Previous studies suggest that sleep deprivation (SD) decreases glucose tolerance in humans. The present study examined the ability of 10 males to process a glucose load during two conditions separated by at least 10 d. Condition I consisted of sedentary daily activity and sleep deprivation (SDS). Condition II consisted of daily physical activity and sleep deprivation (SDX). In both the SDS and SDX conditions, subjects were sleep deprived for 60 h followed by 7 h of normal sleep. An oral glucose tolerance test (OGTT) was administered at 10 and 60 h of SD, and after a night of recovery sleep in each condition, and plasma glucose and insulin concentrations were measured. No differences in the total plasma glucose response to the OGTT were observed over the total experimental period during Conditions I and II. However, the insulin response to the OGTT was elevated in the two conditions after 60 h of SD. Furthermore, the sedentary Condition I (SDS) resulted in higher insulin responses at all times compared to exercise Condition II (SDX). It is suggested that SD contributes to the development of an insulin resistance that can be partially reversed by physical activity. The results support the suggestion that SD results in decreased insulin sensitivity at peripheral receptor sites which can eventually lead to insulin exhaustion at pancreatic sites after longer periods of SD.