Enzyme Polymorphism and Adaptive Strategy in the Decapod Crustacea

Abstract

Protein polymorphism in 44 species of decapod Crustacea was assayed by electrophoresis. An average of 26 loci per species and 24 individuals per locus were surveyed. Each species (together with an additional seven species for which electrophoretic data were already available) was ranked on each of 10 niche descriptors: adult size, post-larval vagility, trophic level, trophic generalism, number of species per genus, decapod species diversity, latitude, productivity, euryhalinity and littorality; in addition each was ranked on two composite measures, Levinton's "index of opportunism" and a measure of trophic resource instability. Niche descriptors were intercorrelated over species and found to fall into three groups, an "organismic" group, an "environmental" group, and a third consisting of productivity and latitude. Maximum heterozygosities for 10 enzymes were intercorrelated. Though correlations were weak, the enzymes fell into two groups corresponding approximately to the groups I (central metabolic enzymes) and II (enzymes processing externally derived substrates) of Gillespie and others. Maximum heterozygosities were correlated with niche descriptors over species. Group I maximum heterozygosities were found to be largely positively correlated with "environmental" descriptors in which the common factor appeared to be environmental variability, and somewhat negatively correlated with the "organismic" descriptors such as size and trophic level. Group II maximum heterozygosities, on the other hand, were largely negatively correlated with the "environmental" descriptors (except latitude and productivity) and largely positively correlated with the "organismic" descriptors, almost the reverse of the correlation pattern of the group I enzymes. Neither population size fluctuations (bottlenecks) nor the random processes and normative selection of mutation-selection balance theory can account for the differential correlation of group I and group II enzymes with the "organismic" and "environmental" group of niche descriptors. A hybrid "environmental heterogeneity-trophic diversity" model is proposed to account for the results. Species with "coarse-grain" adaptive strategies show higher group I heterozygosities; because in the decapods these species are more likely to be trophic specialists, their group II heterozygosities are lower. Species on a "fine-grain" path have lower group I heterozygosities, but because they tend to span phyletically broader trophic spectra they have higher group II heterozygosities. Group I polymorphism is greater in more variable environments, as in the Levene-Levins model. Group II heterozygosities, on the other hand, tend to be lower in the more variable habitats because, at least within the decapods sampled, the tendency to trophic specialization and its correlates is greater there, and perhaps because phyletic diversity of potential prey is reduced. Average heterozygosity and percent polymorphic loci tended to have correlations with niche descriptors that were large when the group I and group II correlations were of the same sign, and small when group I and group II correlations canceled one another. Thus, average measures tend to obscure real differences in heterozygosity pattern between species with different adaptive strategies.