102 Mr Meikle on a proposed Improvement 



the specific heat of air under a constant pressure, by its specific 

 heat under a constant volume, viz. 1.1547. Still, however, this 

 multiplier, as obtained from the experiments above mentioned 

 on the specific heat of air, gives the velocity of sound too small. 

 The object of the present article is to suggest a reason for this 

 deficiency. 



The theory of sound, as improved by Laplace, supposes it to 

 be propagated by a wave of air, having an increased temperature, 

 without any addition to its quantity of heat ; but as each portion 

 of the air forming the wave is warm when it communicates mo- 

 tion to the next, it must also impart to it a portion of its heat *. 

 Hencd? sound is propagated by a wave of air, having not mere- 

 ly its temperature increased by compression, but having also an 

 addition to its quantity of heat. In this way, a wave of heat 

 accompanies sound through the air ; and I presume, that to it 

 we owe the excess of the experimental, over the theoretical, ve- 

 locity of liaplace. The theory of this distinguished philosopher 

 lays no stress on the amount of the rise of temperature ; but 

 such amount must depend on the degree of compression, that 

 is, on the iritensUy of sound ; and as the transference of a quan- 

 tity of heat from each portion to the next, will be greater as its 

 excess of temperature is greater, it is clear that the velocity of 

 sound must be greater when it is more intense. I am perfectly 

 aware, that some suppose sounds of all intensities to be propa- 

 gated with the same velocity, and allege as a proof, the undis- 

 turbed succession of musical notes, when heard at a distance. 

 So far as regards the present inquiry, I need only remark, that 

 musical notes, or the differences of their intensities, are mere 

 playthings, when compared with the penetrating report of a 

 cannon issuing from the flames. 



From the account of experiments made in Holland by Dr 

 Moll, with many excellent precautions, and published in the Phi- 

 losophical Transactions for 1824, p. 424, it appears that sound 

 moved slower from Kooltjesberg to Zevenboompjes, than in the 



• Heat cannot be here lost laterally, because sound is propagated, not in 

 an insulated line of air, but rather as in a pyramidal figure, or something like 

 a spherical sector, having the sonorous body for its centre, as is plain from 

 sounds being heard over a considerable lateral extent. A line of air, there- 

 fore, which is not near the outside of the sector, will lose no heat laterally. 



