Published online by Cambridge University Press: 27 March 2009
The results are reported for an experiment in which nitrogen rates from 0 to 672 kg/ ha and potassium rates from 0 to 558 kg/ha were applied annually on an S. 24 perennial ryegrass sward. At each of the five crops cut annually the yields of herbage dry matter and crude protein, and the potassium content of the herbage dry matter were determined. The potassium content of the soil was also measured annually. Models relating each of these variates to the nitrogen and potassium rates were derived, and their validity is demonstrated with reference to the results from earlier experiments. The problems associated with determining the correct rate of potassium fertilizer to apply on swards receiving given rates of nitrogen fertilizer are discussed using the models to predict the effects of different nitrogen-potassium combinations. In order to relate the implications of this discussion more closely to practical grassland management, economic as well as biological criteria are taken into account by incorporating cost factors in the models.