640 
Journal of Agricultural Research voi xxm, no. s 
The large difference in the yield between the burned and unburned 
areas in 1918 is difficult to explain, since the quadrat chartings up until 
June 3 indicated considerably more shoots on the burned areas than on 
the unburned. This apparently should mean higher yields. It seems 
probable that in obtaining the cuttings a portion of the dried material 
remaining from 1917 was included. 
There was a slight difference in yield in 1919 and in 1920, in favor of 
the unburned plots, but it is too small to be considered of much import 
Wtt/r f/V 0 /A/G 
Fig. i.— Graph showing weekly mean maximum and minimum temperatures in 1918 at a depth of 1 inch 
under the surface on burned and unburned plots. 
tance. Conditions were unfavorable in 1920, and the grasses showed a 
poor growth. 
In 1921 the burned area produced the higher yield, the difference 
being 183 pounds per acre. Conditions were subnormal in the early 
season. There was not much rainfall early in the spring and cold weather 
persisted until June. Later on, however, good rains fell and grass 
showed excellent growth. The average difference of 386 pounds per 
acre in favor of the unburned is largely due to the 1918 results. Con¬ 
sidering the data for 1919, 1920, and 1921 only, it will be seen that the 
amounts of hay produced on the two areas are practically equal. 
The general opinion among those who are opposed to burning is that 
much more forage is produced on areas not burned. It is believed that 
this conclusion is erroneous, since the experiments here reported have 
