Island Differentiation Muddied by Island Biogeographers

The effects of initial founding populations have been repeatedly undervalued by population and evolutionary biologists. Chapter 7 in MacArthur and Wilson's (1967) classic Theory of Island Biogeography is called 'Evolutionary changes following colonisation'. The authors begin by stating that “it is appropriate to study how natural selection acts on islands and, in particular, how it acts differently on islands as opposed to mainlands”. They go on to describe the evolution of new populations as passing through three phases: response to the effects of initial small size; adjustment to the novel features of the invaded environment; and finally speciation, secondary emigration and radiation. They claim that “the founder population is no more than the observation that a propagule should contain fewer genes than the entire mother population”. They err in this: the founder propagule will almost certainly have its genes (alleles) in different frequencies than in the parental population (Conrad Waddington called this 'intermittent drift' as distinct from the 'persistent drift' occurring in continuing small populations), while Sewall Wright “attributed most significance to wide random variability of gene frequencies (not fixation or loss) expected to occur simultaneously in tens of thousands of loci...”. It may be sensible to divide the 'founder principle' into a 'founder effect' (which leads to genetic impoverishment, elimination of immigration, and – probably – changed gene frequencies) but is only then followed by 'founder selection'. The MacArthur and Wilson model assumes that a newly colonised population will diverge progressively from its ancestors, and therefore the amount of differentiation could be taken as a measure of time since establishment. This would be wrong and could lead to a misleading estimate of the colonisation history of a particular site.