the Agricultural Schedule) filled up voluntarily by the occupiers 
of all holdings over one acre.* This method of obtaining direct 
returns in writing from the occupier is probably the most accurate 
and reliable method of obtaining information of this type— 
provided the occupier is willing to supply the information and to 
take a little trouble in doing so. No one is in a better position 
than the actual occupier to state the area under the different 
crops growing on the farm or the number of live stock, and he 
has no obvious reason for giving erroneous information. In 
general, the particulars are supplied very willingly, the number 
of actual refusals being relatively small. The results, moreover, 
are not materially affected by these refusals, as in most cases 
it is possible to get approximate information by local enquiry. 
Small errors must naturally arise from mistakes in measurement 
or other incorrect statements, but with the large number of 
returns these inaccuracies should tend to balance one another. 
The main difficulty in connection with the collection of these 
annual statistics lies in ensuring that all holdings are enumerated. 
The frequent changes in occupation, especially in recent years, 
the combination or division of farms, and the withdrawal of 
land for building or other purposes make the task of seeing that 
all the agricultural land in the country is accounted for a very 
serious one. A proportion of holdings do, in fact, escape enumera- 
tion, especially in the case of the smaller occupations under five 
acres, though the total area omitted from the returns in this way 
is a very small proportion of the agricultural area of the country. 
An essential feature of the Agricultural Schedule is that the 
information asked for is such as the occupier is well able to 
supply ; thus, the questions relate only to the acreage under the 
various crops on the farm, the number of live stock of different 
ages, the labour employed, or other facts which the occupier 
either knows or can easily find out. The questions, moreover, 
relate solely to the actual position on the farm at a given date, 
viz., June 4th each year. 
Estimates of production.—Annual estimates of the production 
of the principal crops were first collected in 1884. There are 
two methods by which knowledge of the total production of any 
particular crop can be obtained. Application can be made to 
the grower or an estimate can be made by a competent observer. 
A variation of this is the sampling method by which the production 
of numerous sample plots is ascertained each year and the 
production of the country is calculated on the basis of the results 
obtained from these sample plots. At first sight it would seem 
* The Agricultural Returns Act, 1925, made these returns compulsory 
but did not come into force until 1926. The annual returns were, however, 
compulsory under the Corn Production Act, 1917, from the year 1918 
until the Act was repealed in 1921. 
The information obtained in connection with the Census of Production 
has been obtained voluntarily on each occasion, as agriculture has not 
been prescribed as a trade under the Census of Production Act. 1906.