22 BULLETIX 991, IT. S. DEPARTMENT OF AGRICULTURE. 
low level is disease, of which the most important at Edgeley is rust. 
This generally occurs in wet years when yields would otherwise be 
relatively high. It was the cause of the low yields in 1916 and 1919. 
Aside from disease, the most important factor controlling yields 
at this station is the rainfall during the growing season. 
The precipitation for the months of April, May, June, and July 
for the several years was as follows : 1906, 14.03 ; 1907, 6.44; 1908, 9.18 ; 
1909, 10.50; 1910, 5.08; 1911, T.09; 1912, 15,63; 1913, 9.53; 1914, 13.23; 
1915, 13.81; 1916, 12.48; 1917. 7.54; 1918, 8.58; 1919, 13.45; average, 
10.47 inches. 
There is a close relation between these figures and those of yields 
shown in Table II, except in those cases where disease interferes with 
the production of a good crop by a sufficient rainfall. 
It will be noted that, generally speaking, a rainfall of over 9 inches 
during the growing season is necessary to the production of a good 
crop. 
CONTINUOUS CROPPING COMPARED WITH ROTATION. 
Considerable study has been devoted to the subject of changes that 
may be taking place in the relative yields of crops grown continu- 
ously on the same plats and those grown in rotation with other crops. 
The great fluctuations due to seasons and the relative response to 
methods of cultivation in different seasons tend to obscure results 
in even as long a series of years as that under study. There are, 
however, rather marked indications of comparatively decreasing 
yields under continuous cropping to any one small grain. This 
observation is not confined to this station alone, but is more or less 
general. After the first few years, from four to seven, on new land 
there appears to come a break in the relative yields from land con- 
tinuously cropped to one grain. The most obvious reason for this, 
and one that in some cases clearly accounts for it, is the development 
of weeds. Diseases that are propagated in the soil are probably an- 
other reason. It is not believed that it is due to any impairment of 
the soil. Another bumper crop year such as 1915 was will be very 
interesting on account of the evidence it will furnish on this subject. 
CONCLUSIONS. 
The results attending the use of barnyard manure, various green 
manures (leguminous and nonleguminous), sod crops, and a con- 
tinued and rather extensive test of commercial fertilizers which 
has been conducted at the station but is not considered in the present 
paper, all show that soil fertility is not a limiting or controlling 
factor of major importance in crop production at Edgeley. On the 
other hand, the seasonal variation in yields shows that the chief 
controlling factor is the seasonal rainfall. The full operation of its 
control is interfered with by plant diseases, of which the chief one 
not under control is rust. The nature of the soil and the amount and 
