188S.] Change of Potential of a Voltaic Couple. 



295 



strengthen the solutions, and thus arrive at the " minimum-point. " 

 The method is a little less accurate than the one in which a thermo- 

 pile is employed. 



By means of this method, using a couple composed of magnesium- 

 platinum in distilled water, I have found the following " Minimum- 

 points of Change of Voltaic Potential," in solutions of potassic 

 chloride, potassic chlorate, hydrochloric acid, and chlorine. I selected 

 these substances because they were representative ones, suitable to 

 yield results for comparison, and because they gave extreme and 

 intermediate magnitudes of the effect. The results are compared 

 with those obtained with a Mg + Pt couple and the thermopile. 



Potassic Chloride, KC1. 



Solution at 18° 0. ; " minimum-point" lay between 1 part in 3875 

 and 4650 parts of water • and by the aid of the thermopile, with 

 the solution at 17° C, " minimum-point" between 1 in 3875 and 

 4305. 



Potassic Chlorate, KC10 3 . 



Solution at 19° C. ; " minimum -point" between in 1 in 4650 and 5166 

 and with the pile, solution at 18° C, between 1 in 4920 and 5470. 



Hydrochloric Acid, HC1. 



Solution at 17° C. ; " minimum-point " between 1 in 516,666 and 

 664,285 ; and by aid of the pile, solution at 19° 0., between 1 in 

 516,666 and 574*074. 



Chlorine, CI. 



Solution at 18° C. ; " minimum-point" between 1 in 15,656,500,000 

 and 19,565,210,000; and with the pile, solution at 12*5° C, between 

 1 in 17,000 millions and 17,612 millions. 



These results show the great degree of delicacy of each method, and 

 the extremely large difference of proportion of different substances 

 required to upset the balance. The two methods agree. 



By employing a great variety of dissolved substances, I have 

 found that nearly every such substance has a minimum proportion 

 below which it has no apparent effect upon the electromotive force of 

 a MgPt or ZnPt couple in distilled water ; and this proportion 

 appears to be a constant number, dependent only upon very simple 

 conditions, viz., unchanging composition of the voltaic couple and 

 liquid, a uniform temperature, and employing the same galvanometer. 

 The apparently constant numbers thus obtained may probably be 

 used as tests of the purity or of the uniformity of composition of 

 dissolved substances. 



The " minimum-point" and degree of sensitiveness varies with, 



