280 BOTANICAL GAZETTE [APRIL 
upon the process of starch hydrolysis, any single treatment may be 
regarded as a combination of two separate applications of the salt 
in question, each at a lower concentration. The data above pre- 
sented show that increasing the concentration of any salt almost 
never results in a proportionately increased effect upon hydrolysis. 
Sometimes such an increase completely reverses the direction of the 
effect, as when an increase in the concentration of sodium chloride 
from m/128 to m/32 produces an alteration in the efficiency ratio 
fromo.85to1.42. Another example of this is furnished by calcium 
chloride, for which a change in concentration from m/4 to molecu- 
lar is accompanied by an alteration of the salt effect on diastatic 
action, from 33 per cent acceleration to a quantitatively undeter- 
mined, but nevertheless exceedingly marked, retardation. In most 
cases increased concentration of a single salt does not change the 
direction of the effect, but simply alters its intensity, as when an 
increase in the concentration of sodium chloride, from m/8 to m/2, 
results in decreasing the acceleration of hydrolysis from 82 to 71 
per cent. It is to be expected, therefore, that the bringing of two 
different salt treatments into combination may frequently result 
in the same sort of effect as that produced by increasing the con- 
centration of one of the single salts, the total salt concentration 
remaining the same as before. This is the true condition of affairs 
in many instances, and the specific question which confronts us— 
without attempting any precise quantitative comparisons—is 
whether combinations of two salts produce more or less effect upon 
starch hydrolysis than would either salt alone, at the higher con- 
centration that obtains in the combination. Put in another way, 
the problem is this: if a portion of the amount of a given salt in a 
diastase mixture be replaced with a molecularly equal amount of 
another salt, is the resulting alteration in the effect upon hydrolysis 
merely that traceable to dilution of the first salt together with 
that due to addition of the second (as would be expected if the 
combined treatments were without effect upon each other), or 
is this alteration of a different character? Despite the incomplete- 
ness of the data at hand, study of the tables brings out several 
points bearing upon this question, which are made still more 
clear if graphs are resorted to. The more striking of these points 
will now be mention 
