Describing and predicting potential growth in the pig

Most animal growth models contain an explicit growth function. It determines the pattern of growth over the lifetime of the animal and defines an upper limit to growth rate ( the potential). The criterion of the 'goodness-of-fit' to one or more sets of data is frequently used to select a suitable growth function. Alternative criteria are described here that can be used to choose between forms that describe potential growth. Of the functions reviewed only a few fulfilled all of the proposed criteria. Of these the Logistic and Gompertz functions were favoured because of an economy of parameters and their ability to describe relative growth rate as a simple function of size. The Logistic function was rejected on the grounds of its numerical consequences for growth in pigs over a wide range of degrees of maturity, leaving the Gompertz function to be tested for its ability to make sensible predictions of potential growth. Pre-natal growth data, assumed to occur under non-limiting conditions as long as the mother is not subjected to extremely adverse nutritional conditions or incidence of infection, were used to estimate the values of the two Gompertz function parameters-the growth coefficient and the initial condition-given an estimate of mature size. The values were comparable with literature estimates based on post-natal growth and predictions of growth rate over a wide range of degree of maturity were thus sensible. On these grounds, and because it uses few parameters all with biological meaning, the Gompertz function is proposed as a suitable descriptor of potential growth.