Compétition et association dans les biocénoses de fourmis insulaires

Abstract

In a recent paper (Baroni Urbani, 1971), the author has studied some patterns of arrival of the propagulae on islands. This study, however, does not represent the solution of all the problems involved in settlement of island biotas because we can suppose that several species arriving on a given island will have no or very little chance of survival. In the literature we mostly find a series of conclusions empirically drawn from field data and empahsizing the competition between closely related species as the only explanation of the greater taxonomic diversity between insular biocenoses in comparison with those of the source area. On the other hand, a few theorethical works claim that the rduced ratio of species per genus of insular communities must be considered only as a trivial statistical property of random samples, or even that the species/genus ratio on natural islands is larger thant the one would expect for theoretical ones, because of the greater similarity of related species in colonisation patterns and ecological preferences. In the present paper the author offers a statistical analysis of the degree of coexistence of the ant species of the Tuscan Archipelago (populated mostly by natural ways) and of the small islands of Polynesia (probably completely introduced by man). The results for these two extremely dissimilar situations are quite different, but, in general, competition appears always much more common than association between species. In fact, in the Tuscan Archipelago, statistically significant competition is shown by 7.3% and in the small polynesian islands by 22.1 % of the possible species pairs, while association is shown in the two groups of islands by 2.9 and 0.4% of the species pairs respectively. Concerning intrageneric competition and association, we have ascertained the presence of both, but in different proportions. In the Tuscan Archipelago intrageneric competition represents 0.3 % and and intargeneric association 0.2 % of the observed pairs, while in the small polynesian islands 5.8% of the couples show intrageneric competition and none of them shows intrageneric association. It is interesting to note, moreover, that the ants of the small Polynesian islands have a U-shaped species/genus ratio distribution. If we can consider this sample as representative, we must expect that the mean species/genus ration will remain constant in any other sample regardless of the size. It is evident from these data that synecological factors play a great role in determining the structure of insular biotas, but intrageneric association, although present, as supposed by Simberloff (1970), cannot have a big influence in determining the faunal structure. (abstract provided by author)