The effects of repetitive activity on action-potential shape in Aplysia californica pleural sensory cells are described. Action potentials were evoked by intracellular current injection at frequencies between 7.41 and 0.2 Hz. In contrast to other molluscan neurons having brief action potentials, it was found that at these firing rates the normally brief action potential develops a prominent shoulder or plateau during the repolarization phase. Higher stimulus rates broaden the action potential more rapidly and to a greater extent than lower stimulus rates. Inactivation is slow relative to activation; effects of 3-s 6-Hz trains are detectable after 1 min rest. The amplitude of the plateau voltage reaches a maximum of 50-70 mV at the highest stimulus rates tested. Frequency-dependent increases in action-potential duration measured at half-amplitude normally range between 6 and 15 ms. Cadmium, at concentrations between 0.05 and 0.5 mM, antagonizes frequency-dependent broadening. The increases in duration induced by repetitive activity are more sensitive to cadmium than are the increases in plateau amplitude. Tetraethylammonium, at concentrations between 0.5 and 10 mM, slightly increases the duration and amplitude of single action potentials. During repetitive activity at high stimulus rates the maximum duration and rate of broadening are both increased but the amplitude of the plateau potential is not affected by these tetraethylammonium concentrations. Above 10 mM, tetraethylammonium greatly increases the duration and amplitude of single action potentials as well as the rates of action-potential duration and amplitude increase during repetitive activity. These high tetraethylammonium concentrations also cause the normally smoothly increasing duration and amplitude to reach a maximum value early in a train and then decline slowly during the remainder of the train. The consequences of frequency-dependent spike broadening in these neurons have not yet been investigated but it is clear from these data that repetitive activity in these cells will augment calcium entry and that this increased calcium entry has a complex but predictable dependence on the duration of, and firing rate within, an afferent volley. Because these cells are involved in important adaptive behaviour it is inferred that these behaviours will be complexly affected by the intensity and duration of the stimulation of the receptive fields of the pleural sensory neurons.