FORTY YEARS O F GENECOLOGY
189
consideration in the design of genecological investigations so as to
determine the scale and nature of the sampling programmes to be
adopted to produce information of value in comparative studies.
H. GENETIC BASIS O F ECOTYPIC DIFFERENCES
It is a commonplace of descriptive genecology that ecotypes tend to
differ in characteristics such as growth habit and dimensions of organs
which show continuous variation within populations. Moreover, the
gradation between ecoclinally related populations in these features and
in others such as developmental periodicity is itself usually found to be
continuous, or so finely stepped as to be effectively so when the modulating effects of the environment are superimposed. These two observations are best explained by the hypothesis that ecotypic differences are
mostly polygenically controlled (Mather, 1943).
Discussing in 1953 the genetical structure of natural populations
Mather noted that the direct evidence of polygenic systems mediating
continuous variation was not abundant, most likely because the experimental demonstration of their presence is difficult. Among plant
examples he quoted the species of Layia and Madia mentioned by
Clausen et al. (1940), and the same authors’ comments on the ecotypic
variation of Achillea borealis (Hiesey et al., 1942).
The analysis now available of the genetics of the races of Potentilla
glandulosa from the more recent work of the Carnegie group (Clausen
and Hiesey, 1958) provides much the fullest evidence of the genetic
basis of ecotypic differences. Most of the differential features are controlled multifactorially ; but the number of loci concerned is seemingly
not especially high.
I n the study, crosses were made between individuals of the subspecies
typica (Californian coast), reJlexa (Sierran foothills) and nevadensis (subalpine form and alpine form). The most fully analysed crosses were
between typica and the alpine form of nevadensis, and between rejexa
and the subalpine form of nevadensis. I n the first of these, large F, and
F, populations were studied in a standard garden, and in the second, F,,
F, and F, populations were grown in the standard garden and in addition
cloned F, individuals were compared at three transplant stations at
different altitudes on the Californian transect. Up to nineteen characters
were observed simultaneously in the experiments, and the huge body of
data was processed by a punched card technique. Estimates of the
minimum number of genes associated with the character differences
observed between the different races were based upon comparisons of F,
frequency distributions with binomial distributions. These estimates for
nineteen characters are summarized in Table 11, from the original of
Clausen and Hiesey. For the whole complex of differential features
G2
C.E.R.
189
consideration in the design of genecological investigations so as to
determine the scale and nature of the sampling programmes to be
adopted to produce information of value in comparative studies.
H. GENETIC BASIS O F ECOTYPIC DIFFERENCES
It is a commonplace of descriptive genecology that ecotypes tend to
differ in characteristics such as growth habit and dimensions of organs
which show continuous variation within populations. Moreover, the
gradation between ecoclinally related populations in these features and
in others such as developmental periodicity is itself usually found to be
continuous, or so finely stepped as to be effectively so when the modulating effects of the environment are superimposed. These two observations are best explained by the hypothesis that ecotypic differences are
mostly polygenically controlled (Mather, 1943).
Discussing in 1953 the genetical structure of natural populations
Mather noted that the direct evidence of polygenic systems mediating
continuous variation was not abundant, most likely because the experimental demonstration of their presence is difficult. Among plant
examples he quoted the species of Layia and Madia mentioned by
Clausen et al. (1940), and the same authors’ comments on the ecotypic
variation of Achillea borealis (Hiesey et al., 1942).
The analysis now available of the genetics of the races of Potentilla
glandulosa from the more recent work of the Carnegie group (Clausen
and Hiesey, 1958) provides much the fullest evidence of the genetic
basis of ecotypic differences. Most of the differential features are controlled multifactorially ; but the number of loci concerned is seemingly
not especially high.
I n the study, crosses were made between individuals of the subspecies
typica (Californian coast), reJlexa (Sierran foothills) and nevadensis (subalpine form and alpine form). The most fully analysed crosses were
between typica and the alpine form of nevadensis, and between rejexa
and the subalpine form of nevadensis. I n the first of these, large F, and
F, populations were studied in a standard garden, and in the second, F,,
F, and F, populations were grown in the standard garden and in addition
cloned F, individuals were compared at three transplant stations at
different altitudes on the Californian transect. Up to nineteen characters
were observed simultaneously in the experiments, and the huge body of
data was processed by a punched card technique. Estimates of the
minimum number of genes associated with the character differences
observed between the different races were based upon comparisons of F,
frequency distributions with binomial distributions. These estimates for
nineteen characters are summarized in Table 11, from the original of
Clausen and Hiesey. For the whole complex of differential features
G2
C.E.R.