Maternal deprivation of neonatal rats for 24 h has immediate and persistent effects on hypothalamic-pituitary-adrenal (HPA) regulation. Immediately after deprivation corticosterone (CORT) is elevated. The primary purpose of our experiments was to determine if, by preventing this CORT elevation, the persistent effects could be reversed. In experiment 1, pups were injected with dexamethasone at the onset of the 24-h deprivation period on postnatal day 11 to suppress the rise in CORT. In experiment 2 some aspects of maternal behaviour known to suppress CORT levels were mimicked during deprivation from postnatal days 11-12. The pups were either: (1) left undisturbed; (2) stroked periodically; or (3) stroked and episodically fed. At postnatal day 20 basal and stress-induced adrenocorticotrophic hormone (ACTH) and CORT levels were measured as well as brain mineralocorticoid (MR) and glucocorticoid receptors (GR). Neonatal rats receiving dexamethasone prior to the onset of the deprivation on day 11 did not differ on day 20 from deprived pups that were exposed to elevated CORT levels. There were no detectable changes in the non-deprived pups that were treated with dexamethasone. In contrast, feeding and stroking during the period of deprivation obliterated the persistent effects both with regard to the reduced ACTH response and the decreased GR mRNA in hippocampus and hypothalamus. Stroking alone appears to have no influence. In conclusion, the persistent reduction of the ACTH response to mild stress and the decrease of GR mRNA is not mediated by deprivation-induced elevations in CORT, but appears to be reversible by reinstating specific aspects of the dam's nurturing behaviour.