be entirely accurate in relation to any natural population and it
is of interest to understand the costs of making such an assumption.
For example, it is well accepted that stochastic influences are
particularly strong for small population sizes. The differing views
between the pioneers of population genetics Ronald Fisher and
Sewall Wright on the relative significance of drift and selection
are surveyed in Ewens (2004, Section 1.7), with Fisher having emphasised
the role of natural selection over stochastic effect in large
populations and Wright having given more attention to the effects
of drift in relatively small populations. The relative importance of
drift (and mutation) and selection on biological populations continues
to be debated into the present era as exemplified by Lynch
(2007). We will see that in the context of a growing population, the
degree of influence of stochastic processes is dependent not only
on population size, but rather on a parameter combining population
size and growth rate.
In this context, there have been a number of studies addressing
by theory or simulation studies the effect of variable and in
particular exponential population growth (Slatkin and Hudson,
1991; Keinan and Clark, 2012; Rogers and Harpending, 1992;
Rogers, 2014). Exponential population growth may be considered