Abstract
Pedigree and performance records (N=3197) of a purebred herd of 1322 Nili-Ravi buffaloes maintained at the Livestock Experiment Station, Bahadumagar, Okara, Pakistan were utilized for the estimation of repeatability of 305-day lactation milk yield, total lactation milk yield, lactation length, dry period, service period, gestation period and calving interval. The repeatability values were estimated by using restricted maximum likelihood procedure fitting an' individual animal model. The derivative-free restricted maximum likelihood set of computer programmes was used in this analysis. The repeatability estimates were computed to be 0.21±0.02 for 305-day lactation milk yield, 0.18±0.02 for total lactation milk yield, 0.20±0.03 for lactation length, 0.09±0.03 for dry period, 0.09±0.03 for service period, 0.07±0.02 for gestation period and 0.05±0.02 for calving interval. The low estimates of repeatability for all of the performance traits suggested that most of the observed variation in these traits was due to temporary environmental conditions and culling buffaloes on the basis of early performance records would be of little value. Thus multiple records of buffaloes need to be considered while making culling decisions. Key words: animal model, Nili-Ravi buffaloes, performance traits, repeatability estimates
