Isozyme variation and recent biogeographical history of the long-lived conifer Fitzroya cupressoides

Authors


Andrea C. Premoli, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, Quintral 1250, 8400 Bariloche, Argentina. E-mail: apremoli@crub.uncoma.edu.ar

Abstract

Aim Palaeoenvironmental records of Pleistocene glaciation and associated vegetation changes in Patagonia have led to the hypothesis that during the last glacial maximum (LGM) tree species survived locally in favourable habitats. If present populations originated from spread from only one refugium, such as an ice-free area of coastal Chile (Single Refugium hypothesis), we would expect that eastern populations would be genetically depauperate and highly similar to western populations. In contrast, if the ice cap was not complete and tree species persisted in forest patches on both slopes of the Andes (Multiple Refugia hypothesis), we would expect a greater degree of genetic divergence between populations either on opposite sides of the Cordillera (Cordillera Effect scenario) or towards its present-day southern distributional limit where the ice sheet reached its maximum coverage (Extent-of-the-Ice scenario).

Location We tested this refugia hypothesis using patterns of isozyme variation in populations sampled over the entire modern range of the endemic conifer Fitzroya cupressoides (Mol.) Johnst. (Cupressaceae) in temperate South America.

Methods Fresh foliage was collected from twenty-four populations and analysed by horizontal electrophoresis on starch gels.

Results Twenty-one putative loci were reliably scored and 52% were polymorphic in at least one population. Populations from the eastern slope of the Andes were genetically more variable than those from the western slope; the former had a greater mean number of alleles per locus, a larger total number of alleles and rare alleles, and higher polymorphism. Genetic identities within western populations were greater than within eastern populations. Discriminant analyses using allelic frequencies of different grouping schedules of populations were non significant when testing for the Single Refugium hypothesis whereas significant results were obtained for the Multiple Refugia hypothesis.

Main conclusions Our results indicate that present Fitzroya populations are the result of spreading from at least two, but possibly more, glacial refugia located in Coastal Chile and on the southern flanks of the Andes in Argentina.

Ancillary