SIMPLIFICATION OF MILK RECORDING SCHEME IN LATXA MILKING SHEEP

Simplified methods of estimating milk yield and composition were evaluated on 1021 lactations of Latxa ewes. The standard method for recording milk production and composition in dairy cattle (monthly recording of two daily milkings) was used as a reference method to evaluate the accuracy and precision of the simplified methods studied. Five methods for simplification of quantitative and qualitative milk recording were examined: the 'classical' methods (recording every month) and three 'punctual' methods (recording only twice during lactation). The two classical methods (i.e. milk yield) were (1) a.m. recording only (AFM method), and (2) a.m. and p.m. recording alternately (ATM method). Punctual methods (i.e. milk composition) were (1) first- and second-month recordings (P12), (2) first- and third-month recordings (P13), and (3) second- and third-month recordings (P23). The main evaluation criterion used was the 'variance of error' (VE: ratio variance of the differences between actual and estimated production divided-by variance of actual production). Among classical methods, ATM performed the best (VE less-than-or-equal-to 3%) while P13 show the highest accuracy among punctual methods (VE, 5-30%). A recording program including an ATM scheme for milk yield and a P13 punctual scheme for milk composition is recommended for Latxa milking sheep.