The pattern of mtDNA variation and geographic distribution indicate an initial population expansion, probably immediately after divergence from the wolf-like ancestor, around 100,000 years ago. The partition of mtDNA haplotypes that followed was most likely the result of habitat reduction and fragmentation at the onset of the deglaciation approximately 15,000 years ago. Our samples confirmed very low levels of genetic diversity but it appears that the impact of historical habitat fragmentation on Ethiopian wolf gene flow did not occur long enough ago for a significant reduction in diversity of the mitochondrial genome .

The genetic structuring of Ethiopian wolves today is the result of processes acting over the last 18,000 years, leading to their final retreat into modern mountain refuges. The species however maintained overall genetic diversity, as alleles that might be lost in one population could be fixed in another, preserving more heterozygosity than a simple population of similar total size. The most likely genetic partitioning corresponds to three mountain areas: Arsi/Bale (in the South), Wollo/Shoa and Simien/Mt.Guna, a pattern that closely matches the geographic history of Afroalpine fragmentation. Although there is a degree of clustering of haplotypes from both sides of the Rift Valley, the lack of reciprocal monophyly does not support the taxonomic classification of two subspecies

This study highlighted the importance of populations north of the Rift Valley for the maintenance of genetic variability within the species. The populations north of the Rift Valley act as the main reservoir of genetic variability in the species, largely exceeding the variability of the southern populations. Conservation management plans should urgently focus on these areas, where human pressure on remaining habitats has also reached critical levels. Management to restore or maintain variation in genetically at risk populations could be effective and still maintain this pattern by using transplants only among contiguous regions.

Fourteen microsatellite loci were also used to characterize patterns of genetic diversity. The genealogical history of alleles in the populations was best explained by an equilibrium between gene flow and drift. In spite of indications of some gene flow among clusters of populations in the past, it might be desirable to maintain the current genetic structuring of three mountain blocks. Pragmatically, gene flow wihtin mountain blocks could be artificially restored without a severe risk of genetic swamping. While we may not able to recognize whether populations within the identified mountain blocks actually constitute ‘evolutionary significant units’ (populations with independent evolutionary histories), they certainly are useful management units for restoration processes, as amply supported by genealogy and genetic differentiation.

Relevant publications

Gottelli D, Sillero-Zubiri C, Applebaum GD, Girma D, Roy M, Garcia Moreno J, Ostrander E, Wayne RK (1994) Molecular genetics of the most endangered canid: the Ethiopian wolf, Canis simensis. Molecular Ecology 3:301-312.

Dalton , R., Kitchener, A., Sillero-Zubiri, C. (2003) Craniometry for conservation – The skull morphology of the endangered Ethiopian wolf. Unpublished MSC Thesis. Oxford University.

Gottelli D, Marino J, Sillero-Zubiri C, Funk S (2004) The effect of the last glacial age on speciation and population genetic structure of the endangered Ethiopian wolf (Canis simensis). Molecular Ecology 13:2275-2286

Gottelli, D; Wang1, J; Marino, J.; Sillero-Zubiri, C. and Funk S.M. (submitted) Integrating molecular genetic structure to the restoration process of the endangered Ethiopian wolf.