Bulking up: the roles of sex and sexual orientation on attempts to manipulate physical attractiveness.

Socialization processes can either reduce or enhance biological differences between individuals or groups. Arguments, for example, that sex differences in math ability, aggression levels, or nurturance are reflections of sex-role stereotyping are, essentially, claims that socialization is creating or enhancing group differences. Such arguments are often presented with the accompanying claim that gender socialization can lead to stereotypes, prejudice, discrimination, and infringements on personal freedom and thus should be discouraged (e.g., Beall, 1993; Cross & Marcus, 1993; Fausto-Sterling, 1985; Geis, 1993; Lott & Maluso, 1993; Matlin, 1993). Some carry the argument even further with the exhortation to encourage socialization processes that counteract traditional sex roles (e.g., Bem, 1984; Cross & Marcus, 1993; Geis, 1993; Matlin, 1993).

One response to this plea is the recent increase in social and financial support systems that encourage women to engage in more sports, fitness, and body-building exercises that were once thought to be solely masculine activities. As a result of this cultural change, women athletes are participating in greater numbers in more events and are giving performances that were once thought to be possible only by men (e.g., Fausto-Sterling, 1985; Fox, Bowers, & Foss, 1993). It was my suspicion, however, that most non-professionals taking advantage of athletic equipment and training programs were doing so in ways that would enhance, rather than reduce, pre-existing sex differences. I hypothesized that many men and women who use athletic training equipment do so in an attempt to increase their physical attractiveness to potential mating partners (Buss, 1988) and would thus use athletic equipment in ways that would exaggerate physical, biological gender signals.

As a result of hormonal differences beginning at puberty, males and females develop noticeably different body profiles (Ross & Ward, 1982). Males develop a "triangular torso with broad shoulders and narrow hips," whereas females develop an "hourglass shape" (Campbell, 1989, p. 139). These shapes are, generally, easily and quickly identifiable and are used not only to classify individuals as male or female (Lippa, 1983) but also to assess physical attractiveness (Barber, 1995; Beck, Ward-Hull, & McLear, 1976; Franzoi & Herzog, 1987; Furnham, Hester, & Weir, 1990; Gitter, Lomranz, Saxe, & Bar-Tal, 1983; Horvath, 1979; Lavrakas, 1975; Madson, 1995; Singh, 1993a,b, 1995). These distinctive body shapes, like other bio-social signals, have been emphasized and exaggerated by reproductive-age individuals across a wide variety of times and cultures using a variety of techniques (e.g., Eibl-Eibesfeldt, 1989, pp. 63-65). This makes sense from an evolutionary perspective to the extent that an increase in the magnitude of an individual's sexual signals corresponds to an increase in sexual attractiveness and its consequent reproductive payoffs (Barber, 1995).

Given the evolutionary logic and the cross-cultural data suggesting that we olden use socio-cultural techniques to enhance pre-existing sex differences, I postulated that the recent encouragement of women to participate in traditionally male fitness and strength-training activites would not result in a lessening of "culturally imposed" gender differences, but rather allow a co-opting of this new opportunity to exaggerate pre-existing biological differences. Specifically, men were hypothesized to spend a disproportionate amount of their exercise time on anaerobic activities, especially working to build muscles of the upper body, which are, generally, bigger (Ross & Ward, 1982) and pound per pound, twice as strong in men as in women (Fox et al., 1993; Laubach, 1976; Wells & Plowman, 1983). Women, on the other hand, were hypothesized to spend more of their time in aerobic activity and, of their anaerobic exercise, to spend a disproportionate amount of time working to build leg muscles, which, pound for pound, are as strong as those of men (Fox et al., 1993; Laubach, 1976; Wells & Plowman, 1993) and specifically to avoid activities leading to an increase of muscle mass in the upper body.

These predictions were first tested in several samples of individuals presumed to be heterosexual and then in a sample of men presumed to be homosexual. Based on recent findings by Bailey, Gaulin, Agyei, and Gladue (1994), Bailey, Kim, and Hills (1996), Gonzales and Meyers (1993), and Kenrick, Keefe, Bryan, Barr, and Brown (1995), gender was expected to be the key variable influencing courtship-related attitudes and behavior, regardless of sexual orientation. To the extent that working out is a means of increasing one's physical attractiveness, I therefore predicted that gay men, like heterosexual men, would try to enhance male gender signals, resulting in data that would replicate those from heterosexual men and differ from those from heterosexual women.

Method

Participants

A brief questionnaire was administered to five samples of volunteers as they worked out in a men's gymnasium on a small college campus (N = 73; mean age 20), a women's gymnasium on the same college campus (N = 73; mean age 19), a co-ed gymnasium on a different, but nearby college campus (N = 98 males; mean age 21; and 31 females; mean age 20), a predominantly gay male, membership-only gymnasium nearby (N = 61 males; mean age 36), or an age-matched, predominantly heterosexual, membership-only gymnasium nearby (N = 64 males; mean age 36; and 48 females; mean age 28).

Measures

The questionnaire consisted of a one-sided page with five questions. Participants were first asked to note their sex and age. Next, participants were asked to cross off from a list those possible reasons for working out that did not pertain to them and then to rank order (with 1 as most important) those remaining items that did pertain to them. The list of possible reasons for working out included organized athletic training program, weight control, appearance to opposite sex, appearance to same sex, stress relief, general health maintenance, and injury rehabilitation.

The next question asked participants to report the average amount of time spent in various categories of activity during a typical workout. The categories were stretching, aerobic workout, and anaerobic workout of the upper body, lower body, and torso. Last, participants were asked to give any additional information about whether and why they might particularly avoid or spend extra time on any of the five activity types. The purpose of this question was to determine whether any differences between the groups might have been due to extraneous reasons (such as that a particular piece of equipment may have been absent or broken at one gym) and to provide qualitative supplemental information about the participants' reasons for and patterns of working out.

Because questionnaires were distributed and re-collected face to face, anonymity could not be ensured at all phases of the research. For this reason participants were not queried about sexual orientation. For the purposes of data analysis it was assumed that the majority of participants from the gymnasium that targeted its advertising toward the gay community were, in fact, gay, and that the majority of participants from the campus and co-ed membership-only gyms were heterosexual. Post-hoc data simulation trials that varied the hypothetical "true" values of a pure gay sample and the proportion of heterosexuals "contaminating" the gay sample showed that, with sample sizes and standard deviations similar to those of the actual data, sample contamination of up to 30% would not lead to Type 2 (beta) errors. That is, for between-sample comparisons, a reported lack of a significant difference between the "gay" and "heterosexual" samples is, indeed, likely to indicate a true lack of difference between the groups unless more than a third of the sample from the "gay gym" was heterosexual or vice versa.

Procedure

After receiving permission from the respective gymnasium managers, I or a student research assistant (male for the gay gym and the age-matched co-ed gym, female for the college gyms) approached individuals as they entered the weight room and asked if they could spare a minute or two to fill out an anonymous survey about exercise habits. No one refused.

Self-report data from the questionnaire were used to exclude data from those participants who indicated that their primary reason for working out was tied to involvement in a specific (e.g., team) athletic training program or to rehabilitation of an injury; this process reduced the samples by 26 (men), 16 (women), 34 (32 men and 2 women), 4 (men), and 8 (7 men and 1 woman), respectively. Data from an additional 28 individuals (23 men and 2 women from the college samples and 3 men from the older samples) were excluded because the questionnaires were not correctly filled out. The most common error in completing the questionnaire was that participants simply checked relevant items on the list of reasons for working out, rather than ranking them. These culls left net sample sizes of 35 (men), 56 (women), 83 (55 men and 28 women), 55 (men), and 103 (56 men and 47 women), respectively. Other minor errors led to a few cases of partially missing data for some participants.

Because people spent substantially differing amounts of total time in their workouts, the absolute time spent in each of the five activities was converted to a proportion for each participant. Statistical comparisons for hypothesis testing were based on these relative proportions of each individual's workout that was devoted to each type of activity. When comparing these proportions between samples (e.g., as when comparing the proportion of their workout that a sample of men versus a sample of women devoted to exercising the upper body), independent-sample t-tests were used; when comparing these proportions within a sample (e.g., as when comparing the proportion of their workout that a sample of women devoted to upper body exercise versus lower body exercise), matched t-tests were used.

Hypothesis-testing involved preplanned comparisons between the sexes, within each sex, and between men from the membership-only "gay gym" and the age-matched gym. Between-sex comparisons were made only between samples that were matched for age and type of gym (single-sex college, coed college, or membership-only); thus, these three sets of comparisons served as three independent replications. Within-sex comparisons were made separately for each sample, again allowing for three independent replications. Because all hypotheses pertained either to relative time spent in aerobic versus anaerobic workout or to relative time spent in workout of the upper body versus the lower body (and not to time spent stretching or time spent on anaerobic workout of the torso), even when comparisons were made within samples, these relative proportions were free to vary statistically independently of one another. In this way, conversion of raw data to proportions was possible without compromising degrees of freedom and therefore without artificial inflation of p-values.

Results

On average, men had longer workouts (112.7 minutes across 3 samples) than did women (87.2 minutes across 3 equivalent samples), with the additional time, as predicted, devoted specifically to anaerobic (muscle building) activities. Table 1 presents the means and standard deviations (in minutes) devoted to each type of exercise activity for each sample. These data were then converted to proportions for hypothesis testing.

Table 1 Total Time (Minutes) Spent on Various Activities

            Unisex College Gyms        Co-ed College Gym
              Men      Women           Men        Women

Stretch      7.80       7.66           7.07       16.31
            (4.95)     (4.69)         (6.67)     (29.41)
Aerobic     13.24      24.51           9.29       38.71
           (20.73)    (15.11)        (14.93)     (27.89)
Torso       17.91       9.62          16.24       17.40
           (11.78)     (5.09)        (19.30)     (23.11)
Upper       49.80      12.96          53.40       24.64
           (29.18)     (7.81)        (30.74)     (23.84)
Lower       31.15      16.12          34.57       30.71
           (31.49)    (11.31)        (23.61)     (26.59)
Total      119.89      70.87         120.80      129.64
           (78.02)    (28.87)        (63.68)    (103.36)

            Co-ed Private Gym

             Men       Women

Stretch      9.63       9.67
            (4.94)     (5.25)
Aerobic      20.44      27.78
            (15.28)    (19.99)
Torso        12.38      10.09
            (9.40)     (5.50)
Upper        32.06      15.95
            (25.37)    (11.54)
Lower        25.14      17.59
            (23.26)    (11.75)
Total        99.65      81.06
            (51.81)    (25.58)

Note: Means and standard deviations (in parentheses). Columns are listed pairwise for ease of comparison of matched samples.

In terms of the proportion of time devoted to the various activities, men, as predicted, spent a greater proportion of their time working out their upper body, both as compared to time spent on their lower body and as compared to the time that women spent. [The within-sample comparison for men at the unisex college campus yielded t(36) = 5.84, p [is less than] .0001; for men in the co-ed campus gym t(45) = 5.77, p [is less than] .0001; and for men in the private gym t(54) = 2.63, p [is less than] .05. The between-sex comparison yielded t(91) = 8.93, p [is less than] .0001 for the unisex gyms; t(72) = 7.83, p [is less than] .0001 for the co-ed college gym; and t(99) = 3.66, p [is less than] .0001 for the co-ed private gym.] Women, on the other hand, as predicted, spent more time than did men on aerobic activities [t(91) = 7.31, p [is less than] .0001 for the unisex gyms; t(79) = 8.29, p [is less than] .0001 for the co-ed college gym; t(99) = 1.98, p [is less than] .05 for the co-ed private gym], and spent approximately equal amounts of time on anaerobic exercise of their upper versus lower body [t(55) = 1.11, ns for women at the unisex gym; t(27) = 1.59, ns for women at the co-ed campus gym; t(45) = 1.45, ns for women at the co-ed private gym]. Whereas women spent a far smaller proportion of time on workout of their upper body as compared to men (reported previously), there were no significant sex differences in proportion of time spent on workout of the lower body [t(91) -0.08, ns for the unisex gyms; t(72) = 0.73, ns for the co-ed college gym; t(99) = 0.02, ns for the co-ed private gym]. These differences were stronger in the college samples than in the older samples (see Table 2).

Table 2 Across- and Within-Sex Comparisons of the Proportion of Time Spent on Various Activities

               Unisex College Gyms     Co-ed College Gym

                Men     Women           Men     Women

Aerosub         .09    .35(****)       .075     .32(****)
Topratio        .45    .19(****)       .44      .20(***)
Lowratio        .21    .21(#)          .27      .25(#)

Top vs. Lov   (****)    (#)            (***)     (#)
(within-S)

               Co-ed Private Gym

                Men     Women

Aerosub         .21     .34(*)
Topratio        .29     .19(****)
Lowratio        .23     .23(#)

Top vs. Lov     (*)     (#)
(within-S)

In further support of the hypothesis, for both samples of college men (but not the older samples), Spearman correlations showed that the more important "appearance to the opposite sex" ranked as a reason for working out, the more time men spent working out their upper body. This result was mirrored for women at the single-sex campus gym (but not the co-ed campus gym), as the importance of "appearance to opposite sex" as a reason for working out was associated with less time working on the upper body and more time working on the lower body (see Table 3).

Table 3 Significant Correlations of Time on Exercise with Participants' Rating of "Attractiveness to Opposite Sex" as Reason for Working Out

   Unisex College                      Co-ed College
     Gym: Men                             Gym: Men

Total Time Upper Body   -.41(**)   Total Time Upper Body   -.29(*)
Proportion Upper Body   -.34(*)    Total Time Lower Body   -.27(*)
Total Anaerobic         -.30(*)    Total Anaerobic         -.33(*)

Total Workout           -.27(*)

   Unisex College                     Co-ed College
     Gym: Women                         Gym: Women

Topratio                +.29(*)    Total Time Lower Body   +.55(**)
                                   Proportion Lower Body   +.39(*)

Note: Negative correlations indicate more time spent as attractiveness was rated as more important; positive correlations indicate less time spent as attractiveness became more important;

(*) p < .05; (**) p < .01. The correlations are not independent and so should not be given undue meaning; on the other hand, all significant correlations fit with the general hypothesis, and no correlations were significant in a direction that would contradict the general hypothesis.

Data from the men from the "gay gym" were, as predicted, similar to data from the other male samples. Although gay men reported somewhat less total time per workout than did the age-matched sample, it was not significantly shorter [t(104) = 1.59, ns], and proportions of time spent on aerobic activity, upper body workout, and lower body workout were very close [t(104) = 0.74, ns; t(104) = 1.60, ns and t(104) = 0.90, ns, respectively; see Table 4]. Also, like the men in the age-matched sample, men from the "gay gym" devoted a significantly greater proportion of their workout time to upper body exercise than to lower body exercise [t(50) = 4.09, p [is less than] .0001].

Table 4 Comparisons Between and Within Age-Matched "Gay" and "Heterosexual" Samples

                Private Gay     Private Age-Matched
                 Gym (Men)            Gym (Men)

Stretch             7.82                 9.63
                   (5.68)               (4.94)
Aerobic            20.51                20.44
                  (14.99)              (15.28)
Torso              14.34                12.38
                   (9.75)               (9.40)
Upper              41.94                32.06
                  (23.62)              (25.37)
Lower              33.77                25.14
                  (24.68)              (23.26)

Total Time        118.25                99.65        (minutes)
                  (46.76)              (51.81)
Aerosub              .17                 .21(#)
Topratio             .33                 .29(#)
Lowratio             .25                 .23(#)      (proportions)
Top vs. Low        (****)                (*)
(within-S)

Note: Aerosub, Topratio, and Lowratio refer to the proportion of time of an individual's total workout that was devoted to aerobic, upper body anaerobic, and lower body anaerobic activities.

(*) p < .05;

(**) p < .01;

(***) p < .001;

(****) p < .0001;

(#) = not significantly different, in line with prediction of no difference. Results of across-group comparisons (independent groups t-tests) and within-group comparisons (paired t-tests using individual data) appear in the text. The data and within-S tests in column 2 are redundant to those reported variously in Tables 1 and 2 where they were used to compare data across the sexes.

In this sample there was no significant relationship between importance of "appearance to same sex" and workout patterns, just as there was no such relationship with "appearance to opposite sex" in the age-matched sample. Also as in the age-matched sample, appearance to others was a less frequently reported reason for working out than in the younger samples (see Table 5).

Table 5 Percentage Ranking Appearance to Same+ / Opposite# Sex as Reason/hr Working Out

             Single Sex            Co-ed Collge Gym
             College Gyms

              Men(#)   Women(#)      Men(#)   Women(#)

Ranked 1:      10.8      1.8          20.0      10.0
Ranked 2:      27.0      17.5         21.8      13.3
Ranked 3:      29.7      22.8         27.3      23.3

             Gay Gym
                         Age-Matched     Co-ed Gym
             Gay Gym
             Men(+)          Men#          Women#

Ranked 1:      5.5            0.0           8.9
Ranked 2:     18.2            4.3           16.1
Ranked 3:     25.5           10.9           10.7

Note: These numbers reflect the percentage ot each sample alter Cleletmg mose mmwaums whose data were not included in the rest of the analyses because they had noted that their primary reason for working out was either because it was required for eLn organized sport or they were rehabilitating an injury.

For college women, the number one reason for working out was weight control, which is likely to be motivated at least partly, if not primarily, to increase physical attractiveness (Singh & Young, 1995). For gay men and the age-matched heterosexual men, the number one reason for working out was general health, followed by weight control. "Attractiveness to same sex" was, on average, given a higher ranking by gay men than was given to "Attractiveness to opposite sex" by heterosexual men in the older group. Younger participants of both sexes gave higher rankings to "attractiveness" than did older participants of both sexes.

Discussion

Overall, the results largely supported the hypotheses. When given free choice in whether and how to use a new technology that has the potential to reduce biological sex differences, people, in fact, used the technology in ways that enhance those differences: Men tailored their workouts in ways that would enhance the musculature of the upper body, and women tailored their workouts in ways that would enhance leg and hip muscle. These results were consistently statistically significant across three independent samples of males and three independent samples of females, appearing most strongly in the younger, college-aged samples for whom "appearance to opposite sex" was one of the most important reasons for working out.

Also, as predicted, workout style was more closely related to gender than to sexual orientation. Data from men working out at a "gay gym" did not differ on any measures from those of men working out at an age-matched co-ed gym, and data from both samples were different in predictable ways from data from age-matched women. The finding that gay men tailor their workouts to enhance upper body signals of maleness fits with the findings of Bailey et al. (1996) that gay men prefer men with a decidedly masculine appearance as partners.

Although these null results were as predicted, it is unlikely that every participant at the "gay gym" was gay and that every participant from the co-ed gym was heterosexual, leaving the possibility of a Beta, or Type II error. To address this possibility, I ran post-hoc data-simulation trials using sample sizes and standard deviations derived from this data set. The results suggest that the lack of difference between the two samples reflects a true null difference between gay and heterosexual men unless more than a third of the men at either gym were incorrectly categorized. There was no opportunity to replicate comparisons across sexual orientations and no data to compare heterosexual and lesbian women. This aspect of the results requires further investigation.

All in all, the results support the ethological hypothesis: To the extent that the reason for working out is to enhance one's physical attractiveness, workouts are consciously or unconsciously designed in ways that function to enhance pre-existing biological gender signals related to strength and body shape. This rationale for working out is more commonly reported by participants in the younger samples--a finding that is also congruent with the evolutionary approach. Mate-attraction strategies should vary with age (Gangestad, 1993; Kenrick et al., 1995), such that younger individuals should rely more on the strengths associated with youth (e.g., energy and physical attractiveness), whereas older individuals should rely more on the strengths associated with age (acquired resources such as social skills, knowledge, wealth, and status).

The results also support, more generally, the idea that socio-cultural roles (such as those for gender) should not be viewed as a set of restrictions foisted upon defenseless individuals or written upon a tabula rasa, but instead should be viewed as the cumulation of individual choices and, perhaps, evolved predispositions. People do not randomly absorb information or conform to stereotypes but instead are selectively and differentially responsive to various messages, instructions, and role models, depending upon their status, the status of the model, and the content of the message (Buss, 1995; Fagot, 1985). I suspect that much of our technology and culture are generated by, and used in the service of, psychological predispositions that evolved because at one time they enhanced reproductive success and that, thus, individuals will often choose to use technology and other products of culture in ways that mimic and exaggerate biological sex differences, rather than in ways that reduce or transcend them.

References

Bailey, J. M., Gaulin, S., Agyei, Y., & Gladue, B. A. (1994). Effects of gender and sexual orientation on evolutionarily relevant aspects of human mating psychology. Journal of Personality and Social Psychology, 66, 1081-1093.

Bailey, J. M., Kim, P., & Hills, A. (1996). Butch, femme, or straight-acting? Partner preferences of gay men and lesbians. Unpublished manuscript.

Barber, N. (1995). The evolutionary psychology of physical attractiveness: Sexual selection and human morphology. Ethology and Sociobiology, 16, 395-424.

Beall, A. E. (1993). A social constructionist view of gender. In A. E. Beall & R. J. Sternberg (Eds.), The psychology of gender (pp. 127-147). New York: Guilford.

Beck, S. B., Ward-Hull, C. I., & McLear, P. M. (1976). Variables related to women's somatic preferences of the male and female body. Journal of Personality and Social Psychology, 34, 1200-1210.

Bem, S. L. (1984). Androgeny and gender schema theory: A conceptual and empirical integration. In T. B. Sonderegger (Ed.), Nebraska symposium on motivation: Psychology and gender (pp. 179-226). Lincoln: University of Nebraska.

Buss, D. M. (1988). The evolution of human intrasexual competition: Tactics of mate attraction. Journal of Personality and Social Psychology, 54, 616-628.

Buss, D. M. (1995). Evolutionary psychology: A new paradigm for psychological science. Psychological Inquiry, 6, 1-30.

Campbell, A. (1989). The opposite sex. Topsfield, ME: Salem House.

Cross, S. E., & Marcus, H. R. (1993). Gender in thought, belief and action: A cognitive approach. In A. E. Beall & R. J. Sternberg (Eds.), The psychology of gender (pp. 5598). New York: Guilford.

Eibl-Eibesfeldt, I. (1989). Human ethology. New York: Aldine de Gruyter.

Fagot, B. I. (1985). Beyond the reinforcement principle: Another step toward understanding sex role development. Developmental Psychology, 21, 1097-1104.

Fausto-Sterling, A. (1985). Myths of gender: Biological theories about women and men. New York: Basic Books.

Fox, E. L., Bowers, R. W., & Foss, M. L. (1993). The physiological basis for exercise and sport (5th ed.). Madison, WI: WCB Brown & Benchmark.

Franzoi, S. L., & Herzog, M. E. (1987). Judging physical attractiveness: What body aspects do we use? Personality and Social Psychology Bulletin, 13, 19-33.

Furnham, A., Hester, C., & Weir, C. (1990). Sex differences in the preferences for specific female body shapes. Sex Roles, 22, 743-754.

Gangestad, S. W. (1993). Sexual selection and physical attractiveness: Implications for mating dynamics. Human Nature, 4, 205-235.

Geis, F. L. (1993). Self-fulfilling prophesies: A social psychological view of gender. In A. E. Beall & R. J. Sternberg (Eds.), The psychology of gender (pp. 9-54). New York: Guilford.

Gitter, A. G., Lomranz, J., Saxe, L., & Bar-Tal, Y. (1983). Perceptions of female physique characteristics by American and Israeli students. The Journal of Social Psychology, 121, 7-13.

Gonzales, M. H., & Meyers, S. A. (1993). "Your mother would like me": Self-presentation in the personal ads of heterosexual and homosexual men and women. Personality and Social Psychology Bulletin, 19, 131-142.

Horvath, T. (1979). Correlates of physical beauty in men and women. Social Behavior and Personality, 7, 145-151.

Kenrick, D. T., Keefe, R. C., Bryan, A., Barr, A., & Brown, S. (1995). Age preferences and mate choice among homosexuals and heterosexuals: A case for modular psychological mechanisms. Journal of Personality and Social Psychology, 69, 1166-1172.

Lavrakas, P. J. (1975). Female preferences for male physiques. Journal of Research in Personality, 9, 324-334.

Laubach, L. L. (1976). Comparative muscular strength of men and women: A review of the literature. Aviation, Space, and Environmental Medicine, 47, 534-542.

Lippa, R. (1983). Sex typing and the perception of body outlines. Journal of Personality, 51, 667-682.

Lott, B., & Maluso, D. (1993). The social learning of gender. In A. E. Beall & R. J. Sternberg (Eds.), The psychology of gender (pp. 99-123). New York: Guilford.

Madson, L. (1995, May). Effects of physical androgyny on ratings of attractiveness. Poster presented at the Annual Meeting of the Midwestern Psychological Association, Chicago.

Matlin, M. W. (1993). The psychology of women (2nd ed.). Orlando, FL: Harcourt, Brace, Jovanovich.

Ross, W. D., & Ward, R. (1982). Human proportionality and sexual dimorphism. In R. L. Hall (Ed.), Sexual dimorphism in Homo Sapiens: A question of size (pp. 317-361). New York: Praeger.

Singh, D. (1993a). Body shape and women's attractiveness: The critical role of waist-to-hip ratio. Human Nature, 4, 297-321.

Singh, D. (1993b). Adaptive significance of female attractiveness: Role of waist-to-hip ratio. Journal of Personality and Social Psychology, 65, 293-307.

Singh, D. (1995). Female judgment of male attractiveness and desirability for relationships: Role of waist-to-hip ratio and financial status. Journal of Personality and Social Psychology, 69, 1089-1101.

Singh, D., & Young, R. K. (1995). Body weight, waist-to-hip ratio, breasts, and hips: Role in judgments of female attractiveness and desirability for relationships. Ethology and Sociobiology, 16, 483-507.

Wells, C. L., & Plowman, S. A. (1983). Sexual differences in athletic performance: Biological or behavioral? The Physician and Sportsmedicine, 11, 52-63.

Manuscript accepted August 20, 1996

Portions of this article were presented at the International Behavioral Development Symposium on the Biological Basis of Sexual Orientation and Sex-Typical Behavior, May 1995, Minot State University, Minot, ND.

I thank those student assistants who helped with the data collection: Sheila Beales, Andy Heim, and Dawn (Bee) Schlotec.

Dr. Linda Mealey is now at the Psychology Department of the University of Queensland, Brisbane 4072, Australia. Fax: [phone]. E-mail: [e-mail].