34 
Colorado Experiment Station 
titled to consideration. The plants from the three check plots show 
quite as great variations among themselves as we find between these 
and the phosphorus and potassium plots. It must be remembered, 
however, that there is no compelling reason why plants grown on 
three abutting check plots should be identical in composition, es¬ 
pecially if our contention that ditferences in the supply of plant 
food produces ditferences in the composition of the plant, and we 
may anticipate a little by adding, if its seeds, too, be correct. We 
have shown by experiment what we all know, that soil varies in the 
amount of total and of nitric nitrogen from foot to foot. (See Part I 
(Bulletin 208) p. 26.) We have further given, in former publica¬ 
tions on the subject of Fixation of Nitrogen In Our Colorado Soils, 
instances of very great variations of nitrates within a few feet, var¬ 
iations of hundreds or even thousands of parts per million. Identity 
of results cannot be expected, but larger and persistent dilferences, 
such as we find occurring in all of the series of. plants presented in 
Part I, and in this one. are founded on something more than acci¬ 
dent; they have some general and sufficient cause. 
Further, while the etfect of the nitric nitrogen is evident and 
persistent through the two seasons, the total nitrogen in the plants 
in 1915 is only about three-fourths of that present in 1913. In this 
statement we compare 32 samples in 1913 with 36 samples taken in 
1915. As stated, the relative effects of the respective fertilizers are 
discernable throughout both series. The seasons were dissimilar and 
this difference would appear to be the direct cause of the ditferences 
in the nitrogen content up to the end of July, or during the vege¬ 
tative period. The factor which caused this difference can not be 
the total amount of water supplied by way of the soil, for that ap¬ 
plied by irrigation and rainfall during the two seasons was very 
nearly the same. This is on the assumption that all the rainfall 
reached the ground, which is not correct. But if the amount of 
water supplied by the soil had been different, other results would 
still compel us to eliminate this as the cause for the difference. It 
is fortunately the fact that some of our plots in 1913 received dur¬ 
ing the season 30 inches of water, whereas, in 1915, the total re¬ 
ceived was a little over 19 inches. The samples from these plots re¬ 
ceiving 30 inches of water in 1913 contained as much nitrogen as the 
corresponding plots that received 19 inches, so that a difference of 
11 inches of water applied to the soil made no difference in the 
amount of total nitrogen contained in the stems, which we have 
used as representative of the whole plant. The same is true of the 
proteid nitrogen, or, perhaps 1 should say. that the differences are 
so small and irregular that we cannot consider them in this work. 
