Theoretical assumptions regarding the genetic and environmental structure of personality proposed in Cloninger's seven-factor model of temperament and character were verified in a Japanese sample by using the twin method. The Temperament and Character Inventory (TCI) was administered to 296 twin pairs ranging in age from 14 to 28 years old. Among four temperament dimensions (novelty seeking [NS], harm avoidance [HA], reward dependence [RD], and persistence [PS]), HA and PS showed significant additive genetic contributions and no shared environmental effect, supporting the original theoretical assumption. NS and RD could be explained by either genetic or shared environmental factors with nonshared environment. All three character dimensions (cooperativeness [CO], self-directedness [SD], and self-transcendence [ST]) could be explained exclusively by additive contributions and no shared environmental effect. Multivariate genetic analysis indicated that there were no significant associations between NS, HA, and RD, as the theory predicts, and the genetic components of PS, SD, and CO were derived from those of the temperament dimensions. The fourth genetic component, which had a substantial load specifically on ST and overlapped with PS, was identified. Although most of the nonshared environmental effects were trait-specific, the phenotypic correlation between NS and HA could be explained by nonshared environmental overlap.