I cannot suppress another objection to the usual way of proving the--to my mind undoubted--influence of breast-feeding on the duration of life in infancy. Why is the mortality of those children who have not been suckled for a week so large? Is it because they have not been suckled, or because they have only lived altogether for less than a week? Or, again, to be able to be suckled for 40 or 50 weeks, one must have lived for 40 or 50 weeks, but a child who has lived for 40 or 50 weeks, whether it has been suckled or not, has passed over the worst period. It is well-known that the mortality in the first days of life is the highest in the second week, much higher than in the third week, and so on. In short, the mortality changes in such an extremely high degree in the course of the first year of life that this period is much too long for the comparison between mortality of suckled and non-suckled children. One ought to calculate how many of those who have been suckled for 0 weeks, one week, two weeks, one month, three months, six months, and so on, have survived the first week, the second week, the first month, and so on. Only in this manner can be established what is the share of the absence of breast-feeding and what is the share of the innate weakness and tendency to disease in the degree of infant mortality." [Illustration: Map of Bavaria Percentage of under-sized Bavarian recruits (below 1.62 metres in height) in 1875. After Professor Ranke. Figure C 80.] Exhibit C 81-82. [Illustration: Map of Bavaria. Fitness for Military Service in Bavaria, 1902. Figure C 81.] [Illustration: Map of Bavaria Mortality from Pulmonary Consumption in 1901. Figure C 82.] [Sidenote: C 83]