Hypernatremic dehydration was induced in rabbits during a 3- to 5-day period resulting in mean plasma sodium concentrations of 187 meq/liter. The animals were then rehydrated during a 4-h period by intravenous administration of a 2.5% glucose or fructose solution. The water content of four regions of brain sample showed a significant (P less than 0.05) increase in brain water content above normal in the rehydrated groups. Brain water content was significantly (P less than 0.01) greater in those animals with seizures compared with those without seizures, suggesting the importance of water intoxication in the pathogenesis of seizure activity. Changes in muscle Na, K, Cl, and water content were not similar to those of brain, indicating that muscle content of these substances was not an accurate reflection of the brain content specific time. The incidence of seizures was significantly (P less than 0.05) greater when glucose solution was used for rehydration (49%) compared with the use of fructose solution (25%). The mechanism(s) by which fructose resulted in a lower incidence of seizures is not known. The frequency of seizure activity was directly proportional to the rate of administration of intravenous solutions utilized to correct hypernatremia. In addition, the specific carbohydrate of the solution appeared to play an important role in the pathophysiology of the development of seizures.