The experiments distinguished the effects of meaningfulness (M) and pronounceability (P) on acquisition and retention. In the first experiment, the exposure duration was varied and immediate recall obtained. It was found that P strings required a shorter exposure duration than M strings to obtain the same recall level. Random (R) strings required the longest exposure duration. In the second experiment, level of learning was equated by the use of the previously determined exposure times and a retention interval was introduced. Delayed recall was highest for R strings and lowest for P strings. The results were interpreted in terms of differential representation of the chunks in short-and long-term memory due to the nature of the chunks and the opportunity for rehearsal.