untitled

unstable statistic tending to increase as sample size increases. We recommend

against its use.

The level of heterozygosity within a population can be calculated by DNA sequen-

cing or gel electrophoresis. Here we consider only the latter. If the function of a given

gene is to direct the synthesis of a protein, an individual with heterozygous alleles

at that locus will produce two versions of that protein, but only one if homozygous.

Identifying the individual as one or the other is a simple technical exercise. A sample

of its blood plasma is placed on a slab of gel and subjected to a weak electrical

current which causes the plasma proteins to migrate through the gel at a speed specific

to each protein. The protein of interest, say transferrin, can be differentiated from

the other proteins by staining with Nitroso-R salt (1-nitroso-2-naphthol-3,6-disulfonic

acid) and the different versions of that protein, corresponding to different alleles, can

be identified by their banding pattern on the gel. By this means we can readily iden-

tify which alleles are present at the locus controlling transferrin synthesis and hence

whether the individual is heterozygous at that locus. The same applies for other loci

and other proteins. A precise value for Hcan be obtained by examining about

30 loci for each of about 30 individuals, and so Hcan be monitored fairly easily over

time to detect a slump in overall heterozygosity.

So far we have dealt only with single genes in isolation, those that either produce an

effect or do not. Electrophoretic analysis depends on these: a particular protein is

either produced or not according to whether a particular allele is present at the locus

of interest. This is the realm of Mendelian genetics, which deals with qualitative

characters.

Adaptive evolution, however, is based mainly on the selection of quantitative

characters such as foot length or timing of conception. These vary in a continuous

fashion. Such a character is influenced by many genes. The variation in a quan-

titative trait between individuals has two sources, environmental and genetic. The

genetic variance can be partitioned into three components: that resulting from the

effect of alleles within and between loci (additive genetic variance), that resulting

from dominance effects, and that resulting from interaction between loci. The

CONSERVATION IN THEORY 293

20

15

10

5

0

< 0.009< 0.019< 0.029< 0.039< 0.049< 0.059< 0.069< 0.079< 0.089< 0.099< 0.109< 0.119< 0.129< 0.139< 0.149< 0.159< 0.169< 0.179< 0.189

% of species

Mean heterozygosity, H

Fig. 17.1Frequency
distribution of mean
heterozygosity Hfor
169 mammalian species.
(Data from Nevo et al.
1984.)

17.3.2Additive
genetic variance