Group Selection: The Phenotypic and Genotypic Differentiation of Small Populations

Abstract

The biological character or phenotype of a population is determined both by the genotypes of its component individuals and by their interaction with biotic and abiotic factors in the environment. To the extent that there are genetic differences among individuals, differences among populations may be generated by both drift and selection. In most theoretical studies, the populational phenotype is completely characterized by the genotypic or phenotypic mean and variance of its component individuals. In these cases, the population characteristic in question is simply a statistical summarization of its component parts. When this description of the relationship between individual and population phenotypes is accurate, the phenotypic differentiation of an array of populations owing to drift or selection can readily be represented by the frequency distribution of within-population means. This distribution is a complete characterization because the variance within the population in most genetic models is a function of the mean gene frequency and in most quantitative models this variance is assumed to be constant.However, laboratory studies of genotype-genotype and genotype-density interactions (Lewontin and Matsuo, 1963;Sokal and Karten, 1964;Weisbrot, 1966;McCauley and Sokal, 1977) have illustrated many instances in which the biological character or phenotype of a population is not a simple linear function of its component parts. As a result of the interactions among individuals, it is possible, at least in principle, for small genetic differences between groups of individuals to become relatively large differences in

The Group-Level Consequences of Sexual Conflict in Multigroup Populations

“…When dispersal is random, heterogeneity amongst groups is typically assumed to decline, therefore reducing the strength and relevance of group selection in the population [36]. However, classic empirical studies on group selection in the flour beetle ( Tribolium castaneum ) have shown that even high levels of random dispersal (25% per generation) between local populations does not prevent genetic differentiation nor interfere with selection at the group level [37], [38]. Furthermore, the random dispersal of individuals is more the exception than the rule, as the movement of animals is often in response to environmental conditions such as food, resources, or composition of the social group [18]–[20].…”

Is group selection a factor modulating the virulence of RNA viruses?

“…On the basis of theoretical (Levins, 1970 ;Wilson, 1973Wilson, , 1975Levin & Kilmer, 1974 ;Gadgil, 1975 ;Gilpin, 1975) and experimental (Wade, 1976(Wade, , 1977Wade & McCauley, 1980) results it seems clear that in order to detect the existence of any group selection component in the evolution of any population, it is necessary that individual and group selection act in opposite directions. This was the case in our experiment.…”

“…The general conclusion was that group selection would be of significant strength in natural populations only under some restrictive conditions that could be generated in the laboratory. Wade (1976Wade ( , 1977 and Wade & McCauley (1980) carried out experiments with the flour beetle, Tribolium castaneum, demonstrating that group selection could be an important factor influencing population size in organisms with a population structure promoting rapid genotypic and phenotypic divergence between local demes. The evolution of virulence has been modelled extensively under different theoretical conditions (Anderson, 1979 ;Anderson & May, 1979 ;May & Anderson, 1979a, b, 1990Bremermann & Pickering, 1983 ;Bremermann & Thieme, 1989 ;Stewart & Levin, 1984 ;Levin & Svanborg Ede!…”

Population Biology

“…The analogy with the "fundamental theorem of natural selection", that response is proportional to heritability, was introduced by Wade and McCauley (1980). (45).…”