_Statistical Data._--The statistics relating to child mortality are in an exceptionally well-developed state, and no unprejudiced student of sociology can afford to ignore them. The literature of child mortality contains a number of extremely important and thoroughly trustworthy works; the reason for this may be that, in comparison with other difficult problems of population, the study of questions of child mortality is easier, because various disturbing influences which complicate adult death-rates have no bearing upon child mortality. Even simpler is the question of infant mortality.[1] In computations dealing with this matter it is not necessary to make use of the figures of the general census, for the calculations are based simply upon the recorded births and deaths. The calendar year in which the birth took place does not come directly into the question at all. What we record is the rate per thousand at which, in or during a particular year, say 1909, infants have died before attaining the end of the first year of their life; some of these will have been born in the year 1909, others, of course, in the year 1908. Nearly 30 per cent. of all deaths are infant deaths; about 10 per cent. of all deaths are those of children of ages one to five years; about 50 per cent. of all deaths are those of children from birth to fifteen years. The least dangerous section of human life is between the ages of ten and fifteen years. Child mortality, extremely heavy during the year of infancy, diminishes greatly after the completion of the first year, and diminishes enormously after the completion of the fifth year. In the civilised countries of Europe, at the end of the nineteenth and the beginning of the twentieth century, of every 1000 infants, from 100 to 300 die, on the average, every year. The attainable minimum of infant mortality, under conditions practically realisable to-day, may be regarded as about 70 per 1000. In families in which exceptionally favourable conditions prevail, the infant mortality rate is even lower than the figure just stated; in the families of the higher aristocracy and among the royal houses it is as low as one-half, and even as low as one-third, of this “practicable minimum” of 70 per 1000. Since the middle of the eighteenth century there has, in nearly all the countries of Europe, been a decline in infant mortality, in the case alike of boys and of girls; this fall in the infantile death-rate is greater than the fall in the general death-rate. Although the data available justify this general statement, it is necessary to point out that authentic statistical data bearing on the question exist only in the case of small and isolated areas, such as individual towns; national registration of such particulars is of much more recent date than the middle of the eighteenth century.