CHROMOSOMAL EVOLUTION AND SPECIATION
111
of those species, but which outlived their usefulness (because of changes
in the environment or in the genetic composition or breeding system
of the population) and were replaced by new monomorphisms. This
generalization seems to be of great importance because it enables us to
reason, to some extent, from cytotaxonomy to population genetics, and
vice versa, and endows both types of study with a new meaning. Thus
we know almost nothing about chromosomal polymorphism in mammalian populations, but the great number and magnitude of the cytotaxonomic differences between related species of mice and other rodents,
for example, force us to the conclusion that such polymorphisms must
have existed in the past, and therefore that some of them presumably
still exist. To take a concrete instance: Wahrman and Zahavi (1953)
report the chromosome numbers (2n) of the Palestinian spiny mice
Acomys cahirinus and A. russatus as 38 and 66, respectively. It is clear
that numerous evolutionary changes in chromosome number have taken
place in the adaptive radiation of this genus, and it would be surprising if a thorough study of the natural populations of the numerous
African species did not reveal some instances of adaptive polymorphism
in respect of chromosome "fusions" and "fragmentations." An instance
of chromosomal polymorphism (almost the only credible one recorded
in the Mammalia) is reported by Wahrman and Zahavi (1955) in the
rodent genus Gerbillus, which is likewise notable for a wide range of
chromosome numbers (Matthey, 1953, 1954a,b). Up till now, vertebrate
cytologists have been more interested in cytotaxonomy than in population cytogenetics and have frequently been satisfied with an examination of a very few individuals of a species, or even a single one. Thus
their methods have been adapted to reveal cytotaxonomie differences
but not cytological polymorphism in natural populations.
In those groups where both types of study have been carried out the
evidence in general supports the thesis that a parallelism exists between
the extent and kind of the adaptive cytological polymorphisms in populations and the cytotaxonomie differences between species. Thus in the
repleta group of the genus Orosophila there are relatively few chromosomal rearrangements either "within" or "between" species, whereas
in the obscura group there is a much higher frequency of paracentric
inversions, both in a state of flux in natural populations and as stable
cytotaxonomie differences between species (Wasserman, 1954; Patterson and Stone, 1952). And within the virilis group, one subgroup
centered around D. montana shows far more rearrangements both within
and between species than the subgroup which includes D. virilis, D.
novamexicana,
and D. americana
(Hsu, 1952; Patterson and Stone,
1952; Moorhead, 1954). In those grasshopper genera such as Trímero-
111
of those species, but which outlived their usefulness (because of changes
in the environment or in the genetic composition or breeding system
of the population) and were replaced by new monomorphisms. This
generalization seems to be of great importance because it enables us to
reason, to some extent, from cytotaxonomy to population genetics, and
vice versa, and endows both types of study with a new meaning. Thus
we know almost nothing about chromosomal polymorphism in mammalian populations, but the great number and magnitude of the cytotaxonomic differences between related species of mice and other rodents,
for example, force us to the conclusion that such polymorphisms must
have existed in the past, and therefore that some of them presumably
still exist. To take a concrete instance: Wahrman and Zahavi (1953)
report the chromosome numbers (2n) of the Palestinian spiny mice
Acomys cahirinus and A. russatus as 38 and 66, respectively. It is clear
that numerous evolutionary changes in chromosome number have taken
place in the adaptive radiation of this genus, and it would be surprising if a thorough study of the natural populations of the numerous
African species did not reveal some instances of adaptive polymorphism
in respect of chromosome "fusions" and "fragmentations." An instance
of chromosomal polymorphism (almost the only credible one recorded
in the Mammalia) is reported by Wahrman and Zahavi (1955) in the
rodent genus Gerbillus, which is likewise notable for a wide range of
chromosome numbers (Matthey, 1953, 1954a,b). Up till now, vertebrate
cytologists have been more interested in cytotaxonomy than in population cytogenetics and have frequently been satisfied with an examination of a very few individuals of a species, or even a single one. Thus
their methods have been adapted to reveal cytotaxonomie differences
but not cytological polymorphism in natural populations.
In those groups where both types of study have been carried out the
evidence in general supports the thesis that a parallelism exists between
the extent and kind of the adaptive cytological polymorphisms in populations and the cytotaxonomie differences between species. Thus in the
repleta group of the genus Orosophila there are relatively few chromosomal rearrangements either "within" or "between" species, whereas
in the obscura group there is a much higher frequency of paracentric
inversions, both in a state of flux in natural populations and as stable
cytotaxonomie differences between species (Wasserman, 1954; Patterson and Stone, 1952). And within the virilis group, one subgroup
centered around D. montana shows far more rearrangements both within
and between species than the subgroup which includes D. virilis, D.
novamexicana,
and D. americana
(Hsu, 1952; Patterson and Stone,
1952; Moorhead, 1954). In those grasshopper genera such as Trímero-