136
4 Molecular Evolution
tions are uniform and thus apparently founded by
single animals. The populations in various southern states of the USA, on the other hand, are
completely identical and probably originate from
the same population as that in southern France
[294].
Single-sexed reproduction by diploid parthenogenesis is found in various animal groups. Here,
the progeny of a single individual are genetically
identical; the populations consist of one or more
clones of genetically identical individuals.
Because there is no segregation or recombination, new alleles arising by mutation remain heterozygous. If a population changes from bisexual
to parthenogenetic reproduction, the heterozygosity at first increases, but then decreases again
in the course of evolution due to the accumulation of non-functional alleles. Occasional bisexual reproduction allows new alleles into the
population. In island populations of the parthenogenetic Australian snail Thiara ballonensis,
which has no bisexual neighbours, 41 electromorphs were found for 12 enzymes, each of the 13
examined populations displaying its own uniform
pattern [396]. Protein polymorphism in populations which alternate cyclically between bi- and
monosexual reproduction has been studied in
detail in waterfleas in the genus Daphnia
[281, 358] and in aphids [413].
4.3.5 Dependence of Protein Polymorphism
on the Size and History of a Population
According to the neutral theories, the average
heterozygosity should increase with the size of the
population. However, the electrophoretically
determined H value rarely exceeds 0.30, even in
very large populations. On the one hand, this
may be due to the occurrence of new variants
which have the same electrophoretic mobilities as
already existing electromorphs and therefore
remain undetected, but, on the other hand, it is
also predicted by certain of the neutral theories.
In any case, reduced polymorphism is expected in
a population founded by a few individuals (founder effect) or after the sudden reduction in size of
a population (bottleneck effect); recovery of the
H value will require many thousand generations
[263]. Evidence for this theory is found in several
cases where the history of particular populations
is well known. For example, the population of the
northern elephant seal Mirounga angustirostris
decreased in 1890 to about 20 individuals; the
effective population was in fact smaller than this
because only certain males can participate in mating. In 1974, investigations of the several thousand individuals of the newly expanded population showed no polymorphisms at 24 bloodprotein loci [35]; in contrast, in the southern
elephant seal, which has never been so greatly
decimated, 5 out of 18 loci were polymorphic.
The relatively limited extent of protein polymorphism in the large mammals (with the exception of man) is perhaps the result of frequent
bottleneck effects.
The founder effect explains many cases of
unexpectedly low polymorphism [294]. The mosquito Aedes aegypti first reached Asia in the nineteenth century; the Asiatic population correspondingly shows significantly lower polymorphism (H = 0.086) than the African and American
populations (H = 0.129-0.155) [401]. Animal
populations from small islands, lakes or ponds, or
from caves are usually less polymorphic than their
parent populations. Thus, the island populations
of the rodent Peromyscus polionatus on Santa
Rose island, off the coast of Florida, has an H
value of 0.018 compared with that of 0.067 for the
mainland population [294]. Conspicuous in this
regard is the relatively high polymorphism of the
laboratory rat, the 300 strains and sub-strains of
which were bred during the course of the century
from a few wild animals; polymorphism was
found in 9 of 22 loci examined [189].
4.3.6 Quantitative Genetic Variability
The quantitative polymorphism of proteins has
been relatively seldom investigated compared
with structural polymorphism, because the analysis using crossing experiments and concentration
measurements of single protein species requires
much more experimental effort. There is, however, no doubt that heritable variation of enzyme
activities or the concentration of non-enzymic
proteins is widespread, and that null variants are
the exception [70]. Thus, for example, in 26 out
of a total of 35 examined enzymes from the laboratory mouse, heritable activity differences were
found in proportions up to 1 : 8.8 [47]. At 14 loci
of two natural populations of Drosophila melanogaster, 58 variant alleles with lower or zero activities were recorded; with one exception, these
produced no reduction in viability or fertility in
the heterozygous state [50]. The quantitative variability of the alcohol dehydrogenase locus of
D. melanogaster has been thoroughly examined
[387]. The frequency of mutations which change
4 Molecular Evolution
tions are uniform and thus apparently founded by
single animals. The populations in various southern states of the USA, on the other hand, are
completely identical and probably originate from
the same population as that in southern France
[294].
Single-sexed reproduction by diploid parthenogenesis is found in various animal groups. Here,
the progeny of a single individual are genetically
identical; the populations consist of one or more
clones of genetically identical individuals.
Because there is no segregation or recombination, new alleles arising by mutation remain heterozygous. If a population changes from bisexual
to parthenogenetic reproduction, the heterozygosity at first increases, but then decreases again
in the course of evolution due to the accumulation of non-functional alleles. Occasional bisexual reproduction allows new alleles into the
population. In island populations of the parthenogenetic Australian snail Thiara ballonensis,
which has no bisexual neighbours, 41 electromorphs were found for 12 enzymes, each of the 13
examined populations displaying its own uniform
pattern [396]. Protein polymorphism in populations which alternate cyclically between bi- and
monosexual reproduction has been studied in
detail in waterfleas in the genus Daphnia
[281, 358] and in aphids [413].
4.3.5 Dependence of Protein Polymorphism
on the Size and History of a Population
According to the neutral theories, the average
heterozygosity should increase with the size of the
population. However, the electrophoretically
determined H value rarely exceeds 0.30, even in
very large populations. On the one hand, this
may be due to the occurrence of new variants
which have the same electrophoretic mobilities as
already existing electromorphs and therefore
remain undetected, but, on the other hand, it is
also predicted by certain of the neutral theories.
In any case, reduced polymorphism is expected in
a population founded by a few individuals (founder effect) or after the sudden reduction in size of
a population (bottleneck effect); recovery of the
H value will require many thousand generations
[263]. Evidence for this theory is found in several
cases where the history of particular populations
is well known. For example, the population of the
northern elephant seal Mirounga angustirostris
decreased in 1890 to about 20 individuals; the
effective population was in fact smaller than this
because only certain males can participate in mating. In 1974, investigations of the several thousand individuals of the newly expanded population showed no polymorphisms at 24 bloodprotein loci [35]; in contrast, in the southern
elephant seal, which has never been so greatly
decimated, 5 out of 18 loci were polymorphic.
The relatively limited extent of protein polymorphism in the large mammals (with the exception of man) is perhaps the result of frequent
bottleneck effects.
The founder effect explains many cases of
unexpectedly low polymorphism [294]. The mosquito Aedes aegypti first reached Asia in the nineteenth century; the Asiatic population correspondingly shows significantly lower polymorphism (H = 0.086) than the African and American
populations (H = 0.129-0.155) [401]. Animal
populations from small islands, lakes or ponds, or
from caves are usually less polymorphic than their
parent populations. Thus, the island populations
of the rodent Peromyscus polionatus on Santa
Rose island, off the coast of Florida, has an H
value of 0.018 compared with that of 0.067 for the
mainland population [294]. Conspicuous in this
regard is the relatively high polymorphism of the
laboratory rat, the 300 strains and sub-strains of
which were bred during the course of the century
from a few wild animals; polymorphism was
found in 9 of 22 loci examined [189].
4.3.6 Quantitative Genetic Variability
The quantitative polymorphism of proteins has
been relatively seldom investigated compared
with structural polymorphism, because the analysis using crossing experiments and concentration
measurements of single protein species requires
much more experimental effort. There is, however, no doubt that heritable variation of enzyme
activities or the concentration of non-enzymic
proteins is widespread, and that null variants are
the exception [70]. Thus, for example, in 26 out
of a total of 35 examined enzymes from the laboratory mouse, heritable activity differences were
found in proportions up to 1 : 8.8 [47]. At 14 loci
of two natural populations of Drosophila melanogaster, 58 variant alleles with lower or zero activities were recorded; with one exception, these
produced no reduction in viability or fertility in
the heterozygous state [50]. The quantitative variability of the alcohol dehydrogenase locus of
D. melanogaster has been thoroughly examined
[387]. The frequency of mutations which change