Variance components of birth weight (BW), weaning weight (WW) and pre-weaning average daily gain (ADG) in Barki lambs were estimated by Multitrait derivative free restricted maximum likelihood (MTDFREML). Six different animal models were fitted by including or excluding genetic maternal effects, maternal permanent environmental effects and covariance between direct and genetic maternal effects in order to determine the most appropriate model. Genetic maternal effects appeared to be the most important effects under the conditions of this study. The most appropriate model was that allowing for the genetic direct effects, both the genetic and permanent environmental components of the dam effects and the covariance between direct and maternal genetic effects (Model 6). From this model estimates of the heritability for genetic direct effect (h²) were 0.29, 0.23 and 0.22 for BW, WW and ADG, while for maternal effects (m²) were 0.05, 0.39 and 0.08 for the same traits. The genetic correlations between direct and maternal effects (r_am) were positive for BW and ADG (0.78 and 0.37, respectively) and negative for WW (-0.34). The genetic direct correlation between BW and WW and between BW and ADG were 0.44 and -0.40, respectively. The corresponding genetic maternal correlation estimates were 0.54 and -0.34. The genetic direct and genetic maternal correlations between WW and ADG were 0.35 and 0.05. The study concluded that maternal effects have to be accounted for when estimate genetic parameters for BW, WW and ADG. The r_am value obtained suggested that ADG is a good selection criterion for carrying out a joint selection on lamb's growth capacity (direct effects) and ewe's suckling ability (maternal effects).