6 BULLETIN 339, U. S. DEPARTMENT OF AGRICULTURE. 
because there have sometimes been great variations in the yields 
produced by the same quantity of water on the same crop upon — 
adjoining farms. These variations have made it very evident that 
dependable results can be secured from agricultural experiments 
only by basing conclusions upon the average of results from a large 
number of experiments. 
WEATHER CONDITIONS IN IDAHO DURING THE INVESTIGATION. 
It has not been considered practicable, although it would have 
been desirable, to install a rain gage in connection with each tract 
ao 
¢ 
R 
SX og 
‘f 
x 
8 
oN 6 
Sy 
“ 
N 
v 
. 
— Average/tnial. | for 4yor fo Saat. the. 
Fic. 1.—Comparison of average annual precipitation with that occurring during the growing season. - 
experimented upon. The United States Weather Bureau has a 
large number of cooperative observers’ stations scattered quite 
aeionia throughout the territory involved in the investigation, 
and the precipitation recorded at the station nearest each tract 
has been used. This may be slightly inaccurate, but the errors 
are compensating and small at the most. 
The normal annual precipitation in the districts of southern 
Idaho which have been investigated ranges from slightly less than 
10 inches per annum at Ouldlie to 17.75 inches at Hailey. Hailey 
is situated at a comparatively high altitude, and the precipitation 
in this place is consequently bem the average for irrigated southern 
Idaho. The average precipitation in the Src eared districts 
