Royal Institution of Great Britain. lit 



entirely to the diminution of the force of the pressure of tite ^mi6« 

 sphere in the direction opposed to the current. 



Another illustration of the effbct was given, intended to lead directly 

 to a clear understanding of the phenomena of the original experi- 

 ment. A tin cone, about eight inches long, and three inches dia- 

 meter at the base, had about half an inch in length cut off from the 

 apex, so as to leave a small aperture there ; a round hole half an 

 inch in diameter was made in the side, about four inches from the 

 narrow etid. Upon blowing forcibly into this cone from the nar- 

 row end, the current of air continually passed from a smaller to a 

 larger space, and on bringing a flame to the lateral aperture, the 

 whole of it was urged into the cone, and could be seen by looking 

 in at the mouth of the instrument. By pressing this instrument 

 into a flattened form, it was then shown to be an exact represen- 

 tation of part of the original tube apparatus, and the relation of 

 the effects in both at once simply referred to the same cause. 



M. Clement's experiment, in which he measured the forces per- 

 pendicular to and between the discs, by means of gauges, were 

 then referred to *, and also several other forms of the apparatus, 

 as well as other effects dependent upon the same cause. Mr. Fa- 

 raday also referred to the supposed explanation of the original 

 effect given by those who attribute it to the friction of the air, or 

 to the current which sets in all round upon the original current of 

 air. Friction can have nothing to do with it, because the force 

 which supports the discs may be resolved into one perpendicular 

 to its surface ; whereas friction can only be at right angles to this 

 direction. With regard to the influx of air, by which the effect is 

 supposed to be produced, it is itself an effect only, produced by 

 the same cause as that which supports the disc, and which may 

 easily be distinguished from it. In place of using a disc, it was 

 shown, that if a cylinder of the same diameter as the disc, and 

 closed at one or both ends, had the closed end applied in place 

 of the disc, it was equally well supported, notwithstanding it 

 might be many inches in length. It was also shown, that when 

 the influx of the neighbouring air upon the disc Was allowed, (wo 

 discs might be supported instead of one. The one by the causes 

 already described, the other by the sweeping of the neighbouring 

 air over the disc as it passed on, to follow the direction of the 

 stream originally put in motion. 



Dr. Granville laid several curious objects upon the library table s : 

 ♦ See Quarterly Journal, New Series, vol, ii, 



