Test of association of morphological variation with heterozygosity in the snail Cepaea nemoralis

Abstract

It has been suggested that the polymorphism for shell colour and pattern in Cepaea may be unrelated to the visible appearance of the shells, but maintained by heterosis or heterozygote advantage. A possible way to test this supposition is to examine whether phenotypic variance is lower in the morph exhibiting the dominant phenotype than in that with the recessive phenotype. This has been tested in two samples of shells involving totals of 1792 and 5423 individuals, respectively. Overall, there is a significantly greater variance in the recessive than the dominant class (P <0.05). Correlation between measurements on the same shell was also examined. No evidence of difference between phenotypes was found. If the difference in variance is real it may indicate either: (i) that visible appearance is not important and the polymorphism is maintained by a heterozygous balance; or (ii) evolution of such a balance in a polymorphism maintained over a long time span by selection acting on visible attributes of the shells.

A two–stage model for Cepaea polymorphism

“…The conventional means to measure heterosis would now be to examine how variance in some ¢tness character changes with multi-locus heterozygosity (see, for example, Zouros & Foltz 1987;Koehn et al 1988), but no experiments of the scale required have yet been made. As far as I know, the only attempt to measure heterosis in C. nemoralis is that of Cook & Gao (1996). In Cepaea, shell size and shape a¡ect ¢tness (Wolda 1970).…”

“…As far as I know, the only attempt to measure heterosis in C. nemoralis is that of Cook & Gao (1996). In Cepaea, shell size and shape a¡ect ¢tness (Wolda 1970).…”

Heterosis in Cepaea

“…In colonies studied for several years in enclosures, large individuals survived less well when the population of their offspring reached a high density than when they were not crowded (Cook & Cain, 1980). In the Cambridge snails, differences in variance are more marked in the comparisons involving colour than banding whereas, in the Irish ones, the reverse was true (Cook & Gao, 1996). Genes linked to the colour and banding loci are implicated, rather than direct effects of the morphs themselves.…”

“…Analysis was by the method used by Cook & Gao (1996). Variance ratio ( F ) values were calculated for each comparison by dividing the larger variance by the smaller.…”

“…Because the dominant classes for colour and banding include heterozygotes for the alleles or chromosome sections involved, whereas the recessive phenotypes do not, they might show lower variance for shell dimensions. Using material from the Cain collection at Manchester Museum Cook & Gao (1996) found that, for over 100 samples from western Ireland, the dominant classes had lower variance than the recessive classes regardless of whether the comparison was for colour, banding or the double recessive vs. the rest. Using a one‐tailed test (i.e.…”