208
J. HESLOP-HARRISON
gressively give rise to higher and higher proportions of phenotypes
adapted to the same habitat. I n other words, ecotypes should evolve,
even in the face of the cross breeding.
Some additional factors need to be taken into account. Obviously the
intensity of the disruptive selection will be a paramount factor in determining the rate of differentiation. This probably means that this kind of
response is only likely to occur at any significant rate where extreme
habitats are encountered; and it is noteworthy that some of the best
examples of clear-cut ecotypic differentiation are found in habitats
where selection is patently intense. The most familiar situation is the
asymmetrical one, where a generally favourable habitat adjoins or is
interpenetrated by another distinctly less favourable: a coastal belt of
extreme conditions adjacent to an equable hinterland, or an area of
serpentine soil in an otherwise edaphically normal region. The response
here should be such as to produce a specialized but narrow group of
phenotypes (at least so far as the adaptive characteristics are concerned),
while permitting the survival of a much wider range in the adjoining less
demanding environment. There is, of course, no reason to expect divergence in non-adaptive characteristics.
It is to be noted that once divergence has been initiated other circumstances will subsequently develop to encourage it. Thus there will necessarily be a movement away from randomness so far as seed dispersal is
concerned, since each surviving individual is likely to seed more intensively its own immediate neighbourhood and so its own appropriate
habitat than more remote parts of the alternative habitat. I n so far as
its adaptive characteristics are heritable, this will ensure that a higher
proportion of adapted than non-adapted seedlings arise in each habitat.
The same will apply with respect to pollen dispersal; mere contiguity in
the individual habitats will guarantee some assortative mating. In
consequence, a higher proportion of recombinants giving more extreme
phenotypes will be generated, so allowing a progressive directional
change towards better and better adaptation. In a habitat where selection is intense the product would ultimately be an assemblage of biotypes lying outside of the original distributional range altogether.
The situation discussed above in which disruptive selective pressures
arise within a population because of its occupancy of a heterogeneous
environment represents a special case, since it is envisaged that the"
population is capable of a t least surviving in all of the diverse habitats
ab initio. This cannot always be true. A capacity for phenotypic plastic
adaptation may be important in permitting the invasion of some types
of unusual habitat (p. 213), but beyond any extended tolerance range
which this may provide, immigration into an environment offering
novel and perhaps intense selective pressures will demand new geneti-
J. HESLOP-HARRISON
gressively give rise to higher and higher proportions of phenotypes
adapted to the same habitat. I n other words, ecotypes should evolve,
even in the face of the cross breeding.
Some additional factors need to be taken into account. Obviously the
intensity of the disruptive selection will be a paramount factor in determining the rate of differentiation. This probably means that this kind of
response is only likely to occur at any significant rate where extreme
habitats are encountered; and it is noteworthy that some of the best
examples of clear-cut ecotypic differentiation are found in habitats
where selection is patently intense. The most familiar situation is the
asymmetrical one, where a generally favourable habitat adjoins or is
interpenetrated by another distinctly less favourable: a coastal belt of
extreme conditions adjacent to an equable hinterland, or an area of
serpentine soil in an otherwise edaphically normal region. The response
here should be such as to produce a specialized but narrow group of
phenotypes (at least so far as the adaptive characteristics are concerned),
while permitting the survival of a much wider range in the adjoining less
demanding environment. There is, of course, no reason to expect divergence in non-adaptive characteristics.
It is to be noted that once divergence has been initiated other circumstances will subsequently develop to encourage it. Thus there will necessarily be a movement away from randomness so far as seed dispersal is
concerned, since each surviving individual is likely to seed more intensively its own immediate neighbourhood and so its own appropriate
habitat than more remote parts of the alternative habitat. I n so far as
its adaptive characteristics are heritable, this will ensure that a higher
proportion of adapted than non-adapted seedlings arise in each habitat.
The same will apply with respect to pollen dispersal; mere contiguity in
the individual habitats will guarantee some assortative mating. In
consequence, a higher proportion of recombinants giving more extreme
phenotypes will be generated, so allowing a progressive directional
change towards better and better adaptation. In a habitat where selection is intense the product would ultimately be an assemblage of biotypes lying outside of the original distributional range altogether.
The situation discussed above in which disruptive selective pressures
arise within a population because of its occupancy of a heterogeneous
environment represents a special case, since it is envisaged that the"
population is capable of a t least surviving in all of the diverse habitats
ab initio. This cannot always be true. A capacity for phenotypic plastic
adaptation may be important in permitting the invasion of some types
of unusual habitat (p. 213), but beyond any extended tolerance range
which this may provide, immigration into an environment offering
novel and perhaps intense selective pressures will demand new geneti-