measurements, and from latest obtainable data, or from records in the offices of the city engineers of the respective cities. Fall River is an exception to this rule, as the boundaries of Wards 6 and 9 in that city have never been accurately defined. The city of Washington, in the table, includes the area and population inclosed within the actual municipal boundaries, and not the total area and population of the District of Columbia. The islands in the East River, with an area of five hundred and twenty acres, and which are geographically situated in Wards 12, 19, and 23, are included as part of New York.
The most interesting feature of the foregoing table is that relating to the distribution of population according to area; but in this one must not be deceived. The population to each acre or to each square mile of a city can not well be compared with like data for another city, unless the exact area of dense population is known—as, for instance, a city may comprise fifty square miles of territory and have 500,000 population, which would give a population of 10,000 to each square mile, but the population may be compressed into twenty-five square miles, when the actual distribution would be 20,000 persons to the square mile; while another city of like area and like total population, but with the population distributed more evenly over the whole area, would be in a much better sanitary condition than the first city named, although in statistics the population per square mile would be the same when the whole area is considered.
Twenty-two of the cities named in the foregoing table have a
population of over 100,000 each, the total being 8,737,648, which is 13'95 per cent of the total population of the country. The population to the square mile of these twenty-two cities is 15'92 to the acre; but the differences in ratios of population to area are very great, ranging from four in St. Paul, five in Minneapolis, nine in Omaha, ten in New Orleans and Buffalo, eleven in Chicago and Denver, and twelve in St. Louis, to thirty in San Francisco, thirty-one in Washington, forty-eight in Brooklyn, and fifty-nine in New York. These figures represent population to a square acre. So skillful a statistician as Dr. Billings is of course careful to remark that the ratios indicated by the statistics published "give no information as to the difference in density of the population in the actually built-up portions," and he cites that in New York the number of persons per acre ranges from four hundred and seventy-four in Ward 10 to three in Ward 24, while in Chicago there is a range from one hundred and sixteen in Ward 16 to two in Wards 28 and 23. These instances show the extremes, and teach us emphatically that any comparison of population to the square acre or to the square mile for the purpose of drawing conclusions relative to sanitary and other conditions must be
VOL. XL. — 42
