102 



Potts — Rowland's New Method. 



this, A/T is constant, i. e., A is independent of the current 

 flowing through the condenser. 



The variation in A/T due to changes of temperature was 

 obtained as follows : in an opening made in the side of the 

 box containing the condenser, a mercury thermometer was 

 placed. The temperature indicated by the thermometer was 

 of course not that of the inside of the condenser but that of 

 the outer edge. The condenser, however, was kept within a 

 degree or two at least of the desired temperature for some six 

 or more hours before being used. 



This method gave sufficiently accurate results, as there was 

 no occasion for an accurate determination of the temperature. 





Temp. 











cent. 



r=481C 





Date, etc 



t 



R 2 



R 3 



R 4 



ll-16-9£ 













19-3 



709-0 



2557 



203-6 



102 











1 



22-0 



707-8 



2195 



202-6 



11-2-99 











7 



22-2 



707-8 



2614 



202-6 



1 



16-5 



•709-9 



3271 



203-0 



12-8 











o 



16 -1 



709-6 



2862 



203-0 



106 













20-2 



2985- 



2985 



203-3 



1 



4'5 



302-4 



2696 



407-3 



Table VI. 



Calcu- 

 lated. . 

 Rx R'i A T ~ 

 R'x-R, T 



34-90 65-35 30-45 '1167 258-3 



34-97 54-88 19-91 "0816 242"7 



34-40 44-08 9-68 "0543 178'6 



34-44 50-04 15-60 -0816 19T3 



34-93 48-27 13-34 -0569 234-5 



34-55 45-87 11-32 "0752 150-9 



Table VI gives the results of this investigation, and in fig. 4 

 they are plotted with temperatures as abscissas and A/T as 



250 



200- 



150- 



100 



50- 



0° 



5° 



CI l-sciss<ie re/brese??~t tem^ero'turc t 

 C/rt/indtes represent %. 



10° 



15° 



20° 



