We test the proposal (McGeachie and Grounds 1985) that myogenesis following severe (crush) injury is prolonged compared with minor (cut) injury. Forty-four mice were injured with a cut and a crush lesion on different legs, and tritiated thymidine was injected at various times after injury (0 to 120 h), samples of regenerated muscle were taken 9d after injury and autoradiography was used to determine the initiation of muscle precursor replication, and duration of proliferation after the two different injuries. In both lesions replication of potential myoblasts was initiated 30 h after injury. Myogenesis was essentially completed in cut lesions by 96 h after injury, although the peak was finished by 60 h. In contrast, significant muscle precursor replication in crush lesions was still occurring 96 h after injury, and myogenesis was almost finished by 120 h. The pronounced difference in duration of myogenesis in different lesions strongly supports the original proposal. The extended duration of myogenesis in crush lesions, in conjunction with tritiated thymidine reutilisation, appears to account for conflicting experimental results in support of the concept of a circulating muscle precursor cell.