Alpha 2-adrenergic binding sites in the medulla oblongata of tree shrews and rats were detected and quantified by in vitro-autoradiography with the alpha 2-antagonist 3H-rauwolscine (3H-RAUW). The autoradiographic pattern of the radioligand binding in the tree shrew medulla oblongata resembles that which has been described by others for the human myelencephalon. This pattern coincides well with the occurrence of catecholaminergic structures detected by immunocytochemistry with antibodies against phenylethanolamine-N-methyltransferase and tyrosine hydroxylase. In contrast to the rat, where only the nucleus tractus solitarii and the nucleus dorsalis nervi vagi were labeled, five discrete nuclei specifically bound 3H-RAUW in tree shrews. The highest number of binding sites was detected in the nucleus dorsalis nervi vagi (nX; Bmax: 333 fmoles/mg) and the nucleus tractus solitarii (NTS; 311 fmoles/mg), followed by the nucleus nervi hypoglossi (nXII; 297 fmoles/mg), the nucleus reticularis parvocellularis (FRS; 230 fmoles/mg), and the area of the catecholamine cell groups A1 and C1 (area C1; 202 fmoles/mg). Maximal binding in the two labeled nuclei of the rat was 158 fmoles/mg. The discrete nuclei of the two species also showed different affinities for 3H-RAUW with Kd ranging from 0.17 to 0.83 nM in tree shrews and 1.80 to 1.95 nM in rats. Competition experiments revealed that the radioligand bound specifically to alpha 2-binding sites. In the tree shrew, nX, nXII and the area C1, also have a relatively high affinity for the alpha 1-antagonist prazosin which is a quality of the adrenoceptor subtype alpha 2B. Furthermore, in the area C1, 3H-RAUW binding was inhibited by the dopamine antagonist haloperidol. There are thus species related as well as regional differences with respect to the number, the affinity, and the pharmacological properties of alpha 2-binding sites in the medulla oblongata. In tree shrews, alpha 2-adrenoceptors can be autoradiographically quantified in regions which are not labeled in the rat, although former data predicted the existence of such receptors, e.g., in the area of the adrenaline cell group C1.