Page images
PDF
EPUB
[subsumed][merged small][ocr errors][subsumed]

[Lane's electrical machine, with discharging electrometer. From his paper in the Phil. Trans. 1767, p. 451. For Cavendish's form of discharging electrometer, see Art. 405.]

[graphic]

the usual degree, it discharged itself, as was before said, at '04 of an inch, so that the usual degree of electricity was about a mean between these two*.

It seemed as if the electricity spread about of an inch further with the stronger degree of electricity than with the weaker, but the experiment was not accurate enough to determine it with certainty.

330] I made an experiment of the same kind to determine whether the electricity spread to the same distance on crown-glass as on this. It seemed to spread about 8 of an inch on it, that is, rather less than on the plate H, though its thickness was, of the two, rather less. But whether this difference is real, or owing to the error of the experiment, I cannot tell.

331] There seems no reason, from the foregoing experiments, to think that the charge of any of these plates is sensibly greater than it would be if the electricity was disposed uniformly on their coated surfaces, as their charges agree very well together without such a supposition. If we suppose that the charges of any of them are sensibly greater than they would be if the fluid was disposed uniformly, it will be necessary to suppose that there is a still greater difference between the distance to which the electricity spreads on the surface of thin plates and that of thick ones than what we have assigned. But I shall speak more on this subject at the end of Art. [365].

332] But though it appears from the foregoing experiments that the charges of plates of glass of different thicknesses with coatings of different shapes and sizes bear the same proportion to each other that they ought to do by theory, yet their charge is many times greater in proportion to that of a globe than it ought to be on a supposition that the electricity does not penetrate to any sensible depth into the substance of the glass, as will appear by the following experiment.

333] In order to compare the charge of the plate D with the globe of 12 inches used in the former part, I made two plates coated as a Leyden vial, the charge of each of which was about

* [By Macfarlane's experiments (Trans. R. S. Edin. Vol. xXVIII. Part II. 1878) the electromotive force required to produce sparks between flat disks at those distances would be 14, 11-8, and 9 units respectively.]

that of D, each consisting of two plates of glass cemented together and coated on their outside surfaces with circular pieces of tinfoil about 1 inch in diameter*.

I then compared the charge of each of these double plates with that of the globe in the same manner that I compared together the charges of different bodies in the former part, the only difference being that, in trying either of these double plates, I made a communication between the lower coating of the plate and the ground, the wires Mm and Dd (Fig. 14) being contrived so that they were sure to fall on the upper coating †.

By this means the charge of each of these double plates was found to be just equal to that of the globe. The charge of the plate D was then compared with that of the two double plates together, and was found to be less than that in the proportion of 263 to 272, and consequently the charge of the plate D is to that of the globe as 26-3 to 13-6.

334] Before we go further it will be proper to consider what effect the three circumstances taken notice of in Art. 277 will have in altering the proportion of the charge of the double plate to that of the globe. With regard to the two first, it appears that the charge of the globe and double plate will neither of them be sensibly different from what they would be if they were placed at an infinite distance from the jar by which they are electrified, and moreover, in trying the globe, the repulsion of the redundant fluid in the globe increased the deficience of fluid in the trial plate as much as the attraction of the trial plate increased the quantity of redundant fluid in the globe, so that it required the same size to be given to the trial plate as it would have done if the globe and trial plate had exerted no attraction or repulsion on each other; and in trying the coated plate, the coated plate could not sensibly increase the deficience in the trial plate, nor could the attraction of the trial plate sensibly increase the redundance in the coated plate, so that neither of these two causes had any tendency to alter the proportion of the charges of the globe and coated plate to each other.

If they had been made of a single piece of glass, the coatings must have been so small as would have been inconvenient unless the glass had been of a greater thickness than could have been easily procured. [Arts. 446, 451, 649, 653, 654.] + [Note 17.]

[Arts. 455, 456, 478.]

335] But the third cause will have a sensible effect, for in trying the globe the floor and sides of the room near it would be made undercharged, which would increase the charge of the globe, whereas in trying the coated plate the floor would not be made sensibly undercharged, nor, if it was, would it have any sensible effect in increasing the charge of the plate.

So that the charge of the globe bore a sensibly greater proportion to that of the coated plate than it would have done if it had been placed at an infinite distance from any other bodies.

How much the charge of the globe should be increased hereby I can not tell, but I should imagine it should be at least by th part, for if the room had been spherical and 16 feet in diameter (about its real size) and the globe placed in its center, it should have been increased as much as that*, and as the globe was really placed three times as near to the floor as to the ceiling†, I suppose the effect to have been still greater.

* Let the globe Bbß, whose centre is C, be insulated in the hollow globe Ddô concentric with [it]. Let the inner globe be pos. electrified by the canal BE not

a

D

B

communicating with the outer globe, and let the outer globe communicate with the ground. The quant. defic. fluid in the outer globe must be equal to the redundant in the inner globe, and the attraction of the outer globe on the canal BE is to the repulsion of the inner one thereon as

[blocks in formation]

and therefore the quantity of redun. fluid in the inner globe is to that which it would contain if the outer globe were away as

[blocks in formation]

If room was spherical, 16 feet in diameter, globe in middle of it, its charge should be increased in ratio of 16 to 15 by reason of undercharged floor, &c.

[This is the only indication of the height of the room. The circles were suspended by silk s rings from a horizontal bar (Art. 466) 87.5 inches from the floor. By Art. 474 the platform 14 inches high diminished the height of the bodies in the

336] In order to find out, if possible, how much the charge of the globe was increased hereby, I made four flat plates of a mixture of rosin and bees wax*, about 4 inches square and 22 thick, and coated each of them with circles of about 18 inches in diameter, and compared the charge of each of them separately with that of a circular plate of tin, 9-3 inches in diameter. I then compared the charge of two of these plates together with that of a tin circular plate 18 inches in diameter, and lastly I compared the charge of all together with that of a circle of 36 inches diametert.

337] By a mean of the different experiments it appears that the charge of each of the rosin plates was alike, and that the charge of any one of them was to that of the circle of 9-3 inches as 10:34 to 9-3, that the charge of the circle of 18 inches was to that of two of the rosin plates together as 20·19 to 21·96, and that the charge of the circle of 36 inches was to that of all four plates as 4375 to 42.06.

But the charge of the four plates together will not be exactly four times the charge of one plate singly, as some allowance must be made for the charge of the wire connecting their upper surfaces, and, besides that, the charge of the plates when placed close together will not be quite so great as if placed at a distance from each other+.

By trying the charge of all four rosin plates together by the machine, Fig. 20, both when placed close together and at as great a distance from each other as I could, I found their charge when close together to be to their charge when placed at a distance nearly as 41 to 412, and, from some other experiments I made, I am inclined to think that the charge of each of the wires which connected the upper coatings of the plates was to that of one plate alone as 28 to 930§.

ratio of 2 to 3. Hence the height of the center of the bodies from the floor was 42 inches, and the height of the room 4 x 42 inches, or 14 feet. This would agree with the height of the top of the circle of 18 inches being 51 inches from the floor (Art. 472).]

* These plates are non-conductors of electricity, and may be charged as Leyden vials. The manner in which I made them will be described in the following pages [Arts. 373, 514]. My reason for making them of these materials is that the charge of such a plate is much less than that of a plate of glass of the same dimensions.

It must be observed that, in the two last mentioned comparisons, the rosin plates were placed close together and their upper surfaces connected by a piece of wire.

[Art. 557.]

$ [Arts. 555, 558, also 443.]

« PreviousContinue »