i4^ M. R. Koiihaascb^s Theory of the 



maving fovcCj it is, to use an image, as if, whilst the resistance 

 to the attack increased each moment, it became tired out by the 

 diu'atiun of the same, so that gradually it became actually ex- 

 hausted. 



In a graphic representation, if we take the time calculated 

 from the moment when the external influence commenced as 

 abscissa, and the magnitudes of the changes of form as ordinates, 

 then to different bodies differently shaped curves will correspond. 

 Whilst with steel, the curve, rising at first almost perpendicularly, 

 will then appear bent almost at right angles and have a pretty 

 sharp corner, we find that Weber^s curve for the silken thread rises 

 much more slowly, and approaches its asymptote at a greater 

 inclination ; and lastly, if we bear in mind the properties of a 

 spiral coil of wax, the tapers cut from which, when placed on 

 our Christmas trees, grew always crooked again, there can be no 

 longer a doubt that the curve in question possesses but a gradual 

 curvature, and rises quickly but for a very short period of time. 



According to this, if no loss of electricity took place, there is 

 nothing in the form of the residue curve (R, Plate VI. fig. 2), 

 which was produced in the bottle h, that is contradictory to the 

 hypothesis which regards the molecular forces of the glass as the 

 cause of the slow formation of the residue and of its limit ; in- 

 asmuch as they, in this case as well as in that of elasticity, per- 

 mit but a slow realization of any changes in the state of equili- 

 brium. 



Although, it must be confessed, this explanation by means of 

 a mechanical, resisting force rests ultimately on only one analo- 

 gous action, we nevertheless abide by this manner of representa- 

 tion, on account of the difficulty there is to find a parallel case 

 for the peculiar phfenomenon of so slow a motion in comparison 

 to the acting forces. 



By this manner of representing the electric moment of the 

 glass, where throughout the interior of the same the electricity 

 on every particle is brought into a different position, we can now 

 explain why the thicker plate furnishes a greater residue. In 

 order to see this, however, we must again make a small digression. 



The question is virtually this. We have two thin, insulated, 

 metallic plates, which, being charged equally with opposite elec- 

 tricities, are placed parallel and at a very small distance from 

 one another in comparison to their own dimensions. Will the 

 action of these plates on a point between them decrease very 

 much when iLt; distance between them is increased, e. (j. doubled, 

 but still remains very small in comparison to their dimensions ? 



Here, where a strict calculation appears inadmissible, a few 

 indications may serve to decide the question. 



Let a plane, circular surface with the radius R be conceived, 



