The Census of 1891 and Rural Depopulation. 45 
the means of drawing such a picture of the occupational dis- 
tribution of the people as shall be fairly true in its main lines, 
though little value can be attached to the detailed features. It 
is not wise to demand from a material a result for the produc- 
tion of which it is unsuited.” Accordingly, the Registrar-Gene- 
ral holds that the endeavour, made for the first time in the present 
census, to obtain statistics of the numbers of employers and 
employed has led to anything but satisfactory results. This 
distinction — adopted in deference to the suggestion of econo- 
mists — does not, perhaps, possess any great importance for agri- 
culture ; but the general difficulties which confront the attempt 
to attain accuracy do not fail to present themselves. It is only 
broad conclusions which can be drawn with any certainty, and 
these, perhaps, are mainly useful because they corroborate evi- 
dence furnished from other sources. 
The Agricultural Class is combined in the census returns 
with the fishing industry, but the Registrar-General observes that 
“ the latter is too small to affect seriously the total.” The sub- 
divisions of the class have undergone valuations in different cen- 
suses, and, in drawing a comparison between the figures of one 
and another, it is desirable to confine the attention to those sub- 
divisions which are similar. Happily, they represent the more 
important constituent elements of the agricultural community. 
In Tables I. and II. (p. 46) are contained the classification 
adopted in the last three censuses of the various subdivisions, 
and the figures for the last five censuses of the total of the agri- 
cultural order — of the farmers, of their male relatives, of the 
bailiffs, and of the labourers. 
From the first of these tables it will be seen that the changes 
made in the subdivisions of the agricultural class proper in the 
last three censuses do not vitiate or obstruct comparison between 
the numbers at the respective periods of the farmers and 
graziers, of their male relatives living with them in the house 
and returning no other occupation, and therefore presumably 
engaged in farming, of the farm bailiffs, and of the agricultural 
labourers and shepherds. For all practical purposes these classes 
constitute the vast majority of the agricultural population, and 
the broad significance of the figures relating to them can hardly 
be misinterpreted. 
The agricultural order as a whole has 'fallen steadily since 
the beginning of the period comprised in Table II. In 1891 
it was composed of 1,336,945 persons. These, the Registrar- 
General states, were “ almost exclusively males,” and “repre- 
sented 6T per cent, of the population” of the age of ten or 
upwards. In 1881 the class contained 1,383,184 persons, and 
