Figure C 64 shows the +parallelism between+ the +average length of breast-feeding and the average time between births+ within the families. A half to three-quarters of the mothers who suckled either long enough or very long show an interval between births of from 1-1/4 to 3 years, whereas of those who did not suckle at all, or only did so insufficiently, only one-third belong to this group, and figure largely in the column of lower birth intervals. Dependence of Infant Mortality on the Average Duration of Breast-Feeding and the Average Length of Time Intervening between the Successive Births of the Children in a Family. [Illustration: Figure C 64.] [Sidenote: C 65] Figure C 65 enables us to examine into the +influence exercised by a longer or shorter interval after the preceding birth on the vitality of a child+, according as to whether the child was not breast-fed at all or only moderately or amply so. The black oblongs demonstrate that the average infant mortality falls regularly and decisively according to the length of time between the birth of the children considered and their predecessors. The average mortality of infants who are born in rapid succession--under one year, one to one and a quarter years, amounts to over 25 and to 22 per cent. respectively, whereas the average mortality of children with at least two years' interval amounts only to 11 per cent. "At the same time, however, it is observed that the influence of the length of suckling is still greater than that of the length of time elapsing between births. Even with an interval of three or more years, the mortality of children who were insufficiently or not at all breast-fed was above 20 per cent. The children who had been suckled for at least three-quarters of a year were only very slightly influenced by this factor in all groups, except that with a birth interval of less than one year, where the influence of short birth intervals is not counterbalanced even by long extended breast-feeding." [Sidenote: C 66] Figure C 66. "The +infant mortality within the families+ dealt with +falls materially and evenly as the average birth intervals lengthen+. With an average birth interval of less than one year, one-third of the children die in the first year, but only 7 per cent. where the average birth interval was over three years; but here also the influence is strongly modified by the mode of feeding. With the non-suckling families the mortality is almost 25 per cent., even with a birth interval of more than two years. On the other hand, when the duration of suckling is sufficient, short birth intervals almost disappear (see Table 2), and with an average birth interval of 1-1/4 to 2 years and a suckling duration of at least half a year the mortality remains on an extremely small scale." [Sidenote: C 67-73]