25 



causes separately; but as it is the results appear to be significant. 

 We may assume that the crop index would increase as the amount 

 of manure used increases within very wide limits if no disturbing 

 factor enters. But in this case the maximum yield appears to be 

 reached at a certain point, although the amount of manure used 

 still increases beyond that point. 



It is known that summer tillage has some tendency to promote 

 oxidation of organic matter in the soil, and thus to destroy humus 

 and set free nitrates in the soil. The oftener the land is put into 

 a tilled crop the greater this influence should be. Finally we reach 

 a point where oxidation is so rapid as to reduce the amount of 

 organic matter in the soil more rapidly than it is being added, in 

 spite of the fact that more organic matter is added than before, and 

 the crop index thus begins to fall. Unfortunately the variation in 

 percentage of intertilled crops stops just at the point where it would 

 be most interesting to know what the next few terms of the series 

 would be. However, we may tentatively conclude that when the 

 percentage of intertilled land increases beyond about 37 per cent 

 in the locality in question the tillage tends to lower the crop index, 

 while up to that point its influence is not deducible from these data 

 because whatever influence tillage exerts is complicated by the 

 influence of intensity of stocking. Presumably the two influences 

 are similar up to 37 per cent of tilled crop area. 



Fortunately, a survey of farms on the black prairie lands of 

 Illinois supplies the missing terms of the series of crop index aver- 

 ages in Table III. The data for the Illinois farms are given in 

 Table IV. 



TABLE IV. Influence of area of intertilled crops and animal units 

 on crop yield in Marshall silt loom Illinois. 



