Some comments on the shape of the gain–stocking rate curve

Abstract

SUMMARYIn this paper the method proposed by Jones & Sandland (1974) for testing the linearity of the animal gain–stocking rate relationship is examined. It is shown that the method cannot be used to establish such linearity. An examination of the biases in estimation of optimum stocking rate and gain per animal and per unit area at the optimum stocking rate arising out of assuming linearity where the true relationship was quadratic led to the following conclusions. Quite large biases can arise. Selection of stocking rates straddling the true optimum reduces these biases. If experimental stocking rates are lower than the true optimum stocking rate the optimum stocking rate is overestimated and the animal gain at optimum is underestimated and vice versa if the experimental stocking rates exceed the true optimum rate.