332 On Elect ricif^ excited in Paper. 



winch and handle. These were fixed parallel to each other 

 in a frame in such a manner that the distance between them 

 could be increased by means of catgut strmgs, yet they should 

 still retain their parallel position. 



A band of paper about seven inches broad was put round 

 both the cylinders, which was stretched by drawing them 

 asunder with a force of about thirty-five or forty pounds. The 

 cylinders were now filled with boiling water, and the holes by 

 ■which the water was put in closed with corks. Two pieces of 

 Indian rubber, together nearly as long as the paper was broad, 

 fastened on wooden springs, were applied to the paper as it 

 passed over one of the cylhiders, while the latter was kept 

 turning by means of the winch. A small conductor was 

 brought near that part of the paper which had been rubbed 

 as it parted from the curved surface of the cylinder. This 

 gave out a constant stream of electric fluid visible in the day- 

 light (and even I think sufficient to charge a small phial), but 

 too small to encourage the constructing the machine on a 

 larger scale. 



Though this did not succeed very well, it might still be 

 worth making a model ; the cylinders of which should be of 

 wood instead of metal, and somewhat after the design ex- 

 plained in the section annexed ; because paper can be excited 

 much more intensely over wood than metal. The paper and 

 cylinder on which the rubber is applied, may be kept warm 

 by means of a tin vessel filled with boiling water himg near it, 

 and the water in this vessel might be kept boiling by a lamp. 



W. M. G. 



A B. Wooden cy- 

 linders. 



cc. Paper band. 



E. Indian rubber 

 and spring, the 

 tail of which is 

 pulled by the 

 string / 



D. Tin vessel of 

 boiling water. 



N.B. The tin cylin- 

 ders were fixed 

 in this manner, 

 but tliey were of 

 ecjual size, 



LXXI, An 



