The restorative and energy conservation theories of sleep predict that physical exercise will result in increased slow - wave sleep (SWS) and sleep duration. Restorative theories hypothesize that exercise creates a need for restoration, with a concommitant increase in SWS and sleep duration; while the energy conservation hypothesis holds that sleep, and SWS in particular, is a state of energy conservation, such that, any increase in energy expenditure caused by exercise should result in an increase in these components of sleep. Exercise may be acute (a particular exercise session), or chronic (physical fitness due to habitual exercise). Accordingly the effects of exercise on sleep may be a consequence of a particular exercise session, or physical fitness. The present literature does not show clear support for a facilitative effect of exercise on SWS or sleep duration. There is evidence however, that the effect of exercise may depend on the fitness and age of the subjects and the intensity and duration of the exercise. This thesis reports three experiments designed to assess the hypothesised facilitative effect of exercise on SWS and sleep duration in young and older fit subjects under varying conditions of intensity and duration of exercise. Physical fitness resulting from aerobic training, has been shown to be associated with higher levels of SWS and longer sleep duration in young subjects. It was of interest to determine if this finding extended to an older, population. The effect of aerobic training in young and older subjects was compared in the second experiment. In the first experiment the sleep of 11 fit young subjects was compared across 4 conditions: a no exercise condition, a one hour walk, a one hour run, and a 6 hour walk. In the second, the sleep of four groups of subjects, 10 younger fit and 10 younger unfit subjects (average age 22 years), and 12 older fit and 9 older unfit subjects (average age 41 years) was compared under no exercise conditions, and the sleep of the fit subjects was compared under a no exercise and a severe (11 hour training run) exercise condition. Finally, in the third experiment, the sleep of 8 marathon runners (average age 40.75 years) was studied following a no exercise condition, a 11 hour training run and a competitive marathon. The results indicated that exercise did not increase SWS or sleep duration in any condition in any experiment. Instead, the second and third experiments indicated that intense exercise disrupts sleep. In the second experiment aerobic fitness in both age groups was associated with increased SWS, decreased sleep onset latency, and a tendency to increased sleep duration. This latter result replicates and extends previous findings of the effects of fitness on sleep. The negative results clearly fail to support either body restorative or energy conservation theory in their present form. The validity of the observation that aerobic fitness affects sleep is extended since it has been found in both young and older subjects. However, this finding does not offer support for either theory as other studies have recently shown that the fitness effect is limited to aerobic training.