MRS. H. AYRTON ON THE MECHANISM OF THE ELECTRIC ARC. 
311 
cease only when the arc becomes normal again. In other words, when an arc 
of given length, with a given current flowing, exists between given carbons, neither 
the resistance nor the P.D. between the carbons has any fixed value, except when the 
arc is, and continues to be normal. In all other cases each varies, within certain 
limits, according to the time that has elapsed since either the current, or the length 
was altered, and according to what change was then made. 
That the P.D. does actually undergo alterations of the kind just described, after 
a change of current, is evident from fig. 8, the curves in which were plotted from 
actual experiments made in 1893. For these curves, the current was suddenly 
altered when the arc was quite normal, and was then kept constant at the new value 
while the P.D. continued to alter, the time change of P.D. being noted. The first 
change of P. D.—the rise or fall that must have instantaneously accompanied the 
Fig. 8. Experimental curves showing simultaneous time-changes of P.D. and current with a constant 
length of arc of 1 millim. Solid carbons, 18 millims. and 15 millims. in diameter. 
change of current, when the resistance outside the arc was suddenly diminished— 
was too quick to detect. Indeed, it was only after assisting in carrying out these 
experiments that it occurred to me that we ought to have seen it, and that, on trying, 
I found I could sometimes detect it and sometimes not.* 
The rapid change in the P.D. while the area of the crater and the cross-section of 
the mist were altering, is very marked, however, as well as the slow rise or fall of 
P.D. accompanying the diminution or increase in the cross-section of the arc due to 
the change in the cross-section of the carbon ends. 
Why Measurements of the Resistance of the Arc made under the same Apparent 
Conditions Differ in Value and even in Sign with Different Experimenters. 
The dependence of the resistance of the arc on its previous history, as well as on 
the actually existent length and current, has an important bearing on the question 
* It is only when the carbons are cored that it can be detected. The reason will be explained later. 
