Feb. n. 1918 Effect of Season and Crop Growth on Soil Extract 323 



by the well-known tumbler method. All the soils showed a satisfactory 

 capacity for ammonification. 



At the close of the season of 191 5, soils 12 and 14 were noticeably 

 lower in nitrifying power than the others of the group. At the close of 

 1 91 6, soil 12 alone showed a significant lower range in nitrification. With 

 these exceptions the soils gave results which indicated normally active 

 bacterial floras. The detailed data are not considered sufficiently il- 

 luminating to justify their inclusion in this paper. 



SOIL CONTAINERS AND INSTALLATION 



The soil containers selected were 30 inches wide, "60 inches long, and 

 18 inches deep. They were made of No. 24 galvanized iron and were 

 thoroughly coated with asphaltum varnish. When filled, each con- 

 tained approximately i ,700 pounds of soil. The design of the containers 



I X /4- iron s>{rao -y 



Perforated '/z' pipe for sub irrigation, 

 hofe-i 2" ap>ort. 



Fig. 1. ^Design of soil containers. 



is shown in figure i . It will be observed than an outlet is placed in the 

 bottom at one end. This outlet not only prevented the accumulation 

 of excess water, but also gave additional aeration to the lower soil. 

 The escape pipes may possibly be seen in the illustration of the wire 

 house inclosing the containers (PI. 14, A). 



Distilled water was the only moisture used at any time. This was 

 added partly between the rows of grain by a long-spouted sprinkling 

 can and partly by subirrigation through the two perforated pipes running 

 from end to end of the containers. So far as possible, only enough water 

 was added at one application to keep the soil at optimum moisture. It 

 was found possible to observe the moisture condition quite closely by 

 drawing out a core of soil with an i8-by-^-inch cheese trier, and then 

 replacing the soil. 



The containers numbered 28, 2 for each soil, and were arranged on 

 level mud sills in two duplicate sets, as indicated in the accompanying 

 diagram (fig. 2). By the above arrangement the smallest possible 



