Sept. 9,1918 Deep Tilling and Dynamiting in the Great Plains 487 
and decreases, so this method is more satisfactory in the averages as 
determined by it than might at first be thought possible. It is this 
method that has been adopted for presentation. With each pair of 
comparable yields in Tables I to XII is given the percentage of the yield 
of plot E to the mean of the yield of B and E. These percentages are 
averaged at the right, and the probable error of this average is given. 
In Table XIII these average ratios and probable errors are assembled 
and further averaged by crops and by stations. The data in this table 
are presented graphically in figures 2 and 3. 
In these tables and charts a percentage of 100 shows that there was no 
difference in the yield of the two plots, a percentage above 100 shows 
that the higher yield was from the subsoiled plot, and a percentage below 
100 shows that the higher yield was produced on the plot not subsoiled. 
It is recognized that the use of ratios weights the results, and that aver¬ 
ages of such ratios are not the same as the ratios of the averages. They 
are not, therefore, accurate quantitative expressions of the results and 
are not a measure of the economic value of a method as compared with a 
control. They are not, however, in the present study misleading in 
direction, and serve a useful purpose of facilitating comparisons between 
things otherwise difficult of direct comparison. 
JUDITH BASIN TIEUD STATION 
The field station at Moccasin, Mont., in the Judith Basin, is located on 
a heavy clay soil of limestone origin. The soil is apparently very rich 
in available fertility. It is underlain at a depth of approximately 3 feet 
with a limestone gravel that is closely cemented with lime materials. 
The gravel subsoil, which extends to a depth of about 30 feet, is practi¬ 
cally free from soil. While it is so closely cemented that it does not 
unduly drain the soil, it is not of a character to allow the storage of avail¬ 
able water or the development of roots within it. 
Comparative results of fall plowing and of subsoiling are available for 
spring wheat, winter wheat, oats, barley, corn, and flax for the seven 
years 1910 to 1916, inclusive. Crops were raised on this land in 1908 
and 1909, but the first subsoiling was not done until the fall of 1909. 
Hail in 1912 destroyed all crops except winter wheat and flax. The 
yields of these crops were considerably reduced by the storm, but as the 
damage was relatively the same on each plot, the yields are used. The 
winter wheat crop of 1916 was lost by winterkilling. 
The E plot for each of the crops was subsoiled in the fall of 1909, 1910, 
1911, 1913, and 1915. 
The yields and the ratios as previously described are presented in 
Table I. None of the average differences observed are significant, the 
departures from the mean being very small and either less or only slightly 
greater than their probable errors. This is true of all crops except bar¬ 
ley, which, from the results at this station alone, would appear to be 
