In this study we utilized electrophysiological and pathway tracing methods to investigate the projections from the medial prefrontal cortex to the midbrain raphe nuclei of the rat. Initial pathway tracing experiments using retrograde (horseradish peroxidase conjugates with wheatgerm agglutinin or choleratoxin B subunit) and anterograde (Phaseolus vulgaris-leucoagglutinin) markers demonstrated a direct, bilateral projection to the dorsal raphe nucleus and median raphe nucleus from the medial prefrontal cortex, and the origin of this projection was localized predominantly in the ventral medial prefrontal cortex (infralimbic/dorsal penduncular cortices). Using chloral hydrate-anaesthetized rats, extracellular recordings were made mostly from 5-hydroxytryptamine neurons in the dorsal raphe nucleus, but non-5-hydroxytryptamine dorsal raphe neurons were also studied, as was a small number of 5-hydroxytryptamine neurons in the median raphe nucleus. In an initial study, electrical stimulation of the ventral medial prefrontal cortex caused a post-stimulus inhibition in the majority (49/56) of dorsal raphe 5-hydroxytryptamine neurons tested (mean duration of inhibition, 200+/-17 ms); in some cases (8/56) the inhibition was preceded by short-latency (26 +/-3 ms) orthodromic activation, and a small number of cells was antidromically activated (6/56). Both single spiking and burst-firing 5-hydroxytryptamine neurons in the dorsal raphe nucleus responded in the same way, and median raphe 5-hydroxytryptamine neurons were also inhibited (5/5). In contrast, few (2/12) of the non-5-hydroxytryptamine dorsal raphe neurons tested were inhibited by ventral medial prefrontal cortex stimulation. The effects of stimulation of the dorsal and ventral medial prefrontal cortex were compared on the same raphe 5-hydroxytryptamine neurons (n=17): ventral medial prefrontal cortex stimulation inhibited 16/17 of these neurons while only 8/17 were inhibited by dorsal medial prefrontal cortex stimulation. Finally, the inhibitory effect of ventral medial prefrontal cortex stimulation on 5-hydroxytryptamine cell-firing was not altered by 5-hydroxytryptamine depletion with p-chlorophenylalanine or by systemic administration of the selective 5-hydroxytryptamine1A receptor antagonist WAY 100635. The latter findings indicate that the inhibition is not due to release of raphe 5-hydroxytryptamine which could theoretically arise from anti- or orthodromically activated 5-hydroxytryptamine neurons. Our results show that stimulation of the ventral medial prefrontal cortex causes a marked post-stimulus inhibition in the vast majority of midbrain raphe 5-hydroxytryptamine neurons tested. It seems likely that the projection from ventral medial prefrontal cortex to the midbrain raphe nuclei mediates the responses of 5-hydroxytryptamine neurons to cortical stimulation. These data are relevant to recent discoveries of functional and structural abnormalities in the medial prefrontal cortex of patients with major depressive illness.