144 
Walter Stiles 
the higher the valency the greater the antagonistic effect, while 
Szucs also found from experiments in which the concentration of 
the electrolyte (calcium nitrate) was kept constant and the concen¬ 
tration of dye varied, that the antagonistic effect also depended on 
the concentration of the dye. The time required for the cells to acquire 
a standard intensity of colour is not inversely proportional to the 
concentration; it was found instead- that the product of time and 
concentration of dye in the external solution increased with increasing 
concentration of dye. But no doubt the relation between time 
required for the absorption of a definite quantity of dye and the 
concentration of the dye in the solution, will depend on the quantity 
of dye selected. The inadequacy of this method of investigation has 
already been emphasized in an earlier chapter (Chapter x) and need 
not be further stressed here. 
That the retardation of absorption of dye brought about by the 
presence of electrolyte is not due to an action between the dye and 
electrolyte outside the cells, becomes evident from a number of 
considerations. Practically the same concentrations of potassium, 
calcium and aluminium nitrates are required to precipitate methyl 
violet, so it would appear that the action of these three salts in 
bringing about molecular association of the dye must be approxi¬ 
mately the same in each case, and consequently the coefficient of 
diffusion of the dye would be approximately the same in every case. 
Again, it was shown that yeast cells become stained with a 0-005 
per cent, solution of neutral red in less than 10 minutes, while in 
the same solution containing aluminium chloride not until after the 
lapse of 48 hours. If this were due to molecular association and 
consequent reduction in the coefficient of diffusion, it would mean, 
assuming Herzog’s formula (see Chapter iv), that the molecular weight 
of neutral red in presence of aluminium chloride is more than 
2-3 x io 7 , which is obviously an impossibly big magnitude. 
A consideration of the results collected in Table XLVI makes it 
clear that diluter solutions of electrolytes are relatively more active 
in antagonising the entrance of the dye than more concentrated 
solutions. According to Sziics his results with potassium nitrate agree 
well, and those with calcium nitrate and aluminium nitrate approxi¬ 
mately, with the equation 
1 
t = aC” 
where C is the concentration of salt and t the time taken for the dye 
to enter. Because of the similarity of this expression to the adsorption 
