## A A

= 0.30. To determine the aver-

age expected heterozygosity of the subpopulations requires observed allele frequencies for each subpopulation. In subpopulation one 13 of the 20 alleles are white and seven of the 20 alleles are blue. If p is the frequency of the white allele and q the frequency of the blue allele, then p1 = 13/20 = 0.65 and q1 = 1 -p1 = 0.35. In subpopulation two the situation is the exact opposite with p2 = 7/20 = 0.35 and q2 = 1 - p2 = 0.65. The average expected heterozygosity in the two subpopulations is then

HS = 1[2(0.65)(0.35) + 2(0.35)(0.65)] = 0.455. In the total population average allele frequencies are p = 1(0.65 + 0.35) = 0.50 and q = 1(0.35 + 0.65)

= 0.50. (Notice that obtaining the average of the subpopulation allele frequencies is equivalent to combining all the alleles in the total population and then estimating the allele frequency, as in 13 + 7

p = ——— = 0.50.) The expected heterozygosity of the total population is then HT = 2(0.5)(0.5) = 0.5.

After calculating the different observed and expected heterozygosities in Fig. 4.6, it is apparent that they are not all equivalent. There are differences between the observed and expected heterozygosities at the different hierarchical levels of the population. Recall from section 2.5 that the difference between observed and Hardy-Weinberg expected genotype frequencies was used to estimate the fixation index or F. In that case there was only a single population and we were only concerned with how alleles combined into diploid genotypes compared with the expectation under random mating. The fixation index can be extended to accommodate multiple levels of population organization, thereby creating measures of deviation from Hardy-Weinberg expected genotype frequencies caused by two distinct processes. With multiple subpopulations there is a possible excess or deficit of heterozygotes due to non-random mating within subpopulations and a possible deficit of heterozygotes among subpopulations compared to panmixia. In the latter case the fixation index will show how much allele frequencies have diverged among subpopulations due to processes that cause population structure compared with the ideal of

Table 4.6 The mathematical and biological definitions of fixation indices for two levels of population organization.

0 0