238 WHEAT PRODUCTION IN NEW ZEALAND 
example, with reference to stock, it is assumed that one- 
third of its use is devoted to producing this particular 
crop. Manager’s wages are over-estimated, as is also 
horse-feed in two eases. But to be in error on this side 
is a matter of small consequence for our purposes. 
An average of the three estimates is probably the best 
figure from which to draw conclusions. The simple 
arithmetic average is 3s. 6d. per bushel, but the weighted 
average is more important. Weighting according to the 
number of bushels the average is 3s. 5d. It is not that 
there is any actual case in which the cost of production 
is this amount, but the statement merely implies that 
the average cost of production is 3s. 5d. per bushel. 
Land C would be considered as marginal land, but the 
estimated cost. of production here is too high on account 
of a relatively low yield in the particular year. On this 
a conservative estimate of the yield would be 24 bushels 
per acre, when the cost would be 3s. 6d. per bushel. 
Our investigations then lead us to the conclusion that 
the average marginal cost of production is 3s. 6d. per 
bushel. Now, since the average market price for the 
ten years 1904-13 was 3s. 7d., it is obvious that unless 
a higher price can be maintained, the margin of culti- 
vation will not be extended further. 
On the higher grade lands the cost of production is 
slightly above 3s., and very often below this amount. 
Price for the most part is quoted at the nearest railway 
station, and the purchaser bears the cost of carriage. 
This leaves a clear profit of some 6d. per bushel for most 
land above the margin. 
9, Conclusion. 
The consideration of the importance of these estimates 
in forecasting future tendencies in the wheat industry 
is left till the concluding chapter. It was for this 
purpose, primarily, that the investigation on these par- 
