ELECTROMOTIVE FORCES OF THE ELECTRIC ARC. 327 



Effect of Varyimj the Nature of the Electrodes. 



The results of experiments to determine whether the resistance and back E.M.F. 

 depend on the size and nature of the electrodes are given in Table IV. The effect 

 of the size of the electrodes is not well marked in the case of " Conradty " and 

 " Apostle " carbons, the changes in size being probably too small to make the effect 

 very evident; the resistances of the arcs are, however, slightly larger with the 

 smaller carbons. With " Le Carbone " Electrographitic solid carbons the impedance 

 of the arc between two 11-millirn. carbons is about 17 per cent, higher than that 

 between two 9-millim. carbons. Observations on the arc between these two 11-millim. 

 carbons, which was very unsteady and difficult to maintain, lead the author to think 

 that the observed difference in impedance is not due to the change in size of the 

 electrodes, but to the material of the two sizes of carbons being different, though 

 nominally the same. It is also to be noted that the power-factor, 01)2 of the arc 

 between the 11-millim. carbons, is the only one which has not been found equal to + 1, 

 to within the limits of experimental error, at a frequency of 100,000 per second. 

 To be quite certain that this was not owing to some error the experiment was 

 repeated, but with practically identical results. 



It may be mentioned that these Electrographitic carbons are not ordinary arc 

 lamp carbons, but were specially made for the experiments by " Le Carbone." They 

 are supposed to consist of pure graphite, and they are said to be made by expelling 

 the remaining impurities from carefully prepared carbon by heating it in an electric 

 furnace until the impurities are volatilised. 



Both the resistance and the hack E.M.F. of the arc depend greatly on the make of 

 carbon, that is to say, on the composition of the electrodes, since it is very improbable 

 that any two makers' carbons have identical chemical composition. The experiment 

 of soaking a pair of solid "Conradty Noris " carbons in potassium carbonate, drying, 

 and re-determining the resistance, shows that the effect of introducing this potassium 

 salt was to reduce the resistance from 3 '81 ohms to 2 '92 ohms, and to increase the 

 back E.M.F. from 12 volts to 15 volts for the same arc length and current. A 

 similar effect is produced by drilling out one of the carbons and inserting a glass rod 

 as a core, probably due to the introduction of sodium into the vapour column. The 

 lower resistance and higher back E.M.F. of arcs between cored carbons than of those 

 between solid carbons is also probably due to a similar cause, namely, the presence of 

 potassium silicate in the core. In fact, it seems probable that the whole of the 

 observed differences between solid and cored arcs, and between arcs for which 

 different makes of carbons are used, not only in resistance and back E.M.F., but also 

 in all their physical properties, are due to the different amounts of the traces of 

 foreign substances present in the arc. 



The author believes that if it were possible to obtain perfectly pure carbon 

 electrodes, then the resistance of the arc between them would be very high, so high 

 that it might be impossible to maintain a true arc between them at all. He is of the 



