ee e. 
EFFECT OF NITROGENOUS FERTILIZERS. 129 
EFFECT OF NITROGEN IN THE FERTILIZER UPON THE PRO- 
TEIN OF THE CROP. 
In most of the special nitrogen experiments and in a few 
of the soil test experiments, the effects of nitrogen of the 
fertilizers upon the composition of the crops have been studied 
by chemical analysis of samples taken in such a manner that 
they could be considered representative of the whole crop. The 
produce of each plot was weighed in the field at the time of 
harvesting, and the samples were gathered and. weighed sep- 
arately at that time. These samples were put into small bags, 
and were weighed again just before the analyses were 
made. From the results of the analyses, together with 
the weight of the sample and the total weight of the yield, it is 
possible to estimate not only the percentages of dry matter, 
nitrogen compounds (protein), and other ingredients, but also 
the total amounts of the several ingredients per acre. In the 
following discussion of the effects of nitrogenous fertilizers on 
the composition of the crop, tables are given showing both 
the percentage composition and the actual yields per acre 
calculated from the results of the analyses. The figures in 
these tables are averages of the results of all the similar experi- 
ments for all the years in which they were made. They are the 
figures given in bold faced type in the corresponding tables in 
the series at the end of the discussion. The tables there give 
the details of all the experiments in which analyses have been 
made ; one set of tables gives the percentage composition of the 
' crops analyzed, and the other set gives the yields per acre of 
nutrients calculated from the percentage composition and the 
weight of the crop. 
Effect upon the Composition of Mixed Grasses—In the ex- 
periments with mixed grasses the samples were taken while 
the hay was in the windrow, just before it was carted from the 
field. Large samples, from 12 to 20 pounds, were gathered 
into coarse sacks by taking small quantities from different 
parts of the plot. These samples were immediately cut into 
pieces one or two inches long, and mixed and sub-sampled, 
the final samples, from 3 to 5 pounds, being carefully weighed. 
In some instances duplicate, and in others triplicate samples 
were taken and analyzed separately, in order to prove the 
accuracy of the sampling. There were occasional slight varia- 
tions in the composition of the different samples from the same 
plot, but the results on the whole were very uniform. 
9 
