The Atya-Mke Shrimps of the Indo-Pacific Region (Decapoda: Atyidae) FENNER A.. CHACE, JR. SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY ? NUMBER 384 SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION Emphasis upon publication as a means of "diffusing knowledge" was expressed by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph Henry outlined a program that included the following statement: "It is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge." This theme of basic research has been adhered to through the years by thousands of titles issued in series publications under the Smithsonian imprint, commencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the following active series: Smithsonian Contributions to Anthropo/ogy Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology Smithsonian Studies in Air and Space Smithsonian Studies in History and Technology In these series, the Institution publishes small papers and full-scale monographs that report the research and collections of its various museums and bureaux or of professional colleagues in the world cf science and scholarship. The publications are distributed by mailing lists to libraries, universities, and similar institutions throughout the world. Papers or monographs submitted for series publication are received by the Smithsonian Institution Press, subject to its own review for format and style, only through departments of the various Smithsonian museums or bureaux, where the manuscripts are given substantive review. Press requirements for manuscript and art preparation are outlined on the inside back cover. S. Dillon Ripley Secretary Smithsonian Institution S M I T H S O N I A N C O N T R I B U T I O N S T O Z O O L O G Y ? N U M B E R 3 8 4 The Atya-like Shrimps of the Indo-Pacific Region (Decapoda: Atyidae) Fenner A. Chac SMITHSONIAN INSTITUTION PRESS City of Washington 1983 A B S T R A C T Chace, Fenner A., Jr. The Atya-\ike Shrimps of the Indo-Pacific Region (Decapoda: Atyidae). Smithsonian Contributions to Zoology, number 384, 54 pages, 24 figures, 1983.?The shrimps recognized in this study comprise six freshwater species that have often been referred, untenably and sometimes in synonymy, to the genus Atya. Three of the species are here assigned to the re- established genus Atyoida: A. bisulcata, confined to Hawaii; A. pilipes, ranging eastward from eastern Indonesia and the Philippines to the Marquesas and Gambier islands; and A. serrata, possibly limited to Madagascar and smaller islands in the tropical western Indian Ocean. Two closely related forms compose the new genus Atyopsis: A. moluccensis, ranging through Thailand and Malaya to Indonesia and perhaps westward to Sri Lanka and northeastward to the Philippines; and A. spinipes, apparently inhabiting the eastern Lesser Sunda Islands and extending northward through the Philippines to Okinawa and eastward as far as Samoa. Atya striolata, occupying streams along the east coast of Australia, is assigned to the new genus Australatya. Keys are provided to the genera and species and, for each of the latter, there are a complete synonymy, review of the literature, references to published illustrations, a diagnosis, color notes if available, size limits, the known range and material examined, variations observed, ecological information, life-history notes if any, common names, and economic importance. Special attention is paid to the heteromorphism of the chelipeds in Atyoida, especially as displayed by a series of several hundred specimens of A. pilipes from Palau, Caroline Islands. OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. SERIES COVER DESIGN: The coral Montastrea cavemosa (Linnaeus). Library of Congress Cataloging in Publication Data Chace, Fenner Albert. The Atya-like shrimps of the Indo-Pacific region (Decapoda: Atyidae) (Smithsonian contributions to zoology ; no. 384) Bibliography: p. Supt. of Docso. no.: SI 1.27:384 1. Atyidae. 2. Crustacea?Indian Ocean. 3. Crustacea?Pacific Ocean. I. Title II Series QL1.S54 no. 384 591s [595.3'843] 83-600083 [QL444.M33] Contents Page Introduction 1 Acknowledgments 1 General Considerations 2 Key to Atya-Yike Genera of Atyid Shrimps 2 Genus Atyoida Randall, 1840 4 Key to Species of the Genus Atyoida 4 Atyoida bisulcata Randall, 1840 5 Atyoida pilipes (Newport, 1847) 10 Atyoida serrata (Bate, 1888) 17 Genus Atyopsis, new genus 26 Key to Species of the Genus Atyopsis 27 Atyopsis moluccensis (De Haan, 1849), new combination 27 Atyopsis spinipes (Newport, 1847), new combination 35 Genus Australatya, new genus 43 Australatya striolata (McCulloch and McNeill, 1923), new combination 43 Literature Cited 48 in The Atya-\ike Shrimps of the Indo-Pacific Region (Decapoda: Atyidae) Fenner A. Chace,Jr. Introduction A protracted study of the caridean shrimps collected during the Albatross Philippine Expedi- tion of 1907-1910 has made it clear to me that the Indo-Pacific freshwater shrimps commonly referred to the genus Atya Leach, 1816, for up to 100 years or more are not only quite distinct from the typical Afro-American species of that genus but also that they are not even congeneric among themselves. When Hobbs and Hart (1982) re- cently reviewed the typical members of the genus Atya, they reached a similar conclusion but lim- ited their study to Leach's genus in the restricted sense and urged me to publish my opinion sepa- rately. This, then, is a sequel to the Hobbs and Hart revision and extends their undertaking to the Atya-\ike shrimps inhabiting the lands of the Indian and western Pacific oceans. In keeping with its role as a sequel to the Hobbs and Hart review, the present report follows the general format of its predecessor but in less detail; I do not believe that current taxonomic or distri- butional knowledge of the Indo-Pacific species is sufficient to substantiate precise analysis. Each of the species accounts consists of a synonymy and Fenner A. Chace,Jr., Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, DC. 20560. list of references as complete as I have been able to compile, a review of the more important ele- ments in that literature, an appraisal of the pub- lished illustrations of the species, a concise diag- nosis, information on color in life when available, size, the known range and a list of the material examined, what is known of the variability, ecol- ogy, and life history of the species, common names, economic importance, and occasional gen- eral remarks. Each species is illustrated rather fully, but no consummate verbal descriptions have been attempted. ACKNOWLEDGMENTS.?My carcinological Smithsonian colleagues, particularly C.W. Hart, Jr., Horton H. Hobbs, Jr., and Raymond B. Manning, furnished hortatory and practical sup- port almost daily during the year devoted to this study. L.B. Holthuis of the Rijksmuseum van Natuurlijke Historie, Leiden, was similarly help- ful, both by correspondence and by direct com- munication during visits in August and Septem- ber 1982; in particular, he furnished a photo- graph of the larger of the type-specimens of Atya moluccensis and no less than a score of the refer- ences cited. R.W. Ingle of the British Museum (Natural History) kindly provided indispensable material of two of the species. Lawrence G. Abele of Florida State University donated a fine series of Archaeatya that was important to an under- standing of that eastern Pacific genus. The staff 1 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY of the Smithsonian Library, especially Jack F. Marquardt and Carolyn S. Hahn, strove mightily in their successful efforts to locate and deliver numerous obscure publications. Lilly King Man- ning graciously shared her expertise on the ar- rangement and reproduction of illustrations. Fi- nally, C.W. Hart, Horton Hobbs, L.B. Holthuis, and W.D. Williams of the University of Adelaide reviewed the penultimate draft of the manuscript. All of these, and others, contributed materially to the final draft of the study. I am most grateful to them as well as to the authorities of the National Museum of Natural History for providing space and facilities that are essential to the completion of this and other post-retirement commitments. GENERAL CONSIDERATIONS.?A study of this kind is bound to reveal provocative gaps in our knowledge of the animals involved. Some of those that appeared as this project unfolded can be solved only by additional collecting, others by comparative population analyses, and still others by laboratory observations that should be easily made on animals, like these, that seem to be aquarium-hardy and long-lived. As a descriptive zoologist, I have neither the talent nor the desire to pursue any of the following problems, but I do have an interest in promoting their solution. What is the actual range of Atyoida serrata? Is it really confined to the high islands of the tropical western Indian Ocean, as currently available in- formation would seem to indicate? Is Atyoida pilipes truly absent west of approxi- mately 120? east longitude and, therefore, from the larger Indonesian islands and the Asiatic mainland? Is it a single taxon or are other species or subspecies discernible in parts of its extensive Pacific range? Is the endopod of the first pleopod in old males of Atyopsis spinipes different from the one in A. moluccensis of comparable size? In other words, are the two forms species or subspecies? Also, does A. moluccensis occur in the Philippines? If so, to which form does the incomplete holotype of Atya spinipes belong? What are the life-history adaptations of shrimps that inhabit swift mountain streams? Is it true, as observed by Hunte (1978:135) in Ja- maica, "that species in high-gradient streams had larvae which required high salinity for develop- ment, and those in low-gradient streams had larvae with low salinity requirements?" Are the larval stages actually abbreviated during torren- tial periods, as suggested by Edmondson (1929:16)? If larvae are estuarine or marine, how do they manage to migrate the hundreds of ar- duous kilometers upstream to the adult habitat? Are the species of Atyoida protandrous, either totally or partially, as maintained by Carpenter (1978:349)? Is the heteromorphism displayed by the species of Atyoida of survival importance, as suggested by Fryer's statement (1977:125) that Atya innocous (Herbst, 1792), a species with chelae adapted for either sweeping the substrate or filter feeding, may compete successfully with Atya scabra (Leach, 1815), which is specialized for filter feeding? Or does the apparently sex-associated heteromorph- ism in Atyoida negate that possibility? Finally, what is the mechanism for color change in Atyoida bisulcata ? Key to Atya-like Genera of Atyid Shrimps (Carpus of second pereopod deeply excavate and little longer than broad) Rostrum with series of median dorsal spines; pleurobranchs on somites bearing 4 anterior pairs of pereopods only. (Pterygostomian margin of carapace rounded; telson with posterolateral angles not overreaching setigerous posterior margin; chelae not heteromorphic; 3rd pereopod of large males not armed with massive spur on merus; epipods on 3 anterior pairs of pereopods only; 1st pereopod without arthrobranch.) 2 NUMBER 384 Rostrum without median dorsal spines (except Atya crassa (S.I. Smith, 1871)); pleurobranchs on all 5 somites bearing pereopods. (Third max- illiped with 2 arthrobranchs.) 3 2. Rostrum with more than 12 dorsal teeth; antennular peduncle with distolateral spine of proximal segment reaching midlength of 2nd seg- ment; 3rd maxilliped with 1 rudimentary arthrobranch Atyella Caiman, 1906 (Lake Tanganyika) Rostrum with no more than 10 dorsal teeth; antennular peduncle with distolateral spine of proximal segment not even nearly reaching mid- length of 2nd segment; 3rd maxilliped with 1 well-developed and 1 rudimentary arthrobranch. (Third maxilliped without stout terminal spine; endopod of 1st pleopod of male expanded in proximal half, not tapering regularly from base to tip; appendix masculina spinose near distal end only; mastigobranchs on 4 anterior pereopods.) Micratya Bouvier, 1913b (West Indies; Costa Rica; and Panama) 3. Third maxilliped with terminal spine (sometimes partially concealed by setae); endopod of 1st pleopod of male tapering from base to tip; females exceeding males in size, possibly protandrous. (Telson with posterolateral angles not overreaching setigerous posterior margin; 3rd pereopod of large males not armed with massive spur on merus.) 4 Third maxilliped terminating in numerous stout setae but lacking single terminal spine; endopod of 1st pleopod of male expanded into broad lamina, not tapering from base to tip; males equalling or exceeding females in size, not protandrous. (Pterygostomian margin of carapace dentate or angular; chelae not heteromorphic; appendix masculina spinose over more than distal half of length; 1st pereopod with arthrobranch.) 6 4. Pterygostomian margin of carapace broadly rounded; chelae not hetero- morphic, adapted for filter feeding; appendix masculina with spinose area distinctly distal to apex of appendix interna; epipods on 3rd and 4th pereopods much reduced, vestigial; no mastigobranchs Australatya, new genus Pterygostomian margin of carapace narrowly rounded to sharply acute; chelae heteromorphic, adapted for filter or substrate feeding; appendix masculina with spinose area overlapped by at least distal part of appen- dix interna; well-developed epipods on 4 anterior pairs of pereopods; mastigobranchs on all 5 pereopods 5 5. Pterygostomian margin rounded; 1st pereopod without arthrobranch Archaearya Villalobos, 1960 (Costa Rica; Isla del Coco; and Perlas Archipelago, Gulf of Panama) Pterygostomian margin dentate or angular; 1st pereopod with arthrobranch Atyoida Randall, 1840 6. Telson with posterolateral angles not overreaching setigerous posterior margin; 3rd pereopod of large males without massive spur on merus; SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY endopod of 1st pleopod of male semicircular, spinulose, flexible; well- developed epipods on 4 anterior pairs of pereopods; mastigobranchs on all 5 pereopods Atya Leach, 1816 (America: Pacific drainage from northern Mexico to Peru, Atlantic drainage from northern Mexico and West Indies to southern Brazil; Africa: Atlantic drainage from Senegal to northern Angola) Telson with posterolateral angles overreaching setigerous posterior margin; 3rd pereopod of large males armed with massive spur on merus; endopod of 1st pleopod of male rhomboidally oval, submarginally spinose, rigid; epipods on 3rd and 4th pereopods reduced, vestigial; no mastigo- branchs Atyopsis, new genus Genus Atyoida Randall, 1840 Atyoida Randall, 1840:140 [type-species by monotypy: Atyoida bisulcata Randall, 1840:140; gender: feminine]. Atyoidea.?Gibbes, 1850a:25 [erroneous spelling]. Ortmanma Rathbun, 1901:120 [type-species by original des- ignation: Ortmannia henshawx Rathbun, 1901:120, footnote (= Atyoida bisulcata Randall, 1840:140); gender: feminine]. Alya.?Bouvier, 1904c: 136, footnote [erroneous spelling for Atya but used only in the combinations Alya serrata and A. bisulcata]. Pseudatya Roux, 1928:209 [type-species by monotypy: Pseud- atya beauforti Roux, 1928a:209 (= Atya pilipes Newport, 1847:160); gender: feminine]. Vanderbiltia Boone, 1935:159 [type-species by monotypy: Van- derbiltia rosamondae Boone, 1935:160 ( = Atya pilipes New- port, 1847:160; gender: feminine]. DIAGNOSIS.?Body pigmented, eyes well devel- oped; rostrum not strongly compressed laterally, median dorsal carina typically unarmed, ventral keel with 0-4 teeth; anterior margin of carapace armed with antennal spine, pterygostomian mar- gin sharply or bluntly acute; supraorbital spines absent; telson with setigerous posterior margin overreaching posterolateral angles; 3rd maxil- liped with uncinate terminal spine; pereopods without exopods; 1st and 2nd pereopods with chelae heteromorphic (with or without palm), fingers tipped with brushes of setae, carpus of both appendages excavate distally, little if at all longer than broad; 3rd pereopod without meral spur in large males; branchial complement con- sisting of 5 pleurobranchs, 3 arthrobranchs, 1 podobranch, 5 epipods, and 5 mastigobranchs; 1st pleopod of male with endopod tapering sin- uously but rather regularly to slender apex; 2nd pleopod of male with appendix masculina sub- cylindrical, spinose over more than distal half. RANGE.?Except for the almost certainly erro- neous record of A. serrata from the Cape Verde Islands by Bate (1888:699) Atyoida is known only from the high islands of the Indo-Pacific region: A. serrata perhaps being restricted to Madagascar and smaller islands in the tropical western Indian Ocean between 40? and 60? east longitude, A. pilipes to eastern Indonesia and the Philippines eastward from about 120? east longitude across the vast expanse of the Pacific Ocean as far as the Marquesas and Gambier islands, and A. bi- sulcata to the Hawaiian islands of Oahu, Maui, and Hawaii. Key to Species of the Genus Atyoida Rostrum bent somewhat ventrad; pterygostomian margin bluntly acute, not spinous; chelae dimorphic, adapted for filter-feeding or sweeping, not for selective picking alone. (Rostrum unarmed or bearing single ventral tooth, less frequently 2 teeth, very rarely 3; fixed teeth on posterior margin of telson inconspicuous.) A. pilipes Rostrum nearly horizontal; pterygostomian margin sharply acute, spinous; NUMBER 384 chelae polymorphic, adapted for filter-feeding, sweeping, or selective picking 2 2. Rostrum usually unarmed ventrally; posterior, setigerous margin of telson armed with 2-7 conspicuous fixed teeth A. bisuJcata Rostrum with 2-4 ventral teeth; fixed teeth on posterior margin of telson inconspicuous A. serrata Atyoida bisulcata Randall, 1840 FIGURES 1, 2 Atyoida bisulcata Randall, 1840:140, pi. 5: fig. 5 [type-locality: "Sandwich Islands"].?Stimpson, 1860:28.?A. Milne- Edwards, 1864:151.?Ortmann, 1895:407 [part].?Bou- vier, 1905:95, 114, 127, 130; 1925:277, 297.?Holthuis, 1955:26, 27.?Smith and Williams, 1982:343-345, 349, 358, 359. Alyoidea bisulcata.?Gibbes, 1850a:25 [erroneous spelling]; 1850b: 196. Atyoida bisulcata?.?Dana, 1852:540, pi. 34: fig. la-i. Atya bisulcata.?Bate, 1888:692, 700, fig. 71, pi. 120.?Thall- witz, 1891b:26.?De Man, 1892:362, 363.?Sharp, 1893:111.?Bouvier, 1904a:447-449; 1904b:378-380; 1904c: 137; 1905:98, 100, 107, 127, 130.?Rathbun, 1906:919.?Bouvier, 1909:333.?Caiman, 1910:786, 789- 791, figs. 1, 3A', 3A".?Blaringhem, 1911:192.?Bouvier, 1912b:692, 694:1913a:460; 1918:136.?Bouvier and de Charmoy, 1919:317.?Bouvier, 1925:262, 277, 289, 290, 297-299, 301, 309, 321, 322, 334, 350, 357, figs. 668- 671.?Roux, 1925:154.?Edmondson, 1929:5-12, 14, 15^ 19-35, figs. UJ-$, 2a-d,r-t, 4a,d,e-o, pi. 1A,C; 1930:4-7.? Radir, 1930:351-353.?Edmondson, 1935:17, 18, fig. 6j,k; 1936:3, 5, 13, fig. 1.?Woltereck, 1937a:246, 247; 1937b:322-325.?Gurney, 1942:86-88, fig. 20cl.?Roth- Woltereck, 1942:262-264, 268.?Holthuis, 1954:2, 3.? Ullman, 1967:56, 57, 60, 2 figs.?Williams, 1977:412.? Carpenter, 1978:343, 349, 350.?Costa, 1980:697.?Smith and Williams, 1982:343, 344. Ortmannia henshawi Rathbun, 1901:21, footnote [type-local- ity: "Kaiwiki, Hilo, Hawaii"]; 1906:919.?Edmondson, 1929:5-15, 19-30, 34, 35, figs. lc,d/-k, 2e-q, 3, 4b,c, pi. 1B,D.?Radir, 1930:351-353.?Edmondson, 1935:17.? Woltereck, 1937b:322, 323, 325.?Gurney, 1942:86, 87, fig. 20c2, DI , 2.?Holthuis, 1954:2, 3; 1955:27.?Carpen- ter, 1978:349.?Costa, 1980:697.?Smith and Williams, 1982:344, 349. Ortmannia Henshawi.?Bouvier, 1904a:447; 1904b:378, 379; 1905:100, 102, 107, 127-130; 1909:333.?Caiman, 1910:789-791, figs. 3B', B".?Blaringhem, 1911:192.? Cuenot, 1911:400.?Bouvier, 1911:1821, 1823, 1824; 1912a:920, 922; 1912b:692-694; 1913a:459, 460; 1918:136.?Bouvier and de Charmoy, 1919:317.?Bou- vier, 1925:263, 277, 278, 280, 287, 288, 293, 297-299, 321, 322, 333, 345, 349-351, 356-358, figs. 640-644. Ortmannia, mutation Henshawi.?Bouvier, 1904a:448. Atya bisulcata, Henshawi variation.?Bouvier, 19O4b:379. Ortmannia, modification Henshawi.?Bouvier, 1904b:379,380. Atya bisulcata, mut. Henshawi.?Bouvier, 1904c: 136; 1905: 107. Atya brevirostris.?Bouvier, 1904c: 137 [part].?Seurat, 1934: 50, 51 [part]. [Not A. brevirostris De man, 1892.] Ortmannia Henshawi mut. bisulcata.?Bouvier, 1905:98, 100, 111, 114, 127, 130. mutation bisculcata of 0. Henshawi.?Bouvier, 1905:109 [er- roneous spelling]. Ortmannia Alluaudi Bouvier, 1905:108, 116 [part]. Atyoida {Atya) bisulcata.?Bouvier, 1925:303. Atyoida henshawi.?Roux, 1925:154. Atya bisculcata.?Edmondson, 1929, fig. lj,k, 26 [erroneous spelling]. Ortmannia Henshavi.?Woltereck, 1937a:246, 248 [erroneous spelling].?Roth-Woltereck, 1942:262-264, 268, 277. Ortmannia henssavi.?Woltereck, 193 7b: 323 [erroneous spell- ing]- REVIEW OF LITERATURE.?Ten years after Ran- dall (1840) described this genus and species from Hawaiian material collected by Thomas Nuttall, Gibbes (1850a, 1850b) introduced the first of several misspellings of the binomen. Two years later, Dana (1852) tentatively assigned to the species material taken at Oahu by the U.S. Ex- ploring Expedition and was the first of numerous carcinologists to question the propriety of the genus, but Stimpson (1860) and A. Milne-Ed- wards (1864) continued to recognize it. Bate (1888), however, concurred with Dana's doubts and relegated the species to Atya, where it was left by Thallwitz (1891b), De Man (1892), and Sharp (1893). Ortmann (1895) reinstated Atyoida be- cause of the nonatyoid chelae in Hawaiian ma- terial examined by him, and he mistakenly syn- onymized A. tahitensis with A. bisulcata. The Atya- like chelae of Randall's type-specimen led Rath- bun (1901) to propose a "nom. nov.," Ortmannia henshawi, for the form denoted by Ortmann and to treat Atyoida as a synonym of Atya; inasmuch SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY as Ortmann's action must be construed as a misidentification rather than the creation of a homonym?if the two forms are considered to represent separate taxa?Rathbun must be deemed to have proposed a new species, instead of a replacement name, and the type-series must include the material examined by her, as well as that mentioned by Ortmann and discussed by her. Bouvier (l904a,b,c) decided that A. bisulcata and O. henshawi were forms of the same species of Atya and he (1904a:449) considered "la mutation Henshawi" to be a "forme atavique" representing a species "en voie d'evolution." Rathbun (1906:919) accepted Bouvier's explanation and referred to O. henshawi as an "atavic form of Atya bisulcata.'''' Caiman (1910) reported that one spec- imen in a lot of 88 collected by the Challenger at Honolulu has three Atya-type chelae and the fourth "distinctly of the Ortmannia-shape;" he also noted that "Miss Rathbun displaced Atyoida on the ground that the surviving type-specimens of Randall's Atyoida bisulcata, the type-species of Atyoida, have chelae of the Atya-type. If, however, O. henshawi, the type-species of Ortmannia, is only a form of A. bisulcata, the two genera are synony- mous and the older name should be used." Cuenot (1911) believed that Ortmannia henshawi and the "mutation atyienne" Atya bisulcata rep- resent a link between the tropical American spe- cies of Ortmannia (= Potomirim Holthuis, 1954) and the American and Indo-Pacific species of the true Atya. In his valiant attempt to bring order out of chaos in the extremely difficult family Atyidae, Bouvier (1925) chose ("pour suivre les regies de la nomenclature,") to treat Ortmannia henshawi and Atya bisulcata separately, the latter albeit as a "mutation atyienne" of the former. By far the most complete coverage of the history, variability, responses to changes in salinity, temperature, light, and pH, as well as regeneration, molting, larval development, associated fauna, and means of dispersal of the species called Atya bisulcata (and Ortmannia henshawi) is to be found in Edmondson (1929). In a continuation of these studies, Ed- mondson (1930) experimented with the effects of ultraviolet light on regeneration, Radir (1930) demonstrated hermaphroditism, and Edmondson (1936) determined the effect of X-rays on regen- eration. Little attention has been paid to this unusual animal during the past half century, although Ullman (1967) indicated its adaptabil- ity to aquarium conditions, and Carpenter (1978) called attention to possible protandry in the spe- cies. PUBLISHED ILLUSTRATIONS.?Accompanying Randall's original description of the species in 1840 is a crude drawing of the third pereopod, showing its slender form in comparison with that of this appendage in the species of Atya. Dana (1852), in his atlas (1855), offered a toto illustra- tion and sketches of the mouthparts, chela, and dactyl of the fourth pereopod. Bate (1888) illus- trated the entire shrimp in lateral view, the an- tennule, antenna, mouthparts, and first and fifth pereopods, as well as the zoeal telson. Caiman (1910) presented a lateral view of an ovigerous female from the Challenger collection and the first and second chelae of both atyoid and ortman- nioid forms. Bouvier (1925) showed the anterior end of the animal in lateral view, the rostrum in dorsal view, the preanal spine in ventral and lateral views, the chelipeds, and the dactylar setae of a chela. In the review of the Hawaiian atyids by Edmondson (1929), there are photographs in lateral view of the two forms and of the chelipeds of both, as well as drawings of rostra in lateral view, cheliped variation and setae, pereopod re- generation, and first pleopods. Larval antennules and telson are shown in Edmondson (1935). The same author (1936) depicted the effect of X-rays on cheliped regeneration. Gurney (1942) showed cheliped "mutation." Finally, there are two pho- tographs of the shrimps in an aquarium in Ull- man (1967). DIAGNOSIS.?Rostrum continuing general trend of dorsal surface of carapace in adults, not bent noticeably ventrad, usually unarmed ventrally; pterygostomian angle of carapace sharply acute; telson with 2-7 conspicuous fixed teeth on setig- erous posterior margin; chelae polymorphic, adapted for filter-feeding, sweeping, or selective picking. COLOR NOTES.?A label accompanying the five specimens from a canyon five miles (8 km) east of NUMBER 384 FIGURE 1.?Atyoida bisulcata (all from ovigerous female from Pepeekeo, Hawaii, carapace length 8.8 mm): a, cephalothorax and appendages, left aspect; b, rostrum, left aspect; c, anterior carapace and appendages, dorsal aspect; d, telson and uropods; e, posterior end of telson; /, right mandible; g, right 1st maxilla; A, right 2nd maxilla; i, right 1st maxilliped; 7, right 2nd maxilliped; k, left 3rd maxilliped; /, same, distal end; m, right 1st chela and carpus; n, right 2nd chela and carpus; o, right 3rd pereopod; p, same, dactyl; q, same, flexor surfaces of propodus and dactyl; r, dieresis of lateral branch of right uropod. (Magnifications: a, X 4.3; c, d, X 5.2; b, g-k, m-o, X 10.8; tf brevirostris variety de-mani.?De Man, 1915:406. Atya brevirostris var. de Mani.? Bouvier, 1925:294, 295. Atya spimpes.?Roux, 1928a:209 [lapsus for A. pilipes]. Pseudatya beauforti Roux, 1928a:209-213, figs. 1-9 [type-lo- cality: Batjan].?Woltereck, 1937b:324.?Holthuis, 1955: 27.?Johnson, 1961:121.?Smith and Williams, 1982:349. Atya serrata.? Adamson, 1935:16.?Edmondson, 1935:14-18, figs. 5i, 6g-i,l.?Estampador, 1937:485.?Woltereck, NUMBER 384 11 1937b:322, 323 [part].?Miyake, 1938:107, 111.?Adam- son, 1939:36.?Balss, 1957:1530.?Estampador, 1959:19. [Not Atya serrata Bate, 1888.] Ortmanma alluaudi.?Adamson, 1935:16.?Woltereck, 1937b: 322, 323 [part].?Adamson, 1939:36. [Not 0. Alluaudi Bouvier, 1905.] Ataya breverostris var. de mani.?Blanco, 1935:31 [erroneous spelling]. Vanderbiltia rosamondae Boone, 1935:160, pis. 41, 42 [type- locality: Venus Point Reef, Tahiti, Society Islands].? Holthuis, 1953:113-118, fig. 1; 1955:27.?Smith and Wil- liams, 1982:349. Vanderbiltia rosamundae.?Holthuis, 1952:22 [erroneous spell- ing]- Vanderbiltia mirabilis Holthuis, 1953:114 [manuscript name published as a synonym]. Atyoida pilipes.?Smhh and Williams, 1982:345, 349, 358, 359 [part]. REVIEW OF LITERATURE.?Fortunately, the original designation of the type-locality of Atya pilipes by Newport (1847:160) as "Apia, Upoln, New Zealand" was promptly corrected to "Apia, Upolu, Navigator or Samoan Group" by Dana (1852:533). Stimpson (1860:28) described Atyoida tahitensis, which was generally accepted as a dis- tinct species for about 65 years. A. Milne-Ed- wards (1864:150) considered Atya pilipes to be a valid species, but he seems to have been unaware of Dana's correction of the type-locality and cited a third version: "Apia, Upolce, la Nouvelle-Ze- lande;" he also referred to Atyoida tahitensis as a species "presque semblabale a YA. bisulcata." Schmeltz (1869:152) noted that topotypic mate- rial of this species occurred at more than 1000 feet (305 m) above sea level at Upolu. Miers (1876:79) expressed little doubt "that Samoa is the correct locality of the type-specimen" of Atya pilipes, a belief that he reiterated four years later (1880:382), but at that time he confused the relationships by remarking that "the types both of A. spinipes and of A pilipes, Newport, are small and in bad condition; and it is probable that they are not specifically distinct." Schmeltz (1881) almost certainly misidentified Ponape specimens as "Atyoidea bisulcata" Filhol (1885) listed Atya pilipes among New Zealand crustaceans but a year later (1886) he indicated that its presence in New Zealand was doubtful. Atya brevirostris was described by De Man (1892) from Flores and Timor. Ortmann (1895) treated Atya pilipes as a synonym of A. spinipes and Atyoida tahitensis as identical with A. bisulcata, but he recognized Atya brevirostris as distinct. L.B. Holthuis (personal communication) has called my attention to the possibility that the material recorded by Thomp- son (1901:21) from Ponape may be the same as that listed by Schmeltz (1881:15) in the Godefroy Museum catalog. Doflein (1904:104) similarly identified a single specimen from the Marshall Islands as Atyoida bisulcata. Bouvier (1905) and Bordage (1908) synonymized Atya brevirostris with A. serrata. Bouvier (1925) accepted Mier's state- ment that Atya pilipes is a synonym of A spinipes, and synonymized Atya brevirostris, A. brevirostris var. de Mani, and Atyoida tahitensis with Atya serrata. Roux (1925, 1926a) synonymized Atya serrata and A. pilipes and presumably confused material of two species under the latter name. Two years later, Roux (1928a) maintained the same synon- ymy, reported one male out of 76 specimens with ortmannioid chelae (perhaps the first Indo-Aus- tralian specimen found with chelipeds of this form), and described Pseudatya beauforti from Batjan. The same author (1934) reported that Atya pilipes was much more common than A. spinipes at Manus, Admiralty Islands, and that all of the specimens (3 males, 75 females, 108 oviger- ous females) bore atyoid chelae. The material from the Marquesas Islands identified as Atya serrata and Ortmannia alluaudi by Adamson (1935, 1939) must be assignable to Atyoida pilipes accord- ing to the ranges accepted herein, and Edmond- son (1935), Estampador (1937), and Miyake (1938) followed suit with other material beyond the range of A serrata. Boone (1935) created an enigma that defied interpretation for more than 15 years with her inaccurate description of Van- derbiltia rosamondae from Tahiti, a species that Holthuis (1953) was able to synonymize with A serrata, the name then used for all material of Atyoida, except the Hawaiian A bisulcata. Al- though Holthuis (1955) synonymized Pseudatya with Atya, he did not indicate the senior synonym of P. beauforti, the type-species of Pseudatya, and it 12 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY remained for the same author (1980) to include it in the synonymy of Atya pilipes. Holthuis and Rosa (1965) indicated the recorded distribution of Atya pilipes (including A serrata). Finally, Smith and Williams (1982:349) listed the synonyms of Atyoida pilipes (including Atya serrata) and its known distribution. PUBLISHED ILLUSTRATIONS.?The earliest pic- tures of Atya pilipes are probably those offered by Ortmann (1890), consisting of rather crude rep- resentations of the rostrum in dorsal and lateral views and of the first pereopod. More detailed illustrations of the anterior end of the animal in dorsal view, the anterior carapace in lateral view, the telson, and the first and third pereopods of Atya brevirostris are available in De Man (1892). Nobili (1900) gave stylized sketches of the rostrum in dorsal view, showing the difference between Atya brevirostris and the variety De Mani. Roux (1928a) provided figures of the rostrum in lateral and dorsal view, of the first, second, third, and fifth pereopods, and of the first and second pleo- pods, illustrating the juvenile appearance of the species, which he called Pseudatya beauforti. The quite inaccurate illustrations that contributed to the confusion about the identity of Vanderbiltia rosamondae were published by Boone (1935); they show the carapace and anterior appendages in dorsal and lateral view, the tail fan, the "petasma," the third maxilliped, all 5 pereopods, and the second pleopod. Edmondson (1935) showed the anterior adult carapace and rostrum in lateral view and the larval pereopods, antenna, and telson of the species that he called Atya serrata. Substitutions for the incorrect illustrations of the type-specimen of Vanderbiltia rosamondae?includ- ing the anterior end in lateral view, the rostrum in dorsal view, the telson and right uropod, the antennule and antenna, and the first, second, third, and fifth pereopods?were offered by Hol- thuis (1953). Finally, Johnson (1958) published a distribution map for Atya pilipes (including A. serrata). DIAGNOSIS.?Rostrum usually bent distinctly ventrad, unarmed or with 1 ventral tooth, less frequently with 2, very rarely with more; ptery- gostomian angle of carapace bluntly acute, not spinous; fixed teeth on posterior margin of telson inconspicuous; chelae dimorphic, adapted for fil- ter-feeding or sweeping, not for selective picking or scraping alone. COLOR NOTES.?None. SIZE.?The available material includes males with carapace lengths of 3.0 to 5.8 mm, nonovi- gerous females, 3.1 to 7.9 mm, and ovigerous females, 4.6 to 8.8 mm. The maximum carapace length corresponds to a total length of about 34 mm. The largest specimens recorded in the liter- ature that seem to be reliably identified with A. pilipes are the ovigerous females mentioned by Roux (1928b:218) from Sumba that measured 36 to 38 mm overall, virtually the same as the "lVfc inch" cited by Newport (1847) for the holotype. Specimens of 32 to 41 mm from New Guinea were probably accurately identified by De Man (1915:406), but those of 37 to 45 mm recorded from Celebes by Schenkel (1902:500) may be slightly suspect, and certainly the largest speci- mens (57 to 61 mm) mentioned by Ortmann (1890:467) from "Siidsee" and Samoa are prob- ably referable to Atyopsis spinipes, a species that Ortmann subsequently (1895:409) treated as a senior synonym of Atyoida pilipes. Holthuis (1980:70) indicated total lengths of 20 to 45 mm for A. pilipes (including A. serrata), presumably accepting the Celebes record of Schenkel (1902:500). DISTRIBUTION AND SPECIMENS EXAMINED.?The species representing the somewhat variable con- cept here proposed for Atyoida pilipes has been reliably reported from the Philippines and eastern Lesser Sunda Islands at about 120? east longitude eastward through the Pacific high islands, as far north as Rota in the Marianas at about 14? north, as far south as Rapa in the lies Tubuai at about 271/fe? south, and as far east as Magareva in the lies Gambier at about 135? west. It seems to be absent from the Asiatic mainland. Material has been examined from the following localities. Numbers in parentheses following the specimens listed are measurements, in mm, of postorbital carapace lengths. NUMBER 384 13 FIGURE 3.?Atyoida pilipes (all from ovigerous female from Samoa, carapace length 6.8 mm): a, cephalothorax and appendages; b, rostrum, left aspect; c, anterior carapace and appendages, dorsal aspect; d, telson and uropods; e, posterior end of telson;/, left 3rd maxilliped; g, same, distal end; h, left 1st chela and carpus; i, left 2nd chela and carpus; j , left 3rd pereopod; k, same, dactyl; /, same, flexor surfaces of propodus and dactyl; m, distal part of lateral branch of right uropod. (Magnifications: a, X 5.2; c, d,fj, X 10.8; b, e, g-i, k-m, X 21.5.) PHILIPPINES: (1) Mountain stream back of Romblon, Romblon Island, 16 (6.9), 26 Mar 1908, Albatross Philippine Expedition. (2) Nonu- can River, Mindanao, 18c5 (5.0-5.6), 119 (5.8- 8.3), 15 ovig 9 (5.8-8.8), 6 Aug 1909, Albatross Philippine Expedition. CAROLINE ISLANDS: Arakabesan Island, Palau?(1) Nger Dis, 7?21'OO"N, 134?27'00"E, hand col- lected below Japanese reservoir in riffle area among vegetation, 1 ovig 9 (6.9), 9 Aug 1976, J. June. Babelthuap Island, Palau?(2) Stream south of village of Olei, Ngarehelong Peninsula, 14 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY FIGURE 4.?Atyoida pilipes (all from female from Samoa, carapace length 7.2 mm): a, right mandible; b, same, distal aspect; c, right 1st maxilla; d, right 2nd maxilla; e, right 1st maxilliped; / , right 2nd maxilliped. (Magnifications: all X 21.5.) 7?42'53"N, 134?37'24"E, water fresh down to last falls, then brackish, yellowish in color, V\-\x/z m/sec current, bottom volcanic rock with pot- holes, banks rock and mud, overhanging vegeta- tion (liverworts, extensive tree roots), sta 122, 636* (3.2-5.0), 37$ (3.1-6.8), 35 ovig 9 (5.5-7.4), 13 juv (2.6-2.9), 23 Aug 1955 (1000-1300 hours), liquid rotenone, R.R. Harry, H.A. Fehlmann, F.M. Bayer, Rikrik, Y. Sumang (George Vander- bilt Foundation Expedition). (3) Upper reaches of Hgeremeskang River, Hgiwal municipality, 7?32'50"N, 134?35'47"E, stream 1-3 m wide, current about \xh m/sec, bottom sand, gravel, rock, banks soil and rock without vegetation, water temperature 26? C, sta 241, 16* (4.6), 6 ovig 9 (6.3-7.6), 18 Oct 1955 (1430-1600 hours), roti- cide, HAF, YS (GVF). (4) Ilmaw stream, Hgetkip village, Airai municipality, 7?21'37"N, 134?- 31'19"E, stream 1-2 m wide, white, slightly tur- bid water, about \xh m/sec current, bottom rock, gravel, sand, trailing roots, banks rock and soil without vegetation, water temperature 26? C, sta 250, 306* (3.5-5.2), 179 (3.3-5.2), 37 ovig 9 (4.8- 7.4), 11 juv (2.5-2.7), 25 Oct 1955 (1430-1700 hours), roticide, HAF, YS (GVF). (5) Iklong stream (tributary to Alsemith River), east of Ngchemliangel village, Aimeliik municipality, 7?31'32"N, 134?29'54"E, stream 1-3 m wide (pools 9-12 m), water white, turbid (heavy rain during night), about 2 m/sec current, bottom gravel, solid rock, many cascades, banks soil and rock without vegetation, water temperature 25? C, sta 268, 16* (7.2), 2 ovig 9 (7.1, 7.4), 1 Nov 1955 (0930-1200 hours), roticide, HAF, YS, Rengiil (GVF). (6) Imengelngal stream (tributary to Al- semith River), northeast of Medorm and Ngchemliangel villages, Aimeliik municipality, 7?27'30"N, 134?30'40"E, slightly brownish and turbid, about \xh m/sec current, bottom gravel, solid rock, banks soil, solid rock, without vegeta- tion, water temperature 25.5? C, sta 269, 1 ovig 9 (7.8), 1 Nov 1955 (1300-1515 hours), roticide, HAF, YS, Rengiil (GVF). (7) Arakitaoch stream, strand zone, sta 170, 16* (5.4), 31 Oct 1956, A. Fehlmann (GVF). (8) Same, cascade zone, 556* (3.9-5.8), 399 (3.3-7.9), 102 ovig 9 (4.8-8.2), 4 Nov 1956, AF (GVF). (9) Same, source region, 46* NUMBER 384 15 FIGURE 5.?Atyoida pilipes (all from Palau specimens; a-i, male, carapace length 5.0 mm; j , juvenile, carapace length 2.6 mm; k, female, carapace length 4.2 mm; /, ovigerous female, carapace length 7.0 mm): a, rostrum, left aspect; b, telson and uropods; c, posterior end of telson; d, right 1st chela and carpus; e, right 2nd chela and carpus; / , right 3rd pereopod; g, same, dactyl; h, exopod and endopod of right 1st pleopod; i, right appendix masculina and appendix interna, mesial aspect; j , juvenile rostrum; k, adolescent rostrum; /, variant of adult rostrum. (Magnifications; b,f, X 10.8; a, c-t, g, h,j-l, X 21.5; i, X 53.8.) (4.0-5.2), 29 (4.0, 5.8), 5 ovig 9 (4.7-7.0), 5 Nov 1956, AF (GVF). (10) South fork of Arakitaoch River, cascade zone, sta 170-A, 49a gastiva Ortwann."?Urita, 1921:216 [erroneous spelling of Atya gustavi Ortmann]. Atya molluscensis.?McCuWoch and McNeill, 1923:57 [erro- neous spelling]. [?]Ortmania sp.(?).?Blanco, 1935:36, pi. 3: figs. 33-40 [er- roneous spelling]. [?]Atya serrata.? Blanco, 1935:31, pi. 1: figs. 1-4. Atya spinifera.?Edmondson, 1935, fig. 5j,k [erroneous spell- ing] [?]Ortmannia sp.?Johnson, 1961:146. REVIEW OF LITERATURE.?The original descrip- tion of Atya spinipes by Newport (1847) so avoids mention of the characters that are currently of specific or even generic diagnostic importance that it is little wonder that there has been, and still is, confusion about the true identity of the species, even by those who have examined the holotype. A. Milne-Edwards (1864) believed that the species he named Atya armata differed from A. spinipes (= Atyopsis spinipes) in having the third pereopods more spinulose and the major meral spine proximal rather than distal to the minor one; but these differences may be observed in comparing large males and females of the same species (Figures 16^, 18c). Schmeltz (1869:x) re- corded specimens as Atya armata from more than 1000 feet (305 m) at Upolu. Miers (1880:382, pi. 15: figs. 3-6) was the first to indicate what may be true differences between Atya spinipes (from Samoa) and A. moluccensis {? Atyopsis moluccensis) (from Java), but he failed to recognize the differ- ences between Newport's type specimens of Atya spinipes and A. pilipes. It is apparent from the lengths indicated for the specimens, as recorded by Ortmann (1890:467), that at least those from the "Siidsee" and some of those from Samoa are probably A. spinipes rather than A. pilipes. From the size (total length 77 mm, cephalothorax 19 mm) and the presence of only three ventral teeth on the rostrum, the specimens from northern Celebes named "Atya dentirostris" by Thallwitz (1891a: 101, 102) seem to belong to the concept of Atyopsis spinipes postulated here. Nothing in his other description and illustrations, published in the same year (Thallwitz, 1891b:26, pi. 1: fig. 7), refutes this assumption. It would seem, from the small number (3-7) of ventral rostral teeth indi- cated, that some of the specimens recorded from Sumatra by De Man (1892:359, 360) and those from Celebes, Salajar, Timor, and Flores belong to Atya spinipes, rather than to A. moluccensis. Ort- mann (1895:407-409) followed Miers (1880) in recognizing Atya moluccensis and A. spinipes and in synonymizing A. pilipes with the latter. The total length (53 mm) would suggest that the form described as Atya brevirostris var. De Mani from Kepulauan Mentawi by Nobili (1900:475) might be Atyopsis spinipes. The Halmahera specimens assigned to Atya moluccensis by De Man (1902:893) probably represent Atyopsis spinipes, as indicated by the few (3-5) ventral rostral teeth. The size (total length 37-45 mm) of the ovigerous females NUMBER 384 37 recorded from Tomohon, Celebes, as Atya breviros- tris by Schenkel (1902:500) suggests the possibility that they might belong to Atyopsis spinipes. The record of PAtya spinipes from the Gambier Islands, Tuamotus, by Nobili (1907:353) must be ques- tionable, because only Atyoida pilipes is known with reasonable certainty from that area. The two females with three or four ventral rostral teeth recorded from "Sagamibai" as PAtya moluc- censis by Balss (1914:26) would seem to be iden- tifiable with Atyopsis spinipes, but the locality must remain suspect for the time being. There is little doubt that the species called Atya mollucensis in his two papers on habits and synonymy by Cowles (1915a, 1915b) are refereable to Atyopsis spinipes. The record of "Atya gastiva Ortwann" from the Kagoshima Wan region of Kyushu, Japan, by Urita (1921:216) is annoyingly ambiguous. Bou- vier (1925:305) believed that the differences be- tween Atya spinipes and Atya moluccensis (length of the rostrum and position of the spur on the third pereopod of the adult male) were insufficient specific characters, that Atya spinipes and A. pilipes would eventually be demonstrated to be distinct, and that the former should be adopted as a synonym senior to Atya moluccensis. In the species that Roux (1926a:218, 219) called Atya moluccensis from Sri Lanka, Burma, Strait of Malacca, and New Caldeonia, the number of ventral rostral teeth varied, usually from 6 to 10, but might be reduced to 2 to 4 in large specimens. The Phil- ippine specimens called Atya serrata and Ortmania sp. by Blanco (1935:31, 36, pi. 1: figs. 1-4, pi. 3: figs. 33-40) are difficult to identify; the first may be Atyopsis spinipes, and the other may be a juve- nile of the same species. Edmondson (1935) rec- ognized Atya spinipes and A. moluccensis as distinct species. Finally, Yu (1974) recorded Atya spinipes under the name Atya moluccensis from Taiwan. PUBLISHED ILLUSTRATIONS.?The first illustra- tions of Atya spinipes are the lateral views of the rostrum and of the third pereopod of a large Samoan male furnished by Miers (1880) to show the characters that distinguish the species from Atya moluccensis. Those drawings are considerably better than those of the rostrum and first pereo- pod of Atya gustavi offered by Ortmann (1890). Somewhat better than the latter are the dorsal view of the carapace, lateral view of the rostrum, and the third pereopod of Atya dentirostris in Thall- witz (1891b). The dorsal view of the rostrum of Atya brevirostris var. De Mani by Nobili (1900) is so stylized as to be of little interest. Somewhat better is the series of five lateral views of the variable rostrum and the first pereopod of Atya brevirostris in Schenkel (1902), that may or may not be referable to A. spinipes. Cowles (1915a) presented outline drawings of the first pereopod extended and flexed, a fine toto figure of a female in lateral view (enlarged 50%), the anterior portion of a female in lateral view (enlarged 100%) showing the chelae expanded for filter feeding, and two hairs from the fingers of a cheliped (enlarged much more). The same author (1915b) offered a series of outline drawings of the merus of the third, fourth, and fifth pereopods of five speci- mens varying in total length from 42 to 70 mm, showing spine modification with growth in Phil- ippine material. De Man (1915) gave excellent illustrations of dorsal views of the anterior portion of the carapace and lateral views of the rostrum of a young male and of an ovigerous female, as well as of partially hooked hairs on the dorsal surface of the carapace. Blanco (1935) offered somewhat stylized drawings of the anterior end in lateral view, both chelipeds, and the posterior margin of the telson of the Philippine species that he called Atya serrata and of the rostrum in lateral view, the antennal scale, the antennule, the first pereopod, dactyls of the third and fifth pereopods, the uropodal dieresis, and the posterior margin of the telson of a young Philippine specimen that he called Ortmania sp.(?). A lateral view of the rostrum and merus of the third pereopod of a male, probably from the Fijis, and both done in his usual stylized manner, were published by Edmondson (1935). Kubo (1938) gave poor half- tones of dorsal and lateral toto photographs and offered outline illustrations of the anterior part of the carapace in lateral view, the telson, the man- dible in two aspects, the first, second, third, and fifth pereopods, the first pleopods of the male and 38 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY the female, the appendix masculina, and the endopod of the second female pleopod, as well as a distribution map of the species that he called Atya moluccensis. Finally, a diagram of the nervous system was included in Gabriel (1963). Most recently, Yu (1974) contributed good line draw- ings of the entire shrimp in lateral aspect, the carapace in lateral aspect, the telson and uropod, the chela and carpus of the first pereopod, the third pereopod, and the second pleopod in both male and female. DIAGNOSIS.?Rostrum abruptly terminating in h n FIGURE 20.?Atyopsis spinipes (all from male from "Cabugao" River, Catanduanes Island, Philippines, carapace length 13.8 mm): a, entire specimen, left aspect; b, anterior carapace and appendages, right aspect; c, same, dorsal aspect; d, telson and uropods; e, posterior end of telson; /, 6th abdominal somite and median tubercle on 5th, ventral aspect; g, preanal carina, left aspect; h, right mandible; i, right 1st maxilla; j , right 2nd maxilla; k, right 1st maxilliped; /, right 2nd maxilliped; m, right 3rd maxilliped; n, same, distal end. (Magnifications: a, X 2.1; b-d,f, i-m, X 5.2; e,g, h, n, X 21.5.) NUMBER 384 39 comparatively broad apex, armed ventrally with 2-6 discrete teeth; endopod of 1st pleopod of male more than 1% times as long from proximal articulation to base of retinaculate projection as maximum width, not including marginal spines. COLOR NOTES.?Photographs were taken by F. M. Bayer of the larger male and larger female collected by the George Vanderbilt Foundation Expedition in 1955 at station 250 on Babelthuap Island, Palau, and are now in the USNM collec- tions. The color has deteriorated in the transpar- ency, but the color pattern is still clear. The male (carapace length 12.1 mm) had a broad pale streak in the dorsal midline extending from the tip of the rostrum to the end of the telson; ante- riorly, the streak was about as wide as the base of the rostrum and it extended posteriorly at this width, broadening slightly on the posterior third of the carapace and reaching as an even slightly broader stripe over the abdomen to the base of the telson, thence tapering a little to coincide finally with the extent of the posterior margin of the telson. Laterally, the carapace was marked with 6 or 7 light longitudinal stripes outlined with dark pigment, the 4 ventral ones broadest and anastomosing near the anterior and posterior margins of the carapace; the abdomen bore about 6 similar stripes anteriorly, merging into 4 on the fifth somite and into 2 on the anterior half of the sixth. The 3 posterior pairs of pereopods were light colored with dark spots on the basis or ischium and near the proximal end of the merus, a band at the distal end of the carpus, 2 bands on the propodus (proximal and distal thirds), and a less distinct band darkening the extreme distal end of the propodus and the proximal end of the dactyl. The female (carapace length 14.7 mm) had a similar middorsal streak that became ob- scure at its anterior and posterior extremities and was completely obliterated on the gastric region, apparently by the underlying stomach contents. The female was otherwise much darker than the male, the lateral stripes on the carapace and abdomen being much less distinct, and the 3 posterior pairs of pereopods almost uniformly dark colored. The photograph of a female specimen from the Ryukyu Islands, called Atya moluccensis by Kubo (1938, fig. 22), is too poor to permit comparison of the color pattern with that in the Palau speci- mens, but the lateral stripes seem to be more prominent than they were in the Palau female photographed by Bayer. SIZE.?The specimens studied include males with carapace lengths of 6.2 to 13.8 mm, nono- vigerous females, 5.3 to 14.7 mm, and ovigerous females, 8.1 to 19.6 mm. The largest specimen has an overall length of about 71 mm. Of the specimens listed by Johnson (ms) from localities presumed to fall within the range of Atya spinipes (= Atyopsis spinipes), the largest male, from Sa- moa, had an overall length of 65.5 mm (compa- rable to a carapace length of about 18 mm), a Philippine ovigerous female had a total length of no more than 36.0 mm (equivalent to a carapace length of about 10 mm), and the largest female, also ovigerous and also from the Philippines, had a length of 53.7 (similar to a carapace length of about 15 mm). The largest male that can be positively identified with this species seems to be the "old male, 70 millimeters in length, with 3 teeth on the ventral border of the rostrum" re- corded by Cowles (1915b: 148) from "a mountain stream near Manila." These measurements might suggest that Atyopsis spinipes is a smaller species than A. moluccensis, the largest male of the former being little more than 80 percent as long as the maximum length in the latter, and the smallest ovigerous females being only about 60 percent as long in A. spinipes as they are in A. moluccensis. DISTRIBUTION AND SPECIMENS EXAMINED.? Specimens available for this study came from the Philippines, Palau, and the Fijis. The most relia- ble records in the literature suggest that A. spinipes ranges from the Philippines and eastern Lesser Sunda Islands at about 120? east longitude north- ward to Taiwan and as far as Tokuno-shima in the Ryukyus and eastward as far as Samoa. The occurrence of the species in the Sagami Nada region of Japan, reported by Balss (1914), as well as the presence of "Atya gastiva " near Kagoshima Wan, recorded by Urita (1921), were not men- 40 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY v u g FIGURE 21.?Atyopsis spinipes (all from male from "Cabugao" River, Catanduanes Island, Philippines, carapace length 13.8 mm); a, right 1st pereopod; b; right 2nd pereopod; c, right 3rd pereopod; d, same, dactyl; e, same, flexor surfaces of propodus and dactyl; / , right 4th pereopod; g; same, dactyl; h, left 5th pereopod; t, same, dactyl; j , endopod of right 1st pleopod, posterior aspect; k, same, anterior aspect; /, right appendix masculina and appendix interna, mesial aspect. (Magnifications: a-cj, h, X 5.2; e,j-l, X 10.8; d, g, i, X 21.5.) tioned by Kubo (1938), and these records should probably be questioned until they are positively confirmed. Verification is also needed of the Ma- gareva record in Nobili (1907) and Seurat (1934) in order to eliminate the possibility of confusion between Atyposis spinipes and Atyoida pilipes, which seems to be established on the Gambier island. Material has been examined from the following localities. Numbers in parentheses following the specimens listed are measurements, in mm, of postorbital carapace lengths. PHILIPPINES: British Museum (NH) 43-6, no locality, holotype of Atya spinipes, 1$? (~15), Mr. Cumings. No locality, 1 ovig $ (19.6), E.A. Mearns. (1) Tayabas River, Luzon, 16" (13.5), 25 Feb 1909, Albatross Philippines Expedition. (2) "Cabugao" River, Catanduanes Island, 16* (13.8), 9 June 1909, Albatross. (3) "Varadero Mountain," Mindoro(?), 1$ (13.4), 23 Jul 1908, Albatross. (4) Malaga River, Hinunangan Bay, Leyte, 1$ (11.2), 4 juv (3.1-4.8), 30 Jul 1909, Albatross. (5) Man- anga River, Cebu, 19 (11.7), 25 Aug 1909, Alba- tross. CAROLINE ISLANDS: Babelthuap Island, Palau?(1) Stream south of village of Olei, Ngarehelong Peninsula, 7?42'53"N, 134?37'24"E, water fresh down to last falls, then brackish, yellowish in color, V\-\xh m/sec current, bottom volcanic rock with potholes, banks rock and mud, overhanging vegetation (liverworts, extensive tree roots), sta NUMBER 384 41 122, 3c5 (8.0-9.2), 59 (4.5-11.0) 4 ovig 9 (9.6- 11.9), ljuv(2.5),23Aug 1955 (1000-1300 hours), liquid rotenone, R.R. Harry, H.A. Fehlmann, F.M. Bayer, Rikrik, Y. Sumang (George Vander- bilt Foundation Expedition). (2) Ilmaw stream, Hgetkip village, Airai municipality, 7?21'37"N, 134?3ri9"E, stream 1-2 m wide, white, slightly turbid water, about IV2 m/sec current, bottom rock and soil without vegetation, water tempera- ture 26? C, sta 250, 26 (6.2, 12.1), 1$ (14.7), 1 ovig 9 (12.4), 25 Oct 1955 (1430-1700 hours), roticide, HAF, YS (GVF). (3) Airisong stream, 2.4-3.2 km north of Hgetkip village, Airai mu- nicipality, 7?22'52"N, 134?3r30"E, stream 1-3 m wide, clear, white water, \xh m/sec current, bottom solid rock, sand, gravel, banks solid rock and soil without vegetation, water temperature 27? C, sta 251, 16 (10.5), 26 Oct 1955(1200-1600 hours), roticide, HAF, YS (GVF). (4) Imengel- ngal stream (tributary to Alsemith River), north- east of Medorm and Ngchemliangel village, Ai- meliik municipality, 7?27'3O"N, 134?30'40"E, slightly brownish and turbid, about \xh m/sec current, bottom gravel, solid rock, banks soil, solid rock, without vegetation, water temperature 25.5? C, sta 269, \6 (12.0), 1 Nov 1955 (1300- 1515 hours), roticide, HAF, YS, Rengiil (GVF). (5) Arakitaoch stream, strand zone, sta 170, 1 ovig 9 (11.3), 31 Oct 1956, A. Fehlmann (GVF). (6) Same, cascade zone, 26 (8.6,9.9), 49 (7.3-9.0), 1 ovig 9 (10.3), 4 Nov 1956, AF (GVF). (7) South fork of Arakitaoch River, cascade zone, sta 170- A, 26 (8.0, 11.8), 19 (11.2), 1 ovig 9 (8.1), 26 Nov 1956, AF (GVF). (8) Same, source zone, 26 (8.3, 10.3), 19 (8.5), 27 Nov 1956, AF (GVF). Koror Island, Palau?(9) Tibdul stream draining into Iwayama Bay, 7?2O'1O"N, 134?30'06"E, stream 1-2 m wide, water white, clear, current about 2 m/sec (many falls and cascades), bottom rock and gravel, banks rock, soil, sawgrass, water tem- perature 26.5? C, sta 216, 19 (8.8), 11 Oct 1955 (1500-1800 hours), roticide, HAF, YS. Fiji ISLANDS: NO locality, 1 ovig 9 (13.3), Aug 1928, H.S. Ladd. VARIATION.?Of the 47 specimens examined, 44 have the rostrum intact; of these, 5 have 2 ventral rostral teeth, 10 have 3, 15 have 4, 11 have 5, and 3 have 6. Apparently none of the 15 males, ranging in carapace length from 6.2 to 13.8 mm, is large enough to display the spur on the merus of the third pereopod that is character- istic of old Atyopsis males. Of all the specimens in which the third pereopod is present, it bears 2 spines in 20 specimens, 3 spines in 12, 4 in 6, and 1 spine in 2 specimens. The merus of the fourth pereopod bears 4 spines in 33 specimens, 3 in 10, 5 in 2, and 2 spines in just 1 specimen. The fifth pereopod bears 3 meral spines in 21 specimens, and 2 in 13 specimens. The dactyl of the third pereopod bears 4 flexor spines in 28 specimens, 5 in 14, and 3 in 6. The dactyl of the fourth pereopod bears 5 flexor spines in 19 specimens, 3 and 4 in 8 specimens, each, and 6 in 2 specimens. In adult specimens, the number of spines on the flexor margin of the dactyl of the fifth pereopod varies from 24 to 34 spinules, without any real numerical superiority for any particular count. There are from 13 to 19, usually 14 to 16, spines on the dieresis of the lateral branch of the uropod. Finally, the endopod of the first pleopod of all but the smallest males varies from 1.7 to 2.2 times as long as wide (not including marginal spines) (Figures 2\j,k, 22). ECOLOGICAL NOTES.?Schmeltz (1869) indi- cated that the species he called Atya armata oc- curred at an altitude of more than 1000 feet (305 m) at Upolu. Cowles (1915a) noted that Atyopsis spinipes (called by him Atya moluccensis) is abun- dant in mountain streams, uncommon in lowland ones. Individuals are so prone to clinging to roots, grasses, etc., in swift water that they will grasp any suitable object, even when removed from the water. In moving water, they spread their chelae in a filter-feeding attitude; in still water, they remain motionless and do not sweep the substrate for food as some atyids do. Cowles (1915a: 14) never observed them making burrows of any kind. Shokita and Nishijima (1976) stated that the population on Tokuno-shima, Ryukyus, occurred "in the shallow grass beds of middle streams, where the water flow is relatively rapid." The ecological information gathered by H.A. Fehl- mann and other members of the George Vander- bilt Foundation Expeditions to Palau indicates 42 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY 9.0-4 9.2-2 9.9-5 10.3-5 10.5-4 11.8-6 12.0-2 12.1-6 13.5-5 FIGURE 22.?Atyopsis spinipes, variation in shape of endopod plate of right 1st pleopod of males from Palau (a, m, George Vanderbilt Foundation sta 250; bj, g, GVF sta 122; c, d, i, k, GVF sta 170-A; e, h, GVF sta 170;>, GVF sta 251; /, GVF sta 269) and Philippines (?, Tayabas River, Luzon). Left numerals (decimals) indicate carapace lengths of specimens in mm, right numerals (italics) denote number of ventral rostral teeth. that Atyopsis spinipes frequents streams in those islands that vary in width from 1 to 3 meters, have a current velocity of about V* to 2 meters per second over a rock bed sometimes covered with gravel, sand, or mud between banks of rocks and soil, with or without liverworts, tree roots, or other vegetation; the water varied from white to brownish, clear to turbid, and in temperature from 25.5? to 27?C. LIFE HISTORY NOTES.?None. COMMON NAMES.?See this section under Atyop- sis moluccensis (p. 34). ECONOMIC IMPORTANCE.?Cowles (1915a: 11) remarked that Philippine atyids in general, of which the species of Atyopsis are the largest, "are of almost no commercial value, although they are sometimes eaten when food is very scarce." REMARKS.?If, as proposed here, Atyopsis, con- tains more than one species (see "Remarks" under A. moluccensis, p.34), the true identity of the unique holotype of A. spinipes becomes more im- portant than it would be if the genus were mon- otypic. Thanks to the kindness of R.W. Ingle of the British Museum (Natural History), I have had the opportunity to examine that specimen, but the resulting conclusions are hardly decisive. Newport's species should be distinguishable from A. moluccensis by the form and dentition of the rostrum, the position of the massive spur on the merus of the third pereopod of old males, and the form and proportions of the endopod of the first pleopod in the male. The type specimen of A. spinipes is dry and brittle. The posterior portion of the carapace is firmly wedged into the first abdominal somite, making actual size determi- nation difficult; my best estimate is that its postorbital length must be about 15 mm and the total length about 55 mm. The rostrum is broken jaggedly so that its intrinsic shape is indetermin- able, and the ventral margin, which bears the diagnostic series of teeth, is almost completely obliterated, thereby eliminating the rostrum as a clue to the identification. Secondly, I believe that the specimen is a female; if true, the other two diagnostic characters could not be ascertained from the specimen no matter what its condition. Under these unfortunate circumstances, it seems best to assume that the specimen from the Phil- ippine collection of "Mr. Cumings" (spelling on specimen label) belongs to the form that seems to be most prevalent in the Phlippines. All 10 of the Philippine specimens in the Smithsonian collec- tions belong to the form here called Atyopsis spi- nipes, as apparently also do those reported by Cowles (1915b: 147, 148) "from a mountain stream near Manila" that have 3 to 6 teeth on the ventral margin of the rostrum. Except for the remark by Blanco (1935:30) that the rostra in his 3 specimens from the Dumaguete River, south- eastern Negros, are "armed with numerous little teeth on ventral keel" and the fact that Johnson (ms) included the Philippines in the range oiAtya spinipes moluccensis, there seems to be no indication that the form here assigned to Atyopsis moluccensis NUMBER 384 43 occurs in the Philippines. (Johnson mentioned only Newport's type specimen and 6 specimens from a "mountain stream near Mariveles, Manila Bay, Luzon," received from Cowles, as all of the Philippine material seen by him.) Genus Australatya, new genus TYPE-SPECIES.?Atya striolata McCulloch and McNeill, 1923. ETYMOLOGY.?The feminine name Australatya is derived from the Latin adjective australis, south- ern, used as a prefix to the name Atya. DIAGNOSIS.?Body pigmented, eyes well devel- oped; rostrum not strongly compressed laterally, median dorsal carina unarmed, ventral keel with 4-8 teeth; anterior margin of carapace armed with antennal spine, pterygostomian margin rounded; supraorbital spines absent; telson with setigerous posterior margin overreaching postero- lateral angles; 3rd maxilliped with uncinate ter- minal spine partially concealed by dense setae; pereopods without exopods; 1st and 2nd pereo- pods with chelae monomorphic, without palm, fingers tipped with brushes of long setae appar- ently adapted for filter feeding, carpus of both appendages excavate distally, little longer than broad; 3rd pereopod without meral spur in large males; branchial complement consisting of 5 pleurobranchs, 3 arthrobranchs, 1 podobranch, 5 epipods (reduced posteriorly), no mastigo- branchs; 1st pleopod of male with endopod ta- pering to slender apex; 2nd pleopod of male with appendix masculina subcylindrical, spinose on less than distal third of length. RANGE.?Australatya is confined to rivers and streams along the east coast of Australia. REMARKS.?Although Australatya resembles Atyoida more closely than it does Atyopsis or Atya, it seems clear to me that the species originally called Atya striolata is excluded from that genus by the broadly rounded pterygostomian margin of the carapace, the reduced epipods on the third and fourth pereopods, the absence of mastigo- branchs, the monomorphic chelae apparently adapted primarily for filter feeding, and the elon- gate appendix masculina spinose on less than the distal third of its length. Only one species is known. Australatya striolata (McCulloch and McNeill, 1923), new combination FIGURES 23, 24 Atya striolata McCulloch and McNeill, 1923:55, pi. 9: figs. 3, 4, pi. 10: fig. 3 [type-locality: Norton's Basin, Nepean River, New South Wales, Australia].?Anderson, 1926:253.?McNeill, 1926:325.?Roux, 1926b:253.? McNeill, 1929:148.?Riek, 1953:111; 1959:252, 254.? J.A. Bishop, 1967:117, 119.?Williams, 1968:145, 146; 1981:1121.?Smith and Williams, 1982:343. Atyoida striolata.?Smith and Williams, 1982:345, figs. 1-7. REVIEW OF LITERATURE.?The original descrip- tion by McCulloch and McNeill (1923) is accom- panied by accounts of the type-locality and be- havior of the species. Roux (1926b) added a few details and measurements to the initial descrip- tion, and the editor, Anderson (1926), recorded an additional locality for the species in a footnote to Roux's article. A third locality for the species was mentioned by McNeill (1929). The first Queensland record was published by Riek (1953), who gave a brief description and color notes based on the new material and remarked that the specimens differed slightly from those in the type- series. The same author (1959:254) referred to Atya striolata as an atyid "relict element." J.A. Bishop (1967:119) also mentioned that Atya striol- ata might be a "relict species" and recorded spec- imens as "facultative troglobes" from caves in New South Wales. In his treatise on Australian freshwater invertebrates, Williams (1968:146) of- fered a key to the Australian atyid genera and made the enigmatic remark that "Atya striolata is the only species [of Atya] known with any detail but another is recorded in the literature." Finally, Smith and Williams (1982) nearly preempted the purpose of this review by reinstating Atyoida for A. bisulcata, A. pilipes, and A. striolata; they fully redescribed the latter from topotypic material, noted that there is little variation throughout the range of the species, demonstrated conclusively that A. striolata is a protandrous species, and argued that protandry may have clear adaptive 44 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY value by increasing the number of larvae avail- able to attempt the precarious ascent of swift coastal streams. PUBLISHED ILLUSTRATIONS.?McCulloch and McNeill (1923) gave an excellent toto drawing in lateral view of the holotype, showing the color pattern, but substituting the three posterior per- eopods from a smaller paratype of those missing in the holotype; they also depicted the telson in dorsal view and reproduced a photograph of the type-locality. Apparently the only other pub- lished illustrations of Australatya striolata are those in the extensive series in Smith and Williams (1982), consisting of outline drawings of the hol- otype in lateral view (lacking four posterior per- eopods), the telson in dorsal view, an antennule, antenna, labrum, paragnath, left and right man- dibles, both maxillae, two anterior maxillipeds, second pereopod, all five pleopods, and uropod, all from the holotype; the third maxilliped, first pereopod, and third to fifth pereopods of a female paratype; the first and second pleopods of a male topotype; a geographical distribution map for the species; and histograms indicating protandry in two populations of the species. (Several of the illustrations presented herein duplicate those of- fered by Smith and Williams, but they were prepared before I was aware of the latter contri- bution, and it seems best to include them for what they may be worth.) DIAGNOSIS.?Rostrum nearly horizontal, armed ventrally with 4-8 teeth; pterygostomian margin of carapace broadly rounded; telson with single median tooth on posterior margin; chelae mono- morphic, adapted for filter feeding; appendix masculina unusually elongate, spinose on less than third of length. COLOR NOTES.?McCulloch and McNeill (1923:56) offered the following detailed account of the color pattern with the original description of the species. Green in life, closely speckled with microscopic blackish- brown dots. A broad, light yellowish, median band extends from the tip of the rostrum to the end of the telson, which is closely speckled and sharply defined by blackish borders. Five narrow longitudinal stripes along each side of the carapace, some of which are more or less interrupted; these are light in colour without darker speckles, and have dark margins. Two similar stripes along each side of the abdomen. Uropods pale green basally, changing to light blue termi- nally; the outer with a light distal spot with an ill-defined darker border, the inner with a similar but less distinct spot. Antennular peduncles with a light stripe on the upper surface, the remainder green. Limbs and antennae translu- cent green; pencils of the fingers darker at the bases and tips. Riek (1953:112) gave the following color de- scription of specimens from Cave Creek near the Queensland-New South Wales border. Dull reddish-brown speckled with microscopic black dots. A broad, almost white, median band extending from the ros- trum to the telson and sharply defined by a black border of irregular thickness. Five longitudinal stripes along each side of the carapace, somewhat irregular in outline and bordered in black similar to the median dorsal band. Two similar stripes occur along each side of the abdomen. SIZE.?The 4 males available to me from the type-locality vary in carapace length from 7.8 to 8.7 mm and in total length from 32 to 36 mm. Measurements available from the literature sug- gest that males range in carapace length from 5.0 to 10.4 mm (total length, 22-43 mm); females with carapace lengths of 7.6 to 13.0 mm (total length, 33-57 mm), with ovigerous specimens as short overall as 30 mm (= carapace length of about 6.8 mm). The species is almost certainly protandrous, as shown by histological studies con- ducted by Smith and Williams (1982). DISTRIBUTION AND SPECIMENS EXAMINED.?The species occurs in rivers and streams along the east coast of Australia from Cooktown, northern Queensland, to the Shoalhaven River, New South Wales (Smith and Williams, 1982, fig. 5). The following lot has been examined. The numbers in parentheses are measurements, in mm, of postorbital carapace lengths. NEW SOUTH WALES, AUSTRALIA: Norton's Basin, Nepean River, paratypes (?) of Atya striolata, 46 (7.8-8.7), 23 Nov 1919, A.R. McCulloch. VARIATION.?Smith and Williams (1982:355) compiled data on 33 proportional attributes of 20 male, 24 female, and 2 juvenile topotypes, as well as of material from 18 other populations, and concluded that "the species is remarkably uni- form over its whole known range," except that NUMBER 384 45 FIGURE 23.?Australatya striolata (all from male paratype(?) from Norton's Basin, Nepean River, New South Wales, carapace length 8.7 mm): a, entire specimen, left aspect; b, anterior carapace and rostrum, left aspect; c, anterior carapace and appendages, dorsal aspect; d, telson and uropods; e, posterior end of telson;/, right mandible; g, right 1st maxilla; h, right 2nd maxilla; i, right 1st maxilliped; ^ , right 2nd maxilliped; k, right 3rd maxilliped; /, same, distal end. (Magnifications: a, X 2.6; c, d, X 5.2; b, g-k, X 10.8;