Temperature of the Carbons of the Electric Arc, fyc. 35 



Therefore we have a radiation from the +pole of the arc, in our 

 arbitrary units, corresponding to 



245 

 244-48 x-- = 13613 in position B, 



4'4 



and to 



255 

 456-37 x ^-^ = 30420 in position A. 



' 



The geometrical mean of these two numbers, 



x/13613 x 30420 = 20350, 



is the true radiation in our arbitrary units. To find the correspond- 

 ing temperature, we mnst substitute this value in the equation on 

 p. 31, i.e., 



This is most easily solved by trial, T coming out as 3520. 



To this about 100 must be added, owing to the curvature of the 

 radiation curve (for full reason, see Wilson and Gray's paper already 

 quoted, p. 387), giving approximately 



3600 abs., or 3300 C. 



as the temperature of the hottest part of the + pole of the electric 

 arc, a result surprisingly near Violle's estimate, 3500 C. 



The Comparative Radiation from Different Parts of the Arc. 



After the above experiments had been finished, an attempt was 

 made to obtain comparative values of the radiation, and hence the 

 temperatures, of different parts of the carbons of the arc. For this 

 purpose, a radio-micrometer of the ordinary form was employed, on 

 to the receiving-surface of which radiation could fall through a large 

 pin-hole. 



An image of the carbons was then formed by an arrangement 

 shown diagrammatically in the figure (fig. 5), in which L is the arc 



Fio. 5. 



',^ ====== 



^H 



lamp, inside a lantern with the condenser removed, M! is a concave 

 mirror, M 2 a convex mirror, both silver-on-glass, and R is the radio- 



i> 2 



