MRS. H. AYRTON ON THE MECHANISM OF THE ELECTRIC ARC. 
303 
a 
rest, leaving a mist-covered tip which is longer and slenderer, because its sides are 
hotter and burn away more readily, the larger the 
crater and the shorter the arc. 
Hence, with a small crater and a long arc, the nega¬ 
tive carbon remains fairly flat, as in a , (fig. 2); whereas, 
as the crater becomes larger, its action alone shapes the 
negative carbon as dotted in b, (fig. 2), and the extra 
heating due to the mist combined with the protection 
which the mist offers as shown in the full line. With 
a short arc, on the contrary, a small crater alone would 
produce an end as dotted in c, (fig. 2), while the com¬ 
bined effect of the crater and mist produce the end 
outlined by the full line. Finally, with a large crater 
and a short arc, the crater alone would j^roduce an end 
as dotted in el, (fig. 2), while the crater and mist 
together would shape the negative carbon as given by 
the full line in d. Experience shows that the negative carbon does shape itself in 
this way under the various conditions. 
g. 2. Negative carbons, a, long 
arc, small crater; b, long arc, large 
crater; c, short arc, small crater ; 
d, short are, large crater. 
Why die Area of the Crater is not Directly Proportioned, to the Current, but 
Depends edso on the Length of the Arc. 
Suppose that the current and the distance between the ends of the carbons have 
been kept constant long enough for all the conditions of the arc to have become 
steady, so that it is “ normal” and that then the resistance in the outside circuit is 
suddenly diminished. At the first instant the P.D. between the carbons must be 
increased, a larger current will have to flow through a vapour film of the old 
dimensions, and consequently the heat developed in it per second will increase. The 
temperature of the existing vapour film cannot rise, because there is no increase of 
pressure, consequently it must expand, and spread over a larger area of solid carbon. 
The moment the film had expanded in the slightest degree, it would begin volatilising 
carbon from a part of the surface hitherto inactive, and thus a larger quantity of 
vapour per second would be volatilised. At the next instant, therefore, the quantity 
of carbon volatilised per second would have increased, and the resistance of the vapour 
film would have become lower, and its tendency to expand would, therefore, be 
diminished on both accounts. Thus, at each instant after the change of current the 
volatilising surface would increase, but more and more slowly, till its area was such 
that the heat developed per second in the vapour film only just sufficed, after all 
losses from conduction, &c., to keep up the volatilisation. After that, the vapour 
film would cease to expand, and the surface of volatilisation would have reached its 
maximum area for the new current. 
