This study was designed to examine effects of three practice models, blocked, variable, and combined, on the acquisition, retention, and transfer of new motor skills. 67 subjects (M age = 9.5 yr., SD = .3) from the fourth year of primary school (31 boys and 26 girls) were assigned at random to three different practice groups (Blocked = 22, Variable = 23, Combined = 22) to study acquisition of two skills, dribbling a soccer ball and kicking a soccer ball at a stationary target using the dominant foot. All participants received a pretest and posttest, a transfer test, and a retention test 2 wk. later. Analysis showed significant improvement after practice of kicking skills by the three groups but not in the dribbling skills, for which only the combined practice group showed any notable improvement. At the end of acquisition, the combined practice group had significantly better performance on the dribbling task than the other two groups. However, the only differences noted in performance of kicking the ball with the dominant foot were by combined practice and blocked groups.