CHROMOSOMAL EVOLUTION AND SPECIATION
125
thereby giving rise to aneuploid gametes. Orientation is more likely to
be regular, and hence fusions are more likely to establish themselves, in
species with a pronounced distal localization of chiasmata. The fact
that centric fusions only seem to have established themselves in about
5 out of 177 species of spiders which are known cytologically (Suzuki,
1954) is probably due to the prevalence of proximal chiasma localization in this group of animals (in most species of spiders all the chromosomes are acrocentric). Why so many species of spiders should have
the chiasmata localized in this way, at least in the males, is a question
that cannot be answered at present.
The second precondition for the "success" of a centric fusion is a
strictly genetic one, i.e., either an adaptive position effect or the bringing together in a linked combination of two or more interacting genes
or gene complexes (that position effects may sometimes be of importance in natural populations is indicated by the work of Levitan, 1954,
on Orosophila
robusta).
We are now in a position to attempt an answer to the question: why
do approximately 95% of the grasshopper species belonging to the subdivision Cryptosacci retain a karyotype consisting of 23 acrocentric elements (in the diploid set of the males)? And, in particular, why do
they differ in this respect from a number of other groups, such as the
crickets and the Australian eumastacid grasshoppers of the subfamily
Morabinae (White, unpublished data), in which a large number of
centric fusions seem to have established themselves?
Differences in chiasma localization may furnish a partial explanation.
Where, as in species of Chorthippus,
metacentric chromosomes which
have clearly arisen by centric fusion have nonlocalized chiasmata (or,
as in certain Morabinae, proximally localized ones) the pattern of chiasma
distribution has probably changed since the establishment of the fusion.
Even if the frequency and distribution of the chiasmata is favorable
to the establishment of centric fusions, the majority of those that occur
spontaneously are unlikely to succeed in evolution because they confer
no adaptive superiority on individuals heterozygous for them. Only
when the genetic contents of the two fusing elements interact to produce
a complex which forms a heterotic combination with the unfused homologues in the population, or with some of them, can we expect the
fusions to persist in the population (it does not seem imperative to suppose with Sheppard, 1953, that there is any necessary connection with
conspicuous visible polymorphisms such as those of Cepaea and the
grouse locusts, although such polymorphisms are the rule rather than
the exception in grasshopper species). Distal localization of chiasmata
is again important here, since the absence of proximal chiasmata pre-
125
thereby giving rise to aneuploid gametes. Orientation is more likely to
be regular, and hence fusions are more likely to establish themselves, in
species with a pronounced distal localization of chiasmata. The fact
that centric fusions only seem to have established themselves in about
5 out of 177 species of spiders which are known cytologically (Suzuki,
1954) is probably due to the prevalence of proximal chiasma localization in this group of animals (in most species of spiders all the chromosomes are acrocentric). Why so many species of spiders should have
the chiasmata localized in this way, at least in the males, is a question
that cannot be answered at present.
The second precondition for the "success" of a centric fusion is a
strictly genetic one, i.e., either an adaptive position effect or the bringing together in a linked combination of two or more interacting genes
or gene complexes (that position effects may sometimes be of importance in natural populations is indicated by the work of Levitan, 1954,
on Orosophila
robusta).
We are now in a position to attempt an answer to the question: why
do approximately 95% of the grasshopper species belonging to the subdivision Cryptosacci retain a karyotype consisting of 23 acrocentric elements (in the diploid set of the males)? And, in particular, why do
they differ in this respect from a number of other groups, such as the
crickets and the Australian eumastacid grasshoppers of the subfamily
Morabinae (White, unpublished data), in which a large number of
centric fusions seem to have established themselves?
Differences in chiasma localization may furnish a partial explanation.
Where, as in species of Chorthippus,
metacentric chromosomes which
have clearly arisen by centric fusion have nonlocalized chiasmata (or,
as in certain Morabinae, proximally localized ones) the pattern of chiasma
distribution has probably changed since the establishment of the fusion.
Even if the frequency and distribution of the chiasmata is favorable
to the establishment of centric fusions, the majority of those that occur
spontaneously are unlikely to succeed in evolution because they confer
no adaptive superiority on individuals heterozygous for them. Only
when the genetic contents of the two fusing elements interact to produce
a complex which forms a heterotic combination with the unfused homologues in the population, or with some of them, can we expect the
fusions to persist in the population (it does not seem imperative to suppose with Sheppard, 1953, that there is any necessary connection with
conspicuous visible polymorphisms such as those of Cepaea and the
grouse locusts, although such polymorphisms are the rule rather than
the exception in grasshopper species). Distal localization of chiasmata
is again important here, since the absence of proximal chiasmata pre-