348 Professors A} r rton and Perry on the 



periments. The following are samples of the readings of the 

 voltmeter, the distance between the carbons in each case 

 being fixed and the current alternately trebled and diminished 

 to one third. 



Difference of 

 Current potentials between 



approximately. carbons, in volts. 



1 26-5 



3 26-5 



1 , 24-5 



3 25-5 



1 26'5 Distance between 



3 26*5 carbons re- 



1 25-5 adjusted. 



3 26-5 



3 32 A somewhat 



1 30 greater dis- 



3 39*5 tance between 



1 30 the carbons. 



3 34 



1 34 



3 36 



1 38 



3 30 



1 28 



3 30 



3 30 Distance between 



1 28 carbons re- 



3 30 adjusted. 



1 28 



3 30 



It would appear therefore that for a fixed distance between 

 the carbons- the difference of potentials necessary to maintain 

 the arc is nearly but not quite independent of the current, 

 the electromotive force requiring to be slightly increased when 

 the current is very much increased. 



2. The second part of the investigation was for the purpose 

 of ascertaining in what way the differences of potentials 

 between the carbons varied with the length of the arc when 

 the current was kept constant. For this purpose the arc was 

 projected on a distant scale by means of a lens, the magnifying- 

 power of the arrangement being calculated, first, by comparing 

 the distance between the scale and the lens with the distance 

 between the arc and the lens, secondly by putting close to 

 the arc a piece of carbon of known thickness and measuring 



