between Metallic Masses having different Temperatures. 43 1 



I. PHENOMENA OF SOUND. 



4. Musical sounds do not always accompany the vibrations 

 above described. There is one condition or modification of the 

 apparatus which generally ensures their production. If a groove 

 be made either in the bar or block, as shewn in the sections 

 Fig. 2, in the direction of the axis of the bar, and separating the 

 points of contact with the block, upon which the bar oscillates, 

 we shall rarely fail in producing the sound. These sounds gene- 

 rally commence with a deep base note, which rises as the experi- 

 ment proceeds, and as the equilibrium of temperature of the two 

 metals approaches ; sometimes rising suddenly an octave in the 

 most fitful manner, and occasionally redescending. Mr TRE- 

 VELYAK, in his paper just alluded to, has treated of the sounds 

 thus produced, and seems to consider the phenomena introduced 

 by the condition of the groove, of an essentially different class 

 from the others. He assumes that the only effect of the groove 

 is, that it may allow a current of heated air to pass through 

 it ; yet he admits that this current is not sufficient alone to 

 produce musical notes, because no such occur when vibrations 

 do not take place ; nor yet, according to him, do the vibra- 

 tions suffice, since they require the supposed current of air 

 introduced by the groove to complete the effect. In his enu- 

 meration of the sources of musical notes, Mr TREVELYAN has 

 omitted to state that the mere accumulation of impulses of any 

 kind up to a certain number in a second, produces alone a 

 musical sound depending upon that number. This Mr FARA- 

 DAY readily shewed to be the true cause of the musical sounds 

 produced in this experiment, namely, the number of contacts 

 of the hot with the cold metal in a second, and he illustrated 

 the fact by putting a cold bar of metal in vibration by means of 



