We used Figure as an example of how the frequency of an allele (in fruit flies) does not c...

We used Figure as an example of how the frequency of an allele (in fruit flies) does not change in unselected (control) populations but does change in response to selection. However, look again at the unselected control lines in Figure The frequency of the allele in the two control populations did change a little, moving up and down over time. Which assumption of the Hardy–Weinberg model is most probably being violated? If this experiment were repeated, what change in experimental design would reduce this deviation from Hardy–Weinberg equilibrium?

Figure Frequencies of the allele in four populations of fruit flies over 50 generations The black squares and circles represent control populations living on normal food; the magenta squares and orangecircles represent experimental populations living on food spiked with ethanol. From Cavener and Clegg (1981).