Two- and three-dimensional micro-computed tomography was used to quantify human molar tissues from a whole-crown perspective. This study tested how the sexes differ in their crown tissue components, how dental tissues are distributed throughout a molar and along the molar row, if measurements from different planes of section can predict the whole-crown condition, and whether hypocone or hypoconulid expression influences crown tissue component measurements.
Males significantly differed from females by having relatively thinner enamel in their lower molars and absolutely thicker enamel (and greater enamel quantity) in their upper molars, in particular in the UM1 and the protocone, which was attributed to function; however, females have a greater proportion of enamel in their molars compared to males. Furthermore, males had significantly greater dentine quantity, enamel-dentine junction (EDJ) length and surface area (few significant shape differences were found), basal-crown diameter and area, more widely-spaced and taller dentine horns, and taller dentine crowns. Combining the sexes for some analyses is therefore not appropriate, and comparisons of crown components of fossil hominin should take into account the possibility that some of the variation may be a result of sex rather than species differences.
Discriminant function analysis identified crown components of UM2 (volume), LM1 (distal section) and UM1 (buccal section) as being the most significant sex discriminators with an accuracy of 75% to 100% correct classification.
Significant metameric differences in crown components were discerned. Although molars differed from each other in EDJ size and shape, considerable overlap precludes the separation of molars. Furthermore, enamel thickness components increased from M1 to M3 while other crown components decreased. The mesial-to-distal increase in enamel thickness within molars is part of the greater trend of increasing thickness along the molar row.
Enamel and dentine volume were reliably predicted from different virtual planes of section, demonstrating their utility in reconstructing the whole-crown condition in worn and broken specimens. Furthermore, hypocone and hypoconulid expression significantly influenced the measurements of dentine area, EDJ length, and bi-cervical diameter in the distal and lingual sections of M2 and M3. Enamel thickness components were not significantly affected by trait expression in two- or three-dimensions.