Data from the experiments on feed intake, digestibility
and live weight gain were subjected to the analysis of variance
(ANOVA) in a randomized complete block design using
the general linear model procedure of SAS (1998). Indivi-
dual differences between means were tested using Tukey HSD
test.