from humeral edge of glenoid facet to sternal lip of scapular facet 8.8 7 7.2-8.7 8.1 Shaft width just sternal to procoracoid process 5.8 7 4.5-4.8 4.7 Family Scopidae Bonaparte, 1853 The fossil specimens can be referred to the Scopidae on the basis of the following characters: (1) distal end of tarsometatarsus essentially similar to that in the Ciconiidae but much smaller than in any known living or fossil stork; coracoid with (2) distinct ovoid furcular facet, (3) extensive pneumatization, (4) lack of excavation on the ventrolateral surface between the head and the glenoid facet, and (5) no procoracoid foramen. Character 2 separates the Scopidae from all avian taxa except the Balaenicipitidae and Pelecaniformes (within which the Phaethon- tidae are exceptional in lacking this character and the modern Fregatidae have it obliterated by fusion with the clavicle). Character 3 separates the Scopidae from the Phalacrocoracidae and Anhingidae, character 4 from the Pelecanidae and Sulidae, and character 5 from the Balaenicipitidae. Genus Scopus Brisson 1760 Scopus xenopus, new species Figs. 1-2 Holotype. — Distal end of left tarsometatarsus with posterior parts of inner and outer trochleae abraded, SAM-PQ-L43396. Paratype. —Scapular half of right coracoid lacking part of the head and the tip of the procoracoid process, SAM-PQ-L28440S. Type-locality and horizon.—“‘E”’ Quarry, Langebaanweg, southwestern Cape Province, South Africa. Varswater Formation (early Pliocene). The holotype is either from bed 3aS of the Pelletal Phosphate Member or from the Quartzose Sand Member, whereas the paratype is definitely from the latter (see Hendey 1981b). Measurements.—See Table 1. Etymology.—Greek, xenos, strange, and pous, foot; in reference both to the peculiarities of the tarsometatarsus and to the fact that frogs and tadpoles of the genus Xenopus Wagler are one of the principal food items of the living species Scopus umbretta (Kahl 1967). Diagnosis.—Somewhat larger than Scopus umbretta (Table 1). Inner trochlea 738 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Stereo pairs of distal end of left tarsometatarsus, plantar aspect. A, Scopus xenopus new species, holotype, SAM-PQ-L43396; B, Scopus umbretta, USNM 18898. Figures are 3x. (II) of tarsometatarsus more distally produced, almost level with the middle trochlea (III), and consequently with a noticeably longer shaft; scar for hallux larger, extending farther distally, with a more distinct, rounded proximal articular surface; middle trochlea angled more medially; distal foramen relatively larger. Furcular facet of coracoid narrower and more elongate. Discussion.—In the preliminary sorting of avian remains from Langebaanweg (Rich 1980), the holotype of Scopus xenopus was identified as a pelecaniform of uncertain affinity; as such it was listed under the Pelecaniformes by Hendey (1981b:48) as ““Fam., gen. and sp. not det.’’ This determination was not unjustified and carries interesting functional and systematic implications. The tarsometatar- sus of Scopus xenopus resembles that in the Pelecaniformes because of the more distally situated inner trochlea that is nearly even with the middle trochlea. Such a condition obtains in most of the Pelecaniformes, in some of which the inner trochlea extends farther distally than the middle one. This adaptation is associated with swimming behavior; increasing specialization for terrestrial locomotion in- volves elevation of the inner trochlea rather than distal displacement. The living Hamerkop is an aquatic “wading” bird that typically walks in shallow water while foraging (Kahl 1967). The scant fossil evidence suggests the possibility that Scopus xenopus may have been better adapted for swimming locomotion and thus may have had different habits from S. umbretta. If so, S. xenopus and S. umbretta may not have had an ancestral-descendent relationship. That the tarsometatarsus of S. xenopus bears a resemblance to that in the Pelecaniformes may have phylogenetic significance. Cottam (1957) noted many VOLUME 97, NUMBER 4 739 Fig. 2. Fossil and recent species of Scopus: right coracoid, ventral aspect (A, B); distal end of left tarsometatarsus, cranial aspect (C, D). A, Scopus xenopus new species, paratype SAM-PQ-L28440S; B, Scopus umbretta, TM 42863; C, Scopus xenopus, new species, holotype SAM-PQ-L43396; D, Scopus umbretta, USNM 18898. Figures are 2 x. striking similarities between the Shoebill (Balaeniceps rex) and the Pelecani- formes; a number of these same characters are also present in Scopus. The highly derived order Pelecaniformes must have had its origins in some less specialized group. Such a group may have included the ancestors of Scopus and Balaeniceps, a possibility that merits detailed investigation. Acknowledgments I am grateful to T. M. Crowe and the FitzPatrick Institute, University of Cape Town, for the opportunity to work in South Africa, and to P. Haarhoff and Q. B. Hendey for providing facilities at the South African Museum (SAM) and for much information and assistance. I am indebted to Clive Booth (SAM) for the photo- graphs and to R. K. Brooke, P. Haarhoff, P. V. Rich, and D. W. Steadman for comments on the manuscript. Comparative material examined was from the collections of the SAM, the National Museum of Natural History, Smithsonian Institution (USNM), and the Transvaal Museum (TM). Literature Cited Brodkorb, P. 1963. Catalogue of Fossil Birds. Part 1 (Archaeopterygiformes through Ardeiformes).— Bulletin of the Florida State Museum, Biological Sciences 7:179-293. Cottam, P. 1957. The pelecaniform characters of the skeleton of the Shoe-bill Stork Balaeniceps rex.— Bulletin of the British Museum (Natural History) Zoology 5(3):51-71. Hendey, Q. B. 198la. Geological succession at Langebaanweg, Cape Province, and global events of the late Tertiary.—South African Journal of Science 77:33-38. 1981b. Palaeoecology of the late Tertiary fossil occurrences in “E’ Quarry, Langebaanweg, 740 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON South Africa, and a reinterpretation of their geological context.— Annals of the South African Museum 84:1-104. 1982. Langebaanweg. A record of past life.—South African Museum: Cape Town. 71 pp. Kahl, M. P. 1967. Observations on the behaviour of the Hamerkop Scopus umbretta in Uganda.— Ibis 109(1):25-32. Ligon, J. D. 1967. Relationships of the cathartid vultures.—Occasional Papers of the Museum of Zoology, University of Michigan 651:1—26. Olson, S. L. 1979. Multiple origins of the Ciconiiformes.— Proceedings of the Colonial Waterbird Group 1978. Pp. 165-170. —. [In press.] The fossil record of birds. Jn D. Farner, J. King, and K. C. Parkes, eds., Avian Biology. Vol. 8.—Academic Press: New York. ,and A. Feduccia. 1980. Relationships and evolution of flamingos (Aves: Phoenicopteridae). — Smithsonian Contributions to Zoology 316:1-73. Peters, D. S. 1983. Die ““Schnepfenralle’” Rynchaeites messelensis Wittich ist ein Ibis.—Journal fiir Ornithologie 124(1):1-27. Rich, P. V. 1980. Preliminary report on the fossil avian remains from late Tertiary sediments at Langebaanweg, (Cape Province), South Africa.—South African Journal of Science 76:166—170. Snow, D., editor. 1978. An atlas of speciation in African non-passerine birds. British Museum (Natural History): London. 390 pp. FitzPatrick Institute, University of Cape Town, Rondebosch 7700, South Af- rica. Permanent address: Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 USA. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 741-743 EVIDENCE OF A LARGE ALBATROSS IN THE MIOCENE OF ARGENTINA (AVES: DIOMEDEIDAE) Storrs L. Olson Abstract.—A single pedal phalanx from early late Miocene marine deposits in Argentina provides the first Tertiary record of an albatross (Diomedeidae, genus indet.) from South America. The specimen is from a species considerably larger than represented by the two Tertiary albatross fossils hitherto reported from the Southern Hemisphere. Although the modern forms of albatrosses (Diomedeidae, Procellariiformes) are concentrated mainly in the Southern Hemisphere, there is a much better fossil record for the family in the Northern Hemisphere. Tertiary albatrosses have been reported from California, Florida, England, and France, with numerous unpub- lished specimens being known from Oregon, North Carolina, and South Carolina (Brodkorb 1963; Olson, in press, a). Only two albatross fossils have been reported from the Southern Hemisphere: an incomplete rostrum from the late Miocene of Victoria, Australia, the holotype of Diomedea thyridata Wilkinson (1969); and the proximal end of a tarsometatarsus from the early Pliocene at Langebaanweg, Cape Province, South Africa (Olson 1983; in press, b). Both of these specimens are from relatively small species, the size of D. melanophris. A third fossil now documents the former existence ofa larger species of albatross in the Tertiary of the southern oceans. The specimen consists of the left basal phalanx of the fourth pedal digit (vertebrate paleontological collections of the National Museum of Natural History, Smithsonian Institution, USNM 336381). It was obtained at Punta Ninfas, on the south side of the entrance of Golfo Nuevo, opposite Peninsula Valdéz, Chubut Province, Argentina. The fossil was derived from beds of so-called Rionegrense age at a level 50 m (165 feet) above the present base of the cliff. Various outcrops in northeastern Argentina attributed to the Rionegrense because of similarities in lithology may not actually be contempo- raneous, however, due to their discontinuous nature and the frequency of Neogene marine transgressions in this area (Zinsmeister et a/. 1981). Fortunately, a volcanic tuff near the top of the Rionegrense horizon at the site of collection of the fossil is the source of the only radioisotopic age determination yet obtained for any of the Rionegrense beds. The mean of three glass concentrates from this tuff dated by the *°K-*°Ar method was 9.41 x 10° years, which correlates “with the late Miocene Tortonian marine stage in Europe and the Chasicoan Land Mammal Age in South America”’ (Zinsmeister ef al. 1981). The large size, very elongate, slender proportions, and occurrence in a marine context, identify the fossil with the Diomedeidae. The specimen (Fig. 1) measures 53 mm from the dorsal margin of the proximal articular surface to the distal end; the width and depth of the shaft at the midpoint are 4.4 x 5.7 mm. Thus, in overall size the fossil is larger than any of the living albatrosses except D. exulans (56.0-62.0 mm, n = 4) and D. epomophora (57.0 mm; n = 1), with the closest 742 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Left pedal phalanx (p 1, d IV) of Diomedeidae gen. et sp. indet. from the Miocene of Peninsula Valdéz, Argentina (a, c); compared with Diomedea albatrus USNM 567025 (6, d). A, B, lateral aspect; C, D, ventral aspect. All figures natural size. approach among lesser species being D. albatrus (41.0—45.0 mm; n = 3) and D. cauta (48.0 mm; n = 1). Postcranial specimens of D. leptorhyncha (=D. irrorata auct.) were not available for comparison. Because the fossil is more laterally compressed, it appears much more slender than in any of the species of albatrosses examined, including the two species of Phoebetria. Thus the affinities of the specimen within the family cannot be refined. Nevertheless, it constitutes the first Tertiary record of an albatross from South America. Acknowledgments I am grateful to Phyllis Dean and David W. Steadman for information and comments on the manuscript. The photographs are by Victor E. Krantz, Smith- sonian Institution. _ Literature Cited Brodkorb, P. 1963. Catalogue of Fossil Birds. Part 1 (Archaeopterygiformes through Ardeiformes).— Bulletin of the Florida State Museum, Biological Sciences 7 (4):179-293. Olson, S. L. 1983. Fossil seabirds and changing marine environments in the late Tertiary of South Africa.—South African Journal of Science 79(10):399—402. —.. [In press, a]. The fossil record of birds. Jn J. King, D. Farner, and K. Parkes, eds., Avian Biology, volume 8.—New York: Academic Press. —. [In press, b]. Early Pliocene Procellariiformes (Aves) from Langebaanweg, Southwestern Cape Province, South Africa.— Annals of the South African Museum. VOLUME 97, NUMBER 4 743 Wilkinson, H. E. 1969. Description of an Upper Miocene albatross from Beaumaris, Victoria, Australia, and a review of the fossil Diomedeidae.— Memoirs of the National Museum of Victoria 29:41-51. Zinsmeister, W. J., L. G. Marshall, R. E. Drake, and G. H. Curtis. 1981. First radioisotope (Potas- sium-Argon) age of marine Neogene Rionegro beds in northeastern Patagonia, Argentina. — Science 212:440. Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 744-747 TANAOCHELES STENOCHILUS, A NEW GENUS AND SPECIES OF CRAB FROM GUAM, MARIANA ISLANDS (BRACHYURA: XANTHIDAE) Roy K. Kropp Abstract.—A new xanthid crab, Tanaocheles stenochilus, described from Apra Harbor, Guam, Mariana Islands, is the type-species of a new genus of crab which lives in association with the scleractinian coral Leptoseris gardineri. Three specimens of the xanthid crab subfamily Trapeziinae were collected among colonies of the scleractinian coral Leptoseris gardineri van der Horst, 1921, at moderate depth (27 m) from Apra Harbor, Guam. These specimens are not referable to any of the known genera of the subfamily and are described herein as new. Crabs were measured to the nearest 0.1 mm with dial calipers. Measurements are given as carapace length x carapace width. Specimens are deposited in the National Museum of Natural History, Smithsonian Institution (USNM) and the Allan Hancock Foundation (AHF). Family Xanthidae MacLeay, 1838 Subfamily Trapeziinae A. Milne Edwards, 1862 Tanaocheles, new genus Diagnosis. —Carapace wider than long, convex in both directions; posterolateral border longer than anterolateral border, latter with spines. Eye large, not wholly contained within cuplike orbit. Front broad, sinuous, without teeth or spines; carapace sharply deflected near frontal margin. Chelipeds unequal, but morpho- logically similar, very long, relatively thin; fingers of chelae short, spoon-tipped. Walking legs long, thin; dactylus of each long, with many spiniform and stout setae, claw bladelike. Abdomen of male with 7 distinct segments. Etymology.— From the Greek “‘tanao,”’ outstretched, in combination with the Greek “‘chele,”’ claw; in reference to the very long chelipeds characteristic of this genus. Gender feminine. Type-species.— Tanaocheles stenochilus, new species, by monotypy. Remarks.— The nine genera of the Trapeziinae can be separated into two groups on the basis of the number of distinct segments of the male abdomen. Tanaocheles is included among the genera in which the male abdomen has seven distinct segments. These genera are Tetralia Dana, 1851; Ectaesthesius Rathbun, 1898; Calocarcinus Calman, 1909; and Philippicarcinus Garth and Kim, 1983. Tanao- cheles and Philippicarcinus both differ from the other three genera by having approximately hexagonal carapace outlines. Tanaocheles differs from all four genera by having long, very slender chelipeds, with meri much longer than broad, and spoon-tipped fingers. VOLUME 97, NUMBER 4 745 Tanaocheles stenochilus, new species Fig. 1 Holotype.— Male, 4.5 mm X 6.2 mm; Apra Harbor (Western Shoals), Guam, Mariana Islands; on Leptoseris gardineri; 27 m; 23 Jul 1981; Coll. V. Tyndzik; USNM 210636. Paratypes.—Female, 5.6 mm X 7.3 mm; USNM 210637. Male, 3.2 mm x 4.3 mm; AHF 819. Both same collection data as holotype. Description. —Carapace wider than long; smooth, regions poorly marked, with a few scattered plumose setae. Lateral margins rounded; spineless posterolateral part about twice as long as anterolateral part; anterolateral part bearing two spines, posterior sharp, anterior blunt, rounded. Orbital margin raised. Front broad, slightly sinuous, without median notch; carapace sharply deflected at frontal mar- gin then produced anteriorly forming narrow, flattened rim. Inner suborbital angle not meeting front, second antenna not excluded from orbit. Chelipeds unequal, very long, thin; merus and carpus of both morphologically similar, major thicker. Merus long, smooth, rounded, with no distinct margins; with 3 sharp anterior spines, 2 at proximal end, one near distal end; no posterior spines. Carpus smooth, with a sharp or rounded anterior spine. Manus long, rounded, without spines. Major manus gradually thickening distally, much thicker than minor; fingers short, dark, spoon-tipped; outer edges of occlusal surfaces with 3 broad, sharply edged teeth, inner edges rounded, without teeth, leaving space when fingers close. Minor manus long, thin, of uniform thickness; fingers short, spoon-tipped, occlusal surfaces morphologically similar to those of major, but with 2 low teeth. Carpus and manus with interlocking knobs on upper and lower surfaces at joint. Walking legs long, thin, with scattered long plumose and nonplumose setae. Merus of legs 1—3 with anterior margin minutely denticulate, sometimes with spines; merus of leg 4 smooth. Carpus of legs. 1—4 with 2, 2, 1, 0 anterodistal spines respectively; anterodistal margin produced into knob fitting into socket on propodus. Propodus long, thickening slightly distally. Dactylus long, claw broad, bladelike; anterior margin with 2 rows of stout, spiniform setae, distal setae largest; posterior margin lined with many long, spiniform and short, stout setae. Propodus- dactylus joint with locking mechanism on posterior surface; distal margin of propodus produced into broad flange fitting into groove made by raised knob on proximal part of dactylus. Color notes.—Carapace and chelipeds reddish blond, fingers of chelae dark brown. Walking legs white with scattered orange blotches and spots or broken orange lines. Remarks.—The female paratype differs from the holotype by having a third small, sharp spine immediately posterior to the large spine on the anterolateral border of the carapace and by having 3 sharp proximal spines on the anterior ~ surface of the merus of the cheliped. The male paratype has a sharp anterior spine on the anterolateral border of the carapace instead ofa blunt spine, and a proximal spine on the posterior surface of the merus of the cheliped. Etymology. — From the Greek “‘stenos,”’ narrow, in combination with the Greek “cheilos,” a lip or rim; in reference to the narrow rim at the frontal margin of the carapace. 746 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Tanaocheles stenochilus, male holotype, carapace length 4.5 mm: a, Carapace and chelipeds; b-e, Walking legs 1-4; f, Dactylus of leg 1; g, Abdomen; h, Gonopod. b-e, g slightly enlarged; f| h enlarged. Habitat.— Found free-living in association with the living coral Leptoseris gar- dineri. Distribution.— At present known only from the type-locality. Acknowledgments I thank Vaughn Tyndzik for collecting the new crab and Richard Randall for identifying the host coral. Thanks also to L. G. Eldredge and R. B. Manning for VOLUME 97, NUMBER 4 747 comments on the manuscript, and to Lilly Manning for help with the figures. This work was supported in part by a grant to G. J. Vermeij by the Biological Ocean- ography Program of the National Science Foundation. This paper is Contribution number 200 from The University of Guam Marine Laboratory. Literature Cited Calman, W. T. 1909. Ona new crab taken from a deep-sea telegraph-cable in the Indian Ocean. — Annals and Magazine of Natural History (8) 3:30-33. Dana, J.D. 1851. On the genera Trapezia and Tetralia.— American Journal of Science and Arts (2) 11:223-224. Garth, J. S., and H. S. Kim. 1983. Crabs of the family Xanthidae (Crustacea: Brachyura) from the Philippine Islands and adjacent waters based largely on collections of the U.S. Fish Commission steamer Albatross in 1908-1909.—Journal of Natural History 17:663-729. MacLeay, W.S. 1838. On the brachyurous decapod Crustacea brought from the Cape by Dr. Smith. In Illustrations of the Annulosa of South Africa; being a portion of the objects of natural history chiefly collected during an expedition into the interior of South Africa, under the direction of Dr. Andrew Smith, in the years 1834, 1835, and 1836; fitted out by ““The Cape of Good Hope Association for Exploring Central Africa,” pp. 53-71. London. Milne Edwards, A. 1862. Monographie des Crustacés fossiles de la famille Canceriens.— Annales des Sciences Naturelles, Zoologie, series 4, 18:31-85. Rathbun, M. J. 1898. The Brachyura collected by the U.S. Fish Commission steamer Albatross on the voyage from Norfolk, Virginia to San Francisco, California, 1887-1888.— Proceedings of the United States National Museum 21:567-616. Department of Zoology, University of Maryland, College Park, Maryland 20742. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 748-760 PROTEOCEPHALIDEAN CESTODES FROM VENEZUELAN SILURIFORM FISHES, WITH A REVISED CLASSIFICATION OF THE MONTICELLITIDAE Daniel R. Brooks and Graciela Rasmussen Abstract. — Specimens of proteocephalidean cestodes collected in siluriform fish- es from the delta of the Orinoco River represent five species: Proteocephalus cf. kyukyu, Nomimoscolex kaparari, Goezeella siluri, Amphoteromorphus praepu- tialis, and Choanoscolex abscissus. The last species is reported in new collections for the first time since its description in 1896; it possesses cortical gonads and is thus a monticelliid. The delta of the Orinoco is a new locality for all five species. Megalodoras irwini is a new host for P. kyukyu, Brachyplatystoma vaillanti is a new host for N. kaparari, Pseudoplatystoma fasciatum is a new host for C. abscis- sus, and Pseudocetopsis othonops is a new host for G. siluri and A. praeputialis. Examination of these specimens allowed character analysis for a suite of mor- phological characters leading to a phylogenetic hypothesis for the monticelliids which is superior to the one proposed earlier by the first author. Classification of the cestode order Proteocephalidea has been hampered by a lack of critical character analysis and by incomplete information regarding many taxa. Brooks (1978) summarized published data used to classify proteocephali- deans based on a phylogenetic systematic analysis. In that study, the relationships of the genera comprising the family Monticelliidae LaRue, 1911, were marked by extensive parallel evolution. Postulates of parallel evolution stemming from phylogenetic analyses at the supra-specific level may be the result of (1) real parallel evolution, (2) para- and polyphyletic groupings or (3) inadequate character anal- ysis. Phylogenetic trees produced by cladistic analysis, and the classifications derived from them, are explicit hypotheses predicting certain distributions of character- istics among species. They can be tested by finding new characters and determining their distributions. If the new data are incongruent with previous observations, they may force a modification of the original phylogenetic hypothesis. This study represents the first test of Brooks’ (1978) hypothesis of monticelliid relationships. Collection of proteocephalidean material from the delta of the Orinoco River during the winter of 1978 has permitted re-evaluation of some characters. This, coupled with extension of the analysis to the species level to test the monophyly of the generic groupings, has reduced the apparent parallel evolution shown by the monticelliids. Materials and Methods Hosts were collected by hook and line or trawl and examined immediately for parasites. Cestodes were removed from hosts, studied alive, then relaxed in river water, fixed in steaming AFA and stored in 70% ethanol. Whole mounts were VOLUME 97, NUMBER 4 749 prepared by hydrating specimens, staining them with Mayer’s hematoxylin, de- hydrating them, clearing them with methyl benzoate, and mounting them in Canada balsam. Some specimens were embedded in paraffin, serially cross sec- tioned at 8 wm, and stained with hematoxylin-eosin for study. Figures were drawn with the aid of a drawing tube. Results Proteocephalus cf. kyukyu Woodland, 1935 Host.— Megalodoras irwini Eigenmann (Siluriformes: Doradidae), new host. Locality.— Vicinity of Isla Tres Canos, delta of the Orinoco River, Venezuela, new locality. Site of Infection.— Anterior third of intestine. Deposition of Specimens.— University of Nebraska State Museum, No. 22455; James R. Adams Parasitology Collection, University of British Columbia. Woodland (1935) described Proteocephalus kyukyu from immature specimens found in the gut of Pseudodoras niger and P. brunnescens. His specimens exhibited simple scolices, a very long unsegmented portion of strobila, and a few segments at the posterior end. Even the posteriormost segments lacked formed reproductive organs although anlagen of ovaries and cirrus sacs were present. Freze (1965) considered this a species inquirenda. We found over 100 specimens in the intestine of one Megalodoras irwini, all of which agreed with Woodland’s (1935) description. The immaturity of the strobilae in specimens collected from the Amazon (Woodland) and Orinoco (pres- ent study) suggests that this might be a normal aspect of the biology of this species. The implication of such a thought is that P. kywkyu might be hyperapolytic, with maturity and reproduction occurring after the proglottids have left the strobila. However, we found no such detached proglottids in the infected host. Until more is known we concur with Freze and consider this a species inquirenda. Nomimoscolex kaparari Woodland, 1935 Host.—Brachyplatystoma vaillanti (Cuvier and Valenciennes) (Siluriformes: Pimelodidae), new host. Locality.— Vicinity of Isla Tres Canos, delta of the Orinoco River, Venezuela, new locality. Site of Infection.— Anterior third of intestine. Deposition of Specimens.— University of Nebraska State Museum No. 22451. We collected three immature proteocephalideans exhibiting scolices with four simple suckers with two muscular papillae, one on each side of the anterior margin. Woodland (1935) reported that N. kaparari possessed eight “unguiculate protru- sions,’ two on each sucker. They appear to be the same as structures found on the suckers of N. alovarius Brooks and Deardorff, 1980, a parasite of Pimelodus clarias in the Magdalena River of Colombia (see Brooks and Deardorff 1980). The papillae of NV. alovarius are positioned on the anterior portion of the suckers rather than on the margins of the anterior portions. Nomimoscolex kaparari has previously been reported in Brachyplatystoma filamentosum and Pseudoplaty- stoma tigrinum in the Amazon River. 750 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON int alee U8 Figs. 1-4. Goezeella siluri. 1, Scolex; 2, Mature proglottid; 3, Gravid proglottid; 4, Cross section of young gravid proglottid. Scale in millimeters. Goezeella siluri Fuhrmann, 1916 Figs. 1-4 Host.— Pseudocetopsis othonops (Eigenmann) (Siluriformes: Cetopsidae), new host; immature specimens in Brachyplatystoma vaillanti. Locality.— Vicinity of Los Castillos, of Isla Tres Canos, and of La Portuguesa, delta of the Orinoco River, Venezuela, new localities. Site of Infection. — Posterior fifth of intestine. Deposition of Specimens. — University of Nebraska State Museum, No. 22452, 22454; James R. Adams Parasitology Collection, University of British Columbia. Collection of a number of G. si/uri permitted examination of living worms and of sectioned material. By observing live specimens, we were able to confirm that the metascolex is formed by an expansion of the neck tissue posterior to the scolex. Contracted specimens do not show this clearly (e.g., see Brooks and Deardorff 1980). A vaginal sphincter was found in all proglottids, and the vagina was found to pass anterior or posterior to the cirrus sac. The single specimen reported by Brooks and Deardorff (1980) from Ageneiosus caucanus in Colombia had all vaginae passing anterior to the cirrus sac. Cross sections of G. siluri show cortical testes, ovaries, vitellaria, and uteri. The vitelline follicles converge towards the ventral midline of the proglottid, and the uterine wall is very thin (Fig. 4). VOLUME 97, NUMBER 4 751 eeeeige% 0! egesce wry 00S Figs. 5-8. Amphoteromorphus praeputialis. 5, Scolex; 6, Mature proglottid; 7, Gravid proglottid; 8, Cross section of young gravid proglottid. Scale in micrometers. Amphoteromorphus praeputialis Rego, Dos Santos and Silva, 1974 Figs. 5-8 Host.— Pseudocetopsis othonops, new host. Site of Infection.— Anterior fifth of intestine. Locality.— Vicinity of Los Castillos, of Isla Tres Canos, and of La Portuguesa, delta of the Orinoco River, Venezuela, new localities. Specimens Deposited.—University of Nebraska State Museum, No. 22449, 22453, 22457, 22458; James R. Adams Parasitology Collection, University of British Columbia. Our specimens agree well with the original description of this species in Cetopsis caecutiens (Licht.) from the Amazon River near Maicuru, Brazil. There appears to have been some confusion in labelling some of the diagrams in the original description (see Rego et al. 1974). Their figures 23-24, labelled as scolices of Monticellia siluris (=G. siluri), show contracted specimens with uniloculate suck- ers; G. siluri has biloculate suckers (see Fig. 1). Figure 25 by Rego et ai. is an unclear photomicrograph of a specimen called M. siluris, but the nature of the sucker faces cannot be determined. Finally, their figures 26-27 are labelled pho- tomicrographs of M. siluris, but judging from the shape of the uterus, they are specimens of A. praeputialis (see Figs. 6—7). The metascolex of A. praeputialis is formed by a proliferation of tissue around each sucker rather than by the expansion of the neck seen in G. si/uri (compare Figs. 1 and 5). In addition, all gonads except the vitellaria are medullary in A. praeputialis, and the uterus is very narrow (Figs. 6—7) and thick-walled (Fig. 6). eZ PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON [gtaw, es & So 32 20° @ e2o8 S9% & cE FeO 20S LORS B ay Oo (9) § “ @2908 8 a) 28 e® 3 S8oe s 0.958 LPs e) E 3 ° 3° t 11 Figs. 9-11. Choanoscolex abscissus. 9, Scolex; 10, Mature proglottid; 11, Gravid proglottid. Scale in micrometers. We examined 18 Pseudoceptopsis othonops from four different localities during this study. Both A. praeputialis and G. siluri were found in all four localities. Fifteen of the 18 fish (83%) were infected with at least one species of cestode. Nine of those 15 (60%) carried A. praeputialis and 12 of 15 (80%) carried G. siluri; 6 of 15 (40%) carried both species. Amphoteromorphus praeputialis always oc- curred in the anterior fifth of the intestine whereas G. si/uri always occurred in the posterior fifth of the intestine, regardless of the presence of other species. We found no other helminths in P. othonops. Based on our phylogenetic analysis (presented later), G. siluri appears to be a colonizer of P. othonops, while A. praeputialis exhibits an apparent coevolutionary relationship with its cetopsid hosts. Thus, it is surprising that G. si/uri occurred more often in P. othonops than did A. praeputialis. The small sample size precludes any stringent conclusions from being drawn, but this does illustrate the point that degree of coevolution cannot necessarily be drawn from host specificity or incidence data (see also Brooks 1979). Choanoscolex abscissus (Riggenbach, 1896) LaRue, 1911 Figs. 9-11 Host.— Pseudoplatystoma fasciatum (Linnaeus) (Siluriformes: Pimelodidae), new host. Locality.— Vicinity of Isla Tres Canos, delta of the Orinoco River, Venezuela, new locality. VOLUME 97, NUMBER 4 753 Proteocephalids Monticelliids Fig. 12. Evolutionary transformations in scolex suckers among proteocephalideans. Site of Infection.— Anterior third of intestine. Deposition of Specimens.— University of Nebraska State Museum, No. 22456. This species was described by Riggenbach (1896) from specimens collected in ““Silurus sp.” (probably a pimelodid) from the Paraguay River. To our knowledge, it has never been reported in new collections until now. Unfortunately, our spec- imens are only of marginal quality, so we will not attempt a redescription at this time. However, we have been able to make a number of observations which help us place this species in a more suitable taxonomic context. Riggenbach reported (1) approximately 100 testes per proglottid, and our specimens have 130-173 testes, (2) a large number of uterine diverticula, and our specimens show 70-80 total diverticula, (3) a flat but folliculate ovary, which our material confirms, (4) a genital pore in the anterior fourth of the proglottid, in general agreement with our measurements, which show the genital pore in the anterior 16—25% of mature proglottids and 23-25% of gravid proglottids, and (5) a metascolex, which our observations of living and relaxed fixed material show to be formed by an ex- pansion of the neck region, similar to that of G. si/uri. In addition, we were able to provide a few cross sections which provided very little detail beyond the recognition that the gonads are cortical. This necessitates removal of C. abscissus from the Proteocephalidae and placement of it in the Monticelliidae. Discussion The new data supplied by study of the present specimens permit re-interpre- tation of some of the characters listed by Brooks (1978), and the use of some new characters. 1. Metascolex (Brooks’ character 5). This study demonstrates that the monti- celliids exhibit at least two different types of metascolex. One type is exhibited 754 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Co a = 2 : © € » (A) Fig. 13. Evolutionary transformations in vitelline configuration, viewed in cross section, among monticelliid cestodes. by Goezeella siluriand Choanoscolex abscissus, in which the metascolex is formed by an expansion of the neck. The other type is exhibited by Amphoteromorphus praeputialis and its relatives. In the latter species, the metascolex is formed by proliferation of tissue around each sucker on the scolex. Within the Proteoce- phalidae, metascolices of the Corallobothriinae are formed by a general prolif- eration of scolex tissue around the base of the suckers. Thus, at least three different structures have all been called “‘the proteocephalidean metascolex.” Re-coding of this character eliminates three cases of parallel evolution from the cladogram depicted by Brooks (1978). 2. Uterine structure. Amphoteromorphus praeputialis possesses thick-walled, narrow uteri, which differ from those of most other proteocephalideans. Illustra- tions of Zygobothrium megacephalum and the other species of Amphoteromorphus suggest that this structure is representative of this small group of monticelliids. 3. Scolex papillae. Three species, Nomimoscolex kaparari, N. alovarius, and Myzophorus woodlandi, are known to possess papillae on the suckers or sucker margins of the scolex. For N. kaparari, there are two papillae on the margins of each sucker; N. alovarius also has two papillae per sucker, but they are situated on the suckers themselves; and M. woodlandi has four papillae on the margin of each sucker. . 4. Ovarian structure. Most monticelliids possess highly follicular ovaries (see Figs. 2, 6), which differ from those of other proteocephalideans. This includes those of the members of the Acanthotaeniinae, which have compact ovaries with digitiform lobes, and of the Corallobothriinae, whose compact ovaries may be lobulated. 5. Sucker structure. The primitive state for proteocephalidean scolex suckers is the uniloculate condition. In one proteocephalid lineage and in the monticellids, bi-, tri- and tetraloculate suckers have arisen. The transformation series for these changes is shown in Fig. 12. VOLUME 97, NUMBER 4 WS VY yey & wy Y VvYvY VYry Ywvyvyryr YYYY YY 4 5 omc <5 8 YY & iy 7 . RY sO 4 oe és YS ys Ss » ° 2 SS S YS VEO IIMS SM SII II SDT SSS TS ¢ 2 & Ry & YS SY & & of YS SS SY Ss YY SS MS YS SIT MYST SI SE wy Eo OF 1g ‘ IE EI IT 8 SS SES OSE GLE CF BVP IS GI SI VS SE By 7 wy yY Y iy YQ, / Ya 2) 7 Fig. 14. Cladogram depicting phylogenetic hypothesis for 34 species of monticelliids. Abbreviations for generic names include: N = Nomimoscolex; My = Myzophorus; E = Endorchis; O = Othinoscolex; Ep = Ephedrocephalus; R = Rudolphiella, A = Amphoteromorphus; Z = Zygobothrium; M = Monti- cellia; S = Spatulifer, C = Choanoscolex, G = Goezeella. Numbers accompanying slash marks on branches refer to derived traits (Synapomorphies) listed in Appendix. 6. Embryonation of eggs. None of the monticelliids which we examined have any evidence of embryonated eggs in utero. This is a condition found in at least some tetraphyllidean, trypanorhynch and lecanicephalan cestodes. The presence of embryonated eggs in utero may be of use in distinguishing some groups of cestodes. At present, not enough is known to draw conclusions, but we include the character for future reference. 7. Ovarian position in cross section (Brooks’ character 2). Primitively, the ovary is medullary. In some monticelliids, the ovary is partially cortical, and in others it is almost entirely cortical. 8. Vitelline configuration in cross section. There are four different basic con- figurations found in monticelliids. The plesiomorphic condition is derived from outgroup comparisons. These are shown in Fig. 13. 9. Testes number. The primitive state for monticelliids appears to be between 100 and 150 testes per proglottid based on both outgroup comparisons with other proteocephalideans and on functional outgroup analysis. Other states include: (1) 80-120, with a mean of 100; (2) 40-60; (3) 150-200; and (4) 200-400. 10. Position of vaginal opening relative to cirrus sac. The primitive condition among proteocephalideans is the presence of vaginal openings irregularly alter- nating between anterior and posterior to the cirrus sac. In some monticelliids, the vagina is reported to pass either only anterior or only posterior to the cirrus sac. The cladogram corresponding to the most parsimonious representation of the above data, and data previously recorded by Brooks (1978), for 34 species of monticelliids is shown in Fig. 14. The high rate of parallel evolution found by 756 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON [ One n, Fig. 15. Cladogram from Fig. 14 with letter codes indicating host identity superimposed. Refer to text for host names. Brooks (1978) can be seen to have been due to the para- and polyphyletic nature of some of the generic groupings, in particular Nomimoscolex, Myzophorus, Am- photeromorphus, and Monticellia. An indicator of the degree to which characters showing parallel evolution have been used to construct a phylogeny is the con- sistency index (CI) (Kluge and Farris 1969; Farris 1970). The higher the CI (1.0 iS maximal), the lower the parallel evolution. For Fig. 14, the CI is 0.57, which is average for phylogenetic analyses of various kinds of organisms (see Mickevich 1982). However, virtually all of the parallel evolution can be attributed to two characters, vaginal position and testes number. If we remove those characters, the CI for the tree is 0.83, indicating very little parallel evolution in the majority of other characters. If the results of this study are corroborated by future work, it will be necessary to revise the taxonomy of the monticelliids considerably to make it consistent with their phylogeny. However, until we have made a thorough examination of available material, we refrain from making such nomenclatural changes. Host-Parasite Relationships The search for monticelliids in neotropical siluriforms cannot be termed ex- tensive. On the other hand, recent collections in Brazil (Rego, Dos Santos, and Silva 1974), Colombia (Brooks and Deardorff 1980) and Venezuela (present re- port) have produced only two new species in Nomimoscolex, one of Spatulifer and one of Amphoteromorphus. This suggests that although monticelliids are VOLUME 97, NUMBER 4 757 Fig. 16. Transformation series for best-fitting hypothesis of host-parasite relationships. Species enclosed in dotted circle belong in the family Pimelodidae. highly divergent morphologically, they may not be particularly speciose. In ad- dition to the monticelliids, only nine species of proteocephalids are known from South America, one of Corallotaenia (see Brooks and Deardorff 1980), one of Megathylacus and seven of Proteocephalus, including four in siluriform fishes (P. fossatus, P. jandia, P. kyukyu, P. platystomi). Thus, proteocephalidean diversity in South America may not be striking. The six species reported in this study appear to have relatively broad distributions in South America. The cladogram in Fig. 15 represents the phylogenetic relationships of the mon- ticelliids considered in this study with letters superimposed at the ends of branches indicating the identity of piscine hosts used by each cestode species. The letters refer to the following hosts (families listed in parentheses): . Brachyplatystoma flavicans (Castelnau) (=B. rousseauxiil) (Pimelodidae) Brachyplatystoma filamentosum (Lichtenstein) (Pimelodidae) . Brachyplatystoma vaillanti (Cuvier and Valenciennes) (Pimelodidae) . Luciopimelodus pati (Cuvier and Valenciennes) (=Pimelodus pati) (Pime- lodidae) Pimelodus clarias (Bloch) (Pimelodidae) Pseudoplatystoma fasciatum (Linnaeus) (Pimelodidae) . Pseudoplatystoma tigrinum (Cuvier and Valenciennes) (Pimelodidae) . Phractocephalus hemiliopterus (Schneider) (Pimelodidae) Ageneiosus brevifilis Cuvier and Valenciennes (=Pseudogeneiosus brevifilis, P. zungaro) (Ageneiosidae) Ageneiosus caucanus Steindachner (Ageneiosidae) . Platystomatichthys sturio (Kner) (Pimelodidae) Pinirampus pirinambu (Cuvier and Valenciennes) (=Pirara bicolor) (Pimelodidae) py je (eal teal | Wo) () ee) pe A 758 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON M. Cetopsis caecutiens (Lichtenstein) (Cetopsidae) N. Pseudocetopsis othonops (Eigenmann) (Cetopsidae) O. “Silurus sp.”’ (prob. Pimelodidae) P. Pseudopimelodus zungaro (Humboldt) (Pimelodidae) Q. Rhamdia sp. (Pimelodidae) R. Calophysus macropterus (Lichtenstein) (Pimelodidae) Note that the bulk of the hosts belong to the family Pimelodidae. Only six of the 47 (13%) of the host records for this group of cestodes are from other silu- riforms. Figure 16 depicts a transformation series of hosts, derived using the “nearest-neighbor” method of Mickevich (1982), which provides the most par- simonious explanation of host-parasitic relationships. If those transformations are consistent with host phylogeny, only nine of the 47 (20%) host relationships postulated by the cladogram are attributable to host-switching; all the rest are due to coevolution. If this is the case, the monticelliids have coevolved with a fairly small number of siluriforms, and may well have speciated more often than their hosts. One of the most interesting host switches is that for Goezeella siluri in Cetopsis caecutiens and Pseudocetopsis othonops (family Cetopsidae). When G. siluri is found in other hosts (pimelodids and ageneiosids), it occurs in the anterior third of the intestine. We found immature specimens of G. si/uri in the anterior third of the intestine of Brachyplatystoma vaillanti. But when it inhabits cetopsids, it is found in the posterior third of the intestine. This does not appear to be a function of competitive exclusion by Amphoteromorphus praeputialis, which in- habits the anterior third of the intestine of cetopsids, because G. si/uri lives in the posterior third of the gut even if A. praeputialis is absent. We suggest that the site selection by G. si/uri in cetopsids is a function of the location in the gut of a host cue (either physical or physiological) to which G. si/uri responds and is not a function of the presence of other helminths. Summary The previous phylogenetic hypothesis for monticelliids (Brooks 1978) overes- timated the amount of parallel evolution exhibited by the group. This was due to faulty character analysis and the use of non-monophyletic generic groupings. The present study, while far from a complete analysis, nonetheless presents a more stable phylogenetic hypothesis. At present, monticelliid cestodes appear to be a distinctive group of cestodes which has coevolved primarily with pimelodid catfish. Literature Cited Brooks, D.R. 1978. Evolutionary history of the cestode order Proteocephalidea. — Systematic Zoology 27:312-323. . 1979. Testing the context and extent of host-parasite coevolution.— Systematic Zoology 28: 299-307. , and T. L. Deardorff. 1980. Three proteocephalid cestodes from the Colombian siluriform fishes, including Nomimoscolex alovarius, sp. n. (Monticelliidae: Zygobothriinae). — Proceedings of the Helminthological Society of Washington 47:14-21. Farris, J.S. 1970. Methods for computing Wagner trees.—Systematic Zoology 19:83—92. Freze, V.I. 1965. Proteocephalata in fish, amphibians and reptiles. Jn K. I. Skrjabin, ed., Essentials of cestodology, vol. 5. 597 pp. (IPST English translation). VOLUME 97, NUMBER 4 ; 759 Kluge, A. G., and J.S. Farris. 1969. Quantitative phyletics and the evolution of anurans. —Systematic Zoology 18:1-32. Mickevich, M. F. 1982. Transformation series analysis.—Systematic Zoology 31:461—478. Rego, A. A., J. C. Dos Santos, and P. P. Silva. 1974. Estudos de cestoides de peixes do Brasil.— Memorias do Instituto Oswaldo Cruz 72:187-204. Riggenbach, E. 1896. Der Genus Ichthyotaenia.— Revue Suisse de Zoologie 4:165-275. Woodland, W. N. F. 1934. On the Amphilaphorchidinae, a new sub-family of proteocephalid ces- todes, and Myzophorus admonticellia, gen. et sp. n., parasitic in Pirinampus spp. from the Amazon.—Parasitology 26:141-149. 1935. Some more remarkable cestodes from Amazon siluroid fish.— Parasitology 27:207- 225. Department of Zoology, University of British Columbia, 6270 University Blvd., Vancouver, B.C. V6T 2A9, Canada. Appendix 1.—Synapomorphy list for cladogram of monticelliid cestodes inhabiting neotropical siluriform fishes. . Uniloculate suckers . Thin-walled uterus . Weakly-developed internal muscle layer . Compact ovary . Vagina alternating between anterior and posterior to cirrus sac . Apical organ present . Vitellaria partially cortical . 100-150 testes per proglottid . Vitelline bands arranged like two parentheses on either side of the proglottid . Ovary medullary . Testes medullary . Uterus medullary . No metascolex . No papillae on suckers . Vagina posterior to cirrus sac only . 50-60 testes . Ovary follicular . Vitellaria totally cortical . Internal muscle layer strongly developed . Vagina anterior to cirrus sac only . Triloculate suckers . No apical organ . Vitellaria in four quadrants, two dorsal and two ventral, along lateral margins . Vagina posterior to cirrus sac only . No apical organ . Metascolex I (Amphoteromorphus type) . Vitellaria in four quadrants, two dorsal and two ventral, along lateral margins . Partly cortical ovary . Vitellaria in two ventral bands on either side of proglottid . 150-200 testes . Elongate eggs . 100 testes . Vagina anterior to cirrus sac only . Short, thin egg filaments . Long, thick egg filaments . Vagina posterior to cirrus sac only . Uterus thick-walled, tubular . Biloculate suckers . Spinose scolex . 40 testes . Vitelline bands arranged like two parentheses on either side of the proglottid NOK RR RR RS i RS ODomaoarnanhwnr OO WOAYADM &WNY NN WN WN WWWWWWWWWwWwonNNNN NY NY OMAAINDMAHBWNKH OO ONAN f A i= ©) 760 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 42. 150-200 testes 43. Tissue growth around suckers so pronounced that only an anterior and posterior opening remain 44. Ovary partially cortical 45. Two papillae on margin of each sucker 46. 40-60 testes per proglottid 47. Alate ovarian lobes 48. Two papillae on tops of each sucker 49. Vagina posterior to cirrus sac only 50. Vagina anterior to cirrus sac only 51. Four papillae on margins of each sucker 52. Vagina anterior to cirrus sac only 53. Vagina posterior to cirrus sac only 54. 200-250 testes per proglottid 55. Vagina posterior to cirrus sac only 56. Vagina anterior to cirrus sac only 57. Ovary cortical 58. Testes cortical 59. Uterus cortical 60. Vitelline bands converging towards midline ventrally 61. 200 testes per proglottid 62. Spinose suckers 63. 60 testes per proglottid 64. 200 testes per proglottid 65. Metascolex II (Goezeella type) 66. 100 testes per proglottid 67. 40-50 testes per proglottid 68. 150-200 testes per proglottid 69. 200-400 testes per proglottid 70. Biloculate suckers PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 761-772 TWO NEW SPECIES OF GIGANTIONE KOSSMANN (ISOPODA: EPICARIDEA: BOPYRIDAE) FROM THE WESTERN NORTH ATLANTIC Daniel L. Adkison Abstract.— Gigantione mortenseni n. sp. is described from three dromiid crabs, Dromidia antillensis Stimpson, Hypoconcha sabulosa (Herbst), and H. spinosis- sima Rathbun. Gigantione uberlackerae is described from an undescribed axiid species, Axiopsis (Paraxiopsis) sp. Gigantione is a well defined taxon having several unusual characteristics re- flecting the distinctiveness of Gigantione from other bopyrid genera. The unusual characteristics for the female of Gigantione are: (1) the oral cone is visible in dorsal view, (2) the maxilliped is edged with plumose setae, (3) the basal segment of the first antenna is enlarged and applied to the oral cone, (4) the coxal plates and lateral plates are pubescent, and (5) the uropods have a discrete peduncle. The male of most species has biramous pleopods and uropods. The following generic diagnosis is presented to include several of the above unusual characters. Gigantione Kossmann Generic diagnosis. — Female: body nearly circular in outline; distortion slight to moderate; all regions and segments distinct. Head with oral cone visible in dorsal view; anterior lamina medially narrow, usually laterally developed into lobes; basal segments of first antenna greatly enlarged and applied to oral cone; maxil- liped without palp, margin at least medially fringed with plumose setae; posterior lamina, one pair of lateral projections and median projection. Coxal plates present on at least anterior pereomeres, pubescent. Dorsal bosses on pereomeres 1—4. Pleomeres 1-5 with lateral plates, lateral plates pubescent; pleomere 6 set deeply in pleomere 5. Pleopods, 5 biramous pairs; first pair larger than others with rami leaf-shaped; other pleopods digitate, decreasing in size posteriorly. Uropods bi- ramous. Male: all regions and segments distinct. Pleomeres usually with lateral margin directed ventrally. Pleopods, 5 pairs. Pleopods and uropods present, generally biramous. Remarks.—In the phylogeny of the Bopyridae, Gigantione represents a prim- itive branchial form (Shiino 1952, 1965; Markham 1974). In the Bopyridae, advanced forms are characterized by the reduction in some or all of the following characters: pleopod rami reduced, number of pleopod pairs reduced, oostegites reduced, fusion of pleomeres, fusion of pereomeres. For the male, advanced forms are characterized by some or all of the following: head fused to pereomere, pleo- pods reduced in number or rami, pleomeres fused. Both the female and male of Gigantione have all regions and segments distinct, five pairs of biramous pleopods (generally biramous for males), and uropods generally biramous. 762 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Gigantione mortenseni, new species Figs. 1—5 Gigantione bouvieri Nierstrasz and Brender a Brandis, 1931:152—153, figs. 7-12. Material examined.—Infesting Dromidia antillensis Stimpson, 1859, Gulf of Mexico. Station SAM 475-10; 30°08’N, 87°07'W; 32 m; Apr 1975; R. L. Shipp collected; D. L. Adkison determined host; 1 2 (ammature, no male) USNM 172464.—Station MAFLA ITI-D(37); 29°48'N, 86°03'W; capetown dredge; 43 m; Jun 1974; T. S. Hopkins collected; D. L. Adkison determined host; 2 2 (both immature, no males, both from right gill chamber of same host) USNM 172461.- Bellows station 39; 28°26'N, 84°21'W; capetown dredge; 43 m; | Jul 1977; S. B. Collard and D. L. Adkison collected; D. L. Adkison determined host; | 2 (gravid), 1 6 USNM 172467.—Station MAFLA II-N(49); 28°24'N, 84°21'W; 46 m; Jun 1974; T. S. Hopkins collected; D. L. Adkison determined host; 2 2 (1 ovigerous, 1 immature), 2 6 (1 juvenile) RMNHL I-5911.—Station MAFLA I-B(62); 27°50'00’N, 83°31'00"”W; diver collected; 28 m; 16 Aug 1976; T. S. Hopkins collected; D. L. Adkison determined host; 1 2 (gravid, no male) USNM 172463.— Station EJ-65-272; 27°37'N, 83°28'W; 37 m; 31 Aug 1965; B. Presley collected; N. Whiting determined host; | 2 Gmmature), 1 6 FSBC I-22928.—Station BLM 3102014; 26°24'N, 83°49'W; capetown dredge; 98 m; 15 Jul 1976; T. S. Hopkins collected; D. L. Adkison determined host; | 2 (gravid), 1 ¢ZMC.— Hourglass station L, EJ-65-347; 26°24'N, 83°22'W; 55 m; 13 Nov 1965; B. Presley collected; W. G. Lyons determined host; 1 2 (gravid), 1 6 FSBC I-22929.—Hourglass station L, EJ-65-380; 13-14 Dec 1965; H. Wahlquist collected; W. G. Lyons determined host; 2 2 (1 gravid, 1 juvenile), 1 6 FSBC JI-22927.—Station BLM 15-I-A(62); 26°21'N, 82°57'W; capetown dredge; 37 m; 28 Jul 1975; T. S. Hopkins collected; D. L. Adkison determined host; 1 2 (gravid), 1 ¢ DISL.—Station 65-188; 25°50’N, 82°44'W; 36 m; 30 Jun 1965; E. Joyce determined host; 1 2 (gravid, holotype) USNM 172456; 1 6 USNM 172457; 2 2 Guvenile), 1 ¢ USNM 172458.—Station EJ-65-209; 25°29'N, 82°17'W; 27 m; 10 Jul 1965; A. Provenzano determined host; 1 2 (non-gravid), 1 ¢é USNM 172466.—Oregon II station 21317, 21318; 24°51'N, 82°20'W; trawl; 28 m; 15 Dec 1976; D. L. Adkison collected and de- termined host; 1 2 (non-gravid, no male) USNM 172460.—Station EJ-67-147; 24°38’N, 82°59'W; 28-33 m; 27 Apr 1967; W. G. Lyons and M. Moe collected; W. G. Lyons determined host; 4 2 (juvenile), 1 é USNM 172462. Atlantic Ocean off Florida. —Station EJ-73-63; 27°55'N, 80°03’W; 46 m; 14 May 1973; D. Barber collected; N. Whiting determined host; 2 2 (1 gravid, 1 immature), 2 6 (1 juvenile) USNM 172459.— Yucatan Peninsula, Mexico. Station EJ-68-64; 20°03’N, 87°04'W; box dredge; 37 m; 20 Jul 1968; D. K. Camp collected; D. L. Adkison determined host; | @ (gravid), 1 6 USNM 172465.—Haiti; 1974; Pr. A. Veillet collected; D. L. Adkison determined host; 1 9 (non-gravid), | é MNHN Ep. 114. Infesting Hypoconcha sabulosa (Herbst). — North from Thatch Cay, St. Thomas, Virgin Islands; approx. 18°21'N, 64°52’W; 15 Mar 1906; T. Mortensen collected; D. L. Adkison determined host; 1 2 (gravid), 1 ¢ ZMC.—Station EJ-74-191; 28°37'N, 80°11'W; 40 m; 14 Oct 1974; D. Barber collected; R. H. Gore determined host; 1 2 Gmmature), 1 6 USNM 172468. Infesting Hypoconcha spinosissima Rathbun.—Station MAFLA III-D(37); 763 VOLUME 97, NUMBER 4 | Fig. 1. Gigantione mortenseni, female: a, Dorsal view, holotype; b, First antenna (stippled area, point of attachment); c, Second antenna (stippled area, point of attachment); d, Posterior lamina, right maxilliped and spur removed (spur indicated with stippling); e, Maxilliped (setae omitted); f, Pereopod 6; g, Oostegite 1, internal view. Figures from holotype, a, d, g. Figures from USNM 172468, b, c, f. Figure from Nierstrasz and Brender a Brandis specimen ZMC, e. Scale = 2.0 mm. 764 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Gigantione mortenseni, female: a, Pleon, ventral view (right pleopod 3 and left pleopod 1 missing); b-f, Pleopod, exopods on right; b, Pleopod 1; c, Pleopod 2; d, Pleopod 3; e, Pleopod 4; f, Pleopod 5. Figure from Nierstrasz and Brender a Brandis specimen ZMC, a. Figures from USNM 172468, b-f. 29°37'N, 80°03'W; capetown dredge; 43 m; Jul 1974; T. S. Hopkins collected; D. L. Adkison determined host; 1 2 Guvenile, no male) USNM 172470.—Station 2105; 26°24’N, 83°49’W; capetown dredge; 98 m; 2 Feb 1979; T. S. Hopkins collected; D. L. Adkison determined host; 1 2 (juvenile, no male) USNM 172467. Description. — Female (Figs. 1-4): Mature specimens; length, 5 to 9.5 mm; width across pereomere 4, 5.7 to 8.8 mm; distortion angle 0-—30°. Head wider than long; anterior lamina narrow, laterally expanded into lobes. Eyes often present, visible in anterolateral view. First antenna, segmentation indistinct, 2 to 4 segments, usually 3 segments; basal segment greatly expanded. VOLUME 97, NUMBER 4 765 Fig. 3. Gigantione mortenseni, immature female: a, Dorsal view; b, Maxilliped; c, Maxilliped; d, Oostegite 1, external view; e, Oostegite 1, internal view; f, Second pereopod, coxal plate and oostegite; g, Pleon, ventral view; h, Pleon of a, ventral view. Figures from USNM 172470, b and g (specimen less mature than USNM 172467). Figures from USNM 172467, a, c-f, h. Scale A = 1.0 mm (Fig. a); B = 0.5 mm (Figs. b-f); C = 1.0 mm (Fig. h); D = 1.0 mm (Fig. g). Second antenna, 5 or 6 segments; with numerous short setae. Second antenna about twice length of first. Maxilliped fringed with long plumose setae. Posterior lamina margin tuberculate, 1 pair of lateral projections and a median projection with | pair of tubercles. 766 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Gigantione mortenseni, variation of female: a, From left gill chamber, USNM 172463; b, From left gill chamber, FSBC I-22929; c, From right gill chamber, USNM 172459; d, From right gill chamber, DISL. Scale = 5.0 mm. Pereon with dorsal bosses present on pereomeres | to 4 and rarely pereomere 5. Coxal plates present on all pereomeres, pubescent; projection elongate, increas- ing in length posteriorly; on expanded side, coxal plates longer and narrower than on reduced side, coxal plate projection often reduced to small lobe on pereomere 1. First oostegite, posterior plate short, broader than anterior plate; anterior plate with anteromedian edge fringed with short plumose setae, internal ridge with projections on lateral half of ridge. Pereopods of similar size, increasing slightly in size posteriorly, with basal carina. Pleon of 6 segments, lateral plates on all but pleomere 6; lateral plates pubescent; lateral plates on expanded side with narrow projection similar to that of coxal plates. Pleopods of 5 biramous pairs; first pair much the largest, leaf-shaped, with dorsal margin digitate, exopod wider than endopod. Other pleopods digitate to multilobed often appearing triramous; pleopods decreasing in size posteriorly, endopod longer than exopod. Uropods biramous, rami nearly equal in size, base ~ VOLUME 97, NUMBER 4 767 Fig. 5. Gigantione mortenseni, male: a, Dorsal view; b, Left antennae; c, Maxillipeds; d, Pleon, ventral view; e, Right uropod; f, Pereopod; g, Complex spine on carpus. Figure from FSBC I-22927, c. Figures from USNM 172468, a, b, d—g. Scale = 0.5 mm. with prominent tubercle on ventral surface at about point where uropod becomes visible in dorsal view. Male (Fig. 5): Based on mature specimens; length 2.1 to 2.7 mm, width across pereomere 4, 0.8 to 1.0 mm. Head, approximately trapezoidal; expanded lobes present at posterolateral cor- ner of head. Eyes present, generally pigmented. First antenna, 3 segments; second antenna, 5 segments; second segment 3 times length of first. Maxillipeds as figured. Pereomere | anterior margin convex; pereomeres 2—4 nearly straight; pereo- meres 5-7 progressively directed more posteriorly. Midventral tubercle absent. Pereopods, all of similar size, carpus with a complex spine. Pleon of 6 segments; all but last with lateral area directed posterolaterally. Midventral tubercles absent. Pleopods, 5 biramous pairs, endopod usually elon- gate and larger than exopod. Uropods biramous. Etymology.—This species is named for Dr. Th. Mortensen who collected the first known specimens of the new species. 768 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Distribution. —Gigantione mortenseni is known from Haiti, the Virgin Islands, the Gulf of Mexico, and the Atlantic Ocean off Florica. Discussion. —Nierstrasz and Brender a Brandis, 1931, described and illustrated a pair of bopyrids from Hypoconcha sabulosa as Gigantione bouvieri Bonnier, 1900, a species known from the xanthid crab, Pilumnus hirtellus (L.). Based on the phylogenetic distance between Hypoconcha and Pilumnus, the forms do not appear to be conspecific. While Nierstrasz and Brender a Brandis (1931) thought the differences between their specimen and the description of Gigantione bouvieri were minor, I believe these differences are specific and differentiate the two species. The differences noted for the female G. mortenseni are: (1) the coxal plates are very slender, (2) coxal plates do not cover dorsal bosses, (3) the first pair of pleopods are smaller (though the relative development of the pleopods is age-related to some degree), and (4) the uropods have a basal projection. The differences for the male of G. mortenseni are the presence of biramous pleopods and uropods. The complex spine on the pereopods of the male of G. mortenseni is unusual and may be a specific character. The pleopods of the female may also be more digitate but this cannot be determined without examination of specimens of G. bouvieri. Gigan- tione mortenseni also has a distinct frontal lamina that was not mentioned by either Bonnier (1900) or Nierstrasz and Brender a Brandis (1931), or illustrated by Nierstrasz and Brender a Brandis (1931). Examination of the material of Nierstrasz and Brender a Brandis shows it to have a distinct frontal lamina. During maturation of the female G. mortenseni, projections of coxal plates and lateral plates appear early in development, but the coxal plates and dorsal bosses do not become discrete units until the adult form is nearly attained. The uropods take on the adult form with the basal projection early in development but after the projections on the coxal and lateral plates are present. The relative length of the oostegites and the complexity of the pleopods are the best estimates of ma- turity. Pleopod 1 becomes larger and broader with age, with the tubercles on the dorsal edge becoming larger and more numerous. Pleopods 2 to 5 become more complex with maturity, being a simple V-shape in the juvenile to complexly tuberculate in the adult and appearing triramous. While the pleopods are becoming more complex, oostegites 2 to 5 are increasing in length. At the simple V-shape stage of the pleopods, the oostegites are small plates that do not reach the midline of the specimen. The oostegites continue to increase in length until in a mature nongravid female the oostegites nearly reach the pereopods of the opposite side. The maturation of the male is not known due to the lack of immature specimens. The only important variation noted (Fig. 4) for the mature female is a variation in the outline, distortion angle, and development of posterior coxal plates and pleon lateral plates. This variation in gross form does not affect the specific characteristics listed above. Gigantione uberlackerae, new species Figs. 6-8 Material examined.—Infesting Axiopsis (Paraxiopsis) n. sp. Florida Middle- grounds, eastern Gulf of Mexico; Bellows station 21; 28°34'N, 84°17'W; capetown dredge; 30 m; 10 Jul 1977; S. B. Collard and D. L. Adkison collected; B. F. VOLUME 97, NUMBER 4 769 Fig. 6. Gigantione uberlackerae, female, holotype: a, Dorsal view; b, Pleon, ventral view. Scale = 2.0 mm. Kensley determined host; 1 2 (non-gravid, holotype) USNM 172449, 1 6 USNM 172450 (host USNM 172451). Description. — Female (Figs. 6 and 7). Length 2.9 mm; width across pereomere 4, 3.2 mm; distortion angle less than 5°; specimen nearly mature, oostegites nearly fully developed. Head with distinct median notch; anterior lamina reduced; no lateral projec- tions. Eyes absent. First antenna, segmentation indistinct apparently on 5 seg- ments; second antenna, 8 segments; second antenna three times length of first. Maxilliped fringed with plumose setae. 770 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 7. Gigantione uberlackerae, female, holotype: a, Posterior lamina; b, First antenna; c, Second antenna; d, Maxilliped; e, Oostegite 1, external view; f, Oostegite 1, internal view; g, Pereopod 1; h, Pereopod 4. Pereon with dorsal bosses on pereomeres | to 4; coxal plates on all pereomeres pubescent, increasing in length until pereomere 5, thereafter decreasing in length; tergally defined on pereomeres | to 5. Pereopods all of similar size, posterior pairs with small basal carinae. First oostegite, posterior plate wider than anterior plate; interior ridge with 2 lateral projections. Pleomere 1 broader than pereomere 7; lateral plates on pleomeres 1 to 5; pubescent, lateral plates narrower on right side. Pleopods, 5 biramous pairs, first pleopods leaf-shaped, pleopod rami becoming more lanceolate on posterior pairs; margins with only scattered tubercles. Uropods biramous, right uropod without exopod. Male (Fig. 8): Length 1.7 mm; width across pereomere 3, 0.7 mm. Head, unpigmented eyes present. First antenna, 4 segments; second antenna, 8 segments; first antenna approximately 14 length of second antenna. Maxillipeds as figured (Fig. 8e). Pereomeres 5 to 7 with lateral area progressively directed posteriorly. Midven- tral tubercles absent. Pereopod slightly larger anteriorly. Pleon decreasing in width posteriorly; pleomere 6 without lateral development, deeply set in pleomere 5. Pleopods, 5 biramous pairs, decreasing in size posteriorly; exopod reduced to lobe. Uropods biramous. Etymology.—This species is named for Joan Uberlacker, who has found several interesting bopyrids in her studies. Distribution. — Gigantione uberlackerae is known from one pair collected on the Florida Middlegrounds. VOLUME 97, NUMBER 4 ; 771 Fig. 8. Gigantione uberlackerae, male: a, Dorsal view; b, Antennae; c, Pereopod 1; d, Pereopod 7; e, Left maxillipeds; f, Pleon, ventral view; g, Left uropod. Scale = 0.5 mm. Discussion. — The female of Gigantione uberlackerae n. sp. is distinguished from the nine other members of the genus by: (1) the presence of an anteromedial notch in the head, and (2) the pleopod having few tubercles. Only three species of Gigantione: G. uberlackerae, G. rathbunae Stebbing, 1910, and G. sagamiensis Shiino, 1958, do not have lateral projections on the head of the female. The female of G. rathbunae differs from the female of G. uberlackerae by at least pleopods 2 and 3 being tuberculate. The male of G. rathbunae also differs in having uni- ramous pleopods and uropods. The female of C. sagamiensis differs from G. uberlackerae by the presence of an enlarged first pleopod, pleopods 2 to 5 having a V-shape with some tubercles, short uropodal rami, and tuberculate lateral plates on the expanded side. The male of G. sagamiensis has uniramous uropods. Gigantione uberlackerae is the first Gigantione reported from a non-brachyuran host. It is for this reason that I have decided to describe this species based on only one pair. The holotype is non-ovigerous and may be somewhat immature. Oostegites 2 to 5 overlap those on the opposite side. Based on study of the maturation of Gigantione mortenseni, pleopods of the holotype appear to be nearly mature. The first pleopod of G. uberlackerae is not enlarged as in the other members of the genus. The lack of an exopod of the right uropod of the female is the only apparent abnormality noted for Gigantione uberlackerae. VH2 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The holotypes of both species and several paratypes are deposited in the col- lection of the Smithsonian Institution (USNM). Paratypes are deposited in col- lections of the following museums: Universitetets Zoologiske Museum, Copen- hagen (ZMC); Rijksmuseum van Natuurlike Historie, Leiden (RMNHL); and Muséum National d’Histoire Naturelle, Paris (MNHN). Acknowledgments I wish to express my appreciation to Dr. Richard W. Heard for pointing out the many specimens of Gigantione mortenseni. I wish to express my thanks to David K. Camp, Florida Department of Natural Resources (FSBC); Thomas S. Hopkins, Dauphin Island Sea Laboratory, Alabama (DISL); Robert L. Shipp, University of South Alabama; Torben Wolff, Universitetets Zoologiske Museum, Copenhagen (ZMC); and Roland Bourdon, Muséum National d’Histoire Natu- relle, Paris, for loan of material in their collections. Literature Cited Bonnier, J. 1900. Contribution a l’étude des épicarides. Les Bopyridae.—Travaux de la Station Zoologique de Wimereux 8:1—476. Kossmann, R. 1881. Studien iiber Bopyriden: I. Gigantione moebii und Allgemeines iiber die Mund- werkzeuge der Bopyriden. — Zeitschrift fiir Wissenschaftliche Zoologie 35:652—680. Markham, J. C. 1974. A systematic study of parasitic bopyrid isopods in the West Indian faunal region.—Ph.D. dissertation, University of Miami, Coral Gabies, Florida, 344 pp. Nierstrasz, H. F., and G. A. Brender 4 Brandis. 1931. Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-1916. 57. Epicaridea II.— Videnskabelige Meddelelser fra Dansk Naturhis- torisk Forening 91:147-226. Shiino, S. M. 1952. Phylogeny of the family Bopyridae.— Annual Report Prefectural University of Mie Section 2, Natural Science 1:33-56. . 1965. Phylogeny of the genera within the family Bopyridae. — Bulletin du Muséum National d’Histoire Naturelle, Paris, ser. 2, 37(3):462-465. Department of Biology, Tulane University, New Orleans, Louisiana 70118. Present address: Department of Veterinary Pathology, Texas A&M University, College Station, Texas 77843-4463. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 773-787 COMMENTS ON THE SKATES OF THE TROPICAL EASTERN PACIFIC: ONE NEW SPECIES AND THREE NEW RECORDS (ELASMOBRANCHII: RAJIFORMES) John D. McEachran and Tsutomu Miyake Abstract.— Recent collections of benthic fishes off the coasts of Ecuador, Peru, and the Galapagos Islands yielded a new species and three new records of skates from the eastern central Pacific. Bathyraja peruana, n. sp. is described from eight specimens, including the paratype of B. aguja Kendall and Radcliffe. This species is distinguished from other Bathyraja species of the eastern Pacific in morpho- metrics, meristics, spination, and pigmentation. A specimen resembling B. rich- ardsoni but differing from it in a number of proportional measurements is dis- cussed. Bathyraja longicauda and Breviraja nigerrima (=Malacoraja nigerrima) are recorded from Peru for the first time. A neotype for M. nigerrima is designated and described. As presently known, the skate fauna of the eastern central Pacific, southern Baja California to northern Peru, is species poor, consisting of Psammobatis aguja, (Kendall and Radcliffe), (=Bathyraja aguja), B. spinosissima (Beebe and Tee-Van), Raja badia Garman, R. ecuadoriensis Beebe and Tee-Van, R. equatorialis Jordan and Bollman, R. velezi Chirichigno, and Sympterygia brevicaudata (Cope). In the eastern central Atlantic, zoogeographically similar to the eastern central Pacific (Briggs 1974), there are 32 species of skates (Stehmann 1981). Beebe and Tee-Van (1941) and Hildebrand (1946) suggested that the paratype of B. aguja represents an undescribed species of skate and recent collections mainly by the R/V Anion Bruun (cruises 16, 18A, and 18B) support their suggestion and indicate that the paucity of skates in the eastern central Pacific may, in part, be due to sampling error. Herein we describe the eight species of skate from this area and discuss three other species which are either new records for the area or new species. Materials and Methods Specimens examined were obtained from the Museum of Comparative Zoology, Harvard University (MCZ), the Smithsonian Oceanographic Sorting Center (SOSC), and the National Museum of Natural History, Smithsonian Institution (USNM). The SOSC specimens were deposited at the Museum of Comparative Zoology (MCZ), the Texas Cooperative Wildlife Collection (TC WC) and the USNM. One specimen of the new species was dissected to reveal the structure of the neuro- cranium, scapulocoracoids and claspers. Most specimens, including the holotype and paratype of B. aguja, were radiographed to verify anatomical observations based on dissections and to count vertebrae and pectoral radials. Methods for making measurements and counts follow McEachran and Compagno (1979, 1982). 774 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Bathyraja peruana MCZ 1364 (Holotype, 283 mm TL, immature male), a, Dorsal view; b, Ventral view. Bathyraja peruana, new species Figs. 1—4; Tables 1, 2 Raja aguja Kendall and Radcliffe, 1912:78-79 (in part), Fig. 2. Holotype.—MCZ 1364, 283 mm TL, juvenile male (paratype of Raja aguja, collected off Point Aguja, Peru, 5°47’S, 81°24'W, 980 m, 12 Nov 1904, R/V Albatross, Station 4653. Paratypes.—MCZ 61113, 173 mm TL, juvenile male, 70°49’S, 80°38'W, 605— 735 m, 5 Jan 1966, R/V Anton Bruun, Cruise 18B, Station 754.—TCWC 3515.1, 970 mm TL, 980 mm TL, mature females, 15°11.5'S, 75°43’W, 1010 m, 25-26 Aug 1966, R/V Anton Bruun, Cruise 18A, Station 739A.—TCWC 3516.1, 1053 mm TL, mature female, 15°11'S, 75°44’W, 1060 m, 25-26 Aug 1966, R/V Anton Bruun, Cruise 18A, Station 739.—USNM 267045, 203 mm TL, 469 mm TL, juvenile males, 610 mm TL, juvenile female, 3°15’S, 80°55’W, 945-960 m, 10 Sep 1966, R/V Anton Bruun, Cruise 18B, Station 770. Diagnosis.— Disc lozenge-shaped, anterior margin more or less straight; preor- bital length 2.41 to 3.40 times orbital length and 0.10 to 0.14 times total length; disc lacking thorns; tail with 18 to 26 midrow thorns, lacking thorns between dorsal fins; ventral surface of disc without dermal denticles. Description.—Disc 1.15 times as broad as long (1.22 to 1.36 in paratypes); VOLUME 97, NUMBER 4 775 _ - ef a hte 4 =, - < ~~, eae . ws pre Fig. 2. Bathyraja peruana TCWC 3516.1 (Paratype, 1053 mm TL, mature female). maximum angle in front of spiracles 105° (100° to 125°); margin of disc nearly straight from tip of snout to level of nares, slightly convex to level of first gill slits, nearly straight to outer corners which are abruptly rounded, posterior margin slightly convex. Axis of greatest width 0.80 (0.70 to 0.82) times distance from tip of snout to axil of pectoral fins. Pelvic fins deeply incised; anterior lobe narrow, tapering distally, length 0.88 times posterior lobe (0.64 to 0.95, with ratio greater in larger specimens). Tail slender, little depressed, with narrow lateral fold along ventrolateral surface originating at axil of pelvic fins, widening distally and ex- tending to near tip. Length of tail from center of cloaca to tip 1.07 times distance from tip of snout to center of cloaca (0.88 to 1.40, with ratio smaller in larger specimens). Post dorsal fin tail length less than one-half length of base of second dorsal fin. Preorbital length 3.13 (2.41 to 3.40) times orbit length; preoral length 1.69 times internarial distance (1.23 to 1.72, with ratio smaller in larger specimens). Inter- orbital distance 1.04 times orbit length (1.02 to 1.17); orbit length 1.62 times spiracle length (1.16 to 1.78, with ratio smaller in larger specimens). Anterior nasal flap (nasal curtain) coarsely fringed along distal margin; posterior nasal flap PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 776 -odAjeiegd z “odA}O[OH, |; OEL OL CL LL S[eIpes [e10100d jo JSOqunyy 0°89 IL 89 69 69 v9 69 69 69 9BIQSLIOA [epneo [esiopaid Jo Joquinyy SEE ve ce ve €€ (43 ce €€ ve sBIQOLOA JUNI) JO IoquINNy SPE Ce 9€ ve 9€ ce ve Ce €€ (mef 1addn) smo1 y1001 Jo Ioquinyy (ys OS LY LV TS IS €s gs GS dij [epnes 0} vorojo—sourisiq Lv 6P €v or Lv (OS SP Lv or UISLIO [epned 0} BORO[D—dOUrISIG Or vv [Life Or 6£ (4 €v Or Or UY [BSIOP 1S] 0} BOROTI—9s0URISIG OS (ys €S 1S 8P 6P OV (47 Sr BOBO]D 0} NOUS —ddUR}SIG CSI pO SSI Gal 191 all O7I STI 9TI 9qO] dIAjod 1oLIa\sod jo yuaT vit STI 0°01 601 O7I LUI ell 611 VU 9qO] dIAjad IOLI9}uUe Fo YIsUIT eel a | vel 6 v1 8°01 COI VT Ell LU sdutuado |[I3 yy UsaMjJoq JOULISIG C61 (GANG 0°07 6°07 € 91 pL 161 191 € 61 sdutuado |[I8 Is1y U9aMJoq dULISIG 8 v7 81 61 cl vl UT 60 TI sdutuado ][I8 YY JO YIPI C7 1€ v7 97 €7T 61 imal ia 07 sdutuado ][I3 psry JO YIPIA\ IPG LG I 07 07 61 cl Tea pal sdutuado |IId IsIy JO YIPIM 78 08 88 €8 08 v8 6L 69 £8 S[Lsou U9aMJeq JdULISIG I€ v7 SZ e€ gE lv €€ 97 Le€ yINou! 0} dIeN L8 76 0°6 v6 Gal 18 18 9°9 78 WPIM YyINOP] lee ale VL 9L 09 19 89 C9 VL sgpoeiids usamjaq 20ULIsIq Te ce €€ L€ CC CC LAG S16 97 yisuay apoesids GS 09 61 GS 61 IS 8'P xs es yisug] spoestds pue 11q10 Laat 1° tL Sr 61 lv vv ev v'Y S}IQIO U99MJOq dULISIG Uv ev Ob ev 6'€ 6€ ev SE (52 JoJWIRIP 11q1O C6 68 £6 L8 v0l TI 8°01 68 OTT yisuay [eseucid TE TE €€ O€ (43 9¢ (43 TE 9€ YIPIM WNUWIXeU 0} jNOUg OTT 101 601 VI 6 EI Cr TEI 8°01 lvl ({e1001d) yIsua] NOUS ane ¢ Ol 601 601 vel eel rail 8°6 eel ({e11q1001d) yduI] NOUS pS vS 1S 6S TS €S 1S 6P ¢¢ yisus] OsIq 69 OL OL IL 69 OL €9 €9 v9 YIPIM OSI €SOl 086 0L6 019 69P €07 ELI €87 (uit) YIsUZ] [210], rs) ) fs) 3 ? ° 9° ° X9S x zI9ISE zI SIS zISISE zSVOLIT zSVOLIT zSPOLIT E1119 ip9El IMOL IMOL IMOL WNSf WNSf WNSf) ZOW ZOW "y13Ug] [2101 JUV0I0d UT passaidxa o1e SUONIOdOIg ‘DUDNAad DiDAAYIJDg JO SON|VA INSLIOU pue s}USWoINSvOW [eUOTLIOdOIg — "| 9IQeL VOLUME 97, NUMBER 4 777 Table 2.—Neurocranial and scapulocoracoid measurements of Bathyraja peruana USNM 267045, 469 mm TL juvenile male. Measurements are expressed as percentage of nasobasal length or scapulocor- acoid length. Nasobasal length (mm) 51.7 Cranial length 187 Rostral cartilage length 87 Prefontanelle length V7 Cranial width 115 Interorbital width 34 Rostral base 30 Anterior fontanelle length 36 Anterior fontanelle width 19 Posterior fontanelle length 42 Posterior fontanelle width 4 Rostral appendix length 31 Rostral appendix width 16 Rostral cleft length 19 Cranial height 23 Width across otic capsules 49 Least width of basal plate Di] Greatest width of nasal capsule 42 Internasal width 21 Scapulocoracoid length (mm) 33.2 Scapulocoracoid height 69 Premesocondyle 29 Postmesocondyle 71 Postdorsal fenestra (largest) length 15 Postdorsal fenestra (largest) height 10 Predorsal fenestra length 10 Predorsal fenestra height 10 Rear corner 49 poorly developed and weakly fringed (smooth to weakly fringed). Upper and lower jaws slightly arched (moderately arched in larger specimens). Teeth with short pointed cusps (larger specimens with longer cusps) and quincunx arrangement. Distance between first gill slits 2.32 (2.08 to 2.64) times internarial distance; distance between fifth gill slits 1.41 (1.25 to 1.89) times internarial distance; length of first gill slits 1.48 times length of fifth gill slits (1.05 to 1.32 with ratio smaller in larger specimens). First dorsal fin about equal in size and shape to second; distance between dorsal fins equal to about one-half of base of first; second dorsal fin separated by short distance from poorly developed epichordal caudal fin lobe. Upper surface of disc and tail covered with dermal denticles, dense along an- terolateral margin, over cranium and along midline to tip of tail but sparse over branchial region and on base of pectoral fins; denticles with four-pointed, star- shaped bases and posteriorly directed spine. Disc devoid of thorns (some speci- mens with several midrow thorns just anterior to axil of pectoral fins). Tail with 25 small midrow thorns (18 to 26) with oval bases. Color.— After storage in alcohol, holotype is uniform brownish gray; paratypes are similarly colored, except areas around nares, mouth, base of pelvic fins and base of tail are lighter, ranging from brown to yellow. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 778 UDUILIOF DAIOU SNSBA— KX “UDUIVIOJ AIOU jeosuAZieydossojs— x] “UsWIeIOJ DAIOU [eIORy JO YOuRIQ IejnqipuewoAy—]TA ‘UOUILIOJ DAIOU IVd[YIOI— AT “USUTRIOJ JAIOU IOJOULO[NIO — TI] ‘UdWIRIOJ 9AIOU ONdO—]] “Weys [e11so1—s1 ‘gpou [esoi1—u ‘aseq [e11so1—qui ‘x1puodde [esOI—PI “USUILIOF 9AIOU snpunjoid—joud ‘ssosoid jei1q1001d — dod ‘ssa001d ye11q10isod —djsod ‘usuresoy [eu [e1IGQ1091d — dod “o][auBIUOJ Jo110}sod —jd ‘uowesoy sneydurAytied —tad *y901s ondo—so ‘uouiesoy [euRd [eseUO}IGIO—9dUO ‘oINssy [B1IGQ1IO—Jo ‘g[Apuod [e}IdI9N90—90 “UDdUUIRIOJ DAIOU [PIOVT Jo yoursq 910—JqO ‘ajnsdeo [eseu—ou “Youe yesn(—ef ‘uowie1oy UTOA [e}IG1010}UT—O}UT “UsUTBIOY Aioue pose [eUIOJUI—9dI ‘JooRJ Je[NqrpueWoc Variables which differ significantly between samples from Alaska and the U.S.S.R. are indicated by asterisks (* P < 0.05, ** P < 0.01, *** P < 0.001; by one-way analyses of variance). ¢ Data are from Table II of Miller et a/. (1983), or from raw data used in that study. 4 Probability of difference between means (two-tailed), based on f-tests (symbolism as for footnote b; NS = not significant (P > 0.05)). minutilla; Miller 1983a), notably in their sequential increase in duration and decrease in frequency and modulation rate within song units, and their rhythmic amplitude and frequency modulation (which are coupled). Other features of song structure in these three species are similar, but none so strikingly. Information about songs of other calidridines is too scanty to permit comprehensive com- parisons, though song in several species is very different (Miller 1983b). Vocalizations of only two species each of snipe, woodcock, and godwits have been analyzed, and these differ strongly from song structure described here (Glutz von Blotzheim et al. 1977; Mal’chevskii 1980; Warham and Bell 1980; Cramp 1983; Miller 1984). However, buzzy elements occur in several of the taxa men- tioned, and also characterize nuptial displays in several species of Calidridini 810 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Miller 1983b). Quantitative analysis of buzzy elements which occur within song, or independently of it, could therefore provide insight into adaptive radiation and affinities of this group of Scolopacidae. Available evidence is clearly too limited to judge which features of L. scolopaceus song are ancestral and which are derived. Detailed descriptions of vocalizations in the taxa mentioned are badly needed. Acknowledgments We thank F. A. Pitelka, B. McCaffrey, and T. Hall for assistance. Recordings in Alaska were obtained during ecological research funded by the United States Department of Energy in contracts to F. A. Pitelka. Other research support was from the Natural Sciences and Engineering Research Council of Canada (indi- vidual operating grant to E. H. Miller), and the Friends of the British Columbia Provincial Museum (equipment and research grants to E. H. Miller). Literature Cited Cramp, S. (ed.) 1983. Handbook of the birds of Europe, the Middle East and North Africa. The birds of the Western Palearctic, Vol. 3.—Oxford University Press, Oxford, 913 pp. Glutz von Blotzheim, U.N., K. M. Bauer, and E. Bezzel (eds.). 1977. Handbuch der Vé6gel Mittel- europas, Band 7, Charadriiformes (2. Teil). —Akademisch Verlagsgesellschaft, Wiesbaden. Gochfeld, M., J. Burger, and J. R. Jehl, Jr. 1984. The classification of the shorebirds of the world. Pp. 1-15, in J. Burger and B. L. Olla, eds., Behavior of marine animals, Vol. 5, Shorebirds: breeding behavior and populations.— Plenum Press, New York. Johnsgard, P. A. 1981. The plovers, sandpipers, and snipes of the world.— University of Nebraska Press, Lincoln, 493 pp. Johnson, N. K. 1980. Character variation and evolution of sibling species in the Empidonax diffi- cilis—flavescens complex (Aves: Tyrannidae). — University of California (Berkeley) Publications in Zoology 112:1-151. Lanyon, W. E. 1978. Revision of the Myiarchus flycatchers of South America.—Bulletin of the American Museum of Natural History 161:427-628. Mal’chevskii, A. S. 1980. (New data on sound communication in birds).— Vestnik Leningrad Uni- versity Biology 4:58—64 [in Russian]. Miller, E. H. 1983a. Structure of display flights in the Least Sandpiper.— Condor 85:220—242. . 1983b. The structure of aerial displays in three species of Calidridinae (Scolopacidae).— Auk 100:440-451. 1984. Communication in breeding shorebirds. Pp. 169-241, in J. Burger and B. L. Olla, eds., Behavior of marine animals, Vol. 5, Shorebirds, breeding behavior and populations. — Plenum Press, New York. , ——,, W. W. H. Gunn, and R. Harris. 1983. Geographic variation in aerial song of the Short- billed Dowitcher (Aves, Scolopacidae).— Canadian Journal of Zoology 61:2191-2198. Mundinger, P. 1979. Call learning in the Carduelinae: Ethological and systematic considerations. — Systematic Zoology 28:270-283. Pitelka, F. A. 1950. Geographic variation and the species problem in the shore-bird genus Lim- nodromus.— University of California (Berkeley) Publications in Zoology 50:1—108. Stein, R. C. 1963. Isolating mechanisms between populations of two populations of Traill’s Fly- catchers. — Proceedings of the American Philosophical Society 107:21—50. Strauch, J. G., Jr. 1976. The cladistic relationships of the Charadriiformes.— Ph.D. thesis, University of Michigan, Ann Arbor. ——. 1978. The phylogeny of the Charadriiformes (Aves): A new estimate using the method of character compatibility analysis.— Transactions of the Zoological Society of London 34:263- 345. Tikhonov, A. V., and S. Yu. Fokin. 1979. (Acoustic signalling and the behavior of shorebirds in early ontogenesis. I. Prenatal stages of development).—Nauchnye Doklady Vysshei Shkoly Biologicheskie Nauki 10:33—40 [in Russian]. VOLUME 97, NUMBER 4 811 . 1980. (Acoustic signalling and behavior of shorebirds in early ontogenesis. II. Signalling and the behavior of nestlings).—Nauchnye Doklady Vysshei Shkoly Biologicheskie Nauki 10:45-— 54 [in Russian]. —. 1981. (Acoustic signalization and behavior during the nesting period in snipe).— Byulleten’ Moskovskoga Obshchestva Ispytatele1 Prirody Otdel Biologicheskii 86:31—42 [in Russian]. Veprintsev, B. M. 1982. (Birds of the Soviet Union. A sound guide). (2 long-playing discs).—All- Union Studies for Recorded Sound. U.S.S.R. [notes and narration in Russian]. Warham, J., and B. D. Bell. 1980. The birds of Antipodes Island, New Zealand. —Notornis 26:121- 169. (EHM) Vertebrate Zoology Division, British Columbia Provincial Museum, Victoria, B.C. V8V 1X4, Canada, and Biology Department, University of Victoria, Victoria, B.C. V8W 2Y2, Canada. (WWHG) Box 738, Bobcaygeon, Ontario KOM 1A0, Canada. (JPM) Vertebrate Biology, Academy of Natural Sciences, 19th and the Parkway, Philadelphia, Pennsylvania 19103, U.S.A. (BNV) Institute of Bio- logical Physics, U.S.S.R. Academy of Sciences, Pushchino-on-Oka, Moscow Re- gion 142292, U.S.S.R. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 812-826 TAXONOMY OF SOME SPECIES OF PHALLODRILUS (OLIGOCHAETA: TUBIFICIDAE) FROM THE NORTHWEST ATLANTIC, WITH DESCRIPTION OF FOUR NEW SPECIES Christer Erséus Abstract.—New material of species of Phallodrilus Pierantoni from Georges Bank (SE of Massachusetts), the continental shelf off New Jersey, Maryland, and Virginia, and littoral sediments in Barbados is described. Four of the species, P. boeschi, P. flabellifer, P. caviatriatus, and P. davisi, are new to science. New taxonomic information is also provided for P. sabulosus Erséus, P. coeloprostatus Cook, and P. biprostatus (Baker and Erséus) n. comb. Peosidrilus Baker and Erséus is considered a junior synonym of Phallodrilus, as the definition of the latter is modified to include a few species possessing true penes. The relationships between Phallodrilus and some other genera within the subfamily Phallodrilinae are dis- cussed. In 1979 the marine tubificid genus Phallodrilus Pierantoni, 1902, was revised by the present author (Erséus 1979a) to include 23 species. Subsequently, several species have been added (Giere 1979; Erséus 1980a, b, 198la, 1983a, 1984a; Baker 1981; Giani and Martinez-Ansemil 1981), whereas two species have been transferred to other genera (Erséus 1981b, 1982). Thus the genus has comprised no less than 39 species to date. This makes Phallodrilus by far the largest genus within the subfamily Phallodrilinae, and yet many species are still undescribed (present paper; Erséus, unpublished). However, Phallodrilus is a heterogeneous taxon; for instance, it has hitherto been defined (Erséus 1979a, with slight emen- dation by Baker 1981) to include both species with and without an alimentary canal. In addition, although the general configuration of the male efferent duct is the same in all the species (vas deferens entering atrium apically or sub-apically, and prostate glands always two per atrium), the terminal part of the duct exhibits considerable variation throughout the genus. In some species the atrium opens directly on the body surface, whereas in others, there are various kinds of pseu- dopenes developed (with or without copulatory sacs). In some, the ““‘pseudopenial”’ structure has evolved so far as to approach the situation of a true penis. One such species is P. minutus Hrabé, 1973 (see Erséus and Kossmagk-Stephan 1983), another one P. vanus, recently described and discussed by Erséus (1984a). One phallodriline species with a conspicuous (true) penis was regarded as a member of a separate, monotypic genus by Baker and Erséus (1979: Peosidrilus biprostatus), despite the fact that the rest of the male duct and the penial setae of that species clearly are Phallodrilus-like. The present paper treats some additional material of Phallodrilus from the east coast of the U.S. and from Barbados. Some of the worms represents new species, all exhibiting various degrees of elaboration of the male terminalia in a way which bridges the gap between Phallodrilus and Peosidrilus. It is therefore proposed here VOLUME 97, NUMBER 4 813 that Peosidrilus becomes a junior synonym of Phallodrilus. The paper should, however, be regarded only as a partial revision of the genus; in a forthcoming paper (Erséus, in press), the gutless species of Phallodrilus will be scrutinized and taxonomically revised. Material and Methods Most of the material used in this study was collected during benthic surveys on the continental shelf off the east coast of the U.S. These projects are (1) the Georges Bank Infauna Monitoring Program, (2) a Bureau of Land Management baseline marine study, and (3) the ““Philadelphia Dump Site”’ study, the latter two made by Virginia Institute of Marine Science off the coasts of New Jersey through Virginia. The specimens were stained with haematoxylin or paracarmine and mounted whole in Canada balsam. The material of P. caviatriatus n. sp. was collected by the author during a stay at the Bellairs Research Institute of McGill University, St. James, Barbados. The worms were fixed in Bouin’s fluid; the anterior end of one specimen was subse- quently sectioned and stained in Heidenhain’s haematoxylin and eosin, whereas the remaining individuals were stained in paracarmine and mounted whole in Canada balsam. The type-material of the new species has been deposited in the United States National Museum of Natural History (USNM), Washington, D.C. Phallodrilus sabulosus Erséus, 1979 Phallodrilus sabulosus Erséus, 1979a:188-—189, figs. 1-2.—Erséus and Loden 1981: 820-821, fig. 1A. New material examined.—Two specimens from W of New Jersey, U.S.A., 39°22.1'N, 74°15.6'W, 15.5 m, medium to coarse sand, Aug 1976. Remarks.—Only one of the new specimens is complete. It is 5.5 mm long and consists of about 45 segments, but its posterior end is not fully developed. Both worms are mature, but pre-copulatory. The penial setae (see Erséus 1979a: figs. 1-2) are one at each side, ventrally in XI; the spermathecal setae (Erséus 1979a: - fig. 1) also one at each side, ventrally in X. This species was previously known only from Florida (Erséus 1979a; Erséus and Loden 1981). Thus the present record considerably extends its known dis- tribution to the north. Habitat. —Sublittoral, largely coarse sands, known from 3—15.5 m depth. Distribution. —East coast of the U.S.A., from New Jersey to Florida. Phallodrilus coeloprostatus Cook, 1969 Fig. 1 Phallodrilus coeloprostatus Cook, 1969:16-17, fig. 5.—Brinkhurst and Jamieson 1971:565, fig. 8.38D—F.—Erséus 1979a:189-190, fig. 4. New material examined.—Four specimens from Georges Bank, SE of Massa- chusetts, U.S.A., May 1983: two from 40°39.6'N, 67°45.8'W, 78 m; one from 40°39.8'N, 67°46.1'W, 78 m; and one from 40°39.5'N, 67°45.4'W, 78 m; all sites 814 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Phallodrilus coeloprostatus: A, Free-hand drawing of anterior seta; B, Free-hand drawing of posterior seta; C, Free-hand drawing of penial seta; D—E, Spermathecae; F, Lateral view of male duct. a, atrium; pp, pseudopenis; pr |, anterior prostate gland; pr 2, posterior prostate gland; ps, penial seta; vd, vas deferens. coarse sand. Four specimens from off Maryland, U.S.A., 38°16'56’N, 74°15’41°W, 28 m, sediment unknown, Apr 1979. Description of material from Georges Bank. — Length (fixed specimens) 3.6—4.2 mm, 48-54 segments. Diameter at XI, 0.17—0.23 mm. Clitellum extending over X-—XI in two of the specimens (see below; location of genital pores), poorly de- veloped in the others. Somatic setae (Fig. 1 A—B) bifid, with upper tooth shorter and thinner than lower; teeth slightly more diverging in post-clitellar setae than in anterior setae. Bifids 35-45 wm long, 1.5—2 wm thick, 3—4 per bundle anteriorly, (2)3 per bundle in post-clitellar segments. Ventral setae of X (2 specimens), or IX (1 specimen), or VIII (1 specimen) modified into penial bundles (Fig. 1F, ps), each of which contains 9-13 straight, ectally hooked (Fig. 1C) penial setae, 35- 40 um long, about 1.5 um thick. Male pores as small bursae in line with ventral somatic setae, posteriorly in VIII, IX or X. Spermathecal pores in lateral lines in most anterior part of VII, VIII or IX. Pharyngeal glands poorly visible, but present in V—VI of at least one of the specimens. Male genitalia (Fig. 1F) paired. Vas deferens not observed in full length in available material, but enters atrium more or less apically. Atrium comma- VOLUME 97, NUMBER 4 815 A B E Fig. 2. Phallodrilus boeschi, free-hand drawings of setae: A, Anterior seta; B, Posterior, dorsal seta; C, Penial seta. shaped to almost horseshoe-shaped, about 85 wm long, 25-35 wm wide, with very thin outer lining, and thick, granulated inner epithelium. Anterior prostate gland very large, lobed, attached to ental end of atrium near entrance of vas deferens. Posterior prostate smaller, stalked, attached to ectal part of atrium. Ectal part of atrium abruptly tapering and terminating in small styliform pseudopenial papilla, which protrudes from wall of male bursa. Spermatheca (Fig. 1D-E) 100-150 wm long, 12-37 wm wide, consisting of short ducts and slender, thin-walled ampullae, latter with random masses of sperm. Remarks.— Morphologically the new specimens from Georges Bank basically conform to the original description of P. coeloprostatus from Cape Cod Bay by Cook (1969), and they are very similar to the specimens of that species reported from off New Jersey by Erséus (1979a). However, the position of the genital organ in the new material is remarkable. In none of the Georges Bank specimens are the spermathecae and male ducts located in segments X—XI as is normal for a tubificid, and as was known for all previously reported individuals of P. coelo- prostatus, but they are found in VIJ—VIII, VIII-IX, or [X—X. This forward shift is even more strongly expressed by the position of the testes (not mentioned in description above): in only one of the four specimens, the testes are located in the spermathecal segment (which is the normal tubificid situation); in all the other worms the testes are located more anteriorly, in the segment next or second next to that bearing the spermathecae. The four worms from off Maryland exhibit a more moderate tendency towards forward shift of the genitalia; in two individuals the genitalia are in IX—X (with testes in IX), whereas the other two are “normal,”’ with genitalia in X—XI (testes in X). It must be concluded here that the specimens with an aberrant location of the genitalia should not be regarded as a separate species, but as variant of P. coeloprostatus. For some reason not yet understood, the otherwise so firmly fixed position of the tubificid genitalia has loosened in some populations of this species. In all other morphological respects, the specimens are identical. Forward shift of genital organs has been reported for individuals of marine Tubificidae before (e.g., Erséus 1981b, Giereidrilus inermis; 1983a, Bathydrilus sandersi and B. argentinae; Erséus and Davis, 1984, a new species of Adelodrilus). 816 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON In brief, P. coeloprostatus (sensu lato) can be described as follows: Length 3.6— 10 mm, 38-60 segments. Somatic setae, with upper tooth shorter and thinner than lower, 3—5 per bundle anteriorly, (2)3—4 per bundle posteriorly. Penial setae 8-13 per bundle, in fan-shaped formations, and with ectal ends clubbed and hooked (Fig. 1C). Spermathecal pores in lateral lines. Pharyngeal glands extending into VI. Atria more or less horseshoe-shaped, 85—130 um long, ending in small, protruding pseudopenial papillae within a pair of shallow male bursae. Prostates very large, lobed. Habitat.—Sublittoral, largely coarse sands, known from 3.4—78 m depth. Distribution. —East coast of the U.S.A., from Massachusetts (Cape Cod Bay and Georges Bank) to Maryland. Phallodrilus boeschi, new species Figs. 2-3 Holotype. —USNM 96042. Type-locality. — Off Virginia, U.S.A., 37°06.1'N, 74°33.4’W, 180-200 m, mixed sand, Mar 1976. Paratypes.—USNM 96043-96046. Four specimens from off Maryland, U.S.A., 38°17.5'N, 74°41.0'W, 29 m, medium to coarse sand, Mar 1976. Other material examined. — Author’s collection: three specimens from the same site as the paratypes; one specimen from off New Jersey, U.S.A., 39°43.6’N, 73°34.8'W, 37 m, medium sand, Mar 1976; and one specimen from Georges Bank, SE of Massachusetts, U.S.A., 40°39.5'N, 67°43.3’W, 72 m, medium to coarse sand, Jul 1982. Etymology.—The species is named for Dr. D. F. Boesch (now at Lumcon, Chauvin, Louisiana), principal investigator of the Bureau of Land Management baseline study from which most of the material of this species originates, and who first drew my attention to the large collection of marine oligochaetes in the Virginia Institute of Marine Science. Description. — Length (fixed specimens) 2.8—4.2 mm, 28-39 segments. Diameter at XI, 0.11-0.15 mm. Clitellum extending over 2 X—XIII. Somatic setae (Figs. 2A-B) bifid, 25-45 um long, 1—2 um thick, 3—5 per bundle anteriorly, 2—3(4) per bundle posteriorly. Bifids with upper tooth much shorter and thinner than lower. Anterior setae (Fig. 2A) with lower tooth straight, almost parallel to upper tooth. Posterior dorsal setae (Fig. 2B) with lower tooth characteristically curved down- wards, somewhat prolonged. Setae of mid-body and posterior ventral bundles intermediate. Ventral setae of XI modified into penial bundles (Figs. 3A—B, ps), each of which contains about 10-15 (exact number very difficult to establish) small, straight setae, 20-26 wm long, maximally 1 wm thick. Penial setae ectally clubbed, with a strongly bent apical tooth (Fig. 2C). Male pores in line with ventral somatic setae, posterior to middle of XI. Spermathecal pores near lateral lines, in most anterior part of X. Pharyngeal glands inconspicuous, but extending as far back as into VII or VIII(1X?). Male genitalia (Fig. 3) paired. Vas deferens coiled, 7-9 wm wide, much longer than atrium. Atrium small, more or less comma-shaped, about 35 um long, 12-20 um wide, with very thin (up to 1 wm) muscular lining; ectal part of inner epithelium of atrium granulated, cilia not seen. Prostate glands compact and stalked. Anterior prostate attached with broad stalk to middle of atrium. Posterior VOLUME 97, NUMBER 4 817 vd prips pr2 = sl 50 um ps MES pp Fig. 3. Phallodrilus boeschi, lateral view of spermathecae and male ducts in segments X-XI: A, Specimen from Georges Bank; B, Specimen from off Virginia (holotype). s, spermatheca; other ab- breviations as for Fig. 1. prostate more ectally attached to (posterior face of) atrium. Atrial opening a simple pore lateral to penial setae, but in some specimens ectal end of atrium somewhat protruded to form small pseudopenis (Figs. 3A—B, pp). Spermathecae (Figs. 3A— B, s) 50-65 um long, maximally 20-30 um wide, with indistinct and very short ducts, and thin-walled ampullae; latter with random masses of sperm. Remarks.—This species is closely related and superficially very similar to P. coeloprostatus. However, it is easily distinguished from the latter by the shape of its posterior, dorsal setae (Fig. 2B), and by the small size of its atria; length only about 35 wm as opposed to 85-130 um for P. coeloprostatus. Habitat.—Sublittoral, medium to coarse sand, known from 29-200 m depth. 818 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Distribution. —East coast of the U.S.A., from Massachusetts (Georges Bank) to Virginia. Phallodrilus flabellifer, new species Fig. 4 Holotype.—USNM 96047. Type-locality.— Georges Bank, SE of Massachusetts, U.S.A., 40°39.6'N, 67°45.8'W, 78 m, coarse sand, May 1983. Paratypes.—USNM 96048-96049. Two specimens from Georges Bank: cne from 40°39.5'’N, 67°46.2'W, 79 m, medium to coarse sand, May 1983; and one from 40°39.5'N, 67°45.4'W, 78 m, coarse sand, May 1983. Other material examined.—Author’s collection: one specimen from the type- locality. Etymology.—The name flabellifer (Latin for “‘fan-bearing’’) alludes to the fan- shaped arrangement of the penial setae of this species. Description. — Length (fixed, complete specimens) 3.0—3.3 mm, 29-34 segments; One specimen, which is not complete but very stretched, is 4.5 mm long, but still only consisting of first 28 segments. Diameter at XI, 0.11—0.16 mm. Clitellum extending over XI and at least anterior half of XII, but poorly developed in available specimens. Somatic setae (Fig. 4A) bifid, with upper tooth thinner and slightly shorter than lower. Bifids 35-45 um long, 1—1.5 wm thick, 2 per bundle throughout body. Ventral setae of XI modified into penial bundles (Fig. 4C, ps), each of which containing 13-15 straight, ectally hooked (Fig. 4B) setae, arranged in fan-shaped formation. Penials 30-37 um long, about 1.5 wm thick. Male pores in line with ventral setae posteriorly in XI. Spermathecal pores in line with ventral setae in most anterior part of X. Pharyngeal glands in III-IV(V), poorly developed. Male genitalia (Fig. 4C) paired. Vas deferens about 5 wm wide, thin-walled and very long, entering apical end of atrium. Atrium somewhat spindle-shaped, only slightly curved, 60-80 um long, 17-25 wm wide, with 1-2 wm thick outer lining of muscles, and inner granulated and ciliated epithelium. Lumen of atrium wide. Prostate glands mod- erately large, anterior one attached to apical end of atrium, near entrance of vas deferens, posterior one to ectal end of atrium. Atrium opening directly on body surface immediately anterior to ectal ends of penial setae. Spermathecae with distinct ducts, 35-50 um long, 11-14 wm wide, and thin-walled ampullae, 70- 120 um long, 23-35 wm wide; latter with loose and random sperm. Remarks.—The fan-shaped penial bundles of P. flabellifer are reminiscent of those of P. coeloprostatus and P. boeschi, but the new species differs from the others in two important respects: (1) its somatic setae are arranged in bisetal bundles throughout the body, whereas most bundles of P. coeloprostatus and P. boeschi each contain 3, 4 or 5 setae; and (2) its atria are spindle-shaped and hollow (lumen and ciliation easily seen), whereas the atria are of other shapes and with very narrow lumens in the other two species. Phallodrilus obscurus Cook, 1969, which was redescribed on the basis of fully mature specimens by Erséus (1980a), is possibly also a close relative of P. fla- bellifer. That species, however, has a very high number of somatic setae (up to 7 VOLUME 97, NUMBER 4 . 819 A B a ps l 100 um l Fig. 4. Phallodrilus flabellifer. A, Free-hand drawing of somatic seta; B, Free-hand drawing of penial seta; C, Lateral view of spermatheca and male duct in segments X—XI. Abbreviations as for Fig. 3. per bundle anteriorly) and its penial setae are more or less parallel within the bundles, not arranged in a fan as in P. flabellifer. Habitat.—Sublittoral, largely coarse sand, known from 78-79 m depth. Distribution.— Known only from Georges Bank, SE of Massachusetts. Phallodrilus biprostatus (Baker and Erséus, 1979), new combination _ Peosidrilus biprostatus Baker and Erséus, 1979:506—508, figs. 1-2 [Note that due to a typographical error the legend of fig. 2 was deleted from a part of the circulation of this publication. It should read: “Fig. 2. Peosidrilus biprostatus: Lateral view of spermathecea and male genitalia in segments X and XI. Ab- breviations: a, atrium; p, penis; pr | and pr 2, prostate glands; psa, penial sac; pse, penial setae; s, spermatheca; vd, vas deferens.’’].—Erséus and Loden 1981: 819-820. Remarks.— This species was described from sublittoral sediments off New Jer- sey and North Carolina (Erséus and Baker 1979), and subsequently reported from a similar habitat off the E coast of Florida (Erséus and Loden 1981). The species is characterized by a pair of large penes with bulbous bases, sepa- rated from the atria proper by constrictions, and P. biprostatus was, because of this striking feature, assigned to a monotypic genus, Peosidrilus Baker and Erséus, 1979. However, in other important characters, such as the penial setae (apically 820 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON clubbed and hooked), atria and prostate glands, the species is very similar to most of the Phallodrilus species treated in the present paper. It is now concluded that the penis of Peosidrilus biprostatus is just an elaboration of the pseudopenial structure found in several Phallodrilus forms, and consequently, the species is here transferred to the latter genus. Habitat.—Sublittoral, medium to very coarse sands, known from 5.5—73 m depth Distribution.—East coast of the U.S.A., so far recorded from New Jersey to Florida, but the species has recently been found on Georges Bank, SE of Mas- schusetts (Mr. D. Davis, pers. comm.). Phallodrilus caviatriatus, new species Rigas Holotype.—USNM 96050. Type-locality.— River Bay, NE Barbados, West Indies, inner end of bay at mouth of a stream, low water mark, coarse sand, probably often subject to brackish- water conditions; 27 Oct 1979. Paratypes.—USNM 96051-96052. One sectioned and one whole-mounted specimen from the type-locality. Other material examined.—Author’s collection: one specimen from the type- locality. Etymology.—The name caviatriatus refers to the hollow (=Latin “cavus’’) nature of the atrium of this species. Description. — Length (fixed specimens) 3.9—4.6 mm, 40—45 segments. Diameter at XI, 0.23-0.29 mm. Clitellum extending over 4, X—*4 XII in whole-mounted paratype, poorly developed in other specimens. Somatic setae (Fig. 5A) bifid, with upper tooth thinner and slightly shorter than lower, and often with thin ligament connecting tip of lower tooth with setae shaft. Bifids 35-50 um long, 1.5—2.5 um thick, 2—3 per bundle anteriorly, 2 per bundle in postclitellar segments (occa- sionally ““bundles”’ unisetal, or missing). Ventral setae of XI modified into penial bundles (Fig. 5D, ps), each of which containing about 13-18 (exact number often difficult to establish) straight, ectally clubbed setae, 40-55 um long, about 2 um thick. Ectal “‘clubs” of penials bearing a strongly curved apical tooth, but only seen in side view (see Fig. 5B). Penial setae located immediately ventral to male openings, at bases of penes. Male pores in line with ventral setae, posteriorly in XI. Spermathecal pores slightly dorsal to line of ventral setae, in most anterior part of X. Pharyngeal glands extending into V, and most anterior part of VI. Male genitalia (Fig. 5C—D) paired. Vas deferens 9-12 um wide, slightly shorter than atrium, entering apical, ental end of atrium. Atrium cylindrical, hollow, characteristically bent at middle, with conspicuous circular muscles, 5—9 um thick, and ciliated inner epithelium; epithelium much thinner in ental part than in ectal part of atrium; atrium about 150-200 um long (exact length difficult to establish), ental part 45-60 um wide, ectal part 30-35 um wide; ectal part tapering into slender duct leading to wedge-shaped, rounded penis (cf. Fig. 5C). Two compact, small prostate glands present, one at entrance of vas deferens, another located posterior to, and attached to, atrial duct; stalks of prostates not well developed. Sperma- VOLUME 97, NUMBER 4 821 50 um Fig. 5. Phallodrilus caviatriatus: A, Free-hand drawings of somatic setae; B, Free-hand drawings of penial setae, as seen from different angles; C, Section through terminal part of atrium and penis; D, Lateral view of spermatheca and male duct in segments X—XI. p, penis; other abbreviations as for Fig. 3. thecae (Fig. 5D, s) slender, consisting of 80-90 wm long ducts, and 70-115 um long, 40—45 wm wide ampullae; ducts with bulbous outer portion, 30—45 um wide, and narrow inner portion, only about 10-12 um wide; ducts gradually widening into ampulla, having thick walls and containing random sperm. Remarks.—The hollow ental half of the atrium is the most striking feature of P. caviatriatus, and it clearly distinguishes the species from other Phallodrilus. The penes are similar to, but not quite as pendant and deeply located within the penial sacs as, those of P. biprostatus. Habitat.— Intertidal coarse sand. Distribution.— Known only from the type-locality in Barbados. 822 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 100 um | Fig. 6. Phallodrilus davisi: A, Free-hand drawings of setae; B, Lateral view of spermatheca and male duct in segments X—XI. Abbreviations as for Fig. 5. Phallodrilus davisi, new species Fig. 6 Holotype.—USNM 96053. Type-locality.— Georges Bank, SE of Massachusetts, U.S.A., 40°39.6'N, 67°45.8'W, 78 m, coarse sand, May 1983. Paratypes.—USNM 96054-96055. Two specimens from Georges Bank: one from 40°39.5'N, 67°46.5’W, 79 m, coarse sand, May 1983; and one from 40°39.4’N, 67°46.9'W, 80 m, medium to coarse sand, May 1983. Other material examined.— Author’s collection: one specimen from 40°39.3'W, 67°46.2'W, 79 m, medium to coarse sand, May 1983. Etymology.—This species is named for Mr. Dale Davis (Battelle New England Marine Research Laboratory), who very kindly provided the material of Phal- lodrilus from Georges Bank. Description. — Length (fixed specimens) 3.4—4.4 mm, 36—41 segments. Diameter at XI, 0.12-0.18 mm. Clitellum extending over at least XI-'’4 XII (not fully developed in some specimens). Setae (Fig. 6A) bifid, with upper tooth thinner and shorter than lower. Setae 40-70 wm long, 1.5—2.5 um thick, (2)3—4 per bundle throughout body except for ventral bundles of XI, which are absent. Male pores in line with ventral setae in posterior part of XI. Spermathecal pores in line with ventral setae in most anterior part of X. Pharyngeal glands in III-IV, but poorly developed. Male genitalia (Fig. 6B) paired. Vas deferens about 4—5 um wide, long and thin-walled, entering atrium subapically. Atrium sigmoid- to comma-shaped, 65-85 um long, 21-32 wm wide, with very thin outer lining of muscles and thick inner granulated and (at least VOLUME 97, NUMBER 4 823 partly) ciliated epithelium; granulation of atrium confined to drop-shaped aggre- gations (each one an epithelial cell?). Anterior prostate gland large, attached to apical end of atrium. Posterior prostate smaller than anterior one, stalked and attached to most ectal end of atrium. Atrium terminating in cylindrical, ectally somewhat pointed penis, latter pendant within penial sac. Penis 25-30 um long, 6-12 um wide at middle, hollow, and appearing stiff, but cuticular lining not thickened into sheath. Spermathecae (Fig. 6B, s) slender, about 125 um long, maximally 25—40 um wide, consisting of ectal ducts, gradually widening into thin- walled ampullae; lumen of latter often incised at one or more places; sperm as random masses. Remarks. —Phallodrilus davisi is the only known species of Phallodrilus which combines the possession of true, pendant penes with the lack of penial setae. Habitat. —Sublittoral, largely coarse sands, known from 78-80 m depth. Distribution.— Known only from Georges Bank, SE of Massachusetts. Modified Definition of the Genus Phallodrilus Pierantoni, 1902 Diverse group of predominantly marine tubificids. Hair setae absent. Penial setae generally present. Spermathecal setae occasionally present. Male and sper- mathecal pores paired, generally in segment XI and X, respectively (exceptions, see P. coeloprostatus above, and P. postspermathecatus Erséus, 1980b). Coelomo- cytes, if present, small and sparse, not of the “‘rhyacodriline-type.”’ Male efferent ducts paired. Vasa deferentia ciliated, not dilated and storing sperm, but (generally) narrow and clearly set off from atria. Vasa deferentia join apical or subapical, ental ends, but never ectal halves of atria. In some species copulatory structures present, as protrusible pseudopenes or non-cuticularized true penes. Two pairs of prostate glands present; the anterior pair always attached to atria near entrances of vasa deferentia, posterior pair generally attached to ectal ends of atria; prostates often stalked. Spermathecae paired, generally longer than wide. Sperm in loose masses or bundles in spermathecae; spermatozeugmata and external spermatophores absent. Type-species.— Phallodrilus parthenopaeus Pierantoni, 1902. Discussion The definition of the genus Phallodrilus provided above has been modified to include some species with true penes (biprostatus, caviatriatus and davisi; all treated in the present paper). It excludes, however, some other phallodriline species with penes, such as the members of the genera Aktedrilus Kn6llner, 1935, and Bacescuella Hrabé, 1973 (see Baker and Erséus 1982; Erséus 1978, 1980c, 1984a— c; Erséus and Kossmagk-Stephan 1982; Finogenova and Shurova 1980; Hrabé 1973; Strehlow 1982). In these genera the spermathecae, when present, are un- paired, middorsal, and the penes are certainly not homologous to those of the three Phallodrilus species mentioned above. In addition, in Bacescuella sperm are transferred by means of external spermatophores, a mode not known for any other genus within the Phallodrilinae. In the previous definition (Erséus 1979a), it was stated: “alimentary canal absent in some species.” This sentence is now deleted, as the gutless species of Phallodrilus will be removed from that genus in a separate revision (Erséus, in press). 824 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A great taxonomic problem with Phallodrilus is that the male efferent ducts of most of its species conform to a basic phallodriline plan, from which the male ducts of most other genera within the subfamily can be derived. This means that Phallodrilus, as defined here and previously, is a paraphyletic rather than a strictly monophyletic taxon (a “‘paraphyletic” group does have and includes a common ancestor, but it does not include all the descendants of that ancestor). Unfortu- nately, there does not seem to be any way out of this dilemma. A widening of Phallodrilus to a level where one would be confident that it was a strictly mono- phyletic group would imply that virtually all of the 15 phallodriline genera were united to form one large genus, which would hardly be beneficial for the classi- fication of the very diverse subfamily Phallodrilinae. On the other hand, a splitting of today’s Phallodrilus into smaller, monophyletic genera is, at present at least, impossible. One can of course recognize several species groups within Phallodrilus, which within themselves are monophyletic and can be defined on the basis of their synapomorphies, but if all these groups were raised to the level of genera, the remaining species would just become a large, most probably polyphyletic group, defined only by its plesiomorphic characters. It is, therefore, considered appropriate to accept that Phallodrilus remains para- phyletic, but to define it in a way which allows for some short lineages of derived species; P. biprostatus, P. caviatriatus and P. davisi are examples of such derived species. It is a much different situation with large, coherent groups, such as Ak- tedrilus (references given above) and Adelodrilus Cook, 1969 (see Erséus 1983b; Erséus and Davis, 1984). Aktedrilus now contains 16 species, Adelodrilus 14, and they are both easily recognized and classified as genera on the basis of their shared derived characters (synapormophies) (Aktedrilus: unpaired dorsal sper- matheca, well developed penes, lack of penial setae, and littoral habitat; Adelo- drilus: dilated vasa deferentia storing sperm, and highly modified penial setae). Most of the Phallodrilus species treated in the present account, viz. P. coelo- prostatus, P. boeschi, P. flabellifer, P. biprostatus, and P. caviatriatus, share the feature of small, multiple penial setae with characteristically clubbed and apically hooked tips (Figs. 1C, 2C, 4B, 5B; Baker and Erséus 1979, fig. 1C). It is noteworthy that these are all northwest Altantic species and that there are no other known Phallodrilus with this kind of penial setae in other areas of the world. It seems feasible to regard these species as a monophyletic group within the genus and that it has evolved and radiated in the northwest Atlantic. This group, however, is not the only one within the Phallodrilinae with the “clubbed and hooked”’ penial setae. Such setae are characteristic also for most species within Adelodrilus (see Erséus 1983b) and the monotypic Bermudrilus Erséus, 1979 (see Erséus 1979b). The geographical records of the latter genera are largely confined to the East coast of the U.S.; one species is known also from Bermuda and a few species are described from Europe (Erséus 1983b). It should be noted that the knowledge of marine Tubificiade from the northwest Atlantic is perhaps better than that from any other part of the world, but if Adelodrilus and Bermudrilus were present in, for instance, the Pacific Ocean, one would have expected these genera to turn up in the fairly extensive collections that now have been made there (the work by H. R. Baker in British Columbia; Erséus 198 1a, 1984a, and unpublished). It therefore appears likely that the Adelodrilus-Ber- mudrilus lineage originated in the first area, or at least within a fauna the de- VOLUME 97, NUMBER 4 825 scendants of which today inhabit this region. This would imply that the Adelo- drilus-Bermudrilus assemblage split off relatively late from their ancestral stock within (most probably) Phallodrilus, a stock probably closely related to the Recent group of Phallodrilus treated in the present paper. The high number of closely related species of Adelodrilus within one restricted area (the U.S. east coast con- tinental shelf; Erséus 1983b, Erséus and Davis, 1984) demonstrates how suc- cessful and radiative speciation has been within this particular group. Acknowledgments Specimens from Georges Bank were collected as part of the Georges Bank Benthic Infauna Monitoring Program, supported by contract number 14-12-0001- 2912 from the U.S. Department of the Interior Minerals Management Service to Battelle New England Marine Research Laboratory. I am much indebted to Dr. D. F. Boesch, Dr. R. J. Diaz (Virginia Institute of Marine Science), and Mr. D. Davis, for providing the oligochaete material and important information; Mr. Davis also kindly read and commented upon the first draft of the manuscript; to Dr. F. Sander (former director of Bellairs Research Institute of McGill University), for working facilities in Barbados; to Mrs. Barbro Lofnertz, for skilful technical assistance; and to the Swedish Natural Science Research Council, for financial support. This paper is a contribution from the Bellairs Research Institute of McGill University, Barbados. Literature Cited Baker, H. R. 1981. Phallodrilus tempestatis n.sp., a new marine tubificid (Annelida: Oligochaeta) from British Columbia.— Canadian Journal of Zoology 59:1475-1478. , and C. Erséus. 1979. Peosidrilus biprostatus n.g., n.sp., a marine tubificid (Oligochaeta) from the eastern United States. — Proceedings of the Biological Society of Washington 92:505-— 509. , and 1982. A new species of Bacescuella Hrabé (Oligochaeta, Tubificidae) from the Pacific Coast of Canada.— Canadian Journal of Zoology 60:1951-1954. Brinkhurst, R. O., and B. G. M. Jamieson. 1971. Aquatic Oligochaeta of the world. Oliver & Boyd, Edinburgh. 860 pp. - Cook, D. G. 1969. The Tubificidae (Annelida, Oligochaeta) of Cape Cod Bay with a taxonomic revision of the genera Phallodrilus Pierantoni, 1902, Limnodriloides Pierantoni, 1903 and Spiridion Knollner, 1935.—Biological Bulletin 136:9-27. Erséus, C. 1978. Two new species of the little-known genus Bacescuella Hrabé (Oligochaeta, Tu- bificidae) from the North Atlantic.— Zoologica Scripta 7:263-267. —. 1979a. Taxonomic revision of the marine genus Phallodrilus Pierantoni (Oligochaeta, Tu- bificidae), with descriptions of thirteen new species.— Zoologica Scripta 8:187—208. 1979b. Bermudrilus peniatus n.g., n.sp. (Oligochaeta, Tubificidae) and two new species of Adelodrilus from the Northwest Atlantic.— Transactions of the American Microscopical Society 98:418—427. . 1980a. Redescriptions of Phallodrilus parthenopaeus Pierantoni and P. obscurus Cook (Oli- gochaeta, Tubificidae).— Zoologica Scripta 9:93—96. . 1980b. New species of Phallodrilus (Oligochaeta, Tubificidae) from the Arctic deep sea and Norwegian fjords. —Sarsia 65:57-60. . 1980c. Taxonomic studies on the marine genera Aktedrilus Kn6llner and Bacescuella Hrabé (Oligochaeta, Tubificidae), with descriptions of seven new species.— Zoologica Scripta 9:97— 111. . 198la. Taxonomic studies of Phallodrilinae (Oligochaeta, Tubificidae) from the Great Barrier 826 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Reef and the Comoro Islands with descriptions of ten new species and one new genus. — Zoologica Scripta 10:15-31. 1981b. Taxonomic revision of the marine genus Heterodrilus Pierantoni (Oligochaeta, Tu- bificidae). — Zoologica Scripta 10:111-132. 1982. Atlantidrilus, a new genus of deep-sea Tubificidae (Oligochaeta). —Sarsia 67:43—46. 1983a. Deep-sea Phallodrilus and Bathydrilus (Oligochaeta, Tubificidae) from the Atlantic Ocean, with descriptions of ten new species.— Cahiers de Biologie Marine 24:125-146. ——. 1983b. New records of Adelodrilus (Oligochaeta, Tubificidae), with descriptions of two new species from the North-west Atlantic.—Hydrobiologia 106:73-83. —. 1984a. The marine Tubuficidae (Oligochaeta) of Hong Kong and Southern China.— Asian Marine Biology 1:135-175. 1984b. Two new species of Aktedrilus (Oligochaeta, Tubificidae) from the West Atlantic. — Bulletin du Muséum National d’Histoire Naturelle, Paris, 4° sér., 5, section A, n° 4:(1983)105 1— 1057. —. 1984c. Interstitial fauna of Galapagos XX XIII. Tubificidae (Annelida, Oligochaeta).— Mi- krofauna Marina 1:191-198. —-. Taxonomy and phylogeny of the gutless Phallodrilinae (Oligochaeta, Tubificidae), with de- scriptions of one new genus and twenty-two new species. — Zoologica Scripta 13. [In press] ,and D. Davis. 1984. Three new species of Adelodrilus (Oligochaeta: Tubificidae) from Georges Bank (NW Atlantic). — Proceedings of the Biological Society of Washington 97:834-843. , and K.-J. Kossmagk-Stephan. 1982. A new species of Aktedrilus (Oligochaeta, Tubificidae) from the North Sea coast of the Federal Republic of Germany.— Zoologischer Anzeiger 209: 91-96. , and 1983. Redescription of Phallodrilus minutus (Oligochaeta, Tubificidae) based on ncw material from the North Sea coast of Germany.—Sarsia 68:229-231. —, and M. S. Loden. 1981. Phallodrilinae (Oligochaeta: Tubificidae) from the east coast of Florida, with description of a new species of Adelodrilus.— Proceedings of the Biological Society of Washington 94:819-825. Finogenova, N. P., and N. M. Shurova. 1980. A new species of the genus Aktedrilus (Oligochaeta, Tubificidae) of the littoral zone of the Sea of Japan. Pp. 65—69 in Pribrezhnyi plankton i bentos severnoi chasti yapanskogo morya.— Akademia Nauk SSSR: Vladivostok [in Russian]. Giani, N., and E. Martinez-Ansemil. 1981. Observaciones acerca de algunos Tubificidae (Oligo- chaeta) de la peninsula ibérica, con la descripcion de Phallodrilus riparius n.sp.— Annales de Limnologie 17:201-209. Giere, O. 1979. Studies on marine Oligochaeta from Bermuda, with emphasis on new Phallodrilus species (Tubificidae).— Cahiers de Biologie Marine 20:301-314. Hrabé, S. 1973. A contribution to the knowledge of marine Oligochaeta, mainly from the Black Sea.— Travaux du Muséum d’Histoire Naturelle ‘Gr. Antipa’ 13:27-38. Strehlow, D. R. 1982. Aktedrilus locyi Erséus, 1980 and Aktedrilus oregonensis n.sp. (Oligochaeta, Tubificidae) from Coos Bay, Oregon, with notes on distribution with tidal height and sediment type.— Canadian Journal of Zoology 60:593-596. Swedish Museum of Natural History, c/o Department of Zoology, University of Géteborg, Box 25059, S-400 31 GGteborg, Sweden. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 827-833 AN EXTINCT SPECIES OF LEIOCEPHALUS FROM HAITI (SAURIA: IGUANIDAE) Gregory Pregill Abstract.— At least five species of the endemic West Indian lizard genus Lei- ocephalus are known only by fossils excavated from cave deposits in the Greater and Lesser Antilles. Fossils of yet another extinct species: were recently found in a crate of matrix that was collected over 50 years ago from caves near St. Michel de l’Atalye, Haiti. This new form, L. anonymus, has a partially closed Meckelian groove, a condition structurally intermediate between the fully open groove of the extinct species L. apertosulcus, and the completely closed and fused groove characteristic of all other living and fossil species of Leiocephalus. One of the first assemblages of fossil vertebrates reported from the West Indies came from cave deposits in north central Haiti, northeast of the town of St. Michel de l’Atalye in the Département L’Artibonite. Descriptions of these caves and the bones of mammals found therein were published in a series of papers by Gerrit S. Miller (1922, 1926, 1929). Alexander Wetmore (1922) described some of the fossil birds, and Arthur J. Poole (1929) wrote additional descriptions of several more sites in the vicinity. Yet except for a comment on lizards by Hecht (1951) and Rimoli’s (1976) review of the fossil rodents of Hispaniola, almost nothing has been written on the fossil material since its initial discovery. All of the fossils and unsorted matrix from the St. Michel caves that were collected in the 1920s had been deposited with the Division of Vertebrate Pa- leontology, National Museum of Natural History, Smithsonian Institution. Some of the matrix remained in two shipping crates, each bearing a hand-written label identifying the contents as “‘small material from caves of Haiti, contains small shells, lizard jaws, etc. Has been sorted for small mammal remains.” Each label had been initialed “P,”’ presumably for Arthur J. Poole. Inside one crate was a _ note stating ““Cave 1. Discarded material carefully worked over— Poole.” “Cave 1” possibly refers to the first of three caves that Poole (1929:59) called the “‘San Francisco group . . . situated about three miles to the west of L’Atalaye and about one mile east of the village of St. Michel.” While casually picking through this matrix recently I found that most of the fossils of mammals and birds had indeed been removed, yet abundant remains of lizards were still present. Among them were bones of the iguanid lizard Leiocephalus, a genus endemic to the West Indies and presently confined to the Greater Antillean Islands of Cuba, Hispaniola, and the Bahamas. Elsewhere in the West Indies at least five extinct species are known: Jamaica (Etheridge 1966a), the Dominican Republic (Etheridge 1964), Puerto Rico (Pregill 1981), and the Leeward Islands of Barbuda, Antigua and Anguilla (Etheridge 1964; Pregill, in prep.). The fossils from St. Michel share certain similarities with the extinct species Leiocephalus apertosulcus that was described by Etheridge (1965), based on ma- terial collected from cave deposits in the Cerro de San Francisco, La Estrellata 828 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Province, Dominican Republic. Yet consistent differences are apparent, especially in the partially closed Meckel’s groove in the St. Michel lizards. The dentary is structurally intermediate between the completely open groove of L. apertosulcus and the closed and fused condition obtained in all other species of the genus. These and other features of this new Leiocephalus are described below. Measurements were made with an eyepiece reticle on a Wild M8 steromicro- scope and read to the nearest 0.1 mm. Comparative skeletal material of Leiocepha- lus included representatives of 20 of 22 living species in the collections of the National Museum of Natural History, Smithsonian Institution (USNM), Museum of Comparative Zoology (MCZ), and Richard Etheridge (REE—San Diego State University). Notes and measurements on all fossil species of Leiocephalus were available from a previous study (Pregill 1981); the type material of L. apertosulcus was in hand. Estimates of snout to vent length of fossil individuals, hereafter referred to as SVL, are derived from the simple procedure of comparing ratios of measurements with modern individuals of known size (Pregill 1981). Leiocephalus anonymus, new species Fig. 1 Holotype.—USNM 340182, a right dentary in the collection of the Division of Vertebrate Paleontology. Locality and horizon.— An unspecified cave(s) near St. Michel de l’Atalye, Dé- partement L’ Artibonite, Haiti. Probably collected by Arthur J. Poole in December, 1927, in cave sediment of undetermined age, certainly no older than latest Pleis- tocene. Etymology.—Greek, anonymus, nameless; in reference to the fact that fossils of this species lay buried in a crate for over fifty years, unnamed and unknown. Referred material.—Dentaries: the type plus 18 left, 8 right nearly complete, and 30 fragments (USNM 340183-340185); Maxillae: 7 left, 7 right plus 17 fragments (USNM 340186-—340188); Premaxilla: 1 (USNM 349192): Frontals: 5 complete and 3 partial (USNM 340189); Parietal: 1 (USNM 340190); Pterygoid: 1 partial right (USNM 34191). Diagnosis. — A large species of Leiocephalus having Meckel’s groove open except for a short distance in the middle of the dentary, usually between the twelfth and fifteenth teeth where the upper and lower borders converge and touch. The open portions of Meckel’s canal expose a well-developed intramandibular septum. The pineal foramen is located wholly within the frontal bone. Description of Holotype.—Tooth row 15.2 mm; length overall 18.6 mm, and height posterior to last tooth 3.4 mm; 24 pleurodont teeth of which numbers 5, 7, 8, 10 and 18 missing, front to back; estimated snout to vent length of individual in life 120 mm. Teeth 1 through 4 plus number 6 simple, recurved and pointed; each slightly larger than tooth preceding; ninth tooth with small central cusp distinct from tooth shaft, and incipient lateral cusp on posterior side; tenth tooth and all those posteriad except for number 11 (missing crown) and number 24 (replacement tooth) with tricuspid crowns in fleur-de-lis shape typical of Leiocephalus. Anterior cusp of each tooth generally overlapping slightly posterior cusp of preceding tooth; shafts parallel-sided; approximately 35% of each tooth protruding above parapet VOLUME 97, NUMBER 4 829 Fig. 1. Holotype of Leiocephalus anonymus, new species, a right dentary (USNM 340182) in lingual (above) and labial (below) view. of jaw; teeth on holotype more abraded than those of some referred dentaries but their tricuspid condition clearly evident; on mesial side of dentary below dental shelf, Meckel’s groove exposed anterior to twelfth tooth and posterior to fifteenth; between twelfth and fifteenth teeth, upper and lower borders of Meckel’s groove converge and make contact, but do not fuse; within Meckel’s canal well- developed intramandibular septum extending from symphysis posterior to level of last tooth; septum deeply emarginate at its posterior border. Labial face of dentary convex throughout most of its length except for antero- ventral surface, where it is moderately flattened; horizontal row of six mental foramina perforating upper half of iabial face between second and fourteenth teeth; in front of surangular process on dorsolabial surface, a deep wedge-shaped scar extending forward to level of twenty-second tooth, marking former position of anterolabial arm of coronoid. Other dentaries.— Twenty-six other dentaries referred to this species are similar to the holotype in most respects. Some of these are nearly complete, others are badly worn, missing teeth or various processes and borders. Teeth missing from the holotype are present collectively on the referred specimens. The transition from simple to tricuspid teeth takes place rather abruptly between teeth 9 and 11. The number of teeth increases ontogenetically from 19 in the smallest dentary (tooth row of 9.2 mm) to 25 in the largest (tooth row 16.4 mm). These dentaries came from individuals estimated at 73 mm and 130 mm SVL, respectively. The average tooth row length for the 26 dentaries is 13.9 mm, an individual of ap- proximately 110 mm SVL. The scar of the coronoid overlap on the dorsolabial surface spans a distance equivalent to | to 4 posteriormost teeth (x = 2.1), and the number of mental foramina ranges from 4 to 8 (X = 5.5) placed between the second and fourteenth teeth. The most important diagnostic feature is the persistence of a partially closed 830 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Meckelian groove near the middle of the dentary. In most cases contact is made between the upper and lower borders of Meckel’s groove; in three specimens the borders actually fuse, and in two specimens Meckel’s groove is continuously open. Maxillae.— Fourteen nearly complete maxillae have tooth rows ranging from 11.3 mm (93 mm estimated SVL) to 15.4 mm (126 mm estimated SVL). The average tooth row length is 13.4 mm (110 mm estimated SVL). Tooth counts vary from 18 to 22, the number increasing with the size of the jaw. In all specimens the transition from simple to tricuspid teeth takes place at the eighth or ninth tooth. The teeth are otherwise similar to those on the dentary. The supradental shelf is curved medially at the anterior end where it forms the premaxillary process. A prominent palatine process protrudes medially just pos- terior to the middle of the shelf. On the labial side of the maxilla a tall nasal process rises almost vertically above the dental row. Anterodorsally, the nasal process turns sharply mediad above the fenestra exonarina. On the nasal process immediately above the tooth row is a horizontal series of from 6 to 9 maxillary foramina placed between the third and thirteenth teeth. In one specimen the foramina are placed between the second and eleventh teeth. An additional 1 to 3 foramina penetrate the nasal process above the first 2 to 6 maxillary foramina. Premaxilla.—A single premaxilla is referred to this species by its large size, and similarity to the premaxilla of other Leiocephalus. The rectangular incisive process is 3.6 mm wide, and bears seven simple, pointed teeth, as in all species of the genus. There is minimal inflection between the incisive process and the posteriorly directed nasal process, which is 1.4 mm wide at its base and 3.9 mm in length. The nasal process is thus relatively narrow and long in comparison to that of other species such as L. carinatus. Frontal. — Five of the frontal bones are nearly complete; three others are missing portions of the postfrontal or nasal processes. The largest has a midsagittal length of 12.1 mm, an interorbital width of 4.2 mm, and measures 12.2 mm across the parietal border. These same measurements on the smallest complete specimen are 9.9 mm, 3.3 mm, and 9.5 mm, respectively. Overall, the frontal bone of L. anonymus has a gentle dorsal curvature with a rugose surface. In all specimens the pineal foramen lies wholly within the frontal bone, near the middle of the parietal border. Two deep semicircular excavations anteriorly indicate the former overlap of the nasal bones. Ventrally, the olfactory canal is a smooth, deep fossa occupying the anterior third of the bone and bordered on either side by sharply descending subolfactory processes. Parietal.—The single parietal referred to this species is missing only the distal half of the left supratemporal process. Its width at the frontal border is 10.2 mm. The width across the ends of the supratemporal processes is approximately 12.7 mm, accounting for the broken left tip. The midsagittal length is 4.7 mm. The bone probably came from a subadult individual estimated at 97 mm SVL. The parietal table is rugose, like the frontals, and the sides converge postero- medially above the parietal fossa. On each of the anterolateral processes there is a distinct notch that in life articulates with the postorbital. Pterygoid.—A single right pterygoid is missing the distal half of the quadrate process but otherwise is complete. There are no pterygoid teeth. The bone mea- sures 5.9 mm across the tips of the palatine and ectopterygoid processes. A deep VOLUME 97, NUMBER 4 831 basipterygoid notch is present for articulation with the basipterygoid process of the basisphenoid. In Leiocephalus the basipterygoid notch is accentuated by a pronounced ventral shelf and the ectopterygoid process is excavated on the dor- somedial surface. Remarks. — Because of the unusual form of the dentary, Leiocephalus anonymus need only be distinguished from the extinct species L. apertosulcus. Indeed, one might argue that L. anonymus is simply a population of L. apertosulcus in which Meckel’s groove has become partially closed. But the two differ in other respects, and some features stand apart as much or more than those of other Leiocephalus. Unfortunately, only a few comparably diagnostic bones are known for both taxa. For example, the premaxilla is unknown in L. apertosulcus but most other major skull elements have been identified (Etheridge 1965). In L. anonymus the pre- maxilla is known, but the species is otherwise represented by comparatively few cranial bones. For now, whether or not these two species are each other’s closest relatives is an interpretation that depends largely on whether the open condition of Meckel’s groove is primitive or derived. In all species of Leiocephalus Meckel’s groove is closed and fused except for a small opening anteriorly. In two fossil species, L. etheridgei and L. partius (both from Puerto Rico), Meckel’s groove is open anteriorly as a narrow sulcus extending back a distance of 5 to 6 teeth (Pregill 1981). Of fifty dentaries referred to L. apertosulcus by Etheridge (1965) a continuously open groove was present in all but two specimens, wherein the upper and lower borders of the groove made contact below the twelfth to fourteenth teeth. Both species have a well-developed intramandibular septum (IMS), a structure found in other Leiocephalus only in the extinct species L. partitus, and the Antigua Bank form L. cuneus. Loss of the IMS in Leiocephalus and other tropidurines is thought to be a derived state correlated with closure and fusion of Meckel’s groove (Pregill 1981). However, the septum is present in L. partitus and L. cuneus, two species having the Meck- elian groove closed. In these, the persistence of the IMS may be the result of a reversal secondarily derived, or simply a retention of the primitive condition. I prefer the former interpretation because the complex dentary/postdentary artic- ulation is constructed about a closed septumless Meckle’s groove in most tropi- durines and all extant Leiocephalus, clearly a monophyletic genus (Etheridge 1966b; Pregill, in prep.). By this same reasoning, the open and nearly open groove of L. apertosulcus and L. anonymus may also be a reversal and thus a shared derived character. However, Etheridge and de Queiroz (in prep.) suggest that an open Meckel’s groove is plesiomorphic for iguanids. Hence, the character polarity of an open Meckel’s groove and the presence of an intramandibular septum in Leiocephalus is still conjectural. (A review of the various tropidurine genera under way by Richard Etheridge, Darrel Frost, and myself may cast light on this inter- esting character complex.) Another feature in which L. anonymus differs from L. apertosulcus is that the pineal foramen is located completely within the frontal bone. In L. apertosulcus and all other species except L. semilineatus and L. raviceps, the foramen is located at the frontoparietal suture, the primitive condition for iguanids (Etheridge and de Queiroz, in prep.) By comparison to L. apertosulcus, the frontal of L. anonymus is also proportionately wider between the orbits. This can be confirmed by visual inspection, and verified by comparing their ratios of interorbital width to mid- 832 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON sagittal length. The ratio averages 0.350 (0.339-0.377) in the five complete frontals of L. anonymus, and 0.315 (0.300-0.329) in frontals of L. apertosulcus. In fact, the difference in these ratios increases ontogenetically in this sample of individuals, estimated at from 85 mm to 130 mm SVL. Differences also exist in the shape of the supratemporal processes of the parietal. In Leiocephalus the sides of the parietal table converge posteromedially with increased body size (Etheridge 1966b), whereas the shape of the supratemporal processes remains more or less unchanged during growth. Those of L. apertosulcus are flared dorsoventrally and are deeply emarginate; in L. anonymus the processes are narrower and not as excavated. Differences in the maxilla and pterygoid bones between these two species are subtle; in fact, neither bone varies significantly among Leiocephalus. Pterygoid teeth are absent from both, but they have also been lost in most other species of the genus. The maxillae differ principally in that the transition to tricuspid teeth takes place variably between teeth 5 and 8 in L. apertosulcus, and rather abruptly at tooth 8 or 9 in L. anonymus. The transition to tricuspid teeth in Leiocephalus is subject to some ontogenetic increase (i.e., more posteriad), but there is an observable pattern throughout the genus. In some species the transition is consistently anterior at about the fourth or fifth tooth, as in L. cuneus. In other species such as L. cubensis the transition is as far posteriad as the tenth or eleventh tooth. The latter is probably the more derived condition (Pregill, in prep.). Leiocephalus apertosulcus may have obtained a larger maximum adult size than L. anonymus. My calculations concur with Etheridge’s estimates that L. aperto- sulcus was at least 150 mm snout-—vent length, and possibly as much as 200 mm. Leiocephalus anonymus, on the other hand, probably obtained an average max- imum size between 120 and 130 mm SVL. These figures may be influenced by sampling bias in the fossil record. It is curious, however, that both these species plus the other extinct forms L. jamaicensis and L. cuneus achieved a snout—vent length greater than any living species. Among living members only L. carinatus and L. melanochloris approach 120-130 mm. Characters in which the two species are similar include the transition to tricuspid teeth on the dentary, the length of the coronoid overlap on the dentary’s dorsolabial surface, and the number of mental and maxillary foramina. Of these, only the transition to tricuspid teeth has any phylogenetic utility. The other features vary individually and ontogenetically within Leiocephalus and are of little or no use as systematic characters. As with maxillary teeth, the transition to tricuspid teeth on the dentary occurs as far anteriorly as the third or fourth tooth in some species, and posteriorly at the eleventh or twelfth tooth in others. The distribution of Leiocephalus in the West Indies today is obviously relictual, yet the causes of extinction within the group are unclear. Some extinctions may be related to changes in climate and sea level at the end of the Pleistocene 10- 12,000 years ago (Pregill and Olson 1981), whereas others are probably much more recent and due to environmental disturbances wrought by early human colonization in the Antilles (Steadman et a/., 1984). Neither Miller (1922, 1926, 1929) nor Poole (1929) speculated much about the ages of the bone-bearing sediments from St. Michel, but like all other cave deposits thus far known from the West Indies, they are certainly no older than Late Pleistocene. Supposedly, some of the St. Michel sites were at least pre-Amerindian, others apparently were VOLUME 97, NUMBER 4 833 contemporaneous with early human occupation of the region. I did find bones of Rattus in the crates of matrix from which the fossils of L. anonymus were removed. Unfortunately, collecting techniques were often haphazard in earlier days and surface sediments could have been mixed with several deeper layers, or contam- inated in other ways. Hence, we cannot be totally confident that L. anonymus actually survived into post-Columbian time, that is, after Rattus had been intro- duced. A recent extirpation is not unlikely in view of the compelling evidence for human-caused extinction of small vertebrates in the West Indies (Steadman et al., 1984). Acknowledgments I wish to thank Robert Purdy (Division of Vertebrate Paleontology, USNM) for providing access to the fossil material, and Pere Alberch (MCZ), George Zug (USNM), Richard Etheridge (San Diego State University), and C. R. Schaff (MCZ) for the use of comparative specimens. Darrel Frost and Richard Etheridge com- mented on the manuscript. Figure 1 was drawn by Lynn A. Barretti. This study was supported by NSF grant DEB 8207347. Literature Cited Etheridge, R. 1964. Late Pleistocene lizards from Barbuda, West Indies.—Bulletin of the Florida State Museum 9(2):43-75. —. 1965. Fossil lizards from the Dominican Republic.—Quarterly Journal of the Florida Academy of Sciences 28(1):83-105. . 1966a. An extinct lizard of the genus Leiocephalus from Jamaica.— Quarterly Journal of the Florida Academy of Sciences 29(1):47—-59. . 1966b. The systematic relationships of West Indian and South American lizards referred to the iguanid genus Leiocephalus.—Copeia 1966:79-91. , and K. de Queiroz. Cladistic relationships within the Iguanidae and the origin of this family [manuscript in prep.]. Hecht, M. K. 1951. Fossil lizards of the West Indian genus Aristelliger (Gekkonidae).— American Museum Novitates No. 1538, 33 pp. Miller, G. S., Jr. 1922. Remains of mammals from caves in the Republic of Haiti.—Smithsonian Miscellaneous Collections 74(3):1-8. 1926. Exploration of Haitian caves.—Smithsonian Miscellaneous Collections 78(1):36—40. 1929. A second collection of mammals from caves near St. Michel, Haiti. —Smithsonian Miscellaneous Collections 81(9):1—30. Poole, A. J. 1929. Further exploration of Haitian caves.— Explorations and Field-work of the Smith- sonian Institution in 1928:55-62. Pregill, G. K. 1981. Late Pleistocene herpetofaunas from Puerto Rico.— University of Kansas Mu- seum of Natural History Miscellaneous Publication No. 71, 72 pp. , and S. L. Olson. 1981. Zoogeography of West Indian vertebrates in relation to Pleistocene climatic cycles.— Annual Review of Ecology and Systematics 12:75-98. Rimoli, R.O. 1976. Roedores fosiles de la Hispaniola. — Universidad Central Del Este Serie Cientifica III, San Pedro de Macoris, R. D., 95 pp. Steadman, D. W., G. K. Pregill, and S. L. Olson. 1984. Fossil vertebrates from Antigua, Lesser Antilles: Evidence for Late Holocene human-caused extinctions in the West Indies. — Proceed- ings of the National Academy of Sciences 81:4448-4451. Wetmore, A. 1922. Remains of birds from caves in the Republic of Haiti.—Smithsonian Miscel- laneous Collections 74(4):1—4. Department of Herpetology, San Diego Natural History Museum, P.O. Box 1390, San Diego, California 92112. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 834-843 THREE NEW SPECIES OF ADELODRILUS (OLIGOCHAETA: TUBIFICIDAE) FROM GEORGES BANK (NW ATLANTIC) Christer Erséus and Dale Davis Abstract.—Species of Adelodrilus Cook (subfamily Phallodrilinae) are reported from Georges Bank, SE of Massachusetts (U.S.A.). Adelodrilus inopinatus, n. sp., A. correptus, n. sp. and A. pilatus, n. sp. are described, and taxonomic notes are given for A. anisosetosus Cook, A. multispinosus Erséus, and _ A. cristatus Erséus. All six species are interstitial forms inhabiting sublittoral, coarse sands. In the course of the ““Georges Bank Benthic Infauna Monitoring Program,” performed by Battelle New England Marine Research Laboratory for the Minerals Management Service, U.S. Department of the Interior, six species of the marine, meiofaunal tubificid genus Adelodrilus Cook, 1969 (subfamily Phallodrilinae) were encountered. Three of these are new to science and they are described in the present paper, which also provides taxonomic notes on the other three species found. Adelodrilus was taxonomically revised by Erséus (1978), and subsequently treat- ed in supplementary works by Erséus (1979, 1983), and Erséus and Loden (1981). With the new forms described here, the list of Northwest Atlantic Adelodrilus species is increased to ten (Erséus 1983, fig. SA). Material and Methods The material used in this study was sorted from grab samples collected on Georges Bank, SE of Massachusetts (U.S.A.). The worms were fixed in formalin, subsequently stored in 70% isopropanol, finally stained with paracarmine or haematoxylin and mounted whole in Canada balsam. A few specimens of A. inopinatus, n. sp. were also found in material originating from a Bureau of Land Management baseline marine study undertaken by Virginia Institute of Marine Sciences. The type-series of the new species are deposited at the U.S. National Museum of Natural History (USNM), Washington, D.C. Adelodrilus anisosetosus Cook, 1969 Adelodrilus anisosetosus Cook, 1969:13-15, fig. 3.—Erséus 1978:138-139, fig. 3; 1979, figs. 3F-J. New material examined.— Authors’ collections: two specimens from off Mas- sachusetts (Georges Bank), 40°39.5'’N, 67°41.9'W, 75 m, medium to coarse sand (Feb 1983). Remarks. — Adelodrilus anisosetosus, which is the type-species of the genus, was previously known only from Cape Cod Bay (Massachusetts) (Cook 1969, 1971). The new material conforms with the original description with one exception: VOLUME 97, NUMBER 4 835 A 20 UM Fig. 1. A-B, Adelodrilus multispinosus: A, Dorsal, posterior somatic setae; B, Outline of whole worm (whole-mounted), 4.1 mm long; C, Adelodrilus inopinatus, outline of whole worm (whole- mounted), 4.3 mm long. in the new worms the spermathecal pores are located between the lines of the ventral setae and the lateral lines, not in line with the ventral setae as stated by Cook (1969). The species shares the feature of sharply single-pointed, somewhat enlarged setae in posterior dorsal bundles with A. multispinosus and A. inopinatus, n. sp. (both treated below). Adelodrilus anisosetosus is distinguished from these two species by the more or less equal size of the smaller penial setae within the bundle (Erséus 1978:fig. 3), as opposed to those of multispinosus and inopinatus, which are about twice as many (from 15 to about 28 per bundle) and of much varying size within the bundle (Erséus 1979:fig. 2; fig. 3). Habitat. —Sublittoral, largely coarse sand, 10-79 m. Distribution.— East coast of the U.S.: Cape Cod Bay and Georges Bank. Adelodrilus multispinosus Erséus, 1979 Fig. 1A-B Adelodrilus multispinosus Erséus, 1979:421-—423, figs. 2, 3B—E. New material examined.— Authors’ collections: seven specimens from off Mas- sachusetts (Georges Bank): one from 40°35.0’N, 67°11.7'W, 140 m, medium to coarse sand (Jul 1982); one from 40°39.5’N, 67°44.7'W, 72 m, medium to coarse sand (Jul 1982); one from 40°39.6'N, 67°45.8’W, 78 m, coarse sand (May 1983); and four from 40°39.5'N, 67°41.9’W, 75 m, medium to coarse sand (Feb 1983). The junior author checked a large number of additional specimens for consistency regarding body shape and setal characters. Remarks.— This species was originally described from off New Jersey (Erséus 1979); the new material thus extends the known range northward to off Massa- chusetts. The new worms are variable in size (2.5—8.2 mm long, 36-67 segments), but all are uniformly wide throughout their body length (Fig. 1B). The latter feature distinguishes it from the closely related A. inopinatus, n. sp. (see below). 836 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON It was noted in the original description that setae in the dorsal bundles of the posterior end of the body are sharply single-pointed. However, some of the setae in these bundles do have a reduced, very thin upper tooth, and it is possible that the single-pointed versus bifid nature of the setae simply is a matter of angle from which they are viewed (Fig. 1A); 1.e., all somatic setae of A. multispinosus may actually be bifid. Up to about 28 smaller penial setae per bundle were counted in the new in- dividuals. In some worms the lengths of the penial setae slightly exceed the values given originally (giant penial seta up to about 175 um long, anteriormost smaller penial seta up to about 108 um long). Habitat.—Sublittoral, medium to coarse sands, 25-140 m. Distribution. —East coast of the U.S.: off New Jersey and Massachusetts (Georges Bank). Adelodrilus inopinatus, new species Figs. 1C, 2-3 Holotype.—USNM 96056. Type-locality.— Georges Bank, SE of Massachusetts, U.S.A., 40°39.5'N, 67°46.2'W, 79 m, medium to coarse sand (Feb 1983). Paratypes.—USNM 96057-96058. Two specimens from type-locality. Other material examined.— Authors’ collections: seven specimens from off Massachusetts (Georges Bank): four from type-locality; one from 40°39.5'N, 67°46.5'W, 72 m, coarse sand (Nov 1981); one from 40°39.0'N, 67°46.1'W, 72 m, medium to coarse sand (Feb 1982); and one from 40°34.2'N, 67°12.3’W, 140 m, coarse sand (May 1982). The junior author checked a large number of addi- tional specimens from Georges Bank for consistency regarding body shape and setal characters. Senior author’s collection: seven specimens from off Maryland: five from 38°17.5'N, 74°41.0’W, 29 m, medium to coarse sand (Mar 1976); and two from 38°08.0'N, 74°13.0'W, 53 m, medium to coarse sand (Mar 1976). Etymology. —The name inopinatus (Latin ““unexpected’’) alludes to the fact that this species was long regarded as a variety of A. multispinosus by the senior author. Description (based on material from Georges Bank). — Length (fixed specimens) 3.4-5.1 mm, 45-56 segments; diameter at XI in whole-mounted, slightly com- pressed specimens, 0.19—0.24 mm. Posterior end of body distinctly narrower than anterior (Fig. 1C). Clitellum extending over '2 X—XII; tall epidermal cells present ventrally and partially extending up lateral sides, anteriorly in X (Fig. 2F). Somatic setae bifid, with upper tooth shorter and thinner than lower tooth (Fig. 2A); in all ventral bundles, and in anterior and mid-body dorsal bundles. Posterior, dorsal setae (Fig. 2B) sharply single-pointed; ectally strongly curved; distinctly larger than bifid setae. Somatic setae 45-75 um long, 1.5—3.5 um thick; (2)3—4 per bundle anteriorly, (2)3(4) per bundle posteriorly. Ventral setae of XI highly modified into penial bundles, each containing: (1) one giant, slightly sigmoid seta (Figs. 2E; 3:gs), 130-168 um long, 8—14 um wide at middle, with broad inner end and ectal “spoon” ending with single-pointed, but rounded tip; (2) about 5 intermediate, smaller setae (Figs. 2D; 3:ss, Part), largest (anteriormost) about 85-115 um long (inner end difficult to see), 4-6 um thick at middle, these setae ectally provided with slightly hooked tips followed by many tiny spines; (3) high number (at least VOLUME 97, NUMBER 4 837 20 um 44 Fig. 2. Adelodrilus inopinatus: A, Bifid anterior somatic seta; B, Dorsal, posterior somatic seta; C, Smaller penial seta; D, Intermediate penial seta; E, Giant penial seta (different view than in Fig. 3); F, Spermatheca. 20 in many specimens) of small, ectally hooked setae (Figs. 2C; 3:ss, Part), 30— 45 um long, 1—-1.5 um thick. Small penial setae generally erect; giant and inter- mediate penial setae more or less parallel to long axis of worm. Ectal ends of penial setae located within deeply folded and complex copulatory sacs. Male pores paired, in line with ventral somatic setae posteriorly in XI. Spermathecal pores paired, in line with ventral setae in X near intersegmental furrow IX/X. Pharyngeal glands in (IV)V—VII. Male genitalia (Fig. 3) paired. Vas deferens thin-walled and broadly dilated; longer than atrium, up to about 20 um wide; filled with spermatozoa; entering atrium subapically. Atrium oval; 45-70 um long, 27-32 wm wide; with thin outer lining and thick, granulated inner epithelium; cilia not observed; opening into inner end of copulatory sac. Anterior prostate gland large, consisting of many lobes; attached to ental end of atrium, near junction _ with vas deferens. Posterior prostate smaller, attached by long stalk to ectal end of atrium. Spermathecae (Fig. 2F) with ducts 45-60 wm long, 12—25 um wide, ectally dilated and entally narrow lumen; ampullae up to about 200 um long, 12- 30 wm wide, slender and thin-walled, sperm as random masses. Remarks. — Adelodrilus inopinatus is similar and very closely related to A. multi- spinosus; the penial bundles of both species each contain one giant, somewhat spoon-shaped seta, and a row of about 20-25, gradually smaller setae. However, there are morphological differences between the two, differences which proved to be consistent when the junior author checked hundreds of specimens for body shape, spined versus non-spined intermediate penial setae, and shape of posterior dorsal somatic setae. Adelodrilus inopinatus is always tapered toward its posterior end, not as stout as A. multispinosus (Fig. 1B—C); its intermediate penial setae have spines ectally; and its posterior, dorsal setae are ectally much more curved and pointed than those of A. multispinosus (Figs. 1A, 2B). It should be noted that the genitalia of one of the worms from Georges Bank 838 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 100 ym | Fig. 3. Adelodrilus inopinatus, lateral view of male genitalia in segment XI. a, atrium; gs, giant penial seta; pr 1, anterior prostate gland; pr 2, posterior prostate gland; ss, smaller penial seta; vd, vas deferens. area shifted forwards to segments VI-VII. This is a very unusual position for the genital organs of a tubificid, and it must be considered an intra-specific anomaly without taxonomic implications. The material of A. inopinatus from off Maryland differs from the above de- scription only in some dimensional features. Habitat. —Sublittoral, largely coarse sands, 29-140 m. Distribution. —East coast of the U.S.: off Maryland and Massachusetts (Georges Bank). Adelodrilus correptus, new species Fig. 4 Holotype.—USNM 96059. . Type-locality.— Georges Bank, SE of Massachusetts, U.S.A., 40°39.6’N, 67°45.8'W, 78 m, coarse sand (May 1983). Paratypes.—USNM 96060-96061. Two specimens from off Massachusetts (Georges Bank), 40°39.5'N, 67°45.4'W, 78 m, coarse sand (May 1983). Other material examined. — Authors’ collections: three specimens from off Mas- sachusetts (Georges Bank): one from type-locality, and two from 40°39.8'N, 67°46.1’W, 78 m, coarse sand (May 1983). Etymology.—The name correptus is Latin for “pronounced short”; this is a very small species of Adelodrilus. Description. — Length (fixed specimens) 2.2—2.9 mm, 25-32 segments; diameter VOLUME 97, NUMBER 4 839 A B 100 ym OF Fig. 4. Adelodrilus correptus: A, Free-hand drawing of somatic seta; B, Free-hand drawing of tip of smaller penial seta; C, Giant penial seta; D, Laterial view of spermatheca and male genitalia in segments X—XI. s, spermatheca; other abbreviations as for Fig. 3. at XI in whole-mounted, slightly compressed specimens, 0.1 1—0.16 mm. Clitellum extending over '2 X—XII. Somatic setae bifid, with upper tooth thinner and shorter than lower tooth (Fig. 4A); 35-50 um long, 1—-1.5 wm thick; (2)3—4(5) per bundle anteriorly, (2)3—4 per bundle in post-clitellar segments. Ventral setae of XI mod- ified into penial bundles, each containing: (1) one giant, slender, spoon-shaped seta (Figs. 4C; D:gs), with single-pointed, somewhat hooked tip, with ectal blade comprising slightly more than 3 of seta; (2) a row of about 15-20 smaller setae (Fig. 4D:ss), single-pointed and slightly hooked (anteriormost and larger ones), or somewhat clubbed with an apical tooth (posterior, smallest ones; Fig. _ 4B). Giant seta 65-85 um long, 3.5—4.5 wm wide at middle (blade wider). Largest of smaller penial setae 45-60 um long, about 2 wm thick; smallest penial setae about 25-30 um long, about 1 um thick. Ectal ends of penial setae located within thin-walled, folded copulatory sacs. Male pores paired, in line with ventral somatic setae, posteriorly in XI. Spermathecal! pores paired, in line with ventral setae in X near intersegmental furrow IX/X. Pharyngeal glands in IV—VI. Male genitalia (Fig. 4D) paired. Vas deferens thin- walled and broadly dilated; longer than atrium, 15-20 wm wide; filled with sper- matozoa or at least in middle and ectally; entering atrium sub-apically. Atrium oval or ovoid; 28-35 um long, 21—28 wm wide; with thin outer lining and thick, granulated and ciliated inner epithelium; opening into inner end of copulatory sac. Anterior prostate gland large, attached to apical end of atrium. Posterior prostate smaller, attached by long stalk to middle-to-ectal part of atrium, some- what opposite to entrance of vas deferens. Spermathecae small and slender (but not always as narrow as the one depicted in Fig. 4D); ducts 18—23 um long, 14—- 19 um wide; ampullae thin-walled, 44-57 um long, 10-25 um wide; sperm as random masses or compact bundle. 840 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Remarks. —Adelodrilus correptus is similar to the European species A. pusillus Erséus, 1978 and A. kiselevi (Finogenova, 1972), in terms of penial setae and (small, slender) spermathecae. Adelodrilus correptus, however, has a higher num- ber of smaller penial setae per bundle (correptus: about 15-20; pusillus: 5-10; kiselevi: 3—7), and the posteriormost setae within these bundles of smaller penial setae in correptus are ectally clubbed and hooked (Fig. 4B), a feature not observed in pusillus and kiselevi (Erséus 1983:table 1). Habitat.—Sublittoral, coarse sand, 71—80 m. Distribution. —East coast of the U.S.: Georges Bank. Adelodrilus pilatus, new species Fig. 5 Holotype.—USNM 96062. Type-locality.— Georges Bank, SE of Massachusetts, U.S.A., 40°39.5'N, 67°45.4'W, 72 m, coarse sand (Nov 1982). Paratypes.—USNM 96063-96064. Two specimens from Georges Bank: one from 40°39.0'N, 67°46.1’W, 80 m, medium to coarse sand (Nov 1982); and one from 40°38.5'N, 67°46.1'W, 78 m, heterogeneous coarse sand (Nov 1982). Other material examined.— Authors’ collections: seven specimens from Georges Bank: three from type-locality; two from 40°39.6'N, 67°47.6'W, 72 m, medium to coarse sand (Nov 1981); one from 40°39.5’N, 67°46.2’W, 78 m, medium to coarse sand (Feb 1983); and one from 40°40.6’N, 67°46.1'W, 77 m, medium to coarse sand (Nov 1982). Etymology.—The name pilatus is Latin for ““‘armed with a heavy javelin,” and refers here to the much enlarged, sharply single-pointed seta in the ventral bundles of segment X in this species. Description. — Length (fixed specimens) 2.4—3.7 mm, 26-35 segments; diameter at XI in whole-mounted, slightly compressed specimens, 0.10—0.17 mm. Clitellum extending over 2 X—XII. Most somatic setae bifid, with upper tooth thinner and shorter than lower tooth (Fig. SA); 30-50 um long, 1—1.5 um thick; 3—4 per bundle anteriorly, 2-3 per bundle posteriorly. Ventral setae of [IX and X sharply single- pointed and enlarged, those of X (Fig. 5B:se) larger than those of IX; 2-3 per bundle in IX, 2 per bundle in X. One of two setae in each ventral bundle of X larger, 65-80 wm long, 3.5—7 um thick at node; smaller seta in each ventral bundle of X and ventral setae of IX 50-65 um long, 2.5—3 um thick at node. Ventral setae of XI highly modified into penial bundles, each containing: (1) one giant, slightly sigmoid, ectally flattened and widened seta (Fig. 5B:gs), 115—140 um long, 7-9 um thick at middle (much wider ectally); (2) two or three intermediate setae, largest 40-65 um long, 3—4 wm thick, more or less single-pointed and slightly curved ectally; (3) about 5 to 9 (exact number difficult to establish) thin, small setae (Fig. 5B:ss), ectally clubbed and with small apical hooks; 20 wm long, about 1 wm thick. Giant seta and intermediate setae parallel or somewhat oblique to long axis of worm; smaller penial setae generally erect. Ectal ends of penial setae located within thin-walled, folded copulatory sacs. Male pores paired, located close to each other, posteriorly and ventrally in XI. Spermathecal pores paired, in line with ventral setae in X near intersegmental furrow IX/X. Pharyngeal glands in IV-VI. Male genitalia (Fig. 5B) paired. Vas deferens thin- VOLUME 97, NUMBER 4 841 A B Fig. 5. Adelodrilus pilatus;. A, Free-hand drawing of somatic seta; B, Lateral view of spermatheca and male genitalia in segments X—XI. se, ventral seta of X; other abbreviations as for Fig. 4. walled and broadly dilated; about 15 wm wide; containing sperm in large bundle; entering close to apex of atrium. Atrium ovoid; 40-55 um long, 20—30 um wide; with thin outer lining and thick, granulated and ciliated inner epithelium; opening into inner end of copulatory sac. Anterior prostate gland attached by stalk to apex of atrium, at entrance of vas deferens. Posterior prostate attached by long stalk to ectal part of atrium. Spermathecae (Fig. 5B:s) slender; ducts long and hollow, about 50-60 um long, 15-20 wm wide; ampullae thin-walled and elongate, 75- 100 um long, 25-35 wm wide, sperm as random masses or compact bundle. Remarks. — This species is closely related to the European A. cooki Erséus, 1978, in having single-pointed setae in the ventral bundles of segments X and XI. - However, in A. cooki these setae are far less enlarged than those of A. pilatus and they are not appreciably thicker than the normal, bifid setae. In addition, the anteriormost of the intermediate penial setae is 4. pi/atus is much stouter than the corresponding seta in the penial bundles of A. cooki (Fig. 5B; Erséus 1978: fig. 2D). Habitat. —Sublittoral, largely coarse sand, 71-80 m. Distribution. —East coast of the U.S.: Georges Bank. Adelodrilus cristatus Erséus, 1983 Adelodrilus cristatus Erséus, 1983:78-79, fig. 4. New material examined.— Authors’ collections: two specimens from Georges Bank, 40°39.8’N, 67°46.1’W, 78 m, coarse sand (May 1983). Remarks.—This species was recently described from off Maryland and New Jersey (Erséus 1983); the new material thus extends the known range northward to off Massachusetts. 842 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The two specimens conform fully to the original description, even in dimen- sional respects. Pharyngeal glands which were originally not observed, are poorly developed, but present in segments (IIJ)IV—V of the present material. Habitat. —Sublittoral, largely coarse sand, 21-78 m. Distribution.—East coast of the U.S.: off Maryland, New Jersey, and Massa- chusetts (Georges Bank). Discussion All the six species of Adelodrilus reported here were found in an area of Georges Bank characterized by bottoms of coarse sands. This underlines the view that Adelodrilus is largely a genus of interstitial tubificids. Only the species described from the Black Sea, A. kiselevi and A. borceai (Hrabé, 1973), and the deep-sea A. voraginus (Cook, 1970) and A. fimbriatus Erséus, 1983 are known from finer sediments. As has been discussed elsewhere (Erséus 1978, 1984), Adelodrilus is closely related to Phallodrilus Pierantoni, and it is likely that the genus has evolved from Phallodrilus-like ancestors similar to species today found along the east coast of North America (P. coeloprostatus, P. boeschi, P. biprostatus and P. flabellifer, see Erséus 1984). The very high species diversity of Adelodrilus along this coast, including complexes of very closely related forms (anisosetosus-multispinosus- inopinatus and voraginus-fimbriatus) is noteworthy, and it can be expected that additional species are yet to be encountered in the area. Acknowledgments We are indebted to Dr. R. J. Diaz (Virginia Institute of Marine Sciences), for providing the supplementary material of A. inopinatus; to Dr. James Blake (Bat- telle N.E. Marine Research Laboratory), for valuable criticism of the manuscript; and Battelle New England Marine Research Laboratory, for financial support. The work was also supported by the Swedish Natural Research Council. Specimens from Georges Bank were collected as part of the Georges Bank Benthic Infauna Monitoring Program, supported by Contract No. 14-12-0001-2912 from the U.S. Department of the Interior Minerals Management Service to Battelle New England Marine Research Laboratory. Literature Cited Cook, D. G. 1969. The Tubificidae (Annelida, Oligochaeta) of Cape Cod Bay with a taxonomic revision of the genera Phallodrilus Pierantoni, 1902, Limnodriloides Pierantoni, 1903 and Spiridion Knollner, 1935.—Biological Bulletin 136:9—27. . 1970. Bathyal and abyssal Tubificidae (Annelida, Oligochaeta) from the Gay Head-Bermuda transect, with descriptions of new genera and species.— Deep-Sea Research 17:973-981. . 1971. The Tubificidae (Annelida, Oligochaeta) of Cape Cod Bay. II: Ecology and systematics, with the description of Phallodrilus parviatriatus nov. sp.— Biological Bulletin 141:203—221. Erséus, C. 1978. Newspecies of Adelodrilus and a revision of the genera Adelodrilus and Adelodriloides (Oligochaeta, Tubificidae).—Sarsia 63:135-144. . 1979. Bermudrilus peniatus, n.g., n. sp. (Oligochaeta, Tubificadae) and two new species of Adelodrilus from the Northwest Atlantic.— Transactions of the American Microscopical Society 98:418-427. VOLUME 97, NUMBER 4 843 1983. New records of Adelodrilus (Oligochaeta, Tubificidae), with descriptions of two new species from the Northwest Atlantic.—Hydrobiologia 106:73-—83. —. 1984. Taxonomy of some species of Phallodrilus (Oligochaeta, Tubificidae) from the North- west Atlantic, with description of four new species.— Proceedings of the Biological Society of Washington 97:812-826. , and M. S. Loden. 1981. Phallodrilinae (Oligochaeta, Tubificidae) from the east coast of Florida, with description of a new species of Adelodrilus.— Proceedings of the Biological Society of Washington 94:819-825. Finogenova, N. P. 1972. New species of Oligochaeta from Dniepr and Bug Firth and Black Sea and revision of some species. [In Russian.]—Trudy Zoologicheskogo Instituta, Akademiya Nauk SSSR, Leningrad 52:94-116. Hrabé, S. 1973. A contribution to the knowledge of marine Oligochaeta, mainly from the Black Sea.—Traveaux du Muséum d’Histoire Naturelle ““Grigore Antipa”’ 13:27-38. (CE) Swedish Museum of Natural History, % Department of Zoology, Uni- versity of G6teborg, Box 25059, S-400 31 Géteborg, Sweden; (DD) Battelle New England Marine Research Laboratory, 397 Washington Street, Duxbury, Mas- sachusetts 02332 (Present address: University of Hawaii at Manoa, Department of Zoology, Edmondson Hall, 2538 The Mall, Honolulu, Hawaii 96822). PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 844-848 THE DEEP-SEA AMPHIPOD PARACYPHOCARIS PRAEDATOR (GAMMARIDEA: LYSIANASSIDAE) ASSOCIATED WITH THE PELAGIC SHRIMP OPLOPHORUS NOVAEZEELANDIAE AS AN EGG-MIMIC Thomas E. Bowman and Robert A. Wasmer Abstract.—Three specimens of the amphipod Paracyphocaris praedator were found among the attached eggs of the shrimp Oplophorus novaezeelandiae. The posterior pereon of the amphipod was greatly distended and filled with oily glob- ules, causing it to resemble in form and color the eggs of the shrimp. It is suggested that P. praedator is an egg-mimic that feeds upon the Oplophorus eggs. The prehensile pereopods of Paracyphocaris, and perhaps those of other cyphocarid genera, prevent it from being dislodged by the host’s grooming. Paracyphocaris is one of a group of genera of Lysianassidae designated by Barnard (1969) as “cyphocarids.”” The genus comprises three species: the type- species P. praedator Chevreux, 1905, P. brevicornis Birstein and Vinogradov, 1955, and P. distincta Birstein and Vinogradov, 1963. All captures of the three species have been made in the deep-sea pelagial. Paracyphocaris distinctais known only from the Philippine Trench, and P. brevicornis has been found at two lo- calities, the Kurile-Kamchatka Trench (Birstein and Vinogradov 1955) and near the Solomon Islands (Birstein and Vinogradov 1960). Paracyphocaris praedator has a wide known distribution, having been recorded from seven localities in the North Atlantic (Summarized by Shoemaker 1945) and two localities in the Indian Ocean, NW of Rodriguez Island and NE of the Chagos Archipelago (Birstein and Vinogradov 1964). Its known range is extended herein to include the South Pacific off southern Chile, where three specimens have been found among the eggs at- tached to the pleopods of the pelagic shrimp Oplophorus novaezeelandiae. Paracyphocaris praedator Chevreux Fig. 1 Paracyphocaris praedator Chevreux, 1905:1, figs. 1-3; 1935:25-—27, pl. 10, fig. 3, pl. 11, figs. 2-3.—Stephensen, 1923:54; 1933:10.—Schellenberg, 1926:216; 1927: 667-668, fig. 61.—Birstein and Vinogradov, 1960:170-171, fig. 1; 1964:156.— Shoemaker, 1945:189, text-fig. 2.—Gurjanova, 1962:71-73, figs. 11A, 11B. Material.—Complete specimen about 8.7 mm in length and specimen lacking pleonite 3 and urosome judged to be about same length (USNM 211073), from among ova of 2 different specimens of Oplophorus novaezeelandiae De Man (USNM 211072), Eltanin cruise 25, sta 303, off Santiago, Chile, 33°1 1'-33°14’S, 72°40'-72°38'W, 400-475 m, 25 Sep 1966. Complete specimen about 9.5 mm in length (USNM 213334) from among ova of Oplophorus novaezeelandiae (USNM 213333), Eltanin cruise 25, sta 322, off Osorno, Chile, 41°01’—41°08’S, 78°59’— 78°56'W, 380-475 m, 7 Oct 1966. VOLUME 97, NUMBER 4 845 Fig. 1. A, Oplophorus novaezeelandiae, ovigerous °, pereopod 5 propus and dactyl. B—F, Para- cyphocaris praedator from Oplophorus marsupium: B, Habitus, lateral, most appendages omitted; C, Antenna 1; D, Antenna 2, segments 1 and 2 of peduncle omitted; E, Telson and uropod 1; F, Uropod 3; G, Outer and middle circles, outlines of distended area of pereon of P. praedator viewed ventrally; middle circle is specimen in B; inner circle, outline of egg of O. novaezeelandiae. Scale in mm applies to B and G. Diagnostic characters.— The Chilean specimens match the illustrations of Chev- reux (1905) and Shoemaker (1945), as far as could be seen without dissection. The characters that distinguish P. praedator from its congeners are given in the key that follows. Key to the Species of Paracyphocaris 1. Antenna 2 shorter than antenna 1. Pereopod basis wider than long Seve bd Oot Bio ETE SOLE EMT AEA AD (onan Oem mera eens Seat P. brevicornis — Antenna 2 longer than antenna 1. Pereopod 4 basis longer than wide .. 2 2. Antenna 1 flagellum 5-segmented. Pereopod 1 propus 2.3 as long as wide. Epimeron of pleonite 3 rounded posteriorly. Telson 2.3 x as long AST Wil CC Meier Nai Rs ae Une Cn Mi rR a TY) P. praedator 846 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON — Antenna | flagellum 4-segmented. Pereopod 1 propus 1.7 as long as wide. Epimeron of pleonite 3 pointed posteriorly. Telson 1.5 x as long as WIE). Be RECO ALS, AEE A ee oe tA eee P. distincta Chevreux’s illustrations do not show the pectinate inner margins of the rami of uropod 1 (Fig. 1F). Shoemaker did not illustrate the uropods, but his Bermuda specimens have pectinate margins on the rami of uropod 1. Nature of the Paracyphocaris-Oplophorus Relationship Both Chilean specimens have the body greatly distended ventrally in the region of pereonites 5—7. The distension amounts to more than the normal depth and width of the body (Fig. 1B, G). The distended part is filled with oily globules and is dull orange after 18 years in alcohol. The eggs of the Oplophorus are similar in appearance, being filled with oily globules of yolk, and the color in alcohol is a paler yellowish orange. The eggs are oval and somewhat smaller than the distended part of the amphipod (Fig. 1G), but the long diameter of the egg is about equal to the length of the distended part in the complete specimens of Paracyphocaris. Thus the swollen region of the Paracyphocaris matches the Oplophorus egg in texture, color, and size, and we suggest that P. praedator may be an egg-mimic, the first example among the Amphipoda. Elsewhere among the Crustacea egg- mimicry is found in copepods of the family Choniostomatidae (Hansen 1897) and perhaps some cryptoniscid isopods (Nielson and Stromberg 1965). In the choniostomatid Sphaeronella monothrix (Bowman and Kornicker, 1967) not only does the female mimic the eggs of its ostracode host, Parasterope pollex, but it lays its eggs in groups of about 15, enclosed by a membrane. Each group, or ovisac, is about the size of an egg of Parasterope. The presence of Sphaeronella in the Parasterope marsupium appears to inhibit egg-laying by the host. In the Caspian Sea the amphipod Cardiophilus baeri, which usually occurs in the mantle cavity of the cockle Cerastoderma lamarcki, has been found in the marsupium of the amphipods Corophium curvispinum and C. spinulosum by Osadchikh (1977), who believes that early development of C. baeri takes place there. Osadchikh did not suggest that Cardiophilus either mimics or feeds on the Corophium eggs or young. Paracyphocaris does not inhibit egg-laying by Oplophorus. The two Chilean shrimps from station 25 carried six and ten eggs respectively, rather loosely packed. In comparison, a South Atlantic O. novaezeelandiae of about the same length in the Division of Crustacea collections has 20 closely clustered eggs. This difference, together with the distended condition and appearance of the amphipod guts, suggests that the amphipods feed on the eggs of the shrimp. Thus the Paracy- phocaris gains from its association a protected and concealed habitat where energy requirements are minimal and also a readily available source of rich food. Seven of the 12 genera grouped as “cyphocarids”’ by Barnard (1969:300) are characterized by the prehensile nature of from 2—4 of the pereopods posterior to pereopod 2. In a recent discussion of Gammaridea having prehensile pereopods, Vader (1983) divided them into two groups: (1) abyssopelagic or bathypelagic species, which he considered to be probably all predators; (2) those living in direct contact with the mouthparts or appendages of large Crustacea. Vader placed the cyphocarids in the first group. Paracyphocaris is not necessarily predaceous, how- VOLUME 97, NUMBER 4 847 ever; the prehensile pereopods may be adapted for resisting dislodgment by the grooming activities of the host rather than for seizing prey. Nothing is known about grooming in the Oplophoridae, but the armament of the last pereopod of O. novaezeelandiae (Fig. 1A) suggests that it may be used in grooming. The posterior margin of the carpus bears two rows of setae with scalloped margins directed toward the medial axis of the shrimp. Similar setae on the thoracic appendages of Pandalus danae have been demonstrated by Bauer (1975) to func- tion in grooming, although this shrimp does not brush its eggs with the 5th pereopods or with the chelate legs as in palaeomonids, hippolytids, and alphaeids (Bauer 1975, 1979). If Oplophorus grooms its eggs, as we suspect, the Paracy- phocaris must be able to resist dislodgment, and its prehensile pereopods 3-6 seem ideally suited for preventing it from being ejected. The majority of the known specimens of P. praedator did not have a distended gut when collected and were probably free-living rather than inhabitants of mar- supia. No ovigerous females have been collected, and details of the life history are unknown, but in general storage of large amounts of food is more likely to precede than to follow breeding. We suggest that after gorging itself, Paracypho- caris leaves the shrimp marsupium and takes up a pelagic life, during which it lives on the stored food and perhaps does not feed. Mating could take place during the pelagic phase or in the shrimp’s marsupium. For the latter a second amphipod would have to enter the marsupium. We could speculate at length on further aspects of the life history, but this would contribute little in the absence of factual evidence. Finally, we suspect that the other cyphocarid genera that have prehensile pereo- pods may prove to be associated with the marsupia of deep-sea pelagic shrimps. These genera are Cebocaris, Crybelocephalus, Crybelocyphocaris, Mesocyclocaris, Metacyclocaris, and Metacyphocaris. Diagnoses and references for these genera are given by Barnard (1969). There is some evidence for this suspicion. K. H. Barnard (1932) reported that the ventral surface of smaller specimens of Meta- cyphocaris helgae was bulging, and Shoemaker (1945) found this condition in an immature female of the same species. Acknowledgments We are grateful to Drs. J. L. Barnard and Wim Vader for their comments on the manuscript. This work was partially supported by Contract PC-206882 to RAW from the Cooperative Systematics Research Program of the Smithsonian Oceanographic Sorting Center, funded by grant DPP-7920835 from the National Science Foun- dation, Division of Polar Programs, B. J. Landrum, Principal Investigator. Literature Cited Barnard, J. L. 1969. The families and genera of marine gammaridean Amphipoda.— United States National Museum Bulletin 271:1—535. Barnard, K. H. 1932. Amphipoda.—Discovery Reports 5:1—326, pl. 1. Bauer, R. T. 1975. Grooming behaviour and morphology of the caridean shrimp Pandalus danae Stimpson (Decapoda: Natantia: Pandalidae).— Zoological Journal of the Linnean Society 56(1): 45-71. . 1979. Antifouling adaptations of marine shrimp (Decapoda: Caridea): gill cleaning mech- 848 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON anisms and grooming of brooded embryos. — Zoological J ournal of the Linnean Society 65:28 1- 303. Birstein, J. A., and M. E. Vinogradov. 1955. Pelagic gammarideans (Amphipoda— Gammaridea) of the Kurile-Kamchatka Trench. — Trudy Instituta Okeanologii Akademiya Nauk SSSR 12:210- 287 [in Russian]. , and _ 1960. Pelagic gammarideans from tropical regions of the Pacific Ocean.— Trudy Instituta Okeanologii Akademiya Nauk SSSR 34:165-241 [in Russian]. , and . 1963. Deep-sea pelagic Amphipoda of the Philippine Trench. — Trudy Instituta Okeanologii Akademiya Nauk SSSR 71:81—93. [in Russian, English summary]. , and 1964. Pelagic gammarideans from the northern part of the Indian Ocean. — Trudy Instituta Okeanologii Akademiya Nauk SSSR 65:152-196. Bowman, T. E., and L. S. Kornicker. 1967. Two new crustaceans: the parasitic copepod Sphaero- nellopsis monothrix (Choniostomatidae) and its myodocopid ostracod host Parasterope pollex (Cylindroleberidae) from the southern New England coast.— Proceedings of the United States National Museum 123(3613):1—28, pl. 1. Chevreux, E. 1905. Paracyphocaris praedator, type d’un nouveau genre de Lysianassidae. — Bulletin du Musée Océanographique de Monaco 32:1-6. _ 1935. Amphipodes provenant des campagnes scientifiques du Prince Albert Ier de Monaco. — Résultats des Campagnes Scientifiques Accomplies sur Son Yacht par Albert Ier, Prince Sou- verain de Monaco 90:1—214, pls. 1-16. Gurjanova, E. F. 1962. Gammarideans of the northern part of the Pacific Ocean (Amphipoda— Gammaridea). Part 1.—Fauna of the USSR, Zoological Institute, Akademy of Sciences USSR 74:1-441. Hansen, H. J. 1897. The Choniostomatidae, a family of Copepoda, parasites on Crustacea Mala- costraca. 190 pp, 13 pls., Copenhagen. Nielsen, S.-O., and J.-O. Strémberg. 1965. A new parasite of Cirolana borealis Lilljeborg belonging to the Cryptoniscinae (Crustacea Epicaridea).—Sarsia 18:37-62. Osadchikh, V. F. 1977. A finding of Cardiophilus baeri in the marsupium of corophiids (Amphipoda, Gammaridae).—Zoologichesii Zhurnal 56(1):156-158 [in Russian]. Schellenberg, A. 1926. Amphipoda 3: Die Gammariden der Deutschen Tiefsee-Expedition. — Wis- senschaftliche Ergebnisse der Deutsche Tiefsee-Expedition auf dem Dampfer “Valdivia” 1898— 1899 23(5):195-243, pl. 5. 1927. Amphipoda des nordischen Plankton.—Nordisches Plankton 20(6):589-722. Shoemaker, C.R. 1945. The Amphipoda of the Bermuda Oceanographic Expeditions, 1929-1931.— Zoologica (New York) 30(4):185-266. Stephensen, K. 1923. Crustacea Malacostraca, V. Amphipoda I.—Danish Ingolf-Expedition 3(8): 1-100. ——. 1933. Amphipoda. The Godthaab Expedition 1928.—Meddelelser om Gronland 79(7): 1-88. Vader, W. 1983. Prehensile pereopods in gammaridean Amphipoda.—Sarsia 68:139-148. (TEB) Department of Invertebrate Zoology (Crustacea), NHB 163 Smithsonian Institution, Washington, D.C. 20560; (RAW) Department of Biology, Columbia Union College, Takoma Park, Maryland 20912. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 849-863 TWO NEW SPECIES OF LEPIDONOTOPODIUM (POLYCHAETA: POLY NOIDAE: LEPIDONOTOPODINAE) FROM HYDROTHERMAL VENTS OFF THE GALAPAGOS AND EAST PACIFIC RISE AT 21°N Marian H. Pettibone Abstract.— New collections from the hydrothermal vent areas at 21°N off West- ern Mexico and the Galapagos Rift included specimens of Lepidonotopodium fimbriatum Pettibone, 1983, and two new species: L. riftense and L. williamsae. The species are described and compared, including observations on their early stages. The species of Lepidonotopodium are unique among the Polynoidae in having well-developed bracts encircling the notopodia. Among the unusual fauna clustered around the hydrothermal vents in the Ga- lapagos region and the 21°N site on the East Pacific Rise, scale-worms of the family Polynoidae are well-represented. The abundant available material was collected by the Galapagos Rift Biology Expedition in 1979 and the OASIS group of Scripps Institution of Oceanography in 1982. Two reports on the polynoids from these areas have appeared: Lepidonotopodium fimbriatum in the new subfamily Lepidonotopodinae by Pettibone (1983) and Branchipolynoe symmy- tilida, commensal in the mantle cavities of the deep-sea vent mussels, in the new subfamily Branchipolynoinae by Pettibone (1984). Based on additional material received from J. F. Grassle and I. Williams of the Woods Hole Oceanographic Institution (WHOJ), with preliminary sorting by I. Williams, the present study supplements the first of the reports above. In addition to more specimens of L. fimbriatum, numerous others are referred to two new species of Lepidonotopo- dium. The various methods of collecting and sorting of the material have produced many adults, as well as many minute young specimens of polynoids, as noted by - accompanying collecting labels: clam bucket wash, clam bucket with mussels, crab trap wash, bottom of Instant Ocean aquarium, washes of, and rubble samples with, Alvinella, clams or Calyptogena, vestimentiferans or Riftia, and mussels. As is well known, many species of polynoids are noted for their close associations with other invertebrates. The closeness of the association of the species of Lep- idonotopodium with the invertebrates referred to above cannot be determined. However, possible hosts are the ampharetid polychaetes Alvinella pompejana, described by Desbruyéres and Laubier (1980), the giant clams Calyptogena mag- nifica, by Boss and Turner (1980), the giant vestimentiferans Riftia pachyptila, by Jones (1981), and the deep-sea vent mussels, as yet unnamed, but to be de- scribed by Kenk and Wilson (in press). The latter definitely serves as the host for Branchipolynoe symmytilida, since the polynoids have been found in the mantle cavities of the mussels, as noted above. Many of the specimens have foreign material attached to the parapodia, setae, and elytral surfaces, including epibiotic bacterial masses of different morphological 850 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON types—coccoid, filamentous, terminal filaments, apical tufts or “hairs,”’ as ob- served also for Alvinella pompejana by Desbruyéres et al. (1983). Types and additional specimens are deposited in collections of the Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian In- stitution (USNM). Family Polynoidae Subfamily Lepidonotopodinae Pettibone, 1983 The genus Lepidonotopodium Pettibone, 1983, with L. fimbriatum Pettibone, 1983, as the type-species, is emended to include two additional new species from the hydrothermal vents in the Galapagos area and the East Pacific Rise at 21°N. Lepidonotopodium Pettibone, 1983, emended Diagnosis.— Body short, flattened, up to 30 segments (first achaetous). Elytra and prominent elytrophores 11 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, 19, and 21, with dorsal cirri on posterior segments. Prostomium deeply bilobed; median antenna with ceratophore inserted in anterior notch, with short style; without lateral antennae but with small frontal filaments on anterior lobes of prostomium; paired palps cylindrical, smooth, with filamentous tips; without eyes. First or tentacular segment fused to prostomium, not visible dorsally; tentacu- lophores lateral to prostomium, each with pair of tentacular cirri, single aciculum, without setae; without facial tubercle. Second segment with first pair of elytra, biramous parapodia and ventral or buccal cirri attached to basal parts of parapodia lateral to mouth; styles longer than following ventral cirri. Parapodia biramous, with notopodia shorter than neuropodia. Notopodia subconical, with well-de- veloped bracts encircling notopodia anteriorly and dorsally. Neuropodia diago- nally truncate, deeply notched dorsally. Distal tips of notopodia, notopodial bracts and neuropodia fimbriated with slender papillae. Notosetae and neurosetae nu- merous, spinous. Dorsal cirri on segments lacking elytra with cylindrical cirro- phores attached on posterodorsal sides of notopodia, with tapered styles; ventral cirri short, tapered, attached near middle of neuropodia. Dorsal tubercles on cirrigerous segments large, inflated. Elytrophores, dorsal tubercles and their bases with numerous ciliated ridges. Ventral segmental papillae lacking or with 2 to 4 pairs of long papillae on some middle segments (1 1-14). Pygidium dorsal, bulbous, wedged between parapodia of posterior small segments, with or without pair of anal cirri. Pharynx with 7 or 9 pairs of papillae, 2 pairs of jaws with few (5-9) or numerous basal teeth. Lepidonotopodium fimbriatum Pettibone Fig. 1 Lepidonotopodium fimbriatum Pettibone, 1983:393, figs. 1—S. Additional records.—Pacific Ocean off western Mexico, 20°50’N, 109°06’W, OASIS Alvin dives in 1982:— Dive 1213, 2617 m, 19 Apr, Alvinella wash, 1 young specimen (USNM 81998).—Dive 1214, 2633 m, 20 Apr, vestimentiferan wash, 2 young specimens (USNM 81999).—Dive 1223-11, 2616 m, 7 May, coarse VOLUME 97, NUMBER 4 851 A vA . : Z O f | Ap es or B D E Fig. 1. Lepidonotopodium fimbriatum, A, B, elytra from young specimen of 9 mm long and 25 segments, USNM 81998; C-E, border papillae and jaw of extended pharynx from young specimen of 7 mm long and 24 segments, USNM 82000: A, Left first elytron from segment 2, with detail of few microtubercles and micropapillae; B, Left elytron 10 from segment 19, with same; C, Dorsal border papillae; D, Ventral border papillae; E, Jaw. Scales = 0.5 mm for A, B; 0.2 mm for C-E. fraction, 15 adults and 12 young specimens (USNM 96482).—Dive 1226-7, 2616 m, 10 May, Riftia, Calyptogena and Alvinella wash, 4 adults and 14 young spec- imens (USNM 82000), and fine fraction, 2 young specimens (USNM 96483). Supplementary description.— The 19 adult specimens agree with the previously described types except for the larger size of some of them: the largest specimen, 37 mm in length and 16 mm in width with 30 segments, the last 2 of which are minute, compared with the adult types, 23—24 mm in length, 12—13 mm in width, with 28 segments, the last segment small. Nine of the adults have similar ventral _ segmental papillae on segments 11 and 12 (fig. 3A, in Pettibone 1983) and 10 lack them. Some of the young specimens have the usual 11 pairs of elytra with 22-26 segments, the last segment small, 5-9 mm in length and 3.5—5 mm in width. When present, the elytra show some signs of the characteristic raised macrotu- bercles but lack the full number of 2 macrotubercles per elytron (Fig. 1A, B; figs. 1, 2E-G, in Pettibone 1983). The elytra are thinner and more delicate, with scattered microtubercles and slightly larger round micropapillae; some fine “hairs” or “‘bacteria’”’ are attached to the microtubercles (Fig. 1A, B). The pharynx was extended on a young specimen, 7 mm in length, 4 mm in width, with 24 segments; it is slightly different from that of the figured paratype: 7 dorsal papillae with the median one extra large and 6 ventral papillae, with 2 middle ones small and adjacent lateral ones large, and jaw with 9 basal teeth (Fig. 1 C—E; 7 pairs of papillae, jaws with 5-7 basal teeth on paratype, fig. 3B, C, in Pettibone 1983). The remaining young specimens have 6-10 pairs of elytra, with 12—21 segments, 852 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the last 1 or 2 segments very small, 1.5—S5 mm in length and 1.2—3 mm in width. The elytra are rather thin and lack the raised macrotubercles. Distribution.—East Pacific Rise at 21°N, in 2616-2633 meters, appear to be associated with ampharetid polychaetes, Alvinella pompejana, giant clams, Ca- lyptogena magnifica, and vestimentiferans, Riftia pachyptila. Lepidonotopodium riftense, new species Figs. 2-4 Material.—East Central Pacific, from dives of the A/vin on 3 vent areas along the Galapagos Rift in 1979: MUSSEL BED, 00°47.89'N, 86°09.21'W:—Dive 880, 21 Jan, 2493 m, 8 para- types (USNM 96008). — Dive 887, 12 Feb, 2488 m, 27 paratypes (USNM 96009). — Dive 888, 13 Feb, 2483 m, 8 paratypes (USNM 96010).—Dive 991-1, 8 Dec, 2490 m, clam bucket wash, 13 paratypes (USNM 96011). GARDEN OF EDEN, 00°47.69'N, 86°07.74'W:—Dive 884, 25 Jan, 2482 m, clam bucket with mussels, 5 paratypes (USNM 96016).—Dive 993-7, 10 Dec, 2518 m, crab trap wash, 30 paratypes (26 young, USNM 96017). ROSE GARDEN, 00°48.25'N, 86°13.48’W:—Dive 890, 15 Feb, 2447 m, bot- tom of Instant Ocean washings, 11 paratypes (USNM 96013).—Dive 894, 19 Feb, 2457 m, amphipod trap, holotype (USNM 80635) and 3 paratypes (USNM 80636, 80637).—Dive 896-22, 21 Feb, 2460 m, Instant Ocean washings, 3 para- types (USNM 96014).—Dive 984-32, 1 Dec, 2451 m, mussel washings, 2 young paratypes (USNM 96012).—Dive 990-41, 7 Dec, 2451 m, vestimentiferan wash, young paratype (USNM 96015). Pacific Ocean off Western Mexico, 20°50’N, 109°06'W, OASIS Alvin dives in 1982:—Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 90 paratypes, including many young (USNM 96021).—Dive 1215-5B, 21 Apr, 2616 m, slurp sample, young paratype (USNM 96485).—Dive 1218-15, 24 Apr, 2618 m, clam and crab trap wash, adult and 15 young paratypes (USNM 96486).—Dive 1219-1B, 1219- 10, 25 Apr, 2612 m, Riftia and clam wash, coarse and fine fractions and slurp sample, 19 paratypes, including young (USNM 96018).—Dive 1221-15, 4 May, 2618 m, Riftia and Calyptogena wash, coarse and fine fractions, 117 paratypes, including many young (USNM 96022).—Dive 1222-5b, 6 May, 2614 m, rubble sample from Calyptogena residue, 4 paratypes (USNM 96019).—Dive 1223-11, 7 May, 2616 m, coarse and fine fractions, 2 adults and 35 young paratypes (USNM 96487).—Dive 1225-6A, 9 May, 2618 m, rubble, coarse fraction, 2 young para- types (USNM 96488).—Dive 1225-7, coarse and fine fractions, 4 adults and 9 young paratypes (USNM 96489, 96490).—Dive 1226-7, 10 May, 2616 m, Riftia, Calyptogena and Alvinella wash, coarse fraction, 8 young paratypes (USNM 96020) and fine fraction, 6 young paratypes (USNM 96491). Description.—The holotype, from the Galapagos Rift (USNM 80635), has a length of 6 mm, a width of 4.5 mm, including the setae, with 23 segments. A slightly smaller paratype (USNM 80636) has a length of 5 mm, a width of 4 mm, with 22 segments. The largest paratype from the Galapagos Rift (USNM 96010) has a length of 10 mm, a width of 5 mm, with 24 segments. The largest paratype from the East Pacific Rise (USNM 96018) has a length of 13 mm, a width of 7 mm, with 25 segments. No color remains. The body is short, suboval in outline, flattened dorsoventrally, slightly tapered and rounded anteriorly and posteriorly. VOLUME 97, NUMBER 4 853 Fig. 2. Lepidonotopodium riftense, A-C, E-G, holotype, USNM 80635; D, paratype, USNM 96009; H, paratype, USNM 96021: A, Anterior end, dorsal view; B, Same, ventral view; C, Dorsal view of left segments 9 and 10; D, Ventral view of extended pharynx; E, Isolated jaw from pharynx; F, First right elytron, with detail of micropapillae; G, Left middle elytron, with same; H, Three elytra showing variable numbers of oval projections, with detail of micropapillae and projections. Scales = 1 mm for A-D; 0.1 mm for E; 1 mm for F-H. The 11 pairs of elytra are attached eccentrically on prominent elytrophores, with dorsal cirri on the posterior segments (10 pairs of elytra on smaller juveniles with 19 segments). The elytra are oval to subreniform, imbricated, and cover the dorsum. They are opaque, smooth, with branching “veins” emanating from the place of attachment to the elytrophores. Near the posterior and lateral borders of 854 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON > ft \| | . | will itn My | | | "TT | Hi AAWANV ANNAN AS = = in SS W WV \ C ES © Ze eS yl \i< $9BUCTE A ‘ (uw ! Cua a y ARGOS Se rs Sus ( we % 7 Fig. 3. Lepidonotopodium riftense, holotype, USNM 80635: A, Right elytrigerous parapodium from segment 9, anterior view, with detail of distal papillae; B, Right cirrigerous parapodium from segment 10, posterior view; C, Enlarged view of right notopodium of elytrigerous parapodium, anterior VOLUME 97, NUMBER 4 855 the elytra, scattered globular micropapillae with short to longer tapered or cylin- drical bases are found on the surface (Fig. 2F, G). Numerous fine branched “‘hairs”’ or “bacteria’’ are found on the borders and surfaces of the elytra. On some specimens from the East Pacific Rise, the elytra have a variable number of oval projections near the posterior border, similar to the well-developed macrotuber- cles of L. fimbriatum (Fig. 2H). The dorsal cirri are attached dorsoposteriorly on the notopodia; they have cylindrical cirrophores with a blister-like lobe on the posterior side; the styles are tapering and extend to about the tips of the neurosetae (Figs. 2A, C; 3B). The dorsal tubercles on the cirrigerous segments are large and inflated. The surfaces of both the elytrophores and dorsal tubercles have bands or tufts of cilia (Figs. 2C; 3A—-D). The prostomium is bilobed, the anterior lobes subtriangular, each with a small frontal filament; lateral antennae are absent (Fig. 2A, B). The median antenna is inserted in the anterior notch, having a short cylindrical ceratophore and short tapering style. The palps are cylindrical, with tapered tips, about one and a half times the length of the prostomium. Eyes are lacking. The tentaculophores of the tentacular segment are lateral to the prostomium and lack setae; the styles of the 2 pairs of dorsal and ventral tentacular cirri are similar in length, smooth, tapered, and subequal in length to the palps. The second or buccal segment bears the first pair of large elytrophores, biramous parapodia, and ventral or buccal cirri attached basally on prominent cirrophores lateral to the ventral mouth; their styles are similar to the tentacular cirri and longer than the following ventral cirri (Fig. 2A, B). The mouth is enclosed in upper, lateral and posterior lips between segments | to 3 (Fig. 2B). The opening of the extended pharynx is encircled by 9 pairs of bulbous papillae, with the lateral one small or with only 7 pairs of subequal papillae (Fig. 2D). The 2 pairs of dorsal and ventral hooked jaws are minutely serrated with numerous teeth (about 26; Fig. 2D, E). The biramous parapodia have shorter notopodia located on the anterodorsal sides of the longer neuropodia (Fig. 3A). The notopodium is subconical, with a projecting acicular lobe hidden anteriorly by the very numerous notosetae and enclosed by well-developed flaring anteroventral and posterodorsal bracts attached basally to the acicular lobe (Fig. 3A, C, D). The neuropodium is diagonally truncate _and deeply cleft on the upper part. The presetal acicular lobe projects dorsally beyond the shorter rounded postsetal lobe (Fig. 3E, F). The distal margins of the notopodial bracts and neuropodial lobes are fimbriated with short papillae, with “hairs” or “bacteria”’ attached. The notosetae are very numerous, forming thick radiating bundles. They vary in length, are mostly slightly stouter than the neu- rosetae and have relatively few (7-8), widely-spaced spines along one side and blunt rounded tips; many of them had long strands of foreign material attached subdistally (“bacteria’”’?; Fig. 4A). The neurosetae are numerous, forming fan- shaped bundles. They are all similar, with slightly hooked bare tips and 2 rows —_— view; D, Same, posterior view, with detail of distal papillae, notoaciculum dotted; E, Enlarged view of right neuropodium, anterior view, with detail of distal papillae, neuroaciculum dotted; F, Same, posterior view. Scales = 0.5 mm for A, B; 0.2 mm for CF. 856 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Lepidonotopodium riftense, A, B, holotype, USNM 80635; C, paratype, USNM 80636; D- J, young paratype of 13 segments, 1.5 mm long, 1.5 mm wide, USNM 96008: A, Three notosetae, one with strand of attached “bacteria’’; B, upper, middle and lower neurosetae, with detail of middle part; C, Right parapodium from segment 11, anterior view, showing elongated segmental papilla; D, Anterior end, dorsal view; E, Right middle elytron, with detail of some micropapillae; F, Right posterior elytron from segment 13 (incompletely developed); G, Right elytrigerous parapodium, anterior view, acicula dotted; H, Right cirrigerous parapodium, posterior view; I, Short and longer notosetae, with detail of tip; J, Upper, middle and lower neurosetae, with detail of middle part. Scales = 0.1 mm for A, B, G-J; 0.5 mm for C; 0.2 mm for D-F. of numerous spines along one side; the upper neurosetae are longer, with longer spinous regions (Fig. 4B). The ventral cirri are short, tapered, smooth and attached on the middle of the posterior face of the neuropodia (Fig. 3A, B). Segmental or nephridial papillae are not obvious. On about half of the adult specimens there are 2 pairs of elongate papillae on segments 11 and 12, attached VOLUME 97, NUMBER 4 857 basally on the ventroposterior sides of the neuropodia and extending to the tips of the lower neurosetae (Fig. 4C). Such papillae are lacking on the holotype. On some of the juveniles the papillae are present but shorter and incompletely de- veloped. The pygidium is visible dorsally as a bulbous lobe wedged between the parapodia of the 2 posterior smaller segments, with a pair of long ventral anal cirri, similar to the posterior dorsal cirri. Numerous young specimens were collected at the Galapagos Rift from washings of mussels and in crab and amphipod traps and from the East Pacific Rise from washings of Riftia, Calyptogena and Alvinella, coarse and fine fractions, rubble, clam and crab trap wash, and slurp sample. Juveniles with 17 to 20 segments, with less than 11 pairs of elytra, have lengths of 2.5 to 4 mm and widths of 2 to 3 mm. Young specimens with 10 to 15 segments have lengths of 1 to 2 mm and widths of 1 to 1.5 mm. A young specimen of 13 segments, 1.5 mm long and 1.5 mm wide, has 7 pairs of elytra (USNM 96008; Fig. 4D-—J). Compared with the adult specimens, the anterior end is similar except that the palps are bulbous on the proximal half and filiform distally (Fig. 4D). The elytra are oval, delicate, with few scattered oval micropapillae; the posterior elytra on segment 13 are very small and incompletely developed (Fig. 4E, F). The bracts on the notopodia are not as well developed and the neuropodia have more prominent subtriangular presetal lobes (Fig. 4G, H). The notosetae and neurosetae are fewer in number and more slender (Fig. 4I, J). Etymology.—The species is named riftense for its association with the rifts in the region of the hydrothermal vent areas of the Galapagos and 21°N. Distribution.—East Central Pacific in Galapagos Rift, in 2447-2518 meters, associated with the unnamed deep-sea mussels and vestimentiferans, Riftia pa- chyptila. Also in East Pacific Rise at 21°N, in 2612-2633 meters, associated with ampharetid polychaetes, Alvinella pompejana, giant clams, Calyptogena magni- fica, and vestimentiferans, Riftia pachyptila. Comparisons. — Lepidonotopodium riftense is close to L. fimbriatum (see Key to the species of Lepidonotopodium). It differs in its smaller size (up to 13 mm in length, 7 mm in width, up to 25 segments, compared to up to 37 mm in length, 16 mm in width and up to 30 segments). The elytra are more delicate, with only scattered micropapillae, compared with the thick, leathery elytra with the char- - acteristic raised macrotubercles and numerous microtubercles of L. fimbriatum. The jaws of L. riftense have numerous minute basal teeth, compared with the relatively few and larger basal teeth of L. fimbriatum. Lepidonotopodium williamsae, new species Figs. 5-7 Material.—East Central Pacific, from dives of the Alvin on 3 vent areas along the Galapagos Rift in 1979: MUSSEL BED, 00°47.89'N, 86°09.21’W:— Dive 880, 21 Jan, 2493 m, mussel washings, paratype (USNM 96026). GARDEN OF EDEN, 00°47.69'N, 86°07.74'W:—Dive 884, 25 Jan, 2482 m, clam bucket from mussels, paratype (USNM 96027). ROSE GARDEN, 00°48.25’N, 86°13.48’'W:—Dive 984-32, 1 Dec, 2451 m, mussel washings, holotype (USNM 96023), paratype (USNM 96024), 2 paratypes (USNM 96025). 858 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON = A Zz LY Uf Fig. 5. Lepidonotopodium williamsae, holotype, USNM 96023: A, Anterior end, dorsal view; B, Anterior end, ventral view; C, Dorsal view of left segments 13 (elytrigerous) and 14 (cirrigerous); D, Dorsal view of posterior end showing right parapodia of segments 21 to 25 and left parapodia of segments 24 and 25; E, Right first elytron from segment 2; F, Right fifth elytron from segment 9, with detail of microtubercles and scattered micropapillae; G, Right eleventh elytron from segment 21, with detail of same. Scales = 2 mm for A—D; 2 mm for E-G. Pacific Ocean off Western Mexico, 20°50’N, 109°06’W, OASIS Alvin dives in 1982:—Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 64 paratypes (USNM 96032).—Dive 1219-10A & B, 25 Apr, 2612 m, Riftia and clam wash, coarse fraction, 5 paratypes (USNM 96028).—Dive 1221-15, 4 May, 2618 m, Riftia and Calyptogena wash, coarse fraction, 5 paratypes (USNM 96033).—Dive 1223-17, 7 May, 2616 m, rubble, 3 small paratypes (USNM 96029).— Dive 1225-7, 9 May, 2618 m, fine fraction, paratype (USNM 96484).—Dive 1226-7, 10 May, 2616 m, Riftia, Calyptogena and Alvinella wash, coarse fraction, 2 paratypes (USNM 96030).—Dive 1227-3, 11 May, 2616 m, slurp sample, paratype (USNM 96031). Description.—The holotype, from the Galapagos Rift area, has a length of 26 mm, a width of 13 mm, including setae, with 25 segments. The adult paratypes from this area range in size from 11 to 26 mm in length, 6 to 14 mm in width, with 25 segements. The largest paratype from 21°N has a length of 36 mm, a VOLUME 97, NUMBER 4 859 width of 14 mm, with 26 segments. The adult paratypes from this area range in size from 10 to 36 mm in length, 5 to 14 mm in width, with 24 to 26 segments. They are brownish to tan in color, elongate-oval, flattened ventrally and arched dorsally, rounded anteriorly and posteriorly. The 11 pairs of elytra are attached eccentrically on prominent elytrophores, with dorsal cirri on the posterior seg- ments (Fig. 5A, C, D). The elytra are large, imbricate, covering the dorsum, oval, subreniform to subtriangular in shape (Fig. SE—G). They are thick, stiff, opaque, without raised macrotubercles. Their surface is nearly covered with rounded to conical microtubercles, along with scattered globular micropapillae; fine “hairs” or “‘bacteria” may be attached to the microtubercles (Fig. 5F). The dorsal cirri have cylindrical cirrophores, wider basally, attached dorsoposteriorly on the no- topodia; the styles are tapering and extend slightly beyond the neurosetae (Figs. 5C, D; 6B). The dorsal tubercles on the cirrigerous segments are elongated and inflated (Figs. SC; 6B). The surfaces of both the elytrophores and dorsal tubercles have numerous ciliated ridges (Figs. 5C, D; 6A, B). The prostomium is bilobed, the anterior lobes prominent, cylindrical, extending anteriorly, each with a small frontal filament; lateral antennae are absent (Fig. 5A, B). The median antenna is inserted in the anterior notch, having a short cylindrical ceratophore and subulate style extending to about the tips of the palps. The palps are cylindrical, smooth, tapering, about one and a half times the length of the prostomium. Eyes are lacking, although a pair of tannish spots on a whitish area on the posterior half of the prostomium, each with a darker spot, may suggest a pair of “eyes” (Fig. 5A). The tentaculophores of the tentacular segment are lateral to the prostomium, each with a single aciculum, without setae. The styles of the 2 pairs of tentacular cirri are subequal in length to the palps, the ventral tentacular cirri slightly shorter than the dorsal ones (Fig. 5A, B). The second or buccal segment bears the first pair of large elytrophores, biramous parapodia and ventral or buccal cirri attached basally on prominent cirrophores lateral to the mouth; their styles are similar to the tentacular cirri and longer than the following ventral cirri (Fig. 5A, B). The ventral mouth is enclosed in upper, lateral and posterior lips between segments | and 2 (Fig. 5B). The opening of the large muscular pharynx is encircled by 7 pairs of bulbous papillae, subequal in size (Fig. 7B). The 2 pairs of dorsal and ventral jaws are fused medially and the basal parts are serrated with 5-7 teeth (Fig. 7B-D). The biramous parapodia have shorter notopodia located on the anterodorsal sides of the longer neuropodia (Fig. 6A). The notopodia are subconical with projecting acicular lobes hidden by the numerous notosetae and enclosed antero- dorsally by well-developed large flaring bracts (Fig. 6A—C). The neuropodia are diagonally truncate and deeply notched on the upper part. The presetal acicular lobes project dorsally beyond the rounded postsetal lobes (Fig. 6A, B, D). The distal margins of the notopodial acicular lobes, notopodial bracts and neuropodial lobes are fimbriated with slender papillae to which numerous “hairs” or “‘bacteria”’ are attached. The notosetae are very numerous, forming radiating bundles; they vary in length from short to longer, much stouter than the neurosetae, with spinous rows on the distal part and blunt tapered bare tips; most of the notosetae have numerous fine “hairs” or “‘bacteria” attached (Fig. 6E). The neurosetae are nu- merous, forming fan-shaped bundles. The few upper supracicular neurosetae have 2 rows of prominent spines and tapered bare tips (Fig. 6F). The very numerous 860 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Baan. UKs RIS 4438, Fig. 6. Lepidonotopodium williamsae, holotype, USNM 96023: A, Right elytrigerous parapodium, anterior view; B, Right cirrigerous parapodium, posterior view; C, Right notopodium, posterior view, with detail of distal papillae, notoaciculum dotted; D, Right neuropodium, anterior view, with detail of papillae, neuroaciculum dotted; E, Short and long notosetae and one thickly covered with “hairs” or “bacteria”; F, Two upper neurosetae, with detail of tip and middle part; G, Middle and lower neurosetae. Scales = 1 mm for A—D; 0.1 mm for E-G. VOLUME 97, NUMBER 4 : 861 F Fig. 7. Lepidonotopodium williamsae, A, paratype, USNM 96024: B, C, paratype, USNM 96032: D, small paratype, USNM 96028; E-G, young paratype of 13 segments, 1.5 mm long, 1.5 mm wide, USNM 96029: A, Ventral view of left segments 1 1-15; B, End view of distal end of extended pharynx; C, Dorsal jaws of same, spread apart; D, Dorsal and ventral jaws of smaller specimen, spread apart; E, Dorsal view of anterior end; F, Right middle elytron; G, Right elytrigerous parapodium, posterior view. Scales = 2 mm for A, B; 0.2 mm for C; 0.2 mm for D, E; 0.1 mm for F, G. subacicular neurosetae have slightly hooked bare tips and small areas with finely spinous rows on the cutting edge (Fig. 6G). The ventral cirri are short, tapered, smooth, attached on the middle of the posterior face of the neuropodia (Fig. 6A, B). Segmental or nephridial papillae are not obvious. On about half of the adult specimens, there are 4 pairs of elongate ventral papillae on segments 12, 13, 14, and 15, attached on the middle of the bases of the neuropodia and extending to near the bases of the ventral cirri (Fig. 7A). The pygidium is visible dorsally as a bulbous lobe wedged between the parapodia of the 3 posterior smaller segments (23-25); anal cirri are lacking (Fig. 5D). 862 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Numerous young specimens were collected on the East Pacific Rise at 21°N, in vestimentiferan wash and in rubble. The smallest specimens have a length of 1.5 to 2 mm, a width of 1.5 mm, with 11 to 16 segments (Fig. 7E—G). The anterior end is similar to the adults in shape except that the palps, tentacular cirri and buccal cirri are wider basally, with longer distal filamentous regions (Fig. 7E). The notopodial bracts are not as well developed (Fig. 7E, G). The notosetae are similar to the adults except that the larger ones are more slender. The neurosetae are similar to the upper neurosetae of the adults, having 2 prominent spinous rows. The elytra are delicate, lacking microtubercles and micropapillae (Fig. 7F). Etymology.—The species is named for Isabelle Williams of the Woods Hole Oceanographic Institution in recognition of her careful work in separating and retrieving the numerous polynoids on which this study is based. Distribution. —East Central Pacific on the Galapagos Rift, in 2451-2493 meters, associated with the unnamed deep-sea mussels. Also on the East Pacific Rise at 21°N, in 2612-2633 meters, associated with ampharetid polychaetes, Alvinella pompejana, giant clams, Calyptogena magnifica, and vestimentiferans, Riftia pachyptila. Comparison. — Lepidonotopodium williamsae may be separated from the other two species of Lepidonotopodium by the following key to the species. Key to the Species of Lepidonotopodium 1. Prostomium with prominent cylindrical anterior lobes (Fig. 5A). Elytral surface covered with numerous conical microtubercles and scattered mi- cropapillae, without raised macrotubercles (Fig. 5E—G). Notosetae much stouter than neurosetae, with spinous rows on distal part (Fig. 6E). Upper neurosetae with 2 rows of prominent spines (Fig. 6F); lower neurosetae minutely spinous (Fig. 6G). Pharynx with 7 pairs of papillae, subequal in size; jaws with 5-7 basal teeth (Fig. 7B—D). Without anal cirri (Fig. 5D). Some specimens with 4 pairs of elongate ventral papillae on segments 12-— DESOTO EUAN). ok etiet caret eV, cn fon SPs a L. williamsae, new species — Prostomium with subtriangular anterior lobes (Fig. 2A; fig. 2A, in Petti- bone, 1983). Notosetae similar in width to neurosetae, with relatively few (4-9) widely-spaced spines along one side (Fig. 4A; fig. 4E, in Pettibone, 1983). Upper neurosetae not markedly differing from lower ones (Fig. 4B; fig. 4F, in Pettibone, 1983). With pair of anal cirri (fig. 2D, in Pettibone, 1983). Some specimens with long ventral papillae on segments 11 and 12 (Fig. 4@> fig: 3A; in Pettibone, 1983)... oc ck oc. oe oe Se 2 2. Elytra with 2 raised macrotubercles per elytron, numerous microtubercles and scattered globular micropapillae (figs. 1, 2E—G, in Pettibone, 1983). Pharynx with 7-9 pairs of papillae, unequal in size; jaws with few basal teeth ©—9: Fig: | @—E: fig: 3B; ©)in Pettibone, 1983) — 2. eee th) Ee ee AIT a UR Aaa ON eT ae, Ree: mare ga L. fimbriatum Pettibone, 1983 — Elytra without raised macrotubercles, with scattered clavate micropapillae, with or without variable number of small projections (Fig. 2F—H). Pharynx with 7-9 pairs of papillae; jaws with numerous basal teeth (Fig. 2 DE) oe HR OS Ad, Ee SSD Sar os L. riftense, new species VOLUME 97, NUMBER 4 863 Acknowledgments I wish to thank J. F. Grassle and I. Williams of the Woods Hole Oceanographic Institution for the material on which this study is based, as well as the members of the Galapagos Rift Biology Expedition in 1979 and the OASIS group of Scripps Institution of Oceanography in 1982, including K. Smith. The manuscript ben- efited from the review of Meredith L. Jones. This paper is contribution no. 52 of the Galapagos Rift Biology Expedition and contribution no. 23 of the OASIS Expedition, both supported by the National Science Foundation. Literature Cited Boss, K. J., and R. D. Turner. 1980. The giant white clam from the Galapagos rift, Calyptogena magnifica species novum.— Malacologia 20(1):161-194. Desbruyéres, D., F. Gaill, L. Laubier, D. Prieur, and G. H. Rau. 1983. Unusual nutrition of the “Pompeii worm”? Alvinella pompejana (polychaetous annelid) from a hydrothermal vent en- vironment: SEM, TEM, !3C and '°N evidence.— Marine Biology 75:201—205S. , and L. Laubier. 1980. Alvinella pompejana gen. sp. nov., Ampharetidae aberrant des sources hydrothermales de la ride Est-Pacifique.—Oceanologica Acta 3(3):267-273. Jones, M. L. 1981. Riftia pachyptila, new genus, new species, the vestimentiferan worm from the Galapagos rift geothermal vents (Pogonophora). — Proceedings of the Biological Society of Wash- ington 93(4):1295-1313. Kenk, V. C., and B. R. Wilson. (In Press). A new mussel (Bivalvia, Mytilidae) from hydrothermal vents in the Galapagos rift zone— Malacologia. Pettibone, M. H. 1983. A new scale worm (Polychaeta: Polynoidae) from the hydrothermal rift-area off Western Mexico at 21°N.— Proceedings of the Biological Society of Washington 96(2):392-— 399. . 1984. A new scale-worm commensal with deep-sea mussels on the Galapagos hydrothermal vent (Polychaeta: Polynoidae).— Proceedings of the Biological Society of Washington 97(1): 226-239. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 864-881 TWO NEW SPECIES OF THE STPHONOECETES COMPLEX FROM THE ARABIAN GULF AND BORNEO (CRUSTACEA: AMPHIPODA) J. Laurens Barnard and James Darwin Thomas Abstract.—A new species, Siphonoecetes arabicus, is described from the Arabian Gulf, and a new species B. wongi is described in the new genus Bor- neoecetes from Borneo. The former species does not fit recently described sub- genera, thus indicating a necessary reassessment of those taxa. A new species of Siphonoecetes from the Arabian Gulf and a new species of Borneoecetes, new genus, from the shallow continental shelf off north Borneo are described. The description of the Arabian species seems crucial at this time because of a massive oil spill in the Arabian Gulf in late 1982, a year after it was first collected by Dr. John C. McCain. Dr. McCain informed us that his survey of the benthos on the Arabian side of the Gulf near Dhahran showed this species to comprise nearly 30 percent of the total amphipod specimens in the sand biotope. This new species, Siphonoecetes arabicus, spins a tube from amphipod-silk, glues coarse grains, especially shell fragments, to the tube and simultaneously glues the tube to dead shells of bivalves or inside dead gastropod and bivalve mollusk shells, calcareous polychaete tubes, or into detached movable-finger crab claws, all of appropriate sizes. Owing to its small adult size, less than 3 mm in body length (smaller than 6 mm in body-antennae length), one assumes the competition for snail shells with hermit crabs is quite limited. Some of the variety of abodes selected by S. arabicus for gluing down its tubes are shown in Fig. 1. Although Harada (1971) found S. tanabensis able to drag its unfixed tubes around on the bottom by use of antenna 2, we initially judged that the heavy shells to which S. arabicus cements its tube would be too massive and too heavy to be moved by the very feeble appearing amphipod. However, Dr. J. Just of the Zoological Mu- seum, University of Copenhagen, informs us that he has observed siphonoecetines able to drag very heavy loads and believes that most of the abodes we show in Fig. 1 could be moved by the amphipods. Other species of Siphonoecetes are also known to inhabit shells of various sorts, for example, Siphonoecetes pallidus in Dentalium sp. (Sars 1895) in deep waters off Norway (100-300 m), S. australis also from Dentalium from Australia (Steb- bing 1910), S. conchicola in Olivella sp. (Gurjanova 1951) from the Japan Sea, and S. sabatieri in snails of the genera Bittium and Rissoa (Chevreux and Fage 1925). But, apparently, some species always form simple silk tubes with agglu- tinated shell and mineral fragments, such as S. colletti (see Chevreux and Fage 1925). Several papers being published by Dr. Just give many more details of habitats of various species in the Siphonoecetes complex. We follow the diagnostic method of Myers and McGrath (1979) who described Siphonoecetes from the British Isles and provided a key to the seven species of VOLUME 97, NUMBER 4 865 2.0mm Fig.1. Abodes of Siphonoecetes arabicus: a, Tube of amphipod silk and agglutinated shell fragments glued to bit of coral rubble; b, Same glued to bivalve; c, Same glued inside movable finger of crab claw; d, Same glued inside snail-shell; e, Same glued inside crab claw; f, Same glued inside similar snail-shell from different angle. Europe. Because of the closeness of Siphonoecetes and Borneoecetes we use the same diagnostic method for B. wongi. Harada (1971) gave a key to all species of Siphonoecetes known at that time; we have therefore expanded the diagnoses from the Myers-McGrath format to include the characters used by Harada in his world treatment of the genus. A paper by Just (1983) has been utilized in our work to classify these species 866 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON at generic level; his work may be consulted for extensive diagnoses and relation- ships of genera. Explanation of Figure Legends.—Upper case letters in illustrations are ex- plained in the following list; lower case letters to right of capital letters or in the body of an illustration are explained also in the following list; lower case letters to the left of capital letters are provided for subsidiary figures to note illustrated specimens listed in “‘Material.”’ For each page of figures one main specimen is called “‘unattributed” and lacks letter designation. A, antenna: B, body; C, coxa; F, accessory flagellum; G, gnathopod; H, head; J, pleopod; K, spine or seta; L, labium; M, mandible; O, outer plate or ramus; P, pereopod; R, uropod; S, max- illiped; 7, telson; U, labrum; V, palp; W, pleon; X, maxilla; Y, urosome; Z, oostegite. d, dorsal; m, medial; 7, right; s, setae removed; u, unflattened. Siphonoecetes arabicus, new species Figs. 2-5 Diagnosis.—Size of adults about 2.25—2.80 mm. Body delicate, fragile, unpig- mented in alcohol. Rostrum pointed, extending as far as ocular lobes, latter pro- tuberant, quadrate. Eyes small. Antenna | barely exceeding apex of article 4 on antenna 2, flagellum with 5 well-developed articles tipped with tiny sixth. Articles 4 and 5 of peduncle on antenna 2 equal in length, flagellum with 4 articles, first two articles long and equal in juveniles, article 1 becoming slightly dominant in tropical adults, articles 3—4 vestigial, spines absent from article 1. Coxa 1 scarcely broader than long, anteroventral corner weakly and bluntly attenuate, all setae confined to small distance on margins. Gnathopod 1 with slender, unexpanded propodus (article 6) bearing 2—3 major posterior spines. Gnathopod 2 propodus ovate, palm sloping, equal to half of posterior margin, no extra spines present other than spine defining palm. Uropod | peduncle with about 4 dorsolateral setae evenly distributed, ventral apex extended as strongly serrate or fimbriate lamina. Uropod 1 rami unequal in length and thickness, inner shorter and thinner than outer, inner not swollen; outer ramus with dorsal spines, when present, evenly distributed, not mingling with apical spines. Uropod 2 peduncle apicoventral lamella strongly fimbriate; inner ramus more than 90 percent as long as outer ramus. Inner rami of uropods 1 and 2 each with 1 apical spine, apices of both rami with hollows bearing armaments. Uropod 3 peduncle with 2-3 apical spine- setae shorter than peduncle, ramus distinct, with 2—4 setae, middlemost (plumose) about twice as long as peduncle. Description and Notes on Illustrations.—Most of morphology shown in illus- trations; description largely based on holotype male. Eyes weak, formed of several irregular elements with scattered brownish pigment. Lateral and medial surfaces of peduncle on antenna 1 almost equally setose; accessory flagellum vestigial, composed of evanescent boss bearing 4 diverse setae. Lateral surfaces of pedun- cular articles on antenna 2 setose but lacking spines, medial surface of article 4 in both sexes with proximal to distal spine formula of 1-2, formula on article 3 in male = 1-2, in female = 0-1. Female antenna 2 relatively smaller than male but when adjusted to same magnification as in illustrations, articles 3—4 of pe- duncle of similar size, article 5 relatively longer in female, flagellum relatively 867 ga. = i =e ne oe . / I ya An 5 Ate) PS sy BS, Ae | tl res: Ree Ns, XQ W Alr female “‘f”’ VOLUME 97, NUMBER 4 Fig. 2. Siphonoecetes arabicus, unattributed figures, holotype, male “‘a” 2.46 mm; f 2.75 mm; m male ““m” 2.51 mm. 868 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON longer and thinner than in male; flagella with 4 articles; spinal differences slight (see illustrations). Epistome unproduced; upper lip articulated to epistome, subcircular, weakly truncate or slightly excavate below. Mandibular incisors toothed, right and left laciniae mobiles with 2 and 3 teeth respectively, right and left rakers = 2 and 3, each molar with flake and seta, palp composed of one long article flared and bifid apically when properly flattened, notch bearing second tiny article tipped with seta. Labium as shown. Inner plate of maxilla 1 evanescent, outer plate with 7 spines (one hidden in illustration), medial pair bifid, palp ordinary. Inner plate of maxilla 2 with 3 medial setae, no facials. Inner plate of maxilliped with 2 main stubby apical spines, outer plate with sharp medial chisel-spines, palp article 4 small, stubby, bearing long nail and accessory nail. Coxae all disjunct; coxa 1 of subadults generally with only 2 anterior setae and 6 on anteroventral corner (groups of 3-3 lateral and medial). Propodus (article 6) of gnathopod 1 in both sexes with tiny tooth apically representing palm, posterior margin with 2 spines, long in female, short in male. Gnathopod 2 without strong sexual distinction, lobe of carpus (article 5) with one stout spine, palm weak and defined by one stout spine, palm near hinge in female with cusp. Pereopods 3-4 alike, article 4 enveloping reduced article 5, article 6 attached to oblique apical margin of article 5, dactyl simple (bearing subapical meatus). Pereopods 5-6 subsimilar, small, article 5 with facial stridulating (or non-skid) surface, pereopod 7 elongate, lacking rough surfaces, dactyls of pereopods 5—7 with outer cusp. Large gills present on coxae 3-6 in male but in female gills 3—4 reduced greatly and gill 5 slightly reduced, thin oostegites present on coxae 2-5, formulae of anterior setae = 3-3-3-4, distal setae = 2-2-3-4, posterior setae = 2-2-2-3 (on female “‘f”’). Epimera 1-3 each with midposterior seta in notch, epimeron 1 with 4 large ventral setae (one missing in illustration) and one posteroventral setule, epimeron 2 with 4 large ventral setae and 2 smaller posteroventral setae, epimeron 3 with 2 small ventral setae. Pleopods all subsimilar, each peduncle with medial lobe bearing 2 coupling hooks, formula of lateral peduncular setae on pleopods 1-3 = 3-2-2, outer rami all with 7 articles, formula of articles on inner rami = 7-7-6. Urosomites 2 and 3 fused dorsally, urosome otherwise naked. Inner ramus of uropod | much smaller than outer ramus, dorsal minor spination of rami variable from side to side as shown. Uropod 2 as shown. Larger specimens and females generally with more setae on uropod 3 than male holotype, female “f’’ ramus with 5-6 setae, apex of peduncle with 5, medial peduncle with 1 seta; male “‘m”’ 2.51 mm with 4 apical setae on peduncle and 5 on ramus; female ““k”? 2.07 mm with 3 setae on peduncle apex and 3 on ramus. Plaques on telson with small corneal hooks (not well enough resolved in oil-immersion microscopy to illus- trate). Female generally much more setose and spinose than male, especially on gnathopods. Holotype.—USNM No. 195105, male “a,” 2.46 mm. Type-locality.— Arabian Gulf, off Manifa, 27°30'35”N, 49°00'15”E, Sta 1S1, 2- 3 m, 21 Nov 1981, coll. Dr. J.C. McCain; fine medium grain sediments, near coral reef, 70°F, salinity 42%o. Material and Distribution.—56 specimens from one station in the Gulf of Ara- bia, kindly supplied by Dr. John C. McCain of Tetra-Tech, Ltd., Dhahran, Saudi VOLUME 97, NUMBER 4 Fig. 3. Siphonoecetes arabicus, all figures, holotype, male “a” 2.46 mm. 869 870 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Siphonoecetes arabicus, unattributed figures, holotype, male “a” 2.46 mm; f = female “f” 2.75 mm. VOLUME 97, NUMBER 4 871 =a SSE EO So Ser S—— Fig. 5. Siphonoecetes arabicus, unattributed figures, holotype, male ‘“‘a’’ 2.46 mm; f = female “‘f”’ 2.75 mm. 872 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Arabia; male “d’’ 2.72 mm from crab claw; female “‘f’’ 2.75 mm (illustrated) from gastropod; female “‘k” 2.07 mm from crab claw; male ‘“‘m’’ 2.51 mm (head illus- trated) from gastropod. Ecology.—Dr. McCain informs us that this species represents about 30 percent of the specimen count for amphipods in shallow waters along the shores of Saudi Arabia in the Arabian Gulf. The species builds a thin parchment tube of amphipod silk and attaches the tube inside the shells of Dentalium sp., Caecum sp., or other gastropods and often on the curved or flat surfaces of other shelled organisms such as pelecypods or the broken calcareous fragments of many other, unidentified shell-forming organisms. Relationship. — This species does not fit precisely any of the subgenera described by Just (1983) as shown in the accompanying table. Four characters are used by Just to delineate those subgenera: (1) the presence or absence of spines on article 1 of the flagellum on antenna 2; (2) the normal ventral apex of uropod 1 bearing a straight comb as contrasted with a process; (3) the presence or absence of an apical hollow on the inner rami of uropods 1 and 2; (4) the presence of a strong row or essentially no row of apical setae on the peduncle of uropod 3. Siphonoecetes arabicus: has only 1 spine on article 1 of the flagellum of antenna 2 and thus would barely be in the subgenus Orientoecetes but would remain very close to the other subgenera; has the normal uropod 1 and thus is not Central- oecetes; has the hollowed apex on the inner rami of uropeds 1—2 and thus has affinities with the subgenus Siphonoecetes and not Centraloecetes nor Orientoe- cetes; has 3 apical setae on the peduncle of uropod 3 and thus lies in between any of the subgenera (Table 1). By running this species through the key of Harada (1971) the similarities appear to be with S. erythraeus and S. orientalis, both of which belong to Orientoecetes. This species differs from S. erythraeus Ruffo (1959:20) from the Red Sea, in (1) the sharply bifid and splayed apex of the mandibular palp and the presence of a minute article 2 on the mandibular palp; (2) the denser setation of uropod 3 (7 setae versus 4); (3) the presence of only one (versus 2) apical spine on the inner ramus of uropod 2; (4) the much stouter and better developed hand of female gnathopod 2. This species does not appear to be S. orientalis Walker (1904, pl. 7, fig. 49), from Ceylon, because gnathopod 2 of both sexes has a distinct palm and broader hand than illustrated by Walker. The flagellum of antenna 1 in S. orientalis has 14 articles. The spines on the flagellum of antenna 2 seem to be short, thick, evenly curved and regularly distributed. Siphonoecetes australis Stebbing (1910), from Australia, has a broadened hand on gnathopod 1 and a strongly shortened and unornamented inner ramus on uropod 1. The rostrum is very blunt. This species is very similar to S. sabatieri de Rouville (see Chevreux and Fage 1925:362) from the Mediterranean Sea, but differs in (1) the absence of thick spines mixed among the setae on the posterior margin of the hand on gnathopod 2; (2) the lack of one thick spine each on the peduncle and ramus of uropod 3; (3) the presence of only 1 (versus 2) apical spine on the inner ramus of uropod 1; (4) the different form of coxa 1 in which the plate is smaller, more quadrate, less attenuate and with the setae confined to less marginal length; many other characters of S. sabatieri are as yet poorly known. All other species of Siphonoecetes have either a longer rostrum, no eyes, extra VOLUME 97, NUMBER 4 873 Table 1.—Characters of subgenera in Siphonoecetes, compared to Siphonoecetes arabicus. Siphono- Centralo- Oriento- Character ecetes ecetes S. arabicus ecetes Antenna 2 flagellum spines on article 1 (male) 0 0 0 + Uropod 1 normal + 0 a a Uropod 2 inner ramus with hollowed apex + 0 =r 0 Uropod 3 apical setae on peduncle 0 5+ 3 0 posterior spines on gnathopod 2 besides the spine defining the palm, more weakly fimbriate peduncles of uropods 1-2, more strongly shortened inner ramus of uropods | or 2, the presence of pigmentation, long setae on the peduncle of uropod 3 or distinct spines or uropod 3 (two kinds of spine-setae, thick and thin). Now that our species has come to light, we detect that the following attributes and characters of several species of Siphonoecetes must be better described so that new distinctions can be attributed to those species: the flattened and bifid aspect of the mandibular palp; diversity in male and female gnathopods; and spine counts both on the apices and margins of uropods 1-2. Borneoecetes, new genus Diagnosis.— Rostrum normally pointed. Coxae 3—4 with simple distal margin. Large gills present on coxae 3-6; oostegites present on coxae 2—5 in female. Gnathopod 1 simple; gnathopod 2 subchelate. Dactyls of pereopods 3—7 with accessory tooth. Urosomite 2 free. Telson free, but proximal margin hidden by urosomite 2. Uropod | biramous, inner ramus reaching three-fourths along outer, both rami of similar structure, with strong apical spine(s); uropod 2 uniramous. Uropod 3 with distomedially expanded peduncle and setiferous ramus. Type-species.— Borneoecetes wongi, new species. Etymology.—Named for Borneo, the type-locality. Relationship.—Dr. Jean Just has given us extensive help from his recent ex- perience with undescribed species in the vicinity of this taxon. At first we con- sidered this species to be Bubocorophium tanabensis (Harada, 1971), but Dr. Just assures us that the shape of the mandibular palp, the uniarticulate palp of maxilla 1 and unfused urosomites make such identification impossible. The latter char- acter makes impossible the assignment of our species to Bubocorophium Karaman (1981). The only other available genus with uniramous uropod 2 is Rhinoecetes Just (1983) but our species differs from that genus in the ordinary head lacking anterodorsal depression. Our species “‘more or less” lacks spines on article 1 of the flagellum on antenna 2 (it often has one seta thickened sufficiently to be called a spine) but we have decided to ignore the spine problem for the moment as spines can only be evaluated synoptically by a single observer looking at all known species or by rendering unequivocal diagrams to define spines. Borneoecetes wongi, new species Figs. 6-9 Diagnosis.—Size of adults about 2.00-2.75 mm. Body delicate, fragile, unpig- mented in alcohol. Rostrum pointed, extending as far as ocular lobes; latter round- 874 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A b> ZB JL — Main large setae on epimeron | = 3, epimeron 2 = 7. Setae of uropod 1 peduncle, dorsolateral = 6, apicolateral = 5; peduncle of uropod 2 = 3 and 1 proximal; peduncle of uropod 3 = 2, ramus = 2. Juvenile “i, 0.98 mm.—Flagellum of antenna 1 with 3 large articles tipped with tiny iea Antenna 2 as stout as adult, first 2 articles equal in length as in Japanese adults of B. tanabensis. Example of weaker setation than in adults, coxae 1-7 = 6-9-4-3-2-1-0, with anterior and posterior divisions on coxae 5 and 6 being 1-1 and 0-1. Propodus (article 6) of gnathopod 1 with only | main posterior spine, gnathopod 2 with only | palmar defining spine. Epimera 1-3 all with large posterior crotch seta, main ventral setae = 1-2-0. Peduncle of uropod 1 with only 2 apicolateral setae, outer ramus with 3 apical spines, none dorsally. Only spine on uropod 2 on apex of ramus. See illustrations of uropod 3 and telson apex, peduncle of uropod 3 with one apicomedial seta, ramus with one long seta, vestige of inner ramus appearing through chitin of peduncle. Each apical plaque of telson with 3 weak hooks. Holotype.—USNM No. 195136, male “n’’ 2.03 mm, illustrated. Type-locality.— Borneo, north coast of Pameh off Sungei Buloh, 4°N, 113°57’E, 2-3 m, 1983, temperature 31.5°C, salinity 32%o, Wong Tat Meng Sample MCF 40 (71 specimens). Material.—Type-locality, male “‘n’’ 2.03 mm (main illustrations), female 2.59 mm, ?male “‘q’ 2.55 mm, male “r” 2.14 mm, male “s”’ 1.47 mm, ?sex op 1.61 mm, female “‘v’ 2.71 mm (illustrated), juvenile “1” 0. 98 mm (illustrated), juvenile “‘z’’ 0.90 mm. Also 15 specimens from MCD 53 and 9 specimens from MCD 47, Bea from same area in depths of 3—4 m and 2-3 m respectively. Etymology.—The species is named for Dr. Wong Tat Meng of Universiti Sains Malaysia. Remarks.—We immediately noticed the partially fused ramus of uropod 3 in our material and Harada (1971) shows the ramus completely fused to the peduncle in Bubocorophium tanabensis. Dr. Just informs us that this occurs infrequently in populations of the many species of the group he has examined. This fusion of uropod 3 to the peduncle was confusing to Harada who therefore identified the segment of urosomite 3 as the peduncle and the remainder of the appendage as the ramus, whereas one can distinguish in Harada’s figures the fused ramus and the medial peduncular protrusion armed with 2 setae. Hence, one must correct Harada’s key to the species of Siphonoecetes so as to state that the medial pro- trusion on the peduncle of uropod 3 actually bears 2 setae; this is crucial to separating B. tanabensis from its neighbor, S. conchicola Gurjanova, from the Japan Sea, 5-12 m, associated with Olivella shells. That species is shown by Gurjanova to have about 5 peduncular setae and 7-8 ramal setae (one of which appears actually to be a stout spine). Harada’s figures 8-5 and 8-6 show the peduncle with 1-3 medial and 1—4 ramal setae. Harada discusses the probability that Gurjanova’s illustration of uropod 2 can be interpreted to show only one ramus, in which case conchicola indeed belongs to Bubocorophium. “F8) 99 VOLUME 97, NUMBER 4 879 Fig. 9. Borneoecetes wongi, unattributed figures holotype male “n’’; i = juvenile “i” 0.98 mm; v = female “v” 2.71 mm. 880 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Harada noted that Barnard (1969) accepted Siphonoecetes as having one or two rami on uropod 2 (a condition Barnard had taken from his interpretation of S. conchicola). Karaman later established Bubocorophium to accommodate tana- bensis and ‘‘maybe”’ conchicola. Just (1983) has continued the division of si- phonoecetines in greater detail and verifies the validity of the characters used in our diagnosis of Borneoecetes above. We find the palp of maxilla 1 to be uniarticulate unlike any other siphonoecetine (Just has found a 2-articulate palp in 30 other species). Whether or not this has validity as a generic character must await Dr. Just’s further studies. Acknowledgments We thank the collectors of the material, Dr. Wong Tat Meng of Universiti Sains Malaysia, Penang; and Dr. John C. McCain of the University of Petroleum and Minerals, Dhahran, Saudi Arabia, who collected his material under auspices of Tetra-Tech, Ltd. We thank Linda B. Lutz of Mobile, Alabama for drawing Figure 1; and Janice Clark of the Smithsonian Institution for laboratory assistance. Car- olyn Cox Lyons of New York City kindly inked our drawings. We thank Dr. Jean Just of the Zoological Museum, University of Copenhagen, who took much time from his sabbatical leave in Australia to help us, for his advice and for copies of his manuscripts in press. The second author was supported by NSF grant DEB 8121128. Literature Cited Barnard, J. L. 1969. The families and genera of marine gammaridean Amphipoda.— United States National Museum Bulletin 271:1—535, figs. 1-173. Chevreux, E., and L. Fage. 1925. Amphipodes.—Faune de France 9:1—488, figs. 1-438. Gurjanova, E. 1938. Amphipoda, Gammaroidea of Siaukhu Bay and Sudzukhe Bay (Japan Sea).— Reports of the Japan Sea Hydrobiological Expedition of the Zoological Institute of the Academy of Sciences, USSR, in 1934 1:241-404, figs. 1-59. . 1951. Bokoplavy morej SSSR i sopredel’nykh vod (Amphipoda-Gammaridea).— Akademiia Nauk SSSR, Opredeliteli po Faune SSSR 41:1-1029, figs. 1-705. Harada, E. 1971. A new amphipod of the genus Siphonoecetes from the shallow bottom of southern Japan, with reference to the diagnoses of the genus and its species.— Publications of the Seto Marine Biological Laboratory 18:355-378, figs. 1-11. Hirayama, A. 1984. Taxonomic studies on the shallow water gammaridean Amphipoda of West Kyushu, Japan II. Corophiidae.— Publications of the Seto Marine Biological Laboratory 29:1- 92, figs. 1-100. Just, J. 1983. Siphonoecetinae subfam. n. (Crustacea, Amphipoda, Corophiidae) 1: Classification. — Steenstrupia 9(6):117-135, 12 figs. Karaman, G.S. 1981. Revision of some genera of family Corophiidae with description of three new genera. — Poljoprivreda i Sumarstvo 26:13-24. Myers, A. A., and D. McGrath. 1979. The British and Irish species of Siphonoecetes Kroyer (Am- phipoda-Gammaridea).—Journal of Natural History 13:211-—220, figs. 1-16. Ruffo, S. 1959. Contributo [13] alla conoscenza degli Anfipodi del Mar Rosso.—Sea Fisheries Re- search Station, Haifa, Bulletin 20:1—36, figs. 1-6. Sars, G. O. 1895. Amphipoda.—An account of the Crustacea of Norway 1: i—vili, 1-711, plates 1- 240 + 1-8 suppl. pls. Stebbing, T.R.R. 1910. Crustacea. Part V. Amphipoda. Scientific Results of the Trawling Expedition of H.M.C.S. “Thetis.” — Australian Museum, Memoir 4, 2:565-658, pls. 47-60. Walker, A. O. 1904. Report on the Amphipoda collected by Professor Herdman, at Ceylon, in VOLUME 97, NUMBER 4 881 1902.—Report to the Government of Ceylon on the Pearl Oyster Fisheries of the Gulf of Manaar, Supplementary Report 17:229-—300, plates 1-8. (JLB) Division of Crustacea, NHB-163, Smithsonian Institution, Washington, D.C. 20560; (JDT) Newfound Harbor Marine Institute, Route 3, Box 170, Big Pine Key, Florida, 33043. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 882-908 A REVIEW OF THE NEOTROPICAL PREDACEOUS MIDGE GENUS PAR YPHOCONUS (DIPTERA: CERATOPOGONIDAE) Gustavo R. Spinelli and Willis W. Wirth Abstract.—A key is presented for identification of females of the 38 species of the Neotropical genus Paryphoconus Enderlein, including the following 11 species described as new: apicalis, brunneipennis, ecuadorensis, fittkaui, flinti, fusciradi- alis, latipennis, leei, maya, paranaensis, and steineri. Ceratobezzia flavida Johann- sen and Bezzia telmatophilus Macfie are transferred to Paryphoconus (New Com- binations). Paryphoconus lanei Wirth is a junior synonym of P. flavidus (Johannsen) and P. travassosi Lane a junior synonym of P. subflavus Macfie (New Synonymy). Previous erroneous identifications and synonymy are corrected and new distri- bution records are presented. Since the publication of Wirth and Ratanaworabhan’s revision of the Steno- xenini in 1972 a large amount of new Neotropical material of the genus Pary- phoconus Enderlein has accumulated in the U.S. National Museum of Natural History. In connection with concurrent studies of the Neotropical species of Bezzia Kieffer we have discovered several species that bridge the taxonomic gap between the two genera and necessitate a re-evaluation of the tribe Stenoxenini. At the same time we have separated a significant number of undescribed species of Paryphoconus and found a number of erroneous or new synonymies that are reported here. As a result of our studies we now recognize 38 species of Pary- phoconus, of which 11 are described as new. Unfortunately we have been unable to find good characters for separation of species in the male sex, which shows considerable sexual dimorphism from the corresponding females. Our re-evalu- ation of the tribe Stenoxeninz1 is still incomplete, pending completion of our studies on Bezzia and revision of generic concepts in the Palpomyia-Bezzia complex. For explanation of our terminology see the papers by Wirth (1952) and Downes and Wirth (1981). All material examined is from the collection of the National Museum of Natural History in Washington, D.C., unless otherwise noted. Genus Paryphoconus Enderlein Paryphoconus Enderlein, 1912:57. Type-species, Paryphoconus angustipennis En- derlein, by original designation. References.—Mayer, 1952:582 (egg; classification); Lane, 1956:299 (revision; key); Mayer, 1959:233 (pupa); Lane, 1961b:453 (key); Wirth and Ratanawora- bhan, 1972:1369 (revision). Diagnosis.— Moderate to large size species; body slender, dorsoventrally flat- tened. Eyes bare, widely separated. Female antenna much elongated, segments with dense setose hairs, these sometimes conspicuously flattened (hastate). Palpus 5-segmented; 3rd segment slender, without sensory pit. Mesonotum narrowed in VOLUME 97, NUMBER 4 | 883 front, sometimes conically produced, usually with an erect anteromedian spine or tubercle. Legs long and slender; femora unarmed; femoral and tibial vestiture of very fine, erect hairs; last 3 tarsomeres relatively short; 4th tarsomere cordiform; 5th tarsomere without ventral batonnets. Female claws short and equal, each with small internal basal tooth. Wing long and narrow, often with color pattern; r-m crossvein distinct but short; costa long, extending nearly to wing tip; 1 very narrow radial cell, posterior branch of radius closely approaching costa; medial fork broadly sessile. Female abdomen slender, with paired gland rods and eversible glands. Male genitalia: 9th sternite short, 9th tergite tapering with well-developed apicolateral lobes; gonocoxite slender, gonostylus elongated; aedeagus with high, rounded basal arch and slender, pointed distomedian process projecting poste- riorly from just behind tip; parameres separated, with contiguous clavate apices. Discussion.—Females of Stenoxenus Coquillett can be distinguished readily from those of Paryphoconus by the following combination of characters: thorax broadly rounded in front, without anterior spine; palpus 4-segmented; and wing with vein M2 elbowed at the base. Males of Stenoxenus closely resemble those of Paryphoconus in wing venation, but lack the anterior mesonotal spine, have the palpus 4-segmented (only | segment distad of the one bearing the sensory organ), and the parameres are fused on the distal portion. In only a few species of Paryphoconus does the female costa extend to the wing tip as it does in all Stenoxenus; the non-elbowed media and 5-segmented palpus will readily distin- guish these. Similarly, the females of some Paryphoconus species with the anterior mesonotal spine lacking can be distinguished by wing venation and their 5-seg- mented palpus. We have seen a few males with the parameres separate distally but with 4-segmented palpus, that we cannot associate with any known females of either genus; we are unable to interpret this combination of characters on the basis of the limited material available. Possibly they may represent species that may bridge the gap between the two genera, or one or the other of the characters may prove to be unreliable. Reared specimens in which the sexes can be definitely associated by pupal characters are badly needed for all species. Included Species. — aemulus Macfie, 1940b:180. Guyana. amapaensis Land, 1961a:450. Brazil. angustipennis Enderlein, 1912:57. Brazil. anomalicornis Kieffer, 1917:333. Colombia and Paraguay. apicalis n. sp. Ecuador. barrettoi Lane, 1946:203. Brazil. batesi Lane, 1961b:455. Brazil. brunneipennis n. sp. Brazil. ecuadorensis n. sp. Ecuador. enderleini Lane, 1956:301. Brazil. fittkaui n. sp. Brazil. flavidus (Johannsen), 1943a:761 (Ceratobezzia). Guyana. flinti n. sp. Mexico. fusciradialis n. sp. Brazil. fuscus Lane, 1946:206. Brazil. golanensis Lane, 1961b:456. Brazil. 884 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON grandis Macfie, 1939:6. Brazil. guianae Macfie, 1940c:180. Guyana. kiefferi Lane, 1956:302. Brazil. lanei Wirth, 1959:236. Brazil. Syn. of flavidus (Johannsen). latipennis n. sp. Colombia. leei n. sp. Colombia. macfiei Lane, 1946:203. Brazil. maya n. sp. Belize. mayeri Wirth, 1959:236. Brazil. neotropicalis (Lane), 1948:229 (Macropeza). Brazil. nigripes Macfie, 1939:8. Argentina. nubifer Macfie, 1939:5. Brazil. oliveirai Lane, 1956:303. Brazil. paranaensis n. sp. Argentina. paulistensis Lane, 1961b:457. Brazil. sonorensis Wirth and Ratanaworabhan, 1972:1374. Mexico. steineri n. sp. Peru. subflavus Macfie, 1940a:23. Guyana. telmatophila (Macfie), 1940b:77 (Bezzia). Brazil. terminalis (Coquillett), 1904:90 (Ceratopogon). Nicaragua. travassosi Lane, 1956:304. Brazil. Synonym of subflavus Macfie. unimaculatus Macfie, 1940c:179. Guyana. wirthi Lane, 1961b:458. Brazil. wygodzinskyi Lane, 1946:208. Brazil. Key to the Species of Paryphoconus (Females) 1. Wing entirely hyaline, without dark markings (may become gradually and slightly unfuscated toward apex) .......................000- 2 - Wing entirely or partly infuscated, at least a distinct dark marking Over r-m crossvein and/or at wing tip ........................5. 8 Wing Hyaline Without Markings 2 Legs entirely yellowish or whitish; mesonotum with yellowish ground color smalllispecies; wing 2—3)mml long)... ae 3 - Legs with at least hind tibia darkened at apex; mesonotum dark brown or blackish; larger species, wing 3-7 mm long ............ 4 3. Mesonotum corps yellow and with strong anterior spine .... Siete eta. SALAS TS an ped ek, © rere none) ana gare mayeri Wirth - Mesonotum yellow except for large blackish mark in front of wing LOOL) anitenornspinerabSenini eee ae wygodzinskyi Lane 4. Large species, wing 5—7 mm long; legs brown, fore femur yellow, hind |temur blackishyonidistall/4) eee eee batesi Lane - Smiallerspecies,, wing 392—47°0 mim long. eee 5 ): All femora and tibiae dark at apices ............ neotropicalis (Lane) - Legs yellow, hind! tibia panthy dark 9205 .\.-..)5050 Gee 6 6. Hind tibia brown at base and apex; hind tarsus pale at base; antenna slender, ight ‘brown’ ....2.-. 3500-400 oo ee ee ee paulistensis Lane VOLUME 97, NUMBER 4 i 885 8(1). 3). 14. T3)5 16. Hind tibia brown on distal '4; hind tarsus brown to base; antennal Seomentsishorthalleblackgint ae wine Mls ino ts wea WS wo. 7 Mesonotum without anterior spine; antenna longer, antennal ratio J TLS oe Rea oa RE cee roe sonorensis Wirth and Ratanaworabhan Mesonotum with prominent anterior spine; antenna short, antennal A UIORO alee tents eae epnth tee nf ite LL Bos eli paranaensis, n. sp. Wineainivscatedsoverentine SUTMACEN 5 ono se yee eae 9 Wing with dark spots or shading but not darkened over entire surface 9: Sud) arte Lourie ic eit Sou Rare Le eee) las ae ne 7 Wing Entirely Infuscated Remoradand tibiae brownish black..." .........5....5..-5-.-.. 10 Femora and tibiae yellowish, at most with narrow apices dark .... 11 Hind femur with subapical pale band; tarsi brownish; wing 5 mm Honma (ioral e) harcewocrce tee aA cg Ahi corai, Ge sane 5 aim cts cae eat fuscus Lane Hind femur without subapical pale band; tarsomeres 1-3 pale yellow; SATS SCM OM Sse irate 1 la ed eenncr es Siac Ce wegen eo Me oe aemulus Macfie angeispecies, wingss—/smmm ONG... .......0.505ne nena le 12 Smaller species, wing 1.5—3.0 mm long; (wing without well defined @istalidanksband) ta kke teens adit an finnuated | Ameya eres atae wel: 13 Wing with more intense dark band at apex: tibiae and hind femur with narrow apices dark; base of hind basitarsus pale .. barrettoi Lane Wing darker on anterior margin and along veins, without well-defined distal dark band; hind tibia dark on distal 4; hind basitarsus all [SURON AO Aes eens gee As er ee AGA PA et aaz ie LRA OES Sev EPIL ORE RISERS. F grandis Macfie Small species, wing 1.5 mm long; wing broad, uniformly infuscated; antenna entirely dark brown; legs with vestiture normal, the setae SNOB a Sock oh Pe 98 A f oth ecg tp EY LEN Spee REE POD AG SRO TR Do 14 Larger species, wing 2—3 mm long; wing narrower; antenna with base paler; legs with longer, fuzzier, pilose vestiture .................. US Mid and hind femora with prominent broad brown bands; wing infuscation much darker in radial cell, paler posteriorly; spermatheca larges O30 mmniim diameter 245). 95..) ese. 2 fusciradialis, n. sp. Femora unbanded; wing infuscation in radial cell scarcely darker than on posterior portion of wing; spermatheca smaller, 0.053 by ORO OF rants eS ve ALAR. fa ba a CS MER. Sd latipennis, n. sp. Fore tibia all brown; wing pale brownish, gradually darker toward COStaim “atest treele: ce tiergtat las ena A rat. cate Mead sendin ££ leei, n. sp. Fore tibia yellowish to pale brown; wing pale brown, veins darker Restle Fh Sik Win chan, iNe la tiNe vo che iene pent esc out Me ls! 16 Legs uniform yellowish to pale brown, apex of hind tibia darker, hind tarsus pale; antenna long and slender, antennal ratio 1.45, seg- ments 3—10 white, 11—15 brown; anal angle of wing poorly developed DER eto ee EOS ene een ee a ee ee brunneipennis, n. sp. Legs dirty yellowish brown, hind femur, tibia, tarsus dark brown; antenna pale brown, segments 3—10 with pale bases, antennal ratio 2.38; wing with anal angle broad ............... ecuadorensis, n. sp. AN OD). WD Bde 24. Jo) (229). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Wing with a single dark spot or area Over r-m crossvein ......... 18 Wing with additional markings or without a spot at r-m crossvein 21 Dark Spot Over R-M Crossvein Mesonotum yellow, or yellowish with dark markings ............ 19 Mesonotum dark brown or blackish; (wing 2.5 mm long) . wirthi Lane KEAanTge Species, wine oon Ong areas unimaculatus Macfie Smaller species, wing 2:3—3.0 mm long ................555355= 20 Legs entirely yellowish; mesonotum without anterior spine; marking on r-m crossvein small and inconspicuous ........... enderleini Lane Hind tibia with dark apical band; mesonotum with strong anterior spine; marking on r-m crossvein prominent ..... flavidus (Johannsen) Wing with dark mark at r-m crossvein and also dark on entire anterior Margin, wing tip pale; (thorax blackish; large species, wing 6 mm LONG) pice peta, ee Re ee ne eee eae goianensis Lane (part) Wing with dark marking at apex. 25. .0 2 4..5 054 00 See DD, Wing without dark mark over r-m crossvein; mesonotum blackish 23 Wing with dark mark over r-m crossvein; mesonotum various ... 25 R-M Crossvein Pale: Dark Mark at Wing Tip Wing with a single dark mark, at apex; legs yellowish except apex of hind tibia brown; small species, wing 2.0-2.9 mm long .......... 24 Wing with dark apical shading and a mark on lower branch of cubital fork (vein Cul); legs blackish; larger species, wing 5.5 mm long ... i hg oe etter o dk oe OR aN ae pete ad Shah eS LT ef nigripes Macfie Antenna all black; wing 2.9 mm long ............... guianae Macfie Antenna with segments 3-10 white, 11-15 black; wing 2.0 mm long SE OcRR Beas Se Fen Seen Min eae rade terminalis (Coquillett) Wing infuscated at base, over r-m crossvein, and at broad distal portion, forming more or less 3 bands, these sometimes connected by infuscation along veins; mesonotum yellowish, usually with dark brown markings; large species, wing 4—7 mm long .............. 26 Wing markings confined to r-m crossvein, wing tip, and rarely nar- rowly along veins; mesonotum various; size various, usually smaller 29 Angustipennis Group; Large Species With Banded Wing Mesonotumventinelyayellowishise 5556 eee oliveirai Lane Mesonotum with dark brown markings ........................ 27 Antenna stout with prominent, dense, long, black, hastate setae Poe gh, EER ASIN? ot a LA Rs pee ae Ce REE oe angustipennis Enderlein Antenna long and slender with slender setae .................... 28 Antenna uniformly dark brown) 2. 278. ee ee flinti, n. sp. Antenna with segments 3-10 pale yellow; wing more or less infus- cated between the 3 dark bands ................. macfiei Lane (part) VOLUME 97, NUMBER 4 | 887 DI): 30. 31(29). 35(32). 36. Ss 38. Mesonotum yellowish or pale brown, with dark brown markings .. 30 Mesonotum uniform dark brown or blackish ................... 31 Yellowish Species Larger species, wing 5.5 mm long; mid and hind tibiae dark at apices; wing without infuscation on vein Cul .................. kiefferi Lane Smaller species, wing 2.2-3.0 mm long; only hind tibia with dark apex; wing with faint dark mark along vein Cul ..... subflavus Macfie Mesonotum Blackish Antenna with 5 distal segments contrasting white, proximal segments blackish with prominent hastate setae ........ anomalicornis Kieffer Antenna with segments uniformly colored or with distal 5 darker .. 32 Wing With Dark Marks Over R-M Crossvein and at Tip JANINE Foxes, watery GOs) wean MOY socsassncscacutosucscacdeucs 33 Smialllspeciesswihee-5—>5) Ohms ONG ams ie eee ree 35 Legs completely dark brown; wing infuscated around entire margin Ra REE R MONI TF oes a aE Toes wah coi Gis cat? ol goiaensis Lane (part) Legs with tibiae yellowish at least in part ...................... 34 All tibiae extensively marked with brown; wing very broad as in TOUS VAM AULA "Se ee kee CAE ae ee SO ae amapaensis Lane Only hind tibia marked with brown; no dark mark on vein M1; small darkemarksinubaseofsanalicelli@ 4a see macfiei Lane (part) Antenna short, all black; wing veins all infuscated .... nubifer Macfie Antenna long and slender, more or less pale at base; veins not in- UsCated! @xeg ot ison Aiovel/Ore CU ow seco cee osc csscasocsocscosnese 36 Hind tibia dark brown; (legs dirty yellowish, fore and mid tibiae brownish, hind tarsus wholly brown; wing white, narrow apex in- fuscated, faint dark marks on r-m and Cul; antenna with segments SallO ware, WSIS) lomoNmo ON) eae beaaeeosoobnoceee cook maya, N. sp. Hind tibia with only distal half or less brown ................... 37 Wing distally with broad brown band, radius and costa, r-m and Cul more or less infuscated; antenna with proximal segments brownish on apices, antennal ratio no more than 2.24 .................... 38 Wing white with narrow tip infuscated, r-m infuscated, Cul very slightly infuscated; antenna white on 3-10, pale brown on 11-15, extremely long and slender with slender setae, antennal ratio 2.60; (legs pale yellow, narrow apex of hind tibia brown, basal '2 of hind basitarsus pale; halter dark brown) .................. fittkaui, n. sp. Wing whitish with narrow tip infuscated; hind tarsus all brown; antennal ratio 2.24; 4 pairs of gland rods ............ Steineri, N. sp. Wing gray overall with broad, ill-defined infuscation distally; legs pale yellow, narrow apex of hind tibia dark brown; hind basitarsus pale at base; antennal ratio 1.65; 1 pair of gland rods .. apicalis, n. sp. 888 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Paryphoconus aemulus Macfie Paryphoconus aemulus Macfie, 1940c:180 (male; Guyana).— Lane, 1946:206 (fe- male; Brazil).— Wirth and Ratanaworabhan, 1972:1371 (female notes; figs.; rec. Brazil). Distribution. — Brazil, Guyana. Notes.—The species is placed in the key on the basis of female characters given by Lane (1946) and Wirth and Ratanaworabhan (1972) from Brazilian specimens. The species resembles P. fuscus Lane and barrettoi Lane in its completely infus- cated wing, but it is a smaller species (wing 3.3 mm long), has legs entirely dark except for the trochanters and extreme bases of the femora, and tarsomeres 1-3, which are pale yellowish. The antennae are unusually long, slender, and whitish in color. Spermatheca oval. Paryphoconus amapaensis Lane Paryphoconus amapaensis Lane, 1961a:450 (female; Brazil; fig. wing).— Wirth and Ratanaworabhan, 1972:1371 (notes). Distribution. — Brazil. Notes. — Wing length 6 mm. A large, shining dark brown species. Legs pale, fore and mid femora dark brown distally, tibiae brown except distally on mid leg; hind leg with only knee brownish; tarsi pale, distal tarsomeres brownish on fore and hind leg. Wing unusually broad, with expansion at margin of mediocubital fork; extensive infuscation narrowly around wing margin and across r-m crossvein to base of vein M3+4. Halter pale. No new data. Paryphoconus angustipennis Enderlein Fig. 13a Paryphoconus angustipennis Enderlein, 1912:58 (female; Brazil; figs.).— Macfie, 1939:5 (Brazil; fig. wing). — Lane, 1946:202 (Brazil; figs.).— Lane, 1956:300 (Bra- zil, Argentina).— Wirth and Ratanaworabhan, 1972:1371 (redescription; figs.; distribution). Distribution. — Argentina, Brazil, Colombia, Peru, Venezuela. New Record.—ARGENTINA: Misiones, Arroyo Piray Guazu, San Pedro, 22 Nov 1973, O.S. Flint, Jr., 5 females. Notes.— Wing length 5.8-6.5 mm; a large yellowish brown species with dark brown abdomen; legs yellow, distal tarsomeres and narrow tip of hind tibia dark brown; halter knob brownish. Antenna with torus yellowish, flagellum dark brown, the segments elongated, cylindrical, antennal ratio 1.35; last 5 segments with dense black hastate setae. Wing (Fig. 13a) with extensive dark brown pattern as figured; costal ratio 0.91. Spermatheca large, elongate oval, measuring 0.194 by 0.097 mm including the short neck. Male much smaller, wing length 2.1 mm; costal ratio 0.86. Genitalia figured by Wirth and Ratanaworabhan (1972). The female from Mexico reported by Wirth and Ratanaworabhan was misidentified, and is here described as P. flinti n. sp. VOLUME 97, NUMBER 4 889 B C D Fig. 1. Paryphoconus apicalis, female: a, Antenna; b, Femora and tibiae of (top to bottom), hind, mid, and fore legs; c, Tarsi of (top to bottom) hind, mid, and fore legs; d, Spermatheca. Paryphoconus anomalicornis Kieffer Fig. 13b Paryphoconus anomalicornis Kieffer, 1917:333 (female; Colombia, Paraguay; figs.). — Wirth and Ratanaworabhan, 1972:1372 (redescribed; figs.; distribution). Distribution. —Brazil, Colombia, Honduras, Mexico, Paraguay, Venezuela. New Records.— BRAZIL: Amazonas, Rio Irapirapi, Cachoeira, 11 Jan 1963, E. J. Fittkau, at light, 1 female; Rio Solimoes, 15 Sep 1961, Fittkau, at light, 3 females. Notes.— Female wing length 3.1—4.2 mm. A shining black species with yellow legs, hind tarsus and extreme tip of hind femur dark brown. Wing (Fig. 13b) slightly infuscated along anterior margin; r-m crossvein and an apical area darker brown, slight infuscation narrowly along apices of veins M1, M2, M3+4, and Cul; costal ratio 0.88. Antenna with proximal flagellar segments black, bearing dense black hastate setae; distal 5 segments much elongated, contrasting whitish; antennal ratio 1.60. Spermatheca small and ovoid with long slender neck, mea- suring 0.097 by 0.072 mm, neck 0.011 mm. The females from Rio Solimoes are larger (wing 4.2 mm long) and darker than typical P. anomalicornis, with wing deeply infuscated along veins and around margins, the femora somewhat infuscated in midportions, and the hind tibia dark except at the base. Paryphoconus apicalis, new species Figs. 1, 13c Female.—Wing length 2.36 (2.23-2.45, n = 5) mm; breadth 0.62 (0.60-0.65, n= 5) mm. Head: Dark brown including scape and pedicel. Eyes separated for a distance equal to diameter of 5 ommatidial facets (0.092 mm). Antenna (Fig. la) dark brown, 2 or 3 basal segments slightly pale; lengths of flagellar segments in pro- portion of 38-26-24-26-24-24-25-27-77-72-72-74-80; antennal ratio 1.65 (1.53- 1.75, n = 5); verticils and setae sparse and fine, no black hastate setae. Palpus 890 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON short, yellow, with segments in proportion of 8-14-18-8-14; 3rd segment with scattered sensilla. Thorax: Uniformly dark brown; mesonotum with a well-developed, erect, blunt, anterior spine. Legs (Fig. 1b, c) pale yellow, apex of hind tibia dark brown; tarsi brown except tarsomeres 1 and 2 of mid leg and basal '2 of hind basitarsus pale; hind tarsal ratio 2.00 (n = 5). Wing (Fig. 13c) slender, anal angle poorly developed; membrane gray with broad, well-defined infuscation distally; veins more or less infuscated, especially costa, radius, and r-m crossvein; radial cell very narrow; costal ratio 0.91 (0.90-0.93, n = 5). Halter dark. Abdomen: Brown. One pair of well-seclerotized gland rods; spermatheca (Fig. 1d) pyriform, measuring about 0.060 by 0.046 mm, with 0.014 mm neck. Male.— Unknown. Distribution. — Ecuador, Panama. Types. — Holotype female, Ecuador, Pastaza, Cononaco, 30 May 1976, J. Cohen, at light. Paratypes, 4 females, as follows: ECUADOR: same data as holotype, 2 females. PANAMA: Darien Prov., Santa Fe, 11 Sep 1967, A. Broce, 2 females. Discussion.—This species is very similar to P. fittkaui, maya, and steineri; characters to separate all of them are found in the key. Paryphoconus barrettoi Lane Paryphoconus barrettoi Lane, 1946:203 (female; Brazil; fig. wing, spermatheca). Distribution. — Brazil. Notes.—No new records. The specimens from Mexico and Venezuela described as P. barrettoi by Wirth and Ratanaworabhan (1972:1373) were misidentified and are actually P. grandis Macfie (which see). Paryphoconus barrettoi is a large species, wing 6.5 mm long; wing infuscated throughout, with distinct darker band on distal 4; base of wing very narrow, anal angle not developed; costal ratio 0.90. Antenna dark brown, with dense dark setae; antennal ratio 1.8. Thorax reddish brown, blackish at sides of mesonotum and in front of wing; mesonotum, scutellum, and postnotum densely covered with short pale pile; mesonotum with short anterior tubercle. Legs yellowish brown; narrow apices of femora and fore and mid tibiae, extreme base and distal 4 of hind tibia, apices of basitarsi, and all of tarsomeres 2—5, dark brown. Paryphoconus brunneipennis, new species Figs. 2, 13d Female.—Wing length 2.23 mm; breadth 0.67 mm. Head: Pale brown including scape, pedicel, and palpus. Eyes separated for a distance equal to diameter of 5 ommatidial facets (0.090 mm). Antenna (Fig. 2a) long and slender; brown, except antennal segments 3-10 and narrow bases of 1 1- 15 pale; lengths of flagellar segments in proportion of 57-38-38-39-40-41-37-36- 75-70-78-102-146; antennal ratio 1.45; verticils and setae sparse and fine, no black hastate setae. Palpus with lengths of segments in proportion of 8-14-20-13- 19; 3rd segment with scattered sensilla. Thorax: Brown with abundant small black mottlings; mesonotum with a well- developed, pale, pointed anterior spine. Legs (Fig. 2b, c) yellowish brown, narrow VOLUME 97, NUMBER 4 891 Fig. 2. Paryphoconus brunneipennis; a—d, female; e-f, male: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca; e, Parameres; f, Genitalia, parameres omitted. apex of hind tibia infuscated; tarsi yellowish, basitarsus of mid leg with a single row of palisade setae; hind tarsal ratio 2.40. Wing (Fig. 13d) broad posteriorly, anal angle poorly developed; membrane infuscated over entire surface; veins darker, especially on r-m crossvein area; radial cell narrow to tip; costal ratio 0.90. Halter pale. Abdomen: Brown. Apparently 4 pairs of sclerotized gland rods; spermatheca (Fig. 2d) subspherical with short neck, measuring 0.100 by 0.085 mm. Male.—Wing length 1.77 mm. Similar to female with usual sexual differences. Antenna with lengths of flagellar segments in proportion of 43-30-26-26-25-24-23-24-26-31-90-117-190. Wing with costal ratio 0.85. Genitalia (Fig. 2f): Ninth sternite narrow without caudomedian excavation; 9th tergite tapered distally, with 2 very well-developed apicolateral lobes. Gonocoxite with a triangular mesal protuberance; gonostylus shorter than gonocoxite, pointed. Aedeagus and parameres (Fig. 2e) typical of the genus. Distribution. — Brazil (Amazonas), Colombia. Types.— Holotype female, Brazil, Amazonas, Rio Solimoes, 15 Sep 1961, E. J. Fittkau, at light. Allotype male, Colombia, Valle, Rio Raposo, May 1965, V. H. Lee, light trap. Discussion. — This species is similar to P. ecuadorensis, n. sp. and P. /eei, n. sp., but differs from these by the pale coloration of the proximal antennal segments; other differences are found in the key. This group of similar species, all with the general wing infuscation as in P. fusciradialis and P. latipennis, but with narrower 892 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON wings, also are distinguished by the long pilose vestiture of the legs giving them a fuzzy appearance. This fuzziness is most noticeable in P. brunneipennis. Paryphoconus ecuadorensis, new species Figs. 3, 13e Female. —Wing length 2.20 mm; breadth 0.74 mm. Head: Pale brown. Eyes separated for a distance equal to diameter of 4 om- matidial facets (0.060 mm). Antenna (Fig. 3a) entirely pale brown, bases of seg- ments 3-10 very slightly paler; lengths of flagellar segments in proportion of 38- 22-21-20-19-20-21-23-92-78-83-79-103, antennal ratio 2.38; verticils and setae sparse and fine, no black hastate setae. Palpus pale brown; lengths of segments in proportion of 8-13-20-11-14; 3rd segment with scattered sensilla. Thorax: Pale brown; mesonotum with a very small, pale, anterior spine. Legs (Fig. 3b, c) dirty yellowish brown; hind leg darker; basitarsus of mid leg with a single row of palisade setae; hind tarsal ratio 2.60. Wing (Fig. 13e) broad, anal angle well developed; membrane infuscated over entire surface, veins dark brown; costal ratio 0.90. Halter pale brown. Abdomen: Pale brown, gland rods not visible; spermatheca (Fig. 3d) oval, without sclerotized neck, measuring 0.074 by 0.053 mm. Male.— Unknown. Distribution. — Ecuador. Type.— Holotype female, Ecuador, Pichincha, 113 km via Puerto Quito, 2700 ft, 30 Aug 1976, J. Cohen, CDC light trap. Discussion.—Paryphoconus ecuadorensis can be distinguished from P. brun- neipennis by the color of the antenna and legs, and by the well-developed anal angle of the wing. Paryphoconus fittkaui, new species Figs. 4, 13f Female.—Wing length 2.90 (2.80-—3.10, n = 3) mm; breadth 0.72 (0.71-0.75, n= 3) mm. Head: Dark brown including scape and pedicel. Eyes separated for distance equal to diameter of 4.5 ommatidial facets (0.070 mm). Antenna (Fig. 4a) very long and slender and slender setae; pale brown except segments 3—10 and narrow bases of 11-15 paler; lengths of flagellar segments in proportion of 32-21-21-23- 21-21-20-20-99-88-8 1-84-110, antennal ratio 2.60 (2.58—2.62, n = 3). Palpus short, yellow, lengths of segments in proportion of 10-15-23-13-21; 3rd segment with scattered sensilla. | Thorax: Dark brown; mesonotum with a very small, blunt anterior spine. Legs (Fig. 4b, c) pale yellow, narrow apex of hind tibia brown; proximal 2 of hind basitarsus pale, the rest brown; tarsi of fore and mid leg pale except tarsomeres 3-5 of fore leg infuscated; hind tarsal ratio 1.82 (1.80—-1.85, n = 3). Wing (Fig. 13f) narrow; membrane white, narrow tip of wing infuscated; costa, radius, and r-m crossvein more or less infuscated, rest of veins when infuscated, only at tip and very slightly; radial cell very narrow; costal ratio 0.88 (0.87-0.89, n = 3). Halter dark brown, apex of knob blackish. Abdomen: Dark brown. Four pairs of slender gland rods; spermatheca (Fig. 4d) VOLUME 97, NUMBER 4 893 By Cc Fig. 3. Paryphoconus ecuadorensis;, a—d, female; e-f, male: a, Antenna; b, Femora and Tibiae; c, Tarsi; d, Spermatheca; e, Parameres; f, Genitalia, parameres omitted. oval with well-sclerotized neck, measuring 0.065 by 0.050 mm with neck 0.023 mm. Male.—Unknown. Distribution. — Brazil (Amazonas). Types.— Holotype female, Brazil, Amazonas, Rio Marauia, Jan—Feb 1963, E. J. Fittkau, at light. Paratypes, 2 females, Amazonas, Rio Cueiras, Dec 1960, Fittkau, at light. Discussion. — This species can be recognized by the peculiar, very long antenna with slender setae, and very high antennal ratio (2.60). Paryphoconus flavidus (Johannsen), new combination Fig. 13g Ceratobezzia flavida Johannsen, 1943a:761 (female; Guyana).— Johannsen, 1940b: 767 (fig. antenna, thorax). Paryphoconus lanei Wirth, 1959:236 (male, female; Brazil; fig. male genitalia). — Mayer, 1959:233 (pupae; fig. respiratory horn).— Wirth and Ratanaworabhan, 1972:1374 (redescribed; figs.; distribution). Distribution. — Brazil. New Record.—BRAZIL: Mato Grosso, Cuiaba, 2 Apr 1972, W. H. Whitcomb, 12 females. 894 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Paryphoconus fittkaui, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca. Notes.—A small yellowish species, hind tibia usually with a narrow dark brown apical band; wing (Fig. 13g) whitish hyaline with a prominent dark spot over r-m crossvein and a faint dark line along vein Cul. Antenna dark brown with numerous strong hastate setae on segments 3-10; antennal ratio 1.78. Wing length 2.3 mm; costal ratio 0.89. Spermatheca ovoid with short tapering neck, measuring 0.105 by 0.072 mm. The synonymy of P. /anei and P. flavidus was established by direct comparison of the holotypes, that of P. flavidus by courtesy of Dr. Quentin D. Wheeler of Cornell University. Paryphoconus flinti, new species Figs. 5, 13h Female.—Wing length 4.30 (4.10—4.50, n = 4) mm; breadth 1.03 (0.99-1.05, n= 4) mm. Head: Yellowish, including scape, with numerous small dark mottlings. Eyes separated for a distance equal to diameter of 5 ommatidial facets (0.10 mm). Antenna (Fig. 5a) slender, pedicel brown, flagellum entirely dark brown; lengths of flagellar segments in proportion of 40-22-22-23-23-22-22-22-54-48-52-57-100; antennal ratio 1.50 (1.43-1.57, n = 2); verticils and setae sparse and fine, no black hastate setae. Palpus moderately long, 3 proximal segments brown, last 2 dark brown; 3rd segment with scattered sensilla; lengths of segments in proportion of 10-25-30-23-30. Thorax: Yellowish brown with numerous small dark mottlings; mesonotum with a small blunt, black anterior spine. Legs (Fig. 5b, c) dirty yellowish brown, tip of tibiae dark brown. Wing (Fig. 13h) brownish at base, at middle, and on broad apex, plus some infuscation along veins in between; costal ratio 0.91 (n = 4). Halter yellowish. Abdomen: Dark brown, terga shining black. Five pairs of well-sclerotized gland rods; spermatheca (Fig. 5c) elongate oval with short neck, measuring 0.138 by 0.088 mm. Male.— Unknown. Distribution. — Belize, Mexico. VOLUME 97, NUMBER 4 895 Cc D Fig. 5. Paryphoconus flinti, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca. Types. — Holotype female, Mexico, Tabasco, Rio Puyacatengo, E. Teapa, 28 Jul 1966, Flint and Ortiz. Paratypes, 3 females, as follows: BELIZE: Cayo District, Western Hwy. MP 66, Jun 1969, W. and D. Hasse, light trap, 3 females. MEXICO: Chiapas, Rio Tulija, 48 km S Palenque, 17 May 1981, C. M. and O. S. Flint, Jr. 1 female. Discussion. — This species can be distinguished from related species by the dark brown tips of the tibiae; it also differs from P. angustipennis by the sparse, fine, antennal setae and verticils, and from P. macfiei by the antennal coloration (seg- ments 3—10 white). Paryphoconus fusciradialis, new species Figs. 6, 131 Female.—Wing length 1.55 (n = 3) mm; breadth 0.60 (n = 3) mm. Head: Dark brown. Eyes broadly separated, for a distance equal to diameter of 7 ommatidial facets (0.110 mm). Antenna (Fig. 6a) entirely brown, bases of segments 3-10 very slightly paler; lengths of flagellar segments in proportion of 34-22-20-20-20-20-20-20-57-57-57-57-80; antennal ratio 1.74; verticils and setae sparse and fine, no black hastate setae. Palpus short, pale; lengths of segments in proportion of 5-10-20-12-14; 3rd segment broad, bearing a group of 5-6 sensilla on midportion. Thorax: Dark brown; mesonotum without anterior spine. Legs (Fig. 6b, c) yellowish except mid '3 of mid and hind femora, and distal '3 of hind tibia dark brown; tarsi yellowish, last 3 segments slightly infuscated; hind tarsal ratio 2.23 (2.15-2.30, n = 3). Wing (Fig. 131) with membrane infuscated over entire surface, anterior veins and radial cell brown; venation as figured; costal ratio 0.88 (n = 3). Halter pale brown. Abdomen: Dark brown. One pair of very short gland rods present; spermatheca (Fig. 6d) exceptionally large, spherical with short neck, measuring 0.130 mm in diameter. Male.— Unknown. Types.—Holotype female and 2 female paratypes, Brazil, Para, Belem, Apr 1970, T.H.G. Aitken, light trap in APEG Forest. 896 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON D Fig. 6. Paryphoconus fusciradialis, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Sper- matheca. Discussion.—This small species can be distinguished from the closely related P. latipennis n. sp. by the broad dark brown bands of the mid and hind femora, by the deep infuscation of the radial cell, by the extremely short gland rods, and by the voluminous spherical spermatheca. This species and P. Jatipennis have the general habitus of some species of the genus Bezzia, and may in fact be primitive members of the genus Paryphoconus indicating closer relationships of the two genera than previously thought. Paryphoconus grandis Macfie Fig. 13) Paryphoconus grandis Macfie, 1939:6 (female; Brazil; fig. wing). Paryphoconus barrettoi Lane of Wirth and Ratanaworabhan (misident.), 1972: 1373 (female; Mexico, Venezuela; figs.). Distribution. — Belize, Brazil, Mexico, Venezuela. Specimens Examined.—BELIZE: Cayo Dist., Western Hwy., 3 Jul 1969, W. Hasse, | female. MEXICO: On railway car from Mexico in quarantine at Nogales, Arizona, 24 Apr 1963, C. H. Spitzer, 2 females. VENEZUELA: Est. Portuguesa, Guanare, 10-13 Sep 1957, B. Malkin, 1 female (Calif. Acad. Sci.). Notes.— Wirth and Ratanaworabhan (1972) misidentified their material which they described as P. barrettoi Lane. This description and figures should therefore be compared with Macfie’s original description of P. grandis, which is in agree- ment. The large size (wing 5—7 mm long) and long costa (costal ratio 0.98) on the brownish wing with darker brown anterior margin and obscure dark lines along the tips of the medial and cubital veins are diagnostic (Fig. 13j). Halter with dark brown knob. Antenna short and uniformly black, without strong setae; antennal ratio 1.37. Legs yellowish brown, knees blackish; extreme tips of fore and mid tibiae, distal 2-3 tarsomeres on fore and mid legs, distal '4 of hind tibia and all of hind tarsus blackish. Thorax dark brown, with dense pubescence and short hairs; anterior tubercle of mesonotum small. Spermatheca large, oval; measuring 0.173 by 0.122 mm with narrow neck 0.018 mm long. VOLUME 97, NUMBER 4 . 897 Table 1.—Numerical values for species of Paryphoconus. Species Wing length (mm) Costal ratio Antennal ratio aemulus 3.3 0.89 1.18 amapaensis 6.0 — 0.99 angustipennis 5.8-6.5 0.91 1.35 anomalicornis 3.1-4.2 0.88 1.60 apicalis 2.2-2.5 0.91 1.65 barrettoi 6.5 0.90 ~ 1.8 batesi 6.5 0.95 — brunneipennis DP) 0.90 1.45 ecuadorensis Dol 0.90 2.38 enderleini 3.0 — 1.50 fittkaui 2.9 0.88 2.60 flavidus 2.3 0.89 1.78 flinti 4.3 0.91 1.50 fusciradialis 1.5 0.88 1.74 fuscus 5.0 0.87 3.10 goianensis 6.0 0.93 — grandis 5.0-7.0 0.98 — guianae 2.9 0.87 2.00 kiefferi 5.5 — 1.42 latipennis 1.5 0.91 2.05 leei 2.4 0.90 2.00 macfiei 6.5 0.90 — maya 2.5 0.88 1.95 mayeri 2.4 — _ neotropicalis 3.2 _ 1.10 nigripes 5.5 — 0.92 nubifer 3.0 0.87 1.20 oliveirai 4.0 — 1.55 paranaensis 4.6 0.99 0.71 paulistensis 4.0 — — sonorensis 3.8 0.98 1.18 steineri 2.8 0.90 2.24 subflavus 2.5 0.90 1.40 terminalis 2.0 0.87 1.70 unimaculatus 6.0 — 2.00 wirthi 2.5 — — wygodzinskyi 2.6 — — Paryphoconus guianae Macfie Fig. 13k Paryphoconus guianae Macfie, 1940c:180 (female; Guyana).—Lane, 1956:302 (Brazil record).—Lane, 1961b:456 (Brazil records).—Wirth and Ratanawo- rabhan, 1972:1377 (erroneous synonym of terminalis). Distribution. — Brazil, Ecuador, Guyana. New Record.—ECUADOR: Pastaza, Cononaco, 30 May 1976, J. Cohen, at light, 1 female. Notes.—A small black species; wing hyaline with tip faintly infuscated (Fig. 13k). Wing length 2.9 mm; costal ratio 0.91. Antenna all dark brown; antennal ratio 2.00. The species is almost identical with the widespread species, P. ter- 898 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON minalis (Coquillett) and Wirth and Ratanaworabhan erroneously made P. guianae a junior synonym in 1972. Paryphoconus terminalis is a smaller species (wing 2.0 mm long), and segments 3-10 of the antennae are whitish. Paryphoconus latipennis, new species Figs. 7, 131 Female.—Wing length 1.50 (1.30-1.60, n = 4) mm; breadth 0.54 (0.48-0.58, n = 4) mm. Head: Dark brown, except frontoclypeus and palpus pale brown. Eyes broadly separated, for a distance equal to diameter of 8 ommatidial facets (0.140 mm). Antenna (Fig. 7a) entirely brown; lengths of flagellar segments in proportion of 35-23-20-19-21-23-22-24-70-70-70-72-102; antennal ratio 2.05 (n = 2); verticils and setae sparse and fine, no black hastate setae. Palpus short, lengths of segments in proportion of 7-14-17-10-15. Thorax: Dark brown; mesonotum with a small, erect, anterior spine. Legs (Fig. 7b, c) yellowish, apex of hind tibia and distal 3 tarsomeres brown; hind tarsal ratio 2.42 (2.30—2.50, n = 4). Wing (Fig. 131) infuscated over entire surface; broad posteriorly, anal angle poorly developed; radial cell broad, vein R4+5 more or less arcuate; costal ratio 0.91 (0.90-0.93, n = 4). Halter pale. Abdomen: Brown. One pair of slender, lightly sclerotized, gland rods; sper- matheca (Fig. 7d) oval with short neck, measuring 0.053 by 0.046 mm, with neck 0.010 mm long. Male.—Wing length 1.30 mm; breadth 0.40 mm. Similar to female with usual sexual differences. Antenna with lengths of flagellar segments in proportion of 38- 24-22-21-20-20-20-20-20-20-25-75-100-135. Wing with costal ratio 0.88. Gen- italia (Fig. 7f): Ninth sternite with very deep caudomedian excavation; 9th tergite with 2 rounded apicolateral lobes. Gonocoxite short, with pointed mesal protu- berance and more or less developed ventromesal lobe; gonostylus elongated, slen- der. Aedeagus with very high basal arch and slender pointed tip. Parameres (Fig. 7e) separate, slender, each with distal portion slightly expanded in a bladelike process. Distribution.—Colombia, Venezuela. Types.— Holotype female, Colombia, Rio Raposo., Apr 1963, V. H. Lee, light trap. Allotype male, same data except May 1965. Paratypes, 3 females, as follows: COLOMBIA: same data as types except Oct 1964, 1 female, and Feb 1965, 1 female. VENEZUELA: Guarico, 12 km S Calabozo, 6 Feb 1969, P. and P. Span- gler, light trap, 1 female. Discussion.—The reduced size distinguishes P. /atipennis from all other con- geners with entirely infuscated wings except P. fusciradialis; characters to separate these two species are found in the key and the discussion of the latter species. Paryphoconus leei, new species Figs. 8, 14a Female.—Wing length 2.37 (2.32—2.45, n = 3) mm; breadth 0.66 (0.64-0.68, n= 3) mm. Head: Dark brown including scape and pedicel. Eyes separated for a distance equal to diameter of 3.5 ommatidial facets (0.060 mm). Antenna (Fig. 8a) brown, VOLUME 97, NUMBER 4 | 899 é D Fig. 7. Paryphoconus latipennis; a—d, female; e-f, male: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermathecae; e, Parameres; f, Genitalia, parameres omitted. bases of flagellar segments 4—10 pale; lengths of flagellar segments in proportion of 45-26-24-24-24-25-25-28-84-77-77-80-125; antennal ratio 2.00 (n = 2); ver- ticils and setae sparse and fine, no black hastate setae. Palpus short, last segment deeply infuscated; 3rd segment broad (palpal ratio 1.25) with scattered sensilla; lengths of palpal segments in proportion of 10-15-16-11-16. Thorax: Uniformly blackish; mesonotum with very small anterior spine. Legs (Fig. 8b, c) yellowish, fore tibia all brown; basitarsus of mid leg with single row of palisade setae; hind tarsal ratio 2.38 (2.28—-2.46, n = 3). Wing (Fig. 14a) narrow, especially toward apex; membrane infuscated over entire surface, gradually darker toward costa; radial cell narrow; costal ratio 0.90 (n = 3). Halter dark. Abdomen: Very dark brown. Four pairs of well sclerotized gland rods; sper- matheca (Fig. 8d) oval with very short neck, measuring 0.078 by 0.060 mm. Male.—Unknown. Distribution. —Colombia. Types.— Holotype female, Colombia, Valle, Rio Raposo, Aug 1964, V. H. Lee, light trap. Paratypes, 2 females, same data except Mar 1964 and Jun 1965S. Discussion.— This species can be distinguished from other related species by the narrow distal portion of the wing, and by the entirely brown fore tibia. Paryphoconus macfiei Lane Figs. 9, 14b Paryphoconus macfiei Lane, 1946:203 (female; Brazil; fig. wing).— Wirth and Ra- tanaworabhan, 1972:1371 (in key). 900 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON D Fig. 8. Paryphoconus leei, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca. Female.—Wing length 3.45 (3.35-3.55, n = 5) mm; breadth 0.87 (0.84-0.90, n= 5) mm. Head: Yellowish brown including scape and pedicel, with numerous small dark mottlings. Antenna (Fig. 9a) with segments 3-10 pale yellow, 11-15 dark brown; lengths of flagellar segments in proportion of 35-20-21-22-23-23-23-23-41-41- 47-50-80; antennal ratio 1.25 (1.12—1.37, n = 5); verticils and setae sparse and fine, no black hastate setae. Palpus moderately long, 3 proximal segments brown, last 2 dark brown; 3rd segment with scattered sensilla; lengths of segments in proportion of 10-18-28-18-28. Thorax: Yellowish to brown with small dark mottlings; mesonotum with very small, blunt, anterior spine. Legs (Fig. 9b, c) dirty yellowish brown; tibiae much infuscated, only tip of hind tibia dark brown; tarsi yellowish brown, last 2 tar- someres infuscated; hind tarsal ratio 2.20 (2.15—2.30, n = 5). Wing (Fig. 14b) as in P. angustipennis, but infuscation also lighter between the 3 bands; costal ratio 0.90 (0.90-0.91, n = 5). Halter pale. Abdomen: Brown, terga shining black. Five pairs of sclerotized gland rods; spermatheca (Fig. 9d) very elongate oval with short neck, measuring 0.156 by 0.060 mm. Male. —Unknown. . Distribution. — Brazil, Ecuador, Venezuela. Specimens Examined.—BRAZIL: Mato Grosso, Cuiaba, 4 Apr 1972, W. H. Whitcomb, 10 females. ECUADOR: Pastaza, Cononaco, 30 May 1976, J. Cohen, light trap, 3 females (wing bands more distinct in this series). VENEZUELA: Bolivar, Anacoco, Rio Cuyuni, 1—23 Aug 1979, Exp. Sta. La Salle, 1 female (wing bands distinct, only distal band prominent). Paryphoconus maya, new species Figs. 10, 14c Female.—Wing length 2.45 (2.35-2.51, n = 5) mm; breadth 0.64 (0.61-0.66, n=5) mm. Head: Dark brown. Eyes separated for a distance equal to diameter of 4 om- matidial facets (0.060 mm). Antenna (Fig. 10a) with flagellar segments 3—10 white, VOLUME 97, NUMBER 4 901 B D Fig. 9. Paryphoconus macfiei, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca. 11-15 brownish; lengths of flagellar segments in proportion of 30-23-20-20-20- 20-20-23-71-66-63-70-84; antennal ratio 1.95 (1.88-—2.00, n = 5); verticils and setae sparse and fine, no black hastate setae. Palpus short, pale yellow; lengths of segments in proportion of 7-13-16-14-16; 3rd segment with scattered sensilla. Thorax: Dark brown, almost blackish; mesonotum with small, blunt, anterior spine. Legs (Fig. 10b, c) dirty yellowish, fore and mid tibiae brownish hind tibia dark brown (specimens from Mexico with only distal 2 of hind tibia dark brown); tarsi brownish, hind tarsus wholly brown (base of hind basitarsus slightly pale in some specimens); hind tarsal ratio 1.90 (1.83—2.00, n = 5). Wing (Fig. 14c) narrow; membrane white, narrow apex infuscated; costa, radius, r-m, and Cul with faint dark mark; M1, M2, and M3+4, when infuscated, only at tips and then very slightly; costal ratio 0.88 (0.87-0.89, n = 5). Halter dark brown, end of knob blackish. Abdomen: Dark brown. Apparently with 3 pairs of sclerotized gland rods; spermatheca (Fig. 10d) oval with relatively long neck, measuring 0.058 by 0.049 mm with neck 0.018 mm long. Male.— Unknown. Types. — Holotype female, Belize, Cayo District, Western Hwy MP 66, Jun 1969, W. and D. Hasse, light trap. Paratypes, 8 females, as follow: BELIZE: same data as holotype, 6 females. MEXICO: San Luis Potosi, El Salto, 8 May 1964, F. S. Blanton, light trap, 2 females. Discussion.—The species can be distinguished from related species by the col- oration of the legs, especially the dark brown hind tibia. The antennal coloration is similar to that of P. fittkaui n. sp., but P. fittkaui is larger (wing length 2.9 mm) and the antennal ratio is much higher (2.60). Paryphoconus nubifer Macfie Fig. 14f Paryphoconus nubifer Macfie, 1939:5 (female; Brazil; fig. wing).— Lane, 1946:206 (Brazil record). — Lane, 1956:300 (in key). — Wirth and Ratanaworabhan, 1972: 1374 (female diagnosis; figs.; distribution). 902 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON C D Fig. 10. Paryphoconus maya, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca. Distribution.— Argentina, Bolivia, Brazil, Paraguay, Uruguay. New Records.—ARGENTINA: Misiones, Arroyo Piray Guazu, N San Pedro, 22 Nov 1973, O. S. Flint, Jr., 4 females. PARAGUAY: Rio Aquidoban, Cerro Cora, 29 Nov 1973, O.S. Flint, Jr., 1 female. Notes. — Wirth and Ratanaworabhan’s (1972) record from Panama is erroneous, and this species does not occur north of Bolivia and Brazil. The species is readily distinguished by the extensive dark mark around the wing tip and the prominently infuscated veins (Fig. 14f). The antenna is unusually short and entirely black. Paryphoconus oliveirai Lane Fig. 14g Paryphoconus oliveirai Lane, 1956:303 (female; Brazil).— Wirth and Ratanawo- rabhan, 1972:1374 (female redescribed; pupa; figs.; Brazil). Distribution. — Brazil, Colombia. New Record.—COLOMBIA: Amazonas, Leticia, 7 Sep 1969, D. H. Messer- smith, at light, 1 female. Note.— Wing as in Fig. 14g. Paryphoconus paranaensis, new species Figs. 11, 14h Female.—Wing length 4.60 (4.45—4.70, n = 3) mm; breadth 1.18 (1.15—1.20, n= 3) mm. Head: Dark brown. Eyes separated for a distance equal to diameter of 4 om- matidial facets (0.065 mm). Antenna (Fig. 1 1a) short, entirely dark brown; lengths of flagellar segments in proportion of 35-19-20-19-18-20-19-18-24-21-23-21-27; antennal ratio 0.71 (0.65—0.79, n = 3); verticils and setae sparse and fine, no black hastate setae. Palpus short, pale brown; lengths of segments in proportion of 13- 25-17-14-22; 3rd segment short and broad (palpal ratio 1.15), bearing scattered sensilla on midportion. Thorax: Dark reddish brown; mesonotum with prominent, pale, pointed an- terior spine (visible only on holotype). Legs (Fig. 11b, c) bright yellow, deep shade; VOLUME 97, NUMBER 4 903 D Fig. 11. Paryphoconus paranaensis, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Sper- matheca. knees blackish, narrow apex of hind tibia dark brown; tarsi dark brown except fore and mid basitarsi pale brown; hind tarsal ratio 2.20 (2.15—2.25, n = 3). Wing (Fig. 14h) pale, gradually darkening at distal '2 of costa to wing tip; costal ratio 0.99 (n = 3). Halter stem pale brown, knob dark. Abdomen: Dark brown, terga not shining. Spermatheca (Fig. 11d) elongate oval with moderately long neck; measuring 0.095 by 0.060 mm, neck 0.012 mm long. Male.— Unknown. Distribution. — Argentina. Types.— Holotype female and 2 female paratypes, Argentina, Misiones Prov., Arroyo Piray Guazu, NE San Pedro, 22 Nov 1973, O. S. Flint, Jr. Discussion. — This species 1s very similar to Paryphoconus sonorensis Wirth and Ratanaworabhan, from which it differs especially by the short antenna (antennal ratio 0.71 vs 1.18 in P. sonorensis). In addition, all the type material of P. sono- rensis lacks the prominent anterior mesonotal spine found in the holotype of P. paranaensis. D Fig. 12. Paryphoconus steineri, female: a, Antenna; b, Femora and tibiae; c, Tarsi; d, Spermatheca. 904 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a. ANGUSTIPENNIS g. FLAVIDUS bh. ANOMALICORNIS e. ECUADORENSIS k. GUIANAE f. FITTKAUI . |. LATIPENNIS Fig. 13. Wings of Paryphoconus species, females: a, angustipennis from San Pedro, Argentina; b, anomalicornis from Comayagua, Honduras; c, apicalis from Cononaco, Ecuador (holotype); d, brun- neipennis from Rio Solimoes, Brazil (holotype); e, ecuadorensis from Pichincha, Ecuador (holotype); f, fittkaui from Rio Marauia, Brazil (holotype); g, flavidus from Cuiaba, Brazil; h, flinti from Tabasco, Mexico (holotype); i, fusciradialis from Belem, Brazil (paratype); j, grandis from Cayo Dist., Belize; k, guianae from Conaco, Ecuador; 1, /atipennis from Rio Raposo, Colombia (paratype). Paryphoconus sonorensis Wirth and Ratanaworabhan Fig. 141 Paryphoconus sonorensis Wirth and Ratanaworabhan, 1972:1374 (female; Mexico (Sonora), U.S.A. (Oklahoma); figs.). VOLUME 97, NUMBER 4 905 Distribution. — Belize, Mexico, Panama, U.S.A. New Records.— BELIZE: Cayo Dist., Western Hwy. MP 66, Jun 1969, W. and D. Hasse, light trap, 2 females. PANAMA: Dolega, 12 Dec 1952, F. S. Blanton, light trap, 2 female; Naranjal, 10 Oct 1952, Blanton, light trap, 1 female. Notes.—Wing as in Fig. 141. Paryphoconus paranaensis from Argentina is al- most identical with P. sonorensis but that species has much shorter antenna (antennal ratio 0.71 in P. paranaensis, 1.18 in P. sonorensis), and there is a prominent pointed, pale, anterior spine on the mesonotum in P. paranaensis. The two species are widely separated in their distribution, P. sonorensis being essen- tially a Central American species, while P. paranaensis is restricted to Argentina. Paryphoconus steineri, new species Figs. 12, 14) Female.—Wing length 2.80 (2.60—3.00, n = 2) mm; breadth 0.71 (0.65-0.78, n= 2) mm. Head: Dark brown. Eye separation not visible on slide preparations available. Antenna (Fig. 12a) dark brown, 4 proximal segments slightly paler; lengths of flagellar segments in proportion of 40-26-24-26-26-25-25-25-96-88-88-92-122; antennal ratio 2.24; verticils and setae sparse and fine, no black hastate setae. Palpus short, yellow; lengths of segments in proportion of 9-15-20-14-18; 3rd segment with scattered sensilla. Thorax: Uniformly blackish. Mesonotum apparently without anterior spine. Legs (Fig. 12b, c) pale yellow, distal 3 of hind tibia dark brown; tarsi dark brown except tarsomeres | and 2 of mid leg and proximal 4 of basitarsus of hind leg pale; hind tarsal ratio 1.85 (1.80—1.90, n = 2). Wing (Fig. 14j) with anal angle poorly developed; membrane whitish with narrow tip infuscated; costa, radius and r-m crossvein all infuscated; M1, M2, M3+4 and Cul infuscated only distally; radial cell very narrow; costal ratio 0.90 (n = 2). Halter yellowish, end of knob darkish. Abdomen: Dark brown. Four pairs of sclerotized gland rods. Spermatheca (Fig. 12d) elongate oval, with long neck broad at base; measuring 0.070 by 0.045 mm, plus neck 0.025 mm long. Male.—Unknown. Distribution. — Panama, Peru. Types.— Holotype female, Peru, Cuzco, Quince Mil, 26 Jan 1979, W. E. Steiner. Paratype, | female, Panama, Cocle Prov., Hermita, 18 Sep 1952, F. S. Blanton, light trap. Discussion.— This species is named for Warren E. Steiner, in appreciation of his interest in collecting ceratopogonids for us. For separation from related species see the discussion under P. apicalis n. sp. Paryphoconus subflavus Macfie Fig. 14k Paryphoconus subflavus Macfie, 1940a:23 (female; Guyana).— Wirth and Rata- naworabhan, 1972:1376 (female redescribed; figs.; distribution). Paryphoconus travassosi Lane, 1956:304 (female; Brazil). —Wirth and Ratana- worabhan, 1972:1378 (female redescribed; figs.; Brazil). New Synonymy. 906 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON d. MAYERI j. STEINERI e. NIGRIPES k. SUBFLAVUS {. NUBIFER 1. TERMINALIS Fig. 14. Wings of Paryphoconus species, females: a, /Jeei from Rio Raposo, Colombia (holotype); b, macfiei from Cuiaba, Brazil; c, maya from Cayo District, Belize (holotype); d, mayeri from Ta- bacorama, Brazil (holotype); e, nigripes from Cuiaba, Brazil; f, nubifer from San Pedro, Argentina; g, oliveirai from Leticia, Colombia; h, paranaensis from San Pedro, Argentina (paratype); 1, sonorensis from Sonora, Mexico (paratype); j, steineri from Cuzco, Peru (holotype); k, subflavus from Rio Raposo, Colombia; 1, terminalis from Tocumen, Panama. Distribution. — Brazil, Colombia, Guyana. Note.— Wing as in Fig. 14k. The characters which Wirth and Ratanaworabhan (1972) used to separate P. subflavus and P. travassosi, the dark line along vein Cul, and the infuscation on the fore and mid tibiae, are entirely unreliable, and in our opinion P. travassosi is a junior synonym. VOLUME 97, NUMBER 4 907 Paryphoconus telmatophilus (Macfie), new combination Bezzia (Probezzia) telmatophila Macfie, 1940b:77 (male; Brazil; figs.). Bezzia telmatophila Macfie.— Lane, 1958:35 (male redescribed; fig. genitalia; Bra- zil). Distribution. — Brazil. Notes on Types.—Through the courtesy of Richard Lane and the Trustees of the British Museum (Nat. Hist.) Macfie’s holotype of Bezzia telmatophilus was examined and found to belong to the genus Paryphoconus. The male that Lane (1958) described as the allotype of P. telmatophilus is not congeneric with Macfie’s species and appears to represent the male (taken at the same time and place) of Lane’s new species Bezzia brasiliensis, which is a junior synonym of Nilobezzia schwarzii (Coquillett). Because of the difficulty of distinguishing species of Paryphoconus in the male we are treating this species as a species inquirenda in Paryphoconus. Paryphoconus terminalis (Coquillett) Fig. 141 Ceratopogon terminalis Coquillett, 1904:90 (female; Nicaragua). Paryphoconus terminalis (Coquillett).— Wirth and Ratanaworabhan, 1972:1377 (combination; erroneous synonym guianae; redescribed female; figs.; distribu- tion). Distribution. — Belize, Brazil, Costa Rica, Ecuador, Honduras, Nicaragua, Pan- ama, Peru. New Records.— BELIZE: Cayo Dist., Western Hwy. MP 66, Jun 1969, W. and D. Hasse, light trap, 2 females. ECUADOR: Pastaza, Cononaco, 30 May 1976, J. Cohen, at light, 1 female. Notes.— Wing as in Fig. 141. Wirth and Ratanaworabhan’s synonymy of P. guianae Macfie under P. terminalis was erroneous. Characters for separating the two species are given in the key and in the discussion under P. guianae, P. terminalis appears to be one of the most widespread species of the genus, and is fairly common in collections. Acknowledgments We acknowledge with gratitude the assistance of the following persons and institutions for the loan or gift of specimens: Paul H. Arnaud, Jr., California Academy of Sciences, San Francisco; Franklin S. Blanton, University of Florida, Gainesville; J. Antony Downes, Agriculture Canada, Ottawa; Ernst J. Fittkau, Zoologische Staatssammlung, Miinchen, West Germany; Oliver S. Flint, Jr., Smithsonian Institution, Washington, D. C.; Richard Lane, British Museum (Nat- ural History), London; Vernon H. Lee, formerly with the Rockefeller Foundation in Cali, Colombia; Warren E. Steiner, Smithsonian Institution; and Quentin D. Wheeler, Cornell University, Ithaca, New York. We are especially grateful to Ian Roper and Alan L. Dyce of the McMaster Laboratory, CSIRO, Sydney, Australia, for the wing photographs. 908 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Coquillett, D. W. 1904. New Diptera from Central America.— Proceedings of the Entomological Society of Washington 6:90-98. Downes, J. A., and W. W. Wirth. 1981. Chapter 28. Ceratopogonidae. pp. 393-421. In J. F. McAlpine et al., eds., Manual of Nearctic Diptera. vol. 1, 674 pp.—Agriculture Canada Re- search Branch Monograph 27. Ottawa. Enderlein, G. 1912. Paryphoconus, eine neue Chironomiden-Gattung aus Brasilien.—Stettiner En- tomologische Zeitschrift 1912:57-60. Johannsen, O. A. 1943a. Two new species of American Ceratopogonidae (Diptera).— Annals of the Entomological Society of America 36:761—762. . 1943b. A generic synopsis of the Ceratopogonidae (Heleidae) of the Americas, a bibliography, and a list of the North American species.—Annals of the Entomological Society of America 36:763-791. Kieffer, J.J. 1917. Chironomides d’Amérique conservés au Musée National Hongrois de Budapest. — Annales Historico-Naturales Musei Nationalis Hungarici 15:292-364. Lane, J. 1946. New neotropical Ceratopogonidae (Heleidae) (Diptera, Nemocera).— Review of Ento- mology 17:202-215. —. 1948. Novos ceratopogonideos do Brasil (Diptera, Ceratopogonidae (Heleidae)).— Arquivos Faculdade Higiene e Satde Publica, Universidade de Sao Paulo 1:225-239, 4 plates. 1956. On “‘Paryphoconus” and ““Stenoxenus” (Diptera, Ceratopogonidae).— Revista Brasi- leira Biologia 16:299-308. 1958. On Neotropical Bezzia.— Revista Brasileira Entomologia 8:25-36. 196la. Insecta Amapaensis, Diptera: Ceratopogonidae and Anisopodidae.—Studia Ento- mologica 4:449—452. 1961b. The genera Stenoxenus and Paryphoconus in the Neotropics (Diptera: Ceratopo- gonidae).—Studia Entomologica 4:452-458. Macfie, J. W.S. 1939. A key to the species of Ceratopogonidae akin to Macropeza Mg. (Diptera).— Transactions of the Royal Entomological Society, London 89:1—12, 1 plate. 1940a. A report on a collection of Ceratopogonidae (Diptera) from British Guiana. —Ento- mologists Monthly Magazine 76:23-32. 1940b. Ceratopogonidae (Diptera) from North-eastern Brazil.— Proceedings of the Royal Entomological Society, London (B) 9:73-79. 1940c. Ceratopogonidae (Diptera) from British Guiana and Trinidad. Part 1.—Proceedings of the Royal Entomological Society, London (B) 9:179-195. Mayer, K. 1952. Die Macropeza-Gruppe der Heleiden (Diptera: Heleidae).—Beitrage der Ento- mologie 2:582—-585. . 1959. Die Puppen brasilianischer Heleiden (Diptera). — Deutsche Entomologische Zeitschrift 6:230-233. Wirth, W. W. 1952. The Heleidae of California— University of California, Publications in Ento- mology 9:95—266. . 1959. Newspecies and records of Heleidae from Brazil (Diptera). — Deutsche Entomologische Zeitschrift 6:234—237. Wirth, W. W., and N. C. Ratanaworabhan. 1972. A revision of the tribe Stenoxenini (Diptera: Ceratopogonidae).— Annals of the Entomological Society of America 65:1368-1388. (GRS) Becario Externo, Consejo Nacional de Investigaciones Cientificas y Tec- nicas de la Republica Argentina; Instituto de Limnologie, Museo de la Plata, La Plata, Argentina; (WWW) Systematic Entomology Laboratory, IIBIII, Agricultural Research Services, USDA, % National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 909-926 ACANTHOHAUSTORIUS PANSUS, A NEW SPECIES OF SAND-BURROWING AMPHIPOD FROM LOOE KEY REEF, FLORIDA KEYS, WITH REDESCRIPTION AND DISTRIBUTION DATA OF ACANTHOHAUSTORIUS BOUSFIELDI FRAME, 1980 (AMPHIPODA: HAUSTORIIDAE) J. D. Thomas and J. L. Barnard Abstract.— A new species of sand-burrowing amphipod, Acanthohaustorius pan- sus, 1s described from carbonate sands of the Florida Keys Reef Tract. This is the first time the predominantly cold-water genus Acanthohaustorius has been re- ported south of Virginia coastal waters. Acanthohaustorius bousfieldi Frame, 1980, originally described from Long Island Sound, is refigured and reported from shell-hash sediments off the central Atlantic coast of Florida. A revised key to the species of Acanthohaustorius is presented. Investigations by the authors have revealed a previously undescribed species of Acanthohaustorius from Looe Key Reef in the Florida Keys, and a range extension to Hutchinson Island, Florida, of A. bousfieldi Frame, 1980. To date, no species of Acanthohaustorius had been reported south of Virginia. This pre- dominantly cold-water genus is found mostly in quartzose shallow water sediments and its occurrence in shell-hash sediments from central Florida and coralline sands of the Florida Reef Tract is significant. The genus Acanthohaustorius was previously composed of six species: A. spi- nosus (Bousfield, 1962); A. millsi Bousfield, 1965; A. intermedius Bousfield, 1965; and A. shoemakeri Bousfield, 1965; all described from American North Atlantic waters. Acanthohaustorius bousfieldi Frame, 1980, and A. similis Frame, 1980, were described from offshore bottom sands of New York Bight and Long Island Sound. Reports of undescribed Acanthohaustorius species have been made by Camp et al. (1977) from Hutchinson Island, Florida (three species), and by Rob- ertson and Shelton (1978) from the northwestern Gulf of Mexico. We believe that a number of undescribed haustoriids exist in the southern Atlantic and Gulf of Mexico waters, and that any major taxonomic treatment of the American Haus- torlidae should be delayed until the southern Atlantic and Gulf of Mexico com- ponent species can be included. Haustoriids are exceedingly spinose and setose and in the past the location and numbers of certain spines or spine groups have been used as species characters. However, the numbers of these spines and spine groups vary not only with size of the specimen (larger = more spinose) but with geographic distribution of a species (number of spines decreases with southerly distribution). One taxonomic difficulty is the determination of those characters or set of characters that will withstand the normal range of specific variation. Our studies show that spine morphology on pereopods 5-7, spination of uropod | (especially the peduncle), and the configuration of the telson are characters that vary little and thus offer 910 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON some help in sorting out the species. We examined a single character in selected haustoriids, the number of comb-setae on mandibular palp article 3, and found a direct correlation among genera and species between the number of comb-setae and specimen length, thus rendering this particular character of no use in sepa- rating species. Explanation of Figure Legend.— Upper case letters refer to body parts as fol- lows; lower case letters to left of capital letters refer to specimens mentioned in text; lower case letters to right of capital letters and in body of drawings are in following list: A, antenna; B, body; C, coxa; E, eye; G, gnathopod; H, head; I, inner plate or ramus; J, pleopod; K, seta or spine group; L, labium; M, mandible; N, palp; P, pereopod; R, uropod; S, maxilliped; T, telson; U, upper lip; W, pleon; X, maxilla; Y, epimeron; Z, molar; d, dorsal; f, flattened; 1, left; 0, other view; r, right. Acanthohaustorius pansus, new species Figs. 1-5 Holotype.—USNM 195103, female “‘a,”” 4.82 mm, with 7 eggs. Type-locality.— Looe Key Reef, Florida Keys, 24°32.5’N; 82°24.0’W, in coarse coralline sand in front of forereef, 8 m, 22 May 82, J. D. Thomas collector. Diagnosis. — Posterior margin of coxa 3 deeply concave, posteroventral corner sharp and protuberant. Article 5 of pereopod 6 with horizontal ventral margin armed with about 4 spines evenly spaced, facial spines in 2 pairs forming square, posterior margin sparsely armed; posterior margin of article 6 with only one spine group proximal to group on apex. Coxa 7 blunt posteriorly. Peduncle of uropod 1 with 5 large dorsolateral spines, 2 ventral spines, 3 medial spines; inner ramus variable in size, short to subequally as long as outer ramus. Telson cleft to base, lobes separated by broad gape. Material.—Female “‘a” 4.82 mm; female ““b” 4.30 mm; male “‘g”’ 4.60 mm; juvenile “jy”? 2.11 mm. Description of female (ovigerous), 4.82 mm.—Body broad, robust, barrel- shaped. Head 1.42 times wide as long, rostrum short. Eyes: in life white; in freshly preserved material (1-2 hours), clear with 4—5 pinkish-orange corneal inclusions; not discernible in preserved material. Antenna |: article 1 of peduncle as deep as wide, bearing 7 plumose setae on lateral margin; article 2 narrow, equal in length to article 1; flagellum 8-articulate; accessory flagellum with 2 long, subequal articles. Antenna 2: peduncular article 4 broad, with posterior lobe, posterolateral margin with 21 elongate glassy spines, medial margin with 7 elongate spines, entire posterior margin with long plumose setae, (not figured for clarity), 4 facial plumose setae near posterodistal margin; article 5 slightly expanded distally; flagellum composed of 6 articles, article 1 longest. Upper lip broad, apex smooth. Right mandible: incisor bifid, raker row with 6 elongate and | short spines; molar well developed, triturative, molar surface indented, bearing one jointed penicillate seta; palp article 2 with 3 setae; palp article 3 with 11 apical spines, and 11 marginal comb-spines. Left mandible: lacinia mobilis short; with 7 raker spines. Maxilla 1: inner plate with medial plumose setae, outer plate marginally pubescent, bearing 11 apical spines; palp slender, armed with spines and plumose setae. Maxilla 2: inner plate slender, VOLUME 97, NUMBER 4 911 SS \j ee ss \ aN é wily : Vie SN } \ | \ \ Aff \ Fig. 2. Acanthohaustorius pansus, unattributed figures, female “‘a,” holotype, 4.82 mm; b = female “b” 4.30 mm. 914 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Yr Fig. 3. Acanthohaustorius pansus, unattributed figures, female “‘a”’ holotype, 4.82 mm; j = juvenile “9? 2.11 mm. VOLUME 97, NUMBER 4 915 anterior margin densely lined with plumose setae, anteroventral corner with sev- eral setae; article 4 with expanded posterodistal margin, anterior margin spine count = 1-1-1-1-2-2-4, ventral margin with cluster of 2 spines, posterior lobe with 4 single spines, facial setae in 3 clusters, anterior cluster = 1-2-2-2, middle clus- ter = 1-2-2, and one facial spine near posterior margin; anterior, posterior, and medioventral margins with numerous plumose setae; article 5 subquadrate, nar- rowing slightly at insertion with article 4, anterior margin spine formula = 1-3- 4-4, anteroventral corner with group of 5 spines, ventral margin with 3 single spines, posterodistal margin with 3 spines, posterior margin with 2 single spines, each with one single long seta, 4 single facial spines on lateral surface, medial surface with pair of spines inserted near ventral margin; article 6, length 4 x width, posterior margin with group of 4 spines near apex, apex with 9 spines and one immersed penicillate seta representing dactyl. Pereopod 7: coxa small, rounded posteriorly, posterior margin with 3 short setae; article 2 large, circular, anterior margin with plumose setae and 7 elongate spines, anterodistal corner with 6 spines, posterior margin with 2 widely separated setules; article 4 produced posteroven- trally, anterior margin with numerous plumose setae and 1-2-3 spines, posterior margin with moderately spaced plumose setae and 3 long spines at posterodistal margin, ventrolateral margin with 2 spines, medial surface with 4 setal clusters of 1-3-4-2, ventromedial margin with 3 spines, medial margin near posteroventral corner with 4 spines. Pleosome: epimeron 1, ventral margin slightly excavate, 8 marginal setae, pos- terior margin with 3 setae, posteroventral margin with single setule; epimeron 2 ventral margin with 8 setae, 3 setae in oblique row near posterior margin; epimeron 3 largest, with moderate tooth, setal formula = 1-1-4-5-4-3-4; Pleopods: pleopods 1 and 2 similar, peduncles with slightly expanded medioventral lobes, outer rami with 15 articles, inner rami with 11 articles; pleopod 3 outer ramus with 15 articles, inner ramus with 12 articles. Uropod 1: peduncle 3 x long as wide, outer margin with 3 dorsolateral spines, ventral margin with 2-3 spines (variable L-R), apex with 1 large interramal spine surrounded on either side by one shorter spine; outer ramus longer than inner, with 1-2-2 dorsofacial spines and 5 apical and 2 me- diomarginal spines; inner ramus 0.66 times outer, with 1 mediomarginal and 3 apical spines, and 4 mediomarginal and 3 apical setae. Uropod 2: peduncle and rami subequal, covered with numerous long setae. Uropod 3: peduncle shorter than rami; outer ramus longer than inner, 2 articulate, with 2 lateral setae on article 1, articles 1 and 2 with numerous apical setae; inner ramus slightly shorter than outer, with 3 mediomarginal setae, apical setae numerous. Telson cleft to base, consisting of two widely separated lobes joined by thin web, lateral margins of lobes with large penicillate seta and 2 long setae, posterior margin with 7-9 long setae. Gills laminar, present on pereopods 2-6; brood plates setose, plate 2 smallest, plates 3-5 larger, subequal. Male “‘c,” 4.30 mm.—Similar to female in most respects but generally having more plumose setae, and clusters of facial spines on articles 4 and 5 of pereopods 5, 6, and 7. Pereopod 5: article 2 with 7 posterior setae; article 4 with 2-2-3 anterior facial spines, 2-4 middle facial spines, and 2-1 posterior facial spines; article 5 with 7-8-6 anterior facial/marginal spines, and 4-3-4 posterior facial/ 916 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ( ges ESS x Js g Wad = zal == Fig. 4. Acanthohaustorius pansus, unattributed figures, female ‘‘a” holotype, 4.82 mm; b = female “b” 4.30 mm; g = male “g” 4.60 mm. VOLUME 97, NUMBER 4 917 Zoo Fig. 5. Acanthohaustorius pansus, unattributed figures, female “a” holotype, 4.82 mm; g = male “g? 4.60 mm. marginal spines. Pereopod 6: article 4 facial spines = 2-2-2, 3-3, and 1. Pereopod 7: article 4 with group of 3 facial setae on lateral margin. Specimens under 3 mm in length have fewer spines and spine groups, and fewer clusters of spines in each group. Mandibular palp article 3 shows reductions in number of comb setae and apical spines. Juvenile ‘j,’’ 2.11 mm.—Left mandibular palp article 3 with 1 comb seta and 4 apical setae; pereopod 6 article 5 lacking facial spines; article 6 lacking posterior spines. Juvenile “‘k,’’ 2.82 mm.—Left mandibular palp article 3 with 3 comb setae and 7 apical setae; pereopod 6 article 5 lacking facial spines; article 6 with a single posterior marginal spine. 918 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Variations.— All specimens exhibited some degree of variability. The ratio of the inner ramus to the outer in uropod 1 was especially variable. The percent length of the inner ramus to the outer was measured in the following specimens: male “‘g,” 25 percent; female ““b,” 50 percent; male “cc,” 58 percent; and the holotype, 70 percent. This wide variation sheds doubt on the validity of this particular meristic character which has been used to separate species within the genus Acanthohaustorius. Etymology. — From the Latin “‘pansus,”” meaning “spread out.” This is in ref- erence to the completely separated lobes of the telson in which this species is unique. Relationships. —Acanthohaustorius pansus appears closest to A. millsi but differs in: 1) widely separated lobes of the telson, 2) smaller inner ramus on uropod 1, 3) blunt posterior margin of coxa 7, 4) sparser facial or posterior armament of pereopod 6, either on articles 5 or 6, 5) presence of ventral spines on peduncle of uropod 1, 6) smaller number of comb setae on article 3 of the mandibular palp. Acanthohaustorius pansus differs from A. shoemakeri in 1, 3, 4 (article 6 only), and 6; from A. intermedius in 1| and 4; from A. similis in 1-4. Distribution.—From Looe Key Reef, Florida Keys, to Eastern and Northern Gulf of Mexico, 8—40 m, in fine to medium fine sediments. Material examined.—Bureau of Land Management MAFLA Stations: 2318, 29°05'00.8”N, 83°45'00.5”W, medium sand, 20 m.—2419, 29°46'59.8’N, 84°05'00.2”W, medium fine sand, 10 m.—2424, 29°13'00.7’N, 85°00'01.4’W, medium sand, 27 m.—2855, 30°08’02.1”N, 86°30’00.0” W, medium sand, 40 m.— 2856, 29°54'01.3”N, 87°24'00.2’”W, fine sand, 30 m.—2960, 25°40'N, 82°20'W, fine sand, 27 m. Acanthohaustorius bousfieldi Frame Figs. 6—9 Material. — Female “‘f,”’ 9.50 mm, with 8 eggs; male ““m,”’ 6.62 mm Hutchinson Island, Florida, 27°21.6’N, 80°28’W, 12 m, coarse, poorly sorted shell hash. Diagnosis.— Posterior margin of coxa 3 deeply concave, posteroventral corner sharp and protuberant. Article 5 of pereopod 6 with horizontal ventral margin armed with about 3 unevenly spaced spines, facial spines in rectangle of pairs; and singles, posterior margin well armed; posterior margin of article 6 with more than 2 spine groups other than proximal group. Coxa 7 blunt posteriorly. Peduncle of uropod 1 with short dorsolateral spines along middle two-thirds of peduncle, basal, and distal spines stout, no ventral or medial spines (one medial setule); inner ramus as long as outer, not variable. Telson cleft to base, lobes not separated. Description of Female, 9.50 mm.—Body large, barrel shaped. Head 0.71 x long as wide, rostrum short, eyes not visible. Antenna 1 article 1 of peduncle with mid-ventral margin slightly produced, oblique row of plumose setae on dorso- lateral face; article 2 shorter than article 1, dense growths of plumose setae on anterior and posteroventral margins; flagellum 11-articulate; accessory flagellum with 2 long, subequal articles. Antenna 2: article 3 of peduncle with 3 plumose setae and 3 spines on posterior margin; article 4 large, posterior margin expanded, anterodistal quadrant with 9 plumose setae, posterior margin with 24 blunt spines on lateral surface and 7 elongate spines on medial surface, 4 facial setae located proximally, distal face with 4 long, plumose setae, medial surface with 6 enlarged VOLUME 97, NUMBER 4 919 Fig.6. Acanthohaustorius bousfieldi unattributed figures, female “f’’ 9.50 mm; m = male ““m”’ 6.62 mm. plumose spines embedded in sockets anteroproximally, posterior margin with numerous plumose setae; article 5 with 15 plumose setae near anterior margin, posteriorly lined abundantly with plumose setae, flagellum 8-articulate, article 1 longest. Upper lip broad, apex smooth. Lower lip: inner lobes truncate, setose anteriorly, outer lobes anteromedial margins armed with short spines, pubescence well de- veloped on anterior and lateral margins. Right mandible: incisor bifid, lacinia mobilis small, subacute and serrate; raker row with 1 short and 8 elongate spines; molar triturative, with inserted penicillate seta; palp articles 1-3 with ratio 18: 65:72, article 3 with 21 comb-spines and 11 apical spines. Left mandible with 8 long raker spines. Maxilla 1: inner plate small, 11 plumose setae on medial margin; outer plate marginally setose, with 16 apical spines; palp with numerous plumose setae on lateral margin, apex with thin spines and one thick penicillate seta; coxal 920 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON baler lobe with 3 distinct lines of pubescence. Maxilla 2: inner plate slender, with plumose setae along medial margin, with row of oblique facial setae, apex with abundant long, hooked spines; outer plate enlarged aboral surface of medial margin with 15 thick, submarginal spines, medial margin bordered with numerous plu- mose setae, lateral surface with dense pubescence. Maxilliped: inner plate with 10 medial plumose setae, apex with 2 blunt spines and penicillate seta on aboral surface plus oblique row of 7 plumose setae on oral surface, lateral margin pu- bescent; outer plate much broader than inner, with curved spines and dense setae on distomedial margin; palp article 2 greatly expanded, reaching to end of article 3, medial margin densely setose; palp article 3 geniculate, arising from raised process on article 2, bearing facial row of 9 recurved setae, apex with 14—15 thick spines. Coxae 1-4 forming even curve below, no disjunctions, coxae progressively longer and broader. Gnathopod 1: simple; coxa small, slightly excavate, ventral margin with 6 plumose setae, anterior margin with 9 medial setules; article 2 elongate, anterior margin bare, posterior margin with long setae; article 5 expanded distally, posterior margin densely setose; article 6 slender, with numerous hooked setae; dactyl simple, with long nail. Gnathopod 2: chelate; coxa small, 7 plumose setae on ventral margin, a single spine near apex; article 2 elongate, posterior margin distally setose; article 5 similar to gnathopod | except posterodistal surface with 3 groups (5), (7), (16-17), of spoon-shaped pectinate spines; article 6 with dense hooked setae. Pereopod 3: coxa scythe-shaped, posterior margin with 5 plumose setae, apex with slender spine, anteroventral margin with single setule; article 2 truncate, anterior margin with 4 setules, posterior margin with long setae and enlarged plumose setae at posteroventral margin; article 4 anterior margin with 8 plumose setae, posterior margin with 18 setae and 6 thick, elongate plumose setae, posterodistal apex with single long spine; article 5 with circle of 14 spines and 2 thick plumose setae; article 6 with circle of 8 spines and 3 plumose setae, 2 setae near apex. Pereopod 4: coxa enlarged, rounded anteriorly, posterior margin with 2 setae; article 2 anterior margin with 6 setules, posterior margin with sparse setae, single enlarged plumose seta on posterodistal margin; article 4 anterior margin with 5 plumose setae, posterior margin with 10 plumose setae interspersed with short spines, posterodistal surface with oblique row of 5 plumose setae; article 5 with circle of 6 spines and 3 plumose setae; article 6 with circle of 10 spines and 2 plumose setae, one setule near apex representing dactyl. Pereopod 5: coxa bilobed, hind margin with 18 plumose setae; article 2 sub- circular, anterior margin with plumose setae, 4 short setae on anteroproximal margin, posterior margin with 15 plumose setae and one setule; article 4 expanded posterodistally, posterodistal margin truncate, anterior margin with interspersed long and short plumose setae and single spines, ventral margin with 4 anterior and 7 posterior spines, posterior margin with plumose setae only, facial spine formulas = 1-3-3-4 anterior, and 2-6-3 posterior, mediofacial setae = 3-8-10 an- terior and 2 posterior; article 5 similar in shape to 4, anterior margin lacking setae, facial spine formulas = 10-8-6 anterior and 6-3-6 posterior, 5 plumose setae embedded in posteroventral spine set; article 6 anterior spine formula = 2-4-3-3, posterior margin bare, apex with 5 spines and one penicillate seta. Pereopod 6: coxa small, bilobed, hind margin with 6 plumose setae; article 2 truncate, hind margin with 12 plumose setae and one setule, medial surface with 1-2-2-2 and 5 VOLUME 97, NUMBER 4 921 =F, >, = LD rc \\ vila LX) IN) Fig. 7. Acanthohaustorius bousfieldi, unattributed figures, female “f’ 9.50 mm; m = male “m” 6.62 mm. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 922 FPS A yyy ‘Az LESS ZZ ae) AES Lo Sa i > lige O mame Fig. 8. Acanthohaustorius bousfieldi, unattributed figures, female “VOLUME 97, NUMBER 4 Sa S ——— Fig. 9. Acanthohaustorius bousfieldi, unattributed figures, female “f” 9.50 mm. 924 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON facial setae; article 4 expanded posterodistally, anterior margin spine formula = 1-1-1-1-2-3-4-4, midventral margin with 2 spines, posterior margin with 1-1-2-4 spines, anterior and posterior margins with numerous plumose setae, 1-2-3-3 anterofacial, 1-3-3 midfacial, and 2 posterofacial spines, medial surface with about 13 long, plumose setae along anteroventral border; article 5 quadrate, anterior spine formula = 4-4-5-5-6, ventral margin with 1-1-1 spines, posterior margin with 1-1-2-3 spines plus 2 plumose setae, facial spines 2-2-1 anterior and 1-1, (1- 1-1, on right pereopod) midfacials; article 6 elongate, narrow, anterior margin bare, posterior margin with 2-3-3, apex with 9 spines and one penicillate seta. Pereopod 7: coxa small, hind margin with 3 setae; article 2 large, subcircular, upper half of anterior margin with plumose setae, lower half with basally expanded, distally acute very thin spines, posterior margin with one seta; article 4 postero- ventral lobe angular, extremely produced, anterior margin spines = 1-6, plumose setae and setae also present on anterior margin, posterior spine formula = 1-2, posterior margin bordered with plumose setae, ventral margin with 13 spines, lateral facial setae in 3 clusters 3-6-5, medial surface with 6 clusters 5-8-10-7-11-3; article 5 tapering somewhat at proximal margin, anterior spine formula 2-4-6-7- 7-6, posterior margin lacking spines, posteroventral corner with 7 spines, facial setae abundant on lateral surface, medial surface with 5 spines near midventral margin and | cluster of facial setae; article 6 anterior spine formula = 4-4-5, posterior spine formula = 1-4-4-5, apex with 7 spines and penicillate seta, 3-1 facial setae present proximally. Epimeron | smallest, excavate ventrally, with 16 setae, posterior margin with 7 setae. Epimeron 2 with slightly excavate posterior margin, ventral margin with 1-3-1-2-2-1-1-2-1 setae, posterior margin with 8 setae. Epimeron 3 rounded ven- trally, hind margin with large spinous process, ventral margin with 2-5-7-7-5-5- 4-4 setae. Pleopods 1-3, outer rami with 19-20-19 articles; inner rami with 13- 14-14 articles. Uropod 1: peduncle 3 x long as wide, outer margin with | basofacial, 6 stubby, and 3 apicolateral spines; outer ramus subequal to peduncle, with 2-2-3 outer marginal spines, 7 apical spines, and 1 medial spine; inner ramus slightly shorter than outer, outer margin bare, inner margin setal formula = 1-2 and 1 spine, apex with 5 spines and 2 long setae. Uropod 2: peduncle 2.5 long as wide, with oblique row of facial setae; inner and outer rami subequal, slightly longer than peduncle, distally setose. Uropod 3: peduncle short, 1.5 x long as wide, apically spinose; outer ramus 2-articulate, 2.8 x peduncle, each article apically setose; inner ramus subequal to outer. Telson: cleft to base, lobes not separated, outer margin with concavity bearing 4 setae plus penicillate seta and setule, posterior margin with 7 dorsal and 4 ventral setae. Gills laminar, present on peraeopods 2-6; brood plates setose, plate 2 smallest, plates 3 and 4 larger, subequal. Male ‘“‘m,”’ 6.62 mm.—Similar to female, tending to be more setose and spinose, especially articles 4 and 5 of pereopods 5-7. Ecology.—Found in coarse, poorly sorted shell hash, 12 m. Remarks.—We have compared our specimens with the type-material of A. bousfieldi in Smithsonian collections and answered our questions. Although our specimens seemed to have fewer short spines on the peduncle of uropod 1 than did the original illustrations, the allotype male of A. bousfieldi (USNM 172428) VOLUME 97, NUMBER 4 925 also has the smaller number of short spines we noted. We depict the telsonic lobes as being appressed together, whereas Frame depicted them as being separated by a gape. On the type-specimens in undissected form the telsonic lobes are relatively closely appressed, perhaps less so than in our drawing but more than in Frame’s drawing. The allotypic male also has the small number of ventral spines on article 5 of pereopod 6 seen in our material. The sharpness of coxa 7 matches closely in our respective specimens. We conclude that our southern specimens are identi- fiable with the northern A. bousfieldi but smaller and slightly less spinose and setose in all stages. Distribution. — Hutchinson Island, Florida. Key to the species of Acanthohaustorius 1. Telson U-cleft less than one-half to base, lobes truncate, posterior margins straight, outer margins with slight concavity; pereopod 6 article 5, distal margin oblique; setae on inner ramus of uropod 2 arranged in clusters. . 5 oc 00 6 8 Oe Ae Reap Sen 5 oe ea ORES ott ee eae ee we Seo eee A. Spinosus — Telson cleft to base or nearly so; lobes of telson posteriorly rounded; pereopod 6 distal margin of article 5 horizontal; setae on inner ramus of MRO OCH MMSCTLCGIGIM PINS 5. ears a sdde Alem bo na awl ol ae dea ew etna eeeas 2, 2. Telson of two widely separated and distinct lobes; pereopod 6 article 5 with 4 or fewer single facial spines, article 6 posterior margin with 1-2 SIMSTS SRO IES IOI CS aye ryan Wa airy ek hos he ielas coionialyehan pelea tae NOE A, pansus — Telson lobes not widely separated; pereopod 6 article 5 with more than 4 facial spines, posterior margin article 6 with 3—5 spine clusters ...... 3 3. Coxa 3, posteroventral lobe weak; epimeron 3 posterior margin lacking HOO LMM OICOMCA VA EY, ar: 2h. fae enh ge¥ee ees Segoe Mieieensne nc Sho tao Rye A. intermedius — Coxa 3, posteroventral lobe strong; epimeron 3 with large tooth, posterior IO APACS ON CA VAD Sot = ie sare usid ote eile ue teare Suaye od ore cas aeieieeae ne a were 4 4. Peduncle of uropod 1 with 5-9 short, stubby spines; coxa 7 posterior IMATE WDA CUES a ot, op het eats cuca Spa es SAP Re oekerhen Wines, ccna Byler othe A. bousfieldi — Peduncle of uropod | lacking short stubby spines; coxa 7 posterior margin ASUS 3S SEC A ee eee a Br nnn Ee etic ete Sa, En an et 5) 5. Ventral margin of article 5 pereopod 6 with spines in 3—4 groups; uropod Sinn eraramiulssone-NalleOutcime as ee ae ae eee eee A. shoemakeri — Ventral margin of article 5 pereopod 6 with continuous row of spines; GO) OC Matra TIMIe SUS CUA sheet) lt cect as hs cal ema UO A Se Annie te 8 ¢ 6 6. Peduncle of uropod | with 3-4 dorsolateral spines, ventral spines absent 3 oo ol bee Sips eee RaS chee Oe EN Be noes OTe net OES DIRS a oD OAR 1 ee es A. millsi — Peduncle of uropod | with 8-10 dorsolateral spines, 1—2 ventral spines TESTE Al lias ONESCMIG Le ache ee eee teem: heey OREN ERS ee coils ae ladte on aU A. similis Acknowledgments Both authors were supported by a grant from the Scholarly Studies Program of the Smithsonian Institution. The first author was also supported by grant DEB 8121128 from the National Science Foundation, and contract NA82AAA01157 from the Sanctuary Program Division of the Office of Ocean and Coastal Resource Management. We would like to thank Sara LeCroy of Applied Biology, Inc. for 926 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON supplying the Hutchinson Island material which was collected under a contract from Florida Power and Light. Gary Goeke and Richard Heard of the Gulf Coast Research Lab kindly provided the Bureau of Land Management MAFLA material. Carolyn Cox Lyons of New York City inked the plates. Literature Cited Bousfield, E. L. 1962. New haustoriid amphipods from the Canadian Atlantic region.— Bulletin of the National Museums of Canada 183:63-75. 1965. Haustoriidae of New England (Crustacea:Amphipoda).— Proceedings of the United States National Museum 117:159-240. Camp, D. K., N. H. Whiting, and R. E. Martin. 1977. Nearshore marine ecology at Hutchinson Island, Florida: 1971-1974. V. Arthropods.— Florida Marine Research Publications 25:1-63. Frame, A. B. 1980. Two new species of sand-burrowing amphipod crustaceans from Long Island Sound and the New York Bight (Amphipoda:Haustoriidae).— Estuaries 3:75-83. Robertson, P. B., and C. R. Shelton. 1978. Two new species of haustoriid amphipods (Crustacea: Amphipoda) from the northwestern Gulf of Mexico.—Contributions in Marine Science 21: 47-62. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 927-941 REVISION OF OPHIOPAEPALE LJUNGMAN, 1872 (ECHINODERMATA: OPHIUROIDEA), WITH A REDESCRIPTION OF O. GOESIANA LJUNGMAN, 1872, AND NOTES ON O. DIPLAX (NIELSEN, 1932), NEW COMBINATION Richard L. Turner Abstract.—Ophiopaepale Ljungman, 1872, is revised. Schizoderma Nielsen, 1932, is a junior synonym. Ophiopaepale now includes two species, which form a geminate pair from tropical American waters: O. goesiana Ljungman, 1872, type-species, West Indies and Gulf of Mexico, redescribed herein; O. diplax (Niel- sen, 1932), n. comb., type-species of Schizoderma, Gulf of Panama and other eastern tropical Pacific localities. Synonymy was required by the discovery of fragmented dorsal arm plates in O. goesiana and the presence of tables comprising the disc granulation of both species. Ophiopaepale is distinguished from other ophiuroid genera by the division of each ventral arm plate into proximal and distal halves. Nielsen (1932) erected Schizoderma as a new monotypic genus of ophioder- matid brittle star from the Gulf of Panama. He distinguished it from the closely related, monotypic, West Indian genus Ophiopaepale Ljungman, 1872,' by the presence in Schizoderma of fragmented dorsal arm plates and dorsally elongated genital slits. Nielsen was aware of the geologic history of the Isthmus of Panama and its role in formation of geminate species (1932:245); it is, therefore, surprising that he did not more closely examine the holotype of O. goesiana Ljungman, 1872, for comparison with his new material. Since Nielsen’s (1932) original de- scription, there has only been brief mention of Schizoderma by Clark (1940), Fell (1960), and Spencer and Wright (1966). Examination by the present author of recently collected O. goesiana from the southwest Florida shelf revealed the pres- ence of fragmented dorsal arm plates along most of the arm length. This condition has been confirmed in the holotype. Based on this observation, an amended diagnosis of Ophiopaepale is given, Schizoderma is relegated to synonymy, and O. goesiana is redescribed. Family Ophiodermatidae Ljungman, 1867 Ophiopaepale Ljungman, 1872 Ophicpaepale Ljungman, 1872:615—616 (diagnosed), 639 (key).—Lyman, 1882: 17-18 (diagnosed).— Nutting, 1895:78 (listed).—Bather et al., 1900:278 (listed ' This work is mistakenly dated 1871 by most authors who credit Ljungman with the several new genera, species, and varieties described therein. The error arises from the fact that the work was published (in 1872) in a volume which reported Proceedings of the Academy for 1871. 928 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON in systematic review as Ophiopaepale Lym.).— Meissner, 1901:906, 909, 912 (listed in systematic review), 919 (diagnosed).— Delage & Hérouard, 1903:150 (diagnosed). — Clark, 1915:346 (listed). — Matsumoto, 1915:83 (listed in system- atic review).—Clark, 1921:34 (zoogeographic list).—Nielsen, 1932:338-339 (compared to Schizoderma).—Fell, 1960:27 (key).—Spencer & Wright, 1966: 98 (listed in systematic review).— Madsen, 1983:30 (listed). Schizoderma Nielsen, 1932:243 (listed), 327 (key), 335 (diagnosed), 338-339 (compared to Ophiopaepale) (type-species, Schizoderma diplax Nielsen, by monotypy).— Fell, 1960:27 (key).—Spencer & Wright, 1966:98 (listed in sys- tematic review). Diagnosis.— Disc arcuate pentagonal; abactinal surface flat or sunken; granu- lated with tables on both surfaces. Radial shields connected interradially by a few large marginal plates. In each interradius, | pair of long genital slits extending at least to edge of disc and visible when viewed from above. Arms long, slender, with 3—4 arm spines. Dorsal arm plates fragmented; ventral arm plates divided transversely into proximal and distal halves. | tentacle scale. Type-species.—Ophiopaepale goesiana Ljungman, 1872, by monotypy. Remarks. — Ophiopaepale, as redefined herein, includes two species. In addition to features described by Nielsen (1932), both species have fragmented dorsal arm plates, and the disc granulation consists of tables. Each table consists of a base and spire (Fig. 1A, B) but is larger and morphologically more complex than that of aspidochirote holothuroids. Species-specific differences in table morphology have not yet been found in Ophiopaepale, but they differ from the non-tabular granules of Ophioderma (Fig. 1C, D). The species differ in number of marginal disc plates, length of genital slits and arrangement of associated ossicles, granu- lation of the oral frame, relative length of arms and vertebrae (Fig. 2A, B; cf. Ophioderma, Fig. 2C), fragmentation pattern of dorsal arm plates, morphology of tentacle scales, and length of arm spines. Ophiopaepale goesiana Ljungman, 1872 Figs. 1A, 2A, 3A—-E, 4A—C, 5A—E Ophiopaepale goesiana Ljungman, 1872:615-617 (diagnosed and described), 650 (zoogeographic list).—Lyman, 1875:3 (listed); 1878:228 (material); 1882:18, 313, 315 (listed), pl. 37, figs. 4-6; 1883:228 (listed), 233 (material, part).— Agassiz, 1888:111 (listed), fig. 393.—Nutting, 1895:81 (briefly described). — Verrill, 1899a:8 (material); 1899b:373 (listed).— Meissner, 1901:919 (listed). — Koehler, 1904:54 (listed), 57-59 (redescribed), figs. 1-4; 1907:287 (listed); 1914: 48 (material). — Clark, 1915:346 (material). — Clark, 1921:44—-50 (zoogeographic & station lists).— Nielsen, 1932:338—339 (compared to Schizoderma diplax).— Clark, 1941:106 (material), 139 (station list).—Clark, 1954:378 (zoogeographic list). — Fell, 1960:27 (listed).—Spencer & Wright, 1966:98 (listed). Non Ophiopaepale goesiana Lyman, 1883:233 [part; (MCZ 797), 1 dry specimen, disc diameter 1.0 mm, Blake sta 127, is a young Ophiomusium or Ophio- sphalma]. Material.—HOLOTYPE (Naturhistoriska Riksmuseet Stockholm 1447): al- coholic specimen, disc diameter (dd) 5.7 mm; Anguilla; 1870, 180 fm. “WIL CQ'Q “Ieq afBog “azIjog “ABD MOg LLU) ‘wnssaiddv putsapoiydC “q ‘ezijagq “ABD MOg sLLUeD ‘Wnasau19 DiuusapolydC ‘D ‘6SPS ZOW ‘Xvi dIp ajodando1ydo “g -9780€A WNSN ‘DuDvIsa08 ajpdandoiyd¢C “VW “SMAIA (MOI W10}10q) [eseq pue “(MOI J[PPTUZ) Opts “(MOI do}) [eoIde :sajnueis osIq “| “314 st a ea —Q = =) Z . (oy ea = =) | e) 2 930 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Vertebrae from base of arm, near disc, abactinal views. A, Ophiopaepale goesiana, Holotype; B, Ophiopaepale diplax, MCZ 5459; C, Ophioderma cinereum, Carrie Bow Cay, Belize. Scale bar, 0.5 mm. Non-type material: (MCZ 248), 1 dry spec., dd 10.4 mm; off Barbados; Blake, sta 272 and/or 297, 1878-1879, 76 fm and/or 123 fm.—(MCZ 249), 1 dry spec., dd 5.1 mm; Gulf of Mexico; Blake, sta 45, 1877-1878, 25°33'N, 84°21’W, 101 fm.—(MCZ 250), 1 dry spec., dd not measurable; off St. Vincent; Blake, sta 269, 1878-1879, 124 fm.—(MCZ 251), 1 dry spec., dd 5.8 mm; off Barbados; Hassler Expedition, 100 fm.—(MCZ 253), 2 dry spec., dd 6.5, 9.5 mm; off Grenada; Blake, sta 253, 1878-1879, 92 fm.—(MCZ 793), 2 alcoholic spec., dd 4.3, 5.4 mm; off Montserrat; Blake, sta 155, 1878-1879, 88 fm.—(MCZ 794), 3 alcoholic spec., dd 7.6-10.4 mm; off Havana; Blake, 1877-1878, 80-242 fm.—(MCZ 795), 3 alcoholic spec., dd 5.5—7.3 mm; same data as MCZ 250.—(MCZ 796), 1 alcoholic spec., dd 9.8 mm; St. Cruz, off Frederickstadt; Blake, sta 132, 1878-1879, 115 fm.—(MCZ 962), 1 dry spec., dd 7.7 mm; same data as MCZ 793.—(MCZ 6552), 1 dry spec., dd 8.1 mm; Old Bahama Channel, off Cayo Coco, Camaguey Province; Harvard-Havana Expedition, Atlantis, sta 3400, 28 Apr 1939, 180 fm.—(USNM 6491), 2 alcoholic spec., dd 5.7, 6.5 mm; off Barbados; Blake, sta 293, 1878— 1879, 82 fm.—(USNM 7176), 3 alcoholic spec., dd 6.3—8.7 mm; Albatross, sta 2163, 30 Apr 1884, 23°10'31’N, 82°20'29’W, 133 fm.—(USNM 10184), 1 al- coholic spec., dd 6.9 mm; Albatross, sta 2319, 17 Jan 1885, 23°10'37°N, 82°20'06"W, 143 fm.—(USNM 10187), 1 dry spec., dd 5.2 mm; off Havana; Albatross, sta 2329, 17 Jan 1885, 23°11'03”N, 82°18'45’W, 118 fm.—(USNM 12396), 5 dry spec., dd 6.4—9.9 mm; Albatross, sta 2327, 17 Jan 1885, 23°11'45’N, 82°17'54”W, 182 fm.—(USNM 12411), 7 alcoholic spec., dd 6.2—9.8 mm; Al- batross, sta 2159, 30 Apr 1884, 23°10'39’N, 82°20'08”W, 98 fm.—(USNM 12487), 1 dry spec., dd 6.7 mm; off Havana; Albatross, sta 2166, 1 May 1884, 23°10'36'N, 82°20'30”"W, 196 fm.—(USNM 12506), 1 dry spec., dd 8.6 mm; off Havana; Albatross, sta 2160, 30 Apr 1884, 23°10'31”N, 82°20'37”W, 167 fm.—(USNM 12513), 1 dry spec., dd 9.9 mm; off Havana; Albatross, sta 2161, 30 Apr 1884, 23°10'36"N, 82°20'28”W, 146 fm.—(USNM 15286), 5 alcoholic spec., dd 7.7— 10.1 mm; A/batross, sta 2322, 17 Jan 1885, 23°10'54’N, 82°17'45’W, 115 fm.— (USNM 15290), 3 alcoholic spec., dd 6.3-10.2 mm; Albatross, sta 2348, 20 Jan 1885, 23°10'39’N, 82°20'21”W, 211 fm.—(USNM 15306), 2 alcoholic spec., dd VOLUME 97, NUMBER 4 931 6.7, 7.4 mm; Albatross, sta 2337, 19 Jan 1885, 23°10'39’N, 82°20'21”W, 199 fm.—(USNM 15312), 6 alcoholic spec., dd 9.0-10.2 mm; Albatross, sta 2336, 19 Jan 1885, 23°10'48’N, 82°18'52”W, 157 fm.—(USNM 15350), 3 dry spec., dd 7.6-9.4 mm; Albatross, sta 2320, 17 Jan 1885, 23°10’39”N, 82°18'48”"W, 130 fm.—(USNM 15394), 1 dry spec., dd 10.2 mm; off Havana; Albatross, sta 2346, 20 Jan 1885, 23°10'39’N, 82°20'21”W, 200 fm.—(USNM E3663), 1 dry spec., dd 9.4 mm; off Puerto Rico; Johnson Smithsonian Expedition, sta 679, 1933.— (USNM E4294), 5 alcoholic spec., dd 7.8-10.4 mm (2 not measurable); off Ha- vana, Morro Castle, bearing SW by W, about 2'2 mi; Bahamas Expedition, sta 2, 24 May 1893, 110 fm.—(USNM E6302), 5 dry spec., dd 7.2—-9.8 mm; Havana; State University of Iowa Biological Expedition to the Florida Keys and the West Indies, sta 2.—(USNM E11852), 1 dry spec., dd 7.7 mm; Oregon, sta 5624, 25 Sep 1965, 10°52'’N, 66°08'’W, 56 fm.—(USNM E28509), 3 alcoholic spec., dd 9.6-10.9 mm; Little Bahama Bank; Gerda, sta 251, 5 Feb 1964, 27°25’N, 78°41'00"W to 78°37'30"W, 293-311 m.—(USNM E30823), 1 alcoholic spec., dd 7.7 mm; Gulf of Mexico; U.S. Bureau of Land Management, Southwest Florida Shelf Ecosystems Study, Year 2, Cruise II (BLM 321-II), sta 35, 26 Jul 1981, 25°44.8'N, 84°21.0’'W, 159 m, otter trawl.—(USNM E30824), 2 dry spec., dd 5.3, 6.6 mm; same data as USNM E30823, triangle dredge.—(USNM E30825), 2 dry spec., dd 10.4, 11.1 mm; Gulf of Mexico; BLM 321-II, sta 36, 2 Aug 1981, 25°16.8'N, 83°57.4’W, 127 m, otter trawl.—(USNM E30826), 1 alcoholic spec., dd 10.4 mm; same data as USNM E30825, triangle dredge.—(USNM E30827), 1 dry spec., dd 10.0 mm; Gulf of Mexico; BLM 321-III, sta 32, 6 Feb 1982, 26°16.7'N, 84°04.1'W, 137 m, otter trawl.—(USNM E30828), 1 alcoholic spec., dd 9.3 mm; same data as USNM E30827, triangle dredge. Diagnosis.—3 plates at margin of disc in each interradius; middle plate wider than long. Genital slit extending to edge of disc; not bordered by specialized papillae and scales. Oral shields cordiform, naked centrally. Arm length 8-10 times disc diameter. Dorsal arm plates moderately fragmented; distal border composed of 2—4 subequal pieces. Distal tentacle scales spiniform. Morphology.—The disc is arcuate pentagonal and firmly fused to the 5 arms. Strongly calcified margins of the disc support the thin integument of the central abactinal and interradial actinal surfaces. Dense granulation covers all abactinal and most actinal disc plates. The sunken abactinal integument consists of crowded, imbricating, polygonal scales, which are smaller near the center of the disc and increase gradually in size toward the disc margin; the largest scales are between the paired radial shields (Fig. 3C). The diameter of the disc granules changes similarly. The radial shields are large, convex, smooth, broadly rounded and thick distally, becoming thin and more scale-like proximally; the adradial margins of members of a pair are divergent and extend proximally as inward processes as illustrated by Lyman (1882); a similar but shorter process occurs at the proximal end of the abradial margin of each radial shield. Interradially, 3 large ossicles in series give a prominent margin to the disc (Fig. 3C). The middle ossicle is trap- ezoidal to rectangular, with its longer side parallel to the disc margin; it overlaps at each end another shorter ossicle which in turn overlaps the radial shield. Although the marginal ossicles and radial shields are obscured by granulation, the tumid condition of these plates gives the disc margin a lumpy appearance that reveals their location. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. General morphology and color patterns. Ophiopaepale goesiana: A, Abactinal view, Ho- lotype, disc diameter 5.7 mm; B, Same, actinal view; C, Abactinal surface, partly digested with alkali, USNM E30825, disc diameter 11.1 mm; D, Abactinal view, USNM 12411; E, Same, USNM 6491, disc diameter 6.5 mm. Ophiopaepale diplax: F, Abactinal view of an interradius digested with alkali, showing interradial plate, radial shields, and associated genital bar and scales, MCZ 5459, disc diameter 13.8 mm. VOLUME 97, NUMBER 4 933 The central actinal surface is flat and sharply demarcated from the steeply sloping, triangular, concave interradial surfaces. Most of the oral frame, including the bottom edge of the dental plate, is densely granulated, and the shapes and orientation of the major ossicles are not visible without removal of the granules. Only the central part of the oral shield is naked. There are 3—5 robust, symmetrical, lanceolate teeth borne in a vertical row, each tooth usually on a separate piece of the fragmented dental plate. In addition, the lowest piece bears a pair of smaller, asymmetrical, lanceolate teeth of equal or unequal size; if equal, they are aligned horizontally and are a pair of “infradental papillae’’; if unequal, they are aligned vertically or obliquely, the smaller lying beneath the larger, giving the impression of an unpaired “infradental papilla’’; in some cases, a third tooth is present on the lowest piece of the dental plate. Tooth papillae are absent. There are 6-8 oral papillae (S—7 on the holotype), the distalmost borne on the adoral shield and the others on the oral plate. The proximal oral papillae are asymmetrical, lanceolate, and point obliquely toward the center of the actinal surface; they are more spi- niform and longer nearer the jaw apex and more lamelliform distally. The dis- talmost 2 oral papillae are rectangular, lamelliform; the next to last is the largest oral papilla, longer than wide (almost 3:1 in the holotype), and closing the oral gap; the distalmost is somewhat smaller, longer than wide, often with a concave adradial edge and not closing the oral gap. An additional oral papilla or enlarged granule sometimes lies between the 2 rectangular papillae. The oral plates are elongate triangles, with the shortest sides of a pair in contact abradially (Fig. 4B); the other 2 sides of each plate taper to an apex near the junction of the 2 distalmost oral papillae. The adoral shields join in front of the oral shield and extend to the first ventral and lateral arm plates of adjacent arms; the proximal and distal margins, formed by contact with the oral plate and oral shield, respectively, are somewhat parallel. The oral shield is cordiform. On the interradial actinal surface, dense granulation covers small, thin, imbricating scales, through which the gonads can be seen when the granulation is removed from alcoholic specimens. The granulation extends onto the genital scales and ends at the edge of the genital slit. In each interradius there are 2 genital slits, each of which extends from the adoral shield to the edge of the disc, where the broadened, distalmost genital scale and the genital bar meet and form a specialized groove (Fig. 4C; see also Lyman 1882, pl. 37, fig. 4). The end of the groove can be seen in abactinal view, but it does not appear to curve upward onto the abactinal surface. No specialized granules, papillae, or scales occur along the genital slits. The arms are long (8—10 times the disc diameter), slender, and taper uniformly to acute tips. The arms are usually widest at segment 3 (rarely segment 4; segment 2 on the holotype). Arm segments are wider than long (5:3 in larger specimens, 6:5 in smaller ones on the fourth free segment from the disc), but the vertebrae are longer than wide (Fig. 2A; see also Lyman 1882, pl. 37, figs. 5, 6). The segments are about as high as wide proximally, but distally the arms are flattened. Each segment is swollen distally because the lateral arm plates angle outward and the dorsal arm plate bulges upward just before the distal margin (Fig. 5A, C); this arrangement gives a beaded appearance to the arm, especially in smaller specimens and on the distal part of the arm. All dorsal arm plates are fragmented, strongly convex, and trapezoidal, with the distal margin wider. The distal margin is straight 934 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 5 c VOLUME 97, NUMBER 4 935 or somewhat undulating and composed of 2—4 rectangular pieces, wider than long. The lateral margin is straight and composed of 2-3 additional rectangular pieces. Centrally there are a few pieces of variable shape. Typically, the dorsal arm plate consists of 9 pieces in 3 rows of 3 pieces each, all closely fitted as if made of fused porcelain tiles (Fig. 5C). The junctions of adjacent pieces are difficult to see without treatment with alkali and careful illumination. Each lateral arm plate bears rows and columns of many fine glassy tubercles on its surface. The first ventral arm plate is hexagonal, wider than long, and undivided. The remaining ventral arm plates are divided into two pieces: a short proximal piece lying between the lateral arm plates of a given segment; a wide distal piece lying between the paired lateral arm plates of two adjacent segments. The proximal piece is square or subrectan- gular on basal segments, with a convex distal margin and straight or slightly convex lateral margins; on distal segments, the proximal piece becomes very elongate with strongly convex lateral margins (Fig. 5E). The distal piece is somewhat octagonal; the distal margin is straight or slightly concave centrally, where it meets the proximal piece of the next segment, and enters a convex curve distolaterally, where it meets the lateral arm plates of the next segment; the distolateral margins curve sharply into the nearly parallel lateral margins, ending at the tentacle pores; the proximal margin is strongly concave at the lateral arm plates and less concave at its junction with the proximal piece. On distal segments, the distal piece becomes more semicircular or sometimes bilobed. Proximal ventral arm plates lack paired pores characteristic of some other ophiodermatid species. There are 3 arm spines, usually appressed and pointing slightly upward. Segment | usually bears 2 (rarely 1) spines on each side; segment 2 usually 2—3 spines; segment 3 almost always 3 spines. On the distal half of the arm, 4 arm spines are sometimes present. Arm spines are straight, subconical, subacute, becoming more flattened and acute on distal segments. The upper 2—3 arm spines of distal segments typically have dentate upper edges (Fig. 5B). The upper arm spine of basal segments is robust, '2 to 1 times the segment length; those on the rest of the arm are about 2 the segment length; the middle arm spine is smaller, and the lower arm spine is the smallest. Tentacle scales are single. Tentacle scales of the basal arm segments are lamelli- form, oval, lanceolate, or trapezoidal, with the point of attachment to the lateral arm plate narrower than the maximal width. Scales of distal segments are pro- gressively more spiniform and are morphologically similar to, but flatter than, the arm spines (cf., Figs. 4A and 5E). Tentacle scales are 3 to 2 the length of the lower arm spine. Granulation extends onto the base of the arm. Abactinally, the first free arm segment is heavily covered with granules, although most of the surface of the distal row of fragments is bare, as is occasionally the surface of more proximal fragments; the granulation extends onto the lateral arm plates. — Fig. 4. Ophiopaepale goesiana: A, Actinal view of granulated specimen, USNM E30827; B, Oral frame, granules removed, MCZ 795; C, Side view of arm with interradial integument removed to show orientation of genital ossicles; genital bar and an accessory scale (light stipple); genital scale (white); base of arm, radial shield, and interradial plates (heavy stipple); USNM E30825. Ophiopaepale diplax: D, Oral frame, granules removed, MCZ 6173; E, Same view as C with genital scale removed, MCZ 5459. Scale bar, 1 mm. 936 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Ophiopaepale goesiana: A, Side view of proximal arm segments, USNM E6302; B, Same, distal arm segments, USNM 15394; C, Abactinal view of proximal arm segments, USNM 6491; D, Same, distal arm segment, USNM E6302; E, Actinal view of distal arm segment, USNM E6302. Ophiopaepale diplax: F, Abactinal view of proximal arm segments, MCZ 5690; G, Actinal view of distal arm segment, MCZ 5458. Scale bar: A, C, F, 1 mm; B, D, E, G, 0.5 mm. The next 2—3 segments rarely have more than a few scattered granules on the dorsal arm plate; granules often occur at the junction of the dorsal and lateral arm plates. Granulation does not extend farther on the arm. The first (undivided) ventral arm plate has a central naked region. The next 2—3 segments bear granules only at the junction of the distal piece of the ventral arm plate and the lateral arm plates; on the lateral and proximal regions of the lateral arm plates; and sometimes near the bases of the arm spines. The rest of the arm is naked actinally. Coloration.—The color pattern of alcoholic and dried specimens is variable (Fig. 3D, E). Generally, the upper surface of the disc bears yellow-white spots on a darker, orange-yellow background. Large spots overlie the radial shields and marginal disc plates, and a dark crescent outlines the proximal edge of the unpaired interradial marginal plate. Other spots do not correspond to the locations of disc scales. The center of the disc frequently has a large irregular or subcircular spot. The spotted pattern continues onto the actinal interradial surface of the disc. The VOLUME 97, NUMBER 4 37 oral frame and underside of the arm are uniformly pale yellow or yellow-white. The upper surface of the arm is banded. Band width is 1'2—2 segments. Banding results from alternating dark and light background colors coinciding with regions of small and large light spots, respectively. The 2—4 distal pieces of each dorsal arm plate bear larger light-colored spots than other pieces. In many specimens, the pieces of the dorsal arm plate are outlined by the dark underlying intervertebral muscles, which show through the thin edges where pieces abut. Verrill (1899a) reported that Nutting’s (1895) description of white spots on a brown disc is the color in life. Type-locality.— Anguilla, Lesser Antilles, 180 fm. Distribution. — Eastern Gulf of Mexico and throughout the Greater and Lesser Antilles; 56-211 fm (or as deep as 242 fm) on coral (Koehler 1914), “‘pentacrinus grounds” (Nutting 1895), and other coarse and hard substrata (Pierce and Pat- terson 1879). The bathymetric range of 38-250 fm given by Lyman (1883), Verrill (1899a, b), Meissner (1901), and Clark (1915) apparently includes a misidentified juvenile of Ophiomusium or Ophiosphalma from Blake sta 127 (38 fm) (MCZ 797). In addition to material examined by me, Clark (1921) reported O. goesiana from Blake sta 298 at 120 fm. Lyman (1883) and others reported it from Blake sta 101 at 175-250 fm, but the material (MCZ 252) has been lost. Remarks.— Although Koehler (1904) examined and illustrated Ljungman’s ho- lotype, his fig. 3 showed entire dorsal arm plates. An earlier figure (Agassiz 1888) of non-type material showed fragmentation only for the first segment free of the disc. It is difficult to interpret from Nielsen’s (1932) discussion if he examined Ljungman’s holotype or relied on the published figures mentioned above to con- clude that dorsal arm plates of O. goesiana are entire. Ophiopaepale diplax (Nielsen, 1932), new combination Figs. 1B, 2B, 3F, 4D-E, 5F—G Schizoderma diplax Nielsen, 1932:243 (listed), 335-339 (described), fig. 39.— Clark, 1940:343 (station data and material).— Fell, 1960:27 (listed). —Spencer & Wright, 1966:98 (listed). Material.—PARATYPE (Universitetets Zoologiske Museum, Copenhagen): al- coholic specimen, dd 9.5 mm; Gulf of Panama, South of San José, Islas Perlas; Dr. Th. Mortensen’s Pacific Expedition 1914-16, 27 Jan 1916, 25 fm. Non-type material: (MCZ 5456), 5 dry spec., dd 8.1-15.9 mm; Gulf of Cali- fornia, Santa Inez Bay; Templeton Crocker Expedition, Zaca, sta 142-D-2/3/4, 11 Apr 1936, 30-50 fm.—(MCZ 5457), 3 dry spec., dd 7.8-15.6 mm; Gulf of California, Santa Inez Bay; Templeton Crocker Expedition, Zaca, sta 143-D-1, 13 Apr 1936, 20 (or 29?) fm.—(MCZ 5458), 3 dry spec., dd 7.6-11.0 mm; Gulf of California, Santa Inez Bay; Templeton Crocker Expedition, Zaca, sta 146-D- 1, 17 Apr 1936, 35 fm.—(MCZ 5459), 3 dry spec., dd 5.8-13.8 mm; Gulf of California, Arena Bank; Templeton Crocker Expedition, Zaca, sta 136-D-1/14, Apr 1936, 45 fm.—(MCZ 5460), 5 dry spec., dd 6.6—9.7 mm; Lower California, Gorda Banks; Templeton Crocker Expedition, Zaca, sta 150-D-8/9, 22 Apr 1936, 40-60 fm.—(MCZ 5690), 1 dry spec., dd 6.0 mm; Colombia, Port Utria; Uni- versity of Southern California, Hancock Pacific Expeditions, Velero, sta 856, 25 Feb 1938, 15-30 fm.—(MCZ 6172), 9 dry spec., dd 5.7-10.4 mm; Costa Rica, 938 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 14 mi SE of Judas Pt.; N.Y. Zoological Society, Department of Tropical Research, Eastern Pacific Zaca Expedition 1937-38, sta 214-D-1/2/3/4, 1 Mar 1938, 9°19'32’N, 84°29'30’W to 9°17'40’N, 84°27'30"W, 42-61 fm, 4-ft Blake dredge. — (MCZ 6173), 10 alcoholic spec., dd 6.9-10.6 mm; same data as MCZ 6172. Diagnosis.—1 plate at margin of disc in each interradius; longer than wide. Genital slits extending onto abactinal surface of disc; bordered by flattened, square papillae on large accessory scales. Oral shields triangular, naked; adoral shields naked. Arm length 4—7 times disc diameter. Dorsal arm plates greatly fragmented; distal border composed of 2 large lateral and 1 small central pieces. Distal tentacle scales lamelliform. Remarks.—The description of O. goesiana given above generally fits O. diplax. The notes given below emphasize major differences, and Nielsen’s (1932) thorough description 1s modified based on examination of new material collected outside the type-locality. In non-type material and one paratype, no distinct band of larger granules borders the disc; granules increase gradually in size from center to margin as in O. goesiana rather than abruptly as Nielsen (1932) described. Radial shields of adjacent radii extend well beneath a single interradial plate and nearly meet at the midline (Fig. 3F); the interradial plate is longer than wide (1.e., with a short side along the disc margin). On most specimens, granules cover the bottom edge of the dental plate and border the sutures of the oral frame; but the centers of the oral and adoral shields and, contrary to Nielsen (1932), oral plates are naked; the oral plates are granulated only in some larger specimens. Spination of the dental and oral plates and adoral shields is the same morphologically and meristically as in O. goesiana except that the 2 distalmost oral papillae of O. diplax are shorter (Fig. 4D). The paired oral plates are more extensively in contact; the adoral shields are wider and more tapering (i.e., proximal and distal margins divergent); oral shields are triangular, with proximal sides in some specimens concave. Genital slits appear to extend onto the abactinal surface of the disc because of the greater protrusion of the distal genital scale and of specialized scales on the genital bar at the distal end of the slit (Fig. 4E); flattened, square papillae border the distal 3 of the margin of the slit (Nielsen 1932, fig. 39c). Arm length is 4—7 times the disc diameter. The arm is widest at segment 3—4, sometimes 5; this might reflect the relatively shorter arm length and shorter proximal vertebrae (Fig. 2B) compared to O. goesiana; i.e., a greater number of segments lie beneath the disc. Dorsal arm plates are more greatly fragmented (20-25 pieces; Nielsen 1932) and are bordered distally by 2 large lateral pieces and 1 small, keystone-like central piece (Fig. SF; cf, Nielsen 1932, fig. 39c). The distal pieces of the ventral arm plates are proportionately wider than those of O. goesiana. The number of arm spines is more variable, generally 3—4 on proximal segments free of the disc (not regularly 4 as found by Nielsen 1932); distally, 4 arm spines are usually found. A fourth, uppermost arm spine, if present, is short; no arm spine exceeds 2 the length of a segment. Tentacle scales are proportionately larger than those of O. goesiana, and those on free segments of the arm do not become progressively spiniform (Fig. 5G). Coloration is similar to that of O. goesiana, but the banding extends onto the ventral arm plates in the distal half of the arm; a small central white spot occurs on the disc of many specimens, and a distinct white spot occurs abactinally on the arms every few segments. Ophiopaepale diplax ranges from ~“VOLUME 97, NUMBER 4 939 Santa Inez Bay, Gulf of California, to Port Utria, Colombia, in 25-61 fm and may occur as shallow as 15 fm. It was taken on mud, shell, and rock by the Zaca (Beebe 1938; Clark 1940). Discussion. — Ophiopaepale was included in the same family or subfamily with Ophioderma and separated from Ophiura (=Ophioglypha) and Ophiolepis by early workers (Ljungman 1872; Verrill 1899a, b). It was later removed from association with Ophioderma and placed with Ophiura and Ophiolepis (Bather et al. 1900; Meissner 1901; Delage and Hérouard 1903; Koehler 1907), but Koehler (1914) returned it to the Ophiodermatidae. Matsumoto (1915) separated Ophiopaepale from the other three genera entirely, placing it in his new family Ophioleucidae; his treatment was accepted by H. L. Clark (1915, 1941) and A. H. Clark (1921, 1954). Nielsen (1932) assigned his new genus Schizoderma to the Ophioderma- tidae but stated his willingness to assign it with Ophiopaepale to the Ophioleucidae. Most recently, Fell (1960) and Spencer and Wright (1966) treated Ophiopaepale and Schizoderma as ophiodermatids, not ophioleucids or ophiurids; and Madsen (1983) excluded Ophiopaepale from the Ophioleucinae, which he assigned to the Ophiuridae. Familial assignment of these genera was unspecified in other works cited in the synonymies given above for Ophiopaepale and O. goesiana. The differences in systematic treatment have largely been due to the authors’ interpretations of the morphology and spination of the arms, structural relation- ship of the arms to the disc, and spination of the oral frame. Ophiopaepale is retained in the Ophiodermatidae in the present work. Although the arms are long and slender, they are widest basally, at segments 3—5. Arm spines are few (3-4), but some other ophiodermatids also have relatively few arm spines (Fell 1960). The arms are inserted laterally into the disc, as in ophiodermatids; the appearance of a ventral, less firm insertion is only due to the elongated and upturned genital slits. The disc is almost fully granulated; the granules are tables (morphologically similar to those of some aspidochirote holothuroids) rather than rounded granules of some Ophioderma species, but no other comparative evidence of granule mor- phology is available on which to base a conclusion. Spination of the oral frame is identical to that of Ophioderma. Reliance on the number of infradental papillae alone is presently unwarranted for familial assignment because the nature of infradental papillae often has been misinterpreted; moreover, previous authors have not agreed on the number (indeed, the presence) of infradental papillae in ophiodermatid brittle stars. Madsen’s (1983) inclusion of the closely related Ophi- oleucidae as a subfamily of Ophiuridae complicates assignment of Ophiopaepale. Re-examination of disc granulation, oral armature, and other features in families of the Chilophiurina would be well advised. Acknowledgments I thank A. Franzén, Naturhistoriska Riksmuseet Stockholm; F. J. Madsen, Universitetets Zoologiske Museum, Copenhagen (UZM); D. L. Pawson, USNM; R. M. Woollacott and M. S. Brancato, MCZ; and K. D. Spring, Continental Shelf Associates, Inc. (CSA), for the loan of material in their care. T. A. Smoyer, Harbor Branch Foundation, Inc. (HBF), did much of the photography. K. J. Eckelbarger and J. E. Miller (HBF) gave technical assistance with scanning electron micros- copy. Ophioderma spp. illustrated herein were collected by J. E. Miller. K. R. 940 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Jensen (UZM) translated the Swedish description of Ophiopaepale goesiana. I thank J. E. Miller and J. H. Dearborn, University of Maine, for helpful advice on the manuscript. Material from BLM-321 was collected during the Southwest Florida Shelf Ecosystems Study under contract number AA851-CT1-45 to CSA from the U.S. Department of the Interior, Bureau of Land Management. During part of this study, I was a Harbor Branch Institution Postdoctoral Fellow while on sabbatical leave from Florida Institute of Technology. Contribution number 380 of HBF. Literature Cited Agassiz, A. 1888. Three cruises of the United States Coast and Geodetic Survey Steamer ““Blake”’ in the Gulf of Mexico, in the Caribbean Sea, and along the Atlantic coast of the United States, from 1877 to 1880, vol. 2.—Bulletin of the Museum of Comparative Zoology at Harvard College 15:1—220. Bather, F. A., J. W. Gregory, and E. S. Goodrich. 1900. The Echinoderma. /n E. R. Lankester, ed., A Treatise on Zoology, part 3.—Adam & Charles Black, London. 344 pp. Beebe, W. 1938. Eastern Pacific Expeditions of the New York Zoological Society, 14. Introduction, itinerary, list of stations, nets and dredges of the Eastern Pacific Zaca Expedition, 1937-1938.— Zoologica 23:287—298. Clark, A. H. 1921. Report on the ophiurans collected by the Barbados-Antigua Expedition from the University of Iowa in 1918.— University of Iowa Studies in Natural History 9(5):29-63. 1954. Echinoderms (other than holothurians) of the Gulf of Mexico.—Fishery Bulletin of the Fish and Wildlife Service 55:373-379. Clark, H. L. 1915. Catalogue of Recent ophiurans: based on the collection of the Museum of Comparative Zoology.— Memoirs of the Museum of Comparative Zoology at Harvard College 25:165-376. 1940. Eastern Pacific Expeditions of the New York Zoological Society, 21. Notes on echi- noderms from the west coast of Central America. — Zoologica 25:331-352. —. 1941. Reports on the scientific results of the Atlantis Expeditions to the West Indies, under the joint auspices of the University of Havana and Harvard University. The echinoderms (other than holothurians).— Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 15(1): 1-154. Delage, Y., and E. Hérouard. 1903. Traité de Zoologie Concréte, 3. Les Echinodermes.— Reinwald, Paris. 495 pp. Fell, H. B. 1960. Synoptic keys to the genera of Ophiuroidea.— Zoology Publications from Victona University of Wellington 26:1-44. Koehler, R. 1904. Ophiures nouvelles ou peu connues.—Mémoires de la Société Zoologique de France 17:54-119. 1907. Revision de la collection des ophiures du Muséum d’Histoire Naturelle de Paris.— Bulletin Scientifique de la France et de la Belgique 41:279-351. 1914. A contribution to the study of ophiurans of the United States National Museum. — United States National Museum Bulletin 84:1-173. Ljungman, A. V. 1872. Férteckning 6fver uti Vestindien af Dr A. Goés samt under korvetten Josefinas expedition 1 Atlantiska Oceanen samlade ophiurider. — Ofversigt af Kongliga Vetenskaps-Aka- demiens Férhandlingar, 1871 28:615-658. Lyman, T. 1875. Zoological results of the Hassler Expedition, 2. Ophiuridae and Astrophytidae, including those dredged by the late Dr. William Stimpson. — Illustrated Catalogue of the Museum of Comparative Zoology at Harvard College 8:1-34. 1878. Reports on the results of dredging, under supervision of Alexander Agassiz, in the Gulf of Mexico, by the United States Coast Survey Steamer “Blake,” Lieutenant-Commander C. D. Sigsbee, U.S.N., commanding, 2. Ophiurans and astrophytons.— Bulletin of the Museum of Comparative Zoology at Harvard College 5:217—238. 1882. Report on the Ophiuroidea dredged by H.M.S. Challenger during the years 1873-— 76.—Report on the Scientific Results of the Voyage of H.M.S. Challenger during the Years 1873-76, Zoology 5:1—386. “VOLUME 97, NUMBER 4 941 . 1883. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Caribbean Sea in 1878-1879, 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, 20. Report on the Ophiuroidea.— Bulletin of the Museum of Comparative Zo- ology at Harvard College 10:227—287. Madsen, F. J. 1983. A review of the Ophioleucinae stat. rev. (Echinodermata, Ophiuroidea) with the erection of a new genus, Ophiostriatus.—Steenstrupia 9:29-69. Matsumoto, H. 1915. A new classification of the Ophiuroidea: with descriptions of new genera and species. — Proceedings of the Academy of Natural Sciences of Philadelphia 67:43-92. Meissner, M. 1901. Echinodermen (Stachelhduter), 3. Schlangensterne, E. Systematik.—Bronn’s Klassen und Ordnungen des Thier-Reichs 2(3):902—966. Nielsen, E. 1932. Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-16, 59. Ophiurans from the Gulf of Panama, California, and the Strait of Georgia.— Videnskabelige Meddelelser fra Dansk naturhistorisk Forening 1 Kobenhavn 91:241-346. Nutting, C. C. 1895. Narrative and preliminary report of Bahama Expedition.— Bulletin from the Laboratories of Natural History of the State University of Iowa 3:1—251. Pierce, B., and C. P. Patterson. 1879. List of dredging stations occupied by the United States Coast Survey steamers “Corwin,” “Bibb,” “Hassler,” and “Blake,” from 1867 to 1879.—Bulletin of the Museum of Comparative Zoology at Harvard College 6:1-15. Spencer, W. K., and C. W. Wright. 1966. Asterozoans. In R. C. Moore, ed., Treatise on Invertebrate Paleontology, part U. Echinodermata, 3.—Geological Society of America, Boulder, and Uni- versity of Kansas Press, Lawrence, pp. 4-107. Vermll, A. E. 1899a. Report on the Ophiuroidea collected by the Bahama Expedition from the University of Iowa in 1893.—Bulletin from the Laboratories of Natural History of the State University of Iowa 5:1—86. —. 1899b. North American Ophiuroidea.— Transactions of the Connecticut Academy of Sci- ences 10:301-—386. Department of Biological Sciences, Florida Institute of Technology, 150 W. University Boulevard, Melbourne, Florida 32901, and Harbor Branch Institution, Inc., RR 1, Box 196-A, Ft. Pierce, Florida 33450. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 942-950 A NEW SPECIES OF CALLANTHIAS (PISCES: PERCIFORMES: PERCOIDEI: CALLANTHIIDAE) FROM THE SOUTHEASTERN PACIFIC OCEAN William D. Anderson, Jr. and G. David Johnson Abstract.—Callanthias parini, a new species from the Nazca Ridge, is distin- guished from all other species of Callanthias in having the following combination of characters: tubed lateral-line scales 25 to 30, unilaterally, 52 to 59, bilaterally; circum-caudal-peduncle scales 21 to 24; epipleural ribs 10 to 12; and segmented dorsal- and anal-fin rays—each 10. The new species is dimorphic in lengths of its dorsal-, anal-, and caudal-fin rays, and appears to be protogynous. Callanthias can be distinguished from the closely related Grammatonotus by the presence of two well-developed opercular spines and in having more segmented dorsal- and anal-fin rays, branched caudal-fin rays, and tubed lateral-line scales (segmented dorsal- and anal-fin rays almost always 10 or 11, branched caudal-fin rays 15 (8 + 7), and tubed lateral-line scales 21 to 42 in Callanthias). During the course of our revisionary studies of the percoid genera Callanthias and Grammatonotus, we have been fortunate to receive from N. V. Parin (via J. E. Randall) ten specimens of an undescribed species of Callanthias collected from the Nazca Ridge in the southeastern Pacific. Callanthias Lowe, 1839, has been assigned by various authors to the Serranidae. Gilbert (1905) described Grammatonotus, assigned it to the Serranidae, and stated that he believed it to be closely allied to Callanthias. Our findings confirm the close relationship of these two genera, but we agree with Gosline (1966) that they should be removed from the Serranidae because they lack the innovative spe- cialization, three spines on the opercle, that characterizes that family (Johnson, 1983). Bohlke (1960) placed Grammatonotus in the Grammidae (~Grammati- dae). Springer (1982) considered both Callanthias and Grammatonotus to be representatives of the Grammatidae, but acknowledged that “‘there is little evi- dence to unite” the five genera he included in that family. Fourmanoir (1981) elevated the Callanthiinae, considered by Ogilby (1900), Katayama (1960), and Katayama et al. (1982) as a subfamily of the Serranidae, to the familial level; Johnson (in press) concurred with this and included both Callanthias and Gram- matonotus in the Callanthiidae. Methods and Abbreviations Counts and measurements were made following Hubbs and Lagler (1958) except as noted. Lateral-line scales were counted on both sides of each specimen. Other scale counts, with exception of those around the caudal peduncle, were made on either side, depending on condition of the specimen. Mid-side lateral scales were counted along a horizontal line extending from middle of gill opening to middle of structural base of caudal fin. In making counts of rows of cheek scales, rows “VOLUME 97, NUMBER 4 943 of scales above the lateral line, and scales above and below the lateral line, small scales at orbit and at bases of dorsal and anal fins were excluded. Rows of cheek scales were difficult to count because of missing scales and the irregularity of the rows. The count of scales below lateral line was made along a posterodorsal series from the origin of the anal fin. Gillrakers on first gill arch and pseudobranchial filaments were counted on the right side. The first vertebra with a haemal spine was considered the first caudal vertebra; the urostylar vertebra, the last. On ra- diographs it is frequently difficult to determine which vertebra is the first bearing a haemal spine, but the first haemal spine in percoid fishes is almost always just posterior to the first pterygiophore of the anal fin. Measurements were made with needlepoint dial calipers to nearest 0.1 mm. Measurements from anterior end of snout were taken from premaxillary sym- physis; those involving orbit (snout length, orbit diameter, interorbital width, and postorbital length of head) were of bony orbit. Measurement of orbit was of horizontal diameter. Body depth was measured from dorsal-fin origin vertically to ventral midline of body. Pectoral- and pelvic-fin lengths were of longer (either left or right) fin. Lengths of caudal-fin lobes were taken from middle of fin base to distal tips of longest rays. The distance from the more posterior rib of last pair of pleural ribs to first haemal spine was measured on radiographs; this measure- ment was taken perpendicularly at the greatest point of separation between rib and haemal spine; damaged or grossly distorted specimens were not measured. Abbreviations used include: ANSP (The Academy of Natural Sciences of Phila- delphia), BPBM (Bernice P. Bishop Museum, Honolulu), CAS (California Acad- emy of Sciences, San Francisco), GMBL (Grice Marine Biological Laboratory, College of Charleston), IOAN (Institute of Oceanology, Moscow), SL (standard length), USNM (National Museum of Natural History, Smithsonian Institution, Washington, D.C.), and ZMMU (Zoological Museum, Moscow State University). Callanthias parini, new species Figs. 1, 2 Callanthias, sp., Parin et al., 1981:14 (brief description of material collected by R/V Ikhtiandr on the Nazca Ridge). Material examined.—Ten specimens, 148 to 179 mm SL; all from the Nazca Ridge in the southeastern Pacific. HOLOTYPE: USNM 265444, 175 mm SL; 21°25’S, 81°37'W; 325 m; Aka- demik Knipovich trawl 27; 4 Sep 1980. PARATYPES: 4 specimens, 148 to 179 mm SL; same data as for holotype, BPBM 29399, 1 specimen, 148 mm SL; CAS 54643, 1 specimen, 179 mm SL; ZMMU P-15572, 2 specimens, 157 & 174 mm SL. ANSP 152995, 1 specimen, 153 mm SL; 21°27'S, 81°39'W; 335 m; Ikhtiandr trawl 43; 26 Sep 1979. IOAN uncat., 1 specimen, 177 mm SL; 21°27'S, 81°41'W; 330 m; Ikhtiandr trawl 15; 7 Sep 1979. IOAN uncat., 1 specimen, 164 mm SL; 21°25’S, 81°38'W; 320 to 330 m; [khtiandr trawl 24; 12 Sep 1979. IOAN uncat., 2 specimens, 163 & 168 mm SL; 21°29’S, 81°41'W; 335 m; Ikhtiandr trawl 83; 13 Nov 1979. Diagnosis.—Dorsal-fin rays XI, 10. Anal-fin rays III, 10. Tubed lateral-line scales 25 to 30; sum of tubed lateral-line scales from left and right sides of each 944 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Data on morphometric characters for Callanthias parini. Standard length is in mm; other measurements in percentage of standard length. Character n Range x Holotype Standard length 10 148-179 166 175 Head, length 10 33.0-35.6 34.2 33.7 Snout, length 10 6.3-8.4 7.5 6.3 Orbit, diameter 10 11.3-13.2 12.3 11.3 Postorbital length of head 10 13.0-15.3 14.6 15.2 Upper jaw, length 10 13.3-15.4 14.2 13.7 Interorbital width 10 6.3-8.7 7.8 7.7 Body, depth 10 33.4-36.9 35.0 35.0 Predorsal length 10 29.1-33.1 31.1 30.3 Caudal peduncle, length 10 18.8-22.8 20.6 20.3 Caudal peduncle, depth 10 10.6-12.6 11.6 11.8 Pectoral fin, length 9 25.3-28.3 DY DP Dial Pelvic fin, length 8 20.9-25.3 23.1 24.4 Anal fin, length of base 10 23.0—28.8 26.4 26.8 Distance from most posterior pleural rib to first haemal spine 10 2.4-3.9 3.2 3.4 specimen 52 to 59. Circum-caudal-peduncle scales 21 to 24. Dimorphism evident in lengths of dorsal-, anal-, and caudal-fin rays; these structures appreciably shorter (in most cases) in specimens less than 160 mm SL than in larger specimens. Description.— Morphometric data appear in Tables 1 and 2. Meristic data are given below; counts for the holotype are followed by asterisks. Pectoral-fin rays 21 or 22, counts bilaterally symmetrical (21* in 3 specimens, 22 in 7). Pelvic-fin rays I, 5*. Branched caudal-fin rays 8 + 7*. Dorsal procurrent caudal rays 7 or 8 (7* in 8, 8 in 2); ventral procurrent caudal rays 7*. (Most posterior dorsal pro- current ray supported by most posterior epural and apparently also by the fifth hypural; most posterior ventral procurrent ray articulating with haemal spine of penultimate vertebra.) Branchiostegal rays 6*. Pseudobranch with 22 to 27 fila- ments (22 in 1, 23* in 3, 24 in 4, 25 in 1, 27 in 1). Gillrakers, including rudiments, on first gill arch 8 to 10 + 22 to 24—total 30 to 34 (upper limb: 8 in 1, 9 in 8, 10* in 1; lower limb: 22 in 3, 23 in 3, 24* in 4; total: 30 in 1, 31 in 2, 32 in 3, 33 in 3, 34* in 1). Tubed lateral-line scales 25 to 30 (left: 26 in 3, 27 in 1, 28 in 4, 29* in 1, 30 in 1; right: 25 in 1, 26 in 3, 27 in 3, 28 in 1, 29* in 2); sum of tubed lateral-line scales from left and right sides of each specimen 52 to 59 (52 in 2, 53 in 2, 54 in 2, 55 in 1, 56 in 1, 58* in 1, 59 in 1). (Range of differences between left and right counts of lateral-line scales for individual specimens 0 to 3.) Mid-body lateral scales 38 to 46 (38 in 1, 39 in 1, 42 in 1, 43 in 1, 44 in 2, 45 in 1, 46* in 1). Rows of scales between lateral line and mid-base of spinous dorsal fin 1 or 2*. Scales between dorsal-fin origin and lateral line 2 to 4*. Scales between anal-fin origin and lateral line 15 to 19 (17*). Rows of cheek scales 7 to 10*. Circum- caudal-peduncle scales 21 to 24 (21 in 1, 23 in 4, 24* in 2). Vertebrae 24 (10 precaudal + 14 caudal)*. Procurrent spur (Johnson 1975) absent; parhypural and 5 hypurals present; hypural 1 fused with hypural 2 and hypural 3 fused with hypural 4; epurals 2 or 3 (2 in 1, 3* in 9); uroneurals 1 pair. 945 “VOLUME 97, NUMBER 4 ¢9[eul ‘Hd é é ‘Hd ‘Hd é o[Buloy o[BUloy o[BWlay X9§$ pSP< CEE BO p9S< 90E< 6€t< €9p< Ovs< € ce Bd 6 6¢ 9 8 87C 9qO| UY-[BpNed 19MOT 8 6E< lbp e9 LOS< VEt< L8t< p8p< = STE TOe 9 L8C< 9qo] uYy-jepneds 1sdd/) LLY OLY L 9D LLv (Gas C 8 1 6p y 6 Bd 1 0v CLE uy Jeue possoidoq ICC €0¢ a 6 L1¢ LEC VIC 6 €C 6 €1 6! Ell Ae1 1JOs [Bue YyIUOT, VVC OCC OCC € CC 9ST O'ET 9S vSI CST L Yl ABI JOS [BUL YIN 9°6¢ COE B9 £6¢ v 6C CTE C 6C 0'6¢ 06! C81 CLI Awd yJOs [euv YINOF 8°6¢ y 6c 9 6 6¢ € Of 9°8C COE 8° 8C V8I< Csi TLI Awd yOs Jeue p1ry ], im £9¢ 9 LT C 8C CLC ¥9 L87é 8 SC i VLI v9l ABI YOS JBUL SIT] LUE = €€¢ OCC oa: 09C< OCC CSI Ovl Cel oulds jeue psy | 6 81 Spies v 81 = 60¢ 6 IC Vy 8i OCI OCI L6 suds [eue puosag ia 96 ee OL C6 c'6 C8 99 LL 6S ourds eur jsi14 €9L € CL CSL VEL 8 6L LSL 6 6L y99 199 y99 uy [es1op posso1doq 0-0¢ ¢ 0c 6 0¢ Lgl VCC v0C LCC Lel Sel = ABI WOS [eSIOp YUL CCC CIC VCC 9'0¢ 9 OVC L&T OVC LSI 9ST Srl ABI 1JOS [BSIOP YIUIN OVC Lc 9 VSC JET LG OSC B95 C9 B9 SLI VLI v9! ABI JJOS [BSIOP YIWIAIG = COE B95 OCE vOt Ve VIE L 6c 9 £61 TV6l C81 Av1 JJOS [BSIOP p1TY 9°6¢ OTE (ants 6 OF = 9CE BO Cc Of 9 181 C6l LLI ABI JJOS [BSIOP 1SIT{ 09C< = LLC CLE CLT Bd =) re 891 = Cc rl sulds [es1op yUsAI|q 60C< ioe LSC< 19¢ = a a 09T 991 Ori suds [esiop YUIN CCC = 6 CC 8 CC O'ET = = Srl CSI SII oulds [es1op YUAIg VLI = V 81 O81 VLI = OLI cll VI vO! oulds [esiop yyy LI aa | Oe! ae 9TI 6 v1 = TOl v6 06 ourds [esiop psy. V8 9°83 88 C8 L8 0) LL OL GEE 69 oulds [esiop puosag LS £9 LS = €9 €9 SP 6£ cs cs oulds |esiop 1s114 7y3uaT 6LI LLI CLI LI 891 pol £91 LSI €cl srl yisueT] piepuris Ep9vS “yeoun ¥VPVSIT TLSSI-d yeoun “yeoun “yeoun ZLSSI-d $667S1 66£67 JoovIeYyD SVO NVOI WNsn OWIWZ NVOI NVOI NVOI OWINZ dSNV Wadd ‘Hd Aq poleorpur oie ontposydeutsoy Ajqissod oie yey) suswiioadg “ystio}se ue Aq pojouop si odAjojoy oY], “YyISUs] PrepuL}s Jo 9de}UIdIEd UT s}USWOINSvOU JOYIO {WIUI UT SI YIZUST PIepuLISg JoyOUR “TS WW QO] UY} 10}e013 ssOy) ‘YdIOW 9UO dsSOdWIOD Tg WU QO] UY} sso] sUsUTOadg “pouTUIeXxS soInjon4}s dU} JO [[e JSOU[e IOF EJeEp dy} UT JUOPTAD sI WIsIYdIOWIP ‘Wud sp1yjUuD]JDD JO SusUIIdeds Jo xos puke Ske uy puke sUY fo s}usWIOINSes|\—"Z IQR L 946 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Callanthias parini, new species, holotype, USNM 265444, 175 mm SL. Epipleural ribs on first 10 to 12 vertebrae (10 in 1, 11 in 1, 12* in 7). Pleural ribs on vertebrae 3 through 10. Configuration of predorsal bones and anterior neural spines difficult to describe in conventional symbolization of Ahlstrom et al. (1976) because predorsal bones do not actually interdigitate with neural spines; the two predorsal bones oriented more or less obliquely with their proximal ends usually terminating anterior to distal end of anteriormost neural spine. Posteriorly, dorsal and anal fins each associated with single trisegmental pterygiophore. Mouth terminal and oblique; jaws almost equal. Maxilla reaching posteriorly to near middle of eye. Premaxilla protrusible. No supramaxilla. Posterodorsal border of maxilla not covered by infraorbital bones. Interorbital slightly convex to flattened. Anterior nostril in short tube somewhat remote from eye; posterior nostril elliptical, near eye, much larger than anterior nostril, bordered by fleshy rim which is quite well developed anteriorly. Section A, of adductor mandibulae simple, without anterodorsal extension. Posteriormost point of head extending somewhat past vertical through anterior end of dorsal-fin base. Distal margins of preopercle, interopercle, and subopercle without serrations. Opercular spines two, ventral spine better developed. Lateral series of rakers on first gill arch long and slender, medial series of rakers on first gill arch and rakers on other arches short. Gillrakers on all arches with small teeth. Premaxilla with outer series consisting of conical teeth and a few small exserted canines at anterior end of jaw; inner band of villiform to conical teeth, band expanded near symphysis; no teeth at symphysis. Dentary with series of conical teeth, teeth larger anteriorly; patch of villiform to conical teeth next to symphysis; one to few exserted canines at anterior end of jaw; no teeth at symphysis. Vomer with small conical teeth in crescent- shaped or chevron-shaped patch. Palatine with series of small conical teeth; series of teeth may be expanded anteriorly into narrow band. No teeth on tongue or pterygoids. Body compressed, moderately deep. Scales ctenoid; posterior field of a scale with single row of primary and secondary scalelets. Secondary squamation essentially absent. Anterior end of lower jaw, gular region, branchiostegal mem- branes, and usually small section on lateral part of snout without scales; scales on maxilla largely restricted to distal end; remainder of head heavily scaled. Midlateral series of modified scales on body (Fig. 2). Lateral line ascending abrupt- VOLUME 97, NUMBER 4 947 Fig. 2. Callanthias parini, new species; scanning electron photomicrographs of a modified mid- lateral scale from the caudal peduncle of CAS 54643, 179 mm SL; scale bar = 1000 um in A, 100 pm in B. 948 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ly from origin near opercle to within one or two scale rows of dorsal-fin base, continuing posteriorly to terminate at or near base of ultimate dorsal soft ray. Tubes in lateral line straight and simple, each extending for most of length of a scale. Lateral scale rows essentially parallel to horizontal part of lateral line. Membranes of dorsal and anal fins without scales. Pectoral-fin base with scales. Pelvic axillary scales present; pelvic-fin bases covered by mid-ventral triangular cluster of scales. Scales extending onto caudal fin. Dorsal and anal spines relatively slender. Dorsal fin not incised at junction between spinous and soft portions. Distal margin of anal fin rounded. Pectoral fin: distal margin rounded; longest rays in dorsal part of fin, reaching posteriorly to at least vertical from base of tenth dorsal spine to as far as vertical from base of second segmented dorsal ray; two dorsalmost and one to three ventralmost rays unbranched, others branched. Pelvic fin reaching posteriorly at least as far as between bases of first and second anal spines to as far as base of first segmented anal ray. Caudal-fin lunate with dorsal and ventral lobes produced in larger specimens. Coloration.—In alcohol, body and head straw colored, fins pallid to straw col- ored; no distinctive patterns of pigmentation remaining. We received a color transparency of one of the specimens from N. V. Parin via J. E. Randall. In that photograph, head and body mostly orange; iris mostly pale anterodorsally and mostly melanistic elsewhere; dorsal and anal fins yellow-orange; pectoral and pelvic fins orange; caudal fin dull orange. Dimorphism.—Our specimens are separable into two distinct morphs based on lengths of dorsal-, anal-, and caudal-fin rays (see Table 2). Individuals less than about 160 mm SL have short fin rays and larger specimens have long fin rays. Determinations of sex reported in Table 2 are based on examination of histological sections of the gonads by Mr. William A. Roumillat. Unfortunately, the viscera were not well preserved; consequently, sex could not be determined for each specimen unequivocally. Despite this, it seems likely that Callanthias parini is a protogynous hermaphrodite and that the two morphs represent different sexual stages. Comparisons.— The other nominal species of Callanthias are: C. allporti Giin- ther, 1876, and C. australis Ogilby, 1900, both from the Australian-New Zealand region; C. japonicus Franz, 1910, from Japanese waters; C. /egras Smith, 1947, from off southern Africa; C. platei Steindachner, 1898, from the southeastern Pacific; and C. ruber (Rafinesque, 1810), from the eastern Atlantic and Mediter- ranean Sea. (Callanthias crosnieri Fourmanoir, 1981, is a Grammatonotus.) Cal- lanthias parini may be distinguished by number of tubed lateral-line scales from C. allporti, C. australis, C. japonicus, and C. platei (25 to 30 in C. parini vs. 33 to 42 in the others); by sum of tubed lateral-line scales from left and right sides (from individual specimens) from C. ruber and the species previously listed (52 to 59 in C. parini vs. 43 to 50 in C. ruber and 68 to 83 in the others); and by numbers of circum-caudal-peduncle scales, epipleural ribs, and epural bones from C. legras (C. parini with 21 to 24 caudal-peduncle scales, 10 to 12 epipleural ribs, and 2 or 3 (almost always 3) epural bones; C. /egras with 15 to 17 caudal-peduncle scales, 13 to 16 epipleural ribs, and 2 epural bones). Also useful in separating C. parini from C. australis, C. japonicus, and C. platei are numbers of dorsal- and anal-soft rays (C. parini with 10 in each fin, the other species almost always with 11 in each fin). In addition, dimorphism in length of some fin rays (see above and Table 2) has been demonstrated only in C. parini. “VOLUME 97, NUMBER 4 . 949 Distribution. —Callanthias parini is known only from the Nazca Ridge off the coast of Chile in depths of 320 to 335 meters. Etymology.—It is a pleasure to name this species for N. V. Parin who provided the specimens of the new species. Remarks. —Callanthias and Grammatonotus appear to be sister groups. They share three characters (Johnson, in press) that appear to be unique among percoid fishes: structure of the nasal organ (devoid of lamellae), presence of modified scales with unique ornamentation along the body mid-laterally, and lateral line running along base of dorsal fin and terminating near base of ultimate dorsal soft ray or continuing posteriorly on dorso-lateral surface of caudal peduncle. Callan- thias can be distinguished easily from Grammatonotus in having more opercular spines, segmented dorsal- and anal-fin rays, branched caudal-fin rays, and tubed lateral-line scales (two well-developed opercular spines in Callanthias, one in Grammatonotus; segmented dorsal-fin rays 10 or 11—very rarely 12—in Callan- thias, usually 9—rarely 8 or 10—in Grammatonotus; segmented anal-fin rays 10 or 11—very rarely 9 or 12 in Callanthias, 9 in Grammatonotus; branched caudal- fin rays 15 (8 + 7) in Callanthias, 13 (7 + 6) in Grammatonotus; tubed lateral- line scales 21 to 42 in Callanthias, 14 to 18 in Grammatonotus). Also, in Cal- lanthias the most posterior dorsal procurrent caudal-fin ray articulates with the most posterior epural and apparently in most cases also receives support from the fifth hypural; whereas, in Grammatonotus support for this element is almost always from the fifth hypural and only occasionally from both the most posterior epural and the fifth hypural. In Callanthias the most posterior ventral procurrent caudal-fin ray always articulates with the haemal spine of the penultimate vertebra, but in Grammatonotus this fin ray is almost always supported by the haemal spine of the penultimate vertebra and by the parhypural. The distance from the most posterior pleural rib to the first haemal spine is usually appreciably greater in Callanthias than in Grammatonotus (Callanthias: N = 172, range in % SL = 1.05 — 4.79, X in % SL = 3.20; Grammatonotus: N = 28, range in % SL = 0.68 — 2.14, x in % SL = 1.63). Acknowledgments John E. Randall (BPBM) forwarded the specimens described herein and a color transparency of one of them to us from N. V. Parin of the P. P. Shirshov Institute of Oceanology, Moscow, who very generously allowed us to deposit types in museums in the United States. William A. Roumillat, South Carolina Marine Resources Research Institute, Charleston, made and examined the histological sections of gonads. James F. McKinney (GMBL) provided the radiographs and the photograph for Fig. 1, and Robert Ashcraft (Medical University of South Carolina) made the scanning electron photomicrographs for Fig. 2. Carole C. Baldwin (GMBL) typed the manuscript and assisted in a number of other ways in its preparation. Carole C. Baldwin and John E. Randall reviewed the manu- script. This is GMBL contribution number 63. Literature Cited Ahlstrom, E. H., J. L. Butler, and B. Y. Sumida. 1976. Pelagic stromateoid fishes (Pisces, Perciformes) of the eastern Pacific: kinds, distributions, and early life histories and observations on five of these from the northwest Atlantic.— Bulletin of Marine Science 26:285—402. 950 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Bohlke, J. E. 1960. Comments on serranoid fishes with disjunct lateral lines, with the description of a new one from the Bahamas.—Notulae Naturae No. 330:1-11. Fourmanoir, P. 1981. Poissons (premiére liste).—Résultats des campagnes MUSORSTOM. I— Philippines (18-28 Mars 1976):85-102. Franz, V. 1910. Die japanischen Knochenfische der Sammlungen Haberer und Doflein.—Abhand- lungen der math.-phys. Klasse der K. Bayer. Akademie der Wissenschaften, Supplement 4, no. 1:1-135. Gilbert, C.H. 1905. The deep-sea fishes of the Hawaiian Islands. Jn D. S. Jordan and B. W. Evermann, eds. The aquatic resources of the Hawaiian Islands.— Bulletin of the United States Fish Com- mission 23, part 2, section 2:575-713. Gosline, W. A. 1966. The limits of the fish family Serranidae, with notes on other lower percoids. — Proceedings of the California Academy of Sciences, 4th Series, 33(6):91-112. Giinther, A. 1876. Remarks on fishes, with descriptions of new species in the British Museum, chiefly from southern seas.— Annals and Magazine of Natural History, (4), 17:389—402. Hubbs, C. L., and K. F. Lagler. 1958. Fishes of the Great Lakes region.—Cranbrook Institute of Science, Bulletin 26, Bloomfield Hills, Michigan. Johnson, G. D. 1975. The procurrent spur: an undescribed perciform caudal character and its phylogenetic implications.—Occasional Papers of the California Academy of Sciences 121: 1-23. . 1983. Niphon spinosus: A primitive epinepheline serranid, with comments on the monophyly and intrarelationships of the Serranidae.—Copeia 1983:777-787. . [In press]. Percoidei: development and relationships. Jn H. G. Moser et al., eds., Ontogeny and Systematics of Fishes. —Special Publication No. 1, Supplement to Copeia, American Society of Ichthyologists and Herpetologists. Katayama, M. 1960. Fauna Japonica Serranidae (Pisces). — Biogeographical Society of Japan, Tokyo, 189 pp. Katayama, M., E. Yamamoto, and T. Yamakawa. 1982. A review of the serranid fish genus Gram- matonotus, with description of a new species.— Japanese Journal of Ichthyology 28:368—374. Lowe, R.T. 1839. Asupplement toa synopsis of the fishes of Madeira. — Proceedings of the Zoological Society of London, 7:76-92. Ogilby, J.D. 1900. Contribution to Australian ichthyology.— The Proceedings of the Linnean Society of New South Wales for the year 1899, 24:154-186. Parin, N. V., G. A. Golovan, N. P. Pakhorukov, Yu. I. Sazonov, and Yu. N. Shcherbachey. 1981. Fishes from the Nazca and Sala-y-Gomez underwater ridges collected in cruise of R/V “Ikh- tiandr.” In Fishes of the open ocean.—Institute of Oceanology, Academy of Sciences of the USSR, Moscow (1980), pp. 5-18. Rafinesque, C. S. 1810. Caratteri di alcuni nuovi generi e nuove specie di animali e piante della Sicilia, con varie osservazioni sopra 1 medesimi. Palermo, 105 pp. Smith, J. L. B. 1947. Brief revisions and new records of South African marine fishes.—Annals and Magazine of Natural History (11)14:335-346. Springer, V.G. 1982. Pacific plate biogeography, with special reference to shorefishes.— Smithsonian Contributions to Zoology, No. 367, iv + 182 pp. Steindachner, F. 1898. Die Fische der Sammlung Plate. Fauna Chilensis.—Zoologische Jahrbiicher. Supplement 4:281-337. (WDA) Grice Marine Biological Laboratory, College of Charleston, Charleston, South Carolina 29412; (GDJ) Department of Vertebrate Zoology, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 951-960 NOTES ON THE FRESHWATER SHRIMPS OF ISLA DEL COCO WITH THE DESCRIPTION OF MACROBRACHIUM COCOENSE, NEW SPECIES Lawrence G. Abele and Won Kim Abstract.— Five species of freshwater shrimp are reported from Isla del Coco: Macrobrachium hancocki, M. americanum, M. cocoense n. sp., M. sp., and Ar- chaeatya chacei. Habitat notes are presented for each species, and chela dimor- phism in 4A. chacei is noted. Isla del Coco is an isolated oceanic island located at 5°32'57” N and 86°59'17” W about 500 km southwest of Costa Rica. It is a small island about 23.3 km in circumference with abundant fresh water. The highest point is reported to be about 850 m, although at least one report states 518 m as the greatest elevation (see Hertlein 1963). Hertlein (1963) reviewed the biogeography of the island and presented a checklist of the fauna and a bibliography. In the list of 50 decapod crustaceans he included two freshwater species: Macrobrachium americanum Bate and M. hancocki Holthuis. A third freshwater decapod, Archaeatya chacei, was described by Villalobos (1959). In 1973 one of us (LGA) had the opportunity, provided by the Smithsonian Tropical Research Institute, to visit Isla del Coco and to make some general collections there. In addition to the three previously reported freshwater shrimp species, a single male of an undescribed species of Macrobrachium was collected from a stream emptying into Wafer Bay. A male specimen of another Macro- brachium was collected from the same stream and is described but not named. In the descriptions that follow, the abbreviation tl refers to total length including. the rostrum, cl to carapace length from the posterior dorsal margin to the posterior orbital margin, and USNM to the National Museum of Natural History, Wash- ington, D.C. Drawings were made with the aid of a Wild M-5 camera lucida by the second author. Macrobrachium cocoense, new species Figs. 1-2 Material.—Isla del Coco, Costa Rica, stream on east side of Wafer Bay; 1 6; 14 Aug 1973; coll. L. G. Abele, J. Rodgers. Description of holotype.—Rostrum convex posteriorly, very slightly upturned anteriorly, overreaching antennular peduncle but falling short of distal end of scaphocerite; armed dorsally with 11 teeth, including 2 postorbital; teeth about equally spaced from second to ninth tooth; distinct space between ninth and 2 anteriormost teeth; armed ventrally with 6 teeth, spaced more closely posteriorly; lateral ridge sloping slightly upward anteriorly, merging with orbital margin pos- teriorly. Carapace covered with minute spinules (=very short setae) most distinct in anterior and anterolateral region; lower orbital angle rounded; hepatic spine dis- 952 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Macrobrachium cocoense, male holotype, lateral view. Scale = 50 mm. Pleuron of third abdominal somite damaged. Rostrum scale = 10 mm. tinctly smaller than antennal spine; distinct hepatic and branchiocardiac grooves present; region behind orbital margin slightly depressed. Abdomen covered with very small spinules except dorsal part of first 3 somites; pleura covered with very short setae similar to those on carapace; pleura of first 3 somites broadly rounded posteroventrally; pleuron of fourth somite angular; pleuron of fifth somite somewhat triangular; sixth somite with acute ventral and subdorsal posterior margins, about 1.4 times as long as fifth. Telson covered with minute spinules, about 1.5 times as long as sixth somite, with one pair of dorsal spines at midlength and only one (right) spine at * length of telson; posterior margin distinct, ending in median acute point; apex falling far short of tips of inner of 2 pairs of posterior spines; several feathered setae present between inner spines. Cornea rounded and broader than eyestalk; ocellus distinct. Antennule with stylocerite extending beyond midlength of penultimate segment of antennule; ultimate segment slightly longer than penultimate segment in dorsal view; antepenultimate segment lengthened laterally and ventrally. =) Fig. 2. Macrobrachium cocoense, male holotype: A, Right third pereopod; B, Dorsal view of anterior region; C, Right first pereopod; D, Chela of right first pereopod; E, Left third maxilliped; F, Left second pereopod; G, Right appendix masculina; H, Left uropod; I, Telson; J, Fingers of left second pereopod. Scale: A, C, J = 10 mm, B = 20 mm, D = 3 mm, E, H, I= 5 mm, F= 50 mm, G = 2 mm. 958 - VOLUME 97, NUMBER 4 rind ts REE SS SSRIS SL as TOP TI ee x =SSTSS SS 954 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Scaphocerite overreaching rostrum, almost 3 times as long as broad; outer margin almost straight, ending in short spine falling far short of distal margin of inner blade. Third maxilliped falling slightly short of middle of scaphocerite; antepenulti- mate segment with many very small spinules laterally, about 1.4 times as long as penultimate; penultimate about 1.3 times as long as ultimate; penultimate and antepenultimate segments with many long setae; exopod extending to distal margin of antepenultimate segment. First pereopods reaching with chela and part of carpus beyond scaphocerite; fingers slightly longer than palm; carpus 2.9 times as long as chela and 1.4 times as long as merus; merus and ischium with dense spinules ventrally, increasing in size proximally. Second pereopods similar in shape, unequal in size, overreaching scaphocerite by carpus and % length of merus. Left pereopod with fingers 0.36 times as long as palm, closing over their entire length; movable finger with large basal tooth about 7, of length from tip, 5 small denticles present between base of edge and large tooth, edge entire distal to large tooth; immovable finger with large tooth somewhat proximal to large one of movable finger, armed with 5 denticles between base and large tooth; both fingers covered with thick pubescence; palm elongate and cylindrical, about 11 times as long as high, with several lon- gitudinal rows of spinules becoming larger lateroventrally; carpus longer than palm, though being shorter than entire chela, about 9.5 times as long as broad, narrowing slightly proximally; merus slightly swollen, broadest in middle, about 0.7 times as long as carpus, about 2 times as long as compressed ischium; carpus, merus, and ischium with several longitudinal rows of spinules being larger mid- ventrally. Third through fifth pereopods similar, covered with very short, dense spinules on all segments except dactylus. Third pereopod reaching with part of dactylus beyond scaphocerite; propodus 3.7 times as long as dactylus, 1.7 times as long as carpus and 0.8 times as long as merus. Fifth pereopod reaching somewhat beyond middle of scaphocerite; propodus almost 4.5 times as long as dactylus, 1.9 times as long as carpus and equal in length to merus; inferior margin of propodus with strong spines and rows of transverse, long setae. First pleopods lacking appendix interna; endopod less than half as long as exopod. Appendix masculina present on second pleopod, about 2 times as long as appendix interna, with strong, stiff setae on superior surface and tip. Uropods ovate, extending well beyond telson; lateral ramus slightly longer than mesial; diaeresis with one movable spine mesial to immovable spine at lateral angle; minute spinules on surface of uropods. Holotype.—The male holotype (tl = 166.5 mm, cl 39.6 mm) is deposited in the USNM. Habitat.—Macrobrachium cocoense was collected with a dip net in a slow- flowing, almost stagnant, freshwater stream. The depth was about 1 m, width 3- 4 m and the bottom was mud and rotting vegetation. Visibility was virtually zero. Color.—Carapace and abdomen predominately light brown mottled with black; abdomen bronze posteriorly; chelipeds black. Remarks.—Macrobrachium cocoense is morphologically similar to two other eastern Pacific species in the genus, M. tenellum (Smith) and M. rathbunae Hol- thuis. We have examined specimens of both of these species, including the type- specimen of M. rathbunae. Macrobrachium cocoense can be easily distinguished VOLUME 97, NUMBER 4 955 Fig. 3. Macrobrachium sp., lateral view. Scale = 50 mm. Abdomen disarticulated at third somite. Rostrum scale = 10 mm. from M. tenellum by the following characters: in M. cocoense the fingers of the second pereopod are about 0.36 as long as the palm, while in M. tenellum they are about 0.8 as long; in M. cocoense the rostrum is almost straight, while in M. tenellum it is distinctly curved upward. The present species can be distinguished from M. rathbunae by the following characters: in M. cocoense the carpus of the first pereopod is about 2.9 times the chela length, while in M. rathbunae it is about 2.0 times the chela length; in M. cocoense the second pereopods extend beyond the scaphocerite by the carpus and about 0.6 of the merus while in M. rathbunae they extend beyond the scaphocerite by the carpus only; in M. cocoense the fingers of the second pereopod are about 0.36 as long as the palm, while in M. rathbunae they are about 0.5-0.6 as long; in M. cocoense there are dense, strong spinules on pereopods 2-5, while in M. rathbunae they are fewer and weaker. Etymology. — From the type-locality. Macrobrachium sp. Figs. 3-4 Material. —Isla del Coco, Costa Rica, stream on east side of Wafer Bay; 14 Aug 1973; coll. L. G. Abele. Description.— Rostrum armed with 12 teeth dorsally, including one postorbital and one above orbital margin; armed ventrally with 5 teeth. Carapace smooth. Abdomen smooth; pleura covered with indistinct short setae, sparse ventrally. Telson sparsely covered with very small spinules. 956 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Lae eS SSS N N VN Fig. 4. Macrobrachium sp.: A, Right first pereopod; B, Left third maxilliped; C, Chela of left first pereopod; D, Left third pereopod; E, Telson; F, Right uropod; G, Left appendix masculina. Scale: A, D= 10 mm, C= 3 mm, B, E, F, G=5 mm. Right stylocerite damaged, with 2 lateral spines. Third maxilliped having segments with long setae ventrally becoming dense proximally. First pereopods overreaching scaphocerite by length of chela and having seg- ments almost naked, ischium with sparse long setae ventrally. Left second pereo- pod overreaching scaphocerite by length of carpus and 4 length of merus; fingers about 0.6 times as long as palm, covered with pubescence; palm elongate, cylin- drical, about 7 times as long as high, with several longitudinal rows of spinules larger and more distinct lateroventrally; carpus about 6.8 times as long as broad; merus 1.6 times as long as ischium; carpus, merus, and ischium with several longitudinal rows of spinules becoming larger lateroventrally. Right second pereo- pod regenerating. Third through fifth pereopods similar, covered with very short, sparse spinules on narrow area of ventral surface of merus, carpus and propodus. Third pereopod with propodus about 2.5 times as long as dactylus. Fifth pereopod with propodus 3.4 times as long as dactylus. Uropods with sparse spinules on dorsal surface. Remarks. The present specimen was collected in the same stream as M. cocoense and is virtually identical in size (tl 166 mm, cl 39.6 mm). We considered the possibility that this specimen is conspecific with M. cocoense, but without more material we hesitate to make a decision. We also compared it to type-material of M. rathbunae, but there are numerous differences in the form of the second pereopod and in the shape of the rostrum. The specimen is also damaged, with an aberrant right stylocerite and a regenerating right second pereopod. It seems best to describe the specimen, without naming it, until more material becomes available. VOLUME 97, NUMBER 4 957 Table 1.—Comparison of size, sex, and chela type in A. chacei from Isla del Coco Island. CL (mm) 1.1- 16- 2.1- 2.6- 3.1-— 3.6- 4.1- 4.6- Gender Chela type 1.5 2.0 2.55 3.0 3.5 4.0 4.5 5.0 Total Male Ortmannioid 1S 9e ee) 78 Atyoid Suetesane tit pa 85 Female Non-oviger- Ortmannioid 1 7a el Vi faa | 3 ous Atyoid (Geis) lis ie es eae lea 61 96 Ovigerous Ortmannioid 3. GG NY Atyoid 6 4 10 27 Total lh” 1753-62 26 We) zl) 208 individuals % Female 100 100 66 32 36 91 100 100 Macrobrachium hancocki Holthuis, 1950 Macrobrachium hancocki.—Holthuis, 1952:111, pl. 29, figs. a—e. Material examined.—Isla del Coco, Wafer Bay, west side in swamp; 8 44, 13 22 (8 ovigerous); 15 Aug 1973; coll. L. G. Abele (LGA 73-63).— Wafer Bay, east side in stream; 4 44, 13 92 (5 ovigerous); 14 Aug 1973; coll. L. G. Abele (LGA 73-61). Measurements. — Males cl 7.5-16.9 mm, females cl 7.1-13.8 mm, ovigerous females cl 7.2-11.5 mm. Type-locality.—Esparta, Rio Barranca, Costa Rica. Distribution. — This species has been reported from the Pacific drainage of Costa Rica, Panama, and Colombia as well as from Isla del Coco and the Galapagos Archipelago. Remarks.—Macrobrachium hancocki is common on Isla del Coco. A few in- dividuals were observed in a small stream emptying into Chatham Bay. Specimens were seen along the edges of the stream in slow-moving or still areas. This species was extremely common in a swamp on the eastern shore of Wafer Bay. The water was brown, almost stagnant with rotting vegetation on the bottom. Large males, but not small males or females, were a beautiful deep blue in color (see also Holthius 1952:112). Additional collecting around Wafer Bay revealed that M. hancocki is most abundant in swamp waters 8-15 cm deep that are slow moving or stagnant. Abele and Blum (1977) present data on the biology of this species from the Archipielago de las Perlas, Panama. Macrobrachium americanum Bate, 1868 Macrobrachium americanum.—Holthuis, 1952:128, pl. 31, figs. d, e. Material examined.—Isla del Coco, Wafer Bay, east side in stream; 1 2; 14 Aug 1973; coll. L. G. Abele (LGA 73-61). 958 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Measurements.— Female cl 28.0 mm. Type-locality.—Lake Amatitlan, Guatemala. Distribution.—Macrobrachium americanum is widely distributed in Pacific drainages from Lower California to northern Peru, including Isla del Coco and the Galapagos Archipelago. Remarks.—The single female was collected from the larger of two streams on the eastern side of Wafer Bay. Holthuis (1952) reported specimens from a stream entering Chatham Bay. Abele and Blum (1977) reported M. americanum from large pools in streams in the Archipielago de las Perlas, Panama. Archaeatya chacei Villalobos, 1959 Archaeatya chacei Villalobos, 1959:331, figs. 1-25. Material examined.—Isla del Coco, Chatham Bay, freshwater stream; 57 46, 97 22 (25 ovigerous); 13 Aug 1973; coll. L. G. Abele (LGA 73-59).— Wafer Bay, west side, swamp; 28 66, 20 9° (2 ovigerous); 14 Aug 1973 (LGA 73-63).— Wafer Bay, east side; 6 99; 14 Aug 1973 (LGA 73-61). Measurements.— Males cl 2.1-3.1 mm, females cl 1.2—4.5 mm, ovigerous fe- males cl 3.4—5.0 mm. Panamanian specimens: males cl 2.6—4.6 mm, females cl 2.4-5.6 mm, ovigerous females cl 4.2—5.4 mm. Type-locality.—Isla del Coco. Distribution. —Isla del Coco (Villalobos 1959); Costa Rican mainland (Smalley 1964); Archipielago de las Perlas, Panama (Abele and Blum 1977). Remarks. —Individuals of A. chacei were collected from all freshwater streams sampled in Chatham and Wafer Bay watersheds from the sea-stream interface to the greatest altitude sampled (250 m). The species was found in virtually all microhabitats: side pools of streams, moving about on rocks, in fast-flowing riffle areas and among submerged parts of riparian vegetation at or near the shore. Individuals were relatively uncommon in a small stream flowing into Chatham Bay but were abundant in a larger stream flowing southwest in the same area. In the Wafer Bay area local variation in abundance was even more pronounced. A swamp on the east side had almost zero visibility, rotting debris, and flowing water. Macrobrachium hancocki was very abundant there, and A. chacei rare. On the west there was a swamp with clear, still water, where A. chacei was quite abundant and M. hancocki rare. Chela heteromorphism has been described for some members of the Atyidae by Bouvier (1925), Edmondson (1929) and, most recently, by Chace (1983). Chela morphology basically has two forms: the ortmannioid form, in which a palm is present because the dactylus is shorter than the propodus, and the atyoid form, which lacks a palm because the dactylus and propodus are subequal in length. Generally the ortmannioid form has short setae on the chelae and the atyoid form long setae. The two forms can be quite distinct although the chelae of some individuals are difficult to classify. We examined this phenomenon in A. chacei and its relationship to size and sex (Table 1). There are 208 complete specimens, 41% male and 59% female. Among the males 92% are ortmannioid and 8% atyoid, while among the females 42% are ortmannioid and 58% are atyoid. Among the 27 ovigerous females, which include the largest specimens, 63% are ortmannioid and 47% atyoid in form. VOLUME 97, NUMBER 4 959 Table 2.—Comparison of size, sex, and chela type in A. chacei from Archipielago de las Perlas, Panama. CL (mm) 2.1— 2.6- 3.1- 3.6- 4.1- 4.6-— 5.1- 5.6- Gender Chela type Wes) 3.0 3.5 4.0 45 5.0 5.5 6.0 Total Male Ortmannioid 2 Ol 198 7a. | 306 Atyoid 1 1 _2 308 Female Non-oviger- Ortmannioid 1 11 29 20 15 13 11 OZ ous Atyoid One alts 15 of 1 ee) 111 Ovigerous Ortmannioid 2 4 Atyoid 1 ul pial! Total i 25 125 140 Of Ay 6 D) 426 individuals % Female lO 48 D7 13 19 S4 100 Noo The smallest animals appear to be females and no individuals with a cl less than 2.1 mm can be identified as males. Males account for 34-68% of the indi- viduals from cl 2.1—3.5 mm. At cl 3.6—4.0 mm males are rare, accounting for 9% of the individuals. The 16 largest individuals cl 4.1—-5.0 mm, are females. Although similar data have been used to suggest protandric hermaphroditism we hesitate to speculate without seasonal and histological data. We compared the Isla del Coco specimens to individuals collected at Archi- pielago de las Perlas, Panama, and found no consistent differences except that the Panamanian specimens are larger as noted by Abele and Blum (1977). The sex ratio data may not be valid because these collections were used to measure egg numbers and stomach contents of females, and consequently females were re- moved from the vials. However, the data are valid to examine size, sex and chela type and to compare these data to those from Isla del Coco. There are 308 complete males, and of these 306 or 99% have ortmannioid chelae. Among the 118 females 108 or 92% have ortmannioid chelae, and 8% have atyoid chelae. The differences between the chela types are not as clear cut in the Panama material as in that from Isla del Coco. The size and sex data differ little from those of Isla del Coco. Other than the absence of animals with cl below 2.0 mm the general trend is the same. Males account for the majority of individuals in the middle size classes but are rare or absent in the larger size classes. Nothing is known concerning the significance of the chela dimorphism. Acknowledgments We thank the Smithsonian Tropical Research Institute and especially its di- rector, Dr. Ira Rubinoff, for providing support and arranging the trip to Isla del Coco. Drs. Fenner A. Chace, Jr., and Brian F. Kensley provided comments on parts of the manuscript. Partial support was provided by the National Science Foundation (BSR 82-15400). 960 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Abele, L. G., and N. Blum. 1977. Ecological aspects of the freshwater decapod crustaceans of the Perlas Archipelago, Panama.— Biotropica 9(4):239-252. Bouvier, E. L. 1925. Recherches sur la morphologie, les variations, la distribution géographique des crevettes de la famille des Atyides.—Encyclopédie Entomologique, series A, 4:1—370. Chace, F. A., Jr. 1983. The Atya-like shrimps of the Indo-Pacific region (Decapoda: Atyidae).— Smithsonian Contributions to Zoology 384:1—54. Edmondson, C. H. 1929. Hawaiian Atyidae.—Bernice P. Bishop Museum Bulletin 66:3-36. Hertlein, L.G. 1963. Contributions to the biogeography of Cocos Island, including a bibliography. — Proceedings of the California Academy of Sciences, 4th series, 32(8):219-289. Holthuis, L. B. 1952. A general revision of the Palaemonidae (Crustacea Decapoda Natantia) of the Americas. II. The subfamily Palaemonidae.— Occasional Papers of the Allan Hancock Foun- dation 12:1-396. Smalley, A. E. 1959. The genus Potimirim in Central America (Crustacea, Atyidae).—Revista de Biologia Tropical 11(2):177-183. Villalobos, A. F. 1959. Un nuevo género de Atyidae (Crustacea, Decapoda), procedente de la Isla del Coco.— Anales del Instituto de Biologia de la Universidad Nacional Autonoma de Mexico 30:33 1-347. Department of Biological Science, Florida State University, Tallahassee, Florida 32306. PROC. BIOL. SOC. WASH. 97(4), 1984, pp. 961-969 STUDIES IN THE HELIANTHEAE (ASTERACEAE). XXXIV. REDELIMITATION OF THE GENUS ANGELPHYTUM Harold Robinson Abstract. — The genus Angelphytum Barroso of eastern South America, originally monotypic, is expanded to include a total of 14 species. The character of eradiate heads originally used to delimit the genus proved unreliable, occurring also in species of Zexmenia, Wedelia, and Aspilia. As redefined, Angelphytum includes species related to Dimerostemma but having fertile liguliform or disciform pe- ripheral flowers in the heads. Three new species, A. bahiense, A. hatschbachii, and A. reitzii are described, ten new combinations are made, and a key is provided. The present effort continues a series of studies attempting to resolve and in some cases describe the numerous members of the Wedelia relationship of the subtribe Ecliptinae in Brazil. The related group includes many species, often xylopodial, in the open or crystalline habitats of eastern Brazil. It remains one of the most poorly resolved elements of the family Asteraceae in the area. The papers of the present series have sometimes retained the artificial traditional generic concepts such as Wedelia and Aspilia (Robinson 1984a, b), but in cases such as the recent treatment of Dimerostemma (Robinson 1984c), an apparently logical and natural generic limit has been attained. Again, in the present study of An- gelphytum, logical and workable natural limits are established. The studies of both Angelphytum and Dimerostemma show the expected result of proving that many species of the Ecliptinae in Brazil previously assigned to widely distributed genera actually belong to local Brazilian genera. Angelphytum was originally described by Barroso (1980) to honor the Argentine botanist Angel L. Cabrera. The genus included a single species from Mato Grosso, Brazil. It was compared with Zexmenia and Dimerostemma because of the winged margins of the achenes, but was separated by the lack of rays in the flowering heads. As indicated by Barroso, the heads had differentiated peripheral flowers with triquetrous achenes, but their corollas are disciform and bisexual rather than liguiform. In comparison, Dimerostemma has rays which are totally asexual, while Zexmenia has ray flowers with functional female structures. Angelphytum was not originally directly compared with either Wedelia or Aspilia, but the latter two differ traditionally from the genera mentioned by Barroso by having a strong constriction at the top of the achene below the pappus and by the essential lack of wings on the achene. At the time of the original description of Angelphytum, there was no reason to doubt seriously the naturalness of the monotypic genus or the value of its distin- guishing character. Only Zexmenia foliosa Rusby in Jones of Bolivia, a plant of very different habit, had at that time been described in the related group of genera with the same type of peripheral flowers in the head. Still, the form of the peripheral flowers of the head in the Heliantheae, and especially in the Ecliptinae, is evidently 962 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of sporadic occurrence (Robinson 1981), and is beginning to come under suspicion as an a priori generic character. It seems rather fortuitous that in the last few years two additional previously undescribed South American members of the Ecliptinae that also lack radiate corollas have been sent for identification. As in the case of Zexmenia foliosa, these species are not congeneric with the type-species of An- gelphytum. The first of the new species, from Mato Grosso, Brazil, has precisely the same floral arrangement as the type of Angelphytum, but the achenes, including the triquetrous peripheral achenes, lack wings and are constricted apically under the pappus. Also, the corollas, even though disciform, have funnelform throats rather than cylindrical throats with slightly campanulate bases as in Angelphytum. The new Mato Grosso species has shown additional characters relating it to members of the genus Wedelia that are common in the area, and the species has been named Wedelia hatschbachii (Robinson, 1984a). The second new eradiate Ecliptine species, collected in Ecuador, differs significantly from the first in the apparent lack of triquetrous peripheral achenes of the type found in Angelphytum. In this case the eradiate condition seems to derive from a complete lack of any flowers outside of the normal disk flowers. The Ecuadorian plant also differs in having black anther appendages and strong fiber sheaths along the veins of the corolla throats, characters that are very rare in Wedelia and quite common in Aspilia. This latter set of characters seems useful in spite of evidence of some artificiality in the separation of Wedelia and Aspilia. The combination of the characters in the Ecuadorian plant has been interpreted as evidence of relationship to Aspilia, a genus in which the rays are usually present but sterile, and the species has been named Aspilia pastazensis (Robinson, 1984b). The existence of Zex- menia foliosa of Bolivia, and the discovery of the two new species belonging to Wedelia and Aspilia, all lacking rays, in addition to the original doubts about the value of the character, furnish convincing evidence that the genus Angelphytum as originally described is untenable. The discovery of the untenability of Angelphytum as originally described would be unwelcome except for the fact that the name thus becomes available for a much broader natural element that would otherwise be without a name. The additional species transferred here to Angelphytum have fertile rays and have previously been placed in Zexmenia. As in the case of Wedelia hatschbachii, the bisexual peripheral flowers with triquetrous achenes in the type-species of An- gelphytum seem to relate to the fertile-rayed condition. The species transferred here to Angelphytum are not considered to have any direct relationship to Zex- menia which is typified by the Mexican and Central American Z. serrata Llave. As indicated in the unpublished thesis of Rindos (1980) on relationships within Zexmenia, some subgroups, including the type, have apically constricted achenes and are closely related to if not congeneric with Wedelia. Other comparatively distinct elements recognized in Zexmenia by Rindos prove to belong to Lasian- thaea (Becker, 1979), Lundellianthus (Robinson, 1978), or in one case, imme- diately related to Otopappus from which it differs only by the lack of the primary technical character (Anderson et al. 1979). From all these latter elements the present concept of Angelphytum is distinct by being xylopodial herbs rather than shrubs, by having cylindrical rather than funnelform upper throats on the disk corollas, by having inflated collars on the anthers, and by having generally con- VOLUME 97, NUMBER 4 : 963 torted or coiled abaxially glanduliferous rather than gradually curving non-glan- duliferous branches on the disk styles. The species of Angelphytum do not have the strong fiber sheaths on the veins of the disk corolla throats seen in Lasianthaea DC., the intricately interconnected paleae seen in Lundellianthus H. Robinson, or a viny habit and wings of the mature achene extending without significant inter- ruption to the tips of the pappus awns as seen in Otopappus Benth. Most Brazilian species that have previously been placed in Zexmenia prove to belong to Angelphytum in a manner similar to the way Brazilian species once placed in the Andean genus Oyedaea DC. have proven to belong to Dimerostem- ma. Nevertheless, a few species in the area that have been placed in Zexmenia, such as Z. foliosa of Bolivia, Z. rudis Baker of Brazil, Z. apensis (Chod.) Hassl. of Paraguay and adjacent Brazil, and two species closely related to the latter, Z. goyazensis (Gardn.) Benth. and Hook. and Wedelia goyazensis Gardn. of Brazil, are shrubby species without xylopodia, with more funnelform throats on the disk corollas, mostly with partially to completely blackened appendages on the anthers, and with less curved branches on the disk styles. The members of the Z. apensis group show glands abaxially on the style branches and evenly distributed papillae on the inner surfaces of the corolla lobes as in Angelphytum, but they are further distinct in the more pilosulous outer surfaces of the disk corolla lobes. Evidence indicates that the above species are not particularly close to Angelphytum, and they are probably best retained under the somewhat narrowed umbrella concept of Zexmenia. One species from northern Argentina, Wedelia brachylepis Griseb., that has recently been transferred to Zexmenia seems best returned to Wedelia on the basis of habit and floret structure. The actual closest relationship of Angelphytum seems to be with Dimerostem- ma. The latter has a similar herbaceous habit with a xylopodium in most species, has similarly shaped disk corollas with cylindrical upper throats, scarcely to non- setuliferous outer lobe surfaces, and evenly papillose inner lobe surfaces, has similarly enlarged anther collars, has similarly contorted usually abaxially glan- duliferous branches of the disk styles, and has similar usually broadly winged disk achenes. In fact, the principal distinction of Angelphytum, the fertile rather than sterile peripheral or ray flowers, would not alone be evidence that the division between the genera was natural. The characteristic well differentiated outer series of involucral bracts in Dimerostemma compared to the undifferentiated or irreg- ularly differentiated outer bracts in Angelphytum, however, seems to confirm the phyletic value of the distinction based in the fertile peripheral flowers. One other group of species in the area of Brazil has some of the characters of Angelphytum and may prove related. These are the many xylopodial perennial Wedelia species of the planalto that are quite unlike the shrubby non-xylopodial typical members of that genus in the wetter parts of tropical America. Neverthe- less, these Brazilian species have the achenes with constricted coroniform apices and have fertile ray flowers as in typical Wedelia, and for that reason they continue to be placed in that genus. In addition to having xylopodia, these species resemble Angelphytum in the shape of their disk corollas and in often having distinctly thickened anther collars. Still, none of the species have style branches of the disk flowers as curved as those of Angelphytum, and the branches are non-glandulifer- ous abaxially. Also, the corolla lobes often have prominent hairs on the outer 964 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON surface and vary in both overall shape and pattern of internal papillosity. The detailed floral characters seem to support fully the naturalness of the generic distinction from Angel/phytum based on the form of the achenes. The planalto species do seem to be linked by intermediate forms with some of the typical types of Wedelia such as W. hookeriana Gardn. or W. alagoensis Baker that are found in eastern Brazil. Angelphytum G. M. Barroso, Bolet. Soc. Argent. Bot. 19 (1—2):7-11. 1980. Perennial herbs with few to many erect, scarcely to moderately branched stems from a xylopodium. Leaves alternate to opposite, linear and subsessile to ovate on distinct petioles, uni- to trinervate. Inflorescence with 1—many heads; peduncles elongate. Heads broadly campanulate with many flowers; involucre without or with an irregularly distinct outer series of bracts. Peripheral flowers of heads ca. 10-14, fertile, forming trigonous achenes, usually bearing rays, bearing bisexual disciform corollas in the type species. Disk corollas ca. 25-70, with a short glabrous basal tube weakly demarcated at the top from the slightly campanulate base of the throat; upper throat cylindrical and mostly glabrous, without obvious fiber sheaths on the veins; lobes ovate-triangular, with glands and few or no setulae externally, with low uniform papillae covering the entire inner surface; anther collars distinctly thickened and abruptly constricted at the top; anther appendages with glands on the outer surface; style branches strongly curved or contorted, with glands abaxially. Achenes not constricted above under the pappus, with broad wings or lobes on the margin not continuing uninterrupted to the tips of well developed awns; awns usually stout and tapering, triquetrous, sometimes lacking, squamellae variably present. Type-species.—Angelphytum matogrossense G. M. Barroso. Key to the species of Angelphytum 1. Leaves mostly or completely alternate, sometimes opposite near the base 2 2. Heads without rays, with peripheral flowers bearing bisexual disciform CORO WAS es eaten ye pe oO cue cake ohne eee Seema A. matogrossense 2. Heads with Gistinet TaVS: me > QMS 7? > > = a: = = & = : = S s 2 = D aiiieD 1A . Bcc acai NYINOSH o are o a is ne NIAN | ee 2 LIWS S3IUVUSITLIBRARIES,, = = = < S: .S = =) eC = = ==) Zz = z a 7 oy 2 OS SS poe Oo: Ae O 3 Bi ro) I YYY So WYN x Bi “ Q 2 g 2 add 2 Re 8 Z E 2 = S = 2 °% ic » & z Sei . 2 é eS a e cS = : SMITHSONIAN INSTITUTION, NOILNLILSNI_NVINOSHLIWS Sa1¥VusIT_LIBRARI ES SMITHSONIAN _ INSTITUTION NOILNLILSNI_NVINOSHLINS ar a SB We a ts Ff uw a os yet fg 0 = me aly cm Cais oe = 7 7, % c = a < = xy eH < a | Go 5 a S oc S ce S| oc S | “4 rs ES Fs a = rr F = ry = | Jae = 2 a 2 je S 2 so NVINOSHLINS S3IYWYEIT LIBRARIES = SNOLLALILSNIT NVINOSHLINS S31uvu a1 LIBRARIES_ SMITHSONIAN “= a aS i = = | E = s zg = = = G%. = > > n> > | > 2 = 5 py, = S aa is 2 = = = Lis - 2 = = = = - es i YP fl? | 7) = Ww: A 7) m w” = wn.” : ; < Ww = 77) = ” < is = | SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS S3IYVUGIT_ LIBRARIES, SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLIWS ee a ree a ee z aoe es Zz xX a fy Z =| z = S = > | Ms. pa a | if, 2 NN AZ? g 2 SS3 rN ee = INS 2 Mf (YO 2 c 2 = = = | , = “SS > = > = : > = > = : (Tp) st z (7p) z (7p) za Tp) 2 NVINOSHLINS, SAtYVYAIT LIBRARIES SMITHSONIAN _INSTITUTION pNOLLALILSNI_ NYINOSHLIWS _LIBRARI ES SMITHSONIAN ee wl a uu ra ul a Wl a w | — = aa = ae 7 < = < = < = z 3 ee : Ss . : fod (= (ea (= y oc Cc < : a rz .S = = SI ma = o a S = : 5 See ae S = . =I a = 4 = =!) Z SMITHSONIAN INSTITUTION NOILALILSNI _NVINOSHLIWS S2/NVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLIWS ; = = c see z ce 5 rs S oe ) SS OR ease SS \e) PSL w 2 Phe jo" | 2 a Ws ON 8 ake Bd we See awe) Ewe a Ww Le ae Sie af = = s ES = Pe ie a z a z ‘s = GY: IGM ARS F NSE yp = =. rE 2 = 2 = > = Nes = uel Ww 5 172) rani ” Si tad RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31uyvysII Zt Ze A aw “ AS a cal I = bt > = Uff /*. > = 2 = 2 = OY KG 5R E m 2 m n’ ; m “” w = Ww = wa = RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHUINS SaIluyvygil a oi we 7) = 7) 4 z= \ x 5 AY: = ~ & = Yi; S \\ 6 2 S AN \S: 6 ", 2? 6 y , *. we BN 2 E V2 : ae wv OS SS ose s >" ln OS = wn vob z 7p) 2 “” ; ILSNI_NVINOSHLINS S3IYVYSIT LIBRARIES INSTITUTION NOILA = “ = 2) S n a Ge = ow 2 ay o a _<£ oa x a A x = ao =a fea} a (a) oO _ re) fan ro) eS z =) Zz ena) Ff 4 4 RIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3IuVUSIT LIBRA! 6 — = A Came z Ric She = = ee = = -vZZ > =D) Spas =) > — wW Lp vw ay Po) = Po) _— 3 a7) 2 “if? ™ B c= on = = i w ze wn , & rap) ILSNI_NVINOSHLINS S3IYVYEIT LIBRARIES SMITHSONIAN NOILAL = 22) Zz Res ep) z wn bs = = oe NS = = =a Uf 3 z BE F S ae ye ay EW 2 = 2 ay = S z . 2 2 Sai Ww) . RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYSIT LIBRA! (7) > 7p) > 4 2) > x. ty. = oc ies o 2, <3 = < a < 2 = Vip 3 : S : =— Ache oO — Oo — ro) a za peal} 4 - uy = JLSNI_NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN NOILAL! a ee a i Zz hs Li z es] i eu] = bad = ~ = ceed > Ace! S > = > bk > i= = = a =) 2 = m y 2 m Br m 2 n = w = 7) = RIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S3liuvugi7_LIBRAF = Ae 77) z 2) , 2 * = < = < = < =4 Zs 3 z 4 = Awe. MQ = ro) AY z = ro) N \E 2 NG ‘2 g 2S 2 aE NE = z. EWN = ae: > = Pad = a i ee Oe CO Ro | e eee —_ ae A | 0. SS ee RN, w a (fo heey ha, or Ps SMITHSONIAN INSTITUTION LIBRARIES CONT 88 01205 2205 will aay Net ene { my | t 0 ji ‘ ib h t elt ie sf ‘ ' bar ytarin ; bat fs in lhe een tw) i , , Pend AON cad ‘ ‘ Bh F . é ' ' ’ ' 5 wy : i