A further analysis may be made of the body-weights of these men to find if they represent optimum weights or not. For this purpose we may compare them with the normal averages given in the Medico-Actuarial Mortality Investigation tables for corresponding ages and heights, and note the expectancy of life as there established.4 Although the desirability of underweight has been especially emphasized by a number of writers,6 more particularly in Dr. E. P. Joslin's admirable treatise on diabetes mellitus,6 and in connection with the general problem of overeating, it is surprising that as a rule little stress is laid upon the fact that it is distinctly disadvantageous for young people to be underweight, although distinctly advantageous for people over 35 to be underweight. Thus, with young people between 20 and 24 years of age the mortality for the average weight is exceeded in all cases with an underweight of 5 or more pounds. From a careful analysis of the Medico-Actuarial Mortality Investigation records, which was made available through the kind assistance of Mr. Arthur Hunter, the actuary of the New York Life Insurance Company, it has been shown that this striking difference between young people and those over 35 years of age was due to the fact that a large number of these young underweights died of tuberculosis. It should be recognized, however, that the tuberculosis was not noted at the time of the examination and was either latent or developed subsequently. It is furthermore to be noted that the criteria of various examining physicians regarding acceptance for life insurance differ widely.

1 A portion of this gain may logically be attributed to the normal tendency of college students to gain during the fall and winter. 2 Benedict. Carnegie Inst. Wash. Pub. No. 77, 1907, p. 526. 3 Benedict. Harvey Society Lectures, 1906-1907, p. 199. Lecture delivered January 12, 1907.

4 Medico-Actuarial Mortality Investigation, II. Influence of Build on Mortality. The Association of Life Insurance Medical Directors and the Actuarial Society of America. New York. 1913. 5 Welch, Trans. Actuarial Society of America. 1916, 17, p. 17.

6 Joslin, Diabetes Mellitus. Philadelphia, 1917, 2d ed.. p. 66.

With Squad A at their minimum weight, the most marked case of underweight was that of Tom, with a deficiency of 12 kg., or approximately 25 pounds. Pec had an underweight of 8.7 kg., or 19 pounds, and Moy had an underweight of 8 kg. (approximately 17 pounds). Since Pec was 44 years old, we find from the tables of the Medico-Acturial Mortality Investigation that his underweight was advantageous, for it gave him a better expectancy of life. Moy and Tom, who were younger, may be considered, on the same authority, as having an expected mortality greater than normal. (See table 9, p. 207).

According to the standards established by the Medico-Actuarial Mortality Investigation, therefore, the underweight of most of our subjects would be disadvantageous. As previously stated, Mr. Hunter found that a considerable proportion of the underweights included in these tables died of tuberculosis. Whether the low weight was due to the latent tuberculosis or was simply a contributing factor can not now be determined. The tables of the Medico-Actuarial Mortality Investigation were, however, based upon deficiencies in weight which were due to normal uncontrolled causes, and the underweight of our subjects was the result of a designedly reduced diet. While the fact should not be lost sight of that underweight with young individuals, not produced by design, leads to a high mortality, one can not say definitely that the underweight due to the low diet given our subjects was an actual disadvantage.