Hypotheses to explain assortative pairing in gammarids (Amphipoda) include size-related habitat segregation and male-male competition. We investigated the influence of micro-habitat heterogeneity on size-assortative pairing in Gammarus pulex, using field-based studies and complementary laboratory-based studies. Firstly, we studied the distribution of males and females of different size and the intensity of size-assortative pairing in micro-habitats differing in particle size at three different study sites. In one site both males and females distributed among micro-habitats in relation to their size, with larger individuals in vegetation and smaller individuals in fine gravel. In a second site, there was no difference in the size of both males and females between two micro-habitats of different particle size, whereas in a third site with two micro-habitats of different particle size the difference was significant for males, but not for females. However, size-assortative pairing was significant within each micro-habitat in each study site. In the first site, pooling data from the three microhabitat subsamples resulted in a higher correlation coefficient between male size and female size, compared to the ones obtained within each micro-habitat, whereas no such difference was observed for the two other sites. Secondly, in experimental conditions, we compared the strength of size-assortative pairing under combinations of both heterogeneous and homogeneous environmental conditions and male-biased or female-biased sex ratios. Size assortative pairing was significant only when the sex-ratio was male-biased. Environmental heterogeneity had no effect on size-assortative pairing. We conclude that male: male competition is the major determinant of size-assortative pairing in G. pulex and that micro-habitat segregation plays only a marginal, if any, role.