The project was designed to study genetic variation in species of tropical rainforest trees. Shorea leprosula (an emergent species) and Xerospermum intermedium (an understorey species) were chosen, to see if they differed in breeding patterns. Genetic variation was studied by measurement of isozyme differences and also morphological variation. The isozyme patterns of 10 species of Shorea were compared with the indices of morphological similarity. The provisional estimate for proportion of polymorphic loci works out at 0.5--0.6 for Shorea leprosula, a high figure, and 0.3 for Xerospernrum intermedium. Both species are outbreeders. For Shorea leprosula, there is strong indication of spatial heterogeneity in isozyme frequency together with evidence that genetic similarity between populations is inversely related to the distance which separates them. In Xerospermum intermedium, variation in leaf size and morphology between individuals is highly significant and there is a distinct tendency for trees growing together to be more alike. Since both species are out breeders, this degree of spatial heterogeneity was unanticipated. It implies interruption of free gene exchange and this may help us to account for the origin of the great species diversity in the rainforest. Comparative isolation of populations, the means of fruit dispersal, the means of pollen distribution and possibly also the pattern of flowering time may jointly contribute to relative isolation. It appears that one of the major enigmas of evolution is at last accessible to systematic study.