29 
dred times as much water lias been evaporated from the leaves. 
Some variation is to be expected in different kinds of plants, and in 
the same kind under different conditions; but as a whole, the varia¬ 
tion is less than might be supposed. The subject has been experi¬ 
mented upon by many. According to the;experiments of Hellreigel, 
quoted by Storer (Agriculture, 1;14), the cereals require more than 
the leguminous crops, weight for weight; and of the cereals, oats 
more than wheat, the amount varying from 376 pounds for oats to 
273 for peas, for each pound of dry matter produced. If the ratio 
be considered as three hundred to one for alfalfa, then to produce 
three cuttings of 3,000 to 4,000 pounds each per acre during the 
season, a depth of from 12 to 16 inches must have been available 
for transpiration from the leaves for the growth of the crop. There 
is, in addition, some growth of the roots in the soil which would re¬ 
quire an additional amount of water. If this growth be considered 
as confined to five months, say 153 days, as used in some of the pre¬ 
ceding tables, the net duty of a cubic foot per second could not ex¬ 
ceed 225 acres, even if there was no waste. Such water as the plant 
secures from other sources than irrigation, as rainfall, from ground 
water, etc., would lessen proportionately the amount which it would 
be necessary to supply by irrigation, but when it is considered that 
this does not allow for any of the losses by evaporation or seepage 
from the ditches, evaporation from the surface of the soil, or losses 
by downward filtration, which in the aggregate are large, we may 
consider such a duty as exhibiting good, if not the best practice. 
Bringing these results together for comparison, there results the 
following table, showing the depths which, on the average, have 
been applied to the lands under the two canals under consideration, 
and to the valley as a whole. The second part of the table gives 
the corresponding duties. There is some water applied once to the 
lands which returns to the river. In the average it seems to equal 
to the flow of ninety cubic feet per second between the measuring 
flume in the canon and the last canal taken from the river, as based 
'on a series of measurements taken in the fall, by the State Engineer’s 
department, and by this section of the Experiment Station, and by 
one determination in the spring by this section. There are reasons 
to think it is approximately constant. This water is again taken from 
the river by the various canals, and should be counted with the water 
which is applied to the land of the whole valley, for unless this in¬ 
flow could be utilized, it is evident that a correspondingly greater 
amount of water would be needed to enter the canon, or else the 
acreage in crops would be correspondingly less. Counting the re¬ 
turn as ninety cubic feet per second for the time covered by the 
table, the duties of a cubic foot per second as measured at the heads 
of the canals, is given by the numbers in the line where seepage is 
included. 
