MRS. H. AYRTON ON THE MECHANISM OF THE ELECTRIC ARC. 
333 
deduce them from what we already know. For instance, when the positive carbon 
alone is cored, it must have the same form A B C, as when both are solid, since the 
change of cross-section due to a given change of current is not materially altered by 
coring the positive carbon alone. Coring the negative carbon alone diminishes the 
negative value of SV C /SA, and must diminish it most when the current is least, for it 
is then that the metallic vapour from the core will be expended on the smallest 
Current. 
Fig. 15. Suggested curve connecting the part of SV/8A that depends on changes in the cross-section of 
the arc with the direct current for a constant length of arc. 
quantity of hard carbon mist, and will consequently have most effect. Hence the 
curve for a cored negative and solid positive carbon must resemble D E F (fig. 15), 
and the current for which SV^/SA becomes positive, if any, will depend upon the 
length of the arc and the frequency for which the curve is drawn. Finally, with both 
carbons cored, SV f /SA is even more positive than when the negative only is cored, so 
that the curve must resemble G H K (fig. 15), since the same reasoning as before 
shows that the cores have least effect when the current is largest. 
To find the full curves connecting SV/SA with A, for each pair of carbons, vve have 
only to add each ordinate of each curve in fig. 13 to the corresponding ordinate of the 
curve for the same carbons in fig. 15. Curves resembling those that would be thus 
obtained for one length of arc and frequency of alternating current are given in 
fig. 16. The exact distance of each above or below the zero line, and the exact points 
where it cuts that line must, of course, depend upon the length of arc and frequency 
