106 Development and Activities of Roots of Crop Plants. 



mented by heavy, water-tight, cylindrical, galvanized-iron containers 2 feet 

 in diameter and 3.5 to 4 feet deep, while the one for corn measured 3 feet in 

 diameter and 5 feet deep. Suflficient soil to fill the containers was secured from 

 both the upland and lowland crop plats north of Lincoln (p. 40) on February 

 20 and stored in large piles in a dry place until March 30-31, when it was placed 

 in the containers. These soils were chosen because they were from the same 

 plats as those used in the preceding experiments and both their mechanical and 

 chemical composition had been determined (tables 8 and 9). Care was exer- 

 cised to keep the soils from the several levels to 4 feet separated. When 

 taken from the field they were in such excellent condition regarding water- 

 content that it was unnecessary to add more water when working them over to 

 break up lumps, etc. During storage they were kept well covered to reduce 

 water-loss. 



FiQ. 41. — Average daily temperature (long broken lines), 

 humidity (upper solid line), and average daily 

 evaporation at Lincoln, 1921. 



On March 30-31 a large trench was dug in the field where the 1921 crops 

 were grown (p. 80), and the containers, 16 in number, placed close together in 

 a row in such a manner that the tops were about 4 inches above the surface 

 of the soil. The prepared soil from the deeper levels was then poured slowly 

 into the containers, where it was tamped into place. When filled, the soil in 

 each container (with certain exceptions to be noted) occupied the same relative 

 position as regards depth that it had occupied in the field. Barley was grown 

 in the lowland soil; potatoes in that from the upland. 



Sodium nitrate was added at certain sealed levels at the same rate as in 

 the preceding experiment (viz, 400 p. p. m. of NaNOs based on the wet weight 

 of the soil). In this series, however, the proper amount for the weighed soil 

 at any level was dissolved in about 0.5 hter of water, which was poured upon 

 the soil after it had been placed in the container. This facilitated rapid 

 diffusion, and it seems probable that within a period of 2 weeks the nitrate was 

 rather uniformly distributed throughout the soil-mass. Triplicate samples 



