Group Selection in Structured Populations

Publisher Website
Google Scholar
Abstract
A population model proposed by Wilson (1975) is reanalyzed. The model describes a population divided into completely isolated groups during one phase of the life history but whose members intermingle freely during another phase. Sufficient conditions are demonstrated that an altruistic genotype should be favorably selected, including its increase from mutation frequency. Neither the model nor the conditions for group selection appear to involve any biological absurdities, although these conditions appear to be more stringent than was originally suggested by Wilson. This treatment suggests that a rigid distinction between individual and group selection would be artificial.