446 



Fishery Bulletin 101(2) 



cing allele sizes of 175-181 bp for silver seatrout and 

 175-193 bp for sand seatrout. Amplification of the silver 

 seatrout and sand seatrout samples with the remaining 

 three microsatellite and two intron loci did not provide fur- 

 ther evidence of hybridization. SOC044 and CNE612 allele 

 size ranges for both species fell within the range exhibited 

 by the weakfish samples, and the SOC014 and both intron 

 loci did not amplify in either the silver seatrout or sand 

 seatrout samples. 



Individuals with unusually small SOC050 alleles from 

 the Georgia 1997 sample fell into one of four general 

 classes. Seven individuals had silver seatrout mtDNA and 

 two small SOC050 alleles and were presumably pure silver 

 seatrout. The inclusion of these individuals in the collection 

 may not be surprising because both weakfish and silver 

 seatrout are common in the South Atlantic Bight (Bigelow 

 and Schroeder, 1953; Hildebrand, 1955) and are difficult to 

 distinguish during their early life history stages. Although 

 the latter species is known to inhabit deeper waters as 

 adults (Ginsburg, 1931), both species are inshore summer 

 spawners (Devries and Chittenden, 1982; Shepherd and 

 Grimes, 1984). 



Three individuals possessed sand seatrout mtDNA and 

 two small SOC050 alleles and were presumably pure sand 

 seatrout. Some researchers have suggested that weak- 

 fish and sand seatrout represent separate populations 

 of a single species (Moshin, 1973; Weinstein and Yerger, 

 1976; Cowan, 1985; Ditty, 1989), and others treat them as 

 separate species (Schlossman and Chittenden, 1981) with 

 distributions confined to the western Atlantic (weakfish) 

 and the Gulf of Mexico (sand seatrout). Paschall (1986) 

 was unable to distinguish between the two species using 

 allozyme electrophoresis. In contrast, results presented 

 here are consistent with the existence of two distinct spe- 

 cies, with weakfish and sand seatrout co-occurring off the 

 east coast of the United States at least as far north as 

 Doboy Sound, Georgia. This distribution pattern is consis- 

 tent with the phylogeographic patterns of 19 freshwater, 

 coastal, and marine species distributed along the U.S. East 

 Coast and the Gulf of Mexico that exhibited geographically 

 concordant forks in their intra- or interspecific mtDNA 



phylogenies (or in both phylogenies) (Avise, 1992). In the 

 present situation, apparently distinct Gulf (sand seatrout) 

 and Atlantic (weakfish) species may have reestablished 

 contact in a hybrid zone (see below) through movement of 

 the Gulf species into the Atlantic. 



Three individuals had weakfish mtDNA and a single 

 small SOC050 allele and were presumably hybrids of 

 weakfish and sand seatrout or silver seatrout (with female 

 weakfish parentage). In addition, two individuals pos- 

 sessed sand seatrout mtDNA and a single small SOC050 

 allele and were presumably hybrids of weakfish and sand 

 seatrout with female sand seatrout parentage. These data 

 suggest that hybridization occurs between weakfish and 

 sand seatrout and that the genetic exchange is not gender 

 restricted. Because of the overlap in microsatellite allele 

 sizes seen between silver seatrout and sand seatrout, hy- 

 bridization between weakfish and silver seatrout could not 

 be excluded. The lack of suspected hybrids with silver seat- 

 rout mtDNA, however, suggests that hybridization did not 

 involve this species. The possibility exists that the putative 

 hybrids are in fact weakfish with rare mtDNA haplotypes 

 common to the three Cynoscion species studied here. This 

 seems unlikely because only one 12S/16S mtDNA//?,sn 

 I pattern was noted among 40 weakfish in the species 

 identification study of Cordes et al. (2001). Furthermore, 

 analysis of 20 weakfish taken from each of the four loca- 

 tions outside of Georgia with the 12S/16S marker revealed 

 no new mtDNA patterns. Also, the mtDNA haplotypes seen 

 in sand seatrout and silver seatrout seem to vary in size 

 and can not be clearly related to the weakfish haplotype 

 by the addition or deletion of presumed restriction sites. 

 This condition is more in keeping with mtDNA of different 

 species, although the apparent size differences may be gel 

 artifacts due to unresolved bands <100 bp in one or more 

 of the species. 



Re-evaluation of the remaining 1996-97 SOC050 data re- 

 vealed occasional occurrences of small alleles in individual 

 fish m all but the New York samples (Table 3). Examina- 

 tion of the 12S/16S mtDNA region of these individuals 

 identified a single silver perch (Bairdiella chrysoiiro) in 

 the Chesapeake Bay 1997 sample (silver perch mtDNA 



