16, 6 Trelease: The Growth of Rice 617 
poorest salt mixtures to the soil, since the grain yield (20) ob- 
tained with these mixtures was slightly higher than the aver- 
age yield (18) from the control cultures. 
Grain yields and individual salts.—In fig. 4 the data of the 
relative yields of grain, given in the seventh column of Table 
3 and plotted on the triangular diagram of fig. 3, have been 
replotted in such a way as to bring out relations that may hold 
between yield values and the proportions of the individual salts. 
In all three graphs of fig. 4 the relative yields of grain are 
plotted as ordinates. In the uppermost graph, A, the abscissas 
represent the cultures arranged as a series in the order of the 
magnitude of the ammonium sulphate content of their fertilizer 
mixtures, beginning with the culture receiving the highest 
amount of this salt and ending with the control culture having 
no fertilizer at all. In the middle graph, B, the abscissas re- 
present the cultures arranged in the order of the magnitude 
of the potassium sulphate content of the fertilizer mixture, 
while in the lowest graph, C, the cultures are similarly arranged 
in the order of the magnitude of the primary calcium phosphate 
content of the salt mixture. The most striking general features 
of these graphs are that the uppermost, although showing minor 
fluctuations, indicates a strong general downward slope to the 
right, while the two lower graphs fluctuate widely, with some 
indication of a general upward slope toward the right. Each 
of the three graphs will now be considered in greater detail. 
The uppermost graph was plotted in order to bring out any 
general relation that might exist between the proportion of 
ammonium sulphate (supplying the important element nitrogen) 
and the yield of grain. As just mentioned, the abscissas, from 
left to right, give the cultures arranged in order of the magni- 
tude of the ammonium sulphate content of their fertilizer mix- 
tures, and the ordinates represent the corresponding relative 
grain yields. The vertical dotted lines in this figure divide 
the cultures into nine groups; all of the cultures of each group 
received the same amount of ammonium sulphate, but they dif- 
fered with respect to the proportions of the other two salts in 
the fertilizer mixture. A single culture (to the left of the first 
vertical dotted line) had 0.8 of its total salt addition due to 
ammonium sulphate; the second group of two cultures had 0.7 
due to this salt, etc.; and the group of eight cultures, at the 
extreme right, had 0.1 due to this salt. Finally, the control 
culture, having no fertilizer at all, is plotted for the sake of 
