Relationships were examined between relative yields of pasture with respect to phosphorus (P) and sulphur (S) applications, and clover herbage P and S concentrations and P/nitrogen (N) and S/N ratios in a P x S factorial fertiliser trial on a mown clover/ryegrass sward. The relationships between the yield of a treatment relative to the maximum yield obtainable with high rates of P alone (RY[P]) and both % P and P/N were affected by the application rate of S. S deficiency generally lowered the % P required for any given RY[P] but raised the P/N required. This latter effect was because S deficiency markedly lowered clover herbage N concentration. P deficiency had a similar but smaller effect on % S and S/N required for any RY [S]. S/N was much less sensitive to changes in clover S status than P/N was to changes in clover P status. This resulted from the strong stimulatory effect of S on clover % N. Relationships were also established between clover P and S adequacy indices and RY [PS] which is the herbage yield with P and S in physiological balance relative to the maximum yield with P and S in physiological balance. These relationships differed from those involving RY[P] and RY [S], requiring a higher value of the adequacy index for any specified RY. Nutrient concentrations (particularly of S and N) and ratios changed during a year, indicating that more than one sampling per year would be A96068 Received 16 September 1996; accepted 6 December 1996 necessary to determine whether the current fertiliser policy was succeeding in maintaining adequate and balanced nutrition throughout the year.