Document Type


Publication details

Robson, BA 2002, 'Chromosomal speciation via monobrachial homology in native Australian Rattus species', PhD thesis, Southern Cross University, Lismore, NSW.

Copyright BA Robson 2002


Despite assertions to the contrary in the literature, proven cases of "chromosomal speciation" (ie. speciation directly due to chromosomal rearrangements) are difficult to find in animals.

The reasons are two-fold:-

First, any chromosomal rearrangement differentiating species, that in turn causes post-mating isolation as hybrid sterility, must initially go through a heterozygous phase and survive. Theoretical models (Lande, 1979) suggest that the probability of such occurring is very small.

Secondly, where species differ by chromosomal rearrangements that cause meiotic problems, it can be very difficult to determine whether the chromosomal rearrangements caused the speciation event, or occurred subsequent to the speciation event.

Baker and Bickham (1986 ) proposed a model in which chromosomal rearrangements that do not cause meiotic problems in the heterozygous state, might nevertheless act as a post-mating isolating mechanism. viz - monobrachial homology. In this model, two populations become fixed for centric fusions with only one chromosome in common ("Monobrachial Homology"). Centric fusions per se appear in many cases to not reduce fertility in the heterozygote. However, mice (Mus) heterozygous for chromosomes showing monobrachial homology show severe reduction in fertility. Baker and Bickham (1986) therefore proposed that chromosomal rearrangements leading to monobrachial homology could, in and of themselves, cause speciation.

The second problem of actually proving such cases exist remains problematical. One would expect to detect such a speciation event only in very recently derived species, where the appropriate chromosomal changes have occurred, but there has been insufficient time for other genetic differences to accumulate to observe the actual cause of speciation.

This thesis presents convincing evidence that speciation has occurred by monobrachial homology in a group of Australian Rattus - the R.sordidus complex.

The genus Rattus is represented in Australia by eight endemic species. Many detailed studies have been carried out on these rats, including morphological, ecological, and genetic studies. Four of the species form a very closely related clade, the sordidus species complex. The member of this complex are morphologically similar and allozyme studies show them to be very closely related genetically, to the point where R. villosissimus, R. colletti and R.sp. nov. share at alleles 55 allozyme loci (Baverstock and Adams, unpublished), while R. sordidus differed at only two loci (Baverstock and Adams, 1987). However, each species has a completely unique chromosomal karyotype, the differences consisting mostly of monobrachial homology.

Previously unpublished laboratory hybridisation data are reported and analysed. These data demonstrate that crosses between members of the group are fully fertile, but the resulting F1, hybrids suffer very reduced fertility, as expected from meiotic configurations in the hybrids.

Genetic differentiation among members of the R.sordidus group was assessed and compared to genetic differentiation among subspecies of closely related Australian species using the fastest evolving gene known in vertebrates - the mitochondrial control region. Analysis of nucleotide variation in a 355 base pair fragment of the control region was assessed using PCR, TGGE, and DNA sequencing. Members of the R.sordidus group were remarkably similar for this region. Indeed, sequence variation across all four members of the R.sordidus group was less than that found between subspecies of the other four species of Australian Rattus. Most striking was the comparison between R.villosissimus and R.sp.nov. These two species exhibited only 2 mitochondrial haplotypes each, one common and one rare. The common haplotype in each species was shared between the two species, therefore these species share 50% of their alleles. Furthermore, in each of these species, the rare allele differed by only a single base change from the common allele.

The results very strongly support the hypothesis of chromosomal speciation in the sordidus species complex by way of rearrangements resulting in monobrachial homology.

The speciation appears to have been very recent. The sequence divergence rate of the control region was calculated based on the divergence rates from a number of other species. Using these rates and the dating of the geological isolation of South Australian island populations of R. fuscipes greyii, a molecular clock was calibrated. The speciation of the sordidus species complex was determined to be within the last 200,000-600,000 years.