OBJECTIVES: The objective of this research is to evaluate variation of mandibular arch forms at different heights and to determine if there is any correlation between occlusal arch and sub-gingival arch forms. METHODS: 44 subjects were selected based on the inclusion criteria and their CBCTs then were imported to Mimics software (Materialise NV, Belgium) and traced. Each tooth was sliced midsection at 4 different heights: Occlusal, CEJ, Apex and 5mm apical to the apex. At occlusal level the midpoint of the crown was calculated mesiodistally as well as buccolingually. Subsequently, midpoints between the buccal and lingual plates were located for every tooth between and including the first molars. The points were connected forming 4 splines, which then were exported to Geomorph software (cran.r-project.org, Geomorph package, Dean Adams author, Iowa State 2015) for shape statistical analysis. RESULTS: The variation in the arch form among subjects is significantly smaller in the Occlusal and CEJ level. The variation at apical and basal bone levels are higher than the variation at CEJ and occlusal levels. However, variation between apical and basal bone levels are minimal. CONCLUSIONS: Mandibular dental arch form demonstrate more variation apically. It may be concluded that dental arch form variation should be considered when using standardized arch forms for different patients. The subgingival arch forms cannot be predicted from occlusal arch form. Occlusal arch shape and form may not be an indication of basal bone arch form. More information needed for detection of correlation between occlusal arch and sub-gingival arch forms.