Taxometric Comparison of the Atlantic Slope and Ohio River Populations of Etheostoma caeruleum Storer

Abstract

Morphometric and meristic data of Etheostoma caeruleum Storer from the Gauley, Kanawha, Monongahela and Potomac drainages were compared using discriminant analysis. Ten of 21 characters from the four drainages had group means which were significantly different (P s .05), and the group centroids were significantly different (Ps.05). The discriminant analysis, which maximizes group differences, could correctly assign 67 % of the specimens to the a priori group. Unexpected variation was observed in anal spine and pelvic ray counts from two drainages. INTRODUCTION Etheostoma caeruleum Storer, the rainbow darter, inhabits moderate-sized streams with gravel or rubble bottoms E of the Rocky Mountains. The darter is distributed throughout the Ohio and Mississippi river drainages and is found in all of the Great Lakes drainages except Lake Superior (Stauffer and Hocutt, 1980). Although widely distributed throughout the central Appalachians, it had not been reported from the New river drainage or from any Atlantic slope drainages until recently. The darter was first reported above Kanawha Falls by Hocutt et al. (1973) from the East River, W. Va. Subsequent collections in the New River found it well distributed throughout the main stem and its tributaries (Hocutt et al., 1979), excluding the Greenbrier River (Hocutt et al., 1978). The darter is absent from the Atlantic slope except for recent collections from two forks of the South Branch of the Potomac River (Stauffer et al., 1978). The ichthyofauna of the Potomac River has been well studied in the past (Lee et al., 1976, and references therein). The discovery of a previously unreported species as distinctively colored as Etheostoma caeruleum is, therefore, surprising. Knapp (1964) examined the meristics and morphometrics of Etheostoma caeruleum from the Great Lakes, Mississippi and upper Ohio rivers. The purpose of this paper is to present meristic and morphometric data for specimens from the Potomac, Kanawha, Monongahela and New river drainages and compare the populations using discriminant analysis. METHODS AND MATERIALS The initial selection of characters to be examined in the analysis was based on Knapp's (1964) study. Other characters were included to aid in group separation. Meristic characters examined were nape scalation, number of lateral bands, total lateral line scales, pored lateral line scales, unpored lateral line scales, scales above lateral line, scales below lateral line, least caudal peduncle scales, dorsal spines, dorsal rays, total dorsal fin elements, pectoral rays, pelvic spines, pelvic rays, anal rays, anal spines and caudal rays. Morphometric characters which were standardized by dividing by standard length included: body depth at dorsal origin, caudal peduncle depth, head length, snout length, orbit length, upper jaw length, snout to dorsal fin insertion, and snout to pelvic fin insertion. Additionally, various ratios of standardized characters were examined (i.e., orbit length to snout length, orbit length to upper jaw length). All counts and measurements were made following the method outlined by Hubbs and Lagler (1974) except number of lateral bands and snout tip to pelvic fin insertion. Number of lateral bands was determined by counting all bands whether complete or in-