Retention of nonsense syllables was tested in two groups of Ss: those having initial learning in the morning and those having initial learning at night immediately before sleep. These groups were subdivided so that recall was tested either 8 or 24 hrs after initial learning. As Jenkins and Dallenbach first demonstrated, retention was superior after 8 hrs for Ss when learned at night compared to those who learned in the morning. Retention with night learning was equal after 24 hrs to that observed after 8 hrs. Surprisingly, retention scores after morning learning were superior after 24 hrs to those observed after 8 hrs. Some possible interpretations of this result are advanced.