between Metallic Masses having different Temperatures. 433 



ployed. Mr TREVELYAN practically found the influence of throw- 

 ing the mass of matter to the sides of the bar, so as to render its 

 equilibrium more unstable, and thus lengthening the period of its 

 vibrations ; and Mr FARADAY well observed, that the pressure of 

 the finger upon the upper part of the bar, while in a state of vi- 

 bration, by shortening the period of oscillation, raised the note to 

 a great extent, and rendered the sounds musical, when before they 

 were not so. In the bars which I have generally employed, the back 

 was hollowed out, as shewn in the section Fig. 4, which served 

 both to diminish the quantity of matter near the axis, and to 

 contain a portion of mercury, the bright surface of which I found 

 a convenient test of the existence of oscillations too minute to 

 be readily perceived by the eye, as was generally the case when 

 the sounds were musical. They could always be detected like- 

 wise by approaching the point of the finger gently to the bar. 

 This will even detect vibrations which produce no visible tre- 

 mor in a clear globule of mercury. 



6. I trust I have now shewn that all the phenomena of 

 sound are ultimately resolvable into those of simple vibration. 

 There are one or two facts connected with, and indicated by, the 

 musical tone, which we shall more particularly consider in treat- 

 ing of the modification of the vibrations. These are especially 

 the sudden changes in the note of the instrument, which general- 

 ly rises as the experiment proceeds, and the influence of the 

 groove in raising the note, that is, rendering the vibrations more 

 frequent. 



7. We have only a remark to make upon the absolute velo- 

 city of these oscillations, which the sound produced affords the 

 only accurate way of estimating. With one of the ordinary bars 

 such as I have been in the habit of using, the highest note which 

 I have distinctly observed and compared, was A above the mid- 

 dle C of the pianoforte, which corresponds to 480 vibrations 

 in a second. From this velocity (and it is often very much 



