132 THE ELECTRIC ARC 



rents keeping the length of the arc constant, and by Blondel 1 

 with the voltage constant. Fig. 41 reproduces two of the 

 curves given by Ayrton. 2 As may be seen, the increase 

 in light is nearly proportional to the current flowing. Rey 

 measured the candle power with currents up to 250 amperes 

 and states that the light increases more rapidly than the 

 current. 3 



The larger the carbons the less the candle power, due to 

 shading of the light from the crater by the carbons. 4 It has 

 also been found that there is a loss of candle power when 

 the pressure of the surrounding gas is more than one atmos- 

 phere. This is due to the presence of fumes of nitric acid. 5 



Intrinsic Brightness of Crater. For many purposes the 

 usefulness of a light depends not only on the total amount 

 of light, but also on the amount given per unit area of the 

 luminous surface. A concentrated light is often blinding 

 and may be injurious to the eye. In this respect all forms 

 of the arc are far from being ideal and the carbon arc is 

 the worst of all lights. Blondel states that the intrinsic 

 brightness of the crater is 160 candles per sq.mm. Petavel 6 

 found a value of 147 candles per sq.mm. for a current density 

 varying from 0.06 to 0.26 ampere per sq.mm. This is 

 something like 200,000 times the brightness of a candle 

 flame. 



Flaming Arcs. The previous discussion applies to the 

 older form of arc where by far the greater part of the light 

 comes from the anode. In the flaming arcs the chief 



1 Eel. Elec., 10, 297; 1897. 



2 Mrs. Ayrton's "Electric Arc," p. 366.- 



3 Eel. Elec., 32, ii i ; 1902. 



4 Schreihage, Centralbl. f. Elektrot., 10, 604; 1888. 



5 Wilson, Astrophys. Journ., 2, 212; 1895. 

 8 Proc. Roy. Soc., 63, 469; 1899. 



