The evidence supporting the suggestion that kindling is a good model of human temporal lobe epilepsy is briefly reviewed. Parallels between the human condition, involving both partial and secondarily generalized seizures, and kindling in rats and other animals are drawn and contrasted.