A still better method of district-comparison, is to arrange in a series the death-rates prevailing in several localities for persons _over five years of age_, and side by side with this column, another for the death-rates of children _under five years of age_. The first column will of course indicate very well the relative sanitary conditions of the districts; but the differences between them will be expressed far more clearly, and, as it were, in a magnified form, in the column of infantine death-rates. Thus, for instance--to repeat the comparison just instituted between the Northumberland and the London district; the death-rate for all ages over five is about 12 in the former district, and nearly 15 in the latter; a difference quite sufficient to establish the inequality of their sanitary conditions. But, how much more strongly is this disparity expressed in the comparison of the infantine death-rates--26·5 for the healthier district, 107·57 for the unhealthier one! Nothing can be more conclusive than the evidence afforded by statistics, as to the dependence of high infantine mortality on the general causes of endemic unhealthiness. My own observation within the City gives complete confirmation to this view, showing me that the diseases specified in my table (diarrhœa, bronchitis and pneumonia, hooping-cough, croup and measles, hydrocephalus and convulsions) however various in nature they may seem, and however apt you may be to dissociate their occurrence from the thought of local causation, yet unquestionably multiply their victims, in proportion to the otherwise demonstrable unhealthiness of a place, owe most of their fatality to local causes, and may, therefore, to a great extent he disarmed of their malignity. The last column gives the total of those which have preceded it, and shows, out of 9493 deaths, 3923, all from acute disease, in intimate dependence on local and obviable causes. It will be a moderate computation with respect to these deaths, if we estimate that two-thirds of them might have been hindered. And yet it is not only by _acute_ disorders, that preventable death succeeds in ravaging the population. If we turn to the examination of _chronic_ ailments producing death, we may quickly recognise many indications of their preventability, and may satisfy ourselves that here also the general mortality might be very largely reduced. Look, for instance, at the whole immense class of scrofulous diseases, including pulmonary consumption, a class probably causing, directly or indirectly, at least a quarter of our entire mortality; and consider the vast influence which circumstances exert over its development.