Published online by Cambridge University Press: 18 August 2016
Net energy efficiencies were calculated from data of an experimental herd with respect to type of cow, lactation number, stage of lactation and diet. The trial consisted of 71 Holstein-Friesians, 71 Jerseys and 71 Holstein-Jersey F1-crosses in 1st, 2nd and > 2nd lactation. Data were collected during 210 days of lactation, from calving to week 30 and included total dry matter intake, energy content of foods, milk yield, milk solids, body weight, body condition scores and several body measurements. The cows were divided into four feeding groups : high and low energy content of roughage as well as high and low proportion of concentrates. Net energy efficiency was calculated as the ratio of milk energy to total net energy intake after subtracting energy utilized for maintenance and body tissue change. Body tissue change was calculated either via body-weight changes or body condition-score changes. Due to the distribution of the efficiencies being skewed, efficiencies were transformed applying a Box-Cox transformation. Transformed net energy efficiencies were analysed using a repeated measurements design considering the sequential nature of the observations. Mixed models with a compound symmetry structure for the variance components were applied. Final models contained the fixed effects of type, lactation number, feeding group and the covariates week of lactation and its square. Holstein-Jersey crosses were more efficient than purebreds, second lactation cows were least efficient, cows given low energy roughage and a lower proportion of concentrates were more efficient than cows on the other diets. Least efficient were the cows belonging to the high energy roughage and higher proportion of concentrates group. The coefficients of determination of the final models were between 0·357 and 0·492.