Plants of Lolium perenne and Festuca rubra were grown in sand culture receiving all nutrients as a complete nutrient solution containing 1.5 mM NH4NO3, and subjected to one of three defoliation treatments: undefoliated, defoliated on one occasion, or defoliated weekly. N-15 labelling was used to determine the rate of N uptake, allowing the amount of N remobilized from storage for the growth of the two youngest leaves (subsequently referred to as 'new leaves') growing over a 14 d period after defoliation to be calculated. The total plant N uptake by both species was reduced, compared with undefoliated plants, by both a single and repeated defoliation, although neither caused complete inhibition of uptake. Regularly defoliated L. perenne had a greater reduction in root mass, concomitant with a greater increase in N uptake per g root than did regularly defoliated F. rubra. In both species, the amount of N derived from uptake recovered in the new leaves was unaffected by the frequency of defoliation. Both L. perenne and F. rubra mobilized nitrogen to the new leaves after a single defoliation, mobilization being sufficient to supply 50 and 41%, respectively, of the total nitrogen requirement. In regularly defoliated plants, no significant nitrogen was mobilized to the new leaves in L. perenne, and only a small amount was mobilized in F. rubra. Plants achieved greater leaf regrowth when only defoliated once. We conclude that increasing the frequency of defoliation of both L. perenne and F. rubra had little effect on the uptake of nitrogen by roots which was subsequently supplied to new leaves, but depleted their capacity for nitrogen remobilization, resulting in a reduction in the rare of growth of new leaves. (C) 1997 Annals of Botany Company.