14 



ClRCULAB NO. Ill, BUREAU OF PLANT INDUSTRY. 



3 acres, while an equal area of Columbia produced less than 1 baJe. In 

 L912 both fields were planted to Durango cotton, which produced an 

 excellent crop on the same land where the Columbia had failed the year 

 before. The behavior of the Columbia hi a near-by experimental 

 planting was much the same in 1912 as in 1911. Even where the 

 plants had ample exposure instead of field conditions the yield was 

 altogether inferior to that of adjacent plants of the Durango cotton. 

 (Fig" 2.) 



Fig. 1.— Durango cotton plant at El Centro, Cal., with leaves removed to show habits of branching. 



Comparisons between some of the best plants in the two rows showed 

 over three times as many open bolls on the Durango plants and a 

 general yield of about twice that of the Columbia. These contrasts 

 appeared the more remarkable because the Columbia is a very pro- 

 lific variety in the South Atlantic States. 



The differences of behavior can be understood when the habits of the 

 two varieties are compared in detail. The failure of the Columbia 

 cotton to develop a large crop of bolls under the Imperial Valley con- 

 ditions is evidently connected with the tendency to produce a dense 

 canopy of leaves. This is quite different from the behavior of the 



