At the adult neuromuscular junction, acetylcholine (ACh) receptors are highly localized at the subsynaptic membrane, whereas, embryonic myotubes before innervation have receptors distributed over the entire surface. Thus sometime during development, ACh receptors accumulate to the nerve contact area. This nerve-induced receptor accumulation can be reproduced in Xenopus nerve-muscle cultures, which provides us with a unique opportunity to investigate the underlying molecular mechanism of this event. Anderson and Cohen (1977) have shown that nerve-induced receptor accumulation is, at least partly, due to migration of pre-existing receptors. It is, thus, plausible that freely diffusing receptors in the membrane are trapped at the nerve-contact region and form clusters. We tested this diffusion trap model. First, receptors in the background region are indeed predominantly mobile and those in the cluster are immobile. Second, the diffusion of receptors in the membrane is fast enough to account for the rate of receptor accumulation. Third, when receptors were immobilized by a lectin, Concanavalin A, the nerve no longer induced receptor accumulation. Thus the diffusion trap model seems adequate to accommodate these observations. Aside from this diffusion mediated mechanism, it is conceivable that newly formed receptors are preferentially inserted at the nerve contact site and these new receptors become immobilized at the site of insertion. To test this hypothesis we stained new receptors separately from old ones and quantitatively compared their distribution. For this purpose we developed a method to quantify fluorescence micrographs. We found that the ratio between old and new receptors was similar at all nerve-induced clusters examined and at the diffusely distributed region.(ABSTRACT TRUNCATED AT 250 WORDS)