In all England and Wales, then, the proportion of women who may be expected to remain unmarried is, roughly speaking, one in six; in London it is one in five. The important question arises, Are these chances equally distributed? On the face of it, it would seem not; but people readily point out that the greater ratio of middle-aged spinsters in Kensington, Hackney, and Islington, as compared with Shoreditch or Bethnal Green, is easily explained by the number of servants who naturally, if unmarried at this age, congregate in the richer districts, but would, if distributed among the working-class districts, make the ratios fairly equal. The explanation sounds so plausible, that, were it not that experience has convinced me that in the educated middle class there is a surplus of women over men above the average, I should have accepted it without further inquiry. But by a study of the Census for 1861 (in many respects an ideal one so far as the tabulation of facts is concerned) and of the unpublished official returns of 1881 for Shoreditch, Bethnal Green, Whitechapel, Hackney, and Kensington, I find that, supposing all the middle-aged indoor domestic servants to be single, they nevertheless are not more than one-third of the single women in each district. Of the outdoor domestic servants, such as charwomen, the percentage under 25 years of age is so very small that it may fairly be assumed that the great majority are married women or widows, and that the exceptions to this rule will be balanced by the exceptions to the rule that the middle-aged indoor domestic servants are single women. Shoreditch and Bethnal Green (with almost exactly equal populations) give us together a ratio of 11·6 unmarried women between 35 and 45 to 100 married women at that age as the normal for a working-class district without any upper middle class. Kensington (including Paddington), with a population of 270,000, contains 70 per cent. of working-class inhabitants; the surplus women, whether servants or otherwise, are to be found in the houses of the 30 per cent. of middle and upper-class inhabitants. Roughly speaking, then, to every 70 working-class married women in Kensington we may assign 8 unmarried women, and to the remaining 30 married women between 35 and 45 years of age we must assign 54 unmarried women. To every 76 working-class married women in Hackney we may assign 9 unmarried women at this age-period, leaving 18 unmarried women to the remaining 24 married women. One-third of these being domestic servants, if we subtract them, we have left in Kensington in Classes G and H 36 unmarried women to 30 married women, and in Hackney 12 unmarried women to 24 married women. It follows, therefore, that in Kensington, excluding domestic servants, more than 50 per cent. of the women between 35 and 45 in the servant-keeping classes are unmarried, while in Hackney about 33 per cent. of the same class are unmarried.