Evolutionary Homoplasy among Species Flocks of Central Alpine Coregonus (Teleostei: Salmoniformes)

Evolution of phenotype and mode of speciation were examined for 19 Coregonus populations within and among eight lakes in the Central Alpine region of Europe. These populations reflect a mosaic of morphological, ecological, and life history traits, and thus represent numerous forms (qualitatively described according to relative body size and ecology). Each population had been previously evaluated for six microsatellite DNA loci, and herein for five meristic counts, four fin pigmentation characters, three body color variables, three measures relating to spawning ecology, an estimate of breeding tubercles, and average weight (= size). The two matrices (genotype vs. phenotype/ecology) were then tested for significance amongst themselves, and against evolutionary and geographic models. Microsatellite data associated significantly with a species flock model in which in situ diversification occurred from a common ancestor in each lake or cluster of neighboring lakes. However, phenotype/ecology associated significantly with a model invoking multiple invasions of lakes by pre-existing forms. The latter supports historic perspectives on the evolution of Central Alpine Coregonus and suggests forms within lakes have adapted to specific within-lake environmental niches that are replicated across lakes. This convergence of form and function has long clouded an understanding of coregonine biodiversity, and we suggest it represents a homoplasious condition (i.e., a similarity due not to genealogy but to iteration). This compendium of homoplasious and homologous characters is actually quite unique in nature, and the evolutionary diversification of Central Alpine Coregonus can now be evaluated according to the relative contribution of each character type.

Fig. 1.

Lakes in the Central Alpine region of Europe from which Coregonus populations were sampled. Names and locations of sampling sites are in Appendix 1

Fig. 2.

Unrooted, majority-rule consensus tree based on chord distances among 19 Coregonus populations from the Central Alpine region of Europe. Abbreviations for populations are listed in Table 1. Numbers at nodes indicate bootstrap percentages >50%

Fig. 3.

Minimum spanning tree connecting 19 populations of Coregonus from the Central Alpine region of Europe superimposed onto the first three axes of a principal components analysis of 20 phenotypic/ecological variables. Abbreviations for populations are listed in Appendix 1. Open circles are large forms, closed circles represent small forms, and those closed circles with an ‘x’ represent intermediate forms.

Fig. 4.

Minimum spanning tree connecting 20 phenotypic/ecological characters collected from 19 populations of Coregonus in the Central Alpine region of Europe. The tree is superimposed onto the first three axes of a principal components analysis. See text for character abbreviations.

Fig. 5.

UPGMA clustering of taxonomic distances separating 19 populations of Coregonus in the Central Alpine region of Europe. Distances were derived from a principal components analysis of 20 phenotypic/ ecological variables utilizing Burnaby's procedure for size removal. Abbreviations for populations are listed in Appendix 1

Contributor Notes

(MRD, MED) Department of Fishery & Wildlife Biology, Colorado State University, Ft. Collins, Colorado 80523–1474; (PCB) Institute of Plant Sciences/Phytopathology, Federal Institute of Technology, ETH-Zentrum, LFW Universitaetstr. 2, CH-8092 Zurich, Switzerland. marlis.douglas@colostate.edu Send reprint requests to MRD.

Accepted: 29 Dec 2004