INTRODUCTION

In 1948, Kinsey et al. reported that 37% of white American males in their study of 20 000 individuals had had some homosexual experience. Kinsey was the first researcher in this field to distinguish between exclusive homosexuality and bisexuality, and proposed a scale to classify a person's sexuality covering the whole spectrum from exclusive heterosexuality (Kinsey Scale 0) to exclusive homosexuality (Kinsey Scale 6). An individual's rating is not constant, but may change throughout his or her lifetime, although Kinsey (1948, 1953) showed that by the age of 25 in men and 40 in women, a person's Kinsey rating had usually stabilised. Kinsey's estimate for the prevalence of predominant homosexuality (Kinsey scale 4-6) in white males at age 25 was 7.6%.

The Kinsey et al. study has subsequently been criticised (e.g. Pillard et al. 1986, Fay et al. 1989) for bias by the inclusion of substantial numbers of subjects with prison experience, and substantial numbers from college campuses. Also, the data are applicable only to men distinguished by their willingness to co-operate with an investigation of sex behaviour (Kallman 1952).

More recent attempts to determine the base rate of homosexuality in Western society include McConaghy et al.'s 1979 study of medical students at the University of New South Wales. Such a sample was chosen so that nonpatients could be investigated in which the homosexual and heterosexual groups were drawn from the same population. The results of the study showed that 60% of the students reported that they were aware of some homosexual feelings during adolescence, and more than 40% were still aware of such feelings. Although this study has the advantage of being highly representative of the given population (84% of all the medical students participated), most subjects were in their late 'teens or early twenties, so many may not have reached their final adult sexual orientation. Also, a sample of medical students can hardly be generalised onto the Australian population as a whole.

Probably the most accurate estimate of the prevalence of male homosexuality available is that of Fay et al. (1989). From the secondary analysis of a sample of over 3000 adults interviewed in 1970, they suggested the following lower-bound estimates: "20.3% of adult men in the United States in 1970 had sexual contact to orgasm with another man at some time in life; 6.7% had such contact after age 19; and between 1.6 and 2.0% had such contact within the previous year."

In estimations of prevalence, as in most other areas of study related to homosexuality, we find the peculiar fact that very little has been mentioned of female homosexuality. It is generally agreed that the prevalence of lesbianism is somewhat lower than that of male homosexuality (e.g. Reinisch, 1991).

Whatever the precise figures, it is clear that a significant number of the population in Western societies have some homosexual inclinations.

Krafft-Ebing (1922) gave, on the basis of the cases published up to 1877, the first exhaustive scientific exposition of homosexuality. He states that "the more painstakingly one investigates the so-called acquired cases of homosexuality, the more clearly it appears that the real decisive factor is a predisposition." Thus, as early as the turn of the century, a genetic basis of homosexuality was suspected, although the appearance of Freud's "Three Essays on Sexuality" and the birth of the psychoanalytic school signalled a shift of emphasis in explanations of sexual orientation. The various theories that have been proposed to explain homosexuality will be described in more detail later.

For such an apparently evolutionary unfavourable trait, it is not surprising that many attempts to explain its aetiology have concerned themselves with abnormal parent-child relationships, childhood seductions and the like, rather than purely genetic explanations.

Many researchers claim to have shown that homosexuality, or, at least, male homosexuality, is familial in nature (e.g. Henry 1941 (also see Zuger 1978), Pillard et al. 1981, Pillard & Weinrich 1986). For example, Pillard and Weinrich found that, in the Boston area of the United States, "heterosexual men had about as many homosexual brothers as would be predicted given national prevalence figures for homosexuality, but homosexual index men had about four times as many homosexual brothers, although in both cases most brothers were heterosexual." No significant difference was found in the number of homosexual sisters of homosexual men compared to heterosexual men. However, some studies have found evidence for the familial nature of female homosexuality (e.g. Kenyon 1968).

Of course, if homosexuality is familial, this could be explained by either a genetic or an environmental theory. Generally, many of the theories advanced are not mutually exclusive. For example, a theory which supposes that child-parent interactions are the crucial factor for determining sexual orientation does not tell us why such interactions are as they are; the way that a child interacts with his environment, and the extent to which the same environmental event can cause different subjective experiences in different children is largely controlled by the genetic makeup of each child. Further consideration to this subject is given in a later section.

What is required to give us a better understanding of the aetiology of homosexuality is direct evidence for or against the involvement of genes. A common paradigm to research such a question is the Twin Study method. A brief outline of the rationale of this method is given in the next section, followed by a summary of all of the twin studies of homosexuality conducted to date.