LAUSR

Abstract Body weights (BW) of 2,611 Muzaffarnagari lambs at birth and 3, 6, 9, and 12 months of age were used to compare five alternative growth curves. Four three-parameter functions (Brody, von Bertalanffy, Gompertz, and logistic) and the four-parameter Richards function were fit for each lamb. Estimates of the asymptotic final BW (A), a maturing-rate parameter (k), and the degree of maturity at birth (u0) = BW/A were estimated for each function. A shape parameter (m) was included in the Richards function and used to identify the degree of maturity (u) at the point of inflection of the growth curve (uI). Among three-parameter functions, the Brody function had the smallest pooled residual sum of squares (RSS) across all lambs and the smallest RSS for 56% of individual lambs. For the Brody function, the growth rate is maximum at birth and declines continuously as animals age. In comparison with the Brody function, fitting the Richards function with a variable inflection point did not significantly reduce RSS. Comparisons of Brody and Richards functions indicated that the Richards function converged to a Brody function for 17% of the lambs. The predicted growth rate was maximum at an average uI of 0.26 for the Richards function and at an average of u0 = 0.10 for the Brody function. Parameters k and m interacted to define the maturing rate in the Richards function, and the predicted age at u= 0.50 (t50) was proposed to measure growth efficiency. Correlations between estimates of A, u0, and t50 for the Brody and Richards functions were all 0.89, confirming limited benefit from fitting the Richards function. For the Brody function, correlations among estimates of A, k, and u0 were modest (-0.59 to 0.39). Use of these parameter estimates in selection to modify the growth curve would be relatively straightforward. Correlations among A, k, and u0 from the Richards function were also modest (-0.60 to 0.27), but the correlation between k and uI was 0.82. Selection to modify the Richards parameters would therefore be more challenging, and the Richards function likely over-parameterized the growth curve. We therefore concluded that the Brody function was the most appropriate growth function to describe BW changes from birth to 1 year of age in Muzaffarnagari lambs.