Comprehensive Description
provided by Smithsonian Contributions to Zoology
Nocomis biguttatus (Kirtland)
Hornyhead chub
Semotilus biguttatus Kirtland, 1840:344, pl. 5, fig. 1 [type locality: Yellow Creek, Mahoning River, Ohio River system, Ohio].
Ceratichthys cyclotis Cope, 1864:277 [lectotype, ANSP 2113; type locality: Michigan, herein designated].
Ceratichthys stigmaticus Cope, 1864:278 [type locality: Michigan].
Hybopsis kentuckiensis.—Fowler 1918:33, plate VIII [redescription].
Nocomis biguttatus.—Hubbs 1926:28, pl. I, fig. 1; pl. II, fig. 3.
Nocomis nebrascensis Girard, 1856:190 [type locality: Sweet Water River, Platte River system].
NOMENCLATURE.—Certain nomenclatural problems involving Nocomis biguttatus commenced when Rafiinesque (1820:48) described Luxilus kentuckiensis, which was a composite account, probably involving species of Nocomis and perhaps species of Notropis. This early history is briefly reviewed by Lachner and Jenkins (1967:558–559). The next available name for a species of Nocomis was Kirtland's description of Semotilus biguttatus (1840:344). His specimens were taken in Yellow Creek, a tributary of the Mahoning River, Ohio. His description and illustration of the nuptial male (Plate 5, Figure 1) and coloration easily identify the above synonyms with the species of Nocomis that Kirtland had found. One problem remains which could involve the security of the entrenched name Nocomis micropogon (Cope). Cope (1864) described two species of Nocomis on page 277, Ceratichthys cyclotis and Ceratichthys micropogon (as well as another one, Ceratichthys stigmaticus on page 278). Lachner and Jenkins (1971) discuss the hybrid nature of Cope's type specimen (ANSP 5061) of Ceratichthys micropogon and conclude that it probably represents a hybrid Nocomis micropogon × Notropis cornutus, based on a number of morphological characters. Lachner and Jenkins (1970) restricted the name Ceratichthys micropogon Cope to that presumed parent of the type specimen. But Cope's description of Ceratichthys cyclotis involves both Nocomis biguttatus and N. micropogon. The type material came from Michigan, at Grosse Isle, and from Waterford, Oakland County, from Clinton River, and from Bruce, Macomb County. Hubbs (1926:29) indicated that Cope's description may have been based on the above two species. Cope recorded the dentition as variable, “in some specimens of this species, especially among the half-grown, I find a tooth of the second row of pharyngeals.” Hubbs further commented that Fowler's (1918:33) redescription of the species amounts to a restriction of the name biguttatus. Fowler, however, records the dentition as that of micropogon, 4–4.
Examination of the extant types of Ceratichthys cyclotis (ANSP 2113–2117, 5 specimens) reveals two species. We designate ANSP 2113, a nuptial male, 132 mm SL, a lectotype of Ceratichthys cyclotis Cope, 1864, and place it in the synonymy of Nocomis biguttatus Kirtland, 1840, because of its dentition, 1,4–4,1, and the widespread occurrence of tubercle scars, dorsally, on the head, extending from the posterior area of the snout to the occiput. The following numbers are designated paralectotypes: ANSP 2114, 101 mm SL and ANSP 2115, 60 mm SL. Both specimens have two rows of pharyngeal teeth, and indications of a once large dark spot at the base of the caudal fin, or post-orbital head tubercles. ANSP 2116, 62 mm SL, may have had two rows of pharyngeal teeth, but the arches are in poor condition; since the specimen is otherwise faded and shows no tubercle marks, it may represent either N. biguttatus or N. micropogon. (We suspect that it is N. biguttatus.) ANSP 2117, 142 mm SL, is referred to N. micropogon because it has a dentition of 4–4, and the tubercle spots are located in the internasal and subnasal areas. With the above lectotype restriction of Ceratichthys cyclotis Cope, 1864, the name is not available for Ceratichthys micropogon Cope, 1864.
SPECIMENS STUDIES.—Nocomis biguttatus has an extensive distribution primarily in the central basin of the United States (Figure 8). It is the key species in understanding the evolution, ecology, and distribution of the species group. The specimens listed below represent material from which morphological data were recorded. An attempt was made to study the larger specimens available in museum collections from samples widely scattered within the various river systems of the area.
MISSISSIPPI BASIN
WHITE DRAINAGE.—Arkansas: KU 3242 (1), OSUMZ 5994 (2), TU 10130 (5), 44685 (2), 44710 (2), 44817 (1), 46953 (3), 49386 (2), 49473 (1), 49804 (1), UF 14704 (3), UMMZ 123903 (1), 128305 (1), 169890 (1), 170895 (1), USNM 42684 (3),42677(1), 193359 (10).
Missouri: KU 11014 (2), UMMZ 102586 (1), 102712 (2), 111459 (1), 151091 (7), 151320 (1), 151367 (1), 152061 (2), USNM 42814 (1), 42882 (1), 204018 (10), 204855 (34).
ST. FRANCIS DRAINAGE.—Missouri: UMMZ 139507 (1).
MERAMEC DRAINAGE.—Missouri: CU 10783 (4), UMMZ 148338 (1), 148389 (2), 148441 (1), 149630 (2), 149664 (1), 149816 (2).
MISSOURI BASIN
GASCONADE DRAINAGE.—Missouri: CU 24323 (3), UMMZ 102721 (1), 111436 (2).
OSAGE DRAINAGE.—Missouri: UMMZ 86532 (1), 102748 (3), 102804 (1), 111355 (1), 150102 (4), 150092 (2), 150377 (1), 150465 (1), 150881 (1), 150917 (2), 150942 (1), 151944 (1), 152694 (1).
Kansas: KU 2470 (2), 2498 (10), 2605 (7), UMMZ 61214 (1), 126847 (1), 160538 (1).
MISSOURI DRAINAGE.—North tributaries: UMMZ 147948 (1), USNM 35876 (1).
KANSAS DRAINAGE.—Kansas : KU 426 (1), 427 (1), 429 (1), 438 (1), USNM 3551 (1).
MISSOURI DRAINAGE.—Nebraska: USNM 86694 (1).
Iowa: UMMZ 114088 (3).
UPPER MISSISSIPPI BASIN
Missouri: UMMZ 148252 (1).
Illinois: UMMZ 144502 (1), USNM 171692 (2).
Indiana: UMMZ 81365 (4), USNM 66457 (3).
Iowa: TU 14498 (6), UMMZ 87066 (1), 87075 (1), 101200 (1), 101214 (1), 101234 (1), 101244 (1), 101270 (1), 101288 (1), 101308 (2), 101366 (1), 101376 (1), 101387 (2), 101399 (1), 146829 (2), 146849 (2).
Wisconsin: UMMZ 74187 (2) 74362 (1), 75579 (3), 75618 (1), 75648 (2), 75808 (1), 75925 (1), 76699 (2), 76854 (2), 76922 (2), 77031 (1), 77056 (1), 77184 (1), 77238 (1), 77483 (2), 77558 (2), 77574 (1), 77933 (6), 78956 (1), 96137 (4), 96148 (1), 96181 (4), 96227 (2).
Minnesota: USNM 133094 (4).
South Dakota: UMMZ 166896 (6), 167000 (6), 167131 (1).
OHIO BASIN
Kentucky: CU 37297 (5), UL 10479 (1), 12465 (1), 12620 (6).
Indiana: UMMZ 6662 (1), UMMZ (IU 9549) (1), 99904 (8), 99952 (2), 112746 (1), 112747 (1), USNM 66449 (1), 69245 (2).
Ohio: UMMZ 86964 (1), 87203 (2), 107667 (1), 118369 (1).
Pennsylvania: CU 33875 (4), 40731 (1), 46677 (2), USNM 166418 (5), 166419 (2), 166420 (7).
GREAT LAKES BASIN
New York: CU 659 (3), USNM 166437 (2), 166452 (3), 166456 (6), 166457 (7), 166459 (2), 166460 (6), 166463 (2), 166464 (6), 166465 (7), 166468 (8), 166469 (4), 166475 (19), 166478 (2).
Ontario: UMMZ 85516 (1), 89023 (3), 89078 (1).
Ohio: UMMZ 121801 (2), 121810 (5), 121832 (3), 121853 (2), USNM 40096 (3).
Indiana: UMMZ 114919 (1).
Michigan: UMMZ 55272 (1), 60230 (2), 61387 (1), 68285 (1), 69039 (1), 69060 (1), 69085 (3), 69123 (1), 69129 (3), 69143 (5), 70558 (1), 79096 (1), 79881 (1), 80731 (1), 81500 (2), 82609 (2), 82975 (1), 83087 (1), 83146 (2), 83231 (1), 83267 (1), 83411 (1), 83543 (1), 83771 (1), 83913 (1), 84650 (2), 84681 (2), 84758 (2), 84802 (3), 84951 (1), 86292 (2), 89454 (2), 89595 (2), 92108 (1), 96987 (1), 97242 (2), 98410 (2), 101908 (1), 101958 (1), 104120 (1), 106876 (2), 106902 (1), 110352 (1), 113321 (1), 116249 (1), 116335 (1), 116972 (1), 117014 (4), 117137 (1), 131946 (1), 136822 (1), 136944 (1), 137148 (1), 137217 (1), 137287 (1), 137649 (1), 137672 (1), 137692 (16), 137701 (1), 137731 (3), 138070 (5), 139797 (1), 139804 (2), 139834 (2), 140215 (1), 162738 (8), 164288 (19), 164376 (12), 167826 (5), 185397 (3), 185398 (2), 185400 (3), 185406 (3), 185411 (2), 185414 (5).
Wisconsin: UMMZ 64516 (3), 64522 (2), 64663 (3), 64902 (2), 72884 (1), 72975 (4), 73892 (4), 73963 (2), 74087 (1), 74159 (1), 74285 (1), 74894 (5), 74940 (5), 76713 (1), 76763 (1), 77794 (3), 78532 (1), 87029 (2), 97150 (1), 108084 (1), USNM 143015 (3), WSU466 (7).
The plots on the map for the distribution of Nocomis biguttatus (Figure 8) are based on two sets of data. One set includes the specimens listed above, and hundreds of additional collections examined in these museums, which, while of no immediate systematic value, were important in the locality data they provided and contributed to a better understanding of the distribution, ecology, and zoogeography of the group. Because of the thousands of specimens involved, it is prohibitive to list all of these collections herein. Another set of distributional data were taken mainly from comprehensive state faunal surveys. These records, often extremely abundant, were used to round out aspects of the distribution of a local nature, from which we saw no material. The references from which these data were taken, listed by author with the state or province involved, are as follows: Cross (1967, Kansas), Ellis (1914, Colorado), Bailey and Allum (1962, South Dakota), Keleher (1956, Manitoba), Hinks (1943, Manitoba and North Dakota), Cleary (1956, Iowa), Forbes and Richardson (1920, Illinois; N. micropogon apparently is confined to the Wabash system in Illinois), Gerking (1945, Indiana), Greene (1935, Wisconsin), Taylor (1954, Michigan), Trautman (1957, Ohio), Raney (1938, Pennsylvania), Underhill (1957, Minnesota), Simon (1946, Wyoming), and Copes and Tubb (1966, North Dakota).
Locality records for collections in the Royal Ontario Museum, Toronto, Canada, were taken from a list provided by W. B. Scott. Distributional records for New York were checked against the data published in the New York State Biological Survey Reports.
DIAGNOSIS.—Nocomis biguttatus is a wide-ranging species, most closely related to N. asper in having two rows of pharyngeal teeth. It differs from N. asper and N. effusus in never having tubercles on the scales of the body, laterally. It is a coarser scaled species than N. effusus. The red postocular spot is intermediate in its development in N. biguttatus, compared with its two species-group relatives. In N. biguttatus the red spot is well developed only in adult males, and is brassy colored or not evident in adult females. In N. asper the red spot is conspicuously developed in adults of both sexes. In N. effusus the red spot is present, but faint, in the large males only.
Comparison of Species in the Biguttatus Group
The primary differentiation among the three species is in the tuberculation of the body, the reduction of teeth on the pharyngeal arch from two rows to one row, the development of coarser and finer scaled species, the elongation of certain body proportions, as length of snout and head, and the development of certain nuptial colors in the males.
Counts of the head tuberculation revealed that the three species have about the same number of head tubercles and the same increments in tubercle numbers with increase in body length (Tables 3, 4, and 5). Where large samples of nuptial males were available from particular drainages, great variability existed in the number of head tubercles within any of the groups classified by a range of 10 mm in body length. The group of Nocomis biguttatus 70 to 79 mm in size has specimens in which the number of head tubercles exceed others by three times or more. The same variation existed in the larger specimens. All three species had a high variability in the number of head tubercles within any given size-group. Preliminary counts suggested differences in the number of head tubercles among the three species and among drainage populations of N. biguttatus, but as more adult specimens became available these differences proved unreal, for they were associated primarily with body size. N. biguttatus is apparently a small species over most of its range, excluding those of the White River drainage, and it is much smaller than N. effusus or N. asper. Only one specimen was examined which exceeded 150 mm SL (Table 5)—a gravid female, 162 mm from Weber Lake, Lake Michigan drainage. Many males of N. biguttatus in the Great Lakes drainage are mature, and with head tubercles enlarged, at 100 mm, and the females are gravid at 90 mm. Apparently, N. biguttatus in the White River drainage attains the largest size among the drainage populations and among the members of the species group.
In one collection, USNM 204018, eleven specimens from the North Fork of the White River, Ozark County, Missouri, five males exceeded 200 mm SL (compare data in Tables 3, 4, and 5, and Lachner and Jenkins 1967, Table 1). One of these specimens also had about 155 head tubercles, the highest number found among the three species.
Nocomis effusus may be a larger species than N. asper; five nuptial males of the former attained body lengths of 161 to 202 mm and averaged 179 mm. Seventeen nuptial males of the latter species ranged from 143 to 182 mm and averaged 167 mm.
Although tuberculation on the body, laterally and on the nape, increases with increase in body length, there are two notable differences in the development of body tubercles among the three species. First, Nocomis biguttatus does not develop body or nape tubercles, excluding one or two supernumerary occipital tubercles that may be present in the largest nuptial males. The body tubercles only appear on N. asper and N. effusus. They also are absent on all numbers of the micropogon group and leptocephalus group. Secondly, the body tubercles on N. asper have differentiated significantly from those on N. effusus in the number of tubercles per scale. The total number of tubercles on the tuberculated lateral body scales was about the same in both species (Table 6, N. asper, seventeen specimens range from 28–112 total tubercles, average 82.2; N. effusus, five specimens range from 25–116 total tubercles, average 86.6). Only 3 percent of the tuberculated scales in N. effusus had two tubercles per scale, and no scales had more than two tubercles. About 23 percent of the tuberculated scales in N. asper had two tubercles (Tables 8 and 9), 2 percent had three tubercles, and one specimen had one scale with four tubercles. There is a great range in N. asper in the number of tuberculated scales having either one or two tubercles per scale. Nuptial males may have only 26 scales bearing one tubercle and other larger specimens have 82 to 86. Some specimens have less than 10 scales bearing two tubercles and other specimens may have 30 to 50 scales bearing two tubercles (Tables 8, 9). The smaller nuptial males of both species averaged fewer tuberculated scales (Table 7). In N. asper, the 10-mm-size groups from 140 to 180 mm SL, had average values for number of tuberculated scales of 64, 75, 85, 85, and 112.
Tuberculation on the nape was highly variable but the larger specimens of both species had more tubercles. The nape tuberculation is probably more extensively developed and more tubercles are present in Nocomis asper than in N. effusus. The body tubercles in N. effusus are about the size of the average head tubercles, whereas in N. asper the body tubercles are much smaller than in N. effusus, about one half the size of the average head tubercles. The head tubercles are about equal in size in all three species of the biguttatus group.
There is nearly a complete separation of Nocomis asper and N. biguttatus from N. effusus in the pharyngeal dentition. The former two species have two rows of teeth with the following counts: N. asper, 1,4–4,1 (22 pairs of arches), 0,4–4,1 (1), 1,4–4,2 (1); N. biguttatus, 1,4–4,1 (47), 1,4–1,0 or 0,4–4,1 (5), and 4–4 (1, questionable). N. effusus has one row with 4–4 (28 pairs of arches) and 0,4–4,1 (1).
A summary of the differences in squamation among the three species is given in Table 2 involving three different scale counts, the lateral line, circumferential, and caudal peduncle. The data for Nocomis biguttatus were grouped by four general areas: (1) the Great Lakes drainage including the Ohio and upper Mississippi River drainage, (2) the Missouri drainage, (3) the lower Mississippi drainage area, and (4) the White River system. We were particularly interested in the White system because of its geographic location between the Arkansas River to the south (and adjacent headwaters) and the northern and eastern systems of the Missouri, Mississippi, and Ohio basins. Although the number of scales is of minor value in the identification of the species, the character has systematic value in showing trends and relationships among the species and among populations of N. biguttatus. N. effusus is finer scaled, which is shown particularly in the high values of the circumferential count (average 36.1 scales) and the caudal peduncle count (average 19.7). N. asper is somewhat more coarse-scaled than N. biguttatus of the White drainage. The northern and eastern populations of N. biguttatus, in the Great Lakes, Ohio, and upper Mississippi drainages, were consistently lower in all three scale counts when compared with those from the Missouri, lower Mississippi, and White drainages (and also with those counts of N. asper and N. effusus).
The vertebral numbers (Table 2) were higher in Nocomis effusus (modally 41) compared with those of N. asper (modally 40) and N. biguttatus (modally 40).
Our first observations of Nocomis asper, based on a few large males, indicated that it was a long-head, long-snout form. As more material accumulated, particularly larger specimens, it became possible to test statistically these preliminary observations. Because of the contiguous distribution N. asper has with N. biguttatus, its close relationship with this species, and probable evolution from biguttatus stock, more detailed comparisons of head lengths were made between these two species. The relationship between standard length and head length of N. asper and N. biguttatus from the White River drainage is shown in Figure 3, based on a sample of measurements of juveniles and adults distributed as widely over the drainages as the collections permitted.
Statistical tests were performed to compare the rela tionship between standard length and head length of Nocomis asper with each of the following populations of N. biguttatus: The White drainage, the Missouri drainage, and the Great Lakes drainage including the Ohio and Upper Mississippi drainages. Tests of differences of slopes and heights (intercepts) of pairs of regression lines were performed by analysis of covariance (Simpson, Roe. and Lewontin, 1960). F values yielding probabilities greater than 0.95 were considered significant. The samples tested were randomly selected from several hundred measurements of specimens from each population; these specimens were widely distributed within the area tested. The regression equations are as follows:
N. asper, y = .0149X + 26.49, n = 124;
N. biguttatus (White drainage), y=.0063X +26.64, n=70;
N. biguttatus (Missouri drainage), y=.0141X + 26.48, n=60;
N. biguttatus (Great Lakes), y=.0114X+26.12, n=100.
The slope of the regression of head length (as percent of SL) on standard length of Nocomis asper is significantly different from N. biguttatus of the White drainage (F = 8.71; 190 degrees of freedom). The slope of the regression of head length (as percent of SL on standard length of N. asper was not statistically different from that of N. biguttatus of the Great Lakes drainage (F=.40; 220 degrees of freedom), but the heights were significantly different (F=36.38; 221 degrees of freedom). The slope and height for data of N. asper were not statistically different from, those of N. biguttatus from the Missouri drainage (slope: F = .04, 180 degrees of freedom; height: F=
Many body proportions of the species of Nocomis show allometric growth (Lachner and Jenkins 1971; Lachner and Wiley 1971), an increase in body length resulting in an increase in size of structure. The relationship of snout length to standard length of N. asper and N. biguttatus from the Great Lakes drainage (Figure 4) shows the same general growth trends and differences as head length. Considerable variability occurs in the size of the proportional character within each population and within the 10-mm-size SL groups. Such proportional characters, however, are useful in describing populations and in demonstrating differentiation among populations and species of Nocomis.
The three species show consistent differences in life in the intensity of development of the red postocular spot, in the color of the fins, and in the development of the dark lateral body band. These comparisons pertain mainly to the nuptial males. The red postocular spot is large, conspicuous, and intensely developed in Nocomis asper on specimens of subadults to adults (and present in mature females at sizes of 100 mm SL or greater). It is well developed only in the larger, mature males of N. biguttatus and is brassy colored in some adult females. It is small and weakly developed in the largest nuptial males of N. effusus, often absent in the smaller nuptial males, and absent in all other individuals.
The most intense and noticeable life colors occur in adult males of Nocomis effusus, in which the fins—particularly the caudal, pelvic, and anal—are bright orange or reddish orange. The pectoral fin is bright in N. asper, being predominantly pinkish yellow to pinkish orange. The pelvic, anal, and caudal fins are mainly pinkish olive or light pinkish orange. N. biguttatus develops the weakest fin colors among the three species; the pectoral fins are orange, the pelvics and anal fins are pale orange, and the caudal fin is light orange red.
The dark lateral band on the body is prominent in nuptial males of Nocomis biguttatus, weakly developed in N. asper, and almost completely absent in N. effusus. The young to subadult males and most all-size stages of the females have subdued colorations and they are, other than the differences discussed above, fairly similar in their general color pattern. More detailed descriptions of color and color patterns are given by Lachner (1952) and Lachner and Jenkins (1967).
Distribution and Ecology
The distribution of the three species of the biguttatus group is shown on Figure 8. The extensive distribution of Nocomis biguttatus in the central basin contrasts markedly with the restricted distributions of N asper and N. effusus. N. asper has a center of distribution in the Ozark highland section of the Arkansas River drainage. It is absent in the western tributaries of the Neosho system. Hall (1952:57) reported it only from the upland tributaries of the east side of Grand (Neosho) River. Branson (1967:133) found it “very abundant in most of the eastern clear tributaries but is absent from the main (Neosho) stream.”
Three isolated populations of Nocomis, two in the Arkansas drainage which possibly represent N. asper, and one in the Red drainage, definitely N. asper, are apparent evidence of what was once a much greater distribution of the species. In the upper Arkansas drainage of Kansas, Cross (1954) reported Hybopsis biguttatus (=N. asper?) as rare in the South Fork of the Cottonwood River and that juveniles were common in Cedar Creek, both of the Neosho River system. These localities are plotted on the Kansas map by Cross (1967:87). Cross and Braasch (1969) did not capture the species in their 1967 survey whereas it was present in 1952 (1954 report) in one half of all collections made. We have examined the following Nocomis collections from the Neosho River system in Kansas: KU 2704 (1), KU 2718 (16), UMMZ 116059 (1), and UMMZ 120801 (1). All of these specimens were young and juvenile sizes under 83 mm SL, and we could not specifically identify them because of their small size. These specimens are indicated on Figure 8 by a question mark and they may represent either N. asper or N. biguttatus. A single, isolated western record (Figure 8) of Nocomis in north-central Oklahoma, Arkansas River drainage, is tentatively referred to N. asper. One specimen, UOMZ 15475, 71 mm SL, was taken near Turkey Island, Panner County, on 8 April 1933 by George A. Moore, where according to Hubbs and Moore (1940:92) “large springs emerge in the bed of the main river.” A third isolated population occurs in south-central Oklahoma (two plots on Figure 8) where we have examined seven collections from the Blue River, Johnston County, Red River drainage. The Blue River is the only portion of the Red drainage inhabited by Nocomis. A good series of large specimens had the typical body tuberculatum of N. asper and otherwise agreed with the characters of this species. N. asper was taken from the Blue River in 1947 by Linder (1955:175, as Hybopsis biguttata). R. E. Jenkins and F. F. Snelson, Jr., found N. asper very common in the Blue River system in 1967.
The disjunct occurrence of these three populations from the main one in the Ozark upland, Arkansas drainage, closely parallels isolated populations of Notropis pilsbryi. Cross (1967: 110) shows the distribution of Notropis pilsbryi in the upper Neosho system where it occurs as a relic, occupying clear, flowing water with a substrate of rubble and gravel. Notropis pilsbryi once occurred with Nocomis asper at the isolated western station of the Arkansas drainage near Turkey Island, Panner County (Hubbs and Moore 1940: 94, Figure 1; Gilbert 1964: 133, Map 2). It was reported from the Red drainage at three localities, each in a separate tributary by Hubbs and Moore (1940: Figure 1). The species was recently captured from the Blue River, a fourth locality in the Red drainage. These isolated populations of Nocomis asper and Notropis pilsbryi appear to be remnants of a fauna that was formerly much more widely distributed. They do not seem to be products of introductions or accidental transfer of data, as was suggested by Hubbs and Moore (1940: 94) for Notropis pilsbryi. Other evidence suggests a natural distribution of these species, such as the close compatibility between them during spawning. Notropis pilsbryi is one of the common nest associates of Nocomis asper. While the latter species constructs a gravel mound in the reproductive process, the former species occupies the nest as a spawning site, often in large numbers, with practically no antagonism shown between the two. Both species prefer clear, flowing streams with clean gravel, rubble, and sandy substrates.
The wide distribution of Nocomis biguttatus shows interesting gaps, some of which are related to the immediate erosional history of the central basin, while others are associated with the geology and glacial history of the area. Climatic factors, such as droughts, also have affected the general distribution and occurrence of Nocomis biguttatus. The hornyhead chub is common on the Ozark upland of Arkansas and Missouri with some scattered populations at lower elevations. The species is fairly common in the low elevation areas of the upper Osage drainage. Deacon and Metcalf (1961) did not collect this species in the Wakarusa River system draining northeastern Kansas. They report (page 316) that it was common in early collections but that it has not been taken since 1924. Its disappearance is probably related to increased siltation and intermittency of stream flow. Metcalf (1966: 105) found the hornyhead chub to have one of the most restricted ranges of any fish in the Kansas River drainage. He lists earlier records and museum specimens, some dating back more than 100 years. Some of these earlier records appear on the distribution map of Figure 8. The occurrence of Nocomis biguttatus in the Missouri River drainage today is limited to a few, widely separated populations. Metcalf points out that its extirpation in the Missouri drainage is much like that described in Ohio by Trautman (1957) where fine siltation due to cultivation has covered coarser sediments. The water has become more turbid, with reductions in flow and volume, and the streams are now intermittent in some areas. The ecological conditions, such as clear, constantly flowing streams with gravel-rubble bottoms that once provided suitable habitat for this gravel-mound, nest-building chub to maintain a wide distribution, are now largely in a state of attrition. Extirpation of the species through change of habitat, and increased competition from species more highly adapted to silt-bearing waters, is now leading to its confinement in numerous scattered localities.
Our single plot (Figure 8) in northwest Missouri is an 1884 record of Jordan and Meek (1885:12) from Hundred and Two River. The plot in eastern Nebraska is based on a specimen from Papillion River (USNM 86694) received from the Museum of Comparative Zoology, Harvard, and was taken by the Hassler Expedition prior to 1872. It is not known whether the species occurs there today. The few scattered records in the Platte River system in Colorado and Wyoming are documented by Ellis (1914) and Simon (1946). The occurrence of the hornyhead chub in the upper Cheyenne system of Wyoming (two plots, Figure 8) is based on several specimens. Nocomis biguttatus has some populations in northwest Iowa and southeast South Dakota of the Missouri drainage.
Farther west across the Continental Divide in the Colorado drainage, Cope and Yarrow (1875:651) reported Nocomis biguttatus, based on collections taken by H. C. Yarrow and H. W. Henshaw (Wheeler Survey) in 1872, and the locality was given as Harmony, Utah. Two extant collections of Yarrow and Henshaw, USNM 16986 (8 specimens) and ANSP 19854 (1 specimen), were examined by us and these represent Nocomis biguttatus. This locality is within the hydrographic boundary of ancient Lake Bonneville, between the southern boundary of Lake Bonneville and the headwaters of the Virgin River (R. R. Miller, personal communication), a present tributary of the Colorado River. Miller and Hubbs (1960:34) consider the locality as unquestionably erroneous and we concur, because extensive field work by them throughout the Great Basin has never revealed any Nocomis. Miller and Hubbs (1960) also cite other erroneous locality data associated with collections of the Wheeler Survey of 1871–1874.
The hornyhead chub is widely distributed within the glacial regions of the upper Mississippi River drainage. It is largely absent from the Driftless Area of Wisconsin(Greene 1935:75, Map 26), as well as the larger western rivers. It is widespread in the upper Red River of the North drainage but apparently absent in the Lake of Woods drainage, both of the Hudson Bay basin. In the northern Ohio River basin Nocomis biguttatus is widely distributed, but avoids the nonglaciated areas. The species has a restricted distribution in the Allegheny River system of northwestern Pennsylvania. It is absent in the more upland sections of the Allegheny and Monongahela River systems of western Pennsylvania.
Lachner and Jenkins (1967:576) reported Nocomis biguttatus south of the Ohio River in Kentucky based on two collections from Elkhorn Creek, Franklin County, lower Kentucky River drainage. Three additional collections have been examined from Elkhorn Creek or its South Fork, UL 10479, 12465, and 12620. These five collections are represented on Figure 8 by a single plot. We suspect that this isolated population is not natural because of the absence (but high survival potential) of the species elsewhere in the system and south of the Ohio drainage. The area has been an intensive angling site and the population probably represents introduction by bait fishermen or, accidentally, through stocking programs or pond culture.
Several post-glacial routes assisted the entry of Nocomis biguttatus into the Great Lakes drainage. These routes are reviewed by Lachner and Jenkins (1971). The hornyhead chub is widespread in drainages of southern Lake Ontario and all or most of Lake Erie and Lake Michigan. It is present on the Atlantic slope only in the Mohawk-Hudson drainage of New York. It is absent from northern Lake Huron and the Georgian Bay region. Its presence in the Lake Superior area is restricted to streams of the southwestern portion.
The distribution and ecology of Nocomis biguttatus is compared with N. effusus and N. micropogon by Lachner and Jenkins (1967). Over much of its range, east of the Mississippi River, N. biguttatus is sympatric with N. micropogon. The ecology of N. biguttatus is treated further in Lachner (1952). As a whole, the species of the biguttatus group are tributary forms, avoiding the main rivers. Clear water streams, small-to-moderate in size, with moderate-to-low gradients, and with clean substrates of gravel, rubble, and some sand are preferred. Generally, the preferred streams have a moderate balance of riffles and pools, but long pools of slack water often exist. Turbid waters or fine silt-laden situations are usually not inhabited. Streams with some aquatic vegetation appear to be preferred more by N. biguttatus and least by N. effusus.
The spawning behavior of Nocomis asper was observed by the authors at two typical habitats of the Arkansas River drainage in Oklahoma—at Spavinaw Creek, Delaware County, 7 June 1967, with water temperature about 21° C. and at Flint Creek, Delaware County, 12 June 1967, with water temperature about 21.5° C. These small, clear-water streams, 20 to 40 feet in width, have substrates of 70 to 95 percent small-to-large gravel, the remainder of the bottom being sand and rubble. The riffles are shorter in length than the pools; the linear relationship is about 1:2. Most of the riffles are about one-half to two feet in depth, the slack water about two feet deep and the pools three to four feet in depth, the deepest exceeding five feet. The riffles are generally shorter in length than the width of stream. Many nests of N. asper were completed, located in the moderately flowing waters, at the head of riffles, and in pools. The active nests, where spawning was observed, also were occupied by the spawning cyprinids Notropis pilsbryi and Dionda nubila. The common stream inhabitants in Spavinaw Creek were, in addition to N. asper, the cyprinids Dionda nubila and Notropis pilsbryi, the cottoid Cottus carolinae, the smallmouth bass Micropterus dolomieui, and the percid fishes Etheostoma flabellare lineolatum and Etheostoma spectabile.
The general habitat and ecology of Nocomis asper elsewhere in the Ozark upland of the Arkansas River drainage, particularly in the Neosho River system, is described in relation to other fishes and problems by Cross (1954), Minckley and Deacon (1959), Deacon and Metcalf (1961), Branson (1967), and Cross and Braasch (1969).
Evolution and Zoogeography
We have postulated (Lachner and Jenkins 1967) that the biguttatus species group is probably primitive within the genus, the assumption based mainly on the general tuberculation pattern of the head and the presence of two rows of pharyngeal teeth. A biguttatus stock (s) was probably the precursor of the three groups of Nocomis. The adaptive specialization of head tuberculation among the species of the micropogon and leptocephalus groups involves elaboration and/or reductions of tubercle number, size, and distribution. The head tuberculation within the biguttatus group is about the same in the three species in regard to size, distribution, and numbers when related to body length.
The minor tooth row of the pharyngeal arch is absent in the micropogon and leptocephalus groups and the loss is considered a derived condition. The primitive condition of two tooth rows, 1,4–4,1, persists in Nocomis biguttatus and N. asper, but the minor row is absent in N. effusus. The loss is regarded as a development independent to the loss in the micropogon and leptocephalus groups.
The evolution of the three groups of Nocomis are further compared and discussed by Lachner and Jenkins (1967, 1971). Within the biguttatus species group the body tuberculation in N. asper and N. effusus possibly represents evolution of another type of tubercle pattern, restricted entirely to this group and these two species of Nocomis. N. asper has more tubercles per tuberculate scale, laterally, on the side of the body than N. effusus. The occurrence of tubercles on the nape and on the scale rows above the lateral line is more widespread in N. asper compared with N. effusus. The body tubercles are much smaller in N. asper than those of N. effusus. The reduction in tubercle numbers per scale and distribution on the body, and the increase in tubercle size in N. effusus is regarded as a derived condition and correlates with the evolutionary trends of head tubercle development in the other groups of Nocomis. N. biguttatus is probably the more primitive of its species group, having a general tubercular pattern only on the head. N. effusus, because of its reduced dentition and specific body tubercle pattern, is placed at the advanced end of the biguttatus line. It also has the finest scales, narrowest scale margins, and shows the least development of the dark lateral band. N. asper is intermediate between N. biguttatus and N. effusus in character development, having a close affinity to N. effusus in its body tuberculation, and a close affinity to N. biguttatus in its coarser scales, dark lateral body band, well-developed red postocular spot, and two rows of pharyngeal teeth.
The zoogeography of the eastern and southern forms of Nocomis are discussed by Lachner and Jenkins (1971) and Lachner and Wiley (1971). The occurrence of the wide-ranging N. micropogon and its distributional relationships with N. effusus in the southwestern Ohio River basin is treated by Lachner and Jenkins (1967). The interesting aspects of the zoogeography of the biguttatus species group is that two of the species, N. asper and N. effusus, have restricted distributions and have been long isolated by a vast lowland; the nominal form is extremely wide ranging. The three species are allopatric.
Nocomis effusus probably evolved within the region of the Cumberland, Green, and lower Tennessee drainages. It is unlikely that it had a much wider range because of its inability to traverse the Mississippi River valley. Its preference for moderate-to-low-gradient streams inhibited upstream movement. The closest relative and probable ancestor of N. effusus, N. asper, is not far west of the range of N. effusus, in the Ozark region of the Arkansas River drainage and a restricted portion of the Red River drainage. The lower portions of the Mississippi River tributaries below the mouth of the Tennessee River are not inhabited by members of the biguttatus species group. The range of N. biguttatus begins north of that of N. asper and is largely north and west of N. effusus.
The Ozark upland has additional faunal relationships with the Tennessee and Cumberland drainages, many species of which have almost always been confined to these two areas. Stages of main river development have provided means for species to establish a southern trans-Mississippi River distribution, and for isolation and subsequent differentiation on either side of the river. Trans-Mississippi dispersal of many upland species was probably quite limited and, once a large river was transversed, they would tend to seek conditions upstream within newly occupied systems.
The important early relations among the Mississippi and Ohio Rivers and some of their main tributaries are discussed by Fenneman (1938:87–90) . (Also, see Figure 30 of Lachner and Jenkins 1971). The Ohio River once entered the Mississippi appreciably southward of the present mouth of the Ohio. The point of juncture was assumed to have been somewhat below Helena, Arkansas, which is about fifty air miles above the mouth of the Arkansas River. At about that time, the Tennessee River had a more southern confluency with the Ohio, and the Cumberland was probably a tributary of the former. These conditions were extant during part of the Pleistocene, presumably until near-present relations were established in late Pleistocene. Large rivers that were probably often turbid in the past, as the Mississippi, are regarded as both barriers to dispersal for upland, clear-stream fishes, and/or as limited, direct routes of dispersal.
The species of Nocomis distinctly disfavor ecological conditions that were probably present in the region of the lower Mississippi, Ohio, and Tennessee Rivers. Assuming N. effusus was derived from N. asper stock, however, there had to be eastward dispersal from the Ozark region into the southwestern Ohio River basin. It is most probable that such dispersal occurred before post-Pleistocene conditions. From the mouth of the Arkansas River to the present mouths of the Tennessee River and Cumberland River involves upriver movement over a considerably greater distance and through conditions probably more difficult than those extant when the rivers were confluent farther south. In addition, the evolution of N. effusus probably required a greater time span than the age of the present drainage relationships.
Dispersal of Nocomis effusus among the three drainages of the southwestern Ohio basin was through main rivers and/or stream capture. It is possible that N. effusus used the Ohio River after the close of the Pleistocene, as mouths of the Tennessee, Cumberland, and Green drainages are close to each other. They may have been closer or formed a single river during earlier stages of the formation of the Ohio basin. The fact that N. effusus is known from the lower Tennessee and Cumberland drainages lends credence to river dispersal.
It would be more presumptuous to assume that both Nocomis asper and N. effusus were derived independently from a biguttatus-type stock. If so, the invasion of biguttatus stock into the southwestern Ohio basin would not be difficult to envision.
Problems related to the usage of stream capture in Zoogeographic considerations are discussed by Lachner and Jenkins (1971). Although the known geological evidence for stream captures in the southwestern Ohio basin is meager, Wright (1936:246) stated that evidence of capture disappears with active dissection of the capture area. When geological evidence for stream capture is not ctied below, the possibility of stream capture is still invoked when biological evidence exists, and the streams involved are adjacent to each other. The drainage in which Nocomis effusus evolved is not known. Stream capture, however, probably occurred between the Cumberland and Green and the Cumberland and Tennessee drainages, thus permitting crossing of their divides. Considerable portions of both the Green River and Barren River systems, Green drainage, are adjacent to the range of N. effusus in the Cumberland. Campbell (1896:671–672) stated that these divides have migrated toward the Cumberland. Such migration probably involved stream capture. N. effusus would likely have been transferred during headwater capture since it typically inhabits small streams. Many additional small and moderate stream species are shared by the Cumberland and Green drainages. There also are close fanual relationships between the Cumberland and Tennessee drainages. One theater of stream capture may have been in their lower portions where the Duck River system of the Tennessee is bounded almost entirely on the north by the Cumberland.
Nocomis asper probably evolved in the Ozark plateau of the Arkansas River drainage. It could have been more widespread at one time, as evidenced by relict populations at Turkey Island, western Arkansas drainage, in the Blue River of the Red River drainage, and by the population in the upper Neosho River system of Kansas. The occurrence of these relict populations is interestingly paralleled by at least one species, N. pilsbryi. Although the Arkansas and White drainages have closely adjacent headwater tributaries and their mouths are practically together, there is no known recent exchange of Nocomis between these drainages.
The Nocomis forms from the Arkansas River and White River drainages have many characters common to each other. We suspected intergradation in body tuberculation and acquired large specimens needed to solve this problem. None from the White River drainage showed any evidence of body tuberculation. Other characters were also found to have differentiated from those of the Arkansas drainage. Drainage exchanges, however, between the White River and the Arkansas River are suggested by Hubbs and Moore (1940:95) involving Notropis pilsbryi, and by Branson (1964: 746) involving Notropis galacturus; Branson (1967: 148–152) discusses stream piracy and possible faunal exchanges between these two drainages. Nocomis does not appear to be involved in any recent White-Arkansas drainage exchange for the differentiation of the two forms is complete and constant in all tuberculate specimens.
The population of Nocomis biguttatus in the White drainage is morphologically similar to that of the Missouri basin. The distribution of the hornyhead chub in the Missouri basin consists of scattered, local populations, some of which are at a low level of survival, and others are known only by collections of early explorers. The present range of N. biguttatus in the Missouri basin indicates an early, wider distribution. A progressive restriction in the distribution is attributed to the erosional history of the area, recent agricultural practices, and to disfavorable climatic conditions.
The present distribution of Nocomis biguttatus indicates that refugia existed south of the glacial limits in the Mississippi-Missouri basins during Pleistocene times. With glacial retreat, N. biguttatus invaded the upper Mississippi and Ohio basins. Its entrance in the Great Lakes at several different, broad postglacial connections and times, and entry into The Red River of the North, is discussed by Greene (1935), Radforth (1944), Gerking (1945), Underhill (1957), Trautman (1957), Hubbs and Lagler (1958) and Bailey and Allum (1962). The absence of N. biguttatus from above the falls of the Genesee River system in New York and in the upland Allegheny River system indicates that it entered the Lake Ontario drainage from the west rather than through the Allegheny-Genesee (Cuba) postglacial outlet. Entry of N. biguttatus into the Mohawk-Hudson drainage was probably during the Mohawk glacial outlet. Although N. biguttatus and N. micropogon today are sympatric over extensive portions of their ranges, N. biguttatus probably had central basin refugia during glacial times and N. micropogon had a more easterly refuge (Lachner and Jenkins 1971: Figure 30).
- bibliographic citation
- Lachner, Ernest A. and Jenkins, Robert E. 1971. "Systematics, distribution, and evolution of the Nocomis biguttatus species group (family Cyprinidae: Pisces) with a description of a new species from the Ozark upland." Smithsonian Contributions to Zoology. 1-28. https://doi.org/10.5479/si.00810282.91
