140
Μ. J. D. WHITE
D. persimilis; Hsu (1952) for the virilis group; Carson (1954) for D.
bocainensis and its relatives]. This might indicate that a species in statu
nascendi is likely to pass through a relatively monomorphic stage,
acquiring a new assortment of polymorphisms later. Or it may simply
indicate that adaptation to new environments and the passage of time
are likely to lead to the replacement of old polymorphisms by new ones.
However, the possession by very closely related species of different polymorphisms is just what we should expect if speciation were frequently
preceded by a stage in which a species has developed different polymorphic systems in different parts of its range, which are incompatible
in the sense that they could not coexist in the same population without a
breakdown of the adaptive mechanisms.
A possible example of such a situation has been considered by Wallace
(1953), who points out that certain combinations of three overlapping
inversions ("triads") in a population will lead to a breakdown, by crossing-over, of the coadapted genetic complexes within each sequence.
The data on D. pseudoobscura
support the hypothesis, since there are
almost no populations in which the three members of a triad coexist in
high frequency. On the other hand, in D. robusta all three sequences
constituting a triad are often found in high frequency in the same population (Levitan et al., 1954). The discrepancy may well be due to differences between the patterns of crossing-over in the two species. Wallace
has put forward a very plausible hypothesis to account for the evolutionary separation of D. pseudoobscura
and D. persimilis, based on the
"triad" principle, and it seems likely that the particular mechanism
which he envisages is only one of a number of generally similar ones,
based on various kinds of incompatibility between different polymorphic
systems, which may lead to speciation. Thus polymorphic systems may
be genetically
incompatible (because of selection against individuals
bearing some chromosomes derived from each of two such systems);
or they may be cytologically
incompatible (if the meiosis of such individuals leads to the formation of gametes carrying deficiencies or
duplications). But there is as yet no reason to believe that incompatibilities of this general type are the only cause, or even the main cause,
of speciation in animals.
Many instances are now known where a species possesses a very
elaborate system of chromosomal polymorphism in one part of its range
and a much simpler one in another part. Thus populations of the grasshopper Trimerotropis suffusa from the Sierra Nevada and Cascade ranges
of the Western United States show almost ten times as much structural
heterozygosity for pericentric rearrangements as ones from the Rocky
Mountain area of Colorado and Wyoming (White and Morley, 1955).
Μ. J. D. WHITE
D. persimilis; Hsu (1952) for the virilis group; Carson (1954) for D.
bocainensis and its relatives]. This might indicate that a species in statu
nascendi is likely to pass through a relatively monomorphic stage,
acquiring a new assortment of polymorphisms later. Or it may simply
indicate that adaptation to new environments and the passage of time
are likely to lead to the replacement of old polymorphisms by new ones.
However, the possession by very closely related species of different polymorphisms is just what we should expect if speciation were frequently
preceded by a stage in which a species has developed different polymorphic systems in different parts of its range, which are incompatible
in the sense that they could not coexist in the same population without a
breakdown of the adaptive mechanisms.
A possible example of such a situation has been considered by Wallace
(1953), who points out that certain combinations of three overlapping
inversions ("triads") in a population will lead to a breakdown, by crossing-over, of the coadapted genetic complexes within each sequence.
The data on D. pseudoobscura
support the hypothesis, since there are
almost no populations in which the three members of a triad coexist in
high frequency. On the other hand, in D. robusta all three sequences
constituting a triad are often found in high frequency in the same population (Levitan et al., 1954). The discrepancy may well be due to differences between the patterns of crossing-over in the two species. Wallace
has put forward a very plausible hypothesis to account for the evolutionary separation of D. pseudoobscura
and D. persimilis, based on the
"triad" principle, and it seems likely that the particular mechanism
which he envisages is only one of a number of generally similar ones,
based on various kinds of incompatibility between different polymorphic
systems, which may lead to speciation. Thus polymorphic systems may
be genetically
incompatible (because of selection against individuals
bearing some chromosomes derived from each of two such systems);
or they may be cytologically
incompatible (if the meiosis of such individuals leads to the formation of gametes carrying deficiencies or
duplications). But there is as yet no reason to believe that incompatibilities of this general type are the only cause, or even the main cause,
of speciation in animals.
Many instances are now known where a species possesses a very
elaborate system of chromosomal polymorphism in one part of its range
and a much simpler one in another part. Thus populations of the grasshopper Trimerotropis suffusa from the Sierra Nevada and Cascade ranges
of the Western United States show almost ten times as much structural
heterozygosity for pericentric rearrangements as ones from the Rocky
Mountain area of Colorado and Wyoming (White and Morley, 1955).