Changes in membrane capacity and excitation-contraction coupling caused by glycerol movements have been investigated in toad sartorius fibres using a standard glycerol-Ringer solution containing 400 mM glycerol.1. The rates of glycerol movement, in and out of fibres, were determined by measuring diameter changes in single fibres. Glycerol equilibrated across the surface membrane within 20-25 min after changes in extracellular glycerol concentration.2. The reduction in membrane capacity, which occurs when glycerol-loaded fibres are returned to Ringer solution, was slower than, and not dependent on, changes in fibre volume.3. To obtain the maximum reduction in membrane capacity, it was necessary to expose fibres to glycerol-Ringer for 50-60 min and then to return them to Ringer for at least 60 min. If preparations were not kept in Ringer for the full 50-60 min, the reduction in membrane capacity could be partially or completely reversed by returning the fibres to glycerol-Ringer.4. When fibres were exposed to glycerol-Ringer there was an initial transient contracture: twitches and K contractures were rapidly inhibited, and then slowly recovered over the next 40-50 min. In some preparations, eventual potentiation of twitches was seen.5. When returned to Ringer solution after 60 min in glycerol-Ringer, preparations lost twitches and K contractures within 5-10 min. The time course of this effect was very similar to the time course of the recovery of the normal volume after the initial swelling.6. The briefer the exposure to the glycerol-Ringer, the slower the loss of twitches and K contractures on return to Ringer. In contrast to the loss of membrane capacity, the lesion of excitation-contraction coupling was essentially irreversible. Exposure times as brief as 5 min were eventually effective in producing paralysed fibres which, however, still responded to caffeine.7. The differences in the glycerol load-times required to produce decoupling of excitation and contraction, and disconnexion of the transverse tubules, show that the two effects are separable and that the first does not depend on the second.8. It is proposed that the two lesions obtained in glycerol-treated fibres may be related to distension of the transverse tubular system. It is shown in an Appendix that the outward movement of glycerol from sarcoplasm to transverse tubules would be expected to produce some early swelling of the tubules and this is related to the loss of contraction. Furthermore, much greater swelling would occur if a slow-loading compartment (presumed to be the sarcoplasmic reticulum) dumped glycerol into the transverse tubules: it is thought that this is related to the disconnexion of the transverse tubules.