15N labelling and variations in natural 15N abundance offer possibilities to quantify the importance of mycorrhizas in plant N uptake. When atom %15N or 15N natural abundance of mycorrhizal and non-mycorrhizal plants are compared in experiments, one, however, has to account for dilution effects of seed N and the interaction between changes in sslabelling of the available pool and N uptake. This has been considered in studies of N2 by diazotrophs, but has not been explicitly corrected for in studies of N uptake by mycorrhizal plants. An example is given, where ectomycorrhizal plants obtained an atom %15N different (P < 0.01) from that of non-mycorrhizal plants, but where this could be ascribed to dilution effects of seed N and differences in N uptake rates interacting with changes in 15N labelling, rather than to variations in use of N sources or to isotope fractionation during uptake. Careful N and l5N budgets based on seed N and sequential harvests, as well as frequent measurements of atom % or abundance of 15N in available pools of N, are thus necessary in studies of this kind. Regressions of 15N on total N are a useful alternative to comparisons of means of atom % only.