The aim of the current study was to investigate three volleyball skills in three training conditions of blocked, random and serial to reveal their effects in three phases of acquisition, retention and transfer. 60 amateur male students (mean age = 21.5 years, SD= 1.2) were voluntarily selected and randomly divided into three training groups (blocked, random and serial). A pretest was used to match the subjects. In the acquisition phase, the subjects of each group repeated three skills of pass tiller, forearm pass and overhead service for 378 times for 9 sessions. In three sessions in the training phase, 45 acquisition tests were completed. 48 and 72 hours after the last session, subjects participated respectively in 15 retention and transfer tests. Analysis of Variance (ANOVA) and Tukey post hoc test were used to analyze data, to compare the means of pretest and tests in acquisition, retention and transfer phases of the three training methods. The results revealed a significant difference (P<0.001) in retention and transfer phases in favor of random (with more interference) and serial (with moderate interference) training methods compared to the block method (with lower interference) in all three skills. In the acquisition phase, blocked group had higher interference (P<0.001). There was no significant difference between random and serial methods in acquisition, retention and transfer phases.