HARD RED WINTER WHEATS IN DRY AREAS 9. 
EXPERIMENTAL METHODS 
The results reported here were obtained in field plats ranging. 
from one-hundredth to one-tenth acre in area at the different sta- 
tions or during different seasons. Nursery experiments were con- 
ducted in rows at most of the stations for the preliminary testing of 
new varieties and selections, but are not reported here. The experi- 
mental conditions, such as crop sequence, size of plat, width of alleys, 
and number of replications, vary somewhat at different stations. The 
results obtained at one station, therefore, are not necessarily com- 
gems with those obtained at any other station. In many cases, 
owever, they probably are directly comparable. The results from 
different varieties at the same station were nearly always obtained 
under similar conditions and may be compared directly. Any known 
exceptions to this fact arestated. In all cases the crops were grown 
with only the natural rainfall and under conditions approximating 
good farm practice for the district. Many of the experiments were 
conducted on summer-fallowed land. Summer fallowing is the com- 
mon method of seed-bed preparation for winter wheat in the Great 
Basin and frequently is practiced in the Great Plains. At some sta- 
tions during recent years two plats of each variety have been sown 
on fallow and two plats on corn ground or fall-plowed stubble. 
NUMBER AND SIZE OF PLATS 
Previous to 1912 nearly all varieties were grown in single plats, 
and check ay of a standard variety were sown at intervals of about 
every third or fifth plat. In this bulletin the yields from the single 
plats are not computed to the average of the checks, and the actual 
yields only are ay ae Most of these tenth-acre plats were 2 by 
8 rods in area. Since 1915 nearly all experiments have been repli- 
cated by sowing two to five plats of each variety. The number of 
replications may have varied from year to year and at different sta- 
tions. The replicated plats were mostly the width of the grain drill 
and 8 rods in length, or about one-fortieth or one-fiftieth of an acre 
eachin area. In the textual discussions of the yields of the varieties 
at each station, the number of plats sown to each variety is stated. 
INTERPRETATION OF THE RESULTS 
In the tables showing the results obtained at the different stations 
the actual acre yield of each variety is reported for each year in which 
it was grown. Where the plats were replicated the recorded yields 
are the averages for all plats of a variety. Probable errors of these 
averages, or means, of the results from replicated plats are not given. 
The yields were tabulated from the annual reports of the field-station 
men previously mentioned. 
Because of the great variability of the yields of wheat from year 
to year the average yield of a variety for the entire period grown 1s 
the most important. The yield of each variety has been compared 
with the yield of Kharkof each year, and the average plus or minus 
difference is recorded. The probability of these differences recurring 
is. expressed in odds obtained by the ‘‘Student” * method for deter- 
mining the probable error of the mean, as follows: 
Mean - Standard deviation = Z 
2 Student, The probable error of a mean. Biometrica 6: 1-25. 1908 
96528°—25t 2 
2 
" 
ee eer ere rere Be) i 
