Factors affecting milk yield and ewe's lactation curves estimated with test-day models

Abstract

Test-day records of purebred Tsigai, Improved Valachian and Lacaune ewes were analysed with a general linear model in order to investigate the effects of flock-test day, lactation number, days in milk, litter size and month of lambing. In total, 121576 (Tsigai), 247 902 (Improved Valachian) and 2196 (Lacaune) test-day records gathered over the period 1995-2005 were included in the analyses. Average daily milk yields were 0.604 ± 0.279 kg (Tsigai), 0.595 ± 0.243 kg (Improved Valachian) and 1.053 ± 0.475 kg (Lacaune). The significant (P < 0.05) or highly significant (P < 0.01) effects of flock-test day, lactation number (except for Lacaune), days in milk, litter size (except for Lacaune) and month of lambing (either fixed effects or covariates) tested by Fisher's tests were shown. The model explained about 50% of daily milk yield variability, with coefficients of determination as follows: 0.479 for Improved Valachian; 0.487 for Tsigai; 0.537 for Lacaune. Differences in estimated least-squares means were tested using multiple-range Scheffe's tests. A lower daily milk yield was found for the first lactation, single litter and lactations starting in March in comparison with daily milk yield for the second and third lactations (except for Lacaune), multiple litter and lactations starting in January and February (except for Improved Valachian). Ali-Schaeffer regression adopted for sheep was used for the fitting of lactation curve according to breed.