284 
Journal of Agricultural Research 
Vol. I, No. 4 
which they grew (see Table I) and a dissimilarity in groups arranged by 
the source of soil (see Table II). The averages by localities (see Table 
III) differ greatly, being 86, 99, and 35 per cent for California, Kansas, 
and Maryland, respectively. The averages by soils are very uniform, 
being 71, 69, and 85 per cent, respectively. (See Table IV.) 
These averages do not show the great variations actually found in the 
different regions in any one year, for seasonal variations of the individual 
localities tend to equalize the averages. In Table I, for example, while 
the samples grown in California and Kansas in 1910 in each of the three 
soils were for the most part flinty, those grown in the three soils in 
Maryland were all more or less starchy or mealy. Similar figures are 
noted in 1911, when the Kansas samples grown on all three soils yielded 
wheat which was practically 100 per cent flinty, while on the same soils 
in Maryland the percentage of flinty kernels was less than half as great. 
CHEMICAL CONSTITUENTS 
In considering the composition of the wheat it will be. seen that many 
of the organic and inorganic constituents undergo as great variations as 
have already been noted with respect to the physical characteristics. 
On the other hand, there are a number of these constituents which 
showed very little variation, or no regularity in such variations as exist. 
Among those showing but little variation may be mentioned the gliadin 
number and the potash in the ash, and among those showing no pro¬ 
nounced regularity in the variations are the fat, fiber, pentosans, and 
sugars. With those exhibiting variations of a regular character belong 
particularly the nitrogen and protein, the ash, the phosphoric acid, and 
the phosphoric acid in the ash. 
PROTEIN 
As the protein of wheat is its most important constituent, it will be 
of more than usual interest to note the changes produced by difference 
of soil and by change of environment. As a rule, there was a remarkable 
uniformity each year among the samples grown in any one locality, 
independent of the soil upon which they grew. Thus, in 1910, 1911, and 
1912 the protein in wheat grown in California was almost uniformly 
low, about 13 per cent; in Maryland it was also low, about 11 per cent; 
while in Kansas it was high, nearly 18 per cent. This fact is more 
clearly brought out in Table III, which shows the average protein con¬ 
tent to be 13.11, 18.83, an d 12.43 per cent for California, Kansas, and 
Maryland, respectively. 
In Table IV, where the results are arranged according to source of 
soil, it will be seen that the wheats grown on California soil in all three 
localities had an average protein content of 13.88 per cent, those grown 
on Kansas soil, 13.94 P er cent, and those on Maryland soil, 15.44 P er cent. 
This shows a rather striking uniformity and again emphasizes the rela- 
