2. London is very largely a city of wealthy and well-to-do people, most of whom must be looked upon as sojourners rather than dwellers in the city. Among such as these, who can command every luxury and necessary of life, including change of air, death-rates ought to be low. It is manifestly unfair to contrast the death-rate of St. George’s, Hanover Square, or Kensington, with the death-rate of a town packed with the wage-earning class. 3. The mobility of the London population is so great that it must vitiate any statistics bearing on the health of the inhabitants. “Londoners” are a mixture of races, recruited from every clime from China to Peru. They are, as the phrase goes, “Here to-day and gone to-morrow,” and probably no one fact quickens their departure more than ill-health. I am told by the proprietor of Kelly’s Post Office Directory that the annual correction of addresses amounts to about ten per cent. of the whole, so that the London population shifts on an average completely every ten years, even among classes who have far more stability than the labouring classes. It is also well to point out that these changes in the Directory do not represent all the changes, because in trade it is common for new individuals to trade under an old and established name. I find, on comparing the Directories of 1880 and 1889, that in my own street of 96 houses there have been 87 changes of names, and that 96 houses are now credited with the addresses of 140 individuals, whereas in 1880 the individuals numbered 120. 4. Still more important, as vitiating the value of the “death-rate,” is the abnormal age distribution in London. In London (and especially in the central portions of it) there is a great deficiency of young children and old people, among whom the death-rate is always highest; the population of London is largely composed of selected adults imported from the country, among whom the death-rate ought to be low. 5. The continued low death-rate of London is very largely accounted for by the diminishing birth-rate. Thus the birth-rate for the ten years 1877-86 averaged 34·4 and the death-rate 21·2, while for the year 1887 the birth-rate was 31·6 and the death-rate 19·5. This is a diminution of 2·8 per 1,000 of population in the birth-rate. This, in a population of 4,250,000, means a deficit of 11,900 children; and as out of every 1,000 children born in London in 1887, 158 died before they were one year old (_i.e._, 13 per 1,000 more than in England as a whole, and 66 per 1,000 more than in the county of Dorsetshire), it is evident that this diminution of the birth-rate entails a deficit of 1,940 in the total deaths occurring in London in the year. It is clear from this that in taking account of a diminishing death-rate we have to take into consideration the diminishing birth-rate also.