In this work I have developed a chromosomal speciation model for animals of limited vagility or subdivided population structure which combines and extends aspects of Mayr's (1942, 1954, 1963) genetic revolution, the chromosomal fixation model of Wright (1941), and the chromosomal speciation models of White (1968, 1969; White et al., 1967) and Key (1968). The theory suggests that speciation may occur in certain species under a variety of geographic conditions without the requirement for prolonged or absolute allopatric isolation if these species have suitable population structures and genetic systems. Basic requirements for chromosomal speciation are:
Given the initial fixation event in a non-optimal environment for the ancestral stock (which may likely be associated with a Mayrian genetic revolution, since the conditions required for the fixation of the chromosomal difference are also ideal for the revolution), the chromosomally differentiated local population will occasionally be endowed with a superior local adaptation. If this happens, the
chromosomally differentiated population may survive its early interactions with the ancestral stock. Then the chromosomal semisterility of the heterozygotes will
If the chromosomally derived population has achieved a superior adaptation and/or is assisted by chance events so that it can grow large enough to form a defined zone of parapatric hybridization with the parental stock (assuming that premating isolation has not been evolved), the reduced fecundity of chromosomal heterozygotes in the hybrid zone will lower the population pressure in the zone, thereby encouraging net migration--and hence gene flow--into the hybrid zone. This will add considerably to the effectiveness of the heterozygote semisterility as a barrier to gene flow into the "pure" population beyond the hybrid zone, because the zone will function as a "sink" to trap, even many of those genes which may once or twice successfully introgress into chromosomal homozygotes. Once the "sink" effect becomes operative, ^t would also tend to prevent the further evolution of premating isolation by the parapatrically hybridizing populations by trapping genes that were only partially effective. Therefore, notwithstanding the lack of effective premating isolation or complete hybrid sterility, the populations removed from the hybrid zone could be sufficiently isolated genetically to evolve independently as good biological species.
Since
such chromosomal speciation may initiate a cascading process which will amplify the probability that derived species will in turn give rise to even more derived species. Because of this amplification effect and the fact that the first species to be derived from an ancestral stock is likely to fill the ecological niches adjacent to the ancestral stock, either by itself or by its own further derivatives, before the ancestral stock can give rise to a second derived species (= "lateral inhibition"), cascades of species will tend to show sequences of karyotypic derivation that are more linear than fan-like. Also, because of the genetic inertias of the larger ancestral populations, the karyotypically derived species will either be ecologically derived or specialized with respect to its ancestral stocks.
Karyotypically diverse rodent radiations were reviewed which seem to exemplify the predicted phenomena of the chromosomal speciation theory (Perognathus: Heteromyidae, Spalax: Spalacidae, and Sigmodon: Cricetidae, among the best known), and in which these phenomena cannot be readily explained by other cytogenetic mechanisms. These were compared with a karyotypically diverse radiation (Peromyscus: Cricetidae) which can be readily accounted for by other mechanisms. Unfortunately, the published information on these radiations is too limited to provide more than suggestive evidence that they have resulted from chromosomal speciation as described in the present model.
However, the much more detailed analysis of the karyotypic variation and cryptic speciation in the Sceloporus grammicus complex, which is presented in the present report, exemplifies almost all of the phylogenetic and phenomenological predictions of the chromosomal speciation model; and, unlike the rodent radiations discussed that lack good comparative backgrounds against which the evolutionary significance of their karyotypic variability can be judged, the radiation of the other crevice-using Sceloporus provides an ideal background for comparison. As far as can be determined from the present speciation in the grammicus complex seems to be strictly associated with chromosomal differentiation, And, in comparison to the rock-crevice-using torquatus group, which has radiated in a habitat which provides ample opportunity for geographic speciation, and which is diverse enough to allow size differentiation for sympatry, the grammicus complex has proliferated almost as many species in its much more limited and geographically continuous habitat. Relative to the torquatus radiation, it would then seem that the grammicus have proliferated many more species than would be expected if speciation could occur only by the classical allopatric models. Also, data to be presented in later reports indicate that similar but older radiations of chromosomal speciation have occurred in other branches of Sceloporus, which in large part account for the great species diversity of Sceloporus in comparison to the chromosomally conservative and conservatively speciating remaining sceloporines. Clearly, much work still needs to be done before the details of all of the speciation in Sceloporus are fully understood, but at least the present data seem to provide more than suggestive evidence for some mode of chromosomal speciation that depends more on intrinsic characteristics of the species than does classical allopatric speciation. However, as is true for most evolutionary studies, the evidence does not conclusively prove that the speciation has occurred exactly as I have reconstructed it.