ELECTROMOTIVE FORCES OF THE ELECTRIC ARC. 329 



Seat of the Back E.M.F. Search Carbons in the Arc. 



The fact that the arc has a back E.M.F. which appears to increase with the amount 

 of foreign substances present in the vapour column, at once leads to the question 

 whether this E.M.F. is located at one or the other of the electrodes, or distributed 

 along the vapour column. In order to obtain an answer to this question, some 

 experiments were made on a 6-millim. 9 '91 ampere solid arc by introducing a search 

 carbon, 2 millims. diameter, into the arc, and measuring not only the direct P.D. 

 between the search carbon and each of the main carbons, but also the impedance to 

 the high-frequency testing current of that part of the arc between it and each ot 

 the main carbons. 



In the experiments, three different positions of the search carbon were employed, 

 (1) with its centre 1 millim. from the positive electrode, (2) central in the arc, (3) with 

 its centre 1 millim. from the negative electrode. The fine point to which the search 

 carbon burns was always kept, so far as possible, just reaching to the axis of the 

 main carbons. The results of these experiments are given in Table V. 



The introduction of a search carbon into an arc always greatly disturbs the 

 conditions of the arc, and the present case was no exception. The introduction of 

 the search carbon increased the direct P.U. arc by 4'0 volts, and the impedance of 

 the arc lamp by 0'44 ohm. So that the introduction of the search carbon, either by 

 deflecting the arc and so increasing its length, or by chilling the vapour column, 

 increases its resistance by an amount which approximately accounts for the observed 

 increase in P.D. arc. The back E.M.F. of 'the arc, as a whole, was but little affected 

 by the introduction of the search carbon. This distortion of the arc by the search 

 carbon probably also accounts for the observation that the measured impedance of 

 the arc as a whole is not equal to the sum of the impedances of the t\vo parts 

 comprised between the search carbon and the main electrodes. 



Owing to the correct method of apportioning between the two electrodes, the 

 resistance and self-induction of the loop formed by the carbons, holders, and frame ot 

 the lamp, being unknown ; and owing to the fact that the measured quantities are 

 only roughly approximate, due to the disturbing effect of the search carbon, no 

 attempt was made to apply the small correction to the observations for the self- 

 induction and resistance of the carbon holders and lamp frame, and the observed 

 impedances were treated as resistances, and the back E.M.F.'s calculated as usual. 

 Further, the three arcs which had the same length and current will be considered as 

 having been identical, though such was not strictly the case. 



On these assumptions, consider the resistance between the positive electrode and 

 the search carbon when the search carbon is 1 millim. from the positive electrode, 

 and then 5 rnillims. from the positive electrode (i.e., 1 millim. from the negative). 

 The change in resistance due to this -change of 4 millims. in the position of the 

 electrode is 1 72 ohms. Taking next the measurements made between the negative 



VOL. ccm. A 2 u 



