Previous experiments have indicated that interference with somatosensory feedback from convulsive movements may lessen the severity of audiogenic seizures in susceptible rodents. For further investigation of this phenomenon, mice were partially immobilized with tubocurarine chloride to attenuate convulsive movements and somatosensory input associated with such movements. In Experiment 1, seizures of mice injected with .15 mg/kg were evaluated behaviorally and compared with seizures of saline-injected litter-mates. The likelihood of clonic-tonic seizures in curarized mice was as high as that of control mice, although convulsive movements were somewhat less violent and seizure fatalities were markedly reduced. In Experiment 2, seizures of mice given .25 mg/kg were evaluated with electroencephalography, and records were compared with those of controls. Despite the near absence of behavioral signs of convulsions, electroencephalograms of curarized mice showed that audiogenic seizures readily occurred. The findings suggest that audiogenic seizures are centrally "programmed" and do not require feedback from convulsive movements. However, it may be possible to disrupt the central "program" by introducing appropriate somatosensory input not normally encountered during audiogenic seizures.