Transfer of N from Phaseolus bean to intercropped maize was studied in glasshouse experiments using N-foliar-15 feeding and N-isotope-15 dilution methods. Nodulated and non-nodulating Phaseolus genotypes were included in separate treatments to help distinguish between benefits due to transfer of fixed N and competition for N in the growth medium. When intercropped with bean foliarly fed with ((NH4)-N-15)2SO4, maize was enriched with N-15, showing that N had been transferred. The amounts of N transferred were small, and always < 5% of the N in the N2-fixing beans. There was a decrease in shoot-N in maize intercropped with N2-fixing bean compared to maize intercropped with the non-nodulating beans. Non-nodulating bean transferred comparable amounts of N to intercropped maize plants although their total N content was less than a quarter of that in the N2-fixing beans. For the isotope dilution experiments, N-15-fertilizer was incorporated into a soil-based compost together with sucrose to stabilise the N-15-enrichment of available N. When plants grew vigorously no transfer of N from bean to maize was detected by isotope dilution, and again shoot N of maize intercropped with N2-fixing beans was less than that of maize with non-nodulating beans. In a further experiment, growth of maize and bean plants was reduced by severe insect attack and up to 15% (between 9 and 15 mg N pot-1) of the N in N2-fixing beans was estimated by isotope dilution to have been transferred. Small (5-10 mg N pot-1) but significant increases in total N yield were found in the maize intercropped with N2-fixing bean compared to maize intercropped with non-nodulating bean. In this experiment treatments with or without vesicular-arbuscular mycorrhiza were established but showed no significant differences in N-transfer from unioculated plants. As transfer of N from the beans to intercropped cereals showed such little benefit under conditions of severe N limitation, our results indicate that many careful field experiments are required before we can conclude that N-transfer from Phaseolus to intercropped cereals is significant in agriculture.