Richness values computed in each sampling site were rarefied
to remove the effect of sample size according to the individuals
that were disregarded (Gotelli & Colwell 2001) using the
software PAST (Hammer et al. 2001). Next, the rarefied richness
values computed for the different stand types were compared
by analyses of variance (ANOVA) with permutation tests
(Pillar & Orlóci 1996) The test criterion was the sum of square
Euclidian distances between groups of sampling units. The
same test was used to check for differences in the density of
native plants and frequency of zoochorous diaspores between
the plantations and the native forest.