Early studies in which it was found that learning followed by sleep was better remembered than learning followed by wakefulness were inter-preted as giving support for the Interference Theory of Forgetting. More recent studies have shown better retention over the first half of the night's sleep (slow-wave sleep) than over the second half (REM sleep), and conclusions have been drawn that a Decay Theory of Forgetting is more strongly sup-ported. Those studies, however, confounded the type of sleep following learning with sleep prior to learning. When prior sleep was controlled in the present study, there was no support for a first half-night sleep benefit, and, contrary to Decay Theory, there was a second half-night benefit for high imagery material. The strong detrimental effect of sleep prior to learning was inconsistent with the Interference Theory of Forgetting and suggested, instead, the importance of the consolidation process for long-term memory.