236 PHILOSOPHICAL TRANSACTIONS. [aNNO]751. 



a noii-electric, by becoming itself the conductor of electricity, that is, by its 

 keeping out the air, and suffering the electricity to pervade the vacuum. 



Mr. W. was desirous of knowing, for the further illustration of his proposi- 

 tions, whether the Leyden experiment could be made through the vacuum. For 

 this purpose he made the before-mentioned exhausted tube part of the circuit, so 

 necessary to this experiment. In this experiment it is absolutely necessary that 

 the whole quantity, or nearly so, of the accumulated electricity, should be dis- 

 charged in the same instant of time. Accordingly on making the experiment, 

 at the instant of the explosion, a mass of very bright embodied fire was seen to 

 jump from one of the brass plates in the tube to the other ; but this did not 

 take place when one of the plates was farther distant from the other than 10 

 inches. When the distance was greater, the fire then began to diverge, and lose 

 part of its force ; and this force diminished in proportion to its divergency, which 

 was nearly as the distance of the 2 plates. 



The difficulty however of applying the Torricellian vacuum to these experi- 

 ments has been happily got over by Lord Charles Cavendish, our worthy vice- 

 president. This noble lord, who to a very complete knowledge of the sciences 

 joins that of the arts, and whose zeal for the promotion of true philosophy is 

 exceeded by none, has applied it in the following manner, and his lordship put 

 his apparatus into Mr. W.'s hands. This apparatus consisted of a cylindrical 

 glass tube of about -^ of an inch in diameter, and of 7-^ feet in length, bent 

 somewhat like a parabola in such a manner, that 30 inches of each of its extre- 

 mities were nearly straight, and parallel to each other, from which an arch 

 sprung, which was likewise of 30 inches. This tube was carefully filled with 

 mercury ; and each of its extremities being put into its basin of mercury, so 

 much of the mercury ran out, until, as in common barometrical tubes, it was 

 in equilibrio with the atmosphere. Each of the basins containing the mercury 

 was of wood, and was supported by a cylindrical glass of about 4 inches in dia- 

 meter, and 6 inches in length ; and these glasses were fastened to the bottom of 

 a square wooden frame, so contrived, as that to its top was suspended by silk 

 lines the bent tube filled with mercury ; so that the whole of this apparatus with- 

 out inconvenience might be moved together. The Torricellian vaccuum then 

 occupied a space of about 30 inches. In making the experiment, when the room 

 was darkened, a wire from the prime conductor of the common electrical ma- 

 chine communicated with one of the basins of mercury, and any non-electric 

 touching the other basin, while the machine was in motion, the electricity per- 

 vaded the vacuum in a continued arch of lambent flame, and as far as the eye. 

 could follow it, without the least divergency. 



That the electricity was not furnished from the glasses employed in these ope- 



