Tree shrews constitute an interesting animal model to study the impact of stress or aging on the hippocampal formation, a brain structure known to be affected under such environmental or internal influences. To perform detailed investigations of the hippocampal formation, adequate knowledge of its anatomy should be present. Until now, the hippocampal formation of the tree shrew has not yet been studied extensively. The main objective of this study, therefore, was to describe the subfield boundaries in various levels of the dorsoventral hippocampal axis of the tree shrew (Tupaia belangeri) in detail. The secondary aim was to clarify whether a separate CA2 field can actually be distinguished in the tree shrew hippocampus, a fact that was denied in former reports. In addition, we aimed at investigating whether or not a CA4 subfield can be identified in the tree shrew's hippocampus. The immunocytochemical distribution of microtubule-associated protein 2 and the calcium-binding proteins, parvalbumin and calbindin, and the characteristics of Nissl staining in adjacent sections were compared. Because of the rather dorsoventral orientation of the long hippocampal axis in tree shrews, staining patterns were analyzed mainly in horizontal sections. The subiculum and the hippocampal CA1 and CA3 areas were easily identified. Moreover, we were able to demonstrate the existence of a distinct CA2 subfield in the tree shrew's Ammon's horn, contrary to previous reports. However, our results indicate that a CA4 field in the tree shrew hippocampal formation cannot be identified with the methods that we used. Therefore, supposed CA4 pyramidal neurons should be included into the CA3 field.