Three different techniques were used to study the effects of 72 hr deprivation of paradoxical sleep on percentages total sleep, light slow wave sleep, deep slow wave sleep and paradoxical sleep during deprivation and recovery periods in rats. The paradoxical sleep deprivation methods used were the classical platform, the pendulum and the multiple platform techniques. The different groups were continuously recorded over a base-line day, three deprivation days and four recovery days. All three groups showed clear suppression of paradoxical sleep throughout the deprivation period, although small differences between the pendulum and the multiple platform technique emerged. The distribution of sleep across the day and the amount of deep slow wave sleep were affected differently. Besides an exception immediately after deprivation, no important differences were detected on the recovery days. The recovery is characterized by an immediate rebound of paradoxical sleep, completed within two days, as well as rapid re-normalization of sleep percentages. A small rebound of deep slow wave sleep was recorded at the end of the dark period during the first three recovery days. Various sleep and non-sleep related variables of the PS deprivation techniques are discussed. It is improbable that the platform-pendulum controversy is due to differences in the amount of PS deprivation or the other sleep parameters measured here. Rather it looks as though non-specific platform effects override the effects of PS deprivation.