Clearly, many aspects of the action of neuromuscular blockers remain poorly understood at the molecular level. In the case of competitive blockers, blockade of EPPs by competitive binding to the ACh receptor site accounts for the most clinically important aspect of blockade. Although train-of-four fade produced by curare and some other competitive agents probably represents a presynaptic action, the molecular mechanisms underlying this effect have not been addressed. Depolarizing blockade is inherently more complicated than competitive blockade. Simple depolarization and inactivation of the mechanism for generation of the action potential probably account for the major clinical effect seen in phase I block. Furthermore, the relative balance between activation of channels and desensitization may also provide a qualitative explanation for phase II block and tachyphylaxis. However, effects that are more likely to be explained by presynaptic actions of depolarizing blockers have also been described, and it is not yet possible to assess quantitatively whether the rates of onset of the different postsynaptic actions can account for most aspects of depolarizing block. This discussion has raised several issues which need to be addressed in future studies. 1. What are the presynaptic effects of cholinergic drugs? Do these drugs act through a specific receptor or, on other ion channels in the terminal membrane, or do they operate by mechanisms distinct from effects on membrane conductance? Can any of the observations be explained by indirect effects mediated through postsynaptic ACh receptors, e.g., K+ release? 2. What are the factors that influence variability in sensitivity to neuromuscular blockers among species, muscles within species, and during development? Many of the potential factors, e.g., differences in safety factor, resting conductances, ACh receptor type, and differences in the presence and absence of presynaptic receptors, have been outlined, but definitive tests of the contribution of any particular mechanisms are lacking. 3. Does modulation of desensitization play a role in any components of neuromuscular blockade? 4. Can trapping of blocking agents in ion channels in some cases account for slowly reversible components of blockade? 5. Can closed-channel block provide an effective mechanism of neuromuscular block?