Abstract
In a survey of walleye (Stizostedion vitreum vitreum) from several lakes in central Canada, starch-gel electrophoresis of muscle myogens showed only the three phenotypes previously reported from Lake St. Clair. Measurements of body ratios and counts of interneural and interhaemal spines and of vertebrae gave no morphological basis for distinguishing between the three protein phenotypes. Although phenotype frequencies varied from lake to lake, within lakes the observed frequencies appeared to agree well with those calculated according to the Hardy–Weinberg rule assuming control by two nondominant alleles. The geographical variation was also consistent with the idea of reinvasion of the sample area from two separate refugia.