Own Study

Body measurements

Tanner (1951) constructed a formula for the ratio between the bi-iliac/ biacromial breadth for discriminating between the sexes. LindegArd (1953) doubted whether the relationship between two correlated variables could be expressed as a simple ratio and instead worked out a system for comparing body dimensions by means of regression analysis. Studying representative groups of nearly 200 men of about 21 years of age, and of nearly 300 women of about 24, Lindegird (1953) found that the sexes differed in their arithmetic means for different measurements, and also in various ratios between these measurements, as seen from regression analysis. Among other things he noted that, given the same length of the tibia, the radius was longer in men than in women-in other words, that women had shorter arms in proportion to their legs than did men. He also noted that the biacromial breadth was greater in relation to the length of the radius and tibia in men than in women, whereas the bi-iliac breadth was greater in relation to the length of the radius in women than in men,

I calculated the following relationships for all 43 of the present subjects: the length of the radius/length of tibia, biacromial breadth/length of radius, biacromial breadth/length of tibia, bi-iliac breadth/length of radius, and bi-iliac breadth/length of tibia, according to Martin's (1928) method, making each measurement twice in succession, and using the average to represent the patient in question. The numerical values are set out in Table A, in the appendix, p. 124. After this I plotted the values for the various cases around LindegArd's regression lines in graphs containing Lindegird's values for once and twice the standard deviation for the variable represented on the ordinate. The results are shown-in figures

Figs. 1-10. Relationship between various body dimensions in transsexuals distributed around regression lines for relationships in question in a population studied by LindegArd (1953). Broken lines represent once and twice the standard deviation from the regression lines for the variable rcpresented on the ordinate. Figs. 1, 3, 5, 7, and 9 are for the male transsexuals, and 2, 4, 6, 8, 10 for the female transsexuals. 1-10. As seen, as a rule about two-thirds of the observations fell within one standard deviation from the mean, i.e., the measurements were distributed about the same as in the general population.

The women differed from the men in their distribution around the regression lines as follows:

Figure 4, showing the regression of the biacromial breadth on the length of the radius, shows that only 3 of the 13 values for the women fell under the regression line, while 10 fell above it. Thus the female group tended to have a larger biacromial breadth in relation to the length of the radius than do women in the general population. The same was true of the ratio between the biacromial breadth and the length of the tibia (fig. 6), 9 values falling above the regression line. But all the measurements fell within two times the standard deviation from the mean.

Lindegird also found that the sexes differed in amount of fat and muscle, but there was no point in examining either of these conditions in this series, as some of the subjects had been taking hormones, which might affect both fat and muscle. The nutritional state at the time of examination might also influence these factors.