The present study attempted to replicate, in a real-life analogue, the findings of Rachlin and Green (1972) regarding the reversal of preference for a small, immediate reward or a large, delayed reward in time, as well as the utility of prior commitment as a means of self-control. One hundred three college students were presented with a two-link chain of events. In the initial link, subjects could either commit themselves to a large, delayed reward or proceed to a terminal link, T days later, in which they could choose between the large, delayed reward or a small, immediate reward. According to Rachlin and Green, self-control is evidenced by both a reversal in preference from small to large reward in time and an increase in commitment to the large reward as T increases. In the present study, a definite and consistent preference was found for the large, delayed reward across all values of T. These results replicated neither Rachlin andGreen (1972) nor Burns and Powers (1975) in their attempt to replicate Rachlin and Green's results. In post hoc analyses of the data, a methodological problem was discovered which altered the procedure used in this study from that used by Rachlin and Green. Upon correction of delay values, tentative support was found for Rachlin and Green's matching equation for finding the value ratio for two alternative choices differing in amount and delay of reinforcement.