i6o 
Journal of Agricultural Research 
Vol. XIV, No. 4 
omitted to avoid confusion. The position of any culture represented 
on the diagrams may readily be found by means of its culture number, 
which always indicates the row and the number of the culture in the 
row. On the left margin of each diagram the rows are numbered con¬ 
secutively from base to apex. The cultures of each row, represented by 
the points of intersection of the lines, may be considered numbered con¬ 
secutively from left to right. The number of the last culture in each 
row is given on the right margin of the diagram. The lower triangular 
diagram of figure i represents the average yields of buckwheat tops from 
series A, carried out during the physiological growth period extending 
from germination to the flowering stage. The upper diagram represents 
the corresponding yields from series B, conducted from the flowering 
stage to the maturity of the seed. 
(a) Comparison op the effects of the various osmotic salt 
PROPORTIONS FOR THE TWO DIFFERENT PHYSIOLOGICAL GROWTH PERIODS. 
Consideration of the relative dry weights of tops. —It Will be 
observed that in the diagram of series A (fig. i) the main area of low 
yields, including eight of the nine cultures lying within the range of 
low dry weights of tops, extends along the right margin to the apex of 
the triangular diagram, while in series B the main area of low yields 
borders on the left margin, extending to the base of the triangle at the 
lower left, and includes six cultures. It is to be noted that only a 
single culture (R4C5) included in the low area of series A is also included 
in the low area of series B. There is, moreover, with the exception of 
this one culture, scarcely any overlapping of the areas of low yields on the 
diagrams of the two series. The lowest average yields of tops do not 
occur with corresponding cultures of the two series. The lowest aver¬ 
age yield produced by series A and by series B occurred with cul¬ 
tures R1C4 and R1C1, respectively. The former is characterized by the 
lowest osmotic proportions of monopotassium phosphate and medium 
proportions of calcium nitrate and magnesium sulphate, while the latter 
is characterized by the lowest proportions of monopotassium phosphate 
and calcium nitrate, and by the highest proportion of magnesium sul¬ 
phate. 
The main area of high yields on the diagram of series A occupies a 
central region lying to the left of the vertical axis of the diagram and 
includes four cultures on the left margin. Two secondary high areas 
are also indicated for cultures R2C4 and R3C5. On the diagram of 
series B the main area of high dry weights occupies a central region 
lying mainly to the right of the vertical axis of the diagram and extends 
to the right margin at culture R3C6. An outlying high area is indicated 
at the lower right, and a smaller one about cultures R2C2 and R3C2; but 
these two cultures, together with culture R8C1 at the apex of the dia¬ 
gram, mark the lower limit of the range of high yields. The main areas 
of high yields of the two series are thus seen to lie on opposite sides of 
the diagrams, as do also the main areas of low dry weights. 
