ELECTROMOTIVE FORCES OF THE ELECTRIC ARC. 337 



(4.) Considerable thermo-electric forces at the contacts between the electrodes and 

 the vapour column, whose resultant is the greater part of the observed back E.M.F. 

 of the arc. The force at the positive opposing the flow of the current, and that at 

 negative helping it, account for the much greater conversion of electric energy into 

 heat at the positive electrode. 



In the above, it will be noticed that the volatilisation or sublimation of carbon is 

 not supposed to have any influence on the back E.M.F. except in so far as it limits 

 the temperature the electrodes can obtain. 



The introduction of impurities up to a certain limit into the arc will cause the 

 conductivity of the vapour column to increase, and for a fixed current its temperature 

 will probably decrease, so the thermo-forces and the back E.M.F. may be expected to 

 increase as has been observed (see Table IV.). 



There is another possible explanation of the larger back E.M.F. when impurities 

 are present, viz., the thermo-electric force of the junction carbon-vapour may be 

 larger the greater the quantity of impurity present. If this is the case, may not the 

 explanation of the drop in P.D., when the arc hisses, or when oxygen or hydrogen 

 come in contact with the positive crater, as found by Mrs. AYRTON, be that the <jas 

 combines with the impurity and reduces the back E.M.F., and not that it combines 

 with the carbon as suggested by Mrs. AYRTON ? 



In conclusion, the author wishes to express his thanks to Professor AYRTON and 

 Mr. MATHER of the Central Technical College for the very valuable assistance and 



O i/ 



advice they have given him during the course of these investigations ; he also wishes 

 to thank the many students who have from time to time helped him with the 

 experiments, and especially Messrs. DEL MARR, LYNN, BROWN, WATSON, and VINKS. 



ArPENDIX I. 



On the Relation betivcen the P.D. Current and Length of the Arc. 



Mrs. AYRTON has pointed out that the relation between the P.D., the current, and 

 the arc length, may for the solid carbons which she used be accurately expressed by 

 an equation of the form V = a + /3l + (y + SI)/ A, where V and A are the P.D. and 

 current respectively, and a, ft, y, and S are constants, between the limits over which 

 she varied the current and length. It was thought, therefore, that it would be of 

 interest to determine the values of the constants , ft, y, S for the solid " Conradty 

 Noris " carbons used in so many of the experiments in this paper. The two series of 

 results in Curves III. and V. give P.D. when the length is kept constant and 

 the current varied ; and also the P.D. when the current is kept constant and the 



VOL. ccni. A, 2 x 



