On evolution / John Maynard Smith.

  • Maynard Smith, John, 1920-2004.
Date:
1972
Catalogue details

Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Credit: On evolution / John Maynard Smith. Source: Wellcome Collection.

    22/144 (page 10)
    is a conflict between 'group selection' favouring conventional behaviour and 'individual selection' against it. It follow^s that if we argue that conventional methods have evolved because they reduce the risk that individuals will be injured, we may tacitly be assuming that group selection is more effective than individual selection. For reasons given in the next section, this is probably not so. The main purpose of this paper is to see whether an explanation of conventional fighting can be given in terms of selection at the individual level. But first, I want to explain in a little more detail my reluctance to accept group selection as an explanation, s Group selection and kin selection Most discussions of natural selection, and almost all mathe¬ matical models of it, are concerned with individual selection. A population consists of individuals of two ( or more ) dif¬ ferent genotypes, say A and a. On the average, A individuals leave more offspring than a individuals. Depending on the par¬ ticular mechanisms of inheritance, this will cause a calculable change in gene frequency each generation. This is 'individual selection'. In contrast, suppose that a species were divided into a number of groups, reproductively isolated from one another, some consisting entirely of A individuals and some of a indi¬ viduals, and suppose also that from time to time populations be¬ come extinct and that other populations split to form two. It might be that A groups are more likely to survive and split than a groups. If so, in time all groups will consist entirely of A individuals. This remains true even if, in a mixed group of A and a indi¬ viduals, a replaces A by individual selection. To give an exam¬ ple, if a individuals are cannibalistic, it might be that a groups are liable to extinction, but that in mixed groups a would win over A. The process whereby all groups become A groups is called group selection. Note that it will not work if migration from group to group is at all common, because groups of A indi¬ viduals are constantly being 'infected' by migrant a individuals, and any 'infected' group will become an a group by individual selection. However, no such migration occurs between different species, so group selection can certainly act at the species level, by the extinction of some species and the survival of others. It may also act within a species which ( because of the 10