184 



DR CARGILL G. KNOTT ON THE 



bers, together with first differences, and the ratios of corresponding pairs of 

 differences — which quantit es were necessary for the final reductions. 



Experiment of May 14, 1886. 



Resistance of Hydrogenium = 1 "4245 Ohms. 



Temperature. 



Ft - FLd. 



Differences. 



P^ - Fd. 



Differences. 



Ratio. 





63 





164 











23 





49 



•47 



40° 



86 





213 











23 





50 



•46 





109 





263 











19 





51-5 



•37 





128 





314-5 







50° 



153 



25 



365-5 



61 



•49 







21 





55 



•38 



60° 



174 





420-5 







and so on. 



The temperatures in the first column are not to be regarded as accurately 

 determined. They merely indicate, to a sufficient approximation, the tempera- 

 ture of the moment. The ratios in the last column should give the ratios of 

 the corresponding electromotive powers of hydrogenium and palladium referred 

 to platinum. The mean of all for any one experiment will give a good value for 

 this ratio at the mean temperature of 60° C, the cold junction being 20° C, 

 and the highest temperature being 100° C. Instead of taking a single mean, 

 however, I divided the experiment into two portions, and found the mean of 

 all differences below 80° C. and the mean of all differences above 80° C. In 

 this way I got two values for the ratio — one corresponding to temperature 

 50° C, and the other to temperature 90° C. These ratios, which are repre- 

 sented by the symbol de'/de, are given in the accompanying table, each 

 specimen of hydrogenium being distinguished by its resistance estimated in 

 terms of the resistance of the originally pure palladium wire ; that is, by the 

 ratio B'/R 



Table comparing the Changes in Resistance and in Thermoelectric Power of Palladium Wire charged 



with different amounts of Hydrogen. 



R'/R 



1 



1-043 1-121 



M89 



1-26 



1-302 



1-395 



1-504 



1-78 



de l de \ 90° 



1 

 1 



•88 

 •88 



•75 

 •76 



•77 

 ■70 



■70 

 •65 



•63 



•61 



•47 

 •53 



•44 

 •44 



-1-58 

 -1-58 



'/- ' 'de (mean) 



1 



•88 



•755 



•735 



•675 



•62 



•50 



•44 



-1-55 



