Objectives: The clinical significance of the existence of a retromolar canal and of its neurovascular content is not yet clear.The aim of the present study was to assess the visibility, diameter and course of the mandibular retromolarcanal (MRC) using cone beam computed tomography (CBCT) scan -had been taken for pre-operative radiographic evaluation of impacted mandibular third molars- compared to panoramic radiographs. Study Design: Subjects eligible for study enrollment were those who underwent preoperative CBCT scan for the extraction of impactedmandibular third molars were determined to be extremely close to the mandibular canal on panoramic radiographs. Radiographs were screened for the presence and course of retromolar canals, and linear measurements. Results: 947hemimandibles in 632 patients were examined.A total of 253 MRCs (144 left, 109 right) were detected with CBCT images (26.7%). Only 29 of these canals were also seen on the corresponding panoramic radiographs. Most MRCs had a vertical course (type VI, 28.46%), followed by slightly curved (type I, 26.09%). The visibility of the MRC on the OPGs, according to the increase in the diameter, was not statistically significant for both sides (p>.05).Statistically difference were found for the width at the point of origin from the mandibular canal (p: .037), the mean distance from the MRC to the second molar (p: .042) and height of MRC when compared the gender. Conclusions: The findings suggest that the MRC isn’t a rare anatomical structure. This study therefore clearly establishes the incidence and importance of the MRC. The detection of the presence of the MRC using CBCT may be crucial for extraction of mandibular third molars. Key words:Accessory innervation, cone beam computed tomography, mandibular anatomy, panoramic radiographs, retromolar canal, retromolar foramen.