51 
The Census of 1891 and Rural Depopulation. 
in the former, but only 101 to 100 in the latter. Up to 10 
years of age there is no difference ; from 10 to 15 the proportion 
of females becomes much higher in the urban than in the rural 
districts; between 15 and 20 it is 107 to 100 in the towns, and 
only 87 to 100 in the country ; between 25 and 45 the migration 
of males, setting in later than that of the other sex, reduces the 
inequality very considerably ; but, 45 past, the disproportion 
increases in the urban population until, in the last age-period — 
that of 85 years and upwards — the towns have 99 per cent, 
more women than men, while in the rural districts the dispro- 
portion increases indeed, but not at so early an age or by such 
rapid stages. In the last period it is only 41 per cent., as com- 
pared with the urban 99. Of this difference the Registrar- 
General furnishes two explanations. One is, that the conditions 
of town life as contrasted with life in the country are more fatal 
to men than to women of advanced age, and this explanation is 
supported by actual statistics. The other is probable rather 
than proven. It is that, as they grow old, men leave the towns 
and retire into the country more generally than women. They 
are probably incapacitated for their work at an earlier stage, 
while women can pursue their lighter occupations for a longer 
period of their working life. And so, alike at the beginning 
and at the decline of life, the females come sooner and remain 
longer in the towns, while the males leave the country later and 
find their way back to it at an earlier age. 
Among the remaining particulars of interest brought out by 
the Occupations Returns are a decline in the number of corn- 
millers from 23,462 in 1881 to 22,759 in 1891, and an increase 
of gardeners of 20 - 9 per cent, in the same interval, and of wood- 
men of 15 - 9 per cent.; while the Agricultural Returns exhibit 
for the corresponding period an increase of some 65 per cent, in 
the area used for market-gardens or nursery-grounds, and of 
less than 10 per cent, in the acreage of woods and plantations. 
V. Decrease in the Populations of certain Counties. 
The returns are, as we have seen, defective. But they 
suffice to establish certain broad conclusions. It is impossible to 
ignore the general drift of the evidence that they furnish ; and the 
conclusion to which they lead may be reached independently. 
In the earlier portion of the census report we are informed that in 
13 registration counties of England and Wales the population 
declined between 1881 and 1891. The counties were, with few 
exceptions, those which had also fallen off in numbers between 
1871 and 1881. In the earlier period Flintshire, Merioneth- 
