Abstract
Collins (1985) has described two separate mouse strains, obtained by selective breeding, which differ in having high (HI) or low (LO) degrees of paw preference on a standard test. In this paper I argue that the differences between these strains may not be due to a specific gene (or genes) but, instead, probably reflect differences in the total heterozygosity of the strains, such that the HI strain is more heterotic than the LO strain. Greater degrees of heterozygosity are argued to buffer against fluctuating asymmetry and hence result in a greater degree of paw preference.