FORTY YEARS OF GENECOLOGY
171
this origin. A l l this leads to the conclusion that the comparison of “seedpopulation” and “plant-population” variances would itself be worth
while in genecological trials of perennial plants where technically
possible-as it would be in periodicity studies on clonally divisible
grasses and the like.
A word is necessary here concerning the other important distinction
which has to be made in a genecological survey - between the genetical
and non-genetical components of inter-population variation. The common practice is to.attempt to eliminate the effects of direct environmental modification of individuals by cultivating population samples
side by side in a standard garden. This method (which ante-dates genecology by at least a century) may be quite adequate to permit the kinds
of distinction required, given an appropriate design of lay-out and
analysis. In some circumstances, however, it may be unsatisfactory.
The most important deficiency of the method is that in eliminating
environmentally imposed variation it may obscure genetically determined differences in the capacity to react adaptively to special environments. This is particularly dangerous where the aim is not merely to
observe morphological differentiation but to test physiological responses.
For example, consider the case of two genecologically differentiated
populations, one with the capacity to react adaptively to intense sunlight, the other without. Tests made after side-by-side cultivation in a
r‘neutra17’ environment under moderate illumination could fail altogether to reveal any difference in response to intense light. One might go
so far as to say that an “ecotype” is never adapted to its special milieu
when cultivated in an experimental garden; it merely carries the ability
to become so adapted under the appropriate evocative environment. It
may in cultivation show differences from plants from other habitats, but
the differences need not relate at all closely to its true adaptive capacities. The work of Bjorkman and Holmgren (1963), described in detail in
a later section, is exemplary in giving f d attention to the matter of
pre-conditioning.
Another deficiency of the comparative cultivation methad lies in the
possibility that the test environment while suppressing some environmentally induced characteristics may evoke others never expressed in
the natural habitats. A sample from a phenotypically uniform natural
population may turn out to be genotypically highly diverse when observed in a different environment, and this artificial enhancement of
variance could prejudice attempts to detect and evaluate genecological
differentiation. Numerous examples of this effect have now been recorded, particularly in connection with the environmental control of
developmental periodicity ; Sinskaia (1958) and Clausen and Hiesey
(1958b) discuss some of its implications for genecology. It may be noted
171
this origin. A l l this leads to the conclusion that the comparison of “seedpopulation” and “plant-population” variances would itself be worth
while in genecological trials of perennial plants where technically
possible-as it would be in periodicity studies on clonally divisible
grasses and the like.
A word is necessary here concerning the other important distinction
which has to be made in a genecological survey - between the genetical
and non-genetical components of inter-population variation. The common practice is to.attempt to eliminate the effects of direct environmental modification of individuals by cultivating population samples
side by side in a standard garden. This method (which ante-dates genecology by at least a century) may be quite adequate to permit the kinds
of distinction required, given an appropriate design of lay-out and
analysis. In some circumstances, however, it may be unsatisfactory.
The most important deficiency of the method is that in eliminating
environmentally imposed variation it may obscure genetically determined differences in the capacity to react adaptively to special environments. This is particularly dangerous where the aim is not merely to
observe morphological differentiation but to test physiological responses.
For example, consider the case of two genecologically differentiated
populations, one with the capacity to react adaptively to intense sunlight, the other without. Tests made after side-by-side cultivation in a
r‘neutra17’ environment under moderate illumination could fail altogether to reveal any difference in response to intense light. One might go
so far as to say that an “ecotype” is never adapted to its special milieu
when cultivated in an experimental garden; it merely carries the ability
to become so adapted under the appropriate evocative environment. It
may in cultivation show differences from plants from other habitats, but
the differences need not relate at all closely to its true adaptive capacities. The work of Bjorkman and Holmgren (1963), described in detail in
a later section, is exemplary in giving f d attention to the matter of
pre-conditioning.
Another deficiency of the comparative cultivation methad lies in the
possibility that the test environment while suppressing some environmentally induced characteristics may evoke others never expressed in
the natural habitats. A sample from a phenotypically uniform natural
population may turn out to be genotypically highly diverse when observed in a different environment, and this artificial enhancement of
variance could prejudice attempts to detect and evaluate genecological
differentiation. Numerous examples of this effect have now been recorded, particularly in connection with the environmental control of
developmental periodicity ; Sinskaia (1958) and Clausen and Hiesey
(1958b) discuss some of its implications for genecology. It may be noted