Reciprocal transplant experiments have been made to investigate the .intra-specific variation in two clonal species, Primula vulgaris and RBDUDculus repens. Primula transplants performed best when returned to their native populations, indicating that they were differentia~ed in response to local conditions. There was marked variation in the degree of local specialisation of plants in different primrose populations and possible causes of this variation are discussed. Although buttercup transplants also showed great variability, there was no evidence that they were specialised, either between, or within, local populations. The lack of genetic specialisation in RBDUDculus repens may be due to its spreading growth form, widespread distribution and low level of seedling recruitment. In glasshouse experiments, the presence or absence of neighbours affected many parameters of buttercup growth. Within a genet the effect of edaphic and biotic heterogeneity was integrated, so that ramets in favourable conditions supported interconnected ramets in less favourable sites. Plants of R. repens vary phenotypically in different environments but appear to respond to heterogeneous local conditions by phenotypic plasticity of individual ramets rather than genetic specialisation. The assumption that differences between transplants are solely indicative of genetic specialisation has been questioned. Virus infection was detected in 7 of 14 primrose populations surveyed. Infected plants showed no symptoms of disease, yet they produced significantly fewer but larger leaves than uninfected plants. Differences between transplants which could easily be attributed to genetic variation may be due to differential virus infection. Furthermore, viruses may ultimately contribute to genetic differentiation and have a role as selective forces in the environment.. Phenotypic differences between ramets of the same genet of R. repens were maintained and even increased after 26 week's growth in a cammon environment. It is clearly imPortant in transplant experiments to use comparable phenotypes and virus-free plants when determining the role of genotype in the match between organism and environment.