The atrophy was evaluated by using muscle biopsies, computed tomography and measuring tape. Ten well functioning below-knee amputees with a mean age of 56 years participated. The amputees had 33 per cent type I fibres in the amputated leg compared to 38 per cent in the non-amputated leg, the difference not being significant. In the amputated leg there was less type IIA fibres (p less than 0.01), more type IIB (p less than 0.05) and type IIC fibres (p less than 0.01) than in the non-amputated leg. Atrophy (single fibres or groups of 4-5 small fibres) of both type I and type II fibres were seen in most patients. The mean muscle fibre area of the vastus lateralis of the amputated leg was 74 per cent of that of the non-amputated leg (p less than 0.05). In both legs the area of type II fibres was larger than that of type I fibres. The cross-sectional area of the whole thigh in the amputated leg was 86 per cent (p less than 0.05), of the quadriceps muscles 66 per cent (p less than 0.01) and of the hamstring muscles 80 per cent (p less than 0.05) of the area of the non-amputated leg. It was not possible to demonstrate if there was a reduction in the total number of muscle fibres. This indicates that the main cause for the reduction of muscle volume is the reduction in fibre sizes. There was no significant difference in circumference measurements when using computed tomography or measuring tape.