A Comparison of Treated and Untreated Male Crossdressers
Jack L. Croughan, M.D.,1,2 Marcel Saghir, M.D.,1 Rose Cohen,1 and Eli Robins, M.D.1 In an interview study of 70 male members of crossdressing clubs, multiple comparisons between treated and untreated subjects showed that the two groups are more similar than dissimilar. The only areas of comparison in which the treated group significantly differed from the untreated group were in fantasizing themselves as females while masturbating, having ever engaged in heterosexual intercourse while crossdressed, currently preferring both heterosexual intercourse and homosexual behavior while crossdressed, and having experienced more adverse consequences from crossdressing. Further, where comparisons are possible, our results are similar to those found in prior studies. All of the subjects were male and the average age of onset is prior to 10 years, with virtually all subjects first crossdressing if not in childhood then by middle adolescence. The course is chronic with only occasional and usually brief remissions, although there are instances in a minority of subjects of periods of remission lasting several months to a few years within the context of more than two decades of otherwise continuous crossdressing behavior. The interval between onset and first treatment, if any, is several years. Early in its development, crossdressing is virtually always associated with sexual arousal and sexual behavior, usually masturbation. Later, in adult life, it is more frequently associated with heterosexual intercourse and only rarely with masturbation as subjects approach middle age. There is a trend toward a more asexual nature to the crossdressing during late adult life. Crossdressing is infrequently associated with sadomasochism and not at all with exhibitionism. Rates of unipolar depression and alcoholism were increased in this sample. The results do not support a significant positive association between crossdressing and obsessive-compulsive neurosis. The present study confirms previous findings that crossdressing lacks a familial component either with respect to crossdressing itself or in association with another disorder. This study was supported in part by grant MH 20520-01. 1Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri 63110. 2Jewish Hospital of St. Louis, 2165, Kingshighway, St. Louis, Missouri 63110. KEY WORDS: crossdressers; interview; treatment; diagnosis; family; transvestism. INTRODUCTION The majority of publications on transvestism focus on discussion of theories of etiology or proposals for classification of crossdressing behavior. Some reports provide data about patients who have voluntarily sought treatment or who have been referred by the judicial system for evaluation. There are two prior series of articles that report on the results of interviews with subjects who were not selected on the basis of being patients or court referrals. Rather, they were members of crossdressing social organizations (Buhrich and McConaghy, 1977; Buhrich, 1976, 1977a, 1978) or respondents to a request for information in a transvestite publication (Prince and Bentler, 1972). To our knowledge, however, there have not been reports that have compared treated and untreated subjects. In addition, there is a series of reports on the results of personality tests with transvestites (Bentler and Prince, 1969, 1970; Bentler et al., 1970) that essentially shows no significant differences between subjects and controls. With the exception of a few case reports, however, we did not find any studies describing either the results of systematic psychiatric interview with subjects or the occurrence of psychiatric illness in their families. The purpose of this paper is to present comparative results of a personal interview study of treated and untreated male crossdressers. We want to determine whether conclusions based on results from prior studies of patients seeking treatment should be modified to account for differences in results obtained from subjects who have not sought treatment. In addition, we present information about psychiatric diagnoses in these subjects and their families. METHODS OF PROCEDURE Subjects either were a member of one of two national organizations of crossdressing social clubs or were referred as nonmembers by a member of one of the clubs. The clubs provided their members with advice and emotional support for crossdressing as well as a means of socialization. Exclusion criteria prior to interview eliminated homosexual drag queens and those who had received a transsexual conversion operation. Subjects denying these exclusion criteria were systematically interviewed using both closed and open-ended formats and were asked questions regarding the following areas: demographic characteristics including age, marital status, usual and current occupation, annual income, and residence; medical histories including a review of systems, hospitalizations and operations, and psychiatric hospitalizations; and symptoms of psychiatric illnesses including anxiety neurosis, hysteria, phobic neurosis, obsessional neurosis, depression, mania, alcoholism, drug abuse, antisocial personality, and schizophrenia. Psychiatric diagnoses were made using specific criteria (Feighner et al., 1972). Subjects were also asked about family histories of psychiatric illness including nervous breakdowns, attempted suicides, completed suicides, alcoholism, frequent trouble with the law, drug abuse, and sexual problems including homosexuality and crossdressing. Subjects were asked about environments of rearing including presence and absence of parents, broken homes by separation, divorce, and death, and global assessments of the subjects' opinions about the degree to which they were happy during childhood in that environment. They were also asked questions concerning discipline and role modeling for each parent or parental figures. Specific arrest histories were obtained including types and frequencies of arrests relating to sexual problems. Questions relating to early childhood traits including sissiness and early thoughts regarding gender dysphoria were asked. Subjects were also asked questions with regard to thoughts and behaviors relating to transsexualism. Extensive data were obtained regarding onset and characteristics of crossdressing as well as the association of this behavior with sexual arousal, masturbation, and heterosexual and homosexual behavior. Information about articles of clothing or accessories used, frequency of the use, fantasies associated with crossdressing, and awareness of parents, siblings, wives, children, and friends with regard to the subject's crossdressing were ascertained. Questions were asked regarding the relationship of alcohol and drug use to crossdressing and the extent to which crossdressing had directly or indirectly interfered with subjects' jobs, marital and family relationships, and other social relationships. Subjects were asked about the extent to which they had sought medical and psychiatric help for problems relating to crossdressing and the extent to which they had experienced remissions and exacerbations of crossdressing and related sexual behavior. A life history of heterosexual and homosexual experiences was obtained both in association with and in dependent of crossdressing. RESULTS A total of 70 males were interviewed. Club members accounted for 85% of the subjects, with 60% from one club and 25% from another The other 15% of the subjects were not members of any crossdressing organization or club. They were referred for interview by members of the two organizations. Table I. Sociodemographic Characteristics Treatment No treatment (N = 34) (N = 36) Average age at interview 41.9 42.9 Current marital status Married 50% 61% Divorced 30% 14% Single 21% 25% Religion of rearing Roman Catholic 21% 19% Protestant 62% 69% Jewish 6% 8% None 12% 3% Current SMSA Los Angeles 15% 31% San Francisco 21% 22% Chicago 26% 25% Denver 15% 3% Other Midwest Area 18% 8% Refused 6% 3% Foreign 0% 8% Highest educational level Attended high school 9% 6% Graduated high school 53% 64% Graduated college 26% 25% Graduated professional/graduate school 12% 6% Occupational rank Unskilled 9% 11% Semiskilled 6% 11% Clerical 9% 23% Skilled 38% 29% Professional/managerial 38% 26% Annual income (1972) $6,000-9,999 21% 22% $10,000-14,999 23% 28% $15,000-19,999 23% 19% $20,000-29,999 18% 8% $30,000-50,000 12% 11% Unemployed 3% 1l% Table II. Onset and Frequency of Crossdressing Treatment No treatment (N = 34) (N = 36) Average age at onset 8.3 11.3 Full or partial crossdressing once/week or more at < 10 years old 29% 19% 10-19 years old 56% 47% 20-29 years old 53% 56% > 30 years old 71% 58% Past one year 80% 86% Subjects were separated into treated and untreated subgroups for comparison purposes. A subject was placed into the treated group if he had on one or more occasions been seen by a physician, counselor, or other mental health professional for problems relating to his crossdressing. Duration, extent, and results of treatment were not used in determining allocation to the comparison groups. Using the above definition, 34 subjects (49%) were placed in the treated group and 36 subjects (51%) in the untreated group. Half of those in the treated group were self-referred for treatment (N = 17), the other half primarily sought help as a consequence of legal pressures by the courts (N = 6) or requests by wives (N = 7), parents (N = 2), or friends (N = 2). The sociodemographic characteristics of the subjects are displayed in Table I. The sample is of middle age and 95% white. Approximately one out of four had never married and most had been reared as Protestants. The subjects are of higher than average educational level and occupational rank, with higher than average income. There were no significant differences between the treated and untreated groups. Table II provides information about age of onset and subsequent frequencies of crossdressing. The difference in average age of onset between the treated and untreated groups is not significant. All but four subjects (6%) first crossdressed by age 14 years. About half of each group wore just undergarments the first time, whereas about 10% in each group were fully dressed. The remainder dressed in varying combinations of clothes at onset. Both groups (treated and untreated) reported similar frequencies of crossdressing during subsequent 10-year intervals. Table III. Masturbation, Fantasies, and Crossdressing Treatment No treatment (N = 34) (N = 36) (%) (%) Proportion of subjects who masturbated while crossdressed during adolescence 88 69 Every time crossdressed 40 36 A majority of the times 17 24 A minority of the times 43 40 Predominant fantasies during masturbation while crossdressed as adolescenta Crossdressing self 12 14 Self as female 29 6 Heterosexual thoughts 29 25 Homosexual thoughts 0 3 No fantasies 29 53 Proportion of subjects who masturbated while crossdressed as an adult 88 69 Every time crossdressed 47 52 A majority of the times 23 24 A minority of the times 30 24 Predominant fantasies during masturbation while crossdressed as adultb Crossdressing 15 19 Self as female 35 8 Heterosexual thoughts 23 22 Homosexual thoughts 3 0 No fantasies 24 51 ap < 0.06. bp < 0.05. The frequencies of masturbation with crossdressing as well as the predominant fantasies associated with masturbation and crossdressing at different age intervals are shown in Table III. During adolescence, more than half and beyond adolescence about 3/4 of both the treated and untreated groups masturbated at least a majority of the times they crossdressed. During adolescence (p < 0.06) as well as later (p < 0.05) subjects in the treated group more often fantasized themselves as females than did those in the untreated group. The extent to which subjects engaged in various forms of sexual behaviors while crossdressed are shown in Table IV. Approximately half had engaged in heterosexual intercourse at some time while crossdressed, whereas about one in four had participated similarly in homosexual behavior. None had been involved in exhibitionism in public. There was a trend for more of the subjects in the treated group to have ever engaged in the sexual behaviors listed while crossdressed, with a significant difference between the groups for heterosexual intercourse (P < 0.05). Within the year prior to the interview, the differences between the groups with regard to preferred sexual activity while crossdressed were also significant (p < 0.05). Most of the treated group preferred heterosexual intercourse, whereas about half of the untreated group preferred no sexual activity. Table IV. Sexual Behavior and Crossdressing Treatment No treatment (N = 34) (N = 36) (%) (%) Proportion of subjects who engaged in some form of sexual behavior while crossed-dressed 97 92 Masturbation 94 83 Heterosexual intercoursea 62 36 Homosexual behavior 32 22 Sadomasochism 6 3 Current preference while crossdressedb Masturbation 3 3 Heterosexual intercourse 59 42 Homosexual behavior 12 0 Any of above 0 8 No sex 26 47 ap < 0.05. bp < 0.05. Table V. Current Precipitating Factors within Past Year Treatment No treatment (N = 34) (N = 36) (%) (%) Seeing clothing 21 17 Depressed, bored 12 6 Feeling good 0 8 Tension, conflict 15 14 Desire for sexual arousal and relief 0 11 None 53 44 Table VI. Nonsexual Psychiatric Diagnoses Treatment No treatment (N = 34) (N = 36) (%) (%) Anxiety neurosis 9 3 Obsessional neurosis 3 8 Significant interference 0 3 Mild, occasional interference 3 6 Depression, unipolar 35 25 Definite 24 19 Probable 12 6 Alcoholism 26 22 Definite 18 11 Probable 9 11 Additional heavy drinkers 6 11 Drug use 26 19 Marijuana only, psychological dependence 3 3 Polydrug (nonnarcotic), no interference 24 17 Antisocial personality 18 8 Definite 6 6 Probable 12 3 Undiagnosed Paranoid schizophrenia 6 3 Table VII. Family History of Psychiatric Diagnosis All family members combined Father Mother Brothers Sisters Treatment No treatment (N = 69) (N = 69) (N = 62) (N = 73) (N = 143) (N = 130) (%) (%) (%) (%) (%) (%) Depression 3 7 5 4 6 4 Schizophrenia 3 - - - - - Alcoholism 14 4 5 - 9 5 Sexual deviations 1 - 7 1 - - Crossdressing 1 - 2 1 1 < 1 Homosexuality - - 5 - 2 - First-degree family members > 1 above diagnoses 25 12 16 6 - - Total - - - - 18 10 As shown in Table V, only about half of the subjects reported that there were factors in the prior year that they felt increased their desire to crossdress. Of those who did report one or more factors, however the majority reported that the sight of women's clothing often provoked and increased their desire to crossdress. The treatment group more often reported a form of dysphoric mood (depression, boredom, tension, conflict) as a precipitant, whereas the untreated group more often reported an increased desire to crossdress in association with feeling good or a desire for sexual arousal and relief. The differences approached but did not reach significance (p < 0.10). All but two of the subjects had tried to stop crossdressing on at least one occasion. However, over half of both groups tried stopping only once in their lifetime, and less than 30% in each group tried three or more times. There were no differences between the treated and untreated groups regarding abstinence from crossdressing. The longest average period of abstinence for both groups was about 1 year. The extent to which subjects reported nonsexual psychiatric diagnoses are shown in Table VI. Some subjects received more than one diagnosis. All diagnoses were made on the basis of lifetime prevalence (Feighner et al., 1972). The frequencies of unipolar depression and alcoholism were elevated. Mild increases in antisocial personality and possibly also paranoid schizophrenia and obsessional neurosis were observed (Goodwin and Guze, 1979). Table VII shows the lifetime prevalences of these same psychiatric disorders as well as the sexual deviations in first-degree family members. The data do not reveal any evidence for crossdressing as a familial disorder. In addition, there was no significant increase in homosexuality. The usually observed general population prevalence ratios for males and females of 1:2 for depression and 3-5:1 for alcoholism were observed in this family history data also (Goodwin and Guze, 1979). The proportions of subjects who experienced a variety of adverse consequences of crossdressing are shown in Table VIII. Subjects were recorded as having adverse consequences if they had been arrested, divorced, had experienced some significant interference in their occupation, education, or social relationships with others, or had experienced negative thoughts because of crossdressing. Categories were not mutually exclusive. In all, over 95% had either experienced at least one of the consequences listed or had sought treatment specifically for their crossdressing behavior. Treated subjects reported that they had experienced significantly more adverse consequences than subjects who had not sought treatment (P < 0.05). Table VIII. Adverse Consequences of Crossdressing Treatmenta No treatmenta (N = 34) (N = 36) (%) (%) Arrested Crossdressing 38 8 Child molestation 3 0 Divorce 27 8 Interfered with occupation 24 8 Interfered with education 18 3 Interfered with social relationship with other men 62 28 Interfered with social relationship with women 41 17 Subject objects 24 8 Family objects 62 53 Has lost friends 18 14 Other object 74 36 Feels guilty 18 25 ap < 0.05 DISCUSSION All of the subjects interviewed were males. None of the clubs had female members, although wives of members were often encouraged to accompany their husbands to the meetings and to club functions as guests. We are not aware of analogous crossdressing organizations that provide similar opportunities to females. This might be because of lack of need. It is much more possible in our culture for women to wear obviously masculine clothing without fear of recrimination. Thus, the need for mutual support and advice as well as a protective environment may be nonexistent for females who prefer to wear more masculine clothing. Also, possibly because of biologically or culturally induced differences in mechanisms of sexual arousal in males and females, there does not appear to be a female entity of crossdressers corresponding to males who crossdress for sexual arousal and relief. Such women are either rare, nonexistent, or simply do not bring attention to themselves by seeking help, experiencing interference, or coming into significant conflict with their family, friends, or society. Those females who do crossdress and seek help or experience conflict are reported to be either homosexual or transsexual (Lukianowicz, 1959a; Randell, 1959; Benjamin, 1966; Lester, 1975). None seem to correspond to the group of heterosexual males who crossdress primarily for purposes of sexual arousal. Regarding other sociodemographic characteristics, the data in Table I illustrate the fairly broad range and distribution of the variables described for the treated and untreated samples. Comparable information has been previously reported (Turtle, 1963; Buhrich and McConaghy, 1977; Buhrich, 1977a, 1978; Prince & Bentler, 1972), although our sample was 2-5 years older, somewhat more often married, and of slightly higher socioeconomic status. Except for cases of crossdressing associated with psychosis (Ward, 1975; Lukianowicz, 1959a, 1962), all authors report an early age of onset (Lukianowicz, 1959b; Buckner, 1970; Benjamin, 1966; Stoller, 1968; Turtle, 1963; Randell, 1975; Prince and Bentler, 1972), on the average by 10 years of age and almost always by 15. None of the four subjects in the current study whose crossdressing began after adolescence, however, was diagnosed as psychotic. A comparison of the frequency of crossdressing at different age intervals did not reveal any significant differences between the treated and untreated groups. The slightly higher proportions prior to 20 years for the treated group is probably a reflection of the earlier age of onset. Exact comparisons with other reports are compromised because of differences in data-reporting formats. However, a fairly specific comparison is possible between Buhrich's data on crossdressing frequencies in the prior 2 years (Buhrich & McConaghy, 1977) and our frequencies in the prior 1 year. Of the subjects in the present study, 64% engaged in full crossdressing and 59% in partial crossdressing on at least a weekly basis during the 1 year prior to interview. Corresponding figures for full and partial crossdressing from Buhrich's study are significantly lower, 29% for each. Regarding sexual arousal and behavior associated with crossdressing, Buhrich reported that all of his subjects had shown arousal to women's clothes. (Buhrich & McConaghy, 1977). Five of our subjects (7%) denied ever experiencing arousal with crossdressing. These same subjects also denied any sexual behavior, including masturbation, hetero- or homosexual activities, and other forms of sexual or sexually related behavior such as sadomasochism, child molestation, rape, or exhibitionism, while crossdressed. Manifestations of arousal most often involve masturbation or heterosexual intercourse. Of the subjects in the current study, 79% reported masturbating with crossdressing during adolescence or as an adult. The comparable figure for the period of adolescence from Buhrich's study is 53% (Buhrich & McConaghy, 1977). All but four of our subjects (6%) had masturbated at some time while crossdressed. Nearly half of our subjects (49%) had engaged in heterosexual intercourse while crossdressed, and about one in four (27%) had participated in some form of homosexual behavior on at least one occasion while crossdressed. There were 4% who had participated in sadomasochistic behavior while crossdressed. Comparable figures were not found in Buhrich's (Buhrich McConaghy, 1977) or other reports. Within the year prior to interview, 37% of our subjects stated that they preferred not to engage in any sexual activity. This occurred predominantly in the untreated group and in older subjects. Buhrich reported; similar observation of a decrease in arousal as subjects aged, with somewhat lower figure of 27% reporting no arousal in the previous months (Buhrich and McConaghy, 1977; Buhrich, 1977a). The fact that Buhrich's subjects were, on the average, 3-4 years younger might partly explain his lower figure. With the exception of isolated cases associated with episodes of psychosis (Lukianowicz, 1959b; Ward, 1975), the course of crossdressing is consistently described as chronic and unremitting (Lukianowicz, 1959b Buhrich, 1978; Lebovitz, 1972). This was certainly evident in the present study, where, on the average, subjects had been crossdressing with few exceptions on at least a weekly basis for over two decades, with occasional brief periods of remission usually lasting from a few months to a few years. A comparison with Buhrich's data (1978) reveals that 54% of his transvestites attempted to discard permanently all their women's clothes with 40% discarding them on more than one occasion. Corresponding figures from the present study are 83% and 53%. Buhrich also comments that his subjects invariably began crossdressing again usually within several weeks. As previously noted, our subjects were somewhat older and, thus, would have had more time in which to try to stop crossdressing, thereby perhaps explaining in part our higher percentages. Treatment appeared to play virtually no role in bringing about periods of abstinence This might have been a consequence of both the types of treatment employed and the duration of behavior. Most of the subjects received psychotherapy alone. This approach has not been found to be very successful; instead, electric aversion has been advocated (Marks et al., 1970). Further more, the average elapsed time between onset and first treatment for those who received treatment was about 20 years, perhaps making their behavior more fixed and resistant to change. Regarding possible associations of crossdressing with other psychiatric illnesses, we did not find other studies using systematically applied diagnostic criteria for nonsexual psychiatric disorders. There are isolated reports of crossdressing behavior accompanying the course of schizophrenia and manic-depressive illness (Lukianowicz, 1959b, 1962; Ward, 1975). Benjamin (1966) comments that the presence of psychotic behavior and frequency of diagnoses of psychosis including schizophrenia were observed in, at most, 6-8 patients out of 150 male crossdressers. Bentler et al. (1970) comment, on the basis of projective tests, that transvestites do not score like schizophrenics. However, he also interprets his findings as indicating that his subjects might have a latent thought process disturbance. Crossdressing associated with delusions of menstruation and pregnancy have also been mentioned (Lester, 1975, p. 169). In addition to these comments on crossdressing and psychosis, there are a few reports about possible associations with nonpsychotic disorders. A possible association of crossdressing with obsessive-compulsive neurosis has been discussed many times (Slater and Roth, 1969; Lester, 1975; Lukianowicz, 1959a; Randell, 1975; Buhrich, 1978). Lukianowicz (1959a) and Randell (1975) have also remarked that crossdressing has been described as being infrequently associated with psychopathy. Benjamin (1966) writes that his subjects' rate of alcoholism was low. In the current study, there was an increase of both unipolar depression and alcoholism compared with general population prevalences for these disorders. There might also have been mild increases of sociopathy, obsessional neurosis, and schizophrenia in our subjects (Goodwin and Guze, 1979). Thus, our findings appear to be consistent with prior reports in that there does not appear to be any obvious relationship between crossdressing and other nonsexual psychiatric diagnoses. The only category that was increased in our sample but not commented on in prior reports was depression. Dysphoria (i.e., a negative mood state) was reported as a precipitant of crossdressing, but depression as a syndrome seems unlikely as an etiological factor. The onset of crossdressing in virtually all of the subjects preceded the onset of depressive illness. Beyond etiological considerations, however, the observed rates of depression and alcoholism in our sample suggest a need for treatment of these disorders. There are two reports of instances of familial transvestism (Liakos, 1967; Buhrich, 1977b). In the present study, there were three subjects who had one first-degree family member each, a father, sister, and brother, who was also reported to have crossdressed. Other authors have commented on the nonfamilial character of crossdressing (Randell, 1975; Buhrich and McConaghy, 1977; Buhrich, 1977a; Friedemann, 1966; Edelstein, 1960). Family history data in the current study also failed to substantiate a positive association between crossdressing and other psychiatric disorders. Regarding the adverse consequences experienced by our subjects, the data suggest that subjects receiving treatment for crossdressing experience or at least report more problems associated with crossdressing. We are not aware of other reports in the literature with comparable data. REFERENCES Benjamin, H. (1966). The Transsexual Phenomenon. Julian Press, New York. Bentler, P. M. (1968). A note on the treatment of adolescent sex problems. J. Child Psychol. Psychiat. 9: 125-129. Bentler, P. M., and Prince, C. (1969). Personality characteristics of male transvestites: III J. Abnorm. Psychol. 74: 140-143. Bentler, P. M., and Prince, C. (1970). Psychiatric symptomatology in transvestites. J. Clin. Psychol. 26: 434435. Buckner, A. T. (1966). Deviant-group organizations. In Benjamin, H. (ed.), The Transsexual Phenomenon, Julian Press, New York. Buckner, H. T. (1970). The transvestic career path. Psychiatry 33: 381-389. Buhrich, N. (1976). A heterosexual transvestite club: Psychiatric aspects. Aust. New Zeal. J. Psychiat. 10: 331-335. Buhrich, N., and McConaghy, N. (1977). The clinical syndromes of femmiphilic transvestism. Arch. Sex. Behav. 6: 397-412. Buhrich, N. (1977a). The discrete syndromes of transvestism and transsexualism. Arch. Sex. Behav. 6: 483-495. Buhrich, N. (1977b). A case of familial heterosexual transvestism. Acta Psychiat. Scand. 55: 199-201. Buhrich, N. (1978). Motivation for crossdressing in heterosexual transvestism. Acta Psychiat. Scand. 57: 145-152. Congalton, A. A. (1969). Status and Prestige in Australia. F. W. Cheshire, Melbourne, Australia. Edelstein, E. L. (1960). Psychodynamics of a transvestite. Amer. J. Psychother. 14: 121-131. Feighner, J., Robins, E., Guze, S. B., Woodruff, R. A., Winokur, G., and Munoz, R. (1972). Diagnostic criteria for use in psychiatric research. Arch. Gen. Psychiat. 26: 57-63. Friedemann, M. W. (1966). Reflection on two cases of male transvestism. Amer. J. Psychoter. 20: 270-283. Goodwin, D. W., and Guze, S. B. (1979). Psychiatric Diagnosis (2nd ed.). Oxford University Press, New York. Lebovitz, P. S. (1972). Feminine behavior in boys: Aspects of its outcome. Amer. J. Psychiat. 128: 103-111. Lester, D. (1975). Transvestism. In Unusual Sexual Behavior - The Standard Deviations, Charles C Thomas, Springfield, Ill. pp. 169-179. Liakos, A. (1967). Familial transvestism. Brit. J. Psychiat. 113: 49-51. Lukianowicz, N. (1959a) Survey of various aspects of transvestism in the light of our present knowledge. J. Nerv. Ment. Dis. 128: 36-64. Lukianowicz, N. (1959b). Transvestism and psychosis. Psychiatr. Neurol. 138: 64-78. Lukianowicz, N. (1962). Transvestite episodes in acute schizophrenia. Psychiatr. Q. 36: 44-54. Marks, I., Gelder, M., and Bancroft, J. (1970). Sexual deviants two years after electric aversion. Brit. J. Psychiat. 117: 173-185. Prince, A., and Bentler, P. (1972). Survey of 504 cases of transvestism. Psychol Rep. 31 903-917. Randell, J (1959) Transvestism and transsexualism Brit Med J. 2: 1448-1451 Randell, J. (1975). Transvestism and transsexualism. Brit. J. Psychiat. (spec. no.) 9: 201-205. Slater, R., and Roth, M. (1969). Clinical Psychiatry. Bailliere, Tindall & Cassell, London. Stoller, R. J. (1967). "It's only a phase." J. Amer, Med. Assoc. 201: 98- 99. Stoller, R. J. (1968). Sex and Gender. Science House, New York, pp . 176- 193 . Thomson, W. (ed.). (1968). Sex and its Problems. E and S Livingstone, London. Turtle, G. (1963). Over the Sex Border. Victor Gallancz, London. Ward, N. G. (1975). Single case study - successful lithium treatment of transvestism associated with manic-depression. J. Nerv. Ment. Dis. 161: 204-206.