airageetinshe Me EA er ree me tinteuls argiinsie PGMA me rietirdemeieteutre ere ee aN reutina fetinontesr eae nt : since mineha ee Farnhnes AEA BARN ATE HB ats ODS ASD thetio et dating avn eiictcasiebin tindintie Olt tet: ry eh ink Aes ein MN BIN Meine ate Pet hl hres Bia mene eit eR ott Ruane Peace Date fa nn tT a acerca wari aip neh tomtom WMA mt a toner Nam mndiy wine e EY ae ¥ Gannon nen WAdincte geen pang ne Ponneiee tay ee te Nem sence Serine Weetanabitrren tnt pate Few Colinton notes pins > apelin B22 NAN ofp elit) Cae aan on Be a cae arenn Palle haba Cisanearet oe at batenay aK Lm USA te an ema omn a miapnson tet ee pen 2 hates reteon ane vonupaan On fot 7 AILEY diene * a SMM BAAIney jth tg be rath Co MA EGE the Nes Biles cD Bitaeneney elo neh plan Mebane yas (ot 32 Penang ede sere r erate Seer ro Ava18 95 Pprerrry Pen een he tree eee a SeO Neate A RT FOLDER EARLM AM PERDANA LAA Wa tee At UT nacan ta Ae Aba oh URNA UNCER NODE AS IANS tn Den 8m ptr a Tent Ate cn euy caer tnts Ae se ener nD I META teat SE OIS-RAMY eis aan Raa Ne SN NN “eaten moana ns Dated Po ao Spa PURSUE NIT Mn NICER A mend ayn 8 Seon U8 ony MIDALRARORD ay rim DRPULY SVAN NEE NEE Se IE rere prerrrererr ne. it SC lo areca cia atk ANNES fl ans SOA agit oe a erinatt Nh : Spat ae eat tow hogan ye Teen Cnyaitmatnn a Dasimitnanceabierinnioat anim ahah Kart Pua re a teecsnektn cx nt Ace ape onne . Sos . yn a dT OUR Atom an am AN DeNne Fe TA a necting themed NBD ya eto Oo Tne ne cea oats patna ACN Ee aN pura tens art tine ln e aon ieast ip nia ane Tew Wain = enki an ILO, ieknan wy ok 4 atte oe BAe Ceyntnt ence ame me eran Le Sa AP oh mt en Pan Bh mh Ae Ata Beet sara een ahaa umn gets ohh! iy eae nee henna od IYVYGIT LIBRARIES SMITHSONIAN IE NO L INSTITUTION NOILALILSNI ~ WASTE, NO C) Lut z NVINOSHLIWS # 5 — S = . ww. oe rs) eS ) os Ve C 5 iE sa = D SG OE =, ‘ & RY = > Ee > i= mal NS - : is = i 2 SQ = # z i z Ee me NOILNLILSN! SSIYVYEIT LIBRARIES SMITHSONIAN w rt w Ps Roe WY , zs = fs = at Ac = q = ae a4 (Z eA ARS =A zZ j g B BV 8 g FE a = \ Z F > = > = ae, S zZ Wy rad “Y) a v SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS = f 2 y 2 ge ¢ : = “S : ZN : - < = < = WK : = a = se 1 YS. O a je) ea fo) Lo a ea st Toe 4 Fa Sg AS STITUTION - NOILALILSNI NVINOSHLINS a IYVYGIT LIBRARI ES_SMITiheC : ee S) a o) — ro) iy i E 2 = a ; Gy ° a = oe) ay A $ = ‘a = > & f/f? Ww m op) mM uw * , n z w = w z u aed OLE RAL ES SMB Soran: INSTITUTION NOILALILSNI _ NVINOSHLINS | p sy e (o2) hen, > a < ° = I 4 = = ,; Zz Sj L fi dif. © NS os 5 WS = iy, S , lp le 2 eS 3 2 SY 2 I fi 2 c y, fe Ve = ~ “Ss z = aN g bi hf // = = 2 = “NY 5 s = =a = 3 Y Zz Y). ae = i¢p) a STITUTION (SNS Sakae luVvud IT LIBRAR! ES SMitrSONiang uw & a & ul ac = oc a Pe : < = = < Ge s S a 5 a : S fe) a o = 3 Pad ra | 2 = 2 11YVYUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS 7 3 a = J Z @ es a = w : 2B = ~ = os is = = = x ae a = a ( o = o z o ge STITUTION NOILALILSNI S3tIYVeYaI? LIBRARIES SMITHSONIAN o n ra n 2 Lae ” ; ae < = S = YS: er z = z a 2 Sas = 1 e ra a Orr < ‘ : g g 3 : g : S = a i Z : > y = > = > : as ae 7) az 7) 2 ( 31uyvugia LIBRARIES SMITHSONIAN NVINOSHLIWS = a > a > ox. g : 3g : [SS | rs x} - < a WAX a a . i = foal = fan) = Sa ye a ee OTS Ne Nome {G11 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI a = Le Zz a o — “0 aers oo 0 S 2 a 20 > rae > rat > z i: 2 = 2 z Oo = TION NOILNLILSNI SAIYVYUREIT LIBRARIES wn” = n = ay: w = < = os Ne = ay 5 = Bly 5 = g = SUYE = S Z = = “yy = 2 > = > a = . > a 7p) mo 7) » 2 SMITHSONIAN INSTITUTION NOILOLILSNI 2 nf = Hf nO o ae = z rs : = i cc A cc rs) S Z 3 ay Fd = Zz -} TION NOILALILSNI NVINOSHLINS SSluYvudii LIBRARIES eS a: = 3 = w s aad wo 5 2 We = x0 - » \ WS po > E 7 SA Wr ” ts m SSS Boks m = a = = n = w 417 ot BRARI ES SMITHSONIAN INSTITUTION NOILNLILSNI ’ & uw y = s z = . ty = =a z \, 4 fi? SN3 — NS: “EB NO 2 E Wy" 2 = : SS = = SS = NOILNLILSNI NVINOSHLINS S3I1YVYEIT LIBRARIES p 2 z w Ww 6. ai = < : a fe a -j 0 Oo fe) na = = 4817 LIBRARIES SMITHSONIAN S3IUVYGIT LIBRARIES Saluyvudi INSTITUTION TION NOILALILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES w : = s et a8 SB \. & me. y = Yi 3RARIES SMITHSONIAN ILALILSNI ILALILSNI NVINOSHLIWS INSTITUTION NOILNLILSNI INSTITUTION INSTITUTION SMITHSONIAN BRARIES Qs NS . NOILNLILSNI [LALILSNI INSTITUTION NOILNLILSNI NVINOSHLINS Saluvaudit LIBRARIES SMITHSONIAN Ol. = a oO ~ ae | = ke Y . & SMITHSONIAN a < a oO Ww a = = wy NVINOSHLIWS Peau g YY) uj oc < oc fan) eee SMITHSONIAN _ Mie _ A i: > a 2 v rm (ep) NVINOSHIIWS = = Es uw oO = z. A z (ep) es. ~ =f ae = oO Pal iS) kK SD | - = Ww . 2 SMITHSONIAN < Fh [@) (ep) 26 = = Ww NVINOSHLING INS Eg QO & Le <7 Neo Sale INSTI >a Fe INSTI Soda] INSTI Sale INSTI Se fe: 7 - - an i Ral a) ; 7 7 . : = au = 2 e | 7 : | : - a 7 : ; + . | = - | 7 . oha® 7 : : 7 _ ~ : _ | | iy a J ‘ ~ - a 7 a ra as : : | | " a : | | + 7 = - . : | . — a 7 . ; sce el aT aia a + (} wakes pp. 369-382 12 October 1976 arp rei { th ea am ‘hey one es, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SHALLOW-WATER SEA CUCUMBERS (ECHINODERMATA: HOLOTHUROIDEA ) FROM CARRIE BOW CAY, BELIZE? By Davm L. Pawson Department of Invertebrate Zoology, Smithsonian Institution Washington, D.C. 20560 In a recent article on the echinoderms of British Honduras (Belize), Devaney (1974) described a fauna of 34 species, of which only one, the synaptid Euapta lappa (Muller), was a holothurian. During the Smithsonian Institution’s In- vestigations of Marine Shallow Water Ecosystems (I.M.S.W.E.) Project, based largely at Carrie Bow Cay, Belize (16°48’N, 88°05’W ), participating investigators have made general collections of invertebrates, and these include 12 species of holothurians, of which 3 species of the genus Leptosynapta are new. In this paper the new species are described, other species are briefly characterized and some illustrated, and a key to all species is provided. A briet de- scription of Carrie Bow Cay habitats can be found in Dahl (1973). All specimens were collected from Carrie Bow Cay, unless otherwise indicated in the species accounts. They were sampled by wading and skin diving in less than 2 m water depth. Kier (1975) described the echinoids of Carrie Bow Cay, and added a further 13 species to the 7 listed by Devaney. Thus, the total number of echinoderms now reported from Belize is 58. As is to be expected, the fauna is typically Caribbean in character. For the holothurians some interest- ing new range extensions are reported. No dendrochirotid "Contribution No. 30 of the Investigations of Marine Shallow Water Ecosystems (I.M.S.W.E.) Project, Smithsonian Institution. 31—Proc. Brox. Soc. Wasu., Vou. 89, 1976 (369 ) 370 Proceedings of the Biological Society of Washington (“cucumaria’-type) species have been collected at Belize. In several other Caribbean areas the dendrochirotids may comprise up to 30% of the holothurian fauna. This hiatus is inexplicable, but may be related in some way to the reduced vagility of the dendrochirotids (which lack a pelagic larval stage), or to the fact that more collecting needs to be done in areas with hard substrates. I wish to thank the coordinator of the I.M.S.W.E. Program, Dr. k. Ruetzler, for making the material available for me, and Mr. k. Sandved for photographs. I am grateful to the several individuals named in the text who collected the holo- thurians described here. The I.M.S.W.E. Program has re- ceived support from the Smithsonian Institution and the Exxon Corporation. Key To HoLorHurRIANS KNOWN FROM BELIZE 1 (12) Body wall usually thick. Tentacles 20. Tube feet present. 2 (3) Anus surrounded by 5 calcified “teeth” eee ee Actinopyga agassizii (Seleaica) 3 (2) No such teeth present. { (5) Burrowing form, body fusiform, with inconspicuous feet. Light brown with dark brown blotches dorsally — eae ; _ Holothuria arenicola Semper 5 (4) Aut iiecoeing: fob feet more or less conspicuous. 6 (11) Tube feet placed upon conspicuous warts, at least dorsally. 8) Skin thin, rough to touch. Grey mottled with brown — ees Ee ee Holothuria impatiens (Forskaal ) 8 (7) Skin thick, more or less smooth. Light to dark brown, mottled. Ossicles in body wall tables and C-shaped bodies. 9 (10) C-shaped bodies (when present) approximately as long as tables are high Isostichopus badionotus (Selenka ) 10 (9) C-shaped bodies approximately twice as long as tables are high —. —-sTsostichopus) macroparentheses (Clark ) 11 (6) Warts absent. Skin very thick, smooth, rigid. Dark brown dorsally, yellowish laterally and reddish to pink ven- Gy: lh rns —...... Holothuria mexicana Ludwig 12 (1) Body wall thin, ody: worm- jie tentacles less than 20. Tube feet absent. 13 (22) Skin sticky to touch, due to presence of projecting ossicles in form of anchors supported by anchor plates. No wheels in body wall. 14 (17) Arms of anchors smooth; vertex with minute knobs. Anchor Sea cucumbers from Belize 371 plates with well developed bridge for support of anchors. Active, conspicuous forms; generally non-burrowing. 15 (16) Grey to brown, often longitudinally striped. Body wall often with characteristic hemispherical “bumps.” Stock of anchor branched — Euapta lappa (Miller) 16 (15) Green and white mottled or brown and white mottled. No “bumps” on body wall. Stock of anchor unbranched __ Sane oo. Synaptula hydriformis (Lesueur ) 17 (14) Arms of anchors with teeth; no knobs on vertex. Anchor plates lack true bridge. Inconspicuous burrowing forms. 18 (19) Anchors and anchor plates very large, anchors usually ex- ceeding 600 wm in length, plates usually exceeding 400 (TA 00 eee a a a aR Leptosynapta imswe Pawson 19 (18) Anchors and anchor plates do not exceed 200 wm in length. 20 (21) Anchors and anchor plates very small, considerably shorter than 130 um. Miliary granules resemble rosettes — ee entosynapta nannoplax Pawson 21 (20) Anchors and anchor plates larger, usually exceeding 140 um in length. Miliary granules more or less C-shaped, with enlarged ends, not resembling rosettes — aT OIE se Leptosynapta roseogradia Pawson (13) Skin more or less smooth, but with numerous scattered papillae containing aggregations of wheels. Brick red to POH gee ee ce ee ah Chiridota rotifera (Pourtales ) bo bo Actinopyga agassizii (Selenka ) Figures 1C, E Actinopyga agassizii—Deichmann, 1930, 78, pl. 5, figs. 21-29. Material examined: Lagoon, west of island, coral sand and _ turtle grass, 19 April 1972, collected R. J. Larson, 1 specimen; lagoon, April 1974, 1 specimen; Coral berm, 3 May 1974, collected by B. Spracklin, 1 specimen; back reef, 1-2 m, April 1975, collected by M. Carpenter, 2 specimens. Remarks: Grows to about 20 cm. Five conspicuous calcareous “teeth” surround the anus. Skin thick, leathery. Numerous tube feet dorsally and ventrally. Color variable, but usually with mottled brown and yellow predominating. Tentacles yellow. Common on sand in grassy areas in shallow water around islands of the West Indies, from Barbados to Florida. One record from Ber- muda. The record from Belize is a new range extension into the Western Caribbean. Holothuria impatiens (Forskaal) Holothuria impatiens—Deichmann, 1930:64, pl. 3, figs. 17, 18. Material examined: Just inside reef crest, 7 May 1974, collected by F. H. C. Hotchkiss and k. Sandved, 2 specimens; lagoon, west side 372 Proceedings of the Biological Society of Washington al Fic. 1. A, Holothuria mexicana, ventral; B, Holothuria arenicola, dorsal; C, Actinopyga agassizii, dorsal; D, Isostichopus macroparen- = 4 theses, anterior dorsal; E, juvenile Actinopyga agassizii, ventral. of island, coral sand, turtle grass, under conch shells, 19 April 1972, collected by R. J. Larson, 1 specimen. Remarks: Grows to about 20 cm. Tube feet few, scattered, placed on distinct warts, at least dorsally. Skin thin, rough to touch. Color grey with brownish patches. A “tropicopolitan” species, often uncommon or rare where it occurs. Usually found on sand or in grassy areas. Holothuria arenicola Semper Figure 1B Holothuria arenicola—Deichmann, 1930:66, pl. 4, figs. 1-9. Material examined: Under conch shells, foot of pier, 29 April 1974, collected by F. H. C. Hotchkiss and K. Sandved, 1 specimen; lagoon, Sea cucumbers from Belize 373 west side of island, near shore, 1 May 1974, collected by F. H. C. Hotchkiss and K. Sandved, 1 specimen; lagoon along west shore of island, low tide, 30 April 1972, collected by R. J. Larson, 1 specimen; back reef, 1-2 m, 1975, collected by K. Sandved, 1 specimen; same locality, 22 April 1975, collected by M. Carpenter, 2 specimens; same locality, 27 April 1975, collected by M. Carpenter, 1 specimen. Remarks: Grows to about 30 cm. Body slender, fusiform, with very small tentacles. Adapted to a burrowing habit, and often found con- cealed under rocks and shells. Skin relatively thin and smooth to the touch. Ground color usually light brown, with dark brown patches in 2 series on dorsal surface. Color can vary considerably, probably de- pending upon chemical properties of the habitat; rust-colored to al- most black specimens occur in some areas of the Caribbean. A “tropicopolitan” species, commonly encountered in suitable habi- tats. Holothuria mexicana Ludwig Figures LA, 2A Holothuria mexicana.—Deichmann, 1930:74, pl. 5, figs. 15-20. Material examined: Lagoon, south end of island, 1 specimen; May 1974, collected by F. H. C. Hotchkiss, 1 fragment; back reef, im- mediately behind crest, 1-2 m, April 1975, collected by M. Carpenter, 2 specimens. Remarks: Grows to about 50 cm. Skin very thick, smooth, extremely hard when contracted. In life, body dark brown to blackish dorsally, with flanks yellowish brown, and ventral surface frequently reddish or pink. The reddish to pink ventral coloration disappears in alcohol. The specimen from station 1 above is unusual in being very light brown dorsally. The smaller specimen from the back reef is 110 mm in total length. The ground color is greyish white; ventral feet are light brown with dark brown endplates; dorsal surface with light brown feet and a double row of 4 large dark brown blotches. Deich- mann (1930:74) noted that young specimens of this species have similar coloration, but she did not mention the presence of large dark blotches dorsally. Ranges from Cuba to Curacao in shallow water (to 20 m) in grassy areas or on muddy sand. This species has been frequently confused with the closely related H. floridana Pourtales. Isostichopus badionotus (Selenka) Stichopus badionotus——Deichmann, 1930:80, pl. 5, figs. 30-36. Material examined: Lagoon, west side of South Water Cay, mixed sand and Thalassia, 1 m depth, 4 May 1974, collected by F. H. C. Hotchkiss and K. Sandved, 1 specimen; Coco-Plum Cay, at edge of mangrove swamp, 2 May 1972, collected by R. J. Larson, 1 specimen; 374 Proceedings of the Biological Society of Washington lagoon, west side of island, coral sand, turtle grass, under conch shells, 19 April 1972, collected by R. J. Larson, 1 specimen. Remarks: Grows to about 20 cm. Skin thick, with low warts dorsally and 3 crowded rows of tube feet ventrally. Color highly variable, from light brown to black, usually with numerous large spots or blotches of dark brown on a lighter background. Common on muddy sand or in grassy areas, in shallow water, from Bermuda to Panama. Isostichopus macroparentheses (Clark) Figure 1D Stichopus macroparentheses H. L. Clark, 1922:61, pl. 4, figs. 1-7; Deichmann, 1930:82, pl. 5, figs. 37-43; H. L. Clark, 1933:110 | macraparentheses |. Tsostichopus badionotus.—Deichmann, 1963: 106. Material examined: Back reef, 1-2 m, April 1975, collected by M. Carpenter, 1 specimen. Remarks: Grows to about 12 cm. In the field, virtually indistinguish- able from I. badionotus, and can be positively identified only after examination of the spicules. Deichmann (1963) was inclined to re- gard I. macroparentheses as a juvenile stage of I. badionotus which has exceptionally large C-shaped ossicles. The difference between these C-shaped ossicles in the two forms is dramatic, particularly when one compares juveniles of I. badionotus with I. macroparentheses. 1 can- not agree with Deichmann’s contention, and thus prefer to retain I. macroparentheses as a separate species. In life, “bright brown, with very dark rings around the bases of the papillae which have yellow tips” (Clark, 1933:110). The present specimen is 30 mm long. Color photograph shows light brown body with dark brown dorsal feet and light to dark brown ventral feet; tentacles are more or less colorless. The species is also known from Antigua, Jamaica and the Tor- tugas, in shallow water. Euapta lappa (Muller) Figure 2B Euapta lappa—H. L. Clark, 1924:464, pl. 1, figs. 5-7; 1933:118. Material examined: Lagoon, 2 specimens; just inside reef crest, 6 May 1974, collected by F. H. C. Hotchkiss and K. Sandved, 1 speci- men; reef flat, south end, east side of island “Penicillus” rock zone, 1 specimen; lagoon, west side of island, coral sand, turtle grass, under conch shells, 19 April 1972, collected by R. J. Larson, 2 specimens. Remarks: Grows to about 100 cm. Color grey to brown, often longitudinally striped. Body wall prickly to touch, owing to presence Sea cucumbers from Belize 375 Fic. 2. A, young Holothuria mexicana, lateral; B, Euapta lappa, anterior dorsal; C, Synaptula hydriformis, dorsal. of projecting spicules (anchors). Tentacles plume-like. Body often displays rows of characteristic conspicuous hemispherical protuberances. An active reef-dwelling form often associated with lumps of dead coral. Can swim to a limited extent. Ranges the entire Caribbean in shallow water. Synaptula hydriformis (Lesueur ) Figure 2C Synaptula hydriformis—H. L. Clark, 1924:473, pl. 3, figs. 5, 6, pl. 4, fig. 4; 1933:119. Material examined: Foot of boat pier, bulk sample no. 2, 29 April 1974, collected by F. H. C. Hotchkiss and K. Sandved, 8 specimens: just inside reef crest, 6 May 1974, collected by F. H. C. Hotchkiss and Kk. Sandved, 1 specimen; back reef, 1-2 m, April 1975, col- lected by kK. Sandved, 1 specimen. Remarks: Grows to about 10 cm. Prickly to touch due to projecting spicules (anchors). Two color phases, mottled green and white and mottled brown and white, can occur. Some correlation exists between habitat type (red or brown or green algae frequently) and color of body wall, but exceptions occur; for example, in Bermuda the brown form is conspicuous among clumps of the green Penicillus. Green form only known so far from Carrie Bow Cay. Viviparous; apparently can breed all year round in Bermuda. Ranges from Bermuda to Brazil. Usually found in weed in shallow water. 376 Proceedings of the Biological Society of Washington Leptosynapta Verrill, 1867 The Carrie Bow Cay collections contain a total of 11 complete specimens and several fragments of synaptids which can be referred to the genus Leptosynapta. Surprisingly, the specimens represent 3 new and distinctive species. While I am reluctant to describe 3 fur- ther species in a genus which is already quite large (approximately 25 species) and requires revision, it is clear that under currently accepted taxonomic criteria for the family Synaptidae, the present species must be regarded as new. It seems likely that further collecting in sandy areas of the Caribbean will reveal a large and diverse fauna of burrowing synaptids; such habitats have received very little at- tention from collectors in the past. The 3 species described below do not appear to be closely related to one another, but each shares some important features with other western Atlantic congeners. Leptosynapta imswe, new species Figure 3 Diagnosis: Anchors and anchor plates of one kind, very large, an- chors usually exceeding 600 jm in length, plates exceeding 400 wm in length. Miliary granules numerous, in form of C-shaped rods with enlarged ends. Material examined: Holotype (USNM_ E15854, specimen 57 mm long), lagoon, sand, north end, west side of Carrie Bow Cay, Belize, 27 April 1974, collected by F. H. C. Hotchkiss and K. Sandved. Para- types 3 complete specimens and 11 fragments (USNM E15855) from same locality as holotype. Etymology: The species is named for the Smithsonian Institution I.M.S.W.E. program (Investigations of Marine Shallow Water Eco- systems ). Description: Total length ranges from 7 to approximately 90 mm; probably species exceeds 120 mm in life. Specimens uniformly whitish, body wall translucent when expanded; color in life pink to light brown. Conspicuous anchors project. through body wall rendering specimens very prickly to touch. Tentacles 12, each with 5-6 pairs of digits and a terminal digit; digits increase in length distally, and terminal digit is longest (Fig. 3F). Inner (oral) surfaces of tentacles with double row of well developed sensory cups. Body wall deposits comprise large anchors and anchor plates of one kind, and numerous miliary granules. Anchors and plates at an- terior, middle and posterior of body wall essentially similar, although developmental stages of these ossicles more numerous posteriorly. An- chors average 630 wm in length (standard deviation 2.49; standard error 0.83), 374 um in width, arm-tip to arm-tip (standard deviation 2.01; standard error 0.67), 126 wm in width of stock (standard de- Sea cucumbers from Belize 377 = — AY _ 100um Den) G0 fal ) £5 9 6 2) es 100m 50 um Fic. 3. Leptosynapta imswe n. sp. A, Anchor; B, Detail of anchor arm; C, Detail of anchor stock; D, Miliary granules from body wall; E, Ossicles from tentacle stem; F, Outline of tentacle; G, Anchor plates; H, Ossicles from tentacle digits. viation 1.26; standard error 0.42). Arms carry up to 10 conspicuous sharp teeth. Stock unbranched, but equipped with numerous small sharp projections (Fig. 3A-—C). Anchor plates elongate, approximately oval, with numerous toothed perforations (Fig. 3G); central perforations tend to be larger than others. No true bridge for support of anchors, but in area of support, anchor plate strengthened by having an_ ir- regular “pseudo-bridge” in form of a double layer of calcite. Anchor plates average 447 wm in length (standard deviation 3.09; standard error 1.03), and 246 wm in greatest width (standard deviation 0.84; standard error 0.28). Miliary granules numerous everywhere in body wall, highly variable in shape, but generally tending to have enlarged, recurved ends. Granules up to approximately 30 wm in length (Fig. 3D). Stems of tentacles with ossicles similar to miliary granules of body wall but tending to be slightly smaller (Fig. 3E). In tentacle digits 378 Proceedings of the Biological Society of Washington ossicles tend to be more elongate, length up to 45 ym, and some have pertorated ends (Fig. 3H). Remarks: This is one of the few species of Leptosynapta sensu lato which has very large anchors and anchor plates. L. acanthia H. L. Clark, known only from Bermuda, has anchors and plates which are similar in size to those in L. imswe, but the plates have fewer per- forations, and further, L. acanthia has numerous small anchors 140— 210 wm in length and plates of approximately the same size (Clark, 1924:478); such deposits are apparently lacking from L. imswe. L. multipora H. L. Clark also has large anchors and plates but lacks the numerous miliary granules. Leptosynapta roseogradia, new species Figure 4A—D Diagnosis: Anchors of one type, usually less than 200 um long, anchor plates of one type, usually less than 160 um long. Miliary granules C- or bracket-shaped, numerous. Radial pieces of calcareous ring perforated for radial nerve. Material examined: Holotype (USNM_ E15856, specimen 36 mm long), lagoon, sand, north end, west side of Carrie Bow Cay, Belize, 27 April 1974, collected by F. H. C. Hotchkiss and K. Sandved. Para- types 3 specimens (USNM E15857) from same locality as holotype. Etymology: The species is so named to indicate its superficial simi- larity to Epitomapta roseola. Description: Total length 16, 26, 36, and 40 mm. Specimens white to yellowish, body wall thin, translucent; color in life pink to light brown. Tentacles 12, each with 5-6 pairs of digits and a terminal digit, which is longest. Sensory cups present in small numbers on oral surface of tentacles. Radial pieces of calcareous ring perforated for passage of radial nerve. Ciliated funnels all of one type, small, approximately 100 »m in length. Deposits in body wall anchors, anchor plates and numerous miliary granules. Anchors and plates at anterior, middle and posterior of body essentially similar, but differing in dimensions: Anterior anchors length 167 wm S.D. 1.36 S.E. 0.43 plates length 154 um S.D.1.29 S.E. 0.42 Mid-body — anchors length 192 wm S.D.2.6 S.E. 0:93 plates length 182 pm $.D.3.6 S.E. 1.29 plates width 142 um S.D. 2.46 S.E. 0.89 Posterior anchors length 148 wm S.D. 1.32 S.E. 0.44 plates length 144 pm S.D. 1.42 S.E. 0.45 Anchors (Fig. 4A) have 1-3 serrations on arms; stock with numerous small teeth. Anchor plates (Fig. 4D) approximately oval, with 7 Sea cucumbers from Belize 379 > a 1) i ) eS as ey 2) Oo ( —— - ~ en ey SOpm 30pm _ 50 um ie: 2) Le E = ~ Faas J vay € é 4 4 or ( -) \ (f > wo | ea ; | < 4 WIS ie ) Ey Ey! (O C Ad hee YOM \wapO7 SEU eae a (19) \ e a / ( es - LX) Sah S WI 9 G/ Zo (rs QUvV/ C0, ¢@ 7)O/ () 0 100 urn 40 um _ EF ie on H sy, ox ae : BY pa TOD pw Dd sae Mh TR AV eo oA. \ tir we Ci Qs SF &9 TA NN “US ais) Sas Be Qn, | ‘ aan, \ \ \ PAG Gosia B leh ¢ . Sige oe i. a as AS i eP jae i S ee Nae ce : yy, ; op GP & ce miyl) 3} Yipee a) eet) Oe) WRAL gi Rast 40 um Se) ge J et: 40um 50um Fic. 4. Leptosynapta roseogradia n. sp. A, Anchor; B, Miliary gran- ules from body wall; C, Ossicles from tentacle stems and digits; D, Anchor plates. Leptosynapta nannoplax n. sp. E, Anchor plates; F, Anchor; G, Miliary granules from body wall; H, Ossicles from stems and digits of tentacles. major perforations, always toothed, also with varying numbers of smaller perforations. Plate distinctly narrower at articular end, with several small perforations. No true supporting bridge for anchor. Body wall with minute miliary granules (Fig. 4B) more or less C- or bracket- shaped, with enlarged ends. Size variable, largest granules approxi- mately 25 um in length. Tentacle stems and digits contain granules similar to those in body 380 Proceedings of the Biological Society of Washington wall, and digits in particular have more elongate rods (Fig. 4C) up to 60 um in length. Remarks: This is another distinctive species of Leptosynapta. The anchors and anchor plates strongly resemble those of Epitomapta roseola (Verrill) from Woods Hole, Bermuda, and Jamaica, but E. roseola has imperforate radial pieces in the calcareous ring, ciliated funnels of two types, and anchor plates which are considerably smaller (110 wm, Heding, 1928:237) than those of L. roseogradia. Within the genus Leptosynapta, L. tenuis (Ayers) from the north- eastern United States shares some features with this new species, but differs in having larger anterior and posterior anchors, smaller an- terior anchor plates, and differently shaped miliary granules and tentacle rods. Further, in L. tenuis, the anterior and posterior anchors are distinctly different in appearance, a feature not observed in L. roseogradia. Leptosynapta nannoplax, new species Figure 4E—H Diagnosis: Anchors and anchor plates very small, anchors usually less than 125 pm in length, anchor plates usually less than 90 pm in length. Miliary granules resemble rosettes. Material examined: Holotype (USNM_ E15858, specimen 31 mm long), lagoon, sand, north end, west side of Carrie Bow Cay, Belize, 27 April 1974, collected by F. H. C. Hotchkiss and K. Sandved. Paratypes 2 specimens (USNM E15859) from same locality as holotype. Etymology: The specific name is derived from Greek nannos small, and plax, a plate, in reference to the diminutive ossicles in the body wall. Description: Total length of 3 specimens 26, 27 and 31 mm. Species may exceed 50 mm in life. Specimens white, body wall translucent; color in life pink to light brown. Twelve pinnate tentacles, with 3-5 pairs of digits and a terminal digit which is the longest. Sensory cups present on oral surface of tentacles. Body wall ossicles comprise very small anchors and anchor plates of one kind, and minute miliary granules. Anchors and plates at an- terior, middle, and posterior of body similar, but differing slightly in dimensions: Anterior anchors length 86 pm S: Di: 0379 25. E025 plates length 83 um S.D.0.51 S.E. 0.16 Mid-body — anchors length 107 «wm S.D. 4.36 $.E. 153 plates length 90 um S.D.1.91 S.E. 0.64 plates width 61 um S.D; 1,62, -S.B0jol Posterior anchors length 124 wm S.D.0.70 S.E. 0.22 plates length 88 wm S.D.0.71 S.E. 0.24 Sea cucumbers from Belize 381 Anchors carry up to 4 or 5 distinct serrations on arms; stock is toothed but not strongly so (Fig. 4F). Anchor plates elongate oval, usually with 7 large dentate perforations, and several smaller holes at anchor support area (Fig. 4E). No true bridge for anchor support, but usually a double layer of calcite is more or less well developed. Miliary granules scattered, usually less than 40 um in length (Fig. 4G). They bear close resemblance to rosettes found in other groups of apodous holothurians. Tentacle stems and digits contain numerous deposits similar to miliary granules of body wall; in addition, digits in particular contain elongate rods up to approximately 70 um in length (Fig. 4H). Remarks: Very few species of Leptosynapta have anchors and _ plates of such small size, and apparently none have miliary granules which resemble rosettes in combination with small anchors and plates. L. micropatina Heding from Tobago has anchors and plates of the same order of size as L. nannoplax, but does not possess the rosette-like miliary granules. Further, the anchor plates of L. micropatina usually contain smaller perforations at the wider ends of the plates in addition to the larger dentate perforations; these are lacking from L. nannoplax. In view of the relatively small size of L. nannoplax, it is conceivable that this is the young form of a known species; if this is the case, then it is to be expected that the rosette-like miliary granules disappear or are supplanted by granules of a different type as the animal grows. Chiridota rotifera (Pourtales ) Chiridota rotifera—Heding, 1928:293, figs. 59, 60; H. L. Clark, 1933: 122), Material examined: Foot of pier, bulk sample no. 2, 29 April 1974, collected by F. H. C. Hotchkiss and K. Sandved, 5 specimens; subtidal, bulk sample, 28 April 1974, collected by F. H. C. Hotchkiss and K. Sandved, 1 specimen; back reef, in conch shell, 1 m depth, 27 April 1975, collected by K. Sandved, 1 specimen. Remarks: Grows to about 10 em. Body smooth to touch, with more or less conspicuous aggregations of calcareous spicules (“wheel pa- pillae”) scattered in rows along the interradial areas, especially dor- sally. Ground color light brick-red to pink. Viviparous. Common on sand under rock in shallow water; also occurs in sandy beaches. Ranges from Brazil to Bermuda. LITERATURE CITED Crark, H. L. 1922. The holothurians of the genus Stichopus. Bull. Mus. Comp. Zool. Harvard 45(3 ) :40-74, 2 pls. 1924. The holothurians of the Museum of Comparative Zoology, The Synaptinae. Ibid. 45(13):459-501, 12 pls. 1933. A handbook of the littoral echinoderms of Porto 382 Proceedings of the Biological Society of Washington Rico and the other West Indian islands. Sci. Surv. Porto Rico Virgin Islands, N. Y. 16(1):1-60, 7 pls. Dahl, A. L. 1973. Surface area in ecological analysis: quantification of benthic coral-reef algae. Marine Biology 23:239-249, 8 figs. DeEICHMANN, E. 1930. The holothurians of the western part of the Atlantic Ocean. Bull. Mus. Comp. Zool. Harvard 71(3):43- 226, 24 pls. ——. 1963. Shallow water holothurians known from the Carib- bean waters. Stud. Fauna Curacao 14:100-118. Devaney, D. M. 1974. Shallow-water echinoderms from British Hon- duras with a description of a new species of Ophiocoma (Ophiuroidea). Bull. Mar. Sci. 24(1):122-164, 16 figs. Hepinc, S. G. 1928. Synaptidae: Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-1916, no. 66. Vidensk Medd. dansk naturh. Foren. Kjobenhayn 85:105-323, pls. 2, 3. Krer, P. M. 1975. The echinoids of Carrie Bow Cay, Belize. Smith- sonian Contrib. Zool. 206:1—45, 8 figs., 12 pls. Na 4 \ Se UT few “T abe tf abl ty Ww No. 32, pp. 383-392 12 October 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW TROGLOBITIC CRAYFISH (DECAPODA, CAMBARIDAE) FROM PENINSULAR FLORIDA By Horton H. Hosss, Jr., anp Davin S. LEE Department of Invertebrate Zoology, Smithsonian Institution, Washington, D.C. 20560, and North Carolina State Museum of Natural History, Raleigh, N.C. 27611 Several previously undescribed species of troglobitic cray- fishes have been reported recently from Florida. Hobbs (1971) described Procambarus (Leconticambarus) milleri, the first new form recorded since 1942 (when he reviewed the crayfishes of the state). At this time (1971), he sum- marized the known ranges, provided a key, and illustrated the diagnostic features of all of the troglobitic species known to occur in Florida. Since then, three additional crayfishes have been named: Procambarus (O.) orcinus Hobbs and Means (1972:394), P. (O.) horsti Hobbs and Means (1972: 401), and P. (O.) erythrops Relyea and Sutton (1975:8). Accompanying the description of the latter was a revised key to the troglobitic crayfishes of the state prepared by Hobbs. The species described here brings the total number of known Floridian troglobitic crayfishes to 11, all except two of which are members of the genus Procambarus. We take pleasure in naming this crayfish in honor of Richard Franz, a colleague, friend, and outstanding student of cave ecosystems. His continued interest and support of this study are greatly appreciated. We wish to thank Sylvia Scudder, Richard Bradley, Bar- bara Lee, and Richard Franz for their assistance in collect- ing the specimens on which the following description is based. We also wish to acknowledge the assistance of John E. Cooper who provided useful information leading to 32—Proc. Biou. Soc. WasH., VoL. 89, 1976 (383) 384 Proceedings of the Biological Society of Washington Franz’s initial location of Orange Lake Cave, the source of our specimens. For their critical reading of the manuscript, we are grateful to Thomas E. Bowman, Martha R. Cooper, and Margaret A. Daniel. Procambarus (Ortmannicus) franzi, new species Diagnosis: Albinistic, eyes without pigment or faceted cornea. Rostrum with, or more often without, marginal spines; median carina absent. Carapace with cervical spine cephaloventral to row of spines or tubercles flanking caudal margin of cervical groove. Areola 12.8 to 17.2 times as long as broad and constituting 38.7 to 41.0% of total length of carapace (47.1 to 49.5% of postorbital carapace length). Suborbital angle absent. Postorbital ridge with cephalic spine. Hepatic area with many small tubercles, some spiniform. Antennal scale ap- proximately twice as long as wide, broadest slightly distal to midlength. Ischia of third and fourth pereiopods of first form male with simple hooks, that on third overreaching basioischial articulation and that on fourth highly arched, almost reaching basioischial articulation, and op- posed by low eminence on basis; coxa of fourth pereiopod with promi- nent oblique boss. First pleopod of first form male reaching coxa of third pereiopod, asymmetrical, provided with subapical setae; distal extremity bearing subspiculiform mesial process directed caudally at approximately right angle to shaft of appendage and curved laterally; cephalic process rather short, acute, somewhat hooding central pro- jection cephalically, and directed caudodistally; caudal element con- sisting of inconspicuous, caudolaterally situated caudal knob, and prominent, corneous adventitious process caudomesially, latter distally rounded, convex mesially, and somewhat concave laterally; and corneous beaklike central projection, most conspicuous of terminal elements, directed caudodistally subparallel to cephalic process. Annulus ven- tralis freely movable, subspindle-shaped, slightly more than twice as broad as long, and completely exposed, not (partly) hidden by pro- jections from sternum immediately cephalic to it; cephalic area with convex elevated area bearing submedian J-shaped furrow; sinus origi- nating in furrow and, following sinuous curve caudally, terminating to side of median line slightly caudal to midlength of annulus. Postannular plate slightly more than half as wide and half as long as annulus with cephalic region somewhat inflated. First pleopod in female moderately well-developed. Holotypic Male, Form I: Cephalothorax (Fig. la, 1) subcylindrical. Abdomen narrower than thorax (10.1 and 13.1 mm). Greatest width of carapace greater than height at caudodorsal margin of cervical groove. Areola 14.7 times as long as wide with 1 or 2 punctations across narrowest part. Cephalic section of carapace approximately 1.5 times as long as areola, length of latter 40.7% of entire length of New cave crayfish from Florida 385 Fic. 1. Procambarus (Ortmannicus) franzi, new species (all illustra- tions are of holotype except c and e of morphotype, and d of allo- type): a, Lateral view of carapace; b, c, Mesial view of first pleopod; d, Annulus ventralis; e, f, Lateral view of first pleopod; g, Basal podo- meres of third, fourth, and fifth pereiopods; h, Epistome; i, Antennal scale; j7, Caudosinistral view of first pleopods; k, m, Mesial view of terminal part of first pleopod showing subterminal setae; /, Dorsal view of carapace; n, Dorsal view of distal podomeres of cheliped. 386 Proceedings of the Biological Society of Washington carapace (47.9% of postorbital carapace length). Rostrum’ with gently convergent margins, lacking marginal spines or tubercles but with broad-based acumen clearly delimited by mesially curved cephalic ends of rostral margins. Acumen reaching midlength of penultimate segment of antennular peduncle; dorsal surface shallowly excavate and sparsely punctate. Subrostral ridge rather weak and evident in dorsal aspect along caudal three-fifths of rostrum. Postorbital ridges well- developed, grooved dorsolaterally, and terminating cephalically in small spine, more posterior spines lacking. Caudodorsal margin of cervical groove with row of spiniform tubercles, ventralmost member of row (cervical spine), larger than others. Suborbital angle virtually obsolete. Brachiostegal spine moderately strong. Entire dorsolateral and lateral surfaces of carapace studded with tubercles, some of them spiniform. Abdomen only slightly longer than carapace (25.9 and 25.3 mm). Pleura of third through fifth abdominal segments truncate ventrally, those of fourth and fifth segments subangular caudoventrally. Cephalic section of telson with single fixed spine in caudodextral corner and 2 in caudosinistral corner. Cephalic lobe of epistome (Fig. 1h) sub- rhomboidal but asymmetrical with cephalomedian projection, 3 acute prominences on cephalosinistral margin, and 2 on _ cephalodextral; main body of epistome with shallow cephalomedian fovea continuous with caudally disposed submedian sulcus abutting epistomal zygoma. Ventral surface of proximal podomere of antennular peduncle with prominent submedian spine near midlength. Antenna with moderately large spine on basis and small spiniform tubercle on ischium; flagellum extending caudally beyond telson by at least length of latter. Antennal scale (Fig li) almost twice as long as broad, widest distal to mid- length, and lamellar (part) about twice as wide as thickened lateral part. Third maxilliped extending anteriorly to, or slightly beyond, level of tip of rostrum; ischium with distolateral extremity produced in acute prominence, and lateral half of ventral surface with scattered short setiferous punctations; exopod reaching distal end of carpus. Right chela (Fig. 1n) subovate in cross section, not strongly de- pressed. Mesial surface of palm with several irregular rows of 10 or 11 strongly elevated tubercles; remainder of palm also tuberculate, more mesial tubercles stronger than more lateral ones on both dorsal and ventral surfaces. Both fingers with well defined submedian longi- tudinal ridge flanked proximally by tubercles and more distally by setiferous punctations. Opposable margin of fixed finger with row of 8 tubercles along proximal half of finger, third from base largest, and with single, large, more ventrally situated one slightly proximal to midlength; band of minute denticles extending almost from proximal end of finger (between and ventral to row of tubercles) to base of corneous tip, band distinctly broader distal to level of ventral tubercle. Lateral surface of fixed finger with tubercles, decreasing in size dis- New cave crayfish from Florida 387 TABLE 1. Measurements (mm) of Procambarus (Ortmannicus) franzi. Holotype Allotype Morphotype Carapace Height 10.2 UHI 10.0 Width VS 12.9 12.2 Total length 25.3 26.9 25.2 Postorbital length 21.5 22.1 20.9 Areola Width 0.7 0.7 0.8 Length 10.3 10.4 10.2 Rostrum Width 3.6 3.6 3.5 Length 5.2 5.9 5.6 Chela Length of mesial margin of palm 9.0 7.0 7.0 Width of palm 5.6 4.5 4.6 Length of lateral margin Z5ul: 21.8 20.2 Length of dactyl 14.7 13.4 11.7 Abdomen Width 10.1 10.2 9.8 Length 25.9 26.7 25.3 tally, along proximal half followed by row of setiferous punctations, latter reaching base of corneous tip of finger. Opposable margin of dactyl with row of 9 tubercles, fifth from base largest, along proximal third of finger; minute denticles between and dorsal to row of tubercles and forming broad band immediately distal to last tubercle of row, band continuing to base of corneous tip of finger. Mesial surface of dactyl similar to lateral surface of fixed finger with tubercles diminish- ing in size distally. Carpus of cheliped longer than broad, tuberculate on all surfaces, conspicuously so mesially where 1 tubercle more prominent than others; additional strong tubercle present on dorsomesial distal angle. Shallow oblique sulcus on dorsal surface flanked by small subsquamous tubercles. Ventrodistal margin of podomere with 3 tubercles: weak lateral one, moderately strong one on articular knob, and prominent acute one mesially; row of 3 tubercles extending proximally from latter. Merus of cheliped strongly tuberculate except for proximal parts of mesial and lateral surfaces. Dorsal surface with linear series of tubercles basally continuous with broadening band of them distally; tubercles on ventral surface abundant, not limited to usual 2 rows; rows poorly 388 Proceedings of the Biological Society of Washington defined but each consisting of 20 to 23 tubercles. Ischium with row of 6 tubercles ventromesially, distalmost slightly larger than proximal 5. Hooks on ischia of third and fourth pereiopods (Fig. 1g) as de- scribed in “Diagnosis.” Coxa of fourth pereiopod with prominent oblique (almost vertical), somewhat inflated caudomesial boss; that of fifth pereiopod with very weak prominence corresponding to caudo- mesial boss on fourth pereiopod. Sternum between third and fourth pereiopods rather deep with conspicuous mat of plumose setae extending mesially from ventrolateral margins. First pleopods (Fig. 1b, f, j, k, m) as described in diagnosis; mesial process of left member not bent nearly so strongly laterally as that on right pleopod. Uropod with both lobes of basal podomere bearing spines; distomedian spine on mesial ramus far removed from distal margin of ramus. Allotypic Female: Differing from holotype in following respects: rostrum with marginal spines and reaching proximal end of ultimate podomere of antennular peduncle; subrostral ridges evident in dorsal aspect only along basal fourth of rostrum; tubercles of spiniform row flanking caudal margin of cervical groove much less conspicuous and cervical spine smaller; cephalic section of telson with 2 spines in each caudolateral corner; cephalic lobe of epistome symmetrical with weak prominences at angles of rhomboid; opposable margin of fixed finger of chela with row of 9 tubercles, fourth from base largest, that of dactyl with row of 13 tubercles, fifth from base largest; ventromesial margin of ischium of cheliped with row of 7 tubercles. Annulus ventralis (Fig. ld) only moderately deeply situated in sternum (see “Diagnosis” for details). Sternum immediately cephalic to annulus lacking ornamentations. First pleopod reaching slightly cephalic to caudal margin of annulus when abdomen flexed (see “Mea- surements’ ). Morphotypic Male, Form II: Differing from holotype in following respects: rostrum with small marginal tubercle dextrally (sinistral margin as in holotype), and reaching almost midlength of ultimate podomere of antennular peduncle; subrostral ridges evident in dorsal aspect from caudal margin of orbit to acumen; postorbital ridge with well defined posterior tubercles, sinistral one almost serrate; cervical spine very small, hardly larger than adjacent tubercles; epistome simi- lar to that of allotype but anterolateral angles less sharp, and weak prominences present closely flanking cephalomedian prominence; ce- phalic section of telson with single fixed spine in each caudolateral corner; lateral ramus of sinistral uropod deformed, possessing bilobed distal segment; opposable margin of fixed finger of chela with row of 6 tubercles, that of dactyl with 11; ventromesial surface of ischium of cheliped with row of 8 tubercles and few others nearby; hooks on ischia of third and fourth pereiopods much reduced, that on third not overreaching basioischial articulation; boss on coxa of fourth pereiopod also much reduced. First pleopod (Fig. lc, e) without corneous ter- New cave crayfish from Florida 389 minal elements; mesial and cephalic processes much shorter and heavier, caudal element much less sharply defined, and central projection shorter, and decidedly more tumescent (see “Measurements” ). Type-locality: Orange Lake Cave, 0.4 mi S of junction of U.S. Hwy. 441 and State Route 318 off Hwy. 411 (T.12S, R.21E, Sec. 33/34), Marion County, Florida. Orange Lake Cave is a small, horizontal cavern located on the north side of a quarry of several acres. The status of the entrance prior to quarrying operations is unknown. The two present openings (2 m xX 15 m; 15 m x 3 m) lead into a small room. The main passage (approximately 1 m wide) leads north and northeast from about 20 m and terminates in a large, irregular room (15 m long, 5 m wide, and 3-7 m high) where the water level fluctuates markedly. On 8 Sep- tember 1974, water was encountered within 5 m of the entrance, and the floor of the entire passage and that of the room in the back of the cave were inundated. On 18 May 1975 and 19 November 1975, the water was about a meter lower and was confined to the rear of the big room. On other visits (5 January 1975, 28 September 1974, etc.) water levels were intermediate between these extremes. Water tempera- ture remained at 21—22° C throughout the collecting period. In the spring, as many as 3700 bats (Myotis austroriparius) use the back room of this cave as a maternity site. Throughout the remainder of the year, the bats are not usually present, but Richard Franz re- ported 6000 individuals. having been found there on 19 November 1975. The bats are probably responsible for the major source of energy for the Orange Lake Cave crayfish population. Nearly all of the crayfish were observed in the pool directly under the bat roosts. The total number of crayfish counted on 18 May 1975 and 19 No- vember 1975, when they were confined to one pool, was 32 and 23, respectively. A large percentage of the population consisted of juveniles. On two occasions (October 1974 and 5 January 1975), several small, white crayfish, assumed to be a part of this same population, were seen in a small solution cavity in the floor of the quarry. The water level was about 1.5 m below the bottom of the quarry. Other openings in the sides and bottom of this same quarry apparently do not contain water. Disposition of Types: The holotype, allotype, and morphotype are deposited in the National Museum of Natural History (Smithsonian Institution), numbers 146992, 146993 and 146994, respectively, as are the paratypes consisting of 141], 5@, and partly decayed remains of 141. Size: The largest of the specimens is a female having a carapace length of 29.0 mm (postorbital carapace length 23.5 mm). Perhaps the (partly) decayed form I male was a little larger. The only other first form male is the holotype possessing corresponding lengths of 25.3 and 21.5 mm. Range and Specimens Examined: This crayfish is known only from 390 Proceedings of the Biological Society of Washington the type-locality, and the only specimens available are those included in the type-series: 141, 8 September 1974, R. Franz and D. S. Lee, coll.; 14II, 29, 1 October 1974, R. F. and D. S. L., coll.; 32, 19 May 1975, B. Lee and D. S. L.; 141, 12, 19 November 1975, R. F., R. Bradley, and S. Scudder, coll.; 1411, 8 December 1975, R. F. coll. Variations: The most conspicuous variation observed is in the ros- trum. Except in a few specimens in which there are marginal spines present and in the morphotype in which such a spine is present on one side, all of the other specimens lack marginal spines. Instead the rostral margins curve mesially at the base of the acumen, merging with the flattened dorsal surface of the latter. The subrostral ridges which are continuous with the margin of the acumen may be visible along the length of the rostrum from the level of the caudal margin of the orbit to the tip of the acumen, or the rostral margins may obscure, in dorsal view, part of them between the orbit and the base of the acumen. The other slight variations noted are in the number, dis- position, and sizes of the tubercles, both on the carapace and on the chelipeds, but only those differences occurring on the postorbital ridges seem worthy of mention. Even in a single individual, the tubercles on the paired ridges are not identical; one ridge may have an arrange- ment of tubercles that might be described as almost serrate while the tubercles on the other are so reduced as to be easily overlooked. The only consistency is the presence of a small apical tubercle. Relationships: Procambarus (Ortmannicus) franzi is the seventh troglobitic member of the subgenus to be described, all of them from the subterranean waters of Florida, and all at least as closely related to one another as to any epigean crayfish. Three of the previously described species, P. (O.) pallidus (Hobbs, 1940:394), P. (O.) horsti and P. (O.) orcinus share in common a somewhat laterally displaced cephalic process on the constricted distal part of the first pleopod of the male, and the sternum immediately cephalic to the annulus ven- tralis in the female is produced into caudally projecting prominences that are not present in the remaining ones: P. (O.) lucifugus lucifugus (Hobbs, 1940:398), P. (O.) 1. alachua (Hobbs, 1940:402), P. (O.) erythrops, and the species described here. The range of P. (O.) franzi to the south and east of P. (O.) l. alachua, with which it seems to share more features than with the other species, tends to support an assumption of close kinship. Speci- mens that have been identified as intergrades between the two sub- species of P. (O.) lucifugus were obtained from several caves approxi- mately 20 mi S of Orange Lake Cave (see Warren, 1961:7). The characters cited above distinguish it from P. (O.) pallidus, P. (O.) horsti, and P. (O.) orcinus, and the absence of pigment in the eye serves readily to separate it from P. (O.) 1. alachua and P. (O.) erythrops. The tapering rostrum distinguishes it from P. (O.) lL. luci- fugus, and the terminal part of the first pleopod of the male is unique. Common Name: Because this crayfish appears to have such a limited New cave crayfish from Florida 391 range, the possibility of its ultimately being placed on the list of en- dangered species is very real. Inasmuch as a common name will be- come mandatory when its status of probable safety is considered, we propose that it be known as the Orange Lake Cave Crayfish. LITERATURE CITED Hosss, Horton H., Jr. 1940. Seven new crayfishes of the genus Cambarus from Florida, with notes on other species. Proc. U.S. Nat. Mus. 89(3097 ):387-423, figs. 14-22. ——. 1942. The crayfishes of Florida. Univ. Fla. Publ., Biol. Sci. Ser., 3(2):1-179, 24 pls. ——. 1971. A new troglobitic crayfish from Florida. Quart. Jour. Fla. Acad. Sci. 34(2):114—-124, 19 figs. ———., anD D. Bruce Means. 1972. Two new troglobitic cray- fishes (Decapoda, Astacidae) from Florida. Proc. Biol. Soc. Wash. 84( 46) :393-409, 2 figs. RELYEA, KENNETH, AND BrucE Surron. 1975. A_ new troglobitic crayfish of the genus Procambarus from Florida (Decapoda, Astacidae). Tulane Stud. Zool. Bot. 19(1-2):8-16, 4 figs. WarrREN, Ricuarp D. 1961. The obligative cavernicoles of Florida. Fla. Speleol. Soc., Spec. Pap. 1:1-10, 2 figs. 392 Proceedings of the Biological Society of Washington Wo No. 33, pp. 393-404 12 October 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NEW ENTOCYTHERID OSTRACODS FROM KENTUCKY AND TENNESSEE By Horton H. Hosss, Jr., AND MARGARET WALTON Department of Invertebrate Zoology, Smithsonian Institution, Washington, D.C. 20560, and Mountain Lake Biological Station Pembroke, Virginia 24136 Four new ostracods are described from crayfishes collected in the Cumberland and Tennessee River basins in Kentucky and Tennessee. One, a member of the genus Ascetocythere, is believed to be restricted to a burrowing crayfish, whereas the other three, assigned to the genus Dactylocythere, infest stream dwelling members of the genus Cambarus and _per- haps also two species of the genus Orconectes. The most recent keys and summary of the genera Ascetocythere and Dactylocythere are those of Hart and Hart (1974). Except for four lots of specimens, from three localities, which were collected by Perry C. Holt, (1 lot), Joseph F. Fitzpatrick, Jr.. and H. H. Hobbs, Jr., (2), and Daniel J. Peters, Jean E. Pugh, and H. H. Hobbs, Jr. (1), all were donated to us by Raymond W. and Judith Way Bouchard who also supplied us with identifications of the hosts collected by them. We gratefully acknowledge the gift from the Bouchards and Perry C. Holt, and the assistance given to us by the other collectors. We also wish to thank Margaret A. Daniel, C. W. Hart, Jr.. and Raymond B. Manning for their critical reading of the manuscript. Ascetocythere riopeli, new species (Figure la—d) Male: Eye pigmented and located between one-fourth and one-fifth shell length from anterior margin. Shell (Fig. la) ovate with greatest 33—Proc. Brot. Soc. Wasu., Vou. 89, 1976 (393) 394 Proceedings of the Biological Society of Washington Fic. la—d, Ascetocythere riopeli, new species; le-h, Dactylocythere apheles, new species. a, e, Shell of male; b, g, Copulatory complex of male; c, h, Clasping apparatus of male; d, f, Shell of female. Scales in mm. height one-third shell length from posterior margin where about 1.4 times height at level of eye; margin entire. Submarginal setae present except dorsally, closer together anteriorly and posteriorly than ven- trally. Copulatory complex (Fig. 1b, c) with bulbous ventral portion of peniferum bearing following processes: that borne on cephaloventral margin, and serving as conduit guiding penis, swollen distally and New entocytherid ostracods 395 flared with margin produced in several small prominences; adjacent ventrally directed process slender, subsinuous with cephaloventral ex- tremity slightly broadened and turned laterally; vestigial caudal process situated near caudal base of sinuous process, its subacute apex di- rected cephalically. Penis very prominent, L-shaped, with spermatic and prostatic elements gaping for some distance at and on both sides of midlength of penis, two converging in basal part of cephaloventral process of peniferum and emerging on mesial surface of swollen distal region. Clasping apparatus only slightly curved, tapering from base, its postaxial border entire; preaxial border with 3 teeth on distal third followed by 2 apical ones. Dorsal finger moderately heavy, its setiform tip overreaching midlength of uniformly curved ventral finger, tip of which directed anteroventrally. Triunguis Female: Eye pigmented and situated approximately one- fourth shell length from anterior margin. Shell (Fig. 1d) elongate ovate, highest about one-third shell length from posterior margin where 1.3 times height at level of eye. Submarginal setae disposed as in male but closer together anteroventrally than elsewhere. Genital complex consisting of small, weakly sclerotized prominence with concave apical margin and embedded in heterogeneous mass projecting posteroven- trally from posterodorsal part of body. Measurements (in mm): 4 males and 1 female. Holotype Males Allotype Length (range ) 0.39 0.38-0.41 0.41 Average 0.39 Height (range ) 0.22 0.22-0.24 0.22 Average 0.22 Type-locality: Poor Fork of the Cumberland River, at Upper Cumber- land School, 0.6 mi from Virginia state line, Letcher County, Ken- tucky. The specimens were retrieved from a collection of crayfishes containing Cambarus (Jugicambarus) dubius Faxon, C. (J.) distans Rhoades, C. (Puncticambarus) robustus Girard, and C. (P.) buntingi Bouchard. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), numbers 155317 and 155318, respectively, as are the paratypes. Range and Specimens Examined: Six specimens from two localities in Kentucky: Type-locality, 30 March 1974, R. W. and J. W. Bouchard, coll. and Marrowbone Creek at State Route 195 and Elkhorn Creek at State Route 197 (combined collection), Pike County, Kentucky, 29 March 1974, R.W.B. and J.W.B., coll. Hosts: Although five species of crayfishes are represented from the collections containing this ostracod (Cambarus (Jugicambarus) dubius, 396 Proceedings of the Biological Society of Washington C. (J.) distans, C. (Puncticambarus) robustus, C. (P.) buntingi, ana Orconectes juvenilis (Hagen) ), it is highly probable that this ostracod is confined to C. (J.) dubius. Entocytherid Associates: In the type-locality, Ascetocythere riopeli was found in a collection containing Dactylocythere spinata Hobbs and Walton (1970:860), Dactylocythere sp.P, and Donnaldsoncythere donnaldsonensis (Klie, 1931:334). In the Pike County locality, there were no associates. Relationships: Ascetocythere riopeli is a member of the Asceta Group and is perhaps more closely allied to As. sclera Hobbs and Hart (1966: 42.) than to any other member of the Group. The absence of a cephalic process on the peniferum of the male and the presence of a slender sinuous sclerotized process projecting ventrally from the bulbous area provide a unique combination of characters separating this species from its congeners. Etymology: This ostracod is named in honor of our mutual friend, James L. Riopel, Director of the Mountain Lake Biological Station of the University of Virginia, who, for a number of years, has encouraged us in our studies of the entocytherids. Dactylocythere apheles, new species (Figure le-h) Male: Eye pigmented and located between one-fifth and one-sixth shell length from anterior margin. Shell (Fig. le) subovate; greatest height slightly posterior to midlength where almost 1.4 times that at level of eye; ventral margin weakly convex. Submarginal setae some- what closer together anteriorly and posteriorly than ventrally, and absent dorsally. Sternal spine lacking. Copulatory complex (Fig. lg, h) with slender, distally tapering finger guard provided with anteriorly eccentric tip; peniferum moder- ately slender with ventral part curved anteroventrally and tapering rather rapidly to acute anteroventral angle; ventral part of tip corneous. Peniferal groove opening anteroventrally, almost closed in apical region, its least diameter approximately one-tenth that of corresponding di- ameter of vertical ramus of clasping apparatus. Penis L-shaped with arms subequal in length. Accessory groove represented by short tri- angular clear area posterodorsal to base of penis, its apex barely or not reaching level of spermatic loop. Clasping apparatus C-shaped with short proximal area subparallel to longer distal segment (horizontal ramus); apparatus of almost uniform diameter throughout and with margins entire; distal part with distally radiating grooves extending toward 5 or 6 apical denticles. Dorsal and ventral fingers comparatively slender, latter about twice length of former and gently curved with distal half directed anteriorly. Triunguis Female: Eye pigmented, situated as in male. Shell (Fig. lf) strongly arched dorsally and with almost straight ventral margin; New entocytherid ostracods 397 greatest height slightly posterior to midlength where about 1.5 times that at level of eye. Submarginal setae disposed as in male. Posterior margin of shell entire, smoothly rounded. Genital complex consisting of weakly sclerotized, slender, tubular papilla directed posteroventrally; usual J-shaped rod and amiculum lacking. Measurements (in mm): 3 males and 4 females. Holotype Males Allotype Females Length (range) 0.41 0.40-0.41 0.39 0.36-0.39 Average 0.40 0.38 Height (range) 0.22 0.21-0.22 0.21 0.21-0.22 Average 0.22 0.21 Type-locality: Spring seep flowing across old road to Chimneys at Indian Gap (Little Pigeon River drainage), Sevier County, Tennessee. There the type-series was collected with Uncinocythere zancla Hobbs and Walton (1963:456) on 13 November 1971, by R. W. Bouchard and J. D. Way. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), numbers 155319 and 155320, respectively. Paratypes are in the collection of H. H. Hobbs III and the Smithsonian Institution. Range and Specimens Examined: Seven specimens collected at the type-locality (Little Pigeon River basin). Host: Cambarus (J.) carolinus (Erichson). Entocytherid Associate: Uncinocythere zancla. Relationships: Dactylocythere apheles has its closest affinities with Dactylocythere leptophylax (Crawford, 1961:238). The male of both species possesses a peniferum with a reduced accessory groove, at least occasionally hardly discernible; the margins of the clasping ap- paratus are entire, and the distal extremity bears at least 4 denticles set off proximally by distally diverging striae. The females of both species lack the usual J-shaped rod and amiculum. Members of the new species differ from those of Dt. leptophylax in possessing a simple as opposed to a distally bifid or trifid finger guard; the ventral extremity of the peniferum is smooth, lacking tuberculiform prominences, and the distal part of the clasping apparatus is not expanded. Etymology: Apheles, G. = smooth; referring to the absence of teeth or prominences on the preaxial border of the clasping apparatus and on the ventral margin of the peniferum of the copulatory complex of the male. Dactylocythere brachydactylus, new species (Figure 2a—d) Male: Eye pigmented and situated about one-fifth shell length from anterior margin. Shell (Fig. 2c) subovate with greatest height about 398 Proceedings of the Biological Society of Washington one-third shell length from posterior margin where approximately 1.3 times height at level of eye; ventral margin straight to weakly convex. Submarginal setae present around entire shell, slightly closer to- gether anteriorly and posteriorly than dorsally or ventrally; other setae widely scattered over surface of shell. Sternal spine lacking. Copulatory complex (Fig. 2b, d) with short, robust finger guard produced in bifurcate extension from cephaloventral side and with posterior angle at level of base of extension; peniferum comparatively heavy with ventral portion subtruncate and apical part directed antero- ventrally. Peniferal groove conspicuous, its least diameter from one- to three-fourths as wide as least diameter of vertical ramus of clasping apparatus. Penis L-shaped with rami subequal in length. Accessory groove slender and long, reaching dorsally distinctly beyond level of dorsal extremity of spermatic loop. Clasping apparatus more L- than C-shaped, rami indistinctly delimited; postaxial margin of vertical ramus with slight angle near midlength, and preaxial margin of hori- zontal ramus with 1 or 2 minute teeth near 3 terminal denticles and marked by oblique striae extending proximally from teeth; remaining borders of apparatus entire. Dorsal and ventral fingers slender, former about one-half length of gently and evenly curved ventral finger, latter directed anteriorly. Triunguis Female: Eye pigmented, situated between one-fifth and one-sixth shell length from anterior margin. Shell (Fig. 2a) much more highly vaulted than in male, highest about one-third shell length from posterior margin where approximately 1.5 times height at level of eye; ventral margin slightly concave near midlength. Disposition of setae on shell similar to that on male. Posterior margin of shell entire and rounded. Genital complex consisting of slender, non- corneous tubular prominence, directed posteroventrally, flanked poste- rodorsally by similarly directed inflated sclerotized projection capped by hyalin coat and with heterogeneous material adhering to antero- ventral margin; usual J-shaped rod and amiculum lacking. Measurements (in mm): 10 males and 10 females. Holotype Males Allotype Females Length (range) 0.47 0.44-0.47 0.49 0.47-0.49 Average 0.46 0.48 Height (range) 0.27 0.25-0.27 0.29 0.27-0.29 Average 0.25 0.29 Type-locality: Goose Creek at State Route 66 southwest of Dand- ridge, Jefferson County, Tennessee. There the host was Cambarus (C.) bartonii (Fabricius); specimens of Orconectes virilis (Hagen) were also collected in this locality and may have been infested with New entocytherid ostracods 399 this ostracod. No other ostracod was infesting the host. Collections were made there on 16 March 1969 by R. W. Bouchard. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), numbers 155321 and 155322, respectively. Paratypes are in the collections of H. H. Hobbs III, D. J. Peters, and the Smithsonian Institution. Range and Localities: Little Tennessee, French Broad, Nolichucky, and Tennessee river basins in Cocke, Jefferson, and Roane counties, Tennessee. TENNESSEE: Cocke County—(1) Jake Best Creek on road S from Citico Creek, SE of Acorn (Host, Cambarus (C.) bartonii (Fabricius) ). Jefferson County—(2) Type-locality; (3) Long Creek at State Rte. 32 and U.S. Hwy. 25 E, N of White Pine (Host, Cambarus (Hiaticam- barus) longirostris Faxon); (4) Spring Creek on U.S. Hwy. 70 W of junction with State Rte. 113 near Oak Grove (Host, Cambarus (C.) bartonii); (5) Embayment of Spring Creek off U.S. Hwy. 70 and State Rte. 9, near Douglas Lake (Hosts, Cambarus (C.) bartonii, Orconectes virilis (Hagen), Orconectes forceps (Faxon)). Roane County—(6) Caney Creek at Buttermilk Road, off I-40 SE of Brad- bury (Hosts, C. (C.) bartonii, C. (H.) longirostris). Collections from localities 1-4 were made by R. W. Bouchard, that from 5 by D. A. Etnier, and that from 6 by Bryant. Hosts: Although this ostracod was found in collections of crayfishes containing Cambarus (C.) bartonii, C. (H.) longirostris, Orconectes forceps, and O. virilis, it is probable that it is actually associated with only the former two. Entocytherid Associates: In the Cocke County locality it was asso- ciated with Donnaldsoncythere donnaldsonensis (Klie, 1931:334); in the Long Creek locality in Jefferson County, it was found with Un- cinocythere simondsi (Hobbs and Walton, 1960:17); and in the Roane County locality, with Dactylocythere falcata (Hobbs and Walton, 1961:379). Relationships: Dactylocythere brachydactylus has its closest affinity with Dt. chelomata (Crawford, 1961:242). The similarities in the genitalia of the males are striking; however, members of the former possess a much longer accessory groove in the peniferum, one ap- proaching the length of that in Dt. mecoscapha (Hobbs and Walton, 1960:19) and Dt. macroholca Hobbs and Hobbs (1970:6). In addi- tion, the denticles along the distal preaxial border of the horizontal ramus of the clasping apparatus are not nearly so well-developed. The females of the two species may be distinguished by the absence of a J-shaped rod and amiculum in the genital complex of Dt. brachydacty- lus. Etymology: Brachys G. = short + Dactylus, G. = finger; refer- ring to the comparatively short finger guard on the copulatory complex of the male. 400 Proceedings of the Biological Society of Washington Fic. 2a—d, Dactylocythere brachydactylus, new species; 2e-h, Dacty- locythere demissa, new species. a, e, Shell of female; b, h, Clasping apparatus of male; c, f, Shell of male; d, g, Copulatory complex of male. Scales in mm. Dactylocythere demissa, new species (Figure 2e-h ) Male: Eye pigmented and situated about one-fourth shell length from anterior margin of shell. Shell (Fig. 2f) elongate ovate (about 1.8 times as long as high) with greatest height about one-third shell length from posterior margin where about 1.3 times height at level New entocytherid ostracods 401 of eye; ventral margin almost straight. Submarginal setae present ex- cept dorsally, closer together anteriorly and posteriorly than ventrally. Sternal spine absent. Copulatory complex (Fig. 2g, h) with robust finger guard pro- duced in bifurcate extension from cephaloventral side and with poste- rior angle at level of base of extension; peniferum comparatively stout with ventral portion truncate, anteroventral angle acute and directed anteriorly. Peniferal groove distinct, rather uniformly narrow with least diameter about one-fifth that of least diameter of vertical ramus of clasping apparatus. Penis somewhat L-shaped with distal ramus hardly more than half length of proximal ramus. Accessory groove short, not nearly reaching ventral extremity of spermatic loop, with end adjacent to penis forming inverted (ventrally directed) loop—unique in the genus in this respect. Clasping apparatus L-shaped with postaxial border distinctly angular at junction of horizontal and vertical rami; vertical ramus with postaxial border concave but otherwise entire, its preaxial border almost straight to broadly but shallowly concave; horizontal ramus with postaxial border entire, its preaxial border with 2 small teeth on distal half, and apex of ramus with 3 anterodorsally directed denticles. Dorsal finger comparatively heavier than very slender ventral finger and almost one-half as long; latter gently curved throughout most of its length and with distal portion directed anteriorly. Triunguis Female: Eye as in male. Shell (Fig. 2e) distinctly more highly vaulted than in male (about 1.5 times as long as high) with greatest height approximately one-third shell length from posterior margin where about 1.3 times height at level of eye; ventral margin slightly concave anterior to midlength. Disposition of submarginal setae on shell similar to that on male but also with widely spaced ones dorsally. Posterior margin of shell often with eccentric prominence in vicinity of amiculum. Genital complex consisting of J-shaped rod with long vertical arm inclined subparallel to posterodorsal margin of shell and with conspicuous but short amiculum borne on curved portion of rod; amiculum with short supportive hyaline arcs and often slightly protruding from between caudal margin of valves of shell. Measurements (in mm): 10 males and 10 females. Holotype Males Allotype Females Length (range) 0.42 0.40-0.43 | 0.42 0.40-0.43 Average 0.41 0.42 Height (range ) 0.22 0.21-0.24 0.25 0.23-0.26 Average 0.23 0.25 Type-locality: Pokepatch Creek at County Road 4385, southwest of Pleasant Hill, Cumberland County, Tennessee (Caney Fork River drainage). Specimens were obtained from a collection of crayfishes (containing Cambarus (Veticambarus) pristinus Hobbs, Cambarus (Jugicambarus) parvoculus Hobbs and Shoup, and Cambarus (De- 402 Proceedings of the Biological Society of Washington pressicambaius) sphenoides Hobbs) made by R. W. Bouchard on 7 August 1969. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution) number 155323. Paratypes are in the collection of H. H. Hobbs II, J. D. Peters, and the Smithsonian Institution. Range and Localities: Caney Fork and Obey (Cumberland River), Emory, and Tennessee river basins on the Cumberland Plateau. TENNESSEE: Bledsoe County—(1) McGill Creek at County Rd. 5881 S of Brayton (Hosts, C. (D.) sphenoides Hobbs, C. (J.) parvoculus Hobbs and Shoup); (2) Roaring Creek off County Rd. 5881, SW of New Harmony (Hosts, C. (D.) sphenoides, C. (J.) parvoculus); (3) Henderson Creek at County Rd. 5881 NE of Summer City (Hosts, C. (D.) sphenoides, C. (J.) parvoculus); (4) Moccasin Creek at County Rd. 5881 SW of Milo (Host, C. (J.) parvoculus); (5) Glade Creek at State Rte. 30, NW of Pikeville (Host, C. (J.) parvoculus). Cumber- land County—(6)Type-locality; (7) Caney Fork River at U.S. Hwy. 70 (Hosts, C. (D.) sphenoides, C. (V.) pristinus Hobbs); (8) Clear Creek at County Rd. 4794, 3 mi W of junction with U.S. Hwy. 127 (Host, C. (J.) crinipes Bouchard); (9) No Business Creek at U.S. Hwy. 127, N of Isoline (Hosts, C. (D.) sphenoides, C. (J.) distans Rhoades); (10) Litthe Obed River at U.S. Hwy. 127, N city limits of Crossville (Hosts, C. (D.) sphenoides, C. (J.) distans, C. (J.) par- voculus); (11) Daddy’s Creek at U.S. Hwy. 127 near Big Lick (Host, C. (J.) distans); (12) White Oak Creek, 3.9 mi E of White County line and 0.1 mi S of U.S. Hwy. 70 (Hosts, C. (D.) sphenoides, C. (V.) pristinus); (13) Fox Creek off County Rd. 4252 (Catoosa Wild- life Management Area) (Hosts, C. (J.) distans Rhoades, C. (J.) parvoculus); (14) South Fork of Elmore Creek at County Rd. 4252 (Hosts, C. (D.) sphenoides, C. (J.) distans, C. (J.) parvoculus); (15) Caney Fork River at County Rd. between U.S. Hwy. 70 and 7O0N, NE of Pleasant Hill (Hosts, C. (D.) sphenoides, C. (J.) parvoculus, C. (V.) pristinus). Grundy County—(16) Firescald Creek at County Rd. 4398, in Altamont (Hosts, C. (D.) sphenoides); (17) Piney Creek at State Rte. 108, at Altamont (Host, C. (D.)- sphenoides). Putnam County—(18) Dripping Springs Creek at State Rte. 62, SE of Monterey (Hosts, C. (D.) sphenoides, C. (J.) obeyensis). Rhea County—(19) Whites Creek at State Rte. 68, NW of Grand View (Hosts, C. (D.) sphenoides, C. (J.) parvoculus). White County —(20) Pole Bridge Branch at County Rd. 4385, S of DeRossett (Hosts, C. (D.) sphenoides, C. (J.) parvoculus). All of the specimens from the above localities, except stations 8, 12, and part of those from 6 were collected by R. W. Bouchard or R.W.B. and J. D. Way. Hosts: As may be noted above, in most of the localities this os- tracod was associated with Cambarus (D.) sphenoides, frequently with C. (J.) parvoculus, also with C. (V.) pristinus in the Caney Fork drainage, and occasionally with C. (J.) distans. New entocytherid ostracods 403 Entocytherid Associates: The entocytherids most frequently sharing the hosts with Dactylocythere demissa are Donnaldsoncythere don- naldsonensis, and Dactylocythere brachystrix Hobbs and Walton (1966:2). Occasionally accompanying it are Dt. pachysphyrata Hobbs and Walton (1966:3), Dt. speira Hart and Hart (1971:113), and Entocythere sp. In one locality each, it was found in association with Dt. arcuata (Hart and Hobbs, 1961:173) and Dt. spinata Hobbs and Walton (1970:853). Relationships: One of the most distinctive features of this ostracod is the very short accessory groove of the peniferum, which, in its length, is similar to that in Dactylocythere coloholca Hobbs and Hobbs (1970:7), Dt. exoura Hart and Hart (1966:5), and Dt. speira. In other respects, it also has as much in common with these species as with any of its congeners. In none of the three, however, is the finger guard of the copulatory complex produced in a bifid tip. While De. demissa appears to be more closely allied to Dt. speira than to the other two in possessing a similar clasping apparatus and a rather broad accessory groove, the peniferum of the former is truncate ven- trally, the accesscry groove is disposed in a distinct ventral loop, and the J-shaped rod of the genital complex of the female has an almost straight shaft with a gently curved ventral part rather than being strongly curved throughout, almost forming a spiral. In addition, Dt. demissa possesses a distinct amiculum that is lacking in Dt. speira. Etymology: Demissus, L. = hanging down; so named because of the long, almost straight J-shaped rod in the genital complex of the female; also the looped ventral part of the accessory groove in the peniferum of the male suggests a collapse of the groove. LITERATURE CITED CrawrForp, E. A. 1961. Three new species of the genus Entocythere (Ostracoda, Cytheridae) from North and South Carolina. Amer. Mid]. Nat. 65( 1) :236-245, 21 figs. Hart, C. W., Jr.. anp Dasney G. Harr. 1966. Four new entocy- therid ostracods from Kentucky, with notes on the troglobitic Sagittocythere barri. Notulae Naturae, Acad. Nat. Sci., Philad. 388, 10 p., 13 figs. ———, AnD Horton H. Hosss, Jr. 1961. Eight new troglobitic ostracods of the genus Entocythere (Crustacea, Ostracoda) from the eastern United States. Proc. Acad. Nat. Sci., Philad. 113(8):173-185, 32 figs. Hart, Dasney G., anp C. W. Harr, Jr. 1971. New entocytherid ostracods of the genera Ankylocythere, Dactylocythere, En- tocythere, Geocythere, and Uncinocythere—with a new di- agnosis of the genus Entocythere. Proc. Acad. Nat. Sci., Philad. 123(5):105-125, 13 figs. ——, AND 1974. The Ostracod Family Entocytheridae. Acad. Nat. Sci., Philad., Monograph 18: ix + 238 p., 62 pls. 404 Proceedings of the Biological Society of Washington Hosss, Horton H., Jr., AND C. W. Harr, Jr. 1966. On the entocy- therid ostracod genera Ascetocythere, Plectocythere, Phymocy- there (gen. nov.), and Cymocythere, with descriptions. of new species. Proc. Acad. Nat. Sci., Philad. 118(2):35-61, 37 figs. ——, anp H. H. Hosss m. 1970. New entocytherid ostracods with a key to the genera of the subfamily Entocytherinae. Smithsonian Contrib. Zool. 47:19 p., 9 figs. ———,, AND Marcaret Watton. 1960. Three new ostracods of the genus Entocythere from the Hiwassee drainage system in Georgia and Tennessee. Journ. Tenn. Acad. Sci. 35(1):17- 23, 20 figs. ———., AND 1961. Additional new ostracods from the Hi- wassee drainage system in Georgia, North Carolina, and Tennessee. Trans. Amer. Micros. Soc. 80(4):379-384, 8 figs. ———,, AND 1963. Three new ostracods (Ostracoda, Entocy- theridae) from the Duck River drainage in Tennessee. Amer. Mid]. Nat. 69(2):456-461, 10 figs. ———., AND 1966. A new genus and six new species of entocytherid ostracods (Ostracoda, Entocytheridae). Proc. U.S. Nat. Mus. 119(3542):1-12, 2 figs. ———-, AND 1970. New entocytherid ostracods from Ten- nessee and Virginia. Proc. Biol. Soc. Wash. 82(68):851- 864, 3 figs. Kuz, W. 1931. Campagne spéologique de C. Bolivar et R. Jeannel dans !Amérique du Nord (1928). 3. Crustaces Ostracodes. Biospeol.: Archiv. Zool. Exp. et Gen. 71(3):333-344, 20 figs. La i. 34, pp. 405-410 12 October 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW SEA CUCUMBERS (ECHINODERMATA: HOLOTHUROIDEA) FROM THE EASTERN UNITED STATES By Davi L. Pawson Department of Invertebrate Zoology, Smithsonian Institution, Washington, D.C. 20560 During the examination of holothurians collected by the Northeast Center, National Marine Fisheries Service, Woods Hole, Massachusetts, two new species were found; they are described here. Both species will be discussed further in a report on distribution patterns of holothurians off the eastern United States (Wigley and Pawson, in prep.). I am grateful to Dr. Roland L. Wigley, National Marine Fisheries Service, for giving me access to the collections in his care, and for his help in many other ways. Type-material is deposited in the National Museum of Natural History, Smithsonian Institution. Order APODIDA CHIRIDOTIDAE Chiridota Eschscholtz, 1829 Chiridota wigleyi, new species Figure 1F—H Diagnosis: Wheel papillae very scarce, apparently restricted to dorsal interradii. Radial and interradial areas of body wall with numerous curved rods with bifurcated ends; rods average 63 wm in length. Ten- tacles with curved rods with branching ends; tentacle rods average 85 um in length. Material examined: Holotype (USNM E15904, 40 mm total length) DELAWARE cruise 62-7, station 24, 15 June 1962, 40°20’N, 70°15’W, 90 m, silty sand, bottom temperature 7.8° C. Paratypes 20 specimens (USNM E15905, 25-114 mm total length) same locality data as holotype. 34—Proc. Bro. Soc. WasH., Vou. 89, 1976 (405 ) 406 Proceedings of the Biological Society of Washington The species was collected at 20 stations off the eastern United States, in an area bordered by latitudes 39°47’N and 40°30’N and longitudes 69°31’W and 71°46’W; bathymetric range 70-301 m; bot- tom temperature 6.6-11.6° C; bottom type sand to sandy silt to sand- silt-clay. Etymology: The species is named for Dr. Roland L. Wigley, in recognition of his many contributions to our knowledge of the marine fauna of the northwestern Atlantic. Description: Total length 25-114 mm. All specimens contracted to varying degrees, cylindrical, dark reddish-brown in alcohol. Tentacles 12 with 5-7 pairs of digits. Ossicles in body wall include wheels and curved rods. Wheels aggregated loosely into poorly defined papillae which are sparsely scattered in dorsal interradii. In one specimen of 100 mm total length only 4 wheel papillae present, containing 28, 40, 18 and 17 typical chiridotid wheels averaging 86 ym in diameter. Curved rods with bifurcated ends (Fig. 1H) scattered in radial and interradial areas of body wall. Average length of rods 63 um (range 50-80 um; standard deviation 3.73). Radial longitudinal muscles con- tain numerous elongate miliary granules which vary greatly in length up to a maximum of approximately 100 um. Granules approximately cylindrical, often slightly thickened near center (Fig. IF). Tentacle stems and digits contain rods (Fig. 1G) resembling those of body wall, but with more complex terminal branches and greater variation in size. Average length of tentacle rods 85 wm (range 55-115 um; standard deviation 8.7). Remarks: In possessing numerous rods in the body wall, this new species is immediately distinguished from other temperate North At- lantic and Arctic species Chiridota laevis (Fabricius), C. pellucida (Vahl), C. spirourna Heding, C. groenlandica Heding and C. abyssi- cola von Marenzeller, all of which lack such rods (Heding 1928, 1935). The only other known North Atlantic species, C. rotifera (Pourtales), has smaller rods in the body wall (length averaging 30 um), and also has very numerous wheel papillae scattered in all radii. C. rotifera is tropical, occurring in shallow water in Bermuda and the West Indies (Clark, 1933). Order DENDROCHIROTIDA CUCUMARIIDAE Ocnus Forbes, 1841 Ocnus diomedeae, new species Figure 1A-E Diagnosis: Ossicles in body wall rudimentary cups comprising a pri- mary cross with projecting knobs and averaging 41 wm in length, over- lying knobbed plates of one type which are highly variable in shape and size; average length of plates 72 wm. Tube feet contain elongate smooth or knobbed plates. New eastern U.S. sea cucumbers 407 Fic. lA-E. Ocnus diomedeae n. sp. A, Plates from tentacles; B, Os- sicles from tube feet; C, Cups from body wall; D, One radial piece and two interradials from calcareous ring; E, Plates from body wall. F-H, Chiridota wigleyi n. sp. F, Miliary granules from radial muscles; G, Rods from tentacles; H, Rods from body wall. Material examined: Holotype (USNM E15906, 15 mm total length) ALBATROSS Iv cruise 66-9, station 1004, 15 July 1966, 42°11’N, 65°50’W, 247 m, gravel. Paratypes: 19 specimens (USNM E15907, 8-17 mm total length) same locality data as holotype; 5 specimens (USNM E15908, 9-19 mm total length) ALsBaTross Iv cruise 66-9, station 1005, 16 July 1966, 42°13’N, 65°42’W, 229 m, gravel; 12 specimens (USNM E15909, 10-17 mm total length) Axpatrross tv cruise 68-12, station 72, 15 August 1968, 42°22’N, 65°55’W, 192 m, gravel. Etymology: Species named for collecting vessel. Diomedea is generic name for the albatross. 408 Proceedings of the Biological Society of Washington Description: Body 2-3 times as long as broad, mouth and anus ter- minal. Total length 8-19 mm. Body wall thin, stiff, packed with ossicles. Tube feet restricted to radii, in 5 more or less double rows. One or 2 feet may be present in mid-dorsal interradius. Tube feet apparently only partly retractile. Tentacles 10, richly branched; ventral pair of tentacles smaller than others. Body and tube feet light orange to white in alcohol; tentacles light yellow. Calcareous ring simple (Fig. 1D) with undulating posterior margin. Retractor muscles attach to radial muscles at about middle of body. Polian vesicle single. Respiratory trees well developed, extending to anterior of body cavity. Gonad a tuft of several caeca at about mid- body. In females caeca contain yolky eggs approximately 700 um in diameter. Ossicles in body wall knobbed plates and rudimentary cups. Plates (Fig. 1E) highly variable in shape and size, averaging 72 um in length (range 120-300 um; standard deviation 14.64). Basic plate apparently 4-holed, but very few 4-holed examples seen. Cups (Fig. 1C) minute, averaging 41 um in length (range 38-44 um; standard deviation 1.81), comprising central rod with terminal bifurcation (primary cross pat- tern), each furca carrying small knobs. Tube feet with rudimentary end plates; walls of feet contain nu- merous elongate smooth or knobbed plates (Fig. 1B) which are highly variable in shape and size. Tentacles packed with curved perforated rods and small perforated plates (Fig. 1A); plates and rods usually smooth, or with few small knobs. Remarks: Rowe (1970) erected the new genera Aslia and Pawsonia to accommodate some species which had originally been referred by Panning (1949) to Ludwigia (see Pawson, 1963). Rowe concluded also that some of the species which had originally been assembled under the preoccupied genus-name Ludwigia should provisionally be re- ferred to Ocnus. These species included planci (Brandt), lactea (Forbes), glacialis (Ljungman), hedingi (Panning) and some others, and it is with this group of species, and therefore with the genus Ocnus, that O. diomedeae appears to have its strongest relationships. In possessing rudimentary cups, O. diomedeae is immediately distin- guished from most Ocnus-species. O. hedingi (Panning), O. syracusanus (Grube) and O. glacialis (Ljungman) have similarly reduced cups, but in these species the knobbed plates in the body wall are either more complex or are of a different type, and the ossicles in the tube feet are distinctly different. LITERATURE CITED Criark, H. L. 1933. A handbook of the littoral echinoderms of Porto Rico and the other West Indian islands. Sci. Surv. Porto Rico Virgin Islands N. Y. 16(1):1-60. Hepinc, $.G. 1928. Synaptidae. Vidensk. Medd. dansk naturh. Foren. Kbh. 85:105-323. New eastern U.S. sea cucumbers 409 1935. Holothurioidea. Part I. Apoda, Molpadioidea, Gephy- rothurioidea. Danish Ingolf-Exped. 4(9):1-84. Panninc, A. 1949. Versuch einer Neuordnung der Familie Cucumari- idae (Holothurioidea, Dendrochirota). Zool. Jb. 78(4):404— A470. Pawson, D. L. 1963. The holothurian fauna of Cook Strait, New Zealand. Zool. Publs. Victoria Univ. Coll. 36:1-38. Rowe, F. W. E. 1970. A note on the British species of cucumarians, involving the erection of two new nominal genera. Jour. Mar. Biol. Ass. U.K. 50:683-687. 410 Proceedings of the Biological Society of Washington i 5, pp. 411-420 12 October 1976 ae PROCEEDINGS OF THE p1stOGICAL SOCIETY OF WASHINGTON A RECLASSIFICATION OF IRIS SPECIES BEARING ARILLATE SEEDS By Joun J. TAYLOR University of Montana, Missoula, Montana 59801 In his revision of the genus Iris L., Rodionenko (1961) placed in the subgenus Iris only those species with aggrega- tions of multicellular hairs (beards) on the outer perianth segments. Beardless rhizomatous and nonrhizomatous species were placed in other subgenera, or were transferred to genera other than Iris. In his system, subgenus Iris consisted of sec- tion Hexapogon emended to include all species with arillate seeds, and section Iris containing the remaining non-arillate species. Hexapogon consisted of the subsections Regelia, Pseudoregelia and Oncocyclus. Regelia, however, had the same circumscription as subsection Hexapogon (Bunge) Bentham emend. Lawrence (1953), and Rodionenko’s use of the name Regelia was superfluous (see also under section Hexapogon below). Data published since the Rodionenko revision of the genus make even more obvious the heterogeneous nature of sub- section Hexapogon (Bunge) Bentham emend. Lawrence and its synonym Regelia (Dykes) Rodionenko. The following new system for the arillate species in Hexapogon is proposed. Tris subgenus Iris Section Iris (This type section of the subgenus lacks aril- late species. ) Section Hexapogon (Bunge) Baker Section Regelia Lynch Section Oncocyclus (Siemssen) Baker Section Psammiris (Spach) J. Taylor Section Pseudoregelia Dykes 35—Proc. Brou. Soc. WasH., Vou. 89, 1976 (411) 412 Proceedings of the Biological Society of Washington The essential features of this reclassification are (1) the segregation of species previously included in section Hexa- pogon (Bunge) Baker emend. Rodionenko (1961) into five distinct taxa, and (2) the designation of a type of the section Regelia Lynch. A compendium of literature relevant to sec- tion Iris is also included. Iris subgenus Iris Iris sect. Pogiris Tausch in Hort. Canalius, 1 (1823); idem in Schultes Additamentum Mantissum, 2:369 (1824), quoad typus.—lIris subgen. Euiris Alefeld in Bot. Zeit., 21:296 (1863), p.p.; non Boissier Fl. Orient., 5:118 (1884); non Klatt in Linnaea, 33:591-604 (1866).— Tris sect. Euiris Bentham in Bentham and Hooker, Gen. Pl., 3:687 (1882).—iris sect. Pogiris Tausch emend. Lawrence in Gent. Herb., 8(4):353 (1953).—Iris sect. Iris emend. Lawrence in Gard. Irises, 142 (1959). Type of the subgenus: I. germanica L. Species composing subgenus Iris are characterized by distinctly rhi- zomatous stems, and flowers with beards of multicellular hairs on the outer, and occasionally the inner, perianth segments. KryY TO THE SECTIONS OF SUBGENUS IRIS 1. Seed without an aril; capsule usually dehiscent very near apex, but if dehiscent considerably below apex, then some internal placen- talion muptired at Mmaburiiy 22224 ee Section Iris 1. Seed with conspicuous creamy-white aril; capsule dehiscent sub- apically or laterally; mature placentation intact —..-----_- 2 2. Both inner and outer perianth segments with conspicuous more or less dinear “beards ets. 3 3 3. Spathe valves (bracts) 3 or 4; rhizome usually compact, slowly creeping; seed with small aril; chromosome number of counted SPECIES? sm) FS Oe ee Section Hexapogon 3. Spathe valves 2; rhizome stoloniferous, readily spreading; seed with large aril; chromosome number of counted species n = LG ea eee eee eee NOM Bar yee tice, eee eee Section Regelia 2. Only outer perianth segments bearded, or if all segments bearded, then beards on outer segments scattered 4 4. Beard on outer segments scattered; spathe 1-flowered; seed with L801 a1 6 | eee a ee Re Oe reer Mee Section Oncocyclus 4. Beard on outer segments linear; spathe usually 2-flowered; seed with aril much smaller than seed —.. 5 5. Rhizome creeping or spreading by stolons; flowers essentially monochromic though sometimes lightly veined ~~. a eee He Section Psammiris Reclassification of arillate Iris 413 5. Rhizome compact, gnarled; flowers usually dichromic, spotted ERIE Ions Paneth See Be ee Section Pseudoregelia Iris section Iris Iris subgen. Pogiris Tausch, l.c.; idem in Schultes, l.c—Ivis subgen. Pogoniris Spach in Ann. Sci. Nat. ser. 3, 5:103 (1846); idem in Hist. Nat. Veg., 13:48-68 (1846), p.p.; Klatt in Bot. Zeit., 30:515 (1872), p.p.; Baker in Journ. Linn. Soc. Lon. Bot., 16:143 (1877), p.p.; Randolph in Bull. Amer. Iris Soc., 109:4 (1948), p.p.—lIris subgen. Euiris Alefeld in Bot. Zeit., 21:296 (1863), p.p.; non Boissier, l.c.; non Klatt, /.c—lIris subsect. Pogoniris (Spach) Bentham in Bentham and Hooker, l.c.; Pax in Engler and Prantl, Nat. Pflanzenfam., 2(5):145 (1888), p.p.; Lawrence, l.c.—Iris sect. Pogoniris Baker, l.c., p.p.— Iris subgen. Eupogoniris Randolph, l.c. Type of the section: I. germanica L. In his reclassification of Iris, Lawrence (1953, 1959) divided sub- section Pogoniris into the series Pumilae and Elatae. To the former were assigned those species which are less than 3 dm tall, and either acaulescent or if caulescent, then not branched. Series Elatae com- prised the taller species which are distinctly caulescent and branched. Although he cited these taxa, Rodionenko did not include them in his system. Recent collections of bearded irises have included forms which could justify emending series Pumilae to include the low-growing species which are nevertheless caulescent and branched, e.g., I. furcata MB, I. timofejewii Woron., and perhaps others. Anyone systematically treat- ing these non-arillate species should include such emendation if the taxon is retained. Because of the numerous morphological, karyological, genetic and distributional differences between the arillate and the non-arillate spe- cies, Rodionenko (1961) removed arillate species from section Iris, based on I. germanica L., and gave them sectional rank collectively in Hexapogon. His circumscription of section Iris is retained here. Iris section Hexapogon (Bunge) Baker in Gard. Chron. ser. 3, 5:787- 788 (1876); Boissier, Fl. Orient., 5:119 (1884). Iris subgen. Hexapogon Bunge ex Alefeld in Bot. Zeit., 21:296 (1863), nom. illegit.; Baker in Journ. Linn. Soc. Lon. Bot., 16:147 (1877), nom. illegit—Iris subsect. Hexapogon (Bunge) Bentham in Bentham and Hooker, Gen. Pl., 3:687 (1882); Pax in Engler and Prantl, Nat. Pflanzenfam., 2(5):145 (1888). Type of the section: I. falcifolia Bunge. Species included within the section: I. falcifolia Bunge 1847 I. longiscapa Ledebour 1853 414 Proceedings of the Biological Society of Washington The name Hexapogon, first used by Bunge (1847) in describing the new species I. falcifolia and I. filifolia Bunge (an invalid synonym of I. longiscapa Ledeb.), was taken up by Alefeld (1863) for a sub- genus circumscribing the two Bunge species and I. susiana L. Because I. susiana was the type for Spach’s earlier subgenus Susiana (1846), Alefeld’s use of Hexapagon was illegitimate (Article 63, International Code of Botanical Nomenclature (ICBN), 1969). Baker (1876) trans- ferred I. susiana to section Oncocyclus and first validly published Hexapogon as a section comprising I. falcifolia and I. filifolia Bunge only. (The following year (1877) he raised Hexapogon to the rank of subgenus, but his use of the name at that rank was illegitimate (Article 64, ICBN).) In 1892, Baker transferred the Hexapogon species to subgenus Pogoniris, and introduced the name Regelia for a subgenus containing three species described by Eduard Regel. Although it is apparent that Baker intended to segregate Regelia from Hexapogon and Pogon- iris, he did not adequately characterize Regelia, and it remained a nomen nudum until validly published as a section by Lynch in 1904. A type was not designated. Dykes considered I. falcifolia and I. filifolia Bunge to be synonymous, and transferred them (1913) from Pogoniris to section Regelia. Be- cause one or the other of these Hexapogon species must be considered the type for the earlier section Hexapogon (Bunge) Baker, Dykes’ use of the name Regelia was illegitimate (Article 63, ICBN). Rodion- enko’s use of Regelia for a subsection (1961) was illegitimate for the same reason. Lawrence did not acknowledge section Regelia Lynch, and combined Regelia Dykes, including Hexapogon, with Psammiris species in subsection Hexapogon (1953, 1959). I. falcifolia and I. longiscapa are xeritic species of the Turkmenian and Uzbekian deserts in southcentral U.S.S.R., and of a few similar but restricted localities in Iran and Afghanistan. They are characterized by weakly or non-stoloniferous rhizomes, very narrow leaves, thin and leafless stems, and spathes of 3 or 4 bracts enclosing 2 to 5 small flowers with all perianth segments bearded. Iris longiscapa has been examined karyologically (Randolph and Mitra, 1961), and is the only bearded iris species yet counted with the diploid chromosome number 18. Iris section Regelia Lynch, Bk. Iris, 56 (1904); Fedtschenko in Ko- marov, Fl. USSR, 4:539 (1935); nec Dykes, Gen. Iris, 123 (1913), p.p., nom. illegit. Iris subgen. Regelia M. Foster ex Baker, Handbk. Irideae, 1 (1892), p.p., nom. nudum.—Iris subgen. Pogoniris Randolph in Bull. Amer. Iris Soc., 109:4 (1948), p.p. affin. sect. Regelia Lynch; non Spach (1846), lc.; non Baker (1876), l.c—Iris subsect. Regelia (Dykes) Rodionenko, Gen. Iris L., 198-199 (1961), p.p., nom. illegit. Reclassification of arillate Iris 415 Type of the section: I. korolkowii Regel. Species included within the sectiton: I. afghanica Wendelbo 1972 . darwasica Regel 1884 . heweri Grey-Wilson & Mathew 1974 . hoogiana Dykes 1916 . korolkowii Regel 1873 . kuschkensis Grey-Wilson & Mathew, ? edit. . lineata M. Foster 1887 . stolonifera Maximowicz 1880 Le ee te Se Be Ee | To my present knowledge, a type of section Regelia as circumscribed by Lynch has not previously been designated. After a study of that author’s protologue, and in an attempt to preserve both the originally intended and the current usage of Regelia (Article 7 B, ICBN), I have selected and here designated I. korolkowii Regel as the type of this section. The Regelia species are essentially montane species characterized by more or less stoloniferous rhizomes bearing unbranched scapes each with a 2-bracted spathe containing 2 (rarely 3) flowers with all perianth segments bearded. They differ from all other arillate sections except Oncocyclus, with which they show the greatest affinity, in general distribution, plant habit, spathe and/or floral morphology, and, except for chromosome number in Psammiris and Pseudoregelia, in karyotype. Intersectional hybrids between Regelia and the remain- ing arillate sections (except Oncocyclus) have been difficult or im- possible to obtain. The rare hybrids produced have been sterile. In contrast, there is marked fertility in intersectional diploid hybrids between Regelia and Oncocyclus, and, excepting the large metacentric chromosome of Regelia species, a striking similarity of haploid karyo- types in the two sections. The taxa are differentiated, however, by the weakly or non-stoloniferous rhizomes, 1-flowered spathes, and widely scattered beards on the outer perianth segments of Oncocyclus species. Iris section Oncocyclus (Siemssen) Baker in Gard. Chron. ser. 3, 5:788 (1876). Oncocyclus Siemssen in Bot. Zeit., 4:706-707 (1846).—Iris subgen. Susiana Spach in Ann. Sci. Nat. ser. 3, 5:110 (1846); idem in Hist. Nat. Veg., 13:70-71 (1846).—Iris subgen. Oncocyclus (Siemssen) Alefeld in Bot. Zeit., 21:296 (1863); Baker in Journ. Linn. Soc. Lon. Bot., 16:142 (1877).—Iris subsect. Oncocyclus (Siemssen) Bentham (1882), l.c.; Pax (1888), l.c.—lIris subgen. Pogoniris Randolph in Bull. Amer. Iris Soc., 109:4 (1948), p.p. affin. gen. Oncocyclus Siemssen; non Spach (1846), l.c.; non Baker (1876), l.c. Type of the section: I. paradoxa Steven. 416 Proceedings of the Biological Society of Washington Species included within the section: I. acutiloba C. A. Meyer 1831 . antilibanotica Dinsmore 1933 . atrofusca Baker 1894 . atropurpurea Baker 1889 . auranitica Dinsmore 1933 . barnumiae M. Foster & Baker 1888 . biggeri Dinsmore 1933 . bismarckiana (Dammann) Regel 1890 . bostrensis Mouterde 1954 . camillae Grossheim 1950 . calcaria Dinsmore inedit. . cedretii Dinsmore ex Chaudhary 1972 . damascena Mouterde 1967 . demawendica Bornmueller 1902 . ewbankiana M. Foster 1901 . gatesii M. Foster 1890 . grossheimii Woronov 1928, ? hyb. nat. . hauranensis Dinsmore 1933 . haynei (Baker) Mallet 1904 . heylandiana Boissier & Reuter 1877 . hermona Dinsmore 1933 . iberica Hoffmann 1808 . kirkwoodii Chaudhary 1972 . lineolata (Trautvetter) Grossheim 1950 . lortetii Barbey 1881 . lupina M. Foster 1887 . lycotis Woronov 1915 . maculata Baker 1876 . manissadjanii Freyn 1896 . meda Stapf 1885 . nigricans Dinsmore 1933 . paradoxa Steven 1844 . petrana Dinsmore 1933 . polakii Stapf 1885, p.p. . sari Schott 1876 . schelkownikowii Fomin 1907, ? hyb. nat. . schischkinii Grossheim 1950 . sofarana M. Foster 1899 . sprengeri Siehe 1904 . susiana Linnaeus 1753 . urmiensis Hoog 1900 . yebrudii Dinsmore ex Chaudhary 1972 a a a a a a ee ee Of the arillate irises, the section Oncocyclus is the most refractory to systematic treatment. The situation has resulted from the numerous intraspecific phenotypic variations among populations and _ collections Reclassification of arillate Iris A417 of apparently valid species, the frequency of natural hybridization among sympatric species and the use of specific epithets for hybrids and their nothomorphs, and variations in interpretation and use of differentiating criteria among taxonomists. It is possible, therefore, that the list immediately above lacks the names of valid Oncocyclus species, and includes names which may be synonymous with others. It should be considered a tentative listing in lieu of more thorough examinations of putative species and hybrids, and of systematic anal- yses based on characteristics perhaps somewhat more instructive of this section than gross morphology and habitat alone. The Oncocyclus species are found in dry desert and montane habitats from the eastern Mediterranean coastal region east and northeast into Iran, Afghanistan and Turkmenian U.S.S.R. They are characterized by weakly or non-stoloniferous rhizomes, more or less falcate radical leaves, and a scape bearing a 2-bracted spathe containing a single flower. Both inner and outer perianth segments vary remarkably in outline among the species, from the much reduced sepals of I. para- doxa, the type for the section, to the narrow and acutely pointed seg- ments of the I. meda-I. acutiloba complex and the wide rounded perianth of I. susiana and similar cultivated species. All, however, possess a beard of multicellular hairs scattered widely and often densely along the haft and onto the blade of the sepal. Iris section Psammiris (Spach) Taylor comb. nov. Tris subgen. Psammiris Spach in Ann. Sci. Nat. ser. 3, 5:110 (1846); idem in Hist. Nat. Veg., 8:69-70 (1846). Type of the section: I. humilis Georgi. Species included within the section: I, bloudowii Bunge ex Ledebour 1833 I. humilis Georgi 1775 ? I. mandschurica Maximowicz 1880 I. potaninit Maximowicz 1880 Because there has been some reluctance to reject I. flavissima Pallas as a later synonym for I. humilis Georgi, a brief nomenclatural history of this species is included here. Messerschmidt found a low-growing yellow iris near the Trans- baikalian town of Ulan-Ude. Its description (“Iris humilis angusti- folia. .... ”) was first published by Ammann (1739) after number 133, page 101. Gmelin (1747) described after his number 31 “iris foliis ensiformi- bus, caule bifloro. Tab. V. Fig. 2” and included as a synonym “Iris humilis angustifolia. .... Mess. Amm. (Stirp. rar.) Ruth. post 133.” Gmelin stated that his number 31 (as Ammanni) had also been ob- served by Messerschmidt in the hills near Ulan-Ude. It is obvious that Gmelin considered his number 31 to be the same as Messerschmidt- Ammann number 133. 418 Proceedings of the Biological Society of Washington Georgi (1775) again observed the iris in the southern Baikal region, and published the following description of it under the name I. humilis: “Tris humilis, angustifolia, lutea. Messerm. Amman. (Stirp. rar.) p. 101. Radix flavo alba fibrosa. Caulis 3. ad 6. pollicum, saepius biflorus, foliis gramineis, duplo longioribus.” It is apparent that Georgi con- sidered his I. humilis to be the same as Messerschmidt-Ammann number 133 (the only “iris humilis angustifolia” on page 101), and his diag- nosis is compatible with Gmelin’s figure 2, plate V. In 1776, Pallas published the name I. flavissima, and included as a synonym: “Tris foliis ensiformibus. ... . Gmel. Flor. Sibr. I. p. 31. tab. V. fig. 2. cum synon. Ammanni.” Thus I. flavissima is the same as Gmelin number 31 and therefore Messerschmidt-Ammann number 133, and I. humilis Georgi, also conspecific with the latter, is the first validly published name for this species. Spach (1846) first published Psammiris as a monospecific subgenus based on I. arenaria Waldst. & Kit., a later synonym of I. humilis Georgi. Baker (1877) later combined psammirises with Pogoniris, where they remained until Lawrence (1953) transferred them to subsection Hexa- pogon. Although most forms of I. humilis have more or less stolon- iferous rhizomes and radical leaves which are tinged with anthocyanin pigments at vernal emergence (characteristics sometimes assumed to show affinity with common garden forms of some Regelia species), the psammirises differ from other arillate species in karyotype, general distribution range, plant habit and/or floral and rhizome morphology. The psammiris karyotype (Simonet, 1934; Gustafsson and Wendelbo, 1975) is quite distinct among arillate irises, unlike even those of Regelia and Pseudoregelia species with similar chromosome numbers (Randolph and Mitra, 1961; Zakhariyeva and Makushenko, 1969). The few intersectional hybrids produced experimentally between Psammiris and other arillate species are sterile. Psammirises are found in open meadows and on hillsides from south- central Europe east into transcaucasian U.S.S.R., and on exposed mountain slopes and in grasslands and shaded, dry mountain valleys from the western Altai region north and east into Mongolia, Manchuria and transbaikalian U.S.S.R. They are the most widely distributed of the arillate species, and the only arillate irises native to central Europe. Although there are some differences in flower stalk length and rhizome morphology among Psammiris species and geographical forms of species, all produce narrow radical leaves which frequently dry to leave fibrous vestiges near the rhizome, and thin scapes bearing 1 or 2 2-bracted spathes each with 1 or 2 short-lived flowers with more or less elongated perianth tubes. The withering flowers become charac- teristically helically twisted (Spach, 1846; Ugrinsky, 1922). Iris section Pseudoregelia Dykes, Gen. Iris, 129 (1913). Iris subgen. Pseudoevansia Baker, Hndbk. Irideae, 2 (1892), nom. nudum.—lIris sect. Pseudoevansia Lynch, Bk. Iris, 55 (1904), nom. Reclassification of arillate Iris 419 nudum.—lIris subsect. Pseudoregelia (Dykes) Lawrence in Gent. Herb., 8(4):356 (1953); Rodionenko, Gen. Iris L., 199 (1961). Type of the section: I. kamaonensis Wallich ex D. Don. Species included within the section: I. goniocarpa Baker 1876 I. hookerana M. Foster 1887 I. kamaonensis Wallich ex D. Don 1841 PI. leptophylla Lingelsheim 1922 I, sikkimensis Dykes 1908, ? hyb. nat. I. tigridia Bunge ex Ledebour 1829 Baker (1892) created the subgenus Pseudoevansia to include several small central Asian species which he believed were related to the crested Evansia species. His differential diagnosis was based exclu- sively on the rudimentary crests terminating the beard of pseudoevansias, structures now known to have little taxonomic significance. Lynch (1904) retained Pseudoevansia as a section, but failed to describe the taxon further. The name remained a nomen nudum. Dykes (1913) examined Pseudoevansia, transferred some included species to other taxa, and fully characterized and published the name Pseudoregelia as a section circumscribing the remaining species. Both Lawrence (1953, 1959) and Rodionenko (1961) retained the name for a sub- section with the same circumscription. The pseudoregelias are essentially montane often alpine species found from the Indian Himalayas north and east into Nepal, Tibet, Mongolia and eastern Siberia, and west into the Altai region. They superficially resemble Psammiris species in dwarf plant habit and floral mor- phology, but differ significantly in rhizome morphology and in karyotype (Simonet, 1952). The known Pseudoregelia species exhibit no close affinities with other arillate irises. Fertile intersectional hybrids involving pseudoregelias have not been produced. LITERATURE CITED ALEFELD, F. G. C. 1863. Ueber die Gattung Iris L. Bot. Zeit. 21: 296-298. AMMANN, J. 1739. Stirpium rariorum in Imperio Rutheno sponte proventientium icones et descriptiones collectae ab Ioanne Ammano. Acad. Sci. St. Petersburg, 210 pp. Baker, (E.) J. G. 1876. A synopsis of the known species of Iris. Gard. Chron. ser. 3, 5:787-788. 1877. Systema Iridacearum. Journ. Linn. Soc. Lon. Bot. 16:142-147. 1892. Handbook of the Irideae. George Bell & Sons, Lon- don, 247 pp. BuncE, A. von. 1847. Beitrag zur Kenntniss der Flor Russlands und 420 Proceedings of the Biological Society of Washington der Steppen Central-Asiens. Mem. Sav. Etr. Petersb. 7:177— 536. Dykes, W. R. 1913. The genus Iris. University Press, Cambridge, 245 pp. Georci, J. G. 1775. Bemerkungen einer Reise in Russichen Reiche im Jahre 1772. 1:196. Ge Lin, J. G. 1747. Flora Sibirica sive historia plantarum Sibiriae. 13h, GustaFsson, M., aNnpD P. WENDELBO. 1975. Karotype analysis and taxonomic comments on irises from SW and C Asia. Bot. Notiser. 128:208-226. LAwRENCE, G. H. M. 1953. A reclassification of the genus Iris. Gent. Herb. 8(4) :346-371. —. 1959. Garden irises. The American Iris Society, St. Louis, 575 pp. Lyncu, R. I. 1904. The book of the iris. J. Lane, London and New York, 214 pp. Patuias, P. S. 1776. Reise durch verschiedene Provinzen des rus- sichen Reichs. 3:716. RanboupH, L. F., anp J. Mirra. 1961. Karyotypes of Iris species indigenous to the USSR. Amer. Journ. Bot. 48:862—870. RopionENKo, G. 1961. The genus Iris L. Acad. Sci. USSR, Moscow, 215 pp. (in Russian). ReceL, E. 1873. Iris korolkowii Rgl. Gartenfl. Deutsch. Russ. und der Schweiz 8:225-226, Tab. 766. SIMONET, M. 1934. Nouvelles recherches cytologiques et genetique chez les Iris. Ann. Sci. Nat. Bot. 10° serie 16:229-383. — —. 1952. Nouveaux denombrements chromosomiques chez les Iris. Compt. Rend. Acad. Sci. (Paris) 235:1244-1246. Spacu, E. 1846. Revisio generis Iris. Ann. Sci. Nat. ser. 3, 5:89- HUE Ucrinsxy, K. A. 1922. Die gesamtart Iris flavissima Pall. Eine monographische studie. Fedde’s Repert. Beiheft. 14:14. ZAKHARIYEVA, O. I., AND L. M. MaxusHenxo. 1969. Chromosome numbers of monocotyledons belonging to the families Lili- aceae, Iridaceae, Amaryllidaceae and Araceae. Bot. Zhur. 54: 1213-1227 (in Russian). at sabes 6, pp. 421-432 12 October 1976 alts ba ot int tet Het ae PROCEEDINGS OF THE pIULOGICAL SOCIETY OF WASHINGTON AFFINITIES OF PARANIPHARGUS LELOUPARUM MONOD, A BLIND ANCHIALINE AMPHIPOD (CRUSTACEA) FROM THE GALAPAGOS ISLANDS By J. LAuRENS BARNARD Department of Invertebrate Zoology, Smithsonian Institution Washington, D.C. 20560 Paraniphargus lelouparum Monod, 1970, is the first blind hypogean amphipod known from the Galapagos Islands. A new genus is described for this species to recognize several distinctions it has from Paraniphargus. The possibility that the species has sternal gills is refuted, thereby clarifying its relationships. Terms: “Gammaridan” refers to Gammaridea in the hypo- thetical group level of Section. “Mark” (“M.” refers to a position a stated distance from the proximal end on a scale of 0 to 100. “Melitid gnathopod 1” refers to a small, mitten- form gnathopod with transverse palm, elongate wrist and pubescence on one or more of articles 4-6. “Hadziid gnatho- pod 2” refers to an enfeebled female gnathopod with elongate wrist, the palm and posterior margin of the hand confluent and both armed with sparse groups of stiff, apically bent, elongate setae. These setae are also found in melitids, such as Psammoniphargus Ruffo, but occur on the posterior margin of the hand outside the palm. Uropod 3 is described in the following terms: Dispariramus, outer and inner rami dissimilar; Aequiramus, outer and inner rami similar in length, shape and patterns of armament; Magniramus, inner ramus extending as far as outer ramus; Variramus, inner ramus not as long as outer ramus but medial margin with armaments; 36—Proc. Bro. Soc. WasH., Vou. 89, 1976 (421) 422 Proceedings of the Biological Society of Washington Parviramus, inner ramus much shorter than outer ramus and lacking medial armaments, inner ramus usually very short and scalelike. The presence of a conspicuous article 2 on the outer ramus results in a classification of dispariramus; such uropod 3 can be magniramus, variramus or parviramus; an aequiramus uropod 3 is always magniramus but a magniramus uropod 3 can be either dispariramus or aequiramus. Galapsiellus, new genus Type-species: Paraniphargus lelouparum Monod, 1970 (here desig- nated ). Etymology: Contrived. Masculine. Diagnosis: Coxal gills 2-6, ovate, weakly pedunculate, not 2-articu- late. Sternal gills absent. Males and females almost identical except for penial processes of males and weak sausage-shaped oostegites 2-5 on female. Body subvermiform, all coxae very short, of similar length. Uropod 3 parviramus, outer ramus l-articulate, peduncle greatly elon- gate, about as long as longest ramus on uropods 1-2 and nearly as long as outer ramus of uropod 3. Telson fully cleft, lobes apically turgid, each bearing apicomedial spine. Gnathopods of both sexes enfeebled, gnathopod 1 of melitid form, wrist elongate, anteriorly pu- bescent, hand weakly trapezoidal, palm scarcely oblique, short, article 4 swollen and pubescent. Gnathopod 2 broader and longer than gnatho- pod 1, wrist similarly elongate, not pubescent, article 4 similar, not pubescent, article 6 almost twice as long as article 6 of gnathopod 1, about 1.2 times broader, palm oblique. Palms of gnathopods sparsely setose, lacking spines except at defining corners. Wrists of gnathopods unlobate. Mandibular palp article 3 linear, bearing only E setae (apical). Lower lip with weak inner lobes. Medial setae on maxillae absent or sparse. Pleopods biramous. Urosomites free, naked, or with at most one dorsolateral setule on each side. Galapsiellus lelouparum (Monod) Paraniphargus lelouparum Monod, 1970:13-25, figs. 6-45. Description: Blind. Head almost truncate anteriorly, with weak but broad and truncate anterior lobe (less accentuated than shown by Monod). Article 2 of pereopods 5-7 unexpanded, elongate, weakly pyriform; posteroventral corners right angular (or weakly sharpened in Monod’s specimens). Basofacial spine of peduncle on uropod 1 situated at Mark 45, apex of peduncle extended and_proboscoid. Dacty] of maxilliped with strong apical nail (not shown by Monod). New material: Five specimens from JLB GAL 103, Isla Santa Cruz, Galapagos Islands, Academy Bay, mangrove tidepool 300 m from sea near lower bodega of Charles Darwin Research Station, tidepool of Affinities of blind Galapagos amphipod 493 anchialine variety, not connected to sea at surface but fluctuating with tidal level, 23 January 1964, collected by J. L. Barnard. Associated fauna, Ampithoe sp., Cheiriphotis megacheles (both amphipods) and palaeomonid shrimps. Observations: One specimen is a definite male, bearing small penial processes on sternite 7 of the thorax. No sternal gills are present. Pre- sumably, therefore, the sausage-shaped appendages noted by Monod on pereonites 2-5 represented brood lamellae of a female, probably attached to the coxae but appearing to Monod to be attached to the sterna. One of these had a seta, also suggesting their identity as oostegites. Monod’s depiction of this species is excellent. New illustrations are therefore not required. Distribution: Isla Santa Cruz, Galapagos Islands, phreatic and an- chialine. CLASSIFICATION OF GAMMARIDANS Monod (1970) noted the potential affinities of Paraniphargus lelouparum as belonging to the broadly conceived groups proposed by Stephensen (1933) which included such diverse genera as Niphargus, Neoniphargus, Paraniphargus, Metaniphargus (= Hadzia), Uroctena, Austroniphargus, Melita and Crangonyx. He noted also the remarks of Schellenberg (1931) showing how close Paraniphargus lies to Melita and how Stephensen (1933) placed Paraniphargus into a subgroup containing Niphargus, Neoniphargus, Niphargopsis, and Metaniphargus (= Hadzia). These genera are now better divided into the following groups, some of which are briefly characterized (see in part, Bousfield, 1973). I. Crangonychoids (Crangonychidae, Bousfield, 1973). A primitive superfamilial group characterized by either true sternal gills, paddle- shaped calceoli or the presence of densely packed bifid-trifid spines on the palm of male gnathopod 2. Including, for example, most fresh- water genera of Australia (Neoniphargus Stebbing, Uroctena Nicholls); South Africa (Paramelita Schellenberg); Falklandella Schellenberg and Phreatogammarus Stebbing from Falkland and New Zealand, respec- tively; Pseudocrangonyx Akatsuka and Procrangonyx Schellenberg (= Eocrangonyx Schellenberg) from east Asia; plus the Holarctic cran- gonyxes. Numerous other genera. II. Gammaroids (Gammaridae, Bousfield, 1973). Sternal gills ab- sent. Palm of male gnathopod 2 not densely lined with bifid or trifid spines. Coxal gill 7 present or occasionally absent in apomorphic forms otherwise derivable from gammaroids; or marked by plesiomorphic characters such as tympanic calceoli in males. Eighty-five genera, nu- merous groupings, examples, Gammarus J. C. Fabricius, Chaetogam- marus Martynov, Acanthogammarus Stebbing, Amathillina Sars, Saro- throgammarus Martynov, Pontogammarus Sowinsky, Micruropus Stebbing. 424 Proceedings of the Biological Society of Washington A. Anisogammarids. Palms of male gnathopods with peg-spines. Anisogammarus Birstein, Spinulogammarus Tzvetkova, Eogammarus Birstein, Bathyceradocus Pirlot, ?Metaceradocoides Birstein and Vino- gradova (and see below). B. Mesogammarid. Apomorphic form. Mesogammarus Tzvetkova. C. Micruropus alaskensis Bousfield and Hubbard. Apomorphic (new genus, Bousfield, in prep. ). D. Eoniphargus Ueno. Apomorphic form with tympanic calceoli. III. Gammarelloid Groups. One or both gnathopods scarcely or not prehensile, but coxal gill 7 usually retained. Divisible into groups characterized by Gammarellus Herbst, Cheirocratus Norman, Horn- ellia Walker, Megaluropus Hoek, Argissa Boeck, Melphidippa Boeck and Macrohectopus Stebbing. IV. Hadzioids (Hadziidae, Karaman, 1943; Melitidae, Bousfield, 1973). Sternal gills absent. Coxal gill 7 absent. Gnathopods subchelate except for special form of female hadziid gnathopod 2. Calceoli absent. Loss of any other marker attributes showing immediate descent from Crangonychoids and Gammaroids. Probably polyphyletic and subject to further elaboration. A. Melitids, to contain Melita Leach, Eriopisa Stebbing, Dulichiella Stout, Melitoides Gurjanova, Psammoniphargus Ruffo, and several new genera. Uropod 3 dispariramus. Lower lip with inner lobes. Female gnathopod 2 with distinct palm Jacking groups of bent setae. 1. Subgroup to contain Psammogammarus S. Karaman, characterized by loss of sexual dimorphism on gnathopod 2, possibly Eriopisa longiramus Stock and Nijssen to be distinguished generically by variramus uropod 3. Divergent from ancestors of Eriopisa. 2. Subgroup to contain Paraniphargus Tattersall characterized by loss of medial setation on maxillae and possibly by loss of sexual dimorphism in gnathopod 2; retaining enlarged coxae unlike sub- group 1. Derivative from Melita. 3. Subgroup to contain Galapsiellus, new genus, characterized by re- duction of medial setation on maxillae, reduction of anterior coxae, partially mittenform gnathopod 2 lacking sexual dimor- phism. Derivative from Eriopisa. B. Hadziids, to contain Hadzia Karaman, Dulzura J. L. Barnard and new genera (in part, see Zimmerman and Barnard, in press). Uropod 3 dispariramus. Lower lip lacking inner lobes. Female gnatho- pod 2 lacking palm but dactyl closing against margin furnished with sparse groups of stiff, elongate, apically bent setae. Other genera re- moved to weckeliids and ceradocids. 1. Metacrangonyx Chevreux. Group characterized by entire telson and miniaturized uropod 3. C. Eriopisellids, to contain Eriopisella Chevreux, Netamelita J. L. Bamard, Indoniphargus Straskraba, and Microniphargus Schellenberg. Both sexes with fully mittenform gnathopods. Affinities of blind Galapagos amphipod 4295 1. Bathyonyx subgroup to contain Bathyonyx Vejdovsky. Charac- terized by especially reduced maxillary spination. 2. Niphargus pulchellus Sayce. Australia. D. Salentinellids, to contain Salentinella Ruffo, Parasalentinella Bou. Gnathopods mittenform but article 2 of pereopods 5-7 expanded and lobate. E. Austroniphargids, to contain Austroniphargus Monod and a new genus (Austroniphargus starmuhlneri Ruffo). Characterized by ham- mer-like gnathopods bearing lobular wrists. Urosomites coalesced. Pos- sibly of direct crangonychoid descent. F. Pseudoniphargids, to contain Pseudoniphargus Chevreux. Gnatho- pod 1 of melitid form, hammer-like. Gnathopod 2 enlarged, palm oblique, wrist short, scarcely lobate. Telson almost entire. G. Niphargids, to contain Niphargus Schiddte, Pontoniphargus Dan- cau, Haploginglymus Mateus and Mateus, Niphargopsis Chevreux, Niphargellus Schellenberg and Carinurella Sket. Characterized by hammer-like gnathopods resembling mittens but enlarged, or apo- morphically derived from enlarged hammer-like gnathopods lacking palmar spines, wrists unlobate. H. Ceradocopsids, to contain Ceradocopsis Shellenberg (= Maera- cunha Stephensen), ?Metaceradocoides Birstein and Vinogradov. Uro- pod 3 miniaturized, retaining conspicuous article 2 on outer ramus. I. Nuuanuids (see McKinney and Barnard, in prep.), to contain Cottesloe J. L. Barnard, Gammarella Bate, Nuuanu J. L. Barnard, and new genus. Uropod 3 miniaturized as in category 9 but article 2 of pereopod 7 broadly expanded, hatchet-shaped. J. Weckeliids, formerly hadziids, to contain Weckelia Shoemaker, Alloweckelia Holsinger and Peck, Mexiweckelia Holsinger and Minckley, Mexiweckelia particeps Holsinger and Minckley (new genus, Holsinger, in prep.). Uropod 3 aequiramus. Female gnathopod 2 enfeebled, palm distinct, lined evenly with weakly bifid spines, posterior bent setae not fully developed on palm. Inner lobes of lower lip weak. Possibly derivative from ceradocids (to follow) or directly from crangonychoids, K. Ceradocids, to contain all other fully marine gammaridan genera lacking coxal gill 7, bearing gills 2-6, bearing fully subchelate gnatho- pods, nonvermiform body, normal oostegites and pleopods, uropod 3 basically magniramus and aequiramus except in apomorphic genera. Including Paraweckelia Shoemaker, a former hadziid. 1. Ceradocins. Infra group a. Ceradocins, Ceradocus Costa, Ceradomaera Ledoyer, Paraweckelia Shoemaker, Ceradocoides Nicholls. Group IVJ weckeliids, probably descend at this point. Infra group b. Paraceradocins. Antenna 2 with articles 4-5 elongate, slightly thickened. Paraceradocus Stebbing, Quadrivisio Steb- bing. 426 Proceedings of the Biological Society of Washington Infra group c. Anelasmopus Oliveira, Elasmopoides Stebbing, Maeropsis Chevreux, Maera Leach. Infra group d. Elasmopus Costa. Infra group e. Parelasmopus Stebbing, Mallacoota J. L. Barnard, Ifalukia J. L. Barnard, Beaudettia J. L. Barnard. 2. Parapherusa group. Uropod 1 and uropod 3 aberrant. Para- pherusa Stebbing. 3. Maerellins. Peduncle of uropod 3 elongate. Maerella Chevreux, Jerbarnia Croker. L. Kergueleniolids, to include Kergueleniola Ruffo (= Kerguelenella Ruffo). Like ceradocopsids but body vermiform. V. Bogidiellids, to include Bogidiella WHertzog, Bollegidia Ruffo. Body vermiform, coxal gills reduced to 4 pairs, brood plates with dense setae confined terminally, some brood plates geniculate. VI. Pseudingolfiella Noodt. PROBLEMS OF CONVERGENCE The higher classification of gammaridean Amphipoda is fraught with difficulties, among them convergence. The problem of the third uropod in hadziids is just one of many for which morphologists alone may never have answers. For example, crangonychoids (Bousfield, 1973) are marked either by sternal gills or bifid spines densely lining one or more of the gnathopodal palms. Because loss of structure is the trend in gammaridean evolution, the loss of these gills and spines could result in a host of descendents unrecognized in other super- families. Characters frequently do not occur universally in a higher taxon of amphipod, though closely related or inclusive taxa can often be recognized by marker attributes. Some crangonychoids lose sternal gills, some lose coxal gill 7 and others lose bifid palmar spines, but if the first and last are present, or if an apomorphic species can otherwise be adjoined in an obvious evolutionary sequence, then one may recog- nize a crangonychoid. Eoniphargus Ueno, for example, a Japanese hypogean gammaroid, might have crangonychoid ancestry except that the male calceoli are tympanic, rather than paddle-shaped. By that character, Eoniphargus is also not a member of the greater hadzioids and probably should be derived from a gammaroid (greater Gam- maridae) ancestry, even though coxal gill 7 is absent. A geographic companion, Awacaris Ueno, is clearly analogous to Eoniphargus in scores of attributes, though uropod 3 is aequiramus, whereas uropod 3 of Eoniphargus is dispariramus and parviramus. Awacaris therefore appears related to the marine ceradocins and the bogidiellids with aequiramus uropod 3 but its gnathopods are far more apomorphic than those of Eoniphargus, which has the apomorphic uropod 3. However, it also shows plesiomorphy in the almost imperceptible remnant of article 2 on the outer ramus. Uropod 3 of Phreatogammarus (crangony- choid) also is aequiramus in contrast to all of its congeners. The dis- pariramus uropod 3 greatly resembles that of Notogean syncarids and Affinities of blind Galapagos amphipod 497 could be conceived as a plesiomorphic attribute, but the aequiramus uroped 3 would have to be derived from that plesiomorphic condition by loss of article 2 on the outer ramus without concomitant reduction of the inner ramus. Most of the ceradocid marine gammaroids possess the aequiramus uropod 3 but it is also present in the weckeliids and in the bogidiellids, kergueleniolids, awacarids and phreatogammarids. The aequiramus uropod 3 also has the rami more or less equal in thickness and armaments. The greater proportion of gammaridans carry the dispariramus uropod 3, or its derivatives, with the outer ramus incipiently or fully biarticulate and the inner ramus of diverse shapes and Jengths. Ancestry in taxa with severe reduction in any component of uropod 3 cannot necessarily be traced. The aequiramus uropod 3 can be shown to be ancestral to the fully parviramus stage (as seen in the marine Beaudettia J. L. Barnard) as much as can the dis- pariramus uropod 3. The probability is high that a consistent evolutionary trend in uropod 3 dominates the amphipods but much elucidation is required before this trend can be perceived. The question is very basic to the origins of Gammaridea because either the gammarid-like amphipods or the photidcorophiid amphipods such as Gammaropsis, characterized by fleshy telson and generally by aequiramus uropod 3, are presumed to be the most primitive living gammarideans. The ultimate question is whether or not the dispariramus uropod 3, associated generally with freshwater amphipods, is the more primitive and therefore signals a freshwater, Notogean, syncarid ancestry. In this evolutionary sequence one would assume that sternal gills mark the primitive state and that Phreatogammarus, isolated in New Zealand, would be the most plesio- morphic of the aequiramus and sternobranchiate forms, perhaps antic- ipatory to invasion of the sea. On the other hand, the fleshy telson of corophioids, coupled with the typical aequiramus uropod 3 of marine gammarideans may mark the ancestral stock, in which case the dis- pariramus uropod 3 is an apomorphic development. AFFINITIES OF GALAPSIELLUS Galapsiellus differs from Paraniphargus in the elongate peduncle of uropod 3, the size reduction of gnathopod 2 in both sexes, and the distad position of the basofacial spine on uropod 1. Gnathopod 2 ap- proaches the mittenform-shape found in Eriopisella and its allies but is significantly larger than gnathopod 1 and the hand is weakly ex- panded apically in contrast to the eriopisellid genera. Reduction of gnathopod 2 in male gammaridan amphipods is a common generic character. It typifies the eriopisellids, another group of genera known as weckeliids and other scattered genera of the gammaridan group. This characteristic is especially prevalent in anchialine or phreatic or anoculate groups. Monod was undoubtedly correct in assuming a relationship of G. lelouparum to Paraniphargus. That genus, with two species, occurs 428 Proceedings of the Biological Society of Washington on western Pacific and Indian Ocean islands in freshwaters and pre- sumably the species are basically hypogean although the type-species, P. annandalei Tattersall, 1925, was found among matted rootlets of a jungle stream at 152-244 m altitude in the South Andaman Islands, whereas the second known species, P. ruttneri Schellenberg, 1931, was found in a well in east Java. Paraniphargus has the primitive gnathopod 2 exhibited by con- tiguous marine amphipods such as Melita, an enlarged appendage with short and weakly lobate wrist, and has the primitive uropod 3 with shortened peduncle. Paraniphargus can be derived from a widely distributed tropical and temperate marine genus, Melita, with more than 50 species, often found in estuaries or in anchialine situations. Paraniphargus differs from Melita in the complete loss of medial setation on the maxillae. Paraniphargus is not well-known but one would suspect that it is also characterized by a loss of sexual dimorphism in gnathopod 2, which in Paraniphargus is a blend between male and female conditions of Melita. The loss of article 2 on the outer ramus of uropod 3 and the loss of all but E setae on mandibular palp article 3 are but ex- tensions of conditions almost fully expressed in several marine species of Melita. Paraniphargus maintains the fleshy inner lobes typical of Indo-Pacific species of Melita. Galapsiellus carries the trends of reduc- tion and sexual stabilization in gnathopod 2 to an extreme almost typical of eriopisellids. This group of genera, containing Eriopisella Chevreux, Netamelita J. L. Barmard, Indoniphargus Straskraba and Microniphargus Schellenberg, is characterized by fully mittenform gnathopods. Gnathopod 1 retains the melitid (mittenform) form but gnathopod 2 is reduced to the same size and has some of the same characteristics, such as medial or posterior pubescence and elongate wrist. In Microniphargus and Indoniphargus the gnathopods are axially reversed in contrast to Eriopisella as marked by the presence of a pubescent posterior lobe on the wrist of gnathopod 1, which in Eriopis- ella occurs on gnathopod 2. Eriopisellids also tend to be blind and anchialine and some of them, Microniphargus and Indoniphargus, have penetrated fully into phreatic waters of Belgium and India. Galapsiellus is not on eriopisellid because its gnathopods are some- what more primitive, though, like Paraniphargus, it could lie ancestral to eriopisellids. This group and the eriopisellids could also lie ancestral to niphargids. Galapsiellus bears a remarkable resemblance to the enigmatic Bathy- onyx Vejdovsky, 1905, from Lough Mask, a lake of Ireland. The body of Galapsiellus is more vermiform but the odd shape of the anterior coxae, weakly sinuous posteroventrally, is similar. In Galapsiellus much of this sinuosity can actually be eliminated by pressing the coxae very flat. The large, almost truncate head of Bathyonyx is very similar to that of Galapsiellus. Bathyonyx also has mittenform gnathopods with elongate wrists but they are closer to the eriopisellid kind than to Affinities of blind Galapagos amphipod 429 the galapsiellin kind because they are identical in size or gnathopod 1 is very slightly the larger. Bathyonyx is characterized by the retention of medial setae on maxilla 2 (maxilla 1 inner plate is unknown), the outer plate of maxilla 1 has the spines reduced to 4, uropod 3 is parviramus but the peduncle is short and the outer ramus is 2-ar- ticulate, the mandibular palp article 3 is more tumid and setose, and the telson is elongate. Geographically and temporally Galapsiellus bears no_ relationship to crangonychoids, though apomorphic crangonychoids would be dif- ficult to detect once the sternal gills, coxal gill 7, urosomal setation and bifid palmar spines were lost. Crangonychoid distribution appears to have been a product of Pangaea in the early Mesozoic. Crangony- choids are strictly of freshwater provenance and probably were widely distributed over the coalesced continental masses. They have survived primarily in Notogea, South Africa, and Nearctica with an outpost in the Falkland Islands. In Palearctica they were largely replaced by modern gammaroid genera but never reached Neotropica. There is no nearby source of crangonychoids to postulate an origin for Galapsiellus and the Galapagos Islands are too youthful to retain relicts of Pangaea. Galapsiellus can also be derived from the widespread marine genus Eriopisa. The similarity between these genera is even greater than between Galapsiellus and Paraniphargus, because Eriopisa possesses the shortened anterior coxae not typical of Paraniphargus. Many species of Eriopisa have female gnathopod 2, and occasionally male gnathopod 2, more strongly reduced than in Paraniphargus, although none of them has gnathopod 2 as enfeebled as in Galapsiellus. Eriopisa is more plesiomorphic than either Paraniphargus or Galapsiellus in maxil- lae and uropod 3 and therefore could be ancestral to Galapsiellus but not to Paraniphargus. Uropod 3 of Eriopisa bears a moderately to well-developed article 2 and the medial margins of one or both maxillae bear setae. Eriopisa may not be directly descendent from ancestors like Melita because uropod 3 of the most primitive species of Eriopisa, E. longiramus Stock and Nijssen, is magniramus, a con- dition plesiomorphic to the parviramus uropod 3 of Melita. Eriopisa longiramus and E. caeca (S. Karaman) should be reestablished in Psammogammarus S. Karaman and differentiated from Eriopisa by loss of sexual dimorphism in gnathopod 2 and the evenness of spination on the palm of gnathopod 2 in the female. Eriopisa longiramus may fur- ther be distinguished generically as uropod 3 is almost magniramus, like Pontoniphargus Dancau. Galapsiellus bears a resemblance to the Maerella subgroup of the Ceradocus group because of the elongate peduncle on uropod 3 and the shape of the telson. Male gnathopod 2 of Jerbarnia in that group has undergone an elongation reminiscent of Galapsiellus but otherwise the number of evolutionary steps between Jerbarnia (Micronesian marine) and Galapsiellus is far greater than between Eriopisa and Galapsiellus. 430 Proceedings of the Biological Society of Washington The occurrence of a phreatic amphipod in the Galapagos is striking because of the isolation of the archipelago from the mainstream of gammaridean dispersal and evolution. Nearby South America has only Ingolfiella Hansen (suborder Ingolfiellidea) and Pseudingolfiella Noodt (suborder Gammaridea) occurring in its phreatic waters. Galapsiellus is very remote from those taxa and clearly has a marine origin. In the Galapagos Islands, Galapsiellus has been collected (Monod and herein) in brackish anchialine waters presumed to be a mixture of intruding seawater and phreatic freshwaters percolating downslope to the sea from the highlands of Santa Cruz Island. Brackish sink- holes and emergent aquifers in mangrove swamps are often sufficiently fresh to be potable (5 ppt) and are so used by inhabitants of the island (pers. observ.). The presence of Ampithoe and Cheiriphotis in my sample suggests that Galapsiellus lived in far saltier water than 5 ppt as those genera are strictly marine. The mangrove pond I sampled may actually be a stratified pool of differing salinities so that my broadly cast sample may have covered several salinity regimes. If Galapsiellus is an emergent phreatic genus, the specimens of Gala- psiellus may actually have been dying of exposure to high salinities as I caught them. On the other hand, the genus may be euryhalinic and this may help to explain its immigrational adaptability from the sea. CONCLUSION Galapsiellus is considered to be an apomorphic melitid with the same kind of phreatiform adaptations found in the eriopisellids and niphargids. The best ancestral fit lies near Eriopisa although close morphological similarity occurs with Paraniphargus. ACKNOWLEDGMENTS I am indebted to J. H. Stock, E. L. Bousfield, J. R. Holsinger, and T. E. Bowman for help with but not necessarily full agreement on the classificatory system outlined herein. LITERATURE CITED BousFieLp, E. L. 1973. Shallow-water Gammaridean Amphipoda of New England. Comstock Publishing Associates, Cornell Uni- versity Press, Ithaca, 312 p. KaraMaNn, S. 1943. Die unterirdischen Amphipoden Sudserbiens. Srpska Kral’evska Akademiia, Posebna Izdan’a, Prirodn’achki i Mathematichki Spici, Okhridski Zbornik 34:161-313, figures 1-215. Monop, TH. 1970. V. Sur Quelques Crustacés Malacostracés des Iles Galapagos.... Mission Zoologique Belge Iles Galapagos et en Ecuador (N. et J. Leleup, 1964-1965) 2:11-53, figures 1-104. Affinities of blind Galapagos amphipod 431 SCHELLENBERG, A. 1931. Amphipoden der Sunda-Expeditionen Thiene- mann und Rensch. Archiv Hydrobiologie, Supplement 8:493- 511, figures 1-3. STEPHENSEN, K. 1933. Fresh- and Brackish-water Amphipoda from Bonaire, Curacao and Aruba. Zoologische Jahrbiicher, Sys- tematik 64:414—436, figures 1-8. TATTERSALL, W. M. 1925. Fresh water Amphipoda from the Anda- man Isles. Records Indian Museum 27:241-247, figures 1-13. Veypovsky, Fr. 1905. Ueber einige Siisswasser-Amphipoden. III. Sit- zungsberichte Ko6nigeliga Bohmischen Gesellschaft der Wis- senschaften, Mathematisch- Naturwissenschaftliche Classe, 1905 (28 ):1-40, plates 1-2. ZIMMERMAN, R. J., AND J. L. BARNARD. in press. A New Genus of Marine Hadziid (Amphipoda, Crustacea) from Bimini and Puerto Rico. Proceedings Biological Society, Washington. 432 Proceedings of the Biological Society of Washington y 17, pp. 4383-438 12 October 1976 ied Loss hy, re PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW SPECIES OF ARICIDEA (POLYCHAETA: PARAONIDAE) FROM FLORIDA By Betsy BROWN College of Marine Studies, University of Delaware, Lewes, Delaware 19958 A new species of the genus Aricidea (Paraonidae) from the Indian River lagoon of the east coast of Florida is de- scribed below. Specimens were collected by Dr. David K. Young and his staff of the Smithsonian Institution’s Fort Pierce Bureau (FPB). Additional specimens from the west coast of Florida, collected by Dr. John L. Taylor and Mr. Stuart L. Santos and deposited in the National Museum of Natural History (USNM), were examined and referred to the new species. I am grateful to Dr. Young for the loan of specimens from the Indian River, to Ms. Ruth Swanson, Dr. Donald Maurer, and Dr. Les Watling for useful criticism during preparation of the manuscript, and to Sandy Steele for careful typing of the final draft. I particularly wish to thank Dr. Marian H. Pettibone of the Smithsonian Institution for the loan of specimens, for considerable assistance with some of the litera- ture and systematic problems, and for critically reviewing the manuscript. Types are deposited in the Smithsonian Insti- tution (USNM). This is contribution 109, University of Dela- ware, College of Marine Studies. PARAONIDAE Cerruti Aricidea Webster Aricidea philbinae, new species Material examined: Holotype (USNM 53172) and one paratype (USNM 53173): Indian River lagoon adjacent to Hutchinson Island, Ft. Pierce, Florida (20°20.9’N, 80°15.7’W), on Halodule wrightii grass- flats, 1 m, in muddy sand, collected 12 April 1973. Four paratypes (USNM 53174): Indian River lagoon just N of St. Lucie Inlet, St. Lucie, 37—Proc. Biot. Soc. WasH., Vou. 89, 1976 (433) 434 Proceedings of the Biological Society of Washington Florida (27°10.9’N, 80°10.3°W) on H. wrightii grassflats, intertidal to 1 m, in muddy sand, collected 16 November 1974. Other specimens: about 50 animals in Indian River lagoon ranging from Titusville, Florida (just N of Haulover Canal) (28°44.1’N, 80°45.5’°W) to just N of St. Lucie Inlet, St. Lucie, Florida, intertidal to 1 m, collected 1974-1975, on H. wrightii grassflats, several bottom stations, D. K. Young and staff, collectors. Tampa Bay, Florida (27°47’N, 82°40’W) (USNM 48936) 26 specimens, collected September 1971, J. L. Taylor, collector. Tampa Bay, Florida (USNM 48936) 3 specimens, collected 27 October 1969, S. L. Santos, collector. Lassing Park, St. Petersburg, Florida (27°45.2’N, 82°37.7'°W) (USNM 50280/1) 9 specimens, col- lected 18 April and 20 July 1970, S. L. Santos, collector. Description: Length up to 12 mm, width up to 0.5 mm, up to 77 segments, several ovigerous. The 4 ovigerous specimens ranged from 9.8-10.6 mm long and 0.36-0.44 mm wide. Body cylindrical, widest in branchial region, tapering and flattened dorsoventrally more pos- teriorly. Prostomium subtriangular (Fig. 1b, c), rounded anteriorly, fused with achaetous buccal segment, forming lateral lips of ventral mouth. Posterior lip of mouth formed by anterior edge of first setiger. Median antenna clavate, short, extending to setiger 2, unequally bifid distally, with smaller subterminal process (Fig. lc) or rarely with asymmetrical terminal process (Fig. 1b). Cilia present on distal tips of bifid antenna. Nuchal slits posterolateral to median antenna. Pig- ment spots sometimes present anterior to nuchal slits, as well as on anterior region of body, particularly on dorsal surface. Branchiae 13- 15 pairs, beginning on setiger 4. Branchiae widest proximally, taper- ing to rounded tips and overlapping slightly middorsally (Fig. la, c). Anterior 7-10 branchial pairs usually longer and more robust. Single small orbicular papilla posterior to and hidden by branchia on branchial segments (Fig. la). Notopodial postsetal lobes small, spherical on first 2 setigers, more elongate on third, subulate in branchial region (Fig. la, c) becoming more slender and cirriform in postbranchial region (Fig. le). Dense bundles of basally thickened and evenly tapered capillary setae in prebranchial and branchial regions, number- ing about 10 in notopodia and about 15 in neuropodia. Anterior post- branchial setigers with finer capillary setae, numbering about 5 in notopodia and 7 in neuropodia. Posterior notopodia with about 3 (2-5) slender capillaries. Modified neuropodial hooks beginning on about setiger 22 (19-24). Neuropodia with about 5 (3-7) very slender capillaries and 5 (4-8) modified neurosetae (Fig. le). Modified neurosetae hooked or curved distally, with terminal aristae (apparently fragile, sometimes absent), with small subterminal spine on concave side (perhaps part of subterminal hood) (Fig. 1d, e). Pygidium with pair of small anal cirri. Etymology: It gives me great pleasure to name this species for May- belline Philbin, who was an unparalleled source of inspiration prior to the preparation of this manuscript. New polychaete from Florida 435 Fic. 1. Aricidea philbinae, new species: a, Parapodium from bran- chial region, posterior view; b, c, Anterior ends of 2 different speci- mens, dorsal view; d, Modified neurosetae from posterior parapodium; e, Posterior parapodium, posterior view. Distribution: East and west coasts of Florida. Intertidal to 1 m. Remarks: Aricidea philbinae is close to Aricidea jeffreysii (Mc- Intosh) which was described by Pettibone (1965:134) for paraonids from Virginia. Cerruti first used Aricidea jeffreysit (McIntosh) when he established the family Paraonidae. Since then, it has been used widely as A. jeffreysii sensu Cerruti by numerous authors. Strel’tsov (1973) presented a thorough revision of the family Paraonidae in which he reexamined many specimens including type material. After reviewing the holotype of Scolecolepis (?) jeffreysii Mcintosh, 1879, from the Davis Strait, he concluded (1973:106, 159) that it should 436 Proceedings of the Biological Society of Washington be an indeterminable Aricidea sp. He did so because the type-specimen was a fragment of 26 anterior segments lacking the modified posterior neurosetae necessary for diagnostic species description. Therefore, Mc- Intosh’s name should not be used and all the subsequent records of A. jeffreysii have needed to be referred to other species. Many of them were referred by Laubier (1967:102) and Strel’tsov (1973:105) to Aricidea ( Acesta) cerrutii Laubier, 1966. Strel’tsov (1973:91, 159) examined 6 paratypes of Aricidea (Acesta) catherinae Laubier, 1967, from the Mediterranean (Banyuls-sur-Mer, France). After examining specimens of each of the following species, he referred them to A. (Acesta) catherinae: the records of A. jeffreysii by Pettibone (1963, 1965); the records by Hartman of Aricidea lopezi (1963:38, as A. lopezi, not Berkeley and Berkeley); and Aricidea zelen- zovi Strel’tsov, 1968. From material on loan from USNM, I examined 3 paratypes of A. catherinae from France (USNM 35914) and numerous specimens identified as A. jeffreysii by Pettibone from Maine (USNM 28940), Massachusetts (USNM 28935, 31496/7) and Virginia (USNM 31498-31500). These specimens agreed for the most part with the de- scriptions of Aricidea catherinae by Laubier (1967:112) and Strel’tsov (1973:91). None of the specimens showed a bifid antenna as found on A. philbinae. The revised distribution of A. catherinae (Strel’tsov, 1973) therefore, is: Atlantic coast of North America (from Gulf of St. Lawrence to Chesapeake Bay), coast of Uruguay, Mediterranean, Barents Sea, region of Kurile Island, off Southern California, in 2 to 1929 m. Aricidea catherinae and A. philbinae agree in the following charac- ters: Median antenna is relatively short; branchiae begin on setiger 4; notopodial postsetal lobes are small, bulbous on first 2 setigers, subu- late and cirriform more posteriorly; absence of modified setae in noto- podia; neuropodial lobes lacking; modified setae found in posterior parapodia, possessing terminal aristae and short, subterminal spines (“hoods”). Median Antenna: The median antenna of A. catherinae is enlarged in the midregion by a varying amount and is thinner distally. This agrees with Laubier (1967:114, Fig. 4a, b, c) and Strel’tsov (1978: 91). Strel’tsov’s (1973) illustration of the dorsal view of the median antenna might lead one to believe that it supports a secondary process. This is not the case, as he states in the text and as I have observed in examination of specimens. On the A. catherinae identified as A. jef- freysii by Pettibone (USNM 28935, 28940, 31496-31500) and ex- amined by me, the swelling of the midregion of the antenna was so reduced as to appear subulate. Branchiae: Strel’tsov (1973:91) found that the number of branchial pairs of A. catherinae varied with the size of the specimen, smaller specimens (0.15 mm width) with 8-12 pairs and larger ones (0.7— 0.9 mm) with up to 25 pairs. A. philbinae has 13-15 pairs of branchiae and a maximum width of 0.5 mm. New polychaete from Florida 437 Setae: The modified neurosetae of A. catherinae and A. philbinae are similar possessing terminal aristae and a short, subterminal spine (or “hood”). The “hoods,” visible on the concave side, are very dif- ficult to observe. In examination of the paratypes of A. catherinae, Strel’tsov (1973) observed a short, spinelike structure below the tip of the seta on the concave side, probably corresponding to a part of the “hood.” My observations of specimens borrowed from the USNM correspond with this. Distinct hoods on the modified setae were de- picted by Laubier (1967:Fig. 5a-d). The appearance on the modified neurosetae as having hoods or short, subterminal spines seems de- pendent upon the optical equipment of the observer. Strel’tsov (1972:150) documented the setal morphology of the fam- ily Paraonidae. He described 3 main groupings concerning the de- velopment of specialized setae within the family. The first group, found on the dorsal parapodial branches of the branchial and _post- branchial segments, includes lyrate setae and aciculate setae bearing spines. Setae of the second and third groups are distributed on the ventral branches of the postbranchial parapodia. Group 2 contains “pseudocompound” setae, hooked setae with a subterminal spine and “hooded” setae. According to Strel’tsov’s illustration (1972:90, Fig. 1) group 2 setae may be distally curved. Group 3 includes thickened setae, setae with a terminal spine, and various hook-shaped setae which lack spines. These may be either smooth or possess thin hairs on their distal ends. The classification of the setae of A. philbinae in this scheme is uncertain. They possess the character of group 3 by having a terminal spine and by being hook-shaped. There is no indication of pubescence on the setae nor do they appear thickened. However, they also possess a hood or a short subterminal spine (as discussed earlier) giving them the character of group 2. In Florida, A. philbinae has been collected in association with Ari- cidea fragilis Webster, 1879 and Aricidea taylori Pettibone, 1965 (Pettibone, 1965:129, 131). These 3 species differ in the following characters: A. fragilis A. taylori A. philbinae Median Antenna Subulate Clubbed Clavate, bifid Branchiae 50-60 pairs 26-29 pairs 13-15 pairs Posterior Stouter basally, Bidentate, Unidentate, Modified tapering abruptly aristae arise curved dis- Neuropodial in midregion to — terminally be- tally, with short Setae capillary tips, tween teeth subterminal spine sometimes frac- (hood ), terminal tured at mid- aristae point 438 Proceedings of the Biological Society of Washington Pettibone (1965:131) described the median antenna of A. taylori as “short, extending at most to first setiger, clubbed (one paratype with distal tip bifid, evidently an anomaly. . .).” The holotype and paratype (USNM 31494/5) were examined and the terminally bifid setae and median antenna were as described. The anomalous paratype could not be located (according to Pettibone in correspondence, one paratype was sent to Dr. Laubier in exchange, the other paratype ap- parently has been lost). The relationship of its setae and antenna to those of A. philbinae could not be determined. LITERATURE CITED HarTMAN, Otca. 1963. Submarine Canyons of Southern California, Part 3. Systematics: Polychaetes. Allan Hancock Pac. Ex- ped. 27(3):1-93. LaupieR, Lucien. 1967. Sur quelques Aricidea (Polychétes, Para- onidae) de Banyuls-sur-Mer. Vie Milieu 18(1A):99-132. PETTIBONE, Martan H. 1963. Marine Polychaete Worms of the New England Region. 1. Aphroditidae through Trochochaetidae. Bull. U.S. Nat. Mus. 227:1-356. 1965. Two new species of Aricidea (Polychaeta, Paraonidae ) from Virginia and Florida, and redescription of Aricidea fragilis Webster. Proc. Biol. Soc. Wash. 78:127—140. STREL tsov, V. E. 1972. [On the morphology of the setae of the polychaete family Paraonidae Cerruti, 1909 (Polychaeta, Sed- entaria).] Doklady Akad. Nauk. SSSR, 202(5):1219-1222. [In Russian]. English translation, Proc. Acad. Sci. USSR, Biol. Sci. Sec., 202:148-150. ——. 1973. [Polychaete worms of the family Paraonidae Cerruti, 1909 (Polychaeta, Sedentaria).] Akad. Nauk. SSSR. Mur- manskii Morsk Biol. Inst., 1-170 [In Russian]. 4 te ; = LY lo. 38, pp. 489-450 12 October 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON OCCURRENCE OF THE CARIBBEAN STOMATOPOD, BATHYSQUILLA MICROPS, OFF HAWAII, WITH ADDITIONAL RECORDS FOR B. MICROPS AND B. CRASSISPINOSA By Raymonp B. MANNING AND PAUL STRUHSAKER Smithsonian Institution, Washington, D.C. 20560, and National Marine Fisheries Service, Honolulu, Hawaii 96812 During the sampling program of the National Marine Fish- eries Service (NMFS) aboard the NOAA research vessel TOWNSEND CROMWELL in 1972, one specimen of an unusual stomatopod was taken by trawl off Maui, Hawaiian Islands, at a depth of 731-786 m. It proved to be an adult female of the rare Bathysquilla microps (Manning, 1961), then known from only three specimens taken in 732-952 m off southeastern Florida and the Bahamas (Manning, 1969a). Bathysquilla microps was the second species of bathy- squillid to be recognized, the first being B. crassispinosa (Fukuda, 1909), originally described from Japan and subse- quently recorded from additional localities in the western Indian Ocean. It occurs in depths between 230 and 310 m. A third bathysquillid representing another genus, Indosquilla manihinei Ingle and Merrett, 1971, was described from a unique specimen taken off Cosmoledo Atoll, Indian Ocean, in a depth of 420 m. Apparently bathysquillids are restricted to moderate depths of the outer shelf and upper slope (Fig. 2). The bathysquillids live in far greater depths than the major- ity of known stomatopods. The deepest record given by Kemp (1913) in his monograph of the Indo-West-Pacific stomato- pods was 370-419 fm (677-767 m) for Squilloides leptosquilla from the “Investigator” collections. Chopra (1939) described 38—Proc. Biou. Soc. WasuH., Vou. 89, 1976 (439) 440 Proceedings of the Biological Society of Washington Eurysquilla sewelli from 695 m in the Gulf of Aden and re- corded two species from 1295 m in the same area. However, these two species, Gonodactylus chiragra (Fabricius) and Oratosquilla investigatoris (Lloyd), are known to live in much shallower water (especially G. chiragra, an intertidal species), and their occurrence at this depth is questionable. In a review of the western Atlantic stomatopods, Manning (1969a) re- corded 14 species from depths greater than 300 m, and only two of these, B. microps (732-952 m) and Squilla intermedia Bigelow (291-615 m), occurred in depths greater than 450 m. In reporting this unusual extension of range, we present here additional northwestern Atlantic records for B. microps, based on collections made since 1969 by the National Marine Fisheries Service, additional records for B. crassispinosa, and comparative diagnoses and sketches for both of these species. Terms and indices used in the accounts below have been discussed in Manning (1969a). All measurements are in milli- meters; total length (TL) is measured on the midline from the anterior margin of the rostral plate to a line between the apices of the submedian teeth of the telson. Station data for the Orecon II collections are on file in the Department of Invertebrate Zoology (Crustacea), National Museum of Natural History, Smithsonian Institution. All specimens are in the Smithsonian. The illustrations were prepared by Lilly King Manning. Bathysquilla crassispinosa (Fukuda, 1909) Figures la—c, 2 Lysiosquilla crassispinosa Fukuda, 1909:61, pl. 5, fig. 4. Bathysquilla_ crassispinosa.—Holthuis, 1967:27 [synonymy to 1967].— Manning 1969a:95, 98.—Holthuis and Manning, 1969:fig. 350-2.— Ingle and Merrett, 1971:197. Material: Shikoku Island, Tosa Bay, Japan: 192 TL (total length) 240 mm.—Madagascar; 18°54’S, 43°55’E; 280-310 m; A. Crosnier, leg.; 24 November 1973: 19 TL 215 mm. Diagnosis: Cornea (Fig. la) subglobular, fully pigmented, set obliquely on stalk. Rostral plate (Fig. la) with distinct median longi- tudinal groove. Antennal protopod with 2 papillae. Dactylus of claw with 9-10 teeth. Carpus of claw with 2 dorsal spines (Fig. 1b). Inter- mediate carinae prominent on thoracic and anterior 2 abdominal somites, unarmed, Third to fifth abdominal somites lacking spinules on posterior Caribbean stomatopod off Hawaii 441 Fic. 1. Front, carpus of claw, and basal prolongation of uropod of: a—c, Bathysquilla crassispinosa (Fukuda), female from Madagascar; d-f, Bathysquilla microps (Manning), female from Hawaii. 442 Proceedings of the Biological Society of Washington margin. Telson with submedian swellings anteriorly, lacking distinct dorsal submedian carinae. Proximal segment of uropodal exopod with distinct distal spine dorsally, with 10-13 movable spines laterally. Basal segment of uropod (Fig. 1c) with spine ventrally at articulation of exo- pod, endopod with inner basal spine. Remarks: Although Manning (1969a:95) suggested that the Japanese and South African populations of this species might be distinct, based on discrepancies between published figures, Ingle and Merrett (1971) compared material from the two areas and could find no differences. We observed no differences between our specimens from Japan and Mada- gascar. On our specimens of B. crassispinosa, the intermediate carinae of the eighth thoracic somite may have a minute, sharp tubercle, but the carinae are not obviously armed posteriorly as in B. microps. The posterior seg- ments of the abdomen are more heavily granulated dorsally than in B. microps. The merus and propodus of the claw are strongly inflated in large males (TL 285 mm from Tosa Bay, Japan; reported by Ingle and Merrett, 1971), and the strongly curved dactylus has an angular pro- jection as in large males of Harpiosquilla (Manning, 1969b), but the antepenultimate tooth of the claw is not reduced as in B. microps. The eyes of B. crassispinosa, unlike those of B. microps and I. mani- hinei, are relatively large and the cormmea is fully pigmented. This must be a reflection of its occurrence in shallower waters than either of the other two bathysquillids. Komai (1938:270) noted that “the ground color is of the bright orange-red very commonly found in deep-sea crustaceans. The color is especially bright on the exposed thoracic and abdominal somites partic- ularly on the anterior and posterior margins of each somite. The cara- pace, raptorial limbs and telson are relatively light in color. On the ventral side the pleopods are crimson-red.” Bathysquilla crassispinosa is a large stomatopod, attaining a total length of almost 300 mm; only two other stomatopods, Lysiosquilla maculata (Fabricius) and Harpiosquilla raphidea (Fabricius), are known to grow to this size. Females of B. crassispinosa 200-297 mm long have been recorded in the literature; males appear to be only slightly smaller, with total lengths ranging from 245 to 280 mm. Our specimens have 2-3 movable and 10-12 fixed propodal spines. The propodal indices are similar to those of B. microps 076-078 in two large males examined (TL 250-285 mm) and 070 in a female 215 mm long. A feature which apparently has escaped notice in the past is the presence, in both species of Bathysquilla, of a minute movable tooth at the inferodistal angle of the outer surface of the carpus of the claw (Figs. 1b, e). So far as we are aware such spines are found in no other stomatopods. Distribution (Fig. 2): Indo-West-Pacific region, from Japan, Caribbean stomatopod off Hawaii 443 the western Indian Ocean, and off South Africa, in depths ranging from 230 to 310 m. Records include JAPAN: Sagami Sea (Fukuda, 1909, 1910); deep part of Sagami Sea (Komai, 1938); off Owase, ca. 300 m (Komai, 1927); Shikoku Island, Tosa Bay (Ingle and Merrett, 1971); Shikoku Island, Tosa Bay, 230-295 m.— MADAGASCAR: 18°54’S, 43°55’ E, 280-310 m.—MOZAMBIQUE: 25°12’S, 34°04’E; 230-295 m (Ingle and Merrett, 1971).—SOUTH AFRICA: Off Natal, north of Durban, 150 fm (275 m) (Calman, 1923; Gordon, 1929; Ingle and Merrett, 1971); off Durban, 29°42’S, 31°29’E, 132 fm (242 m) (von Bonde, 1932; Barnard, 1950; Ingle and Merrett, 1971). Bathysquilla microps (Manning, 1961) Figures ld-f, 2 Lysiosquilla microps Manning, 1961:693, fig. 5, pls. 10-11. Bathysquilla microps.—Holthuis 1967:28 [references]—Manning 1969a: 95, Figs. 26-28. Material: Gulf of Mexico, Bay of Campeche, Mexico; 700 fm (1281 m); Orercon II Station 10957: 1¢ TL 112.5 mm.—Gulf of Mexico, south of Panama City, Florida; 395 fm (723 m); Orecon II Station 10635: 19 TL 175 mm.—North Atlantic, off Surinam; 681-769 m; OrEcON II Stations 10604, 10620, 10796: 14 TL 187 mm, 2? TL 173- 220 mm.—Off French Guiana; 604-769 m; Orecon II Stations 10614, 10616, 10803, 10816, 10822: 4¢ TL 176-255 mm, 292 TL 183-195 mm.—Pacific Ocean, near Maui, Hawaiian Islands; 21°04’N, 156°09’W;; 731-786 m (400-430 fm); 12.5 m shrimp trawl; Towns—END CROMWELL Station 61-66; 26 October 1972: 19 TL 201 mm. Diagnosis: Cornea (Fig. 1d) indistinctly bilobed, set very obliquely on stalk, pigmented part reduced to small, transverse bar. Rostral plate (Fig. 1d) not grooved longitudinally. Antennal protopod with 2 papil- lae. Dactylus of claw with 13-15 teeth (14 in Maui specimen). Carpus of claw (Fig. le) with 1 dorsal spine. Intermediate carinae of body prominent, armed posteriorly on eighth thoracic and usually on second to sixth abdominal somites, also on first somite in Hawaiian specimen. Third to fifth abdominal somites variously ornamented with spinules on posterior margin. Telson with distinct dorsal submedian carinae. Proximal segment of uropodal exopod unarmed dorsally, with 6-8 (6 only in Maui specimen) movable spines laterally. Basal segment of uro- pod (Fig. 1f) unarmed ventrally at articulation of exopod, endopod unarmed on inner margin. Remarks: The Hawaiian specimen of B. microps agrees in almost all respects with our material from the western Atlantic. The only major difference that we observed was that the Hawaiian specimen has armed intermediate carinae on the first abdominal somite, whereas the inter- mediate carinae of this somite are unarmed in all of our Atlantic spec- imens. This single difference in one specimen is not, in our opinion, enough to warrant recognizing a separate Hawaiian subspecies. 444 Proceedings of the Biological Society of Washington aytub9 = © Coo1 | 3OLIAY ONIN LRARNUAAOD YD ‘sprymbsAyyeq Jo UOHNqLysIp UMOUy *Z% “OILY yeuiyluDw (%) VIIINODSOGNI sdosiw = osourdsissosua @ VIIINDSAHLWE | m7 A) Caribbean stomatopod off Hawaii 445 The propodal index for five female B. microps (TL 173-220 m) ranges from 074-078; it is 074 in the female from Hawaii. Propodal indices of males of B. microps from the Atlantic (TL 68-255 mm) range from 074 to 092, the higher indices being found in the smaller specimens. The female from Hawaii has 2-3 proximal movable spines and 12-13 fixed spines on the opposable margin of the propodus of the claw. In Atlantic specimens, there are 2 movable proximal spines and 11-15 fixed spines; of 20 claws examined, 16 were armed with 13-14 teeth. The dactylus of the claw is armed with 13-15 teeth in Atlantic speci- mens, 14 in the female from Hawaii. There are 7-8 movable spines on the proximal segment of the uropod in Atlantic specimens, 6 in the female from Hawaii. Bathysquilla microps may be a slightly smaller species than B. crassispinosa which attains a total length of almost 300 mm. Males of B. microps have total lengths of 68-255 mm, and females ranging from 44.5 to 220 mm are known. The color of the female specimen from Hawaii is similar to that of Atlantic specimens, based on photographs by J. E. Randall, Bernice P. Bishop Museum, D. Opresko, University of Miami, and B. Rohr, NMFS. On the Hawaiian specimen the anterior part of the body, from about the level of the antennae and including the eyes, is white; the rostral plate, like the carapace, is pink. The proximal portion of the propodus of the claw is white, the distal portion is pink, and the dactylus is orange. The proximal half of the walking legs is whitish, the distal part is pink. The remainder of the body is pink with orange carinae on the posterior portion of the body. Atlantic specimens show a similar pattern, but specimens have been taken in two color phases, red and orange. Of the more than 300 shrimp trawl stations effected in the Hawaiian Islands by NMFS during the period 1967-72, the Bathysquilla capture occurred at the deepest successful station. Of the total of 300 stations, about 55 were in the 300-800 m depth range. Thus, it would appear that the minimum depth inhabited by Bathysquilla in the Hawaiian Island is about 750 m. At this depth the bottom temperature is approx- imately 5° C. An account of the recent NMFS surveys and the general ichthyological results are given by Struhsaker (1973). Taxa of fishes associated with the Hawaiian capture of Bathysquilla were: sharks (Apristurus, Etmopterus); the eel families Congridae and Nettastomatidae; Halosauridae; Neoscopelidae; Ogcocephalidae; Ophid- ioidei; Macrouridae (Hymenocephalus, Malacocephalus, Matacocephalus, Nezumia, Trachonurus, Ventrifossa); Cynoglossidae. The ophidioid fishes represent a species previously recorded only from the western Pacific and the Indian Ocean (Cohen, 1974). Also taken in the haul were three or four species of pandalid shrimps and 25 kg of discarded military ordnance. During the 1902 expedition to the Hawaiian Islands of the U.S. Fish Commission steamer ALBATROSS, no specimens of Bathysquilla were 446 Proceedings of the Biological Society of Washington taken. Of the 257 beam trawl stations occupied during that expedition, 88 were in the 700-2800 m depth range. Bigelow (1931) reported on the stomatopods collected during that expedition and recorded only three species which occurred on the upper slope: Echinosquilla guerinii (White) in 55-335 m, Odontodactylus brevirostris (Miers) in 55-439 m, and O. hawaiiensis Manning (as O. japonicus) in 110-278 m. Townsley (1953) recorded E. guerinii from depths of 92-220 m; all of the other stomatopods reported by Townsley from Hawaii were taken in compara- tively shallower water. Distribution (Fig. 2): Western Atlantic, from localities near Florida, in the Gulf of Mexico, and off northern South America, in depths be- tween 604 and 1281 m, and off Maui, Hawaiian Islands in 731-786 m. Records include: BAHAMAS: Santaren Channel (24°24’N, 80°00’W), 732-860 m (Bullis and Thompson, 1965; Manning, 1969a).—FLOR- IDA: Off Cape Canaveral (28°03’N, 78°44’W), 915-952 m, and south- east of Tortugas (24°11’N, 83°21.5’°W), 732 m (Manning, 1961, 1969a; Bullis and Thompson, 1965); south of Panama City (28°12’N, 86°09’W), 723 m.—MEXICO: Bahia de Campeche (21°31’N, 96°46’W), 1281 m.— SURINAM: 07°49’N, 54°22’W, 732 m; 07°47’N, 54°05’W, 769 m; 07°53’N, 54°04’W, 681-732 m.—FRENCH GUIANA: 07°37’N, 53°32’ W, 723 m; 07°35’N, 53°32’W, 604-769 m; 07°32’N, 53°22’W, 641-659 m; 07°18’N, 52°59’W, 668-705 m; 07°06’N, 52°44’W, 668 m.—HA- WAII: Off Maui (21°04’N, 156°09’W), 731-786 m. GENERAL DISCUSSION The discovery of a population in Hawaii of a species previously known from the western Atlantic raises many interesting questions, none of which can be answered with our current state of knowledge. Since bathysquillids represent an old stomatopod stock (see below), it seems likely that the Hawaiian and American populations are relicts of a more widely distributed species, but that the Atlantic populations colonized the Pacific or that the reverse occurred cannot be ruled out. It seems likely that two distinct populations now occur in the two areas, for gene flow between deepwater benthic populations in the western Atlantic and the central Pacific must be minimal, but if genetic differences exist they are not strongly reflected in morphological features. Too, geo- graphically intermediate populations may well exist; the fauna of the upper slope is not that well known in most areas of the world. Several species or species groups of slope fishes exhibit distributions similar to that of B. microps. The macrourid Hymenocephalus aterrimus is known only from the tropical western Atlantic and Hawaii in depths of 450-900 m. Marshall and Iwamoto (1973) state that there are virtually no differences between the populations in the two oceans except for a tendency for a slightly greater interorbital width in the Hawaiian spec- imens. They did not feel that subspecific or other recognition of the populations were required. Iwamoto (1974) recently described a new Caribbean stomatopod off Hawaii 447 subgenus and species of macrourid presently known only from the Gulf of Mexico, the Caribbean (732-1062 m), and the Hawaiian Islands (623-705 m). He apparently did not find any morphological differences in the two populations of Nezumia (Kuronezumia) bubonis. This sub- genus also comprises a second species which occurs in the South China Sea (Iwamoto, 1974). The plectognath fish genus Hollardia is represented in the tropical western Atlantic by H. meadi (Bahamas, Cuba, and Barbados: 135- 450 m) and H. hollardi (Bermuda, Florida Keys, Gulf of Mexico, Carib- bean: 225-915 m). Tyler (1968), in describing H. goslinei from Hawaiian waters (365 m), states that it and H. hollardi are closely related and have differentiated along the same lines in several important respects, while H. meadi differentiated less from the ancestral type, re- maining more generalized than the other two species. He also discussed the peculiar distribution of the genus and speculated that the migration of the hollardiins into what is now the tropical western Atlantic must have beeen eastward through the Pacific with goslinei being the only presently known Pacific remnant with the extinction of eastern Pacific populations after the reemergence of Central America. Finally, the Hawaiian endemic bothid flatfish Chascanopsetta prorigera was recently reported from the central western Atlantic by Gutherz (1967) on the basis of a still unpublished manuscript. Preliminary comparison of material from the two regions indicates to us that the western Atlantic population deserves recognition as a distinct species. However, these two forms are very close morphologically, and together constitute a distinctive group within the genus. Such broad distributional patterns are generally rare within the stomatopods. Of approximately 300 species now recognized, only one, Heterosquilla mccullochae (Schmitt), can be considered to be pantrop- ical, occurring in the eastern Pacific, the western Atlantic, and the Indo- West-Pacific regions. Five other species occur in both the Atlantic and the Indo-West-Pacific region but not in the eastern Pacific. Broad horizontal distribution patterns also appear to be rare, in general, in inhabitants of the upper slope (Ekman, 1953; Briggs, 1974). Briggs (1974:435) has pointed out that “many ancient, phylogenetic relicts have accumulated in the slope habitat,’ and documented this with material recently reported in the literature on a variety of animal groups. The bathysquillids also represent one of these groups, an old stomatopod stock that occurs on the upper slope. The stomatopods are not well represented in the fossil record (Holthuis and Manning, 1969), and there is little evidence to indicate what are old or primitive charac- ters in the group. However, the primary distinction between the single fossil family, Sculdidae, and the four recent families, is the presence of an unsegmented exopod in the sculdids. As Ingle and Merrett (1971) noted, one of the characteristics of Indosquilla manihinei is its uropodal exopod, with the distal segment indistinctly sutured rather than distinctly 448 Proceedings of the Biological Society of Washington articulated as in all other recent stomatopods. It seems likely that the bathysquillids do represent an ancient stomatopod stock. LITERATURE CITED Barnarp, K. H. 1950. Descriptive list of South African stomatopod Crustacea (mantis shrimps). Ann. S. African Mus. 38:838—864, figs. 1-4. BicELow, R. P. 1931. Stomatopoda of the southem and eastern Pacific Ocean and the Hawaiian Islands. Bull. Mus. Comp. Zool., Harvard 72:105-191, figs. 1-10, pls. 1-2. Von BonbeE, C. 1932. Report No. 9 for the year ending 1931. Rep. Fish. Mar. Biol. Surv. S. Africa 9:1-128. Briccs, J. C. 1974. Marine zoogeography. McGraw-Hill. 475 pp. Buus, H. R., Jn., AND J. R. THompson. 1965. Collections by the exploratory fishing vessels Oregon, Silver Bay, Combat, and Pelican made during 1956 to 1960 in the southwestern North Atlantic. U.S. Fish Wildl. Serv., Spec. Sci. Rep.: Fish. 510: iii + 130 pp. Catman, W. T. 1923. Preliminary report on Crustacea procured by the S.S. “Pickle.” VI. S. Africa Fish. Mar. Biol. Surv., Res. Rep. 8 (1922 ):1, Cuopra, B. 1939. Stomatopoda. John Murray Exped., Sci. Rep. 6: 137-181, figs. 1-13. Conen, D. M. 1974. The ophidioid fish genus Luciobrotula in the Hawaiian Islands. Pacific Sci. 28:109-110. EKMAN, S. 1953. Zoogeography of the sea. Sidgwick & Jackson, London. 417 pp. Fuxupa, T. 1909. The Stomatopoda of Japan [in Japanese]. Dobut- sugaku Zasshi 21:54—-62, 4 pls. 1910. Report on Japanese Stomatopoda, with descriptions of two new species. Annot. Zool. Japonenses 7:139-152, pl. 4. Gorpon, I. 1929. Two rare stomatopods of the genus Lysiosquilla Dana. Ann. Mag. Nat. Hist. (10)4:460—462, figs. 1-2. GurHerz, E. J. 1967. Field guide to the flat fishes of the family Bothidae in the western North Atlantic. U.S. Fish Wildl. Serv., Cire. 263:iv + 47 pp. Hournuis, L. B. 1967. Stomatopoda I. Fam. Lysiosquillidae et Bathy- squillidae. Pages 1-28 in H.-E. Gruner, and L. B. Holthuis, eds. Crustaceorum catalogus. Pars 1. Den Haag, Dr. W. Junk N. V. , AND R. B. MaAnninc. 1969. Stomatopoda. Pages 535-552, figs. 343-363, in R. C. Moore, ed. Treatise on invertebrate paleontology, Part R, Arthropoda 4. INGLE, R. W., aNd N. R. Merretr. 1971. A stomatopod crustacean from the Indian Ocean, Indiosquilla manihinei gen. et sp. nov. (Family Bathysquillidae) with remarks on Bathysquilla crassi- Caribbean stomatopod off Hawaii 449 spinosa (Fukuda, 1910). Crustaceana 20(2):192-198, figs. 1-9, pl. 1. Iwamoto, T. 1974. Nezumia (Kuronezumia) bubonis, a new subgenus and species of grenadier (Macrouridae: Pisces) from Hawaii and the western North Atlantic. Proc. Calif. Acad. Sci. 39(22): 507-516. Kemp, S. 1913. An account of the Crustacea Stomatopoda of the Indo- Pacific region based on the collection in the Indian Museum. Mem. Indian Mus. 4:1—217, pls. 1-10, 10 text-figs. Komal, T. 1927. Stomatopoda of Japan and adjacent localities. Mem. Coll. Sci. Univ. Kyoto, Ser. B 3(3):307-354, figs. 1-2, pls. 13- 14, 1938. Stomatopoda occurring in the vicinity of Kii Peninsula. Annot. Zool. Japonenses 17:264—275, figs. 1-3. MANNING, R. B. 1961. A new deep-water species of Lysiosquilla (Crustacea, Stomatopoda) from the Gulf of Mexico. Ann. Mag. Nat. Hist. (13)3:693-697, fig. 5, pls. 10-11. ——. 1969a. Stomatopod Crustacea of the western Atlantic. Stud. Trop. Oceanogr. 8:viii + 380, figs. 1-91. ——. 1969b. A review of the genus Harpiosquilla (Crustacea, Stomatopoda), with descriptions of three new species. Smith- sonian Contrib. Zool. 36:1-—41, figs. 1-43. MarsHALL, N. B., AND T. Iwamoto. 1973. Family Macrouridae. Fishes of the western North Atlantic. Sears Found. Mar. Res. No. 1, Part 6. 698 pp. STRUHSAKER, P. 1973. A contribution to the systematics and ecology of Hawaiian bathyal fishes. Ph.D. thesis, University of Hawaii, Honolulu. xv + 482 pp. Towns.ey, S. J. 1953. Adult and larval stomatopod crustaceans occur- ring in Hawaiian waters. Pacific Sci. 7(4) :399-437, figs. 1-28. TyLer, J. C. 1968. A monograph on plectognath fishes of the super- family Triacanthoidea. Monograph 16, Acad. Nat. Sci. Philadel- phia. 364 pp. 450 Proceedings of the Biological Society of Washington SN] 9. 39, pp. 451-466 12 October 1976 PROCEEDINGS OF THE wiOLOGICAL SOCIETY OF WASHINGTON AN ANALYSIS OF PEROMYSCUS DIFFICILIS FROM THE MEXICAN-UNITED STATES BOUNDARY AREA By Victor E. DIERSING Museum of Natural History, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801 In April of 1964 and 1969 Woodrow W. Goodpaster, collect- ing for the Museum of Natural History, University of Illinois, obtained 15 specimens of Peromyscus of uncertain taxonomic status from the Franklin Mountains, El] Paso County, Texas. Numerous mammalogists examined these 15 specimens, of- fered suggestions for possible affinities, but never were cer- tain enough to assign specific identification. Subsequent ex- amination of these and additional specimens, some of which were from the Franklin Mountains, proves that they are Peromyscus difficilis, that this species occurs in mountainous areas near the Rio Grande River, and that they are referable to a distinct subspecies. This subspecies is called Peromyscus difficilis penicillatus Mearns (Peromyscus boylii penicillatus of Mearns). To establish a means of species recognition, comparative studies using 20 skin and skull features were made of samples of P. polius, P. boylii rowleyi, P. pectoralis laceianus, P. truei truei, P. truei comanche, P. attwateri, and P. difficilis. In addition cranial and external morphology, along with cyto- logical evidence was used in studying intrapopulation and interpopulation variation in five subspecies of P. difficilis: P. d. nasutus, P. d. griseus, P. d. penicillatus, P. d. difficilis, and P. d. petricola. METHODS Over 1000 specimens representing six species of the genus Peromyscus were studied. External measurements were taken 39—Proc. Brot. Soc. WasuH., Vou. 89, 1976 (451) 452. Proceedings of the Biological Society of Washington from the specimen label and 15 cranial measurements were taken by me with a dial caliper and recorded in millimeters. Head and body length was calculated. The 15 cranial mea- surements were: greatest skull length, basilar length, breadth of braincase, mastoid breadth (distance between lateral ex- pansions of mastoid processes), zygomatic breadth (greatest distance between the outer borders of the zygomatic arches ), bullae breadth (least distance between the ventral margins of the bony external auditory meatuses), bulla length (least distance from the antero-ventral margin of the bony external auditory meatus to the point of exit of the eustachian tube from the bony auditory bulla), least interorbital constriction, nasal length, palatal length, palatal foramen length, diastema length, post-palatal length, maxillary toothrow length (actual length of the molar toothrow measured near the crown of the teeth), and skull depth (taken as described by Hooper, 1952:10). All other measurements were taken as illustrated by Hoffmeister (1951:28). Ageing of specimens was based on wear of the upper molar toothrow. Ageing is discussed in “Nongeographic Variation.” SPECIES RECOGNITION Specimens of P. d. penicillatus from the Franklin Moun- tains, El Paso County, Texas, are like specimens of other subspecies of P. difficilis in having a tail usually longer than the head and body, large hind feet, ears of moderate length, a broad braincase, long maxillary toothrow, and moderately inflated auditory bullae. P. difficilis penicillatus differs in coloration from P. boylii rowleyi in having uniform pale gray upper parts with the sides and top of the head a noticeable lighter gray than the gray body, rather than ochraceous sides and brownish upper parts and head; externally in a longer tail, hind feet, and ears; cranially in a broader braincase, greater mastoidal breadth, more inflated auditory bullae, and shorter nasals; chromosomally four pairs of large biarmed autosomes rather than one pair of large biarmed autosomes. See Figure 1 for comparisons of maxillary toothrow length vs. bullae breadth. P. d. penicillatus differs from P. pectoralis laceianus ex- Peromyscus—Mexican boundary area 453 ternally in having a longer tail, hind feet, and ears; cranially in a broader braincase, greater mastoidal breadth, more in- flated auditory bullae, longer maxillary toothrow, and greater skull depth. See Figure 1 for comparisons of maxillary tooth- row length vs. bullae breadth. P. d. penicillatus ditfers from P. attwateri externally in having a shorter head and body and shorter hind feet; cranially in a shorter skull, smaller zygomatic breadth, smaller interorbital constriction, shorter nasals, and shorter diastema; chromosomally four pairs of large biarmed autosomes rather than three pairs of large biarmed autosomes. P. d. penicillatus ditters from P. truei truei and P. truei comanche externally in having a longer tail (much shorter in P. t. truei and slightly shorter in P. t. comanche), and shorter ears; cranially in less inflated auditory bullae, longer palate, and longer maxillary toothrow; chromosomally four pairs of large biarmed autosomes rather than five pairs of large bi- armed autosomes. See Figure 1 for comparisons of maxillary toothrow length vs. bullae breadth. P. t. comanche, con- sidered a distinct species by Johnson and Packard (1974), is here considered a subspecies of P. truei following Schmidly (O73) P. d. penicillatus ditiers from P. polius externally in having a shorter head and body, and shorter hind feet; cranially in a shorter skull, narrower braincase, smaller mastoidal breadth, smaller zygomatic breadth, shorter nasal, shorter diastema, and shorter maxillary toothrow. HABITAT During the progress of this study, specimens of P. dif- ficilis were taken by me at 14 different localities representing 10 counties in Arizona, New Mexico, Oklahoma, and Texas. Specimens were taken as low as 4700 ft, Franklin Mountains, Texas, and as high as 8200 ft, Chiricahua Mountains, Ari- zona. The parameter in common to all areas where P. difficilis was collected was the “rockiness” of the habitat. Adult in- dividuals were invariably taken in rocky situations such as lava beds, talus slopes, areas of numerous jumbled rocks, or 454 Proceedings of the Biological Society of Washington Length Maxillary Toothrow (mm) 82 84 86 88 90 92 94 96 98 100 102 104 106 108 10 Bullae Breadth (mm) Fre. 1. Individual specimens of four species of Peromyscus plotted according to bullae breadth vs. maxillary toothrow length, on a scatter diagram. P. b. rowleyi (indicated by solid circles on the scattergram ), Franklin Mtns., Texas; P. d. penicillatus, Franklin Mtns., Texas, open circles; P. p. lacieanus, vic. Carlsbad, Eddy County, New Mexico, closed triangles; P. t. truei, vic. Winslow, Navajo County, Arizona, open tri- angles. any other extensively rocky area having many fissures. Speci- mens taken outside such areas were usually immature in- dividuals, probably driven into submarginal habitat by popu- lation pressure. In these rocky areas where P. difficilis was abundant, P. boylii was absent. P. boylii was abundant in adjacent areas of lesser rock density, and therefore more vegetated. It then seems that niche segregation of these two species is largely determined by the density of the rocks and the amount of interspace within those rocks. NONGEOGRAPHIC VARIATION Age variation: Analyses were done only on those indi- viduals in adult pelage which had considerable wear on all Peromyscus—Mexican boundary area 455 bh AQ GQ An ne OA NA AR AR fa an GA AAR OH OA AA AK Fic. 2. Karyogram of P. difficilis penicillatus, female, UI 50529, from Mckelligan Canyon, Franklin Mountains, El] Paso County, Texas. lingual cusps of all upper molars with at least moderate wear on all labial cusps. Individual variation: Coefficients of variation for 20 skin and skull characters taken from a sample of 35 adult speci- mens from the Franklin Mountains, E] Paso County, Texas, indicate, that all characters exhibited normal variability (Table 1). External measurements were more variable than cranial measurements. Individual variation of the Franklin Mountains population of P. difficilis was in agreement with reported variation in other species of Peromyscus (Hoft- meister, 1951, Schmidly, 1972, 1973b). Secondary sexual variation: Of the 20 characters given in Table 1, ear length was sexually dimorphic at the .05 level; greatest skull length, and basilar length were sexually di- morphic at the .02 level; mastoidal breadth, and post-palatal length were sexually dimorphic at the .01 level. The 15 other characters were not significantly sexually dimorphic. Males ‘al Society of Washington 456 Proceedings of the Biologic fo) tas) tndop [4 cs°0 196 =L60 IT OT ST 663 ESO 0'OT ST cs°0 LIG «6600 —S FID 0G sso tT0 LT SI WSusp AMo}O0] Arvyp ey «96°C 99T 910 196 61 GcL@ LEO 86 SI (Busy eyeed-ysog Ss'T Foe 170 069 0G 3S = OVO EOL gI (psus| vULSIseIq] CST es¢ = 9T0 99'S 0G gee €30 BL cI Heae | MeeaOr ele 100 sI'b 8T0 = LEY id Ib 0 STO LEY ST wpaug] [ere[ed go'T 16S O€0 EOI 0G SFG 9G0— SFOL ST tpeHo| TBSEN 80 Soo = OT'0 SF 0G CLS = TOC‘ me One Oo <4 OS faba Se} > re a v7) S oJ av atv er ares ae) oy = w : ONE S. AS —= © wd B@ oN oD oe = =ie) a Pa O8 SO Bo Ge Se Le ) s a oO : bre BQ owas Se f & oR On amines HH pee: MY OH m8 42Ao mse Ze Ae FY ) P. d. nasutus, Chiricahua Mountains, Cochise Co., Arizona 97.2 119.5 93:91 21.74 98.44 13.92 12.34 8.34 LOW. -743) 9485 449 7.24 0.54 1.30 0.79 0.22 0.22 0.25 0.37 0.36 0.10 462 644 226 598 2.78 1.66 1.78 2.83 3.45 4.85 2.30 17 16 igs lig 17 17 hy ily 17 17 17 (2) P. d. nasutus, vic. Springerville, Apache Co., Arizona 100.3. 100.0 23.42 21.96 28.91 13.24 12.31 9.47 11.37 7.50 4.26 3.69 4.19 0.76 0.91 0.47 0.20 0.20 0.17 0.31 0.19 0.12 3.68 4.19 326 4.16 1.61 152 1.63 1.81 2.70 2.53 2.74 12 11 12; 12 14 14 14 13 14 14 14 (3) P. d. nasutus, Capitan, Sacramento, and Jicarilla mountains, New Mexico 99.7 107.5 22.86 22.29 28.89 13.16 12.45 9.35 11.23 7.39 4.11 5.02 5.09 0.78 1.60 0.43 0.25 0.21 0.30 0.36 0.22 0.15 oo4 4.73 3Al 7.18 149 1.90 169 321 3.21 29s) 4.07 20 21 21 7 19 20 19 19 20 18 20 (4) P. d. griseus, Carrizozo lava beds, Lincoln Co., New Mexico 94.6 101.9 23.07 21.34 28.19 12.91 12.18 9.30 11.12 7.25 4.16 3.93 5.13 069 1.12 0.50 0.26 0.24 0.27 0.30 0.24 0.09 AIS . 503. 2.95 3.25 L738 1.98 LST 2.68 “271 B27 soe At 44 44 tf 43 44 tt 42 43 tt 44 (5) P. d. penicillatus, Franklin Mountains, Texas and New Mexico 94.2 l11.7 23.76 20.24 27.74 12.85 12.07 9.30 10.40 6.96 4.16 3.25 6.97 0.70 0.74 059 025. 0.20 .019> 029 "022750 3.4 48 6.24 2.94 3.66 2.11 1.91 169 2.01 2.74 3.13 2.57 34 34 34 34 33 34 33 32 34 34 34 (6) P. d. penicillatus, Sierra del Carmen, Coahuila 91.0 110.5 22.42 21.33 27.80 12.96 11.90 8.92 10.76 7.16 4.20 10.9 7.98 0.28 0.52 0.24 0.16 O17 ° 0.26 0.17 0.10 12.0 7.23 2.37 1.87 1.81 1.38 191 2.56 2.42 2.38 12 12 12 3 8 9 8 8 12 12 12 Peromyscus—Mexican boundary area 46] TaBLeE 2. (cont.) Lara} — 6 = bo x = S es) Nn —1 lol ~ ay n— Se Osc pel cay too] — + o +5 PS 4) = 4 ay es = ws ine} 1S) Ors Sea a oe aS Fo Is Lo} ed’ 2 & TS) nu nd atk mw rene oe = ey ; oo So HS = a noo 8 oD S oo ine) x =e) be IIE One Sto SO oe) sa 6a Oo ro) (ay) ‘en | iss] tal a -u S| ~ eo) 3 eg i a ey cape te, apes) |e eda) ial Zo AY a (7) P. d. difficilis, vic. Creel, Chihuahua, Mexico 97.6 121.1 25.00 26.03 29.30 13.28 12.47 9.22 11.70 7.60 4.36 5.79 8.29 0.96 1.94 0.42 023 020 0.24 033 0.22 0.11 5.93 685 382 745 143 1.75 161 2.63 2.80 2.87 2.61 19 19 iL) 19 20 19 19 Lg 20 20 18 (8) P. d. petricola, vic. Sierra Guadalupe, Coah., Cerra Potosi, Nuevo Leon 94.71 23.17 29.89 13.40 12.98 9.70 11.63 7.42 4.45 yee 7.03 1.07 1.60 0.49 020 0.19 0.25 0.22 0.17 0.17 81 4.33 691 1.64 149 146 258 232 229 3.82 14 14 14 6 HD; 1} 13 iil 1b 14 14 (9) P. d. petricola, Concepcion del Oro, Zac., vic. Sierra Encarnacion, Coah. 103.8 116.3 24.76 22.13 29.65 13.32 12.86 9.72 11.66 7.63 4.50 7.50 1055 0.81 064 050 024 025 024 0.23 0.14 0.20 7.22 9.07 3.27 289 169 1.80 1.94 247 1.97 1.83 4.44 17 IEF 1 8 16 17 16 16 17 17 ily Variation was not uniform, with areas of differentiation corresponding to large desert basins or areas of low relief. See Table 2 for measurements of all samples of P. difficilis. Samples from the Chiricahua Mountains, Arizona (sample 1 of Figs. 4, 5, and Table 2); Springerville, Arizona (sample 2); and from the Capitan, Sacramento, and Jicarilla moun- tains, New Mexico (pooled as sample 3); agree with each other and differ from the Franklin Mountains’ population of P. difficilis in having an overall larger skull, broader brain- case, greater mastoidal breadth, longer diastema, and in be- ing darker in coloration. The Chiricahua Mountains’ popula- tion of P. difficilis are unique as a sample in having a longer tail in combination with smaller auditory bullae but are here referred with Springerville and Capitan samples. Sample 4 from the Carrizozo lava beds, Lincoln County, New Mexico, differ from the Franklin Mountains’ population 462 Proceedings of the Biological Society of Washington of P. difficilis in having a shorter tail, longer ears, a longer diastema, longer nasals, and in being much darker in color- ation. Sample 7 from the vicinity of Creel, Chihuahua, Mexico, differs from the Franklin Mountains population in having a longer tail, longer hind feet, and much longer ears, an overall larger skull, actually smaller auditory bullae, and in being darker in coloration. Samples § and 9 from northern Zacatecas, southern Coa- huila, and west-central Nuevo Leon, Mexico, differ from the Franklin Mountains’ population of P. difficilis in having a longer tail, longer hind feet, longer ears, a much larger skull with especially larger mastoidal breadth, greatly in- flated auditory bullae, and in being darker in coloration. Sample 6 from the Sierra del Carmen Mountains, Coahuila, are here referred to the subspecies P. d. penicillatus. Only a small sample is available, but the specimens at hand com- pare with the Franklin Mountains’ population in coloration, all external measurements, and in all cranial measurements, except the specimens from the Sierra del Carmen Moun- tains have the diastema averaging slightly longer and notice- ably smaller auditory bullae than the Franklin Mountains’ population of P. d. penicillatus. Peromyscus difficilis penicillatus Mearns Peromyscus boylii penicillatus E. A. Mearns, 1896, Preliminary di- agnoses of new mammals from the Mexican border of the United States, p. 2, May 25 (preprint of Proc. U. S. Nat. Mus., 19:139, December 21, 1896). Peromyscus boylii rowleyi, W. H. Osgood, 1909, A revision of the mice of the American genus Peromyscus. N. Amer. Fauna, 28:145, April 17. Type locality: Foothills of the Franklin Mountains, near EI Paso, El Paso County, Texas. Type, skin and skull, USNM 20034/35426. Holotype examined. Range: Known from the Franklin Mountains (extending from EI] County, New Mexico), Sierra del Carmen Mountains, northern Coa- huila, Mexico, and a single specimen, cat. no. 21129/37194, U. S. Nat. Mus., collected by E. A. Mearns from Dog Springs, Grant County (now Hidalgo County), New Mexico. Other populations might exist on other isolated mountain ranges in northern Coahuila, western Peromyscus—Mexican boundary area 463 Texas, southern New Mexico, and northwestern Chihuahua. Collecting at Dog Springs in the summer of 1974 produced only Peromyscus boylii and Peromyscus eremicus. Diagnosis: A subspecies of P. difficilis recognized externally by its uniform gray upper parts, white underparts, and long penicillate tail. Paso, El! Paso County, Texas, north for 10 miles into Dona Ana The sides and top of the head is noticeably lighter gray than the gray body. Cranially the smallest subspecies of P. difficilis with short nasals and short diastema. Karyotype of four pairs of large biarmed, two pairs of small biarmed, and 17 pairs of acrocentric autosomes. Comparisons: Peromyscus difficilis penicillatus differs from P. dif- ficilis griseus, P. d. nasutus, P. d. difficilis, and P. d. petricola, in having an exceedingly short skull (reflected in having shorter nasals and shorter post-palatal length), and in its light gray head and body coloration rather than a brown to brownish-black coloration. It dif- fers from P. d. nasutus and P. d. griseus in having a relatively long tail (119% head and body length rather than 98 to 109%), and dif- fers trom P. d. petricola in having shorter ears, smaller auditory bullae, and smaller mastoidal breadth. See Figure 5 for comparisons. Remarks: Osgood (1909:145) in synonymizing P. d. penicillatus (then P. boylii penicillatus) with P. boylii rowleyi says that “the type of penicillatus is an abnormally pale individual, but a series from the Franklin Mountains near the type locality does not differ from typical rowleyi.” In actuality the type of P. boylii penicillatus that Osgood examined was a specimen of P. difficilis and the series that he talks about were P. boylii, which also occur in the Franklin Moun- tains, and which are referable to P. b. rowleyi. Specimens examined: Total 175 from: TEXAS. Ext Paso Co.: 3% mi W Fort Bliss, Trans-Mountain Road, 5 (UI); McKelligan Canyon Park, 107 (UI); Head McKelligan Canyon, 4700 ft, 10 (KU). NEW MEXICO. DoNa Ana Co.: 16 mi N El Paso, 10 (UI); 15 mi N El Paso, 21 (UI); 15 mi N El Paso, 5 mi SE junction I-10 and HWG 404, 4 (UI); Hmatco Co.: Dog Springs, 1 (USNM). COAHUILA. 15 mi S, 25 mi E Bouquillas, 7300 ft, 3 (NTSU); Sierra del Carmen, Campo Madera, 8000 ft, 10 (9 USNM, 1 DMNH); Sierra del Carmen, Oso Canon, 3 (DMNH); Sierra del Carmen, Botellas Canon, 1 (DMNH). For all other specimens of P. difficilis, examined by me, the tax- onomy followed is that given by Hoffmeister and de la Torre (1961). Peromyscus difficilis griseus Benson Specimens examined: Total 185. NEW MEXICO. Lincotn Co.: 6% mi W, 5 mi N Carrizozo, 9 (UI); 5 mi W, 5 mi N Carrizozo, 21 (UI); 5 mi N, 4.75 mi W Carrizozo, 25 (UI); 4% mi N, 3% mi W Carrizozo, 25 (UI); 3 mi N, 2.75 mi W Carrizozo, 70 (UI); 3 mi W Carrizozo, 2 (UI); 3% mi W, 2 mi N Carrizozo, 33 (UI). 464 Proceedings of the Biological Society of Washington Peromyscus difficilis nasutus (J. A. Allen) Specimens examined; Total 259. NEW MEXICO. Sterra Co.: San Andres Mtns., N slope Salinas Peak, 6000 ft, 4 (USNM); San Andres Mtns., Bear Canyon, 2 (USNM):; San Andres Mtns., N slope Salinas Peak, 5000 ft, 3 (USNM). Lincotn Co.; Capitan Mountains, 24 (USNM); Corona, 1 (USNM); Jicarilla Mountains, 15 (USNM). Orero Co.: % mi W High Rolls, 13 (6 UI, 7 JD); 12 mi E Alamo- gordo, 1 (JD); Sacramento Mtns., 1 mi W High Rolls, 2 (UTEP): Russian Canyon, Sacramento Mtns., 5 mi S, 24% mi E Cloudcroft, 1 (UTEP); Dona Ana Co.: 35 mi NE Las Cruces, San Andres Can- yon, San Andres Mtns., 5600 ft, in tunnel of lead mine, 1 (USNM). Eppy Co.: Slaughter Canyon, 4.75 mi S, 11 mi W White City, 1 (UI); 26 mi W, 10 mi S Carlsbad, 1 (ENMU). ARIZONA. ApacHEe Co.: Springerville, 2 (USNM); 4 mi E Springerville, 60 (UI); 4% mi ESE Springerville, 44 (UI); 3 mi SE Springerville, 11 (USNM); 4% mi SSE Springerville, 33 (UI). Cocutse Co.: 1% mi E Fly’s Peak, Greenhouse Canyon, Chiricahua Mtns., 7500 ft, 15 (UI); Fly’s Peak, Chiricahua Mtns., 10 (UI); 4% mi E Buena Vista Peak, Chiricahua Mtns., 8100 ft, 2 (UL); 7 mi W Portal, 1 (JD); 1 mi NW Research Station, Chiricahua Mtns., 1 (UI); 1 mi SW Portal, 7 (UI); 1 mi up Cave Creek from Portal, 1 (UI). TEXAS. Cunperson Co.: Mce- Kittrick Canyon, 7800 ft, 1 (USNM); 2 mi E Pine Springs, 2 (UI). Peromyscus difficilis difficilis (J. A. Allen) Specimens examined: Total 42 from: CHIHUAHUA. 15 mi S, 6 mi E Creel, 7300 ft, 24 (KU); Divisadero, 16 mi S, 13 mi W Creel, 7500 ft, 14 (KU); N rim Barraneca del Cobre, 23 mi S, 1% mi E Creel, 7200 ft, 3 (KU); Churo, 7200 ft, 1 (KU). Peromyscus difficilis petricola Hoftmeister and de la Torre Specimens examined: Total 66 from: COAHUILA. Sierra’ Encar- nacion, 19 (USNM); Carneros, 2 (USNM); Sierra Guadalupe, 25 (USNM): 8 mi S Bella Union, HWY 57, 6 (UI). NUEVO LEON. Along road 1 mi NE microwave tower, E slope Cerra Potosi, 1 (USNM): .75 mi NE microwave tower, E slope Cerra Potosi, 1 (USNM): Rocks along road NE microwave tower, 9850 ft, E slope Cerra Potosi, 1 (USNM); Rock ledge, Cerra Potosi, 1 (USNM). ZACATECAS. 4 mi W Concepcion del Oro, 10 (UTI). CONCLUSIONS Peromyscus difficilis from the Franklin Mountains of Texas and New Mexico and from northern Coahuila are referable to a distinct subspecies for which the name Peromyscus boylii penicillatus Mearns, 1896, is available. Analyses of various cranial, external, and chromo- somal characters are presented to facilitate recognition of P. difficilis penicillatus from closely related species of Peromyscus. Methods are Peromyscus—Mexican boundary area 465 discussed for differentiating P. difficilis penicillatus from geographically peripheral subspecies of P. difficilis: P. d. griseus, P. d. difficilis, P. d. nasutus, and P. d. petricola. The habitat preterence of P. difficilis is described. ACKNOWLEDGMENTS Most importantly I thank Dr. Donald F. Hoffmeister for his con- tinued guidance in the preparation of this manuscript and for his unending service in obtaining the necessary specimens. I am indebted to Woodrow and Lois Goodpaster for their skill in collecting and preservation of numerous specimens; F. F. B. Elder who prepared chromosomes and karyograms; and Mitch Paulson for the drawing of figures used in this report. A special thanks is given to those institutions from which specimens were made available for study (abbreviations used in the manuscript are given in parentheses): Museum of Natural History, University of Illinois (UI); Museum of Natural History, University of Kansas (KU); National Museum of Natural History (USNM); Dallas Museum of Natural History (DMNH); Museum of Vertebrate Zoology, Uni- versity ot California (MVZ); Eastern New Mexico University (ENMU); New Mexico State University (NMSU); University of Texas at E! Paso (UTEP); North Texas State University (NTSU); private col- lection of James E. Diersing (JD). LITERATURE CITED HorrMeister, D. F. 1951. A taxonomic and evolutionary study of the pinon mouse, Peromyscus truei. Ilinois Biol. Monog. 21: ix + 1-104. , AND L. De La Torre. 1961. Geographic variation in the mouse Peromyscus difficilis. Jour. Mammal. 42:1-13. Hooper, E. T. 1952. A systematic review of the harvest mice (genus Reithrodontomys ) of Latin America. Mise. Publ. Mus. Zool., Univ. Michigan 77:1—255. Hsu, T. C., and F. E. Arricut. 1968. Chromosomes of Peromyscus (Rodentia, Cricetidae) I. Evolutionary trends in 20. species. Cytogenetics 7:417-446. Jounson, GERALD L., AND Ropert L. Packarp. 1974. Electrophoretic analysis of Peromyscus comanche Blair, with comments on its systematic status. Occ. Pap. Mus. Texas Tech. Univ. 24: 1-16. Lee, M. R., D. J. Scumipty, ann C. H. Hunery. 1972. Chromo- somal variation in certain populations of Peromyscus boylii and its systematic implications. Jour. Mammal. 53:697—707. Oscoop, W. H. 1909. Revision of the mice of the American genus Peromyscus. N. Amer. Fauna 28:1—285. ScumipLy, D. F. 1972. Geographic variation in the white-ankled 466 Proceedings of the Biological Society of Washington mouse, Peromyscus pectoralis. Southwest Natur. 17(2):113- 138. 1973a. The systematic status of Peromyscus commanche. Southwest Natur. 18:269-278. 1973b. Geographic variation and taxonomy of Peromyscus boylii trom Mexico and the southwestern United States. WD 40, pp. 467-476 12 October 1976 PROCEEDINGS OF THE wiJILOGICAL SOCIETY OF WASHINGTON MORE ON EARTHWORM DISTRIBUTION IN NORTH AMERICA By G. E. Gates! Tall Timbers Research Station Tallahassee, Florida 32303 The earthworm fauna of the United States has been much misunderstood both at home and abroad. Consideration of certain beliefs, rather commonly held in the United States, at least in the past, is the main purpose of the present contri- bution. Primarily involved is the Quaternary climate and the fact (Gates, 1970:9, No. 1) that all earthworms must have been exterminated, at the very least throughout the areas then covered with ice thousands of feet thick. Subsequently, in America as also in Europe, the native earthworms actively followed the retreating glacial ice north- ward (Smith, 1912, who merely stated a rather generally ac- cepted belief). Involved in any such northward migration theoretically there could have been included six genera in five families, three of which are solely American. Much more re- cently, European lumbricids supposedly replaced (in active competition?) native earthworms “as was described by Smith whose observations were supported by Goff, and has been commonly accepted” (Stebbings, 1962: 905). Of the six genera that could have been involved, Stebbings seems to have been concerned only with one, the acanthodrilid Diplocardia. Past misunderstandings, as well as present misconceptions, require em- phasis on the following: Native earthworms of any part of the world were unknown until well after European travels and settlements therein. Almost everywhere Europeans went, except in tropical lowlands and in arctic permafrost, earthworms from Europe eventually were recognized. 1 Mailing address, 251 Silver Road, Bangor, Maine 04401 40—Proc. Brion. Soc. WasH., Vou. 89, 1976 (467 ) 468 Proceedings of the Biological Society of Washington For example, consider Smith’s portion of central Illinois. Even as the first native American earthworm, Diplocardia communis, was being de- scribed, at least four European species, Aporrectodea rosea and A. trap- ezoides (if not also A. turgida and A. tuberculata), Eisenia foetida and Octolasion tyrtaeum, already had become so well domiciled as to be characterized as “frequent” to “abundant”, though Garman (1888) only mentioned three of them. Also, the only agent known to be engaged in transporting earthworms on a large and continuous scale (for centuries if not millenia) is man. Data for the report (Smith, 1928) that provided the basis for “the commonly accepted” belief mentioned by Stebbings were secured from study of a small area of glaciated Illinois centering around Champaign- Urbana. The worms were considered in three undefined classes: 1) “Woodland species”, better characterized as litter feeders. They move about on the surface of the soil while searching for those concentra- tions of organic matter in which they usually abound. No change in wood- land populations during the period involved, 1892-1927, was recognized. Only three species were mentioned (Idem, p. 349) as belonging to the group and two of them obviously are native; Bimastos gieseleri and B. hempeli. The third, now known as Dendrodrilus rubidus, has been found around the world in appropriate climates such as are furnished by South Africa, Australia, New Zealand, southern South America and various oceanic islands. 2) “Stream-bank species”. Of those so regarded by Smith, two are better characterized as limiphagous or limicolous. One, the European Eiseniella tetraedra only very rarely, and the other, the American Sparganophilus eiseni, never is naturally found away from saturated mud. Other worms listed as stream bank forms really are geophagous or litter feeders. Some of the former have shown a tendency to aggregate at or near mildly contaminated sites such as soil near or under cow-manure pats. Some of the litter feeders do adapt to polluted habitats like those along the banks of the sewage-contaminated stream that was studied. Along that stream bank was made the only continuously recorded survey for any part of the Champaign-Urbana area. During 1922-1923, a graduate student dug from 11 sites along the bank 5,134 worms. Number of collections at a site varied from one to 12 but usually was more than three. Information was not provided as to how close to the polluted water the worms were dug nor as to the liability to flooding at each site during any part of the year. Few native forms were obtained, and these at only two of the sites, neither of which had previously been searched! The species were D. communis (2 specimens) and D. singularis (94 specimens). The latter native was not again mentioned although earlier it had been said to be “common” in upland regions of central Illinois (Smith, 1915: 556). Are we supposed to assume without supporting evidence that those natives formerly had been present at each of the other nine sites? The largest number of specimens to be listed (Smith, Earthworm distribution 469 1928: 332, Table 1) was of a geophagous form called Helodrilus caligi- nosus trapezoides (now known as Aporrectodea trapezoides) but which could have comprised at least two other species. Each one of the three is geophagous, as common if not more so away from stream banks, and better characterized by its feeding. 3) “Upland-soil species”, better characterized also as geophagous because of their diet. Most of these do not ordinarily crawl about on the surface unless forced there during rains. Lumbricus terrestris may also sometimes be forced to the surface during rain but it alone feeds and copulates on the surface during the night when conditions are favorable to such activities. Although a number of species (including seven of those found at the stream bank sites) could have been considered here, the discussion was restricted to but two and then only with reference to individuals seen above ground during and after rain. Referring to one of the pair, the American Diplocardia communis Garman (1888), its author was quoted as follows: “Hundreds were seen in this locality, migrating during showers of rain.” Migrating seems a poor word for worms that probably were forced out of their abode, many perhaps to die the next day, as often happens. The other species, the European Lumbricus terrestris, was first seen in the same area (but only after rain and when it already may have become fairly well established) “probably about 1896”. Subsequently, the night crawler was thought to have be- come abundantly stocked (judged by observations after rainfall). Mean- while, the native species decreased until on the last night of recording in March, 1927, only 19 specimens were seen in the streets bordering 24 city blocks. Nevertheless, Smith did state that the American worm still was abundant in 1927 in areas further to the east where the night crawler was rarely seen (after rain). Geophagous earthworms do surface after the soil has been poisoned by dilute solutions of various chemicals, as is well known to those who must collect them without digging. Observations for more than 20 years at a single site in Bangor, Maine, indicate that any particular rain rarely, if ever, brings up each and every species known to be present. On the contrary, different rains produce different species on different occasions and in different percentages. So it can be suggested that the night crawler may have replaced the native D. communis in the lawns of the two Illinois cities for two reasons: First, because the soil-infiltrating, industrial poisons in the rain were more deleterious to the native than to the exotic form. Second, because the grass lawns of the cities may be more like the normal habitat for L. terrestris than for D. communis (cf. Harman, 1960: 66). All species of the genus Bimastos are native to the southern states of this country. One, B. longicinctus, was common in soils and parkings of Urbana when first described in 1915 (Smith, 1915: 537). Another, B. zeteki, was said at the same time to be common in central Illinois. All species of that genus as well as most of two other American genera 470 Proceedings of the Biological Society of Washington were ignored by Smith in his discussion of the subject under consider- ation. Even in 1928, “replacement” of the native species in central Tlli- nois was far from complete. How partial replacement may have been now seems to be indeterminable, in absence of information as to when natives of the three genera did reach central Illinois and how extensive their distribution was before man started bringing exotic species from all around the world to America after 1500 A.D. Certainly, Smith provided little basis for the “commonly accepted” replacements which Stebbings himself hesitated to accept. However, Stebbings’ doubts seem to have been mostly about conditions west of the Mississippi River, a region with which this contribution is not concerned. Smith did say (1928: 347), “The tendency is toward an increasing domination of European species, and a corresponding decrease in abun- dance of some indigenous forms.” However, it probably would have been more accurate to end that sentence in some such way as “. . . in city lawns and along banks of a sewage polluted stream.” Certainly, Smith did no more than suggest that a similar change might be under way in other parts of the state. Accordingly, parts of Stebbings’ discussion are irrelevant, as they seem to be based on an unwarranted (even if widely held?) assumption that the replacements supposedly found by Smith involved much, if not most, of all states in a central part of the country. Furthermore, Smith did not take into consideration the role of man in modification of the environment and its influence on earthworm faunas. Some such factors may be more important than competition between endemics (possibly hemerophobic) and exotics that are strongly hemero- philic (favored by human culture). But what evidence is there for a post-Quaternary northward earth- worm migration? And from where? The answer now suggested: None worthy of much serious consideration. Obviously, migrating natives never reached Canada which lacks a single endemic species and even several American litter-feeders now domiciled elsewhere in the world— or New England, with Massachusetts and Connecticut each having but one record of a native species and each at a site to which the species obviously was introduced—or New York, with one to several isolated records of three litter-feeding natives but no records of any geophagous natives. Murchie (1954) found the sole geophagous native, D. singularis, only at four closely spaced localities of three contiguous Michigan coun- ties. Of the American litter feeders, one was obtained from a single locality in each of five widely separated counties. Another was found at two localities of two widely separated counties. Another, B. longicinc- tus, was found at a single site of one county, just as in one county each of Illinois (Harman, 1960) and Ohio (Olson, 1928). In contrast, at least half a dozen European exotics are widely distributed throughout Michigan. A similar situation is shown by Olson’s (1928) maps for Ohio. Of his three geophagous natives, two were recorded in Ohio only from a small central area. A third did have a greater north-south distribution Earthworm distribution A471 but the area involved was much smaller than that occupied by each of a half dozen European exotics. Four litter feeders were shown as from one of three localities each, but two others were indicated only from 12-13 counties, mostly in the central part of the state. Data for American spe- cies from Michigan and Ohio, like that for Massachusetts and Connecticut, suggest more recent and fewer introductions than of the exotic Europeans. In Indiana, four geophagous natives probably were present (Joyner, 1960) in 1960. Subsequently two more were added. Fortunately, more recent results of three years’ collecting in 46 central Illinois counties are available. Included of course was the Champaign- Urbana area of Smith’s observations. One outstanding demonstration of the effects of man was provided by Harman’s (1960) finding that all but one of Smith’s stream bank species had disappeared since 1927 as a result of increased pollution. The sole survivor was the European manure worm, Eisenia foetida. Harman did think that D. communis (Idem: 66), though ranked sixth in number of times collected, “probably” was becoming less abundant but because of “present restriction to flood plains, cccasional uncultivated areas and along roadside”. Even so, the species was obtained in 20 collections from 16 counties (including Cham- paign) as against 21 collections from 15 counties for the supposedly replacing night crawler. Another geophagous native, D. singularis, mostly ignored in Smith’s discussion, was even thought by Harman (1960: 69) “to be increasing its distribution” in central Illinois where it was secured in 18 collections from 11 counties. Not mentioned by Smith in 1928 was another geophagous native, D. verrucosa, he earlier characterized (Smith, 1915: 536) as “abundant” and which was obtained by Harman (1960) in 16 collections from 15 counties. Even more interesting are the results of a Tennessee study (Reynolds et al., 1974). Their figures show that 15 species of European worms are present in every one of the 95 counties of that state. None of the Ameri- can species, whether litter feeders or geophagous, are as widely distrib- uted as two or three of the European. As European earthworms were introduced directly or indirectly by man into every one of the 95 Tennes- see counties, it is now possible to suggest that each of the native species, whether geophagous or litter feeders also could similarly have been brought into the state, and perhaps, less frequently and more recently. For more than a century, greenhouses, conservatories, etc., may have been importing and distributing exotic earthworms in the soil around the roots of live plants. (“During 1825-1860, wealthy estates in Tennessee and Kentucky had greenhouses. The Belmont mansion, near Nashville, had three buildings each 300 feet long.” Gates, 1966: 251.) Of the 23 European earthworm species now domiciled in North America, 18 of the most widely distributed frequently were intercepted (Gates, in MMS) at American ports of entry during the last 26 years. Each of the others is known only from one, two or several widely sep- arated American sites. Accordingly, and regardless of how Julin’s (1949) 472 Proceedings of the Biological Society of Washington habitat and life classifications are interpreted, each one of those coloniz- ing European species seemingly can be regarded as hemerophilic as they owe so much of their distribution to man. Detailed information that has been desired as to immediately post- Quaternary conditions in relation to earthworm life was not found in the literature. Answers were sought in vain to the following questions: Did not arctic gales, blowing for millenia across thousands of miles of thou- sand-foot thick ice, exterminate earthworms below the southernmost limit of glacial advance? If so, how far from the glacial boundary? Was there permafrost in the soil south of the glacial boundary? If so, at what depths and when did it finally disappear? How soon after disappearance of the ice would the deposited rock flour, sand, gravel and boulders have ac- quired enough organic matter to support geophagous earthworm popula- tions? Did the Appalachian mountain tops, even shortly, have local glaciers? If so, how many centuries were required for geophagous earth- worms to eat their way up to and then down the northern slopes of those mountains in order to digest their way through Tennessee and Kentucky into central Illinois? Originally, this author merely said (Gates, 1967: 174) “for as yet unknown distances below the southern ice face, the climate was too frigid for earthworms to survive.” Later on, it was sug- gested (Gates, 1970: 9, No. 2) that the area of supposed extermination may have included all of the area north of the Appalachians (unfortu- nately, again without attracting interest, discussion or repercussions ). Extermination is now suggested to have extended at least to the tops of the Appalachians if not also somewhat down on the southern side into what now appears to have been one of two earthworm refugia in North America. The other refugium comprises a narrow strip along the Pacific coast about from San Francisco to the Canadian border. Between that strip and the 100th meridian of longitude, or thereabouts, endemic earthworms are lacking. An American species accidentally introduced from elsewhere may occasionally be found. Yet, wherever there is water, European and Asiatic worms flourish in a vast area that includes the region once marked on maps as “The Great American Desert”. Efforts to obtain a geological explanation for the absence of native earthworms in such a large area have all been fruitless. The “rival hypotheses” of Omodeo and Gates discussed by Ball (1976), again involved the amphi-atlantic distribution of the Lumbricidae. Two of its genera are endemic in a southern part of the United States. All other lumbricid genera (to as many as 14 according to which classifica- tion is followed) are endemic in Eurasia, for the family reaches into Korea and Japan. The origin and evolution of the Lumbricidae has had less consideration than that of some other megadrile families. An eastern origin of the family, because of the greater number of genera there, may have been assumed. However, possibility of an American origin but with greater Quaternary exterminations than in Eurasia, perhaps should be Earthworm distribution 473 considered. When a North Atlantic bridge for the lumbricids was first suggested cannot now be stated. Undoubtedly, it was assumed by that master architect of bridge builders, Wilhelm Michaelsen. Such a bridge was acceptable to Stephenson who argued effectively against Michael- sen’s other bridges. “A bridge betweeen Europe and North America in comparatively recent times, over the most northerly part of the Atlantic, is, I think, well attested on geological grounds: it accounts for the presence of endemic lumbricids in the eastern United States” (Stephen- son, 1930: 688). The word “endemic” of that previous sentence requires emphasis. Only because of the presence of endemics on both sides of that ocean was that bridge at first thought to be necessary. Omodeo’s contribution (1963) involved: Lumbricid origin in Eurasia. Migration of existing European species across the north Atlantic to Greenland and America. Survival there, morphologically unchanged, on nunataks, during the glacial period. Migration of American worms along the same bridge to Europe at the same time the European species were crossing to America. The only genus that could be mentioned in that reverse direction was Sparganophilus. It is truly American, but is represented in Europe only at two sites in England and one in France, and there by the same species that in America (Jamieson, 1971: 814) extends from Central America to the Canadian shores of the Great Lakes. In marked contrast, European lumbricids reach all the way across North America both in the United States and Canada. The author of the “hypothesis” attributed to Gates cannot now be mentioned. It may never have been developed in a formally logical way but was merely expressed as a probability (of high degree), as by Beddard in his monograph (1895: 155). With the inclusion of such geographical names as present knowledge permits, the “probability” of Beddard can be stated as follows: Presence of lumbricids invariably identical with those of Europe, in South Africa, the hills of south India, Australia, New Zealand, North and South America and oceanic islands such as St. Helena, Bermuda, St. Paul’s Rock (Indian Ocean), Hawaii, etc., resulted from transportation; and by man. One attribute all such areas have in common is that Europeans have taken to each of those places live plants with their roots surrounded by earth. Before 1895 as well as subsequently, Beddard and others had commented on the earth- worms often contained in such earth. Thousands of earthworms were intercepted by the U.S. Bureau of Plant Quarantine during the last 25 years in unsterilized materials associated with the roots of live plants. Often included in such interceptions were 18 of the 23 European lumbri- cids now domiciled in North America. Both Ball and Omodeo derogated the evidence in support of Beddard’s “probability” that has been accumulated by the present author (cf. Gates, 1966, 1967, 1972a: 62, 1972b, 1976, etc.). Ball (1976) for in- stance, while admitting “that some earthworms have been transported 474 Proceedings of the Biological Society of Washington by man,” states, “we cannot from this logically infer that the entire distribution is a result of such transport.” Fifty years study of earthworm literature never once revealed any such claim for even one species of earthworm. Indeed, the author often has emphasized the need to deter- mine the original home of various widely distributed anthropochores. Also, observations of farmers (Ball, 1976: 410) seemingly are regarded as unimportant, although farmers who make their living through regular turning of the soil seem unusually well qualified to speak with authority on the absence of worms in the fields they tilled. However, persons other than farmers, including anglers as well as qualified natural history ob- servers, have recorded again and again the absence of earthworms in various glaciated parts of Canada as well as the United States. Also noteworthy is the absence of a single endemic earthworm anywhere in Canada. That of course could have been predicted by anyone really familiar with the necessities of earthworm life as well as with conditions prevailing during the Quaternary glaciation and subsequently. Indeed that is what the present author almost did long ago (Gates, 1929). Omodeo not only claimed that European earthworms were restricted to an eastern part of the United States (New England was specifically mentioned) but also that 200 years was insufficient to enable the present distribution. Actually more than 400 years is known to have been available for modern man to provide the present distributions. Columbus, on his second voyage to America, brought with him live plants. The English fishermen had been dumping earthen ballast in Newfoundland before there was any British settlement on the continent. Cortez re- turned live plants from Spain after his conquest of Mexico. Early English and Dutch settlers in New England and New York brought over pear trees, whose history has been followed, in large wooden tubs of earth. Eisen had found European lumbricids common as far west as California during the latter half of the 19th century. Finally, institutional and com- mercial as well as individual activity has been shown (Gates, 1966, 1967, etc.) to be adequate to have produced the present distribution of the European species on this continent. LITERATURE CITED Bau, I. R. 1976. Nature and formulation of biogeographical hypoth- eses. Systematic Zool. 24:407—430. GarMan, H. 1888 On the anatomy and histology of a new earthworm (Diplocardia communis gen. et sp. nov.). Bull. Illinois Sta. Lab. Nat. Hist. 3:47-77. Gates, G. E. 1929. The earthworm fauna of the United States. Science 70:266—267. -———. 1961. Ecology of some earthworms with special reference to seasonal activity. American Midland Nat. 66:61-86 (cf. p. 79). Earthworm distribution 475 ——. 1966. Requiem for megadrile utopias. A contribution toward the understanding of the earthworm fauna of North America. Proc. Biol. Soc. Washington 72:239-254. —. 1967. On the earthworm fauna of the great American Desert and adjacent areas. Great Basin Nat. 27:142-176. ——. 1970. Miscellanea Megadrilogica VII. Megadrilogica 1 (2): 1-14. ———-. 1972a. Burmese Earthworms. An introduction to the system- atics and biology of megadrile oligochaetes with special refer- ence to southeast Asia. Trans. American Phil. Soc. (N.S.) 62 (7):1-326. ——. 1972b. Toward a revision of the earthworm family Lumbri- cidae. IV. The trapezoides species group. Bull. Tall Timbers Res. Sta. No. 12:146. —. 1976. More on oligochaete distribution in North America. Megadrilogica 2 (11):1-6. Gorr, C. G. 1952. Flood plain animal communities. American Mid- land Nat. 47:478-486. [Published after Goff’s death and not as he wrote it.] Harman, W.J. 1960. Studies on the taxonomy and musculature of the earthworms of central Ilinois. Ph.D. Thesis, Univ. Illinois, 107 pp. Hemmpurcer, H. V. 1915. Notes on Indiana earthworms. Proc. In- diana Acad. Sci. 24:281-285. Jamieson, B. G. M. 1971. In: Brinkhurst & Jamieson: Aquatic Oligo- chaeta of the World. Toronto, xii + 860 pp. Joyner, J. W. 1960. Earthworms of the Upper Whitewater Valley (East-Central Indiana). Proc. Indiana Acad. Sci. 69:313-319. Junin, E. 1949. De Svenska daggmaskarterna. Ark. Zool. 42A (17): 1-58. Murcuiz, W. R. 1954. Natural history studies on the earthworms of Michigan. Ph.D. Thesis, Univ. Michigan, 282 pp. Otson, H. W. 1928. The earthworms of Ohio. Ohio Biol. Surv. Bull. 17:47-90. OmopveEo, P. 1963. Distribution of the terricolous oligochaetes on the two shores of the Atlantic. In: Love and Love, North Atlantic Biota and their history. New York, pp. 127-151. REYNOLDs, J. W., E. E. C. CLesscu, anp W. M. Reynoxps. 1974. The earthworms of Tennessee (Oligochaeta). I. Lumbricidae. Bull. Tall Timbers Res. Sta. 17:1-182. STEBBINGS, J. J. 1962. Endemic-exotic earthworm competition in the American midwest. Nature 196:905-906. STEPHENSON, J. 1930. The Oligochaeta. Oxford, xiii + 978 pp. SmirH, Frank. 1912. Earthworms from Illinois. Trans. Illinois Acad. Sci. 5th Annual Meeting, 1912, 3 pp. (Reprints unpaged). 476 Proceedings of the Biological Society of Washington 1915. Two new varieties of earthworms with a key to the described species in Illinois. Bull. Illinois State Lab. Nat. Hist. 10:551-559. ———. 1928. An account of changes in the earthworm fauna of IlIli- nois and a description of one new species. Bull. Illinois Nat. Hist. Surv. 17 (10) :347-362. ¢ — j tea Her me ea | i be & » 41, pp. 477-480 12 October 1976 vy dian ks PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON IDENTIFICATION OF THE AMERICAN CYPRINODONTID FISH HYDRARGIRA SWAMPINA LACEPEDE By Reeve M. Battey Aanp E. O. WILEY Museum of Zoology The University of Michigan Ann Arbor, Michigan, 48109, and Museum of Natural History The University of Kansas Lawrence, Kansas, 66045 Hydrargira with its single included species, swampina, was described from Carolina by Lacépéde (1803:378-380, pl. 10, fig. 3, p. 321). Hydrargira was emended to Hydrargyra by Cloquet (1821:102-103) and by many subsequent authors, in- cluding Valenciennes (in Cuvier and Valenciennes, 1846: 203), Gimther (1866:318), Jordan and Gilbert (1883:331), Garman (1895:96), and Jordan, Evermann, and Clark (1930: 175). Jordan and Evermann (1896:632) retained the original spelling. Most workers since Giinther (1866) have placed Hydrargira in the synonymy of Fundulus Lacépéde (1803:37- 39) whose type-species, as designated by Jordan and Gilbert (1883:331), is Fundulus mudfish Lacépéde (1803:37-39), a synonym of Cobitis heteroclita Linnaeus. Hydrargira swampina Lacépéde, type-species of Hydrargira by monotypy, has been considered a synonym of Fundulus heteroclitus (Linnaeus) by most workers, including Garman (1895:98), Jordan and Evermann (1896:641), Fowler (1916: 416), and Jordan, Evermann, and Clark (1930:75, although they also equated the name with F.. majalis on the same page). Garman (1895:98) incorrectly regarded F. majalis as type- 41—Proc. Bion. Soc. Wasu., Vou. 89, 1976 (A477) 478 Proceedings of the Biological Society of Washington species of Hydrargira. Valenciennes created confusion by first stating (1836:228) that Lacépéde’s description was of the young of one species and the figure of a different species, and later (in Cuvier and Valenciennes, 1846:203) by describing specimens from New Jersey under the name Hydrargyra swampina. Valenciennes’ H. swampina was correctly synony- mized with Fundulus diaphanus by Jordan and Evermann (1896:645). If the customary placement of H. swampina as a synonym of F. heteroclitus is correct, Hydrargira is a synonym of Fundulus. Otherwise, since it antedates other generic and subgeneric names in or closely associated with Fundulus, the name is available. Recently, Griffith (1974:320) “for reasons of priority [has employed] F. swampinus (Lacépéde) rather than F. lineo- latus as used by Rivas (1966).” The only documentation is Griffith’s unpublished thesis (1972 Yale Ph.D. Dissertation) in which he stated (p. 250): “The description and figure given by Lacépéde are unquestionably of the nominal F. lineolatus rather than F. heteroclitus or F. diaphanus as indicated by re- cent synonymies.” Fundulus lineolatus (Agassiz, 1854) is in the F. notti species group (Wiley and Hall, 1975) which is assigned by some workers to the genus or subgenus Zygon- ectes. Wiley and Hall (1975:1) noted Griffith’s substitution and suggested that “a ruling by the International Commission [for suppression of swampina based on Article 23] may be in order.” We recognize neither desirability nor need to carry this issue to the Commission. Lacépéde’s description and illustration (loc. cit.) of H. swampina are poor, but they provide no apparent basis for identification with F. lineolatus. His description differs from F, lineolatus in that swampina has 15 pectoral fin rays whereas lineolatus has 11-14 (Brown, 1958); swampina has 11 dorsal rays contrasting with 7-8 in lineolatus (data corrected from Brown, 1958); swampina has 12 anal rays rather than 8-10 in lineolatus (data corrected from Brown, 1958); and swampina reaches about 100 mm in length as against 80 mm in total length in lineolatus, which rarely exceeds 55 mm in standard length (based on large specimens from White Lake, North Carolina, in the Museum of Zoology, University of Michigan). Identity of American cyprinodontid 479 Lacépéde’s well-pigmented figure differs from F. lineolatus in that swampina has no subocular teardrop whereas lineo- latus has one; swampina lacks the distinctive flank stripes of female lineolatus; swampina lacks rows of dots on the side, characteristic of male lineolatus; the vertical bars of swampina (14 shown in Lacépéde’s figure) are not distinctly thickened as they are in male lineolatus; the dorsal fin originates in front of rather than behind the origin of the anal fin as in lineolatus; and the body of swampina is more robust than the slender lineolatus. Based on these differences, we reject identification of Hydrargira swampina Lacépéde with Fundulus lineolatus (Agassiz). The identity of H. swampina Lacépéde depends solely on the original description and figure because Lacépeéde based his description on manuscript notes given to him by Bose and apparently deposited no types (pers. comm. from M. Martine Desoutter, Museum National d'Histoire Naturelle, Paris). There are three other species of Carolina Fundulus which need be considered: F. majalis, F. diaphanus, and F. hetero- clitus. Fundulus majalis can be eliminated because it typically has 12-14, usually 13, dorsal fin rays whereas swampina has 11, and Lacépéde’s figure shows neither the distinctive dorsal fin spot of male majalis nor the prominent horizontal body stripes of the female. Fundulus diaphanus (and its synonym H. swampina Valenciennes, 1846) can be eliminated because it has 12-15 dorsal rays rather than 11, and it usually has a higher number of pectoral rays (15) 16-17 in diaphanus; 15 in swampina. Fundulus heteroclitus generally agrees with Lacépéde’s meristic and color description. The configuration and height of the fins as shown in Lacépéde’s figure are not accurate for any Carolina species, but the shape and position of the anal fin better represent female heteroclitus than male majalis or male diaphanus. Finally, Lacépéde’s account of abundance and habitat support the identification of swampina with heteroclitus. We conclude that there is compelling evi- dence for the retention of Hydrargira swampina in the synon- ymy of F. heteroclitus and that Hydrargira is a synonym of Fundulus. 480 Proceedings of the Biological Society of Washington We thank Dr. Frank B. Cross who has graciously reviewed the manuscript. LITERATURE CITED Acassiz, L. 1854. Notice of a collection of fishes from the southern bend of the Tennessee River in the State of Alabama. Amer. Jour. Sci. Arts, ser. 2(17) :353-369. Brown, J. L. 1958. Geographic variation in southeastern populations of the cyprinodont fish Fundulus notti (Agassiz). Amer. Mid- land Nat. 59(2):477-488. CLoguEtT, Hippotytre. 1821. Hydrargyre, Hydrargyra (Ichthyol.) in Dictionaire des Sciences Naturelles, XXII: 102--103. Fow.er, H. W. 1916. Notes on fishes of the orders Haplomi and Microcyprini. Proc. Acad. Nat. Sci. Phila. 68:415-439. GarMaNn, S. 1895. The cyprinodonts. Mem. Mus. Comp. Zool. Har- vard 19(1):1-179. GrirritH, R. W. 1974. Environmental and salinity tolerance in the genus Fundulus. Copeia (2):319-331. GtnTuer, A. 1866. Catalogue of the fishes in the British Museum. Vol. 6, London. xv + 368 pp. Jorpan, D. S., AND B. W. EvERMANN. 1896. The fishes of North and Middle America. Bull. U.S. Natl. Mus. 47 pt. 1: lx + 1240 pp. Jorpan, D. S., B. W. EvERMANN, AND H. W. Criarx. 1930. Check list of the fishes and fishlike vertebrates of North and Middle Amer- ica north of the northern boundary of Venezuela and Colombia. Rept. U.S. Comm. Fisheries. 1928 (1930), pt. 2:1-670. Jorpan, D. S., ano C. H. Gitpertr. 1883. Synopsis of the fishes of North America. Bull. U.S. Natl. Mus. 16: Ivi + 1018 pp. LaAcEPEDE, B. G. E. 1803. Histoire naturelle des poissons. Paris. Tome 5, Ixviii + 801 pp. Rivas, L. R. 1966. The taxonomic status of the cyprinodontid fishes Fundulus notti and F. lineolatus. Copeia (2) :353-354. VALENCIENNES, A. 1836. Les poissons. In: Cuvier, G. F., Le régne animal, disciples édition (1836-1849). Paris. 392 pp. 1846. In: Cuvier, G., and A. Valenciennes. Histoire naturelle des poissons. Paris vol. 18:xix + 505 pp. Wuey, E. O., III, ann D. D. Harty. 1975. Fundulus blairae, a new species of the Fundulus nottii complex (Teleostei, Cyprinodon- tidae). Amer. Mus. Novitates No. 2577:1-13. Wy a 42, pp. 481-490 12 October 1976 or i late on PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A SECOND TROGLOBITIC SPECIES OF THE GENUS LIRCEUS (ISOPODA, ASELLIDAE) FROM SOUTHWESTERN VIRGINIA By JAMes A. Estes AND JoHN R. HOLSINGER Department of Biological Sciences, Old Dominion University Norfolk, Virginia 23508 INTRODUCTION The first known troglobitic species (i.e., obligatory caverni- cole) of Lirceus, L. usdagalun, was described by Holsinger and Bowman (1973) from three caves in Lee County, Virginia. Prior to that time, 13 species of Lirceus had been described, all of which were primarily epigean forms (Williams, 1972). The genus occurs in parts of the eastern and middle-western United States and the Great Lakes region of southern Ontario, Canada, where it is represented in a variety of aquatic habi- tats, including springs, seeps, streams, ponds, sloughs, wood- land pools, lakes and drain outlets (Williams, 1972). Several species of Lirceus, other than L. usdagalun, also have been re- ported from cave stream habitats (see Holsinger and Bowman, 1973, for a summary), but all have pigment and small eyes and are considered troglophiles and/or trogloxenes. In August 1974, while doing field work in the caves of south- western Virginia, Dr. David C. Culver and one of us (J. R. H.) discovered another population of eyeless, unpigmented Lirceus in a stream in McDavids Cave, Scott County, Virginia. A second trip to this cave in November 1975 resulted in the collection of additional specimens from the population. Sur- face habitats in Lee and Scott counties also were sampled for the presence of Lirceus to determine the taxonomic relation- ship between epigean and hypogean species. On initial inspec- 42—-Proc. Biou. Soc. Wasu., Vou. 89, 1976 (481) 482, Proceedings of the Biological Society of Washington tion the material from McDavids Cave appeared similar to L. usdagalun, but after closer study it proved to differ consis- tently in a number of morphological characters and is de- scribed below. The description of this new species brings the total number of described species in the genus to 15 and the number of troglobites to two. All of the material examined in this study is deposited in the National Museum of Natural History, Smithsonian Institution (USNM), or in the collection of the junior author (JRH). Lirceus culveri, new species Figures 1-4 Material examined: Virginia. Scott Co.: McDavids Cave, holotype male (USNM 156148), 21 paratypes (USNM 156149) and 4 slide- mounted paratypes (JRH collection), J. A. Estes et al., 28 Nov. 1975; 10 paratypes (2 partly on slide mounts) (USNM 152631), J. R. Hol- singer and D. C. Culver, 3 Aug. 1974. Diagnosis: An eyeless, unpigmented species closely allied morpho- logically with Lirceus usdagalun but distinguished from that species as follows: presence of only 1 heavy spine on small proximal process of gnathopod palm in both sexes; median process of palm of gnathopod small to prominent in mature male; spur of endopodite tip of male second pleopod slightly longer; distomedial margin of peduncle of male second pleopod without small spine and more weakly serrulate; uropod less setose, approximaltely 20 percent length of pleotelson, exopod only about ¥% length of endopod. Largest male, 6.2 mm; largest female, 6.8 mm. Description: Body about 64 percent longer than wide, with subparal- lel sides, not widening posteriorly. Head about 60 percent wider than long; lateral margin narrowly and deeply incised, anterior lobe about twice as wide as posterior lobe. Pereonites subequal, margins covered with fine setae. Pleotelson subtriangular, about 28 percent length of body, only slightly broader than long, narrowing posteriorly; median posterior process poorly delimited. Antenna 1: flagellum 5-segmented, last 2 to 4 segments bearing aesthetascs; first 2 peduncular segments with tufted setae (not shown in Fig. 2A). Antenna 2 relatively long but variable in length, 60 to 90 percent length of body; first peduncular seg- ment with single stiff seta; peduncular segment 6 about 33 percent longer than 5, longer than combined length of first 4 peduncular seg- ments; flagellum with up to 37 segments in largest specimen. Mandible without palp; incisor and lacinia mobilis 4-cuspidate (lacinia mobilis absent from right, present on left); spine row with 11 to 13 plumose spines on left, 15 on right; molar bearing clusters of subapical setae. Maxilla 1: outer plate with 12-13 apical spines, outermost 4 longer than inner ones; inner plate with 5 apical, plumose setae. Maxilliped: inner New cave isopod from SW Virginia 483 Fig. 1. Lirceus culveri, new species. Female paratype (5.9 mm): A, Dorsal view; B, Head and pereonite (setation omitted); C, Pleo- telson and uropods, ventral (setation omitted). Male paratype (6.2 mm): D, Pleopod 3, anterior (setation shown only on one side); E, Left antenna 2, in part. 484 Proceedings of the Biological Society of Washington SSE nN \ : Mi ) Mo Fig. 2. Lirceus culveri, new species. Male paratype (6.0 mm): A, Right antenna 1, dorsal. Male paratype (6.2 mm): B, Terminal segments of antenna 1, dorsal (enlarged); C, Maxilla 1; D, Maxilla 2 (laminae of outer plate); E, Lower lip. Male paratype (5.2 mm): F, Left mandible; G, Left maxilliped, posterior. New cave isopod from SW Virginia 485 Fig. 3. Lirceus culveri, new species. Male paratype (5.2 mm): A, Left gnathopod, medial; B, Left uropod, dorsal. Female paratype (5.2 mm): C, Right gnathopod, medial. Male paratype (6.2 mm): D, pe- reopod 7, Female paratype (5.5 mm): E, Right pleopod 2, anterior. 486 Proceedings of the Biological Society of Washington Fig. 4. Lirceus culveri, new species. Male paratype (6.2 mm): A, Pleopod 1, anterior (setation shown only on one side); B, Exopodite of pleopod 2, anterior (enlarged). Male paratype (6.0 mm): C, Pleopod 2, anterior (tip enlarged). Male paratype (5.2 mm): D, Inner margin of peduncle of pleopod 2. Lirceus usdagalun (4.8 mm male): E, Inner margin of peduncle of pleopod 2 (note comparison with that of L. cul- veri). New cave isopod from SW Virginia 487 plate with row of apical, plumose setae; outer plate with 5-6 retinaculae on inner margin and numerous thick, plumose setae on apex; palpal seg- ments 2—5 with numerous setae. Gnathopod propod of male: palm rather oblique, with small proximal process bearing 1 heavy spine; median process of palm small to well developed, usually prominent in larger specimens. Gnathopod propod of female proportionately smaller than that of male; median process absent. Dactyls of pereopods 2-7 bearing 2 spines each. Male pleopod 1: pedun- cle 65-70 percent length of exopod, with 2-3 retinaculae; exopod nearly 50 percent longer than wide, inner margin nearly straight, unarmed; outer margin of exopod convex, armed with relatively long setae. Male pleopod 2: peduncle about % longer than wide, distomedial margin weakly serrulate, without spines. Exopod of male pleopod 2 somewhat variable, slightly more than %4 length of peduncle; distal segment sub- quadrate, bearing 5 long, very weakly barbed (not indicated in Fig. 4B, C), apical to subapical setae and 7-9 long, naked setae on lateral margin. Endopod 2 of male pleopod 2 suboblong, about twice as long as broad; apex with 3 distinct processes and 1 spur; caudal process broadly rounded, margin partly rugose; cephalic process smaller and less broadly rounded than caudal process, anterior surface rugose; mesial process with short stalk and flattened tip bent anteriad; spur slender, finger-like, slightly curved, sometimes reaching just beyond caudal process. Female pleopod 2 short; lateral margin oblique distally, nearly straight prox- imally; apex with 2 long, terminal setae and 1 shorter subterminal seta. Pleopods 3-5 of both sexes generally similar to those of L. usdagalun. Uropod about 5.5 percent length of body, about 20 percent length of pleotelson; peduncle about as broad as long, a little more than twice as long as exopod; endopod about 7% longer than exopod; rami bearing long, terminal setae. Remarks: The small spine on the distomedial margin of the peduncle of male pleopod 2 of L. usdagalun was not indicated in Fig. 3C in the description of that species by Holsinger and Bowman (1973). A cor- rected illustration of this structure is provided by us in Fig. 4E. Type-locality: McDavids Cave, located approximately 6 km northeast of Clinchport in the Rye Cove karst area of Scott Co., Virginia, is a large, stream-passage cave developed along the strike in the Rye Cove limestone of Middle Ordovician age (Holsinger, 1975). Distribution and ecology: L. culveri is known only from the type- locality, where it has been collected twice from on and around gravels (some fused) in areas of the cave stream marked by riffles. Two other troglobitic crustaceans, the isopod Asellus recurvatus and the amphipod Stygobromus mackini, also inhabit the stream, but they generally occur on rocks and appear much less common than L. culveri. On the August 1974 visit, two small female crayfish, Cambarus bartonii, also were collected from the stream; this species is a troglophile or trogloxene. The August 1974 collection contained five males and five females, two 488 Proceedings of the Biological Society of Washington atyhies® KENTUCKY Clinchport mEwing ‘ cael anit TENNESSEE Fig. 5. Distribution of troglobitic species of Lirceus in southwestern Virginia: L. culveri, open circle (Scott County); L. usdagalun, closed circles (Lee County). of which were ovigerous. The ovigerous females measured 5.6 and 7.8 mm and were carrying 18 and 28 eggs, respectively. The November 1975 collection contained nine males and 17 females, three of which were ovigerous or larviparous. Two of these females measured 4.8 and 4.9 mm and were carrying 11 and 13 newly-hatched young, respectively. The other female, 5.2 mm in length, had eight eggs in the brood pouch. Some of the newly-hatched young and eggs from the females in the latter collection were probably lost from the pouches before counts were made. Etymology: It is a pleasure to name this new species in honor of our friend and colleague, Dr. David C. Culver, whose studies on the ecology of aquatic cave stream communities in the Appalachians have contributed substantially to our knowledge of biospeleology. DIscussION Morphologically and ecologically, L. cuiveri appears to be closely re- lated to L. usdagalun and the two species probably share a relatively recent common ancestor. Although the morphological differences be- tween these species are not great, they do, however, appear to be consistent and of enough significance to warrant separation of the two forms at the species-level. As presently known, L. usdagalun and L. culveri occur in separate karst areas in adjacent tributaries of the upper Tennessee River drainage basin—the former from the Powell River Valley and the latter from the Clinch River Valley. The ranges (see Fig. 5), which are allopatric, are situated approximately 45 km apart and are physically separated by the Powell Mountain and other potential barriers to the dispersal of aquatic cavernicoles. In view of the absence of either of these species from surface seeps or springs and the fact that both appear to be confined ecologically to the substrates of cave streams, there is little, if any, possibility for gene New cave isopod from SW Virginia 489 exchange between populations presumably isolated in separate karst areas. The ancestor to these species was probably an epigean form whose range covered parts of both the Powell and Clinch valleys, but which has since been eliminated from surface habitats. L. usdagalun was originally described from three caves in Lee County by Holsinger and Bowman (1973) but has since been found in a fourth cave—Gallohan No. 2. However, the latter cave is hydrologically inte- grated with nearby Gallohan Cave No. 1 (type-locality of L. usdagalun) and these two caves share the same stream (Holsinger, 1975). The Gal- lohan caves are indicated by the same closed circle on the map in Fig. 5. As previously noted, L. culveri is known only from its type-locality, but the discovery of additional populations of this species from other nearby cave streams which contribute to the extensive subterranean drainage system in Rye Cove will not be surprising. Several epigean species of Lirceus occur in the surface waters of the karst areas in Rye Cove and south-central Lee County. To gain further insight into the distribution of Lirceus in these areas and to check for similarities between the epigean and hypogean species, we searched several springs and small surface streams for populations of Lirceus. At least two, and possibly three, species were found, none of which can be assigned with certainty to any described species in the genus. All of these species differ significantly from L. culveri and L. usdagalun, espe- cially in the structure of the gnathopods, shape of the head and usually the body, and in the proportionate length and setation of the uropods. In addition, the surface species are pigmented and eyed. One of these species, a small form (up to 6.0 mm in length) with yellow banding of the pereonites, was found in springs of both karst areas. A larger, un- banded species (length up to 10.0 mm) was found in the surface stream of a blind valley leading to the entrance of Alley Cave in Rye Cove. Finally, a relatively small (up to 7.0 mm in length), lightly pigmented, eyed form was collected from the stream in Alley Cave. The stream in this cave is recharged from the surface stream in the blind valley and is, in turn, a subterranean tributary to the larger stream in McDavids Cave which contains L. culveri. The form in Alley Cave may be conspecific with the species in the surface feeder stream, but its size is smaller and the body is narrower and more lightly pigmented. This same lightly pig- mented, narrow-bodied form was also found along with the small, yel- low-banded species in a series of small springs which empty into Mill Creek on the eastern side of Rye Cove. It is perhaps of biological inter- est that these springs are the resurgence of the subterranean waters in McDavids Cave. Further study is obviously necessary to ascertain the taxonomic re- lationship between the epigean species in Lee and Scott counties, and also to determine more precisely the relationship between the epigean and hypogean species. Considering the present taxonomic confusion in the genus Lirceus (see Styron, 1969; Williams, 1972; Holsinger and 490 Proceedings of the Biological Society of Washington Bowman, 1973), this may prove to be a difficult task until the system- atics of the genus are adequately revised. ACKNOWLEDGMENTS For their assistance with the field work, we are grateful to George D. Corbett, David C. Culver, Gary W. Dickson, Steven W. Hetrick and David Wapinski. We also thank the cave owners in southwestern Vir- ginia for their continued cooperation in allowing us access to their property for our biospeleological investigations. LITERATURE CITED Housincer, J. R. 1975. Descriptions of Virginia caves. Virginia Div. Mineral Resources Bull. 85, 450 pp. , AND T. E. Bowman. 1973. A new troglobitic isopod of the genus Lirceus (Asellidae) from southwestern Virginia, with notes on its ecology and additional cave records for the genus in the Appalachians. Int. Jour. Speleol. 5:261-271. Sryron, C. EF. 1969. Taxonomy of two populations of an aquatic iso- pod, Lirceus fontinalis Raf. Amer. Midl. Nat. 28:402-416. Wituiams, W. D. 1972. Freshwater isopods (Asellidae) of North America. Biota of Freshwater Ecosystems, Ident. Manual No. 7, U.S. Environmental! Protection Agency, 45 pp. @ Oy WU =e pp. 491-508 12 October 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CYCLOPOID COPEPODS ASSOCIATED WITH TRIDACNIDAE (MOLLUSCA, BIVALVIA) IN THE MOLUCCAS By ArtHur G. HuMEs Boston University Marine Program Marine Biological Laboratory Woods Hole, Massachusetts 02543 In the Indo-West Pacific several cyclopoid copepods are associated with large tridacnid bivalves belonging to the genera Tridacna and Hippopus. Members of the copepod genera Anthessius and Lichomolgus live in the mantle cavity of these bivalves, while Paclabius, according to Kossmann (1877), inhabits the pericardium. Records of copepod associates of Tridacnidae, including the new information given below, are as follows: Anthessius solidus Humes and Stock, 1965 from Tridacna squamosa Madagascar Humes and Stock Lamarck (1965 ) Eniwetok Atoll Humes (1972) Anthessius amicalis Humes and Stock, 1965 from Tridacna squamosa Madagascar Humes and Stock (1965 ) Eniwetok Atoll Humes (1972) New Caledonia Humes (1973) from Tridacna elongata Red Sea Humes and Stock Lamarck (1965 ) from Tridacna maxima New Caledonia Humes (1973) (Roding ) from Hippopus hippopus Eniwetok Atoll Humes (1972) ( Linnaeus ) Anthessius alatus Humes and Stock, 1965 from Tridacna noae Red Sea Humes and Stock ( Roding ) (1965 ) 43—Proc. Brou. Soc. WaAsH., Vou. 89, 1976 (491) 492 Proceedings of the Biological Society of Washington from Tridacna squamosa Madagascar Humes and Stock (1965 ) Eniwetok Atoll Humes (1972) New Caledonia Humes (1973) Moluccas present paper from Tridacna maxima New Caledonia Humes (1973) Eniwetok Atoll Humes (1972) from Tridacna gigas Eniwetok Atoll Humes (1972) ( Linnaeus ) Anthessius discipedatus new species from Hippopus hippopus Moluccas present paper Lichomolgus tridacnae Humes, 1972 from Tridacna gigas Eniwetok Atoll Humes (1972) from Tridacna squamosa Eniwetok Atoll Humes (1972) from Hippopus hippopus Moluccas present paper Lichomolgus hippopi new species from Hippopus hippopus Moluccas present paper Paclabius tumidus Kossmann, 1877 from Tridacna sp. Philippine Kossmann (1877) Islands from Tridacna squamosa New Caledonia Humes (1973) Immediately after collection the Moluccan tridacnids were isolated individually in containers of sea water. Sufficient 95 percent ethyl alcohol was added to make a_ solution of about 5 percent. The adductor muscles were cut and the mantle cavity rinsed thoroughly. The sediment obtained was then strained through a fine net (mesh 74 holes per inch) and the copepods removed. The observations and measurements of the two new species were made on specimens cleared in lactic acid. All 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, As = second antenna, L= labrum, MXPD = maxilliped, and P, = leg 1. The specimens were collected by the author during the AupHa Hewrx East Asian Bioluminescence Expedition which was supported by the National Science Foundation under grants OFS 74 01830 and OFS 74 02888 to the Scripps Institu- tion of Oceanography and NSF grant BMS 74 23242 to the University of California, Santa Barbara. I am indebted to Dr. Kenneth J. Boss, Museum of Compara- Copepods associated with Tridacnidae 493 tive Zoology, Harvard University, for the identification of the bivalve hosts. MyicotmwaE Yamaguti, 1936 Anthessius discipedatus, new species Figures 1-24 Type material: 229, 6284 from one bivalve, Hippopus hippopus (Linnaeus ), length 26 cm, in 3 m, Gomumu Island, south of Obi, Moluc- cas, 1°50’00”S, 127°30’54”E,, 30 May 1975. Holotype @, allotype, and 4 paratype ¢@ 4 deposited in the National Museum of Natural History (USNM), Washington; the remaining paratypes (dissected) in the col- lection of the author. Female: Body (Fig. 1) similar in general shape to the three species of Anthessius from Tridacna described by Humes and Stock, 1965. Length (not including the setae on the caudal rami) 1.46 mm (1.39-1.54 mm) and the greatest width 0.61 mm (0.58—0.64 mm), based on 2 spec- imens in lactic acid. Prosome moderately flattened, with the cephalo- some slightly indented near the level of the maxillipeds. Epimeral areas pronounced and rounded. Segment bearing leg 1 separated incompletely from the head by a transverse dorsal furrow. Ratio of the length to the width of the prosome 1.52:1. Ratio of the length of the prosome to that of the urosome 1.70:1. Segment of leg 5 (Fig. 2} 120 x 348 um. Genital segment 198 x 292 wm, tapered in its posterior half. Genital areas located laterally on the midregion of the segment. Each area (Fig. 3) with 2 small naked setae approximately 10 um long. Three postgenital segments from ante- rior to posterior 86 x 156, 73 « 143, and 83 (114 in middle length)