where Yij, observation of the response variable corresponding to the ith treatment; , the overall mean; Ti, effect of treatmenti; and eij, random error.The milk production study was analyzed according to a completely randomized design with the repeated measures(weeks) command of the Proc. Mixed of SAS (SAS, 2009). The model included treatment, experimental week, and theirinteraction as fixed effects. The within-treatment animal effect was included as a random effect. Determinations of milkproduction and composition measured the week before the beginning of the experiment were used as covariates. Thestatistical model was: