52 The Census of 1891 and Rural Depopulation. 
shire, and Anglesey, had increased, while in the later they had 
diminished; but, on the other hand, Westmoreland, Cambridge- 
shire, and Dorsetshire, which had declined between 1871 and 
1881, exhibited some increase between 1881 and 1891. All 
the counties were agricultural, and comprised eight in Wales, 
those of Shropshire and Herefordshire on the Welsh border, to- 
gether with Huntingdonshire, Rutlandshire and Cornwall. In 
the last case, no doubt, the decline may be attributed largely to 
the decay of the mining industry. 
This actual decline of population in certain counties is one 
piece of evidence pointing to rural efflux furnished by the 
census. A second is to be found in a comparison of the urban 
with the rural population. The population of all the “urban 
districts,” so called, in 1891 amounted to 20,895,504 persons, as 
compared with 8,107,021 in the “rural districts.” The propor- 
tion of urban population had accordingly increased on this basis 
from 212 to 100 in 1881 to 258 to 100 in 1891. The Registrar- 
General, however, observes that this comparison is rendered 
somewhat fallacious by changes of area in some urban districts, 
and by the creation of others. Accordingly, adopting a fresh 
mode of measurement, and regarding the areas as urban in 
1881 which were so in 1891, and then comparing the growth of 
population in the interval, he finds that the urban population so 
distinguished had increased by 15 '4 per cent., while the remain- 
ing or rural population had only grown by 3 per cent. 
But, he proceeds to urge, even this mode of comparison is 
hardly fair. Districts technically urban may be really rural in 
character. The small towns in the rural districts belong to the 
rural whole, although any line of distinction between them 
and the really urban towns must be drawn with a somewhat 
arbitrary hand. Two rough standards of measurement, how- 
ever, may be employed. We may include in the rural divi- 
sion all districts with populations below 10,000; or we may 
draw the line lower and only include those with less than 
5,000. Adopting the first standard, the urban population 
amounted to 17,920,210 in 1891 as compared with 15,382,403 
in 1881, and the rural population to 11,076,315 as compared 
with 10,592,036. The one had grown by 16-54 per cent, and 
the other by 4-57. Applying the other standard, the increase 
in the former would be 16-05 per cent., and in the latter 3-29 ; 
while an adherence to the strictly technical classification would 
show an increase of the urban population of 15-4 per cent., and 
of the rural population of 2 98. 
The results of these different methods do not themselves 
differ greatly ; and they all show that, tested in this way, there 
