194
J. HESLOP-HARRISON
the heterogeneous species population” shows that he looked upon
adaptation to habitat as the outcome of selection ; but, as Baker (1953)
has pointed out, he and other pioneer workers were handicapped by
imperfect conceptions of the nature of the variability of plant populations, and were accordingly unable to formulate at all clearly how selection was likely to operate in bringing about the differentiation of habitat
races. At least in his earlier papers Turesson wrote as if he looked upon
the typical plant species as consisting of a generalized population plus
specialized “radiations” from it, the ecotypes. The generalized population was rich in “biotypes” (defhed by Faegri, 1937, as groups of individuals with identical genetical constitution), while the ecotypes were
“biotype depleted”. This conception was opposed by Faegri (1937), who
objected to the idea that biotypes pre-adapted for alpine or other extreme conditions could survive in an undifferentiated lowland population awaiting, as it were, their opportunity to invade a mountain
habitat, and argued further that if specialization necessitated genetical
impoverishment ecotypes would be populations ultimately destined to
become extinct rather than to be the starting points of new species.
Much of the early difficulty lay in the impression that genetical variation in plant populations must necessarily be largely overt. The species
consisted of biotypes (sometimes referred to as though they constituted
something like self-propagating pure lines, even in outbreeding species) ;
and the biotypes were the units which were selected or rejected in the
course of adaptation to habitat. VC’ithout some appreciation of the subtle
ways the ebb and flow of variation is governed in plant populations no
other standpoint was possible. Current understanding of the way genetical variation may be generated, recombined, exposed, conserved, concealed and lost in the course of sexual reproduction may be dated from
two principal publications - Darlington’s “Evolution of Genetic Systems” of 1939, and Mather’s paper on polygenic inheritance and natural
selection of 1943. More recent discussions of these processes in higher
plant populations are given in Stebbins’ “Variation and Evolution in
Plants” of 1950, and in review papers by Baker (1953, 19591, Stebbins
(1957,1958) and Grant (1958).
In general, adaptive changes in a population must depend upon the
accumulation of minor genic changes, mutations, under the directiye
action of selection. As Mather (1943) has shown, a compromise w i l l
always be involved between fitness for the environment as it exists, and
the flexibility which will permit further adaptive change.
Fitness is best served by the production of progeny optimal for the
immediate circumstances ; flexibility by the continuous generation of
variants some of which may be optimal for environments only to be encountered in the future, or elsewhere on the earth’s surface. The raw
J. HESLOP-HARRISON
the heterogeneous species population” shows that he looked upon
adaptation to habitat as the outcome of selection ; but, as Baker (1953)
has pointed out, he and other pioneer workers were handicapped by
imperfect conceptions of the nature of the variability of plant populations, and were accordingly unable to formulate at all clearly how selection was likely to operate in bringing about the differentiation of habitat
races. At least in his earlier papers Turesson wrote as if he looked upon
the typical plant species as consisting of a generalized population plus
specialized “radiations” from it, the ecotypes. The generalized population was rich in “biotypes” (defhed by Faegri, 1937, as groups of individuals with identical genetical constitution), while the ecotypes were
“biotype depleted”. This conception was opposed by Faegri (1937), who
objected to the idea that biotypes pre-adapted for alpine or other extreme conditions could survive in an undifferentiated lowland population awaiting, as it were, their opportunity to invade a mountain
habitat, and argued further that if specialization necessitated genetical
impoverishment ecotypes would be populations ultimately destined to
become extinct rather than to be the starting points of new species.
Much of the early difficulty lay in the impression that genetical variation in plant populations must necessarily be largely overt. The species
consisted of biotypes (sometimes referred to as though they constituted
something like self-propagating pure lines, even in outbreeding species) ;
and the biotypes were the units which were selected or rejected in the
course of adaptation to habitat. VC’ithout some appreciation of the subtle
ways the ebb and flow of variation is governed in plant populations no
other standpoint was possible. Current understanding of the way genetical variation may be generated, recombined, exposed, conserved, concealed and lost in the course of sexual reproduction may be dated from
two principal publications - Darlington’s “Evolution of Genetic Systems” of 1939, and Mather’s paper on polygenic inheritance and natural
selection of 1943. More recent discussions of these processes in higher
plant populations are given in Stebbins’ “Variation and Evolution in
Plants” of 1950, and in review papers by Baker (1953, 19591, Stebbins
(1957,1958) and Grant (1958).
In general, adaptive changes in a population must depend upon the
accumulation of minor genic changes, mutations, under the directiye
action of selection. As Mather (1943) has shown, a compromise w i l l
always be involved between fitness for the environment as it exists, and
the flexibility which will permit further adaptive change.
Fitness is best served by the production of progeny optimal for the
immediate circumstances ; flexibility by the continuous generation of
variants some of which may be optimal for environments only to be encountered in the future, or elsewhere on the earth’s surface. The raw