Five thousand Norwegians, 2783 females and 2217 males, ranging in age from 16 to 89 years, have been screened for the presence of torus palatinus and/or torus mandibularis and analyzed with regard to size and shape of tori, sex, and age. The tori have been classified according to size as small, medium, and large. No measurements have been made, consistent with the view that these characteristics are non-metrical and should be assessed by means of a standard procedure. The prevalence figures of both tori were fairly low, in accordance with former reports. In torus palatinus there was female predominance, with a sex ratio of 5:3. In torus mandibularis the males were in majority, and the sex proportion was 4:3. In both sexes prevalence of the two tori was highest in the 35- to 65-year age group. Further analysis indicated in both tori a real correlation between prevalence and size and some correlation also between size and shape of torus palatinus but not of torus mandibularis. The figures of concurrence of the two tori were low, denoting a non-significant correlation. Still, the analysis showed that each torus occurred more than twice as frequently in an individual bearing the other torus. Torus morphoanalysis, explaining diverging patterns of clinical conduct in the two tori, and the relative weight of hereditary versus environmental factors as morphogenetic determinants of the condition, including a quasi-continuous genetic or threshold model, have contributed to shed new light on the torus phenomenon. The occurrence and propagation of torus mandibularis complies reasonably well with this model, and the present analysis indicates that even torus palatinus should appropriately be considered a threshold character.