Two models to account for the progressive nature of Wallerian degeneration after transection lesions in wild-type axons. (A) A putative inhibitor of intrinsic self-destruction machinery is constantly delivered from the cell body to the unlesioned wild-type axon (top). After axon transection the inhibitor is no longer supplied and is cleared first from proximal regions of the distal stump by fast axonal transport. This leads to a wave of fragmentation moving proximal to distal along the isolated axon stump. (B) In an alternative model, the wave of fragmentation is propagated not by directional removal of a putative inhibitor but by rapid localised influx of calcium ions beginning at the most vulnerable part of the axon. Once inside, calcium ions not only activate calpains to degrade the local axoplasm, but also diffuse and exceed the threshold of calpain activation in the immediately adjacent region. This leads to further axoplasmic and membrane breakdown and further calcium influx. The pattern is repeated to generate a wave of fragmentation moving along the axon. Model (A) has the attraction that the putative inhibitor would be a good candidate for mediating of the WldS phenotype (e.g., it could be overexpressed in WldS), while model (B) more easily explains why the directionality is reversed in a crush lesion. The calcium influx and diffusion wave could spread also retrogradely (not shown) if the distal end were the first to disintegrate. In model (A), however, it is hard to see how retrograde axonal transport could explain the depletion of an inhibitor that ultimately has to come from the cell body (see text for more details).