Is the hypothesis of survival on glacial nunataks necessary to explain the present-day distributions of Norwegian mountain plants?
Birks, H. J. B.
The occurrence and relative frequency of 109 species of mountain plant in Norway in 75 grid squares were related statistically to 11 explanatory variables representing modern geography, topography, bedrock geology, and climate and, as a historical variable, the relative abundance of unglaciated areas. The numerical analyses involved (partial) constrained ordinations and associated Monte Carlo permutation tests. They were designed to answer the following questions: (1) Are any of the 11 explanatory variables statistically significant in explaining the distribution patterns of different species groups?, (2) How much of the variance in the species data is explained by geography and topography, climate, geology, and unglaciated areas?, (3) Do unglaciated areas make a unique statistically significant contribution to explaining the species group distributions?, and (4) Are unglaciated areas a statistically significant variable in explaining the observed distribution and relative frequency of any of the 109 individual species? The numerical results all suggest that there is no statistically significant contribution from unglaciated areas in explaining the present-day distribution patterns when the effects of modern topography, climate, and geology are considered first. There is thus no need to invoke the nunatak hypothesis of glacial survival to explain contemporary distributions of Norwegian mountain plants. The nunatak hypothesis appears, in light of these statistical results and recent palaeobotanical, biosystematical, and evolutionary studies, to be superfluous.