290 LECTURE XXXI. 



fectly agree among themselves ; and the observation which has been made 

 in France, that a heat, sufficient to set tow on fire, may be produced/ by 

 the operation of a condensing syringe, seems to show that Mr. Dalton's 

 results are somewhat below the truth. In this manner the theory may 

 be completely reconciled with experiments ;* we may estimate the modulus 

 of the air's effective elasticity, which is the measure of its immediate 

 force, from the velocity which is thus observed, and its height will appear 

 to be 39,800 feet, instead of 27,800, which is the supposed height of the 

 atmosphere. This velocity remains unchanged by any alternation of pres- 

 sure indicated by the barometer, but it may be affected by a change of 

 temperature. For when an elastic fluid is compressed, its elasticity is 

 increased in the same ratio as its density ; and the height of a homo- 

 geneous atmosphere equivalent to the pressure, remains the same, conse- 

 quently the velocity calculated from that height remains unaltered ; but 

 the identity of the acceleration, from the effect of heat which has been 

 mentioned, can only be inferred from observation : this identity may, how- 

 ever, be satisfactorily shown, by means of experiments on the sounds of 

 organ pipes, which are intimately connected with the velocity of the 

 transmission of sound through the air, and which are found to remain 

 precisely the same, however the air may be rarefied or condensed. The 

 Academicians del Cimento inclosed an organ pipe, with bellows worked by 

 a spring, in the receiver of an air pump and of a condenser, and they 

 found that, as long as the sound was audible, its pitch remained unchanged. 

 Papint screwed a whistle on the orifice which admits the air into the 

 receiver of the air pump, and I have fixed an organ pipe in the same 

 manner ; and the result agreed with the experiment of the academicians. 

 But if the density of the air is changed, while its elasticity remains unal- 

 tered, which happens when it is expanded by heat, or condensed by cold, 

 the height of the column, and consequently the velocity, will also be 

 altered ; so that for each degree of Fahrenheit's thermometer the velocity 

 will vary about one part in a thousand. Bianconi J has actually observed 

 this difference of velocity according to the different heights of the thermo- 

 meter, and it may be shown less directly by means of the sounds of pipes ; 

 but it has not been accurately determined whether or no the correction on 

 account of the effect of compression in causing heat, remains unaltered, 

 although Bianconi's experiments agree very well with the supposition that 

 no material change takes place in this respect. The velocity of sound 

 must also be in some measure influenced by the quantity of moisture con- 

 tained in the atmosphere : it must be a little diminished by cold fogs, 

 which add to the density, without augmenting the elasticity, and increased 

 by warm vapours, which tend to make the air lighter ; and these two 

 opposite states are probably often produced in succession in wind instru- 

 ments blown by the mouth, the air within them being at first cold and 

 damp, and afterwards warm and moist. 



In pure hydrogen gas, the velocity of sound ought, from calculation, to 

 be more than three times as great as in common air, but the difference does 



* Clement and Desormes, Journal de Physique, 1819, p. 34. 

 t Birch, iv. 379. J Comm. Bonon. ii. I. 365. 



