302 
MRS. H. AYRTON ON THE MECHANISM OF THE ELECTRIC ARC. 
Why the End of the Positive Carbon has its Particular Shape. 
As only the part of the positive carbon that is in actual contact with the vapour 
film can be at the temperature of volatilisation, evaporation can only take place at 
that surface, and hence I suggest that, unless the vapour film is as large as the whole 
cross-section of the positive carbon, it must dig down into the carbon and leave the 
surrounding parts unvolatilised, i.e., the part of the positive carbon against which the 
film rests must become concave. These surrounding parts, however, are heated 
sufficiently by conduction from the evaporating surface and by the hot gases 
surrounding them to burn away, and so there must be a race between volatilisation 
of the centre portion and burning away of the edges, which must, in all cases, 
determine the shape of the surface of volatilisation. When, all other things being 
equal, the gap between the carbons is small, so that the end surface of each carbon is 
well protected from the air, volatilisation will gain over burning and the pit may 
become very deep. When, on the other hand, the gap is large, so that the air can 
easily reach all parts of the carbon except that actually covered by vapour, these 
parts may burn away as fast as, or even faster than the inner portion is volatilised, 
and in that case the surface of volatilisation will be flat, or even slightly convex. 
It is evident, therefore, that this surface cannot, from the very nature of things, help 
being concave when the distance between the carbons is short, and flat or convex 
when it is long. And this is true, whether the volatilisation is due solely to a large 
back E.M.F., as some have supposed, or to the resistance of a thin film of carbon 
vapour, as I have suggested, or partly to one and partly to the other. 
When only a small bit of the end of the positive carbon is being volatilised, the 
outer edge of the carbon will not be made hot enough to burn, and the tip will remain 
relatively blunt. When, on the contrary, the area of volatilisation is large, the edge 
of the carbon will be burnt away and a long tapering end will be formed, terminating 
in the surface of volatilisation. Further, the shorter the arc, the less easily will the 
heat be able to escape from between the carbons, so that the more remains in them 
to produce burning, and, consequently, the longer must be the tapering part. 
Experience shows these conclusions to be true. 
Why the End of the Negative Carbon assumes its Particular Shape. 
The negative carbon is shaped entirely by burning away; the heat that raises it to 
burning temperature being furnished partly by the mist that touches it, and partly 
by radiation from the vapour film lying against the positive carbon. The part that 
the mist rests on is protected by it from the action of the air, and does not, therefore, 
burn away. At the same time this part must be hotter than the remainder of the 
carbon, and so the portion of the carbon near it burns away more readily than the 
