Following a historical sketch of attempts to explain homosexuality, we review evidence indicating that the process of determining human sexual orientation is fundamentally the same in all mammals. In this process, four phenotypic dimensions of sexuality develop from two more or less distinct sex genotypes. Studies are reviewed that indicate how phenotypic deviations from these two genotypes (called sexual inversions) can occur. The causes of sexual inversions are categorized as genetic-hormonal, pharmacological, maternal stress, immunological, and social experiential. From this evidence, we propose a theory of how the entire spectrum of human sexual orientation (vs. simply homosexuality) is determined. A consistent preference for sexual relations with one's own sex (homosexuality), the opposite sex (heterosexuality), or varying degrees of ambivalence about the partner's sex (bisexuality) may be called sexual orientation. Homosexuality should not be confused with occasional homosexual experiences. Homosexual experiences are fairly common, especially early in adolescence (Chilman, 1983, p. 18; Kinsey, 1941) or in the absence of alternative sexual outlets (Aldridge, 1983; Groth & Burgess, 1980) and are no more indicative of homosexuality than occasional heterosexual experiences are indicative of heterosexuality. An individual's sexual orientation refers to distinct preferences consistently made after puberty in the presence of clear alternatives, whereas isolated instances of sexual behavior may or may not reflect one's sexual orientation (Gadpaille, 1972, p. 193).