Comparison of four techniques to estimate milk production in singleton-rearing non-dairy ewes

Accurate estimates of milk production or milk intake are difficult, as all methods interfere to some degree with the natural behaviour of the dam and her young, and potentially alter milk yield itself. The present study compared milk yield obtained by the “oxytocin” method, udder dimensions (UD), the isotope dilution method, and live weight change of the lamb, in an attempt to select the most accurate and convenient way of measuring milk production in non-dairy sheep. In addition, the study investigated which of the three milk-estimation techniques was an accurate predictor of growth rates of lambs. Thirty-seven singleton-bearing and rearing ewes were milked once a week, for seven consecutive weeks, using the “oxytocin” method. Prior to each afternoon milking, the external dimensions of the ewe's udder were measured. Lambs were weighed weekly for the first seven weeks of life and live weight change was calculated. The deuterium oxide (D2O) dilution technique was used to estimate milk intake of the lambs and performed at approximately 7 days post-partum and finishing on approximately day 14. Pearson's correlation coefficients and multiple regression coefficients among techniques were calculated. The UD-models at d7 (R2 = 0.35), d35 (R2 = 0.36) and d42 (R2 = 0.34), were the best models explaining variation in milk yield (concordance correlation coefficient (CCC) = 0.49; 0.53; 0.51; for d7, d35 and d42, respectively). The lamb live weight-change model explained the variation in milk yield best at d28 (R2 = 0.32; CCC = 0.49), at d35 (R2 = 0.22; CCC = 0.36) and at d42 (R2 = 0.28; CCC = 0.44). At d14, the intake of milk by lambs as measured by the D2O technique, did not explain the variation in milk yield. In conclusion, udder dimensions, lamb live weight change and lamb milk intake are relatively poor estimators of the milk yield of singleton-rearing ewes obtained by the “oxytocin” method. Additionally, udder dimensions, milk yield and lamb milk intake do not give an accurate prediction of growth rates of singleton lambs. These results emphasize that there is a difference between ewe milk production potential and lamb milk intake, which need to be considered when estimating milk production in non-dairy animals.