The experiment tested the effects of presentation word orders in a multitrial free-recall task. Three types of presentation were used: (1) randomized; (2) constant order; and (3) maintained order (maintenance of subjects order of recall on the subsequent presentation). In addition, the effects of number of recalls per presentation (1 or 3) were tested for each type of presentation. Results show the constant order group with single recalls to be superior to all other groups on mean number of words recalled. Measured organization indicated superiority for both constant order groups. The results are discussed in terms of a dual function of presentation hypothesis with emphasis on the distinction between primary and secondary organization. (Author)