Comprehensive Description
(
Inglês
)
fornecido por Smithsonian Contributions to Zoology
Nocomis leptocephalus (Girard)
BLUEHEAD CHUB
Ceratichthys leptocephalus Girard, 1856:213 [type locality, Salem, North Carolina; Yakin River system].
Hybopsis kentuckiensis.—Jordan, 1889a: 123 [(part) Roanoke drainage].
Hybopsis leptocephala.—Ross, 1959:12, 16, 25 [New drainage].—Ross and Perkins, 1959 [(part) New drainage],
Hybopsis micropogon.—Ross, 1959:12–13 [(part) Potomac drainage].—Ross and Perkins, 1959:16 [(part) New drainage].
Hybopsis species.—Ross and Perkins, 1959:12, 19, 24, 27 [(part) New drainage].
Nocomis leptocephalus—Jackson and Henderson, 1942:95 [(part) Roanoke River].—Burton and Odum, 1945 [(part) James drainage].
Nocomis micropogon.—Fowler, 1945:77 (Roanoke drainage).—Burton and Odum, 1945 [(part) New drainage].
DIAGNOSIS.—A stout bodied species differing from all other Nocomis in the reduced numbers of head tubercles, almost always fewer than 25 in the typical subspecies of the central Atlantic slope and reduced to 6 for nuptial adults in the southern, Gulf slope form; tubercles absent on snout, subnasal, and lachrymal areas; intestine whorled in almost all drainage populations, in all from the central Atlantic slope; scales large, mean values for circumferential scales for drainage populations less than 30; values for most meristic characters are lower than in all other species of Nocomis.
DESCRIPTION.—Morphometry: Table 25 contains proportional data for 13 characters of 47 specimens from the James, Roanoke, and New drainages. The body features are fairly constant within size-groups among the drainage populations of the Atlantic slope as well as among the three subspecies over the entire range of the species. Most of the variation in characters is associated with changes related to allometric growth.
Compared to other species of Nocomis, the body is short, stout, and comparatively deep. The snout is short (averages 11.6% SL), the interorbital wide (9.3% SL), and the lachrymal short (6.9% SL). The dorsal and anal fins in the adults have a rounded contour; the pectoral and pelvic fins are short and rounded. Compare the deeper and more blunt head and deeper body of the juvenile with N. micropogon and N. raneyi in Figure 17.
The intestine is unlike any other Nocomis, having a pronounced whorl or anterior loop extending to the right side and covering part of the anterior alimentary tract (Figure 8), except in certain collections of the southeastern Atlantic slope subspecies. All populations of the typical subspecies ranging north of the Savannah drainage show constant and relatively uniform development of the intestine. It is typically coiled in small juveniles at 25 to 30 mm SL. Its dark, coiled outline may be seen through the pale belly tissue in these small sizes without cutting open the abdominal cavity.
The slope of the preopercle-opercle suture is directed forward at the angle only slightly in 43 specimens measured and either perpendicular or posteriorly directed in 157 specimens.
Meristic characters: Summaries of meristic characters, usually for five drainage populations of the central Atlantic area, including the range for mean values, are as follows: circumferential scale-rows, 28.1 to 29.4 (Table 12); scale-rows along the lateral line, 38.1 to 39.9 (Table 8); scale-rows above lateral line, 6.1 to 6.8, and scale-rows below lateral line, 4.8 to 5.4 (Table 11); caudal peduncle scale-rows, 16.0 to 16.3 (Table 14); number of pectoral fin rays, 16.5 to 16.9 (Table 16); total number of vertebrae, 38.9 to 39.8 (Table 15). Most of the meristic characters have lower values than for all other species of Nocomis, but they were homogeneous, and showed little divergence and were of little value in solving the systematic boundaries of the infraspecific populations. Specimens from certain tributaries of the New drainage, such as Fox Creek, had higher circumferential and lateral line counts compared with those in other New drainage localities or those of the James and Roanoke drainages. The same characters for specimens in the Tar and Neuse Rivers had somewhat lower values compared with those of the James and Roanoke. We cannot attach any particular significance to these differences.
The breast is almost always fully scaled (Table 18). In a sample of 107 specimens from several drainages, the breast was about one half scaled in 4, three quarters scaled in 69, and fully scaled or nearly so in 34 specimens. N. leptocephalus differs from the micropogon group in the constantly high degree of breast squamation.
Tuberculation: The number of head tubercles in specimens from the James, Roanoke, and upper New drainages, segregated by sex and size groups is summarized in Table 19. A comparison of tubercle numbers by size-groups with the species of the micropogon group is shown in Figure 4. The typical, early tubercle development and pattern is illustrated in Figure 3. The distribution of head tubercles in nuptial males is shown in Figures 23, 24, 26, 27. Tuberculation in N. leptocephalus is characterized by a great reduction in numbers, the lowest for any species of Nocomis; by an enlargement of the tubercle size, the largest for any species of Nocomis; and by the absence of tubercles from the snout, subnasal, and lachrymal areas. The tubercles first appear as large, light spots in the internasal-anterior interorbital area, at about 30 to 40 mm SL. With increase in body length the tubercles increase in numbers and spread to the posterior occipital line in the adults. Some specimens of both sexes may not have visible spots at body lengths of 35 to 45 mm SL. The distribution and numbers of tubercles is the primary character in the separation of the three subspecies of N. leptocephalus. The data on tubercle numbers and distribution above (Table 19) pertains to the typical subspecies, in which the numbers are highest and the adult pattern most extensively developed. Analysis of the subspecific populations of N. leptocephalus will be treated in a separate paper. Tubercle numbers for N. l. leptocephalus are almost always fewer than 25. N. l. bellicus has the lowest number, 7 or fewer in the adults. The early tubercle pattern for N. l. leptocephalus, as shown for small juveniles in Figure 3, is actually the definitive adult pattern in N. l. bellicus.
N. leptocephalus has more tubercles on the pectoral fin rays than species of the micropogon group. Tubercles are present on pectoral fin rays 2 to 6 (8 specimens), 2 to 7 (18), 2 to 8 (23), 2 to 9 (1), and 2 to 10 (2).
Nuptial crests: The development of the crest in N. leptocephalus is similar to its development in the micropogon group. In N. leptocephalus large crests develop on smaller size specimens than in the micropogon group. It’s development is undoubtedly associated with age and maturity. The crests of N. leptocephalus become more greatly developed at a position somewhat more forward on the head than in the micropogon group, the swelling is often very large just posterior to the internasal area. Crests, moderate to large, were present on the following size-groups (in mm): 110–119 (1 specimen), 120–129 (4), 130–139 (5), 140–149 (10), 150–159 (5), 160–169 (4).
Pharyngeal dentition: Tooth numbers were 4–4 in 90 percent of 92 specimens sampled over the range of all three subspecies. Counts of less than 4–4 comprised the other 10 percent of the sample, these being 3–4 or 4–3, 3–3, and 2–4. The structure of the arch and the pitted surface is similar to the arch of N. micropogon.
Coloration: Two different primary color forms exist in living nuptial males of N. leptocephalus. N. l. leptocephalus has both forms as described by Lachner (1952:439). Those in the Roanoke and Neuse drainages have blue on the head laterally, an orangish lateral stripe, and variable amounts of shades of orange in the fins while in the Pee Dee drainage the body laterally is bluish instead of orange. Further observations by the authors and others have shown that the orange-sided form is also in the New and James drainages in the northern part of the range while the blue-sided form also exists in the Santee drainage south of the Pee Dee drainage. Additionally, males of the northern color form, at the height of nuptial coloration, have the lateral stripe abruptly changing from orange to olive-yellow, caudally, from about the vertical from the middle of the dorsal fin. Both forms at the peak of color intensity have a tan to brown nuchal crest and the dorsum above the lateral stripe is colored brown to bluish black.
Nuptial males develop a light to moderate blue on the lateral and ventral head, deepest on the opercle and on the cheek, faintest on the lower jaw and branchiostegal regions. The tubercle region is dark tan to brown in living, large males; the tan turns to dark olive after several hours in formalin. Of ten tuberculate males captured in the upper James drainage during 10–30 May 1963–1965, the deepest shade developed on the lateral head was moderate blue in one specimen, a light blue in five specimens (faint blue-gray or blue color absent on the ventral head) and the blue was not detectable in four specimens. In one specimen the light blue color turned noticeably darker while being examined out of water. The color also turned darker in all specimens with blue upon being in formalin for about fifteen minutes; those without the color developed a light or moderate blue. A postnuptial male with Saprolegnia over its tubercle scars was observed in the stream to have a fairly dark blue head. The blue head color of living and preserved N. leptocephalus is lighter than the bluish black color that sometimes develops on the head of recently preserved nuptial male specimens of N. raneyi and N. micropogon. The coloration of the head of the latter species is unlike the nuptial head color of N. leptocephalus in life.
The other nuptial color component in the northern male N. leptocephalus, radically different from the pink-rosy lower sides of the micropogon group, is the horizontal stripe of faint to dark, coppery to rusty-orange. This is developed in the scale medians from the postopercular bar to about the dorsal fin base level, within 3.5 scale-rows above to 1 scale-row below the lateral line at its widest point (anteriorly) and tapering slightly to the dorsal fin level. The edge of this stripe is fairly sharply delimited below by the whitish lower sides, somewhat less above since the dorso-lateral scale medians have brassy iridescence. The stripe then grades posteriad to the level above the anal fin base into a bright olivish yellow, this shade being more yellowish than the dorso-lateral region just above it, and then grades over the caudal peduncle to a less bright, light yellow-olive, about the shade of the dorso-lateral scale medians.
A deviation from the typical orange stripe was seen in nuptial males from Craig Creek, captured on 15 May 1963 where many fresh nests were present. In three of the five nuptial males taken, the scale medians were pink in the anterior lateral stripe area from the postopercular bar posteriad to the level of the anterior base of the dorsal fin; the most ventral row of scale medians in that area had a suffuse yellow added. The posterior portion of the midlateral stripe was typical olive-yellow. Many nuptial males more recently taken or observed in the upper James, several of which were from Craig Creek, had an orange area or stripe anteriorly on the side of the body.
The head coloration of all sizes except the nuptial males and some postnuptial males is very similar to that of the members of the micropogon group; it differs in that the dorsum is slightly lighter, ranging from tan-olive to dark olive. The inner “color ring” of the iris is wide, often occupying half of its width, and is colored bright orange, red-orange, or red. It is very rarely interrupted by the blackish outer iris color. Most specimens of N. leptocephalus can be separated in the field from the micropogon group by this character alone.
The body color grades from dark above to white on the belly. The color of the median of the scales, laterally on the body, tends to be lighter in some specimens, being more of an iridescent yellow-green than yellow-olive and often with brassy iridescence.
The dark midlateral stripe and basicaudal spot is often absent but prominent in some living and preserved juvenile and adult females and juvenile males. Preserved adult males rarely have a well-developed stripe and, when present, it is usually only on the caudal peduncle. An iridescent yellow-olive, middorsal stripe is generally present in all but the larger adult males; this soon turns darker than the adjacent dorsal area in formalin. A moderately dark lateral stripe and light middorsal stripe were seen in a highly tuberculate male of about 130 mm SL that was building a nest in the upper James drainage. No specimens were observed in life with a light greenish lateral band as seen in N. micropogon and N. platyrhynchus.
The fins are colored yellow-olive, usually with small to large areas of orange or, infrequently, slightly reddish orange. These colors are generally restricted to the rays but in some specimens they extend slightly into the membranes. The membranes are clear to clouded with white and some olive. An accentuation of the fin colors has not been noted in nuptial males. One nuptial male with intense body color had pale dorsal and caudal fins, lacking the yellow and orange, and it had only slight shades of these colors in the other fins. The pelvic fin rays were milky white along the branches. The fins of N. l. leptocephalus have a heavier deposition of melanophores along the rays and ray joints than in the micropogon group; it is heaviest in the dorsal fin, the rays appearing almost entirely black in larger nuptial males.
Some slight but rather consistent differences from the micropogon group are seen in the melanophore pigmentation of alcoholic specimens. The anterior margins of the scales are narrower, sometimes absent (when the scale just anterior entirely covers the scale pocket below), while the posterior margins of the scales are wider and do not have a sharply defined inner edge. Frequently the posterior margins do not extend to the actual posterior edge of the scales. Adult females have wider scale margins than adult males.
REPRODUCTION AND GROWTH.—The nest building, spawning, and reproductive behavior of this species is now well known, based on recent studies by Lachner, which will appear in the sixth paper of this series (p. 2), in which the reproductive behavior of all species of Nocomis will be treated. The age and growth of N. leptocephalus was treated by Lachner (1952); the largest specimen studied was a male in the 115–120 mm SL size class. In Table 19 we report on one male specimen in the 170–179 mm SL size class. After having examined about 30,000 specimens from over its range we encountered one collection of 11 tuberculate males that dwarfed all others known. This collection (TU 29534) was taken in Toxaway River, above Lake Toxaway, Transylvania County, North Carolina, on 14 July 1962. Ten of the specimens ranged from 192 to 214 mm SL, and the smallest was 172 mm SL.
MATERIALS EXAMINED.—The 124 collections from five drainages are given followed by the institution and number of collections housed.
James 25: CU 11, UMMZ 5, USNM 9. Roanoke 45: CU 6, UMMZ 1, USNM 38. Tar 10: CU 9, USNM 1. Neuse 12: CU 4, USNM 8. New 32: CU 11, UMMZ 1, USNM 17, DU 3.
DISTRIBUTION.—Nocomis l. leptocephalus is distributed on the Atlantic slope from the Potomac drainage in Virginia southward to the Santee drainage and west of the Appalachian Divide in the upper New drainage (partly shown in Figure 28).
The only known collections of N. leptocephalus from the Potomac drainage were taken in 1956 and 1958 by R. D. Ross and his associates from Back Creek and Middle River, tributaries of the upper South Fork of Shenandoah River, Augusta County, Virginia (all USNM collections). The single specimen known from the Rappahannock drainage was captured in South River, a tributary of upper Rapidan River, Greene County, Virginia, in 1951 by E. C. Raney and C. R. Robins (CU 46283). The York drainage was first included within the range of N. leptocephalus by Raney (1950:161). Six collections were available, all from tributaries of the Pamunkey River, one of the two major tributaries of the York estuary.
N. leptocephalus occurs widely within the James, Chowan, Roanoke, Tar, and Neuse drainages.
This species also has a wide distribution within the upper New drainage of Virginia but, apparently, it does not occur in its headwaters or the lower New. The farthest downstream record is one specimen taken in 1885 from Wolf Creek, Giles County, which enters the New just upstream from the Virginia-West Virginia state line. As known from collections and angling, this species is common in much of Big Stony Creek, the first major tributary just upstream from the mouth of Wolf Creek. Fox Creek is the uppermost tributary known to be inhabited, whose mouth is just downstream from the Virginia-North Carolina state line, where it was taken commonly in 8 collections from 6 well-spaced localities.
Ecology, Association, and Frequency of Hybridization
REGIONAL ECOLOGY.—The habitats of the four species of chubs are basically similar although there is some ecological variation within their geographic ranges. Generally, the species of Nocomis prefer clear, moderate to warm water streams of intermediate gradients. They are carnivorous fishes common over bottoms composed of gravel, rubble, and boulders, and usually with scant or no higher aquatic vegetation.
The primary ecological distinction between the species groups is their preference for different size streams. The three species of the micropogon group generally inhabit larger streams, becoming the dominant chub in streams averaging wider than 40–60 feet. This preference has been noted throughout most of the range of N. micropogon (Lachner, 1952:434–435 and included references; Gerking, 1945:15, 48; Trautman, 1957:297) and is likewise true for N. raneyi and N. platyrhynchus.
There may be limited upstream movement correlated with the breeding season. Miller (1964:316) noted that sometimes mature N. micropogon did not appear in certain breeding areas until early May in Catatonk Creek, Susquehanna drainage, New York. Raney (1950:161) mentioned an upstream movement of this species to spawning areas in April and May. We note that nuptial males, ripe females, and juveniles of N. raneyi were seen or commonly captured on 29–30 May 1964 in an upstream portion of Johns Creek averaging 15–25 feet in width. Whether the adults remain upstream after spawning in this relatively narrow section is unknown. A number of older adult males die soon after spawning.
Less information is available concerning the distribution and abundance of river chubs in the large river channels. They occur locally in certain rivers where adequate sampling was made. The James River where (atypical) N. raneyi was taken fairly commonly is 60 to 150 yards in width. N. raneyi was also captured in the larger sections of the Tar and Neuse rivers. Several adult N. platyrhynchus were caught by sport fishermen at localities in the main channel of the New River in Virginia where large nests, obviously of this species, were seen. Large nests have also been observed in the deeper waters of the Greenbrier River and the Knapp Creek tributary, West Virginia, over the past several years. Several records are available of N. micropogon from the lower Potomac River down to Great Falls just above tide level. Lachner (1952:434) mentioned that larger adults of this species are taken from deep river holes by fishermen; these and more recent observations were made in western Pennsylvania. H. M. Smith (1907:104) stated that in North Carolina N. kentuckiensis (Rafinesque) shows a preference for larger streams. His records are probably based on specimens of N. raneyi in the east and N. micropogon from the Tennessee drainage.
N. leptocephalus generally is most abundant in small streams, although there is usually much overlap with species of the micropogon group in intermediate stretches. This species is common to abundant in streams averaging from 10 to 50 feet in width, but it rarely occurs in extreme headwaters.
The juvenile and adult chubs commonly occupy both the riffle and pool habitats. These life stages were collected in about equal numbers in moderate to rapid riffles and short to moderate length pools of the upper James, Roanoke, and New drainages. They seldom occur in the long stretches of slack water. N. leptocephalus is frequently common in rapid, small mountain streams that are stocked with trout.
A stream bottom composition partly of gravel and rubble is an important factor in the maintenance of a sizeable population of Nocomis, especially since these materials are essential for nest building. Other bottom types in various combinations are occupied. The largest populations of the four species in Virginia and of N. micropogon in Ohio (Trautman, 1957:297), Indiana (Gerking, 1945:15, 48), and elsewhere are most commonly found where current-swept gravel, rubble, and boulders predominate. The decline or disappearance of populations of the widely distributed N. micropogon with increased turbidity and siltation is generally observed. N. raneyi populations apparently have also declined in more silted sections of the Roanoke drainage.
The well-established populations of N. raneyi, N. micropogon, and N. leptocephalus in typical Piedmont streams indicate their adaptiveness to this type of habitat. Lower gradients, sandy bottoms and somewhat greater turbidity generally characterize streams of this province but gravel, rubble, and sometimes boulder riffles occur. N. raneyi and N. micropogon do not tolerate sandy bottom conditions as extreme as those where N. leptocephalus is sometimes common; this may account for the apparent absence of N. raneyi from several Piedmont tributaries of the Roanoke and its rarity in others such as the middle Dan River.
The rarity of chubs in streams of the Coastal Plain province is probably related to the general absence of riffles, exposed gravel and rubble. The courses of the eastern rivers through the Fall Zone, the transitional belt of varying width (whose eastern limit is the Fall Line) between the Piedmont and Coastal Plain, are generally marked by a series of rapids and cascades which provide some habitat. From the James northward the Coastal Plain rivers and most of the generally small tributaries are tidal or brackish up to the Fall Line. However, the rivers south of the James do not drop to sea level so quickly, and their channels are on crystalline rock with rapids occurring for about 20 miles after entering the Coastal Plain (Fenneman, 1938:13, 129). The presence of riffles in some upper Coastal Plain tributaries may be accredited to drainage of higher terraces during which time some have cut deposits to the crystalline rocks of the Piedmont (Clark, et al., 1912:26). Headward extension with exposure of gravel and rock by erosion, beginning with the post-Pleistocene elevation of the Coastal Plain (Clark and Miller, 1912:200), seems to be still in effect.
N. raneyi and N. micropogon occur in the Fall Zone and sparingly in the upper Coastal Plain province. Jordan (1889a: 109–110) noted that a rapid and rocky portion of Swift Creek, James drainage, just below the Fall Zone has a fauna “essentially that of the upland streams.” The southeastern limit of N. micropogon (excluding the upper Savannah drainage population) is the James drainage, and for reasons stated above, its distribution terminates at or near the Fall Zone. N. raneyi has available a rather wide lowland region in southeastern Virginia and northeastern North Carolina but few streams are probably suitable. It is not known from the Coastal Plain portion of the Chowan and Roanoke drainages although one collection from the Chowan is in or near the Fall Zone. Of the 24 collections of N. raneyi from the Tar and Neuse drainages seven were taken from the Fall Zone and nine below. Of the latter, some were from moderate to rather large streams with riffles, bordered at least in part by swampland, and with lightly brown-colored water. One stream, Nahunta Swamp Creek, Neuse drainage, averages only 10 feet in width. No species of Nocomis is known to inhabit the predominantly slow blackwater type of stream or inner portions of swamps.
N. leptocephalus was taken in the Fall Zone from the York and the Chowan drainage southward to the Neuse and in most cases with N. raneyi. The former species apparently avoids Coastal Plain streams in the central Atlantic region, for we know of only one collection from this area in a Roanoke tributary.
N. micropogon, the only member of the species group inhabiting central United States, shows a distinct preference for recently glaciated regions within the northern and middle, lowland portions of its range in the Ohio River drainage basin. It is absent from the region covered by the Illinoian glacier south of the Wisconsin glacier limit in Indiana (largely the lower Wabash drainage) except for a borderline record (Gerking, 1945:15, 48–49, map 25). The habitat here is generally unsuitable, prevailing conditions being similar to or more limiting than those on the Atlantic Coastal Plain. Jordan (1889a: 160) stated that the lower Wabash tributaries “are mostly sluggish and yellow with clay and mud”. Gerking (1945:17) indicated the same for the lower Wabash flood plain. Trautman (1957:11) noted that most unglaciated and hence undisturbed and almost base-level streams of southeastern Ohio have no well-developed riffles while the glacial streams in this region have relatively high gradients. Some of the latter streams are inhabited by the river chub (Trautman, 1957:296–297, map 61). Its absence from other Ohio River tributaries in Kentucky may be related to similar sluggish conditions.
Higher aquatic vegetation is not common in most areas inhabited by the micropogon group and N. leptocephalus. Generally, habitats suitable for good plant growth are borderline situations for these species. An exception is the well-established growth of water willow, Justicia, in portions of the upper James and New drainages supporting a thriving chub population, and similarly reported in Ohio by Trautman (1957:297). This stout emergent herb often grows in dense stands along the shore and in shallow pools and riffles. Shockley (1949:259–261) found stream areas supporting a growth of Justicia to be more productive than those where it was absent. Submerged portions of water willow usually show retention of much organic detritus and an abundance of aquatic insects. Chubs also are found in streams with good growths of the algae Cladophora and the vascular riverweed, Podostemum. Lachner (1950:230; 1952:434) found a large population of N. micropogon in streams with an abundance of algae, chiefly Cladophora, and vascular plants (several species of Potamogeton), in Lake Ontario tributaries of New York. Smith and Bean (1899:183) reported this species as common over grassy bottoms of the Potomac River.
Lachner (1950) found the food of N. micropogon in western New York to be largely aquatic insect larvae with lesser percentages of Crustacea and Mollusca. In many instances, algae, chiefly Cladophora, and vascular plants were taken in large quantities. Much of the plant material is probably ingested incidentally with animal food since it appears to be undigested. Fishes are rarely included in the diet of chubs. Differences in food ingested by N. micropogon were found among young, juveniles, and adults and, within the juvenile stage, among certain seasons (Lachner, 1950). Comparative food studies of the species of chubs occurring sympatrically might reveal differences, especially since N. leptocephalus has a long, whorled intestine, differing from all other Nocomis. Flemer and Woolcott (1966:83–84) analyzed food intake of 249 specimens of N. leptocephalus from a typical, lower piedmont tributary of the James River just above the Fall Line. They found a great preference for plant food by this species, forming a high percent of total items eaten. They concluded that the plant material was specifically selected by N. leptocephalus.
ASSOCIATION AND FREQUENCY OF HYBRIDIZATION.—The following discussions refer to the extent that the species of chubs occur syntopically (occupy the same macrohabitat) or together (Rivas, 1964). All species occurring in the same drainage have been collected together. However, differences among the species of the micropogon group in their frequency of occurrence with N. leptocephalus are revealed and the apparent noncompatibility between N. raneyi and N. micropogon is discussed. The data support our preceding statements concerning stream size preferences. The relative frequencies of hybridization between certain of the species are reported below. Factors involved in hybrid formation will be discussed in papers 5 and 6 on the study of Nocomis hybrids and reproduction.
The data summarized in Table 26 are derived largely from intensively sampled tributary systems of the James, Roanoke, and New drainages. In order to eliminate bias from small collections and those taken by selective angling, the data indicating syntopy or allotopy (Rivas, 1964) are from collections in which a minimum of ten specimens were taken. The sampling intensity for the four species in intermediate size streams was about the same.
In the James drainage, N. leptocephalus was taken with N. raneyi in 89 percent of the collections (Table 26) from the Craig Creek system (Figure 20). In approximately the center of this system, in five collections from three localities in lower Craig Creek just below Newcastle, N. raneyi was always found to be the more abundant species, appreciably outnumbering N. leptocephalus (429 to 239 specimens). In upper Craig Creek within a radius of about 3.5 miles above Newcastle, N. raneyi was taken in three collections from two localities, but in less numbers than N. leptocephalus (26 to 52 specimens). Two juvenile N. raneyi were taken about 5.5 miles above Newcastle, but none have been found above this area. N. raneyi was found to extend well up Johns Creek to within 0.75 mile of Maggie, the latter being where N. micropogon and hybrids of N. leptocephalus × N. micropogon were taken. In three large collections, two in spring and one in summer, from within 3 miles of Maggie, 58 N. raneyi and 92 N. leptocephalus were taken, the latter species being more numerous in each. The stream in this area averages about 15 to 25 feet wide. N. leptocephalus apparently becomes progressively less numerous below the Newcastle area. A sparse spawning population occurs down into the middle section of lower Craig Creek. Only N. raneyi was taken in four collections from the lower 3 miles of Craig Creek.
The Blackwater River of the Roanoke drainage in Franklin County, Virginia, is a well-surveyed stream in which similar distributional relationships were found between N. raneyi ahd N. leptocephalus as observed in Craig Creek (Figure 29). The Blackwater is a typical Piedmont river; its lower portion is about equal in size to lower Craig Creek. Of the 22 collections (15 localities) in which N. raneyi was taken, all but two are in the main channel from lower North Fork near its mouth in Roanoke River. N. leptocephalus was taken at all but one location. A progressive upstream decrease in the relative numbers of N. raneyi is indicated when the main channel is divided into three sections of approximately equal lengths and by calculating the percentage composition of each species per section. N. raneyi comprised 93 percent of the 246 chubs taken in the downstream section; in the middle section, 76 percent of 296 specimens; and upstream, 21 percent of 505 specimens. The progressive change in relative numbers is also indicated in almost all individual collections. In the two collections from the lower portion of the most downstream, and major, tributary of the Blackwater, N. raneyi comprised only three of the 49 chubs taken. N. leptocephalus is common to abundant in the remainder of the tributary system in which N. raneyi is absent or rare.
The distributional relationships in the intensively collected lower South Fork and upper main channel of the Roanoke River, Montgomery and Roanoke counties (Figure 20) are somewhat different. Although within the Appalachian province, the bottom of these streams is generally much more silted than that of Craig Creek to the north, even though riffles are more frequent in the Roanoke River than in Craig Creek. The Roanoke River at the city of Salem, Virginia, averages about the same in width as lowermost Craig Creek. N. leptocephalus was captured in all nine lower South Fork collections (five localities) in which N. raneyi was taken. The uppermost locality is near the town of Alleghany Springs, where the stream averages about 30 to 40 feet in width. N. raneyi comprised only 3 percent of 510 chubs captured. Only N. leptocephalus was taken in numerous additional collections from lower South Fork and farther upstream. Although N. raneyi showed an expected increase in relative numbers in the downstream direction, it was taken from the upper main channel only at seven localities in ten collections, in which it accounted for 29 percent of the 228 total chub catch. The large majority of the recent collections (through 1970) from the main Roanoke from Salem upstream did not include N. raneyi, although N. leptocephalus was, at least, common. The only collection (USNM 197684) in which N. raneyi was common was from a bottom area with less silt than at all other main Roanoke localities. The upper Roanoke record from below Salem was made in 1888. The Roanoke River from the city of Roanoke downstream is now badly polluted.
The known case of hybridization between these species occurred in the upper Roanoke. Ten specimens of N. leptocephalus, one of N. raneyi, and two hybrids were taken.
Elsewhere in the Roanoke and in the Chowan, Tar, and Neuse drainages about the same distributional relations prevail between N. raneyi and N. leptocephalus as that in the James and Roanoke drainages.
In contrast to N. raneyi, N. micropogon was found less frequently with N. leptocephalus in the James drainage (61% of the comparable collections, Table 26). Included in this percentage are six collections with hybrids of N. leptocephalus × N. micropogon taken with nonhybrid specimens of only one or the other species. The occurrence of one or more hybrid specimens was relatively high, hybrids being taken in 60 percent of 25 collections. Only three of the 15 hybrid collections contained less than ten specimens; thus it appears that this hybrid combination may occur fairly frequently (87 specimens examined). Hybridization occurred in almost all tributaries in which the two species were taken together and, within each tributary, in a frequency ratio about equal to the relative number of collections from that tributary.
Ecological conditions in the upper New River basin of the Blue Ridge and Appalachian provinces are varied but generally montane. Individual tributary systems in Virginia are described by Ross and Perkins (1959), Shoup (1948), and Burton and Odum (1945). In the Virginia portion of the upper New the relations between N. platyrhynchus and N. leptocephalus are rather similar to those between the latter species and N. micropogon in the James drainage. N. leptocephalus was absent from the West Virginia portion of the New River drainage; collections of N. platyrhynchus from this area are not included in the following comparison. The two species were syntopic in 63 percent of 16 comparable collections. Hybrids were taken in three (20% of 15 samples), each from a different tributary system.
The distributional relationships between N. raneyi and N. micropogon in the James drainage have ecological and zoogeographical significance. N. micropogon has been taken about as commonly and in equal numbers throughout the James, exclusive of the Craig Creek system, as N. raneyi has been taken in the Craig system (excluding the stream section a short distance below Newcastle that is eutrophically enriched). N. micropogon, however, is rare within the Craig system, being taken in only four or five collections and in small numbers (see p. 49). Two of the collections are from the headwaters, a habitat marginal for the micropogon group because of the small stream size. N. raneyi is generally distributed downstream and one of the most dominant members of the fauna. As might be expected to occur between closely related ecological homologues, this pattern of distribution and relative abundance suggests that N. micropogon is being largely displaced from the Craig Creek system (as well as the lower Catawba) by competition with the apparently better adapted N. raneyi. N. micropogon may be the more successful of the two species in tributaries upriver from Craig Creek and downriver in the Maury River system. Perhaps here the period of their contact, if any, has been of such short duration that changes in distribution have not occurred. One hybrid N. micropogon × N. raneyi was taken in lower Johns Creek and one was taken in lower Catawba Creek (see p. 49), where observations in the last several years have revealed only high breeding populations of N. leptocephalus and constant breeding numbers of N. raneyi.
SUMMARY.—On the basis of similar ecological preferences, and apparent competition in the Craig Creek system, the three species of the micropogon group are regarded as ecological homologues. N. raneyi almost always coexists with N. leptocephalus in moderate-size streams. There is a definite tendency for N. leptocephalus to occur less frequently with N. micropogon and N. platyrhynchus. The greater frequency of occurrence of N. raneyi with N. leptocephalus suggests that competition is somewhat more intense between the latter species and N. micropogon and N. platyrhynchus.
There are differences among the species of the micropogon group in their frequency of hybridization with N. leptocephalus, this being inversely proportional to the frequency of syntopic occurrence. Hybridization between N. raneyi and N. leptocephalus is rare, only two specimens are known from a habitat more disturbed than that of the other streams which were intensively sampled. N. micropogon and N. leptocephalus hybridize comparatively commonly (87 specimens known). Although the number of collections in which N. platyrhynchus and N. leptocephalus were taken together, and the number of specimens involved, is less than those of the other two combinations of chubs, the frequency of hybridization is appreciably (5 specimens are known to us) less than that of the N. leptocephalus × N. micropogon combination.
Dispersal: Biological and Geological Evidence
The ichthyofauna of central eastern United States presents many stimulating zoogeographical problems, such as the dispersal of the species of the micropogon group, the leptocephalus group and of several of their associates. Possible routes of entry into various drainages have been recognized. These are often indicated by biological evidence of present distributional patterns and relationships and past geological events. In some cases, however, routes have been conjectured only from biological evidence since complementary geological information is meager or nonexistent. While this procedure may not be justifiable since distribution patterns are known to change, the fact remains that populations have crossed drainage divides in some manner. Therefore, attempts are made herein to offer explanations of how dispersal may have occurred. Past and present ecological conditions are considered.
Three kinds of geological events have operated in the dispersal of chubs. These are stream captures, eustatic changes of the Atlantic Coastal Plain, and Pleistocene drainage modifications. An obvious fourth means of dispersal is movement from one drainage system to another through past and existing interconnecting main rivers.
STREAM CAPTURE.—The geological literature on the occurrence of and evidence for captures in western Virginia and adjacent portions of West Virginia is extensive. It appears that numerous captures have occurred which made it possible for stocks of fishes to spread among several drainages. Major papers treating this subject are by Wright (1931, 1934, 1936), Thompson (1939), and Dietrich (1959), and numerous additional references are in their bibliographies. It is interesting to note that Dietrich (1959:30), a geologist, invoked biological evidence from fishes in support of a geological hypothesis. A century ago, Cope (1869) concluded that the faunal similarities among opposing drainages in western Virginia may have resulted in part from headwater transfers. Recently Ross and Carico (1963:7–12) documented two stream captures to support hypotheses on the dispersal of certain fishes between the New and Tennessee drainages. Other authors (Robins and Raney, 1956:31; C. R. Gilbert, 1961:456; 1964:106) mention possible use of stream captures by fishes in western Virginia. Other instances of isolation, subsequent differentiation, and present distributional patterns of fishes in this region can be accounted for primarily by these events.
When invoking stream capture as an agent of dispersal, the biology of the fishes, ecology of the region, and nature of stream capture are factors that should be considered. Some factors are species abundance, interspecific competition, habitual movement and migration of species, size and flow of streams, and other ecological expedients or barriers. Some important geological factors are the magnitude of the theater of stream capture, the duration of the water connection between opposing drainages and the numbers and sequence of captures that occurred in a region. The absence of certain species today from a drainage from which its associated species have successfully entered is perplexing and may be explained by subsequent change in stream conditions, competition, and extirpation.
Another problem in correlating dispersal with stream capture is the uncertainty or lack of information concerning the dates of occurrence. Thompson’s statement (1939:1353) that “captures that can now be definitely recognized belong to comparatively recent geologic times” provides a positive argument for adopting the concept of capture. Many captures described can probably be assigned to post-Harrisburg time, a period of stream rejuvenation which began in the late Tertiary (Wright, 1934:38). In addition, since Wright (1931:246) pointed out that evidence of capture soon disappears with active dissection of the capture area, we may safely assume that many more captures occurred than are presently detectable [see also Ross (1969) and Jenkins, Lachner, and Schwartz (ms)].
EUSTATIC CHANGES OF THE ATLANTIC COASTAL PLAIN.—Dispersal can be equated with eustatic changes of the Atlantic Coastal Plain during or somewhat after Pleistocene time. During such times the several drainages now entering the western shore of Chesapeake Bay, from south of the present Susquehanna River mouth to the James drainage, were sometimes tributary to the extended freshwater Susquehanna, hereafter called the Greater Susquehanna River (Shattuck, 1906:134, pl. 31). This resulted from a general increase in elevation of the central Atlantic Coastal Plain and/or a lowering of sea level (Flint, 1957:270). Lougee (1953:264–265) stated that sea level was at or near its lowest level
during the last glacial climax. It is also possible that a drainage similar to the Greater Susquehanna was in existence during earlier periods of Pleistocene glaciation (Flint, 1957:270).
At the same time it is likely that similar events occurred in North Carolina, in present Albemarle Sound where the Chowan drainage would have been a tributary of the Greater Roanoke River, and in Pamlico Sound where the Tar and Neuse drainages would have been conjoined to form the Greater Pamlico River. Some authors consider the Chowan to be a present Roanoke tributary, but this is not likely since the large lower channels of these drainages are separated by tidal flats (Hubbs and Raney, 1944:5), thus largely or entirely preventing dispersal of most freshwater species between them. Opportunity for dispersal of species considered herein through extended channels is thought not to have occurred earlier than the Pleistocene since there was a Pliocene submergence of the Coastal Plain (Clark and Miller, 1912:215).
PLEISTOCENE DRAINAGE MODIFICATIONS NORTHWARD.—Several means of late Pleistocene dispersal in the Great Lakes and adjacent regions are discussed under the following species accounts.
The vast preglacial Teays River system (Figure 30) served as a reservoir for a large portion of the North American ichthyofauna and provided an early means of dispersal. Its upper and middle portions are of much importance in this study. The upper Teays, approximately, is regarded as the New-Kanawha River system after the occlusion of the middle Teays and formation of the present Ohio River during Pleistocene glaciation; however, some drainage relationships of the Teays with the earlier Ohio, Tennessee and Cumberland drainages are uncertain. Detailed discussion of the history of the Teays system is presented by Leverett (1902), Tight (1903), and others and, of a more general nature, Janssen (1953) and Flint (1957:170–171).
Nocomis platyrhynchus: The basic stock of the micropogon group probably occupied the middle and upper Teays River system (Figure 30), perhaps during late Pliocene and certainly part of Pleistocene times.
The bigmouth chub may represent the form closest to the basic stock of this group and probably evolved in the New River drainage (upper Teays) where it is presently confined. Kanawha Falls, approximately 24 feet in height, and Sandstone Falls just upstream, apparently served as an effective barrier for isolation of the segment of its ancestral stock from that below. The period of inception of Kanawha Falls is not dated, however; it may have been during pre-Pleistocene times. If the ancestor of N. platyrhynchus was below the Falls and was prevented from directly entering the New, it may have made its entry through stream capture between Elk River of the Kanawha drainage and Gauley River of the New, an event for which there is geological evidence (Campbell, 1896:669–670).
The New drainage has a unique fauna, partly characterized by several endemic forms and further marked by the absence of a large fraction of the fauna of the Ohio basin. Ross (1959) listed and discussed the members of this fauna. Forms, in addition to N. platyrhynchus that most likely evolved in the New are the cyprinids Notropis scabriceps and Phenacobius teretulus, and the percids Etheostoma osburni and Etheostoma kanawhae. Most of these species occur widely within the New, from its headwaters to Gauley River, but none are known below Kanawha Falls. It therefore seems that Kanawha Falls has been the major impediment to their establishment in the Kanawha drainage. The several tumultuous rapids, cascades, and low falls in the New River gorge of West Virginia (Campbell and Mendenhall, 1896; Reger, 1926:3, 90) also may limit downstream dispersal.
Despite its endemics, the New drainage has a depauperate fauna in relation to the size of the basin. There is evidence to suspect that the earlier New drainage fauna was similar to that of the Ohio basin and that its depauperate nature probably resulted from extirpation. The Kanawha drainage fauna is essentially that of the middle Ohio basin. Particular examples of probable extirpated forms are species of redhorse suckers (genus Moxostoma, subgenus Moxostoma). The occurrence today of some species of this group in Atlantic slope drainages from the James southward may be best explained as their having crossed the Appalachian divide from the New drainage. Ross and Perkins (1959) discuss factors that possibly combined to limit this fauna. Some of the ecological factors mentioned (ibid.: 10) were “the general absence of aquatic plant life, the narrow poorly developed flood plains, the hard bottom, steep gradient and high velocity which by abrasion kept the stream bottom well scoured.” Speciation in the New also may have been in response to some of these conditions.
N. micropogon: Certain stages in the dispersal of N. micropogon can be readily postulated whereas others are not clear. The present range of N. micropogon south of the Great Lakes drainage may reflect much of its early distribution. A population of the ancestral stock of the micropogon group remaining below Kanawha Falls in the Teays system probably speciated to N. micropogon. Subsequently this species may have occupied the middle Teays drainage of West Virginia, Kentucky, Ohio, and Indiana including the future upper Wabash drainage (Figure 30). Alternatives available for its ingress to the Tennessee and Cumberland drainages are discussed on page 50. N. micropogon probably never occurred west of the Mississippi, even if it once did reach the extreme lower Ohio basin. This would attend the possibility that the lower Teays (early Mississippi) and Ohio were separate systems (Fenneman, 1938:89–90). On the other hand, these large rivers conjoined in the extensive lowland region would still have been more of a barrier than an agent of dispersal.
More than one region may have served as a refugium from which the river chub spread northward during inter- and postglacial times, to repopulate northern Ohio River tributaries and enter the Great Lakes drainage. Evidence does not favor the lower Wabash drainage as having been a refugium for a Teays stock. This drainage was almost entirely overwhelmed by the Illinoian glacier (Gerking, 1945). During glaciation the nonglaciated lower Wabash, although certainly of a character different from that today, probably was adversely affected. Seasonal and diurnal variations of proglacial discharge, sometimes perhaps catastrophic in proportion, combined with sediment and ice loads (Dyson, 1962:77–80; Flint, 1957:174–175) would have greatly restricted fish inhabitants.
Ancestors of populations now occupying drainages of unglaciated Kentucky and West Virginia are likely to have moved northward after glacial recession into northern Ohio River tributaries.
Perhaps the most important center of dispersal for river chub populations now occupying the Ohio, Great Lakes, and Atlantic slope drainages was a southern portion of the Old Allegheny River system, probably the Old Lower Allegheny. The three separate preglacial component drainages of the Old Allegheny system flowed northward to the present Lake Erie basin (Leverett, 1902:129–138). N. micropogon presumably entered the Old Allegheny through stream capture with the Teays. Following interconnection of the Allegheny components with the Ohio River due to glacial advance, the river chub may have spread downriver into Ohio drainage tributaries when uncovered from the ice sheets. Deglaciation of the northern portion of the Allegheny basin would have permitted influx to glacial Lake Maumee (Lake Erie basin), through Conneaut Creek, a present Lake Erie tributary in Pennsylvania whose upper portion was involved in a drainage reversal with the Allegheny in an area of the Old Middle Allegheny (Leverett, 1902:214). Its presence in the lower Allegheny during glacial times would enhance the probability of its reaching the vicinity of Conneaut Creek before its diversion. Dispersal to the upper Allegheny probably postdated the Cuba Outlet connection between the Allegheny and Genesee drainages since N. micropogon is not known from the latter. Ross (1958a: 17–18) discusses a complex distribution problem in this region involving subspecies of Campostoma anomalum.
Once N. micropogon entered the developing Great Lakes basin in the Lake Maumee area, it could have entered future Lake Erie tributaries in Ohio and Ontario and the Maumee drainage of northeastern Indiana. Passage to the Wabash drainage from the Maumee through the Fort Wayne Outlet would follow if this outlet remained open sufficiently long after the Conneaut Creek diversion and/or if ample water connection through the low divide in the Fort Wayne region existed during more recent periods of high water (Gerking, 1945:8).
Dispersal northward could have ensued into the developing Lake Huron basin including its tributaries in Ontario south of Georgian Bay and those of the eastern shore of the Lower Peninsula of Michigan. Some movement to the western shore of the Lower Peninsula may have been around its northern end, but this is unlikely since the river chub apparently did not reach (or has not persisted in) streams of the Upper Peninsula. The eastern shore of Lake Michigan was probably populated after movement across the Lower Peninsula through a later stage of the Grand River connection. Since N. micropogon has not appeared in the Illinois drainage, it probably did not reach the region just north of the upper Illinois until after the Chicago outlet was closed.
Dispersal northeastward from Conneaut Creek could have followed the development of Lake Erie and thence entry of streams of the present western Lake Ontario basin, probably during the Lake Lundy stage. Entrance to the Lake Ontario basin at a later stage would apparently necessitate passage over and establishment below newly created Niagara Falls (Fenneman, 1938:495, 499).
Natural occupation of Lake Ontario streams may have progressed slowly. A notable (52 mile) break occurs in the river chub’s distribution in the southern Lake Ontario drainage, from just west of the mouth of the Genesee River to 16 miles west of the mouth of Oswego River. Two records break the distributional gap, both from the same area and constitute the only occurrence in a tributary of the eastern lakeshore (excepting those from the Finger Lakes drainage). Dispersal along the lakeshore may involve a relatively long period of time for the river chub, especially since for long distances only small streams enter the lake and unsuitable habitats are present (these conditions not permitting establishment of large populations for continuing outward dispersal). Since the New York lakeshore tributaries have been, comparatively, thoroughly collected (Greeley, 1940:42), the easternmost records may indicate local extirpation or unsuitable habitat in the gap area, bait-minnow introduction, or dispersal by natural means. All four alternatives are feasible. Some of the streams do not provide the larger water type of habitat for the river chub; however, this species is common in a relatively short, western Lake Ontario streams and in Salmon Creek, its easternmost stream before the gap. The lower Genesee drainage, whose mouth is just east of Salmon Creek, would provide a large area of apparently suitable habitat. The area is heavily fished by sportsmen. The hardy chubs, frequently used as bait, could have been discharged from a bait pail. Natural means of dispersal would entail the river chub moving northward from the Susquehanna drainage, via the Horseheads Outlet during the Lake Newberry stage (Fairchild, 1934:1097, fig. 8, 1099), through the Finger Lakes drainage to the lakeshore. A problem with the latter hypothesis is that N. micropogon is relatively rare, and the recent records are widely scattered in the Finger Lakes basin, while it is common throughout the adjacent upper Susquehanna. Greeley (1928:84, 97) stated that the Finger Lakes drainage was extensively collected, and reported only a single collection of the river chub from Catharine Creek, the Seneca Lake inlet. He attributed its presence there to a canal connection with the Susquehanna drainage. Although this is possible, it is now known to have a wider, although localized, distribution in the Finger Lakes drainage which may have resulted from the same factors considered for the Lake Ontario shore area.
The river chub probably occupied at least some of the Atlantic slope drainages during glacial times and it had a history somewhat different from that which occurred northward and westward. Evidence for a relatively long eastern occurrence is that its spread throughout all the drainages of its central, eastern range probably required considerable time, and its movement throughout most of the James should predate the entry of N. raneyi into this drainage. Populations south of the Susquehanna may be derived from stocks older than the northern forms. C. R. Gilbert (1964:109) has given biological support in suggesting that Notropis cornutus entered this area at an early time.
N. micropogon possibly entered the Susquehanna by stream capture from the Allegheny drainage. This view conflicts with Bailey’s statement (1945:125) that “some Great Lakes forms found in the Susquehanna system are absent from the upper Ohio basin; thus, stream capture from that [latter] source fails to supply a tenable explanation” for dispersal to the Susquehanna.
N. micropogon probably first gained the central Atlantic slope by entering the Potomac drainage through stream captures from either or both Youghiogheny and Cheat River tributaries, Monongahela drainage (once part of Old Allegheny system). Biological and/or geological evidence for these captures is cited by Ross (1958b: 5–6), Gibbs (1957:205), and Schwartz (1965). Dispersal to the James and Rappahannock was possible during a stage in the development of the Shenandoah River of the Potomac, during the period the Shenandoah attained its drainage by a series of piracies (Stose, 1922:7; Watson and Cline, 1913).
The Greater Susquehanna River may have been an ample means of faunal exchange among all Chesapeake Bay tributary drainages, thus accounting for the presence of the river chub in the York drainage, the minor Chesapeake drainages, and the Appomatox River which now enters the James River estuary. Certain factors, however, militate against assumption that the Greater Susquehanna was the major means of dispersal among its tributaries. This obviously was a large river, certainly so during late Pleistocene periods of precipitation, in which smaller stream and some large stream fishes would not have found suitable conditions. Lower Potomac species such as Ericymba buccata, Percopsis omiscomaycus, and Percina caprodes semifasciata might be expected to have spread more widely through the Greater Susquehanna but are not known south of the Potomac on the Atlantic slope. The presence of Notropis s. spilopterus in this region, only in the Susquehanna and Potomac including their lower portions, is good reason to believe it entered the latter by headland stream capture (Gibbs, 1957:194–205, fig. 3) and not through the Greater Susquehanna. Other factors may confuse this issue. Some species may have entered the upper portions of larger Chesapeake drainages at a relatively recent period and sufficient time for spread through the Greater Susquehanna may not have been available.
Our interpretation of the relationships of the species of river chubs and their populations in the lower New, upper Monongahela, and Potomac drainages is given further credence by consulting the stream history of the area. We have discussed certain morphological similarities that suggest an exchange of river chub forms occurred between the New and Monongahela. A likely avenue of exchange would be a capture that occurred between the upper Greenbrier River of the New and upper Cheat River of the Monongahela. This is supported from geological evidence by Fridley (1933) and Wright (1934:55). Additional biological evidence for a capture in this area is the distribution of Percina oxyrhyncha. This species is widespread in the New-Kanawha system but the only known population in the Ohio system above the Kanawha River mouth occurs in the upper Monongahela (Hubbs and Raney, 1939:1–3). (Additional possible agencies of ingress of N. micropogon into the New drainage are the piracies of Greenbrier River tributaries by three tributaries of Back Creek of the upper James drainage (Thompson, 1939:1352–1353) and the two captures documented by Ross and Carico (1963:7–12) involving Holston River tributaries of the upper Tennessee drainage, and tributaries of the upper New drainage). We have also indicated a relationship between the Potomac and Monongahela populations of N. micropogon. A stock of the Monongahela form of N. micropogon could possibly have entered the Potomac drainage through the capture discussed by Schwartz (1965) for Etheostoma blennioides, and subsequently spread throughout the latter drainage.
The entrance of N. micropogon into, and spread within, the southwestern Ohio basin includes several complex facets. (See Figure 7 in Lachner and Jenkins, 1967, for the distribution of N. micropogon and other Nocomis in this region.) The middle Teays is shown (Figure 30) as it flowed in pre-Pleistocene times (through Ohio, Indiana, and Illinois); this portion may have reformed during Pleistocene interglacial periods (Janssen, 1953). The latter possibility is discussed below since N. micropogon may have evolved during the Pleistocene. It is not known whether the Kentucky and Licking drainages emptied directly into the middle Teays or into the early Ohio River.
We presume that N. micropogon evolved from a stock of the bigmouth chub, N. platyrhynchus of the New River, the latter believed to be the most primitive form in the micropogon group. The early range of N. micropogon within the region of the present southwestern Ohio basin possibly included only the Big Sandy drainage, shown to be a Teays tributary (Leverett, 1902), and perhaps the Licking and Kentucky drainages. Another alternative, its having evolved in the most southwestern parts of the Ohio basin (Cumberland and Tennessee drainages), is unlikely since it would have needed considerable time to attain its wide range in nonglaciated regions east of the Ohio basin, and since much of its range in the southwetsern Ohio basin appears to be recently acquired.
The two alternatives available for the spread of N. micropogon deep into the present southwestern Ohio basin are through possible main river connections with the lower Teays or lower Ohio River, and/or through stream captures with upper or middle Teays tributaries (present middle Ohio basin) in the region about southwestern Virginia and eastern Tennessee and Kentucky. Some of the evidence against the first alternative is the lowland and large river conditions the species would have to traverse in the long bend of the middle and lower Teays. It could have used the more recent and shorter Ohio River after the present course of the Ohio was determined by glacial advance but this route is also unlikely due to river conditions similar to the Teays. If N. micropogon did use a large river pathway downstream to the Green, Cumberland, and Tennessee, upon entering these drainages it could have encountered competition from N. effusus, a species more adapted to the lower gradients. Their distributional relationships in the Cumberland (Lachner and Jenkins, 1967) support the latter hypothesis. These species are allopatric except for being collected together on one possible occasion near a fringe of their ranges. Even if the latter record is valid, this does not mean that close competition is lacking. The absence of N. micropogon from the Green drainage is evidence that it did not use a downriver course to the Cumberland and Tennessee or, if it did, that this species and N. effusus were incompatible in the Green, the latter being dominant.
Ingresses of N. micropogon into the Tennessee and Cumberland drainages were most likely through stream capture since there is considerable biological evidence for this, some geologic evidence, and a bulk of evidence against lower main river entry. The favored sequence of entrance may have been from Big Sandy to Tennessee, then into the Cumberland, and then the first (or another) entrance into the Kentucky drainage. It could have been attained the Kentucky and Licking drainages if these were middle Teays tributaries and probably entered them once they were connected to the Ohio River.
Evidence is given above that the early range of N. micropogon included the Big Sandy drainage, a tributary of the middle Teays and presently a middle Ohio basin tributary. Much of the upper Tennessee drainage of Virginia is adjacent to the upper Big Sandy. An unknown capture(s) is invoked to account for transferal of N. micropogon from the Big Sandy to the Tennessee. This species would not have entered the Tennessee from the adjacent portion of the upper Teays (present New drainage) since its cognate, N. platyrhynchus, occurs therein. N. micropogon has spread throughout most of the middle and upper Tennessee, but is apparently absent from most of its lower portion. Since the habitat in some lower Tennessee tributaries appears quite suitable, its absence may be related to a somewhat slow downriver dispersal rate, large stream size and/ or lower stream gradients. The many uninhabited streams are additional evidence that N. micropogon entered the Tennessee from the northeast and not through the lower Teays or Ohio Rivers.
Stream capture involving the Tennessee and upper Cumberland is most likely. Even if N. micropogon had dispersed upriver in the Cumberland through the large area occupied by N. effusus, the former species would have had to surmount Cumberland River Falls—an impossibility—in order to attain its distribution throughout the upper Cumberland. The upper Cumberland has a depauperate fauna (Kuehne and Bailey, 1961) compared to those of the middle Cumberland and upper Tennessee. N. micropogon was either one of the relatively few species to make the crossing, or to survive thereafter.
The well-established population of N. micropogon in the middle Cumberland was probably derived from washing over the falls and/or from stream capture between Big South Fork of the Cumberland and the Tennessee or between Rockcastle River, or surrounding tributaries of the Cumberland and the Kentucky drainage. Explanation of the presence of the population in the Caney Fork system, middle Cumberland, presents greater difficulties. N. micropogon may have entered Caney Fork directly through stream capture with the middle Tennessee drainage. Possibly, it is a relict population predating farther upriver movement of N. effusus, and now cut off from the main population by N. effusus (and impoundments). It probably is not the result of human introduction since one specimen was collected in 1917, earlier than times of considerable trucking about of bait minnows. The use of minnows as bait in the recently created southwestern Ohio basin impoundments is now heavy.
The Cumberland-Kentucky drainage divide was very likely crossed through stream capture, with the direction of movement largely into the Kentucky. Kuehne and Bailey (1961) give good geological evidence for capture between these drainages and biological evidence for movement in the same direction.
The differentiation of N. micropogon observed in eastern Kentucky has zoogeographic implications. The largest differences in mean values of body circumferential scales are among adjacent drainage populations sampled (Table 12) in the Cumberland (grand mean 32.1), Kentucky (29.9), and Big Sandy (31.9). The differences between the Kentucky and the former and latter are, respectively, 2.2 and 2.0. The next greatest differences from Atlantic slope drainage populations are 1.3 and 1.2; all others are less, usually considerably less than 1.0. N. micropogon is thus relatively consistent in circumferential scale numbers over most of its range, except in eastern Kentucky. A distinct bimodality exists in the frequency distribution of the scale counts from the Kentucky drainage. We have not found a bimodal distribution of meristic characters from any other drainage population of Nocomis. Populational differences in other characters are present but less marked. Lateral line differences approach those of the circumferential scales. The Tennessee drainage population has the most reduced tuberculation and smallest head parts within the species. Too few large specimens are available from eastern Kentucky to determine relationships by these morphometric characters.
Additional differences became evident after segregating the eastern Kentucky circumferential scale counts by tributary systems or areas within the Cumberland and Kentucky drainages (Lachner and Jenkins, 1967, Table 3). The difference (1.3) between the means in scales between the middle and upper Cumberland populations, separated by the Falls, indicates their partial isolation; some individuals may still go over the Falls.
Within the Kentucky drainage, 86 specimens from the South and Middle Fork systems (the two more westerly of the three Forks of the upper Kentucky drainage) have a mean of 30.2. There are no significant differences or trends among the individual samples from the two systems. Thus the western upper Kentucky drainage mean of 30.2 and upper Cumberland mean of 32.7 are appreciably different and represent the greatest scale difference among adjacent populations of N. micropogon and within all other species of Nocomis. The upper Cumberland and Kentucky are in the same or similar physiographic areas; thus these differences probably are genetically based. The same was concluded by Kuehne and Bailey (1961:4) for a species of darter (Percidae) with two subspecies that have similar differences and distributions. It would be difficult to think, because of this difference, that the South and Middle Fork populations are derived from the upper Cumberland population through relatively recent stream capture, were it not for the fact that the populations of the North Fork and Red River systems, Kentucky drainage, also are apparently different from the South and Middle Fork populations. The mean circumferential scale values from the Red River (29.3, 20 specimens), a middrainage tributary, and the North Fork tributaries (29.0, 26 specimens) are lower than those from the South and Middle Forks, and also lower than in any other population of N. micropogon (Table 12). The strongly assymetrical frequency distributions (Lachner and Jenkins, 1967, table 3) for both the North Fork and Red River samples suggest that selection is operating strongly against individuals with scales larger than those found in these populations.
Several implications are derived from the Kentucky drainage data. It appears that a trend toward a large-scaled form began in the Kentucky, probably prior to the entrance into the upper Kentucky of a finer scaled population from the upper Cumberland. The stream capture described by Kuehne and Bailey (1961) occurred between the upper Cumberland and a tributary of South Fork of the Kentucky. Such a capture would interject the finer scaled Cumberland form within the range of a larger scaled differentiate of the Kentucky. Downstream movement within the South Fork and then through the main Kentucky River would lead a fine-scaled form first to the Middle Fork. Two aspects of the number of circumferential scales for the South Fork-Middle Fork sample are notable: (1) The mean is somewhat intermediate between those for the upper Cumberland drainage and the North Fork-Red River samples although much closer to the latter. (2) A distinct bimodality exists for the South Fork-Middle Fork sample. An intermediate mean may be expected with either occurrence or absence of interbreeding between a fine-scaled upper Cumberland form and a large-scaled North Fork-Red River form. The bimodality may indicate divergence of these forms to a level higher than that of a race. The Nocomis of the Kentucky, upper Cumberland, and adjacent drainages warrant further study, particularly to determine if possible differences in fin coloration (see p. 46) correlate with scale differences.
The question of dispersal from the Kentucky River of the large-scaled form cannot be adequately appraised at present since we have little data on adjacent populations. Drainages of nonglaciated Kentucky and West Virginia are believed to have served as refugia for N. micropogon during glacial times. Study of Ohio and Indiana populations may yield clues to the determination of which refugia held their ancestors.
The 15 specimens of N. micropogon from the Licking drainage, between the Kentucky and Big Sandy drainages, have a mean value of 30.7 circumferential scales, distributed as 28 (in 1 specimen), 30(3), 31(10), 32(1). Four specimens (CU 25595) from the Little Sandy River, northeastern Kentucky, have counts of 29(1 specimen), 30(1), 31 (2). The mean for 56 specimens from the Big Sandy drainage is 31.9. The means are similar or somewhat lower from southern Ohio River populations farther upriver from the Big Sandy, in West Virginia. These data indicate that the large-scale Kentucky drainage population may not have spread widely south of the Ohio River.
Why N. biguttatus is not at least moderately distributed in the southwestern Ohio basin is a perplexing question. N. biguttatus generally inhabits streams of lower gradient than does N. micropogon (Lachner and Jenkins, 1967:573). The latter apparently is absent from most of northern Kentucky, a region partly suitable for occupancy by N. biguttatus, and thus it would not be a competitor. One reason may be that the eastward extension of N. biguttatus is probably largely or solely related to postglacial drainage events north of the Ohio River and its habit of keeping within glaciated regions in the East. Its entire range north of the Ohio River is Recent and is almost entirely within or peripheral to regions of Illinoian glaciation; it is thus absent from the major portion of southern Ohio and Indiana (and Illinois). With its current absence from these areas, it is somewhat unlikely that it crossed the Ohio River even though it occurs in the northern Ohio basin. The single southeastern population, known in Elkhorn Creek, lower Kentucky drainage, in all probability stems from bait bucket escapees. The Elkhorn has been a popular smallmouth bass fishing stream for the last century (Towles and McClane, 1965:446), and chubs are favored bait.
The upper Savannah drainage population of N. micropogon was probably derived from the Little Tennessee River of Tennessee drainage by stream capture. Capture here is well documented by geological and additional biological evidence (Ramsey and Woolcott, ms).
The presence of N. micropogon in the Coosawattee system of the Mobile drainage may be attributed to stream capture with the Hiwassee system of the Tennessee. We have not further examined this or considered other possibilities.
N. raneyi: This species probably evolved in the Roanoke drainage from an ancestral stock of N. platyrhynchus after its transfer by stream capture from the New drainage. Ample opportunity for passage to the Roanoke seems to have been available. Wright (1934:61–72), Thompson (1939), and Dietrich (1959) argued that drainage loss by Little River and other adjacent New River tributaries was effected by headward erosion of the lower elevated upper Roanoke in the region of present North and South Forks of the Roanoke River. Wright (1934:61) believed the Roanoke captured an area of approximately 200 square miles formerly drained by the New since the close of the Harrisburg cycle (late Tertiary). Dietrich (1959:29–32) also supported a hypothesis that the headwaters of the Dan River, Roanoke drainage, have captured portions of Big Reed Island Creek of New drainage. N. raneyi is a member of the rich and distinctive Roanoke fauna composed of several endemic and semi-endemic forms, many of which have their heritage to the west.
With the Roanoke drainage as its center of dispersal, N. raneyi spread northward and southward. Ingress to the Chowan drainage may, in part, have been during the Pleistocene when the latter was tributary to the Greater Roanoke. Some events have certainly facilitated dispersal between these drainages since recent collecting has shown the Chowan fauna to be very similar to that of the Roanoke. A major problem lies in how N. raneyi and several other species, now common to the Roanoke, Tar, and Neuse, entered one of the latter two from the Roanoke.
Although specific geological evidence is lacking for stream captures between the Chowan, Roanoke, Tar, and Neuse drainages, apparently because the evidence would be obliterated on the easily eroded Piedmont and Coastal Plain cover, captures may be inferred to have occurred during late Pliocene to Recent times. Clark and Miller (1912:215) stated that coincident with the Pliocene submergence of the Coastal Plain there seems to have been a slight elevation and tilting of the Piedmont west of the shoreline, and that (op. cit.: 200) successive uplifts of the Coastal Plain occurred during the late Pleistocene period of shore terracing. Accompanying rejuvenation and subsequent headward erosion, with lack of structural control in these provinces, may well have resulted in captures. The lower Roanoke presently has no large tributaries entering from the north or south; these were probably lost by capture to the Chowan, Tar, and Neuse. Further, as the Coastal Plain emerged from inundations, swampy conditions must have once been more continuous than at present. Some swamps are not yet discontinuous, but are now becoming so (R. D. Ross, personal communication).
Biological evidence confirms these hypotheses. A large number of species is shared by the four drainages, some of which are small stream forms that are less likely to have utilized river-mouth connections. In addition to transfer with Piedmont captures, the occurrence of common upland species in Coastal Plain streams, such as N. raneyi, suggests the possibility that they may at times have gotten through some swamp connections to adjacent drainages. Members of the more typical Coastal Plain fauna probably attained part of their wide ranges in similar manner.
The entry of N. raneyi into the James drainage from the upper Roanoke is an interesting problem with several aspects, best treated after establishing some pertinent distributional facts. Its limited range in the James is roughly paralleled by that of three additional species of the upper Roanoke drainage that are not known from north of the James drainage or the New drainage. These are a sucker Moxostoma cervinum, a madtom catfish Noturus gilberti, and a darter Percina crassa roanoka. Of the three, M. cervinum appears to have spread farthest within the James, perhaps due to greater vagility. This species is known from the middle and lower Craig Creek system and only downriver, in lower Catawba and Jennings Creeks, and in lower Buffalo Creek of the lower Maury River system. The known range within the James of N. gilberti and P. crassa roanoka is entirely within the Craig Creek system. These three species are taken in about equal abundance in the Craig Creek system and the upper Roanoke drainage. These peculiar distribution patterns stand out after consideration of the entire James and Roanoke fauna. These two drainages share numerous other species, some of which range widely or are more or less restricted to the central Appalachians; all, however, range throughout most of the James and Roanoke drainages or generally occur where their preferred conditions exist. (The single exception is Pimephales notatus, a popular bait and farm pond culture species, widely distributed in the New drainage, whose small range within the upper Roanoke appears to have resulted from introductions.) In order to have achieved their wider distributions these species probably were exchanged at a date earlier than that of the first-mentioned four species and/or at more than one locality. In addition, isolation of some geminate pairs has been sufficiently long for differentiation to the specific or subspecific level (see C. R. Gilbert, 1961:455–456; 1964:107, for discussion of these cases).
The similar and limited ranges of the four species strongly suggest that they entered the Craig Creek area of the James from the upper Roanoke through a relatively recent connection. They do not appear to be restricted by lack of suitable stream conditions elsewhere in the James since their preferred habitat, although unoccupied, is readily accessible. No distinct ecological differences are apparent between the Craig Creek system and other upper and many middle James tributaries. Some evidence (p. 49) indicates that N. micropogon probably occupied the Craig Creek system prior to the entry of N. raneyi into the James. Competition with N. micropogon may be one factor limiting the further dispersal of N. raneyi within the James. As evidenced by the limited distributions of the three other species, however, which do not appear to be restricted by competition, their entrance to the James was probably of such recent date that sufficient time to achieve wide distributions has not elapsed.
Robins and Raney (1956:31) suggested that the populations of M. cervinum and N. gilberti in the Craig Creek system were derived from the Roanoke drainage through stream capture, but that Hybopsis “sp” (=N. raneyi) could have come from the Roanoke or New drainages. Present morphological knowledge of the chubs excludes the latter alternative.
Connections between the Roanoke and James drainages (Figure 20) were probably available. There appears to have been a complex of recent captures in the present valleys of North Fork of Roanoke River and Catawba Creek, involving these streams, a Craig Creek and a New drainage tributary (Wright, 1934:68). Trout Creek, a small upper Craig Creek tributary that heads against upper North Fork and flows through a gap in the Craig Creek-North Fork divide, probably captured a portion of North Fork. This seems to be the only possible route through which fishes could have entered Craig Creek directly from the Roanoke since they are separated elsewhere by relatively unbroken mountain ridges. Another possible capture is one between North Fork and Catawba Creek. Although fishes might cross the existing low divide as it is flooded during a period of excessive rain, the divide may not have shifted much at the expense of either stream (R. D. Ross, personal communication). A third alternative may be connection in Fincastle Valley between Catawba Creek and upper Tinker Creek, the major Roanoke River tributary at Roanoke. Evidence for this capture is given by Wright (1934:62–63).
The distributions of these four species have some significance, bearing on which of these possible routes may have been used. Although N. raneyi is not known from North Fork of Roanake (not taken in five collections containing N. leptocephalus), it may be in its lower portion. At least two of the three other species were taken from North Fork. All four probably occurred in lower Tinker Creek which now suffers from pollution. N. raneyi and M. cervinum occur in lower Catawba Creek, but the two other species are absent or rare therein since they were not taken in the eight known collections from lower and middle Catawba. Thus, although N. gilerti and P. crassa roanoka may have entered the James from either the North Fork or Tinker Creek, their apparent restriction to the Craig Creek system suggests that they entered via a North Fork-Trout Creek capture. This cannot be said for N. raneyi and M. cervinum since they range more widely. Further complexity results from the existence of the atypical population of N. raneyi in the main James, between the Craig and Catawba Creek populations and the Pedlar River population. Whatever factors have affected this population, they probably operated subsequent to the spread of the typical form from Craig or Catawba Creeks to Pedlar River. The bull chub and sucker apparently have not dispersed to James tributaries upriver from Craig Creek. The main channel and tributaries between Craig and (upstream to) Potts Creek, however, are poorly sampled or they have not been collected.
We have considered that perhaps the four species were introduced by man into the James from the Roanoke. This may be suspected owing to the limited ranges and only recent records of the species. Our records of N. raneyi in the James date from 1941, when G. W. Burton (Burton and Odum, 1945) collected a single specimen. The three other species were not reported from the Craig Creek system by Burton and Odum (1945) and our records of these date from 1951. Apparently the only fish collection made in this system before Burton’s was by Cope in 1867 (Cope, 1868). None of Cope’s material has been labeled as from Craig Creek, and none of it represents any of the four species. Very few specimens from Burton’s work are at the University of Michigan Museum of Zoology, probably since only the “necessary specimens [were] preserved” (Burton and Odum, 1945:183). Recent collecting in the Craig system above Newcastle yielded ten native species, in addition to those discussed above, that are relatively common but were not listed by Burton and Odum. Other species taken at only a few stations by Burton were found to be comparatively widespread. The lack of early records of the four species is probably related to insufficient collecting and loss of collections.
There is evidence that the four species were not introduced by man. Since N. gilberti is generally an uncommon species in the Roanoke there is slight chance of it being captured and introduced by fishermen. The chances are also slight that the common but relatively delicate darter would survive an extended period in a bait bucket. Few bait collectors really get into riffles, which are the preferred habitat of these two species. The hardy N. raneyi are more easily taken by bait collectors, but this species is generally uncommon in the upper Roanoke. The establishment of the Craig Creek and atypical populations of N. raneyi downstream almost certainly predates any recent introduction. The five spillway-type dams across the James River between Jennings Creek and Pedlar River have probably been in effect since the latter part of the last century. Their height, from 8–25 feet, would impede movement.
A final and important point is to reexamine the historical and problematical status of M. cervinum in the James drainage in light of recent data. The probable absence of M. cervinum in the James drainage and its early confusion with Moxostoma (Thoburnia) rhothoecum, widely distributed in the James, was discussed by Hubbs (1930:43–44), Raney and Lachner (1946b: 218), and Raney (1950:159). Soon afterward, Robins and Raney (1956:31) reported the capture of M. cervinum from Johns Creek at Newcastle and later it was found to range an appreciable distance downstream. The two suckers occur syntopically in the Craig Creek system. Cope (1868:236) described and gave the type locality of M. cervinum as the headwaters of the Roanoke and James Rivers, Virginia. From Cope’s paper it is clear that he collected in the Craig Creek system and he mentioned no other upper James tributary. M. rhothoecum was not recognized by Cope. Robins and Raney (1956:25–26) found that Cope’s James drainage specimens, designated by Fowler as cotypes of M. cervinum, were M. rhothoecum. It appears then, that Cope’s color plate clearly depicting the diagnostic coloration of M. cervinum, was based on specimens from the Roanoke drainage where he also collected, and where this species is common. The possibility remains, though, that Cope saw M. cervinum from the Craig Creek system but, as was sometimes the case, the specimens were not saved or were subsequently lost.
It should be noted that most of the more downstream records of Thoburnia rhothoeca listed by Burton and Odum (1945:186–187, tables 4 and 5) probably were, in part or entirely, M. cervinum since this species replaces the former in larger streams and is the more abundant of the two in the Newcastle area today. Jordan (1889a: 109), prior to the description of a Thoburnia, reported M. cervinum from Elk Creek, a short, direct tributary of the James River about eight river miles upstream from the mouth of Maury River, and from lower Buffalo Creek, apparently just upstream from the locality at which M. cervinum was recently taken. For several reasons, one being Jordan’s color description of his specimens, Raney and Lachner (1946b: 218) concluded that “they were largely Thoburnia” (M. rhothoecum). Hubbs (1930:43) found at the Museum of Comparative Zoology one of Jordan’s specimens of “cervinum” from Buffalo Creek to be M. rhothoecum and we located three of the same among the old Indiana University collection now at the Museum of Zoology, University of Michigan and six at the National Museum of Natural History. It still appears, however, that characters of both suckers were included in Jordan’s description and that at least some of his “older” specimens were M. cervinum. After leaving the James drainage, Jordan collected in the upper Roanoke where he may have taken M. cervinum and then compounded the colors of at least two species of suckers under his account of M. cervinum from the James. It is strongly indicated that M. cervinum and the other three species were in the James drainage long before fishermen with minnow buckets became common and transportation easy.
N. leptocephalus: The origin of the leptocephalus group is not clear, but after differentiation of the more northern form, N. l. leptocephalus, considerable dispersal occurred that included entry into several drainages. Its further advance northward along the Atlantic slope and westward may have been through some of the same connections utilized by N. raneyi. N. leptocephalus typically inhabits smaller streams than the members of the micropogon group. It would thus be suspected, more than the latter group, to cross drainage divides with greater frequency through small headwater captures. The dispersal of N. leptocephalus raises the possibility that there may have been many undetected or geologically unconfirmed captures in the central Appalachians. Once N. leptocephalus entered a drainage, a wide distribution within the drainage would probably have taken longer than for a large stream inhabitant.
We have no data that indicate whether N. leptocephalus occupied the Neuse, Tar, Roanoke, and Chowan before N. raneyi. Its entry into the James probably predates that of N. raneyi because it is widespread in the James and it occurs farther northward. The bluehead chub may have entered the James drainage by captures along the broad front of adjacency of the Roanoke and Chowan to the James. The Greater Susquehanna River is invoked to account for the presence of N. leptocephalus in the York drainage.
The occurrence of N. leptocephalus within single stream systems of both the Rappahannock and Potomac drainages suggests it got into these by stream captures from the adjacent James drainage. There is no evidence for or against the possibility that these may be due to introductions. A third alternative, through the Greater Susquehanna, is quite unlikely, particularly in the Potomac case, since the streams where N. leptocephalus is known in the two drainages are headwaters, appreciable distances from the drainage mouths. Inspection of topographic maps of the upper Piedmont in the region of tributaries of the Rapidan River, Rappahannock drainage, and those of the Rivanna River of the James, show that headwaters of these streams are opposed to each other although no definite elbow of capture is evident. Clear geological evidence is given by Thompson (1939:1350–1351) for a capture of the Potomac by a Rivanna River tributary in the Blue Ridge near the Albemarle-Augusta County line. If N. leptocephalus used this connection it would have entered South River of the Potomac at a point approximately midway between the two nearby Potomac streams in which it was taken.
Entry of N. leptocephalus into the New drainage was probably through one or more of the recent captures by the Roanoke. This species is known only from Virginia tributaries of the New. Three additional species, Chrosomus oreas, Notropis albeolus, and Notropis cerasinus, are also known in the New almost exclusively from the Virginia portion and may also be recent arrivals. The most upstream tributaries where these four species were taken are Fox and Wilson Creeks, Grayson County, which enter the New just downstream from the Virginia-North Carolina state line. Geological barriers to further dispersal up the New are not known to have existed in the past.
Downstream in Virginia near the Virginia-West Virginia line, C. oreas is known from Big Stony Creek, Giles County (Ross and Perkins, 1959:27). Ross and Perkins (1959:27) and C. R. Gilbert (1964:139, map 3) gave records of N. cerasinus from Big Stony. The range of N. albeolus extends farthest downstream of the four species. C. R. Gilbert (1964:156) reported a collection from the lower Greenbrier River and believed its presence there was due to very recent dispersal or introduction, as it was not collected earlier from this stream and in streams somewhat farther upriver in West Virginia and Virginia just across the state line. The Bluestone Dam, completed in 1949, on the New River one mile above the mouth of the Greenbrier, and its impounded waters extending up to the state line, obstructed recent dispersal. The presence of N. albeolus in the Greenbrier may be related to earlier dispersal or introduction.
Evolution of Nocomis
Our interpretation of the phylogeny of the species of Nocomis is based on morphology, coloration, zoogeography, and life history. The most primitive Nocomis is probably the biguttatus group (Lachner and Jenkins, 1967). The tubercles in the species of this group are rather evenly spaced dorsally over the head. They are absent on the snout. The tubercles are small to moderate in size (but large compared to almost all American cyprinids). The general pattern appears to be the basic one that has become modified in various manners within the micropogon and leptocephalus groups. Generally, the cyprinids with unspecialized breeding habits tend to have small tubercles distributed over the head in no outstanding pattern, although some differences in the patterns usually exist among the species. Development of particular tubercle patterns and larger tubercle size correlate in many of the North American cyprinids known to have specialized reproductive behaviors, such as nest building, special mode of spawning and highly developed territorialities. Examples of these specialized forms, in addition to Nocomis, include Campostoma, Semotilus, Pimephales, and certain Notropis. The two related, nest-building species of Exoglossum are notable exceptions.
The tubercles are slightly to considerably larger in all members of the micropogon and leptocephalus groups compared with those of the biguttatus group. A probable specialization of tubercle distribution, shown only by the micropogon group, is their development on the anterior snout and lachrymal areas. A trend toward this condition is actually seen in the biguttatus group since the dorsal tubercles are developed slightly anterior to the anterior internasal line and, in adults, there are 1–3 tubercles located in the subnasal area but not over the lachrymal bone.
Large body size would be of selective advantage for nest construction and defense. N. biguttatus is, over its range, the smallest species and it may follow that the larger size of all of the other species is the derived condition.
Another important phylogenetic character is the pharyngeal dentition. The general trend in North American cyprinids is probably toward loss of teeth in the major and minor rows and a subsequent loss of the minor row. The dentition of N. biguttatus and the Redspot chub, an undescribed species (part 3) in the Arkansas and Red River drainages, is almost always 1,4—4,1, sometimes 1,4–4,0; loss of the minor tooth row would result in the dental formula of an early stock of the micropogon and leptocephalus groups. N. effusus, considered an advanced form in the biguttatus group (Lachner and Jenkins, 1967) also lost the minor tooth row.
Linear (rowed) arrangement of head tubercles does not appear to be a primitive condition in Nocomis. The head tubercles in the biguttatus group do not align in definite rows when the adult complement is approached or fully developed. Occasionally several tubercles form a longitudinal line over a portion of the head but this appears to be the result of chance. We do not know how Branson (1962:536) construed that most head tubercles occur in rows on the nuptial male of N. biguttatus. The pattern of early tubercle development in juveniles of the biguttatus group also does not indicate a rowed condition since the first several tubercle spots that appear in the many juveniles examined are widely scattered. Only occasionally in small juveniles of N. platyrhynchus and N. raneyi do some of the tubercle spots tend to align into longitudinal rows through the internasal and anterior interorbital areas; in such cases several randomly located tubercle spots are also present. With growth in all of these species, rows are not evident.
In apparently advanced species of Nocomis in which a moderate to low number of tubercles occur, there is a trend for individual tubercles to develop consistently in specific locations and, in some cases, they appear to be arranged linearly. In N. micropogon the tubercles in the anterior middorsal region often develop in four rows, two curving posterolaterad on each side from the anterior internasal area to the anterior edge of the orbits. This trend is most evident in the Tennessee drainage population of N. micropogon which averages the lowest total tubercle numbers within the species. The first tubercle spots that develop in juveniles of this species are included in the internasal rows.
A similar situation but more extreme than in N. micropogon occurs in the leptocephalus group. These forms have the largest tubercles (the largest of North American cyprinids) and the lowest number in Nocomis, and the tubercles generally develop more constantly in specific areas of the head than in N. micropogon. This is particularly shown by the southern form, N. leptocephalus bellicus, adult males of which almost always have only six tubercles and which develop in nearly the same places on all specimens. The posterior internasal and anterior interorbital tubercles of the leptocephalus group are those that particularly develop in the same location of the head. A row typically curves just above the dorsal half of the orbits.
In the micropogon and leptocephalus groups the tubercles are always largest where tubercle spots first develop, in the internasal and interorbital areas. The tubercles in the biguttatus group are largest in the occipital areas; spots first develop in the internasal and interorbital areas, but before many appear in these areas they also develop in the occipital region.
Developmental patterns parallel to those in Nocomis occur in other species of North American cyprinids. The large majority of those species with high tubercle numbers on the head dorsally do not have rowed tubercles. When rows are present, they often occur over rather narrow bones and other supporting structures. An example is the distinct tubercle rows over the mandible in species of the subgenus Pteronotropis, genus Notropis (Bailey and Suttkus, 1952, pi. 2). The rows over the lachrymal bone in all species of the micropogon group is another example. When present on the fins, tubercles occur on the rays. Tubercles often tend to have a linear development near edges or folds of skin such as around the orbits and above the lips. When present on scales, the tubercles are usually at the free, posterior margin, forming a row. All species of Pimephales have a low number of tubercles and they are large in size, comparable to the leptocephalus group. Also in Pimephales, the tubercles develop in rather definite locations and form rows (Hubbs and Black, 1947:13–17). From developmental patterns of tubercles, other morphological aspects and zoogeography, Gibbs (1957:187, 189–192) postulated that the occurrence in rows of head tubercles is primitive within the subgenus Cyprinella, genus Notropis, and that the scattered tubercle condition is derived. Results of a detailed study of tubercle development (Koehn, 1965) in one of the presumably derived species of Cyprinella indicate that its early tubercle pattern is similar to the adult pattern of primitive species. Examination of many nuptial males of Cyprinella indicates that dorsal head tubercles tend to be more linearly arranged in the species with fewer and larger tubercles, which may also be the primitive condition within Cyprinella.
Synthesis of these comments on tubercle development is difficult since there appear to be several factors involved. Limited area for basal attachment would be a simple reason for linear development of tubercles over narrow supporting structures. Tubercle rows about the orbit and other structures could be a protective mechanism during agonistic behavior and nest-building activities. Tubercles developed along the edge of scales would come into more use than if they were more basal, near the imbricated scale base. Genetic fixity for confinement of few tubercles to discrete areas may be associated with specific behavior patterns. Tubercles that develop in rows, such as on the snout of Pimephales and the dorsal part of the head of species of Cyprinella, may serve more efficiently in special breeding behaviors.
The biguttatus group is most closely related to the micropogon group; neither is particularly closely related to the leptocephalus group. The nuptial male coloration, consisting in the biguttatus and micropogon groups of pink and reddish over the lower head and body, is quite unlike the blues and oranges in the leptocephalus group. The whorled intestine of the latter group, readily derived from the simple S-shape intestine of the two other groups, is also unique within Nocomis.
The nuptial crest appears to be a functional and adaptive character. It is prominently developed in the micropogon and leptocephalus groups. Some nuptial male specimens of N. effusus have a very slight head swelling which has not been observed in N. biguttatus; this may be a lesser development of the crest. Possibly the crest arose independently within the micropogon and leptocephalus groups. The fact that the crest is alike in development and general form in these two groups, and that it is unknown in other cyprinid fishes, suggests common ancestry among the two groups and that it was not derived independently.
Crest development, lateral curving of the tips of tubercles and loss of the minor tooth rows relate early stocks of the micropogon and leptocephalus groups. The leptocephalus group would be regarded as the most divergent from a common biguttatus stock.
The crest is perhaps an adaptation for lessening the effects of head impact from certain agonistic behavior. The moderate enlargement of tubercles in two of the three species of the micropogon group and the considerable enlargement of tubercles in the leptocephalus group may correlate with the development of large crests in these groups. A large, hard (cornified) tubercle pressing into the cephalic region upon impact would give more severe shock than smaller tubercles. N. l. bellicus and an undescribed subspecies (part 2) are apparently derived from N. l. leptocephalus by loss of the occipital and medial interorbital tubercles and show a trend toward reduction of the crest posteriorly. However, N. micropogon lacks tubercles in the interorbital and occipital areas where the crest is often well formed.
The direction in which the tips of tubercles point is different in the groups. The head tubercles of most cyprinid species have erect or antrorse tips. The tubercles or their tips in N. biguttatus are mainly antrorse, but a few are more or less erect. The tips of most tubercles in the other two groups of Nocomis are erect or curved laterally.
The development of a hiatus (absence of tubercles) between the anterior snout and internasal tubercles occurs in the micropogon group. The absence of tubercles may have a functional basis, for the development of tubercles, especially large ones, would inhibit the protractility of the upper jaw. A complete absence, or but a few tubercles, occurs between the anterior snout and internasal areas in N. micropogon; tubercles are usually present in this area in N. platyrhynchus and are always present in N. raneyi. The absence of tubercles at the hiatus is correlated with tubercle size in these three species, N. micropogon developing the largest tubercles and N. raneyi the smallest. Within the subgenus Cyprinella, 14 of 16 included species
Summary
The chub genus Nocomis Girard of North America has been characterized and relationships among three species groups in Nocomis were discussed. Two new species were described from the central Appalachian region, both in the micropogon species group. Characters separating Nocomis from other barbeled cyprinids are that the nuptial males have large breeding tubercles on the head that show a phylogeny among the species in their number, distribution, and size; have expansive crests or swellings on the head in two species groups that develop co-incidentally with the spawning period; develop elaborate breeding coloration; construct large moundnests of gravel which they transport in their mouths. The genus has large scales, the body circumferential scales are fewer than 37; the pharyngeal teeth number 1,4–4,1 to 3–3; the scale radii are present only in the posterior field and the total number of radii in the adults ranges from 23 to 54; the total number of vertebrae ranges from 38 to 43; and a single, terminal maxillary barbel.
The three species groups were defined and their characters summarized, namely, the biguttatus group with three species, the micropogon group with three species, and the leptocephalus group with three subspecies. The central Appalachian region is inhabited by four species, N. micropogon (river chub), N. platyrhynchus, new species (bigmouth chub), and N. raneyi, new species (bull chub), of the micropogon group, all of which are sympatric with N. leptocephalus (bluehead chub) in one or more river drainages. A discussion of the nomenclatural history of the genus and nominal species was given and important diagnostic characters Useful in differentiating among the specific and several infraspecific populations were discussed, evaluated, and summarized. The critical specific and subspecific characters, such as cephalic tuberculation involving the distribution, number, size and developmental patterns of tubercles, cephalic crests or swellings and nuptial body coloration are sexually dimorphic, developing elaborately in the males. Many morphometric characters, as well as tubercle numbers and distribution, show an allometric relationship with body size. An account of each species has included a synonymy, diagnosis, description and comparison of meristic and morphometric characters, coloration in life and in preservation, populational differentiation, reproduction, growth and size attained, materials studied and geographic distribution. The species are illustrated, special body features are drawn, distributions are plotted, and important character data are shown in text figures.
N. micropogon and N. leptocephalus are widely distributed, the former occurring over most of northeastern United States and the latter is found chiefly on the Atlantic and Gulf slopes. In the study area, N. leptocephalus is found east of the Atlantic drainage divide north to the Potomac River and west of the divide in the upper New River (New-Kanawha River system). N. micropogon occurs east of the Appalachian divide from the James drainage north ward and throughout the western slope including the Kanawha River system (below the Falls). Both of these species are common and are widely distributed in the James River drainage. N. raneyi is an Atlantic slope form, occurring in the James, Chowan, Roanoke, Tar and Neuse drainages, where it is found commonly and is widely sympatric with N. leptocephalus. These two species and N. micropogon occur together in limited areas of the James drainage. N. platyrhynchus is found throughout the New River system, where part of its range overlaps that of N. leptocephalus.
The ecological requirements of the chubs were reviewed and species preferences discussed. The primary ecological distinction among the species is stream size. Regional ecology was reviewed in respect to occurrence and abundance of the species. Generally, the species of Nocomis prefer clear, moderate to warm water streams of intermediate gradients. They are chiefly carnivorous, common over stream beds of gravel, rubble, and boulders. Well-established populations of three species in typical Piedmont streams indicate an adaptiveness to this shifting, sandy habitat. All species of chubs are rare below the Fall Line in the upper Coastal Plain province.
Associations, interrelationships and frequency of hybridization were compared among the four species. The three species of the micropogon group were regarded as ecological homologues. N. leptocephalus almost always coexists with N. raneyi in moderate-sized streams, where they occur sympatrically, and it occurs less frequently with N. micropogon and N. platyrhynchus. Differences are observed among the species of the micropogon group in their frequency of hybridization with N. leptocephalus. Hybridization is somewhat inversely proportional to the frequency of syntopic occurrence. Only two natural hybrids N. raneyi × N. leptocephalus were found, whereas N. micropogon × N. leptocephalus hybrids are comparatively common (87 specimens known). A population of N. raneyi in a limited area of the James River is considered atypical. The meristic and morphometrical data of this atypical population were compared with typical N. raneyi and N. micropogon from the James drainage. Introgressive hybridization may have occurred between N. micropogon and N. platyrhynchus because it best explains the similarity of populations of the former in the Monongahela and Potomac River drainages with the latter species in the New River system.
The biological and geological evidence providing an explanation of the dispersal of the chubs and other species of fishes in various interdrainage exchanges in the central Appalachian region was comprehensively reviewed. Possible routes of entry into the several drainages were proposed. Present, detailed distributional patterns of the chubs and other species were related with geologic events, and present and past ecological conditions, in an effort to understand the conditions, barriers, and routes affecting dispersal. Three kinds of geological events that operated in the dispersal of chubs are considered: stream capture, eustatic changes of the Atlantic Coastal Plain, and Pleistocene drainage modifications.
The interpretation of evolution in Nocomis is based on morphology, coloration, zoogeography, and life history. The unknown precursor of Nocomis was probably not an elaborate nest builder and probably had fewer sexually dimorphic features. The trend for selection to act more on sexually dimorphic features is characteristic of Nocomis, probably related mainly to the functional aspects of nest construction, nest guarding and territorial maintenance by the male. Nocomis shows an elaboration or adaptive reduction of important diagnostic characters as the extent of tuberculation, development of nuptial crests, and development of particular reproductive colorations. The most primitive Nocomis are probably of the biguttatus group, based on, among other characters, the general pattern, small size and relatively large numbers of cephalic tubercles. This general tubercle pattern appears to be the basic one that has become modified in various manners within the micropogon and leptocephalus groups. The primitive stock of the micropogon group, best exemplified by N. platyrhynchus, shows relationships with biguttatus stock in the widely distributed cephalic tuberculation, small tubercles, fine squamation, and stout pharyngeal arch. Evolution within the micropogon group involved, mainly: the reduction in tubercle numbers; reduction in cephalic tubercle distribution; increase in tubercle size; development of the nuptial crest; and in the differentiation of the pharyngeal arch, size of scales, and certain head and body features. The leptocephalus group, the most advanced, shows a greater reduction in tubercle numbers compared with the micropogon group, a reduction in cephalic tubercle distribution, great increase in tubercle size, development of specific reproductive coloration, stouter body and coarser scales. The forms of the leptocephalus group evolved further in the greater reduction of cephalic tubercle numbers and distribution, and in the differentiation of special reproductive body colorations.
- citação bibliográfica
- Lachner, Ernest A. and Jenkins, Robert E. 1971. "Systematics, distribution, and evolution of the chub genus Nocomis Girard (Pisces, Cyprinidae) of eastern United States, with descriptions of new species." Smithsonian Contributions to Zoology. 1-97. https://doi.org/10.5479/si.00810282.85
Comprehensive Description
(
Inglês
)
fornecido por Smithsonian Contributions to Zoology
Nocomis leptocephalus intergrade interocularis × leptocephalus
SPECIMENS STUDIED.—A total of 155 specimens from the Edisto drainage, as shown in the tabulation on page 2, were studied. For a listing by localities, see Appendix, page 35.
Comparison. The development and occurrence of head tubercles of the Edisto River population is intermediate between N. l. interocularis and N. l. leptocephalus. The four adult specimens (Table 7) had 16 tubercles, a value higher than the mean (8) and maximum number (14) for N. l. interocularis but lower than the mean (17 in northern race, 18 in Santee, 23 in Pee Dee) and much lower than the maximum number (33) in N. l. leptocephalus. Further, intermediacy in tubercle numbers of the Edisto population compared with the above two subspecies is shown in a summary of all material that could be counted (Table 6), and in specimens over 65 mm SL, excluding the supraorbital tubercles, in Table 8. The intermediacy of the occurrence of tubercles posteriorly on the head in juveniles and adults (specimens over 60 mm SL) of the Edisto population is compared with the above two subspecies in Table 9. The mode for the Edisto population occurs on the AOC line, a fairly intermediate position between the AIO area, usual for N. l. interocularis, and the MOC-POC area, usual for N. l. leptocephalus.
The vertebral numbers of 39 (17 specimens), 40 (10), and 41 (1) of the Edisto population are additional intermediate values between the adjacent, southern drainage populations of N. l. leptocephalus (about 39 vertebrae) and N. l. interocularis (about 40 vertebrae).
The adult and subadult specimens from the Edisto drainage are insufficient to compare this population more precisely with the adjacent populations. The above comparisons of three characters show intermediate values for the Edisto population that are not found among the three drainages (Savannah, Altamaha, and Chattahoochee) in which N. l. interocularis occurs nor in drainages from the Santee northward inhabited by N. l. leptocephalus. We presume that the Edisto population has received N. leptocephalus stocks from both north and south, after these two stocks had diverged into N. l. leptocephalus and N. l. interocularis. Thus, the last entry of a N. leptocephalus stock into the Edisto was probably comparatively recent.
Coloration
The general coloration of N. l. leptocephalus is discussed and compared with the other species of Nocomis inhabiting the central Appalachian area by Lachner and Jenkins (1971). Also described are two different primary color forms of N. l. leptocephalus among the living nuptial males. A form with a brassy orange-olive coloration on the body laterally and sometimes forming a stripe occurs in the northern race from the Cape Fear drainage northward. Another form with a bluish body occurs in the Pee Dee-Santee drainages.
The blue form also occurs in N. l. interocularis of the Savannah River drainage. Following is a description of a prenuptial male, 187 mm SL, collected on 15 May 1966 in Mill Creek, a tributary of Lake Toxaway, Savannah River drainage (USNM 200757). This coloration is characteristic of the Savannah drainage population. The specimen was captured with three females from beneath a cut bank near a nest. Its nuchal crest was not fully developed and the head tubercles were normally distributed, but were not enlarged as in some males. The nuchal crest was brown, the rest of the head slate-blue; eye, yellow-brown; opercular membrane, dark gray; dorsum of the body, brown with a few golden highlights on the postdorsal area; posterior margins of the scales dorsally on the sides above the lateral line, black; lateral band, diffuse, a greenish gray or olivaceous color; belly, bluish white; dorsal fin, slate-blue base with dark gray rays, the posterior having a yellow tinge, and the fin membrane, smoky; pectoral fins, smoky yellow, the rays grayish with the first ray darker; pelvics and anal fins, yellow with a gray cast; caudal fin, a burnt orange, the median rays darkest. A large gravid female in the same collection had the nuchal area and dorsum, brown, with a faint, light, predorsal stripe; lateral band, black with a narrow, diffuse, light brown stripe above and extended anteriorly and prominently onto the post-ocular area with a faint bar continuing to below the nostrils; opercular membrane, dark gray; body below the lateral band, yellowish; belly, white; dorsal and caudal fins, rust-orange; the other fins, a brownish yellow.
The following coloration was recorded from a living tuberculate, prenuptial male, 144 mm SL, captured in Amy’s Creek, Habersham County, Georgia, Chattahoochee River drainage, on 2 April 1970 by Lachner and Wiley: dorsal fin, light pinkish orange, melanophores along base of dorsal fin; caudal fin, olive-orange, lighter near the base with melanophores along basal portion of rays; pectoral fin, deep pink-orange; pelvic fin, deep pink-orange with some white on first ray; anal fin, pink-orange, the outer edge lighter; iris, red; body above lateral line grayish with slight bronze; ventral part of head, light reddish; dorsal part of head, slightly tan to pale olive; belly white to milky. The head tubercles were moderately developed, but the head was only slightly swollen. A gravid female, 119 mm SL, captured with the above male, had the same general color pattern in life except that the fins were very light, showing only traces of color, and the body was pale.
Other color data on adult N. l. interocularis are from a color slide of a male, 164 mm SL, collected by E. C. Raney in the Chattahoochee River drainage, Georgia. This tuberculate, prenuptial male had light orange fins, the rays of the dorsal and caudal having a darker burnt orange color. Good comparative color data of nuptial males from the Altamaha drainage are not available.
Observations by the authors and others indicate that the breeding coloration of N. l. bellicus in the Alabama River system of the Mobile Bay drainage, Alabama, is essentially the same as the northern color form of N. l. leptocephalus, e.g., with the blue head and the lateral orange stripe anteriorly on the body changing to yellow caudally. A color description of a large, prenuptial male N. l. bellicus with large tubercles, was recorded from a fresh specimen captured on 16 March 1957 in the Alabama River system, Elmore County, Alabama, by R. D. Suttkus (RDS 2592). The sides of the head were bluish gray; head, dorsally, and entire nape posteriorly to origin of dorsal fin and down on the sides to lateral line, very brassy orange; rest of back and upper sides, bright olive; iris, bright orange-red; caudal fin, olive; leading edge of anal and pelvics, milky; remainder of anal, pelvics and pectorals, olive; dark bar on cleithrum immediately behind opercle. Smaller specimens in same collection have bright orange caudal fin, orangy dorsal fin, and bright orange iris.
A color description of a tuberculate postnuptial male, captured over a nest on 1 July 1965 by R. D. Suttkus (RDS 3717) in the Alabama River system, Clarke County, was recorded in the field as follows: head below lower margin of tubercles, below eye, and below top of opercular opening, pale blue; top of head, golden brown; iris, bright red-orange; back and upper sides, brown with golden olive overlay; center of scales, golden, especially anteriorly on body and more yellow posteriorly on body; predorsal stripe, golden and broad, one scale row in width; postdorsal stripe tapering posteriorly and more yellowish than predorsal stripe; lower sides of body and belly, pale; pectoral fin, yellowish, the anterior one fourth bright yellow, the color pronounced on anterior rays; pelvic fins, yellowish, more so on anterior rays; dorsal fin, mostly olive, the base bluish, and some yellow on leading edge; caudal fin mostly yellowish olive; and anal fin, yellowish anteriorly, with branching areas of rays also yellowish as in pelvic fins.
A subadult N. l. bellicus, 107 mm SL, and a young specimen, 36 mm SL, taken by H. T. Boschung in a tributary of the Cahaba River system, Bibb County, 22 August 1967 (UAIC 2629–1), preserved in 10 percent formalin and ional solution, still retained lifelike colors in January 1968 (H. T. Boschung, personal communication). The larger specimen was colored as follows: body dorsally, dark olive above lateral line, and light olive from lateral line to about two scale rows below; lower body, light to pale; head, dark olive dorsally, light olive laterally, and light blue ventrally; body with no evident lateral band; dorsal fin, bright orange with some olive near base; caudal fin, bright orange; pectoral fins, light olive; pelvic fins, very light olive, with light posterior margins; and anal fin, light olive, with touches of orange along middle portion of rays, and the outer margin, light. The young specimen had the body light brown to dusky above the pronounced dark lateral band and light to dusky below the lateral band; dorsal fin, light orange, the outer margin light to slightly dusky; caudal fin, medium orange, the outer rays light to slightly dusky; and pectorals, pelvics, and anal fins, light to only slightly dusky.
Throughout its range, N. leptocephalus is commonly observed to have variable amounts of orange or reddish coloration in the dorsal and caudal fins (and, frequently, in the other fins) in young, juvenile, subadult male and female, and mature female stages.
There is great difficulty in evaluating color differences (aside from those associated with individual variations and populational differences), because of the gradual color changes related to growth and maturity. Also, among the nuptial males, the changes occur throughout the reproductive season and “peak” colors develop immediately at the time of spawning. The characteristic coloration of the nuptial males compared to that of non-nuptial males and all females is discussed by Lachner and Jenkins (1971). The elaborately developed nuptial coloration common to Nocomis biguttatus, N. effusus, N. micropogon, N. raneyi, and N. platyrhynchus is found in N. leptocephalus, except that the problem of interpreting the systematic significance of the coloration is compounded in the latter species by the presence of two basically different nuptial color patterns. The blue body form of N. l. leptocephalus of the Pee Dee and Santee river drainages is trenchantly different from the brassy reddish colored northern race of N. l. leptocephalus (in all drainages northward from the Cape Fear). These two color forms within N. l. leptocephalus correlate with the tubercle differences noted above, i.e., the northern race has fewer tubercles and is brassy reddish while the southern race has more tubercles and is blue. A similar blue color form appears to be typical for N. l. interocularis. We are not certain if the nuptial males of the Altamaha drainage develop the blue coloration typical of N. l. interocularis of the Savannah drainage. The orange-olive color form reappears in the Alabama drainage and westward in N. l. bellicus. In this subspecies, the nuptial males appear to have brighter orange-red colors in the fins than the northern race of N. l. leptocephalus. Most significant is the fact that great differentiation in nuptial coloration has occurred within N. l. leptocephalus, and probably N. l. interocularis and N. l. bellicus are also completely different in nuptial coloration.
Geographic Distribution and Ecology
The extensive distribution of N. leptocephalus along the Atlantic and Gulf Coast drainages is shown in Figure 4. Collections of the three subspecies and the Edisto intergrades are individually plotted and their distributions are shown in respect to the fall line and the Appalachian Divide. The distribution of N. l. leptocephalus is discussed by Lachner and Jenkins (1971) in detail where the northern race occurs sympatrically with three other species of Nocomis in certain river drainages. N. l. leptocephalus is widely distributed in the James drainage, and in limited headwater localities of the Rappahannock and Potomac drainages, where it occurs sympatrically with N. micropogon. Across the divide in the middle and upper New River system, its distribution overlaps considerably with N. platyrhynchus. In portions of the James, Chowan, Roanoke, Tar, and Neuse drainages, it is more or less commonly associated with N. raneyi.
Figure 4.— The distribution of N. leptocephalus. The three subspecies are plotted with separate symbols. The intergrades, N. l. interocularis × N. l. leptocephalus, in the Edisto drainage are indicated by a single star, but several collections are actually represented by this plot. When two or more collections were made at (or near) a locality, only one was plotted. The single plot in northern Alabama of N. l. bellicus in the Tennessee drainage also represents several collections. The two squares representing N. leptocephalus in southeast Tennessee, Tennessee drainage, are based on small specimens and could not be identified to subspecies with assurance. Data for plots are listed in the text, and those that merit special attention are discussed. The Appalachian Divide and the fall line are indicated.
N. l. interocularis is widely distributed in the Savannah and Altamaha drainages, occurring far below the fall line. All of our records occur above the fall line in the Chattahoochee system. N. micropogon is associated with N. l. interocularisin the upper Savannah drainage, where the former was probably derived by stream capture from the Little Tennessee River, Tennessee drainage. The presence of N. leptocephalus in a restricted area of the Tennessee drainage of North Carolina (probably N. l interocularis) may be associated with stream capture with the Savannah drainage (see discussion by Lachner and Jenkins, 1971). We have examined two collections from this area, totaling eight specimens, all of which were young or juveniles (TU 29663, six specimens, Cedar Creek, tributary to Lake Thorpe, 4.4 mi north of Cashiers, Jackson County; TU 28001, two specimens. Little Tennessee River, Cullasaja River, above Cullasaja Falls, 4.3 mi NW of Highlands, Macon County). Here, N. leptocephalus is again sympatric with the widely distributed N. micropogon of the Tennessee drainage.
The one plot of N. l. interocularis in the upper Coosa River system is based on a typical male specimen, 118 mm SL, captured by R. D. Suttkus in Cartecay River, 7.8 mi SE of Ellijay, Gilmer County, Georgia, on 18 June 1965 (TU 38213). William F. Smith-Vaniz (personal communication) informs us that there is considerable bait fishing in the impounded waters of the area. Populations of N. l. interocularis in the nearby Chattahoochee drainage are a ready source of bait. Several species of fishes, including Nocomis (Figure 4), are apparently prevented from entry into the Coosa River system by the falls at the Fall Line (Smith-Vaniz, 1968:124, 130).
The southwestern subspecies, N. l. bellicus, finding suitable habitat below the fall line, occurs from the Alabama River system westward to certain tributaries flowing into the Mississippi River from the east. It also is common in a restricted area of the Tennessee drainage, where B. R. Wall, Jr., (personal communication) has obtained collections in more than twenty localities of the Bear Creek system (indicated in Figure 4 by one plot mark). Smith-Vaniz (1968:124) cites headwater stream capture in the area involving the Tennessee and the Tombigbee rivers systems and lists four species other than N. l. bellicus that also made entry into the Tennessee River from the Tombigbee River system. The presence of N, micropogon as well as many other species, in the Coosawatee system, Coosa-Alabama drainage, is probably the result of a stream capture with the Hiwassee system of the Tennessee drainage.
The ecology of N. leptocephalus is reviewed by Lachner and Jenkins (1971). Over its large range, the bluehead chub is one of the most abundant species, preferring small stream habitats of about 10 to 50 feet in width and avoiding the larger streams and rivers as well as the extreme headwaters. Clear streams of gravel-rubble bottom composition, particularly current-swept areas, are preferred and necessary for these gravel-mound nest-building chubs (Plate 8). Slack waters are avoided, although the species thrives and is abundant in typical Piedmont streams where lower gradients, sandy bottoms, shifting sands, and some turbidity are characteristic of the habitat, but in these areas some gravel-rubble riffles also occur. The rarity of the bluehead chub along the middle-Atlantic Coastal Plain is probably related to the general absence of riffles or exposed gravel and rubble; however, these ecological conditions also prevail below the fall line of the southern Gulf Coastal area. The widespread distribution of the chub is undoubtedly related to its ability to survive Piedmont and certain Coastal Plain conditions. It probably was able to enter new drainages by way of the mouths of adjacent streams during intermittent periods of sea-level recessions, as well as to enter new drainages by stream capture, because of its ability to survive in relatively small, headwater streams.
Evolution and Dispersal
Lachner and Jenkins (1971) present a detailed account of routes of dispersal in the central Appalachian area.
The enlargement and loss of tubercles probably represents an advanced condition in Nocomis. The adaptive significance of large, stout, nuptial tubercles is not known. They would serve as effective weapons in agonistic behavior but nuptial males of N. leptocephalus, unlike some other species of Nocomis that have smaller, more numerous tubercles on the head, have not been observed to engage in either prolonged or violent fighting or both during the nest-building activities of reproduction. It may be that the large, easily visible tubercles serve as recognition signals to the female (or they may have some other behavioral significance).
The evolution of the leptocephalus complex is not clear with respect to existing data on morphology and distribution. This species complex, although distinct from the micropogon and biguttatus groups, is most closely related to a N. micropogon stock; however, its evolution and dispersal cannot be related to that of the micropogon group. The basic N. leptocephalus stock became isolated on the present Atlantic-Gulf slope probably during the Teays Period. It appears most likely that a N. leptocephalus stock existed, at first, in the upper Santee-Pee Dee drainage area. This stock could have evolved from a N. platyrhynchus type from the Old Teays River (now the upper New River). The high numbers of tubercles, their crowding in the forepart of the head before the eyes, and their occurrence from the internasal to the occipital area are characters of N. l. leptocephalus of the Santee-Pee Dee that are also common to N. platyrhynchus. This southern N. l. leptocephalus stock found dispersal routes to the northeast on the Atlantic slope and westward on the gulf slope at different times and rates via stream capture and by the confluence of fresh waters at the mouths of rivers during Pleistocene sea-level fluctuations.
The westward extension of the N. leptocephalus stock ended abruptly at the Mississippi River. It is not known to occur farther westward. Following the dispersal and establishment of N. leptocephalus stock on the gulf slope, a considerable period of isolation of the Savannah-Chattahoochee stock from that of the Alabama and more western rivers must have been necessary in order for the divergence of N. l. bellicus from N. l. interocularis to have reached the present level of differentiation. The distribution of N. l. bellicus could have been variously effected during geological times, but its morphological homogeneity suggests considerable gene exchange among the several drainages it now occupies or a recent entrance into the western areas via the upper Mobile drainage.
The old escarpment of the Coosa at about the fall line prevented its entrance into this system. Its limited distribution in the Tennessee River in northern Alabama is a well-documented, recent stream capture involving several Mobile drainage forms. The speciation of other fishes on the gulf slope was reviewed by Smith-Vaniz (1968).
The homogeneity of the characters of N. l. interocularis suggests a past history in the Savannah-Chattahoochee area similar in time to that of N. l. bellicus, mentioned above. The recent entrance of N. micropogon into the upper Savannah via the Tennessee River is also discussed under geographic distribution.
The northward dispersal of N. leptocephalus stock from the Santee-Pee Dee area must have occurred during at least two major periods of exchange. The early dispersal of this multituberculate, blue-bodied form could have been northward to the Roanoke-James drainages, and subsequent isolation could have permitted divergence of the northern race toward reduced tuberculation and a brassy orange coloration. The more recent entry of the northern race of N. l. leptocephalus across the Appalachian Divide into the New River system and its capture in the James, Rappahannock, and Potomac drainages are discussed by Lachner and Jenkins (1971).
The intermediate population in the Edisto River suggests very recent exchanges of N. l. interocularis from the Savannah drainage and N. l. leptocephalus from the Santee drainage. The factors related to the development of the blue body coloration in the southern form of N. l. leptocephalus and probably all N. l. interocularis are not understood. It is of interest that both forms of the leptocephalus complex at opposite ends of the geographic distribution diverged similarly in respect to the basic body coloration of the nuptial male.
Plates
- citação bibliográfica
- Lachner, Ernest A. and Wiley, Martin L. 1971. "Populations of the polytypic species Nocomis leptocephalus (Girard) with a description of a new subspecies." Smithsonian Contributions to Zoology. 1-35. https://doi.org/10.5479/si.00810282.92
这是一篇與鯉形目相關的小作品。你可以通过编辑或修订扩充其内容。 