Genetic and phenotypic consequences of translocations between deer (genus Cervus) in Scotland

Abstract

Asian sika deer (Cervus nippon Temminick, 1836) were introduced across Scotland at the tum of the last century and have since established several large feral populations. Their potential to cause economic damage to forestry and hybridise with native red deer (C. elaphus L. 1758) has provoked widespread concern and highlighted a lack of basic ecological and genetic data relevant to their management. A preliminary genetic investigation of two Scottish populations (Great Glen & Argyll) revealed introgression at both sites. The aim of this thesis is to expand and develop this work by investigating the genetic and phenotypic causes and consequences of hybridisation across Scotland.

Tissue samples and cull data were collected from 670 red deer, 266 sika and 8 phenotypic hybrids originating from 9 areas around Scotland with documented sika introductions. The samples were genotyped for 10 micro satellite and one mitochondrial DNA (mtDNA) marker loci. Skeletal material and genetic samples were collected from a further 339 red and 307 sika from Argyll in an intensive study involving an additional 15 micro satellite markers. All markers used were putatively diagnostic, though the interpretation of some rare alleles was problematic.

Samples with several introgressed alleles at different loci provided strong evidence for hybridisation since introduction in Argyll, Galloway and Sutherland and Ross-shire, but different patterns of introgression emerged. In Argyll, hybridisation is rare but followed by repeated backcrossing to give substantial introgression, with 62% of sika-like deer and 33% of red-like deer containing opposite-taxon alleles. Evidence of recent hybridisation was restricted to sample sites at the leading edge of sika colonisation. In Galloway a number of early-generation hybrids were identified. In Sutherland and Ross-shire, introgression was common (>75% of individuals) in sika-like deer at 5 of the 10 micro satellite loci but almost absent at the other loci, suggesting a hybridisation event( s) several generations ago followed by backcrossing and strong assortative mating. Introgression was uncommon in the remaining sika-like populations (<20% of individuals) and rare in red-like populations «5%), but most red-like populations contained rare alleles whose origins were difficult to interpret. These alleles may represent polymorphism, introgression from unreported sika introductions or introgression from introduced wapiti (c. canadensis L. 1758). Across all samples, mtDNA introgression was rare and restricted to Argyll sika-like deer having red mtDNA. Furthermore, all recently generated hybrids contained red mtDNA, suggesting hybridisation events usually involve sika males covering red females.

Genetic distance measures were calculated for populations following the removal of introgressed alleles. Neighbour-joining trees revealed no consistent pattern between hybridising and non-hybridising populations. The Borders sika were confirmed as the most genetically differentiated sika population in Scotland.

The identification of hybrids by stalkers was generally poor and restricted to the most recent backcross offspring. Only 20/0 of the genetically identified hybrids in Argyll showed any obvious visual sign of introgression and around the rest of Scotland, only one sample was correctly identified by the stalker as a hybrid. Detailed measurements of several phenotypic traits confirmed this pattern and after several generations of backcrossing, hybrids and non-hybrids could not be differentiated reliably, even though there were significant increases in some sika-like hybrid measures.

Hybridisation has occurred, and continues to occur, between red and sika across Scotland. The existence of hybrids in an area did not correlate well with the physical size of local red-like and sika-like deer. Similarly, there was no evidence that particular genetic strains of red or sika were more likely to be involved in hybridisation. However, there does seem to be an association with the colonisation process of sika stags. Why this has happened in some cases but not all remains unclear and may be associated with forest design. Nevertheless, there is good evidence that strong assortative mating preferences exist where mate choice is available and that backcrossed offspring quickly attain the parent taxa phenotypes. Under these circumstances it will be difficult to prevent the introgression of opposite taxa alleles by selective shooting.