Ad libitum fed male Sprague-Dawley rats were randomly assigned to be housed individually (n=32) or four in each cage (n=32) from five to 23 months of age. Body weight and body composition by bioimpedance analysis were monitored longitudinally. An equation for analysis of body composition was developed. Deaths and causes of death were recorded. Of the individually housed animals (H1) 16 percent died, while 44 percent of those housed four in a cage (H4) died, the most common causes of death being circulatory failure, nephritis and tumours of the hypophysis.Body weight increased in both groups over time, significantly more for H4 from 6months of age. Both fat free mass and total body fat increased in both groups over time. Fat in percent of body weight increased up to an age of 14months. H4 animals had significantly more fat from 11 to 20 months of age but lost weight, mostly fat when approaching 23months of age. The occiput-sacrum length of the animals increased up to 11months of age and stayed then constant. Animals in both groups grew also in the cross-sectional dimension. The H1 animals continued to grow all the time, while the growth of the H4 animals stopped at the age of 17 months, when the mortality started to become pronounced. It is concluded that housing conditions can modulate the outcome of the ageing of sedentary and otherwise not treated rats in long-term experiments. Further, the feasibility of monitoring body composition longitudinally with bioimpedance analysis was demonstrated.