346 Mr. F. Guthrie on Approach caused by Vibration, 



pieces of card in planes perpendicular to the plane of the vibra- 

 ting body, Mr. Faraday was able to control the accumulation of 

 the powder, and by performing the experiment in vacuo he 

 showed that the internodal deposition no longer took place. By 

 scattering a light powder upon a tuning-fork, one prong of which 

 was vertically above the other, both being horizontal and vibra- 

 ting in a vertical plane, Mr. Faraday showed how the powder 

 collected along the middle line of the prong in little travelling 

 heaps, each heap consisting of moving particles which rose from 

 the fork in the middle of the heap, fell down the side of the heap 

 and were gathered in at the bottom. A general conclusion at 

 which Mr. Faraday arrived was this : whenever the different parts 

 of a surface are vibrated to different degrees, there is always a 

 tendency for the air to flow along the surface of the vibrating 

 body towards the more violently agitated portions from the less 

 agitated. 



§ 4. It is clear that, before examining the possible connexion 

 between these superficial whirlwinds and the fact mentioned in 

 § 1, it is necessary to examine into the existence of air-currents 

 of unclosed paths. The tuning-fork which was most employed, 

 and which I call fork A, vibrated 256 times per second (UT 2 ), 

 and had the following dimensions (Plate IV. fig. 1.) : — 



metre. 

 OY = 00230 



OX = 00172 



OZ = 03255. 



I call the three faces intersecting in the faces a, b, and c re- 

 spectively. Let the symbol H c , &c, denote the position of the 

 fork when c is horizontal, &c. 



§ 5. Experiment 1. — The fork A was set vibrating by drawing 

 the bow across the edge Y ; the plane of vibration was accord- 

 ingly parallel to b. The fork was then brought into the neigh- 

 bourhood of an ascending thread of smoke. The smoke and 

 fork had in succession the three relative positions : — 



(1) Ha. The smoke passed across the face a parallel to Y. 



(2) H a . The smoke passed across the face b parallel to X. 



(3) H a . The smoke passed across the face c parallel to X. 

 In all cases the smoke clung to the surface across which it 



passed in the same manner as if the fork were at rest. 



§ 6. Experiment 2. — A cylindrical glass tube T (fig. 10), 0*4 

 metre long and 0042 metre internal diameter, was fastened in a 

 horizontal position. One end was left open, the other carried a 

 cork, through the centre of which passed a horizontal tube t, 004 

 metre long and 0*0035 metre internal diameter. These tubes were 

 filled with smoke, and the fork A, which had been set vibrating 



