Sheldon, W. H. The varieties of temperament; A psychology of constitutional differences. New York and London: Harper & brothers. pp. 283-284.
The textural component, t, is measured by a rating on the physical refinement or aesthetic quality of the individual, taking an average for the five regions of the body as a whole. For practiced eyes, the aesthetic judging of human bodies is quite as easy, and may be as accurate, as any other rating. The textural component, like gynandromorphy, is gauged to the decimal on a 7-point scale, and the curve for its distribution is even more sharply skewed than that for gynandromorphy. The mean t index for 4000 college students was found to be 1.96. In a study not yet published we have found the mean t index for 2000 men of the general population to be 1.81. So far as the first region of the body (head and neck) is concerned, it is easy to set your sights to match our scale for t. Sit in a public place like a railway station, and carefully grade each adult who passes a selected point. When your mean for a few hundred cases is about 1.8, your scale matches ours. For the other regions of the body, try a public beach in summer. It will be found that both the t component and gynandromorphy are easier to learn to scale than are the primary morphological components, for the former are simple linear variables -- not multidimensional concepts.Sheldon, W. H. (1949). Varieties of delinquent youth; An introduction to constitutional psychiatry. New York: Harper. pp. 20-22.
The t Component. The component of thoroughbredness. A rating on the aesthetic quality of the physical structure. For those who have attended dog shows, horse shows, girl shows, poultry shows, or other competitive exhibits of livestock, the t component is an old familiar friend. It is merely the physical quality of the animal.
As a boy I was, quite naturally, trained to judge poultry and dogs, for my father was a competitive breeder of bothy. Also he was something of a student of prize fighters and of athletes in general. I think he would have considered it a high compliment to have been called a good judge of beauty in women, as I would now. He was a closely interested observer in nearly every branch of natural history. Awareness of the t component from such a point of view and in the rural world in which I lived as a youngster was as natural and as inevitable as awareness of automobiles is today. [. . .]
An essentially quantitative meaning was contained in the term thoroughbred, applied to roosters or to men. Whether or not anybody ever knew in verbal terms what he meant by that concept I cannot say, but it is a fact that scales for thoroughbredness were used--and still are--on which experienced judges consistently agree within 3 points or closer on a 100-point scale.
The t component is really aesthetic pleasingness, I should say, so far as its morphological connotations go. In the case of horse or roosters the (human judges have been able to settle on structural "points" as quantitatively definitive of t. In judging the t component in fellow human beings, however, certain difficulties at once present themselves. One of the most serious difficulties is implied in the question, "aesthetically pleasing to whom?" A West Coast African Negro might have defensibly different standards for aesthetic pleasingness from those of a Chinese or Swede. That lavel of objection to the t component can be answered rightly enough. Negroes are of course to be judges as Negroes, Swedes as Swedes, and 4-5-2's as 4-5-2's; just as Rhode Island Reds and Partridge Cochins were always judged seperately, or pointers and setters. But what about the mongrels--the new hybrides? Every breed of man, as well as of horse or rooster, is mongrel in the sense that it is derived from a blending of predecessor stocks. The point--and I thinnk this may be the crucial point in the problem--is that some blends are good, whether new or old, when measured in terms of the resulting individual's strength, efficiency, intelligence, muscular coordination, and bodily harmony or symmetry, while some other blends are less good.
The t component is intended as a measure not of "purity of ancestry" but of the aesthetic success of that particular biological experiment which the individual himself is. It is not quite a foolproof measure. For judgments of t to have validity the judges must have both aesthetic intelligence and integrity. They must be able to recognize beauty of proportion even though they may have racial and somatotype preferences or prejudices of their own. Achievement of such aesthetic integrity is not as difficult as might be feared. I have found that the average graduate student, when first undertaking a series of ratings of primary t from a racially mixed series of somatotype photographs, achieves a correlation of about +.80 with my ratings. [. . .]
The problem of gauging the quality of human stock is a challenging one. It cannot long be ignored by any political group that desires to survive, and a projected social science which waives the question of the physical quality of its human material must remain either a precipitate of revolutionary opportunism or just a waste of time. [. . .]
Primary t refers to the aesthetic harmony and symmetry, or the beauty, of the physique as a whole as it is presented in the somatotype photograph. This is the t as you see it at a distance. Secondary t is the aesthetic consequence of close contact with the individual. It is a measure of the natural fineness of the physical personality in detail. The features of the face when seen close-up, as at physical examination; eyes, ears, nose mouth, and teeth; skin texture and hair texture; the fineness and excellence of molding of the hands and feet, and also of the genitals; fingernails and toenails; coloring, and the quality of the observable blood vessels. Secondary t is a quantification of the total impression of the covert or underlying quality of the person.
In the present study a simple 5-point scale is used for t. A rating of 1 means excessively low t; 3 is about average for the general population; 5 means excessively high t. This scale is not to be confused with the more precise 7-point scale used in some other studies and describe on page 283 of VT. When a 7-point scale for t is used decimally, 1.1 represents the lowest t ever seen, and 7.0 the perfect ideal. On such a scale the t of the general American population (primary t) turns out to average just about 1.8. To set your sights to match this scale, sit for an hour or two in a busy railroad station and jot down t ratings (disregarding at first the primary-secondary differential) on everything that walks by. When your average falls at 1.8 you are probably rating t in just about he same way that the Constitutional Research Project rates it.