Plot-Less Tests of Interspecific Association

Abstract

Two methods are described in which interspecific association is tested by measurement of distances between individuals. In the first method, pairs of individuals are counted in which each member of the pair is, for its partner, the nearest neighbor of another species. One may then test whether any particular pair of species occurs in the records more or less frequently than would be expected on a basis of chance association. In the second method, the distance form a random point to the nearest individual of each species is compared with the distance from an individual of one species to the nearest individual of the other. The methods are exemplified by application to data from jar rah (Eucalyptus marginata) forest and mulga (Acacia aneura) woodland in Western Australia. In the latter vegetation it is demonstrated that there is a marked association between shrubs and trees in general rather than between particular pairs of species. It appears that many shrub species establish themselves much more readily in the proximity of a nurse plant. This is ascribed to protection from grazing in the early stages of development.