Montane vegetation of the Mt. Field massif,Tasmania: a test of some hypotheses about properties of community patterns

direct gradient analysis was applied to the montane vegetation of the Mt. Field massif, Tasmania. Ecological response surface were constructed, describing the relationship between the mean % cover of each of 100 vascular plant species and two major environmental complex-gradients represented by soil drainage and altitude. The hypotheses tested were that: (1) the ecological responses of species are generally of Gaussian form; (2) the modes of minor species have a uniform random distribution along gradients; (3) the modes of major species are evenly distributed; (4) the frequency distribution over species of modal abundance is either lograndom or lognormal; and (5) alpha diversity has a unimodal trend along environmental gradients. The hypotheses were tested both for the full site of vascular species and for each of five species groups defined by structural form. Hypothesis (1) was rejected: only 45% of species had response surface which appeared unimodal and symmetric. Hypotheses (2) and (4) were rejected for the full set of species, but each was supported for all but one of the structural groups. The modes of herb species were clumped, rather than random and the frequency distribution of modal percentage cover for shrubs was inconsistent with both lograndom and lognormal hypotheses. Contrary to hypothesis (3), the modes of major species were randomly distributed. Although total alpha diversity had a complex trend surface, the patterns for the alpha diversities of each structural group were unimodal, in accordance with hypothesis (5). The results suggest that an adequate model of community variation along environmental gradients must take into account differences in response patterns between species groups.