6 4 
Journal of Agricultural Research 
Vol. XXIII, No. a 
effect of nitrogen in preventing yellow-berry was shown clearly in spite 
of all the disturbing factors, one would be justified in assuming that the 
effect of these chemical groups was inconsiderable, if any. This would 
seem to settle the practical question as to the possibility of preventing 
the characteristic coloration of the grain associated with yellow-berry 
by means other than nitrogen. However, the theoretical question as to 
the nature of the coloration of flinty grain and of that affected with 
yellow-berry remains unsettled. It is still possible that the coloration 
is an indicator reaction in response to physiological alkalinity and acidity. 
All the ammonium salts probably underwent nitrification before they 
were assimilated by the plants. Their effect, therefore, would be the 
same as that of nitrates. It is possible also that the brown color of the 
flinty kernels is caused by the action of the alkaline amino groups which 
might increase with the general increase in the nitrogen content on the 
hypothetical natural indicator present in the hull. 
WEIGHT PER BUSHEL AND WEIGHT PER 1,000 KERNELS 
No correlation can be observed between the different treatments, 
which resulted in different nitrogen contents, and the weight per bushel 
and the weight per 1,000 kernels (Table III). Taking the plots in groups 
of three for each of the nitrogen carriers, it would seem that those which 
had received their nitrogen in the first stage gave a somewhat lower 
weight per 1,000 kernels than those which had received it in the second 
and the third stages. The first stage, however, is not the one which gave 
the highest protein contents. These results, as well as those obtained in 
Kentucky, 6 fail to disclose any direct relation between the protein 
content and the weight per bushel and, especially, the weight per 1,000 
kernels. Cases are on record, however, in which such a correlation has 
been reported. Table IV, compiled from data reported in Bulletin 128 
of the Bureau of Chemistry, United States Department of Agriculture, 7 
will serve as an illustration. This table shows a reverse correlation 
between the nitrogen content and the weight per 1,000 kernels and weight 
per bushel for Kansas and South Dakota, but no correlation for Cali¬ 
fornia, as was the case in the experiments here reported. 
What is the explanation of these seemingly contradictory results? 
Within the limits of the bureau’s experiments^ it has been established 
that the presence of nitrogen at the time of heading is conducive to a 
high nitrogen content. The conditions of these experiments would 
suggest the distinction between a physical and physiological abundance 
of-available nitrogen in the soil. The term “physiological abundance” 
is used to signify a condition in which the crop, being depressed in its 
development, is kept from drawing upon the store of available nitrogen 
in the soil to the full normal extent. Any environmental conditions 
which depress the normal development of the crop cause a physiological 
abundance of available nitrogen. Under such circumstances, the factors 
which are responsible for the abundance of available nitrogen are re¬ 
sponsible also for the subnormal development of the grain. Hence, the 
coincidence of a high protein content and a low weight per 1,000 kernels 
and a low weight per bushel. In these experiments, the abundance of 
8 Davidson, J., and LeClErc, J. A. op. cit. 
7 LeClerc, J. A. tri-local experiments on the influence of environment on the composition 
of wheat. U. S. Dept. Agr. Bur. Chem. Bui. 128, 18 p. 1910. 
