l8o BOTAXICAL GAZETTE [august 



are B; the ordinate at E expresses growth in a mixture in which 

 50 per cent of the dissolved molecules are A and 50 per cent are B. 1 

 It is obvious that the effect of mixing two equally toxic solutions 

 must fall into one of the following categories. 



1. The toxicity is unaltered, that is, the toxic action of the two 

 salts is additive. Each salt produces its own toxic effect precisely 

 as though the other were not present. This is expressed by the 

 horizontal dotted line LJM. 



It is evident that we cannot get increased growth by mixing two 

 such solutions unless the salts have an antagonistic action. The 

 dilution of A from o . 1 M to o . 05 M is exactly compensated by the 

 introduction of molecules of B. Or, to put it in another way, the 

 toxic effect depends on the number of molecules present (if both 

 kinds of molecules are equally toxic and there is no antagonism) 

 and it makes no difference whether the solutions are pure or mixed. 



If the toxic effect depends on ions, rather than on molecules, 

 then, since the number of ions may be somewhat increased by mix- 

 ing solutions, the toxicity may be correspondingly increased; but 

 the amount of this increase would ordinarily be negligible. 



2. The toxicity is diminished, that is, the effect is antitoxic. 

 We then get a curve rising somewhere above the dotted line, such 

 as the unbroken line LKM. 



3. The toxicity is increased. We then get a curve which some- 

 where falls below the dotted line, such as the line interrupted by 

 circles LHM. 



The considerations here set forth apply in all cases where two 

 equally toxic solutions are mixed, whether their concentration is the 



^ ■ In order to avoid unnecessary calculations in making up solutions with the 

 desired molecular proportions, curves similar to those in fig. 2 serve a very useful 

 purpose. The figures on the vertical scale denote molecular proportions, while those 

 on the horizontal scale denote cc. of solutions. Suppose that we are mixing NaCl 1 M 

 and CaCl 2 o 5 M. If we mix 60 cc. of the NaCl with 40 cc. of the CaCl 2 , the molecular 

 proportions are NaCl 75 per cent+CaCl 2 25 per cent. If each of the scales (vertical 

 and horizontal) is too mm. loner thp nrHinato fm^ocur^ri frnm ttkwp Hnwnward) will 



in this case be 25 mm., and the abscissa (measured from right to left) 60 mm. After 

 determining a series of such points, a curve may be drawn from which other propor- 

 tions may be read off directly. Such a curve is shown in the figure (NaCl 1 M - 

 CaCl a 0.5 M); this curve serves equally well for all solutions in which the molecular 

 concentration of one component is twice that of the other. In the same manner the 

 other curve NaCl 1 M - CaCl, o. 1 M applies equally well to all solutions in which the 

 molecular concentrations of the two components are as one to ten. 



