Of the many studies of protein polymorphism in natural populations relatively few have dealt explicitly with the role of natural selection in determining allelic frequencies (Koehn, 1969; Koehn et al., 1971; Schopf and Gooch, 1971; Koehn and Mitton, 1972; Merritt, 1972; Koehn et al., 1973; Williams et al., 1973; Mitton et al., 1973; Ayala, 1974; Ayala and Gilpin, 1975; Mitton and Koehn 1975; Koehn et al., 1976). In most cases the action of selection has not been demonstrated experimentally. Rather, selection has been inferred on the basis of observational data; the correlation between changes in allelic frequencies and environmental parameters is thought to be the result of selection for enzymes having optimal activity under specific environmental conditions. Indeed, Koehn (1969), Koehn et al. (1971) and Merritt (1972) were able to demonstrate that enzymes produced by different genotypes had different optimal conditions and that allelic frequencies in the natural populations studied appeared to vary as a function of these conditions. The logic of the present study takes the following form. If, for a given species, migration among populations is extensive, if allelic frequencies at an enzyme locus vary as a function of environmental gradients, if there is little or no inbreeding and if there is differential survival among genotypes, then selection rather than mutation pressure, gene flow or stochastic processes is likely to be the principal factor determining allelic frequencies. Some or all of these suppositions are identical to those invoked in the studies cited although they were not always explicitly stated as such. This analysis is based on the allelic variation at the phosphoglucose isomerase (E.C.5.3.1.9) locus of the sand crab, Emerita talpoida (Say). This species is essentially panmictic as a result of the extensive pelagic dispersal of its larvae during the zoeal stage of its life cycle (Johnson, 1939; Rees, 1959; Barnes and Wenner, 1968; Dillery and Knapp, 1970; Efford, 1970). Megalopae, juveniles and adults are found at high densities in local aggregations in the intertidal wash zone (Efford, 1965; Barnes and Wenner, 1968). These more sedentary stages are restricted in their distribution to the quartz sand beaches of the western Atlantic Ocean (Wharton, 1942). The largest females produce as many as 5000 eggs per berrying period (Efford, 1970), carrying their eggs for about a month (Dudley and Cox, 1967), and may have five batches of eggs per year (Efford, 1970). Partial evidence suggests the species is a protandric hermaphrodite (Barnes and Wenner, 1968). The length of the larval dispersal stage is estimated by inference to be from 2.5 to 4 months (Efford, 1965, 1967, 1970; Barnes and Wenner, 1968). The longevity of the species has not been established. In North America the distribution of E. talpoida extends from southern Massachusetts to the southern tip of mainland Florida.
Read full abstract