Psychological stress leads to increased heart rate and blood pressure. Our previous work has shown that during conditioned fear, this tachycardia, along with the behavioural freezing response, is mediated at least in part by the medullary raphe, more specifically the nucleus raphe pallidus (RPa) and raphe magnus (RMg). These nuclei receive projections from major limbic brain areas, the most numerous of these coming from the hypothalamus, which is thought to drive the tachycardic response, and the periaqueductal gray (PAG), which has been implicated in the freezing response. To confirm that these direct projections are active during conditioned fear, we conducted double immunolabelling experiments combining fear-induced Fos expression with cholera toxin B retrograde tracing from the RPa and RMg. Rats were either placed for 30 min in a context where they had previously experienced footshocks (conditioned fear, N=5) or left undisturbed in their homecages (control, N=5) prior to sacrifice. Coronal sections of the brains (40 μm thick) were cut in a cryostat and collected every 160 μm from prefrontal cortex (PFC) to the cervical cord. Double labelled neurons ranging from PFC to the rostral brainstem were plotted into a standard rat brain atlas and later counted according to location. Our results show that, relative to the total number of double labelled neurons, most of these were found in the hypothalamus (24.5±6.8%) and PAG (34.2±8.8%). Nevertheless, the pattern of activated projection neurons within those structures did not monotonically correlate with the areas with most projections therein. Hence, within the PAG, double labelling was found in the dorsomedial (14.9±5.0%), lateral (9.9±2.4%) and ventrolateral columns (9.4±1.9%), but with a trend towards more labelling in the dorsomedial rather than ventrolateral column, which is known to have the most projections to the RPa. Similarly, the anterior hypothalamus, the main hypothalamic source of afferents to RPa, had about the same number of double labelled neurons (5.6±2.1%) as the dorsomedial (5.8±1.9%), paraventricular (5.8±1.3%) and perifornical nuclei (4.4±2.0%), the latter two having far less projecting neurons than the former two. The prefrontal cortex was the third most labelled region (15.0±3.8%), which is surprising given that it has relatively few projections to the RPa. Finally, double labelling was also found in the midbrain tegmentum (11.3±4.6%), nucleus cuneiformis (5.3±2.4%) and A5 (2.0±0.7%) area. In control animals not exposed to conditioned fear, very little double labeling was found within these areas. These results confirm that the most important active afferents to RPa and RMg during conditioned fear are the hypothalamus and PAG. They also reveal other sources of input including the prefrontal cortex and midbrain tegmentum. The role of these latter structures on the production of the different components of the fear response mediated by the medullary raphe has not yet been determined. Consequently, these could constitute new targets for the pharmacological prevention of stress-induced effects of fear, including tachycardia.