For future non-invasive optical biopsy by OCT, there is a demand to relate the obtained scans with histological cross sections, which represent the diagnostic gold standard. For this reason, the aim of the performed study was the qualitative matching of OCT cross-sectional images with histological pictures from various image databases. Therefore, the human oral mucosa of forty-seven healthy volunteers was investigated at ten different regions, e.g. lip, floor of mouth, hard and soft palate, by using the recently proposed in vivo endoscopic rigid handheld OCT system (central wavelength: λ = 840 nm, spatial resolution: 11.6 μm axial and 17.4 μm lateral). The results allow answering the question, whether OCT imaging is able to detect sublayers of mucosa and to clearly identify tissue structures by using a qualitative scoring system. In fact, depiction of epithelium, lamina propria and configurations like blood vessels and salivary glands, respectively, is possible, whereas a sufficient visualization of deep submucosa or epithelial sublayers cannot be achieved with the used wavelength range. Despite a well-defined surface profile, evaluating the level of keratinization is challenging. In contrast to histological cross sections impaired by fixation, orientation and organisation of collagen fiber bundles are highlighted better in OCT cross sections due to the in vivo situation. With regard to future clinical studies, correct interpretation of tissue structure in healthy individuals forms the basis to identify and assess pathomorphological changes related to various kinds of oral diseases by means of OCT.