Species which do not enter torpor during fasting and which were efficiently able to spare their body proteins during the first two phases of fasting (which are commonly comprised of 3 successive phases) also increase their daily amount of slow-wave sleep (SWS) during the first two phases. Since in fasting animals the ability to spare proteins was reported to be improved when they were previously fed with a diet enriched with glycerol, it was supposed that, after such a diet, food-deprived rats would increase their daily quota of SWS. In addition, the tolerance to food deprivation, defined as the time elapsed to reach the end of phase II, should also be improved since this tolerance is known to be critically modulated by protein utilization. The daily proportions of wakefulness (W), SWS and paradoxical sleep (PS) were thus studied in Wistar rats after 16 weeks of feeding (i.e., when they were 27 weeks old) with an enriched glycerol diet. These daily W and sleep state proportions were then evaluated until the middle of fasting phase II (MII), i.e., when protein catabolism in the rat appears to be at its lowest level. The rats were able to tolerate more than 5 weeks of food deprivation, which represented an increase of 123% of the fasting tolerance previously reported in rats of the same age but which were fed normally before fasting onset. At MII the daily proportion of SWS was significantly (vs. fed state, p<0.01) increased (due to an increase in the daily mean episode duration), at the expense of W (due to a lowering in the daily occurrence of W episodes). During fasting, the daily amounts of PS remained roughly constant. The hypothesis relating the fasting tolerance to the rate of change observed in the daily amount of SWS still remains valuable even when the normal fasting tolerance is altered. The possible adaptative value connected to the change in the sleep-waking pattern of the present fasting rats is also discussed.