Patatoteretss vt mh a ae oy , Ni : Wey SMITHSONIAN INSTITUTION MUSEUM OF NATURAL HISTORY Noturus albater, new species, a female paratype, 63 mm. in standard length; UMMZ 102781, Missouri. (Courtesy Museum of Zoology, University of Michigan.) UNITED STATES NATIONAL MUSEUM BULLETIN 282 A Revision of the Catfish Genus Noturus Rafinesque, With an Analysis of Higher Groups in the Ictaluridae WILLIAM RALPH TAYLOR Associate Curator, Division of Fishes SMITHSONIAN INSTITUTION PRESS CITY OF WASHINGTON 1969 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 specialists 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. This work forms number 282 of the Bulletin series. FRANK A. TAYLOR Director, United States National Museum U.S. GOVERNMENT PRINTING OFFICE WASHINGTON : 1969 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 - Price $2 Contents Introduction . Acknowledgments. Source of material. Methods. Counts Measurements . : The Family Ictaluridae . The sensory canal system. =e : Relationships of the genera of the Teteduridas ; Genus Noturus Rafinesque . ‘ Key to the subgenera of Noturus . History . Zoogeography Sexual dimorphism . Reproduction. Early development . ‘ Subgenus Schilbeodes Bleeker . axe Key to the species of the subgenus Senbendon. Noturus gyrinus (Mitchill) . aie Noturus lachnert, new species . Noturus exilis Nelson . Noturus leptacanthus Jordan . ; Noturus nocturnus Jordan and Gilbert. Noturus insignis (Richardson) The funebris group. : : Noturus funebris Gilbert aa Swraitl : Noturus phaeus, new species . : Noturus gilberti Jordan and Beeeae Subgenus Noturus Rafinesque. Noturus flavus Rafinesque . . Subgenus Rabida Jordan and cermin Key to the species of the subgenus Rabida . The hildebrandi group . : Noturus hildebrandi (Bailey ane avian) Noturus hildebrandi hildebrandi (Bailey and Taylor) : Noturus hildebrandi lautus, new subspecies . Noturus baileyi, new species . Noturus albater, new species . Theelegans group. . ‘ Noturus elegans, new species . Noturus trautmanz, new species . Noturus eleutherus Jordan . The furiosus group . - Noturus placidus, new species . Noturus stigmosus, new species . 2 Noturus munitus Suttkus and Taylor. VI CONTENTS Noturus furiosus Jordan and Meek . The miurus group . Noturus miurus Jordan . Noturus flavipinnis, new species . Noturus flavater, new species . Hybridization .... Noturus gyrinus (Mitchill) X Noturus miurus Jordan . Noturus exilis Nelson * Noturus miurus Jordan . Phylogeny . Literature cited . . Tables 1-28 Plates 1-21. Index . . . follows Page 184 188 190 201 204 209 210 214 216 221 261 307 311 A Revision of the Catfish Genus Noturus Rafinesque, With an Analysis of Higher Groups in the Ictaluridae Introduction This study of the small North American catfishes in the genus Noturus was undertaken to analyze the species and to determine their relationships. It is based on critical examination of most of the many thousand specimens of Noturus now in museums, upon several hundred skeletonized or cleared and stained specimens of catfishes, and upon comparison with most of the other species in the family Ictaluridae. The species in the family Ictaluridae appear to constitute six genera, forming three major natural groups. Each group includes one mono- typic genus that is blind, unpigmented, and of restricted subterranean range, and one genus with species that are eyed, pigmented, and of widespread distribution in surface waters. The divisions are: (1) an Ictalurus group including the genus Jctalurus Rafinesque and the blind Trogloglanis pattersoni Eigenmann, (2) a Noturus group containing that genus and Prietella phreatophila Carranza, and (3) the Pylodictis group consisting of the large Pylodictis olivaris (Rafinesque) and the eyeless Satan eurystomus Hubbs and Bailey. The 23 species of Noturus include one subspecies and ten species that are described here as new. They are arranged in three subgenera: Rabida Jordan and Evermann, Noturus Rafinesque, and Schilbeodes Bleeker. With Prietella, several of their characters are intermediate between those of the Jctalurus group and those of the Pylodictis group. In certain characters they appear to be the most primitive, but they have specialized in several directions, exemplified by loss, reduction, or fusion of some structures and by increase in others. Hubbs and Raney (1944) studied much of the available material of Noturus exilis, Noturus insignis, and Noturus gilberti, giving some characters and mapping distributions. However, they incorrectly changed the names of ezilis, insignis, and gyrinus to insignis, margin- atus, and mollis, respectively, in the then recognized genus Schilbeodes. Otherwise, no study of the genus Noturus utilizing the existing material has been made since Jordan (1877d) and Swain and Kalb (1883). The species of Noturus are mainly active at night, hiding in cavities or beneath objects during daylight. Consequently, they are often most readily collected by using chemicals, direct current electricity, or by seining after dusk. Some of the species appear to be very spottily 1 2 U.S. NATIONAL MUSEUM BULLETIN 282 distributed and are thus infrequently obtained except in intensive survey work. One, Noturus flavipinnis has not been collected since 1893. The known localities from which it was obtained are now ecologically unsuitable for its existence and it may be extinct. Although not often encountered, the members of the genus become rather widely known because of the painful ‘“‘sting’”’ produced by their pectoral spines. The study of the variation and distribution of populations in Noturus has revealed several interesting patterns and presented problems of interpretation. Some wide ranging species, particularly miurus and insignis, show relatively little structural change geographically, but several structures in Noturus gyrinus gradually increase in length and number from north to south. Similarly a west to east increase is ap- parent in pectoral rays of Noturus leptacanthus and in anal rays of Noturus funebris. The presence of highly variant populations of Noturus nocturnus in the Red River system, which is in the middle of its range, is puzzling. In a number of species, localized and relatively minor variation is noted. Three clusters of species, each with closely related allopatric repre- sentatives, are recognized as species groups; other species groups are suggested but the relationship of the species is not as clear. These allopatric populations are recognized as species because they show a relative morphological uniformity, with consistent divergence from related populations; trends or gradation of characters from one popula- tion toward the other do not exist and intermediates or intergrades are lacking. Natural hybrids between species of Noturus are described but hy- bridization is rare. Sexual dimorphism in the family, once thought to be important in the interpretation of species of Noturus, has been exaggerated; instead differential morphological change, accompanying growth and sexual maturity, has resulted in several unnecessary names in Ictalurus, complicating its nomenclature. Acknowledgments A great many individuals have contributed importantly to this study in many ways, especially by the loan of specimens or making them available for study, provision of laboratory space, aid in col- lecting specimens, or have given freely information about specimens, notes on specimens in their care, or data from writings unavailable to me. To each I am greatly indebted and I sincerely trust that no one has been overlooked: Reeve M. Bailey, Marie-Louise Bauchot, Norman Benson, the late Leon Bertin, Albert P. Blair, James E. Bohlke, E. Milby Burton, Robert S. Campbell, Jorge Carranza, William M. Clay, Robert E. Cleary, Bruce B. Collette, the late CATFISH GENUS NOTURUS RAFINESQUE 3 Fannye A. Cook, Walter R. Courtenay, Jr., Frank B. Cross, Myvanwy M. Dick, Samuel Eddy, D. H. Enlow, Charles E. Farrell, the late Henry W. Fowler, Barry O. Freeman, Glenn Gentry, Shelby D. Gerking, Carter Gilbert, Francois Gouin, P. H. Greenwood, Harold W. Harry, Clark Hubbs, Stanton Hudson, Robert H. Kanazawa, Leslie W. Knapp, Ernest A. Lachner, Robert E. Lennon, L. W. Lowe, Y. J. McGaha, Giles W. Mead, Robert R. Miller, George A. Moore, Harry H. Moore, William M. Palmer, William L. Pflieger, Edward C. Raney, William J. Richards, Carl D. Riggs, Franklin O. Roberts, C. Richard Robins, Perry E. Robinson, C. E. Ruhr, J. Clark Salyer II, Leonard P. Schultz, Albert Schwartz, Frank J. Schwartz, Donald C. Scott, W. B. Scott, Clarence L. Smith, Hobart M. Smith, Philip W. Smith, William F. Smith-Vaniz, Franklin F. Snelson, Jr., Victor G. Springer, the late Margaret Storey, Royal D. Suttkus, Milton B. Trautman, Ethelwyn Trewavas, Denys W. Tucker, Eugene W. Whitney, Martin Wiley, Norman J. Wilimovsky, Howard E. Winn, Arthur Witt, Jr., Loren P. Woods, and Ralph W. Yerger. I appreciate the permission of Carl L. Hubbs to include the de- scriptions of Noturus albater and Noturus flavater. These species were discovered and their previously unpublished names were coined many years ago by Dr. Hubbs. The frontispiece, depicting Noturus albater as drawn by the late Grace Eager for Dr. Hubbs, is repro- duced here by permission of the Museum of Zoology, University of Michigan. The photographs of Noturus lachneri, N. hildebrandi lautus, N. baileyi, and N. elegans from the Duck River system were made by William F. Smith-Vaniz; the others by the University of Michigan Photographic Services. The drawings of the premaxillary teeth, and of the pectoral spines of Noturus lachneri, N. hildebrandi lautus, N. baileyi, and N. elegans from the Duck River system were made by Dorothea Schultz; all others were rendered by William L. Brudon. William F. Smith-Vaniz and Fanny Phillips both greatly aided me in the preparation of the charts and maps. The following individuals have read the manuscript or portions of it at some stage in its preparation, and provided helpful criticisms; I gratefully acknowledge their services: Reeve M. Bailey, Claude W. Hibbard, Carl L. Hubbs, Karl F. Lagler, Robert R. Miller, George A. Moore, and Milton B. Trautman. Finally, I am deeply grateful to Reeve M. Bailey, under whom this study was initiated, for his enthusiasm, encouragement, many suggestions, and helpful criticism. 4 U.S. NATIONAL MUSEUM BULLETIN 282 Source of Material Representatives of the genus Noturus from the following collections have been studied. The abbreviations are those used throughout the text in referring to the material. If not included here, further data pertaining to specimens in the collections indicated by an asterisk may often be found in Taylor (1955). AF Mississippi State Game and Fish Commission, Jackson, Mississippi* ANSP Academy of Natural Sciences of Philadelphia* BMNH British Museum (Natural History) CM Charleston Museum* CNHM Field Museum (formerly Chicago Natural History Museum) * CU Cornell University* DBUF University of Florida* FSU Florida State University INHS Illinois Natural History Survey* IU Indiana University (Note: Gerking (1945) listed his collec- tion of Indiana fishes now at Indiana University. An older collection, formerly at Indiana University, has been partially transferred to the University of Michigan. It bears the abbreviation IU, in parentheses after the ab- breviation UMMZ, since it has not been fully cataloged into the Michigan collection.) KU Museum of Natural History, University of Kansas* MCZ Museum of Comparative Zoology, Harvard University MNHN Museum National d’Histoire Naturelle, Paris NCSM North Carolina State Museum, Raleigh OAM Oklahoma State University* OSU Ohio State Museum (chiefly the former Franz Theodore Stone Laboratory collection), Ohio State University* OU Museum of Zoology, University of Oklahoma* SMF Senckenbergische Naturforschende Gesellschaft, Frankfort am Main SU Division of Systematic Biology, Stanford University* TNHC University of Texas* CU Tulane University Tulsa U Tulsa University* UGa University of Georgia UL University of Louisville U Minn University of Minnesota UMML University of Miami, Institute of Marine Science UMMZ Museum of Zoology, University of Michigan* UMoMZ University of Missouri Museum of Zoology USNM United States National Museum Vanderbilt U Vanderbilt University Specimens of the blind catfishes, in addition to those in Tulane University and the United States National Museum, were loaned by the Witte (now Texas) Memorial Museum, San Antonio, Texas, and the Instituto Mexicano de Recursos Naturales Renovables, Mexico, D.F., Mexico. CATFISH GENUS NOTURUS RAFINESQUE 5 Methods Descriptions of most of the counts and measurements used in this study can be found in Hubbs and Lagler (1947, pp. 8-15 and later editions). The counts of caudal rays and pores in the sensory canal system and some measurements are especially applicable to these fishes. They are not generally used or have been employed in this group for the first time. Counts ANAL FIN RAyS.—All rudimentary rays at the anterior end of the fin were counted. These simple rays decrease somewhat in number with age, i.e., some become branched rays. The number of anal rays was usually determined by making an incision above the anterior end of the fin, along the side of the body, then pulling the skin down to expose the minute rays, and counting them with the aid of strong transmitted light. A few counts were obtained by use of soft x-rays (see caudal fin rays). The full number of anal rays is present and countable at a very early age. Dorsa. FIN RAYS.—Except in Prietella a prominent long hard spine is always present at the anterior end of the dorsal fin and is preceded by a short plate-like spine; neither is usually indicated, because the variation is in the number of soft rays. The full complement of dorsal fin rays is present in the yolk sac stage. PELVIC AND PECTORAL FIN RAYS.—A hard spine always precedes the soft rays of the pectoral fin, but it is also usually excluded from the tabulations. The outer pelvic ray is typically simple, but on occasion may be branched; the frequency of occurrence of its branching was not determined. A minute ray-like structure, separate from and lateral to the base of the outer pelvic ray, was observed in Jctalurus and Pylodictis after clearing and staining. This structure has apparently degenerated into a small round ball or disappeared in Noturus. Shelden (1937) did not mention the element; if it is a degenerated ray or an accessory structure is not known. It is not included in the fin ray counts. Enumerations from both the right and left fins are tabulated. In the genus Noturus, nothing is gained by adding the counts from the two sides. Individual asymmetry is frequent, but no population was found to be predominately asymmetrical. A slight suggestion of correlation in the increase in numbers of both pelvic and pectoral fin rays was noted in several species, particularly in N. flavus, N. albater, N. eleutherus, N. miurus, N. gyrinus, N. nocturnus, N. gilberti, N. phaeus, and N. funebris. In them, a high number of pectoral rays is frequently accompanied by an increase in number of rays in the pelvic fin. In 6 U.S. NATIONAL MUSEUM BULLETIN 282 N. gyrinus this correlation also appears to correspond roughly with an increase in the number of preoperculomandibular pores, forming parallel geographic gradients. A prominent geographic increase in number of pectoral rays, but not in pelvic rays, is found in Noturus leptacanthus. The full complement of paired fin rays is probably present in the yolk sac stage; the normal complement was counted in speci- mens that had just lost the sac. CauDAL FIN RAYS.—The full complement of caudal rays is appar- ently attained early in the first year of life after the yolk sac is lost (fig. 1 and, see p. 32, description of the young of Noturus miurus). In old specimens, there may be degeneration of one or two rays at the anterior end of the fin. This is suggested by occasional short rays (counted) that are neither attached at their base nor continue into the body. The number of branched rays in Jctalurus and Pylodictis is nearly constant, and they are differentiated at an early age, but in Noturus some of the simple caudal rays apparently become branched with age. The variation in number of branched rays is probably because of a continuation of the branching process, in the long rays, throughout life. That the total number of caudal rays does not change after early in the first year of life is evidenced by general observation, by plotting the number of rays according to standard length in several species (see figs. 1 and 4), and from the fact that the rays, once developed, are distinct and do not grade imperceptibly into the surrounding tissues. The anterior rays often lie rather loosely inside the fin mem- brane. The number of rays in each “half” of the fin also remains constant. Enumeration of caudal rays in the species of Noturus, especially in large individuals, is difficult and tedious. A few tabulations were from cleared and stained specimens. Others were made by the use of soft x-rays; in this method, fins were taped closely to the film holder, containing high contrast film, and exposed at 22KV, 150 MAS, with the x-ray unit about 20 inches from the subject. The method most frequently used in determining the number of caudal rays is as follows: the mucus is cleaned from both sides of the fin by scraping with a dull knife; the fin is then rinsed and pressed flat against the glass stage of the microscope in a small pool of liquid; strong light is projected through the fin and a count made; the fish is then turned over to check the enumeration. The caudal fin can be conveniently divided into the following parts: Upper simple caudal rays: All unbranched rays in upper part of fin. Lower simple caudal rays: All unbranched rays in lower part of fin. Branched caudal rays: All branched rays in the fin. CATFISH GENUS NOTURUS RAFINESQUE v $S 5 SRFR ESR » n Caudal Rays in Noturus miurus Jordan Huron River, NW. of Dexter, Michigan a= @-Q Iyoung 3 & 8 n > 1 1<—_—___—__§—— Second Year ond Older -——HH———_ ------- > (Collection taken in August) (Collection takenin May) STANDARD LENGTH-mm. Ficure 1.—Chart showing, by sex and size, the number of caudal rays in Noturus miurus. All specimens are from an area 2 or 3 miles northwest of Dexter, Michigan. Those collected in August were young that had not left the nest (see p. 30). Specimens above 20 mm., which were collected in the spring, had a full complement of caudal rays. No sexual difference in number of rays is evident. In the yearlings and older specimens the rays are distinct and stand out from the surrounding tissue. In the young, counts were made after staining with alizarin red S. Without this preparation the rays graded imperceptibly into the primordial fin tissue. Upper-half caudal rays: Approximately one-half of the caudal fin is above a rather broad gap between the rays that extend from beneath the end of the axial streak (the line formed by the junction of the epaxial and hypaxial muscles). This displacement of the rays, near the middle of the fin, results from their attachment above and below a wide gap between the three lower and three or four upper hypurals. All rays, both simple and branched, that are above this gap are included in the upper half of the fin. Lower-half caudal rays: All rays below the above mentioned gap. Total caudal rays: All rays in the caudal fin. Branched rays in the upper half of the caudal fin (upper lobe rays): All branched rays in the upper half of the fin. Branched rays in the lower half of the caudal fin (lower lobe rays): All branched rays in the lower half of the fin. Caudal fin-ray data where indicated in descriptions, include in order: upper simple rays, upper lobe branched rays, lower lobe branched rays, and lower simple rays, equaling total caudal rays. 8 U.S. NATIONAL MUSEUM BULLETIN 282 SENSORY CANAL SYSTEM.—Variation was found in the number of pores in the preoperculomandibular canal and in the region of the anterior naris. Counts of both sides are tabulated individually. The internasal pores are the openings in the area between the anterior naris and the nasal barbel. If the count is given as one, the infraorbital and supraorbital canals are interpreted as fused anteriorly (pl. 1, fig. 1); if two, the canals are apparently not connected in this region (pl. 1, fig. 2). The variant count of three indicates the unfused condi- tion and an extra pore mediad to the supraorbital canal. If the most anterior preoperculomandibular pore is common to both sides, it is included in both counts (as in Pylodictis). BRANCHIOSTEGAL RAYS.—Counts of both sides, including all rudi- ments, are tabulated individually. Nearly all are from cleared and stained specimens; otherwise the rays usually cannot be counted accurately. GILL RAKERS.—Counts are of the total number, including rudi- ments, on each anterior gill arch. Ranges given in the descriptions of Noturus are based on five or more specimens. Riss.—Data, from both sides, were taken only from cleared and stained specimens. All rudiments are included. VERTEBRAE.—Tabulations do not include elements in the vertebral complex (of which there are five fused vertebrae, see p. 31); all other vertebrae, including one terminal vertebra with attached hypurals, were counted. Caudal vertebrae are those beginning with the first hemal arch with a single spine and include the terminal vertebra. Precaudal vertebrae are those anterior to the first vertebra bearing a hemal spine (excepting elements in the vertebral complex). Vertebrae anterior to the origin of the anal fin are the precaudal vertebrae plus those with hemal spines preceding the first pterygiophore of the anal fin. Hypurats.—The hypurals are enumerated from below upward. Thus, hypurals 2-3 refers to the second and third hypurals as counted from below. They are frequently fused along their adjoining surfaces, except a broad gap exists between the lower three and the upper three or four. Hypural 1 has a hemal arch and has been termed a (the last) hemal spine in several recent papers. Measurements The measurements given are mostly self explanatory, and are described, with the exception of the points noted below, in Bailey and Taylor (1950, p. 33); most are in Hubbs and Lagler (1947, pp. 13-15). The lengths of dorsal and pectoral spines, barbels, and fins, and the depth of the head become relatively lessened with age; other parts seem to remain quite constant. Ranges of eye size, in the descrip- tions of Noturus, are based on five or more specimens. CATFISH GENUS NOTURUS RAFINESQUE 9 ADIPOSE NOTCH OR POSTERIOR END OF ADIPOSE FIN.—This point is variable. If the adipose fin forms a free flap at its posterior end, measurement is to the tip of the flap; if only a notch exists near the junction of the adipose and caudal fins, measurement is to the lowest point in the concavity; if the fin is separated from the caudal fin, the measurement is to the rear end of the adipose tissue. CAUDAL PEDUNCLE DEPTH.—The least depth of the caudal peduncle below the adipose fin. CAUDAL PEDUNCLE LENGTH.—The measurement is from the base of the last anal ray to the midbase of the caudal fin. Dorsau orn1GIn.—The dorsal origin is regarded as the apex of the notch in the imbedded, plate-like spine preceding the dorsal spine. DoRSAL SPINE LENGTH.—Measurement is from the dorsal origin to the tip of the long spine. PECTORAL SPINE TO HUMERAL PROCESS TIP.—The distance is from the anterior base of the erected pectoral spine to the free end of the posterior process (spine) of the cleithrum (humeral process). PREDORSAL LENGTH.—This length is from the dorsal origin to the tip of the snout. SNOUT TO ANAL FIN.—Measurement is from the tip of the snout to the origin of the anal fin. SNoUT TO PELVIC FIN.—The distance is from the tip of the snout to the insertion of the outer pelvic ray. The Family Ictaluridae Members of the family Ictaluridae are found in the fresh waters of eastern North America, from the Hudson Bay and Saint Lawrence drainages of Canada southward through most Atlantic Ocean tribu- taries to the Usumacinta River, Guatemala; in Mexico, they occur in some of the streams flowing into the Pacific Ocean. Several of the economically important species have been introduced into waters in temperate countries throughout the world. Although China was once included in the range of the family, no recent worker on the ichthyological fauna of China has reported any species referable to the Ictaluridae. These old reports probably stem from erroneous labels or misidentifications. The Siluriformes or Nematognathi, to which the Ictaluridae be- long, is characterized by the absence of parietal and opisthotic bones as well as the subopercle (unless the large, broad upper branchiostegal ray represents the subopercle). In most catfishes, the maxilla is greatly reduced; a vestige of this element remains as a small, rod-like structure that projects into the base of the maxillary barbel in the Ictaluridae. 10 U.S. NATIONAL MUSEUM BULLETIN 282 A synopsis of certain characters of the Ictaluridae follows: body lacking scales but covered with thick skin; form moderately elongate; branchiostegal membranes forming a fold across the isthmus; dorsal fin anterior, with a hard spine (except Prietella) and usually with six soft, segmented rays; adipose fin present and of variable length; anterior and posterior nares relatively far apart; jaws toothed except in Trogloglanis; palate toothless; villiform teeth, similar to those of the jaws, in small patches on upper and lower rear pharyngeal arches; upper pharyngeal tooth patch obovate; lower pharyngeal arch S- shaped and bearing a diamond-shaped patch of teeth; palatine rod- shaped; hyomandibular heavy and with a broad head that fits into a groove in the pterotic and sphenotic; post-temporal present, absent, or vestigial; pectoral arch with two ossified radials (actinosts; Starks, 1930, p. 178), which are either separate or fused; entopterygoid (Kindred, 1919, p. 92, pl. 6, fig. 16) or pterygoid (Regan, 1911, p. 562) greatly reduced or absent; ectopterygoid (mesopterygoid of Regan, 1911, p. 562; Starks, 1926, p. 186) variously reduced, below and partially medial to the palatine (in Noturus much reduced in relative size); anterior vertebral complex containing five fused ver- tebral elements; jaws nonprotractile; eye lacking ossicles; gill rakers present anteriorly on all five arches and posteriorly on the third and fourth arches; rear faces of first and second arches smooth, and the fifth bound posteriorly by a membrane; the four pairs of barbels, include two which are mental, one is maxillary, and one is associated with the posterior nares. The pelvic fin in ictalurids has eight, nine, or sometimes ten rays. The pelvic rays are attached to the pelvic girdle; there are no free radials. The pectoral fin possesses a hard spine, as does the dorsal fin, but the number of soft, segmented rays (5 to 12) varies with the species or species group. The anatomy of Amiurus [=Ictalurus] has been described by Macallum (1884), McKenzie (1884), McMurrich (1884), and R. R. Wright (1884a-c). Kindred (1919) described the developing skull and Starks (1926, pp. 185-186) reviewed its ethmoid region. Ex- amination of cleared and stained specimens of representative icta- lurids indicates that the ossified structures are much alike in the species of the family and are essentially as described by Kindred and Mc- Murrich. The differences observed in this study are regarded as primarily of generic or specific significance. Regan (1911) listed some of the major ones, indicating that Noturus differs from other icta- lurids in the absence of the post-temporal and in the shortened lower limb of the cleithrum. These structures are reduced in Noturus and Prietella, and it is probable that they may also be reduced in the other blind species. CATFISH GENUS NOTURUS RAFINESQUE Lt An air bladder is present in all ictalurids except Satan and Troglo- glanis. It is attached to the vertebral complex by connective tissue but is never encased in bone. The testes of Noturus and Pylodictis are grossly similar in shape to those of Jctalurus, as described by Sneed and Clemens (1963). The larger species of Jctalurus and Pylodictis are of considerable economic importance, both in sport and commercial fisheries. Small catfishes are sometimes used for bait, presumably because of their relative hardiness. The young of Jctalurus and members of the genus Noturus are generally reputed to be venomous (Reed, 1907 and 1924b), introduc- ing a poison with their sharp spines. The nature and source of the presumed poison should be carefully investigated. A gland, reputedly the poison gland, is situated just beneath the skin, immediately below the humeral process (posterior extension of the cleithrum) and opens directly to the surface through a large pore. It contains an amber colored, jelly-like material. The pore is reduced in size or closes with age in the larger ictalurids but remains open in Noturus. It is remote from the erect pectoral spine but the spine may closely approximate the pore when folded against the body. Flaky material surrounding the spines and imbedded in their grooves should also be investigated for possible toxicity. Fossil representatives of Jctalurus are known from various Tertiary deposits—at least the Lower Pliocene (Hubbs and Hibbard, 1951)— indicating that the genus has had a long history, and Pylodictis has been recorded from the Pleistocene (Uyeno and Miller, 1962). Although fossils of other groups in the family are yet to be found or recorded, it is presumed that their history has been lengthy. The members of this family are most active at night. The madtoms and stonecats are especially nocturnal in habits; they are seldom seen and infrequently captured by ordinary collecting methods in the day- time. At night they may literally swarm over the area where they appeared to be uncommon in the daytime. These observations point to the cryptic habits of many of the species and to their occasional great abundance in favored habitats. Hooper (1949 and 1951) has indicated something of the abundance of one species in Demming Lake, Minnesota. He estimated at least 8,600 individuals (88 pounds) of Noturus gyrinus in a lake of 12.5 acres. The Sensory Canal System PLATE 1 The sensory canal system provides important indications of inter- relationships in fishes. A short article by Collinge (1895) compared the canals in certain Old World catfishes; Pollard (1892) gave comparative 298-943 O—69 2 12 U.S. NATIONAL MUSEUM BULLETIN 282 information on several forms; and brief descriptions of the system have appeared in anatomical discussions of certain species. In this study of the Ictaluridae, reference has been to R. R. Wright (1884b), Herrick (1901, pp. 230-231), and Kindred (1919). On certain details of struc- ture their findings are not in agreement with my data. In this family the canals are simple tubes. At intervals along their course these give off short lateral branches that extend to the surface and open through pores. The canal system is either imbedded in the large bones or retains cylinders or ringlets of bone (see also McMurrich, 1884, p. 279) for support. The ossified ringlets probably continue posteriorly to the end of the functional lateral line in the Ictaluridae. Stained specimens show that the infraorbital canal and short sections of the canals extending between the large bones are enclosed in these ringlets. Those stained without removing the skin exhibit the ringlets along the side of the body, where they extend to the end of the lateral line. The canal system consists of the preoperculomandibular canal, the infraorbital canal, the supraorbital canal, the lateral cephalic canal, and the lateral body (line) canal (pl. 1). In disagreement with R. R. Wright (1884b, pp. 264-265), I count more pores in the pre- operculomandibular canal and find that this tube is not isolated but connects with the lateral cephalic canal; however, in Prietella only a short section connects with the lateral canal as the preoperculo- mandibular canal is interrupted. Also I find no evidence of an occipital commissure. The preoperculomandibular system traverses the lower jaw and preopercular bones, usually connecting with the lateral cephalic canal by one or more short ossified tubules or ringlets. Along its course it opens to the outside from 10, 11, or 12 pores, the number depending upon the species (tables 1 and 22). Of these pores, the first two to five open below the lower jaw in front of the mental barbels. In Pylodictis and Satan the first pore is median and a common outlet for the preoperculomandibular canals of both sides of the head. In other genera, the distance between the anterior pores is considerably greater, varying from a moderate separation in Noturus, Prietella, Trogloglanis, and in certain species of IJctalurus to a rather wide separation in other forms of Jctalurus. Variations in number of pores in the preoperculomandibular canal seem to be of considerable significance, but the relative spacing is in part an expression of the development of the encasing dermal bones. The canal in Pylodictis and Satan has five pores (counting the median pore) in front of the posterior base of the outer mental barbel and seven posterior to that element. In Noturus, Prietella, and Trogloglanis there are usually only four anterior pores, but the posterior portion CATFISH GENUS NOTURUS RAFINESQUE 13 of the system varies with the species, usually having seven openings; there are six in each of four species. A further lateral shift of the anterior pores takes place in Jctalurus so that anterior to the outer mental barbel there are at most only four pores, in some forms three or even two. This shift leaves six to eight pores posterior to the outer mental barbel, as there are typically ten outlets from the canal (not eight or nine as reported in the literature). Posterior to the outer mental barbel, six to eight pores from the preoperculomandibular system open below the lower jaw and over the opercle. The most posterior of these emerges at about the middle of the upper edge of the opercle and the next to the last typically opens over the midopercle, directly anterior to the pectoral fin. In Prietella, however, the last two pores are nearer the upper edge of the opercle, while the ninth pore is approximately in front of the pectoral spine. In this species the canal is interrupted between the most pos- terior pores. These adjacent pores probably result from the division of one pore; consequently, although Prietella now normally has eleven pores, it likely has been derived from a ten-pored ancestor. The infraorbital canal is rather simple. Anteriorly, it opens by a pore (L1) posterior to or mediad of the anterior naris. The next pore (L2) opens lateral to these and above the maxilla. Posteriorly there are typically five (six in Prietella and one of two specimens of Trogloglanis has six) other openings along its course before it joins the supraorbital canal to form the lateral canal. The entire canal in Prietella typically has eight pores, the last two being isolated by interruption of the canal from the anterior section. Here again, prob- able disruption of the canal at a pore resulted in an extra pore. An- teriorly the canal passes through the first suborbital, also called the lacrymal bone, hence the symbolic terminology used for the first two pores. Posteriorly, the canal is supported by simple tubes of bone and all other pores arise from centers between these. In passing about the eye posteriorly, the canal swings almost directly upward in Ictalurus, Trogloglanis, Prietella, and Noturus; in Pylodictis and Satan the canal passes diagonally backward. It joins the supraorbital and lateral canals in the frontal. The supraorbital and lateral cephalic canals form a simple straight tube which contains five pores. There is no connection between the two sides of the head. Of the pores, the first three exit from the tube traversing the nasal bone and the posterior two from the frontal. In Pylodictis, the anteriormost of the nasal pores (N1) is relatively small and far ahead of the anterior naris, near the upper lip, and the second (N2) or midnasal pore opens mediad or anterior to that structure. The anterior nasal pore (N1) in Satan is about midway between the edge of the upper lip and the naris; in the other genera of the family 14 U.S. NATIONAL MUSEUM BULLETIN 282 Ictaluridae, pore N1 is close to the anterior naris and the midnasal (N2) is in a slightly posteromedial position. Connection of this system anteriorly with the infraorbital system is infrequent. However, in three species of Noturus (lachneri, exilis, and miurus) and in Satan eurystomus the connection is typical. In these, the midnasal (N2) and ‘‘mediad lacrymal”’ (Li1) pores fuse to form an “‘internasal pore” (pl. 1, fig. 1); i.e., topographically the pores involved in fusion lie between the anterior and posterior nares. Posteriorly the lateral cephalic canal connects with the lateral body canal after passing through the upper end of the supracleithrum. The posterior extent of the lateral line varies considerably between species. It ends just behind the head in Noturus, beneath the dorsal fin in some Jctalurus, and extends to the base of the caudal fin in other Ictalurus. Relationships of the Genera of the Ictaluridae During the course of the revision of Noturus an examination was made of most of the recognized species of Ictaluridae, consisting of Pylodictis olivaris (Rafinesque), Satan eurystomus Hubbs and Bailey, Prietella phreatophila Carranza, Trogloglanis pattersoni EKigenmann, as well as over a dozen species of Jctalurus and the members of the genus Noturus described below. Most were represented by ample material, but only three specimens of Satan eurystomus and two of Trogloglanis pattersoni were available, having been described previously by Hubbs and Bailey (1947) and Suttkus (1961). Skeletonized preparations of Pylodictis, Prietella, most species of Noturus, and over half the species of Ictalurus were studied. The family comprises three relatively wide-ranging genera, and three genera that are known only from localized underground waters. The genera from underground waters show similarities to one each of the wide ranging genera: Prietella Carranza to Noturus Rafinesque, Satan Hubbs and Bailey to Pylodictis Rafinesque, and Trogloglanis EKigenmann to Jctalurus Rafinesque. The six genera can be arranged conveniently into three major groups, which are here termed the Noturus group, the Pylodictis group, and the IJctalurus group. No taxonomic name is assigned to them, however, because the suggested relationship requires further study. A comparison of the genera and indication of their relationship is presented in table 1. Certain characters (not listed in table 1), long used as the bases for genera, appear to have value chiefly in delimitation of species. Of these, the shape of the group of teeth on the premaxillary bone and the shape of the caudal fin have been extensively used. Trogloglanis pattersoni has no teeth on the lower jaw or the premaxilla. The premaxillary teeth in the other genera are arranged in a transverse CATFISH GENUS NOTURUS RAFINESQUE 15 rectangular patch. Its width and lateral edges are rather variable, and various posterior extensions of the patch may invade the rear surface of the premaxilla which is variously extended backward. The variable posterior extension may be short and scarcely evident, mod- erate and from the posterior edge of the premaxilla (Ictalurus balsanus), long, narrow, and from near the posterior corner (Noturus flavus; pl. 2, fig. 3), or broad, long, and from the posterior edge (Pylodictis olivaris). The shape of the caudal fin varies from species to species. Variations in the posterior fin margin are: rounded to deeply forked in Jctalurus; slightly emarginate in Prietella, Trogloglanis, and Satan; truncate or rounded to pointed in Noturus; and emarginate to rounded in Pylo- dictis. The degree of development and shape of serrae on the spines varies greatly throughout the species of catfishes. The skeleton is well ossified and the bones are heavy in species that attain a large or rela- tively large size, particularly Pylodictis olivaris, Noturus flavus, and the forms of Jctalurus. The bones are somewhat thinner in small species. The ectopterygoid, broad in both Jctalurus and Pylodictis, is some- what reduced in breadth in Prietella and is narrow in the species of Noturus. Similarly, the post-temporal bone is present in Jctalurus and Pylodictis, but is absent or vestigial in both Prietella and Noturus. Since the bones of both Satan eurystomus and Trogloglanis pattersoni are thin, both of these elements (not examined) could be expected to be reduced. The width of the opercle varies considerably; it is broad in most forms, but narrow in Noturus. The bone is relatively little nar- rowed in Prietella. One epural is typically present in most species, but occasional speci- mens have two. Both Pylodictis and Ictalurus appear to have seven hypurals, with only infrequent reduction to six. Of these, three (1-3) support the lower lobe of the caudal fin and four (4-7) support the upper lobe. Three species of Noturus, subgenus Schilbeodes, possess a modal number of seven (1-3 + 4-7) hypurals. All others normally have six (1-3 + 4-6). Fusion of the hypurals in the specimens of Prietella studied prevents their enumeration. The variation in branchiostegal rays in the family is 8 to 13. Two branchiostegals are typically associated with the epihyal; a third is sometimes at the junction of the epihyal and ceratohyal so that it could be included in the count for the epihyal; all other branchiostegals are associated with the ceratohyal. Pylodictis is the only species that normally has twelve branchiostegals. Two specimens of Satan have eleven, but the type-specimen has ten. In Noturus and Prietella, the variation is eight to twelve with modes from nine to eleven. The limited data from Jctalurus and Trogloglanis indicate a range of eight to eleven branchiostegal rays with modes eight to ten. 16 U.S. NATIONAL MUSEUM BULLETIN 282 The anterior position of the first nasal pore (N1) is unique in Pylo- dictis, but its anterior position is approached in Satan. The number of sensory canal pores anterior to the outer mental barbel (see p.12) appears to be of some evolutionary significance. Another kind of pore, “‘the venom pore,’’ located immediately below the posterior process of the cleithrum is large in most species of Noturus and in young of Jctalurus and Pylodictis; in Prietella and large individuals of the previous genera it is greatly restricted in size, but it appears to be absent in Satan and Trogloglams. IcraLURUS GRouP.—The species in the genus Jctalurus are a cluster of divergent forms having the following characters in common: ten preoperculomandibular pores, eight pelvic rays, and an extremely shortened, high adipose fin which is free from the back posteriorly and remote from the caudal fin. The bullheads, Ameiurus [=Amiurus] are regarded as members of the genus Jctalurus because there is gradation without a significant break from them through several intermediates to the most divergent Jctalurus such as J. punctatus. In young Jctalurus the body may be light grayish blue with black vertical fin margins, mottled, spotted with light or with dark, or brown or blackish above and lighter below. With age and development of breeding characteris- tics these fish assume a color that is almost uniformly dark, especially on the dorsal and lateral surfaces. In some forms the skull is greatly arched, especially in the young; in a few it is flattened or depressed, and in others it is variously intermediate in shape. With the develop- ment of the breeding characters, the skulls broaden and become more depressed. Thus the high arched heads are transformed into shallowly rounded structures and the flattened heads are even broader and more depressed by the change in bone relationships as well as the prolifera- tion of muscle masses laterally and occipitally. Consequently, a breed- ing adult may show little resemblance to its young but instead may resemble the young of another species. The shape of the caudal fin also varies or grades from rounded posteriorly through emarginate, notched, shallowly forked, and deeply forked among the species. The depth of the fork is extensively re- duced with age and breeding condition so that species with deeply forked tails as young have only a deeply notched or very shallowly forked tail when older, again often resembling the young of another species. Only two of the species, Ictalurus furcatus and I. punctatus, have a continuous bony bridge from the supraoccipital to a small bone extending forward from the base of the dorsal fin. The supra- occipital (“occipital’’) process in others is variously developed, from an elongated shelf that is widely separated from the dorsal fin and lacks connecting elements, to a near union with the process from the dorsal fin base. Two of the species, Jctalurus furcatus and I. balsanus, CATFISH GENUS NOTURUS RAFINESQUE ne have an additional posterior chamber to the air bladder; the latter has a moderate backward extension of the patch of teeth on the premaxilla. Their relationship to each other and to the genus Jctalurus should receive further study. Tentatively Jstlarius balsanus Jordan and Snyder is included in IJctalurus because of its resemblance to that group and its probable affinity with Jctalurus furcatus. The two syntypes (ANSP 22082-3) of Gronias nigrolabris Cope (1864) were examined in the course of this study. They do not repre- sent a distinct genus, but are specimens of Jctalurus nebulosus (Le- Sueur) as maintained by Hubbs and Bailey (1947, p. 12). Fowler (1915a, p. 208) regarded them as a distinct species of Ameiurus [=Jctalurus]. Both have eight rays in each pelvic fin; the anal fins have 19 and 20 rays; the pectoral spines are long and serrated posteri- orly. Contrary to report the eyes are present, but are asymmetri- cally developed—undoubtedly a teratological condition. The relationship of Trogloglanis patterson Eigenmann to the Ictalurus group is suggested by several characters. In some respects it resembles the species of Jctalurus which have high arched skulls, a relatively shortened head, a long occipital process, and many gill rakers (table 1). Also, like those species, the mouth is constricted and inferior, but lacks teeth, the dorsal fin is slightly pointed, and the lateral line is long. Otherwise, the presence of eight pelvic rays (the specimen TU 10808 appears to have nine—perhaps an excep- tion—rather than the eight reported) is certainly suggestive of re- lationship with Jctalurus. The posterior flap on the anterior naris - is extreme but an elevation of the posterior rim of the naris is evident in several Jctalurus, especially Ictalurus punctatus in which the con- dition approaches Trogloglanis. Eigenmann (1919) thought Trogloglanis related to the genus Schilbeodes [=Noturus] and Hubbs and Bailey (1947) regarded it as more closely related to Ameiurus [=Ictalurus|. Although Troglo- glanis has undoubtedly diverged considerably from its ancestral ictalurid stock, it has few resemblances to Noturus or to Pylodictis, and it is quite distinct from the other species of blind catfishes. Eigenmann was undoubtedly impressed with the elongated adipose fin (which was once thought to be unique to Noturus), but the more recently discovered blind species also have a similar fin. Indeed, the morphology of the specimens does not suggest derivation from any of these stocks, or from the bullheads (Ameiurus auct) as postu- lated by Hubbs and Bailey (1947). The few similarities to Jctalurus, sensu strictu, which are listed above suggest derivation from a form not greatly unlike J. punctatus. Noturus Group.—Prietella phreatophila Carranza is similar in many characters to the genus Noturus. Several features of the two 18 U.S. NATIONAL MUSEUM BULLETIN 282 genera align them together and separate them from both the Jctalurus and Pylodictis groups although they appear to occupy an intermediate phylogenetic position (table 1). Among the many characters in common are the reduced occipital process, the long adipose fin, the general similarity in head and body shape, divergence from a normal mode of 15 branched caudal rays, a shortened air bladder, an obsolete post-temporal bone, a slightly reduced ectopterygoid, rather widely separated anterior ends of the preoperculomandibular canals, similarly shaped infraorbital canals, a short lateral line, and relatively few vertebrae. Yet they seem to be distinct units. Prietella differs from Noturus especially in the primitiveness or lack of ossification of the first dorsal ray, the degeneration of ossified tissue at the anterior base of the dorsal fin, especially the absence of the plate-like spine, and in features resulting from subterranean life: depigmentation, eyelessness, and interruption of the infraorbital and preoperculoman- dibular canals. The anterior fontanelle is a small round opening in contrast to the elongate slit or opening in all other ictalurids (see Suttkus, 1961, fig. 5). The short, blunt, nonserrated, ungrooved condition of the pectoral spine and eight pelvic rays are duplicated or approached in few kinds - of Noturus. Although Pylodictis and Ictalurus have a range that over- laps their apparent underground relatives, which occur in central Texas, Noturus and Prietella are geographically isolated; Prietella is known only from one locality in the Rio Grande drainage of Mexico, whereas Noturus ranges no farther southwestward than the Nueces River, Texas. They are regarded as distinct genera, but are thought to be related; there is no suggestion that Prietella is related to any of the other recognized groups of ictalurids. Indeed, the superficial similarity of Prietella and certain forms in the subgenus Schilbeodes of Noturus is striking. Due to the similarity of Noturus and Prietella, the following descrip- tion of Prietella phreatophila Carranza (based on thirteen specimens of which three were cleared and stained) is included (see also Carranza, 1954): standard length 34.0 to 52.8 mm.; anal rays 12 (in 1), 13 (1), 14 (6), and 15 (5); pelvic rays 8 (26 sides) ; soft pectoral rays 8 (7 sides) or 9 (19); dorsal rays 6 (2) or 7 (11), see p. 19; internasal pores 2 (24 sides); preoperculomandibular pores 10 (1), 11, (21), and 12 (2); gill rakers 10 (2 sides), 11 (4), 12 (7), 13 (10), and 14 (1); ribs 5 (5 sides) or 6 (1); precaudal vertebrae 7 (1), 8 (1), or 9 (1); caudal vertebrae 21 (1), 23 (1), or 24 (1); vertebrae anterior to anal fin origin 11 (1) or 12 (2); total vertebrae 29 (1), 31 (1), or 32 (1); branchiostegal rays 8 (2 sides) or 9(4). Caudal rays: upper simple rays 12 (1), 13 (3), or 14 (9); upper lobe branched rays 6 (13); lower lobe branched rays 6 (1) or 7 (12); lower simple rays 12 (1), 13 (0), 14 (7), 15 (4), or 16 (1); lower-half CATFISH GENUS NOTURUS RAFINESQUE 19 rays 19 (1), 20 (0), 21 (8), 22 (3), or 23 (1); total rays 37 (1), 38 (0), 39 (0), 40 (2), 41 (7), 42 (2), and 48 (1). Other characteristics are (see also table 1): lower jaw included, almost subterminal; infraorbital and preoperculomandibular canals interrupted, apparently resulting in an extra pore in each; no dorsal spine, instead first dorsal ray segmented and flexible distally but without distinct segments basally; basal half better ossified than in other dorsal rays and the base expanded laterally; no imbedded plate- like spine at anterior end of dorsal fin but the pterygiophore that supports the plate-like spine in other species is present and with distal end broad; cleithrum relatively narrow; posterior process of cleithrum curved inward; pectoral spine present, short, relatively flexible, smooth along all edges, without obvious grooves, its end blunt and continuous with soft tissue which is segmented; one epural; number of hypurals unknown, almost solidly fused, either as a broad plate, or as two plates separated near the middle of the fin base; caudal fin truncate or slightly emarginate; lateral line short, ending just back of head; premaxillary tooth patch very wide and short, without visible posterior extensions; air bladder normal, short, no longer than wide, about as in Noturus. Pyuopictis Group.—Pylodictis olivaris (Rafinesque) and Satan eurystomus Hubbs and Bailey are morphologically similar and closely related, but constitute separate genera. Satan differs from Pylodictis chiefly in the absence of eyes, the obsolescence of the air bladder, the lack of pigmentation, absence of backward extensions of the premaxillary tooth band, the excessive elongation of the adipose fin, and in divergences of the sensory canal system, especially the enlarged pores and the single internasal pore. The twelve preoperculomandibular pores in Pylodictis are approached in Satan. In Satan the type-speci- men has 12 pores on each side counting the anterior median pore. As noted by Suttkus (1961, p. 57) two additional specimens appear to have 11 pores on each side. However, the preoperculomandibular canal is not continuous with the lateral cephalic canal posteriorly, possibly resulting in an extra pore (not counted by Suttkus) on each side, opening from the short canal connecting with the lateral cephalic canal. Thus the basic number of pores may be either 11 or 12. Of these pores, five are anterior to the outer mental barbel. Also the body form, the rather greatly depressed head, the wide mouth, the backward swing of the infraorbital canal, the terminal or subterminal lower jaw, the proximity of the anterior ends of the preoperculomandibular canals, and the presence of nine or more rays in each pelvic fin (the specimen of Satan, USNM 195830, has 9 rays instead of 10 as reported) suggest their phylogenetic relationship (compare, table 1), in agree- ment with Hubbs and Bailey (1947). As noted by Suttkus (1961) 20 U.S. NATIONAL MUSEUM BULLETIN 282 the dermethmoid (supraethmoid, Kindred, 1919) is broadly forked in both Satan and Pylodictis further suggesting close relationship. Uyeno and Miller (1962) indicated other osteological differences of Pylodictis from both Ictalurus and Noturus. Genus Noturus Rafinesque Noturus Rafinesque, 1818a, p. 41 (original description; type-species, Noturus flavus Rafinesque, by monotypy); 1820a, p. 48 (comparison). Pimelodon Vaillant—LeSueur, 1819, p. 155 (probably vernacular and used with the French livrée, thus nonbinominal).—Vaillant, 1896a, p. 28; 1896b, p. 14, pl. 24 (name first available, from LeSueur; type-species, Pimelodon insignarius Vaillant [= Noturus insignis (Richardson)], by monotypy). Pimelode.—LeSueur, 1820, p. 44 (vernacular; used with the French livrée). Schilbeodes Bleeker, 1858, pp. 36, 249, 258 (original description; type-species, Silurus gyrinus Mitchill, by monotypy).—Jordan and Gilbert, 1877b, p. 93 (misspelled as Schilbeoides). Rabida Jordan and Evermann, 1896a, pp. 144-146 (original description in key; furiosus, in parentheses, presumably intended as type-species).—Jordan, 1920, pp. 473, 566 (misspelled as Rabidus; type-species designated: Noturus furtosus Jordan and Meek); 1923, p. 147 (misspelled as Rabidus). The name Pimelodon predates the name Schilbeodes if Pimelodon can be credited to LeSueur as a scientific name. After his description of six species of the genus Pimelodus LeSueur (1819, p. 155) inserted a description, perhaps as an afterthought, of another beginning as follows: “J’indique ici sous le nom de Pimelodon livrée....” In an abstract of this paper LeSueur (1820, p. 44) italicized both parts of the name, shortening Pimelodon to the well-known vernacular ‘‘Pimelode,”’ a name commonly used for Pimelodus; he did not mention Pimelodon in the abstract. The names of the six species of Pimelodus were con- sistently italicized in both papers as was the French name livrée. Pimelodon and most, but not all, of the other vernacular names were not italicized. It is thus apparent that LeSueur coined the word Pimelodon from Pimelodus and intended it only as a temporary vernacular name. Subsequent writers did not mention Pimelodon nor list it as a genus name until Vaillant (1896a,b) published a synonymy and LeSueur’s drawing of Noturus insignis under the name Pimelodon imsignarius. Because it appears that Pimelodon was originally pub- ished as a vernacular name by LeSueur, it is here accepted as first available nomenclatorially in 1896. The meaning and origin of the word Pimelodon is otherwise unknown. Gill (1876, p. 410) used the name Noturi to include the genus Noturus. Noturus is compared with the other genera of the family Ictaluridae in table 1. The characters of the genus are: eyes present; body and fins variously pigmented, never entirely unpigmented; dorsal and pectoral fins always with a hard spine at anterior end; adipose fin CATFISH GENUS NOTURUS RAFINESQUE zt long, its posterior end adnate to the back, sometimes forming a short free flap, and variously connected to, or separated from the upper caudal fin by at most a small notch; caudal fin truncate, rounded, or pointed behind; procurrent caudal rays covered by a thin mem- brane; number of branched (and simple) caudal rays extremely variable; lateral line short; skull arched to greatly depressed; skeleton rather poorly ossified, bones thin; post-temporal bone absent, or reduced and nonfunctional; ectopterygoid narrow, reduced in size; opercle narrow, much longer than broad; teeth present on lower jaw and premaxilla; occipital process very short and narrow, projecting little beyond skull; ossified pectoral radials fused or unfused; pelvic rays usually 8 or 9; gill rakers on the first arch 3 to 10; preoperculo- mandibular pores usually 10 or 11; branchiostegal rays 8 to 12, usually 8 to 11; vertebrae variable, 30 to 42; soft dorsal rays 5 or usually 6; soft pectoral rays 5 to 11, usually 6 to 10; sensory canal system continuous on each side, not interrupted; preoperculomandibular canal with four pores anterior to outer mental barbel, its anterior end usually widely separated from the canal of the opposite side; infraorbital canal bending upward and slightly anteriorly behind eye (pl. 1, fig. 1), with seven pores; first pore of supraorbital canal in front of and adjacent to anterior naris; air bladder always present, mod- erately long to little longer than broad, never with a posterior chamber; epurals typically one, sometimes two; hypurals normally six (1-3 below + 4-6 above), but modally seven (1-3 below + 4-7 above) in three species. Other skull and skeletal characters are much like those of Ictalurus which has been described by Kindred (1919) and MeMurrich (1884). The principal reviews of the genus are by Jordan (1877d) containing a moderately good summary of the recognized forms and characters, which is most useful in identifying the newly described species; Swain and Kalb (1883), giving some new information but introducing confusion; and Jordan and Evermann (1896a), employing a nomen- clature that for the most part still holds today. Characters and distributions were presented for some species by Hubbs and Raney (1944), but these authors made unnecessary changes in nomenclature. Information on the natural history and morphology of the species of Noturus is sketchy. Some of the principal, yet often brief, contri- butions deal with: food habits (Curd, 1960; Forbes, 1880b, 1888a, 1888b; Langlois, 1954; Pearse, 1915, 1918, 1921); spines, poison, and poison glands (Birkhead, 1967; Reed, 1900, 1907, 1924a, 1924b); parasites (Anthony, 1963; Bangham, 1941a, 1941b, 1946; Bangham and Hunter, 1939; Bangham and Venard, 1942; Fischthal, 1947, 1950, 1953, 1956; Larson, 1966; Van Cleave and Mueller, 1934); reproduction (R. M. Bailey, 1938; Fowler, 1917b; Greeley, 1929, 22 U.S. NATIONAL MUSEUM BULLETIN 282 1934; Hankinson, 1908; Langlois, 1954); distribution (Hubbs and Lagler, 1958; Hubbs and Raney, 1944; Jordan and Evermann, 1896a; Trautman, 1959); age and growth (Hooper, 1949); development (Fish, 1932; Ryder, 1886, 1887; Sumner, 1899); morphology (Chrani- lov, 1929; Herrick, 1901; Regan, 1911; Suttkus, 1961). The ecology of most of the species has been briefly discussed by many authors. Several aspects of the life history of Noturus insignis (Richardson) were described by Clugston and Cooper (1960) and a population of Noturus funebris Gilbert and Swain was studied by Thomerson (1966). As indicated by Hubbs and Raney (1944), I feel that there is no very sharp break between the subgenus Schilbeodes and the species forming Rabida. There are probably even fewer important differences between the subgenera Schilbeodes and Noturus. Thus, the species formerly assigned to these three “genera” are combined as Noturus. However, certain resemblances among the species make it practicable to retain three subgeneric groups. Noturus flavus, the largest and one of the most widespread species, is the type-species of Noturus, long recognized as a monotypic genus. N. flavus differs from all other species in the unfused condition of the pectoral radials, but some forms, chiefly N. stigmosus, vary in that direction. The pattern of teeth on the premaxilla, long employed as a generic character, is regarded as of little more than specific import- ance; the shape of the band varies considerably among the species of Noturus and simply reaches an extreme development in flavus; it does not conform to the shape of the premaxilla (pl. 2). N. flavus is not consistently distinguishable from other species in the numbers of pectoral and pelvic rays or of vertebrae. Superficially, N. flavus is most similar to N. gilberti, and their true relationship may be rather close. The present arrangement of subgenera differs from past classifications chiefly in that Noturus is monotypic, Rabida includes only the mottled forms, a very distinctive group of species, and Schilbeodes contains the remaining, more somberly colored species. Three names that have been used to indicate members of this genus were not accompanied by descriptions. They are: Noturus lacanthus.—Jordan (1877c, p. 50). The name is now un- identifiable. Jordan included it in a list of fishes of the Ohio Valley that were not noticed by Rafinesque. Noturus latifrons.—Jordan (1835, p. 802). Gilbert and Swain probably intended to use this name for Noturus eleutherus (see p. 165). Schilbeodes punctatus——Mitchell (1904, p. 154). According to Mitchell (1904, p. 405) this is a lapsus for Schilbeodes gyrinus (Mitchell). CATFISH GENUS NOTURUS RAFINESQUE 23 Key to the Subgenera of Noturus 1. Pectoral spine usually curved, scimitar-like, with both anterior and posterior serrae; anterior serrae small, numerous, sometimes barely visible; posterior serrae larger and their tips (except 1 to 3 near spine base) uniformly re- curved toward spine base; color pattern usually of dark blotches or saddles on back, over lighter background; ossified pectoral radials typically fused; premaxillary tooth patch rectangular with posterior corners rounded or obtusely angulate. .. . . . . Subgenus Rabida Pectoral spine nearly straight io eden ponced: without anterior serrae; anterior edge of spine often with recurved hooks or step-like processes; posterior edge of spine smooth or with serrae which are not regularly turned toward spine base; color pattern typically dark, without prominent darker blotches or padiles On WIDOW sues GL syn afra Pens 2. Premaxillary tooth patch rectangular with distinct, ane? aaa narrow, projections from the posterior corners; ossified pectoral radials never fused. Subgenus Noturus Premaxillary tooth patch rectangular, with posterior corners rounded or acute to obtusely angulate; ossified pectoral radials typically fused. Subgenus Schilbeodes History The early naturalists first discovered members of the genus Noturus in the fresh waters of North America about 1817. Within three years, three of the common species, Silurus gyrinus Mitchill, Noturus flavus Rafinesque, and ‘Pimelodon livrée’’? LeSueur [=Noturus insignis (Richardson)] had been recognized and partially described. LeSueur, however, applied a vernacular name to his fish. At this time, also, Rafinesque proposed the generic name Noturus for flavus. During the next 40 years no new species were recognized but two new names were based on the description by LeSueur. There was considerable speculation at this time as to whether the description of Noturus flavus was based on a young or adult specimen since a contin- uous adipose fin was regarded as a character of larvae or young individuals. Bleeker (1858), believing that Mitchill’s Silurus gyrinus lacked an adipose fin proposed the generic name Schilbeodes for this species. He later placed Schilbeodes in the synonymy of Noturus when informed by Gill that all of the North American catfishes have an adipose fin. The common name stonecat, sometimes used for all species of Noturus, stems from Baird (1860, pp. 420-421) who wrote: ‘The genus Noturus, known provincially as Stone-Cat-fish embraces few species, . . .” The species on which this name was based or the locality of origin was not given. Baird had collected Noturus gyrinus, Noturus exilis, and Noturus insignis by this time. Of these, it is most likely that the name was based on the specimens of N. insignis from Carlisle, Pennsylvania, which became the types of Noturus marginatus. 24 U.S. NATIONAL MUSEUM BULLETIN 282 From 1875 through 1891, Nelson described Noturus exilis, and Jordan and his co-workers described most of the remaining known species. The generic names Rabida and Pimelodon were established in 1896. Since then only two valid species have been named. The common name madtom also originated locally. Jordan (1889, p. 353; 1890, pp. 101, 122) found the name to be in use in Virginia for Noturus insignis and possibly for Noturus gilberti. He said (1890, p. 101) of insignis: “Well known . . . here [Luray, Virginia locali- ties] as elsewhere in Virginia, by the appropriate name of Mad-Tom.”’ Zoogeography Noturus belongs to the large fish fauna of eastern North America. There, its species are found in tributaries to the Atlantic Ocean from the Nueces River, Texas, to the Hudson River, New York, and have spread into the Saint Lawrence and Hudson Bay drainages. Repre- sentatives have been introduced into Germany (unsuccessfully), into the Snake River of Idaho and Oregon, and into several New England streams. The discovery of Prietella, an apparent relative of Noturus, in northeastern Mexico suggests a more widespread distribution of the ancestral Noturus, possibly throughout eastern North America into Mexico. Two ecological factors seem to contribute importantly to the present distribution of Noturus: water sufficiently warm for reproduc- tion and a good supply of oxygen. Avoidance of cold water is indi- cated by the few species living in northern regions and by the infrequent occurrence of any species with cold water inhabiting trout. Of these, Noturus insignis is most commonly reported from trout streams— perhaps submarginal trout streams. Other evidence of a requirement for warm water are late season breeding in the north and general disappearance, after one or two years, from and below the area of large impoundments that release cold water. Although one species seems to prefer lowland areas and water with little or no current the remainder appear characteristic of moderate to rapidly flowing streams. In the latter they are most often found on or just below riffles, where there is an abundance of cover for hiding and the water is well oxygenated. Although wet specimens will often survive for several hours in air, deficient oxygen appears to be a critical factor in controlling distribution. This is most noticeable in late summer and fall when species become almost entirely restricted to a riffle where there is deficient oxygen in the adjacent pools. The present centers of greatest number of species of Noturus, and perhaps of their evolution, lie in the upland regions of the east central United States, roughly in a band from Arkansas and Missouri through CATFISH GENUS NOTURUS RAFINESQUE 25 Kentucky and Tennessee to Virginia and North Carolina. All three subgenera have representatives entering portions of the region, and a number of endemics are found there. The greatest number of species, twelve, are included in the fauna of Tennessee, followed by Kentucky with nine and thence by Alabama, Arkansas, Mississippi, and perhaps Missouri, with eight; away from this area, the number of species gradually decreases. Nine species are included in the faunas of the Ohio and Tennessee Rivers, decreasing to six known from the Cumberland, White, Arkansas, Ouachita, and lower Mississippi Rivers. Because of the favorable ecological habitat available, this general region was probably a refuge for several northern species during Pleistocene glaciation as well as retaining its native fauna. Subse- quently several species dispersed northward, some farther than others. The relatively northern Noturus flavus was undoubtedly pushed southward by glaciation but it has been able to move into nearly all the upper tributaries of the Mississippi River; it crossed (when or where is not clear) into the Arkansas drainage and utilized post glacial stream changes to enter the Great Lakes at a number of points. The members of the subgenus Rabida were perhaps more eastern and southern in distribution and consequently unable to move into the Mississippi drainage above the Ohio River. For some reason, perhaps because of the sediment load carried by the Mississippi, they have not entered any upper portion of that system except for a simple crossover of Noturus miurus from the Wabash drainage to the Kas- kaskia drainage in Illinois. This crossover may not have persisted. Otherwise two species of Rabida, that were probably pushed south- ward during glaciation, traversed the Wabash River and entered the lower Great Lakes by way of the outlet of glacial Lake Maumee. Aside from Noturus gyrinus, the species of the subgenus Schilbeodes have entered only the lower portions of the Ohio (except apparent recent entrances by N. insignis) and Missouri River systems and only one (N. ezilis) has moved extensively into the upper Mississippi River drainage. Thus it seems that the major distribution of the species of Schilbeodes was southward and eastward as today and most were not materially affected by glaciation. Noturus insignis, along the Piedmont, was probably restricted southward but it has successfully moved northward again; it has entered the Mississippi and Great Lakes drainages by relatively recent stream changes or human introduction. Noturus gyrinus occupies the lowlands fringing the eastern uplands, ranging in waters varying from strongly acid to alkaline and into slightly brackish water. Because it lives successfully in quiet, standing waters it has been able to enter the Great Lakes drainage at a number 26 U.S. NATIONAL MUSEUM BULLETIN 282 of points and is the only Noturus that has gotten into the Hudson Bay drainage. An indication of a similar, but more southern, distribution is shown by the ranges of the four species of the furtosus species group in the subgenus Rabida. In this group Noturus furtosus is the sole species of the subgenus that is found in Atlantic coast streams; Noturus munitus ranges farther east along the Gulf coast than do others; these combined with the species in the Ohio and Arkansas drainages point to an ancestor that was widely distributed around, but that did not extensively enter, the eastern upland region. Noturus gilberti in Virginia, and Noturus albater in the Ozarks are relicts without apparent close relatives elsewhere. Sexual Dimorphism Breeding ictalurid catfishes may temporarily show marked sexual differences, but young and immature individuals of both sexes are essentially alike in color and gross morphology. Ripe females are heavy with eggs and, if young adults, may lack the excessive devel- opment of head musculature and dark color displayed by nest-guard- ing males. With age, the female’s musculature and color progressively resemble those of the mature male. Spawning males thus are commonly darker or more drab than spawning females. In some forms the external genitalia are useful in distinguishing the sexes, at least at the time of maximum gonad development. Some of these evidences of sexual dimorphism have played an important part in the interpretation of species of catfishes, especially in Ictalurus, and led to identification of two species of Noturus as one. Prominent, permanent, external, sexual differences have not been found in Noturus, and aside from the genitalia, may not exist in the other genera. There is some evidence suggesting slight external sexual differences in all groups. The only reliable way that was found to determine sex in Noturus was by examination of the gonads, as attempts to sort various species by sex on the basis of external morphology and genitalia met with failure. A few years ago several workers accepted the suggestion that gross sexual dimorphism is a common permanent phenomemon in catfishes and is involved in the proper delineation of species. This belief was largely based on statements by Hubbs (1940, pp. 209-211) and Hubbs and Allen (1944, p. 118) and complicated by the frequent observation of a young appearing female, which often more or less resembles the immature young male, spawning with a more fully developed, dark, puffy cheeked male. These developmental changes have confused the nomenclature of several of the species of Ictalurus with excess names CATFISH GENUS NOTURUS RAFINESQUE 27 based on immature and adult stages. Ontogenetic change is minor and has not complicated nomenclature in Noturus. Although Hubbs (1930, p. 432) had once recognized both Noturus eleutherus and N. furiosus from the Ohio Valley, these were later regarded as conspecific, the result of sexual dimorphism (Hubbs and Lagler, 1941, p. 65; Bailey and Taylor, 1950, p. 31). A re-examination has been made of all Ohio drainage and Great Lakes material previously identified as Schi/beodes furtosus or Schil- beodes eleutherus. Many collections from the Ohio Valley were found to contain two distinct morphological types. It was also noted that samples from the Huron River, Michigan, consisted of only one of the kinds, and collections from the upper Tennessee Valley, the other. Examination of the gonads of these two forms indicated that each contained both males and females, although one sex often predomi- nated in a collection. Combination of several random samples of either kind, however, suggested approximately a 50-50 ratio of males and females in nature. Since no intermediates between the two forms were present and since both males and females were represented in each morphological type, the occurrence of two sympatric species in the Ohio Valley is indicated. One of these, Noturus stigmosus (formerly aligned with Schilbeodes furiosus), in contrast to the other, N. eleu- therus, is characterized by more caudal and anal rays, a longer and more extensively serrated pectoral spine, a longer posterior process of the cleithrum and dorsal spine, a deeper caudal peduncle, usually by 11 preoperculomandibular pores (10 in eleutherus), and by several distinctive features of pigmentation. The difference in number of caudal rays in Ohio Valley specimens of the two species is indicated in figure 4; no prominent sexual difference is apparent in either species. In summary, externa] sexual differences, except perhaps in large breeding ictalurids (which are not well known), are slight. They do not constitute a complicating factor in the recognition of species. Much of the confusion has arisen from lack of knowledge of changes due to growth and sexual maturity. Plots of the number of serrae of the pectoral spine in Noturus stigmosus (fig. 2) and in several other species, suggest a slight sexual difference, the female perhaps averag- ing more. No sexual difference in numbers of fin rays is evident (e.g., figs. 1 and 4). Reproduction Little published observational work has been done on life histories of the species of Noturus. Hankinson (1908) and R. M. Bailey (1938) gave information on Noturus gyrinus; Fish (1932), Greeley (1929 and 1934), and Langlois (1954) recorded time of spawning, observa- 298-943 O—69——3 28 U.S. NATIONAL MUSEUM BULLETIN 282 e «@ a@ a «e © a > AA Atk CAA at aa ad A @ (CO bb NY ee AWA AA a A aoa @sa A Pectoral Spine —-NOWAABDN WD O stigmosus A=" @-2 23 25 27 29 ZI 33 35 37 39 41 43 45 47 49 SI 53 55 57 59 GI Standard Length- mm. eS °o o oOo Oo uJ _ 2 — ® = wn oO al c °o ” = 2 _— o _ ea ® wn Figure 2.—Size change in the number of serrations of the posterior edge of the pectoral spine in Noturus stigmosus. Noturus eleutherus differs little from stigmosus in the number of serrae. In most other species of the subgenus Rabida, it is not possible to distinguish species on numbers of serrae as the overlap is great. There is some suggestion (in the several species plotted) that females average more serrae than do males. tions of young, and egg counts for Noturus flavus; Fowler (1917b) and Clugston and Cooper (1960) reported on Noturus insignis, and Thomerson (1966) gave data on Noturus funebris. Species of Noturus lay comparatively few eggs, undoubtedly the joint result of small body size and large egg size. N. flavus, according to published accounts, lays more than others, probably averaging between 500 and 1000. The madtoms lay few eggs, many species perhaps less than 100 at a time. Spawning occurs in middle or late summer in the north. Since the species are characteristic of warm water, this is to be expected. Probably the late spawning and warm water environment contribute importantly to the relative geographic constancy in fin ray counts of most species. The time of spawning in the south has not been reported, but it probably begins in early spring and extends well into the fall. Hellier (1967) reported young (size not given) of Noturus gyrinus collected in the Santa Fe River, Florida, from April to Sep- tember. Thomerson (1966) was unable to determine the age of a sample of Noturus funebris from Florida, but based its age interpreta- tion on the length frequency distribution of the sample, assuming late season spawning. If the spawning season does begin earlier and CATFISH GENUS NOTURUS RAFINESQUE 29 extends later into the year than in the north, his two smallest groups could easily have hatched the previous year rather than only one group resulting from the previous years spawning activity. The species are nocturnal and all probably spawn in at least relative darkness, under objects or in cavities. All are probably solitary spawners as only one male and one female seem to occupy a nest. The male fertilizes the eggs and then takes over their care and guarding, probably remaining with them and the young until the yolk sac is absorbed and the young are able to fend for themselves. The male, especially, is characterized during the breeding season by swollen areas on the head and upper surfaces, and by a drab color. The apparent ripe female seems to be somewhat less differentiated from nonspawning stages. ReEepropuction In Noturus miurus.—Some information was gained on the reproduction and the young of Noturus miurus in the Huron River, 1.7 miles above Dexter, Michigan, in 1951. The river near the shore where the current was slow, had a bottom of silt and mud. In this area were scattered Chara, Potamogeton, and Sagittaria, numerous tin cans, a fallen tree, tree roots, some brush, boards, much debris, and a sunken boat. In the middle of the stream, where the current was somewhat faster, there was submerged ‘vegetation on a bottom of gravel, sand, rubble, marl, and scattered mollusk shells. The water temperature on August 10 was 78°F. On August 8 and 10 brooding specimens of Noturus miurus were found here in abundance by picking up the sunken tin cans, a relatively high proportion of which contained two adults or one adult with young or eggs. Beer cans which had round openings approximately % to 1 inch in diameter were most often inhabited. The inhabited cans were free of detritus and most often had the open end down- stream. They may have been cleaned by the male which guarded the nest. The area was revisited on July 27, 1952, when N. miurus was again found brooding eggs. A sample of 10 nesting individuals, either with eggs or young, was retained. All were males. They ranged in standard length from 51 to 71 mm.; four were less than 60 mm. long and five were between 60 and 70 mm. Some data were obtained on the relation of size of the male guardian and the number of eggs or young as follows: 55 mm. (34 eggs) ; 55.5 mm. (28 young, brood incomplete); 56 mm. (34 young, brood incomplete) ; 64 mm. (35 eggs); 67 mm. (38 young, brood incomplete) ; 68 mm. (46 eggs). The eggs were large and amber in color. All those in a nest adhered together in an irregular mass until hatched. One guard- ing male swallowed his eggs after placement in a clear glass jar. Nest-guarding males were rather characteristic. The black of the saddles and the top of the head had changed to intermediate grayish 30 U.S. NATIONAL MUSEUM BULLETIN 282 and tended to diffuse indistinctly into the general mottlings of the side. The head became especially broadened, and apparently flattened. The mouth appeared to have enlarged. Several muscle masses became swollen: one was located on the back just anterior to the dorsal fin; another on the top of the head and behind the eye was separated from the swollen mass of the opposite side only by a median depression, and continued indistinctly to below the eye, where it appeared as another enlarged mass on the cheek; the lips and an area around the base of the maxillary barbel were also swollen. The appearance of post- spawning females was not learned, since they did not occupy nests. In a night collection of 99 specimens taken mostly from a Chara bed on August 10, 1951, 44 were males, 55 were females. Twelve of the males resembled nesting individuals. Three of the females lacked the general diffusion of pigment and the swollen muscles that characterized breeding males, but had large eggs; they possibly would still have spawned that summer. No post-spawning females were collected; the remaining 52 females were immature and contained tiny undeveloped eggs. ReEprRopwucTIOoN IN WNoturus stigmosus.—In the Huron River Michigan, reproduction in N. stigmosus seems to occur a little earlier than in N. miurus, and more eggs are laid. The Museum of Zoology, University of Michigan, has two egg masses that were collected July 27, 1932, in which the embryos are well advanced in development. Another lot collected July 27, 1952, was guarded by a male 90 mm. in standard length; it contained 89 eggs. Still another lot taken the same day consisted of an incomplete brood of 61 young; it was also guarded by a male 90 mm. long. A male 67 mm. long, collected on July 19, 1953, was guarding 141 eggs. One of the first two egg masses was reported to have come from gravel under a stone. The others came from cans which had fairly large openings. It is likely that any small cavity of about the size of a Number-2 can or larger with a large open end may serve as a nest. Other adults were found in assorted cans, milk bottles, a coffee can, and sunken boxes. As in N. miurus, occupied cans were free of detritus. The nest-guarding male had the same features: broadening and apparent flattening of the head; swellings of the lips, cheeks, back of head, and predorsal region; and the general diffusion of pigments from the constrasting or sharp markings of younger individuals, that are also characteristic of brooding males of miurus. Early Development Four sac fry from a brood of Noturus miurus that was hatched the night of August 10, 1951, were preserved the afternoon of August 14, 1951, and were stained with alizarin red S to determine ossified CATFISH GENUS NOTURUS RAFINESQUE 31 structures. They range from 9.5 to 11.0 mm. (mean 10.3 mm.) in total length or 8.0 to 9.5 mm. (mean 8.8 mm.) in standard length. At this stage, there are no pigment cells, the yolk sac is large, and considerable ossification has taken place. The top of the skull is unossified but the opercle and the jaws are well ossified. The rod-like basioccipital is prominent; back of it, the vertebral column is well developed. Pectoral and pelvic fins are present, with distinct fin rays; however, the number of rays was not determined because of their mangled condition. The pectoral fin has a poorly ossified spine that has no serrae; the dorsal fin has six soft dorsal rays (at least in three specimens), a well-developed but rather poorly ossified spine, and a short plate-like spine in front. No pectoral radials are visible. The caudal fin at this stage has only 29, 29, and 30 rays, approxi- mately one-half the adult complement; there are no branched rays. There are 14 or 15 (mean 14.8 in 4) anal rays, approximating the number (mean 15.2, see p. 198) in the adults at this locality. The branchiostegal ray complement does not seem to have been attained; there are six to eight rays present. In the vertebral column, it can be seen that five vertebrae make up the anterior vertebral complex (Regan, 1911, pp. 553-554), which also were observed in tiny Noturus gyrinus. The most anterior centrum is longitudinally constricted and becomes a relatively free, flat disk; the next two are somewhat restricted in length; the fourth bears a parapophysis (or transverse process), one arm of which is directed forward at about a 45° angle. Its end turns slightly downward and is attached to the cleithrum. The fifth and successive vertebrae have transverse processes that are directed outward at an angle of about 90° to the vertebral axis. In the next larger size examined (see below), it can be seen that the sixth vertebra is the first that bears ribs. It is also the first free vertebra, as the centra of vertebrae 2-5 are nearly fused. This confirms R. R. Wright’s (1884a, p. 249 and 1884¢c, p. 376) and McMurrich’s (1884, pp. 294-295) observations. The dorsal spine is supported by a heavy strut extending diagonally backward from the fifth vertebra; the plate-like spine in front connects through a large strut to the fourth vertebra and its transverse process. Poster- iorly, two small centra are present in the upturned caudal lobe; however, whether two, or three, or more are fused into the urostyle could not be determined. Two other lots of 10 and 11 specimens of Noturus miurus were stained. The first is a brood at least 12 days old, with a range of from 14.7 to 16.0 mm. in total length and from 12.0 to 12.8 mm. (mean 12.5 mm.) in standard length. The 11 fish in the second brood died at various times, from three to five weeks after collection. They range from 14 to 17 mm. in total length and from 11.5 to 13.4 mm. 32 U.S. NATIONAL MUSEUM BULLETIN 282 (mean 12.8 mm.) in standard length. The body form is like that of the adult; the yolk sac has degenerated and pigmentation is well advanced. The full complement of soft dorsal, soft pectoral, and pelvic rays is present (i.e., 6, 8, and 9 in each fin, respectively). There are nine branchiostegal reys in several. The anterior pectoral-spine serrae are present; there are two posterior serrae on each side in all specimens. The pectoral radials are ossified and distinct with one exception; those of the left side of an individual 12.8 mm. long have the ends fused. The skull is poorly ossified. There are no definite bones in the infraorbital canal system or along the lateral line. The supraorbital, lateral, and preoperculomandibular canals are encased in tubes of bone that almost exceed in diameter the remainder of the bony ele- ments through which the canals pass. The humeral process is long; the maxilla is scarcely visible. Even at these sizes, the anal fin is virtually complete and the caudal fin has over 80 percent of its rays present (i.e., that take up the alizarin stain). The middle rays of the fin are beginning to branch. The following summary gives the range and mean for data from: (a) the first group, averaging 12.5 mm. in standard length, (b) the second averaging 12.8 mm., and (c) large specimens and adults from the same locality. Upper-half caudal rays: (a) 26-28 (27.1), (b) 23-29 (27.1), (c) 30-35 (32.5). Lower-half caudal rays: (a) 22-24 (23.0), (b) 20-24 (22.7), (c) 24-29 (26.4). Total caudal rays (fig. 1): (a) 49-52 (50.1), (b) 43-53 (49.8), (c) 55-64 (58.9). Anal rays: (a) 14-15 (14.6), (b) 14-16 (15.1), (c) 14-17 (15.2). Subgenus Schilbeodes Bleeker Schilbeodes contains all of the dark, nearly uniform colored species of Noturus except flavus. The nine included species show a diversity of characters that precludes an adequate phylogenetic arrangement of the species and may indicate that this grouping is polyphyletic. In addition to the dark color pattern, there are usually 8 or 9 pelvic rays, 6 to 10 soft pectoral rays, and 10 or 11 preoperculomandibular pores. The pectoral spine (pl. 3, figs. 1-9) varies from short and blunt to long and relatively straight; it is deeply grooved in N. gyrinus to very shallowly grooved in N. leptacanthus; there are no anterior serrae, but the anterior edge may have recurved hooks or step-like processes; the posterior edge is variably serrate or smooth; the devel- oped serrae are of variable shape or nearly straight, their tips never uniformly turned toward spine base; the mouth is terminal or inferior; the premaxillary tooth patch is a short, rectangular band without prominent posterior projections. The branchiostegal rays vary from 8 to 12; vertebrae anterior to the first pterygiophore of the anal fin CATFISH GENUS NOTURUS RAFINESQUE 33 10 to 15, most frequently 11 to 14; precaudal vertebrae 7 to 11, usually 8 to 10; ribs 5 to 9, usually 6 to 8; caudal vertebrae showing much variation. The total number of vertebrae often averages higher than in most species of the subgenus Rabida, ranging from 32 to 42. Three species have a modal number of seven (3-+4) hypurals; the remainder have six (3+3), as is typical of the other subgenera. The hypurals show various degrees of fusion. The ossified pectoral radials of each side of the pectoral girdle are typically fused; the anal fin is short to relatively long, containing 12 to 27 rays. The number of branched caudal rays is extremely variable; there are usually 16 or more, except in N. leptacanthus, ranging from 14 to 40. The body varies from short to moderately elongate. The species are of variable size: N. leptacanthus appears to be a small species; N. insignis, N. nocturnus, N. phaeus, and N. funebris attain a relatively large size for members of the genus. The forms are found both in the lowlands and the mountains of the southern and eastern United States, with gyrinus having the greatest range—into southern Canada and south central Texas. This points to a southeastern origin. At least three of the species, N. exilis, N. insignis, and WN. gilberti reach their greatest abundance in uplands. Aside from N. gyrinus, all are typical of graded streams and riffles. N. ezilis, N. phaeus, and N. funebris, and possibly N. leptacanthus, appear to prefer small streams. The earliest known species were placed in the genera Silurus Linnaeus or Pimelodus Lacépéde. The name Pimelodon Vaillant was probably derived from the word Pimelodus and was based on Pimelodon insignarius Vaillant equals Noturus insignis (Richardson); otherwise it bas not been in accepted usage. In later years, most authors have included the species as a unit, together with the species of the sub- genus Rabida, in Schilbeodes, or all with flavus in Noturus. A few that have recognized Rabida as a genus, unnaturally included one or more of the species with pectoral spine serrae in that genus and generally restricted Schilbeodes to include only leptacanthus, nocturnus, and gyrinus, all of which lack prominent pectoral spine serrae. Aside from Noturus lachneri, which is intermediate in several char- acters between Noturus exilis and Noturus gyrinus, Schilbeodes would be satisfactorily monotypic, with Pimelodon including the remainder of the species. The gradations of characters do not suggest the desir- ability of recognition of one group with a terminal mouth and ten preoperculomandibular pores and another with an inferior mouth. Schilbeodes is formed from Schilbe, a genus of fishes lacking an adipose fin, and odes (Greek), meaning likeness. This name was proposed by Bleeker in the belief that the type-species, Silurus gyrinus Mitchill, lacked an adipose fin. U.S. NATIONAL MUSEUM BULLETIN 282 Key to the Species of the Subgenus Schilbeodes Mouth terminal, jaws about equal; typically 10 preoperculomandibular pores, but 11 in some populations of Noturus gyrinus .. . - ta eee Mouth inferior, lower jaw included; piconeeeniaadaiiee paces normally 11, but 10 in some populations, especially Noturus nocturnus . ... . 4 Infraorbital and supraorbital canals typically joined anteriorly (1 internasal pore); head length in standard length 3.6 to 4.4 times; anal rays 16 to 22; vertebrae 86 to41..... a ee tte: tres Infraorbital and supraorbital panels Hmiealled Semaraied icra (2 inter- nasal pores) ; head length in standard length 3.0 to 3.8 times; anal rays 12 to 18; vertebrae 32 to 37; pectoral spine lacking serrae. . Noturus gyrinus Pectoral spine lacking serrae; fins nearly unicolor; caudal rays 56 to 61; distance from rear end of adipose fin to tip of caudal fin stepped 1.2 to 1.6 times in distance from origin of dorsal fin to rear end of adipose Ani hare . . . . . .Noturus lachneri, new species Pectoral spine oan Pecmanent well-developed serrae; fin margins often black; caudal rays 44 to 57, usually 47 to 54; cians from rear end of adipose fin to tip of caudal fin stepped 1.6 to 2.5 times in distance from origin of dorsal fin to rear end of adipose fin... . . . Noturus exilis Pelvic rays normally 8; pectoral spine nearly smooth behind; body and fins often with numerous clusters of darker chromatophores or freckles. Noturus leptacanthus Pelvic rays normally 9 or more; pectoral spine nearly smooth, rough, or serrate behind; no prominent clusters of chromatophores scattered over body an@ hnsy. 7 2 7... Bah Vee eee Entire abdomen and lower pare) of nea aise well miemenied: covered with large, discrete chromatophores; anal base long, with 18 or more, usually 20 or more rays, and separated from caudal fin by only a narrow space; dorsal spine slender and flexible in young and juveniles, becoming stout withage ... 2b Ce a ee Abdomen and lower ate of Read mort anpieietcn. but pigment often on chin and in a narrow cross band anterior to pelvic fins (very large or old specimens, typically above 80 mm. standard length, often with a diffuse, dark pigment covering the lower surface) ; anal base shorter, of 21 or fewer rays; dorsal spine stout, stiff at all sizes . . . . Dee a Cen eLh Several prominent, distinct, sharp serrae along pastetion sage of pectoral spine; anal rays 18 to 25, yin 20 to 22; soft pectoral rays typically 8 OD Oe ae . . . . . Noturus phaeus, new species Pectoral spine Raia eae serrae, or occasionally with one to three irregular size serrae in young, the posterior edge usually only roughened; anal rays 20 to 27, usually 21 to 24; soft pectoral rays typically 9. Noturus funebris Vertical fins with broad light borders; abdomen and lower surface of head immaculate; caudal rays 46 to 54; upper-half caudal rays 22 to 27; spines VeELVAShOLis 4) cae : . . . Noturus gilberti Vertical fins dark or segs ¢ or pecaeioarly ante oe narrow light margins; chin and area in front of pelvic fins usually with pigment; caudal rays 54 to 67, typically 56 or more; upper-half caudal rays 27 to 35, usually 29 Or Moree... . d : Sea eM Pectoral spine tn ater serrae arta fren become Serearilae or reduced in size with age; vertical fins frequently with jet black margins; preoper- _— CATFISH GENUS NOTURUS RAFINESQUE 35 culomandibular pores nearly always 11; anal rays 15 to 21, usually 17 to 19; vertebrae 37 to 42, usually 38 to 40; body relatively slender; head depresded . arse . . . . . Noturus insignis Pectoral spine meets: with ‘distinet peel ‘ts posterior edge usually only roughened; vertical fins usually uniformly dusky, but sometimes light edged and frequently with a submarginal dark brown band; preoperculomandib- ular pores either 10 or 11; anal rays 15 to 20, typically 16 to 18; vertebrae 35 to 38; body heavier; head rounded above .. . . Noturus nocturnus Noturus gyrinus (Mitchiil) TADPOLE MADTOM PuatEs 2 (Fic. 1), 3 (Fie. 1), 5 (FIG. 1); Mae 1 Silurus gyrinus Mitchill, 1817, p. 289; reprinted in Jordan and Evermann, 1896a, p. 281 (original description; Wallkill [R.]).—Mitchill, 1818, pp. 241, 322-323 (description; Wallkill R.).—Rafinesque, 1819, p. 422; 1820b, p. 68; 1820c, p. 362; reprinted in Jordan, 1877c, p. 41, and in Call, 1899, p. 128 (believed to belong to . . . Noturus)—De Kay, 1842, pp. 185-186 (description; New York).—Bleeker, 1858, pp. 249, 258 (relationship; description; type- species of Schilbeodes).—Giinther, 1864, p. 104 (New York).—Gill, 1876, pp. 422, 423 (correctly referred to Noturus by Rafinesque; Bleeker’s 1858 description reprinted).—Jordan, 1919, p. 279 (orthotype of Schilbeodes) . Schilbeodes gyrinus (Mitchill).—Bleeker, 1858, p. 258 (North America).—Eigen- mann and Beeson, 1894a, pp. 81-82; 1894b, p. 45 [and 1905, p. 121] (Indiana records: Kankakee R., Riverside;* Trail Cr., near Michigan City;* others compiled).—Cox, 1896, pp. 608, 612 (Minnesota records).—Evermann and Cox, 1896, pp. 375-426 (distribution; description; Missouri basin records, including Norfolk Cr.,* Norfolk Junction, Nebr.)—Eigenmann, 1896, p. 253 (Indiana records: String L.* [outlet]; Turkey L.,* channel,* and Turkey Cr.* [Syracuse]).—Jordan and Evermann, 1896a, pp. 144-146; 1896b, p. 234.—E. Smith, 1897, pp. 11-21 (Greenwood and Wawayanda Lakes, and Ramapo and Hackensack* Rivers, near New York City).—Cox, 1897, pp. 19-20, 78 (compiled).—Jordan and Evermann, 1898, p. 2790 (anal fin length).—Osburn and Williamson, 1898, pp. 11, 19 (Big Darby Cr. and Masons Run,* Franklin Co., Ohio).—Evermann, 1899, p. 306 (Black Bayou, Miss.; L. Lapourde, La.).—Reed, 1900, pp. 232-233 (poison glands) .—Ever- mann and Kendall, 1900, p. 51 (Florida records).—Blatchley and Ashley, 1901, p. 176 (compiled).— Ramsey, 1901, p. 211 (Winona L.,* Ind.).—Blatch- ley, 1901, p. 252 (L. Maxinkuckee*).—Evermann and Kendall, 1901, p. 480 [and 1902, p. 210] (New York records).—Evermann, 1901, p. 338 (Wallkill R., N.Y.; economics).—Osburn, 1901, pp. 27-28 (ecology; Ohio records including [E. end] Sandusky Bay*).—Blatchley, 1902, p. 184 (compiled) .— Evermann, 1902, p. 95.—Bean, 1903, pp. 93-94, 739 (Passaic R., N.Y.).— Large, 1903, pp. 9-10 [and 1905, pp. 56-57] (ecology; Illinois distribution) .— Fowler, 1906a, pp. 171-173 (associations; Crosswicks Cr. and tribs. of Delaware R., ‘in or above tidewater,’ New Jersey); 1906b, p. 596 (Mill Cr., near Bristol,* Pa.).—Michael, 1906, p. 9.—Fowler, 1907a, p. 14 (Penn- - *Material designated by an asterisk has been re-examined. 36 U.S. NATIONAL MUSEUM BULLETIN 282. sylvania records [in part]: Delaware R., Holmesburg;* Mill Cr., Bristol* and Tullytown); 1907b, p. 279 (specimen, Pennypack Cr., Holmesburg, Pa., figured; affluent of Crosswicks Cr., near Trenton,* N.J.; Mantua Cr., N.J.).—Reed, 1907, pp. 555-564, figs. 3-5 (poison apparatus).—Hankinson, 1908, pp. 188-249, pl. 59 (Walnut L.,* Mich.; ecology; food; reproduction; egg mass described and figured).—Wagner, 1908, p. 32 (L. Pepin, Wis.).— Meek, 1908, p. 141.—Fowler, 1908, pp. 151-152 (Pennsauken Cr., near Pennsauken,* N.J.).—C.W. Nash, 1908, pp. 24-25.—Reed and Wright, 1909, p. 393 (Cayuga L.,* N.Y.).—Forbes and Richardson, 1909 [and 1920, pp. Ixx iii-exv], pp. Ixxiii-cix, 176-201, fig. 49 [not caption, 1st ed.], color plate, map 58 (ecology; comparison; range [in error]; Illinois distribution [see p. 44; one exception noted]).—Forbes, 1909, pp. 387-427, map 58 (lIlli- nois ecology and distribution [one exception]).—Meek and Hildebrand, 1910, pp. 245-246 (Illinois records including: lagoon, Jackson Park, Chicago;* Des Plaines R., Berwyn;* and Indiana records).—Fowler, 19lla, p. 602 (Maurice R. basin [Muddy Cr.], near Elmer;* L. Hopatcong, in Hurd Cove,* N.J.); 1911b, pp. 4, 9 (Delaware*).—Leathers, 1911, pp. 246, 249 (at Rush L.,* and pond, Sand Point,* Huron Co., Mich.).—Bean and Weed, 1911, p. 172 (Little Beaver Dam Br.,* D.C.).—Hankinson, 1911, p. 203 (rush zone, Walnut L.,* Mich.).—Fowler, 1912a, pp. 241, 327 (Notch Cr., Brookdale; Muddy Cr., Elmer,* N.J.)—Wilson and Clark, 1912a, p. 15 (Indiana record); 1912b, p. 40 (Indiana records).—Rizhardson, 1913, p. 411 (Quiver L.,* Ill.).—Shelford, 1913, pp. 85-142 (ecology; L. Michigan, to 25 m. depth; Dead R., N. of Chicago).—Hankinson, 1913, pp. 108, 112 (Kaskaskia R. system, Ill.).—Halkett, 1913, pp. 17, 58 (Canada; hypothetical, Ontario).— Nichols, 1913, p. 92.—Fowler, 1913, p. 92 (Pennsylvania records compiled [in part only]).—Forbes, 1914, p. 18, fig. 14a, map 58 (Illinois distribution [one exception]; teeth figured).—Fowler, 1914a, p. 346 (Delaware R.); 1914b, p. 940 (near Penn’s Manor,* Pa.).—Pearse, 1915, pp. 7-20 (food; ecology; Six Mile Cr., trib., L. Mendota, Wis.).—Fowler, 1915a, p. 208 (New Jersey records: [dam at] Pitman;* [West Br., Rancocas Cr.], Newton’s Bridge;* [trib. Delaware R.], Florence.* Recorded from: Edison?, Pa.; Minneapolis, * Minn.; Brook R.,* Iowa; Miami,* Fla.; other records relisted [in part only)) ; 1915b, p. 5; 1916a, p. 12 (Alloway Cr., Alloway,* N.J.); 1916b, p. 41 (New Jersey record).—Adams and Hankinson, 1916, p. 168 (Oneida L. and tribs., N.Y.).—?Fowler, 1917a, p. 117 (Pennsylvania record).—Wells, 1918, pp. 570-571 (high resistance to adverse conditions in comparison with other fishes; negative to CO,).—Fowler, 1918a, p. 4 (New Jersey records).—A. H. Wright, 1918, pp. 540-542 (in 5 Monroe Co., N.Y. streams).—Pearse, 1918, pp. 274-289 (food; Six Mile Cr. and L. Mendota, Wis.).—Nichols, 1918, pp. 38, 103.—Evermann, 1918, pp. 335, 366 (compiled).—Adams, Hankin- son, and Kendall, 1919, p. 201 (New York record).—Fowler, 1919, p. 57 (Delaware basin in Bucks, Philadelphia, and ?Lehigh Counties, Pa., only).— Evermann and Clark, 1920, pp. 331-332, pl. 5 (reproduction; ecology; food; L. Maxinkuckee,* in main and small lake).—Wilson, 1920, p. 294 (food; ecology).—T. Surber, 1920, pp. 20-21 (figured; records compiled).—Hankin- son, 1920, p. 134 (nests in natural cavities or depressions).—Fowler, 1920b, p. 150 (New Jersey records).—Conger, 1920, p. 11.—Lamont, 1921, p. 3 *Material designated by an asterisk has been re-examined. Le —— CATFISH GENUS NOTURUS RAFINESQUE yA (parasite, Plagiorchis corti, from intestine; L. Mendota, Wis.).—Fowler, 1921a, pp. 389, 398 (Delaware records); 1921b, p. 63 (Bucks Co., Pennsyl- vania records: Delaware R.;* Neshaminy?; Scott Cr.; others relisted).— Osburn, 1921, p. 6 (Portage L., Ohio).—Pearse, 1921, pp. 10-51 (food; L. Pepin, Wis.) —Palmer and Wright, 1922, pp. 356-361 (Okefinokee Swamp;* others compiled).—Dymond, 1922, p. 62.—Pratt, 1923 [and 1935, p. 89], pp. 95-96, figs. 41b—42 ([caption to figure transposed in first edition]; description; range).—Fowler, 1923, p. 28 (relisted).—Hankinson, 1923, p. 32 (New York records).—Wilson, 1923, pp. 253-254 (insect food).—Pearse, 1924, p. 148 (parasites).—Reed, 1924a, pp. 431-451, figs. 2-8 (structure and development of spine); 1924b, pp. 232-256, pl. 8, figs. 7-9, 12; pl. 9, figs. 18-21; pl. 10, figs. 22, 24; pl. 11, fig. 26; pl. 13, fig. 28; pl. 15, fig. 30 (description of pectoral spine and glands).—Hankinson, 1924, pp. 82, 86 (New York records).— Brimley and Mabee, 1925, p. 15 (North Carolina records including Little R., near Wendell*).—A. H. Wright, 1926, p. 81 (Okefinokee Swamp*).— Hubbs, 1926, p. 51.—Cahn, 1927, p. 42 (Oconomowoc R., Stonebank [USNM 87590, Waukesha Co.,* Wis., A. R. Cahn]; food; ?ecology).—Greene, 1927, p. 306.—Greeley, 1927, pp. 49, 57 (not taken in Genesee basin, New York [only]).—Fowler, 1928, p. 225 (relisted).—Potter and Jones, 1928, p. 355 (compiled).—Adams and Hankinson, 1928, pp. 384-385 (life history; ecology; food; effects of poison; Oneida L., New York localities).—Greeley, 1928, pp. 87-107 (food; ecology; New York records).—Eaton, 1928, p. 42.—Hubbs, 1928, p. 250 (‘‘unusual habitat ascribed. .. .”’ by Cahn).—Hubbs and Greene, 1928, p. 390.—Wiebe, 1928, p. 161 (Mississippi R., Minneapolis,* Minn.).—Hildebrand and Towers, 1928, p. 122 (borrow pit, about 6 mi. SW. Greenwood,* Miss.).—Bajkov, 1928, p. 97 (compiled: Hudson Bay drainage).—Dymond, Hart, and Pritchard, 1929, pp. 4, 24 (Canada rec- ords).—Hankinson, 1929, p. 452 (North Dakota records [see p. 45]).— Greeley, 1929, pp. 155-174 (ecology; New York records).—Jordan, 1929, p. 94.—Truitt, Bean, and Fowler, 1929, p. 36 (?Baltimore Co., Md.).— Mueller, 1930, pp. 174-177 (parasites)—Osburn, Wickliff, and Trautman, 1930, p. 174 (Ohio).—Jordan, Evermann, and Clark, 1930, p. 155.—Thomp- son and Hunt, 1930, pp. 27-64, map 33 (Champaign Co., IIl., distribution; Salt Fork system only; ecology).—Cole and Allison, 1931, pp. 119-124 (decreased reaction time from increased hydrogen ions).—Greeley and Bishop, 1932, pp. 76, 87 (ecology; New York records and distribution).— Odell, 1932, pp. 117, 119 (New York records).—Sibley, 1932, pp. 129-130 (food; Black L. and Black R., N.Y.).—Shurrager, 1932, pp. 386-409 (arm of Margaret’s Cr., Athens Twp., Ohio).—Greeley and Bishop, 1933, pp. 88-98, color pl. 8 (ecology; records Hudson R. system; adult male figured) .— R. A. Moore, 1933, p. 17 (description kidneys).—Luce, 1933, p. 118 (near mouth Kaskaskia R., Illinois)—Gowanloch, 1933, pp. 85, 87 (Louisiana) .— Friedrich, 1933, p. 29 (Minnesota record).—Churchill and Over, 1933, pp. 9, 60, fig. 48 (South Dakota distribution).—Hankinson, 1933, pp. 563, 568 (“Vegetal Shallows of Michigan inland lakes’’).—Roach and Wickliff, 1934, p. 374 (killed by oxygen reduction, Buckeye L.,* Ohio).—Van Cleave and Mueller, 1934, pp. 170-322 (parasites, Oneida L., N.Y.).—Jackson, 1934, p. 4 (Manitoba and prairie provinces, Canada).—Wallace, 1935, pp. 143- 164 (parasites).—Greeley, 1935, pp. 86, 96 (ecology; New York records).— *Material designated by an asterisk has been re-examined. 38 U.S. NATIONAL MUSEUM BULLETIN 282 Odell, 1935, p. 132.—Fowler, 1935a, pp. 7, 19, fig. 36 (6 mi. NW. of Bennetts- ville,* S.C.).—Greene, 1935, pp. 144-145, map 60 (ecology; Wisconsin distribution [one exception]).—O’Donnell, 1935, p. 484 (ecology; Illinois distribution).—Ewers and Boesel, 1935, pp. 58, 63 (food; Buckeye L., Ohio).—Aitken, 1936, p. 33 (Iowa).—Fowler, 1936b, p. 150 (Florida rec- ords).—Hoover, 1936, p. 239 (Merrimack R.* and Penacook L., near Concord, N.H.).—Greeley, 1937, pp. 87, 97 (ecology; New York records).—Odell and Senning, 1937, p. 108 (New York records).—Carr, 1937, p. 81.—Toner, 1937, p. 14 (Belleville, Ontario).—Murray, 1938, p. 84 (in Indiana trout streams) .— Blatchley, 1938, pp. 66-67.—R. M. Bailey, 1938, pp. 151-182 (ecology; repro- duction; introduced, Merrimack Co., N.H.: Penacook L.,* 2.5 mi. WNW. Concord; Merrimack R., 1.25 mi. SE.* Concord and 2 mi. SSE.* Concord) .— Schrenkeisen, 1938, p. 166.—Baker and Parker, 1938, p. 161 (Reelfoot L.,* Tenn.).—Bailey and Oliver, 1939, p. 152 (compiled).—Hubbs and Lagler, 1939, p. 26 (comparison; Great Lakes basin).—Trautman, 1939, pp. 279, 287 (Ohio records).—Dymond, 1939, pp. 27-28 (Ontario ecology and rec- ords).—Bangham and Hunter, 1939, pp. 401-434 (parasites, W. Lake Erie).—Kuhne, 1939, p. 68, fig. 43—Baker, 1939a, p. 21 (Reelfoot L., Tenn.*); 1939b, p. 44 (comparison; Reelfoot L.*).—Raney, 1939b, pp. 675, 677 (associations; Cayuga L., Ithaca,* N.Y.).—Fowler, 1940b, p. 8 (Penn- sylvania records relisted [in part only]).—Hubbs and Lagler, 1941, pp. 63-64, fig. 83 (comparison; range; ecology).—Bangham, 194la, pp. 296, 302-305 (parasites; Hardee? Co., Fla.) ; 1941b, p. 445 (parasites; Buckeye L., Ohio).— Harkness and Pierce, 1941, p. 112 (L. Mize, Fla.) —Aitken, 1941, p. 389 (Iowa).—Driver, 1942, p. 254 [and 1950, p. 262].—Bangham and Venard, 1942, p. 29 (parasites, Reelfoot L.).—Eddy and Surber, 1943, pp. 151-162, fig. 29 (comparison; ecology; Minnesota distribution and records: St. Louis R. [system], trib. to L. Superior; L. Itasca; others compiled).—Bennett, 1943, p. 365 (population size estimated, artificial lake, Ill.)—Hubbs and Lagler, 1943, p. 80 (Foots Pond,* Ind.).—Haas, 1943, p. 163 (Fox and Rock R. systems, Ill.) —Hinks, 1943, pp. 58, 61 (figured; Red and Assiniboine R. systems, Manitoba).—Toole, 1943, p. 12 (ecology; Texas).—Smith and Moyle, 1944, p. 115 (relisted) —Radforth, 1944, pp. 6-62, figs. 22-23 (ecology; Ontario and L. Erie distribution; distribution compared with isotherms).—Goin, 1944, p. 146 (near Gainesville, Fla.; about hyacinth roots).—Fowler, 1945, pp. 32-271, fig. 164 (synonymy, distribution, and records, SE. U.S.).—Bangham, 1946, pp. 294, 302 (parasites, Lake-du-Bay, Wis.).—Cuerrier, Fry, and Prefontaine, 1946, p. 26 (St. Lawrence R., near Montreal).—Mélancon, 1946, p. 130.—Brimley, 1946, p. 15 (North Carolina records).—Marshall, 1947, p. 71 (Blue Springs, Fla.)—Baughman, 1950, p. 131 (Texas).—Neil, 1951, p. 765 (in crayfish burrows, Emanuel Co., Ga.).—?Fowler, 195la, p. 91 (Lancaster Co., Pernsylvania records).— Hunt, 1953, p. 15 (Tamiami Canal, Fla.).—Hubbs, 1957, p. 98 (Texas dis- tribution).—Keleher and Kooyman, 1957, pp. 106, 110 (Manitoba distribu- tion; north to central L. Winnipeg).—G. Moore, 1957, pp. 142-144, figs. 2-77b, 2-79F, 2-80.—Hubbs and Lagler, 1957, p. 5; 1958, pp. 89, 91, fig. 175.—C. Hubbs, 1958, p. 8 (Texas distribution).—Slastenenko, 1958a, p. 7 (Canadian distribution); 1958b, pp. 248, 353, fig. 102 (compiled).—W. B. Scott, 1958, p. 19 (Canadian distribution).—Lambou, 1959b, p. 192 (in *Material designated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 39 Louisiana impoundments).—Becker, 1959, p. 97 (ecology; central Wisconsin distribution).—Boudreaux, Strawn, and Callas, 1959, p. 10 (Texas record) .— Symington, 1959, p. 8 (Souris R., Saskatchewan).—Lambou, 1961, p. 57 (Louisiana lakes).—Anthony, 1963, p. 87 (parasites, McDill Pond, Wis.).— Eddy, Moyle, and Underhill, 1963, pp. 113, 115 (Mississippi R., above and below St. Anthony Falls, Minn.).—Birkhead, 1967, pp. 101-109 (compiled). Noturus gyrinus (Mitchill).—Gill, 186la, p. 45 (synonymy).—Abbott, 1871, p. 718 (Stony Brook, near Princeton, N.J.)—Jordan, 1876b, pp. 303-304 (description; Wallkill R., N.Y.).—Jordan and Copeland, 1876, p. 160 (E. Pennsylvania [based on N. insignis?] and SE. New York).—[Klippart], 1877, p. 153 (Indiana).—Jordan, 1877b, pp. 352, 371 (comparison; range {in error]); 1877d, pp. 71-120, pl. 42, fig. 66-67, pl. 43, fig. 69b-69c (com- parison; range and synonymy [in part]; relationship; specimen from Hudson R. figured; Orange, Rockland, and ?Chemung Counties, N.Y.); 1877e, p. 611.—Jordan and Brayton, 1878, p. 93 (relationship).—Jordan, 1878d, p. 337 (description; range [in error]); 1878e, p. 414 (range [in error]).—Bean, 1880, p. 112 (near Piermont,* N.Y.).—Hay, 1881, pp. 514-515 (Mississippi rec- ord).— Jordan and Gilbert, 1883, p. 98.—Swain and Kalb, 1883, pp. 639-644 (synonymy; range [in error]; description; Illinois R.*).—Hay, 1883, p. 74 (Tombigbee, Chickasawha, and upper Mississippi R. systems).—Jordan and Meek, 1885, p. 8 (Chariton R., Chariton,* Iowa).—Graham, 1885b, p. 71 (synonymy; ?Kansas).—Forbes, 1885a, p. 84 [ana 1900, p. 76] (ecology; “throughout Illinois’’?).—Jordan, 1885, p. 802.—Eigenmann and Fordice, 1886, p. 410 (Bean Blossom Cr.,* Ind.).—Jordan and Gilbert, 1886, p. 6 (comparison).—Forbes, 1888a, pp. 455-473 (ecology; food; Pekin,* Peoria, * and Fox R.,* McHenry Co., Ill.; Clear L.* [East Cairo], Ky.).—Forbes, 1888b, pp. 514-536 (food).—Evermann and Jenkins, 1888, pp. 44, 52 (Indi- ana records).—Garman, 1889, p. 80 (Harkness Slough;* Dead Man’s Slough; Willow L.;* Lily L.;* Broad L.;* Wood Slough,* all near Quincy, IIl.).— Henshall, 1889, p. 124 (Ross L., Ohio).—Meek, 1889a, p. 301 (New York records: Cayuga L.;* stream near Montezuma); 1889b, p. 168 (Iowa).— Jordan, 1890, pp. 158-165 (Indiana records: L. Maxinkuckee;* Big Cr., Posey Co.*)—J. Nelson, 1890, p. 671 (lake region, New York; doubted in New Jersey).—McCormick, 1890, p. 126 (Lorain Co., Ohio).—Bean, 1892, p. 20 (description only).—R. R. Wright, 1892, p. 443 (Great Lakes region) .— McCormick, 1892, p. 13 (Lower parts streams entering the lake, Lorain Co., Ohio).—Woolman, 1892a, pp. 253-287 (Kentucky records); 1892b, p. 301 (Florida records).—Meek, 1892a, p. 12 (Iowa); 1892b, pp. 221-246 (Iowa records including Mississippi R., Muscatine;* Cedar R., Waverly* and West Liberty;* West Br. Cedar R., Dumont;* Indian Cr.* [Marion]; Big Sioux R., ?Sioux Falls or Sioux City*); 1892c, p. 108 (Cedar R. basin, Iowa).—Call, 1892, p. 46 (ecology; records relisted).—Meek, 1893, p. 229 (Arkansas [compiled in error?]).—Hay, 1894, pp. 172, 173 (compiled).— Abbott, 1894, p. 479 (compiled).—Kirsch, 1894, p. 87 [and vol. 14, p. 36]; 1896a, p. 48 (weedy bottom in upper Blue R.,* Indiana).—Meek, 1894b, p. 135 (Platte R., Fremont,* Nebr.; Floyd R., Lemars* and Sioux City,* Iowa; Storm L.,* Iowa).—Loénnberg, 1894, p. 113 (Florida records).— Gar- man, 1894, p. 56 (compiled).—Kirsch, 1895, pp. 323-335 (records include Tiffin R., Manitou Beach,* Michigan; Maumee R.,* Fort Wayne, Ind.; *Material designated by an asterisk has been re-examined. 40 U.S. NATIONAL MUSEUM BULLETIN 282 Fish Cr., Hamilton,* Ind.).—Woolman, 1896, pp. 356, 369 (Minnesota rec- ords: Pomme de Terre R., Appleton;* Chippewa R., Montevideo.* North Dakota records including Goose R.,* Hillsboro).— Meek, 1896, pp. 345-349 (Arkansas records: Old R., near Greenway [CNHM 1559,* either Marked Tree or ?Greenway]; St. Francis R.; Little and St. Francis Rivers, near Marked Tree;* Arkansas distribution [mot W. Arkansas and Indian Terri- tory]).—Hay, 1896, pp. 85, 89 (Water Valley, * Ind.; Kankakee R., Momence,* Ill.).—Call, 1896, p. 14 (ecology; Falls of the Ohio, Ky.).—Kirsch, 1986b, p. 105 (records relisted).—H. M. Smith, 1901, p. 134 (L. Mattamuskeet, N.C.).—Hay, 1902, p. 70.—Mitchell, 1904, pp. 154, 161, 405-406 (oral breathing valves described [confused and probably in part as Schilbeodes punctatus, a lapsus]).—Jordan, 1904, pp. 42, 351.—Fowler, 1906a, p. 173 (compiled, New Jersey).—Hahn, 1910, p. 175 (compiled).—Palmer and Wright, 1922, p. 356 (New R., Fla.).— Arnold and Ahl, 1936, p. 233 (figured; description; introduced as aquarium fish into Germany, 1895; failed to reproduce).—Bailey, Winn, and Smith, 1954, pp. 131-160 (records, lower Escambia R., Fla. [see p. 44]; fresh water; synonymy).—Taylor, 1954, p. 44 (synonymy; Michigan records [see p. 44]).—Slack, 1955, p. 40 (Indiana record).—Gerking, 1955, p. 76.—R. M. Bailey, 1955, p. 528 (Bass L., Michi- gan; heat mortality).—Harlan and Speaker, 1956, p. 114, pl. 21.—Cleary, 1956, map 64 (Iowa distribution).—R. M. Bailey, 1956, pp. 335, 364 (key).— Eddy, 1957, p. 152, fig. 382.—Taylor, 1957, p. 192.—Briggs, 1958, p. 260 (Florida, distribution).—Cross and Minckley, 1958, pp. 104, 106 (descrip- tion; records, Osage R., Kansas).—Hancock and Sublette, 1958, p. 46 (Louisiana records, including Bayou Santabarb*).—Hallam, 1959, p. 158.— Stegman, 1959, p. 30 (Kinkaid Cr., Ill.).—Trautman, 1959, pp. 41, 43, 96, 432, 441-445, fig. 114, map 114 (synonymy, description, ecology, distribution, Ohio; range).—F. Cook, 1959, pp. 34, 135, 141, fig. 25D (Mississippi rec- ords).—Davis, 1960, p. 20 (Ouachita R., La.).—Behnke and Wetzel, 1960, p. 141.—Bonn and Holbert, 1961, p. 292 (L. Lavon, Tex.).—Schwartz, 1961, p. 25.—Patrick, 1961, p. 256 (in part; Ottawa and Potomac Rivers).— Suttkus, 1961, p. 63, fig. 5 (comparison; skull illustrated).—Deacon, 1961, pp. 395, 420-421 (Marais des Cygnes* R., Kans.).—Lambou, 1962a, p. 77; 1962b, pp. 197, 199 (Lake Bistineau,* La.).—Tabb and Manning, 1962a, p. 609 (Everglades National Park records); 1962b, p. 59 (records, zone of transition from marine to fresh water, Everglades National Park).—I. A. Carr, 1962, p. 6 (Saginaw Bay distribution).—Louder, 1962, pp. 69, 71 (North Carolina records).—Bailey and Allum, 1962, pp. 90, 118, 121 (char- acters, synonymy, distribution; South Dakota stations 8,* 15, 26c, 39, 44).— Collette, 1962, pp. 168, 169, 172 (in association with Etheostoma).—J. M. Walker, 1962, p. 38 (Louisiana parishes); 1963, p. 47 (Choudrant Bayou, La.).—P. W. Smith, 1963, pp. 254, 257-258 (records, Kaskaskia R., IIl.).— Feldmann, 1963, p. 17 (North Dakota records).—McNaught, 1963, pp. 43, 46 (L. Mendota).—Griswold, 1963, pp. 217, 220, fig. 3 (abundance and growth, Clear L., lowa).—W. B. Scott, 1963, pp. 116, 123 (Ontario records; Canadian distribution) —Hanson and Campbell, 1963, p. 139 (Perche Cr., Mo.).—Larimore and Smith, 1963, pp. 324-345, fig. 47 (Champaign Co., Illinois records, ecology, distribution map).—Anderson, 1964, pp. 45, 51 (South Carolina records).—Keup and Bayless, 1964, p. 121 (Neuse basin, *Material designated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 41 North Carolina; salinities 0 to 0.098).—Becker, 1964a, pp. 21, 25; 1964b, pp. 33-35, 46 (Wisconsin records).—B. T. Walker, 1965, pp. 106, 108 (Bayou D’ Arbonne, station 16;* taken from bottles).—Burton and Douglas, 1965, pp. 93-94 (southern Bayou DeSiard, La.).—Suttkus and Taylor, 1965, p. 177.—H. H. Moore and Braem, 1965, pp. 2, 44 (records [stas. 155* and 156* re-examined] and distribution, Wisconsin tributaries of L. Superior).— Starrett and Fritz, 1965, p. 25 (L. Chautauqua, Illinois)—P. W. Sinith, 1965, p. 9 (Illinois distribution).—Larson, 1966, pp. 99-100 (parasites, L. Itaska, Minn.).—Hellier, 1967, pp. 18-46 (parasites, ecology, distribution, Santa Fe R., Florida; young collected April to September).—Cross, 1967, pp. 197, 216 (figure, description, Kansas distribution). Schilbeodes gyr*nus gyrinus (Mitchill).—Greeley, 1936, pp. 77, 84 (ecology; New York: not in Susquehanna system; in Basher Kill system and Yankee L.).— Odell and Senning, 1936, pp. 93, 95.—Greeley, 1938, p. 69 (Cayuta L. and Cayuta Cr., N.Y.; ecology).—Odell and Senning, 1938, p. 99. Noturus flavus Rafinesque [misidentifications]—De Kay, 1855, p. 66 (New York).—Baird, 1856, p. 26 (Hackensack R. [near Piermont*], N.Y.).— E. W. Nelson, 1876, p. 50 (Illinois, in part) Jordan, 1876b, p. 303 (de- scription; Ohio Valley, in part); 1877a, p. 46 (Indiana records [see Jordan, 1877d, p. 102]: Clear and Pine Lakes, Laporte Co.; St. Joseph R. system; White R. and tribs., near Indianapolis)—Garman, 1881, p. 89 (San Antonio, Texas*).—Eigenmann and Beeson, 1894a, p. 81; 1894b, p. 44 [and 1905, p. 120]; Hay, 1894, p. 175; Blatchley and Ashley, 1901, p. 288, and Blatchley, 1938, p. 66 (Jordan’s Indiana records compiled).—Forbes and Richardson, 1909 [and 1920], p. 195 (Indiana distribution, in part).— Fowler, 1915a, p. 208 (Chariton,* Iowa [part is Ictalurus melas]).—? Bissett, 1927, p. 127 (Manitoba).—?Bajkov, 1928, p. 97 (Hudson Bay drainage, Canada).—?V. W. Jackson, 1934, p. 4 (Manitoba and prairie provinces) .— Hinks, 1943, p. 62, and Dymond, 1947, p. 23 (Manitoba records need con- firmation).—Gerking, 1945, map 62 (in part, compiled Indiana records) .— Baughman, 1950, p. 131 (Texas compiled).—Rostlund, 1952, map 20 (distribution, in part). Noturus sialis Jordan, 1877b, p. 377 (nomen nudum; White R. and small tribs., near Indianapolis, Indiana); 1877d, pp. 73-120, pl. 42, figs. 68, 69 (original description; type figured, and from White R., Indianapolis; about Falls of Ohio; range; synonymy); 1877e, pp. 611, 613 (range) ; 1878a, p. 119 (Ohio R.; White R.; Wabash R.; L. Michigan) ; 1878b, p. 68 (Illinois distribution; Wabash R.; Cache R.,* Johnson Co.; Illinois R., Pekin* and Cairo); 1878c, p. 368 (Indiana); 1878d and 1884, p. 337 (description; range [Kansas?]); 1878e, p. 414 (range).—Jordan and Brayton, 1878, pp. 87-95 (distribution; rela- tionship).—Forbes, 1880b, pp. 74, 79 (food).—Jordan, 1882, pp. 745-801 (Ohio; description). Schilbeodes gyrinus var. sialis (Jordan).—Jordan and Evermann, 1896a, p. 146; Truitt, Bean, and Fowler, 1929, p. 36 (Western specimens of gyrinus possibly distinguishable). Schilbeodes gyrinus sialis (Jordan).—Greeley, 1936, p. 84 (probably valid form) ; 1940, p. 76 (ecology; 58 records [not listed], L. Ontario watershed, N.Y.).— Odell, 1940, p. 95 (records, L. Ontario drainage, New York).—Senning, 1940, pp. 104-105 (records, L. Ontario drainage, N.Y.). *Material designated by an asterisk has been re-examined. 42 U.S. NATIONAL MUSEUM BULLETIN 282 Noturus nocturnus Jordan and Gilbert [misidentifications]—Jordan and Gilbert, 1886, p. 18 (Rio Lampasas, Belton,* Tex.)—Evermann and Kendall, 1894, pp. 80-96 (Texas distribution and records, in part: Lampasas R., Belton;* San Antonio Springs, San Antonio;* Hunter Cr., near Houston). Schilbeodes nocturnus (Jordan and Gilbert) [misidentifications]—Forbes, 1909, p. 398 (Illinois distribution, in part).—Forbes and Richardson, 1909 [and 1920, pp. Ixxviii-lxxxviii], pp. Ixxiii-Ixxxiii, 199 (range and Illinois distribu- tion, in part [Camp Cr., Dallas City,* Henderson Co.]).—Eigenmann, 1919, pp. 398-399 (mouth of underground river, San Marcos, Tex.; ‘‘other places in Texas,”’ in part).—O’ Donnell, 1935, p. 484 (Camp Cr.,* Henderson Co., Il.). Ameiurus natalis (LeSueur) [misidentification]—Evermann and Kendall, 1894, pp. 78, 96 (Garman’s record of Noturus flavus from San Antonio,* Tex.). ?Schilbeodes punctatus.—Mitchell, 1904, pp. 154, 405 (nomen nudum; lapsus; oral breathing valves described). Schilbeodes mollis (Hermann) [misidentifications]—Hubbs and Raney, 1944, pp. 1-26 (synonymy, discussion, and description, in part).—Bailey and Harrison, 1945, p. 68 (ecology; Clear L., lowa).—Gerking, 1945, pp. 13, 74, map 63 (ecology; Indiana distribution [see p. 44]).—Lagler and Salyer, 1945, p. 161 (in food of gartersnake, Thamnophis s. sirtalis).—Carpenter and Siegler, 1947, pp. 6, 53, fig. 45 (compiled).—Hubbs and Lagler, 1947 [and 1949], pp. 71-73, fig. 175 (ecology; comparison; specimen, Jackson Co., Michigan figured; range [Montana, dubious]).—Dymond, 1947, p. 23 (comparison; Canadian distribution).—Fischthal, 1947, pp. 162, 181 (parasites, Wisconsin localities) —Eddy and Surber, 1947, pp. 171-182, diag. 7E, fig. 29 (figured; teeth figured; range; description).—Carlander, 1948, p. 272 (Oak Island and at mouth of Rainy R., L. of Woods, Minnesota area).—Trautman, 1948, pp. 166-173, pl. 1, figs. 3, 6, and 9 (description; tribs. of L. Erie, Ohio; hybridizes with Schilbeodes miurus).—Hooper, 1949, pp. 34-38 (age of popu- lation by vertebrae; Demming L., Minn.).—Cooper and Washburn, 1949, pp. 27, 31 (Merkle L., Mich.; survived heavy winter kill).—[McCabe], 1949, pp. 41, 235 (Howe Pond,* Mass.).—Morgan, 1949, p. 96, fig. 60 (Howe Pond, Spencer, Mass.).—Leonard and Leonard, 1949, p. 304 (Birch Cr., Michigan).— Dickinson, 1949, p. 25 (near Gainesville, Florida).—Lincicome and Van Cleave, 1949, p. 426 (Leptorhynchoides thecatus, a parasite, in specimens from Minnesota, Ohio, and Wisconsin).—Harrison, 1949, p. 339 (Des Moines R. basin, Iowa).—C. B. Nash, 1950, p. 562 (ecology; L. Erie, Long Point, Ohio).—Moore and Cross, 1950, pp. 139-142 (ecology; description; Oklahoma and Swan Cr., Ohio specimens compared; Oklahoma records: [not in Illinois R.?]; oxbow lake,* Poteau; Clear Boggy R.,* Pontotoc Co.; slough along Little R., McCurtain Co.; Mountain Fork R., near confluence with Little R.*).—R. F. Smith, 1950, pp. 63-136, 172, fig. ‘29 and 1953, pp. 21-135, 183-184 (lake and pond records, New Jersey).—Moore and Paden, 1950, pp. 87, 93 (Oklahoma [not Illinois R.]).—Carlander, 1950, p. 109 (compiled) .— Raney, 1950, pp. 170-190 (ecology ; reproduction; coastal plain form; reported, James R. system, Va.).—Reid, 1950, pp. 177-178 (ecology; Orange L., Fla.).— Bailey, 1951, pp. 194, 224 and in reprint, 1951, pp. 194, 226 (comparison; Iowa).—Harlan and Speaker, 1951, p. 97, pl. 18 (Iowa; distribution; natural *Material designated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 43 history).—Legendre, 1951, p. 3 (Quebec).— Mansueti, 1951, p. 302 (Maryland, in associations).—Frey, 1951, pp. 8-39 (description; Lakes Jones, Salters,* Singletary, White, Black, and Waccamaw,* N.C.).—Howden and Mansueti, 1951, p. 95 (compiled).—Hall, 1951b, pp. 14, 17 (L. Murray, Carter* Co., Okla.).—Hooper, 1951, pp. 464-477 (Demming L., Minnesota; introduced in 1940; a great weight of specimens taken after poisoning).—Cross and Moore, 1952, p. 406 (Oklahoma; relisted).—Fowler, 1952, p. 108 (relisted).— G. Moore, 1952, p. [6] (Oklahoma).—Jurgens and Hubbs, 1953, p. [3] (Texas).—Keleher, 1953, pp. 171-172 (distribution; size; Manitoba records: junction of Red and Assiniboine Rivers, Winnipeg; L. Winnipeg, Black Bear Island, 51°46’ N. Lat.).—Freeman, 1953, p. 269 (Savannah R. basin, Aiken* and Barnwell* Counties, S.C.).—Mansueti and Elser, 1953, p. 118 (associa- tions; Chamber’s L., Federalsburg, Md.).—Legendre, 1953, pp. v, xi, 29, fig. 59 (synonymy; comparison; Québec, Canada).—Fischthal, 1953, p. 101 (parasites, Wisconsin).—Bick, Hornuff, and Lambremont, 1953, p. 230 (Louisiana record).—Knapp, 1953, pp. 77-78, fig. 107 (comparison; range) .— Raney and Massmann, 1954, pp. 426-431 (throughout tidewater section, Pamunkey R., Va.; most abundant in coves).—Hall, 1954, pp. 61-62 (doubt- ful, Illinois R., Okla.).—W. B. Scott, 1954, p. 69 (figured; characters; size; food; distribution in E. Canada).—Langlois, 1954, pp. 208, 281 (food; Ohio records).—Cleary, 1954, p. 633 (Shellrock R. and Cedar R. basin, Iowa).— Harrington, 1954, p. 529 (death from carp egg diet).—Legendre, 1954, pp. 8, 14, 43, fig. 59 (key; Québec).—Freeman, 1954, pp. 138, 145, 147-151 (records, in part, Savannah R. drainage, South Carolina: stations 3,* 33,* 35,* 37,* 61*).—Harmic, 1954, p. 50 (Christina Cr., Del.) —Carr and Goin, 1955, pp. 21, 64, pl. 12 (Florida distribution).—Bond and Bisbee, 1955, p. 56 (Oregon and Idaho* records).—R. F. Smith, 1957, pp. 12-189, fig. 17 (lake and pond records, New Jersey; used for bass bait).—Anderson and Freeman, 1957, pp. 104, 109 (Congaree drainage records* [all re-examined]).—Crittenden, 1958, p. 218 (Bay Co., Fla. records). Noturus mollis (Hermann) [misidentification].—Patrick, 1961, p. 256 (Escambia R.). Schilbeodes leptacanthus (Jordan) [misidentifications]—Fowler, 1945, p. 271 (Murdock,* Fla.).—Freeman, 1954, pp. 138, 145 (Savannah drainuge, S.C., station 50* only). Noturus leptacanthus Jordan [misidentification].—? Gunter and Hall, 1965, pp. 23, 49, 57 (Caloosahatchee R., near Fort Myers, Fla.; salinity 0.22 ppt.). Schilbeodes insignis (Richardson) [misidentifications]—Harrison and Speaker, 1954, p. 519 (compiled records: Big Sioux* and Floyd* Rivers; Little Sioux R.;* ?Boyer R., all Iowa). OTHER MATERIAL STUDIED UNITEDSTATES: Atssama: CU 15837, 19281 (Hurricane Cr., 5 mi. E. Holt, Tuscaloosa Co.); USNM 36713 (Black Warrior R., near Tuscaloosa), 200474 (Cahaba R., near Centreville, Bibb Co.); UMMZ 97800, 111204 (vicinity of Au- burn); TU 2619, 19409 (Cahaba R., 8.5 mi. N. Centreville, Hwy. 27, Bibb Co.), 24553. ARKANSAS: UMMZ 128448, 128559; UM MZ (Delavan and Creaser no. 31- 13); USNM 165847; TU 2226; CU 35675. District or CoLtumBi1a: UMMZ (1U no. 6458); USNM 87491. Fiorina: UMMZ 86759 (Paradise Key, Dade Co.), 87896, 92158-9, 92162 thru 92164, 110471, 110669, 136506, 138453, 139232, *Material designated by an asterisk has been re-examined. 298-943 O—69——-4 44 U.S. NATIONAL MUSEUM BULLETIN 282 155060, 155499, 158178, 158549, 158638, 165062, 165113; UMMZ (Rogers coll., 2 lots); CU 3991, 12029, 12786, 15919 (Tamiami Canal, 40 mi. W. Miami), 16732; CNHM 11917; USNM 82246, 88590, 89360, 92909, 133250, 133366, 133390, 133417, 134065, 196731; UMML 4753, 5963, 11356, 13038, 13705, 14055, 16095 (pool, 5 mi. N. Coot Bay Pond, Everglades National Park). Grorc1a: UMMZ 88624, 88669, 134635, 138732, 138740; UMMZ (Fletcher nos. 16-Oconee R., Princeton, Clarke Co.; 31); Tulsa U 6; CU 214, 517, 15457, 17195, 17208, 17262; TU 12160 (Vidlery Cr., at Chattahoochee R., Roswell, Fulton Co.). Ipano: UMMZ 136204 (Snake R., Homedale bridge, Homedale, Owyhee Co.). IL.1- NoIs: UMMZ 105921, 111669, 130053, 135599, 142024, 147029; UMMZ (G. V. Harry coll.; Bauman nos. 42, 68, 73, 75, 76, 77, 79; Mississippi River Survey colls., from Andalusia, Cordova, and Oquawka); CNHM 1380, 1470, 13911 thru 13918, 43268, 43298, 43317; INHS 51, 193, 211, 217, 1099, 2801, 5056, 5057, 5068, 7119, 11884, 13404, 13941, 24721, 24947, 24987, 24990, 24996, 24998, 26023, 26028, 2°090, 26340, 26555, 26690, 26709, 27562, 27600, 27607, 27608, 27610, 27628, 27648, 27755, 28044, 28047, 28077, 28079, 28085, 28089, 28091 thru 28093, 28096, 28099, 28124, 28125, 28128, 28130, 28138, 28148, 28150, 28160, 28161, 28174, 28177, 28180, 28186, 28201, 28225, 28255, 28485, 28486, 28499; INHS (other nos. 484, 3090, 6233); UL 8430, 8991, 9065, 9074, 9102, 9124, 9146; USNM 15266, 201323. Inp1ana: USNM 64941, 64944, 66794 thru 66797, 67769, 68994, 120939, 196701, 196729, 199584; IU 15, 30, 36, 131, 138, 141, 154, 269, 305, 327, 357, 359, 362 thru 364, 366, 425, 446, 461; CNHM 2869, 6401, 7295, 10594, 16837 thru 16839, 42466; UM MZ 66620, 66965; UL 8092, 8104, 8295, 8852, 8873. Iowa: UMMZ 100895, 100942, 101083, 101382, 101553, 142414, 146013, 146839, 146887; UMMZ (Mississippi River Survey colls., from Sabula, Burlington, and near Dubuque); CNHM 1025, 1220, 2109; USNM 54986, 61937, 125567, 174928, 196733. Kansas: KU 3735, 3864, 3890 (Marais des Cygnes R., sec. 6, T. 17 S., R. 20 E., Franklin Co.). Kentucky: UMMZ 121614; UL 5072, 5274, 5495, 5610, 7048, 7118, 7905, 7931, 8770, 9615, 9862; USNM 199589. Lovurstana: UMMZ 165879, 170525, 170837, 184286, 184291; UMMZ (Delavan and Creaser no. 31-53; Taylor nos. 54—3, 55-43, 55-46, 56-38, 56-42, and from L. Bruin, L. St. John, Buck Bayou-T. 9S8., R. 15 E.); USNM 172332, 172655, 172683, 172728, 172786, 172843, 172872, 172997, 173017, 173055; CU 13949; TU 1336, 3363, 3818, 4259, 4568, 5787, 6370, 7791, 11804, 13626. Maryitanp: USNM 30238, 61592, 68165, 74121, 85348, 85691, 89221, 90546, 100246, 100664—100665, 103863, 120967, 192647, 196682 thru 196690, 196730; CU 18697; UMMZ 89394, 136025, 137801; UMMZ (field no. H33-4). Massacuusretts: CU 20590 (Sugden Reservoir, Spencer). Micutgan: UMMZ 55242, 55388, 56041, 56262, 56611, 60014, 60025, 60270, 61742 (Otsego L., Otsego Co.), 61823 (Farm L., Au Sable R.., Otsego Co.), 65668, 66664, 67447 (North Br. Devil R., Alpena Co.), 67486 (Snyder or Mindack L., Alpena Co.), 67516 (Zim L., W. of Alpena, Alpena Co.), 71627 (Devil L., Alpena Co.), 73308, 80674, 81832, 83310, 84845, 85353, 85368, 85395, 89932, 89994, 90127, 90247, 90986, 91003, 91010, 91030, 91067, 91146, 91235, 98003, 98024, 98084, 98131, 101894, 110253, 110327, 110366 (Bear L., Manistee Co.), 110373 (Lemon L., Manistee Co.), 111313, 113160, 116236, 116312, 116339, 116361, 116945, 117019, 117163, 117181, 129157, 129184, 133568, 133582, 136740, 136765, 136807, 136972, 136993, 137013, 137035, 137176, 137188, 137216, 137252, 137271, 137654, 137705, 137737, 137750, 138002, 138056, 138125, 138852, 138863, 139806, 139819, 140157, 140186, 145086, 164283, 164426, 165856 thru 165858, 165861 thru 165864, 165867 thru 165869; UMMZ (Kalamazoo River Survey nos. KA10, KA11, KA19, KA26, KA28, KA29, KA30, KA37, KA89, KA45; field no. S— 1-Hart L., Calhoun Co.; Peterson coll.-Homer Mill Pond, Calhoun Co.; Crowe CATFISH GENUS NOTURUS RAFINESQUE 45 coll.-L. St. Clair, Bouvier Bay, St. Clair Co.; Hankinson nos. 9379J, 9410J, 9414S) ; USNM 68774, 174905, 193202. Minnesota: USNM 86555; UMMZ 80081 (Cloquet R., near mouth, St. Louis Co.), 94835 (Mississippi R., between Cass L. and Bemidji, Beltrami Co.), 97260. Musstsstpp1: UMMZ 113452, 113878, 113894 (Luxapalilla Cr., 5 mi. N. Columbus, Loundes Co.), 157740 (Tombigbee R.., 2.5 mi. N. of Amory, Monroe Co.), 163716; UMMZ (Delavan and Creaser no. 31-74, trib., Noxubee R., 2 mi. E. Macon, Noxubee Co.; Hutchins coll., Bluff L., near State College; Walker nos. 39-31, 39-32, 39-41, 39-46; Stickel coll.) ; AF 3332 (Leaf R.., Scott Co.), 5354 (Trim Cane Cr., Oktibbeha Co.); USNM 165965, 165983, 175384 (Wheeler Cr., N. of Baldwyn, Prentiss Co.), 175285, 175386, 175387 (creek, 0.5 mi. E. Smithville, Monroe Co.); TU 3042, 3756, 4763, 14102; Missis- sippi Game and Fish Commission (Enid Reservoir, Yalobusha Co.; Lake Beulah, Bolivar Co.). Missourt: UMMZ 147169, 148328, 148683, 148732, 148805, 149015, 149050, 149430, 149436, 149456, 149902, 149935, 150199, 150277, 150588, 152504, 152619, 152638, 152953, 152975, 152992, 153086, 153108, 153129, 153152, 153196, 153234, 153250, 153272, 164825 (Missouri R., near St. Joseph, Buchanan Co.); CNHM 754. Nespraska: UMMZ 133947, 134152 (Cedar R., 1 mi. SE. Erickson, Wheeler Co.), 134711, 134746, 134910 (South Fk. Nemaha R., 1 mi. E. and 1 mi. S. Humboldt, Richardson Co.), 135292 (Elkhorn R., 5 mi. W. Stuart, Holt Co.), 135747 (Cache Cr., SW. of Ewing, Holt Co.). New JERSEY: UMMZ 99214, 109536, 109803, 111095, 114839; CU 5345, 5355, 7406. NrEw York: CU 2100, 2334, 2349, 3580 (Oneida L., near Fredericks Cr., Oswego Co.), 5230 (Glenmere L., Orange Co.), 5429, 5493 (Greenwood L., Orange Co.), 5892, 6070 (lagoon of Fall Cr., Ithaca, Tompkins Co.), 6115 (Duck L., Oswego Co.), 9867, 14343, 18356, 20398 (trib. of barge canal, New London, Oneida Co.) ; USNM 64300 through 64303, 196732; UMMZ 99092, 99202, 114199 (Hackensack R., 3 mi. SW. Nyack, Rockland Co.), 114331 (Rutgers Cr., 0.5 mi. NW. West- town, Orange Co.). NortH Carotina: UMMZ 156984; USNM 94236, 118950, 191016, 191038, 191090; CU 4008, 9880, 9888, 11512, 11582, 11688, 14106, 16810, 17011, 19565, 19582, 19853, 41801; UMML 21264. Norra Dakota: UMMZ (Hankinson nos. 9960-Pembina R., Pembina, Pembina Co.; 9961—-Red R.., Pembina, Pembina Co.; 9964-Red R., Grand Forks, Grand Forks Co.); USNM 1508 (Maple R., of North Red R., near Lat. 47° N.), 69295 (Sheyenne R.). Ou1o: UMMZ 87461 (Pymatuning Cr., S. Vernon Twp., Trumbull Co.), 87465 (Pymatuning Cr., SE. Wayne Twp., Ashtabula Co.), 107717 (trib., Little Salt Cr., Liberty Twp., Jackson Co.), 109283 (Sunfish Cr., Newton Twp., Pike Co.), 118150, 118159, 118324, 118329, 118499, 121822; SU 932; USNM 58726, 62852. OxLaHoma: UMMZ 167196 (Bois d’Are Cr., S. of Ada, Pontotoc Co.); Tulsa U (slough along Little R., U.S. Hwy. 70, between Idabel and Broken Bow, McCur- tain Co.); KU 2405 (Lake Murray, Carter Co.). Pennsytvania: ANSP 47354, 48486 thru 48488; USNM 58728 (Erie, Erie Co.). Soura Carotina: CU 15159, 15218, 15319, 15381, 15389, 15712; CM 33.257.3, 33.309.12; USNM 25614, 92204, 162523, 162562, 192682 thru 192734, 200477; UMMZ 143192, 143199. Sourn Daxora: SU 4525 and USNM 76122 (Mitchell, Davison Co.); SU 34889. TENNESSEE: UMMZ 105398; UMMZ (Delavan and Creaser no. 31-11). Texas: UMMZ 110562, 110588, 111008, 120150 (White Rock L., Dallas), 120261, 129782, 129982, 162070, 162084 (Lake Corpus Christi State Park, near Mathis, San Patricio Co.), 166017, 166041, 167220, 170031, 170319, 170395, 170465; UMMZ (Bonham coll.); TNHC 517, 1196, 1580; USNM 166089, 166166, 196670; TU 2980, 14019, 14062, 14084, 21378, 21979, 24591 (Nueces R., 11 mi. W. Batesville, Hwy. 76, Zavala Co.); KU 3609 (near Henrietta, Clay Co.). Vireinia: CU 14403, 14431, 16750, 16886; USNM 85595, 85697; UMMZ 102333. 46 U.S. NATIONAL MUSEUM BULLETIN 282 Wisconsin: UMMZ 64512, 64531, 64649, 64898, 72379, 73568, 73644, 73702, 73717, 73803, 74209, 74347, 74778 (outlet Mary L., Marinette Co.), 74836 (outlet Nocquebay L., SW. of Crivitz, Marinette Co.), 76164, 76691, 77058, 77142, 77630, 77773, 77789, 77875, 78096, 78126, 78210, 78257, 78297, 78310, 78353, 78395, 78448, 78461, 95976, 95984, 96005, 96078, 96175, 96214, 96263; SU 14985; CNHM 35648-9. CANADA: Manitospa: UMMZ 180538 (Millers Cr., sec. 34, T. 30, R. 5 E. Prin. mer.). ONTARIO: UMMZ 130869, 130915, 130939, 130980. QurBEc: UMMZ 136396 (L. St. Louis of St. Lawrence R., at Maple Grove, near Lariviere dock, Montreal). Type.—Silurus gyrinus was described from the Wallkill River. Jordan (1877d, p. 71) did not list the type-specimen among those that he examined, and it is presumably not extant. Dracnosis.—Noturus gyrinus, as well as N. erilis and N. lachnern, in the subgenus Schilbeodes has a terminal mouth (jaws subequal) and typically ten preoperculomandibular pores over much of its range. In contrast to N. exrilis the pectoral spine is deeply grooved and nonserrated, and there are modally eight pelvic rays, except in the central Gulf drainage. From UN. lachneri, gyrinus is characterized by the separation of the anterior ends of the infraorbital and supra- orbital canals (2 internasal pores) and fewer fin rays and vertebrae. Also characteristic are the high procurrent caudal rays, many of which may be branched and a low (1.5, usually 1.2 or less) ratio of the distance from the end of the adipose fin to the tip of the caudal fin stepped into the distance from the origin of the dorsal fin to the end of the adipose fin. Except for the lower, more lightly pigmented surfaces, the body and fins are nearly uniformly colored. Description.—Other counts and measurements are given in tables 17 to 26. Body shape somewhat variable, usually short and chubby, deepest at or in front of dorsal fin, but some individuals, especially in the southern states, are very elongate; head rather deep; skull arched and rounded above; lower jaw terminal; premaxillary tooth patch with lateral edges and posterior corners rounded (not angulate), the overall lateral dimension of the band of teeth notably greater than in other species of Noturus, about 3 or 4 times the anterior- posterior measurement; humeral process distinct, about as long as the width of the shaft of the pectoral spine, its tip usually turned slightly inward; eye very small, 3.5 to 5.0 times in snout; pectoral spine without serrae, but frequently with some smooth bumps on its posterior edge near base; the spine angulate in cross section, very deeply grooved, their depth nearly that of the spine, their length variable, but along most of the spine; imbedded in these grooves and surrounding the spine is a mass of grayish white, flaky material; dorsal spine stout; adipose fin high, widely united to the long procur- rent caudal rays, many of which may be branched; caudal fin some- CATFISH GENUS NOTURUS RAFINESQUE 47 times truncate behind, frequently rounded, but usually ending in an obtuse point, formed by the extreme elongation of the middle caudal rays. The numerous branched caudal rays are very characteristic of this species. Their large number results from splitting or branching of the ends of the long procurrent and principal rays. The number of branched rays in 143 specimens from throughout the range varies from 15 to 40; there seems to be no definite geographic trend. Speci- mens from Michigan average 30.4 branched rays, some from New Jersey 23.8, Indiana specimens 21.7, and samples from Missouri, Texas, Alabama, Florida, and North Carolina average between 27.5 and 31.3. Branched rays in the lower lobe of the caudal fin similarly are low in number in the samples from New Jersey, Maryland, and New Hampshire, averaging 11.1; other samples, as indicated above, average from 12.5 to 13.9; the total mean is 13.3, and the range 9 to 17. Branched rays in the upper lobe of the fin are a little more variable, with means as follows: New Jersey, Maryland, and New Hampshire, 12.3; Great. -akes basin, 16.6; Indiana, 9.4; North Carolina, 16.0; Missouri, 13.6; Texas, 16.8; Alabama, 17.5; Florida, 17.1. The total range is 6 to 23 and the mean 14.3. As a result of the excessive branching of the long rays, the number of simple rays is reduced. The upper simple rays vary from 8 to 26; they may be fewer in number northward; the overall average is 16.9. The lower simple rays seem to increase slightly in number south- ward. They range from 8 to 13 in Michigan (mean 9.8), 10 to 15 in Indiana (mean 12.6), 9 to 15 in Missouri (mean 12.3), 10 to 16 in Texas (mean 12.2), and 12 to 17 in Florida (mean 14.5); the extremes of all samples tabulated are 8 and 18, and the mean is 12.7. The soft dorsal rays vary geographically; they range from four to seven. Skeletons and cleared and stained specimens from the Great Lakes basin, Louisiana, and North Carolina indicate that the ossified pectoral radials (actinosts) are fused (in 30 specimens) on each side; one specimen has ten vertebrae anterior to the anal fin origin and ten have eleven. There are usually ten or fewer gill rakers on the first arch; they range from five to ten in Florida and from seven to nine in Michigan. The largest specimen examined is 105 mm. in standard length. In the same series of 17, USNM 199589, from Goose Creek, near Louisville, Kentucky, are four others ranging from 91 to 98 mm. in standard length. Otherwise, specimens rarely attain lengths of 90 mm. The color in life varies from a dull golden yellow to olive gray. Preserved specimens are usually gray or brown; some are nearly black. The fins, barbels, and upper body surfaces are nearly uni- formly pigmented, but the barbels and fins may be darker or lighter 48 U.S. NATIONAL MUSEUM BULLETIN 282 than the general tone of the body. A dark gray axial streak is present; the lower body surface is variously unpigmented in young, lightly or irregularly pigmented in older specimens. VaRIATION.— The division of this species into eastern and western subspecies or species as postulated by some authors (see synonymy), does not appear to have foundation. It is a wide ranging species that has some local differentiation, but which shows strong north to south gradients. Some of these variations have been mentioned above. Localized variation in the number of dorsal rays exists; gradients in caudal, pelvic, pectoral, and possibly anal rays and preoperculo- mandibular pores are evident. Specimens from northern glaciated areas are heavy bodied and have relatively short spines; those from more southern waters are emaciated and have long hard parts. No trend in internasal pores is evident. The following summarizes in order the number of tabulations, range (in parentheses), and mean for combined samples from (a) Michigan, (b) New Jersey and New Hampshire, (c) North Carolina, (d) Indiana, (e) southeastern Missouri, (f) Texas, (g) Black Warrior River, (h) Florida, and (i) the total for all specimens. Lower-half caudal rays: (a) 34 (22-25) 23.76; (b) 20 (22-28) 25.20; (c) 4 (24-28) 26.00; (d) 29 (23-27) 25.03; (e) 30 (24-28) 26.20; (f) 25 (23-29) 26.00; (g) 4 (26-29) 27.50; (h) 99 (25-32) 28.25; (i) 247 (22-32) 26.50. Upper-half caudal rays: (a) 34 (27-31) 29.44; (b) 21 (27-32) 29.67; (c) 4 (28-31) 29.75; (d) 29 (28-34) 30.93; (e) 30 (28-34) 31.00; (f) 26 (30-36) 32.15; (g) 4 (31-33) 32.25; (h) 100 (30-36) 32.75; (i) 250 (27-36) 31.48. Anal rays: Lake Huron drainage 52 (13-16) 14.65; Lake Erie drainage 13 (14-16) 14.85; Lake Michigan drainage in lower Michigan 12 (13-16) 14.75; Michigan total 77 (13-16) 14.70; New Jersey 19 (15-17) 15.68; North Carolina 4 (16-17) 16.25; Indiana 30 (13-16) 14.73; Missouri 30 (12-16) 14.63; Texas 26 (13-17) 15.04; Florida Parishes, Louisiana 2 (17-18) 17.50; Tombigbee River 14 (15-18) 16.36; peninsular Florida 103 (13-18) 15.50; and total 310 (12-18) 15.17. The other small samples fall in these ranges. Preoperculomandibular pores: Lake Huron drainage 136 (9-12) 10.23; Lake Erie drainage 48 (9-12) 10.42; Lake Michigan drainage in lower Michigan 36 (10-11) 10.47; Michigan total 220 (9-12) 10.31; New Jersey 38 (8-11) 9.79; North Carolina 8 (10) 10.00; Georgia 4 (10) 10.00; Indiana 62 (9-12) 10.21; Illinois 32 (10) 10.00; Missouri 60 (10-11) 10.10; Oklahoma 34 (9-11) 10.03; Texas 108 (9-11) 10.04; Florida Parishes, Louisiana 4 (11) 11.00; other Louisiana 20 (9-11) 10.00; Mississippi drainage, Mississippi 12 (10-11) 10.17; CATFISH GENUS NOTURUS RAFINESQUE 49 Pascagoula River 18 (10-11) 10.94; Biloxi Bay drainage 18 (9-11) 10.56; Bay St. Louis drainage 4 (10) 10.00; Tombigbee River 54 (9-12) 10.91; peninsular Florida 206 (8-12) 10.04; Escambia River 66 (10-11) 10.06; and total 976 (8-12) 10.19. Soft dorsal rays: Most Michigan specimens and a sample from the Ohio River, Indiana, have five rays more frequently than six; two from New Hampshire and six of fourteen from Illinois have five rays; otherwise the samples tend strongly toward six soft dorsal rays. The following summarizes most of the data: Lake Huron drainage 111 (4-6) 5.32; Lake Erie drainage 47 (5-6) 5.40; Upper Peninsula, Michigan 8 (5-6) 5.13; Lake Michigan drainage, Lower Peninsula, Michigan 44 (4-7) 5.61; Michigan total 210 (4-7) 5.40; Indiana 31 (5-6) 5.48; New Jersey 16 (6) 6.00; North Carolina 4 (5-6) 5.75; Missouri 30 (5-6) 5.87; Oklahoma 10 (5-6) 5.70; Texas 49 (5-7) 5.98; Louisiana 12 (6) 6.00; Biloxi Bay drainage 9 (6) 6.00; Tombigbee River 20 (6-7) 6.05; peninsular Florida 102 (5-7) 5.97; Marianna, Florida 6 (6) 6.00; and total 521 (4-7) 5.69. Vertebrae: No prominent variational trend in vertebrae is suggested; however, the higher number in the central Gulf drainage specimens seems to correspond with an increase in numbers of other parts. Summaries by areas: Great Lakes basin 27 (33-36) 34.33; New Jersey 13 (35-36) 35.46; Nebraska 6 (32-34) 33.33; Indiana 31 (33-36) 34.19; Missouri 27 (33-35) 33.63; North Carolina 34 (34-36) 34.71; Texas 26 (33-35) 34.15; lower Mississippi drainage 2 (33-34) 33.50; Florida Parishes, Louisiana 2 (36-37) 36.50; Tombigbee River 5 (36-37) 36.60; Cahaba R., Alabama 2 (36-37) 36.50; peninsular Florida 19 (82-35) 33.37; and total 194 (32-37) 34.30. The geographic variation in caudal, pelvic, and pectoral rays is shown in tables 2 and 3. Gradients in these and in the general mor- phology follow the geographic trend suggested by Hubbs (1940, p. 209). Summarizing the data, Atlantic coastal populations of gyrinus appear to be much like those from the Mississippi Valley from approxi- mately the same latitude, as there seems to be a gradient in most characters from north to south in both areas. The number of soft dorsal rays is reduced in the Great Lakes basin, but six, the normal for Noturus is usually attained elsewhere. There are fewer pelvic, soft pectoral, and caudal rays and fewer preoperculomandibular pores in the north; all of these increase in number southward, but the greatest increase appears to be in the central Gulf drainage, notably the Mobile Bay drainage, Alabama. This is unexplained. The Mobile drainage population does not appear to be isolated. The increase, therefore, may result from recent genetic fusion within the drainage of popula- tions invading from the east and west. There seems to be no important 50 U.S. NATIONAL MUSEUM BULLETIN 282, geographic trend in color; some specimens are nearly black, but in preservative most are slate gray. A rather obvious trend in the relative lengths of spines and robust- ness of the body exists. Most specimens from the glaciated area of the north central states are plump or fat and have very short spines. Their digestive tracts are literally stuffed with insects. Specimens from Houston, Texas, are slender and emaciated; they have a narrow caudal peduncle and very long spines. Their digestive tracts are empty. Most other long spined individuals from the lower Mississippi Valley and Florida that were examined contain little food. Occasional fat, short spined individuals from the same regions are greatly dis- tended by food, chiefly insects. It is believed that the amount of avail- able food directly controls the relative growth of the framework parts. Where little food is available, the hard skeletal parts grow at the ex- pense of the soft tissues; where much food is available, all parts grow uniformly. No quantitative study of the amount of food utilized is available, but comparative growth rates of parts on different quantities of food should be determined. As an indication of the variation, the following stepped proportions are given (number, range, and mean in sequence) from: (a) New Jersey, (b) Michigan, (c) Missouri, (d) Texas, and (e) Florida. Least peduncle depth in predorsal length: (a) 15 (2.1-3.0) 2.6; (b) 15 (2.3-2.7) 2.5; (c) 15 (2.4-3.0) 2.7; (d) 20 (2.5-3.0) 2.7; (e) 21 (2.6-3.1) 2.9. Pectoral spine length in predorsal length: (a) 24 (2.4-3.7) 3.0; (b) 26 (2.43.3) 2.9; (c) 30 (1.9-2.6) 2.3; (d) 39 (1.6-2.5) 2.1; (e) 41 (2.0-2.8) 2.3. Pectoral spine length in peduncle depth: (a) 24 (0.8-1.7) 1.2; (b) 26 (0.8-1.5) 1.2; (c) 30 (0.7-1.0) 0.8; (d) 39 (0.5-0.9) 0.7; (e) 40 (0.6-1.0) 0.8. Dorsal spine length in predorsal length: (a) 14 (3.2-4.5) 3.9; (b) 15 (2.9-4.5) 3.5; (c) 14 (2.2-3.4) 2.8; (d) 20 (2.0-3.0) 2.6; (e) 20 (2.3-3.5) 2h Standard lengths (mm.): (a) 15 (24-60) 39.4; (b) 15 (32-60) 44.5; (c) 15 (83-47) 38.4; (d) 20 (88-50) 44.1; (e) 20 (35-60) 42.5. Distripution.—Noturus gyrinus (map 1) is found from Saskatch- ewan, Manitoba, the Dakotas, Nebraska, Kansas, Oklahoma, and Texas eastward to the Atlantic coast, New York, and the Saint Lawrence River system. It has been introduced elsewhere. It avoids the uplands of the Ouachita Mountains of Arkansas, the Ozark Plateau of Arkansas and Missouri, the Appalachian Highlands, and the Piedmont Plateau. On the Atlantic coast it occurs chiefly below the Fall Line (approximately the line indicated on map 1) and ranges CATFISH GENUS NOTURUS RAFINESQUE 51 @ Noturus gyrinus O- Noturus lachneri Map 1.—Distribution of Noturus gyrinus (Mitchill) and Noturus lachneri, new species. The symbols represent localities from which specimens have been examined. The dashed lines outline the known and hypothetical range of Noturus gyrinus. One line, extending from Texas to Saskatchewan eastward to Quebec and New York, bounds its general range. A second line surrounds the upland region of the eastern United States, an area from which the gyrinus is not known. All unverified literature records that seem to be based on gyrinus are from localities within the range as outlined. N. gyrinus has been introduced into at least two places in New England (shown here), into the Snake River basin of Idaho and Oregon, and into Europe. from the Hudson River, New York, to the tip of peninsular Florida. On the Gulf coast, it is found mostly below a similar line and ranges westward to the Balcones Escarpment and the Nueces River, Texas. N. gyrinus is found throughout much of the lower Red River system, upstream at least to Lake Altus, southwestern Oklahoma (personal communication from Dr. George A. Moore). The single specimen from a lake near Poteau, Oklahoma, in the Arkansas basin, may be the result of an introduction. From the lower Mississippi Valley, 52 U.S. NATIONAL MUSEUM BULLETIN 282 N. gyrinus funnels northward through the lowlands into much of the northern Ohio and upper Mississippi Valleys, and into the Missouri system, where it occurs throughout northern Missouri. It reaches a westward limit in eastern Kansas, Nebraska, and South Dakota. N. gyrinus is distributed in the Hudson Bay drainage from the Red River of the North in the Dakotas at least to Black Bear Island in Lake Winnipeg, Manitoba (Keleher, 1953), and has been reported from the Rainy River (Carlander, 1948; W. B. Scott, 1963) and the Souris River, Saskatchewan (Symington, 1959). In the upper Great Lakes basin, NV. gyrinus is known from the Saint Louis River system and tributaries to western Lake Superior, from the Lake Michigan basin in the Menominee River system, and occurs southward throughout most of the Lake Michigan and Lake Huron drainages of the Lower Peninsula of Michigan. It apparently avoids the northern part of the Lower Peninsula. It occurs in the drainages of the lower Great Lakes to Montreal, Canada, and has occupied the Mohawk and upper Hudson River systems of New York. Elsewhere, N. gyrinus has been introduced into the Merrimack River of New Hampshire, into the Connecticut River system of Massachusetts, and into the Snake River of Idaho and Oregon. Probably there were many introductions of this species along with bullheads (Jctalurus) into other bodies of water. Arnold and Ahl (1936) indicated that N. gyrinus was once introduced as an aquarium fish into Germany, but that it did not reproduce. Specimens of N. nocturnus, N. insignis, N. flavus, and, probably N. exilis have been reported as this species. NoMENCLATURE.—Contrary to the opinion of Hubbs and Raney (1944), I do not think that Hermann’s (1804, p. 309) description of Silurus mollis (reprinted in C. L. Hubbs, 1936, p. 125, and in Hubbs and Raney, 1944, p. 25) can be shown to apply to this species. Hence, their action in changing its name from Silurus gyrinis Mitchill is regarded as invalid. The description is so incomplete that Silurus mollis will probably forever remain a nomen dubium. Some information bearing on the understanding of the description (Hermann, 1804, p. 309) of Silurus mollis follows: 1. ““Observationes Zoologicae . . .”’ was edited and published in 1804 after Hermann’s death; thus the original description may not convey the meaning intended by Hermann. 2. This work has not been entirely overlooked. Several of the catalogs of the British Museum cite it in their bibliographies, but they rarely list or assign its new names. It may be that the descriptions are too incomplete or indefinite to identify. 3. Hermann was a member of the faculty of the University of Stras- bourg. According to Dr. Francois Gouin (personal communication) CATFISH GENUS NOTURUS RAFINESQUE 53 he left no figure or specimen of Silurus mollis in the Strasbourg collec- tion, and may never have handled the specimen. 4. The ‘“Musei Humphrediani” was ‘“‘A [sale] Catalogue of the Large and Valuable Museum of Mr. George Humphrey... .”’ It is dated 1779, and item 33 on page 131 reads “‘a globe fish, another American fish, and a snake, in three bottles.”’ [Copy seen in the British Museum of Natural History.] Lists of European natural history collections do not indicate the disposition of this collection, nor its present location, and I have failed to locate Hermann’s specimen. Humphrey apparently was not the collector of the fish; he was primarily a dealer in shells. 5. The meaningful characters in the description are: the name Silurus, a posterior dorsal fin that is adipose, twelve anal rays, and a very soft and flabby belly. In themselves, these are not diagnostic. In addition, it is stated that there are four cirri of the upper jaw and that the specimen came from America. America could mean either North or South America. The meaning of the four cirri of the upper jaw is un- certain, as no lower barbels are indicated for the specimen. The body of the description “radio primo . . . mollibus candidis,”’ is confused and apparently cannot be meaningfully translated. It indicates among other things something that is soft and (?) white and that a spine is associated with the head, neither character applying to Noturus gyrinus. To summarize, the description of Silurus mollis is indefinite; it may not apply to an ictalurid; and the specimen from which the description was made may be lost forever. Thus, until shown to be otherwise, I agree with Fowler (1945, p. 123) that the nam is unidentifiable; it should be removed from association with Noturus gyrinus. The original description of Silurus gyrinus as published by Mitchill (1817, p. 289) is perhaps inadequate, and led to some confusion among early workers. The later and more complete description given by Mitchill (1818, pp. 241, 322-323) unquestionably identifies his ma- terial with this long known form. Characteristic are the locality, the pectoral formula, the pectoral spine, the tail, and the easily overlooked adipose fin. The type of Noturus sialis Jordan has not been located and probably was destroyed by fire at Indiana University. The name, however, has long been correctly placed in the synonymy of Noturus gyrimus as confirmed by the descriptions and by Jordan’s action. According to Mitchell (1904), the name Schilbeodes punctatus is a lapsus for Schilbeodes gyrinus. Erymo.tocy.—The name gyrinus (Greek) means tadpole, in allusion to its resemblance to a larval salientian. RELATIONSHIP.—Among the species of Noturus, N. gyrinus appears to be more closely related to N. lachneri than to any other. Both 54. U.S. NATIONAL MUSEUM BULLETIN 282 have a relatively low number of paired fin rays, a uniformly dark color pattern, and the pectoral spines lack developed serrae. As a group, with N. gyrinus at one extreme and N. ezilis at the other, the three have in common a terminal or virtually terminal mouth and ten preoperculomandibular pores. ECOLOGICAL CONSIDERATION.—N. gyrinus is typically a species inhabiting quiet or slow running waters, being especially abundant in lakes and their outlets, sloughs, ponds, quiet backwaters, and in the oxbows and base-level mouths of streams. It prefers a soft muddy bottom with extensive vegetation and, according to Hankinson (1908), nests in cavities. It has been taken in collections with the following species of Noturus: leptacanthus, nocturnus, funebris, phaeus, insignis, exilis, flavus, eleutherus, stugmosus, furiosus, and miurus. REMARKS.—Two specimens in the Tulane University collection are identified as Noturus gyrinus, with some doubt. They are TU 3818, Big Branch of Hog Branch (Tickfaw drainage), 2.8 mi. west of Holden, U.S. Hwy. 190, Livingston Parish, Louisiana, and TU 11084, oxbow of West Pearl River, 0.5 mi. north of Yellow Lake Bayou, St. Tammany Parish, Louisiana. Counts from the specimens are included in the summaries of variational data, usually listed as Florida Parishes, Louisiana, and these are included in the tables comparing the species of Noturus. Also included in the summaries are the caudal ray counts of 22+8+13+17=60 for the first, and 20+13+15+16=64 for the Jatter specimen. The two localities are plotted on the map showing the distribution of gyrinus. Both specimens have a slightly included lower jaw. Otherwise the color pattern, body shape, wide caudal fin, short adipose fin, and spines seem typical of gyrinus. The number of fin rays, vertebrae, and pre- operculomandibular pores is as high or higher than data obtained for gyrinus from other portions of its range. The position of the lower jaw and the high counts suggest that the specimens may possibly be hybrids. If so, they may be crosses of Noturus gyrinus and Noturus nocturnus. Discounting the position of the lower jaw, they seem other- wise to be fairly typical gyrinus, but having a high average number of meristic characters. | Noturus lachneri, new species OUACHITA MADTOM PLATES 3 (FIG. 2), 5 (FIG. 2); Map 1 Typr.—USNM 201592 (holotype), collected from the Middle Fork of the Saline River at Arkansas Highway 7 crossing, 11.2 miles north of Mountain Valley, Garland County, Arkansas, May 27, 1967, by Leslie and Betty Knapp. CATFISH GENUS NOTURUS RAFINESQUE 55 PaRATYPES.—ARKANSAS: TU 7074 (1 specimen) and TU 7077 (2), South Fork Saline R., 3 mi. SE. U.S. Hwy. 70 or 7.7 mi. SSW. Owensville, Garland Co., Aug. 7, 1953, E. Liner. USNM 165901 (5) and UMMZ 187095 (2), trib. of Saline R. [presumably Salt Cr., sec. 34, T.158., R. 15 W.], just off U.S. Hwy. 70, about 1 mi. NW. Benton, Saline Co., Apr. 22, 1952, E. A. Lachner, F. J. Schwartz, and W. T. Leapley. DriaGnosis.—WNoturus lachnert is one of the three species of the subgenus Schilbeodes that are characterized by a terminal mouth, subterminal lower jaw, and ten preoperculomandibular pores. It is distinguished from Noturus gyrinus by the single internasal pore and a short head which is projected 3.7 to 4.2 times in the standard length. N. gyrinus seldom has more than 36 vertebrae or 17 anal rays but lachneri has 37 or 38 vertebrae and 16 to 19 anal rays. From the other species, Noturus exilis, N. lachneri is distinguished by lack of serrations of the pectoral spine, more caudal rays, and typically eight rather than nine pectoral rays. Description.—Other counts and measurements are given in tables 17 to 26. Body moderately elongate, about as deep posteriorly as anteriorly; head relatively flattened above, somewhat depressed; jaws about equal, mouth terminal; premaxillary teeth in a transverse band which is about 3.5 times as wide as long, its posterior corners rounded; dorsal spine stout, about half length of longest dorsal ray; pectoral spine nearly straight to very slightly curved, with moderate grooves, devoid of serrae anteriorly and posteriorly; adipose fin relatively short and of moderate height, without a free posterior flap, broadly connected to the procurrent caudal fin; a very shallow indentation at the junction of the two fins; procurrent caudal rays of moderate length, the posterior fin margin broadly rounded; eye small, 2.3 to 2.9 in snout; gill rakers on first arch seven to nine; posterior process of cleithrum (humeral process) nearly straight, slightly shorter than diameter of pectoral spine. The largest specimen, a male, is 69.5 mm. in standard length. All of the specimens have six soft dorsal rays. In the caudal fin, eight of the specimens have eight branched rays in the upper lobe, a count similar to that of Noturus erilis; these same specimens have 23 to 25 upper simple rays; the other specimens have 13, 20, and 22 rays with slight to deep branching in the upper lobe, and 18, 12, and 9, respectively, upper simple rays, a condition suggestive of the ex- cessive branching in Noturus gyrinus. In the lower lobe there are 11 to 14, mean 12.6 branched rays and 11 to 15, mean 13.6, simple rays. In a specimen cleared and stained there is one epural; twelve verte- brae are anterior to the anal fin origin; hypurals 1-3 (those supporting 56 U.S. NATIONAL MUSEUM BULLETIN 282 the lower caudal lobe) are fused, and hypurals 4-5 (in upper caudal lobe) are fused. The pectoral radials are fused on both sides. The color in life of the holotype was reported to be dark brown; when in the water it resembled a darkly colored young bullhead. No bright colors were present. The Tulane specimens are nearly transparent—evidently because of faulty preservation—and have virtually no pigment. The holotype, in preservation, is dark grayish brown above, grading to slightly lighter on lower side, and pale whitish below. Widely scattered chromatophores cover the entire lower surface of the head, an area about the base of the pelvic fins, and the side of the abdomen; the middle of the abdomen is immaculate. Heavy dark pigment is on the chin in front of the mental barbels; a small unpigmented area sur- rounds the vent. The vertical fins are similar in color to the upper body surface; the paired fins and barbels are similarly pigmented but lighter. The remaining paratypes are yellowish brown, and presum- ably extensively faded as the upper chromatophores are not prominent and the abdomen and lower surface of the head are devoid of distinct pigment cells. Typr.—The holotype of Noturus lachneri is an immature female, 40.0 mm. in standard length. It has 17 anal rays, 6 soft dorsal rays, 37 vertebrae (of which, judged from a radiograph, 8 are precaudal, 29 are caudal, and 12 are anterior to the anal fin origin), and 254+8+ 13+14=60 caudal rays. On each side there are 8 pelvic rays, 8 soft pectoral rays, 10 preoperculomandibular pores, and a single internasal pore. There also appear to be six hypurals, of which the lower three (1-3) are fused, and one epural. The head length is stepped 3.9 times in the standard length and the distance from the rear end of the adipose fin to the tip of the caudal fin is stepped 1.35 times in the distance from the dorsal origin to the posterior end of the adipose fin. There appear to be eleven branchiostegal rays on the right side, in agreement with counts obtained from a cleared and stained para- type. Measurements are given in table 27. Distripution.—Noturus lachneri is known only from three localities in the Saline River system, Arkansas, a tributary to the Ouachita River. RELATIONSHIP.—The ten preoperculomandibular pores and terminal mouth suggest a relationship closest to Noturus exilis and Noturus gyrinus. In several characters including number of vertebrae, anal rays, and pectoral rays it is intermediate between the two species. The dark, nearly uniform color, lack of pectoral spine serrae, modal number of eight pelvic rays, and large number of caudal rays are characters similar to those of N. gyrinus, but the fusion of the anterior ends of the infraorbital and supraorbital canals, the short flattened CATFISH GENUS NOTURUS RAFINESQUE 57 head, rather elongate body, and relatively short grooves of the pectoral spine show similarities to N. ezilis. Erymo.ocy.—WNoturus lachneri is named after Dr. Ernest A. Lachner in recognition of his outstanding work and interest in North American ichthyology. Ecotoey.—Little information regarding the habitat of this species is available. The holotype was collected from a stream 30 to 50 feet wide with alternating pools and riffles. The specimen was observed, in daylight, swimming near the edge of a pool in about one inch of water. At this point there was no appreciable current and the bottom was covered with small rocks of a few inches in diameter. During the attempt to capture the specimen it swam beneath several rocks, before it could be finally uncovered and scooped into a net. The series, USNM 165901, came from a small stream that was in flood following a heavy rain. The bottom is described as consisting of gravel, rubble, and sand. Noturus exilis Nelson SLENDER MADTOM PuaTeEs 1 (Fia. 1), 2 (FIG. 2), 3 (Fie. 3), 6 (FIG. 1); Map 2 Noturus exilis E. W. Nelson, 1876, pp. 33, 51 (original description; types from McLean Co., Ill. [probably Mackinaw Cr.])—Jordan and Copeland, 1876, p. 160 (Illinois and Wisconsin*).—Jordan, 1877b, pp. 371-372 (comparison; Illinois and Wisconsin*) ; 1877c, p. 50 (S. Illinois;* not noticed by Rafines- que); 1877d, pp. 73-119, pl. 38, figs. 58-59b (comparison; relationship; range; a type from Illinois R. figured [An error in locality is evident for the other two figures, as the captions are given, page 119, as ‘‘Illinois River” and under the figures as ‘‘Root R., Wis.,”’ the latter an improbable locality.]) ; 1877e, p. 611 (relationship).—Jordan and Gilbert, 1877a, p. 2; [Klippart], 1877, p. 153, and Jordan, 1878c, p. 368 (hypothetical for Indiana; range [in error]).—Jordan and Brayton, 1878, pp. 87, 93 (relationship; range [in error]).—Jordan, 1878a, p. 118 (Illinois R. [system] only); 1878b, p. 67 (McLean Co., Ill., and [Marais des Cygnes R.], Kansas* [Root R., Wis., compiled in error?, see above]); 1878d [and 1884], pp. 335-336 (description; range); 1878e, p.414.—Bean, 1880, p.112 (South Grand R.,* Mo.).—Jordan, 1882, pp. 745-800 (synonymy; description; hypothetical for Ohio; range [in error]).—Swain and Kalb, 1883, pp. 640-644 (identical to Noturus insig- nis).—Jordan and Gilbert, 1883, p. 100 (description; range).—Hoy, 1883, p. 434 (Wisconsin; ?Fox R.).—Jordan, 1885, p. 802.—Forbes, 1885a, p. 84 [and 1900, p. 76] (besides the types, found in creeks in ?DeKalb, [Kappa] Woodford,* and Union* Counties, Ill.) —Cragin, 1885, p. 107 (Kansas).— ?Graham, 1885a, no. 30, p. [2] (Neosho R., Kans. [probably a misidenti- fication of Noturus flavus or nocturnus]); 1885b, p. 71 (Osage R., Kans. [Neosho R. a misidentification?])—Call, 1887, p. 79 (Hinkson Cr., near Columbia,* Mo.).—Jordan, 1889, p. 353.—Meek, 1889b, p. 168 (Iowa; comparison; range) ; 1891, pp. 117-141 (description; Missouri records: Little Dry Fork, near Rolla;* Jones Cr., Dixon;* Little Piney R., Newburg* and *Material indicated by an asterisk has been re-examined. 58 U.S. NATIONAL MUSEUM BULLETIN 282 Arlington; Niangua R., Marshfield;* Spring Br. of Hickory Cr., Neosho;* James R., near Springfield;* Bear and Hinkson Creeks, near Columbia*).— Gilbert, 1891, pp. 146, 152 (Richland Cr., Pulaski, Tenn.;* Cypress Cr., Florence, Ala.).—Call, 1892, p. 46 (Iowa record).—Meek, 1892a, p. 12 (Ames, Iowa); 1892b, pp. 223, 225 (Iowa records); 1892c, p. 108 (Iowa distribution) —Hay, 1894, pp. 172, 174 (description and range [not Indi- ana]).—Meek, 1894a, pp. 75-92 (Arkansas records: Middle Fork of White R., Fayetteville* and Illinois R., Ladd’s Mill,* Washington Co.).—??Kirsch, 1895, pp. 327-335 (Tiffin R., Manitou Beach, Mich. [probably a misidenti- fication of Noturus flavus or gyrinus; subsequent efforts to check or duplicate the record have proven fruitless, and exilis is known neither from the drainage basin nor from a region close to this locality])—Evermann and Kendall, 1895, pp. 469-470 (Indian Cr., S. of Neosho and Spring Br.,*at Neosho Hatchery, Mo.).—Hay, 1902, p. 71 (range [not Indiana]).—Jordan, 1904, pp. 42, 351 (description, range, synonymy).—Cleary, 1956, map 65 (Iowa distribution [but probably not Taylor Co. locality]) Harlan and Speaker, 1956, p. 113, pl. 21 (figured; description, in error; Iowa distribution [but probably not Lake of Three Fires, Taylor Co.]).—Bailey, 1956, pp. 335, 364 (lowa; key).—Taylor, 1957, p. 192.—Eddy, 1957, p. 153, fig. 386.—Clarke, Breukelman, and Andrews, 1958, p. 168 (Lyon Co., Kans.).—S.W. Jackson, 1958, p. 236 (Oklahoma records).—Trautman, 1959, p. 30 (not in Ohio).— Metcalf, 1959, p. 393 (Spring R. drainage, Kans.)—Eddy and Underhill, 1959, p. 342 (Otter Cr.,* Minn.).—Stegman and Minckley, 1959, p. 341 (in gravel, Hutchins Cr., Ill.)—Deacon and Metcalf, 1961, pp. 317-321 (Washington* and Mil!* Creeks, Kans.).—Deacon, 1961, pp. 396, 420-423 (Marais des Cygnes R., Kans.).—Larimore and Smith, 1963, pp. 324-333 (compiled).—P. W. Smith, 1965, p. 8 (Illinois distribution).—Bunting and Irwin, 1965, pp. 294-304 (Tyler Cr., Okla.; as test animal).—Metcalf, 1966, pp. 34-35, 80, 150-151, 167, map 40 (distribution, ecology, history, Kansas R. system; Missouri and Kansas records, including Mill Cr.,* Washington Cr.,* and Indian Cr.,* Kans.).—Cross, 1967, pp. 197, 219, fig. 18B (figure, description, Kansas distribution).—Branson, 1967, pp. 137, 151 (Oklahoma records). Schilbeodes exilis (Nelson).—Jordan and Evermann, 1896a, pp. 145, 147 (descrip- tion; synonymy; range) ; 1896b, p. 234.—Evermann and Cox, 1896, pp. 365- 426 (Missouri and Kansas records compiled [page 388, a Meek record erroneously listed as Little Piney R., Cabool, Mo.]).—?Cox, 1896, p. 608 (Blue Earth R., Mankato, Minn.).—?Cox, 1897, pp. 19-20, 78 (description; Blue Earth R., Mankato, Minn.).—Jordan and Evermann, 1900, pl. 28, fig. 65 (USNM 36261, Osage Fork, Gasconade R., Marshfield,* Mo.).— ??Evermann, 1902, p. 95 (Great Lakes basin [compiled and probably in error]).—Large, 1903, pp. 9-10 [and 1905, pp. 56-57] (comparison; Illinois records: Cane Cr., Freeport;* South Henderson? and Honey* Creeks, Henderson Co.; Iroquois R., near Watseka?; creeks in Union Co.*).— ??Michael, 1906, p. 9 (spelled exiles; range, Michigan record, and inclusion in “Great Lakes’’ compiled [all probably in error]).—Goldsborough and Clark, 1908, p. 33 (one specimen, Guyandotte R.,* W. Va. [USNM 56615, August 15, 1900, W. P. Hay; locality doubtful]).—Forbes, 1909, pp. 387-404 (range and Illinois distribution [in error]).—Forbes and Richardson, 1909 [and 1920, pp. Ixxviii—xciv], pp. lxxiii-Ixxxix, 196-200 (range [in error]; description; figured; Illinois records: Illinois R.* [system]; the Pecatonica [Cane Cr.*], Freeport, *Material designated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 59 Stephenson Co.; Du Page R., Will Co.;* Honey Cr. [Stronghurst*], Henderson Co.; two? creeks in Union Co. [Cave Spring Br., Jonesboro;* Dutch Cr., SW. of Jonesboro;* Clear* Cr.; Big Cr., 6 mi. S. of Anna*]).—Meek and Hilde- brand, 1910, pp. 245-246, fig. 21 (description; range [in error]; Illinois rec- ord).—Shelford, 1913, p. 95 (ecology, positively rheotactic)—Hankinson, 1913, p. 109 (Kaskaskia R., Ill.).—Fowler, 1915a, p. 209 (Brook R., Iowa).— ?T. Surber, 1920, p. 21 (compiled, Minnesota).—Conger, 1920, p. 11.—H. S. Pratt, 1923, p. 96.—C. L. Hubbs, 1926, pp. 51-52 (remarks concerning record from Tiffin R., Manitou Beach, Mich.).—Cahn, 1927, p. 42 (Oconomowoc R.., Stonebank and ?Mukwonago R., Wis.).—Greene, 1927, p. 306.—??Hubbs and Greene, 1928, p. 390 (Great Lakes basin [compiled and in error?]).— Potter and Jones, 1928, p. 355 (compiled).—Hubbs and Brown, 1929, p. 3 (unknown in lower Great Lakes).—Thompson and Hunt, 1930, pp. 27, 44 (habitat; ?Champaign Co., Ill.).—Greene, 1935, pp. 145-146 (L. Erie trib. [in error?]; Cahn’s records of Schilbeodes miurus possibly S. ezxilis; Illinois and Wisconsin records [?Mukwonago R.]; recent origin in Wisconsin).— Aitken, 1936, p. 33 (Iowa).—Schrenkeisen, 1938, p. 167.—??Hubbs and Lagler, 1939, p. 26 (Great Lakes basin [compiled and in error?]).—Kuhne, 1939, p. 68 (Tennessee).—Breukelman, 1940b, p. 381 (Osage R.* system, Kans.).—Hubbs and Lagler, 1941, pp. 63, 65 (range [but, not in Kanawha R. system; recorded from Great Lakes, in error?]; comparison; [not] ‘“‘a complex of subspecies referable to S. insignis’’)—Aitken, 1941, p. 389 (Iowa).—~ Jennings, 1942, p. 365 (Kansas record).—Eddy and Surber, 1943, pp. 151-163 (comparison; Minnesota record doubted).—C. L. Hubbs, 1946, p. 38 (Okla- homa).—G. A. Moore, 1952, p. [6] (Oklahoma).—Hall, 1954, p. 57 (Oklahoma records); 1955, pp. 37, 38 (population reduced, impoundment, Illinois R., Okla.) —G. A. Moore, 1957, pp. 148, 144, figs. 2-76, 2-79D (description; range, in error).—Hubbs and Lagler, 1957, p. 5 (questionable, Great. Lakes) ; 1958, pp. 89, 91, fig. 178 (key; range, in error).—Curd, 1960, pp. 26—29 (food, Tyner Cr. and Illinois R., Okla.).—Birkhead, 1967, pp. 101-110 (comparative toxicity of venoms). Rabida exilis (Nelson).—Jordan, 1929, p. 93.—Jordan, Evermann, and Clark, 1930, p. 156.—Luce, 1933, p. 119 (Illinois record).—O’Donnell, 1935, p. 484 (Illinois records).—H. S. Pratt, 1935, p. 90.—Blatchley, 1938, p. 67 (range and description [in error; not Indiana]).—Driver, 1942, p. 254 [and 1950, p. 262]. Noturus elassochir Swain and Kalb, 1883, pp. 638-644 (original description; type,* USNM 29677, collected by Dr. E. R. Copeland from Illinois R., Napierville, Ill.) —Jordan, 1885, p. 802 (‘identical with Noturus ezxilis, Nelson’’).— Hubbs and Raney, 1944, p. 20 (type-locality: ‘Spresumably Napierville on West Branch of Du Page R., tributary to the Illinois River,”’ IIl.). Noturus insignis (Richardson) [misidentification]——Swain and Kalb, 1883, pp. 640-641 (synonymy and range, in part).—Gerking, 1955, p. 51 (not in Indiana). Schilbeodes insignis (Richardson) ([misidentifications]—Fowler, 1915a, pp. 208-209 (Carthage, Mo.* only).—Hubbs and Raney, 1944, pp. 1-23, map 1 (synonymy, description, distribution, .all in part).—Fowler, 1945, p. 32 (Tennessee R. basin).—?Eddy and Surber, 1947, pp. 171, 182 (comparison; description; Minnesota record compiled).—?Hubbs and Lagler, 1947 [and 1949], pp. 71, 73, fig. 178 (comparison; specimen, Washington Co., Ark. figured; range, in part; Wisconsin and Michigan records doubted).—Har- *Material indicated by an asterisk has been re-examined. 298-943 O—69 5 60 U.S. NATIONAL MUSEUM BULLETIN 282 rison, 1949, p. 338 (Iowa record).—Moore and Paden, 1950, pp. 87, 89 (asso- ciations; natural history; Oklahoma records).—Harlan and Speaker, 1951, p. 97, pl. 19 (description).—Bailey, 1951, pp. 194, 224 and in reprint, 1951, pp. 194, 226 (comparison; Iowa).—Hall, 1951a, pp. 38-39 (Grand L., Okla.) ; 1951b, p. 17 (L. Carleton, Okla.).—Cross and Moore, 1952, p. 407 (records, Poteau R., Okla.).—Lewis and Elder, 1953, pp. 193, 202 (Illinois record).— Martin and Campbell, 1954, pp. 47-53 (riffles of Black R., Mo.).—Cleary, 1954, p. 633 (distribution, Cedar R., lowa).—?Schelske, 1957, p. 38 (Verdigris R., Kans. record [doubtful)). Noturus flavus Rafinesque [misidentifications].—?Jordan and Gilbert, 1886, p. 7 (tribs. to Poteau R. [Poteau R., W. of Hackett,* Indian Territory], near Fort Smith and Lee Cr., above Van Buren, Ark.).—? Meek, 1893, p. 229 (Arkansas records: Fort Smith; Illinois R., Prairie Grove* and Ladd’s Mill;* Jordan’s Cr. and Barren Fork, Dutch Mills).—?Meek, 1894a, pp. 90, 92 (Fort Smith, Ark.).—Large, 1903, p. 10 [and 1905, p. 57] (Kaskaskia R., Douglas Co.,* Ill.).—Forbes, 1909, pp. 397-417, map 57, and 1914, p. 18, map 57; Forbes and Richardson, 1909 [and 1920, pp. lxxxviii—cv], pp. Ixxxiii-c, 195, map 57 (Illinois records [Lake Fork, Douglas Co.;* Buck Cr., McClean Co.*]).— ?Hubbs and Ortenburger, 1929, p. 96 (compiled).—Luce, 1933, p. 118 (Kas- kaskia R. system,* Ill.).—O’Donnell, 1935, p. 484 (Douglas Co.,* IIl.).— ?Moore and Paden, 1950, p. 87 (Meek’s records for Illinois R. and Barren Fork).—?Rostlund, 1952, p. 274 (Arkansas records of Meek).—?Cross and Moore, 1952, p. 406 (Poteau R., Okla.). Noturus gyrinus (Mitchill) [misidentifications].—? Meek, 1894a, pp. 75, 92 ([con- fused] ?Flat and Machine Creeks, at Smithville, Ark., and [by drainage as] Illinois R. system). Schilbeodes mollis (Hermann) [misidentification——? Moore and Cross, 1950, p. 141 and ?Moore and Paden, 1950, pp. 87, 93 (reference to Meek’s record: Illinois Ee oArk:): Schilbeodes nocturnus (Jordan and Gilbert) [misidentification].—Cross and Moore, 1952, p. 406 (Poteau R., Okla. records: Fourche Maline,* sec. 11-12, T. 5 N., R. 19 E., Latimer Co., in part; Poteau R.,* sec. 29, T. 5 N., R. 27 E., Le Flore Co., in part). OTHER MATERIAL STUDIED UNITED STATES: Atasama: UMMZ 96452, 113927 (Cedar Cr., 3 mi. S. Russellville, Franklin Co.), 165876; UMMZ (TVA nos. 2, 4B, 10, 14, 16-Clark Spring Br., trib. Flint Cr., 0.5 mi. from Austinville, Morgan Co.; Delavan and Creaser nos. 31-75, 31-77-stream, 6 mi. SW. Huntsville, Madison Co.); USNM 190722. Arkansas: UMMZ 97214 (trib. to Little Buffalo R., near Diamond Cave, 3.5 mi. W. Jasper, Newton Co.), 102823, 116376 (trib. to South Spring R.., about 5 mi. NW. Ravenden, Sharp Co.), 123437, 128295, 128312, 128377, 128697, 169894, 170899, 170921; CU 9836 (stream, 10 mi. W. Little Rock, Pulaski Co.); CNHM 1624; SU 32234; Tulsa U (Lee Cr. at Oklahoma line); USNM 165883 (McHenry Cr., sec. 28 and 33, T. 1 N., R. 13 W., Pulaski Co.), 201395, 202485 (Wagon Wheel Cr., N. Alonzo, Independence Co.), 202486 (Choctaw Cr., Hwy. 65, S. Choctaw, Van Buren Co.). In.tinois: UMMZ 114911, 135611 (Hutchins Cr., E. of Wolf Lake, Union Co.); CU 3474 (stream, E. of Anna, Union Co.); USNM 1498, 24841 (Rocky Brook, Union Co.), 29677 (holotype, Noturus elasso- chir, Illinois R., Napierville); CNHM 42141, 42177, 42207, 42239; CNHM (Hickory Cr., Sta. 5 and Morley Cr., Sta. 7, Will Co.). Iowa: UMMZ 146788 *Material designated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 61 (Cedar Cr., about 4.5 mi. W. Churden, Greene Co.), 146862 (East Buttrick Cr., Greene Co.), 159766 (East Br. Otter Cr., Linn Co.); USNM 174929 (Iowa City). Kansas: UMMZ 120598 (Marais des Cygnes R., 5 mi. N. Lebo, Coffey Co.), 122201, 126827, 126851 (Salt Cr., 1 mi. S. Osage City, Osage Co.), 155189, 160355, 160373, 160394 (Spring Cr., 5.5 mi. E. Baxter Springs, Cherokee Co.), 160531 (South Fork of Pottawatomie Cr., 1.5 mi. S. Garnett, Anderson Co.), 160546, 160877; KU 655 (Pottawatomie Cr., near Glenlock, Anderson Co.), 1516, 4335 (Indian Cr., sec. 10, T. 13 S., R. 25 E., Johnson Co.), 4672, 4675 (Washington Cr., 0.5 mi. below Lone Star Lake, Douglas Co.), 8117, 8263 (Mill Cr., sec. 11, T. 125S., R. 10 E., Wabaunsee Co.); KU (Big Bull Cr., 1 mi. NW. Paola, Miami Co.; Long Cr., Osage Co.; stream 3 mi. S. Louisburg, Miami Co.); USNM 172051 (Spring R., Hwy. 96, Cherokee Co.), 174904, 174907. Kentucky: UL 6038, 6048 (Whipporwill Cr., U.S. Hwy. 79, Logan Co.), 6826. Minnesota: UMinn 18150 (Otter Cr., trib. Red Cedar R., Mower Co.). Missourr: UMMZ 102517, 102535, 102580 (Barren Fork, trib. to Eleven Point R., 2 mi. SE. Thomasville, Oregon Co.), 102730, 103054, 103209, 111376, 111413, 111419, 111441, 113626, 113642, 113643, 113663, 120045, 142116, 142220, 142249, 148002, 148304 (Elk Fork of Salt R., 2 mi. S. Paris, Monroe Co.), 148425, 148460, 149134, 149342, 149529, 150050, 150090, 150100, 150201, 150275, 150301, 150334, 150350, 150391, 150468, 150519, 150771, 150802, 150866, 150882, 150921, 1509438, 150964, 150995, 151056, 151077, 151093, 151127, 151206, 151265, 151321, 151428, 151452, 151479, 151502, 151571, 151604, 151632, 151651, 151670, 151691, 151713, 151735, 151755, 151810, 151890, 151912, 151933, 151949, 151971, 152010, 152062, 152090, 152115, 152237, 152338, 152360, 152443, 152473, 152673 (North Moreau Cr., 8 mi. SW. Jefferson City, Cole Co.), 152698, 152841, 152868, 152926 (Bois Brule Cr., 3 mi. S. Carnegie, Cole Co.), 153298 (Driskin Cr., trib. of Indian Cr., Oriole, Cape Girardeau Co.), 162602 (Little Bonne Femme Cr., 6 mi. S. Columbia, Boone Co.); UMMZ (Buffalo Cr., Tiff City, McDonald Co.; Creaser and Clanton nos. 24, 25, 28, 32, 33, 35, 45-Caster R., 8 mi. E. Fredericktown, Hwy. 61, Madison Co., 47, 48, 49, 53- Charles Cr., 2 mi. SW. Farmington, St. Francois Co.); CU 10787; USNM 36246, 36292, 171994, 201394; SU 2566; INHS (Niangua R., Bennett Spring Park, Dallas Co.; Salt R., NE. of Shelbina, Shelby Co.; Gasconade R., 4 mi. S. Jerome, Phelps Co.; Sac R., N. of Stockton, Cedar Co.; Meramec R., 10 mi. NE. Salem, Dent Co.; Pomme de Terre R., 2 mi. SW. Hermitage, Hickory Co.). OKLAHOMA: UMMZ 103134, 103153, 103178, 108408, 109472 (Bandy Cr., 1 mi. S. Wilburton, Latimer Co.), 110093 (Cunnee Tubby Cr., 2.25 mi. N. Wilburton, Latimer Co.), 110872 (Brushy Cr., near NW. end Rich Mountain, Le Flore Co.), 110873 (Shaunty Cr., 2 or 3 mi. SW. Stapp, Le Flore Co.), 110953 (Spring Cr., Camp Garland, Mayes Co.), 116439, 116654 (Spavinaw Cr., below dam at Spavinaw, Mayes Co.), 116777, 127173, 127308, 137863, 137941 (Illinois R., near Gore, Sequoyah Co.), 167213; UMMZ (field nos. C-—4-39, C—-5-39, MM25, Se 4-40; Delavan and Creaser no. 31-34); USNM 165820; CU 17885; Tulsa U (Illinois R., Hwy. 59, Adair Co.); TU 4137. TENNEssEE: TU 19471; UMMZ 96392, 105051, 105296, 120169, 120938 (McBride Br., SW. of Beech Grove, Coffee Co.), 168491, 174459 (trib. to Turnbull Cr., 2 mi. W. Kingston Springs, Cheatham Co.); USN M 190822, 193478, 196800, 197397; UMML 10672; CU 22164 (Sulphur Cr., Hwy. 76, 5 mi. S. Adams, Robertson Co.), 37252, 42033 (Duck R., U.S. Hwy. 41, 1.4 mi. NW. Manchester, Coffee Co.), 47850 (Elk Cr., Stewart Co.), 48238 (trib., U.S. Hwy. 64, 7 mi. E. Pulaski, Giles Co.) ; Vanderbilt U (Turnbull Cr., Dickson Co.; Hurri- cane Cr., McEwen-Bold Springs road, Humphreys Co.) ; ANSP 83004 (Brimstone Cr., 2.75 mi. S. New River, Scott Co.). Wisconsin: USNM 1412 (Oconomowoc R., Lac la Belle); INHS (Cahn coll.—-Honey Cr., trib. Rock R., Watertown). 62 U.S. NATIONAL MUSEUM BULLETIN 282 Typre.—The types of Noturus exilis Nelson have not been located. They may have been destroyed in the Indiana University fire of 1883. The type-locality is in McLean County, Illinois. Diacnosis.—Noturus exilis is the only member of the subgenus Schilboedes typically having the combination: mouth terminal or subterminal, jaws about equal, a single internasal pore, nine pelvic rays, and ten preoperculomandibular pores. It has 17 to 22 anal rays; 8 to 10, usually 9 pelvic rays; 8 to 10, frequently 8, but mostly 9 soft pectoral rays; and 44 to 57 caudal rays. The vertical fin borders are often black or darkened, and the posterior edge of the pectoral spine always has distinct serrae. The shortened caudal fin distinguishes Noturus ezilis from N. insignis and N. nocturnus and the ten preoperculomandibular pores segregate it from other species of the subgenus except gyrinus, lach- neri, nocturnus, and gilberti. In contrast to gilberti, the vertical fins have dark borders, some of the abdomen and lower head has pig- ment, and the spines are longer. In contrast to gyrinus and lachneri, there are more pelvic rays. Derscription.—Other counts and measurements are given in tables 17 to 26. Head decidedly depressed, rather elongate, especially in front of eyes, by elongation of the mesethmoid; lower jaw terminal or subterminal; eye small, 2.5 to 3.5 times in snout; humeral process obscure in some populations, usually shorter than the width of the pectoral spine exclusive of serrae; dorsal spine stout; pectoral spine short, straight, with five to eight, occasionally as many as ten serrae which are long, prominent, and usually straight, but sometimes bent outward or inward or fused at the base; anterior serrae never present on spine, instead there may be retrorse hooks or progressive offsetting of the edge (pl. 3, fig. 3); adipose fin long and low, closely united to the anterior caudal rays, without a notch; caudal fin slightly rounded behind; posterior corners of the premaxillary tooth patch rounded, truncate, or projecting slightly backward. The soft dorsal fin has five (in 1), six (189), or seven (4) rays. Of 187 individuals from over the range the caudal fin usually has 16 to 20 (mean 18.39) upper simple rays, 17 to 20 (mean 19.11) branched rays of which 8 are in the upper lobe (mean 8.02) and 10 to 12 (mean 11.09) are in the lower lobe, and 11 to 15 (mean 13.16) lower simple rays. In the skeletons and stained specimens examined: vertebrae an- terior to the anal fin origin 12 (in 1), 13 (12), 14 (1), or 15 (1); ossified pectoral radials fused (in 49 sides examined). The five or six hypurals, at least in the specimens examined, are seldom fused; hypurals 1-3 were fused (in 1 specimen), 2-3 (in 3), and no fusion observed (in 13). There are five to eight gill rakers on the first arch. CATFISH GENUS NOTURUS RAFINESQUE 63 The largest specimen known, an unusually large individual, was retained in an aquarium at the Museum of Zoology, University of Michigan, for one and one-half years, at which time it had attained 113 mm. in standard length. The color of its sides, dorsal surface, and pectoral fins is a slate-gray; the pelvic fins and lower surface are darkly pigmented but somewhat lighter than the sides. Other specimens are less than 100 mm., rarely more than 90 mm., in stand- ard length and lack the darkened undersurface. General color in life yellowish brown to gray-black. In preserved material: ventral surface much lighter than side or upper surface; top of head, upper barbels, back, and most of side nearly uniformly grayish; an obovate, light yellow spot extends backward from the base of the last two or three dorsal rays; side lightly but uniformly pigmented; mental barbels and lower surface of head and belly light, but all under surfaces becoming darker with age; ventral surface immaculate (individuals to about 90 mm. in standard length); old specimens nearly unicolored and only slightly lighter below; a band of pigment crosses the chin in front of barbels and a faint one lies just in front of the pelvic fins; all median fins lightly pigmented with gray; dorsal, caudal, and anal fins often with black borders; the fins unicolor in some populations; pelvic and pectoral fins of small indi- viduals unpigmented, but becoming darker with age; tips of pectoral rays unpigmented. VARIATION.—There is considerable variation in Noturus exilis, as noted by Hubbs and Raney (1944, p. 21). The degree of pigmentation _ of the fins varies from place to place. They have very dark margins in the White River system and are often unicolor in many other areas. Also in the White River, ezilis is more elongate. Average meristic differences among samples from the Arkansas River system, White River system, and Missouri River system can be demonstrated. Subspecific separation, however, probably is not warranted. In fin ray counts, specimens from Tennessee, Illinois, and Towa appear to be much like those of the Missouri basin. The data are listed as follows: number of tabulations, range (in parentheses), and mean in samples from (a) Arkansas River system, Arkansas, Missouri, and Oklahoma, (b) White River system, Missouri and Arkansas, (c) Missouri River basin, (d) Tennessee, Minnesota, Iowa, Illinois, and the Meramec River, Missouri, and (e) total. Data for the speci- men, USNM 56615, reputedly from Guyandotte River, West Virginia, are omitted here, but included in tables 17 and 19 comparing the species of Noturus. Pelvic rays: (a) 126 (8-10) 9.02; (b) 68 (9-10) 9.06; (c) 100 (8-10) 9.03; (d) 100 (8-10) 9.15; (e) 394 (8-10) 9.06. 64 U.S. NATIONAL MUSEUM BULLETIN 282, Soft pectoral rays: (a) 126 (8-9) 8.62; (b) 68 (8-10) 8.97; (c) 102 (8-10) 8.74; (d) 94 (8-9) 8.89; (e) 390 (8-10) 8.78. Lower-half caudal rays: (a) 63 (21-26) 23.48; (b) 34 (22-26) 24.74; (c) 52 (22-27) 24.48; (d) 47 (21-26) 24.23; (e) 196 (21-27) 24.14. Upper-half caudal rays: (a) 63 (22-28) 24.81; (b) 34 (25-29) 26.50; (c) 52 (24-31) 27.52; (d) 47 (25-28) 26.91; (e) 196 (22-31) 26.33. Vertebrae seem to decrease in number northward. The upper Mississippi and Missouri basin samples are low in number. The counts from the White River system are highest: Iowa, Wisconsin, and Illinois River, Illinois 8 (36-38) 37.13; Missouri River drainage, Missouri and Kansas 36 (36-39) 38.11; Arkansas River drainage, Oklahoma and Arkansas 32 (38-40) 38.88; White River system 24 (39-41) 40.13; Tennessee River system 13 (39-40) 39.38; total 113 (36-41) 38.83. Internasal pores: A sample from Hickory Creek, Illinois has 9 of 70 circumorbital canals unconnected anteriorly (two internasal pores) ; the other 61 are connected. Otherwise the species throughout its range has a single internasal pore. This characteristic was utilized as one of the identification checks for nearly all the lots examined. Of the samples tabulated at random from areas other than Hickory Creek, 466 of 471 canals have a single internasal pore. This variation is not reflected in the number of preoperculomandibular pores (table 22): Variations in caudal ray and anal ray counts are given in tables 4 and 5. Note especially the greater number of anal rays in the White River population, and the low number of caudal rays in the Arkansas River system. DistripuTION.—The range (map 2) of Noturus exilis, a creek species, may have been broken very recently into two or more populations, one in the Ozarks and northward and one in the Tennessee and Cumberland River basins. It was recorded by Goldsborough and Clark (1908, p. 33) from the Guyandotte River, West Virginia [USNM 56615], on the basis of a single specimen, likely the result of a trans- position of locality data. Dr. Frank J. Schwartz has failed to find additional specimens in his survey of West Virginia fishes. Thus the locality is now doubted, and not shown on the distribution map, since other records (two from Indiana) from the Ohio drainage have been re-identified as Noturus flavus, and presence in the Green River, Kentucky (Charles, 1967, pp. 386-389) is doubtful and probably also based on misidentifications of juvenile N. flavus. It does not seem likely that N. exilis has occurred in the Ohio basin within historical times. CATFISH GENUS NOTURUS RAFINESQUE 65 Seog ae Ee FEL aoe Be 50 Bae > #5 eR cs) fis Map 2.—Distribution of Noturus exilis Nelson. Circles represent localities whence I have examined specimens. The dashed line indicates the known range. This line is not con- nected from the region of the lower Tennessee basin to southeastern Missouri because the two populations may be isolated by lowlands. The locality of a single specimen from West Virginia is not shown; it is thought to be the result of a transposition of locality information, and thus in error. Other literature records from outside the indicated range are either known misidentifications or of dubious reliability. The report of ezilis from Root River, Wisconsin, in the Great Lakes basin, by Jordan (1877d), is also doubted. As noted in the synonymy, Illinois River, Illinois, and Root River, Wisconsin, are both given as the locality for the specimen Jordan illustrated. The Illinois River locality coincides with the range as presently known. However, two specimens of a Noturus are listed as Noturus exilis in the United States National Museum catalog books. These specimens (USNM 1420) were collected by S. F. Baird from Root River, Racine, Wisconsin, in July 1853 and entered in the catalog book in February 1859. They may have been sent to Agassiz in March 1859 and not returned, for the specimens have not recently been found. Thus it is 66 U.S. NATIONAL MUSEUM BULLETIN 282 possible that an original misidentification was made and that Jordan never actually saw the specimens. Other Wisconsin material was available to Jordan, as two or three specimens (USNM 1412) were collected by Baird in July 1853 from Lac la Belle. The remarks column of the catalog book bears the notation ‘Jordan ident’? for USNM 1412, meaning that Jordan examined the specimens and [correctly] identified them. No such notation is entered for USNM 1420 which appears on the same page. It now seems likely that Jordan copied or was informed of the Root River locality without examining the speci- mens, and that he or someone else added that locality to the figure captions without closely checking the specimen data or ‘“‘List of Illustrations.’’ Subsequently the range of N. ezilis has been stated to include “Lake Michigan” and “‘the lake,’ both based on the Root River material. The record from the Tiffin River, Manitou Beach, Michigan, has not been re-examined or duplicated. It seems likely that Kirsch misidentified a specimen of Noturus gyrinus or Noturus flavus. Since this specimen does not appear to exist, since extensive collecting at the locality has failed to yield further specimens, and since the nearest positive records come from drainages many miles away, it is thought best to regard it as a probable error in identification. Thus, there is no present admissible evidence that exilis occurs in the Great Lakes basin. In Tennessee, Kentucky, and Alabama, the records of ezilis are based on many specimens from the Duck River, the Cumberland River, and the lower Tennessee River basins upstream to the bend in Alabama. N. ezilis is not present in the upper Tennessee basin. Noturus exilis is abundant in many of the Ozark and sub-Ozarkian streams of Arkansas, Kansas, Missouri, and Oklahoma. In the Arkan- sas drainage, it occurs in tributaries to the Neosho River arising in the Ozarks, in most streams east of the lower Arkansas River in Oklahoma, in the Poteau River, and in many tributaries of the Arkan- sas River in Arkansas. In Arkansas and Missouri, it is found in much of the upper White River system. In Kansas and Missouri, the species is found throughout the Osage River system. It is confined in the Missouri River drainage of Missouri to the Osage and the lower tributaries of the Missouri, but in Kansas at least three popu- lations exist in southern tributaries to the Kansas and Missouri Rivers, either as remnants of a wider distribution or as crossovers from the adjacent Osage system. In the upper Mississippi drainage, it is found in several Mississippi tributaries in Illinois, Wisconsin, Missouri, Iowa, and Minnesota. Although recently collected in southern Minnesota (Eddy and Underhill, 1959) the record of Cox (1896) from Blue Earth River, CATFISH GENUS NOTURUS RAFINESQUE 67 Minnesota, is still doubted. Another dubious record, the sole one in the Missouri River drainage, Iowa, was recorded by Cleary (1956, map 65) as Lake of Three Fires, Taylor County. Subsequent efforts to re-examine the specimen have been fruitless, as the specimen was presumably discarded in the field (personal communication from Robert E. Cleary). Specimens of Noturus flavus and of Noturus nocturnus have been identified as this species; Noturus insignis has been confused with it. Table 9 is intended to facilitate the identification of young individuals of exilis and similar species. NoMENCLATURE.—As indicated under Noturus insignis, the action of Hubbs and Raney (1944) in transferring the name Pimelodus insigne Richardson to the identity of Noturus exilis is invalid. That name belongs to the eastern species, here called Noturus insignis. As shown in the above synonymy, the first available name for the present species is Noturus exilis Nelson. The original description of Noturus exilis was supplemented by Jordan (1877b, p. 372) who gave diagnostic characters from the types. Some of these are: “‘Pectoral spines very strongly serrated behind . . . with five or more prominent recurved hooks ... , jaws nearly equal . . . color nearly uniform; tip of dorsal blackish.” In addi- tion, his figure (1877d, pl. 38, fig. 59b) of one of the types confirms the presence of a subterminal lower jaw and the serrated pectoral spine. There can remain little doubt about the application of the name Noturus exilis. The holotype (USNM 29677) of Noturus elassochir from Napier- ville, Illinois, which was described later by Swain and Kalb (1883), also belongs to this species. It is a large individual, 117 mm. in total length or 99.8 mm. in standard length, and the body form, sub- terminal jaws, spines, and premaxillary tooth band are typical. On each side there are: one internasal pore, ten preoperculomandibular pores, nine soft pectoral rays, and ten pelvic rays. The anal fin has 19 rays, the caudal fin has 19+9+12+11=51 rays, and there are 38 vertebrae. ErymMotocy.—The name ezilis (Latin) means slender, referring to the slim body. RELATIONSHIP.—Noturus ezilis has previously been confused with and considered an intimate relative of Noturus insignis, chiefly on the basis of a relatively long body and anal fin and a similarity in color pattern. These characters are misleading, however, as important differences between the two are to be found in the structure of the pectoral spine, the form of the caudal fin and number of hypurals, the shape of the skull, the difference in position of the mouth, and in the structure of the sensory canal system of the head. Instead, it 68 U.S. NATIONAL MUSEUM BULLETIN 282 seems that N. insignis and N. nocturnus are more closely related; that N. ezilis has an affinity with Noturus lachneri and through that species with Noturus gyrinus. This cluster of species has in common a terminal mouth and similar pore counts. Divergence in the group was probably early as judged by the increase in anal and paired fin rays, reduction in number of caudal rays, and well-developed pectoral spine serrae in WN. ezilis. Ecotocy.—N. ezilis lives entirely in streams of small or medium size in moderate or fast current. It occurs on riffles composed of coarse gravel, rubble, or loose slab rock. Noturus exilis has been taken in collections with gyrinus, nocturnus, flavus, albater, placidus, flavater, miurus, elegans, and the specimen referred to elegans from Piney Creek, Alabama. Noturus leptacanthus Jordan SPECKLED MADTOM PuaTEs 3 (FIG. 4), 8 (FIG. 2); Map 3 Noturus leptacanthus Jordan.— Jordan and Copeland, 1876, p. 160 (nomen nudum; Alabama R.).—Jordan, 1877b, pp. 352-372 (original description ; the single specimen from Silver Cr., 1 mi. above its junction with the Etowah R.); 1877d, pp. 73-102, pl. 41, figs. 64-65 (description; synonymy; type figured).—Jordan and Brayton, 1878, pp. 44-87 (description; Chattahoochee R., near Gainesville, Georgia; record relisted).—Jordan, 1878a, p. 119; 1878d [and 1884], p. 336; 1878e, p. 414.— Swain and Kalb, 1883, pp. 639-644 (description; range [except Enterprise, Miss.]).—Jordan and Gilbert, 1883, pp. 98-99 (description; range [in error]).—Jordan, 1885, p. 802.—Jordan and Gilbert, 1886, p. 7 (comparison).—Jordan, 1889, p. 353 (relationship).— Gilbert, 1890, p. 226 (Georgia record); 1891, pp. 153-157 (Alabama records: Choccolo Cr.,* Oxford; Cahawba R., Helena;* Little Escambia R.,* Pollard; Hawkins Cr., Greenville;* Sand(y) Cr.,* Evergreen)—Woolman, 1892b, p. 301 (description; Florida record).—Palmer and Wright, 1922, p. 356 (compiled).—Bailey, Winn, and Smith, 1954, pp. 114-160 (fresh water; range; synonymy; records, Escambia R.,* Alabama and Florida [all were examined]).—Eddy, 1957, p. 152, fig. 383.—Taylor, 1957, p. 192.—Briggs, 1958, p. 260 (SE. United States to Lake Co., Fla.).—Cook, 1959, pp. 34, 135, 141 (description; Mississippi records).—Parsons and Crittenden, 1959, p. 191 (Chipola R., Florida).—Suttkus, 1961, p. 63, fig. 5 (comparison; skull illus- trated).—Patrick, 1961, p. 256 (Escambia and Flint Rivers).—Boschung, 1961, pp. 275, 282 (compiled).—Collette and Yerger, 1962, p. 225 (ecology; Okaloosa and Walton Counties, Fla.).—Anderson, 1964, pp. 45, 51 (South Carolina records).—Suttkus and Taylor, 1965, p. 177 (associations).— Richards, 1966, p. 829 (associations with Etheostoma inscriptum).—Hellier, 1967, pp. 18-46 (parasites, ecology, distribution, Santa Fe R., Fla.).— Suttkus and Ramsey, 1967, p. 140 (associations with Percina aurolineata, Cahaba R., Ala.). Schilbeodes leptacanthus (Jordan).—Jordan and Evermann, 1896a, pp. 145-146 (description; range); 1896b, p. 234.—Evermann and Kendall, 1900, p. 51 *Material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 69 (compiled).—Reed, 1907, pp. 562-564 (axillary glands and probably spine glands present).—Palmer and Wright, 1922, pp. 356-361 (New R., Fla.; hypothetical, Okefinokee Swamp).—H. S. Pratt, 1923, pp. 95-97.—A. H. Wright, 1926, pp. 80-81.—Viosca, 1936, pp. 42-43 (Little Bogue Falia Cr., La., in association with Ambloplites ariommus); 1937, p. 136 (in association with Necturus beyert).—A. Carr, 1937, p. 81 (comparison; Florida).—Schren- keisen, 1938, p. 167.—Hubbs and Raney, 1944, pp. 3, 9, 26 (comparison; anal fin length).—?Fowler, 1945, pp. 32, 271, 351(synonymy; description; range [except Mississippi R. system]; ?Florida records, in part only: Wekiwa R.; L. Fairview; Palm Springs. Alabama records).—D. C. Scott, 1951, p. 37 (Coosa R.,* near Childersburg, Ala.).—Freeman, 1953, p. 269 (Savannah R. basin, in Aiken* and Barnwell* Counties, 8.C.); 1954, pp. 138, 145-146, 151 (Savannah drainage, South Carolina records, in part: stations 3,* 13, 18,* 24,* 33,* 49, 60,* 90*).—Carr and Goin, 1955, pp. 21, 65 (description; Florida distribution)—G. A. Moore, 1957, pp. 143-144.—-Crittenden, 1958, p. 218 (Bay Co., Florida record).—Birkhead, 1967, pp. 101-110 (comparative toxicity of venoms). Rabida leptacantha (Jordan).—Jordan, Evermann, and Clark, 1930, p. 155.— Harper, 1930, p. 152 (Collins Mill Cr., about 7 mi. NE. of Cuthbert,* Ga.).— Fowler, 1935b, p. 73 (Georgia record).—H. 8S. Pratt, 1935, p. 90 (description? ; range).—??Fowler, 1936b, p. 150 (Crystal R., Citrus Co., Fla.).—Driver, 1942, p. 254 [and 1950, p. 262]. Rabida gilberti (Jordan and Evermann) [misidentifications].—Fowler, 1935a, p. 19, fig. 37 (Great Cypress Swamp,* S.C., in part; “South Carolina’’*). Schilbeodes marginatus marginatus (Baird) [misidentifications]—Hubbs and Raney, 1944, pp. 8, 24 (synonymy, in part; records: ‘South Carolina’ ;* Great Cypress Swamp,* in part).—Fowler, 1945, p. 81 (synonymy, in part). Schilbeodes mollis (Hermann) [misidentification].—Freeman, 1954, pp. 138, 146 (Savannah R. drainage, S.C., station 21* only). OTHER MATERIAL STUDIED UNITED STATES: Atasama: Tulsa U 8; CU 11827, 13786, 14019, 15994, 16105, 16213, 17167, 17488, 17671, 19272; USNM 36827, 162336, 162345, 166024, 166035, 199552, 201248; UGa 139; TU 1130, 1644, 2616, 7476, 14144, 15267, 19072; UMMZ 113911, 123970, 123974, 124133, 128749, 155516, 163545, 166385, 168745. FLoripa: UMMZ 87895 (Santa Fe R., Poe Springs, Alachua Co.), 110481, 110487, 110975, 134570, 155496, 161759, 163320 (trib. of Suwannee R., SE. Genoa, Hamilton Co.), 163326 (Juniper Springs, head of Juniper Cr., Ocala National Forest, Marion Co.), 163495, 163522, 166167, 166186, 166193, 166216, 166295; CU 3126 (Black Cr., near Green Cove Springs, Clay Co.), 12024 (stream, 1.5 mi. W. Cassia, Lake Co.), 12128, 12140, 12499, 12604, 16678, 16691, 16700, 19164; TU 80, 223, 5049, 8552, 21223, 24638, 24667, 24676, 24711, 24754, 24769; UMML 11384. Georeta: Tulsa U 5; UMMZ 88389, 88519, 88672, 134636, 157885, 163942, 163965, 164067, 164079, 165878, 168818; UMMZ (Fletcher nos. 6, 23, 45, 77, 78); CU 4044, 15429, 15762, 15806, 15964, 17195, 17216, 17239, 17258, 17299, 17323, 17371, 17456, 17498, 17629, 17650, 17781; CU (Raney no. 1574); USNM 162364, 162453, 162466, 162469, 168042, 168054; TU 14189; UGa 443, 445, 538, 542. Louisiana: USNM 163979; UMMZ (Delavan and Creaser no. 31-59, stream, 0.5 mi. N. Ethel, East Feliciana Parish); UMMZ 165880 (Chefuncte R., Hwy. 35, 10.5 mi. SW. Franklinton, Tangipahoa-Washington Parish line), 166149; TU 1218 (trib. Tangipahoa R., 1.5 mi. W. Bolivar, Tangi- *Material indicated by an asterisk has been re-examined. 70 U.S. NATIONAL MUSEUM BULLETIN 282. pahoa Parish), 6153 (Richland Cr., 7 mi. NE. Norwood, East Feliciana Parish), 14072 (stream, Hwy. 10, 1.4 mi. W. Darlington, Saint Helena Parish), 14076 (stream, U.S. Hwy. 51, 1.4 mi. N. Kentwood, Tangipahoa Parish). MIssiss1pP1: CU 11619, 11851, 12580, 15627, 15656, 15677, 16245, 16612; USNM 129409, 175388, 201247, 201249; TU 820, 1618, 14086 (stream, U.S. Hwy. 51, 2.4 mi. S. Magnolia, Pike Co.), 14158, 14164, 17731, 19798, 23416, 23456, 23744; TU (McGee Cr., Hwy. 98, Tylertown, Walthall Co.; Copiah Cr., Hwy. 27, 2.4 mi. S. Georgetown, Copiah Co.); AF 3443, 4150 (trib. of Tangipahoa R., Pike Co.), 6271; UMMZ 113778 (trib. of Bogue Chitto R., 3 mi. N. Summit, Pike Co.), 146610 (creek, 1 mi., 1.5 mi., and 2 mi. S. Centreville, Wilkinson Co.), 155381, 155442, 157790, 157812, 161158 (Little Bahala Cr., U.S. Hwy. 51, 3.5 mi. N. Brook- haven, Lincoln Co.), 161182 (Big Cr., U.S. Hwy. 51, 0.5 mi. N. Bogue Chitto, Lincoln Co.), 161191 (stream, U.S. Hwy. 51, 2.5 mi. N. Bogue Chitto, Lincoln Co.), 163715, 166120; UMMZ (Walker nos. 39-31, 39-32, 39-33, 39-37, 39-38, 39-49). SoutH Carouina: CU (Suttkus no. 1210, trib. of Broad R., 5.1 mi. S. Allendale, Allendale Co.); CU 15151. 15315 (trib. of Jackson Br., Hwy. 5, 1.8 mi. S. Sycamore, Allendale Co.), 19081 (trib., North Fk. Edisto R., Hwy. 178, 11.5 mi. NW. Orangeburg, Orangeburg Co.), 19655 (Little R., Hwy. 22, 5.6 mi. E. Calhoun Falls, Abbeville Co.); USNM 162350 (trib., South Fk. Edisto R., U.S. Hwy. 1, about 10 mi. NNE. Aiken, Aiken Co.), 162522, 162555, 162574, 168117 (Keowee R., Hwy. 183, 14 mi. E. Walhalla, Oconee Co.), 192667 thru 192670, 192671 (Goodbys Cr., U.S. Hwy. 176, 2 mi. S. U.S. Hwy. 301, Orangeburg Co.), 192672 thru 192674, 192675 (Edisto R., 2.75 mi. SSE. Branchville, Bam- berg Co.), 192676 thru 192681, 200481; TU 14190 (Seneca R., U.S. Hwy. 76, 8.4 mi. SE. Pendleton, Anderson Co.). Typre.—This species was originally described from a single specimen which was probably destroyed in the Indiana University fire of 1883. The type-locality is Silver Creek, one mile above its junction with the Etowah River, Georgia. Diaenosis.—Noturus leptacanthus is the only member of the subgenus Schilbeodes that typically has eight pelvic rays and an included lower jaw. The pectoral spine is short, weakly if at all grooved, and devoid of serrae, usually not even roughened on the posterior edge. There are extremely large chromatophores (obscured in old individuals) scattered over the body and fins. The adipose and caudal fins are united. There are 47 to 58 caudal rays, 14 to 19 anal rays, 7 to 10, usually 8 or 9 soft pectoral rays, 2 internasal pores, and normally 11 preoperculomandibular pores. Description.—Other counts and measurements are given in tables 17 to 26. The body is moderately elongate, little deeper forward than behind; head rounded above, slightly arched, depressed between the eyes; eye small, 2.0 to 2.7 times in snout; lower jaw included; pre- maxillary tooth patch with rounded posterior corners; humeral process moderate, usually a little longer than the diameter of the pectoral spine; pectoral spine relatively short, with very shallow grooves and without a trace of serrae; dorsal spine stout; adipose fin always well connected to the caudal fin, but a shallow, broad CATFISH GENUS NOTURUS RAFINESQUE a notch is formed at their juncture; upper procurrent caudal rays moderate in length; caudal fin truncate or slightly rounded behind. There are usually 6 soft dorsal rays; in 111 specimens, there are five (in 2), six (108), and seven (1). There are five to eight gill rakers on the first arch. The largest specimen examined is 78.7 mm. in standard length. There are (extremes in parentheses): (17) 17 to 20 (23), mean 19.2, upper simple caudal rays; (14) 15 to 18 (19), mean 16.7, branched caudal rays, of which 7 are usually in the upper one- half of the fin and 9 or 10 are most frequent in the lower one-half; and (12) 14 to 17 (20), mean 15.6, lower simple caudal rays. In 13 stained specimens vertebrae anterior to the origin of the anal fin eleven (in 7), twelve (5), and one not counted; ossified pectoral radials fused on all 26 sides. The general life color is reddish brown or a dark yellowish brown. The pigmentation in preserved material is as follows: top of head dark brown; side of body and dorsal surface uniform with slight darkening at the base of the dorsal and adipose fins; area over air bladder frequently dark gray; lower cheek and side of head pale yellow- ish; upper barbels light tipped, mostly pigmented only at their base; anterior naris pale; lower surface of head, abdomen, and pelvic fin creamy white, almost immaculate, but each with scattered pigment, especially concentrated in front of mental barbels; anal and dorsal fins light with moderate, scattered pigment; pectoral fin with pigment concentrated near base, scattered elsewhere; caudal and adipose fins sometimes with a light margin, darkly blotched with large gray-black chromatophores which often diffuse throughout the fins; large brownish chromatophores scattered over fins, head, and side; mental barbels usually immaculate except at base. VARIATION.—Table 7 gives variation in the number of pelvic and soft pectoral rays. Counts from one or more, usually several, collec- tions from each river system are combined; enumerations were usually made of rays of both fin pairs of the same specimens, and both sides are tabulated individually. The pelvic rays are rather uniformly eight, but there is a strong increase in pectoral rays from west to east. No correlation between pelvic and pectoral ray counts is indicated. The increase in pectoral rays is a moderately uniform gradient, but certain modes are evident. West of the Pascagoula River system, the modal number of rays is eight. From the Tombigbee and Alabama River systems eastward, the mode is nine. There are normally eleven preoperculomandibular pores throughout most or the range. However, 19 of 22 canals tabulated from the Comite River system (the western edge of the known range), south- western Mississippi, have ten pores, and only three sides have eleven G2 U.S. NATIONAL MUSEUM BULLETIN 282, (mean 10.14). Pore counts of 192 canals from other areas show no trend; the range is ten to twelve, and the average is 10.96. The number of vertebrae appear to be lowest in the west, especially the Comite and Chefuncte drainages of Mississippi and Louisiana. In this area the modes are 33 or 34 vertebrae. Elsewhere the modes in the counts obtained are 35. The following summary lists in order the number of specimens, range, and mean number of vertebrae for each river system: Comite 11 (32-35) 33.6; Chefuncte 9 (33-35) 34.0; Pascagoula 8 (34-35) 34.6; Tombigbee 21 (35-37) 35.4; Alabama 15 (34-35) 34.7; Apalachicola 9 (34-36) 35.1; Savannah 20 (35-37) 35.3; Combahee 3 (35) 35.0; Edisto 3 (35-36) 35.3. The color pattern and body form are relatively uniform, and no geographic trend is evident in the other tabulations. The number of counts, the range (in parentheses), and the mean are given in sequence for grouped samples from the following areas: (a) Lake Pontchartrain drainage, (b) Pearl River drainage, (c) the area from the Pascagoula River to the Tombigbee River, (d) Pensacola Bay drainage, (e) the Apalachicola River to the Ochlockonee River, (f) Suwannee River drainage, (g) the area east of the Satilla River, including the Great Cypress Swamp, and (h) the total. Anal rays: (a) 19 (16-17) 16.47; (b) 10 (15-16) 15.80; (c) 6 (15-16) 15.67; (d) 40 (15-18) 15.93; (e) 20 (14-17) 16.00; (f) 15 (14-17) 15.60; (g) 13 (15-19) 16.62; (h) 124 (14-19) 16.04. Lower-half caudal rays: (a) 11 (24-27) 24.91; (b) 10 (24-26) 25.30; (c) 6 (23-27) 24.83; (d) 40 (23-28) 25.35; (e) 21 (22-28) 25.05; (f) 15 (24-27) 25.00; (g) 13 (23-27) 24.69; (h) 116 (22-28) 25.10. Upper-half caudal rays: (a) 11 (24-28) 25.64; (b) 10 (25-28) 26.80; (c) 6 (23-26) 25.50; (d) 40 (24-30) 26.30; (e) 21 (24-29) 26.38; (f) 15 (25-29) 26.53; (g) 13 (24-27) 25.62; (h) 116 (23-30) 26.21. Total caudal rays: (a) 20 (48-57) 51.10; (b) 10 (50-54) 52.10; (c) 6 (47-56) 50.33; (d) 40 (48-58) 51.65; (e) 21 (47-57) 51.43; (f) 15 (50-54) 51.53; (g) 13 (48-53) 50.31; (h) 125 (47-58) 51.34. Distrisution.—Noturus leptacanthus (map 3) occurs in both the Atlantic and Gulf coastal drainages. It is found in most streams from the Amite and Comite Rivers system of Louisiana and southwestern Mississippi eastward to both the Great Cypress Swamp and Edisto River, South Carolina. The Florida distribution, however, is poorly known. Aside from a few uncertain records (included in the above synonymy, but which perhaps refer to Noturus gyrinus), this species is known to extend no farther into peninsular Florida than the Su- wannee River system on the west and the Saint Johns basin in the east. The distribution of N. leptacanthus is strikingly similar to that of Percina nigrofasciata as worked out by Crawford (1956, map 1). The CATFISH GENUS NOTURUS RAFINESQUE 73 SCALE OF MILES Map 3.—Distribution of Noturus leptacanthus Jordan. The circles represent localities from which I have examined specimens. The dashed line outlines the range, which includes those literature records th.t appear to be correct. distribution maps of the two species show identical ranges except that Crawford plotted a single record from the Mississippi River basin, and he showed no record from Great Cypress Swamp, South Carolina. Some specimens referable to Noturus nocturnus have been reported as Noturus leptacanthus. NoMENCLATURE.—The original description of Noturus leptacanthus Jordan (1877b, p. 352) unquestionably applies to the species con- sidered here. The diagnostic characters include the color ‘pale reddish yellow, slightly blotched,’ the small anal fin, the eight soft rays of the pectoral fin, the ‘‘small and slender dorsal and pectoral spines which are devoid of internal serratures,”’ and the “upper jaw much the longer.”’ Erymo.tocy.—The name leptacanthus (from Greek) means [small or] slender spined. It was presumably used in reference to the absence of the roughenings on the pectoral spine. RELATIONSHIPS.—The large chromatophores and smooth slender spines are exceptional in Noturus. The general morphology, however, 74 U.S. NATIONAL MUSEUM BULLETIN 282 points to a relationship with the species in the subgenus Schilbeodes, with an early divergence from most members of that group. Its rela- tionship with any other species does not appear to be intimate. Ecotoay.—Noturus leptacanthus is most frequently taken from small to moderate size streams, and appears to be chiefly a creek species. It prefers moderate to fast current, living on riffles of coarse sand or large gravel. Other species of Noturus that have been collected with leptacanthus are gyrinus, nocturnus, funebris, insignis, and munitus. In addition to the similar range, Percina nigrofasciata appears to be a common darter associate of N. leptacanthus. Although leptacanthus is not collected as frequently as nigrofasciata, because of its cryptic habits, the number of collections in which the two are taken together is striking. Noturus nocturnus Jordan and Gilbert FRECKLED MADTOM PuaTEs 3 (FIG. 5), 8 (FIG. 1); Map 4 Noturus leptacanthus Jordan [misidentifications]—Hay, 1881, pp. 514, 515, and Swain and Kalb, 1883, p. 642 (description; Chickasawha R., Enterprise,* Mississippi).—Hay, 1883, pp. 73-74 (distribution; Enterprise,* Mississippi; Big Black R., near Edwards [probably, USNM 32301,* Mississippi?, 1882, O. P. Hay], Mississippi). Noturus nocturnus Jordan and Gilbert.—Jordan, 1885, p. 802 (nomen nudun; range).—Graham, 1885b, p. 77 (nomen nudum; Arkansas R. or branches near Fort Smith; hypothetical in Kansas).—Jordan and Gilbert, 1886, pp. 6-17 (original description; best specimens, USNM 36461, from Saline R., Benton,* Ark.; other records: Poteau R.,* Okla.; Washita R., about 0.5 mi. above Arkadelphia,* Ark.; Sabine R., 5 mi. S. of Longview,* Tex.; Trinity R., Dallas,* Tex. [others are Noturus gyrinus]).—Jordan, 1889, p. 353 (rela- tionship) ; 1890, pp. 161, 165 (description; Big [Pigeon R.] Cr., Evansville,* Ind.).—Meek, 1891, p. 138 (Ouachita R., near Crystal Springs* and [West Fork] Saline R., about 24 mi. E. of Hot Springs, Ark.); 1893, p. 229 (range; compiled records from Arkansas only); 1894a, pp. 90-92 (Arkansas records compiled).—Evermann and Kendall, 1894, pp. 80-96, pl. 11 (type ascribed in error to Sabine R., Belton [for Saline R.,* Benton], Ark.; type figured; type-locality also indicated as Poteau R., near Fort Smith, Ark.; records compiled).—Hay, 1894, p. 172.—Meek, 1896, pp. 342, 346 (Walnut Cr., Kiamichi [Okla.]; Little and St. Francis Rivers, near Marked Tree [now re- corded: CNHM 1581, Marked Tree,* Ark., S. E. Meek and CNHM 764, St. Francis R., Greenway,* Ark., S. E. Meek], Ark.).—Hay, 1902, p. 70 (com- piled).—Parks and Cory, 1938, p. 21 (compiled).—Bohlke, 1953, p. 43 (syn- types, SU 562,* SU 564,* and USNM 36461*).—Gerking, 1955, pp. 51, 76 (Indiana records; key).—Bailey, 1956, p. 338 (hypothetical, Ilowa).—Eddy, 1957, p. 152, fig. 384 (key; range).—Taylor, 1957, p. 192.—Clarke, Breukel- man, and Andrews, 1958, p. 168 (Lyon Co., Kans.).—Hancock and Sublette, 1958, p. 46 (Louisiana: stations 4,* 5,* 12, 13*).—Cook, 1959, pp. 34, 135, *Material indicated by an asterisk has been re-examined. al ee eee CATFISH GENUS NOTURUS RAFINESQUE 75 142, fig. 25E (description; Mississippi records).—Minckley and Deacon, 1959, pp. 348, 349 (in food Pylodictis, Neosho R.).—Metcalf, 1959, pp. 383, 393 (distribution and records, Kansas and Oklahoma).—Deacon, 1961, pp. 395, 421-422 (Neosho R., Kans., distribution).—J. M. Walker, 1962, p. 38 (Lincoln Parish, La.).—Larimore and Smith, 1963, pp. 324-330 (record, Champaign Co., Ill.).—P. W. Smith, 1965, p. 9 (Illinois distribution).— Norden, 1965, p. 102 (records, Little R., La.).—B. T. Walker, 1965, p. 108 (Bayou D’Arbonne, La., station 16*).—Grosvenor, 1965, p. 273 (color photo~ graph).—Suttkus and Taylor, 1965, p. 177 (associations).—Cross, 1967, pp. 197, 218 (figure, description, Kansas distribution).—?Charles, 1967, pp. 385- 396 (Green R., Ky. [some or all may be Noturus gyrinus]).—Branson, 1967, p. 146 (Oklahoma and Kansas records). Schilbeodes nocturnus (Jordan and Gilbert).—Eigenmann and Beeson, 1894a, p. 82; 1894b, p. 45 [and 1905, p. 121] (compiled).—Eigenmann, 1896, p. 253 (Indiana).—Jordan and Evermann, 1896a, pp. 145-146 (comparison; range; type-locality, Saline R., Benton, Ark.; types,* USNM 36461); 1896b, p. 234 (range; type-locality indicated); 1900, p. 3236, pl. 27, fig. 64 (type, USNM 36461, from Sabine [for Saline] R.).—Large, 1903, pp. 9-10 [and 1905, pp. 56-57] (comparison; 10 Illinois records: 2, Havana;* 3, near Lin- coln [Salt Cr.*]; 3, Kaskaskia R. system [Lanx Ford,* ?Clinton Co.; Shelby- ville,* Shelby Co.; Crab Apple Cr.,* Moultrie Co.]; Spoon R., near Lewis- town;* South Fork Saline R.,* Saline Co.).—Jordan, 1904, p. 351.—Reed, 1907, pp. 555-564 (description of poison glands).—Meek, 1908, p. 141 (range; compiled record).—Forbes and Richardson, 1909 [and 1920, pp. - Ixxviii-ciii], pp. Ixxiii-xeviii, 196-199 (Illinois distribution and range [in error, see Large, 1903, pp. 9-10]; description; figures; synonymy).—Forbes, 1909, pp. 387-415 (range and Illinois distribution [in error]).—Hankinson, 1913, p. 109 (Kickapoo Cr.;* Kaskaskia R.* [Cooks Mills]; Flat Branch, all near Charleston, Ill.).—Eigenmann, 1919, pp. 398-399 (“‘other places in Texas” [partly N. gyrinus]).—H. S. Pratt, 1923, pp. 95, 97 [and 1935, p. 89].—Reed, 1924b, pp. 232, 256 (spine glands).—Greene, 1927, p. 309 (Illinois; hypothetical, Wisconsin).—Jordan, 1929, p. 94.—Luce, 1933, p. 118 ({2 mi. below] Vandalia,* Keyesport, and Venedy Station, on Kaskaskia R., Ill.; ecology).—O’ Donnell, 1935, p. 484 (compiled).—Greene, 1935, p. 219.—Schrenkeisen, 1938, p. 167.—Kuhne, 1939, p. 68 (Tennessee).—Hubbs and Lagler, 1939, p. 26 (key only).—Lamb, 1941, p. 44 (records, San Jacinto R. system, Tex., include Winters Bayou* [6 mi. N. of river]).—Aitken, 1941, p. 390 (hypothetical, Iowa).—Hubbs and Lagler, 1941, pp. 63-64 (comparison; range; ecology; “‘possibly divisible into at least 3 subspecies or species’ [but not in Great Lakes]).—Toole, 1943, p. 12 (figured; descrip- tion; Texas).—Gerking, 1945, pp. 74-75 (Big Pigeon Cr., Evansville,* Ind., only).—Hubbs, 1946, p. 38 (Oklahoma).—Higginbotham, 1947, pp. 462-464 (greatest oxygen consumption in late afternoon).—Hubbs and Lagler, 1947 [and 1949], pp. 71, 73, fig. 176 (comparison; range [not Great Lakes]; speci- men figured, Dunklin Co., Mo.).—Trautman, 1948, pp. 166-171 (description; Great Lakes basin record a hybrid; Oklahoma and Missouri).—Moore and Paden, 1950, p. 87 (ecology; Illinois R., Oklahoma records: Stations 6,* 7,* and 8*).—Baughman, 1950, p. 131 (Texas).—Bailey, 1951, p. 196 (hypo- thetical, SE. Iowa; recorded, NE. Missouri).—Cross and Moore, 1952, pp. 406-407 (type-locality indicated; Poteau R., Oklahoma records: Stations 7,* in part, 15,* 22, 24,* in part, 25, 28, M2,* M3,* Miz,* 29, 33,* M,*).— *Material indicated by an asterisk has been re-examined. 298-943 O—69——6 76 U.S. NATIONAL MUSEUM BULLETIN 282 G. A. Moore, 1952, p. [6] (Oklahoma).—Jurgens and Hubbs, 1953, p. [3] (Texas).—Moore and Buck, 1953, p. 24 (description; Oklahoma records) .— Knapp, 1953, pp. 77-78, fig. 108.—Hall, 1954, p. 59 (Oklahoma records in- cluding Grand R. [about 4 mi. E. of]* Choteau, Mayes Co.).—Cross, 1954, p. 311 (riffle, South Fork Cottonwood R., Chase Co., Kans.); 1955, pp. 475-476 (description; ecology; Chikaskia R.; Kansas records: Neosho R. in Coffey, Labette, and Lyon Counties; trib. to Four-Mile Cr.,* Sedgwick Co.).—Hall, 1955, p. 36 (reservoir population reduced, Illinois R., Okla.).— G. A. Moore, 1957, pp. 148, 144, fig. 2-79E (description; range).—Hubbs, 1957, p. 96 (Texas distribution).—Schelske, 1957, pp. 32-47 (occurrence, Verdigris R. system, Kans.).—Hubbs and Lagler, 1958, p. 91, fig. 176 (not in Great Lakes).—C. Hubbs, 1958, p. 8 (Texas distribution).—Lambou, 1959a, pp. 49-52 (Louisiana records).—Birkhead, 1967, pp. 101-110 (com- parative toxicity of venoms). Rabida nocturna (Jordan and Gilbert).—Jordan, Evermann, and Clark, 1930, p. 156.—Blatchley, 1938, p. 67 (compiled).—Driver, 1942, p. 254 (comparison; range [in error]); 1950, p. 262. Noturus flavus Rafinesque [misidentifications]—Forbes, 1909 and 1914, map 57, and Forbes and Richardson, 1909 [and 1920], map 57 (Illinois distribution, in part [Havana;* Crooked Cr., Ripley ;* “north line’’,* Brown Co.})). Schilbeodes gyrinus (Mitchill) [misidentification]—Forbes, 1909 and 1914, and Forbes and Richardson, 1909 [and 1920], map 58 ({Mackinaw Cr.,* Taze- well Co., Illinois]). Schilbeodes insignis (Richardson) [misidentification]—Hubbs and Raney, 1944, map 1 ({Sanbois R., 7 mi. 8. of Stigler,* Haskell Co., Okla.]). TYPE-SPECIMENS.—USNM 36461 (lectotype) and USNM 201388 (2 paralectotypes), Saline River, near Benton railroad bridge, Arkan- sas, 1884, David Starr Jordan and C. H. Gilbert. SU 562 (4 para- lectotypes) and USNM _ 36383 (28 paralectotypes), Poteau R., Slate Ford, W. of Hackett, Oklahoma, 1884, Jordan and Gilbert. SU 564 (2 paralectotypes) and USNM 36426 (15 paralectotypes), Washita [Ouachita] R., about 0.5 mi. above Arkadelphia, Arkansas, 1884, Jordan and Gilbert. OTHER MATERIAL STUDIED UNITED STATES: Axraspama: UMMZ 167361 (Alabama R., Camden, Wilcox Co.); USNM 200440 (trib., Uphapee Cr., 7.0 mi. SW. Auburn, Lee Co.), 200475 (Cahaba R., Harrisburg bridge, near Centreville, Bibb Co.), 201241 (Tensaw R., Stockton, Baldwin Co.). ARKANSAS: Tulsa U (Lee Cr., at Arkansas- Oklahoma line); UMMZ 167217, 167219, 167305, 169947, 169995; UMMZ (Burnett Springs, 2 mi. N. of Paragould, Greene Co.); TU 10351, 12292, 15594; USNM 172406; CU 42244. Itiinois: UMMZ 105762, 165883; UMMZ (Bau- man nos. 45, 68, 75, 76-77). Inp1ana: UMMZ 113553 (White R., Hindustan Falls, near Shoals). Kansas: UMMZ 111483; KU 2351 (trib., Four Mile Cr., sec. 36, T. 27 S., R. 2 E., Sedgwick Co.); USNM 172048, 200781. Kentucky: UMMZ 126943 (Laurel Cr., near mouth, trib. to Little Sandy R.); USNM 63809; UL 8674 (Rough R., 5.5 mi. N. and 3 mi. W. Leitchfield). Lourstana: UMMZ 113722, 165881, 184025; UMMZ (Taylor coll.: La. nos. 55-21¢e, 55-33, 55-43, 55-46; Comite R., about 1.5 mi. NE. Olive Branch, East Feliciana Parish) ; *Material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE ee CU 15525; TU 765, 882, 1335, 2068, 3932, 4258, 4875, 6252, 6265, 13248, 13347, 13625, 13680, 13714, 14089, 14833, 15844, 16795, 20908; TU (Big Cr., Pollock, Grant Parish); USNM 163119, 172110, 172333, 172528, 172550, 172638, 172654, 172670, 172682, 172716, 172808, 172873, 172884, 172926, 172962, 172980, 172998, 173056, 173100, 173118, 173143, 173169, 173282. Mussisstep1: UMMZ 163717; UMMZ (Walker no. 39-32; Sandy Cr., 15 mi. N. of Iuka, 1 mi. W. of Hwy. 25); AF 3255 (Okahatta Cr., trib. to Chunky Cr., Newton Co.); TU 6255, 14941, 15184, 17729, 19786, 23514. Mussour1r: UMMZ 139501, 139569, 139585, 139649, 139723, 142185, 142203, 148564, 149251, 149739, 149994 (Mississippi R., Canton, Lewis Co.), 150225 (Osage R., 5 mi. SE. of Amoret or 5 mi. NW. of Foster, Bates Co.), 152749, 153130, 153151, 164582; INHS (lower St. Francis R., Butler and Dunklin Counties; Osage R., N. of Schell City, Vernon and Bates Counties; Meramec R., 2 mi. SE. of St. Clair, Franklin Co.; Salt R., 6 mi. N. of Center, Ralls Co.); UMoMZ (A.C. Bauman nos. 50, 54; C.B. Obrecht no. 41-29, Current R., 7 mi. N. Doniphan, Ripley Co.). Oklahoma: UMMZ 156771, 167176, 167177, 167178, 167182 (Chikaskia R., below L. Blackwell, Kay Co.), 167184 (trib. to Washita R., Dougherty bridge, Murray Co.), 167195, 167201 (Shoofly Cr., at bridge N. of Braman, Hwy. 177, Kay Co.), 167214; OAM 4112; Tulsa U (Grand R., at Earbob Cr., Mayes-Wagoner Co. line; creek, 0.5 mi. N. of Okay; Big Skin Bayou, U. S. Hwy. 64; Brushy Cr., 2 mi. S. of Choteau, Mayes Co.; Bird Cr., Avant, Osage Co.; creek, 10.5 mi. S. Antlers, Choctaw Co.; California Cr., 2 mi. S. Delaware; Poteau R., 4 mi. SE. Heavener, Le Flore Co.; Pryor Cr., 1.5 mi. S. Pryor; Caney R., Hulah Dam, Osage Co.; Little Beaver Cr., 1 mi. W. Hardy, Kay Co.; Bird Cr., 1.5 mi. NW. Catoosa; Verdigris R., 1 mi. N. of U.S. Hwy. 66, Rogers Co.; Big Lee Cr., 2 mi. E. Nicut) ; TU 13776 (Chikaskia R., 5.4 mi. NW. Braman, below Lake Blackwell); USNM 165779, 165796; CU 14134. TENNESSEE: UMMZ 168328; USNM 193480, 193481 (Rose Cr., 2 mi. above mouth, Hardeman Co.); Tennessee Game and Fish Comm. (Cumberland R., mi. 232, Old Hickory Reservoir). Texas: UMMZ 161774; UMMZ (Delavan and Creaser no. 31-39, White Rock Cr., Preston Road, 12 mi. N. of Dallas) ; TNHC 1046, 1196 (trib. of Peach Cr., Hwy. 105, 15 mi. E. of Conroe, Mont- gomery Co.), 1388 (West Fk. San Jacinto R., 3 mi. W. of Conroe, Montgomery Co.), 1439 (trib. to Long Cr., 2 mi. N. of Goodrich, Hwy. 59, Polk Co.); TU 3384, 3415, 3477, 3505, 3791 (trib., Trinity R., Hwy. 294, 3.7 mi. W. Elkhart, i Co.), 3853, 14042, 14044, 14063 (stream, Hwy. 105, 8.1 mi. E. Conroe, Montgomery Co.), 14066, 14323, 14346, 21679, 21734, 21797, 21910. Diaenosis.—Noturus nocturnus, subgenus Schilbeodes, has 55 to 64 caudal rays; 15 to 20, seldom more than 18 anal rays; 8 to 10, typically 9 pelvic rays; 7 to 11, modally 9 or 10 soft pectoral rays; 2 internasal pores; and normally either 10 or 11 preoperculomandibular pores. The lower jaw is included. The pectoral spine is roughened behind; there are seldom distinct serrae. The vertical fins frequently have narrow, dark submarginal bands and very narrow, clear margins; the dorsal and lateral surface of the body and the vertical fins other- wise are nearly unicolor. N. nocturnus is superficially similar to N. insignis, N. funebris, and N. phaeus; it is easily segregated from the other species of Schilbeodes by several of the above characters. Distinguishing it from insignis are the more rounded head, shorter body, and the usual absence of serrae on the posterior edge of the pectoral spine. The poorly pig- 78 U.S. NATIONAL MUSEUM BULLETIN 282. mented lower surface of the head and abdomen, the normally stiff dorsal spine, and the shorter anal fin distinguish it from funebris and phaeus. DxscripTion.—Counts and measurements are given in tables 17 to 26. Body only moderately elongate; caudal peduncle tapering posteriorly, slightly narrower than head depth; eye small, 2.0 to 3.5 times in snout; gill rakers on first arch 5 to 7; largest specimen, an exceptionally large individual from Missouri, 121.5 mm. in standard length; most are under 100 mm. in standard length. Head rounded above, slightly depressed forward; lower jaw in- cluded; premaxillary tooth band with posterior corners rounded or, sometimes, obtusely angulate; dorsal spine relatively short, stout, its thickness about as in most Noturus; pectoral spine moderate in length and relatively straight, without anterior serrae; its posterior edge irregularly roughened or sometimes with as many as four irregularly developed serrae; posterior process of cleithrum short, its length about equal to the diameter of the pectoral spine; adipose fin broadly connected with the high procurrent caudal rays, without a distinct notch; caudal fin rounded or slightly pointed behind. In cleared and stained specimens: vertebrae anterior to origin of anal fin 11 (in 2), 12 (7), or 13 (7); hypurals all separate distally in 20 specimens or hypurals 2-3 fused in 4 specimens; ossified pectoral radials tightly fused on 27 sides or incompletely fused on 5 sides. Soft dorsal rays in 131 specimens: 5 (in 3), 6 (126), or 7(2). The caudal fin has 19 to 26, mean 23.02 upper simple rays; 17 to 23, usually 18 to 21, mean 19.6 branched rays of which usually 8, less frequently 7 (in specimens above 40 mm. in standard length), are in the upper-half and 10 to 13 are in the lower-half of the fin; and 13 to 19, mean 16.3 lower simple rays. Body color yellowish brown to dark brown, lighter below, "ind yellowish white on abdomen. Side and upper surface of body and head uniformly pigmented, without blotches or light markings; upper barbels dark; anterior edge and sometimes all of lower barbels pig- mented; lips dark; abdomen pale and immaculate or sometimes with a few scattered chromatophores, becoming covered with a diffuse, dusky pigment with age; a moderate band of pigment across abdomen in front of pelvic fins; branchiostegal membrane pale, sometimes flecked with a few melanophores; chin anterior to mental barbels heavily pigmented; dorsal, anal, adipose, pectoral, and pelvic fins heavily flecked with pigment near their beses, becoming more diffuse outward, with the margins to outer half often immaculate; anal fin often with a marginal to submarginal dark or dusky band and the edge clear; caudal fin typically darker than other fins, usually uni- CATFISH GENUS NOTURUS RAFINESQUE 79 formly dusky. Large specimens, above 90 mm., almost uniformly and heavily pigmented over upper part of body, fins, and abdomen. Typre.—Jordan and Gilbert (1886) listed Noturus nocturnus from six localities in Oklahoma, Arkansas, and Texas, but based their description on the specimens from Oklahoma and Arkansas. Their specimens from the Rio Lampasas, Texas, are not nocturnus but are Noturus gyrinus. Two of the localities have been variously listed (see synonymy) as the type-locality, beginning with Evermann and Kendall (1894). Evermann and Kendall were the first to publish a figure of nocturnus, based on a specimen from the Saline River, Benton, Arkansas. This figure was reprinted by Jordan and Ever- mann (1900, p. 3236, pl. 27, fig. 64) and stated to be from the type, USNM 36461, essentially designating a lectotype. When examined by me, USNM 36461 contained three specimens in agreement with the original catalog record. The largest of the three is herein selected as the lectotype because it most closely approaches the length of 2.3 inches noted on the drawing and is unique in having a dark longitudinal line near the middle of the caudal fin, as shown on the original drawing and many of the published illustrations. The lectotype retains the number USNM 36461. The lectotype is a male, 48.5 mm. in standard length. It has 18 anal rays, 26+8+11+17=62 caudal rays, and 6 soft dorsal rays. On each side there are nine pelvic rays, nine soft pectoral rays, two internasal pores, and eleven preoperculomandibular pores. The poste- rior edge of the pectoral spine is roughened, without serrae. The head length is stepped into the standard length 3.8 times and the distance from the adipose notch to the tip of the caudal fin is stepped into the distance from the origin of the dorsal fin to the adipose notch 1.55 times. VaRiaTion.—WNoturus nocturnus is a variable species and appears to form distinctive localized populations. The most noteworthy of these is one in the Red River system. There, specimens are often light gray to yellowish and sometimes have extremely short spines. In this system the preoperculomandibular pores are often ten instead of eleven. Some Texas specimens are extremely elongate, but show no increase in meristic characters. The single specimen from the Tennessee basin in Mississippi has 20 anal rays, an unusually high number. Color variations are a prominent brownish band near the edge of the anal fin and a rather broad white posterior margin of the caudal fin, but there is no definite geographic trend. The specimens from the Alabama River have a high number of soft pectoral rays, modally ten; other specimens from eastern Gulf tributaries tend to have a higher number of pectoral rays than elsewhere. 80 U.S. NATIONAL MUSEUM BULLETIN 282 Aside from the peculiar pore counts of populations in the Red River system, the gradients in meristic characters appear ta be slight. The following data are arranged in the sequence: number of tabulations, range (in parentheses), and mean by drainage. Vertebrae: San Jacinto R., Texas 8 (35-37) 35.75; Neches R., Texas 2 (37) 37.00; Red R., Oklahoma and Louisiana 24 (35-38) 37.04; Ouachita R., Louisiana 16 (36-38) 36.56; Arkansas R., Oklahoma 11 (35-37) 36.09; Black R., Missouri 1 (38) 38.00; Cache R., Illinois 14 (36-37) 36.57; Alabama R., Alabama 7 (36-37) 36.86; total 83 (35-38) 36.61. The following data, in the same sequence, are combined from broad geographic areas: (a) Texas, (b) Red River system, Oklahoma, (c) Red River system, Louisiana, (d) Arkansas River system, Oklahoma and Kansas, (e) Ouachita River system, (f) Mississippi River drainage above Arkansas River, (g) tributaries to Gulf of Mexico between Mississippi and Mobile Rivers, (h) Mobile River drainage, (i) total. Pelvic rays: (a) 56 (8-10) 9.02; (b) 30 (8-9) 8.93; (c) 24 (9-10) 9.13; (d) 46 (9-10) 9.17; (e) 44 (9-10) 9.05; (f) 62 (8-10) 8.98; (g) 44 (8-10) 9.00; (bh) 20 (9-10) 9.15; (i) 326 (8-10) 9.04. Soft pectoral rays: (a) 54 (8-10) 9.06; (b) 30 (8-10) 9.23; (c) 22 (9-10) 9.23; (d) 46 (9-10) 9.30; (e) 44 (9-10) 9.11; (£) 63 (7-10) 8.98; (gz) 42 (9-10) 9.43; (h) 20 (9-11) 9.85; (i) 321 (7-11) 9.21. Anal rays: (a) 22 (16-18) 16.95; (b) 19 (15-18) 16.68; (c) 12 (17-19) 18.00; (d) 12 (15-18) 16.50; (e) 24 (15-18) 17.13; (£) 32 (15-20) 16.97; (g) 24 (16-19) 17.33; (hb) 10 (16-18) 16.70; (i) 155 (15-20) 17.04. Lower-half caudal rays: (a) 22 (27-31) 28.50; (b) 14 (26-30) 27.64; (c) 12 (28-31) 29.08; (d) 12 (27-30) 28.00; (e) 21 (26-29) 27.82; (f) 29 (26-30) 27.66; (g) 5 (28-30) 29.00; (h) 8 (27-29) 27.75; (i) 123 (26-31) 28.07. Upper-half caudal rays: (a) 22 (29-33) 31.00; (b) 14 (29-33) 31.00; (c) 12 (31-34) 32.25; (d) 12 (28-33) 30.33; (e) 21 (29-33) 30.91; (f) 29 (29-34) 30.59; (g) 5 (30-33) 31.20; (h) 8 (29-32) 30.38; (i) 123 (28-34) 30.89. Total caudal rays: (a) 22 (56-64) 59.50; (b) 19 (55-62) 58.68; (c) 12 (59-64) 61.33; (d) 12 (56-62) 58.33; (e) 21 (56-62) 58.73; (f) 30 (55-63) 58.23; (2) 5 (58-63) 60.20; (h) 8 (56-60) 58.13; (i) 129 (55-64) 58.94. Preoperculomandibular pores: (a) 64 (10-11) 10.89; (b) 225 (9-12) 10.26; (c) 114 (9-11) 10.27; (d) 180 (9-12) 10.86; (e) 203 (9-12) 10.94; (f) 54 (10-12) 10.80; (g) 52 (10-12) 10.92; (h) 20 (10-11) 10.95; (i) 916 (9-12) 10.66, including Calcasieu River, Louisiana 4 (11) 11.00. Preoperculomandibular pores in streams of Red River system: Clear Boggy River, Oklahoma 63 (9-11) 10.30; Kiamichi River, Oklahoma 12 (10-12) 10.33; Little River, Oklahoma 146 (9-11) CATFISH GENUS NOTURUS RAFINESQUE Sh 10.23; Washita River, Oklahoma 4 (10-11) 10.25. The modal number for each population in the Red River system, Oklahoma, is strongly ten; the mode for nearly all the other samples outside the Red River system is eleven. In northwestern Louisiana, specimens from Saline and Kisatchie Bayous have strong modes of ten; other populations in the Red River system, Louisiana, may have modes of ten or eleven but I have been unable to establish trends. The Red River populations are confusing because of the shift in pore count inside the range of the species. I do not know if this charac- ter is a geographic variation of nocturnus confined to the Red River drainage, or if two very similar species are involved. As noted previ- ously, some populations with the low pore count have relatively short spines and light color; others do not. Since there appear to be no other distinguishing characters I have been unable to identify populations or specimens consistently without counting pores, and in the lower portions of the drainage the pores may grade from ten to eleven. No samples have been observed containing two distinct morphological types. DistRIBuTION.—Noturus nocturnus (map 4) is found in the lower and central Mississippi drainage and in other tributaries to the Gulf of Mexico in Alabama, Mississippi, Louisiana, and Texas. Base-level streams and brackish water are probably avoided; otherwise nocturnus occurs throughout the lower Mississippi River system from and below the following: the bend of the Tennessee River in Mississippi, tribu- taries to the Ohio River in Indiana and Kentucky, tributaries to the upper Mississippi River in Missouri, the Illinois River in Illinois, the Osage River in Kansas and western Missouri, the Arkansas River system in Kansas and Oklahoma, and the Red River system in Texas and Oklahoma. It ranges eastward in most Gulf of Mexico tributaries to the Mobile River system, Alabama. Westward from the Mississippi drainage, nocturnus is found in many larger tributaries to the Gulf of Mexico, to the San Jacinto River in Texas. N. nocturnus is notably absent from the Ozark upland (upper White River system) and probably is sparsely distributed in the small eastern tributaries to the Mississippi River from Kentucky to Louisiana. The fauna of the latter region, however, has not been collected as ex- tensively as that of the Ozarks; the predominance of shifting sand streams probably limits the favored habitat here. The reason for the absence in the Ozark streams is unknown. In contrast, nocturnus ranges throughout much of the Ouachita upland. The hybrid Noturus gyrinus < Noturus miurus, and the following species of Noturus have been incorrectly recorded as nocturnus: gyrinus, exilis, flavus, and albater. 82 U.S. NATIONAL MUSEUM BULLETIN 282: $0 0 50 100 = _ 300 SCALE OF MILES Map 4.—Distribution of Noturus nocturnus Jordan and Gilbert. The circles represent localities from which I have examined specimens. All unverified literature records are from well within the area of the figures shown. RELATIONSHIPS.—Noturus nocturnus appears to be most closely related to Noturus insignis, from which it differs in the fewer hypurals, the shorter anal fin, the poorly serrated pectoral spine, the poor development of black margins on the vertical fins, the relatively short chunky body, and the rounded head. These characters, however, are subject to considerable variation from population to population in both species. The two species are strikingly similar in structure of the pectoral spine, the number of fin rays, the tendency to have dark marginal or submarginal bands on the vertical fins, the sensory canal system, and show resemblances in ecology. They seem to form a compact and closely related group allied to Noturus funebris and Noturus phaeus. Ecotoay.—This species lives in moderate size to large streams in riffles or in areas of moderate to fast current over a gravel and boulder bottom. It is seldom found in streams with shifting sand bottoms. The CATFISH GENUS NOTURUS RAFINESQUE 83 water is frequently turbid or slightly turbid, but it probably prefers clear water. It has been taken in collections with gyrinus, leptacanthus, exilis, phaeus, flavus, eleutherus, placidus, munitus, and miurus. Erymotocy.—The name nocturnus (Latin), meaning nocturnal, alludes to the dark color of this species. Noturus insignis (Richardson) MARGINED MADTOM PuatsEs 3 (Fia. 6), 6 (FIG. 2); Map 5 Pimelodon livrée [French vernacular].—LeSueur, 1819, pp. 155-156 (description; no locality); 1820, p. 44 (as Pimelode livrée; abstract of description; no locality).—Vaillant, 1896a, p. 28 and 1896b, p. 14 (synonymy).—Bertin and Estéve, 1950, p. 25 (synonymy; specimen, MNHN 3053, from Phila- delphia, United States). Pimelodus insigne Richardson, 1836, p. 132 (named as follows: ‘“‘The following species have been detected in the United States... . Pimelodus catus ... and insigne (livrée, LeSueur).’’). Noturus insignis (Richardson).—Jordan, 1877d, pp. 72-119, pl. 36, fig. 56, pl. 37, figs. 57-57c (synonymy; range; relationship; comparison; Pennsylvania specimens figured) ; 1877e, p. 611 (relationship); 1878a, p. 118 (Susquehanna R., Pa.; synonymy); 1878c, p. 368 (upper Ohio R. [not Indiana]); 1878d {and 1884], p. 335 (description; range; synonymy); 1878e, p. 414 (range; synonymy).—Jordan and Brayton, 1878, pp. 29, 87, 93 (range; Reedy R.,* Greenville Court-House, S.C.).—Bean, 1880, p. 112 (James R.,* Va.; [Conoy Cr.* and Susquehanna R.*], Bainbridge, Pa.; Potomac R.*).—Cope, 1881, p. 77 (description; range; synonymy; tribs. of Susquehanna R.,* Pa.).— Jordan and Gilbert, 1883, p. 100 (range; description; synonymy).—Swain and Kalb, 1883, pp. 638-644 (range and synonymy [in error]; description; Susquehanna R.*).—True, 1883, p. 258 (South Carolina).—Jordan, 1885, p. 802.—Jordan and Gilbert, 1886, p. 6 (comparison).—Jordan, 1889, pp. 352-353 (comparison and associations); 1890, pp. 101-136 (description, ecology, and associations; Virginia records including Shenandoah R.,* 5 mi. NW. of Luray; Buffalo Cr.,* Buffalo Mills; Blackwater R., Zuni;* Roanoke R., Salem* and Roanoke.* North Carolina records including Tar R., 2 mi. below Rocky Mount;* Little R., Goldsboro;* Neuse R.,* Millburnie; South Buffalo Cr., 5 mi. SE. of Greensboro;* Reedy Fork,* Haw R., Fulks Mill; Little Yadkin R.,* Lindsays Mills; ?Second Cr., trib. to Little R.; ?Jumping Run, 6 mi. N. of Salisbury; Catawba R., near Marion;* Bucks Cr., Pleasant Garden;* Johns R., near Morgantown*).—Bean, 1892, pp. 19-20, pl. 19, fig. 26 (description; synonymy; Susquehanna and Delaware drainages, Pa.; not viviparous).—Evermann and Cox, 1895, pp. 304-309 (compiled). Jordan, 1904, pp. 42, 351 (description; range).—Taylor, 1957, p. 192.— Eddy, 1957, p. 153, fig. 387.—Clugston and Cooper, 1960, p. 9 (age, growth, matur- ity, sex ratios, Bald Eagle Cr., Centre Co., Pa.).—Beyerle and Cooper, 1960, p. 256 (common, Kettle Cr., Pa., with trout).—Patrick, 1961, p. 256 (North Anna R.).—Schwartz, 1961, p. 25 (probable bait introduction into Deep Creek L.,* Md. and Ohio R. tributaries,* W. Va.).—McFadden and Cooper, 1962, pp. 54, 56 (Kettle and Shaver Creeks, Pa., with trout).— Tenney and Woolcott, 1964, pp. 17-20 (response to toxin).— Richards, 1966, *Material indicated by an asterisk has been re-examined. 84 U.S. NATIONAL MUSEUM BULLETIN 282: p. 829 (associations with Htheostoma).—New, 1966, p. 23 (eggs on under- surface of rocks?). Schilbeodes insignis (Richardson).—Jordan and Evermann, 1896a, pp. 145, 147 (description; range; synonymy); 1896b, p. 234.—Smith and Bean, 1899, p. 181 (vicinity of Washington, D.C.).—Jordan and Evermann, 1900, pl. 28, fig. 66 (Consy [Conoy] Cr.,* Bainbridge, Pa.).—W. F. Ross, 1902, pp. 112-120 (Bald Eagle Cr., Pa.; reproduction; ecology; food habits).—Bean, 1903, pp. 95-96, 740 (synonymy; description; Delaware and Susquehanna systems, New York).—Fowler, 1906a, pp. 171-173, pl. 15 (description; associations; Crosswicks Cr., N.J.); 1906b, p. 595 (Susquehanna tribs., Cameron Co., Pa. only).—MecConnell, 1906, p. 174 (Marsh Run of Bald Eagle Cr., Pa.).— Reed, 1907, pp. 555-564 (poison apparatus described).—H. M. Smith, 1907, p. 70 (synonymy; description; range; North Carolina distribution).—Fowler, 1907a, p. 14 (Pennsylvania records: Delaware R. basin, in the Schuylkill* and [Delaware R.], Holmesburg;* Susquehanna basin, Carlisle;* Conestoga Cr.* and Paradise,* Lancaster Co.); 1910, p. 927 (outlet, Lakemont Park L., Altoona,* Pa.); 1912b, p. 54 (Conowingo Cr., near Conowingo,* Md.).— Nichols, 1913, p. 92 (near New Yurk City).—Fowler, 1913, p. 92 (Pennsyl- vania records, in part; Cambria Co. [error for Blair Co.; Port Allegany in error]) ; 1914a, p. 357 (Deer Cr., Md.) ; 1915a, pp. 208-209 (in part; Assanpink Cr.,* Trenton, N.J.; Gynn Oak,* Md.; Pennsylvania records: [Delaware R.], Milanville;* Dingmans Ferry;* Delaware Water Gap;* Lopez.;* others relisted).—McAtee and Weed, 1915, pp. 5-10 (Maryland records: Rock Run;* Cabin John Run;* the Canal; Potomac R.;* Sycamore Cove).—Fowler, 1915b, p. 1 (hypothetical, Delaware Co., Pa.).—?Welsh, 1916, p. 54 (Little Peedee R., S.C.).—Radcliffe and Welsh, 1916, pp. 40-41 (Maryland records: Tenmile Cr., near Boyds;* Chesapeake and Ohio Canal, below Buzzards Hole*).—Fowler, 1917a, pp. 110, 117 (Harihokake Cr., near French Town,* N.J.; Pennsylvania records include Saucon Cr.,* Lanark; Martins Cr., near Bangor*) ; 1917b, pp. 34-36, plate (reproduction; nest figured; Middle Fork Tohickon Cr., Pa.); 1918c, p. 90 (Pennsylvania records).—Nichols, 1918, pp. 38, 103 (description; near New York City).—Fowler, 1919, p. 57 (Pennsyl- vania records including Jordan Cr., near Helfrichs Spring,* Lehigh Co.) ; 1920a, p. 299 (Paupack L. outlet, Pike Co., Pa.); 1920b, p. 150 (New Jersey counties); 1921la, p. 387 (Pennsylvania); 1921b, p. 63 (Bucks Co., Pennsyl- vania records including Haycock Cr.*); 1922, p. 16 (Depot Br., Greenville Co., 8.C.).—Breder and Crawford, 1922, p. 322 (as insignia; Oxon Run, District of Columbia).—H. S. Pratt, 1923, p. 96.—Fowler, 1923, pp. 9, 16 (Virginia records) ; 1924, p. 269 (Susquehanna R., near Towanda, Pa.; used for bait); 1925, p. 24 (Big Conewago Cr., Pa.).—Brimley and Mabee, 1925, p. 15 (Boones Pond, N.C.).—Coker, 1925, p. 59 (Paddys Cr., trib. to Catawba R., N.C.; ecology); 1926, p. 108 (associations; Paddys Cr., N.C.).—Fowler, 1927, p. 191 (Pine Cr., Galeton,* Pa.).—Fowler and Carlson, 1927, p. 66 (Pennsylvania records).—Hubbs and Greene, 1928, p. 390 (L. Ontario drainage).—Pickens, 1928, p. 30 (South Carolina).—Eaton, 1928, pp. 41-42 (Keuka L., N.Y.).—Greeley, 1928, pp. 87-105 (ecology; economics; New York records).—Hubbs and Brown, 1929, p. 4 (not in Great Lakes W. of Niagara Falls).—Truitt, Bean, and Fowler, 1929, pp. 36-37, fig. 6 (range; description; synonymy; Maryland counties).—Hildebrand, 1932, p. 53 (description; North Carolina record).—Greeley, 1934, p. 108, pl. 9 (colored figure, adult male, Catatonk Cr., N.Y.).—Stewart, 1935, p. 85 (region of *Material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 85 Lewisburg, Pa.).—Greeley, 1935, pp. 86, 96 (ecology; records, Mohawk R. system, N.Y.).—Odell, 1935, p. 132.—Greeley, 1936, pp. 46-85 (ecology; associations; reproduction; habits; economics; distribution and abundance, Delaware and Susquehanna systems, N.Y.; records).—Odell and Senning, 1936, pp. 93, 95 (New York records).—Hoover, 1936, p. 239 (Merrimack R., near Concord, N.H.).—Greeley, 1937, pp. 87, 97 (ecology; Wallkill and Rondout systems, N. Y.; Rondout R., Mill Hook) ; 1938, p. 69 (Chemung and Cohocton Rivers, N.Y.).—Odell and Senning, 1938, p. 99 (Waneta L.).— Bailey, 1938, pp. 151-182 (comparison; ecology; New Hampshire records: Soucook R., Loudon;* Merrimack R., Merrimack Co.; Pemigewasset R.,* 2 mi. S. Bristol).—Schrenkeisen, 1938, pp. 166-167.—-Bailey and Oliver, 1939, p. 152 (only in Merrimack drainage of New Hampshire).—Hubbs and Lagler, 1939, p. 26 (comparison; Great Lakes basin).—E. W. Surber, 1939, p. 331 (bait fish; Shenandoah and Potomac River systems).—Koster, 1939, p. 201 (associations; Danby Cr., N.Y.).—Van Duzer, 1939, p. 65 (associations; near Ithaca, N.Y.).—Fowler, 1940a, p. 11 (South Br. Rariton R., N.J.); 1940b, pp. 8-9 (Pennsylvania records).—Hubbs and Lagler, 1941, pp. 63-64 (comparison; range; ecology).—Fowler, [1944], p. 52 (Tobyhanna Cr., Pa.).— Hubbs and Raney, 1944, pp. 18-24 (synonymy, in part, including references to basis of the name insignis).—E. W. Surber, 1946, pp. 183-191 (effects of DDT).—Brimley, 1946, p. 15 (comparison; North Carolina distribution; Little R., near Bunnlevel).—Bertin and Estéve, 1950, p. 25 (synonymy).— R. F. Smith, 1950, pp. 58-59, 172 (Farrington L., N.J.; bass bait).—Cheng and James, 1960, p. 164 (new species Cestode from specimens, Sinking Cr., Va.). Rabida insignis (Richardson).—Jordan, 1929, p. 93.—Jordan, Evermann, and Clark, 1930, p. 156 (range; synonymy).—Fowler, 1933, p. 8 (Maryland records including Winters Run,* Harford Co.); 1935a, pp. 6-20 (range; South Carolina records: North Saluda R.* and Middle Saluda R.,* Green- ville Co.; Turkey Cr.,* Edgefield Co.; Thompson Cr.,* Chesterfield Co.; 6 mi. NW. of Bennettsville* [Hwy. 9], Marlboro Co.).—H. 8. Pratt, 1935, p. 90.—Fowler, 1936c, p. 193 (as insigne; Mill Cr., Old Fort, *N.C.); 1938, p. 103 (Pennsylvania records); 1939, p. 60 (New Jersey record).—Driver, 1942, pp. 252, 254 [and 1950, pp. 262-263]. Schilbeodes insignis insignis (Richardson).—Hubbs and Lagler, 1957, p. 5; 1958, pp. 89, 91, fig. 177. Noturus insignis insignis (Richardson).—R. D. Ross, 1959b, p. 4 (key; Shenan- doah R. system).—Flemer and Woolcott, 1966, pp. 78, 87 (distribution and food, Tuckahoe Cr., Va.). Pimelodus lemniscatus Valenciennes, in Cuvier and Valenciennes, 1840, pp. 144- 145 (original description; specimen received with LeSueur’s drawing and indicated by him, “Mem. du Mus. d’hist. natur. de Paris, t. V, p. 155,” under the name ‘‘Pimelode livrée’’)—De Kay, 1842, p. 187 (description; Southern States).—Storer, 1846, p. 405 [and in reprint, 1846, p. 153] (de- scription; synonymy; range).—Bleeker, 1858, pp. 192, 206 (United States) .— Girard, 1859, pp. 158-159 (LeSueur’s specimen believed from Pennsyl- vania).—Bertin and Estéve, 1950, p. 25 (synonymy; holotype, MNHN 3053, Philadelphia, United States). Noturus lemniscatus (Valenciennes).—Girard, 1859, pp. 158-159 (incorrectly as- signed to Pimelodus).—Gill, 1861a, p. 45 (synonymy).—Gimnther, 1864, p. 104 (synonymy; description; North America).—Jordan and Copeland, 1876, *Material indicated by an asterisk has been re-examined. 86 U.S. NATIONAL MUSEUM BULLETIN 282. p. 160 (Southern States).—Jordan, 1876b, p. 303 (description; range) ; 1877c, p. 50 (present in the [upper] Ohio; not seen by Rafinesque in Ohio Valley). Noturus marginatus Baird.—[Kennicott], 1861, p. 23 (nomen nudum; Carlisle,* Pa.).—Baird, in Cope, 1869, pp. 237, 241 (original description; Sinking Cr.* of the Kanawha; head of James R.; the Susquehanna*).—Cope, 1870 [and in reprint, 1877], pp. 484, 495 (Catawba* and Yadkin* Rivers, N.C.).— Jordan, 1876b, p. 303, and Jordan and Copeland, 1876, p. 160 (range [in error}]).—?Jordan and Gilbert, 1877a, p. 2 (Ohio Valley [not Indiana]).— Jordan, 1877b, pp. 371-372 (range [in error]; synonymy).—[Klippart], 1877, p. 153 (upper Ohio R.; not seen in Indiana).—Jordan and Evermann, 1896a, p. 147 (type-locality indicated as Pennsylvania, based on USNM 1571).— Fowler, 1915a, pp. 208-209 (cotypes,* ANSP 8431-2). Schilbeodes marginatus (Baird)—Hubbs and Raney, 1944, pp. 3-26, map 1 and legend (distribution; description; type-locality restricted to Carlisle, Pa.; synonymy).—Burton and Odum, 1945, pp. 187, 191 (Virginia records).— Bailey and Taylor, 1950, p. 31 (two subspecies).—Fowler, 195la, p. 91 (Lancaster Co., Pa., records); 1951b, p. 101 (Chester Co., Pa., records) ; 1952, p. 109 (synonymy; New Jersey records).— Robins and Deubler, 1955, pp. 14, 16 (Tioughnioga R., N.Y.; associations with Lota).—G. A. Moore, 1957, pp. 148, 145. Schilbeodes marginatus marginatus (Baird)—Hubbs and Raney, 1944, pp. 1-24, map 1 (synonymy, in part; distribution; description; locality records; Great Cypress Swamp* [complex]; [not “South Carolina’’]).—Fowler, 1945, pp. 32, 81, 123, 172, fig. 163 (specimen figured, Assanpink Cr.,* Trenton, N.J.; dis- tribution; synonymy, in part; records including Glade Cr.,*Roanoke, Va.; Mill Cr., Old Fort,* N.C.; specimen, Edisto R., Sandy Island Bridge, S.C., described).— Raney and Lachner, 1946a, p. 675 (in associations).—Hubbs and Lagler, 1947 [and 1949], pp. 71, 73, fig. 177 (comparison; range; ecology; Broome Co., N.Y.).—Carpenter and Siegler, 1947, pp. 6, 54, fig. 46 (com- parison; New Hampshire rivers)—Raney, 1950, pp. 170-188 (ecology; synonymy; distribution, James R. basin, Va.).—Fowler, 1950a, p. 100 (Penn- sylvania record); 1950b, p. 104 (Berks Co., Pennsylvania record).—Howden and Mansueti, 1951, p. 95 (Northwest Br., Anacostia R. system, Md.).— Freeman, 1952, p. 36 (South Carolina records); 1953, p. 269 (Savannah R. basin, Aiken* and Barnwell* Counties, 8.C.).—Raney and Massmann, 1954, pp. 426-428 (Lester Manor, and at or above Fall Line, Pamunkey R., Va.).— Harmic, 1954, pp. 25, 45 (Delaware records).—Freeman, 1954, pp. 138, 145, 147, 151 (Savannah drainage, S.C., stations 33* and 40*).—Fischthal, 1956, p. 230 (New York records; list and incidence of parasites).—Anderson and Freeman, 1957, p. 104 (Congaree drainage, South Carolina records* [all reexamined]).—Ross, 1959a, pp. 6-20 (records, Potomac and Shenandoah drainages). Noturus occidentalis Gill [misidentification]—Giinther, 1864, p. 105 (description, in part; young presented by Smithsonian Institution;* ‘intermaxillary teeth” figured).—Jordan, 1877b, p. 372 (of Giinther, probably Noturus marginatus).—Swain and Kalb, 1883, p. 640 (of Giinther, equals Noturus insignis [but not Platte R.]). Noturus flavus Rafinesque [misidentifications].—Cope, 1865, p. 277 (Susquehanna R.).—Uhler and Lugger, 1876, p. 151 (description [in error]; Maryland records).—Jordan, 1877d, p. 99, and 1878a, p. 118 (Potomac R.).—Swain and Kalb, 1883, p. 639 (Potomac and Patapsco).—?J. Nelson, 1890, p. 671 *Material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 87 (New Jersey [but not range, description, or synonymy]).—Truitt, Bean, and Fowler, 1929, p. 37 (Maryland records of Uhler and Lugger are likely Schilbe- odes insignis).—Fowler, 1945, p. 123 (Maryland records of Uhler and Lugger [probably incorrectly] ascribed to Schilbeodes gyrinus). Noturus gyrinus (Mitchill) [misidentifications].—Cope, 1869, p. 237 (Delaware Water Gap,* Pa.); 1881, p. 77 (description; range, in part; trib. to the Dela- ware R. in Pennsylvania).—Bean, 1892, p. 20 (Pennsylvania distribution, in part).—Patrick, 1961, p. 256 (White Clay R., only). Schilbeodes gyrinus (Mitchill) [misidentifications].—Fowler, 1907a, p. 14 (Penn- sylvania records, in part: Delaware Water Gap,* Monroe Co.; Dingmans Ferry,* Pike Co.; Loyalsock Cr.,* near Lopez, Sullivan Co.); 1912b, pp. 47, 54 (Gynn Falls,* near Baltimore, Md.; Big Neshaminy Cr., Etterton,* and ? near Long Pond, Bucks Co., Pa.); 1913, p. 92 (Pennsylvania records relisted) ; 1914a, p. 351 (?Delaware R., Torresdale; Lime Kiln Run,* Lehigh Co., Pa.).—?Fowler, 1915a, p. 208 (Torresdale, Pa.). Pimelodon insignarius Vaillant, 1896a, p. 28 (synonymy; name inscribed on LeSueur’s drawing of Pimelodon livrée) ; 1896b, p. 14, pl. 24 (synonymy; name on LeSueur’s colored drawing; drawing first published).—Jordan, Ever- mann, and Clark, 1930, p. 155 (synonymy; basis of name Pimelodon). Rabida gilberti (Jordan and Evermann) [misidentification].—Fowler, 1935a, p. 19 (Great Cypress Swamp,* 8.C., in part). Schilbeodes exilis (Nelson) [misidentification]—Hubbs and Lagler, 1941, p. 65 (Kanawha R. system). Schilbeodes marginatus atrorus Hubbs and Raney, 1944, pp. 1-25, pl. 1, map 1 (original description; distribution; synonymy and included records; ecology; holotype,* UMMZ 139452, and 8 paratypes,* UMMZ 139453, Wolf Cr., U.S. Hwy. 52, between Bastian and Novis, Bland Co., Va.; figured; other paratypes* and nontypes* listed).—Fowler, 1945, p. 32 (distribution).— Raney and Lachner, 1946b, p. 226 (ecology; Anglin Cr., Patrick Co., Va.;* in association with Thoburnia hamiltoni).—Raney, 1950, p. 187 (N. limit of range in Roanoke or Chowan systems). Schilbeodes insignis atrorus Hubbs and Raney.—Hubbs and Lagler, 1958, p. 91 (range). Noturus insignis atrorus (Hubbs and Raney).—R. D. Ross, 1959e, pp. 8, 24 (key; New R. system).—Ross and Perkins, 1959, pp. 13-33 (records, New R.., Va.).—Ross and Carico, 1963, pp. 9, 16 (records, North Fk. Holston R., Va.; entered from New R. by stream piracy or introduced by human agency). Tyre.—_MNHN 3053 (holotype), Philadelphia, U.S.A., C. A. LeSueur. OTHER MATERIAL STUDIED UNITED STATES: District or Cotumpia: USNM 43605, 85712, 85725. Georaia: CU 17373 (Brushy Cr., 0.9 mi. S. Wrens, Hwy. 1, Jefferson Co.); USNM 162467, 162468 (Reedy Cr., 3.9 mi. NE. Wrens, Jefferson Co.); UMMZ 94576, 94582 (Athens Cr., N. of Talmo); UMMZ (Fletcher nos. 6, 29—Barber Cr., about 2 mi. S. of Athens, Clarke Co., 51, 82). Maryianp: USNM 64426, 66333, 67043, 67073, 70279, 70415-6, 70448, 73287, 73389, 73890, 74776, 75031, 76743, 84052, 85736, 85797, 89012, 103136, 120968, 131769, 131782, 131795, 131814, 157360, 170970 (Deep Creek L., Garrett Co.), 196692 thru 196697; CU 2029, 9249, 13069; UMMZ 66860, 89343. New Jersey: UMMZ 109804, 109820; CU 5333, 5346, 9827. New York: UMMZ 89165, 94468, 109631, 109656, *Material indicated by an asterisk has been re-examined. 88 U.S. NATIONAL MUSEUM BULLETIN 282 114318; UMMZ (Hankinson no. 272); USNM 166441; CU 283, 834, 2952, 3694, 4120, 5758, 7411, 7736, 7873, 8894 (Keuka L.), 9312 (Keuka L.), 11485, 11486 (Catherine Cr., Schuyler Co.), 18719, 18739, 18755, 18781, 18800, 18810, 18982, 19040, 19051, 19102, 19115, 20119, 20139, 20152, 20196, 20233, 20240, 20278, 20382. Norta Carona: Tulsa U 3; USNM 20322, 40443, 67931, 74813, 92529, 162249, 162890, 162934 thru 162936, 190983, 191000, 191015, 191046, 191058, 191072, 191091, 192755, 192762, 192764, 192787 thru 192796, 200476; CU 4015, 8534, 8540, 9369, 9631, 9654, 9734, 9749, 9776, 9797, 9846, 9857, 10069, 11093, 11096, 11263, 11336, 11426, 11431, 11526, 11591, 11664, 11893, 14080, 14107, 15611, 16809, 16847, 16903, 18392, 19075, 19230, 19241, 19310, 19334, 19404, 19438, 19494, 19513, 19528, 19556, 19581, 19763, 37833; CU (Raney no. 1533); UMMZ 94531, 94542, 94656, 126268, 132788, 138481, 139439, 145270, 147569, 147590, 147611, 156664, 165615; UMMZ (White coll.—Morgan Cr., Orange Co.); NCSM 1482 (Watauga R., Hwy. 105, Watauga Co.), 1489 (Laurel Cr., U.S. Hwy. 321, Watauga Co.), 1501 (Watauga R., county road 1200, Watauga Co.), 2356 (Cove Cr., 2.25 mi. SW. Sugar Grove, Watauga Co.). PENNSYL- vVANIA: UMMZ 94482, 94486, 138221 (1 syntype, Noturus marginatus, Carlisle) ; ANSP 8431-2 (2 syntypes, Noturus marginatus, Carlisle), 8519, 48178, 48543; ANSP (Peters Cr., Lancaster Co.; Briar Cr., Columbia Co.); CU 10358, 16380, 16457, 16506, 18820, 18858, 18878, 18913, 19058, 20463; USNM 1571 (9 syn- types, Noturus marg’natus, Carlisle), 67539, 70378, 74847-8, 161756, 161774, 192651 (Mill Cr., 2 mi. W. Elliottsville, Fayette Co.), 196691; MNHN 340 and 341 (2 syntypes, Nolurus marginatus, Carlisle); MCZ 35955 (1 syntype, Noturus marginatus, Carlisle). SoutaH Carotmna: CU 15183, 15380, 15481, 15632, 19068, 19597, 19680, 19736; USNM 23447, 31078, 40600, 162946, 168128, 192735 thru 192754, 192756 thru 192761, 192763, 192765 thru 192786, 192797 thru 192801; UMMZ 94565, 143200, 143203-4. Vrrainra: CU 8322, 8336, 8365, 8423, 9415, 9688, 10055, 10083, 10503, 10520, 11214, 11370, 11456, 11642, 11656, 11724, 11747, 11774, 11885, 11916, 11948, 12229, 13115, 14554, 16837, 16869, 16900, 17002, 19612, 20301, 20316, 20685 (North Fk. Holston R., Hwv. 91, 2.3 mi. SW. Broad- ford, Smyth Co.), 20706, 20721, 20730; USNM 1470, 40194, 44396, 64443, 93285, 100182, 100192, 100212, 100234, 101323, 101345, 102404, 103410, 104077, 104091 (2 paratypes, Schilbeodes marginatus atrorus, Sinking Cr., 3 mi. above mouth, New R. system, Giles Co.), 104096, 104110, 107454 thru 107456, 107722, 125670, 132068, 162001 thru 162006, 162150 thru 162152, 162195 (trib. of Reed Cr., 3.1 mi. N. Max Meadows, Wythe Co.), 162851 thru 162853, 168150, 177279, 190965 (North Fk. Holston R., just above Saltville, Smyth Co.), 201244-5; UMMZ 94522, 95089, 95151, 95190, 95323, 95346, 95401, 95408, 95423, 126243, 135411, 136151, 138523 (5 paratypes, Schilbeodes marginatus atrorus, Reed Cr., Hwy. 21, 1 mi. SW. Wytheville, Wythe Co.), 139452 (holotype, Schilbeodes marginatus atrorus, Wolf Cr., U.S. Hwy. 52, between Bastian and Novis, Bland Co.), 139453 (8 paratypes, Schilbeodes marginatus atrorus, taken with holotype) ; ANSP 8435 (1 syntype?, Noturus marginatus, Sinking Cr. of Kanawha). West VireIniA: UMMZ 136859; CU 4922, 11140 thru 11142, 11181, 13759; USNM 192652 (Big Sandy Cr., Clifton Mills, Preston Co.). Diacnosis.—Noturus insignis is separated from most species of the subgenus Schilbeodes by a combination of characters: 54 to 67, usually 56 or more caudal rays, normally 9 pelvic rays, usually 9 soft pectoral rays, 15 to 21 anal rays, distinct posterior serrae of the pec- toral spine, which may become reduced in size with age, an included lower jaw, typically 11 preoperculomandibular pores, and 2 inter- CATFISH GENUS NOTURUS RAFINESQUE 89 nasal pores. The ventral surface of the head, except the lower lip, and the abdomen, except a narrow band just anterior to the pelvic fins, are usually unpigmented. The vertical fins of many specimens have a black margin. It is distinguished from Noturus exilis, with which it has been asso- ciated, by the longer caudal fin, the included lower jaw, and the greater number of preoperculomandibular and nasal pores; from Noturus nocturnus, an apparent close relative, by the more depressed skull, better developed pectoral spine serrae, and usually 38 or more verte- brae; and from Noturus funebris and N. phaeus by a shorter anal fin and a poorly pigmented ventral surface. Description.—Other counts and measurements are given in tables 17 to 26. Body elongate; caudal peduncle deep, both anteriorly and posteriorly, deeper than head depth; head depressed; lower jaw in- cluded; premaxillary tooth band usually obtusely angulate at corners; dorsal spine moderate in length, stiff; pectoral spine long, slightly curved, without anterior serrae, but with numerous distinct posterior serrae, which may become irregularly fused in old specimens; humeral process about equal in length to the diameter of the pectoral spine; adipose fin broadly united to the moderately high procurrent caudal fin; caudal fin truncate or rounded behind; eye moderate, 1.9 to 2.6 times in snout; gill rakers six to ten on first arch; size fairly large. Specimens over 100 mm. are frequently obtained; one of the largest is a Virginia specimen 126 mm. in standard length. - The caudal fin has (extremes in parentheses): (21) 23 to 26 (28), mean 24.3, upper simple rays; (15) 17 to 20 (22), mean 18.4, branched rays, of which there are nearly always 7 in the upper-half of the fin and most frequently 10 to 13 in the lower-half; (13) 16 to 20 (23), mean 18.3, lower simple rays. The soft dorsal rays are five (in 3), six (119), and seven (2). The posterior edge of the pectoral spine is serrated, with up to nine serrae. In cleared and stained specimens for North Carolina, Virginia, and New York: vertebrae anterior to the anal fin origin 12 (11), 13 (23), 14 (2). In 82 of the 92 sides of the pectoral arches examined, the ossified radials are tightly fused; the other sides show partial fusion of the elements, perhaps because relatively small specimens were used. The hypurals are separate in 39 specimens; hypurals 2 and 3 fused (in 4), hypurals 4-5 fused (1), and hypurals 6-7 fused (3). General color yellowish to slate-gray, lighter on ventral surfaces. Top and sides of body and head uniformly darkly pigmented; no blotches or light areas present; barbels usually with some chromato- phores, those above dark, the barbels below light colored; chin in front of the mental barbels pigmented, and a narrow bridge of pig- ment crosses abdomen in front of pelvic fins; otherwise, abdomen and 90 U.S. NATIONAL MUSEUM BULLETIN 282 lower surface of head usually immaculate, but sometimes lightly pigmented in very large specimens; all fins lightly pigmented, especi- ally near their bases, but pelvic fins relatively clear; margins of pec- toral, anal, dorsal, and caudal fins frequently with a dark band or with concentrations of pigment that is usually darker than the basal areas of the fins, their extreme edges clear; adipose fin dusky near base, clear distally. VARIATION.—Certain features of the variation in this species have been discussed by Hubbs and Raney (1944). They indicate that im- portant average differences in the lengths of the dorsal and pectoral spines exist between populations and describe color variations. The data given by Hubbs and Raney (1944, pp. 8, 9, 10, 17) for several proportional measurements of Roanoke and Kanawha specimens are virtually overlapped by the data from other areas. N. insignis is, therefore, not unlike other species in the great variation in the length of the spines. As discussed under Noturus gyrinus, it appears that some nongenetic factors are operative in determining the develop- ment of hard parts, and that the relative lengths of the dorsal and pectoral spines are dependent upon these factors. The black-bordered fins of Kanawha and Roanoke River specimens are distinctive, but there is a gradient of this character southward into North Carolina and scattered samples from New York, Penn- sylvania, and Virginia have similar appearing specimens. The eco- logical distribution of black fin borders is apparently like that in Noturus exilis, which in cool, clear streams develops black-bordered fins, but in warm turbid streams lacks any trace of the dark borders. One small specimen of insignis from Georgia has clear white fin borders. No important variations in meristic characters are apparent. The data (number of tabulations, range in parentheses, and mean) are summarized from the following areas: (a) Cape Fear River to Oconee River; (b) Neuse and Tar Rivers; (c) Roanoke, Chowan, and Kanawha Rivers; (d) James River to Potomac River; (e) the area north of the Potomac River to the Susquehanna River; (f) Holston River; (g) Watauga River; and (h) total. Soft pectoral rays: (a) 46 (8-10) 9.07; (b) 40 (7-10) 8.90; (c) 62 (8-10) 9.02; (d) 12 (9) 9.00; (e) 95 (8-9) 8.83; (f) omitted; (g) 20 (9-10) 9.15; (h) 275 (7-10) 8.95. Pelvic rays: (a) 46 (9-10) 9.04; (b) 40 (9-10) 9.30; (c) 62 (9-10) 9.06; (d) 12 (9-10) 9.17; (e) 96 (8-10) 9.02; (f) omitted; (g) 20 (9-10) 9.05; (h) 276 (8-10) 9.08. Anal rays: (a) 23 (15-20) 18.00; (b) 22 (17-21) 18.45; (c) 31 (16-20) 18.48; (d) 6 (17-19) 18.00; (e) 49 (15-19) 17.29; (f) omitted; (g) 10 (17-21) 18.80; (h) 141 (15-21) 17.99. CATFISH GENUS NOTURUS RAFINESQUE 91 Lower-half caudal rays: (a) 24 (27-33) 29.50; (b) 22 (28-33) 30.45; (c) 32 (26-32) 29.78; (d) 6 (30-31) 30.67; (e) 49 (28-33) 29.67; (f) 3 (28-30) 29.00; (g) 9 (26-31) 28.56; (h) 145 (26-33) 29.74. Upper-half caudal rays: (a) 24 (29-33) 31.08; (b) 22 (29-34) 31.95; (c) 32 (29-33) 30.91; (d) 6 (33-34) 33.67; (e) 50 (29-35) 31.62; (f) 3 (29-30) 29.33; (g) 10 (27-32) 29.90; (h) 147 (27-35) 31.35. Total caudal rays: (a) 24 (57-65) 60.58; (b) 22 (57-67) 62.41; (c) 32 (56-65) 60.69; (d) 6 (64-65) 64.33; (e) 49 (58-66) 61.31; (f) 3 (57-60) 58.33; (g) 9 (54-62) 58.33; (h) 145 (54-67) 61.10. Vertebrae: In the counts obtained, the mode for each river system is 39, except the Merrimack. The data are summarized for the fol- lowing river systems: Savannah 4 (38-39) 38.75; Wateree 24 (38-40) 38.88; Neuse 69 (37-42) 38.83; Tar 17 (38-41) 39.29; Roanoke 26 (37-40) 39.19; Holston 16 (88-40) 39.25; New 3 (38-39) 38.67; James 52 (87-40) 38.98; Monongahela 8 (38-39) 38.88; Potomac 34 (38-41) 38.97; Susquehanna 36 (37-40) 38.53; Merrimack 1 (38) 38.00; total 290 (37-42) 38.92. Typr.—A single specimen, MNHN 3053, is the basis of the des- cription of Pimelodon livrée, hence the holotype of Pimelodus insigne, of Pimelodus lemniscatus, and of Pimelodon insignarius. Doctor Leon Bertin and Dr. Victor G. Springer (personal communications) have kindly provided certain critical information about this specimen as follows: The standard length is 74.8 mm. and the total length 88.4 mm.; the caudal fin has approximately 25+13+9+16=63 rays; there are 7 soft dorsal rays and 20 anal rays; on each side there are 2 internasal pores, 9 pelvic rays, and 9 soft pectoral rays. ‘‘Philadelphie (Etats-Unis),” is indicated as the type-locality by Bertin and Estéve (1950, p. 25), but Bertin states in his letter that there could have been a transposition of specimens: “‘Provenance. Une erreur a pu se produire a lorigine entre deux spécimens: un Pimelodus lemniscatus (holotype) et un Pimelodus nebulosus dont les provenances sont Savannah et Philadelphie. Il est possible que Savannah s’applique au P. lemniscatus et Philadelphie au P. nebulosus.’’ However, the black fin margins shown in Vaillant’s figure (1896b) of Pimelodon insignarius are not characteristic of southern specimens. A subsequent check on this confusion suggests that data from MNHN 3052 which is listed as Pimelodus nebulosus from Savannah was inadvertently copied on a label for MNHN 3053, and that Philadelphia is probably the correct locality. Ranee.—WNoturus insignis (map 5) has as an apparent native range part of the Lake Ontario drainage, most of the Atlantic coastal streams from New York to Georgia, where it is most frequently found at or above the Fall Line, and the upper New or Kanawha River system. It occurs in the Merrimack River of New Hampshire 298-943 O—69——7 92 U.S. NATIONAL MUSEUM BULLETIN 282 SCALE OF MILES Map 5.—Distribution of Noturus insignis (Richardson). Circles indicate the localities from which T have examined specimens. The dashed line outlines the known limit of distri- bution and surrounds localities for which there are unconfirmed literature records. N. insignis is believed to be a native inhabitant of tributaries to the Atlantic Ocean from Georgia to New York and the upper Kanawha River system. It was probably intro- duced by man into New England and has crossed into the Lake Ontario drainage, the upper Tennessee River basin, and other tributaries of the Ohio River in Maryland, Pennsylvania, and West Virginia. Eee CATFISH GENUS NOTURUS RAFINESQUE 93 where it has presumably been introduced. In New York the records are from Keuka Lake, Catherine Creek, tributary to Seneca Lake, and a few other small tributaries in the Lake Ontario drainage, from the Mohawk and Hudson River systems, and from the Delaware and Susquehanna basins. In the New River, it is confined to the area above the falls in West Virginia, Virginia, and North Carolina. In Georgia it occurs in the Oconee and Savannah River systems. In recent years specimens of Noturus insignis have been collected from other drainages in the upper Ohio River system and from the Tennessee system. The collections from the Tennessee system indicate that insignis is well established there, where it is known from several localities in the Watauga drainage in North Carolina and from the North Fork of the Holston River, Virginia. I and several other collectors have obtained specimens from the latter stream and found it abundant, at least locally. N. insignis has been collected from three localities in the Monongahela and Youghiogheny systems in West Virginia, Maryland, and Pennsylvania. Although few specimens have been collected, the geographic dispersal suggests that msignis is established in both streams. The historical evidence points to the absence of insignis from the upper Ohio and Tennessee systems until recently, as many earlier collectors did not obtain specimens. Establishment could have been by any of at least three methods, but the first seems most likely: (a) Frequent references in the fishery literature indicate that madtoms are often used as bait for other fishes. They are very hardy and able to survive injury and extreme conditions. Dumping of a few unwanted bait specimens could easily start a new population in a favorable habitat. (b) Accidental dispersal when stocking game fishes. Noturus gyrinus has undoubtedly been introduced into several areas by this method. However, insignis is less likely to be an inhabitant of fish hatcheries and it is expected that it would rarely be captured and dis- tributed along with game fishes. (c) Changes in drainage pattern may account for dispersal of N. insignis into the Holston drainage but they do not appear to be responsible for several of the introductions else- where. An undoubted introduction by man was into Clark Lake, Gogebic County, Michigan (UMMZ 186551). Noturus leptacanthus, N. exilis, N. flavus, and N. miurus have been reported under the name Noturus insignis or one of its synonyms. Erymotocy.—The name insignis (Latin) means remarkable or extraordinary. The probable intention was to emphasize the color and the long adipose fin, features which were at one time considered unique. NoMENCLATURE.—The recent resurrection of Noturus marginatus Baird for this species and the transfer of Pimelodus insigne Richardson 94 U.S. NATIONAL MUSEUM BULLETIN 282: to the form herein called Noturus exilis (Hubbs and Raney, 1944) are invalid. Noturus insignis was first described by LeSueur from material (probably a single specimen) that became the basis of five different names. Two of these, both accompanying LeSueur’s descriptions, were vernacular. Richardson (1836, p. 132) named Pimelodus insigne on the basis of one of LeSueur’s descriptions and vernacular names. Next, Valenciennes (Cuvier and Valenciennes, 1840, p. 144) described LeSueur’s specimen anew, naming it Pimelodus lemniscatus. Finally Vaillant published a synonymy (1896a, b) and LeSueur’s colored drawing (1896b) under the drawing’s label Pimelodon insignarius. None of the descriptions indicated a type-locality. Characters of the single specimen that serves as the type of these names are given above. LeSueur (1819, p. 155) indicated that the first dorsal, anal, caudal, and pectoral fins are bordered with a band of black, also shown in LeSueur’s drawing, published by Vaillant (1896b). However, the 50 caudal rays noted by LeSueur must be interpreted as an estimation. Valenciennes (Cuvier and Valenciennes, 1840, p. 144) described the upper jaw as longer than the lower and the caudal fin with more than 60 rays, including the small ones above and below, in agreement with counts obtained by Bertin from the type. The lo- cality indicated, the characters given, and the original descriptions when considered together leave no doubt of identity with the madtom here called Noturus insignis; the combination of characteristics (2 internasal pores, the high number of caudal rays, the included lower jaw, the color, and a locality in the eastern United States) refers to no other species. Noturus marginatus was described by Baird (in Cope, 1869, pp. 237, 241). Localities listed are “Sinking Creek of the Kanawa,” “head of James River,” and “the Susquehanna.” Jordan and Evermann (1896a, p. 147) listed USNM 1571, a collection containing at least nine speci- mens, as type. Hubbs and Raney (1944, p. 15) accepted Jordan and Evermann as having restricted the type and type-locality. No lecto- type of Noturus marginatus has yet been selected. Baird’s description was based, at least in part, upon specimens from the Susquehanna River at Carlisle, Pennsylvania (USNM 1571). It is uncertain if Baird based his description also on the Virginia specimens or if these were only listed by Cope after comparison with the Academy of Natural Science specimens provided by Baird. I have examined several of the syntypes of marginatus and find that they are all referable to Noturus insignis. Schilbeodes marginatus atrorus, described by Hubbs and Raney (1944) from Wolf Creek, Bland County, Virginia, differs little from other material except in the intensity of the black on the fin borders. CATFISH GENUS NOTURUS RAFINESQUE 95 Pimelodus insignis Schomburgk (1841, p. 180 and plate VI in the Natural History of the Fishes of [British] Guiana, vol. 1) is a primary homonym of Pimelodus insigne Richardson (1836, p. 132). EcoLOGICAL CONSIDERATIONS.—As noted by examination of the distribution map (map 5) of this species, the records ‘are chiefly from high-gradient streams above the Fall Line. Raney and Massman (1954) and Greeley (1928, 1935, 1936, and 1937) have discussed the ecology. N. insignis lives in clear or normally clear water in moderate to swift current, chiefly about riffles of rubble, boulders, or coarse gravel. Noturus insignis has been taken in collections containing Noturus gyrinus, N. leptacanthus, N. gilberti, and N. furvosus. The FUNEBRIS Group The funebris group, subgenus Schilbeodes, includes two allopatric species, Noturus funebris and Noturus phaeus. They are found in the lower Mississippi River system and other Gulf of Mexico tributaries from the Red River and Bayou Teche, Louisiana, to Econfina Creek in western Florida. Neither species is known from any of the streams between the Mississippi and Pearl River systems, in the Florida Parishes of Louisiana and southern Mississippi. It is in this region that intermediates between the two species would be expected, but exten- sive collections indicate their absence. Both are typical of permanent springs and small, clear streams; they are commonly found beneath submerged vegetation at the outlet and downstream from springs. In contrast to other species of Noturus the anal fin is long with 18 to 27 rays (table 17) and the dorsal spine is slender and somewhat flexible, becoming stiff with age. The caudal fin has an intermediate number of rays. Like Noturus insignis the hypurals are modally seven (1-3 +4-7) in both species. X-rays and skeletonized specimens indicate a rather prominent mode of eleven vertebrae anterior to the origin of the anal fin, suggesting that the anal fin is more anterior in position than in other species except Noturus gyrinus and Noturus leptacanthus. Unlike other Noturus, the flange or thin bone connecting the anterior and posterior rami of the transverse process of the fourth vertebra in the vertebral complex is very poorly developed or rudimentary. The lateral boundary of this ledge of bone is very deeply concave in contrast to that in other species where it is very shallowly concave to moderately convex or with a nearly straight margin parallel to the vertebral axis. The two species are almost uniformly darkly colored, with occasional narrow darker bands near the margins of the dorsal, anal, and caudal fins. Especially characteristic of the funebris group are the dark, 96 U.S. NATIONAL MUSEUM BULLETIN 282) usually discrete, chromatophores that are uniformly distributed over the lower surface of the head and abdomen at all sizes, except large old adults. These freckles may become diffuse or lost near the midline of these surfaces in older specimens. Noturus funebris Gilbert and Swain BLACK MADTOM PuaTEs 3 (FIG. 7), 7 (FIG. 1); Map 6 Noturus funebris Gilbert and Swain.—Jordan, 1885, p. 802 (nomen nudum; “Northern Alabama’’).—Gilbert and Swain, in Gilbert, 1891, pp. 158, 157 (original description; types* from: ‘‘in a spring run tributary to North River near Tuscaloosa, Alabama’’).—Bdéhlke, 1953, p. 43 (holotype, USNM 36696* and paratype, SU 3952*).—Bailey, Winn, and Smith, 1954, pp. 131-155 (distribution; Escambia R. basin records*).—Eddy, 1957, p. 153, fig. 385 (key; range).—Taylor, 1957, p. 192.—Briggs, 1958, p. 260 (Florida distribu- tion).—Cook, 1959, pp. 34, 135, 142, fig. 25F (description; Mississippi records).—?Suttkus, 1961, p. 63, fig. 5 (comparison; skull illustrated).— Suttkus and Taylor, 1965, p. 177 (associations)—Thomerson, 1966, p. 397 (length frequency distribution; sex ratio; egg counts; associations, Metts Cr., Fla.). Schilbeodes funebris (Gilbert and Swain).—Jordan and Evermann, 1896a, pp. 145, 147 (description; Tuscaloosa,* Ala.); 1896b, p. 234.—Reed, 1907, p. 555 (not examined for poison glands).—Schrenkeisen, 1938, p. 167 (com- piled).—Carr and Goin, 1955, pp. 21, 64 (description; Florida distribution) .— G. A. Moore, 1957, pp. 143-144 (description). Rabida funebris (Gilbert and Swain).—Jordan, Evermann, and Clark, 1930, p. 156. TYPE-SPECIMENS.—USNM 36696 (lectotype), USNM 161732 (2 paralectotypes), and SU 3952 (1 paralectotype), from a spring run, tributary to North River, near Tuscaloosa, Alabama, summer of 1884, C. H. Gilbert and Joseph Swain. OTHER MATERIAL STUDIED UNITED STATES: Atasama: Escambia R. system: TU 3146, 14180, 16520; CU 11826; Tulsa U 8. Perdido R. system: TU 14184. Fish R. system: TU 3156, 14389. Tallapoosa R. system: USNM 201246; CU 14045; UMMZ 111195-6, 168676. Cahaba R. system: TU 14198. Black Warrior R. system: TU 14141; CU 19271. Tombigbee R. system: TU 24549; CU 16162. Other tribs. Mobile Bay: TU 3091; CU 12636; UMMZ 155463. Fuioripa: Econfina Cr. drainage: USNM 201242 (Cr., about 1000 yards W. U.S. Hwy. 231 on Jackson- Bay Co. line); DBUF (Cr., 0.5 mi. W. Econfina Cr. bridge, W. Bennett, Bay Co.). Choctawhatchee R. system: TU 1084, 1102, 20574, 21330. Alaqua Cr. drainage: DBUF (at Portland, Walton Co.). Yellow R. system: TU 23830, 24010, 24089, 24107, 24456, 24661, 24774. Blackwater R. sytsem: TU 20528; CU 16702; UMMZ 155498, 163521, 166192, 166215; DBUF 1457. Escambia R. system: UMMZ 165131. Escambia Co., unknown drainage: CU 13712. Perdido R. system: TU 1778; USNM 143049. Lovis1ana: Pearl R. system: TU 95, 739, 1666, 3598, 5756, 6238, 7352, 7590, 7826, 8245, 11477, 14266, 14394, 14548, 15549, 23093, 23383; USNM 152985. Mussissippi: Pascagoula Bay drainage: *Material designated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 97 TU 1127, 15442; UMMZ 155443, 155449; UM MZ (Walker no. 39-49). Biloxi Bay drainage: UMMZ (Walker nos. 39-34, 39-37). Bay Saint Louis drainage: TU 18599; UMMZ 163697; UMMZ (Walker no. 39-33). Pearl R. system: TU 72, 111, 826, 1730, 1795, 3879, 4870, 5814, 6253, 7223, 7276, 7301, 14162, 15099, 15202, 16141, 16729, 16743, 17578, 17658, 17949, 23433, 23610, 23716; CU 11860, 16616; AF 1718 (Old Brook, trib. Bogue Chitto R., Lincoln Co.) ; UMMZ 155371, 155382. Dracenosis.—Noturus funebris, subgenus Schilbeodes, has 20 to 27, usually 21 or more anal rays; the anal fin separated from the lower procurrent caudal rays by a small notch; the dorsal spine rather slender and usually flexible (except large specimens); the ventral surface, including both the head and abdomen, profusely sprinkled with large discrete chromatophores; 8 to 11, usually 9 soft pectoral rays; 9 pelvic rays; 11 preoperculomandibular pores; 2 internasal pores; 50 to 58 caudal rays; and an included lower jaw. The body and fins are nearly unicolor, blackish or steel blue. In contrast to Noturus phaeus the pectoral spine serrae are absent or vestigial, irregular, and never strongly developed. DerscripTion.—Other counts and measurements are given in tables 17 to 26. Head rounded above; lower jaw included; body heavy and elongate, not much deeper forward than behind; anal base long; dorsal spine slender, relatively flexible, only stiff in large specimens; dorsal fin usually slightly pointed, the first or second ray longest; adipose fin low, closely united to the procurrent caudal fin; pectoral spine short, stout, straight, without prominent uniformly developed posterior serrae; posterior edge of spine usually only roughened; a few distinct but irregularly developed sharp serrae may be present in young specimens; these become reduced to short blunt knobs or fuse into an irregular roughened mass with age; end of caudal fin rounded to truncate; posterior corners of premaxillary tooth band slightly rounded; eye small, 2.0 to 2.5 times in snout; posterior process of cleithrum short, definite, about equal in length to the width of the pectoral spine. Gill rakers on the first arch five to eight. There are (extremes in parentheses) : (18) 19 to 21 (22) upper simple caudal rays; (15) 16 to 18 (20) branched caudal rays, of which there are usually 7 in the upper half of the fin and 9 to 11 in the lower half; and (13) 15 to 18 (19) lower simple caudal rays. Of 83 specimens counted, all have 6 soft dorsal rays. The stained and cleared specimens have relatively light, poorly ossified bones; particularly noticeable externally is the slender, flexible dorsal spine; vertebrae anterior to anal fin origin 10 (in 2) or 11 (1); pectoral radials fused on six sides; hypurals modally 7, all unfused in 27 specimens counted, mostly from x-rays, or fused as follows, num- 98 U.S. NATIONAL MUSEUM BULLETIN 282 bers 2-3 (in 8), 4-5 (1), both 2-3 and 4-5 (2), and both 2-3 and 6-7 (1). This species reaches at least 119 mm. in standard length. The general background color is dark brownish black or blue-black, usually a gun metal blue color. The upper surface, barbels, fins, and side are darkly pigmented. The ventral surface is somewhat lighter, but sprinkled with large chromatophores. These may become diffused with age, so that the midline of the abdomen appears to be immaculate. The fins are all heavily pigmented, but sometimes edged with white or cream color. A band of black pigment often crosses the outer ends of the rays of both the dorsal and anal fins and may outline the posterior margin of the caudal fin. Typr.—The lectotype of Noturus funebris (USNM 36696) herein selected shows evidence of having been dried at one time, but is the best of the three specimens that I found in USNM 36696. It is a female, 77.3 mm. in standard length or 95.5 mm. in total length. There are 6+17=23 anal rays, 21+7 (+28)=56 caudal rays, 6 soft dorsal rays, and 28 rays in both the upper half and the lower half of the caudal fin. On each side there are 9 pelvic rays, 10 soft pectoral rays, 6 gill rakers on the first arch, 2 internasal pores, 11 preoperculo- mandibular pores, and at least 10 branchiostegal rays. The posterior edges of the pectoral spines have a few irregular, very short, blunt, knob-like serrae or roughenings. The head length is stepped into the standard length 3.85 times. The barbels are moderately long; the dorsal spine is short but slender. VariatTion.—WN. funebris shows rather strong west to east gradients in number of anal rays and vertebrae (tables 6 and 8). On the basis of _ these structures two populations may be recognizable. Specimens from west of the Alabama system seem to have modal numbers of 35 or 36 vertebrae and usually 21 to 23 anal rays. Specimens from east of the Alabama system, chiefly western Florida, appear to have a modal number of 37 vertebrae and 23 to 27 anal rays. The pore and paired fin ray counts show very strong modes with only occasional random variations within a sample. The variation in pectoral rays is given in table 8. The counts of pelvic rays, based on the same specimens, vary as follows: Pearl River 7 to 10, mean 8.94; Wolf River 9 to 10, mean 9.05; Blackwater River 9 to 10, mean 9.22; Yellow River 9 to 10, mean 9.20; Econfina Creek 9 to 10, mean 9.33; no variation from 9 rays is indicated in the counts from all other river systems. Based on the relatively few specimens counted, no significant geographic variation in caudal rays is evident. Distrisution.—Noturus funebris (map 6), a small stream or creek species, is confined to eastern Gulf of Mexico tributaries. It ranges from the Pearl River system of Mississippi and Louisiana eastward CATFISH GENUS NOTURUS RAFINESQUE 99 to Econfina Creek, Florida. Although it lives well upstream in the Alabama and Tombigbee River systems, the centers of abundance appear to be in the lower portions of the stream systems. ErymoLocy.—The name funebris (Latin) means of or belonging to a funeral; the allusion is in reference to the dark color. RELATIONSHIP.—Noturus funebris and Noturus phaeus are recog- nized as forming the funebris species group because of their many similarities. Intimate relationship with other species of Noturus is not apparent. Noturus funebris and Noturus nocturnus are strikingly similar superficially, especially externally. Both species are similarly dark colored and lack pectoral spine serrae. The lower surfaces of nocturnus gradually darken with age and increase in size so that nocturnus is not readily distinguished from funebris except in number of anal rays and in the relative slenderness and flexibility of the dorsal spine. Noturus nocturnus, with its apparent relative Noturus insignis, thus may be moderately closely related to the funebris group. Ecotogy.—Most records are from permanent springs, creeks, and small to moderate size streams. The species is often found under vegetation in moderate to fast clear water over a small gravel or coarse sand bottom. It has been taken in collections with N. leptacanthus, N. gyrinus, N. munitus, and probably with N. miurus. ReEMARKS.—Thomerson (1966) listed other fish associates and described aspects of the biology of a population of funebris from west- ern Florida, based on formalin preserved specimens. He concluded that females did not reach maturity until approximately 100 mm. in standard length; three females, 104 to 119 mm. long, contained 114 to 192 ova. Because of long storage in formalin he was unable to determine the age of the specimens but concluded from a length frequency distribution that the sample, collected March 8, 1961, contained several year classes. The larger size specimens were poorly represented. I question his classes I-III, specimens 25 to 35 mm. in standard length, those 45 to 50 mm., and those 55 to 65 mm. especially. The growth rate of his first three classes is materially shorter than that obtained by Clugston and Cooper (1960) for Noturus insignis. They found an average annual increase of about 45 mm. total length in Pennsylvania. The season of growth is undoubtedly longer in Florida than in Pennsylvania, and the season of reproduction could likewise be prolonged. Thus, it seems logical to predict that Thomerson’s classes I and II are actually year class I, with his year class II resulting from reproduction earlier in the year and year class I from late season broods. It is also possible that his year class III may actually be year class I. If this is true, his remaining year classes would require adjust- ment. As in other moderate size Noturus, funebris probably lives no longer than three or four years. 100 U.S. NATIONAL MUSEUM BULLETIN 282 @ NOTURUS FUNEBRIS O Norturus PHAEUS Taw Tee o aA ES 5 7 j SCALE OF MILES Map 6.—Distribution of Noturus funebris Gilbert and Swain and Noturus phaeus, new species. Symbols represent localities from which specimens have been examined. Noturus phaeus, new species BROWN MADTOM PuaATEs 3 (FIG. 8), 7 (FIG. 2); Map 6 Noturus funebris Gilbert and Swain [misidentifications]—Hancock and Sublette, 1958, p. 45 (Little Bayou Pierre,* La.).—Norden, 1965, p. 102 (records, Little R. system, La.).—?Raney and Suttkus, 1966, p. 102 (associations with Etheo- stoma rubrum, Bayou Pierre, Miss.). Typres.—USNM 202143 (holotype) and USNM 193471 (10 paratop- otypes), collected from the North Fork of the Obion River, at state highway 69, Henry County, Tennessee, July 27, 1954, by C. E. Ruhr and party. *Specimens re-examined. CATFISH GENUS NOTURUS RAFINESQUE 101 OTHER PARATYPES.—The number of specimens, in parentheses, fol- lows the museum number. Kentucky: UL 5143 (1), Terrapin Cr., 3 mi. SW. Bell City, Graves Co., June 30, 1954, W. M. Clay and E. W. Distler. Mississipp1: UMMZ 161058 (66), Mimosa Spa, 1.5 mi. N. of Waterford, or 8 mi. 8. of Holly Springs, Marshall Co., April 2, 1949, W. R. Taylor, Norman Benson, and H. W. Harry. UMMZ 161066 (4), stream, 0.6 mi. S. of Waterford, Marshall Co., April 2, 1949, Taylor, Benson, and Harry. TU 3037 (5), trib. to Yocona R., 14.7 mi. E. Oxford, Lafayette Co., May 23, 1952, R. D. Suttkus. TU 3134 (4), trib. to Yocona R., Hwy. 6, 8 mi. E. Oxford, Lafayette Co., May 24, 1952, Suttkus. TU 3668 (3), trib. to Yocona R., old Hwy. 6, 7 mi. E. Oxford, Lafayette Co., May 24, 1952, Suttkus. TU 14014 (1), Yocona R., Hwy. 334, 7.9 mi. SE. Oxford, Lafayette Co., July 12, 1956, Penn and Black. TU 14024 (1), stream, U.S. Hwy. 45, 3 mi. N. Biggersville, Alcorn Co., July 12, 1956, Penn and Black. TU 14036 (2), stream, Hwy. 7, 0.3 mi. SW. Michigan City, Benton Co., July 11, 1956, Penn and Black. USNM 175389 (1), Shones L., E. of Holly Springs, Benton Co., June 9, 1952, Y. J. McGaha. USNM 175390 (2), Little Spring Cr., Hwy. 7, 2 mi. N. Little Tallahatchie R., Marshall Co., June 16, 1952, MecGaha. USNM 175391 (2), Snow Cr., 5 mi. W. Ashland, Benton Co., May 12, 1951, William L. Hudspeth. TENNESSEE: USNM 190745 (2), Moss Cr., on Rose Creek road, 1.3 mi. N. of Hwy. 57 or about 2.5 mi. ENE. of Pocahontas, McNairy Co., August 31, 1959, W. R. Taylor and Robert Kanazawa. USNM 190777 (10), Spring Cr., Hwy. 125, 1.2 mi. S. of Bolivar, Hardeman Co., September 1, 1959, Taylor and Kanazawa. USNM 193465 (3), Moss Cr., about 4 mi. W. of Selmer, McNairy Co., August 18, 1954, C. E. Ruhr. USNM 193466 (1), Indian Cr., 3 mi. due S. of Grand Junction, Hardeman Co., August 6, 1954, Ruhr. USNM 193467 (13), Deer Cr., at Huron, Henderson Co., July 21, 1954, Ruhr. USNM 193468 (8), Loosahatchie R., Hwy. 76, due N. of Somerville, Fayette Co., August 31, 1954, Ruhr. USNM 193469 (1), Loosahatchie R., 5 mi. due N. of Oak- land, Fayette Co., August 31, 1954, Ruhr. USNM 193473 (4), Spring Cr., about 4 mi. SSW. Bolivar, Hardeman Co., August 4, 1954, Ruhr. USNM 197401 (2), Middle Fk. Obion R., Hwy. 6, Gleason Crossing, Weakley Co., September 16, 1954, C. E. Farrell. KU 8958 (1), Middle Fk. Obion R., Hwy. 22, 4.3 mi. SE. intersection with Hwy. 54, Weakley Co., September 7, 1964, F. B. Cross and party. UMMZ 168521 (1), trib. to South Fork Forked Deer R., 2.6 mi. SE. Pin- son, Chester Co., August 31, 1954, R. M. and D. M. Bailey; UMMZ 177681 (4), same locality, May 20, 1956, R. M. Bailey and W. L. Brudon. Vanderbilt U (no number) (4), North Fork Obion R., Campground Levee Crossing, Weakley Co., September 15, 1954, Farrell. Vanderbilt U (no number) (3), Big Black Cr., Madi- son Co., September 16, 1954, Farrell. OTHER MATERIAL STUDIED UNITED STATES: Lovis1ana: UMMZ 113740 (3 specimens), trib. to Little R., 0.25 mi. S. of Dry Prong, Grant Parish. UMMZ (Delavan and Creaser no. 31-57) (1), stream, at Jena, La Salle Parish. UMMZ (Delavan and Creaser no. 31-54) (4), trib. to Little R., Pollock, Grant Parish. UMMZ (Taylor no. La 55-24) (1), Choudrant Cr., sec. 16, T. 18 N., R. 1 E., Ouachita Parish. UMMZ (Taylor no. La 55-37) (5), stream, sec. 16-17, T. 16 N., R. 3 E., Ouachita Parish. USNM 172639 (20) and USNM 172981 (2), Brushy Cr., at Hwy. 7, 3.5 mi. S. Sibley, sec. 9, T. 17 N., R. 9 W., Webster Parish. USNM 172662 (6), Hairs Cr., just E. of Jena, sec. 7, T. 8 N., R. 4 E., La Salle Parish. USNM 172776 (4), Cheniere Cr., S. Cadeville, sec. 35, T. 17 N., R. 1 E., Ouachita Parish. USNM 102 U.S. NATIONAL MUSEUM BULLETIN 282 172874 (2), Saline R., due E. Bienville, sec. 23, T. 16 N., R. 6 W., Bienville Parish. USNM 172895 (1), creek, sec. 27, T. 22 N., R. 1 E., Union Parish. USNM 172905 (6) and USNM 172944 (5), Meridian Cr., 1 mi. E. Conway, sec. 17, T. 22 N., R. 1 E., Union Parish. USNM 172915 (7), Clarks Cr., sec. 24, T. 17 N., R. 1 E., Ouachita Parish. TU 911 (153) and TU 6260 (159), Big Cr., at Pollock, Grant Parish. TU 1223 (1), trib., Red R., U.S. Hwy. 71, 10 mi. SE. Colfax, Grant Parish. TU 1348 (1), trib., Spring Cr., Hwy. 85, 5.1 mi. E. Hineston, Rapides Parish. TU 1361 (4), stream, 2 mi. N. Minden, Webster Parish. TU 2058 (17), Rocky Br., trib. to Spring Cr., Hwy. 85, 9.1 mi. E. Hineston, Rapides Parish. TU 3457 (68), trib., Bayou Cocodrie, Hwy. 85, Rapides Parish. TU 5234 (3), Corney Br. Beaver Cr., Hwy. 311, 4.4 mi. SW. Wilson, East Feliciana Pa- rish. TU 5774 (1), Indian Cr., 3.1 mi. S. Woodworth, Rapides Parish. TU 5868 (6), Br. of Spring Cr., Hwy. 85, Melder, Rapides Parish. TU 13737 (1), Little Bayou Pierre, Mora Road, 8.5 mi. E. Kisatchie, Natchitoches Parish. TU 14173 (6), trib. to Bayou Teche, Hwy. 85, E. of Melder, Rapides Parish. TU 14330 (17), stream, Hwy. 8, 0.5 mi. SW. Manifest, Catahoula Parish. MussissipPi: Tulsa U (uncataloged) (1), Brushy Cr., Homochitto, Amite Co. TU 2922 (6), TU 15121 (5), TU 19775 (1), TU 19850 (5), and TU 23964 (1), trib., Homo- chitto R., U.S. Hwy. 84, vicinity of Lucien, Franklin Co. TU 7203 (1), trib., McCall Cr., U.S. Hwy. 84, 13.2 mi. E. Bude, Franklin Co. TU 23592 (2), Homo- chitto R., Hwy. 550, 4.9 mi. E. Union Church, Lincoln Co. Diaenosis.—Noturus phaeus, subgenus Schilbeodes, has 18 to 25, usually 20 to 22 anal rays; the anal and caudal fins only slightly separated; the dorsal spine slender and usually flexible, but stiff in large specimens; the lower surfaces of the head and abdomen pro- fusely sprinkled with large chromatophores; 7 to 10, usually 8 or 9 soft pectoral rays; typically 9 pelvic rays, 11 preoperculomandibular pores, and 2 internasal pores; 50 to 59 caudal rays; and the lower jaw is included. The body and fins are nearly unicolor, typically medium to dark brown. N. phaeus differs from N. funebris, its allopat- ric relative, in the constant presence of distinct, well-developed posterior serrae on the pectoral spine and averages fewer pectoral and anal rays over much of its range. DeEscriptTion.—Other counts and measurements are given in tables 17 to 26. Head rounded above; lower jaw included; body heavy and moderately elongate, not much deeper anteriorly than posteriorly; dorsal spine slender, relatively flexible especially in young; dorsal fin pointed, the first or second ray longest; adipose fin relatively low, united to the procurrent caudal fin; pectoral spine short, stout, nearly straight, with three to eight, usually four to six distinct pos- terior serrae in juveniles to adults; serrae relatively uniformly devel- oped, nearly straight to slightly curved, sharp, except in old specimens, their length greater than half the diameter of the spine shaft; end of caudal fin rounded to truncate; posterior corners of premaxillary tooth band slightly rounded; eye small, 2.1 to 2.6 times in snout; posterior process of cleithrum about equal in length to the width of the pectoral spine. CATFISH GENUS NOTURUS RAFINESQUE 103 Gill rakers on the first arch six to nine. There are (extremes in parentheses): (17) 19 to 21 (23) upper simple caudal rays; (16) 17 or 18 (20) branched caudal rays, of which there are typically 7 in the upper half of the fin and 10 or 11 in the lower half; and (13) 15 to 18 (19) lower simple caudal rays. The soft dorsal rays are five (in 2) or six (148). In five stained and cleared specimens there are eleven vertebrae anterior to the anal fin origin; the pectoral radials are fused on each side, except the fusion on one side is incomplete. As judged from x- rays and cleared specimens the hypurals are seldom fused; no fusion was observed in 34 specimens; hypurals 2—3 were fused in 7, and 6-7 in 2. The largest known specimen is 123 mm. in standard length. The general background color is light or dark brown in life as well as in most preserved specimens. No bright colors are present. The upper surface, barbels, fins, and side are nearly uniformly brown; the lower surface may be somewhat lighter, but is sprinkled with large distinct, brown chromatophores. These may become diffused with age, so that some old specimens, noted especially in those from western Tennessee, appear to have the medial portion of the abdomen, as well as other parts of the body, unpigmented. The fins are all heavily pigmented, sometimes light edged. The dorsal and anal fins may have a distal dark band of pigment. Typre.—The holotype, USNM 202143, is a male, 94.6 mm. in standard length. It is light brown colored with narrow pale or whitish margins to the fins. There are 6 soft dorsal rays, 21 anal rays, 20+7+ 10+15=52 caudal rays, 1 epural, 6 (3+3) unfused hypurals, and 36 vertebrae. On each side there are nine pelvic rays, eight soft pectoral rays, eleven preoperculomandibular pores, two internasal pores, and eight serrae posteriorly on the pectoral spines. The serrae are sharp pointed, but two left serrae and one right serration have split distally so that they have two sharp points instead of one. The head length is stepped into the standard length 3.35 times and the distance from the rear end of the adipose fin to the tip of the caudal fin is contained 1.8 times in the distance from the origin of the dorsal fin to the rear end of the adipose fin. Measurements are given in table 27. VariatTion.—Noturus phaeus apparently has relatively little geo- graphic variation in most structures studied, but the variation in pectoral rays (table 8) is prominent. In specimens from Kentucky, Tennessee, and northern Mississippi the modal number of soft pec- toral rays is eight and the mode in Louisiana and southern Mississippi is nine. This would point to a break, or shift, in characters in central Mississippi rather than a change across the Mississippi River low- lands. To the contrary, a break could be expected in the floodplain of the Mississippi River where little suitable habitat for phaeus is 104. U.S. NATIONAL MUSEUM BULLETIN 282 apparent, and is not expected across Mississippi where the distribu- tion is probably continuous. The variation in anal rays and vertebrae is included in tables 6 and 8. Variation from the strong modes is small in the pore counts and number of dorsal rays. The variation from nine pelvic rays is random and very similar to that of most other species of Noturus. The means for the same specimens as the pectoral ray counts range from 8.93 to 9.07, except the three specimens from Beaver Creek, Louisiana, with nine or ten pelvic rays, have a mean of 9.33. Variations in caudal ray counts appear to be slightly irregular, but probably insignificant. The following data include number of counts, range (in parentheses), and mean for combined samples for each river system. Caudal rays: Forked Deer River, Tennessee 13 (51-57) 53.92; Yazoo River, Mississippi 70 (50-58) 53.70; Little River, Louisiana 3 (55-57) 56.33; Brushy Creek, Louisiana 22 (53-59) 55.18; Beaver Creek, Louisiana 3 (51-54) 53.00. Upper-half caudal rays: Forked Deer River 13 (27-31) 28.00; Yazoo River 70 (24-29) 26.50; Little River 3 (28-30) 29.33; Brushy Creek 22 (26-29) 27.82; Beaver Creek 3 (26-27) 26.67. Lower-half caudal rays: Forked Deer River 13 (24-28) 25.92; Yazoo River 70 (24-380) 27.20; Little River 3 (27) 27.00; Brushy Creek 22 (26-30) 27.36; Beaver Creek 3 (25-27) 26.33. Distripution.—Noturus phaeus (map 6) is confined to the lower Mississippi Valley and the headwaters of Bayou Teche, Louisiana. It occurs in most eastern tributaries to the Mississippi River from south- western Kentucky to Louisiana, in the lower Ouachita drainage, and in the Red River system upstream to northwestern Louisiana. The population now in upper Bayou Teche is the result of recent diversion of streams from the Red River. Erymotocy.—The name phaeus (Greek) meaning of the hue or color of twilight, dusky, brown, or grayish brown, refers to the color of this species. ReELationsuip.—Noturus phaeus is closely related to Noturus Junebris. The somewhat similar Noturus nocturnus may be confused with phaeus, but the two differ in several characters, including num- ber of anal rays and relative slenderness of the dorsal spine, in addition to other skeletal characters. N. phaeus has prominent pectoral spine serrae; N. nocturnus never has uniformly developed serrae—when present the serrae typically are few, relatively short, and of irregular sizes. Aside from Noturus funebris, phaeus is probably closest related to msignis and nocturnus among the species of Noturus. Ecotoey.—N. phaeus is most commonly collected from permanent springs and small streams. It is frequently found under vegetation in CATFISH GENUS NOTURUS RAFINESQUE 105 moderate to fast clear water over small gravel or perhaps coarse sand. It has been taken in collections with the following species of Noturus: gyrinus, nocturnus, hildebrandi, stigmosus, and miurus. It is often col- lected with Noturus nocturnus in Louisiana and with Noturus halde- brandi lautus and Noturus miurus in western Tennessee. Noturus gilberti Jordan and Evermann ORANGEFIN MADTOM PuATEs 3 (Fia. 9), 9 (FIG. 1); Mar 7 Noturus gilberti Jordan and Evermann, in Jordan, 1889, pp. 351-353, pl. 43, figs. 2-2b (original description; Roanoke R., Roanoke,* Salem,* and Alleghany Springs, Va.; type, USNM 39931*).—Jordan, 1890, pp. 97, 122, pl. 13, figs. 2-2b (description; Roanoke R., Salem* and Roanoke,* Va.).—Bohlke, 1953, p. 43 ({in error]; lectoholotype, USNM 39931;* paralectotype, SU 1335,* Roanoke R., Roanoke, Va.).—Robins and Raney, 1956, p. 31 (Johns Cr.,* Newcastle, Va.; on moderately swift, cold water riffle; probably reached James R. by stream capture Roanoke tribs.).—Eddy, 1957, p. 154, fig. 388.— Taylor, 1957, p. 192.—Robins, 1961, p. 311 (remnant of older fauna). Schilbeodes gilberti (Jordan and Evermann).—Jordan and Evermann, 1896a, pp. 145, 148 (description; range) ; 1896b, p. 234; 1900, p. 3236, pl. 28, figs. 67— 67b (type, USNM 39931,* Roanoke R., Salem, Va.).—Jordan, 1904, p. 351.— H.S. Pratt, 1923, pp. 96-97.—Schrenkeisen, 1938, p. 167.—Hubbs and Raney, 1944, pp. 1, 24-25 (synonymy; range; comparison; Roanoke R., Glenvar,* Va.).—Fowler, 1945, p. 32 (Roanoke R. system only [not in Santee R. system]).—Raney and Lachner, 1946b, p. 220.—Raney, 1950, p. 187 (“known with certainty only from the Roanoke’’).—G. A. Moore, 1957, pp. 143, 145. Rabida gilberti (Jordan and Evermann).—Jordan, 1929, p. 93.—Jordan, Ever- mann, and Clark, 1930, p. 156.—H. 8. Pratt, 1935, p. 90.—Fowler, 1935a, pp. 6, 19 (Roanoke R., Va., only).— Driver, 1942, p. 254. TYPE-SPECIMENS.—USNM 39931 (lectotype) and USNM 161733 (2 paralectotypes), Roanoke River, Salem, Virginia, 1888, David Starr Jordan and party. USNM 40219 (8 paralectotypes), MCZ 31994 (1 paralectotype), and SU 1335 (4 paralectotypes), Roanoke R., Roanoke, Virginia, summer 1888, Jordan. OTHER MATERIAL STUDIED UNITED STATES: Vrrarnia: Roanoke R. and South Fork Roanoke R. at various points, from 0.25 mi. above Alleghany Springs, Montgomery Co. to 2.0 mi. W. of Salem, Roanoke Co.: CU 9417, 11548, 20298, 20314, 20702, 20767; UMMZ 138522, 165832; USNM 161960, 161961, 168151, 177259, 194752. Mayo R., U.S. Hwy. 58, 2 mi. NW. Stuart, Patrick Co.: USNM 162854. Johns Cr., New Castle, Craig Co.: CU 203538. Craig Cr., Hwy. 311, 2.2 mi. SW. New Castle, Craig Co.: USN M 194740. Craig Cr., Hwy. 616, 1.2 mi. NE. New Castle, Craig Co.: USNM 194753. Diacnosis.—Noturus gilberti is the only species of the subgenus Schilbeodes with very short dorsal and pectoral spines, immaculate undersurfaces, and broad light margins of the vertical fins. The 46 to *Material designated by an asterisk has been re-examined. 106 U.S. NATIONAL MUSEUM BULLETIN 282 54 caudal rays, 9 or 10 pelvic rays, included lower jaw, and short anal fin in combination also distinguish it from other species of the subgenus. Derscription.—Other counts and measurements are given in tables 17 to 26. Dorsal and ventral contours of body not noticeably tapering posteriorly; head slightly depressed; eye small, 2.5 to 3.3 times in snout; lower jaw included; posterior corners of premaxillary tooth band slightly rounded; branchiostegal membranes and lower surface of head covered with large papillae; spines very short, rather obtuse; serrae irregular, sometimes well developed on posterior edge of pectoral spine (pl. 3, fig. 9); posterior process of cleithrum short or obscure, its free length less than the diameter of the pectoral spine; adipose fin low, without free posterior end, but nearly free from the caudal fin which is truncate posteriorly. In six stained specimens there are 14 or 15, modally 14, vertebrae anterior to the origin of the anal fin. This range is identical to the range obtained for Noturus flavus. The pectoral radials (actinosts) are fused on both sides in each specimen. Each has six hypurals and one epural. The hypurals are unfused in four specimens; hypurals 2 and 3 are fused in one and 5 and 6 in another. There are seven to nine gill rakers on the first arch. The 47 indi- viduals counted have 6 soft dorsal rays. In the caudal fin there are 15 to 20, usually 17 to 19 upper simple rays; 15 to 17, almost constantly 16 branched rays of which 7 are in the upper half and 9 in the lower half of the fin; and 14 to 18, usually 15 to 17 lower simple rays. The largest individual examined is the Mayo River specimen, 85 mm. in standard length. Color was described from fresh material by Hubbs and Raney (1944, p. 25): “the first dorsal fin has a blackish base, but is otherwise pale . . .; the caudal fin has an orange border, which is broader above than below.” In specimens collected August 31, 1958, the overall color of the upper body surface was dark grayish brown. The light parts of the fins varied in color from light yellow to bright yellow or orange as follows: pelvic and anal fins light to medium yellow; adipose fin and caudal fin border medium yellowish orange; pectoral and dorsal fins bright yellowish orange to orange. In preserved material, the pelvic fins, mental barbels, anal fin; the lower surface of the head and abdomen, the adipose fin, about two- thirds of the dorsal fin, and a broad upper and lower margin and the tip of the caudal fin are immaculate. The side is brownish, only slightly lighter than the uniformly colored dorsal surface of the body which is dark brown or a gray-black; a black blotch on the middle caudal rays extends nearly to the margin of the fin; the pectoral fin is pigmented at the base and about the spine; its outer half is immaculate; the nasal barbels are black; the maxillary barbels are partially pigmented. CATFISH GENUS NOTURUS RAFINESQUE 107 The large specimen from the Mayo River is similar in color to Roanoke and James specimens but a large, dusky brownish blotch, not prominent, occupies the middle half of the anal fin from the base almost to the margin. The base of the dorsal fin is dark brown, as in other specimens, but a band of the same color extends along the first two rays almost to the margin, thence backward across the third and fourth rays. Typre.—Jordan and Evermann (in Jordan, 1889, p. 352, pl. 43, figs. 2-2b) figured Noturus gilberti and listed USNM 39931 as the type. Jordan and Evermann (1900, p. 3236, pl. 28, figs. 67-67b) published the same figures and were the first to indicate that they were “from”’ the type, USNM 39931, essentially designating a lectotype. The original drawings, bearing this catalog number, indicate that a speci- men 3.5 inches long from Roanoke River, Salem, Virginia was drawn. When examined by me, USNM 39931 contained four specimens, three of N. gilbertt and one of Noturus insignis. The catalog book and original records indicate that four specimens were originally included. It is not now possible to be certain which specimen of Noturus gilberti was drawn, but the largest one, slightly less than 3.5 inches in total length, is selected as the lectotype, retaining USNM 39931. It is in good condition, but slightly brownish, perhaps due to becoming dry during the rendering of the illustrations. The other specimens, origi- nally in this collection, showed no sign of having been dried. The lectotype is a female, 73.0 mm. in standard length. It has 15 anal rays, 17+6+9+17=49 caudal rays, 37 vertebrae, and 6 soft dorsal rays; on each side there are 9 pelvic rays, 9 soft pectoral rays, 10 preoperculomandibular pores, and 2 internasal pores. The pectoral spines are very short and broad. The head length is stepped into the standard length 4.3 times and the distance from the adipose notch to the tip of the caudal fin is contained 1.8 times in the distance from the origin of the dorsal fin to the adipose notch. One specimen MNHN 89-312 and one BMNH 89.10.30.89, both collected by David Starr Jordan from Roanoke River, Roanoke, Virginia, are not listed as paratypes. Although they were evidently collected with the paratype series, USNM 40219, they apparently were separated from the original series and not studied by Jordan and Evermann (see comments under Noturus furiosus, p. 187). VARIATION.—Specimens from the James and Roanoke River systems are similar. Counts from Roanoke River specimens broadly overlap those from the five James River specimens and the one from the Mayo River, except three from the James have 54 caudal rays whereas the highest number obtained from Roanoke specimens is 53. Disrrisution.—N. gilberti (map 7) is confined to Virginia. Since the original description, it has been thought to be restricted to head- 298-943 O—69——8§ 108 U.S. NATIONAL MUSEUM BULLETIN 282 Map 7.—Distribution of Noturus gilberti Jordan and Evermann. So far as known this species is confined to the areas of Virginia indicated by the symbols. waters of the Roanoke River, but in recent years specimens have been collected from the Mayo River and headwaters of the James River system adjacent to the Roanoke River. Specimens of Noturus leptacanthus and N. insignis have been reported as gilbert. Erymo.tocy.—WNoturus gilbertt was named for Charles Henry Gilbert, an outstanding American ichthyologist. RELATIONSHIPS.—The species is very distinct and does not appear to be closely related to any other kind of Noturus. It is assigned to the subgenus Schilbeodes because of its common characters, but in some respects, chiefly color pattern, number of paired fin rays, and vertebral elements, it resembles Noturus flavus. Ecotogy.—Hubbs and Raney (1944, p. 24) indicate that N. gilberta was taken from a rubble riffle. Specimens seem to prefer to hide in relatively fast water riffle habitats, beneath large stones and debris that is not swept away by the current. Noturus insignis is a common associate of gilberti. Subgenus Noturus Rafinesque The subgenus Noturus contains only one species, Noturus flavus Rafinesque, the type-species of Noturus. N. flavus has been placed in Noturus since recognition, except for irregular inclusion in the genus CATFISH GENUS NOTURUS RAFINESQUE 109 Pimelodus Lacépéde. The genus Noturus has either been treated as monotypic or enlarged to include all the recognized species of Noturus. The unfused pectoral radials, the specialization of the premaxillary tooth band, the greater numbers of pelvic and soft pectoral rays, an increase in the numbers of certain skeletal parts, and the relatively larger size attained are the chief characteristics of the subgenus. These characters, in view of the great variation of the same characters in other species of Noturus, are not worthy of use in the placement of Noturus flavus in a monotypic genus. Since flavus may be closely related to certain Schilbeodes, the discussion of characters is chiefly in relation to the members of that subgenus. The moderately long, nearly straight, mostly unserrated pectoral spine (pl. 3, fig. 10) shows no important differences from Schilbeodes; indeed, the species of that subgenus have a wide range of variation in shape and serrations that would virtually include all structures found in flavus. There are no anterior serrae; the posterior serrae are irregu- larly developed, usually indistinct and never uniformly turned toward the spine base. The recurved anterior hooks or steps of the spine can almost be duplicated in many individuals of N. insignis; the short grooves are moderate in depth and mostly distal on the spine. The body is moderately elongate as is also true in several Schilbeodes. The numbers of soft pectoral (9 to 11, modally 10) and pelvic (8 to 10, modally 9) rays are higher in flavus than in any other group; however, the high mean number of pelvic rays of flavus is approached by N. gilberti. These two species are the only forms without pigment on the lower surface of the head and abdomen, and are superficially alike, after preservation, in having broad light (immaculate) margins of the vertical fins; otherwise flavus is yellowish, slate-gray, or olive green above; it resembles exilis in having a yellowish spot beneath the posterior end of the dorsal fin base; there are no prominent dark blotches as in Rabida. The shape of the premaxillary tooth band appears to be of some importance. The posterior extensions constitute undoubted special- izations, but the function is not yet known. Posterior extensions of a similar nature but of dissimilar shape and structure have appeared at least twice in other genera of the family. Undoubtedly all three were derived independently. In N. flavus (pl. 2, fig. 3) the extension is long and narrow and emerges from the lateral corner of the otherwise rectangular band; in Pylodictis olivaris (Rafinesque) the extension is broad and long and emerges from the posterior edge; in Jctalurus balsanus (Jordan and Snyder) the extension is from the posterior edge, but is short and broad. In the genus Noturus, there is considerable specific variation in the shape of the premaxillary tooth band. In N. gyrinus, the transverse extent is greater than in other species; in 110 U.S. NATIONAL MUSEUM BULLETIN 282 Rabida and other Schilbeodes, the band of teeth is a short block, usually indented laterally; there is often a short extension posteriorly of the bone on each side. Sometimes these extensions bear short teeth that may be covered with flesh and visible only after the flesh is re- moved. This is true of large specimens of Noturus exilis. After examining the species of the genus in which skeletons were available, it was found that N. flavus differs from the other species chiefly in the extreme degree of development of the backward toothed process. In adults, however, there is no overlap of this character between flavus and other species of the genus. The width of the band is about 1.5 to 2.5 times the least length. Noturus flavus has two ossified pectoral radials on each side. In other species of the genus these fuse or tend to fuse at an early age but no tendency towards fusion has been noted in flavus. The lower jaw is included. The high number of caudal rays is interpreted as of no particular significance. There are usually eleven preoperculomandib- ular pores, the mode for the genus. The anal fin with 15 to 18, rarely 19 rays is of moderate length. The head is relatively large, and is projected 3.3 to 3.8 times in the standard length. The skeleton is well ossified, perhaps a factor associated with the greater size attained. The vertebral number averages higher than in other species of the genus, but is closely approximated by N. ezilis, N. gilberti, and N. wmsignis. The origin of the anal fin is slightly farther back than in other species, with the exception of N. gilberti. In 14 specimens of N. flavus there are typically 14 or 15 vertebrae anterior to the end of the first pterygiophore of the anal fin. This high number is duplicated in N. gilberti and is approximated in N. insignis and N. extlis. The high number of precaudal vertebrae (11 or 12) is also over- lapped slightly by N. insignis and N. furiosus; the number of caudal vertebrae overlaps those of many members of the other subgenera. Noturus flavus, however, is the only species of the genus that usually has nine ribs (range, 8 to 10); others typically have eight or fewer, seldom nine. There are ten branchiostegal rays as in many other Noturus. There is typically one epural. The six (3+3) or sometimes seven (3-+4) hypurals are seldom fused, at least in small to medium size specimens. In Noturus flavus, as in the subgenus Schilbeodes, the number of branched caudal rays is relatively high (16 to 22, usually 17 to 20), but the rays are only of moderate length. In species of Schilbeodes other than N. gilberti and N. leptacanthus the usual number is 17 or greater; species of Rabida usually have fewer branched rays. N. flavus is relatively northern in distribution and extends farther westward than any other species of Noturus. It prefers fast water in large streams, and is commonly associated with a rocky bottom. CATFISH GENUS NOTURUS RAFINESQUE 111 The name Noturus (Greek) is from notos (the back) and oura (the tail), referring to the connected caudal and adipose fins. Noturus flavus Rafinesque STONECAT PuaTEs 2 (FIG. 3), 3 (FIG. 10), 9 (FIG. 2); Map 8 Noturus flavus Rafinesque, 1818a, p. 41 (original description; the Ohio); 1820b, p. 68; 1820c, p. 362 [and reprinted in Jordan, 1877c, p. 41, and in Call, 1899, pp. 128-129] (description; ‘‘common near the falls’) —Kirtland, 1838, pp. 169, 195 (Mahoning R., Ohio; known as young catfish).—Storer, 1846, p. 406 [and zn reprint, 1846, p. 154] (description copied; synonymy; Ohio; Mahoning R.; L. Erie).—Kirtland, 1847, pp. 336-343, pl. 26, fig. 2 (?descrip- tion; synonymy; L. Erie; Mahoning R.; occasionally taken near Cleveland [The figure represents this form, but the counts given were either copied or the records are complex.]).—Gill, 1861a, p. 45 (synonymy); 1861b, p. 50; and Bleeker, 1862, p. 12, and 1863, p. 103 (type-species of Noturus).— Hayden, 1863, p. 178; and Girard, in Warren, 1875, p. 104 (Yellowstone R.*).—Giinther, 1864, p. 104 (compiled).—Cope, 1865, p. 277 (description [complex]; Swartz Cr., Genesee Co.,* Mich.; Youghiogheny R.* [Pa.]).— J. G. Cooper, 1869, p. 297 (‘‘The river at Fort Benton’’).—Cope, 1869, p. 237 (Miami R., Ind.;* Michigan;* Kiskiminitas R., W. Pa.); 1872, p. 442 (“From the waters of the Platte; identical with those from the Ohio’’).— Jordan, 1875, p. 225; 1878c, p. 368.—Jordan and Copeland, 1876, p. 160.— Gill, 1876, p. 423; and Jordan, 1876a, p. 96; and Jordan and Gilbert, 1877b, p. 85 (synonymy).—?Jordan, 1876b, p. 303 (description; synonymy; range [all complex]).—?E. W. Nelson, 1876, p. 50 (Illinois [ecomplex]).—{Klippart], 1877, pp. 149, 153 (Ohio; Indiana).—Jordan and Gilbert, 1877a, p. 2 (Indiana; Ohio Valley).—?Jordan, 1877a, p. 46 (White R., Ind. [complex]).—Jordan, 1877b, pp. 352, 372, 377 (comparison; synonymy; White R. and tribs., Indianapolis, Ind. [complex]); 1877c, pp. 9-41 (synonymy; nominal species of Rafinesque reviewed); 1877d, pp. 71-119, pl. 36, figs. 54-55 (synonymy; range; general localities [complex]).—Jordan and Brayton, 1878, p. 87.— Jordan, 1878a, p. 118 (synonymy; distribution [complex]).—?Jordan, 1878b, p. 67 (Illinois distribution).—Jordan, 1878d [and 1884], p. 335 (description; range); 1878e, p. 414.—Forbes, 1880a, pp. 42-43 (in food of Micropterus salmoides).—Cope, 1881, p. 77 (range; description; tribs. of the Ohio, Pa.).— Jordan, 1882, pp. 739-800 (description; synonymy; range; Ohio; Ohio R.; White R., Indianapolis).—Jordan and Gilbert, 1883, pp. 100-101 (descrip- tion; synonymy; range [complex]).—Swain and Kalb, 1883, pp. 638-644 (synonymy; distribution [in error]; description).—Bean, 1884, p. 491 (range [in error]; Madrid,* N. Y.).—Gilbert, 1884, p. 199 (Indiana records including [Bloomington*], Monroe Co.).—Graham, 1885a, no. 30; and Cragin, 1885, p. 107 (Rock Cr., Kans.).—Graham, 1885b, p. 71 (Kansas).—Jordan and Meek, 1885, pp. 2-16 (Des Moines R., Ottumwa,* Iowa; Hundred and Two R., near Bedford, Iowa* and Maryville,* Mo.; Blackwater Cr., Browns- ville and Flat Cr., near Sedalia,* Mo.; Grand R., Clinton* and Tabo Cr., Calhoun,* Mo.).—Jordan, 1885, p. 802.—Forbes, 1885a, p. 84 [and 1900, p. 76]; 1885b, p. 108.—Jordan and Gilbert, 1886, p. 6 (comparison only) .— Evermann and Bollman, 1886, p. 335 (Pigeon Cr., near Monongahela City,* Pa.).—Evermann, 1886, p. 3 (Indiana record).—Eigenmann and Fordice, *The material indicated by an asterisk has been re-examined. bb U.S. NATIONAL MUSEUM BULLETIN 282 1886, p. 410 (Bean Blossom Cr.,* Monroe Co., Ind.).—Gilbert, 1886, p. 207 (Kansas records).—Hay, 1887, pp. 250-253 (Kansas records: North Fork Solomon R., Lenora;* Smoky Hill R., Wallace).—Forbes, 1888b, pp. 512, 515 (food relations).—Henshall, 1888, p. 77 (Ohio records).—Evermann and Jenkins, 1888, pp. 44-56 (Indiana records including Deer Cr., Camden*).— Gilbert, 1889, p. 40 (Kansas records).—Meek, 1889b, pp. 167-168 (Iowa; description).—Jordan, 1890, pp. 155-167 (Indiana records including Wabash R., Vincennes* and New Harmony;* West Fork White R., Spencer*).— Bollman, 1890, p. 221 (Kalamazoo R., Battle Creek,* Mich.) —McCormick, 1890, p. 126 (Lorain Co., Ohio).—J. Nelson, 1890, p. 671 (description; synonymy; range [not in New Jersey]).—Gilbert, 1891, pp. 146, 152 (Shoal Cr., Florence,* Ala.).—Bean, 1892, pp. 18-19 (description; range; ecology; economics; Pennsylvania)—McCormick, 1892, p. 13 (description; Lorain Co., Ohio distribution) —Woolman, 1892a, pp. 285, 287 (Kentucky record) .— R. R. Wright, 1892, p. 443.—Call, 1892, p. 55.— Meek, 1892a, p. 12 (Iowa); 1892b, pp. 232-246 (misprinted Nocturus flavus; Iowa records: Cedar R., Cedar Rapids;* Missouri R., Sioux City,* near mouth Big Sioux R.; Big Sioux R., Sioux Falls and Sioux City*); 1892c, p. 108 (compiled); 1893, p. 229 (range [in error]).—Bean, 1893, pp. 27-28 (Iowa; range).—Kirsch, 1893, pp. 262-264 (Tennessee records).—Evermann, 1893, p. 77 (South Fork Cheyenne R., Cheyenne Falls; Belle Fourche R.,* Belle Fourche).— Eigenmann and Beeson, 1894a, p. 81; 1894b, p. 44 [and 1905, p. 120] (Indiana records, some compiled [some are Noturus gyrinus]).—Hay, 1894, pp. 172-175 (description; economics; “‘Great Lakes from shore to twenty fathoms’’; other Indiana records compiled [some are N. gyrinus])—Garman, 1894, p. 56 (Licking R., Ky.).—Eigenmann, 1894, pp. 107-132 (Missouri R., Craig, Montana;* “reported at Medicine Hat’’).—Kirsch, 1894, p. 87 [and vol. 14, p. 36] (Indiana records include Eel R. [Logansport*]).—Meek, 1894b, p. 1385 (Salt Cr. [Havelock*], near Lincoln, Nebr.).—Kirsch, 1895, pp. 327-335 (records include: Maumee* R., Fort Wayne, Ind.; Maumee R., Defiance* and Grand Rapids,* Ohio; St. Marys R., Decatur,* Ind.; Auglaize R., Cloverdale* and Defiance,* Ohio; Blanchard R., Ottawa,* Ohio).— Call, 1896, p. 14 (Ohio basin; Falls of the Ohio; ‘‘occasionally seen in mar- kets”’).—Jordan and Evermann, 1896a, pp. 143-144 (description; synonymy; range [in error]); 1896b, p. 233 (synonymy; range [in error]).—Eigenmann, 1896, pp. 252-253 (reproduction; Indiana record).—Kirsch, 1896a, p. 48 (Indiana record); 1896b, p. 105 (Maumee basin records repeated).—Cox, 1896, p. 608 (Minnesota record).—Evermann and Cox, 1896, pp. 334-426 (synonymy; Missouri R. basin; Nebraska records include: Beaver Cr., York;* Platte R., Grand Island.* South Dakota records include: Beaver Cr., Buffalo Gap;* Belle Fourche R., Belle Fourche.* Wyoming records include: Powder R., Arvada;* Platte R., Douglas.* Several records compiled [Missouri R., St. Joseph copied in error]).—Cox, 1897, pp. 16-78 (description; record compiled).—Osburn and Williamson, 1898, pp. 11, 19 (Franklin Co., Ohio records including Scioto R.*).—Parker, Williamson, and Osburn, 1899, pp. 22, 32 (Ohio record).—Jordan and Evermann, 1900, p. 3236, pl. 27, fig. 63 (L. Champlain, Westport,* N.Y.).—Herrick, 1901, pp. 230-231 (descrip- tion head canals)—Evermann and Kendall, 1901, p. 480 [and 1902, p. 210] (Nine-mile Point,* L. Ontario, N. Y.).—Osburn, 1901, p. 26-27 (descrip- tion; Ohio records [Little Darby Cr. listed in error?]).—Jordan and Snyder, 1901, p. 117 (scattered villi on skin of sides).—Evermann, 1902, p. 95.—Hay, *The material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 113 1902, pp. 70-71.—Bean, 1902, pp. 254, 277 (range; figure; Nine-Mile Point,* N.Y.); 1903, pp. 91-92, 739 (description; range [in error]; synonymy; eco- nomics; Nine-Mile Point,* N.Y.).—Large, 1903, pp. 9-10 [and 1905, pp. 56-57] (description; Illinois distribution; Mackinaw Cr.; [but not Kaskaskia R.]).—Jordan, 1904, pp. 41, 351.—Mitchell, 1904, p. 162 (description oral breathing valves).—Henshall, 1906, p. 3 (Missouri R., Craig,* Mont.).— Michael, 1906, p. 8 (range [in error]; Michigan records compiled).— Reed, 1907, pp. 556-564, fig. 1 (description poison apparatus).—Cockerell, 1908, p. 163 (comparison; Rocky Mountain distribution; Wyoming;* Platte R.;* Missouri R.,* Mont.).—C. W. Nash, 1908, pp. 23-25 (description; L. Erie and L. Ontario).—Meek, 1908, p. 140 (range [in error]; reproduction; Indi- ana).—Forbes, 1909, pp. 387-437 (range [in error]; Illinois distribution [Kaskaskia R. record is N. exilis; certain others are N. nocturnus]; ecology) .— Forbes and Richardson, 1909 [and 1920, pp. Ixxviii-cxv], pp. Ixxiii-cix, 176-201, color plate, fig. 48, map 57 (Illinois distribution [see Forbes, 1909]; ecology; synonymy; description; range [in error]; figured; teeth figured [captions transposed in first edition]).—Meek and Hildebrand, 1910, pp. 244-245 (description; range [in error]; Illinois record).—Hahn, 1910, p. 175 (Bean Blossom Cr., Indiana).—Leathers, 1911, pp. 246-249 (Saginaw Bay, Michigan records).—Fowler, 1912b, p. 47 (Two Lick Cr.* and Cherry Run,* Indiana Co., Pa.).—Shelford, 1913, p. 119 (Illinois record).—Hankinson, 1913, p. 112 (not found around Charleston, Ill.).—C. W. Nash, 1913, p. 252 (streams near Toronto, Canada).—Fowler, 1913, p. 91 (Pennsylvania records including Erie,* Erie Co.).—Halkett, 1913, pp. 17, 57 (Ottawa R., Canada, at Ottawa Fish Hatchery; range [in error]).—Forbes, 1914, p. 18, fig. 14c, map 57 (Illinois distribution [in error]; teeth figured).—Fowler, 1915a, p. 208 (records re-examined and relisted, also Iowa: [one record only] Brook R.,* [location?]); 1918b, p. 7 (relisted).—Evermann, 1918, pp. 306-366 (syn- onymy; luteus spelled lutius; Kentucky and Tennessee records compiled [Tennessee R., Florence is probably error for Shoal Cr.*]).—A. H. Wright, 1918, p. 543 (Salmon Cr.,* Monroe Co., N.Y.).—Fowler, 1919, p. 57 (Penn- sylvania records relisted).—T. Surber, 1920, p. 20 (description; presence in Blue Earth R., Mankato, Minn. questioned).—Conger, 1920, p. 10.— Dymond, 1922, pp. 59, 62 (L. Erie records).—Hankinson, 1923, p. 32; and 1924, pp. 84, 86 (New York record).—H. S. Pratt, 1923 [and 1935, p. 89], p. 95, fig. 41d (teeth figured [transposed name in first edition]; description; range [in error]).— Reed, 1924a, pp. 431-451, fig. 9 (comparison and descrip- tion of spines).—C. L, Hubbs, 1926, pp. 49, 51 (Great Lakes basin; com- parison).—Cahn, 1927, pp. 41-42 (Fox R., near Lannan and Oconomowoc R., Stonebank, Wis. [a specimen, USNM 87510, Wisconsin,* Oct. 16, 1925, A. R. Cahn]; food; ecology).—Fowler and Carlson, 1927, p. 66 (Pennsyl- vania records; habits).—Greene, 1927, p. 306 (Wisconsin*).—Greeley, 1927, pp. 49-57 (economics; New York distribution and records).—Potter and Jones, 1928, p. 355 (compiled).—Hubbs and Greene, 1928, p. 390.—Eaton, 1928, p. 42 (New York record).—Greeley, 1928, pp. 87-105 (New York records).—Hubbs and Brown, 1929, p. 42 (trib. of Kettle Cr.,* near Fultons Bridge, about 5 mi. from L. Erie, Ontario).—Dymond, Hart, and Pritchard, 1929, pp. 4, 24 (Credit R. and ?Burlington Bay, Ontario).—Greeley, 1929, pp. 155-174, color pl. 5 (male figured; reproduction; ecology; economics; New York distribution and records).—Sibley, 1929, p. 184 (food).—Jordan, 1929, pp. 92-93.—Hankinson, 1929, p. 452 (North Dakota records: Little *The material indicated by an asterisk has been re-examined. 114 U.S. NATIONAL MUSEUM BULLETIN 282 Missouri R.* [at Marmarth]; Cannon Ball R.* [just below Mott]; Heart R.* [about 10 mi. 8. of Richardton, Stark Co.]).— Hubbs and Ortenburger, 1929, p. 96 (Elk R.,* 7 mi. N. of Grove, Delaware Co., Oklahoma only).— Jordan, Evermann, and Clark, 1930, p. 155 (range [in error]; synonymy).— Greeley, 1930, pp. 45-82 (New York record).—Thompson and Hunt, 1930, pp. 27-60, map 32 (ecology; Champaign Co., Illinois distribution).—Coker, 1930, p. 180 (lock gate,* Keokuk, Iowa).—Osburn, Wickliff, and Trautman, 1930, p. 174 (Ohio).—Greeley and Greene, 1931, pp. 54-89 (ecology; New York distribution and records).—Greeley and Bishop, 1932, pp. 76-87 (New York distribution and records).—Fish, 1932, p. 351, fig. 67 (L. Erie, along rocky shores; reproduction compiled; description and figure of 20 mm. individual).—R. A. Moore, 1933, p. 17 (description of kidneys).— Greeley and Bishop, 1933, pp. 89, 98 (ecological distribution; New York records).— E. Moore, 1933, p. 20; and Odell, 1933, p. 128 (Sacandaga Reservoir, N. Y.).— Pate, 1933, p. 153 (food; Sacandaga Reservoir distribution).—Churchill and Over, 1933, pp. 9-60, fig. 47 (South Dakota; description).—Greeley, 1934, pp. 96, 104 (reproduction; New York records).—Van Cleave and Mueller, 1934, pp. 170-822 (Oneida L., N.Y.; parasites).—Ricker, 1934, p. 107 (ecology?; mouth of Credit R., Ontario).—Greeley, 1935, pp. 86, 96 (ecology; “9 collections from Mohawk River or tributaries,’ N.Y.).—O’Donnell, 1935, p. 484 (Illinois ecology; distribution [in error])—Greene, 1935, pp. 142-146, map 59 (Wisconsin distribution; ecology; range; dispersal routes; Cahn’s record of Schilbeodes miurus from Wisconsin may be flavus).—?Ewers and Boesel, 1935, p. 66 (in food of Aplites salmoides; Buckeye L., Ohio).— Aitken, 1936, p. 33 (Iowa).—Toner, 1937, p. 14 (L. Ontario and St. Lawrence R.; spawning habits).—Greeley, 1938, p. 69 (economics; New York distribu- tion).— Welter, 1938, p. 67 (Kentucky records).—Blatchley, 1938, p. 66 (description; Indiana distribution, in part: Walnut and Raccoon Creeks) .— Schrenkeisen, 1938, pp. 165-167 (description; range [in error]).—Simon and Simon, 1939, pp. 52, 57 (comparison; Wyoming records: Big Horn R.;* Tongue R.; Belle Fourche R.;* Laramie R.).—Trautman, 1939, p. 281 (more numerous, 1895, in Maumee R. than subsequently).—Hubbs and Lagler, 1939, p. 26.—Raney and Lachner, 1939, pp. 158, 160 (in associations; nesting habits; Shenango R., near Delaware Grove* and French Cr., Mill Village,* Pa.).—Raney, 1939a, p. 275 (Ohio drainage of W. Pennsylvania) .— Kuhne, 1939, pp. 62, 68 (comparison; Tennessee).—Bangham and Hunter, 1939, pp. 401-434 (parasites; W. Lake Erie).—Breukelman, 1940a, p. 372 (NW. Kansas distribution); 1940b, pp. 381, 383 (Kansas distribution: Neosho,* Spring,* Verdigris,* Osage,* Solomon,* and Smoky Hill* Rivers; Hackberry Cr.;* Smoky Hill R., Logan Co.*).—Fowler, 1940b, p. 8 (Penn- sylvania records compiled).—Greeley, 1940, p. 76 (New York records).— Senning, 1940, pp. 104-108 (New York records).—Hubbs and Lagler, 1941, pp. 63-64, fig. 82 (comparison; range [except Texas]).—Shoup, Peyton, and Gentry, 1941, pp. 70, 73 (Tennessee records: Puncheon Camp Cr., Allred, Overton Co.;* Ashburn Cr., 1 mi. below Ashburn [Creek] Schoolhouse, * Clay [or Pickett] Co.; Sewell Cr., E. of Millers Chapel, Pickett Co.;* Town Br., N. of Byrdstown, Pickett Co.* and SW. of Livingston, Overton Co.;* Wolf R., NW. of Byrdstown, Pickett Co.;* Obey R., Eagle Creek Ford,* Pickett Co.; Franklin Cr., S. of Moodyville, Pickett Co.;* Blackmans Fork Roaring R., E. of Gainesboro, Jackson Co.;* Big Indian Cr., W. of Boatland Bridge [Jamestown], Fentress Co.*).—Aitken, 1941, p. 389 (Iowa).—Jen- *The material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 115 nings, 1942, p. 365 (Kansas record).—Simon and Simon, 1942, p. 53 (Wyo- ming records).—Shoemaker, 1942, p. 269.—Driver, 1942, p. 252 [and 1950, p. 262] (comparison; range [in error]).—Palay, 1943, pp. 247-274 (neuro- secretory granules and neurosecretion).—Hinks, 1943, pp. 58, 62 (comparison only).—Haas, 1943, pp. 162-163 (Rock R. drainage, Ill.)—Eddy and Surber, 1943, pp. 151, 161 [and 1947, pp. 170, 180, diag. 7d] (comparison; ecology; range; Minnesota distribution and records).—Radforth, 1944, pp. 6-61, figs. 22-23 (Ontario and L. Erie distribution; comparison with iso- therms; ecology; glacial distributions) —Hubbs and Raney, 1944, pp. 3, 6 (synonymy; classification).—Palay, 1945, pp. 129-143 (description of tractus preoptico-hypophysis).—Fowler, 1945, pp. 13, 32, 351, fig. 162 (distribution [Tennessee R. system only]; specimen from Clinton,* Mo., figured).— Gerking, 1945, pp. 16, 73-74, map 62 (ecology; Indiana distribution [but not some literature records; see also IU Material studied p. 118]).—Cuerrier, Fry, and Prefontaine, 1946, p. 26 (Chateauguay,* St. Francis, and Nicolet Rivers, near Montreal, Canada).—C. L. Hubbs, 1946, p. 38.—Simon, 1946, pp. 94-117, fig. 74 (description; range; Wyoming records).—Mélancon, 1946, p. 131.—Hubbs and Lagler, 1947 [and 1949], pp. 71-72, fig. 174 (comparison; ecology; range [in error])——Raney and Roecker, 1947, p. 172 (eaten by gartersnake, New York).—Fischthal, 1947, pp. 162, 181 (parasites, Vermillion and Yellow Rivers, Wis.).—Dymond, 1947, pp. 22-23 (comparison; Canadian distribution [in part only]: upper St. Lawrence R., near Montreal; Great Lakes, N. to Bruce Peninsula).—Smith, Johnson, and Hiner, 1949, p. 208 (Root R. tribs., Minn.).—Lincicome and Van Cleave, 1949, p. 426 (records of Leptorhynchoides thecatus, a parasite, compiled) —Hooper, 1949, p. 35 (vertebrae used in age determination).—D. C. Scott, 1949, p. 177 (Indiana record).—Harrison, 1949, p. 338 (Des Moines basin, Iowa).—Moore and Paden, 1950, pp. 87, 90 (reference to Meek’s records [which are probably Noturus exilis] from Illinois R. and Barren Fork; Illinois R. system, Oklahoma localities: Illinois R.,* near mouth of Flint Cr., Delaware Co.; Flint Cr.,* from its mouth to 1.5 mi. upstream, Delaware Co.; Illinois R., near Scraper,* Cherokee Co:; Illinois R., E. of Tahlequah,* Cherokee Co.; Illinois R., about 5 mi. NE. of Gore,* Sequoyah Co.; Illinois R., near mouth,* Sequoyah Co.; Illinois R., above Tenkillers Dam,* Sequoyah Co.).—Fischthal, 1950, p- 100 (parasites; Hay Cr., Wis.)—Lachner, Westlake, and Handwerk, 1950, p. 93 (associations; French Cr., Carlton,* Pa.).—Starrett, 1950, pp. 118, 126 (Boone Co., Iowa).—Penn, 1950, pp. 648, 651 (compiled). —Laakso, 1951, p. 100 (Montana record).—Legendre, 1951, p. 3 (Quebec).—Bailey, 1951, pp. 194, 224 [and in reprint, 1951, pp. 194, 226] (comparisons; Iowa) .— Harlan and Speaker, 1951, p. 97, pl. 18 (ecology; Iowa distribution; descrip- tion compiled) —Moore, 1952, p. [6] (Oklahoma).—Larimore, Pickering, and Durham, 1952, pp. 8-25 (ecology and distribution in Jordan Cr., Ill.).— Cleary, 1952, p. 439 (Wapsipinicon R., Iowa).—Rostlund, 1952, pp. 33, 274 (map of distribution [in error]; records: [St. Francis R. listed in error; Wash- ington Co., Ark., records probably based on misidentifications of Notwrus ezilis]; Republican [Arikaree?] R., Nebr., near Colorado border; Maitland R., Huron Co. and Glengarry Co., Ontario; others compiled [some, as Manitoba, probably in error]).—Legendre, 1953, pp. xi, 29, fig. 58 (compari- son; Quebec, Canada).—Beckman, 1953, pp. 69-70, 109 (figure; description; rare [or hypothetical] in E. Colorado bordering Nebraska; recorded, Repub- lican R. near Colorado line).—Fischthal, 1953, p. 101 (parasites, Wisconsin) .— *The material indicated by an asterisk has been re-examined. 116 U.S. NATIONAL MUSEUM BULLETIN 282 ??Martin and Campbell, 1954, pp. 47-53 (Black R., Mo., on riffles [mis- identifications; probably Noturus nocturnus]).—Cleary, 1954, p. 633 (dis- tribution, Cedar and Iowa Rivers, Iowa).—W. B. Scott, 1954, p. 68 (char- acters; size; distribution in E. Canada; life history; food; figured) —Langlois, 1954, pp. 34-278 (food; reproduction and life history notes; Lake Erie distribution and records).—C. Hubbs, 1954, pp. 277, 285 (not in Texas) .— Harrison and Speaker, 1954, p. 519 (Iowa distribution) .—Cross, 1954, p. 311 (ecology; Kansas records).—Legendre, 1954, pp. 14, 43, fig. 58 (key ; Quebec) .— Liegey, Donahue, and Eaton, 1955, pp. 12-19 (Ischua Cr., N.Y.).—Hall, 1955, p. 36 (population reduced by reservoir, Illinois R., Okla.).—Gerking, 1955, pp. 51, 75 (Indiana records; key).—Personius and Eddy, 1955, p. 42 (compiled).—Cleary, 1956, p. 271, map 66 (Iowa distribution).—Harlan and Speaker, 1956, p. 113, pl. 21.—Bailey, 1956, pp. 335, 364 (key, Iowa).— Hubbs and Lagler, 1957, p. 5.—Taylor, 1957, p. 192.—?Paloumpis, 1957, p. 60 (oxbows, Squaw Cr., Boone Co., Iowa).—Eddy, 1957, p. 151, fig. 381 (key; range, in part).—G. A. Moore, 1957, pp. 141, 142, fig. 2-77d (descrip- tion; range, in part).— Underhill, 1957, pp. 21, 27, 29, map 17 (Minnesota distribution; arrival post Lake Agassiz) —Schelske, 1957, pp. 40-46 (occur- rence, Verdigris R. system, Kans.).—Slastenenko, 1958a, p. 7 (Canadian distribution); 1958b, pp. 246, 353, fig. 101 (compiled).—W. B. Scott, 1958, p. 19 (Canadian distribution; ?hypothetical southern Canadian Plains region).—Clarke, Breukelman, and Andrews, 1958, p. 168 (Lyon Co., Kans.).—Hubbs and Lagler, 1958, pp. 89, 90, fig. 174 (key; range, in part).— Minckley, 1959, pp. 428, 432-433 (ecology and records, Blue R. basin, Kans.).— Underhill, 1959, p. 101 (lower Vermillion R., S.D.).—Minckley and Deacon, 1959, p. 348 (in food Pylodictis, Blue R., Kans.).—Metcalf, 1959, p. 383 (southern Kansas records).—Lennon and Parker, 1959, p. 15 (lower Abrams Cr.,* Great Smoky Mountains National Park).—Trautman, 1959, pp. 96, 432-434, fig. 110, map 110 (description, ecology, Ohio distri- bution ; range).— Ross, 1959e, pp. 8, 24 (key; hypothetical, New R. system) .— Hallam, 1959, p. 159 (as Notrus; associations, Ontario; infrequently with trout; commonly in warmer streams) .—Becker, 1959, p. 96 (central Wisconsin distribution; ecology).—Suttkus, 1961, p. 63, fig. 4 (comparison; skull illustrated).—Deacon, 1961, pp. 395, 404, 408-421 (sedentary; ‘lacks air bladder” [erroneous]; distribution and abundance Neosho and Marais des Cygnes Rivers, Kans.).—Deacon and Metcalf, 1961, p. 317 (Wakarusa R., Kans., records).—Bailey and Allum, 1962, pp. 90, 118-122 (characters; synonymy; South Dakota stations* [all their specimens examined]).—Carr, 1962, p. 6 (Saginaw Bay).—Lennon, 1962, p. 6 (rare, Abrams Cr.*).—Clay, 1962, p. 92, fig. 62.—Brown, 1963, p. 25 (Missouri and Yellowstone Rivers, Mont.).—Larimore and Smith, 1963, pp. 324-344, fig. 46 (Champaign Co., Ill., records; distribution map; ecology).—W. B. Scott, 1963, p. 123 (On- tario).—Eddy, Moyle, and Underhill, 1963, pp. 113, 115 (Minnesota distri- bution; not above St. Anthony Falls).—Nursall and Lewin, 1964, p. 128 (description; Milk R., Alberta record).—Paquet, 1964, pp. 95, 99 (Etchemin R., Quebec).—Walburg, 1964, pp. 4, 16, 27 (found Lewis and Clark Lake only first two years of impoundment).—Smith, 1965, p. 8 (Illinois distribu- tion).—Fee and Drum, 1965, p. 321 (Lernea parasite, Des Moines R., lowa).— Rock and Nelson, 1965, p. 138 (mortality from Aeromonas, Rock R.., Ill.).— Johnson, 1965, pp. 350, 352 (Humber and Rouge R. systems, Ontario).—H. H. Moore and Braem, 1965, pp. 2, 44 (records and distribution, Lake Superior *The material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 117 tributaries, Wis., including Amnicon* and Nemadji* Rivers).—Metcalf, 1966, pp. 34, 43-44, 59-62, 80, 150, map 39 (records, ecology, distribution, history, Kansas R. system).—Cross, 1967, pp. 197, 221, fig. 18A (figure, description, Kansas distribution).—Branson, 1967, pp. 137, 151 (compiled). Pimelodus flavus (Rafinesque).—De Kay, 1842, p. 187 (Ohio; “‘young catfish with rudiments of an adipose fin’’).—Bleeker, 1858, p. 210 (Ohio). Noturus species.— Rafinesque, 1818b, p. 107 (one species inhabiting the Ohio). Noturus luteus Rafinesque, 1819, pp. 421-422 (original description; no locality given other than interior of U.S.).—Gill, 1861a, p. 45 (equals Noturus flavus) ; 1876, p. 423 (original description reprinted; name of Noturus flavus changed by Rafinesque to Noturus luteuws).—Jordan, 1882, p. 739 (Ohio).—Jordan and Evermann, 1896a, p. 144 (Ohio R. indicated as type-locality). Noturus occidentalis Gill, 1861a, pp. 45-46 (original description; Platte R.).— Giinther, 1864, p. 105 (Platte R.; description [but not records or figure of teeth]).—Jordan and Copeland, 1876, p. 160 (Platte R.).—Gill, 1876, pp. 423-424, pl. 8 (description; Platte R.).—Jordan, 1877b, p. 372 (equals Noturus flavus). Noturus platycephalus [misidentifications] Giinther, 1864, pp. 104-105 (original description; North America;* ‘‘intermaxillary” teeth figured).—Jordan and Copeland, 1876, p. 160 (North America).—Jordan, 1877b, p. 372 (equals Noturus flavus). Noturus insignis (Richardson) [misidentification]—Swain and Kalb, 1883, pp. 640-641 (Nebraska and Platte R. only). Schilbeodes insignis (Richardson) [misidentifications].—?Fowler, 1906b, p. 595 (Allegheny R., above Port Allegany, McKean Co., Pa.); 1913, p. 92 (records questioned, Port Allegany, Pa.).—Hubbs and Raney, 1944, pp. 12-20, map 1 (synonymy, in part and Indiana records).—Gerking, 1945, p. 75 (compiled Indiana records). Noturus exilis Nelson [misidentifications]—Eigenmann and Fordice, 1886, p. 410 (Bean Blossom Cr., Monroe Co. [SU 3980, White R., Spencer*], Ind.).— Evermann and Jenkins, 1888, pp. 44-54 ({USNM 39600] Tippecanoe R.,* Ind.).—Hay, 1894, pp. 172, 174 (Indiana records compiled; description?) ; 1902, p. 71 (Indiana records compiled).—Hahn, 1910, p. 175 (taken, 1885, in Bean Blossom Cr., Indiana, by Eigenmann; not taken in 1904).—?Charles, 1967, pp. 386-389 (Green R., Ky.). Schilbeodes exilis (Nelson) [misidentifications]—Eigenmann and Beeson, 1894a, p. 81; 1894b, p. 44 [and 1905, p. 120]; and Meek, 1908, p. 141 (Indiana records compiled).—Eigenmann, 1896, p. 253 (Indiana). Rabida ezxilis (Nelson) [misidentifications].—Blatchley, 1938, p. 67 (Indiana records compiled). Schilbeodes gyrinus (Mitchill) [misidentifications].—?Fowler, 1907a, p. 14; 1913, p. 92; 1919, p. 57; and 1940b, p. 8; and ?Greeley, 1927, p. 57 (Genesee R., below Gold, Potter Co., Pa.).—?Keim, 1915, p. 51 (headwaters of Genesee R., Potter Co., Pa.).—Greene, 1935, map 60 (Wisconsin record: [St. Croix R., 11 mi. N. of Danbury,* Burnett Co.]). Stonecat.—?Eschmeyer, 1943, p. 47 (common name only; bay of Norris Reser- voir, Tenn.). Schilbeodes nocturnus (Jordan and Gilbert) [misidentifications]|—Gerking, 1945, pp. 74-75 (in part; Indiana station nos. IU 183,* 186,* 225,* 233,* 253* [3 mi. NE. Osgood instead of 3 mi. NW. West Harrison], 254*). *The material indicated by an asterisk has been re-examined. 118 U.S. NATIONAL MUSEUM BULLETIN 282 Type.—USNM 202494 (formerly UL 12460) (neotype of Noturus flavus Rafinesque, herein designated), collected from Eagle Creek (a tributary to the lower Kentucky River which is tributary to the Ohio River), from pool below bridge at state highway 36, 3.5 miles east of Jonesville in Grant County, Kentucky, August 11, 1959, by William R. Turner. OTHER MATERIAL sTUDIED.—No locality known: BMNH (two syn- types of Noturus platycephalus, ‘Old Collection’; no locality or catalog number with specimens). UNITED STATES: Iuutno1s: INHS 14, 196, 7119, 24989, 26456, 26596, 26597, 26619, 26690, 26697, 27610, 27612, 27645, 27648, 27662, 27669, 27747, 28044, 28061, 28132, 28142, 28156, 28177, 28189, 28193, 28199, 28219, 28225, 28237, 28255; INHS (Salt Fork R., S. of Oakwood, Vermilion Co., and Rock R., near Como, Whiteside Co.; other nos. 834, 979, 980, 1146, 1149, 1158, 1177, 1193, 3098, 3113, 3117); CNHM 1379, 21834, 32738, 42206, 43381, 43402; CNHM (Hickory Cr., Will Co.); Cu 3400, 7450, 7547; UMMZ 169749; UMMZ (Bauman nos. 46, 50, 53); USNM 200479; UL 7791 (Ohio R., Lock 51, Golconda). INDIANA: USNM 36747, 40655, 66542 thru 66545, 66548, 67783, 69029-30, 121977; IU 29, 48, 102, 141, 157, 178, 184, 186, 208, 225, 253, 254, 255, 257, 258, 261, 262, 339, 340, 342, 345, 347, 350, 352, 448, 506; UMMZ 66638, 99924, 99956, 100866, 113555, 126474; UMMZ (IU nos. 2863, 3149, 8880, 9627). Iowa: USNM 61936, 174930; UMMZ 101098, 101271, 102260, 146789, 146827, 146831; TU 10199. Kansas: KU 232.7, 255, 645, 646 (Drum Cr., Montgomery Co.), 647, 648, 650, 651, 652 (Hackberry Cr., Gove Co.), 1521; KU (Bourbon Co., and Solomon R., Cloud Co.); USNM 9428 (Fort Lyon, Kansas, Dr. E. Palmer; this is the same locality and collector as listed for the type of Cliola camura Jordan and Meek, and is in apparent error), 131673, 172064, 200678, 200681, 200682; UMMZ 97049, 97067, 97128, 111480 (Verdigris R., 7 mi. NE. of Inde- pendence, Montgomery Co.), 111724, 120488, 120679, 120773 (Verdigris R., 1 mi. E. of Madison, Greenwood Co.), 122202, 122203, 160331, 160339, 160457 (Rose Cr., 2 mi. W. and 2 mi. S. of Wallace, Wallace Co.), 160504; UMMZ (Vermillion R., N. of Wamego, Pottawatomie Co.). Kentucky: UMMZ 126944, 168014, 168038; UMMZ (Delavan and Creaser nos. 31-86, 31-87) ; CNHM 6567; CU 48609, 48610; UL 5356, 5635, 5658, 5679, 5680, 5776, 5828, 6822, 7688, 8039; USNM 202495 through 202498. Micuigan: USNM 10595, 174908, 193197, 193206; UMMZ 55182, 56575, 64222, 64974, 65856 (Au Sable R., below Foote Dam, Iosco Co.), 73219, 79862 (North Br. Chippewa R., sec. 15, T. 16 N., R. 7 W., Mecosta Co.), 81513, 83390 (Ferris Cr., sec. 2, T. 10 N., R. 5 W., Montcalm Co.), 83864, 83910, 92207, 95007, 100786, 101729 (Saginaw Bay, off Katechay Island), 101895, 111328, 116256, 116291, 116327, 116362, 116881 (shore, L. Huron, sec. 26, T. 11 N., R. 16 E., Sanilac Co.), 116964, 117032, 117052, 131864 (Little Rabbit R., sec. 21, T. 4.N., R. 13 W., Allegan Co.), 131894 (Big Rabbit R., sec. 26 and 35, T. 4 N., R. 13 W., Allegan Co.), 136805, 137158, 137215, 137294, 137653, 137674, 138083, 138105, 138142, 138776 (Green Lake Cr., sec. 19, T. 4 N., R. 11 W., Allegan Co.), 162969, 164302, 164313, 164323, 164332, 164355, 165833, 165842; UMMZ (Kalamazoo River Survey nos. KA19, KA21, KA43, KA45; Hankinson no. 9412S). Mrinnesora: UMMZ 156699 (Cottonwood R.., sec. 22, T. 109 N., R. 33 W., Brown Co.). Missourr: USNM 36291 (Sac R., Greenfield, Dade Co.) ; INHS (Osage R., N. of Schell City; Niangua R., Bennett Spring Park, Dallas Co.; Salt R., N. of Shelbina, Shelby Co.; Grand CATFISH GENUS NOTURUS RAFINESQUE 119 R., 2 mi. SW. Sumner, Chariton Co.; Gasconade R., 4 mi. S. Gerome, Phelps Co.; Pomme de Terre R., 2 mi. SW. Hermitage, Hickory Co.; South Grand R., 2.5 mi. S. Prettyman, Cass Co.); UMMZ 142209, 142234, 142260, 147054 (Mis- sissippi R., Crystal City, about 1 mi. above mouth of Plattin Cr., Jefferson Co.), 148105, 148746, 148755, 148981, 148993, 148214, 149360, 149380, 149433, 150091 (Little Sac R., 0.5 mi. SW. Aldrich, Polk Co.), 150123, 150159, 150436 (Niangua R., 5 mi. SE. Buffalo, Dallas Co.), 150651, 150697, 150770, 150803 (Pomme de Terre R., 4 mi. SE. Eckton, Hickory Co.), 152547, 152621, 152639, 152674, 152750, 164577; UMoMZ (W. L. Pflieger coll.: From Missouri R.: P62-152, Rocheport; P63-48, Bonnots Mill; P63-119, at mouth; P63-130, Lexington. From Mississippi R.: P63-114, 2 mi. E. Fornfelt; P63-122, 3 mi. E. Winfield; P63-133, 2 mi. E. Illmo; P63-134, Neely’s Landing; P63-137, West Quincy; P63-138, Canton; P66-41, Grand Tower Island. P63-117, Locust Cr., Sullivan Co.). Montana: USNM 1482 (Yellowstone R., Fort Sarpy), 37780 (Big Horn R.), 129629 (Missouri R., Judith Ferry), 143088 (Musselshell R.); UMMZ 94433 (Tongue R., Miles City, Custer Co.). Nuwsprasxa: USNM 76124 (Long Pine); UMMZ 134043, 134053, 134555, 134968, 135026, 135074, 135136 (Arikaree R., 1 mi. NW. of Haigler, Dundy Co.), 135286 (Minnechaduza Cr., 3 mi. W. of Valentine, Cherry Co.), 135362, 135367; KU 4160; UMoMZ 5320. New York: CU 2153, 2327, 2460, 2641, 3063, 3551, 4610 (Olean Cr., 0.5 mi. N. of North Olean, Cattaraugus Co.), 4653, 5266, 5271, 5443, 6144, 6145, 8150, 8389 (Genesee R., above Wellsville, Allegany Co.); USNM 70012, 70014, 161948. Onto: USNM 1418, 11062, 62854, 62855, 69419, 69420, 69421, 118937, 161715; CNHM 6564, 9777, 9870-1; UMMZ 55669, 87714, 87738, 87795, 87806, 87816, 87835, 101635, 101651, 107566, 107669, 107760, 107782, 118354, 118406, 159834. OxLa- HOMA: Tulsa U (Illinois R., below Tenkiller Dam, 6 mi. NE. of Gore, Sequoyah Co.; Illinois R., U.S. Hwy. 59, Adair Co.; Neosho R., at Earbob Cr., Mayes- Wagoner Co. line; Caney R., Hulah Dam, Osage Co.) ; UMMZ 164576; TU 15535 (Verdigris R., Hwy. 88, Rogers Co.). Pennsytvania: CU 4188, 5597, 6717, 9423, 9435, 10718 (Allegheny R., 1.5 mi. E. of Port Alleghany, McKean Co.); UMMZ 102376, 102942; USNM 161750 thru 161755; UL 9287, 9325, 9353. Sourn Dakota: USNM 76125 (White R., Chamberlain). TrnnessrE: USNM 24839 (near Knoxville), 118938 (Rock Island); UMMZ 125706, 157444 (Roan Cr., 2.5 mi. ESE. of Doeville, Johnson Co.), 157576 (South Holston R.‘ 0.25 mi. above South Holston Dam, Sullivan Co.), 168261; Vanderbilt U (Turnbull Cr., Dickson Co.; Turnbull Cr., Cheatham Co.); TU 19491. Virarnta: USNM 190943 (South Fk. Holston R., Hwy. 91, 5.8 mi. SW. Lodi, Washington Co.). Wesr Virainia: CU 4537, 5734, 13434; USNM 192648 (Shavers Fk., 0.7 mi. NE. Turkey Knob bridge, Randolph Co.), 192649 (Shavers Fk., 3 mi. SW. Parsons, Tucker Co.), 192650. Wisconsin: UMMZ 64640 (Milwaukee R., Waubeka, Ozaukee Co.), 64836 (Milwaukee R., 1.5 mi. W. Newburg, Washington Co.), 64882 (br. Cedar Cr., 1 mi. W. Jackson, Washington Co.), 73660, 74118 (South Br. Little Wolf R., 11 mi. NW. Waupaca, Waupaca Co.), 76911, 76926, 76959, 77029, 77045, 77055, 77108, 77561, 78807 (Black R., 2 mi. N. Wither, Clark Co.); TU 15746 (Milwaukee R., West Bend, Washington Co.), 15762 (Newburg, Washington Co.). Wyomine: UMMZ 113443 (Wind R., Riverton, Fremont Co.), 159939 (Big Horn R., Basin, Big Horn Co.), 161890 (Piney Cr., 6 mi. W. of Ucross, Sheridan Co.), 161903 (Belle Fourche R., U.S. Hwy. 14, 2 mi. W. Carlile Junction, Crook Co.), 162440 (Little Goose Cr., 2 mi. S. Sheridan, Sheridan Co.); USNM 174932. CANADA: Qussec: UMMZ 136395 (Chateauguay R., first rapid above dam, near Montreal). 120 U.S. NATIONAL MUSEUM BULLETIN 282 Dracnosis.—Noturus flavus, the only member of the subgenus Noturus, differs from other species of the genus in the increased number of paired fin rays (usually 9 or 10 pelvic and 9 to 11, more often 10, soft pectoral rays), in the backward projections of the lateral edges of the premaxillary tooth band, in the relatively larger size attained, and in always having two free pectoral radials on each side. There are 54 to 67 caudal rays. It is distinguished from members of the subgenus Rabida by the reduced serrations of the pectoral spine and by the absence of a mottled pattern, from the subgenus Schilbeodes (except N. gilbertt) by the relatively broad cream colored margin of the caudal fin and the immaculate lower surface of the head and abdomen. It differs from N. gilberti in size, in the premaxillary tooth pattern, in the increased number of caudal rays, and in the longer spines. Description.—Counts and measurements are given in tables 17 to 26. Head depressed; lower jaw included; eye small, 2.5 to 4.7 times in snout; humeral process distinct, shorter than width of pec- toral spine; pectoral spine of moderate length, roughened or some- times with a few serrae behind; shallowly beveled or with recurved hooks along the anterior edge, but without anterior dentations as found in Rabida; dorsal spine stout; adipose fin continuous with caudal fin; caudal fin truncate or slightly rounded behind; premaxillary band of teeth with distinct posterior extensions which vary from a shallow V-shape to long, narrow trailing bands that may be nearly as long as the overall breadth of the premaxillary; lower lip and median under surfaces of head covered with coarse papillae. In both skeletons and cleared and stained specimens of Noturus flavus: ossified pectoral radials always two and unfused (48 sides); vertebrae anterior to anal origin 14 (in 10) or 15 (4). The soft dorsal rays are five (in 2), six (109), and seven (6). In the caudal fin there are (extremes in parentheses): (20) 22 to 26 (29), mean 23.9, upper simple rays; (15) 16 to 20 (22), mean 18.2, branched rays of which 7, frequently 8, are in the upper half and 9 to 12 are most usual in the lower half of the fin; (14) 16 to 21 (24), mean 18.1, lower simple rays. There are usually six or seven gill rakers. The size probably exceeds 250 mm. in standard length. Specimens ranging from 175 to 200 mm. are common in collections; many are above 200 mm. The Museum of Zoology has a specimen that is 240 mm. in standard length. General color in life yellowish, slate-gray, or olive green along sides and on dorsal surfaces. In preservative, top of head, areas about dorsal fin base, and between dorsal and adipose fins dark gray; ven- tral surface immaculate, including mental barbels, lower side of head, CATFISH GENUS NOTURUS RAFINESQUE kt abdomen, pelvic fin, lower edge of caudal peduncle, lower and upper margin and sometimes end of caudal fin, and edge of adipose fin. A gray blotch extends well into the adipose fin, the anterior end of which is immaculate; an irregular dark gray blotch covers most of middle of caudal fin and extends variably to the edge; sides of body yellowish gray; anal fin very lightly pigmented; dorsal fin with a light margin, a distal dark gray band, and an intermediate light gray area that grades into the dark gray base; pectoral fins dark gray except margin; upper edge of pectoral spine dark gray, front edge white; upper barbels gray; a light yellow elliptical or ovate spot at posterior end of dorsal fin. With age, lower surfaces become grayed with a diffuse dark pigment. Typr.—As is well known and was documented by Call (1899, p. 29), Rafinesque did not retain the specimens on which his descriptions were based and IJ have found none nor a reference that any of his specimens are extant. Because no type-specimen of Noturus flavus Rafinesque exists, a neotype of it is selected in order to fix that name to the species for which it has been in long and continued use. As discussed under Nomenclature (p. 127) several characters in the original and later descriptions suggest the possibility that Rafinesque may have described a composite of Noturus gyrinus and Noturus flavus—at least his counts of fin rays fit precisely the counts listed herein for gyrinus and most are outside the range of those obtained for flavus. Further, although the present Ohio River is unlike the river in the early nineteenth century, having been modified con- siderably by man, Noturus flavus is apparently rare below Pennsyl- vania, but Noturus gyrinus is moderately common (personal com- munication from Dr. William M. Clay, based on the extensive survey of the Ohio River Valley Water Sanitation Commission). Aside from flavus collected from the river bordering Illinois, I have been unable to examine material from this lower section of the stream. In his original description of flavus, Rafinesque indicated the locality as the Ohio. Later papers list the Falls of the Ohio as the only definite locality, and it is probable that his specimens were obtained there—hence I assume that to be the type-locality. The specimen selected as the neotype is the best that has become available from near this locality. It is typical of flavus in fin ray counts, number of vertebral elements, shape of the premaxillary tooth band, body shape, and color pattern as described and illustrated herein. Aside from the higher fin ray counts, a slightly smaller size, and Rafinesque’s color description, it corresponds well with his original account. It was collected with one other specimen of flavus, USNM 202495. The neotype is a male 114.2 mm. in standard length. The following data from it are not included in other summaries in this paper: I,6 122 U.S. NATIONAL MUSEUM BULLETIN 282 dorsal rays; 6+12=18 anal rays; 24+8+12+18=62 caudal rays; 1 epural; 3+3=6 hypurals, all free distally; 39 vertebrae of which 14 are anterior to the origin of the anal fin; 10 soft pectoral rays on the left side and 11, including a tiny inner ray, on the right side. On each side there are 10 pelvic rays, 2 internasal pores, and 11 preoperculomandibular pores. The head length is stepped into the standard length 3.7 times, the eye into the snout 3.2 times, and the distance from the rear end of the adipose fin to the tip of the caudal fin into the distance from the origin of the dorsal fin to the rear end of the adipose fin 1.8 times. Further measurements are given in table 27. VARIATION.—The following summary of the data obtained on this species is the result of combining several samples from various geo- graphic areas. It indicates relatively little geographic variation. The data listed consist of the number of counts, range (in parentheses), and the mean for, in sequence: (a) Missouri River, South Dakota and Nebraska, (b) Neosho and Illinois Rivers of Kansas and Okla- homa, (c) Osage River system, Missouri, (d) Missouri and Mississippi Rivers, Missouri, (e) Ohio and Cumberland Valleys, (f) Great Lakes basin, chiefly Huron River, Michigan, and (g) totals. Soft pectoral rays: (a) 68 (9-11) 9.68; (b) 40 (9-11) 9.98; (c) 5 (9-10) 9.60; (d) 52 (9-11) 9.62; (e) 40 (9-11) 9.80; (f) 79 (9-11) 9.85; (gz) 284 (9-11) 9.77... Pelvic rays: (a) 68 (8-10) 9.29; (b) 40 (9-10) 9.18; (c) 9 (9-10) 9.11; (d) 51 (8-10) 9.31; (e) 40 (8-10) 9.40; (f) 80 (8-10) 9.25; (g) 288 (8-10) 9.28. Anal rays: (a) 28 (15-18) 16.18; (b) 17 (15-17) 15.88; (c) 5 (16-17) 16.60; (d) 26 (15-18) 16.38; (e) 19 (15-19) 17.11; (£) 37 (16-18) 17.27; (g) 182 (15-19) 16.64. Lower-half caudal rays: (a) 19 (26-30) 28.16; (b) 20 (28-32) 29.75; (c) 5 (27-30) 28.60; (d) 26 (27-33) 30.08; (e) 18 (27-31) 29.28; (f) 36 (26-32) 28.67; (g) 124 (26-33) 29.15. Upper-half caudal rays: (a) 19 (29-36) 31.89; (b) 20 (30-33) 31.00; (c) 5 (30-33) 31.40; (d) 26 (27-34) 31.81; (e) 18 (29-32) 30.78; (£) 36 (28-33) 30.75; (2) 124 (27-36) 31.11. Total caudal rays: (a) 19 (55-66) 60.05; (b) 20 (58-64) 60.75; (c) 3 (60-63) 61.33; (d) 26 (55-67) 61.38; (e) 18 (57-63) 60.06; (f) 37 (54-64) 59.27; (g) 123 (54-67) 60.24. Vertebrae: Specimens from the southwestern portion of the range, lower Missouri and Arkansas River drainages, appear to have fewer vertebrae with modal numbers of 38 or 39; those from the upper Missouri have slightly more, and those from elsewhere the most, modally 39 or 40; French Cr., Pennsylvania 16 (38-40) 39.38; Ten- nessee R. system 25 (38-41) 39.80; Great Lakes drainage 44 (38-41) | CATFISH GENUS NOTURUS RAFINESQUE 123 39.48; upper Missouri R., Nebraska to Montana 15 (37-40) 38.93; Smoky Hill R., Kansas 5 (38) 38.00; Neosho R., Kansas 4 (38-40) 38.75; Missouri and Mississippi Rivers, Missouri 24 (37-40) 38.50; total 133 (37-41) 39.21. Specimens collected from the channels of the Missouri River throughout Missouri and the Mississippi River below the mouth of the Missouri River generally have smaller eyes than specimens from elsewhere. Although a few small eyed individuals have been obtained from the mouths of tributary streams, the small eye size seems to be restricted to the area of these highly turbid big rivers. Specimens from the Mississippi River above the mouth of the Missouri River have normal size eyes. The eye size is variable, but the eyes are usually obviously small. Occasional individuals in a sample have relatively large eyes, and in these samples a few appear to have eyes of intermediate size. The eye of flavus from the upper Mississippi River and other clear streams is usually stepped into the snout length 2.5 to 3.4 times. In 15 of the Missouri and Mississippi River specimens the eye is stepped into the snout 2.5 to 4.7, mean 3.6 times. Because the eyes of these specimens show considerable size varia- tion, especially within a small sample, from tiny to almost normal size, and because the specimens appear otherwise typical of flavus I am inclined to believe that the eye size is simply a response to the muddy, turbid river waters. Degeneration or loss of eyes within a few years has been observed in bullhead populations that have been subjected to reduced light or darkness. Further study of these or similar specimens should be made to compare the structure of the eye in detail with that of other specimens. Aside from the eye size these specimens do not appear to differ from other populations of flavus in their morphology. Their color is drab, usually medium gray in preservation, but the distribution of pigment does not differ from that of other flavus. Distrisution.—Noturus flavus (map 8) occurs in the Mississippi River system, in the Mohawk and Hudson River systems, and in the Great Lakes-Saint Lawrence River drainage. It has entered the Mohawk and upper Hudson River systems (Greeley and Bishop, 1933, pp. 89, 98; and Greeley, 1935, pp. 86, 96) from the Great Lakes, presumably as a canal immigrant. After the recession of Wisconsin ice, several entrances were made from the Mississippi drainage into the Great Lakes basin; one was undoubtedly into the Lake Michigan basin, another into the Lake Erie basin, and one was into the Lake Superior tributaries. In the Lake Michigan basin, flavus is confined to the area south of the Kalamazoo River in Michigan and to the southern half of Wisconsin and southward. 298-943 O—69——_9 U.S. NATIONAL MUSEUM BULLETIN 282 124 "sUOIEIYIQUSPISIUT UO Paseq 9q 0} Padatjaq JO UMOUY Joyje oie osueI poyeoIpul 94} episyno wo; sp10say *saq¥ys sure[d uso}saMm 94} Woy pue oseUIeIP JOATY UOspNyy 9Y} Wo} [eIOAVS apNpoUr pue ore sIyy UIYIIM []ey Sp1OIAI 9INJeII}] D1WUaY Ne Ajqeuinsoid nq paylaauy—) ‘a8ues [eoay,odAY pue UMOUy 9Y} spuNodns oul] poysep oy, *pouruexe useq savy sUsuIDads aoUaYyM sUOIe}s JUasaIdaI sapoIID ‘anbsouyey snany snangony JO UOIINqIIISIG—g AVIV eOll ( fx S371IW 40 31V9S eee he 002 oo| OS O EE CATFISH GENUS NOTURUS RAFINESQUE 125 From the Lake Erie basin, flavus has spread northward into Lake Huron to the Bruce Peninsula, Canada, and the Au Sable River, Michigan. It has become distributed throughout the Lake Erie and Ontario basins and is known to range eastward to Lake Champlain, tributaries of the Saint Lawrence River, near Montreal, and has been reported from the Etchemin River, Quebec (Paquet, 1964). N. flavus is found in western tributaries to Lake Superior (H. H. Moore and Braem, 1965), undoubtedly arriving from the St. Croix River, a tributary of the Mississippi River. Its spread eastward into Lake Superior is probably limited by cold water. Eddy, Moyle, and Underhill (1963) state that N. flavus is absent from the upper Mississippi River, Minnesota, above St. Anthony Falls, which is an apparent effective barrier to distribution. In the Mississippi River basin, flavus is found throughout most of the Ohio, Cumberland, Tennessee, upper Mississippi, and nearly the entire Missouri River systems. It has crossed from the Missouri, by way of the Osage or Kansas Rivers, into the Arkansas River system. It avoids streams of low gradient and is notably absent, for example, from southern Illinois and other areas of slow runoff. N. flavus has not yet been recorded from the South Platte River and apparently avoids much of the rest of the Platte; otherwise it is known from many of the upper Missouri River tributaries. In the Arkansas drainage, flavus is known from the Illinois, Neosho, and Verdigris River systems. Records of flavus from elsewhere are apparently based on misidentifica- tions. The following is a discussion of the more important of these. The inclusion of Texas in the range of flavus by Baughman (1950, p. 131) and by other recent writers in general references is an error initiated by S. Garman (1881, p. 89). His material, from San Antonio, Texas, was re-identified by Evermann and Kendall (1894, pp. 78, 96) as Ameiurus [= Ictalurus] natalis. A small specimen of Noturus gyrinus in the Museum of Comparative Zoology (MCZ 24900), collected by Edward Palmer in 1880 from San Antonio, Texas, mis- identified as Noturus flavus, was examined by me and is undoubtedly the basis of Garman’s report. No specimen or record of Ameiurus [Ictalurus] natalis obtained by Palmer from San Antonio has been found in that collection, suggesting that Evermann and Kendall did not examine Garman’s material. Records of Noturus flavus by Jordan and Gilbert (1886, p. 7) for Lee’s Creek and the Poteau River, near Fort Smith, Arkansas, are probably based on Noturus ezilis. These authors did not list ezvlis, @ species that is now known from Lee’s Creek and was recorded as Schilbeodes insignis by Cross and Moore (1952, p. 407) from many stations in the Poteau River, Oklahoma. In addition, a collection of Jordan and Gilbert’s paratypes of Noturus nocturnus from the Poteau 126 U.S. NATIONAL MUSEUM BULLETIN 282 River contained a specimen of ezilis. Similarly, Meek’s (1893, p. 229) records of Noturus flavus from Arkansas are probably referable to Noturus exilis. It is noted, for example, that Meek initially listed flavus but not ezilis from the Illinois River. Later, without comment, he listed exilis but not flavus from this system. This identification has been confirmed by examination of some of his material. Examination of the record (Forbes and Richardson, 1909, map 57) from the Kas- kaskia River, Illinois, reveals that it, too, is Noturus ezilis. The record from the Black River, Missouri (Martin and Campbell, 1954), is doubted, as the specimens on which the report was based cannot be found. Extensive surveys of Missouri and Arkansas fishes have not otherwise yielded specimens from any of the following rivers: Black, Current, White, and St. Francis. The Meramec River, Missouri, is included hypothetically in the range of flavus; there are no records. References to Noturus flavus from Manitoba by Bissett (1927, p. 127) and from the Hudson Bay drainage by Bajkov (1928, p. 97) and later compilations of these reports are listed here in the synonymy of Noturus gyrinus. Both reports were based on original material, but neither gave a description or listed specific localities, and presumably their material was not retained. It is of further interest that only Bajkov listed gyrinus, this on the basis of previous references. While there are many records of gyrinus, there are no other records of flavus from the drainage, and it is most probable that both are misidentifica- tions of gyrinus. This is in agreement with many recent writers who have questioned the authenticity of the records and conforms to an opinion expressed to me by Dr. W. B. Scott in a letter in which he out- lined the Canadian distribution of this species. In addition to Noturus exilis and N. gyrinus, some records of Noturus insignis and N. nocturnus have been listed as Noturus flavus or one of its synonyms. Young specimens are most frequently mis- identified. A comparison of the young of several similar forms appears in table 9. | NoMENCLATURE.— Rafinesque apparently gave two names to this species. They are: Noturus flavus, described from the Ohio (1818a, p. 41), and Noturus luteus, described later without statement of locality (1819, pp. 421-422). It appears that they were based on the same material; the descriptions, aside from being in different languages and having a few minor differences, are very similar and give the impres- sion that only one kind of animal is described. Evidence that Rafi- nesque knew only one species is that he recognized only Noturus flavus in later papers. The viewpoint that the names are based on the same animals agrees with Gill’s (1861a, p. 45; 1876, p. 423), and the type- localities are here assumed to be the same. Because his types were not CATFISH GENUS NOTURUS RAFINESQUE 127 preserved, the application of these names may have been subject to some uncertainty. Several species of Noturus, besides the one described here, occur at or near the Falls of the Ohio (Jordan, 1877d, p. 99; and Call, 1896, p. 14). Among them is Noturus gyrvnus, which may have been part of the material on which the name Noturus flavus was based. The other species of the genus that are known from the Ohio Valley seem to be eliminated from consideration in this question on the basis of Rafinesque’s descriptions. Confusion arises when it is realized that Rafinesque gave paired fin-ray counts that fit Noturus gyrinus pre- cisely and not flavus. Further, he described the color as entirely of a rufous yellow (1818a, p. 41) or as entirely yellowish (1819, p. 422), a character that could be assumed to be more applicable to gyrinus. Other parts of the description, however, better fit the form known under Rafinesque’s name. Characters that he listed which apply only to flavus as known and not to gyrinus are: the large size indicated (described simply as small in 1819, presumably in comparison with the larger ictalurids), the truncate tail, and the upper jaw longer. Although it is impossible to know whether he had some specimens of gyrinus mixed with flavus, or simply made errors in determining the fin formu- las (his fin-ray counts for some other species of catfishes are inaccu- rate), it appears that the description better applies to the form long known as Noturus flavus, and that the name must be restricted to that species on the basis of the characters listed above. This viewpoint is confirmed by an examination of Rafinesque’s unpublished drawing (notebook in the Smithsonian Institution) of Noturus luteus (= Noturus flavus), that illustrates a long, slender fish resembling either Noturus insignis or Noturus flavus but which on the basis of the de- scribed color could only be the form considered here as Noturus flavus. Gill (186la, pp. 45-46) named this species Noturus occidentalis on the basis of specimens from the Platte River, Nebraska, the identifi- cation of which is confirmed by his descriptions and figures (1876, pp. 423-424, pl. 8). I have been unable to locate any type-specimen of Noturus occidentalis. Giinther (1864, pp. 104-105) named the species anew, as confirmed by his figure of the ‘‘intermaxillary teeth,” after receiving specimens of another species that he thought to be the Noturus occidentalis of Gil. Giinther’s two syntypes of Noturus platycephalus in the British Museum are labled ‘Old Collection”; his description indicates only the locality, ‘“North America.”’ They are typical of N. flavus. Erymo.tocy.—The name flavus (Latin), yellow, was given in refer- ence to the general yellow color of Ohio specimens. RELATIONSHIP.—The relationships of Noturus flavus and the other species of Noturus are intimate. Noturus is recognized as a monotypic 128 U.S. NATIONAL MUSEUM BULLETIN 282 subgenus closest to some of the elongate forms in the subgenus Schilbe- odes, particularly N. gilberti, N. insignis, and N. nocturnus. ECOLOGICAL CONSIDERATIONS.—It is generally recognized that the distribution of flavus is controlled by stream gradient (for example it is absent from southern Illinois, map 8); it prefers riffles or rapids of moderate or large streams which usually have many large loose rocks. It is also known to be common in Lake Erie and Saginaw Bay where there may be a minimum of current, but much wave action. In feeding, it may work into quiet water. Although it prefers large streams I have taken large flavus in the High Plains from loose sub- merged limestone rocks in the outlet of a spring that was no more than five feet wide and remote from any large stream. Noturus flavus has been taken in collections with gyrinus, nocturnus, exilis, eleutherus, placidus, stigmosus, and miurus. The populations of Noturus flavus, particularly, are reduced in size or eliminated in sections of streams where large reservoirs have been constructed. Presumably this is because of loss of riffle habitat plus lowered water temperature. Although flavus is a relatively northern species of Noturus, it is very infrequently found in water cold enough to maintain salmonids. Subgenus Rabida Jordan and Evermann The subgenus Rabida, a very distinctive group, contains the species of Noturus with serrae (dentations) on both the anterior and the pos- terior edges of the pectoral spine (pl. 4). The anterior serrae are fine, numerous, and usually distinct in all but old specimens. Those of the posterior edge are large and distinctly curved; their tips, except the 1 to 3 basal serrae, are regularly turned toward the base of the spine. The pectoral spine may be short to very long and is usually curved backward, scimitar-like. The grooves of the spine are shallow to moderate in depth and short, mostly occupying the distal half of the spine. Other characteristics of Rabida are: body relatively short and usually chunky; lower jaw included; anal fin with 12 to 19 rays; head relatively large; vertebrae 30 to 39, usually 31 to 37; normally 11 preoperculomandibular pores in most species, but 10 in eleutherus and hildebrandi lautus; 9 pelvic rays except 8 in the hildebrandi group; usually 8 soft pectoral rays but 9 in albater and hildebrandi; caudal rays relatively short to moderate in length, the branched rays (12 to 24, usually 15 to 19) commonly less numerous than in Schilbeodes or Noturus; usually 12 or 13 vertebrae anterior to the first pterygiophore of the anal fin; precaudal vertebrae variable, mostly 7 to 10; caudal vertebrae extremely variable in number; 5 to 8 ribs on each side; branchiostegal rays typically 9 or 10 on each side; the ossified pectoral CATFISH GENUS NOTURUS RAFINESQUE 129 radials usually fused, but some species, chiefly Noturus stigmosus, vary in the amount of union of the two elements; all species typically have one epural and 3+3 hypurals, with variable fusion in each set of hypurals. The skeleton is not well ossified, perhaps because of the small size attained. The premaxillary teeth (pl. 2, fig. 4) occupy a short rectangular area which is about 1.5 to 2.5 times as broad as long, with rounded to obtusely angulate posterior corners. All the species are mottled or covered with irregular dark blotches or saddles. In young or small specimens the adipose fin is distinctively translucent or clear. The background color of the body may be pinkish, yellowish, or brownish, becoming darker with age. The included species have been uniformly listed as a unit in either the genus Noturus, or Rabida, or Schilbeodes; they have never been split into separate genera. However, when Rabida has been recognized, authors have tended to include in it one or more of the species in the subgenus Schilbeodes that have posterior serrations on the spine. Schilbeodes has thus been split unnaturally into two or more groups. With the exception of one that is found in Atlantic coastal streams of North Carolina, the species are restricted to tributaries of the Gulf of Mexico or have invaded the lower Great Lakes after recession of the Wisconsin ice. Most species, except Noturus miurus, seem to require a considerable stream gradient and are found at or near riffles. Noturus furwsus, the North Carolina representative, appears to be a segment of a once wide-ranging species which may have occurred in most preglacial rivers of eastern North America. N. stigmosus and N. miurus have crossed into the Great Lakes basin, probably by way of the Maumee outlet. N. munitus and N. miurus are perhaps recent invaders of Gulf tributaries lying to the east of the Mississippi River. Invasion of this area by N. miurus probably was a simple matter, because of its tendency to enter small and quiet waters. However, invasion by N. munitus (a river species) probably called for large quantities of flowing, fresh water such as a major shift in river drain- age, overflow by an overburdened Mississippi River, or by a fresh- water concentration in what is now the Gulf of Mexico adjacent to the mouths of the Mississippi and Mobile Rivers. Another possibility is that munitus has remained as a remnant of the former more widely distributed ancestor of the furiosus group. With these exceptions the species are confined to the Ohio, the Tennessee, the lower Mississippi, and the Red River drainages. N. miurus has crossed over into the upper Kaskaskia River, Illinois; otherwise no member of the subgenus has entered the upper Missis- sippi or the Missouri River systems. The forms are especially char- 130 U.S. NATIONAL MUSEUM BULLETIN 282 acteristic of upland or high-gradient streams where as many as three or possibly four species may occur together. The name Rabida, from rabidus (Latin), means mad, fierce, or furious. Key to the Species of the Subgenus Rabida 1. Pelvic rays typically 8; posterior process of cleithrum (humeral process) shorter than taamicber of pectoral spine shaft... .. LE ete ae, Pelvic rays typically 9; humeral process length variable, short or longer. 3 2. Soft pectoral rays 7 to 10, most frequently 9 but often 8; anal rays 12 to 17, usually 13 or more; preoperculomandibular pores nena 10, or 11; mouth inferior, lower jaw included. . . .. . . Noturus hildebrandi Soft pectoral rays 7 or 8; anal rays 12 or 13; preoperculomandibular pores 11; mouth approximately subterminal, lower jaw only slightly included. Noturus baileyi, new species 3. Preoperculomandibular pores typically 10; adipose fin nearly separated from caudal fin and forming a free posterior flap; caudal rays 39 to 52, usually fewer than 49; blotch on adipose fin low, confined to its basal half; no midcaudal crescent of dark pigment; soft pectoral rays typically 8; humeral process intermediate, longer than width of ae spine shaft, but never longer than shaft plus serrae. .. . . . . . . Noturus eleutherus Preoperculomandibular pores typically 11... .. sis Hees bse peace 4. Posterior process of cleithrum typically shorter fae ane of pectoral spine shaft, always shorter than shaft plus serrae; predorsal length stepped 1.4 to 2.1, usually more than 1.5 times in distanee from dorsal origin to rear end of adipose fin; adipose and caudal fins moderately to broadly united, the adipose fin not developing afree posterior flap. ...... 5 Posterior process of cleithrum (humeral process) typically longer than diame- ter of pectoral spine including its serrae; predorsal length stepped 1.1 to 1.6, usually 1.2 to 1.5 times in distance from origin of dorsal fin to rear end of adipose’ fin’) 4° 3)00i'.0¢ eMinidih f 87/ 5. Soft pectoral rays most frequently 9; a Sionnneht broad! date Ba at caudal peduncle base and distinct white areas usually on procurrent caudal rays; rear process of cleithrum very short or obscure; anterior dentations of pectoral spine poorly developed or nearly obscure; anal rays 13 to 16, ustially I4*or 15°27"... Sai trs Noterns albater, new species Soft pectoral rays 8 but equeady 9: bar, if present, on caudal peduncle no more prominent than medial feaes on caudal fin which is without large white areas; rear process of cleithrum present or obscure; anterior denta- tions of pectaral spine moderately or poorly developed. ... . yh ee 6. Posterior process of cleithrum present, acute, distinct; anal rays 14 to 19, usually 15 or more; vertebrae 34 to 37; dark pddlles on back broad and prominent; pectoral fin pigmented only near base and about spine; adipose fin with a dark bar. ..... ... . Noturus elegans, new species Cleithrum only roughened posteriorly, without a process; anal rays 13 to 16, usually 14; vertebrae 32 to 34; dark saddles on back narrow and not prominent; pectoral fin well pigmented; adipose fin yellowish, without a bart bet . . . . . Noturus trautmani, new species 7. A midcaudal erescenite one ee bar in addition to a dark brown sub- marginal caudal band; dorsal fin with a brown but never with a jet black blotch and caudal peduncle without a jet black bar; dark bar on adipose i ee CATFISH GENUS NOTURUS RAFINESQUE 131 fin not extending to the margin, or if to the margin the caudal rays are 52 or fewer (the furtosus species group) . . cde Sth A submarginal dark caudal band but no meee! preseeaaic bar, or if a bar is present, a more prominent dark bar crosses caudal peduncle; dark bar on adipose fin always extending to margin; dorsal fin with a distal black blotch (except flavipinnis); caudal rays 51, typically 53 or more (the miurus species group) . . ee 8. Adipose blotch dusky, pone aly aca half oF fal nated ae 50 to 59, usually 52 to 58; no distinctive, discrete, round, peeenee chromatophores on abdomen; anterior serrae of pectoral spines poorly developed, or reduced and Grepulan ea ey a . . . . . Noturus placidus, new species Adipose blotch dark Broa enone upper half of fin in juveniles and adults; anterior serrae of pectoral spine prominent, sharp, and well developed . 9 9. Adipose blotch extending to fin margin; caudal ravs 45 to 52, usually 46 to 50; vertebrae 30 to 33, usually 31 or 32; lower head, abdomen, and lower base of pelvic fins sprinkled with numerous small, rounded, discrete, brown chromatophores; head eee aay 2.7 to 3.3 times in standard length... . . . . Noturus munitus Adipose blotch Hob senohine fin Sate eaedal ee 47 to 60, usually 49 to 58; vertebrae 31 to 36, usually 33 to 35; head length stepped 3.1 to 3.8 ness instandard length. ..... Nie eee re ek) 10. Caudal rays usually 49 to 53; only anterior part lor abdomen with Hintinen round, brownish chromatophores in smaller individuals, these often obscured in large individuals; a pair of large light spots, typically en- closed in dark saddle, just anterior to dorsal fin base. Noturus stigmosus, new species Caudal rays usually 53 to 58; no distinctive round, brownish chromatophores on abdomen; dark saddle at dorsal fin nearly uniform anterior to dorsal fin base, without a pair of light spots. . .. . . . Noturus furiosus 11. Anterior ends of infraorbital and supraorbital canals typically connected (1 internasal pore); a jet black blotch distally on first 3 or 4 dorsal rays; adipose and caudal fins moderately connected ; basicaudal bar seldom prom- inent on caudal peduncle, usually confined to posterior edge of peduncle and base of caudalrays .. . . . . . . Noturus miurus Anterior ends of infraorbital and poeeceniel canals typically distinct (2 internasal pores); basicaudal bar extending rather broadly across caudal peduncie') (033.0 |: ee aaseee Ty 12. Dorsal fin mostly mellawinh, wit frie Pana pigment, alaean a prominent black blotch; adipose and caudal fins moderately connected; basicaudal brownish bar extending vue, across caudal peduncle where it widens into a blotch : . . . . . . Noturus flavipinnis, new species A jet black blotch ee on aaa rays; adipose fin nearly free from caudal fin; basicaudal bar jet black, of subuniform width, broadly crossing caudal peduncle. ....... .. .. .. .Noturus flavater, new species The HILDEBRANDI Group Noturus hildebrandi (Bailey and Taylor) LEAST MADTOM Puatss 4 (Figs. 1-2), 10; Mar 9 Diaenosis.—Noturus (Rabida) hildebrandi with typically eight pelvic rays, a short caudal fin, and reduced anterior serrae of the pectoral spine plus a very short posterior process of the cleithrum 132 U.S. NATIONAL MUSEUM BULLETIN 282 (humeral process) is associated with Noturus baileyi in the hildebrandi species group. Both are readily distinguished from other species of the subgenus Rabida by the presence of only eight pelvic rays. In contrast to Noturus baileyi, there are typically eight or nine, more often nine, instead of seven or eight soft pectoral rays, more anal rays (usually 13 to 15), and the lower jaw is definitely included instead of being almost subterminal. The internasal pores are typically two on each side but the preoperculomandibular pores, normally ten or eleven, vary geographically. Description.—Counts and measurements are given in _ tables 17 to 26. Head depressed; lower jaw prominently included; lip rather thick; eye moderate, about 1.4 to 2.5 times in snout; posterior process of cleithrum very short or obscure, blunt to sharp pointed; pectoral spine small, relatively straight (pl. 4, figs. 1, 2); its anterior serrae small or obscure; its posterior serrae, except the proximal one or two, recurved toward the spine base; dorsal spine small and stiff; adipose fin of moderate height and length, without a free posterior flap, broadly connected with the rather short procurrent caudal rays; caudal fin truncate or with a slightly rounded posterior margin; posterior corners of premaxillary tooth patch obtuse. The largest specimen examined is one typical of the southern subspecies, from the Homochitto River, 53 mm. in standard length. Typical specimens of the northern subspecies reach 48 mm. in standard length. The soft dorsal rays are five (in 4), six (156), or seven (2). The caudal fin has (extremes in parentheses): (14) 16 to 18 (20) upper simple rays; (13) 15 or 16 (17) branched rays of which (5) 7 (9) are in the upper half of the fin and (6) 8 or 9 (10) are in the lower half; and (11) 13 to 15 (17) lower simple rays. There are four to eight gill rakers on the first arch. In six cleared and stained specimens, 12 (in 5) or 13 (1) vertebrae are anterior to the origin of the anal fin; the ossified pectoral radials are fused; and the six hypurals show variable fusion distally; hypurals 2-3 were fused in four specimens from the Homochitto drainage; hypurals 2-3 and 4—5 were fused in two from western Tennessee. The number of caudal vertebrae may also vary as follows: Homochitto drainage 25 (2), 26 (1), or 27 (1); western Tennessee 28 (2). RELATIONSHIP.—The characters listed in the diagnosis are regarded as indicating, at least tentatively, the relationship of Noturus hilde- brandi and Noturus baileyi in the hildebrandi species group. Discusston.—A study of the geographic distribution and char- acteristics of populations of Noturus hildebrandi has revealed a situation that appears to be unique among North American fishes. No other eastern North American fish is known that has its range CATFISH GENUS NOTURUS RAFINESQUE 133 confined to the relatively small eastern tributaries to the Mississippi River in Mississippi, Tennessee, and probably Kentucky. Two rather uniform subspecies are recongized. They are most readily distinguished by color pattern and number of preoperculo- mandibular pores. An apparent change in a number of other char- acters from north to south points to a cline, but because of a prom- inent shift, in northern Mississippi and southwestern Tennessee, in pore counts and color pattern, with relative uniformity of popula- tions elsewhere, subspecies are recognized. The synonymy and discussion pertaining to the subspecies will be found with the descriptions of the two forms. The geographic variation of Noturus hildebrandi is discussed separately to facilitate comparison. GEOGRAPHIC VARIATION.—No specimens are known from the Big Black River system, Mississippi, and, aside from the Coldwater River specimens, only five relatively small poor specimens are available from the Yazoo system. Otherwise samples are generally adequate to indicate variational trends. No significant variation in the number of caudal rays, branched and simple caudal rays, dorsal rays, or internasal pores has been noted in any population. Pelvic rays: The pelvic rays appear to be almost constantly eight in all populations. Very few counts of nine rays on one or both sides of a specimen were obtained; they appear randomly throughout the range. The variation in vertebrae, preoperculomandibular pores, and anal rays is shown in tables 10 to 12. The preoperculomandibular pores in table 11 are the sum of the counts obtained from the two sides of each specimen. In the following summaries the data presented include number of specimens, range (in parentheses), and mean for (a) Homochitto River system, (b) Bayou Pierre, (c) Yalobusha system, (d) Yocona system, (e) Tallahatchie system, (f) Coldwater River, (g) Loosahatchie River, (h) Hatchie system, (i) South Forked Deer system, and (j) Obion system. Preoperculomandibular pores: (a) 129 (10-12) 10.87; (b) 50 (10-12) 10.84; (c) 2 (11) 11.00; (d) 2 (11) 11.00; (e) 6 (10-11) 10.67; (f) 18 (9-11) 9.67; (g) 4 (9-10) 9.50; (h) 48 (9-11) 10.06; (i) 4 (10) 10.00; (j) 58 (9-11) 9.84. Upper-half caudal rays: (a) 65 (22-27) 24.11; (b) 25 (21-24) 23.04; (c) 1 (24) 24.09; (d) 1 (24) 24.00; (e) 3 (23-25) 24.00; (f) 9 (21-26) 23.56; (g) 2 (22-23) 22.50; (h) 25 (22-26) 24.00; (4) 2 (24-26) 25.00; (j) 29 (21-26) 23.76. Lower-half caudal rays: (a) 65 (20-24) 22.03; (b) 25 (21-24) 22.40; (c) 1 (23) 23.00; (d) 1 (22) 22.00; (e) 3 (22-23) 22.33; (£) 9 (19-23) 134 U.S. NATIONAL MUSEUM BULLETIN 282 21.22; (g) 2 (19-22) 20.50; (h) 25 (20-25) 22.20; (i) 2 (23-24) 23.50; (j) 29 (20-25) 22.52. Total caudal rays: (a) 65 (44-51) 46.14; (b) 25 (438-48) 45.44; (c) 1 (47) 47.00; (d) 1 (46) 46.00; (e) 3 (46-47) 46.33; (f) 9 (40-48) 44.78; (g) 2 (41-45) 43.00; (h) 25 (43-50) 46.20; (i) 2 (47-50) 48.50; (j) 29 (42-50) 46.28. Head length (expressed as thousandths of standard length): (a) 40 (281-333) 306.4; (b) 11 (268-305) 287.0; (c) 1 (279) 279.0; (d) 1 (305) 305.0; (e) 2 (279, 304) 291.5; (f) 9 (294-340) 309.1; (g) 2 (275, 293) 284.0; (h) 7 (260-277) 271.0; (i) 2 (264, 277) 270.5; (j) 28 (245-293) 264.2. Soft pectoral rays: (a) 129 (8-10) 8.69; (b) 49 (8-9) 8.76; (c) 2 (8) 8.00; (d) 2 (9) 9.00; (e) 4 (9-10) 9.50; (f) 18 (8-9) 8.39; (g) 4 (8-9) 8.75; (bh) 50 (8-10) 8.96; (i) 4 (8-9) 8.50; (j) 58 (7-9) 8.40. Head length: Typical hildebrandi have a relatively large head and the populations from Tennessee have a smaller head. The variation in head size is irregular and difficult to evaluate since the head length appears to be slightly greater in relation to standard length in juveniles. The head lengths plotted against standard lengths (fig. 3) indicate a merging of individuals smaller than 30 mm. in standard length in all populations. Above this size, all specimens from Tennessee show no overlap with those from the Homochitto and Coldwater Rivers. Specimens from Bayou Pierre are somewhat intermediate with a few individuals falling within the plotted area of the Tennessee specimens. The specimen from the Yalobusha system falls within the area of the RIVER SYSTEM: 18 = HOMOCHITTO BAYOU PIERRE YOCONA, YALOBUSHA, AND TALLAHATCHIE COLDWATER 14 LOOSAHATCHIE HATCHIE AND FORKED DEER Head Length In MM. ny 35 40 45 50 Standard Length In MM. Ficure 3.—Head length plotted against standard length, as a comparison of populations of Noturus hildebrand1. CATFISH GENUS NOTURUS RAFINESQUE 135 Tennessee specimens; all others from the Yazoo are outside this range. Thus, aside from the Bayou Pierre specimens and one from the Yalobusha there is no overlap in head length plotted against standard length in larger individuals. Color pattern: All Mississippi specimens have a pigmentation pat- tern on the body and head generally similar to the typical hildebrandi in the Homochitto system. It consists of prominent dorsal saddles, and extension of the pigment to the lower side of the body. In the Homochitto and Bayou Pierre specimens a blotch or bar extends well into the adipose fin, usually to the margin in larger individuals. In Tennessee specimens, the blotches or saddles are not prominent; the adipose fin is yellowish without a blotch but often with small, scattered chromatophores; heavy pigment is confined to the very base of the fin. Also in Tennessee specimens the lower side of the body is immaculate or with a few large, scattered chromatophores. Specimens from the Loosahatchie River and, less so in the Hatchie system, show more pigment on the lower side than do specimens from the Forked Deer system, yet they are very similar in color pattern to those from the Forked Deer system. The Shones Lake specimen is small with little pigment ventrolaterally and a short blotch extending into the other- wise clear adipose fin. Specimens from the Coldwater River, Missis- sippi, strongly resemble those from the Homochitto system in boldness of the blotching and darkness of pigmentation, but the lower side is not as well pigmented; the adipose fin is yellowish without a blotch except two individuals have a very low intrusion of pigment into the fin. The other four specimens from the Yazoo have a body pattern resembling typical hildebrandi but there is little pigment in the adipose fin except in the Yalobusha specimen which has a short blotch. INTERGRADES.—The following material is regarded as intermediate between Noturus hildebrandi hildebrandi and Noturus hildebrandi lautus. UNITED STATES: Mussissipp1: Yazoo River system: USNM 175392 (1 specimen, 48 mm. standard length), creek, 1 mi. S. Coffeeville, Yalobusha Co. USNM 175393 (1, 50), Pumpkin Cr., 6 mi. SE. Oxford, old Hwy. 6, Lafayette Co. USNM 175394 (1, 30) and University of Mississippi (1, 46), trib., Puss Cuss Cr., 12 mi. NE. Oxford, just N. Hwy. 30,-Lafayette Co. USNM 175395 (1, 33), Shones Lake (Snow Cr.), E. Holly Springs or about 4 mi. W. Ashland, Hwy. 4, Benton Co. CU 42162 (9, 34-50), Coldwater R., U.S. Hwy. 78, 5.5 mi. SE. Olive Branch, DeSoto Co. TENNESSEE: Loosahatchie River system: USNM 197399 (1, 49) and USNM 193476 (1, 43), Loosahatchie R., Hwy. 76, due N. Somerville, Fayette Co. The short, heavy bodied specimens from the Coldwater River in extreme northern Mississippi are most intermediate with preoperculo- mandibular pore counts of Jautus but head lengths agreeing with hildebrandi. The numbers of vertebrae and anal rays are also sugges- 136 U.S. NATIONAL MUSEUM BULLETIN 282 tive of hildebrandi but the color pattern includes some characters of both forms. The two slender bodied specimens from the Loosahatchie River, Tennessee, tend strongly toward typical /awtus in most char- acters but are considered intermediate because of the presence of a short blotch on the adipose fin and considerable extension of pigment onto the lower side. The remaining five specimens from the Yazoo system, Mississippi, are relatively short bodied with head lengths, pore counts, and pigmentation patterns intermediate or tending strongly toward hildebrandi but the anal ray count is relatively high as in lautus. Thus, aside from the Coldwater River specimens, the characters of the few known individuals from the Loosahatchie and Yazoo systems rather closely approach those of the adjacent sub- species. Noturus hildebrandi hildebrandi (Bailey and Taylor) Puates 4 (Fic. 1), 10 (Fie. 1); Mar 9 Schilbeodes hildebrandi Bailey and Taylor, 1950, pp. 31-38, pls. 1-2, figs. E-F (original description; Brushy Cr., Amite Co., Miss.; ecology).—Cook, 1953, pp. 192, 195 (Mississippi)—G. A. Moore, 1957, pp. 148, 145, figs. 2-79C, 2-81 (description; range). Noturus hildebrandi (Bailey and Taylor).—Eddy, 1957, p. 154 (key; range).— Taylor, 1957, p. 192.—Cook, 1959, pp. 34, 37, 135, 148, figs. 25A-B (descrip- tion; Mississippi records).—Suttkus, 1961, p. 63, fig. 5 (comparison; skull illustrated).—Suttkus and Taylor, 1965, pp. 171, 175 (comparison).—Raney and Suttkus, 1966, p. 101 (associations with Etheostoma rubrum, Bayou Pierre, Miss.). TYPE-SPECIMENS.—UMMZ 157620 (holotype), UMMZ 155338 (21 paratopotypes), and USNM 112066 (4 paratopotypes), collected from Brushy Creek, 1 mi. above its mouth in Homochitto River, at village of Homochitto, 4 mi. NE. of Coles, sec. 14, T. 4 N., R. 2 E., Amite Co., Mississippi, April 1, 1948, by Reeve M. and Marian K. Bailey. OTHER MATERIAL STUDIED UNITED STATES: Mississippi: Homochitto River system: USNM 172031 (16 specimens, 29-45 mm. standard length), Brushy Cr., Homochitto, Amite Co. TU 2915 (13, 29-51), TU 11990 (4, 41-45), TU 19774 (7, 40-52), TU 19846 (2, 46-49), trib., Homochitto R., Lucien, U.S. Hwy. 84, Franklin Co. TU 19995 (1, 28), Homochitto R., 3.6 mi. SE. Meadville, Hwy. 98, Franklin Co. TU 23897 (23, 31-49), Homochitto R., 4.9 mi. E. Union Church, Hwy. 550, Lincoln Co. TU 23961 (14, 39-53), Homochitto R., 0.4 mi. E. Lucien, U.S. Hwy. 84, Franklin Co. Bayou Pierre system: USNM 200380 (11, 31-47), Bayou Pierre, about 5 mi. SW. Utica Institute, Hwy. 18, Copiah Co. FSU 9226 (6, 23-28), Bakers Cr., 2.2 mi. SE. Port Gibson, Hwy. 547, Claiborne Co. FSU 9265 (8, 22-40), Bayou Pierre, 8.7 mi. SW. Utica, Hwy. 18, Copiah Co. Diaenosis.—Noturus hildebrandi hildebrandi differs prominently from Noturus hildebrandi lautus in having a larger head stepped 3.0 to 3.7, usually 3.1 to 3.5 times in standard length; modally eleven pre- CATFISH GENUS NOTURUS RAFINESQUE 137 operculomandibular pores; and a prominently blotched color pattern with the pigment extending across the lower side of the body almost to the ventral surface. DerscriptTion.—Typical specimens are short and chunky, the abdominal region and head being relatively large and heavy. In contrast to Jautus the caudal region may be short (note the 25 to 27 caudal vertebrae in cleared and stained specimens) and the numbers of vertebrae and anal rays average fewer (tables 10 and 12). General background coloration yellowish or straw color; side and dorsal surface with dark brown chromatophores forming rather prominent saddle-like blotches, alternating with lighter obovate areas on the back and upper side; midside of body rather densely pigmented; pigment on lower sides slightly lighter and less dense or lower side strongly variegated; first dorsal saddle beginning just before the dorsal fin and extending backward to the third dorsal ray; a second, about as long, lies midway between the dorsal and adipose fins; a third slightly longer, at the middle of the adipose fin extends variably, usually to the margin in adults, onto that fin, which is otherwise nearly immaculate; an irregular bar crosses the upper and lower procurrent rays and the rear portion of the caudal peduncle. Caudal fin dusky, without obvious light areas except rear margin; submarginal dark brown caudal band broad, becoming diffuse ante- riorly; also present, a short medial, narrow vertical crescent of dark brown pigment or intermediate band; soft dorsal fin immaculate except a few scattered chromatophores distally and dark brown ba- sally; membrane over dorsal spine heavily pigmented, the heavy dark brown pigment extending from near the spine tip diagonally downward to the base of the third soft ray; bases of rear dorsal rays with little pigment. A few irregularly placed chromatophores distally on longer anal rays and scattered over the pectoral fin, usually on the longer rays, but covering the pectoral spine; pelvic fins immaculate; undersurface of head (except lightly pigmented area in front of barbels), abdomen, and inner mental barbels unpigmented; outer mental barbels sometimes with a few brown chromatophores; upper barbels with dense dark brown pigment; top of head dark brown; a dark bar crosses the back of the head, extending onto the operculum and upper branchiostegal membrane; another, below the eye, ex- tends forward to the nares; a light area behind the eye has fine chromat- ophores; the cheek has relatively little pigment. Typr.—The holotype, UMMZ 157620, a female, is 41.3 mm. in standard length. It has 13 anal rays, 17+7+9+12=45 caudal rays, and 6 soft dorsal rays. On each side there are eight pelvic rays, nine soft pectoral rays, five serrae on the posterior edge of the pectoral spine, two internasal pores, eleven preoperculomandibular pores, and 138 U.S. NATIONAL MUSEUM BULLETIN 282 six gill rakers. The head length is stepped into the standard length 3.25 times and the distance from the adipose notch to the tip of the caudal fin is stepped into the distance from the dorsal origin to the adipose notch 1.8 times. Table 28 gives further measurements. ErymMo.tocy.—This subspecies was named hildebrandi in honor of the late Dr. Samuel F. Hildebrand, an outstanding American ichthy- ologist. Distrisution.—Noturus h. hildebrandi ranges from the Homochitto River system northward in Mississippi. It intergrades with Noturus hildebrandi lautus in northern Mississippi and southwestern Tennessee. The populations in the Homochitto and Bayou Pierre are relatively uniform and are regarded as typical. Intergrades from the Yazoo drainage in northern Mississippi vary toward Noturus hildebrandi lautus, but those from the southern portion of the system closely approach typical hildebrand: morphologically. Eco.toey.—In southern Mississippi Noturus hildebrandi hildebrandi is typically found in relatively shallow, clear, gravelly or small pebble- bottomed riffles with moderate to swift current. It has been collected frequently with Noturus miurus. Noturus hildebrandi lautus, new subspecies PuaTEs 4 (FIG. 2), 10 (FIG. 2); Mar 9 Types.—USNM 201665 (holotype), USNM 193470 (16) and UMMZ 187093 (4) (totaling 20 paratopotypes, 36-47 mm. standard length), collected from North Fork of Obion River, at Tennessee Highway 69, Henry County, Tennessee, July 27, 1954, by C. E. Ruhr and party. OTHER PARATYPES UNITED STATES: Trnnessee: Forked Deer River system; Obion drainage: USNM 197395 (4, 24-44), North Fork Obion R., Campground Levee Crossing, Weakley Co., September 15, 1954, C. E. Farrell. USNM 197402 (2) and UMMZ 187092 (2) (totaling 4, 35-48), Middle Fork Obion R., Hwy. 6, Gleason Crossing, Weakley Co., September 16, 1954, C. E. Farrell. KU 9751 (2), Middle Fork Obion R., 4.5 mi. W., 1 mi. 8. Paris, Henry Co., April 5, 1965, F. B. Cross. KU 10036 (3), Middle Fork Obion R., Hwy. 22, Weakley Co., September 7, 1964, Cross. OTHER MATERIAL STUDIED UNITED STATES: TENNEssEE: Hatchie River system: USNM 190746 (18, 18-44), Moss Cr., on Rose Creek Road, 1.3 mi. N. Hwy. 57 or about 2.5 mi. ENE. Pocahontas, McNairy Co. USNM 190775 (6, 15-42), Spring Cr., 1.2 mi. S. Bolivar, Hwy. 125, Hardeman Co. USNM 193464 (1, 40), Moss Cr., about 4 mi. W. Selmer, McNairy Co. Vanderbilt U (1, 32), Big Black Cr., Madison Co. South Forked Deer River system: USNM 193479 (2, 42-44), Deer Cr., Huron, Henderson Co. Draenosis.—Noturus hildebrandi lautus differs from Noturus hilde- brandi hildebrandi in having a relatively short head stepped 3.4 to 4.1, usually 3.6 to 3.9 times in standard length; modally ten preoper- CATFISH GENUS NOTURUS RAFINESQUE 139 iS K BAILEYI ‘4 @ Norurus HILDEBRANDI HILDEBRAND! PX) @ NOTURUS HILDEBRANDI: HILDEBRANDI X LAUTUS O Norturus HILDEBRANDI LAUTUS } Vel a \ . Ses Ss 25 0 25 50 100 Map 9.—Distribution of members of the hildebrandi species group of the subgenus Rabida. The symbols represent material examined. culomandibular pores; and the lower side of the body is cream colored, with little pigment. Description.—Although the head of lautus is proportionately short, the body is relatively elongate and slender. The elongation seems to be caudad, as noted in the presence of 28 caudal vertebrae in two cleared specimens, and is reflected in the slight increase in number of anal rays and vertebrae (tables 10 and 12). General background color yellowish with reddish brown super- imposed. Blotches or saddles on back often not prominent, separated by rounded lighter areas; midside of body, down to approximately the axial streak, densely reddish brown; lower side at or from just below axial streak rather abruptly yellowish with only scattered or no brown pigment except a small area just behind pectoral fin and at rear end of caudal peduncle. The general location of the dorsal saddles 298-943 O—69——10 140 U.S. NATIONAL MUSEUM BULLETIN 282 and the distribution of pigment in the dorsal fin of hildebrandi and lautus are similar, except that the saddle-like blotch below the adipose fin does not enter that fin in dautus; the adipose fin is yellowish and immaculate or sometimes with small brown chromatophores scattered throughout; bar at rear end of caudal peduncle similar in shape to that of hildebrandi but ending near bases of the upper and lower procurrent caudal rays—these rays mostly immaculate distally; caudal fin yellowish with one or two narrow, short, brown bars distally and a wider brown bar near base; much of the upper and lower portions of the fin often immaculate; anal fin immaculate or with a few chromatophores distally on posterior long rays; pelvic fins, lower barbels, abdomen, and lower surface of head immaculate; pectoral spine brown with some pigment on adjacent soft rays; head dark reddish brown above with a bar across its reac end to the lower margin of the opercle; a similar band extending forward from beneath the eye to the nares; a small round yellowish spot usually present behind the eye; cheek yellowish; upper barbels heavily pigmented. Ho.totyrpe.—The holotype (pl. 10, fig. 2) is a female, 40.3 mm. in standard length. It has 15 anal rays, 16+7+8+15=46 caudal rays, and 6 soft dorsal rays. The left pectoral fin has eight soft rays and the right has nine including a tiny basal splint. On each side there are eight pelvic rays, ten preoperculomandibular pores, two internasal pores, and seven recurved serrae on the posterior edge of the pectoral spine. The anterior edge of the pectoral spine is roughened; the serrae are greatly reduced in size and indistinct. The head length is projected into the standard length 3.8 times, and the distance from the rear end of the adipose fin to the tip of the caudal fin into the distance from the dorsal origin to the rear end of the adipose fin 2.0 times. Further measurements are given in table 28. Distripution.—Noturus hildebrandi lautus is known only from western Tennessee but it may range into southwestern Kentucky and into northern Mississippi in the Hatchie River system. It inter- grades with Noturus hildebrandi hildebrandi in southwestern Tennessee and northern Mississippi. Erymo.tocy.—The name lautus (Latin) meaning washed or clean and neat refers to the trim, neat, and pleasing color pattern of this subspecies. Ecotocy.—In the Obion River, Tennessee, Jawtus was taken in water 12 to 18 inches deep, over a bottom of shifting sand with some gravel up to one inch in diameter. In the Hatchie system it lives in highly turbid, but perhaps seasonally clear, moderately deep streams with slow to moderate current. Where collected in this system little riffle habitat was available; the bottom consisted of shifting sand and debris with occasional gravel or of mud, sand, and silt with CATFISH GENUS NOTURUS RAFINESQUE 141 considerable debris such as sunken logs and brush. It has been col- lected frequently with Noturus phaeus and Noturus miurus, and sometimes with Noturus stigmosus. Noturus baileyi, new species SMOKY MADTOM PuaTEs 4 (Fig. 3), 11 (Fie. 1); Mar 9 Schilbeodes miurus (Jordan) [misidentification]—Lennon and Parker, 1959, pp. 3, 15 (lower Abrams Cr., Great Smoky Mountains National Park by poison- ing, June 6 to 9, 1957; used in toxicity test). Noturus miurus Jordan [misidentification].—Lennon, 1962, p. 6 (rare, Abrams Cr.). TYPE-SPECIMENS.—USNM 201602 (holotype), USNM 201601 (8 paratopotypes), and UMMZ 187096 (1 paratopotype), collected from the lower portion of Abrams Creek, Great Smoky Mountains National Park, June 8, 1957, by Eugene W. Whitney and party. Diacnosis.—Noturus (Rabida) baileyi is a member of the hilde- brandi species group, which is characterized by typically eight pelvic rays, a short caudal fin, short to almost obscure anterior serrae of the pectoral spine, and a very short posterior process of the cleithrum (humeral process). The color pattern of baileyi, in preservation, is somber, more so than in any other species of Rabida. The back and side of the head and body are almost uniformly pigmented, without blotching and the dorsal saddles are small. Like the typical southern subspecies of Noturus hildebrandi there are 11 preoperculomandibular pores and the head is large but generally unlike hildebrandi, the anal fin is very short with 12 or 13 rays and there are only 7 or 8, modally 8, soft pectoral rays. Description.—Counts and measurements are given in tables 17 to 23 and 25 to 26. Head relatively large, rounded above, only slightly deeper than caudal peduncle; greatest body depth beneath dorsal fin; caudal peduncle tapering slightly posteriorly, not prominently con- stricted; mouth subterminal, with lower jaw only slightly included, less so than in any other species of Rabida; eye moderate, 1.8 to 2.1 times in snout; pectoral spine only slightly curved, relatively straight for a Rabida; anterior serrae fine, numerous; posterior serrae moder- ately large, recurved toward spine base, four or five in all specimens; dorsal spine stout, relatively short, about 2/5 length of longest dorsal ray; posterior process of cleithrum short, acute, its length less than half diameter of pectoral spine shaft; premaxillary teeth in a small rectangular cross band with posterior corners smoothly rounded; adipose fin higher than the short anterior upper procurrent caudal rays, without a notch between them; caudal fin of moderate length, nearly truncate posteriorly; barbels all very short. 142 U.S. NATIONAL MUSEUM BULLETIN 282 The soft dorsal rays are six in all specimens. The caudal fins of these specimens all have 15 branched rays with 7 rays in the upper half and 8 in the lower half. The upper simple caudal rays are 16 (in 3), 17 (1), or 18 (1) and the lower simple caudal rays are 13 (3) or 14 (2). The following branchiostegal counts from the right side of specimens were made without staining or dissection of the elements and thus may be low: 9 (in 1) or 10 (2); the gill rakers on the first arch range from 5 to 8; as judged from x-rays the 6 hypurals are variably fused or lack fusion distally. The individuals range from 42.7 to 49.4 mm. in stand- ard length. The specimens are all of a medium brown color (perhaps faded) and without prominent saddles or blotches but moderate blotches as follows: top of head along midline and an area bounded by the pos- terior margins of the eyes and anterior nares darker brown; a narrow dark saddle, about as wide as eye length, extending from just anterior to dorsal spine back to the third or fourth dorsal ray; another about as wide, but lighter, extends from the dorsal fin to the adipose fin; and a very narrow dark saddle at the base of the adipose fin projects as a faint dark area into the adipose fin, perhaps to the margin; faint yel- lowish areas on back at rear base of dorsal fin, and at both the anterior and posterior bases of the adipose fin. Body and head otherwise appear- ing medium brown, lower surfaces lighter. The medium brown effect resulting from numerous rather large brown chromatophores over a yellowish background; large prominent dark brown or black chromato- phores scattered on side of head and cheek; abdomen, lower head, and pelvic fins yellowish, immaculate. Faint brown pigment on rays, at least distally, of dorsal fin, anal fin, and upper side of pectoral fins; dorsal and pectoral spines and dorsal rays, near base, rather heavily pigmented; caudal fin almost uniform light brownish, or perhaps with a slight darkening near ends of rays; aside from the faint blotch, adipose fin yellowish; lower barbels immaculate; upper barbels, brownish. Typr.—The holotype (pl. 11, fig. 1), USNM 201602, is a male 45.8 mm. in standard length. It has 6 soft dorsal rays, 4+-9=13 anal rays, and 18+7+8-+14=47 caudal rays. On each side there are eight pelvic rays, eight soft pectoral rays, one internasal pore (abnormal), and eleven preoperculomandibular pores. The head length is stepped into the standard length 3.45 times and the distance from the rear end of the adipose fin to the tip of the caudal fin is stepped into the distance from the dorsal origin to the rear end of the adipose fin 2.0 times. There are four posterior serrae on the left pectoral spine and five on the right. Measurements are given in table 28. Distrisution.—Noturus baileyi (map 9) is known only from Abrams Creek, Great Smoky Mountains National Park, a tributary to the Little Tennessee River, Tennessee. CATFISH GENUS NOTURUS RAFINESQUE 143 RELATIONSHIP.—WNoturus bailey: and Noturus hildebrandi are tenta- tively associated in the Aildebrandi species group because of their similar numbers of pelvic and caudal rays as well as the general reduction of anterior serrae of the pectoral spine and the reduced posterior process of the cleithrum in both. ErymoLocy.—WNoturus baileyi is named in honor of Dr. Reeve M. Bailey, under whom this study was made. Ecotocy.—The. lower section of Abrams Creek is a stream of 14.6 miles, between the Little Tennessee River and a 25 foot waterfall. Approximately 2.5 miles of the lower part of this stream are now part of Chilhowee Lake, created by construction of a dam on the Little Tennessee River at Chilhowee, Tennessee. The specimens were collected during a reclamation of Abrams Creek, to increase its sport fishery, prior to closure of the dam at Chilhowee. The locality of capture is not now known but would appear to be above the area of present impoundment if the date of capture, June 8, 1957, is correct. Although the reclamation eliminated the fish fauna (Lennon and Parker, 1959) from lower Abrams Creek, subsequent collecting indicates that many of the fish species, reported by them, have returned but no species of Ictaluridae have been taken. The middle section of lower Abrams Creek consists of alternating series of long pools and moderate riffles. The current is moderate in the pools to swift on the riffles. The bottom consists mostly of various size rocks, varying from pea gravel to boulders. Lennon and Parker (1959, p. 12) gave further data regarding Abrams Creek: ‘The stream below Abrams Falls is characterized by short cascades and very long, deep pools. Most of it cannot be waded . . . . At the camp- ground, the stream averages 63 feet in width at the normal discharge rate of 90 cfs. The gradient is approximately 44 feet per mile. The water is clear, slightly brown, and soft.’ Noturus baileyi was collected with Noturus flavus. Remarks.—A possible source of future confusion concerning the origin of the specimens of Noturus baileyi requires comment. The lot of mixed specimens containing the Noturus baileyi, when received at the U.S. National Museum, included three other species of Noturus: N. exilis (USNM 201603, 1 specimen), N. flavus (USNM 201604, 7 specimens), and N. elegans (USNM 201600, 4 specimens). The published list of species collected in the Abrams Creek survey included only Noturus flavus and Noturus miurus. The specimens of Noturus baileyi, although not prominently marked with a bold pattern, as is typical N. miurus, are thought to be the basis of the listing of miurus. The specimens of N. ezilis and N. elegans are regarded as incorrectly included in the fauna of Abrams Creek for the following reasons: (1) The specimens of N. elegans and N. ezilis are similarly preserved and those of N. baileyi and N. flavus have another preservation, 144 U.S. NATIONAL MUSEUM BULLETIN 282 indicating that the elegans and evzilis were not fixed with the flavus and baileyz. All were evidently mixed at a later time. (2) N. elegans is known to occur in the Green River system, Kentucky and the Duck River system, Tennessee. The Duck River population is boldly marked, much more so than any examined from the Green system. The specimens in question, of N. elegans, USNM 201600 (pl. 12, fig. 2), are identical in color pattern and general morphology to others from the Duck system. (3) N. ezilis is known to have been collected with N. elegans in the Duck system, and is fairly common in the lower Tennessee drainage, ranging upstream as far as Alabama. Both exilis and typical elegans are unknown from the upper Tennessee drainage. Thus, the available evidence points to the specimens of N. flavus and N. bailey: having been collected from Abrams Creek, in the upper Tennessee drainage and suggests that those of N. exilis and N. elegans were collected in the lower Tennessee drainage, probably in the Duck River system. Noturus albater, new species OZARK MADTOM FRONTISPIECE; PLateEs 4 (Fia. 4), 11 (Fic. 2); Mar 10 Noturus nocturnus Jordan and Gilbert [misidentifications] Meek, 1893, p. 229 (Arkansas records, in part: Spring R., Black Rock;* Middle Fork White R.., Fayetteville;* White R., Oxford Bend); 1894a, pp. 75, 92 (Spring R., Black Rock,* Ark.). Noturus miurus Jordan [misidentifications].—Meek, 1894a, pp. 75, 92 (deserip- tion; Arkansas records: Middle Fork White R.* and Main Fork White R.,* Fayetteville; White R., Oxford Bend). Schilbeodes miurus (Jordan) [misidentification]—Hubbs and Ortenburger, 1929, pp. 96-97 (reference to previous Arkansas records, in part). TYPE-SPECIMENS.—UMM~Z 151171 (holotype) and UMMZ 167236 (1 paratopotype), from the White River, at Forsyth, on Hwy. M80, Taney Co., Missouri, collected August 4, 1940, by G. V. Harry and Mitro Pellock. OTHER PARATYPES.—The number of specimens, in parentheses, follows the museum number. UNITED STATES: Arkansas: Middle Fork White R., N. of Carter, Washington Co., UMMZ 128311 (37), July 15, 1939, J. D. Black and Henry Mills. Buffalo R., 5 mi. SE. Saint Joe, U.S. Hwy. 65, 2.5 mi. above Gilbert, Searcy Co., UMMZ 169910 (1), August 17, 1940, R. M. Bailey and M. E. Davis. Little Red R., U.S. Hwy. 65, 1.5 mi. SE. Leslie, Searcy Co., UMMZ 169928 (2), August 17, 1940, Bailey and Davis. West Fork of White R., 4 mi. S. West Fork, Washington Co., UMMZ 170900 (3), September 14, 1940, Bailey and Davis. White R., Fayetteville, CNHM 1623 (4), S. E. Meek. Main Fork White R., *Material indicated by an asterisk has been re-examined. CATFISH GENUS NOTURUS RAFINESQUE 145 Fayetteville, USNM 59132 (5), Meek. Spring R., Black Rock, USNM 73569 (3), ANSP 81482 (1), and SU 1170 (3), Meek. Middle Fork White R., Fayetteville, USNM 116375 (32), 18938, Meek. Big Buffalo R., Hwy. 21, 15.8 mi. W. Jasper, Newton Co., TU 10180 (1), April 30, 1955, R. D. Suttkus and party. West Fork White R., U.S. Hwy. 71, 6.1 mi. 8. Fayetteville, Washington Co., TU 12240 (15), September 2, 1955, Suttkus. White R., Hwy. 68, 8 mi. E. Springdale, Washington Co., TU 12258 (85), September 3, 1955, Suttkus; TU 16553 (35), October 4, 1957, Suttkus; TU 16575 (12), October 5, 1957, Suttkus. War Eagle Cr., Hwy. 23, 5.4 mi. N. Huntsville, Madison Co., TU 22401 (7), October 25, 1959, Suttkus and party. Buffalo Fork White R., Hwy. 74, 15.5 mi. W. of Jasper, Newton Co., TU 22504 (37) and TU 22514 (2), October 25, 1959, Suttkus and party. Devils Fork Little Red R., between Stark and Edgemont, Cleburne Co., TU 22709 (9), October 23, 1959, Suttkus and party. West Fork White R., U.S. Hwy. 71, 1 mi. S. Greenland, Washington Co., KU 4542 (4), April 26, 1959, D. A. Distler, C. F. Cole, and A. L. Metcalf. Richland Cr., Hwy. 45, at Goshen, Washington Co., KU 6311 (1), April 22, 1961, F. B. Cross and B. A. Simco. West Fork White R.., at West Fork, Washington Co., KU 6338 (2), April 23, 1961, Distler. Big Cr., Hwy. 14, Searcy Co., KU 6615 (1), September 2, 1960, Distler and W. N. Berg. Middle Fork Little Red R., U.S. Hwy. 65, S. of Leslie, Searcy Co., KU 6625 (3), September 2, 1960, Distler and Berg. Kings R., Hwy. 74, 2 mi. W. Kingston, Madison Co., KU 6674 (3), September 3, 1960, Distler and Berg. Buffalo R., Buffalo River State Park, Marion Co., KU 8020 (2), April 4, 1964, Cross; KU 9866 (6), April 7, 1965, Cross. West Fork White R., U.S. Hwy. 71, 5 mi. S. Fayetteville, Washington Co., CU 34246 (1), August 19, 1959, R. J., R. V., and J. J. Miller. West Fork White R., U.S. Hwy. 71, 1 mi. S. Greenland, Washington Co., CU 35390 (24), April 11, 1959, C. F. Cole. War Eagle Cr., ford, 5 mi. NNW. Huntsville, Madison Co., CU 35633 (1), October 9, 1958, Cole, C. Cleland, and J. Preston. Trib. of East Fork White R., 2.9 mi. NW. Combs, Madison Co., CU 35803 (2), October 10, 1958, Cole, Cleland, and Preston. Richland Cr., Hwy. 45, 10 mi. E. Fayetteville, Washington Co., CU 36427 (1), September 16, 1958, Duane Rorie and Cole. Bear Cr., U.S. Hwy. 65, 6.0 mi. W. Marshall, Searcy Co., CU 36547 (1), April 7, 1958, Cole and S. L. Finkelstein. West Fork White R., Greenland, Washington Co., CU 37441 (9), March 29, 1961, L. W. Knapp and party. Devils Fork Red R., 1.9 mi. N. jet. county road and Hwy. 92, Cleburne Co., CU 41964 (6), April 26, 1962, Knapp and Miller. Fayetteville, BMNH 98.12.29.178 (1), Meek. Mussourr: James R., 1 mi. S. Galloway, U.S. Hwy. 65, Greene Co., UMMZ 102781-2 (2), August 22, 1930, E. P. Creaser and E. B. Williamson. Big Brushy Cr. and Middle Fork Black R., near Black, sec. 27, T. 33 N., R. 1 E., Reynolds Co., UMMZ 139531 (1), July 2, 1941, Aden C. Bauman. Black R. and mouth of Markham Spring Cr., Hwy. A, near Williams- ville, sec. 23, T. 27 N., R. 4 E., Wayne Co., UMMZ 139690 (2), August 15, 1941, Bauman. Finley Cr. of James R., UMMZ 142217 (1), July 14, 1942, Missouri Conservation Dept. Finley Cr. of James R., at Hawkins Bridge, UMMZ 142274 (1), July 17, 1942, Mo. Cons. Dep. Flat Cr. of James R., Barry Co., UMMZ 142303 (1), July 13, 1942, Mo. Cons. Dep. James R., 7.5 mi. E. Springfield, Greene Co., UMMZ 150996 (1), August 1, 1940, G. V. Harry and Mitro Pellock. Beaver Cr., Hwy. M125, Bradleyville, Taney Co., UMMZ 151086 (7), August 2, 1940, Harry and Pellock. Beaver Cr., Hwy. M80, Kissee Mills, Taney Co., UMMZ 151126 (2), August 3, 1940, Harry and Pellock. James R., Hwy. M13, Galena, Stone Co., UMMZ 151287 (6), August 6, 1940, Harry and Pellock. White R., Hwy. M86, 3 mi. S. of Shell Knob, Barry Co., UMMZ 151345 (4), August 7, 1940, Harry and Pellock. Flat Cr., 0.5 mi. S. McDowell, Barry Co., UMMZ 151407 (2), August 8, 1940, Harry and Pellock. James R., County Road 146 U.S. NATIONAL MUSEUM BULLETIN 2:82 A, 6 mi. S. Marshfield, Webster Co., UMMZ 152013 (3), August 20, 1940, Harry and Pellock. Jacks Fork Current R., Alley Spring State Park, Shannon Co., UMMZ 164823 (1), April 27, 1946, Stanton Hudson. James R., 1 mi. S. Galloway, Greene Co., UMMZ 167170 (13), June 29, 1931, John Delavan and Creaser. Current R. and Spring Br., Blue Spring, near Eminence, Shannon Co., UMMZ (no. C41-16) (1), June 27, 1941, C. B. Obrecht. Current R., at Cave Spring, near Hunter, Carter Co., UMMZ (no. C41-21) (1), July 31, 1941, Obrecht. Eleven Point R., Turner Mill Spring, Oregon Co., UMMZ (no. C41-41) (2), August 25, 1941, Obrecht. Marble Cr., near French Mills, Madison Co., UMMZ (no. CC21) (1), October 12, 1933, Creaser and W. Clanton. Eleven Point R., sec. 11-12, T. 24 N., R. 3 W., 1 mi. below Turners Mill, UMMZ (no. 3) (1), June 20, 1938, A. Hugh Denny. Current R., sec. 5, T. 23 N., R. 2 E.,. UMMZ (no. 20) (1), July 27, 1938, Denny. Black R., Lesterville, Reynolds Co., INHS (uncataloged) (3), October 12-15, 1948, Lowry. Current R., Hwy. 14, 0.7 mi. W. Doniphan, Ripley Co., CU 24243 (2), August 13, 1953, C. R. Robins and W. S. Woolcott. Eleven Point R., Hwy. 14, 25.4 mi. W. Current R., Oregon Co., CU 24266 (5) and CU 24267 (15), August 13, 1953, Robins and Woolcott. Little Black R., Hwy. 14, 2.4 mi. E. Fairdealing, Butler Co., CU 24281 (2), August 13, 1953, Robins and Woolcott. Long Cr., Oasis, Taney Co., CU 32834 (5), November 10, 1956, L. W. Knapp and J. P. Henley. White R., Hwy. 86, near Shell Knob, Barry Co., CU 32861 (15), November 9, 1956, Knapp and Henley; CU 32864 (8), February 23, 1957, Knapp and Henley; CU 32935 (3), July 4, 1957, Knapp and M. Bowman. Finley Cr., U.S. Hwy. 160, S. of Nixa, Christian Co., CU 37453 (6), March 29, 1961, Knapp and party. Finley Cr., at crossing, 5 mi. E. of Sparta, Christian Co., CU 37718 (2), August 28, 1957, Knapp and R. Myers. Flat Cr., at second crossing above James R., Stone Co., CU 38052 (2), September 12, 1957, Knapp and party. James R., at Finley Cr., Stone Co., CU 38433 (1), May 18, 1958, L. and B. Knapp. James R., Road A, Webster Co., KU 5550 (1), June 5, 1960, D. A. Distler and A. L. Metcalf. Current R., Round Springs State Park, sec. 20, T. 30 N., R. 4 W., Shannon Co., KU 7645 (18), April 7, 1963, F. B. Cross and M. L. Wiley. Current R., Owls Bend, Hwy. 106, Shannon Co., KU 7679 (1), April 8, 1963, Cross and Wiley. James R., Road D, W. of Turner, Greene Co., KU 7853 (1), May 11, 1963, Cross. James R., Hwy. M125, 6 mi. S. Strafford, Greene Co., KU 7882 (1), May 11, 1963, Cross. Saint Francis R., Sam A. Baker State Park, sec. 32-33, T. 30 N., R. 5 E., Wayne Co., UMoMZ 7261 (4), September 20, 1949, Perry Robinson. Sinking Cr., Hwy. K, 9.2 mi. from junction Hwy. K with Hwy. 21, Reynolds Co., USNM 202484 (9), June 29, 1964, Leslie W. Knapp. Black R., Hwy. A, 3 mi. W. Williamsville, Wayne Co., USNM 202487 (11), June 29, 1964, Knapp. Bryants Cr., Hwy. 14, 14 mi. E. Ava, Douglas Co., USNM 202488 (2), May 21, 1965, Knapp. Diacenosis.—Noturus (Rabida) albater, aside from Noturus hilde- brandi which has eight pelvic rays throughout its range, is the only species of the subgenus with typically nine soft pectoral rays. In contrast to other species of the subgenus the body is elongate with more vertebrae and the head is proportionately small. The head length in the standard length goes 3.7 times or more, distinguishing albater from other Rabida, except N. elegans and N. hildebrandi lautus. From the members of the elegans group, albater is distinguished by pigmentation, especially the large dark adipose bar, the basicaudal CATFISH GENUS NOTURUS RAFINESQUE 147 bar, and the creamy white area on the upper procurrent caudal rays. In N. albater there are 45 to 54 caudal rays, usually 9 pelvic rays, normally 2 internasal and 11 preoperculomandibular pores, and 13 to 16 anal rays. The posterior process of the cleithrum is very short or obscure and the anterior serrae of the pectoral spine are nearly obscure. Description.—Other counts and measurements are given in tables 17 to 26. Body rather elongate for a Rabida, the greatest depth under the dorsal fin; caudal peduncle slightly tapering posteriorly, as deep as the head back of the eyes; eye 1.5 to 2.4 times in snout; head rounded or somewhat flattened above, little depressed; snout slightly depressed; lower jaw included; premaxillary tooth band with poste- rior corners rounded or obtusely angulate; pectoral spine (pl. 4, fig. 4) relatively short, rather slender, and nearly straight, with the anterior serrae weak or almost obscure; serrae of the posterior edge rather numerous and with tips curved toward spine base (except 1 to 3 proximal serrae); humeral process blunt, poorly developed or nearly obscure, seldom as long as the width of the shaft of the pectoral spine; dorsal spine moderately slender; adipose fin high, with a rounded upper edge, no free posterior flap, and united to the moder- ately long procurrent caudal rays; posterior end of caudal fin rounded or truncate; gill rakers six to eight on the first arch. In 91 specimens, 90 have 6 soft dorsal rays, 1 has 5. In 87 specimens (extremes in parentheses): upper simple caudal rays (15) 17 to 19 (21), mean 18.02; branched caudal rays (one aberrant count of 12), otherwise (15) 16 (17), mean 15.9, of which 7 are usually in the upper part of the fin and 9 in the lower half; lower simple caudal rays (13) 14 to 17 (18), mean 15.5. There are three to eleven posterior serrae of the pectoral spine. The largest specimen measures 89 mm. in standard length; most are under 60 mm. In a stained specimen the ossified pectoral radials are tightly fused on each side and there are 13 vertebrae anterior to the origin of the anal fin. Body color yellowish with prominent black or brown blotches. In preservation, yellowish or brown with contrasting dark brown or black markings; top of head dark brown with a posterior bar that extends to the branchiostegal membrane and onto the operculum; a dark band extends from the snout around the eye, backward to meet the first, outlining a light postorbital spot; anterior nares pale, but usually with some pigment; upper barbels pigmented; lips pale, with some pigment in front of and occasionally on mental barbels; a light band (not prominent) of pigment anterior to pelvics; remainder of lower head, abdomen, and pelvic fins immaculate; anal 148 U.S. NATIONAL MUSEUM BULLETIN 282 fin often with a few brownish blotches; pectoral fin and membrane over spine heavily pigmented, its edge immaculate; dorsal spine and proximal part of dorsal fin brownish; remainder of dorsal fin often with brown pigment distally on rays or some specimens with a dark brown submarginal blotch or band; adipose fin whitish, the base anteriorly with scattered brown pigment and the middle with a dark brown or blackish blotch or bar which usually extends to the margin in adults; this blotch confined to the upper body surface and discrete from the other black saddles; another saddle extends from the front of the adipose fin to a yellowish, obovate area at the base of the last dorsal rays; a lightly pigmented area before the dorsal fin; a squarish anterior saddle beginning just before the dorsal fin may extend to or below the lateral line and backward to the third or fourth dorsal ray; a yellowish white blotch on the upper and usually the lower procurrent caudal rays, their base brown; a black or dark brown vertical bar extends through the caudal peduncle to both margins of the caudal fin; branched caudal rays and adjacent simple rays brownish or yellow- ish brown with dark brown concentric bands at or near end of rays. Typr.—The holotype (UMMZ 151171) is an adult male, 61.2 mm. in standard length and 73.5 mm. in total length. It has 5+9=14 anal rays, 17+6+10+14=47 caudal rays, 37 vertebrae, and 6 soft dorsal rays. The pectoral fins and internasal pores are atypical, the pectoral having eight soft rays on the left and nine (counting a small splint) rays on the right; there is one internasal pore on the left and two on the right side. On each side there are nine pelvic rays, seven recurved serrations on the posterior edge of the pectoral spine, and eleven preoperculomandibular pores. The head length stepped into the standard length is 3.75 and the distance from the adipose notch to the tip of the caudal fin stepped into the distance from the origin of the dorsal fin to the adipose notch is 2.35 .The humeral process and the anterior serrae of the pectoral spine are short or even obscure. Plate 11, figure 2 shows the body form and pigmentation of this specimen; table 28 gives further measurements. DisTRIBUTION (map 10).—Noturus albater is known only from the upper White and Saint Francis River systems of Arkansas and Missouri. It occurs in the two main branches of the White, the Black and the White Rivers and their tributaries. Two records are known from the upper Saint Francis River, Missouri. VaRIATION.—The meristic data taken from seven to ten specimens from the Black River system are overlapped (one exception) by the combined counts on 58 to 88 specimens from the upper White River system. Means for the samples from (a) the White River, (b) the Black River, and (c) the combined population follow in order: pelvic rays (a) 9.04, (b) 8.95, (c) 9.03; soft pectoral rays (a) 8.94, (b) 8.85, CATFISH GENUS NOTURUS RAFINESQUE 149 Map 10.—The distribution of Noturus albater, new species. Circles represent material examined. Unverified records which appear to be based on N. albater lie well within the indicated range. (c) 8.94; preoperculomandibular pores (a) 10.93, (b) 10.90, (c) 10.92; anal rays (a) 14.50, (b) 14.90, (c) 14.54; lower-half caudal rays (a) 24.25, (b) 24.30, (c) 24.26; upper-half caudal rays (range: (a) 23 to 28; (b) 22 to 26) (a) 25.13, (b) 24.20, (c) 25.03; total caudal rays (a) 49.35, (b) 48.50, (c) 49.27; vertebrae (a) 36.78, (b) 37.00, (c) 36.80. Only pectoral and pelvic rays were counted from four Saint Francis River specimens. These coincide with counts from White River specimens. Considering the small size of the Black River sample, no important differences are evident. Rewationsuips.—Noturus albater is a distinctive species without close affinity to any other species of the subgenus Rabida. The increased number of pectoral rays and the shortening of the anterior pectoral 150 U.S. NATIONAL MUSEUM BULLETIN 282 serrae and posterior process of the cleithrum, as well as some similarity in color pattern suggest a remote relationship with typical Noturus hildebrandi. Otherwise the elongate body form and small head indicate some relationship to members of the elegans group. ErymoLocy.—The name albater (Latin), albus (white) and ater (black), meaning white and black, refers to the contrasting light areas of the caudal fin and the dark saddles. EcoLoGicaL CONSIDERATIONS.—Records are from clear or normally clear, moderate or large size streams, having a bottom of sand, gravel, rubble, and large rock. The species lives in moderate to swift current, primarily on the shallower riffles. Noturus albater has been taken in collections with Noturus exilis, N. flavater, and N. miurus. The ELEGANS Group Noturus elegans, new species ELEGANT MADTOM PuLaTEs 4 (Frias. 5-6), 12; Map 11 Noturus miurus Jordan [misidentifications]—?Woolman, 1892a, p. 256 (records, in part, Green R. system, Ky.).—?Garman, 1894, p. 56 (compiled from Woolman). Schilbeodes miurus (Jordan) [misidentification]—?Evermann, 1918, p. 335 (com- piled from Woolman). Noturus species.—Murphy, 1964, p. 71 (Green R., Ky.).—Charles, 1967, pp. 385-394 (Green R., Ky.). TYPE-SPECIMENS.—UMMZ 167597 (holotype) and UMMZ 155526 (30 paratopotypes), from Fallen Timber Creek, at Hwy. 90, 8 mi. SE. of Glasgow, Barren Co., Kentucky, collected April 7, 1948, by Reeve M. and Marian K. Bailey. UMMZ 154650 (11 paratopotypes) and USNM 174903 (4 paratopotypes), same locality, April 10, 1947, R. M. Bailey and N. J. Wilimovsky. OTHER PARATYPES.—The number of specimens, in parentheses, fol- lows the catalog number. UNITED STATES: Kentucky: UMMZ 88038 (1), Harrison Cr., 8 mi. S. Glasgow, Barren Co., August 28, 1929, E. P. Creaser and H. R. Becker. UMMZ 154676 (1), Bays Fork of Barren R., U.S. Hwy. 31 E., 2 mi. NE. Scottsville, Allen Co., April 11, 1947, R. M. Bailey and N. J. Wilimovsky. UMMZ 165258 (3), Green R., Hwy. 198, Yosemite, Casey Co., April 5, 1953, R. M. Bailey, D. M. Bailey, H. E. Winn, and J. J. Keleher. UMMZ 165293 (43) and USNM 174902 (5), Green R., Greensburg, Green Co., April 5, 1953, R. M. Bailey and party. UMMZ 165340 (13), East Fork Barren R., Hwy. 63, 5 mi. NW. Tomp- kinsville, Monroe Co., April 6, 1953, R. M. Bailey and party. UMMZ 165395 (18), Fallen Timber Cr., Hwy. 63, 6 mi. SE. Glasgow, Barren Co., April 7, 1953, R. M. Bailey and party. UMMZ 168229 (7), Russell R., Hwy. 206, 1 mi. E. Columbia, Adair Co., August 24, 1954, R. M. Bailey and family. USNM 163065 (9), trib, Green R., Hwy. 35, 4.5 mi. SSW. Liberty, Casey Co., April 25, 1952, F. B. Schwartz, E. A. Lachner, and W. T. Leapley. USNM 163066 (2), trib., Green R., Hwy. 35, 6 mi. NE. Liberty, Casey Co., April 25, 1952, Schwartz, Lachner, and Leapley. USNM 163090 (17), trib., Russell R., Hwy. 80,7 mi. NE. CATFISH GENUS NOTURUS RAFINESQUE PL Edmonton, Metcalfe Co., April 25, 1952, Schwartz, Lachner, and Leapley. USN M 163091 (2), trib., Green R., Hwy. 80, at Edmonton, Metcalfe Co., April 24, 1952, Schwartz, Lachner, and Leapley. USNM 163092 (2), trib., Green R., Hwy. 35, 0.2 mi. E. Dunnville, Casey Co., April 25, 1952, Schwartz, Lachner, and Leapley. USNM 163093 (2), Russell R., Hwy. 80, about 1 mi. SE. Columbia, Adair Co., April 25, 1952, Schwartz, Lachner, and Leapley. USNM 200480 (3), Trammel Cr., 1.5 mi. above mouth, Warren Co., Peter A. Hackney. USNM 201396 (3), Knob Lick Cr., Hwy. 198, just N. Yosemite, Casey Co., June 12, 1965, Schwartz and R. V. Miller. CU 37616 (1), Calhoun Cr., Hwy. 35, 5 mi. S. Liberty, Casey Co., March 25, 1961, Neal R. Foster and party. CU 45997 (1), Trammel Cr., 3.4 mi. SW. jet. U.S. Hwy. 231 and Hwy. 100, at U.S. Hwy. 231, Allen Co., August 1963, T. Zorach and party. KU 7078 (1), stream, Hwy. 80, 4 mi. W. Russell Springs, Adair Co., June 8, 1960, A. L. Metcalf. TU 19390 (12), Green River, Hwy. 88, 3.8 mi. ENE. Donansburg, Green Co., September 7, 1958, R. D. Suttkus and party. TU 19460 (16), trib. to Barren R., U.S. Hwy. 31 E., 2.6 mi. SW. Lucas, Barren Co., September 7, 1958, Suttkus and party. UL 5185 (1), Bays Fork, U.S. Hwy. 31 E., 1.5 mi. NE. Scottsville, Allen Co., July 8, 1954, W. M. Clay and E. W. Whitney. UL 5581 (42), Salt Lick Cr., 0.75 mi. S. and 1 mi. E. Akersville, Monroe Co., October 29, 1955, Charles and party. UL 5728 (22), Russell R., Hwy. 80, 2 mi. E. Columbia, Adair Co., August 10, 1955, D. Distler and Clay. UL 5746 (2), Goose Creek, Casey Co., May 5, 1955. UL 6155 (9), Long Cr., 2 mi. E. Mt. Union or 4 mi. E. and 7 mi. S. Scottsville, Allen Co., June 8, 1956, Clay and Bernard Carter. UL 7915 (1), Green R., below first bridge on Hwy. 35, S. of Liberty, Casey Co., June 28, 1955, Charles and party. TEN- NESSEE: UMMZ 181071 (1), Trace Cr., Hermitage Springs, Clay Co., April 30, 1954, H. E. Winn and Norman Benson. OTHER MATERIAL STUDIED UNITED STATES: Kentucky: USNM 199585 (1 specimen), locality un- known. TENNESSEE: USNM 197396 (2 specimens), Hurricane Cr., 8S. of Mc- Ewen, Humphreys Co.; USNM 197398 (2), Bigby Cr., near Mount Pleasant, Maury Co.; USNM 197400 (1), Tumbling Cr., Humphreys Co.; USNM 201387 (5), Beaver Dam Cr., Hwy. 50, E. edge Coble, Hickman Co.; USNM 201600 (4), locality unknown; mixed with specimens reputed to have been collected in Abrams Creek, Great Smoky Mountains National Park, but instead probably from Duck River system, Tennessee; see remarks under Noturus baileyt. Draenosis.—Noturus (Rabida) elegans is characterized by an elon- gate anal fin with 14 to 19 rays, small spines with short anterior serrae, a short posterior process of the cleithrum, and typically 9 pelvic rays. It has 46 to 55 caudal rays; 8, sometimes 9, soft pectoral rays; normally 11 preoperculomandibular pores; and 2, infrequently 1, internasal pores. The body is moderately elongate; the caudal and adipose fins are united, but the fins are low at the junction; the soft dorsal rays, except near the dorsal fin base, are without dark pigment. It is distinguished from members of the miurus and furiosus species groups by color pattern, shorter spines and posterior process of the cleithrum, and the connection of the adipose and caudal fins; from Noturus eleutherus by small spines, wide union of the caudal and adi- pose fins, and eleven rather than ten preoperculomandibular pores; 152 U.S. NATIONAL MUSEUM BULLETIN 282 from the probable hildebrandi species group by color pattern and nine rather than eight pelvic rays; from Noturus albater by generally fewer pectoral rays and color pattern; from Noturus trautmani by better developed anterior serrae of the pectoral spine, a bolder color pattern in preservation, a longer posterior process of the cleithrum, a lower adipose and upper caudal fin, a longer anal base, and more vertebrae. DescripTIon.—Counts and measurements are given in tables 17 to 26. Head somewhat depressed; lower jaw included; eye moderate, about 1.5 to 1.8 times in snout; posterior process of cleithrum short but distinct, about equal in length to the width of the pectoral spine including the anterior but not the posterior serrae; pectoral spine relatively short, slightly curved, with recurved posterior serrae and very fine, yet distinct anterior serrae; dorsal spine stout; adipose fin moderately low and relatively long, without a free posterior margin, united to the caudal fin; procurrent caudal rays of moderate length; tail slightly rounded posteriorly, middle rays longest; body elongate, dorsal and ventral surfaces only slightly tapering posteriorly; poste- rior corners of premaxillary tooth band rounded. The largest speci- men is 64 mm. in standard length. The caudal fin has (extremes in parentheses): (16) 17 to 20 (22) upper simple rays; (13) 15 to 18 (19) branched rays of which there are modally 7, often 8, in the upper half and 8 to 10 in the lower half of the fin; and (11) 13 to 17 (18) lower simple rays. There are six to eight gill rakers on the first arch; the pectoral spines bear five to nine recurved serrae; and there are typically six soft dorsal rays. Five cleared and stained specimens from the Green River system have the pectoral radials fused and twelve vertebrae anterior to the anal origin. Background color in aquarium specimens from the Green River system yellowish gray, lightly diffused with pink. In preservation golden yellow or light yellow, dark brown on dorsum and side in one example, with numerous minute chromatophores completely covering the side and upper surface. Pelvic fin and ventral surface of head and abdomen immaculate, except a light band of pigment often on the chin in front of mental barbels; mental barbels immaculate or sometimes with scattered pigment; a dark brown bar on posterior end of head passes downward over the operculum and a branch follows the lateral line to below the dorsal fin; a similar colored band extends forward from below the eye onto the snout, through the nares; upper barbels well pigmented; lower cheek light, with moderate chromatophores; top of head dark; dorsal fin immaculate except dark brown at base of first three rays and over dorsal spine; a dark brown saddle, beginning just anterior to base of dorsal spine, extends CATFISH GENUS NOTURUS RAFINESQUE 153 to the third dorsal ray and midside; another lies between the adipose and dorsal fins; a third, at base of adipose fin, variously extends into the basal half or more of fin and to the distal margin in adults; a very light bar of brown pigment crosses the caudal peduncle, ex- tending onto the upper procurrent caudal rays; three light ellipses separate the four dark dorsal saddles or blotches; a quadrate-shaped light area lies between the first saddle and head; caudal fin usually with two or more concentric brown bands covering most of fin, one of which is submarginal; anal fin usually immaculate, infrequently with scattered pigment; pectoral spine covered with brown chromato- phores, the pectoral rays moderately pigmented near spine and toward base. Specimens from the Green River system (pl. 12, fig. 1) usually are somberly colored, lacking bold dark saddles, in contrast to the prominent patterns of Duck River specimens (pl. 12, fig. 2). Tyre (pl. 12, fig. 1).—The holotype (UMMZ 167597) is a male, 44.1 mm. in standard length. It has 6+11=17 anal rays; on each side there are eight soft pectoral rays, nine pelvic rays, eleven preoper- culomandibular pores, two internasal pores, and seven serrae on the posterior edge of the pectoral spine; the dorsal fin has five soft rays and the caudal fin has 19+7+9+14=49 rays. The head length is stepped into the standard length 3.85 times and the distance from the adipose notch to the tip of the caudal fin is stepped into the distance from the dorsal fin origin to the adipose notch 2.05 times. Other measurements are given in table 28. VARIATION.—Vertebrae in 28 specimens from the Green, 42 from the Barren, and 14 from the Duck River systems are 34 to 36, mean 35.07; 34 to 37, mean 35.83; and 34 to 36, mean 35.29, respectively. The combined mean is 35.49. Other mean variational data are listed, in order, for (a)48 Green River specimens, (b)48 from the Barren River system, (c)14 from the Duck River system, and (d) the com- bined samples; the ranges of variation are given in parentheses but omitted if identical: internasal pores (a)1.92, (b)1.86, (c)1.96, (d)1.90; preoperculomandibular pores (a)(10-12) 10.98, (b) (10-12) 11.04, (c)(10-11) 10.82, (d)(10-12) 10.99; soft pectoral rays (a)8.22, (b)8.23, (c)8.25, (d)8.23; pelvic rays (a) (8-10) 9.06, (b) (8-10) 9.02, (c) (9-10) 9.11, (d)(8-10) 9.05; anal rays (a)(49 specimens; 14-18) 15.73, (b) (14-18) 15.81, (c)(16-19) 17.29, (d)(14-19) 15.96; soft dorsal rays (a)(5-6) 5.94, (b)(6) 6.00, (c)(5-6) 5.86, (d)(5-6) 5.95; upper-half caudal rays (a)(23-29) 25.92, (b) (24-29) 25.92, (c)(25-28) 26.21, (d)(23-29) 25.95; lower-half caudal rays (a) (21-26) 24.06, (b) (20-26) 23.63, (c) (23-26) 24.00, (d) (20-26) 23.86; total caudal rays (a) (46-54) 49.98, (b)(46-55) 49.54, (c)(48-54) 50.21, (d)(46-55) 49.82. The anal fin thus appears to be slightly longer and the color pattern is bolder in 154 U.S. NATIONAL MUSEUM BULLETIN 282 gs Nereis TRAUTMANI @ 5 ge nr 8 > o 4 6 > co) 0 a ae = cate _ we 2 Bees sie £2 £ | € 53,8882: . seeeeeatzieas OE Pee ENYOR £ 8lssste2 § 83283 3 B5se 588 SHS RECENT e2 § a2 a a SsSsseesase & ERsssl FS RFESELE 3 3 i 1 STN } PLEISTOCENE | ie \ VAN / ge ee ! / | 1 | Qireglogion yo- = es fe la; ! R é nea 1 / / eae i! | en Puiocenem, \i) sf lei as aut ‘ eine ee / \ i eo q ! N via “Ss \ T 7 \ \ Wy yt NS \ Vv 7 \ \ 1 ts S SB / N 1 / SN | A \ | MIOCENE 3 yo ty \ esa NY 1h of iS ay % ey a 7 \ ii Ne | i} Phylogeny of the Ictoluridae OLIGOCENE rs > oe th Solid Line ~ Fossil Evidence ws u Broken Line —Hypothetical History aS i Si \ | Only selected Ictalurus shown \ \l V | EOCENE M Ficure 5.—Phylogeny of the Ictaluridae. The species assigned to the subgenus Schilbeodes are more divergent and possibly a heterogeneous assemblage. Prominent generalized characters are found in a number of species. Those with a terminal mouth and ten preoperculomandibular pores (gyrinus, lachneri, and exilis) are very generalized with gyrinus being most primitive as judged from its pelvic fin and deeply grooved pectoral spine. In this area, N. exilis has diverged considerably with nine pelvic rays and a serrated spine. Noturus leptacanthus with eight pelvic rays and a nonserrated spine may be the most primitive of those with an in- cluded lower jaw, but Noturus insignis, Noturus phaeus, and Noturus funebris with modally seven hypurals retain a generalized character that may be as significant. Noturus gilberti is probably the most highly specialized member of the subgenus. Noturus flavus, in the subgenus Noturus, resembles gilberti in a number of characters but is the only member of the genus constantly retaining the primitive unfused pectoral radials; however it has diverged considerably from the generalized condition with specialization of the teeth and increase in the number of paired fin rays and caudal rays. In the interpreted phylogeny (fig. 5), only a few of the species of Ictalurus are indicated, otherwise the forms are arranged in natural groups. The solid lines indicate the history as judged from fossils; the dashed lines show the hypothetical history. Literature Cited ABBOTT, CHARLES C. 1871. Further notes on New Jersey fishes. American Nat., vol. 4, pp. 717- 720. 1894. A naturalist’s rambles about home (2nd ed.). New York: D. Appleton and Co., 485 pp. ApamMs, CHARLES C., and HAaNKINSON, THomas L. 1916. Notes on Oneida Lake fish and fisheries. Trans. American Fish. Soc., vol. 45, pp. 154-169. 1928. The ecology and economies of Oneida Lake fish. Roosevelt Wildlife Ann., vol. 1, pp. 239-548, figs. 175-244, 1 map. ApaMs, CHARLES C.; HANKINSON, THomas L.; and KENpatt, W. C. 1919. A preliminary report on a fish cultural policy for the Palisades Interstate Park. Trans. American Fish. Soc., vol. 48, pp. 193-204. ADDAIR, JOHN 1945. The fishes of the Kanawha River system in West Virginia and some factors which influence their distribution. Abstracts Dissertations, Ohio St. Univ. (Columbus), no. 46, pp. 9-17. AITKEN, WALTER W. 1936. Some common Iowa fishes. Iowa St. Coll. Extension Cire., no. 224, pp. 1-33, 27 figs. 1941. A preliminary report on a check-list of Iowa teleosts. Proc. Iowa Acad. Sci., vol. 47, pp. 387-392. ALLEN, W. R., and Criarx, Minor E. 1943. Bottom-preferences of fishes of northeastern Kentucky streams. Trans. Kentucky Acad. Sci., vol. 2, pp. 26-30. ANDERSON, WILLIAM D. 1964. Fishes of some South Carolina Coastal Plain streams. Quart. Journ. Florida Acad. Sci., vol. 27, pp. 31-54. ANDERSON, WILLIAM D., and FREEMAN, Harry W. 1957. Fishes of the Congaree River drainage, South Carolina. Univ. South Carolina Pubs., Ser. 3 (Biol.) vol. 2, pp. 90-111. ANTHONY, JAMES D. 1963. Parasites of eastern Wisconsin fishes. Trans. Wisconsin Acad. Sci., Arts and Letters, vol. 52, pp. 83-95. ARNOLD, JoH. Pau, and AHL, ERNST 1936. Freundlandische siisswasserfische. Braunschweig, pp. 592, illus. BaiLEy, JOSEPH R., and OLivEeR, JAMES A. 1939. The fishes of the Connecticut watershed. New Hampshire Fish and Game Dept., Surv. Rep., no. 4, pp. 150-189, figs. 57-80. BaILEy, REEVE M. 1938. .The fishes of the Merrimack drainage. New Hampshire Fish and Game Dept., Surv. Rep., no. 3, pp. 149-185, 12 figs. 1951. Sitka? eahitere Soe AN ex . Se ees . rn Oy * ; A me Say Vaiss - er L oe "Tare Dare (St Siby b Tables 1-28 ad & i =< ; ¥ . . e At ’ ‘ ‘ T2528 : ' Y , cS r . i ; \ ; : ; aa oy aul 7 i ; ; : dh : 7 F = ~ : a ; = se ' Sw * ™ — L see sae OU aati sai i a 282 NATIONAL MUSEUM BULLETIN U.S. 262 pessoidep Ajoye1epoyy [I[nys puokeq szoeloid A[ereq {Moireu ‘joys AIO A aovjims uoTyel[NoI}We se [euotyounjuou JO ajelosqg eel} ‘Z L Ajjeutiou sfei yos ‘yuesqe eutdg (7961) ezueiied oes ‘(Z]) Sf 10 ([) ZI 8 Aj]1O11eyue AjWWSt[s pue premdn ssuimg peryeredes A[Sopiy (peydni19e4 -UI [e@Uuso 4Nq) spIs YoRe [] D1] 22914 d poeyore AjoyeIopow eseq ulj [BSiIop wo ssoooid 04 spusyxe faaoqe peoiq ‘suo'T (1) eeqy *Z (3) 9'T (3) ST (6161 ‘uueuUedTq) peyodei g oste ‘(T) 6 10 (T) 8 Aj so11eysod A}WYWSI[S pue premdn ssurmg poyeredes ATopin (1) TT-OL 40 (1) OT-OT s1up79 01304 J peyore A[ye013 04 posseidep AT WSIS ese q UlJ [VSJOp wo Ssoooid YiIM SyOoUuUOD JO [[Nys puofsq A[qeieptsuoo spusyxe foaoqe peoig A[eyelopow Jo peoiq ‘yySue] UL SUC] 0} oyeI1Epoy wniyyleloeidns YWIM Soye[Noiyre ‘yuoseig aealj ‘Z g‘{ A[TeulioN (s1)papu UL QT ueqyo yng ‘Area WOPpTes) ST 8 Aj[io110}Ue A|WYSI[S pue premdn ssuimg poyeredes A[OpiM opis qouo OT snin7} D2] T8418 sseooid [eydi0009 (LT6T ‘uedsey ) euog [e10dula}-jsog epIs yore ‘s]eIpei [e1ojoed peljisso sfe1 [esiog sfei [epneos poyouvig (dequinu [epoul) epIs yove ‘shel O1A[og eke pulyoq [eueo [eyIqsov jul s[euso re[nqipuewojno -1edoaid Jo spue Jollowy (aequinu [epoul) seiod re[nqipueuwio[no1sdoeig leyoe rey (‘@SIMI9YIO pazBoIpUr ssajzuN peutwexe sueuoeds jo Joquinu oj JojaI Sasayyuaied ur siequinN) app1injp}2] {O piauay ayy fo uostvdwuo’--" | ATaVL 263 CATFISH GENUS NOTURUS RAFINESQUE pessoidop A[yeo13 04 poyore ATWYBITS pessoidep Ajeye1spoyp pesseidoep AT[7e015 14s [[nys puofeq soe [oid Ajereq ‘moureu ‘y0ys AloA [[nys puofseq eI] Spuezxe ‘mosreu ‘yI04g [Inxs puoXkeqg [[eM Spuszxe ‘mOireu ‘4ySUs] oyeLOpoul JO sseooid yeyidi00Q9 eovjms uolye[NoI4se se peuoryounj -u0U {[VISIJSOA JO JUOSqY (L161 ‘uesey) suog [e10dule}-jsOg unimyleyoeidns YuIM SoyeNoryre {yuoselg S19y4O [[@ UI pesnyj Ajyensn ‘ieyjoue ut A,[yuenb -o1j pue se1oeds [| ul oedj ‘Z apis youe ‘s[elpel [esiojoed paljisso ({) ee *Z eed ‘Z Sotoeds | UI g*] Uezo ynq QT ATTeuLION 9€ 04 FT Ajyensn ‘Op OF BT ‘el qelreA (I) L°I 40 (Z) 91 g‘] A[[euts0 Ny ske1 [esiog (€) ST (Area WOopyes) ST skei jepneo peyouvig (dequinu [epoul) epis yowe ‘skel olAleg (3) OL 29 (1) 6 OT uezyo ‘6 | (setoeds gT) 6 ‘(setoeds g) g eho puryeq jeueo [eyqioeljuy Aj 1O110}ue A}WSI[S pue premdn ssuimg Ajjetpeuioieysod ssuimg Ajjetpewoiseysod ssuimg SOpis y}OqG 0} UOWUIOD e1iod io yd ueIpowl e oyur uedg euI[piu ye yId e ur urof Ajerei {peyeredos AjTopim ATT ensy Seprs yIOq 0} uowwos e1od Jiollejue tpoulor sjeuso rejnqipuewo[no -1adoeid jo spue JOlloyUy (setoeds LI) 11 *(setoeds Z) IT 10 OT ‘(seroeds 7) epis yoee CT (d4oquinu [epoul) seiod re[Nqipueulo[noiedoeig (}x0} 90S) Opis yore ZI epIs yore ZI s421p07k q sninjo N ( 89STMI9q}0 peteorpul SsoyuNn pouluiexs suswtoeds jo Joquind 0} Jojol Ssasoyiosed ul sioquin Ny) ieyou rey penurjuo')--anp2in7Dj97 fo piauay ayy fo uostandwo-"| FTAVL NATIONAL MUSEUM BULLETIN 282 US. 264 Sioqueyo g {yeselg peyueusiduy quesqy VI 97 OL euviIquicul uIy} e@ Aq pepunoimg ulj [epned YIM poeyoouUu0D Ajpeoiq pue 4oeq 07 oyeupe ‘ysiyg Ajeyelopow pue Suo7y D1] 27911 d quesqy peyuoursiduy quesqy (3) 61 10 BT euviquicwl uty} @ Aq pepunoimg ulj [epneo 0} oseq ye poyOouUOO pues yoeq 03 eyeupe ‘A JO110jSs0d pepunod fysiy pue Suo7y s1up]3 078 O41 J SJoquieyo ¢ JO Z% Syuoseig poyuousid A[snoire A queseig ei0Ul JO [] ‘SnoJeuNnN ensst} Ayselj YOY? UL peppequiy uj [epneo Wolj oJOWeI sseq pus yoeq WOlj sel1j pus 101104Ssod qyim dey ysty @ ‘yi0ys ‘[Teulg snin]DI2 soppe[q ITV uIyg sok yore ySilj ‘sexe [1D skei jepneo JueImMooig uly esodipy leyoereyo (‘ASTMJ9yIO poayeoIpUl SSoyuN pourwexa SusWTOadsS Jo JoquINU 0} JoJe1 Sasayjueied ul SJequinN) penutuo)--anp1in) p40] fo piauay ay fo uostundwuo--*| ATEAVL 265 CATFISH GENUS NOTURUS RAFINESQUE quesqy peyueusiduy quesqy (eg) 6 OF LT eueiq -wouw uty} @ Aq popunoung ulj [epneo 0} eseq ye poyoouuoCD pu yoeq 0} eyeupe ‘A[JO1104Ssod pepunoi ‘ysty pue suo‘ uDIDS Siequieyo Z ‘yueseig poeyuoursid A[snoire A queseig e10UI JO OT ensst} Ayselj YOrY) Ul peppequiy ulj [epneo wod ejOUlel oseq pue Yor wWOIj velj pus JOII0ysod y4yIM deyj pepunoi ysty e fyioys ‘esreyT s1301po7hk q Sioquieyo zg ‘yesolg peyuousid Aj[sno1se, queselg OT 07 § ‘Mog eueiq -wouw uly} @ Aq pepunowmng ulj [epneo woij peyeredes A[}Y4SI[S: IO 0} peyoouuod Aj[SNOLIeA pue xoeq 07 oyeupe ‘ysty Ajeyerepoul 07 MOT ‘Suor7T SnN4iNqO AJ Jeppe[q Ty ulyS sok yore 4Sitj ‘S1OyeI [IH skei [epneo juoImMooig ulj esodipy Jeyoereyo (-A9SIMJayIO paleoIpurl ssejuN paurwexe susuToOeds Jo JaquiNU 04 Joyal saseyiueied ul SIoquiny)) penurjuo')--anp1injpj97] fo piauay ay fo uos1spdwoy--"T YTAVL 266 U.S. NATIONAL MUSEUM BULLETIN 282 TABLE 2.--Geographic variation in paired fin rays of Noturus gyrinus (both sides counted) No. of Pelvic Rays Michigan: Upper Peninsula 8.06 Lower Peninsula 8.01 Quebec 7.50 New Hampshire 8.00 New Jersey 7.92 Maryland 7.00 North Carolina 8.13 Georgia 8.00 Florida: Peninsula 8.24 Marianna 8.00 Escambia R. 8.30 Tombigbee R. 9.04 Pascagoula R. 9.00 Biloxi Bay Drainage 8.78 Bay St. Louis Drainage 8.00 Florida Parishes, La. 9.50 Mississippi Drainage, Miss. Other Louisiana Texas Oklahoma Missouri Illinois Ohio R., Ind. G0 GO 00 OO OO 0 OG oOoono ww SFO WDwWaeo se bat es Total CATFISH GENUS NOTURUS RAFINESQUE 267 TABLE 2.--Geographic variation in paired fin rays of Noturus gytinus Area Michigan: Upper Peninsula Lower Peninsula Quebec New Hampshire New Jersey Maryland North Carolina Georgia Florida: Peninsula Marianna Escambia R. Tombigbee R. Pascagoula R. Biloxi Bay Drainage Bay St. Louis Drainage Florida Parishes, La. Mississippi Drainage, Miss. Other Louisiana Texas Oklahoma Missouri Illinois Ohio R., Ind. Total 298-943 O—69——18 (both sides counted)--Continued No. of Soft Pectoral Rays Total COD OTR Os 238 8 21 39 16 10 rw eS OTD &w 119 9 10 Mean AADWAATAABAGD cooouounw SSONDSONO 282 NATIONAL MUSEUM BULLETIN U.S. 268 96°19 86°09 0g°09 6I°8S 06° LS L6°SS S088 I8°¢s 13°€G uve OL 6 GI 86 8% 9G OT Ob kG Gl Gr 26 go 9 £9 19 09 69 8S 29 9g gS skey [epned jo ‘oN snutk3 sninjoy uz sXoi popnvo fo uonniuva o1ydp19004--"€ ATAVL LAO Bpliold euvISINOT pue eureqely Sexo], LINOSSTY euerpuy ‘*y O14O BUI[OIeD YWON pues Arey ‘esysdurey Men ‘Aosio¢ MON uISseq Soyey yearn eoly 269 CATFISH GENUS NOTURUS RAFINESQUE 8h'0S 60 8s" 6G G8 69 vo “OS 1s “TS "TS 6g" aca 6G 1g 66°89 v6 9¢ L9 VI “LY “SP “6D “LY uvoy [999.L I v br 8t vo «O&Y St. s6p 61- GT 8 I I Sete a te Benge G eee oe : - NSE teat eee Pea MG a Lees BARTS si ek : - 2 ol gq. 8 (SG. ee ; eos : - [aCe ene be Orla 10 pal es eae ; - : : et Oa oh ee ec —— - - eel oe oO MTT VAG Oe e ide ee Sk ey ; 7 mF See eh lee ean 16 =i : : eels a ee ay eee sme) he ge ene ot : : es : : N oe Titans 2 : : : a ; : SA et eee hea BOE Se skey [epned jo ‘on St[Ixe SMIMION uz sko1 yopnvo fo saqunu ui uoNnviuva 914dn19005--"% ATAVL ]e10} puwin ‘y sossouuoy ‘OW ‘'y IddississIw IO@MO] pus OSoUlVlo~- ‘yy tddississipw eddy 18307qNS ‘y epeuocoser ‘y WMossip pues odes¢Q :y Lmossip, “a STIG M Te709qNS "yy OYSoeN “a Sout] ‘y nesj0g ‘yry ‘1g AquepHoW :y sesueyiy esvuleiqg 270 U.S. NATIONAL MUSEUM BULLETIN 282 TABLE 5.--Geographic variation in number of anal rays in Noturus exilis No. of Anal Rays Total 17 1818" 20-21 22 Arkansas R.: McHenry Cr., Ark. 18.71 Poteau R. 18. 67 Illinois R. 18.97 Neosho R. 19.18 Subtotal 18.98 White R.: White R. 20.36 Black R. 21.00 Subtotal 20. 38 Missouri R.: Osage and Missouri R. 19.33 Gasconade R. 20.27 Subtotal 19.57 Upper Mississippi R. 18. 73 Meramec and lower Mississippi R., Mo. 19.92 Tennessee R. 20. 09 Grand total CATFISH GENUS NOTURUS RAFINESQUE 271 TABLE 6.--Geographic variation of vertebrae in Noturus funebris and Noturus phaeus (the fused anterior vertebrae are excluded) N. phaeus Forked Deer River Loosahatchie River Yazoo River Red River Little River Beaver Creek Total N. funebris Pearl River Wolf River Escatawpa River Tallapoosa River Perdido River Blackwater River Yellow River Econfina Creek Total No. of Vertebrae Total 384 385 36 37 38 35.92 35.00 35. 73 35.00 36.25 35.00 35. 60 272 U.S. NATIONAL MUSEUM BULLETIN 2.82 TABLE 7.-Geographic variation in paired fin rays of Noturus leptacanthus River System Amite Tangipahoa Chefuncte Pearl Wolf Biloxi Pascagoula Tombigbee Alabama Perdido Escambia Blackwater Yellow Choctawhatchee Apalachicola Ochlockonee Suwannee St. Johns Satilla Altamaha Savannah Combahee Edisto Great Cypress Swamp Total (both sides tabulated) No. of Soft Pectoral Rays Total Mean ooo © © . 22 . 00 31 21 .50 34 48 12 10 79 92 88 . 00 13 02 50 . 02 00 00 Lt seit . 88 . 00 . 69 CATFISH GENUS NOTURUS RAFINESQUE 273 TABLE 7.--Geographic variation in paired fin rays of Noturus leptacanthus (both sides tabulated)--Continued No. of Pelvic Rays Total River System Amite 8. Tangipahoa 8.00 Chefuncte 8.06 Pearl 8.08 Wolf 8. 00 Biloxi 8.04 Pascagoula 8.02 Tombigbee 8.06 Alabama 8.19 Perdido 8.06 Escambia 8.00 Blackwater 8.01 Yellow 8.00 Choctawhatchee 8.03 Apalachicola 8.04 Ochlockonee 8.00 Suwannee 8.10 St. Johns 8.00 Satilla 8.00 Altamaha 8.00 Savannah 8.13 Combahee 8. 67 Edisto 8.00 Great Cypress Swamp 8.00 Total 274 U.S. NATIONAL MUSEUM BULLETIN 282 TABLE 8.--Geographic variation of anal and soft pectoral rays in Noturus funebris and Noturus phaeus No. of Anal Rays Total 18 19 20 21 22 23 24 25 26 27 N. phaeus Forked Deer River Zine Hatchie River 20.73 Loosahatchie River 20.33 Wolf River 21.00 Yazoo River 20.64 Ouachita River 22.80 Little River alee Red River and Bayou Teche 20.89 Homochitto River 21.29 Beaver Creek 21.00 Total 20.99 N. funebris Pearl] River ae = eo Oi sOMl oe clin w =H 22.02 Wolf River See) rae CA! FS Gl Pot ane s=5 Re 22.36 Biloxi Bay PT Ce ae Soe ae ee a a 21.67 Pascagoula Bay = ie st I OMA AS A eh al aim 22.39 Chickasaw Creek a) etl iy= ao sa) enema 2lso0 Tombigbee River Sis ey et Ok Ae ea aaa ice 22.80 Tallapoosa River mine ne dy | a A Pas Bie Aves er eae 21.00 Perdido River es eA Ne Wis oar Sis! ceria Miah ee 23.75 Escambia County and River Blackwater River Yellow River Alaqua Creek Econfina Creek 23.00 24.95 24.36 23.75 23.67 Total 23 58 45 24 11 CATFISH GENUS NOTURUS RAFINESQUE 275 TABLE 8.--Geographic variation of anal and soft pectoral rays in Noturus funebris and Noturus phaeus--Continued No. of Pectoral Rays Total ic Drainage Qo co —_ oO —_ — N. Phaeus Forked Deer River 8.23 Hatchie River 8.40 Loosahatchie River 8.44 Wolf River 8.00 Yazoo River 8.39 Ouachita River Little River 8.92 Red River and Bayou Teche 8.63 Homochitto River 8.93 Beaver Creek 9.17 Total 8.62 N. funebris Pearl River 9.08 Wolf River 9.00 Biloxi Bay 9.00 Pascagoula Bay 9.40 Chickasaw Creek 9.00 Tombigbee River 9.25 Tallapoosa River 9.17 Perdido River 9.17 Escambia County and River 9.38 Blackwater River 9.39 Yellow River 9.20 Alaqua Creek 8.88 Econfina Creek 9.40 Total 2.82 NATIONAL MUSEUM BULLETIN US. 276 que -soid AjjeoidA y, quoseid Aj[eordA J, quesqy quesqy quesqy yyst, A[yuonb -o1J oSdpo ‘spueq [eulsreuqns yrep JIM SOuIT}OUIOg Jepiog yrep e YUM Joaou tyuoseid Ajyensn iepi0oq qSI] peoiq V IO[OO -tun Ajjeords J, Si1epi0q yzep yyM ATTens/) Sioepiog yrep yim Ajyensp Snoieunu 4ou pus ‘jyuoutmoid ‘JOULISIP OBlIag evlles Ie[nsoin M9j @ YJIM SOUT} -oulos ‘pousysnoy peueysnol Ajuo Ajjensp eelIOS OeZIS oyeIe -pow Aueul 04 Moj VW seLlIeS JNOYIM uotsu9}4 -x9 JoI10jsod 9383 -uo[e ue Suraey 04 pedeys-, A[MoTyeys fe[qelrea espe reoy 6-G Ppepny[oul Jemo'yT parted pepunol Jo 3480 -und} ATjensq Et jenbo Ajreon putyoq pepunoi Jo eyeounz ATTensy G-G [enbe Aj rean pulyeq pepunol Jo oyeouny ATTenspy GG pepnyjour 1aMo-T putyoq pepunoi Jo eyeouny ATjensp G-G Ppepnioul 1eMoT pesseid -ep Ajyeois 04 Al eye1opoyp peyore Jo pepunoi AjjeordA Ty, peyore pues pepunoi A[[e1euer pesseidep ATyeoID pesseidep A[moT[eys snu1sk3 s1ugisu2 snuinqz2ou $1]1% a SNAeT} SQINION fo adups ay? Ul 1NDD0 puD ‘Yyq1M at Jessop jo oseq 10119} -sod ye yoqoT[q 7WYSI'T SUIJ UBIPe| (eB8pe IO119}4 -sod) outds [e@10}00q pueq 43003 Arey[Ixeul -oid Jo pues [eieyeq Seiod [eseuisezUy Smee peoH leyoureyo pasnfuos uaaq aany 3DY] SNINJON fo sa12ads 19y}0 pup snae{j Sninjon fo #unok ayq fo uostsndwoy--"6 ATAVL 277 CATFISH GENUS NOTURUS RAFINESQUE SUIJ OLA -[ed 04 JoI1ejuUe ysn{ uowop -qe ssoio Aeul pueq yrep Mol -reu @ 1de0xe Sulj o1rajed 0} Jots1oyue ysnl uowopqe Sessoio pueq rep MOireu e ydooxo ‘poyuou sutj o1ajed 04 iot1ojue ysnl{ ueulop -qe sessoio A[jensn pueq ylep MOdreu @ poquousid -un 0} poyuousid uswopqe poyuowsiduy [jem Ajsnorze, | ydooxe ‘poyuousiduy -sidun Ajjensy ‘poqyuewsiduy jo oo¥vjims Jopuy pojyuoul poyuoeusid poyuoursidun -B1dun osIm Ajqe[nse1 Soulry pojyuouw | Ajvensn sstmioyjyo | -19e4}0 ‘s[Teqreq -owos fpeyueusid -Bidun ostMmioy}o | ‘sTeqreq [eyuoU 04 [@}USUI OF JOT Ay wsiojrun A[yUenb {‘s[eqreq [ejueu} JOTIoyUe di] JomoT | -10}Ue dI] JomMoOT -o1fJ peoy jo sovj | 0} Jol1ojUe dij] Jomo], | uo yuoursId AAvoy uo poe oul peyuousiduy | -ms iopun omjyuq | uo yuousid oye1opop yqyim AjypeorddAy, -did AjVYSIT | peoy Jo sovjms Jopuy ponurjuo')--snaeyj sninion fo azups ay? ul 4nd90 pup ‘yz1m pasnf{uos uaaq aany yy? SNiNiON fo satzads 1ayj0 puv snae{} Snimjon fo Zunok ayz fo uostundwoy--"6 ATAVL 278 U.S. NATIONAL MUSEUM BULLETIN 282 TABLE 10.--Geographic variation of vertebrae in Noturus hildebrandi (the fused anterior vertebrae are excluded) Number of Vertebrae Total SU, ole ooioa 640 J5mna6 Drainage Homochitto R. Bayou Pierre Yazoo R.: Yalobusha R. Yocona R. Tallahatchie R. Coldwater R. Loosahatchie R. Hatchie R. Forked Deer R.: South Forked Deer R. Obion R. 6 Total 14 66 53 21 33.24 34.00 33.00 33.33 33.44 33.50 34.46 34.00 34.03 33.54 CATFISH GENUS NOTURUS RAFINESQUE 279 TABLE 11.--Geographic variation of total preoperculomandibular pores in Noturus hildebrandi (sum of both sides) Number of Pores Total 1S L920 oh (227) 23 Drainage Homochitto R. 21.73 Bayou Pierre 21.68 Yazoo R.: Yalobusha R. 22.00 Yocona R. 22.00 Tallahatchie R. 21.33 Coldwater R. 19.33 Loosahatchie R. 19.00 Hatchie R. 20. 13 Forked Deer R.: South Forked Deer R. 20.00 Obion R. 19.69 Tot al 280 U.S. NATIONAL MUSEUM BULLETIN 282 TABLE 12.--Geographic variation of anal rays in Noturus hildebrandi No. of Anal Rays 12 Ae: AA IG AB Ay Drainage Total | Mean Homochitto R. Bayou Pierre 13.96 Yazoo R.: Yalobusha R. 16.00 Yocona R. 15.00 Tallahatchie R. 15.00 Coldwater R. 14.33 Loosahatchie R. 15.00 Hatchie R. 115292 Forked Deer R.: South Forked Deer R. 14.50 Obion R. 15: 10 Total 281 CATFISH GENUS NOTURUS RAFINESQUE I'S 99 3'p Ajfensn ‘[°¢ 03 O'F I°€ 04 1°S Alfensn ‘p'¢ 07 2'T 6 Ajerei ‘g Ajyensn ‘) usyo IL Atreordéy “gt 07 OT ATSOW 8g 04 9g Alrensn ‘TE 04 FZ eG 07 6h AlTensn “9g 0} LF GI 10 pF, A[jUenbey your ‘QT OF ET iods0q powre A[IAvoy ‘suo'T skei yUounooid oy4 ysnoiy} premiojy Suipuse}xe pue yoseid A[Tensy— apis. 04} JO BUI[}JOW oY} OJUI SpUs{q IO eSNyjIq uly jo jyey roddn siojue Spozyeouny doy ‘ystumoig ulj [esiop o10jyoq syods poio[oo weal0 Z@ Ssutsojous A[jensn ‘re[ndoi1il espe 1Ol1eyUy snonoidsuoo ‘peoig S[enpratp -UI Jo[[ewis Ul UawWOpqe uo AjJOlIa,Ue Soroyd -oyeulorgo ystumoiq ‘oz,er10sip Auew fYysTyIyM “UI OOT pus 09 Usemjzoq suowloeds Auvul S108 re'T SNSOW 317s g°9 903 O'S 9°¢ 03 G°g A[rensn ‘)°¢ 03 G'S 6 Seullyowos ‘g Ajjensn ‘) AjTorey OL Atreordéy ‘TT 09 6 ATISOW GZ 01 63 Alfensn ‘1% 04 1G 8h % PP Alrensn ‘ZG 01 6E PL 10 eT A[quonboey ysoul ‘9] 04 ZI IomMOIIeN pewre Ajsood o10ul S1041049 Aysup vore oy} ‘yuesqy' S[Tenprarpul Iopj[o ult ydooxe ‘yueurwoid ‘umoiq yred ulj JO J[ey Teseq 0} pouljuoo ATysou feuITINO UT oyUTJopUl JOYyyeI SysTAeIH ulj [esIOp e10Joq syOds yYSI] eATOUT} -Stp ou Soutds [esiop ye espe JOLIoqUy' qulej 10 MOLTEN uowopqe uo sor1oydoy -ewWormyo oyoIOStp Moy SAYSNp JO ule[d 09 ueYy} sso] qsow ‘*wu gy) UswIoods ysoBre] ‘[[eulg fe 2 Gees nteaestd 5 peso yydep e[ounped [epneg yywus] oulds [e10j00g :yoyou esodipe 04 UIsIO jesiop soueysiIp ul poddayg sfei ye1ojyoed yog soiod re[nqipuewo[nor1edoalg sei pepneo jey-reddy skei jepnedg sei peuy' ejounped [epneg sso9ooid [eiouny pue seuldg queoseio 10 pueq [epneopIW elounped [epneo jo puso jw ivg yoq0Tq uly osodipy e[ppes xzrep IIA ulj [eSiOp uO pueg eoBejIMs 1aMO'T (y{8u9] prepueys) ezIg Ieyoe rey dO sNsouwigst}s SNINION pup snisyNe{e sninjon fo uostndwoy--"e, |TAVL NATIONAL MUSEUM BULLETIN 282 US. 282 ulj [epneo woij oo1j A[Teou ‘yySue] Ul oYeIpoulIEzUT ‘YS quesqe Jo ‘4uIey ‘[[eUIg qourysip ATyensn Boqojo[W pues sol ppes eye[noeuumt AlTTeordAy eoBjMsiepun ‘yuesqy ulsreul 0} SULpUs}xe you ynq UTJ JO SPiIY}-OM} JOMOT UO fYSIYOR[q JO uMOIg yIeQq ‘WW OO, Seyovel ‘esrey snso1n{ ulj [epneo wo eo A[reou Sqioys ‘YysIy quowsid yrep Aq pesoloue Aj[oye,duoour uoyo ‘esre, ‘yueseig yourystp Ajjensn SeyojJO[G pus Se[ppeg SUIJ OTA -[ed Jo seseq uo useqo pues uswlopge SULIsAOZ uIlsFeW OF UL} SSOIO’ spusyxe A{TeordAy ‘yoRulq 10 uMOIg YEG "uu gy ysosre| ‘eyelpouliezUy ulj [e@pneo wo1j eelj A[Feou SyySusey] pus jYsIey Ul oyeIOpop emosqo woples ‘yueusid 1eyxrep Aq pesojoue ‘yuou -twoid ‘esrey Aj[ensp yourystp Ajjensn SeyojO[q pue Sel[ppeg SUI} [104 -oed usemjyeq A[joryo {ST@NPLAIpUL IJesIe] ul pemosgo ‘yueseig uldrFeUW OF JOU 4Nq UT} jo jrey ioddn o4yul spuseyxe {YsTy{Oe[q JO UMOIqG yIeEd ‘WU OO], Seyovel fesreyT ulj [epneo 04 peyooeuuod A[oyelopow ‘Buo] A[@ATyeTOI ‘MOTT uiezyed IO[OO I9q -wos oy} Aq pemosqgo A[reou ‘[[eus ‘yurey quoeutwoid you Se[ppes pue seyojzo[q ‘1equiog eye [NOVUIUIT A[[Tensn estmieyjo esovjMsiepun ‘synps Ul JUeSQe "WU OZ 0} Sunod ul yUssSeig s16q [eseq Of peut -uoo fystAeis io Aysng “WH T° LG ysosre] ‘*ulu Gg ueYyy Sse] A[Tensn ‘[Teug (paddais aie sJUDWIINSeoU Sq[npe ul uly esodipy uly [esiop jo yuod ul ysnl sjods 1yaI[ Jo 11eqg Apoq jo usezyed 10[OD (queseid ore ueyo YOIyM S|[eo JOU -sid jo suotyeryueou09 Jo SSulyoyO[q wopuvi ey} JOU) UsuIOpqe uO (Sse[xooe1j) Soroydoyeur -o1go YSIumMOoIq ‘puNnod ‘a[suts ‘pesiedsiq ulj osodipe uo reg (yy8ue] prepueys) ezig ieyoe reyO ‘seseyjueied ul sauiesjxa) sninjoN fo dnoi# snsotiny ay2 fo saigads ayz fo uostvdwoy---F1 ATAVL 283 CATFISH GENUS NOTURUS RAFINESQUE (8°S) 95e OF O'S ee (S27). G" On SHAT (DLL PL GD a oe 93: 0°S (0°) (1'h) 9g 07 Te (1') (g°e) 2°98 09 g°e (6'¢) 0'O0T - 8°6% (8%) LZ 01 SZ (83) (ee) LE 04 8% (12) (09) 8g 93 eS (TS) (9g) Se 10 FE (ee) (LT) 9T 10 ST (FT) eelios snj{d euids jo JojoWleIp ueyy Joyeois A[jensn yysue'y] euids Je10j,00d jo Jayo0uIVIp Z/L ynoqe yaesuo] ‘duo'T snso1n{ £3" 3 6% (PI) PTO Ce) kB 1 Casene encg age. (ian) 0°81 - 61S £98, Ve OGG LGD LZ) 9% 014 &S (eS) (ZS) 0G 93 9F (SPF) (eg) se 10 Le (Oe) (G1) bE 10 eT CT) eviies snjd eulds jo JoJouwleIp ueyy ieyeois A[vensn yysue'7] eulds [e104 -ooed Jo JojyowWVIp Z/T jnoge ysesuo] ‘suo'y Joye ol (972) "Geom gee (Te) ZS - PLE (93) GZ 01 8S {88 (Te) 8% 91 SZ (HS (9G) €¢ 09 6F (LF) (se) pe 10 Ge (Te) (91) St 10 FT (ET) eelies snj{d oud 7° JoyoulVIp ueyy Ajyensn yysueT eutds [e104 -oed jo JoyoulVip Z/T qynoqe ysesuo] ‘suo’'T snsowg14s 01 ge (2°) 98h - OE 01 GZ t28) 1% 1% (9% ZS (OS) (9¢) PE 10 ee (SE) (91) St 10 FT (eT) qyeys outds Jer1ojoed jo JayoureIp 0} Tenbe ynoqe yysUeT] eutds [e10jyoed jo roared, P/{L ueyi ssel Ajyeoiddéy yysue] ‘pedo -[@Aep [[eM yOU *yIOYG panurjuo')--(peddajs aie sjuaweinseaw ‘seseyjueied ur sauieaxa) sninjon fo dnou# snsowny ay7 fo sarzads ayy fo uostpdwoy--''1 ATAVL yysue] outds [e10joeg yysus] [esiopeig ulj [Vpneo diy 07 yoyou esodipe sourysiqg :yojyou esodipe 04 UISIIO [BSIOp dou SIP UJ yysue] outds [esiog yydep ojounped je pne :yysue] Tessopeid uy yysue] pie -pueys UL YydUe|] peop pomsvouwl suowtoeds ‘WW Sy}sdue|] prepueyg sfei j[ey-10mMo'T sei jyey-reddy sfei Teo], “uly Tepned evIqoWoA, sei jeuy (sseooid [e1owny) wniyyte]o jo sseooid 10110}S0g eutds jerojoed JO suotyellos IOlIoyuy: Jeyoe rey 19 298-943 O—69 NATIONAL MUSEUM BULLETIN 282 U.S. 284 (3°01) (eo) Chee GuceGheows (0°) (0°S) Sw j (9°9) (e°9) J 10 9 Mt 10 9 (0'8) (0°8) 642 g do » (9°31) (¢°6) 91 901 Zt 978 (1°13) (Z'1e) O€ 97 81 LE 43S (9°13) (S°TT) 9% °F FIT LI 8 (o°9) (¢°3S) 09 972 9S 97 0G vuerIpuy | Uesiyow snuisk3 snu1k3 IT -OL ol Li 8% 69 ‘eu IpU] pliqéy IL-TT 0& 61 €9 yIoX MON PHgqAH Eben Ie ST 8g wes tyo IW Z Pugh Ea al ve €g ue S1yo Iw IT pNgqhky (L‘TL) CED) SE IOTL SI 01 (o'T) (O'L) I Z 10 | (6°) (6°2) gio » 6% 2 (0°6) (0'6) 6 OI 978 (e°ST) (e°91) LT FT 61 FT (1°81) (9°LT) 1S O}LT FS OF GE (g°Sd) (6°F3) 9% FS 66 97:04 (¢°6S) (6°89) 09 93 8GS #9 01 g¢ eueipuy | UBSIyorW sniniu sniniwu soeiod le[nqtpuewo[noiedosig seiod jesvuieqyuy sei je10ojoed yog sfe1 O1Aleg edulis lemory peyousig eiduis reddy TS1OL :sABI [epneg ioyOVIeEyyD [*pepnpour aie omjyonijs ported JO SyusWe[9 YyIOG wo ejeq °*Sesayjuored ur sueow Aq paMoOT[OF oe UOTELILA UT sosuey °(praqdy 241 YIM peyoeay[[oo ‘suaoumtoads [E ‘PEs66l NNSN) BueIpUy {(1eATY GesseMeETYS ‘sueuTToeds CT ‘2089ET ZINN) UPSIYoIW ‘snuzikF snumjony *(E8S661 WNSN) PUuetpUy ‘puqdy *(FL Soh AD) WOR MON ‘PHGAH *(ZOELOT ZINA) UEFIyOIN ‘Z PHGAH “(PLZLST ZINN) UeSTyONW ‘T PHG4H *(pHq4Y ey1 YITM paroa[joo ‘suow -roeds ‘88S661 WNSN) euerpuy f(4eary uoiny ‘suoumoeds 7) ‘PoQ801 ZWD) ueSIYyoIp ‘snunim snunjoyy :sMO][O} Se aie eIEP eUL) saizads quaspd ayz yz1m ‘snintw sninjoN X SnuIIAS sninjoy ‘sprgky fo uosundwoy--"cT ATA L 285 CATFISH GENUS NOTURUS RAFINESQUE » (6 'P) euoN SsuoN re[n deli] re [nde] re [nseily re[nN sed] 6 91¢@ euids [wiojoed jo SBLISS IOlISJSOg [euluey, [euluNey, [BUIWIe], [B@Ululeyqng [eUIUe}qng |BuUTULa}qng pepnyjouy pepnyouy mel Jamoq euelpuy | uBaryorpy vuelpuy eae MON uBsIYoIpW UBSIYOIW eueipuy | ussryopy leyoRreyo snuiska snuik3 priqAy piqAy @ pugqky I prqdéy sniniu sniniw [ ‘Pepafour ose samnjonns paired jo syuaueye Woq Woy Bie ‘sesoyjuored ul sukow Aq pamoT[oy ase uOTeLeA UI soSuey *(priqdy ay) YIM paroajjoo ‘suowroeds Te ‘pgs66l NNSA) BUBIPU] +(1OATY PessemeryS ‘suawtoads Gy ‘2OgOE ZININN) UeStyoIW ‘snugsAF snanjgoxy “(€8S661 NNSN) Buerpuy ‘puqdy “(pI SZh ND) OK MON “PEGAH *(ZOEL9T ZA) YeStYONN “Z PLGA “(PIZZST ZA) YS1yOIW ‘TE PEGA *(praqXy ou YITM paroayjoo ‘suautoads b ‘88S661 WNSN) euelpuy s(seaTy uoiny ‘susutoeds Zz) ‘ZoQgoT ZW) uestyorp ‘Stunzw snunzoxy :sMmo[[o} se aie eyep oul] penunuo)---sazvads quasnd ay2 y21m ‘sniniu snimioy X snutsh3 sninjoN ‘spughy fo uostspdwoy--s, A TaVL 286 U.S. NATIONAL MUSEUM BULLETIN TABLE 16.--Comparison of the hybrid (UMMZ 137904), Noturus exilis X Noturus miurus, from the Poteau River, Oklahoma, with three species of Noturus [The data are from eastern Oklahoma specimens as follows: Noturus miurus, counts from 12, Poteau River, UMMZ 137902 and measurements from 11, Illi- nois and Poteau Rivers, UMMZ 137864 and 137902; Noturus exilis, measure- ments from 20 and counts from 32, Illinois River, UMMZ 137863; Noturus noc- turnus, counts and measurements from 12, Poteau River, UMMZ 137903. Ranges of variation are followed by means in parentheses. Proportional measurements are stepped. Data from both elements of paired structures are included. } Character miurus Standard length (mm.) 33.0 to 52.4 (44.0) 5 Le Caudal rays: Total 57 to 62 (59.6) 56 Upper simple 21 to 27 (24.9) 24 Total branched 15 to 20 (17.6) 17 Lower simple 16 to 19 (17.1) 15 7to9, usually 8 (8.0) 9-9 1 (1.0) 1-1 11 or 12, usually 11 10-10 (11.0) Soft pectoral rays Internasal pores Preoperculomandibular pores Posterior serrae of pectoral spine | Regular; 5 to 7(6.2) Regular; 7-8 Lower jaw Included Subterminal Pigmentation of lower jaw Light Light Black or darkened borders of fins | Prominent Obvious Head length in standard length 3.2 to 3.5 (3.4) 3.35 In predorsal length: Dorsal spine length 2.4 to 3.4 (2.7) 3.4 Pectoral spine length 1.8 to 2.6 (2.2) (2.6) In distance dorsal origin to adipose notch: Predorsal length Head length Caudal peduncle depth Distance adipose notch to caudal fin tip 1.2 to 1.4 (1.3) 1.5 1.6 to 1.8 (1.7) 1.85 4.1 to 4.8 (4.4) 5.0 1.1 to 1.5 (1.3) 1.65 CATFISH GENUS NOTURUS RAFINESQUE 287 TABLE 16.-Comparison of the hybrid (UMMZ 137904), Noturus exilis X Noturus miurus, from the Poteau River, Oklahoma, with three species of Noturus--Continued (The data are from eastern Oklahoma specimens as follows: Noturus miurus, counts from 12, Poteau River, UMMZ 137902 and measurements from 11, [lli- nois and Poteau Rivers, UMMZ 137864 and 137902; Noturus exilis, measure- ments from 20 and counts from 32, Illinois River, UMMZ 137863; Noturus noc- turnus, counts and measurements from 12, Poteau River, UMMZ 137903. Ranges of variation are followed by means in parentheses. Proportional measurements are stepped. Data from both elements of paired structures are included. | Character Standard length (mm.) Caudal rays: Total Upper simple Total branched Lower simple Soft pectoral rays Internasal pores Preoperculomandibular pores Posterior serrae of pectoral Spine Lower jaw Pigmentation of lower jaw Black or darkened borders of fins Head length in standard length In predorsal length: Dorsal spine length Pectoral spine length In distance dorsal origin to adipose notch: Predorsal length Head length Caudal peduncle depth Distance adipose notch to caudal fin tip 43.1 to 58.5 (50.0) 46 to 51 (48.8) 15 to 19 (17.0) 16 to 21 (18.2) 11 to 16 (13.6) 8 or 9, normally 9 (8.5) 1 (1.0) 9 to 11, usually 10 (10.0) Usually regular; 5 to 10 (6.9) Terminal Never heavy Usually prominent 3.8 to 4.4 (4.0) 3.3 to 4.8 (4.1) 2.6 to 3.6 (3.0) 1.7 to 2.1 (1.9) 2.1 to 2.7 (2.4) 4.8 to 5.9 (5.2) 1.9 to 2.5 (2.2) nocturnus 44.0 to 81.2 (62.9) 56 to 62 (58.3) 20 to 25 (22.3) 19 or 20 (19.4) 15 to 18 (16.6) 9 or 10, usually 9 (9.1) 2 (2.0) 10 or 11, usually 11 (10.9) Not distinct or uni- formly developed Included Heavy Usually indistinct 3.3 to 3.8 (3.5) 1.4 to 1.6 (1.5) 1.8 to 2.1(1.9) 4.1 to 5.2 (4.6) 1.5 to 1.8 (1.6) NATIONAL MUSEUM BULLETIN 282 U.S. 288 90 "ST LL ‘vt g0 ST Geral Size Le “vt 6L FT O08 “ST ST tT 96 “ST va FL OF “ST 66 FT v9 “OT 61 “ST 66 (0G 6L GG 66 “LT vO “LT vO OT GS *6T 60 “LT uvoW skey Jeuy jo 1oquinn sninjoy fo sai1zads ay2 u1 skpi oun fo uo4ngiysip Aouanbasy--"2, ATEAVL sazvan7f siuurdianif sniniu snso1n{ snqiunu snsow3izs snp12v7d sniayjnaza 1upwynD.7 supgaj)a 4210q9]D 14a]10q aera. snapjf 19199]13 snapyd s1igaunf s1u3gisul snuinjoou snyjuvonjidaz S272 149UY ID] snu1kZ setoedg 289 CATFISH GENUS NOTURUS RAFINESQUE 182 L6°L 90 °8 T6°2L 9L°L 66 °L v0 °8 8 °8 668 v6°8 06°L 89 °8 LL°6 ee 6 69 °8 61 6 G6°8 16 6 69 °8 8L°8 G08 OMwW Fw g wo HNON 1 m4 ' ' ' - N — oO oD AN ' ' Ye) oc co a — wo ' ' ' 1 ' X= oO mN - © 10 Nw or = he | ' ' co N oD xs N wn —H mm skey [e10j0eq 40g “ON skey olAlod 4agvanj{f si1uurdian7{f sniniu snso1inf snqiunw snsow 313s snpran)d sniayqnaza 1uUDWINDsI supga}a 42a3Dq]D 1497104 spunsgepTiy. snanjf 19499]12 snapyd s1iqaunf siugisu2 snuinjoou snyquvonida] $4/2%9 149UY ID] snumkg setoedg (pequnod 210M sapts yI0q) snmioN fo sa19ads ay2 u1 squnoo Avi-u1f pasiod fo uo1jnq14ys1p Aouanbasy--"gT ATAVL 290 U.S. NATIONAL MUSEUM BULLETIN 282 i i | TABLE 19.--Frequency distribution of total caudal rays in the species of Noturus Species gyrinus lachneri exilis leptacanthus nocturnus insignis funebris phaeus gilberti flavus hildebrandi baileyi albater elegans trautmani eleutherus placidus Stigmosus munitus furiosus miurus flavipinnis flavater 39 40 41 42 ocala NS mae aa eee een ee 43 19 10 i 20 11 1 1 10 23 26 18 25 4 3 38 44 20 20 18 12 23 26 29 10 2 10 24 25 Number of Caudal Rays 44 45 46 47 48 49 50 51 52 53 54 55 56/ 16 17 13 2m = s = o% 18: 145 4 11 22 7 1 12 15 14 12 13 23 20 18% CATFISH GENUS NOTURUS RAFINESQUE 291 TABLE 19.--Frequency distribution of total caudal rays in the species of Noturus--Con. Number of Caudal Rays Species No. Mean 57 58 59 60 61 62 63 64 65 66 67 lgyTinus 167) 26's 26-28 12 (24-10) “Se 652 8 -<-3 57.96 Se ner eee poles cot Py N ean” hole one a a eee 58.18 Let: ae pa ae Ss Pe OS: es as eR a= 50. 49 eG Sie tet kn eee ca Sow EE rc a as 51.34 | nocturnus Bie 19) 18 20018.5 2 Buse MeN Te 58.94 insignis Beit 4 S85 DIO Teed ed et 61.10 Vunebris Geo s-Ge cs ae maa ine) aa Ie i 54.13 . iphaeus Seay cites es ade) te 54.10 gilberti Ee ee eee ee 50. 58 flavus oS! tes VS sok O12 Oe 5 AL eZ 60. 24 hilde brandi Sima ce eh eee asaya eal 45.99 baileyi See oie Ne A elas Io Bak eee ts 45. 00 albater iii emg ees st ey sate, doe ba 49.27 eons eM OT ot oe ie 6 an S| wee a 49.82 Bes eran See CEN “Wate Wd oe ch Coen ne cs ie 48.94 eleutherus SA SS Cee ee 9 ei RG oink |i oe Wen rae = eM 45.90 placidus HOGER, Va eos abi ak yale Pee ie 54.32 |stigmosus USS pies KA) mea tees ER aul ee 51.04 munitus Senne Ean a | VS me vere a4.) Sree 47.63 ifuriosus Miah A a yas pia ee gl Acieene le ais 55.48 iorue Oy ASO Br woe te Jule a, uy a 59. 43 Wevipinnis pe ahh pam) eel eai ge, "bch Maat 58.54 ) 55.00 flavater SIA ellie ela See ON eh ge Pia ticd ' | iH | i 282 U.S. NATIONAL MUSEUM BULLETIN 292 9€ “6G 80 “SE 66 GE GP 66 8S 0G 91°93 60 “8G Cv 6 OS “Sc G6 SG £0 SG 09 “83 68 8% LE RS 60 “SG 60 °L3 VIL] Gere 68 0 1G “96 €E “9G 161 Sy TE ueoyy 66 86 LZ@ sXey [epned jrey-1eddyq jo 1equny snmjoy fo sazoads ay3 u1 skp jppnvo f{7py-1addn fo uo1nq14381p Aouanbasy--"0Z ATAVL 1azvan){f siuurdianj{ sniniwu snso1n{ snqiunu snsow3is snpion7d sniayynaza 1UDWINDIZ supgajza 4240q]D 1ha710q ipupiqap71y snanjf 1949q]13 snanyd sliqaun{ siugisu1 snuinjoou snyjuvovjdaz $212X%9 149UY 9D] snu1sk3 seloedg 293 CATFISH GENUS NOTURUS RAFINESQUE OF “9G 9V “9% 6S 9G v0 °9% G0 °&% O€ “0G G3 9% OS °33G VV 3 98 °&% 96 SG OF “1G LI “33 ST 66 LG °SS L0°L6 G6 96 VL “66 10°86 OL “SZ VI ‘VG L696 uvoW skey [epned jiey-1emo7T jo soquiny sninjoy fo sarzads ayz ur sXpi popnvoa fypy-1amo7 fo uo1qnqisysip Aouanbasy--"[Z ATAVL sotoedg Jaqpan7{ siuuidian7{ sniniu snso1in{ snqiunu snsowgizs snp1op)d sniayinala 1uDwWyNDsy supdaja 42a37Dq]D 1497104 ipupigap]1y snavjf 1949q9]13 snapyd sligaunf s1ug1su1 snuinzoou snyquvonida] S7]1%9 149UY OD) snuik3 U.S. NATIONAL MUSEUM BULLETIN 282 294 v6 OT 18 OL SOP it 16 °OT 16 “OT 68 “OL 86 “OT 66 °6 16 ‘OT 66 “OT 66 “OT 00 “TT 9v “OT 00 TT LL ‘OT 66 ‘OT FOES 86 “OL 99 OT 88 OT 60 “OL 00 ‘OT 61 ‘OL uvoW v 9 6 a a L 8 v I o> Solog ee so10g [eseuiszU 4aqpanjf siuuidianif sniniu snso1in{f snqiunu snsowdi1zs snp1o0)d sniayynaya 1uDwWINDsy sup3a)a 427Dq]D 1ha]104q SPUD rad snap} f 29429723 snapyd slugaunf s1u3zisul snuinzoou snyjunanidayz $2]2%0 119UYy Id] snu1sk3 setoedg (peiunod aJaM septs yI0q) snmjoyj fo saz9ads ay2 u1 sasod youn9o-ksosuas fo uoynqi4sip Auanbasy--"7Z ATV L 295 CATFISH GENUS NOTURUS RAFINESQUE G6 ‘VE le ve GOS 89°FE OF TE 69 ES 69 SE 99S VG EE 6F SE 08 “98 O08 “ee vS es 1G ‘68 08 “LE 09°SE 60 9€ 66 8 19 9 v8 VE €8 8 VOLE OF VE ugeW eviqeyIeA JO JoquinN (pepnpoxe ole seiqo}l9A JOoIlojue pesny eq) sninjon fo sazaads ay2 uz squnoo jv1gajsaa fo uo1jnq1ysip Aouanbaiy—"eZ ATAVL 4aqpanjf s1uuidianz{ sniniw snso1n{ snqiunu snsow313s snp1o0)d sniayynaza 1uDwynDdsq sup3a7a 1270qQ]D 1£9]10q p PUD ya, snanjf 22499118 snapyd siiqaunf s1ugisu snuinj2ou snyjuvopnjda] S22 %2 1 aUuyond) snus setoedg NATIONAL MUSEUM BULLETIN 282 US. 296 e 6s - - - - T - - - - - ee eg 2 pedatte 3 ee See ty es eee eee ole BF ope eee NET aA en eee eer ee roe in i IG - - - - - - - I - - - - - - - - ¢ G - = - - = = - - I - - - - - ° G - - - - - - - - _ - I - - - - - ee Oe ee a PUR Mth t as teers Rae ee Beene gies ee re ee - - - P L - - - - - - - - I G - - - - - peu Pete Gee ee eet ee Fe Me ete Ge Fah Fe eee ee a Se me ertieiGent oo cen re Pes ME ' ' ' ss co - oO ' ' ' Te O& BGe 8G— 16> =496 —S56° —VG. “66. -Gc sVIqoWIoA [epnecoid jo ‘ON eviqeyWoA [epned jo ‘on (pepn[oxe oie 9eIGIyI9A JOTIojue pesny 24) Jaqpan)f{ siuuidianz{ sniniw snso1inf snqziunwu snsow3sizs snpio0j)d sniayynaja 1UDWINDY supgaja 142]0q]D 1437109 ipupiqap7 24 snanjf 1499]13 snapnyd s1iqaun{ s1ugisu1 snuinzoou snyquponida] $1]1x% 9a 14aUuYy dD] snu1sk3 soloedg snimjoN fo saioads ay2 ut squawaza jpiqaqiaa JDpnva puv yopnoovaad fo uoinqi4ysip AQuanbasy--"$% ATAVL 297 (é)T & = p> Cc - wn - A , fC 2 eB = fe g Pp S o 9¢ A 9¢ : Le | 2 Oo a f fa I 'S : Be ; ; % : : : : ee Sat eal. '> ere ‘ ; ; joc te et _ ‘ . 7. - —— . #e i z : - : F ao2 ; dragnet iuagad fase is + = Daewice st Ga. 7or? =: ae F digest ist winlifxat © ae we Pee ts te Ole DEE Hx 3 a : z tier Dec aig iat pee ia et a a ee eee aT Ec SAS AMt SSS See ee Mee ae S- OHt ae oi net iy : sugar eet & et Fone a ; ot ray Sa =? ae Reo cage ess ean : s a sae bo Sie eh oe ottieg 3 dens ace ce nee Ning oh ae Na bs ache SL ee eee ae eee oe, rt 3 ~ . gay pooh Teterg Seaees oe 7 a ope c= a Sy: . Tis Set (tes Se a ee eae a9 oF : : Gtr tee Seow ae a - . oa ' a aban j =e. eS oeeee f rss te > - es | a ea BEE | | fee gra eg ane iba ng ache aaron —- = + : ; : 7 ae mee a 4 + 3 be Sf ; wee, © Lr SeGiEs 4 eet me wh 33 pccwapies < 3-- SEBSST . eave Mb ieee | Gere - a 2 - iad ' moe edad seaaype( Serie Jeers asa s Veo Ser F a | ese] aye ch gu peter es fnlm ne a 7 - 7 aie — Se = — le - eis oe ae i > Pratien’ {i po pe. Penis Pa we Ohi peetRR Sts ire: ole et: Oa; . oo yp wen Si hisF: oan: ies See ee Seattle fe Py ; ; PAN Pare Be a as eer zs has Peat Bel 54ST) BO peat Rabem LitnwloY i ‘8 be mat ait yh: ink) bre Me tbnartaiisy “+ 7oar ~ A PLATE 1 Sensory canal system of two species of Ictaluridae (slightly diagrammatic). Figure 1. Dorsal view of Noturus exilis Nelson. Ficure 2. Dorsal view of Pylodictis olivaris (Rafinesque). (10=Infraorbital canal; LAT—Lateral canal; L1 and L2=First and second infraorbital canal pores; Nl, N2, and N3=—First, second, and third supraorbital canal pores; OM9, OM10, OM11, OM12=Ninth through twelfth preoperculomandibular canal pores; POM= Preoperculomandibular canal; SO—Supraorbital canal.) Ll and N2 are fused and there are ten preoperculomandibular pores in Noturus exilis. In Pylodictis, these pores are unfused, Nl and N2 are located more anteriorly, the infra- orbital canal extends diagonally backward, and there are twelve preoperculomandibular pores. The canal arrangements in the species of Noturus, Prietella, Trogloglanis, and Ictalurus are essentially as in Noturus exilis, but the number of preoperculomandibular pores and the fusion of L1 and N2 varies (see p. 11). Satan eurystomus Hubbs and Bailey is much like Pylodictis, but N1 is more posterior in position and L1 and N2 are fused. U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 1 Fused Nl SO Anterior naris Posterior naris » Nasal barbel > Maxillary =< barbe! Pectoral Ke Anterior naris Nasal barbel “A Posterior *— naris ~\, Maxillary Xx barbel ¥ Pectoral m,/ fin oe 298-943 O—69——21 Premaxillae of representative species of Noturus, showing their ethmoidal attachment and some of the variations in tooth arrangement. [Drawings by Dorothea Schultz.] Ficure 1. Noturus gyrinus (USNM 174905, Michigan). Ficure 2. Noturus exilis (USNM 174904, Kansas). Figure 3. Noturus flavus (USNM 174908, Michigan). Ficure 4. Noturus stigmosus (USNM 174906, Michigan). ‘ee U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 2 PLATE 2 PLATE 3 Left pectoral spines of members of the subgenera Noturus and Schilbeodes, and of hybrids among the species of Noturus, all drawn to the same scale. [Figure 2 was drawn by Dorothea Schultz; all others by W. L. Brudon.] FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE Ficure 10. FIGURE FIGURE it; g Noturus gyrinus (Mitchill) (female; SL 52.8 mm.). UMMZ 152504, Missouri: Clear Cr., 2 mi. W. of Knobnoster, Johnson Co. Noturus lachneri, new species (male; SL 69.5 mm.). USNM 165901, Arkansas: trib. of Saline R., 1 mi. NW. Benton, Saline Co. Noturus exilis Nelson (female; SL 55.9 mm.). UMMZ 111413, Missouri: Niangua R., U.S. Hwy. 66, E. of Marshfield, Webster Co. Noturus leptacanthus Jordan (female; SL 51.8 mm.). UMMZ 110975, Florida: Spring Cr., 3 mi. SE. of Marianna, Jackson Co. Noturus nocturnus Jordan and Gilbert (female; SL 58.2 mm.). UMMZ 139501, Missouri: St. Francis R., near Saco, Madison Co. Noturus insignis (Richardson) (female; SL 62.3 mm.). UMMZ 109631, New York: Tioughnioga R., Itaska, Broome Co. Noturus funebris Gilbert and Swain (female; SL 70.4 mm.). DBUF, Florida: stream at Portland, 4 mi. W. of Freeport, Walton Co. Noturus phaeus, new species (female; SL 53.5 mm.). UMMZ 161058, Mississippi: Mimosa Spa, 1.5 mi. N. of Waterford, Marshall Co. Noturus gilberti Jordan and Evermann (female; SL 64.5 mm.). CU 20767, Virginia: Roanoke R., Glenvar, Roanoke Co. Noturus flavus Rafinesque (female; SL 71.3 mm.). UMMZ 165842, Michigan: Huron R., below North Territorial Road, Washtenaw Co. Hybrid: Noturus exilis Nelson X Noturus miurus Jordan (male; SL 51.1 mm.). UMMZ 137904, Oklahoma: Poteau R., Slate Ford, Le Flore Co. Hybrid: Noturus gyrinus (Mitchill) X MNoturus miurus Jordan (sex?; SL 108.8 mm.). UMMZ 157214, Michigan: Huron R., N. of Milford, Oakland Co. U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 3 12 3mm PLATE 4 Left pectoral spines of members of the subgenus Rabida, all drawn to same scale. [Figures 2, 3, and 6 were drawn by Dorothea Schultz; all others by W. L. Brudon.] Ficure l. FIGuRE 2. Ficure 3. Ficure 4. Ficur_ 5. Ficure 6. FIGuRE 7. Ficure 8. FIGuReE 9. Ficure 10. Ficure ll. Ficure 12. Ficure 13. Ficure 14. Ficure 15. Noturus hildebrandi hildebrandi (Bailey and Taylor) (female; SL 38.8 mm.). UMMZ 155338, Mississippi: Brushy Cr., 4 mi. NE. of Coles, Amite Co. Noturus hildebrandi lautus, new subspecies (female; SL 40.0 mm.). USNM 193470, Tennessee: North Fork Obion R., Henry Co. Noturus baileyi, new species (female; SL 42.7 mm.). USNM 201601, Great Smoky Mountains National Park: Abrams Cr. Noturus albater, new species (female; SL 53.2 mm.). UMMZ 151407, Missouri: Flat Cr., 0.5 mi. S. of McDowell, Barry Co. Noturus elegans, new species (female; SL 52.2 mm.). UMMZ 155526, Kentucky: Fallen Timber Cr., Hwy. 90, 8 mi. SE. of Glasgow, Barren Co. Noturus elegans, new species (female; SL 59.2 mm.). USNM 201600, probably Duck River system, Tennessee (see p. 143). Noturus trautmani, new species (male; SL 42.4 mm.). OSU 5914, Ohio: Big Darby Cr., 1 mi. S. of Fox, Pickaway Co. Noturus eleutherus Jordan (female; SL 59.4 mm.). OSU 5157, Ohio: Little Miami R., Red Bank, Hamilton Co. Noturus placidus, new species (female; SL 44.0 mm.). UMMZ 167654, Kansas: Neosho R., U.S. Hwy. 50, near Emporia, Lyon Co. Noturus munitus Suttkus and Taylor (female; SL 44.0 mm.). USNM 197708, Mississippi: Pearl R., 2.3 mi. E. of Sandy Hook, Marion Co. Noturus stigmosus, new species (female; SL 57.5 mm.). UMMZ 108063, Michigan: Huron R., above Dexter, Washtenaw Co. Noturus furiosus Jordan and Meek (female; SL 69.2 mm.). UMMZ 165855, North Carolina: Raleigh. Noturus miurus Jordan (female; SL 58.0 mm.). UMMZ 165840, Michigan: Huron R., 1.75 mi. NW. of Dexter, Washtenaw Co. Noturus flavipinnis, new species (female; SL 63.8 mm.). SU 4606, locality unknown; thought to be from the upper Tennessee River basin. Noturus flavater, new species (female; SL 57.1 mm.). UMMZ 152091, Missouri: North Fork White R., Tecumseh, Ozark Co. U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 4 BULLETIN 282, PLATE 5 U.S. NATIONAL MUSEUM "OD euryeg “uoquag “MN TW T “x7 oules “qiy :sesueyIV ‘TO6S9T INNS “(wu ¢°99 TS ‘ayeu fodAqeied) satoads mau “2vauyrv] snunjony ‘Od vOIsN], ‘AqIZ) sse_ jo “TU CT “YL sseO IG YUON sues “6ECOlT ZININA “CwU F°6Z TS Sa]eu) (Tyo) Snussks snanjony ‘7 AUN ‘T Tansy BULLETIN 282, PLATE 6 U.S. NATIONAL MUSEUM ‘OZ o][IAurin “ST AMH SD “CU JBL FeuporeD YON ‘“O6SZFI ZININNA ‘Cwu Zz ‘SL 1S‘ ‘OD sejsnoqg STN “AMH OU OYA YIOY YON snossipy *79Q7ST ZININN * (uu % a[eW) (UOspreYoIYy) srudisus snunjony ‘09 'IS foyeur) uosjayy sajiwa snanjony ‘7 FANS ‘| TANI BULLETIN 282, PLATE 7 U.S. NATIONAL MUSEUM "od TTeysseyyy ‘sdutsdg Ayjopy Jo “g “tur g “psojiore AA JO “N “TU GT ‘edg vsounyy :1ddississipy ‘gsO19T ZININA “(mm [ZZ TS ‘eyeuraz fodAqesed) sarsods mou ‘snavyd snanjony *7 AUNOIY *Oc) BSOY vIULS “6g pur 7/8 sAmPT UOINAL 28 Y92IO SVP HOLT “SESS. ZININA “(WU S76 TS S9]eUIE}) UIeMG pure IOqID stuqaun{ snunjoNy “] TUNDTY BULLETIN 282, PLATE 8 U.S. NATIONAL MUSEUM ‘Od) puowyory ‘eysn3ny jo "Ag ‘TU Z ynoge ‘g/ “AMP “SQ “ID AY2OY :¥181094 ‘8/891 ZININN *( Jo "a “Wr T “ID ysOg :stoull]y] “E8gsol ZININN OD IEE Ml (‘wut G 69 TS ‘ ‘WW $19 TS ‘aew) uepsof snyquvsnjz¢az sninjony ‘qeuiey ayeWla}) WoqiIg) pue uepsof snusnpzou sninjony Z aUNolg ] aun BULLETIN 282, PLATE9 U.S. NATIONAL MUSEUM Od) MeUsTYse A “peoy [e107 -113.], YON MOjaq “Y uOINFT :uesIyIT] ‘Z7F8S9T ZININA “(wu ¢°88 TS fayewoy) onbsouyey snavy sninjoyy *7 AUNT Od) 2youvoy ‘IeAUI “Y eyouvoy :eruris, ‘Z7ZSET ZININA ‘(ww 9°ZZ TS fo]eu) UULWIOAY pur uvp10f 1uaqzid snanjony “| AMNOIY BULLETIN 282, PLATE 10 U.S. NATIONAL MUSEUM ‘0- Aruay ‘69 “AMY “Y YOIGO YOY YON :sassouuay, ‘S9910Z INNSN “(wu ¢'OF TS fe]emo} fadAjojoy) satsadsqns mou ‘snyny] ipuvsgapjiy SninjoNN *7 ANNO ‘0- ony ‘s2]0OD “ANN ‘Tw F “ID Aysnag :iddis -SISSIIA ‘SEESST ZININA ‘(WU O'6E TS f2]eu fodAqesed) (s0]Av], pur Adjieg) tpunsgapjry tpuvsqappry snanjoxy “[ THAOTY 2 reerpee BULLETIN 282, PLATE 11 U.S. NATIONAL MUSEUM *O*Z Aour y, “‘yqAsi0,J “OSI “AM “UY OYA *HMosstpy ‘TZT IST ZININA ‘(ww 7°19 TS fayeu fad A1ojoy) saiseds mou ‘vayvqzy snumjoxy +7 TANDI C | IC) surviqy :yieg JeuoneN urezunopy AYourg eaId ‘ZQ9TOZ INNSN ‘(wu g°cF TS foyeur ‘ad Aq0]0Y) sarsads mau ‘1hapiyg snunjoxy ‘| TANI BULLETIN 282, PLATE 12 U.S. NATIONAL MUSEUM ‘(¢p] ‘d aas) vassouuay, ‘uaqsds roary yong Ajqeqoid 4 009102 WWNSN. ° ( wur 6°6S ‘IS ‘Od UsIreg ‘Mosse Jo “AS Tw g ‘C6 “AML “I JOqUIL], uaTe yg :AYONUEY “76SZ9T ZININA ‘CW [FP TS feyew fedAjojoy) saisads mou ‘sumdaza snanjoay ‘o[BWo}) Ssupsaja sn.injo Ay ‘7 FANDIT ‘| TANI U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 13 Ficure 1. Noturus trautmani, new species (holotype; male; SL 44.1 mm.). UMMZ 187098, Ohio: Big Darby Cr., 1 mi. S. of Fox, Jackson Twp., Pickaway Co. Ficure 2. Noturus elegans? (male; SL 51.2 mm.). UMMZ 131386, Tennessee: Dunn Cr., at Jones Cove, Sevier Co. [The dorsal and anal fins are not dark margined, as shown here; instead a dark background shows through these relatively clear fins.] Ficure 3. Noturus elegans? (male; SL 32.0 mm.). UMMZ 165877, Alabama: Piney Cr., Limestone Co. BULLETIN 282, PLATE 14 U.S. NATIONAL MUSEUM "0—- OjoI9g “dm, ARID “Y O1OIDg :01YO “FZ/¢ NSO “(WU ZFS TS fa[eUI) UepIOL snszyjnaja Snunjony *Z TANI *A[iOlajue Apog WO} paqassip sem UY [PU OYJ, OZ URATT[NG ‘(qUsUIpUNOduUII aI0Jaq) WIep UOYs|OFT YINOG aAoqe "IW 67° “Y UOIS[OFY YO YINog :sassauuay, ‘¢/¢/ ST ZININN “(WU g'cP TS ‘ayeur) UepsOL snsayynaja snanjony “| TANI BULLETIN 282, PLATE 15 U.S. NATIONAL MUSEUM O- UOAT ‘eIIOdwy 3vau ‘Og AMP] “S*-Q) JO *§ asnf “y oysoayy :sesuey ‘¢cg/9T ZININA “(mu O'¢sp TS ‘ayew fadAqQo0j0y) satsads mou ‘snpisnj7g snanjoxy “7 aUNDIg "OD PARMA “Y PMI] :BUlpOIeD YON “F88soT ZININAN ‘Cuw ¢-¢9 Ts ‘ayeu) yooy] pur uepsof snsorsnf snamjony *] TANI BULLETIN 282, PLATE 16 U.S. NATIONAL MUSEUM ‘Oy uOoLIe yy ‘yooyy Apueg jo" “Tu ¢°7 “YY pueag :iddississiqj ‘802261 INNSN “(ww [op TS ‘ayeu fodAqe1ed) s0jAey, pue snyqINg snpunu snunjonr *7 AUNT MOTE “YY UOINFT :uesiYysIPAy ‘ O-) MvUazYysEM ‘proy [eOWIIIT, YON CF8S91 ZININA “(uu 6°99 Tg Sayer fadAjofoy) satoads mou ‘snsowSys snunjony *] ANSI U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 17 Ficure 1. Noturus furiosus Jordan and Meek (male; SL 65.5 mm.). UMMZ 165884, North Carolina: Neuse R., Wake Co. Figure 2. Noturus placidus, new species (holotype; male; SL 43.0 mm.). UMMZ 167653, Kansas: Neosho R., just S. of U.S. Hwy. 50, near Emporia, Lyon Co. U.S. NATIONAL MUSEUM BULLETIN 282, PLATE 18 Ficure 1. Noturus stigmosus, new species (holotype; male; SL 66.9 mm.). UMMZ 165843, Michigan: Huron R., below North Territorial Road, Washtenaw Co. Ficure 2. Noturus munitus Suttkus and Taylor (paratype; male; SL 46.1 mm.). USNM 197708, Mississippi: Pearl R., 2.3 mi. E. of Sandy Hook, Marion Co. BULLETIN 282, PLATE 19 U.S. NATIONAL MUSEUM ‘ysied eS eT ‘euaf :eueisInoy “(LS-1¢ ‘ON Josvaig) pue uvavpqd) ZININA ‘(WU ¢ *O7) MEUILYSe AMA "6S TS ‘oyeur) uepi0f snuniw snunjoNy *Z AUNT ‘yayxoq] eAOqe “TUT C/*] “Y UOINFY :uesrysPy “OFSS9OT ZININA “(ww FZ TS ‘ayeu) UePIOL SNintU SNINJOAT “T ANNO BULLETIN 282, PLATE 20 U.S. NATIONAL MUSEUM "OE Areg “OjplAsseD jo “AN ‘Tu 7] “6EIN “AMH “ID 1e[Y stNossipy ‘ZZ7ETST ZINN “(Cur 6'Z01 TS faye fod. Ayo]0y) satsads mau ‘uaqvany snanjony “7 AANOIT [uy a2 Jo suorziod jusIedsuvsy 243 ysnoiyy ZuImoYys PUNOISyDeq Yep dy} 0} onp aie UY yepned oy} Jo suIsIeW JoMO] pue Jaddn ay} Iv seorE Yep Om} ay, ],] ‘uoquID “ID soulpyy :oossouuay, “TO8E9T INNSN “(ww Z'¢9 TS fayeuaz fadAjojoy) sotsads mou ‘srwurdranyf SNINOAT “T AUNOIY BULLETIN 282, PLATE 21 U.S. NATIONAL MUSEUM “OD PURTTEO “PIONNAL JON “A LUN Z AL ‘ET 298 “Y womnzy :ueStyoIPy PIZZST ZININDA “(Ww ggoT TS x08) uepsof snuniu SNANJONT X ([[YOUJA]) Snursk3 snamgony :prsqAyy *Z aunoly "OD M10] VT “uiog Apeys svau ‘prog are] “yy neai0g eUrOyeTIO “FO6LET ZININA “(wu TTS TS fa]eur) uepsof snuntm snunjony Xx UOS[EN] SUjixa Snanjoxy :plIgAyT *] TUNoI 130, 144, 149 (map), 152, 155, 156, 193, 201, 208 Ambloplites ariommus, 69 Ameiurus, 16, 17 natalis, 42, 125 Amiurus, 10, 16 ariommus, Ambloplites, 69 atrorus, Noturus insignis, 87 Schilbeodes insignis, 87 Schilbeodes marginatus, 87, 88, 94 aurolineata, Percina, 68, 180 baileyi, Noturus, 3, 130, 132, 139 (map), 141, 151, 219 balsanus, Ictalurus, 15, 16, 17, 109, 219 Istlarius, 17 camura, Cliola, 118 catus, Ictalurus, 219 Cliola camura, 118 elassochir, Noturus, 59, 60, 67 elegans, Noturus, 3, 68, 130, 143, 144, 146, 150, 154 (map), 156, 157, 159, 180, 191, 201 elegans group, 146, 150, 154 (map), 159, 162 eleuthera, Rabida, 161, 174, 205 eleutherus, Noturus, 5, 22, 27, 28, 54, 83, 128, 180, 151, 155, 159, 160, 166 (map), 167 (fig.), 169, 172, 173, 179, 180, 184, 191, 194, 200, 201, 204, 216, 219 Rabida, 161, 205 Schilbeodes, 27, 161, 166, 169, 173, 184, 194, 205, 214, 216 Etheostoma inscriptum, 68 rubrum, 100, 136, 191 tippecanoe, 159 eurystomus, Satan, 1, 14, 15, 19, 20 exuis, Notutus, 1,14, 23, 24,25, 33, 34, 46, 52, 54, 55, 56, 57, 65 (map), 81, 83, 89, 90, 93, 94, 109, 298-943 O - 69 - 22 Index [Names of new species in italics, page numbers of principal accounts in boldface] albater, Noturus, 3, 5, 26, 68, 81, 128, | exilis—Continued IO, AWS) AS, Li Zp; 128,143, 144, 150), 155, 201, 208, 214, 215, 216, 220 Noturus X Noturus miurus, 165, 214 Rabida, 59, 117 Schilbeodes, 58, 87, 117 flavater, Noturus, 3, 68, 131, 150, 165, 188, 189 (map), 200, 202, 204 flavipinnis, Noturus, 2, 131, 166, 188, 189 (map), 190, 195, 200, 201, 206, 208 flavus, Noturus, 5, 15, 20, 22, 23, 25, 28, 32, 33, 41, 42, 52, 54, 57, 58, 60, 64, 66, 67, 68, 76, 81, 88, 86, 93, 106, 108, 109, 110, 111, 124 (map), 143, 144, 155, 159, 166, 172, 180, 201, 220 Pimelodus, 117 funebris, Noturus, 2, 5, 22, 28, 338, 34, 54, 74, 71, 78, 82, 89,95, 96; 100 (map), 102, 104, 183, 220 Rabida, 96 Schilbeodes, 96 funebris group, 95, 100 (map) furcatus, Ictalurus, 16, 17, 209, 218, 219 furiosa, Rabida, 184 furiosus, Noturus, 20, 26, 27, 54, 95, 107, 110, 129, 131, 165, 167, 169 (map), 174, 184, 200, 217 Schilbeodes, 27, 174, 184 Schilbeodes miurus, 184 furiosus group, 26, 129, 131, 151, 162, 167, 169 (map), 170, 172, 176, 179, 180, 181, 183, 185, 188, 206 gallowayi, Schilbeodes, 162, 165 gilberti, Noturus, 1, 5, 22, 24, 26, 33, 34, 62, 95, 105, 108 (map), 109, 110, 120, 128, 220 Rabida, 69, 87, 105 Schilbeodes, 105 126, 172, 311 312 Gronias nigrolabris, 17 gyrinus, Noturus, 1, 2, 5, 6, 11, 23, 25\- 27, 98. Si, 82, 39-734 “a5, 51 (map), 55, 56, 58, 60, 62, 66, 68, 72; 74; 7d, 79, 81, 83, S87; 93, 95, 99, 105, 109, 112, 121, 125, 126, 127, 128, 166, 168, 180, 188, 1938, 201, 210, 211, 212, 218, 217, 218, 220 Noturus < Noturus miurus, 81, 210, 214, 216 Schilbeodes, 22, 33, 35, 53, 76, 87, 117 Silurus, 20, 23, 33, 35, 46, 52, 53 gyrinus gyrinus, Schilbeodes, 41 gyrinus sialis, Schilbeodes, 41 hamiltoni, Thoburnia, 87 hildebrandi, Noturus, 105, 128, 130, 131, 134 (fig.), 143, 146, 162, 201 Schilbeodes, 136 hildebrandi group, 128; 131, 139 (map), 141, 148, 152, 154, 162 hildebrandi hildebrandi, Noturus, 135, 136, 139 (map), 140 141, 150 Noturus < Noturus hildebrandi lautus, 135, 139 (map) hildebrandi lautus, Noturus, 3, 105, 128, 135, 136, 138, 139 (map), 146, 162, 180, 219 Ictaluridae, 1, 9, 209, 210, 216, 217, 220 (fig.) Ictalurus, 1, 2, 5,6, 10, 11, 12, 13, 14, 15, 16, 17, 18) 20; 21; 26, 52: "200: 210, 217, 218, 219, 220 balsanus, 15, 16, 17, 109, 219 catus, 219 furcatus, 16, 17, 209, 218, 219 melas, 41, 217 nebulosus, 17 punctatus, 16, 17, 191, 209, 210, 217, 218, 219 Ictalurus group, 1, 14, 16-17, 18, 217 inscriptum, Etheostoma, 68 insignarius, Pimelodon, 20, 33, 87, 91, 94 insigne, Pimelodus, 67, 83, 91, 93, 94, 95 insignis, Noturus, 1, 2, 20, 22, 23, 24, 25, 28, 33, 35, 39, 52, 54, 57, 59, 62, 67, 68, 74, 77, 82, 83, 92 (map), 99, 104, 107, 108, 109, 110, 117, 126, 127, 128, 188, 204, 220 Pimelodus, 95 Rabida, 85 U.S. NATIONAL MUSEUM BULLETIN 282 insignis—continued Schilbeodes, 1, 43, 59, 76, 84, 117, 125, 215 insignis atrorus, Noturus, 87 Schilbeodes, 87 insignis insignis, Noturus, 85 Schilbeodes, 85 Istlarius balsanus, 17 lachneri, Noturus, 2, 14, 33, 34, 46, 51 (map), 53, 54, 62, 68, 216, 220 latifrons, Noturus, 22, 161, 165, lautus, Noturus hildebrandi, 3, 105, 128, 135, 136, 138, 139 (map), 146, 162, 180, 219 Noturus hildebrandi Vie Eo ey dis * ai: Rats a | pele i. ; 5 a ‘ hie’ @ q ha fs Fer An Mh Dyke 1) Fae / PA’ Fa Os Pe tual, be Ride we we hi a i i: 7 é y ha) Ps : { j o) Ar fed id Bho ay) We : bis j rat ris a Se Ss ¢ “ 4 el al ; = v ; j ' 'y wath a eu 4 4 ' : 4a) é 5 ‘ * 1& a we ‘ 4 Tal { dipabales i 4 hs f 2 ua 7 ¥ i" rE r ‘ sey ae j r ‘ \ . ' i ' 30-60 les SC red Stepan 209, re it ya - fi) (e ere : ¥) ” , eas . aes f ‘ { f, - n a : e ARN Crore La an A pane At aE Ser ane f IN rent hey aie ne ve , ; AN Ue aN yaaa i : etl pe ts wo Per ays ’ { itt ‘Ge : ' } Tm aA i i + iY a os A Ae 1” ff ; hat an oT ¥) 7 . ay ver nl ia \ i Ms sul na io. ‘ A) ; \ ra I TT 3 9088 01424 3748 |