.328 H. W. Turpin 
dioxide production in the cropped soil. Since only about a quarter of. 
a pound of water was lost in a week from the uncropped soil, tap water 
was used in all cases, as the small loss by evaporation could not possibly 
introduce a disturbing element in the form of an accumulation of soluble 
salts in the soil. 
Results 
On each date of sampling, the samples were taken in duplicate from 
each of the eight cans. Thus eight samples were obtained from the 
cropped soil and eight from the bare soil. Since all of the four cropped 
cans were treated in identically the same manner, the data for the duplicate 
samples from the cropped cans were averaged. This was done also in 
the case of the bare soil. 
It seemed fair to average the data obtained from the cans in each set 
because in all cases the differences were small. This is shown by the very 
small probable error. The data for the.oat crops of 1917 and 1918 are 
given in tables 1 and 2 (appendix, pages 353 to 356), each figure for carbon 
dioxide in these two tables being the average of eight determinations. 
These summarized results are represented diagrammatically in figures 
46 and 47. 
Effect of crop 
The content of carbon dioxide at the beginning of the experiment was 
0.28 per cent by volume for both cropped and uncropped soil. From that 
time on, as may be seen from figures 46 and 47, the amount of carbon 
dioxide in the uncropped soil in no ease reached that in the cropped soil — 
not even after the removal of the crop. The latter point may perhaps be 
explained by the fact that since the roots of the crop were not removed 
from the soil at harvesting, they somewhat increased the available supply 
of organic matter. The results reported here are directly opposite to 
those of Bizzell and Lyon (1918), who worked with the same Dunkirk 
clay loam under field conditions and found that subsequent to the removal 
of the oat crop a marked decrease in carbon dioxide below that in the 
uncropped soil took place. This was not found to be the case, however, 
with the Volusia silt loam used by these investigators. 
A study of figure 46 shows that in the season of 1917 there was a marked 
increase in the carbon dioxide in the cropped soil from the beginning 
of May, amonth after seeding, until the maximum, 2 per cent, was reached 
