94 Sir W. Snow Harris’s Researches in Statical Electricity. 
surface, as at ¢, fig. 10, it would suffer more obstruction to in- 
ductive change from surrounding electrical particles than when 
placed at the extremity g, where it would be less involved, as 
it were, in counteracting forces. If placed directly at and out- 
side the extremity g, then one of its faces would be still more 
exposed and free; it would therefore take up more electricity in 
these points than at the centre c, in which it was less free; 
a result perfectly in accordance with the phenomena of electric 
force already exemplified (6), Exps. 1, 2, &c. By giving the 
touching disc, therefore, a free action of induction, either by 
such means as resorted to in Exp. 9, just described, or by attach- 
ing to its remote surface a light gilded reed and ball of about 6 
inches in length, as represented in fig. 11, we obtain an equal 
reactive force when transferred to the balance, from all points of 
the surface to which we apply it. I have little doubt but that, 
employed in this way, the method of finding the relative quan- 
tity of electricity on charged surfaces employed by Coulomb is 
very exact. I found, for example, that in charging a rectangular 
plate P, fig. 10, with one, two, three, &c. measured quantities of 
electricity, the reactive forces from an elevated cylinder @ 8, or 
from a tangent plate prepared as in fig. 11, were exactly in the 
same ratio, at whatever point of the surface it was applied. 
It is clear that in the preceding Exp. 9 the several altitudes 
admit of being considered as proof-planes of increasing thick- 
ness: were we to trust the indications of such planes, we might 
be led into very unsound deductions as to the distribution of 
electricity upon the charged plate. Suppose the thickness of the 
"plane had been about the -] of an inch, the reactive force at the 
centre, as compared with the force at the extremity, would 
have given quantities in the proportion of 7 : 33, being nearly as 
1:5. If we take the thickness *25, then the proportion would 
have come out in the ratio of 11:33, or as 1:3; if ‘5, it 
would have been as 15:33, or as 1:2 nearly ; so that, as before 
observed (13), Exp. 7, it would be difficult to say at what limit 
we may arrest our measure in respect of a decrease of the thick- 
ness of the proof-plane, although by no means difficult to find 
the limit for its free inductive susceptibility in the opposite di- 
rection, as just shown. There are many important phenomena 
of electrical charge involved in these considerations, and to which 
the proof-plane, as usually employed, would certainly ill apply. 
If, for example, we impart the same quantity of electricity to a 
square, a circular plate, and a rectangular plane of equal area 
greatly extended as to length, then, as shown by Volta, the in- 
tensity of charge is greatly diminished in the case of the long 
plane, and we could dispose upon it a much greater quantity of 
electricity under the same degree of the electrometer. Yet a 
