1. The properties of denervated and fully innervated but paralysed soleus and extensor digitorum longus (e.d.l.) muscles were compared in mice.2. Muscles were paralysed by single or repeated injections of botulinum toxin into the right lower leg. Contralateral muscles were denervated at the same time by sciatic nerve section. Muscles were excised 3-20 days later and direct and indirect isometric twitch and tetanic tensions measured in vitro. ACh sensitivity was assessed from isometric contractions to various concentrations of bath-applied ACh perchlorate.3. Denervated muscles as expected from the work of others developed sensitivity to ACh earlier and more rapidly than paralysed muscles. A transient peak of sensitivity was reached after 7 days of denervation in soleus and after 8 days in e.d.l. However 2 days later the sensitivity of muscles fully paralysed to nerve twitch and tetanic stimuli was not significantly different from that of denervated muscles.4. In both e.d.l. and soleus the direct tetanus/twitch ratios declined in paralysed and denervated muscles. The decline was initially more rapid in denervated than paralysed soleus, but the ratios for paralysed and denervated muscles were not significantly different after 10 days in either muscle. By this time there were no differences in the absolute strengths of the paralysed and denervated muscles.5. It is concluded that the ACh sensitivity and dynamic properties of fully paralysed mouse muscles become equal to those of denervated muscles once the acute effects associated with nerve degeneration are past. This result is consistent with the idea that normal neuronal control of extrajunctional mammalian skeletal muscle properties is mediated solely by means of activity.