504 Journal of Agricultural Research voi. xiv, no. n 
The results with sorghum, cotton, and broom corn being from a single • 
station can scarcely be compared with those obtained with other crops. 
The grand average ratio of all crops at all stations is 97, with a probable 
error of 0.9. This average, of course, is meaningless in its possible 
application to any crop or any station. It would be influenced by the dis¬ 
tribution of the observations among conditions that were either favorable 
or unfavorable in their response. The only purpose of introducing it here 
is to show the relative lack of any effect, and particularly of any favorable 
effect, of the practice of subsoiling as applying to a wide territory and a 
wide range of crops. 
This lack of effect of the practice when applied generally to all crops 
and to the entire area is further confirmed when the results are studied 
in another way. Exclusive of years of total failure there are 353 trials 
here reported. In 15 of these there was no difference in the yield from 
the two methods, in 153 cases the higher yield was obtained from the 
subsoiled plot, and in 185 cases from the plot not subsoiled. 
COMPARATIVE RESUETS WITH SUBSOIUNG IN FAVORABLE AND UN¬ 
FAVORABLE YEARS 
The relative effect of subsoiling in favorable and unfavorable years is 
a question that naturally arises. There have been a number of cases in 
which the crops by both methods have been a total failure. There have 
been some cases in which the plot not subsoiled produced a small crop 
when the subsoiled plot was a total failure . - There has been no case in 
the history of the expenments when the reverse was true. In order to 
obtain definite information on the subject the results were divided into 
two groups; one group containing the ratios from each station for those 
years when the yield was above the average for that station and the other 
group containing the ratios of those years when the yield was below the 
average. Those groups showed average ratios of 100 for the years of 
better production and 94.4 for the years of poorer production. 
In the years above the average in production 75 trials resulted in favor 
of subsoiling, 81 trials in favor of ordinary plowing, and in 8 trials the 
same yield was obtained from each method. In the years below the 
average in production 78 trials resulted in favor of subsoiling, 104 in 
favor of ordinary plowing, and 7 showed no difference in the yields from 
the two methods. 
These results indicate ihat } on the average , subsoiling , instead of over¬ 
coming the effects of drouth , actually intensifies them . In this con¬ 
nection it should be recognized that, while low yields are in some cases 
caused by fungus diseases, by insect attacks, or by unfavorable temper¬ 
ature or other weather conditions, the one primary predominating cause 
of low yields has been the lack of sufficient soil moisture at some time 
during the growth of the crop. 
