war a) Can Pai 7 a mv a =k 7 7 mt : a an ; in acer “| : ‘ SMITHSONIAN IN STITUTION UNITED STATES NATIONAL MUSEUM PROCEEDINGS OF THE UNITED STATES NATIONAL MUSEUM VOLUME 114 NUMBERS 3467-3475 UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1964 Publications of the United States National Museum The scientific publications of the United States National Museum include two series, Proceedings of the United States National Museum and United States National Museum Bulletin. In these series are published original articles and monographs dealing with the collections and work of the Museum and setting forth newly acquired facts in the fields of anthropology, biology, geology, history, and technology. Copies of each publication are distributed to libraries and scientific organizations and to special- ists and others interested in the various subjects. The Proceedings, begun in 1878, are intended for the publication, in separate form, of shorter papers. These are gathered in volumes, octavo in size, with the publication date of each paper recorded in the table of contents of the volume. In the Bulletin series, the first of which was issued in 1875, appear longer, separate publications consisting of monographs (occasionally in several parts) and volumes in which are collected works on related subjects. Bulletins are either octavo or quarto in size, depending on the needs of the presentation. Since 1902 papers relating to the botanical collections of the Museum have been published in the Bulletin series under the heading Contributions from the United States National Herbarium. Frank A. Taytor, Director, United States National Museum. CONTENTS Pages Bonp, Gorman M. Geographic variation in the thrush 3 Hylocichla ustulata. One figure. No. 3471, published March, 6,-1963.3¢ . -.o-- . 373-387 DreisspacH, R. R. New Eneeien e Epica) ae a Auplopus, from the Americas south of the United States (Hymenoptera: Psammocharidae). Thirteen plates. No. 3468, published March 19,:1963 >. warcenousi> > myedagina 20-211 New species: Auplopus shannoni, A. semialatus, A. splendens, A. femurrubrus, A. villosus, A. nebulosus, A. opacus, A. minus, A. zeteki, A. panamensis, A. sapphirus, A. exilis, A. earinus, A. grossus, A. guatemalensis, A. aquilus, A. incognitus, A. oli- varus, A. purpureus, A. violaceus, A. blatteus, A. roseus, A. nigriculus, A. lineatus, A. quartus, A. venetus, A. minusculus, A. magnus, A. callainus, A. buscki, A. medius, A. hidalgoensis, A. fuscus, A. albifrons, A. semirufus, A. abnormalis, A. niger, A. woodi, A. anthracinus, A. clypeatus, A. aeruginosus, A. argen- tinus, A. marginalis, A. atratus, A. amoenus, A. vulcanensis, A. gaumeri, A. kathryni, A. fulgidus, A. dietzi, A. minutus, A. bequaerti, A. gertschi, A. hondurensis, A. hispidus, A. bermuden- sis, A. cyaneus, A. ferrugineus, A. argentinensis, A. batesi, A. argutus, A. pratens, A. aurarius, A. lasios, A. puniceus, A. alarius, A. striatus, A. eriodes, A. brasiliensis, A. coracinus, A. schausi, A. stagei, A. malinus, A. cordobensis, A. flavicrus, A. subaurarius, A. ater, A. pratentis, A. pygidialis, A. editorialis, A, viridulus. Furnt, Outver S., Jr. Studies of Neotropical caddis flies, I: Rhyacophilidae and Glossosomatidae (Trichoptera). Eight figures. No. 3473, published April 16,1963... . . . 453-478 NEW species: Atopsyche clarkei, A. neolobosa, Mortoniella apiculata, M. angulata, M. hodgesi, Mexitrichia aries, M. atenu- ata, M. elongata, Protoptila bicornuta. Hassroucx, Franx F. Moths of the family Acrolophidae in America north of Mexico (Microlepidoptera). Two hundred and nineteen figures. No. 3475, published terabor tn, FORA tate nical syistet. chsiele Sune moleelsios 20m Cue New species: Acrolophus acornus, A. bicornutus, A. klotsi, A. baldufi, A. luriei, A. vanduzeei, A. forbesi, A. juxtatus, A. chiricahuae, A. sinclairi, A. parvipalpus, A. serratus, A. secu- latus, A. macrophallus, A. vauriet. New subspecies: Acrolophus griseus capitatus, A. macrogaster bt- pectinicornus, A. m. unipectinicornus, A. m. laminicornus, A. laticapitanus heinrichi, A. 1. clarkei, A. sinclairi sinclairi, A. s. nelsont. III IV CONTENTS Pages New combination: Acrolophus griseus griseus, A. macrogaster macrogaster, A. laticapitanus laticapitanus, A. l. occidens, A. l. o. form leopardus. New name: Acrolophus pseudohirsutus. Horrman, Ricuarp L. A revision of the North American annelid worms of the genus Cambarincola (Oligochaeta: Branchiobdellidae). Seventy-nine figures. No3470, published"March’6,°1 963s 2° INS ae, a oie | eee New species: Cambarincola ouachita, C. restans, C. mesochorea, C. holti, C. shoshone, C. virginica, C. osceola, C. ingens, C. fallaz, C. holostoma, C. heterognatha, C. demissa. Howven, Henry F., and Cartwricut, Oscar L. Scarab WI beetles of the genus Onthophagus Latreille North of Mexico (Coleoptera: Scarabaeidae). Eleven figures and nine plates. No. 3467, published January 9,1963 ...... . . 1-185 New species: Onthophagus batesi, O. subtropicus, O. cavernicollis, O. monticolus, O. alluvius, O. knulli, O. schaefferi, O. arnetti, O. brown. New subspecies: Onthophagus polyphemi sparsisetesus, O. orpheus pseudor pheus. New combination: Onthophagus strattulus floridanus. Kissinger, Davin G. Weevils of the genus Maemactes. One figure. No. 3474, published March 19,1963. . . . 479-486 New species: Maemactes punctatus, M. imitator. New combination: Maemactes cribratus. Osraztsov, Nichotas 5S. Some North American moths of the genus Acleris (Lepidoptera: Tortricidae). Seven figures and eighteen plates. No. 3469, published May 7, 1963 cn aye ‘ 213-270 New species: Acleris macdunnought, A. santacrucis, A. klotst, A. clarkei, A. capizziana, A. incognita, A. disputabilis. New subspecies: Acleris emargana blackmoret. New combination: Acleris ptychogrammos, A. braunana, A. kearfottana, A. semiannula, A. implexana, A. senescens, A. walkerana, A. robinsoniana, A. gloverana, A. maximana. New status: Acleris implexana form gallicolana. New combination and status: Acleris logiana placidana. RANDALL, Joun E. Review of the hawkfishes (family Cirrhitidae). Sixteen plates. No. 3472, published May 28, 1963 . : » Je et Ghats aipcceaciieys 389-451 New genus: [socirrhitus. New species: Paracirrhites typee, P. xanthus, P. nisus, P. bicolor, Cirrhitichthys serratus, C. falco. Proceedings of the United States National Museum SMITHSONIAN INSTITUTION - WASHINGTON, D.C. Voiume 114 1963 Number 3467 SCARAB BEETLES OF THE GENUS ONTHOPHAGUS LATREILLE NORTH OF MEXICO (COLEOPTERA: SCARABAEIDAE) Henry F. Howpen! and Oscar L. CartwricHrT? ————— The purpose of this paper is to facilitate the identification of the species of Onthophagus of the United States and Canada and to pre- sent information on their habits and life histories. Twenty-three species and subspecies described from North America, two species introduced from Europe and Africa, and eleven previously unrecog- nized species are included. A key to the species, bibliographical references, complete new descriptions, photographs of both sexes, and maps of distribution are given. Important nomenclatural changes are presented. Acknowledgments As a staff member of the University of Tennessee, Howden was encouraged to make four field trips to Florida and Alabama to study distribution and biology of the group. Later the Entomology Research Institute of Canada supported three trips to Texas, Arizona, and Mexico in 1958, 1959, and 1960. The Smithsonian Institution sent Cartwright, as a staff member of the United States National Museum, to study types at the Museum of Comparative Zoology at 1 Entomology Research Institute, Canada Department of Agriculture, Ottawa, Canada. 2 Division of Insects, United States National Museum, Smithsonian Institution, Washington, D.C. 1 2 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 114 Harvard, the Academy of Natural Sciences of Philadelphia, and Purdue University at Lafayette, Ind. Through the assistance of the American Philosophical Society, longer field trips were made by Howden to Texas in 1954 and Arizona in 1956, and by Cartwright to Arizona in 1956. We are grateful for this support. We also acknowledge our indebtedness and extend our thanks to numerous institutions and individuals for the generous loan of speci- mens. ‘The letters in parentheses in the following list of collections studied are abbreviations used in the text when citing material; the name of the curator responsible for the loan of the specimens follows the abbreviation: Academy of Natural Sciences of Philadelphia (ansp), J. A. G. Rehn; American Museum of Natural History (amMnu), Mont A. Cazier; California Academy of Sciences (cas), H. B. Leech ; Canadian National Collection (cnc), W. J. Brown; Carnegie Museum, G. E. Wallace; Chicago Natural History Museum (cnum), Henry S. Dybas; Cornell University (cu), Henry Dietrich; Illinois Natural History Survey (Inus), M. W. Sanderson; Iowa State College (1s), J. L. Laffoon; Museum of Comparative Zoology (mcz), P. J. Darling- ton; Ohio State University (osu), J. N. Knull; Oregon State College (osc), P. O. Ritcher; San Jose State College, J. Gordon Edwards; United States National Museum (usnM); University of Arizona (ua), Floyd Werner; University of California, Davis (ucp), A. T. McClay; University of Kansas (uxa), R. H. Beamer; University of Michigan (umich.), T. H. Hubbell; University of Nebraska, W. T. Atyeo; Uni- versity of Texas (urex), H. J. Reinhard; University of Tennessee (uT); and Utah State Agricultural College, G. F. Knowlton. Individuals who kindly loaned material for study from their per- sonal collections included: B. Benesh, (ss), L. J. Bottimer (xs), B. K. Dozier, D. K. Duncan, R. J. Frederick (rsF), C. A. Frost, B. Malkin, G. H. Nelson, F. H. Parker, Mark Robinson, W. Rosen- berg (wR), G. B. Vogt, R. E. Woodruff (REw), and F. N. Young (FNY). We are especially indebted to E. B. Britton of the British Museum (Natural History) who loaned type material from the Biologia Cen- trali-Americana specimens, to H. Freude of the Sammlung des Bayer- ischen Staats in Munich for the loan of specimens from the Harold and Sturm collections, to K. Delkeskamp of the Zoologische Museum der Humbolt-Universitit in Berlin for the loan of Harold specimens, and to A. Villiers of the Muséum National d’Histoire Naturelle in Paris for comparing specimens with type material from the Harold and Laporte collections. Mrs. Anne Howden aided the project in many ways, both in the field and in the typing of the manuscript. Photographs were taken by Jack Scott of the Smithsonian staff. BEETLE GENUS ONTHOPHAGUS—HOWDEN, CARTWRIGHT 3 Review of literature The genus Onthophagus was established by Latreille in 1802. Since that time parts of the genus have received considerable attention. The American species north of Mexico were discussed by Horn in 1875. In this work he listed only five species; actually he included five species under one of these names and started much of the confusion of later authors. In 1881, he described three more. In 1887, Bates described and listed many of the species occurring in Mexico and in Central and South America. In Blatchley’s 1910 work on the Coleoptera of Indiana, eight species and varieties are listed; incorrect placement of several of these continued and compounded the subsequent confusion. In 1914, Schaeffer included 18 species in ‘‘A Short Review of the North American Species of Onthophagus,”’ the most recent comprehensive paper on the American species north of Mexico. The list of the Onthophagus of the world published by Boucomont and Gillet in 1927 greatly facilitated the study of the group. Subsequently a number of new North American species were described by Brown (1927; 1929a; 1929b). These papers were followed by a synopsis of the Mexican and Central and South American species by Boucomont in 1932. Boucomont’s paper was intended to be used mainly to identify the Mexican and South American forms, but the inclusion of many species from the United States, in both keys and footnotes, made it useful for all the Americas. The habits of our North American species are still almost unknown. Sim, in 1930, discussed the habits of Onthophagus subaeneus (under the name cribricollis). In 1935, Lindquist mentioned some of the habits of O. pennsylvanicus and O. alluvius (under anthracinus). . \ ) Ficures 39-42.—Structural details of Cambarinccla philadelphica (Leidy). 39, Lateral aspect of typical specimen; 40, jaws of the same specimen, dorsolateral aspect; 41, jaws of another specimen, lateral aspect; 42, reproductive systems, specimen from Frederick Co., Maryland. Minnesota, and south through the Appalachians and Piedmont to the Carolinas. An additional form appears to inhabit the Catskill region and adjacent areas, this is generally similar to the typical form but is apparently consistently smaller over a distinct geographic range. At several localities it occurs together with the large phila- delphica. Whether it represents (1) a depauperate northern phase of the main population, (2) a distinct sibling species, or (3) merely an assemblage of young specimens, remains to be worked out by another investigator able to collect and make field studies in New York. In central Kentucky and Tennessee occurs a form in which the peristomium, particularly the dorsal half, is hypertrophied and frequently widely flared, giving the head a campanulate appearance. This is probably a perfectly good subspecies, but the details of its overlap with philadelphica in eastern Kentucky ought to be worked ANNELID GENUS CAMBARINCOLA—HOFFMAN 345 out prior to a formal recognition. The problem is complicated by the nearby occurrence of chirocephala which also appears to intergrade with philadelphica. Representative specimens of each of the three forms of this species are illustrated, and material is separated appropriately in the lists of specimens examined. There is little or no appreciable difference in the reproductive systems. AFFINITIES.—The species of the Philadelphica subgroup are all closely related and by no means easy to separate. A complication is introduced by the certainty that additional forms will be discovered and defined, particularly in the Appalachian region. Of the named Ficures 43-46.—Structural details of Cambarincola chirocephala Ellis. 43, Lateral aspect of typical specimen, Benton Co., Arkansas; 44, dorsal aspect of jaws, specimen from Benton Co., Arkansas; 45, lateral aspect of jaws, specimen from Logan Co., Arkansas; 46, lateral aspect of reproductive systems, specimen from Logan Co., Arkansas. forms, philadelphica, chirocephala, and macrodonta are most alike, and future work may indeed show them to be only components of a wide- ranging polytypic species. Some of the Kentucky material listed under chirocephala shows a remarkable similarity in almost every respect to philadelphica, and I have been able to demonstrate gradual east-west clinal variation in both the jaws and spermiducal gland of chirocephala. That these two species intergrade seems almost cer- tain. Perhaps the establishment of an arbitrary ratio of jaw widths will help define the ranges of the two, as well as their intermediates. The relationships of philadelphica to macrodonta are also very close. The entire peristomium of the latter is a good distinction between well-preserved material of the two, and if present impressions are cor- rect, the smaller bursa (less than the spermiducal gland in diameter) of macrodonta should aid in recognition of the species. C. macrodonta apparently occurs in South Dakota, philadelphica in Wisconsin. 346 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 Obviously the northern part of the Mississippi Valley is an area of some importance in establishing whether or not these two species overlap or intergrade. Distrisution.—Northeastern United States, from New York west to Wisconsin, south through the Appalachian system as far as South Carolina and Tennessee. Material of the typical form of the species has been seen from the following collections: MARYLAND: FREDERICK COUNTY: Catoctin Creek between Brunswick and Point of Rocks on Md. Hy. 464, October 12, 1952, L. B. Holthuis (PCH 600). MINNESOTA: WRIGHT COUNTY: Clearwater River at Minn. Hy. 152, in the town of Clearwater, July 20, 1958, Holt (PCH 790). NortH CAROLINA: WATAUGA COUNTY: Campus pond at Boone, June 24, 1948, Mike Wright (PCH 128). New YORK: FRANKLIN COUNTY: Chateauguay River just west of Chateauguay on U.S. Hy. 11, May 21, 1951, D. W. Crocker and A. Gustafson (PCH 634). LEWIS county: Mohawk River at West Leyden, August 22, 1952, Crocker (PCH 552); outlet of Brantingham Lake, 3.5 miles north of Lyons Falls, May 21, 1951, Crocker and Gustafson, (PCH 639). tTIoGa county: Catatonk Creek at Candor, August 29, 1952, Crocker (PCH 569). tTompxins county: Owasco Inlet, 3 miles south of Groton, May 18, 1951, E. C. Raney (PCH 628). wWwyomine county: Cattaragus Creek at Arcada, August 29, 1952, Crocker (PCH 570). PENNSYLVANIA: PHILADELPHIA COUNTY: Wissahickon Creek in Philadelphia, September 2, 1957, C. W. Hart (PCH 695). SoutH CAROLINA: YORK couNTY: 4.0 miles south of Rock Hill on 8.C. Hy. 72, April 20, 1958, Holt (PCH 748). VIRGINIA: ALLEGHANY county: Dolly Ann Creek, 2 miles east of Covington on U.S. Hy. 60, June 18, 1948, G. Ailstock (PCH 115). amuerRst county: Pedlar River, 6.3 miles northwest of Forks of Buffalo on U.S. Hy. 60, November 11, 1947, Horton H. Hobbs (PCH 82). augausta county: 7.6 miles north of Steeles Tavern on U.S. Hy. 11, May 11, 1947, Hobbs (PCH 71). BEprorp county: Little Otter River, 3.1 miles east of Bedford on U.S. Hy. 460, Holt and Bobb (PCH 73). BRUNSWICK COUNTY: 2.9 miles north of Edgerton on Va. Hy. 140, May 31, 1949, Holt and Bobb (PCH 217). BucKINGHAM couNTY: 9.6 miles south of Sprouse’s Corners on U.S. Hy. 15, November 9, 1946, Hobbs and Hoffman (PCH 36). CAMPBELL county: Evington, July 13, 1947, Hobbs (PCH 106). cRraiIG couNTY: 1.5 miles south of Paint Bank on Va. Hy. 311, June 24, 1948, Hobbs (PCH 97). CHESTERFIELD COUNTY: Small stream below the lake in Camp Shawandasee, near Chesterfield Courthouse, May 14, 1949, Holt (PCH 211). FLUVANNA coUNTY: 1.5 miles south of Palmyra on U.S. Hy. 15, November 10, 1946, Hobbs and Hoff- man (PCH 114). Grayson county: Stream in pasture north of High Rock, northwest of Independence, July 17, 1949, Holt (PCH 260). HANOVER COUNTY: 3.1 miles north of Ashland on U.S. Hy. 1, June 3, 1949, Holt and Bobb (PCH 230). HIGHLAND COUNTY: Shaw’s Creek, 6.6 miles east of McDowell on U.S. Hy. 250, September 27, 1946, Hobbs (PCH 113). Louisa county: 25.5 miles east of Charlottesville on U.S. Hy. 250, July 2, 1948, Holt and Bobb (PCH 121). MONTGOMERY county: Trillium Dale at Blacksburg, November 17, 1957, Holt, Riggin, Hoffman (PCH 873); Bottom Creek, 0.5 mile northeast of Otey on Va. Hy. 637, May 22, 1958, Holt and Hoffman (PCH 874). NELSON couNTyY: North Fork of Rockfish River, 5.4 miles south of Afton on Va. Hy. 151, September 1, 1946, Hobbs (PCH 112). rocksBRipGE county: Tributary to Buffalo Creek, 5.3 miles south of Lexington on U.S. Hy. 11, May 11, 1947, Hobbs (PCH 75); 4.3 ANNELID GENUS CAMBARINCOLA—HOFFMAN 347 Ficures 47-56.—Structural details of three species of Cambarincola. 47-50, C. chiro- cephala: 47, Male reproductive system, Carroll Co., Arkansas; 48, jaws, caudolateral aspect, same specimen; 49, mzle reproductive system, Parke Co., Indiana; 50, jaws, lateral aspect, same specimen. 51-54, C. philadelphica, heads in lateral aspect: 51, Small northern form, Tompkins Co., New York; 52, typical form, Montgomery Co., Virginia; 53, amplistomate form, Jackson Co., Kentucky; 54, amplistomate form, Macon Co., North Carolina. 55, 56, C. macrodonta: paratypes, Boulder Co., Colo- rado, both specimens show the characteristic entire peristomium. miles north of Glasgow on U.S. Hy. 501, May 11, 1947, Hobbs (PCH 78). rocx- INGHAM couNTY: Swift Run, 3 miles east of Elkton, December 12, 1946, Holt (PCH 14). scorr county: Troublesome Creek, 8.2 miles west of Gate City on U.S. Hy. 58, June 16, 1950, Hobbs and Hart (PCH 366). West Virainia: East River, date and collector not stated (USNM 17705) [probably in Mercer County, southwest of Glen Lyn]. Specimens of the small New York phase (either a depauperate form of philadelphica or possibly a different species) have been examined from the following collections: New YorK: CHENANGO couNTy: Otego Creek, just east of Oneonta, August 27, 1952, Crocker (PCH 561). onrrpa county: Tributary to the N.Y. State Barge Canal just west of Utica, August 23, 1952, Crocker (PCH 536); Oswego River at Rome, August 22, 1952, Crocker (PCH 583). tTroga county: Catatonk Creek at Candor, September 6, 1959, L. R. McManus (PCH 928). TOMPKINS county: Owasco Inlet, 3 miles south of Groton, May 18, 1951, E. C. Raney (PCH 628). WARREN county: Stony Creek, in the town of Stony Creek, August 19, 1952, Crocker (PCH 546). Material of the form with enlarged peristomium, discussed in the foregoing text, has been examined from the following localities: KENTUCKY: JACKSON CouUNTY: 1.8 miles south of Bond, July 29, 1958, Holt (PCH 831). woure county: 1.7 miles west of Compton on Ky. Hy. 15, July 31, 348 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 1958, Holt, (PCH 846); large stream 5.1 miles east of Compton on Ky. Hy. 15, July 31, 1958, Holt (PCH 847). NortH CAROLINA: MACON couNTY: Spillway of Lake Ravenel, Highlands, July 8-15, 1958, Hoffman (PCH 875); Buck Creek drainage 3.5 miles northwest of Highlands, August 3, 1958, Hoffman (PCH 884). TENNESSEE: CUMBERLAND county: White Creek, 0.25 mile northwest of Alloway, May 3, 1959, Holt and Ford (PCH 903). Cambarincola chirocephala Ellis Fiaures 43-50, 57 Cambarincola chirocephala Ellis, 1919, Proc. U.S. Nat. Mus., vol. 55, p. 263, figs. 18, 19—Goodnight, 1940, Illinois Biol. Monogr., vol. 17, no. 3, p. 37 (in part, [ll., Ind., Ky., Mo., ?Ala. records); 1940, Rep. Reelfoot Lake Biol. Station, vol. 4, p. 171 (?Tenn. record).—Holt and Hoffman, 1959, Journ. Tennessee Acad. Sci. vol. 34, p. 103. TyPE SPECIMEN.—Holotype, USNM 177138, from Orconectes virilis collected at Rolla, Phelps County, Missouri, by J. Barley. This specimen is mounted flattened dorsoventrally, showing the jaws clearly; the reproductive systems are obscured. Diaanosis.—A small member of the Philadelphica group, recog- nized by the distinctly disparate anisomorphic jaws; by the reduced size of the spermiducal gland and prostate (both often shorter than the bursa); and by the elongate, slender, gradually enlarged sper- matheca. Description.—Body small, ranging from 1.5 to 2.5 mm. in length, the greatest diameter at segment vir; prosomites of all segments larger than metasomites. Diameter of segment 1 less than that of head; segments rx and x reduced, the caudal sucker abruptly enlarged, equaling or surpassing diameter of head. Head of normal size and form for the genus, larger than segment 1, as long as the first three body segments combined. Peristomium conspicuously set off by a basal constriction, flared, its margin pro- vided with four small lobes on the dorsal side and two broad, lower lobes on the ventral. Jaws disparate in both dorsal and lateral aspects, anisomorphic; broadly triangular. Dorsal jaw largest, 1.5 to 2 times as broad as the ventral, the median tooth very conspicuous, lateral cusps small or obscure. Ventral jaw subtrapezoid, with two small paramedian apical teeth and a median sinus, and a smaller sublateral cusp on each side halfway to the base. Male reproductive system moderate in size, extending dorsal through about three-fourths of segment v1. Bursa elongate, pyriform, 2 or 3 times as long as its greatest diameter, the latter more than half the length of the ventral part of the segment. Atrial portion of bursa somewhat longer than the penial sheath. Ejaculatory duct long and slender, two-thirds to three-fourths as long as the ANNELID GENUS CAMBARINCOLA—HOFFMAN 349 Ficure 57.—Distribution of the three species of the Philadelphica subgroup. ©, C. phila- delphica; ©, “‘amplistomate” variety of philadelphica; (, C. macrodonta; @, C. chiros cephala. bursa, subequal in length to the prostate gland. Spermiducal gland relatively small, subequal to or shorter than length of bursa, its ental end recurved caudomesad prior to entry of the anterior deferent duct. Prostate relatively large, nearly as long and thick as sperm- iducal gland, the terminal bulb large and conspicuous. Spermatheca long and slender, the ectal half curving laterad around the gut, ental half more or less abruptly enlarged to 1.5 to 2 times the diameter of ectal portion, the apex bluntly acuminate, not set off as an evident ental process. Total length of spermatheca somewhat greater than diameter of segment Vil. VaRIATION.—Individual variation in this species is very slight, involving chiefly differences in size of the body. The proportions of the reproductive tracts remain constant for a given locality. There is, however, some evidence that the sexual organs are subject to the influence of geographic factors affecting variation. There appears to be an east to west decrease in the spermiducal gland length as indicated by the illustrations (figs. 47, 49). In Indiana 350 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 and Kentucky material, this organ is fairly robust and somewhat longer than the bursa, while in specimens from Arkansas it is reduced in overall size and is shorter than the bursal length. The prostate remains unaffected over this range, so that in the southwest it becomes porportionately longer and broader with respect to the spermiducal gland. Presumably this variation is of at least potential systematic interest, but whether it is evenly clinal in nature, or broken at some intermediate area (such as the Mississippi River) cannot be ascer- tained on the basis of available material. Specimens from Illinois are presently not at hand for study, and for the time being the matter must rest at this point. There appears to be a similar east to west cline in jaw structure, with the disparity in size becoming most distinct westward into Kansas and Oklahoma. Kentucky material tentatively referred to chirocephala can scarcely be distinguished from philadelphica on the basis of size of the jaws. Distripution.—Cambarincola chirocephala appears to be basically autochthonous to the Ozark highlands, now reaching its greatest abundance in northern Arkansas and Missouri. It extends, however, eastward into Indiana and Kentucky where its range meets that of philadelphica, with which it perhaps intergrades. Most of the Indiana localities are in the Wabash River system; one, interestingly enough, is in the Maumee River system which now drains northeast into Lake Erie. ‘The presence of chirocephala in this drainage reflects either fairly recent stream piracy in the region or transgression of the low divides by crayfish—either possibility seeming equally likely. It will be a matter of interest to establish additional records for the species in northwestern Ohio and eastern Indiana. Goodnight (1940, p. 38) has reported this species from a number of midwestern localities as well as from some very unlikely stations in New York and Virginia, and such extralimital records must be presumed to have been based upon misidentifications. MATERIAL EXAMINED: 22 slides, from the following localities: ARKANSAS: BENTON COUNTY: Sugar Creek at crossing of U.S. Hy. 62, July 6, 1958, Holt (PCH 768). cARRoLL county: South Fork of Dry Creek, 2.8 miles east of Green Forest on U.S. Hy. 62, July 6, 1958, Holt (PCH 766). crRiTTEN- DEN coUNTY: 14.6 miles south of Marked Tree on U.S. Hy. 68, July 5, 1958, Holt (PCH 758). FruLTon county: Salem, July 29, 1941, Horton H. Hobbs (PCH 310). LoGan county: West fork of Mill Creek, 4.4 miles west of Dela- ware on Ark. Hy. 22, July 30, 1941, Hobbs (PCH 189). NEWTON coUNTY: Buffalo River, 14 miles south of Harrison on Ark. Hy. 7, July 29, 1941, Hobbs (PCH 175). sHAaRP couNTy: 3.2 miles southeast of Hardy, July 29, 1941, Hobbs (PCH 176); about 9 miles west of Hardy, July 6, 1958, Holt (PCH 760). INDIANA: ALLEN county: St. Mary’s River at Fort Wayne, Max M. Ellis (USNM 17706). MoNRoE county: Bloomington, May 1915, Will Scott (USNM ANNELID GENUS CAMBARINCOLA—HOFFMAN 351 17709). PARKE couNTY: 5.6 miles west of Bellmore on U.S. Hy. 36, July 26, 1958, Holt (PCH 807). KANSAS: HARPER COUNTY: 10.7 miles west of the county line on Kans. Hy. 14, July 8, 1958, Holt (PCH 776). KENTUCKY: CLARK COUNTY: 9.7 miles east of Winchester on Ky. Hy. 15, July 30, 1958, Holt (PCH 839). sackson county: 0.1 mile south of the Owsley County line on Ky. Hy. 30, July 29, 1958, Holt (PCH 834). MapIsoNn county: Otter Creek, 9.3 miles north of Richmond on U.S. Hy. 227, July 30, 1958, Holt (PCH 841). ows.tEy county: Traveler’s Rest, July 29, 1958, Holt (PCH 835). POWELL couNTy: 1.4 miles east of Slade on Ky. Hys. 11 and 13; also Natural Bridge State Park, both July 29, 1958, Holt (PCH 836, 837). MIssouRI: PHELPS COUNTY: Rolla, J. Barley (USNM 17713, the holotype). WASHINGTON couNTy: Irondale, August 28, 1931, Robert Rice (PCH 168). OKLAHOMA: COMANCHE COUNTY: Blue Beaver Creek in Fort Sill, June 6, 1959, J. W. Berry (PCH 905). TENNESSEE: HUMPHREYS CouNTY: Hurricane Creek, 10.2 miles east of Waverly on U.S. Hy. 70, July 5, 1958, Holt (PCH 756). Remarxs.—Goodnight (1940, p. 37) has emphasized the pronounced elevation of prosomites over metasomites as diagnostic of the species, and most specimens are so formed, but contracted individuals of many branchiobdellids likewise assume a distinctly annulated ap- pearance. The larger size of the dorsal jaw is a good character for recognition of chirocephala over most of its range, but it must be recalled that in Kentucky this character loses must of its significance. The affinities of chirocephala are discussed in connection with C. philadelphica; in brief, the likelihood of a genetic continuum between the two seems good. However, the actual details of this relationship are not clear at this time. Whether chirocephala represents the fairly recent modification of a westwardly migrating philadelphica-stock or whether it is an allopatrically differentiating species now radiating from its place of origin and intergrading with the original parent population remains to be established. The name chirocephala (Gk. chiros, hand, and cephalos, head) pre- sumably refers to the dorsal lobation of the peristomium, but it is not particularly appropriate since none of the material examined is more distinctly lobed than most specimens of even C. philadelphica, and in no way approximates the conspicuous tentaculation of fallax and some other species. Ellis himself misidentified the Indiana material cited as philadelphica, the records being originally published in the same paper as the description of chirocephala! The Indiana worms were merely mounted in such a way that the jaws could not be seen in dorsal aspect. Cambarincola macrodonta Ellis FIGURES 37, 38, 55, 56, 57 Cambarincola macrodonta Ellis, 1912, Proc. U.S. Nat. Mus., vol. 42, p. 481, figs. 1-3; 1919, Proc. U.S. Nat. Mus., vol. 55, p. 257 (Colorado records only?).— 302 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 Hall, 1914, Proc. U.S. Nat. Mus., vol. 48, p. 190.—Goodnight, 1940, Illinois Biol. Monogr., vol. 17, no. 3, p. 31 (description only); 1940, Rep. Reelfoot Lake Biol. Station, vol. 4, p. 171 (?South Dakota record).—Holt and Hoff- man, 1959, Journ. Tennessee Acad. Sci., vol. 34, p. 97, figs. 1-6 (redescription of species). TypE SPECIMENS.—Holotype and two paratypes, USNM 53794, from Cambarus diogenes collected at Boulder, Boulder County, Colorado, by Max M. Ellis. Draanosis.—A moderately small member of the genus, differing from other members of the Philadelphica group by the combination of the slender, elongate outline of both body and head, the large anisomorphic subequal jaws; entire peristomial margin; relatively large bursa in comparison with the long, slender spermiducal gland, and shortened prostate gland. The ectal spermathecal duct is much longer in relation to size of the ental bulb than in the other related species. Derscription.—A distinctly slender and graceful worm, the body only slightly thicker at maximum diameter than at the narrowest part, tapering more distinctly caudally down to the smal] caudal sucker, its diameter less than that of either head or segment 1. Segments less than twice as long as broad, the length almost equally divided into prosomite and metasomite, the former not at all larger in diameter than the latter. Head rather long and narrow, about equal to first three segments combined, its greatest diameter a little greater than that of segment I. Peristomium set off by a basal constriction, not evidently flared, the margin divided into dorsal and ventral halves but neither half with lobes or tentacles. Jaws large and massive, subtriangular in dorsal aspect, aniso- morphic, heterodont, the dental formula 5-4; dorsal jaw slightly larger than ventral in lateral aspect, its median tooth subequal to the paramedian teeth of the ventral jaw. Male reproductive system of moderate size, occupying half of one side of the coelom of segment vi or less, the bursa elongate pyriform, its greatest diameter near the ental third instead of near the midlength as in related forms; ejaculatory duct short, about equal to bursal diameter. Spermiducal gland slender and elongate, oriented almost horizontally in striking contrast to the oblique dorsoventral position taken by the gland in most other related species, increasing slightly in diameter entally, with the ental fourth of the length bent a little ventrad. Deferent lobes small or absent. Prostate about two- thirds as long as spermiducal gland and from one-fourth to one-half its diameter, extending along the dorsal side entally about as far as the entry of the posterior deferent duct. ANNELID GENUS CAMBARINCOLA—HOFFMAN 353 Spermatheca composed of three parts: An elongate, slender, muscular walled ectal duct, which extends about halfway up the side of segment v; a subglobose or ovoid ental bulb; and a small terminal ental process composed of glandular cells. VARIATION.—The material which was studied from three localities in Colorado is relatively homogeneous as regards structure of the jaws and reproductive systems, variations in these details being largely a reflection of difference in size of the worms. It is quite true that branchiobdellids are very variable in size after maturity, suggest- ing continuous growth through life, and actual measurements carry less significance than ratios of the measurements. Various combinations of measurements~’as functions of some standard have been plotted graphically, and most of these show that the material of macrodonta examined maintains a very constant set of proportions despite changes in overall size. One suggestive detail involves relative head size. Three specimens from Fort Collins, Colorado, yield a head diameter to body diameter ratio of .60, .61, and .62. In eleven other specimens from Boulder and Black Wolf Creek, Colorado, the same ratio ranges from .71 to .99, with an average of .83. These values, plotted along the horizontal axis of a chart with body- length intervals on the vertical axis, separate out into two discrete groups. Unfortunately, it cannot be ascertained whether we are dealing here with true geographic variation or with the effects of preservation. Within a fairly wide latitude, the degree of contraction or distention of a branchiobdellid after preservation is influenced by the strength of the alcohol used. The foregoing example is introduced to remark the likelihood that with uniformly preserved worms in good series, a future student of the group will be able to cope with the problems inherent in the study of soft-bodied, muscular animals. AFFINITIES.—Cambarincola macrodonta is without doubt closely related to C. philadelphica, and if the two occurred as sympatric or adjacent allopatric forms, it would be difficult to separate them with confidence. In well-preserved specimens, the peristomial character is most useful, but some individuals of philadelphica often do not show the peristomial lobes, and it is easy to understand how various other species have been misidentified by previous workers as macrodonta. On the basis of my own limited knowledge of the species, I would judge it to be an isolated, conservative remnant of the old late Tertiary pre-philadelphica stock which has become isolated in the foothills of the Rockies by recent climatic events which have produced the now semiarid nature of the Great Plains. DisrrisutTion.—This species has been recorded from 12 States raiging from New Mexico to Virginia, from South Dakota to Louisiana. The specimens which I have been able to re-examine, those identified 354 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 114 by Ellis in the U.S. National Museum collection, fall into three categories: Very small, obviously immature worms; adult but mis- identified specimens; and, finally, a few slides of worms conspecific with the type specimen. The juveniles are at present unidentifiable with certainty. Misidentified adults include C. mesochorea and some other species which cannot be confidently identified but which are not macrodonta on jaw shape. Undoubted specimens of the species are listed as follows: COLORADO: BOULDER couNTY: Boulder, September 1915, Max M. Ellis (USNM 17667). LAaRimEeR county: Fort Collins, L. C. Bragg (USNM 17662). yuMA county: Black Wolf Creek, near Beecher’s Island, October 1915, B. Jaffa (USNM 17664). In addition to these Colorado records, the specimen cited by Ellis from Las Vegas, New Mexico (USNM 17661), appears to be a macrodonta, but is so heavily stained that the sex organs cannot be seen. There is nothing from a geographic point of view to preclude the specimen being macrodonta. Specimens from Muldon and Agricultural College, Mississippi, are very much like Colorado material in every respect, yet I hesitate to admit them to the list of macrodonta localities, at least until more material from Mississippi or from intermediate areas comes to hand. Cambarincola meyeri Goodnight Figures 67, 68 Cambarincola meyert Goodnight, 1942, Trans. American Microse. Soc., vol. 61, no. 3, p. 272, figs. 1-3—Holt and Hoffman, 1959, Journ. Tennessee Acad. Sci., vol. 34, p. 108. TYPE sPECiAMEN.—Holotype, USNM 20597, from Cambarus bar- tonii collected in Raven’s Creek, near Lexington, Fayette County, Kentucky, by Marvin C. Meyer (date unknown). Dracnosis.—A small to moderate-sized member of the Philadel- phica group characterized particularly by the (?) entire peristomium, stout reniform spermiducal gland with a short slender prostate, and peculiar form of the spermatheca. Description.—Length of holotype about 3.5 mm. Body of nor- mal proportions, segmental diameter about three times the length at midbody; segments tv—vr of essentially the same size; prosomites elevated above level of metasomites; caudal sucker quite small, its diameter less than that of head or of segment 1. Head of moderate size, as long as first three body segments and slightly wider than segment 1. Peristomium set off by a basal con- striction, slightly flared but the margin entire except for being divided into a dorsal and ventral half; no evidence of lobes visible, ANNELID GENUS CAMBARINCOLA—HOFFMAN 355 Jaws fairly small for size of head, heteromorphic, the dorsal jaw with a large median tooth and two very small cusps at its base on each side; ventral jaw with two large paramedian teeth and a single small cusp at the outer base of each. Jaws virtually identical in size, the dorsal very slightly wider at the base. Male reproductive system characterized by the small globose bursa (its diameter less than that of the spermiducal gland), by the very slender, tubular prostate which is less than half as long as the sper- miducal gland, and by the stout, acutely reniform shape of the latter, its ental half curved strongly caudad and accentuating the region of the posterior deferent duct. Spermatheca composed of three major portions: 1, An enlarged, muscular ectal region about half as large as the bursa; 2, a strongly constricted cervical region of the ectal duct; and 3, a greatly expanded, thin-walled, subglobose ental bulb. VaRIATION.—The species is known only from two specimens. Goodnight’s original description must have been based upon the paratype which he retained, for the holotype is somewhat larger than the published dimensions, its dorsal jaw 0.11 mm. in width instead of 0.07 as stated by Goodnight for his specimen. Having studied but the single specimen, I cannot say anything further on the subject of variation. AFFINITIES.—C. meyert is undoubtedly a member of the Phila- delphica subgroup, but differs from the others at least in the slen- der, short prostate and strongly curved, heavy spermiducal gland. It is possible that Goodnight’s observation on the peristomium will be confirmed, to constitute another diagnostic feature. Had no specimens been available for study, my inclination would have been to dismiss the name as probably a junior synonym of philadelphica or one of its localized races. Goodnight states that ‘“Cambarincola meyer is closely related to Cambarincola vitrea Ellis, but differs in the structure of the upper jaw.” I agree to a relationship between the two, but only as mem- bers of the same genus; actually they seem to be very dissimilar and fall into different groups of the Philadelphica section. Distrirpution.—Known definitely only from the type locality. Remarks.—The status of this species is by no means as well- established as might be desired. I have at hand a large series of well-preserved specimens from the vicinity of Livingston, Overton County, Tennessee (Holt, leg.), which agree in virtually every detail with the holotype of meyeri, not only in size and shape of the body, but also in small details of the reproductive systems. The concord- ance is such that conspecificity with the type specimen is almost assured. Yet there appears to be a discrepancy in the jaw struc- 356 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 114 ture—isomorphic and bipentadont in the Tennessee worms, aniso- morphic and pentatetradont in meyeri. Two likelihoods can be considered: 1, The original types of meyert may have been compos- ite. Goodnight’s account of the jaws was based on the paratype. Unfortunately, the dentition cannot be made out with certainty in the holotype, owing to its orientation on the slide; 2, the types of meyeri may be aberrant specimens with respect to jaw structure, if we assume that both are anisomorphic. The question is one which can be settled only by the study of a series of fresh specimens from the type locality. On the basis of present knowledge of the group, it seems utterly unlikely that virtual identity in form of the sex organs would be contraverted by a basic difference in jaw structure. For the present, however, I refrain from identifying the Tennessee worms as meyer, remarking only their great similarity to the type of that species, and commending the matter to someone having the opportunity to secure topotypes of the species. Cambarincola fallax, new species Ficures 58-60, 62, 63 Cambarincola philadelphica (in part) Ellis, 1919, Proc. U.S. Nat. Mus., vol. 55, p. 262.—Goodnight, 1940, Illinois Biol. Monogr., vol. 17, no. 3, p. 38.—Holt, 1949, Journ. Morphology, vol. 84, p. 535 et seq. ?Cambarincola okadai Yamaguchi, 1933, Proc. Imp. Acad., vol. 9, no. 4, p. 191; 1934, Journ. Fac. Sci. Hokkaido Imp. Univ., vol. 3, no. 3, p. 190. TYPE SPECIMENS.—Holotype and four paratypes, USNM 29945, from Cambarus longulus subsp. collected in Maiden Spring Creek, about 1 mile east of Wardell, Tazewell County, Virginia, on June 19, 1959, by R. L. Hoffman. Additional paratypes from the same collection, PCH 904. Driaenosis.—A moderate to large species of the Philadelphica group characterized by the combination of homognathous, pentadont jaws and conspicuous elongate peristomial tentacles. Dezscription.—A moderate to fairly large species, up to about 4.0 mm. in length. Body form rather slender, without distinct enlargement in diameter in going caudad to the middle of the length. Prosomites about twice as long as metasomites and very distinctly larger in diameter, imparting a pronounced annulate body profile. Segments 11 to vir usually of about equal diameter. Head moderate in size, about as long as first three body segments combined, its diameter about equal to that of segment 11, equal to or slightly larger than diameter of caudal sucker. Peristomium large, set off by a deep basal constriction, almost half the total head length; the dorsal half often a little flared, with four distinct blunt ANNELID GENUS CAMBARINCOLA—HOFFMAN 307 Ficure 58-61.—Two species of the Fallax subgroup, external appearance. 58, Cambar- incola fallax, new species, paratype from Tazewell Co., Virginia, with peristomium extended; 59, specimen from Seneca Lake, New York; 60, specimen from Pike Co., Georgia. 61, C. holostoma, new species, paratype from Highland Co., Virginia, showing the characteristic entire peristomium. elongate tentacles of varying length. Ventral half of peristomium shorter, set off from dorsal by a deep lateral sinus each side, subdivided into two broad lobes. Jaws about 10 percent of the head length, isomorphic, equal in size, heterodont; the median teeth distinctly larger than the lateral cusps which are nonetheless very distinct even in lateral aspect, dental formula 5-5, 5-3, or 3-3, the smaller figures occurring chiefly in old or large specimens. Male reproductive system typical for the group, the more ectal organs of moderate size and occupying about two-thirds or less of the coelom on one side of segment vi. Bursa elongate pyriform, its greatest diameter at midlength; penial sheath abruptly merging into the much narrower ejaculatory duct; latter half as long as bursa. Spermiducal gland oriented almost dorsoventrally, extending down to or beyond level of penial sheath, generally subreniform in outline with an occasional enlargement homologous to the posterior deferent lobe of other species; diameter of gland about equal to that of bursa, but somewhat smaller in occasional specimens. Prostate gland long, slender, its diameter half that of spermiducal gland near their junc- ture; not extending entally as far as apex of the latter. 3508 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 Spermatheca rather small, generally similar to that of OC. phila- delphica, composed of a slender ectal duct and an ovoid or fusiform ental bulb located about halfway up one side of segment v, no glan- dular ental process detected. VariaTion.—Individual variation in body form and peristomial shape is shown in figures 58-60. The two most similar worms are from opposite extremes of the range. The male reproductive system varies somewhat more than in most other members of this genus, particularly with respect to size of the spermiducal gland, but nothing has been noted to indicate any sort of geographic dispersion. Figure 62 shows the typical proportions of the larger organs. The spermatheca is likewise somewhat vari- able, particularly the appearance of the ental bulb. This, however, is pretty clearly a reflection of the degree of distention by its contents. Distripution.—Known from numerous localities throughout the Appalachian uplift and adjacent Piedmont from western New York south as far as central western Georgia. In the southern part of the range fallax appears to be the most abundant member of the genus, further north the records are more scattered and C. philadelphica becomes dominant. Material has been examined from the following localities: GEORGIA: FANNIN county: 5.7 miles south of Morganton on Ga. Hy. 60, November 6, 1958, K. Simonds (PCH 909). murray county: Holly Creek, 10.6 miles north of the Gordon County line on U.S. Hy. 411, April 16, 1958, Holt (PCH 737). pike county: Large stream on outskirts of Zebulon, April 17, 1958, Holt (PCH 740). wurre county: Small ravine, 1 mile south of Helen, July 11, 1958, Hoffman (PCH 877). NortH CAROLINA: CHEROKEE couNTY: Beaver Creek, 0.5 mile northwest of Andrews, June 9, 1959, Simonds (PCH 912). cnuay county: 3 miles south of Tuni Gap on the Hayesville-Andrews road, June 5, 1959, Simonds (PCH 918). TENNESSEE: MONROE couNTY: Small woodland stream, 7 miles north of Madisonville on U.S. Hy. 411, April 16, 1958, Holt (PCH 738). PoLK county: 2.3 miles south of Ocoee on U.S. Hy. 411, June 9, 1959, Simonds (PCH 907). VIRGINIA: ALBEMARLE County: Tributary to Rivanna River near Stony Point, April 25, 1947, Holt and Hobbs (PCH 45, 46). BucKINGHAM couNTY: 9.6 miles south of Sprouses Corners on Va. Hy. 15, November 9, 1946, Hobbs and Hoffman (PCH 36). cHARLOTTE couNTY: 8.8 miles south of Keysville on Va. Hy. 15, November 9, 1946, Hobbs and Hoffman (PCH 38). ames county: Sinking Creek at Va. Hy. 700, near Newport, July 3, 1950, Holt, Tuten, and Kizer (PCH 407). PATRICK county: Shooting Creek, 1.6 miles south of the Franklin County line on Va. Hy. 40, April 13, 1953, Hobbs (PCH 723). TrazewELL county: Maiden Spring Creek, about 1 mile east of Wardell, June 19, 1959, Hoffman (PCH 904, type series). New York: TomPKINS county: Owasco Inlet, 3 miles south of Groton, May 18, 1951, E. C. Raney (PCH 628). County not located: Reeder’s Creek, 1949, L. C. Goldstein (PCH 245). ANNELID GENUS CAMBARINCOLA—HOFFMAN 359 67 2 “ LA ie ea ee Ficures 62-68.—Structural details of three species of Cambarincola. 62, C. fallax, new species, reproductive systems, specimen from Giles Co., Virginia; 63, dorsal aspect of jaws, same specimen, each jaw tilted slightly in opposite directions, dental formula actually 5-5. 64, C. holostoma, new species, reproductive systems, paratype from Highland Co., Virginia; 65, the same, jaws in dorsolateral aspect, same locality; 66, the same, jaws in lateral aspect, same locality. 67, C. meyeri Goodnight, male re- productive system, from holotype, Fayette Co., Kentucky; 68, the same, spermatheca from holotype, the specimen considerably flattened in mounting. In addition to the foregoing records, there are doubtless many published localities for C. philadelphica which really apply to this species. Cambarincola holostoma, new species Ficures 61, 64-66 Type sPeEcIMENS.—Holotype and four paratypes, USNM 29946, from Cambarus bartonii and C. longulus collected in Crab Run, 4 miles west of McDowell, Highland County, Virginia, on U.S. Hy. 250, by L. B. Holthuis, October 25, 1952. Topoparatypes, PCH 599. Dracnosis.—A member of the Falilax subgroup characterized by the long, slender body, the prosomites of which are not distinctly raised; by the completely entire, flared peristomium; and by the slender, elongate, fusiform spermatheca. 663871—63——_7 360 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 Description.—A small worm reaching a maximum length of,about 2.0 mm. in preserved specimens. Body form slender, the diameter increasing gradually to segment vi which is subequal in bulk to seg- ments vir and viir; caudal sucker somewhat larger than preceding segment and about as broad as the peristomium. Prosomites up to three times as long as the metasomites, but not of greater diameter. Head about as long as the first three body segments combined, and equal in diameter to segment 111, the peristomium set off by a very strong basal constriction and distinctly flared, its margin entire, without any trace of division into dorsal and ventral halves or into smaller lobes. Head otherwise not visibly segmented externally. Jaws similar, dental formula 3-3 with the median tooth long and acute, the general appearance very similar to the jaws of C. fallax but the formula perhaps more often 3-3 than in that species. Male reproductive system basically similar to that of other species of the Philadelphica group. Bursa rather long, equaling the length of the spermiducal gland, the two subequal in diameter. Ejaculatory duct modest in size, its length less than the diameter of the spermiducal gland. Latter of normal proportions, without evident Jobation at entry of deferent ducts. Prostate long, slender, about half the diameter of spermiducal gland, which it joins slightly entally of the entry of the ejaculatory duct. Spermatheca slender and elongate, curving laterad and dorsad around the gut, and becoming slightly wider but maintaining about the same diameter almost to its end at a point near the middorsal area of the segment, a distinct ental bulb not being well-developed. VarIATION.—In the small amount of material examined from three localities, some of it not well-preserved, there appeared to be little or no variation in the diagnostic characters of the species. AFFINITIES.—The relationships of this form with C. fallax and O. philadelphica, with both of which it is sympatric, are unquestion- able. It differs from both, however, in characters of the peristomium and spermatheca. Closer relationship with fallaz is postulated on the basis of jaw form, here considered to be a more fundamental character than peristomial lobation. Disrrisution.—Aside from the type locality, this species is known from two collections from western Virginia, in the James and upper Potomac River drainages. _ VIRGINIA: CRAIG COUNTY: 1.5 miles south of Paint Bank on Va. Hy. 311, June 24, 1948, Horton H. Hobbs (PCH 97). RockinaHam county: Swift Run, 3 miles east of Elkton on U.S. Hy. 33, December 12, 1946, Holt (PCH 14). Remarks.—Further knowledge of the distribution of this localized form may be of interest in providing an insight into the factors influenc- ing its speciation. ANNELID GENUS CAMBARINCOLA—HOFFMAN 361 Ficures 69-73.—Structural details of two species of Cambarincola. 69, C. heterognatha, new species, paratype in lateral aspect; 70, the same, lateral aspect of jaws; 71, the same, reproductive systems. 72, C. demissa, new species, body profile of paratype, Wise Co., Virginia. Part of body wall shown cut away to indicate the size of the reduced reproductive organs in situ; 73, the same, reproductive systems in lateral aspect, same specimen. The collection localities, all of which I have seen, are rather small, swift, mountain brooks, somewhat different in character from the larger and perhaps more placid streams in which fallax occurs most abundantly. Conceivably the peristomium of holostoma represents the development of (or retention of) a more efficient holdfast device than the dissected and lobed mouth of fallax, a matter which would certainly enhance the origin and maintainence of specific differences by ecological factors. The problem is recommended to someone having the opportunity to study branchiobdellids and their distribution in the upper James River system of western Virginia. HETEROGNATHA GROUP A separate group must be erected to represent on a coordinate standing the very unusual and highly specialized new species described 362 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 below. In many of its structural features, such as the disparate jaw size and shape, small body, reduced size of the male reproductive system, and particularly the shortened prostate gland, this species shows a combination of evolutionary specialities which occur only singly in various other forms of the genus. Cambarincola heterognatha, new species FicureEs 69-71, 74 TypE sPECIMENS.—Holotype and paratype, USNM 29947, from Cambarus sp. collected in a tributary to Big Wilson Creek, 4 miles south of Mouth of Wilson on Va. Hy. 16, Grayson County, Virginia, by Horton H. Hobbs, Jr., and C. W. Hart, June 14, 1950. Additional paratypes indicated in the list of specimens examined. Diacnosis.—Immediately recognizable by the remarkably dis- similar jaws alone. In addition, the male reproductive system is re- duced in size and the prostate gland is less than half the length of the spermiducal gland. Description.—A moderate-sized species, body length up to about 3.0 mm., somewhat fusiform in body outline, the greatest diameter at segments vi and vit, the least at x. Head of normal size and proportions, about as broad as long, its diameter less than that of segment 1; peristomium only a little flared, without marginal lobes; head indistinctly subdivided into two halves by a slightly median constriction. Jaws very dissimilar in size, the upper triangular with a large median projection and about eight times the bulk of the lower (!) which is about equally quadrate in shape with two small paramedian cusps. Body outline fusiform, segments vi and vit largest, each somite divided by a distinct complete constriction, the prosomites the larger of the two subsegments and not, or only slightly, elevated over the level of the metasomites. Male reproductive system rather small, extending only halfway up the side of segment vi. Bursa fusiform, widest near the middle, about 1.5 times as long as broad, somewhat constricted at entry of ejaculatory duct. Latter of moderate length and rather slender, less than a third the bursa diameter. Spermiducal gland subreniform, slightly broader entally, its diameter equal to or slightly greater than that of bursa, the length a little greater. Prostate small and slender, less than half as long as spermiducal gland, generally about a third as long, the terminal bulb distinct. Spermatheca elongate and slender, abruptly constricted at the mid- length, the ental half with a rounded tip or a slight ental process, the organ extending dorsad nearly to the dorsal side of the coelomic cavity. ANNELID GENUS CAMBARINCOLA—HOFFMAN 363 Figure 74.—Distribution of Cambarincola heterognatha, new species, an abundant Appalach- ian endemic species. Each spot represents collections for one county. VarraTion.—There is little or no appreciable variation in this species as regards size of body, shape of jaws, and general details of the sex organs. There is some individual divergence in the relative length of the prostate gland, which may vary anywhere from a third to half the length of the spermiducal gland, irrespective of locality. The single collection from Kentucky is interesting in that the prostate of all specimens studied is basally much thicker than noted elsewhere in the range of the species, attaining a diameter at its base at least half that of the spermiducal gland. The Kentucky locality is con- siderably removed from the main distribution of the species, and this minor difference may reflect some significant microevolutionary de- velopment due to isolation. Aside from this one departure, hetero- gnatha must be considered a very homogeneous species despite its considerable geographic range. Disrripution.—The southern Appalachians, from northwestern Virginia and adjacent West Virginia, south and west to central eastern parts of Tennessee and Kentucky. Specimens have been examined from the following localities: KENTUCKY: ADAIR couNTy: 8.9 miles east of Columbia on Ky. Hy. 80, July 28, 1958, Holt (PCH 827). TENNESSEE: CLAIBORNE COUNTY: 3 miles southwest of New Tazewell on Tenn. Hy. 33, April 16, 1951, Horton H. Hobbs and W. R. West (PCH 540). cumBEr- LAND couNtTy: Daddy’s Creek, between Crossville and Pikeville, August 1950, Holt (PCH 419). wasHINGTON couNTy: Hartsell Cove, Buffalo Mountain, 364 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 August 1953, Holt (PCH 495); Sinking Creek in Horse Cove, January 10, 1954, Holt (PCH 580). NortH CAROLINA: WATAUGA COUNTY: 2 miles south of Vilas on N.C. Hy. 194, June 14, 1950, Hobbs and Hart (PCH 350). VIRGINIA: AUGUSTA CoUNTY: South River, 2.4 miles south of Waynesboro, September 3, 1948, Hobbs (PCH 271). crara county: 1.4 miles west of New- castle on Va. Hy. 311, June 23, 1948, Hobbs (PCH 99). FRANKLIN COUNTY: Smart View Picnic Area, Blue Ridge Parkway, September 14, 1958, Holt and Hoffman (PCH 891). GmxEs country: Cascades of Little Stony Creek, June 25, 1952, Holt (PCH 492, paratypes). GRrAYsoN county: Tributary to Big Wilson Creek, 4 miles south of Mouth of Wilson, June 14, 1950, Hobbs and Hart (PCH 339, types). LEE county: Hardy Creek, 11.1 miles west of Jonesville on U-S. Hy. 58, June 16, 1950, Hobbs and Hart (PCH 371). RocKINGHAM couUNTY: Swift Run, 3 miles east of Elkton, December 1946, Holt (PCH 4). smyvH county: White Top Mountain, September 5, 1951, John T. Wood (PCH 531). TAZEWELL CouNTY: Bluestone River, 11.2 miles northeast of Tazewell on U.S. Hy. 460, June 18, 1950, Hobbs and Hart (PCH 393, paratypes); Burkes Garden, June 30, 1947, R. L. Hoffman and H. I. Kleinpeter (PCH 532); Maiden Spring Creek, 1 mile east of Wardell, June 19, 1959, Hoffman (PCH 904). wasHINGTON county: 9.5 miles south of Abingdon on U.S. Hy. 11, April 14, 1951, Hobbs and West (PCH 432); 4.8 miles south of Abingdon, April 14, 1951, Hobbs and West (PCH 435); North Bristol, January 1, 1954, W. A. Whittaker (PCH 581). wyTHE county: Reed Creek, 3.3 miles southwest of Wytheville on U.S. Hy. 11, April 14, 1951, Hobbs and West (PCH 439). West VIRGINIA: GREENBRIER COUNTY: Dry Creek, east side of Kates Mountain at White Sulphur Springs, July 3, 1947, Hobbs (PCH 294). PENDLETON COUNTY: 5.8 miles east of Franklin on U.S. Hy. 33, July 30, 1949, Hobbs and Word (PCH 275). WYOMING couNTy: Barker Creek, 5.3 miles south of Tralee, July 12, 1947, Hobbs and Wilson (PCH 95). From the standpoint of major drainage systems, the vast majority of the preceding records lie within the basins of the upper Tennessee and Kanawha rivers, and southwest Virginia clearly seems to be the center of abundance for this species. Peripherally, the records are distinctly more spotty, although in equally well-collected areas. Towards the northeast, the species occurs in the James and Potomac drainage systems. The single known locality for Kentucky lies in a region which was intensively collected by Dr. Holt during July 1958, a fact which permits the inference that heterognatha may exist in central Kentucky only as a relict, and this fact, together with the generally sporadic distribution of species suggests that the range may be in the process of contraction, perhaps as a result of post-Pleistocene increase in temperature in the southern Appalachians. Nonetheless, it must be emphasized that the distribution of het- erognatha is by no means well-known, and many new localities doubt- less remain to be established. Remarxks.—It has already been observed that hetferognatha is endowed with a remarkable combination of presumably evolutionary specializations. It is one of the easiest of branchiobdellids to recog- ANNELID GENUS CAMBARINCOLA—HOFFMAN 365 nize, owing to the enormous dorsal jaw which dominates the entire anterior half of the head. In no other known form in the family is heterognathy carried to such an extreme. Additionally, the spermatheca offers very good specific characters, particularly its length and pronounced median constriction which are evident in every specimen examined. Presumably heterognatha is a highly modified member of the general philadelphica-chirocephala stock, which began to evolve its characters at a very remote time, or has been able to develop them more rapidly than other species of the genus. DEMISSA SECTION A separate section seems necessary to reflect the status of Cambarincola demissa, a very disjunct species which has little relation- ship with other members of the genus. Judged from body shape, small head and jaws, and general appear- ance, this species is perhaps one modified as a gill-inhabiting form, and therefore especially liable to various structural concomittances of a specialized habitat. Unfortunately, we know nothing definite about the preferred microhabitat of the species. Perhaps the most unusual feature of this animal is the marked reduction in the size of the reproductive systems, indicated in outline in the habit sketch of an entire worm (fig. 72). All of the normal organs are present, with the fine structure characteristic of the genus, but both spermiducal and prostate glands are very small proportion- ately, and the latter—although histologically not differentiated— terminates in a small clear bulb. The penial sheath of the bursa is much smaller, in relation to the atrium, than in any other member of the genus. DEMISSA GROUP A monotypic group with the characters of the section. The only species is a moderately small, corpulent-looking worm, probably branchiophilus, known only from extreme southwestern Virginia. Cambarincola demissa, new species Figures 72, 73 TYPE SPECIMENS.—Holotype and four paratypes, USNM 29948, from Orconectes erichsonianus and O. juvenilis collected in a tributary to the Powell River at Big Stone Gap, Wise County, Virginia, by Horton H. Hobbs, Jr., and C. W. Hart on June 17, 1950. Diacenosis.—A moderate-sized (2.8-3.2 mm. long) cuneate species of Cambarincola, differing from all other members of the genus in the exceptionally small sex organs, the prostate gland likewise differing 366 PROCEEDINGS OF THE NATIONAL MUSEUM Vou. 114 from all others in terminating in a small clear bulb although not histologically differentiated from the spermiducal gland. Descrietion.—BKody strongly enlarged caudally, segments vr and vil about twice as wide as segment 1, and tapering very abruptly to the fairly large caudal sucker which is as broad as the head or segment 1. Segments very short, as little as a third of the body diameter at segment VI; prosomites twice as long as metasomites but not elevated above them. Head as long as first three body segments, but smaller in diameter and thus continuing the anterior attenuation of the body; peristomi- um about a third of total head length, set off by a strong basal constric- tion, its margin apparently broadly lobed but not extended into projections or tentacles (all material slightly macerated). An addi- tional more posterior constriction of the head occurs and imparts a trisegmented appearance to the head in lateral aspect. Jaws relatively quite small, about 7.0 percent of the head length, the dorsal jaw slightly the larger and longer, with a distinct large pro- jecting median tooth; ventral jaw likewise with a median tooth and subsimilar in general form to the dorsal, at least in lateral aspect (none of the available material is mounted in a way to show dorsal or ventral surfaces of the jaws, but careful observation indicates that the dental formula is probably 3-3 or perhaps even 1-1). Male reproductive system very small, confined to the ventrolateral portion of the coelom of segment v1, extending dorsal less than half- way up one side of the segment. Bursa small, cordate, the atrial portion making up most of its bulk, the penial sheath confined to the ental fourth of the bursa and very small by comparison with that in other species. Ejaculatory duct moderately long, its length about equal to that of the bursa or spermiducal gland, its wall muscular but of normal thickness. Spermiducal gland and prostate collectively only about as large as the bursa, their histological structure similar (small, glandular, basophilic cells), but the prostate terminates in a small clear bulb presumably homologous to that so characteristic of the Philadelphica section. Spermiducal gland short and broad, at most only half again as long as the diameter; prostate slender but much shorter than spermiducal gland. Latter entally-rounded, without evident lobation at the entries of the small, slender deferent ducts. Spermathecae equally reduced, extending less than halfway up one side of segment v, the shape somewhat fusiform, expanding laterally from the small ectal portion and maintaining essentially the same size to the abruptly acuminate ental tip. VARIATION.—Owing to the small amount of available material, all of which is slightly macerated from the initial preservation in weak alcohol, it is not possible to dwell at any length on the observed varia- ANNELID GENUS CAMBARINCOLA—HOFFMAN 367 tion. Because of the small size of the sex organs, which are not par- ticularly confined by the gut as in most other species, the spermiducal gland and prostate in particular are liable to considerable freedom of motion in the coelom, and preserved specimens show much variation in the shape of these two structures due to a difference in perspective. The spermatheca seems to have a constant shape, as does the general body outline. The range of this species may be so limited that geo- graphically influenced variability will be found negligible. RELATIoNSHIPS.—There is no other species in the genus with which demissa can be compared. C. branchiophila of the Mesochorea section is known to be a gill form, and has small jaws and a cuneate body form, but its sex organs are of normal size and the major parts of the bursa correctly proportioned. It is difficult to derive demissa from either of the other two major sections of the genus, although if com- pelled to make a choice I should tentatively place it much closer to the Mesochorea section as representing the culmination of evolutionary tendencies in that ensemble. As this form is probably localized among the high mountains of southwest Virginia, there is every reason to presume that many other endemic species remain to be discovered, and some of these may cast some light on the affinities of this curious and disjunct little worm. Distrisution.—Aside from the type locality, C. demissa is known only from the following locality: VIRGINIA: TAZEWELL cCouNTY: Bluestone River, 11.2 miles east of Tazewell on U.S. Hy. 460; June 18, 1950, Hobbs and Hart (PCH 393). Remarks.—The possibility of any host specificity on the part of this species is negated by the fact that the type series came from a collection of two species of Orconectes, the other known material from a collection of two species of Cambarus. Presumably the factors influencing the distribution of demissa are those of simple geography and water conditions rather than the occurrence of the crayfish hosts. I cannot, at this time, imagine what might favor the selection re- sulting in the drastically reduced size of the reproductive systems, unless it be that residence within the branchial chambers of crayfish affords survival with a reduced number of progeny. Perhaps knowl- edge of the ecology of the species will shed some light on this interest- ing evolutionary problem. The name demissa (Latin, humble, unimposing, modest) seems appropriate in view of the general form of the animal as well as its very moderately scaled reproductive systems. Species of uncertain systematic position Under this heading I include two species which are known to be referable to Cambarincola, but which, for one reason or another, cannot 368 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 I> 2 ae (( ee ae ee oa Su LE Q VAG 79 Ficures 75-79.—Structural details of two species of Cambarincola. 75, C. floridana Good- night, body in lateral aspect, camera lucida drawing from holotype; 76, the same, jaws in dorsal aspect, from freehand sketch. 77, C. gracilis Robinson, body outline of holotype; 78, the same, male reproductive system of holotype; 79, the same male reproductive system, with bursa everted, paratype from British Columbia. at present be allocated to a group or even to a section with assurance. The type specimens of both have been studied and drawn; the data on hand are summarized and presented at this time for the benefit of future workers who may be able to collect at the type localities and establish the identities of the names. Cambarincola floridana Goodnight Fiaures 75, 76 Cambarincola floridana Goodnight, 1941, Trans. American Microsc. Soc., vol. 60, p70. Typr spECIMENS.—Holotype, USNM 20570, from Procambarus fallax collected in Taylor County, Florida, by Horton H. Hobbs. This specimen is mounted laterally, but the reproductive system cannot be made out with certainty. RemarKks.—The bursa of this species is of considerably greater size, in relation to the other organs, than in most other species of the genus. Otherwise the general configuration suggests that floridana may be a specialized member of the Philadelphica section. ANNELID GENUS CAMBARINCOLA—HOFFMAN 369 Cambarincola gracilis Robinson Fiaures 77-79 Cambarincola gracilis Robinson, 1954, Journ. Parasitology, vol. 40, p. 466, figs. 1-4.—Holt and Hoffman, 1959, Journ. Tennessee Acad. Sci., vol. 34, p. 103. TYPE SPECIMEN.—Holotype, USNM 26110, from Pacifastacus klamathensis collected at Whitman College, Walla Walla County, Washington, by A. G. Rempel. Paratypes: USNM 26111, from the Klamath River, Siskiyou County, California, and USNM 26112, Burnaby, British Columbia. Remarks.—This species was described and illustrated in some detail, the account being more meaningful than most of the existing descrip- tions. Not only the body form and jaws were drawn and discussed, but the form of the reproductive systems as well, and a reasonable comparison was made with C. inversa and C. macrodonta. It is un- fortunate that equal care in the diagnosis of new species was not expended by several of Miss Robinson’s predecessors. I have examined the type material in the U.S. National Museum, and cannot improve upon the original description except to note that the bursa is of the typical cordate Cambarincola-form, and not a eradual enlargement of the ejaculatory duct as indicated in Robinson’s figure 1. Her figure 3, a reconstruction from serial sections, was made from a worm with everted bursa. The reduced size of the male sex organs is perhaps of some diagnostic importance, the appearance in situ being reminiscent of that of C. demissa. Miss Robinson failed to record the histological appearance of the spermiducal gland and prostate, and, unfortunately, I neglected to note the same detail when examining the type. A re-examination of material of gracilis can readily establish the status and taxonomic position of the species. Assuming that the prostate is not differentiated (it is known to have no terminal bulb), the species will fall into the Mesochorea section and either the Branchi- ophila group or a new one of its own. C. gracilis was recorded from ten localities in California, Oregon, Washington, and British Columbia, a relatively extensiverange. Yet curiously enough, Goodnight (1959) refers gracilis to a category of so- called scarce, localized species in his most recent synopsis, while rank- ing in the main part of the key some of his own species known from but a single locality. The paratype from California cannot be studied with respect to the reproductive system. That from British Columbia, however, is well- mounted, and appears to be correctly identified with the holotype. Drawings made from the holotype and paratype are presented as an aid for future recognition of the species. 370 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 144 Literature Cited Evuis, Max M. 1912. € d Ficure 4.—Hydrobiosinae genus A: a, larval head and pronotum, dorsal view; 4, larval mandibles, dorsal; c, femur through tarsal claw of larval foreleg, posterior face; d, larval prosternum, ventral; ¢, left anal proleg of larva, lateral. NEOTROPICAL CADDIS FLIES I—FLINT 463 Atopsyche species 3 This larva is very similar to that assigned to alconura, but differs in lacking the ventral dark marks on the head. Larva.—Length 10 mm. Coloration as in alconura, except central stripe on frontoclypeus narrower and no dark marks ventrally. MarteriAu.—Venezuela: Rio Cobre of the Catatumbo system be- low La Grita, Mar. 31, 1942, L. P. Schultz, 2 larvae (USNM). Hydrobiosinae genus A FIGURE 4 In the material from Ecuador are two hydrobiosine larvae that differ in many structural characteristics from the larvae of Atopsyche. The most important differences are the fusion of the pronotal halves, the reduction of the prosternite to a minute sclerite, and the disappearance of a segment in the forelegs. No genus other than Atopsyche is known to occur in Ecuador, so I can not even speculate as to the generic placement of these larvae. Larva.—Length 6 mm. Head yellowish, muscle scars darker, posterior margin black (fig. 4,¢@). Labrum as in Atopsyche. Maxilla about the same, except lateral part of stipes almost completely mem- branous except at seta and mesal part strongly sclerotized toward base and closely united with cardo. Labium as in Atopsyche, except dor- sally where mesal strap lacks teeth and hypopharyngeal rods with only 5 or 6 teeth on strongly angulate apical portion. Mandibles with only a single dorsal subapical tooth (fig. 4,5). Pronotum yellowish, pos- terior and lateral margins black; lacking mesal suture (fig. 4,2). Pro- sternum very small (fig. 4,d), in width only one-fifth of the distance between legs. Forelegs chelicerate, but femur not twisted; femoral process much narrower than that of Atopsyche; apparently tibia and tarsus fused (at least only 1 segment obviously present); claw extend- ing beyond tip of femoral process (fig. 4,¢). Abdomen generally as in Atopsyche. Anal proleg lacking basoventral spine; claw sharply curved ventrad, basal segment with a broad curved spine arising apicoven- trally and a normal seta arising laterally at its base (fig. 4,e). MareriaAL.—Kcuador: Stream 5 miles south of Antisana, 13,500 ft., Apr. 28, 1958, R. W. Hodges, 1 larva; stream 11 miles west of Pujili, 12,500 ft., Mar. 15, 1958, R. W. Hodges, 1 larva. Family Glossosomatidae The subfamily Glossosomatinae is present in the New World only in the Nearctic region; the Protoptilinae, also widely distributed in North America, is the only subfamily present in the Neotropics. 464 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 114 Hight genera have been recognized for the South and Central American species. The wing venation used here is explained in Ross (1956, figs. 365-367). Key to Neotropical Genera of Glossosomatidae 1. Wings, especially hind, reduced in size and venation . . Merionoptila Schmid Wings not reduced ..... SE NCHO fo A Bie Ue mee 2. All four branches of M present in Porewine | . . . Culoptila Mosely Only two or three branches of M present in forewing ......... 3 3. \) Two branches‘of.. Min forewing) <0. 2h: eons ee. Go SO ee Mireeibranchesiofvisinwtorewangs cas. 1) ee eee se) 6) ees See 4. Cu, branched apically in hindwing. .... . . . . Mortoniella Ulmer Cu; unbranched in hindwing . . . em 5. Branching of M in forewing at pbaath same fleck as ranching of R445. Scotiotrichia Mosely M branching much nearer wing margin than Ry; . . . Mexitrichia Mosely 6. Rg; branching nearer forewing margin than Roi; . . . . . Canoptila Mosely Rapand R5.3 branching at nearly ‘same level-;*. = .2..*'5. 63 eee a 7.. Cuyin hindwing branching apically...) a | Matoptila Mosely Cuysimple:):.. ay .2erdevieoe 3 ser ark we as Syeacoat hb, spec 44:Pretopiila Banks The larvae of the Protoptilinae are very poorly known at present. Ross (1944) has described some particulars of the larva and pupa of Protoptila lega Ross, and Flint (1962) the immatures of Matrioptila geanae (Ross). The larvae of Mortonielia angulata, M. apiculata, M. hodgesi, and Meritrichia aries are described herein. The larvae of these species and of Protoptila alexanderi Ross differ strikingly from those of the Glossosomatinae in the structure of the anal claw. The larvae of the glossosomatines have one large ventral hook and 1 or 2 dorsal accessory teeth. In the protoptilines the dorsal accessory teeth apparently are divided longitudinally, hence there are 2 or more pairs of accessory teeth, one row on each side of the ventral central hook. In addition to the differences in the anal claw, there are dif- ferences between the subfamilies in the apical spurs of the tibiae, the sclerotization of the thoracic notae, and the symmetry of the mandibles. The larvae appear to show generic differences, although criteria for the separation of the species have yet to be found. The tarsal and anal claws, apical spurs of the tibiae, and ninth tergite show some of the most marked generic differences found so far. The following key is presented only as a guide to the characters found in the previously mentioned species and will probably require much modification when more species and genera become known. Key to the Genera of Protoptiline Larvae 1. Tarsal claws trifid; ninth tergite with 3 pairs of setae; 3 pairs of accessory anal teeth: 24. 7 see Wi ve iy iia ad. @ MatrieptilatRess NEOTROPICAL CADDIS FLIES I—FLINT 465 Tarsal claws with a single apical hook; ninth tergite with 2 pairs of long SOLHLO Wee OE bm Mie tobi hk oo 8 eSB A rei gla Pc ontel stPhe clue. cya) a4 = 2 2. Anal claw with 2 pairs of well-developed accessory teeth; seta of tarsal claw short and very broad (fig.7,e) -°. . . 1 ke we Mortoniella Ulmer Anal claw with 3 or more pairs of accessory teeth; seta of tarsal claw nor- EL ERS SO an ees I) OOD EL ORR hee vet 3 3. Ninth tergite concave apically; seta of tarsal claw arising from base of cuticu- Pani rent? (TB d ost 2:04) Ensen roe) whehen Mexitrichia Mosely Ninth tergite convex apically; seta of tarsal claw arising from posterior face Be cUiicolr Ouigrowul (fig. 7,7)... . . . . «ss Protoptila Banks Genus Mortoniella Ulmer Mortoniella Ulmer, Notes Leyden Mus., vol. 28, pp. 95-97, 1906. Ulmer described the genotype, bilineata, from Ecuador in 1906. Subsequently he added a second species, albolineata, from Brazil. The latter, however, was removed from the genus by Mosely in 1939 and placed tentatively in Antoptila. Martynov (1912) described tranquilla from Peru; however, this description, based on a female and unaccompanied by any illustration, is unrecognizable. Such is the present state of our knowledge of the genus. In the Cornell University collection is a male specimen from Ecuador that perfectly matches Ulmer’s description and figures of the genotype and permits the genitalia to be refigured. A second species present in both the adult and the immature material from Ecuador is congeneric on both genitalic and venational characters. Two additional species of Mortoniella are described from pharate adults that agree closely in genitalic structure. Mewsitrichia wygod- zinskit Schmid from Argentina apparently also belongs in Mortoniella on genitalic considerations, though Schmid’s comments on the venation indicate that Cu, in the hindwing is unforked in this species. If Schmid’s comments are accurate, Mewitrichia Mosely may have to be synonymized with Mortoniella, because the genitalic differences alone do not seem to justify a generic separation. Until more species and their venation become known, however, I prefer to regard the two genera as distinct. Larvae of the three new species described here are known to me. They are all similar, in fact, as yet I can find no specific differences, and they are easily separated from the larvae of the other genera. The structure of the tarsal and anal claws is characteristic. The anal claw has only 2 pairs of accessory teeth, and the seta on the tarsal claw has become greatly modified into a short, broad, thumblike process. It must be admitted, however that Mezxitrichia aries, on which the larval differences are based, may not be congeneric with the genotype and when larvae of other species in this genus become known these apparent differences may not prove to be valid. 466 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 Tentatively then, the genus Mortoniella differs from Mezitrichia in that: Cu, in the hindwing is forked; in the male genitalia the tenth tergum is elongate, the ventral surface of the aedeagus is often sclerotized in the form of paired processes, and there are comparatively simple basoventral processes; in the larvae there are only 2 pairs of accessory teeth on the anal claw, and the seta on the tarsal claw is a short, broad, thumblike process; and in the pupa the distal tooth on the inner margin of the mandibles is long and arises far from the apex. Mortoniella bilineata Ulmer Ficure 5,a Mortoniella bilineata Ulmer, Notes Leyden Mus., vol. 28, pp. 97-98, 1906. A figure of the lateral aspect of the male genitalia is given from a cleared specimen to supplement the figures of Ulmer. The only difference to be noted is in the tenth tergum which Ulmer shows to be slightly longer and slightly different in shape. Otherwise the genitalia, venation, and coloration are an excellent match with the original description. Matertau.—Ecuador: Huigra, 4500 ft., June 13, 1914, 1 0, Parish (CU). Mortoniella apiculata, new species Ficures 5,b-d; 6; 7,a-e This species is closely related to angulata from which it differs most conspicuously in the expanded apex of the dorsal process of the aedeagus and in the greater proportionate length of the genital capsule. Aputt.—Length of forewing 5 mm., length of body 4.5 mm. Wings of specimens in alcohol brown, with a pale streak in membrane at anastomosis; venation as shown (fig. 5,c). Male genitalia (fig. 5,6): Ninth segment almost quadrate in lateral view, very oblique; tenth segment with an elevated basomesal area, extending about half length of lateral arms which are separated by a deep narrow mesal incision (fig. 5,d); dorsal process of aedeagus angulate dorsally shortly before apex, apex widened into a thin triangular structure; lateral processes of aedeagus simple, slightly angulate rods extending about two- thirds the length of dorsal process; ventrally to aedeagus a pair of short rods basally and 2 paired processes near midlength, ventralmost pair longer than dorsal pair; venter of aedeagus sclerotized in form of paired spatulate process. Holotype o, allotype, paratype 9: Ecuador: 1 mile east of Papal- lacta, Jan. 29, 1958, R. W. Hodges (USNM Type 66018). Additional paratypes: Ecuador: 6 moderately sclerotized, U-shaped pouch receiving and supporting © bases of harpes. Lateral aspect: narrowing dorsocaudad to point of articulation with lateral arm of tegumen. Tegumen with lateral arm quite broad, gradually narrowing cephalad to point of articulation with vinculum, approximate ventral half punctate, dorsal half glabrous, ventral margin heavily sclerotized and sinuate, dorsal margin moderately sclerotized and sublinear; dorsal area broad, glabrous, rather weakly sclerotized but not emar- ginate mesad. Harpe simple, slightly variable. Lateral aspect: width medium, sublinear or with apical half curving somewhat ventrad; costa and sacculus fused, comprising approximate basal half of harpe, broadest in apical portion at point of attachment of arm of transtilla, gradually narrowing basad to rather narrowly rounded basal extremity, dorsal margin very heavily sclerotized; glabrous except for heavily punc- tate and setose, considerably expanded, subdigitate, caudal portion of sacculus; cucullus partially separated from costa and sacculus by dorsal constriction and ventral area of reduced sclerotization, com- prising apical half of harpe, commonly curving somewhat ventrad and mesad, markedly constricted dorsad near base, major apical portion rather broad, heavily punctate and setose ectad and entad except for glabrous dorsal portion of basal two-thirds, apex broadly and evenly to irregularly rounded. Transtilla with arm approximately one-third as long as harpe, slender, linear to sublinear, well sclerotized, glabrous, rather closely subparallel with dorsal margin of costa, terminating considerably distad of basal extremity of harpe, apex slightly expanded and rounded. Uncus prominently bifid, somewhat variable but very distinctive. Dorsal aspect: base large, heavily punctate and setose, partially set off from tegumen by lateral areas of reduced sclerotization; lateral margins well sclerotized, sublinear to sinuate, converging distad; angle of bifurcation narrowly rounded to acute; furcae with basal portions approximate, broad, strongly narrowing distad, heavily punc- tate and setose. Lateral aspect: furcae abruptly directed ventrad to ventrocephalad at angle of 90 degrees or more, broadly expanded, laterally flattened, subtriangular, punctate and setose, closely sub- parallel to weakly divergent ventrad, gradually narrowing to acute apices. Gnathos paired, curving ventrocaudad, lateral and apical portions well sclerotized, large mesobasal portion membranous; arms flattened, densely scobinate entad, apical portions overlapping, apices rather narrowly rounded. Anellus membranous, unarmed, juxta absent. NORTH AMERICAN ACROLOPHIDAE—HASBROUCK 561 Aedeagus approximately three-fifths as long as harpe, considerably flattened dorsoventrad, sublinear in dorsal and ventral aspects, ap- proximate apical third curving somewhat ventrad in lateral aspect, asymmetrical in all aspects, glabrous, base broadly and irregularly expanded laterad and ventrad, basal half tubular, apical half opening broadly and irregularly dextrad; apical portion consisting simply of one or two, elongate, slender, moderately sclerotized, dorsosinistral strips terminating acutely or subacutely. Vesica large, membranous, armed with two prominent clusters of cornuti: cornuti variously sized, mostly large, heavily sclerotized and darkened, linear, with apices finely acute; basal cluster largest, located dorsodextrad near base of vesica, containing 4-13 cornuti; apical cluster smaller, located dextrad nearer apex of vesica, containing 2-8 cornuti. The number of cornuti in each cluster varies from specimen to specimen and thus constitutes the chief intraspecific variation to be found in the genitalia of this species (see the tabulation, below, for examples of this variation). In the @ genitalia of all the individuals of popeanellus that have been examined, the presence of two distinct clusters of cornuti in the vesica has proved to be a consistent and positive specific character. However, these cornuti commonly vary in size and number between the two clusters of any one individual as well as between the corre- sponding clusters of any two individuals. Thus, they furnish a good example of an intergrading, intraspecific variation in regard to their sizes, numbers, and ratios between the basal and apical groups. This variation is illustrated by a sampling of 18 specimens listed in the following table. Tape 1.—Intraspecific variation of cornuti in A. popeanellus Cornutt Specimen Basal Apical Total 1 13 5 18 2 10 6 16 3 9 2 11 f 8 a 15 5 8 6 14 6 8 6 14 7 8 3 11 8 7 8 15 9 7 7 14 10 fi 6 13 11 7 5 12 12 iG 5 12 13 7 2 9 14 6 5 11 15 6 3 9 16 5 4 9 17 4 5 9 18 4 4 8 562 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 114 The basal cluster may contain as few as 4 cornuti to as many as 13, the apical cluster may contain from 2 to 8, and the total for both clusters ranges from 8 to 18. The basal cluster usually contains the most cornuti, but occasionally its number is equaled or even exceeded by the number in the apical cluster. Individually, the cornuti vary in size from minute to very large. ‘This variation is essentially nongeographic. Typr.—Two type specimens in the Academy of Natural Sciences of Philadelphia. TypE LOCALITY.—‘‘Texas.”’ DistRIBuTION.—This species has a very wide range. It has been reported from northern South America, Central America, and Mexico, as well as from the United States where it is quite common except in the northwestern states and the extreme north and west. It has been recorded from Arizona, New Mexico, and Texas northward through Oklahoma, Kansas, and Nebraska and eastward to Pennsylvania, New Jersey, North Carolina, and Florida. SPECIMENS EXAMINED.—606, from 106 localities. Large series con- taining both sexes are to be found in the collections of the author, the American Museum of Natural History, and the University of Kansas. At least one quarter of all the specimens accumulated for this study eventually proved to be A. popeanellus, and the number of specimens available for study has made it advisable to reduce the distributional data for this common species simply to the localities and months of occurrence: ALABAMA: Gurley (July); Huntsville (July); Valley Head (Aug.). ARIzoNa: Chiricahua Mountains (July); Douglas (Aug.); San Bernardino Ranch, Cochise Co. (Aug.); Tombstone (Aug.); Tueson (July). Arkansas: Camp Chaffee (Aug.); Hope (Aug.). FiLorma: Key Largo (Aug.); Sanford (June); Winter Park (May, June). Groraia: Bainbridge (Sept.); Catoosa Co. (Aug.); Clarke Co. (Aug.); Screven Co. (July, an unusually small ?); Tallapoosa (July). IL- LINOIS: Champaign-Urbana (June and July, both sexes very common at light); Dubois (July); Havana (June); Homer Park, Homer (June, July); Lawrenceville (Sept.); Murphysboro (Aug.); Oakwood (July); Palos Park (July); Parker (July). Peoria (June, July); Putnam County (June, July, Aug.); Quincy (June). InpIANA: Putnamville (July); Tremont (June). Iowa: Denison (June); Des Moines (July, Aug.); Linn Co. (no date); Sioux City (June, July). Kansas: Barton County (June); Bourbon Co. (no date); Caldwell (May, June, Aug.); Clark Co. (May, June); Decatur Co. (July); Douglas Co. (June and July, 2 0d with mites on abdomen); Gove Co. (no date); Linn Co. (no date); Manhattan (June); Medora (June); Ness Co. (July); Onaga (July); Republic Co. (July); Riley Co. (July); Scott City (June); Scott Co. (June, 1 o& with mite on eye); Sheridan Co. (no date); Topeka (no date). Kenruckxy: Harrodsburg (Aug.). Mississippi: Starkville (July); A. & M. College, State College (Aug.). Mu1ssourt: Brentwood (June); Kirkwood (May, June, July); Mineola (July); Ranken (June) ; St. Louis (July); Webster Groves (July); Willard (June). Nrsrasxka: Lincoln (June, July, Aug.). New Jersey: Vineland (no date). New Mexico: Artesia (July); Carlsbad Caves, Carlsbad (July); Chaves Co. (July); Deming (July); Eddy Co. (July); Las Cruces (Aug.); Lordsburg (Aug.); Mesilla (July); State NORTH AMERICAN ACROLOPHIDAE—HASBROUCK 563 College (July, Aug., Sept.). Norra Carouina: Hickory (July, Aug.); Kinston _ (Aug.); Maxton (May); Raleigh (June, July, Aug., Sept.); Wilkesboro (Aug.). _ Oxto: Granville (July). Oxkuanoma: Bartlesville (July); Clinton (June); Eagle- town (June). PENNsyLVANIA: Finleyville (July); Manoa, Delaware Co. (July). _ SourH Caronina: Myrtle Beach (June, July). TENNEssEE: Memphis (Aug.); | Monteagle (July, Aug.); Nashville (July). Texas: Brownsville (June); Burnet Co. (July); Christoval (June); College Station (Oct.); Comfort (May, Sept.); Corpus Christi (Sept., Oct.); Davis Mountains (Aug.); Devils River, near Del Rio (May); Eastland Co. (April); Kingsville (June); Marathon (July); New Braunfels (May); Palo Duro State Park, Randall Co. (Aug.); Richmond (June); San Angelo (June). Virainta: Salem (Aug.). Remarks.—A. popeanellus, one of the first three acrolophids described from the United States, is undoubtedly our commonest and most widely distributed species and is one of the larger and more robust acrolophids occurring in North America. It also exhibits considerable variation in size and coloration. These four factors explain why popeanellus is also the most heavily synonymized North American acrolophid. A. popeanellus is closely related to klotsi, the two comprising a smal] species group in general related to those acrolophids having elongate labial palpi, setose eyes, laminate antennae, bifid uncus, and paired gnathos. However, popeanellus and klotsi may be distinguished from each other and from all other acrolophids on the basis of their aedeagi and unci. The latter organ, in the popeanellus-klotsi group, 1s distinctive in having its furcae abruptly directed or curved very strongly ventrad. In popeanellus, these furcae, easily observable in dried and undissected specimens, are characteristically expanded into subtriangular plates. The genital characters of popeanellus are consistent throughout my very large series and they are quite distinct from those of all the other species treated here. I have not examined the type specimens of this species. Busck (1903), in his report on Clemens’ types of Tineina in the Academy of Natural Sciences of Philadelphia, noted: Anaphora popeanella Clemens. Two types, both rubbed, one unspread and without abdomen; the other spread, and lacking the head and left wings. Clemens’ No. 11; alar exp., 28 mm. Like the present conception of Anaphora popeanella, as determined by Walsingham and synonymous with scardina Zeller, and with agrotipennella Grote. Aspecimen compared with the type is in the U.S. National Museum. Habitat: Eastern United States. Darlington located one of these type specimens at Philadelphia, and described it (in litt., 1946) as: ‘“‘popeanella Clem. Type. Right wings only, poor condition, much rubbed, sex not determined.” The com- bined information that I have been able to gather in regard to pope- anellus leaves no doubt in my mind as to the proper identity and cor- rect concept of this species. The specimens of popeanellus in the U.S. National Museum agree with my previous concept of this moth. In 564 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 114 addition, at the Museum are several slide preparations of # genitalia, labeled ‘‘popeanellus Clem.,’”’ agreeing with my figures of this species. Clemens apparently named this moth after Captain Pope. Wals- ingham (1887, pp. 155-156) designated this species as the type of Clemens’ genus, Anaphora, and ten years later (1897) reported that the larvae of popeanellus attack the roots of Trifolium pratense in April and May, and that this insect also occurs in the West Indies (Puerto Rico). Grote described the &@ of Anaphora agrotipennella as a new species in July 1872. The locality given was ‘‘Central Alabama” where Grote reported the moth to be very common in June and July. An alar expanse of 27 mm. was listed. Following the description Grote stated: I have only seen males of this species, in which the ornamentation of the fore wings above recalls that of various species of Agrotis, such as A. jaculifera, etc. I have tried to recognize in this species A. Popeanella, Clemens, from Texas, but I have failed to reconcile his description with my specimens, which are not “‘luteous or yellow along inner margin.”’ In A. agrotipennella, at the extremity of the median ochrey shade subterminally, are a few black scale points. These can hardly be the same as the row “of dark brown spots’”’ of Popeanella. Neither can I, from the description, consider the differences of colour and ornamentation as produced by any defect in the condition of Dr. Clemens’ specimens. Grote, describing the 9 of Anaphora agrotipennella in August 1872, further attempted to differentiate his species from Clemens’ popeanella. When Walsingham (1887) placed it as a synonym of popeanella, he stated: An examination of my extensive series shows that both varieties belong to the same species. The anal appendages do not differ, and intermediate variations of colouring are noticeable. There can be no doubt that they differ only in the extension of the pale colour of the fold in the direction of the dorsal margin. Walsingham also added that Zeller had evidently regarded the two as synonymous. Since 1887, agrotipennella has been generally treated by writers as a synonym of popeanellus, although Barnes & McDun- nough (1917) and McDunnough (1939) maintained it as a distinct and valid species in their checklists. I have not seen the type specimen, which is in the collection of the Academy of Natural Sciences, but Darlington, in 1946, reported it to be in perfect condition. All the specimens labeled ‘“‘agrotipennellus’”’ that I have ever seen proved, upon examination of their genitalia, to be examples of popeanellus. Likewise, there is no indication in Grote’s descriptions that his agrotipennella could not be popeanellus. In addition, there is no North American species available, save popeanellus, to rep- resent Grote’s descriptions of agrotipennella. It thus seems best to consider agrotipennellus an old synonym of popeanellus (Clemens) with Walsingham receiving the credit for sinking it. NORTH AMERICAN ACROLOPHIDAE—HASBROUCK 565 Zeller described Anaphora scardina as a questionably new species in 1873. At the conclusion of his discussion of Clemens’ genus Anaphora, Zeller said of his own series of specimens: “Whether I have Clemens’ species among my own before me, I cannot decide with certainty from his descriptions; therefore I am giving the latter under separate names.’’ He then proceeded to describe scardina and bombycina at considerable length. That Zeller was uncertain of the validity of his species, scardina, is evidenced by the fact that immediately beneath its designation at the head of the original description he inserted ‘‘? Popeanella Clemens’ rather than listing it as “n. sp.” He gave the ‘“Vaterland” or locality as Texas and Carolina, where popeanellus is quite prevalent, and also mentioned that the largest example of all in his collection was from an unknown locality, thus indicating the possibility of a mixture of species. Cham- bers (1878), following Zeller’s note of caution, gave the listing ‘A. scardina, Zell. =? popeanella, Clem.” Walsingham (1887), after studying Zeller’s specimens, placed scardina in the synonymy of popeanella, where it has since consistently appeared. However, the type material for scardina, now in the British Muse- um (Natural History), presents a considerably different situation. One of the three photographs furnished by Tams is labeled “‘scardina Zell., type &”” and easily but inconclusively passes for an adult popea- nellus. Likewise, a second photograph labeled “seardina Zell.”’ rep- resents a ventral view of the complete < genitalia of a specimen which is clearly and unmistakably popeanellus. On the other hand, the third photograph, also representing a set of genitalia and labeled “scardina Zell., type o,” certainly does not represent popeanellus, nor does it agree with any other known North American species of Acrolophus. Hence, it is quite probable that Zeller’s scardina is actually a valid species of Acrolophus, although not referable to popeanellus or any other North American species. It also seems fairly certain that Zel- ler’s original series of scardina contained at least two species of Acrolo- phus: one, popeanellus, from Texas (also the locality of Clemens’ orig- inal specimens of popeanellus) and Carolina; the other, apparently designated by Zeller as the type of scardina, from parts unknown. The latter may well be the large example of unknown origin mentioned in Zeller’s original description. Although scardina has long been considered an old synonym of popeanellus, the type specimen indicates it does not belong to that segment of the genus occurring north of Mexico. The name is spelled scardinus to agree grammatically with its present genus, Acrolophus. Walsingham (1887) described Anaphora morrisoni as a new species on the basis of two oc" collected in Florida by Herbert Knowles Morrison. The alar expanse given was 18 mm. He apparently 566 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 114 named the species after the collector, who died in 1885. Since 1887, morrisont has consistently appeared in the literature as a distinct species. Walsingham’s illustrations of the o genitalia of morrisoni, showing the uncus and the cucullus of the harpe in both lateral and dorsal aspects, are essentially the same as his corresponding figures of the o& genitalia of popeanellus. On the preceding page of his revision, Walsingham had separated the two species as follows: 1. Points of the uncus distinctly separate—popeanella. 2. Points of the uncus closely approximate—morrisont. This difference is easily absorbed in the range of intraspecific variation exhibited by popeanellus. The type o& is at the British Museum (Natural History), from which tams has sent two photographs, labeled ‘‘morrisoni Wals., type 3.’ These show, through a dorsal view of the pinned moth and a ventral view of its entire genital capsule, that it is simply a small but typical Floridian specimen of popeanellus. Thus, morrisont (Walsing- ham) should be considered a new synonym of popeanellus (Clemens). Beutenmiiller’s manuscript name,