We tested the effect of episodic (≈ 10 days) corticosterone treatment on: (i) behavioural symptoms during kindling epileptogenesis; and (ii) electrical activity in the CA1 hippocampal area during epileptogenesis, and later on, in the fully kindled state. Male rats received a corticosterone-releasing pellet (100 mg/day) shortly before kindling was started, resulting in elevated hormone levels during the early and middle stages of epileptogenesis. The appearance of moderate behavioural signs of epilepsy and severe tonic–clonic seizures was significantly accelerated in corticosterone-treated animals compared to placebo controls. During epileptogenesis, corticosterone treatment did not affect the amplitude and paired-pulse characteristics of in vivo-recorded CA1 field responses, or the duration of the afterdischarge following tetanic stimulation of the Schaffer collaterals. However, other properties of CA1 cells studied in vitro, in the fully kindled state, were altered by the earlier episodic corticosterone treatment. Thus, in kindled rats, the amplitude of the population spike in the CA1 area was significantly enhanced after prior exposure to high corticosterone levels. Prior episodic steroid treatment resulted furthermore in a significantly increased amplitude of voltage-gated Ca currents, in kindled rats. At that time, corticosterone levels of animals which had received a corticosterone-releasing pellet earlier were no longer elevated compared to the placebo controls; the corticosteroid-treated rats did also not differ from the controls with respect to the mRNA expression levels for the two corticosteroid receptor subtypes in the hippocampus. The data suggest that exposure of animals to a period of stressful experiences during a critical phase in epileptogenesis could impose lasting deleterious effects on the course of epilepsy, even when CORT levels have been normalized again.