494 
Proceedings of the Royal Society 
the arrangement of the negative signs in the middle of the difference 
column, indicating a slight curvature, may he said to he accounted 
for. I find a similar result on reducing some of Macfarlane’s Tables 
in the same way, so that it would scarcely he unreasonable to say 
that for distances beyond T cm., the curve is straight within the 
limits of experimental error. 
If this be allowed, there remains to be accounted for only the 
admitted curvature for distances under *1. Two explanations have 
been suggested, and others might be or perhaps may already have 
been advanced. 
1. That the air in the immediate neighbourhood of a solid con- 
ductor is condensed, and therefore dielectrically stronger. This 
explanation, if I understand it aright, appears to me to fail altogether. 
For the beginning of rupture is conditioned by the strength of the 
layer next the conductor, and, even on this explanation, that would 
be the same at all distances, and the curve ought to be a straight 
line. 
2. That there is a difference of potential between the conductor 
and the air. This hypothesis would still give a straight line 
although not passing through the origin, it would therefore still 
leave the curvature near the origin unexplained, which in the case 
of air at least is the main difficulty. Still this possible cause should 
not be lost sight of. It may be part of the explanation. 
3. It might be that the air near the conductor is electrified, in 
which case the characteristic equation of the potential between the 
plates would become 
M + ilrp = 0. 
Without entering into a discussion of this hypothesis in the 
meantime, I may mention that it is subject to the initial objection, 
that it seems difficult to reconcile it with the fact, proved by the 
best modern results, that there -is nothing of the nature of slow 
conduction or even convective discharge through air at the atmo- 
spheric pressure, so long as the electromotive force does not surpass 
a certain limit. 
4. There is yet another hypothesis which seems to me well 
worthy of examination, viz., that the Specific Inductive Capacity of 
the air, undergoes a rapid variation from the surface of a solid 
