Fertilizing programs

Alternative programs for applications of fertilizer phosphorus to successive crops are computed from the data of long term experiments showing the effects of fresh fertilizer applications and of residues from applications to previous crops on the yields of wheat in Australia and of soybeans in Brazil. The programs are compared on the basis of total economic returns with allowance for the time scale for investments in fertilizer and returns from the successive crops. Gains from the use of optimal programs may be small relative to those obtained with normal farm practice for conditions of moderate fertilizer requirement as represented by the Australian experiments, but may be appreciable for soils with high requirements as in the Cerrados region of Brazil. Estimates of optimal programs may be much affected by variations in yield levels and yield responses to fertilizer due to variations in year to year growing conditions, and it is concluded that long term fertilizer experiments are not likely to provide practicable bases for estimating programs for use by farmers. There is need to develop alternative procedures for estimating optimal programs such as might be based on effects of fertilizer on soil composition.