580 The Philippine Journal of Science 1920 
triangles 8 and 9 are the ones that approach most nearly to 
Shive’s diagram of magnesium injury. Whether the Shive series 
containing the three other salts without any potassium chloride 
might have shown the distribution of this injury recorded by 
Shive, if it had been carried out simultaneously with these four 
sets, is of course uncertain. 
The more-intense green color that characterized the cultures 
with a large supply of magnesium sulphate in series I and II 
was also noted in this series ; the region of deepest green occurred 
at the left margin in the triangle in all four cases. Further- 
more, the color intensity of these greenest plants decreased from 
set to set as the potassium chloride content of the solution in- 
creased. As has been mentioned, the high potassium chloride 
content of these sets was accompanied by lower absolute amounts 
of the three other salts, so that the last observation may well be 
related to the small absolute amount of magnesium sulphate 
present in these solutions. Dwarfing of the root system also 
occurred in this series with high relative amounts of magnesium 
sulphate, as has been noted for series I and II, and this dwarfing 
was much more pronounced with high absolute values of this 
salt than with lower ones. As has been mentioned, striping 
of the leaves and “stooling” were not observed in series III. 
Dry weights .— Keference to the diagrams of fig. 7 shows that 
as the partial concentration of potassium chloride is increased, 
the partial concentration of the other three salts together being 
correspondingly decreased, the area of high top values migrates 
downward and to the right; that is, toward lower relative pro- 
portions of monopotassium phosphate and higher ones of calcium 
nitrate. That this migration of the area of high tops is not due 
merely to decreased partial concentration of monopotassium 
phosphate, calcium nitrate, and magnesium sulphate is suggested 
by the fact that a similar decrease in the total concentration of 
Shive’s solution (see Shive’s paper, figs. 2 and 3) produced a 
markedly different change in the configuration of the triangle. 
The migration of the area in question was from the central to 
the right central portion; that is, toward higher proportions 
of calcium nitrate, but not toward lower proportions of mono- 
potassium phosphate. But other evidence from the absolute 
values presented in Table 12 appears to indicate that the main 
effect here produced by increasing the partial concentration 
of potassium chloride is primarily related to the concomitant 
