This study evaluated tile effects of sleep curtailment and SWS-suppression, respectively, on daytime alertness and subsequent night sleep. Seven subjects participated in four conditions: an undisturbed 8-h sleep (8U; 23.00-07.00 hours), an undisturbed 4-h sleep (4U; 03.00-07.00 hours), a 4-h sleep (4D; 03.00-07.00 hours) that was acoustically disturbed when delta waves appeared, and a condition with no sleep (0). Subjective sleepiness, sleep latency, and simple reaction time (RT) were measured. In addition, sleep quality was rated. 4D contained 50% of the SWS (as well as spectral slow-wave energy; SWE) compared to 8U, whereas the curtailment to 4-h did not significantly decrease SWS. 4D had lower subjective quality than the other two sleeps. The main difference in daytime sleep latency was between the 8U and the 0 conditions. Rated alertness was highest after the 8U sleep. The two 4-h sleeps did not differ significantly with respect to rated sleepiness or sleep latency. However, the effects of the 4U sleep were closer to those of the 8U sleep and the effects of the 4D sleep were closer to those of the no sleep condition. RT performance was significantly better during the 8U condition. Recovery sleep after 4D sleep contained significantly more SWS than recovery after 4U and 8U. The effects on SWE during recovery were less clear. It was concluded that sleep duration might be more important for daytime alertness than SWS content and that loss of SWS during one night is recovered during the following night.