Intelligence and Miscellaneous Articles. 535 



rizontal tubes may be emploj'ed by means of a stopcock with three 

 branches, to produce a vacuum in the apparatus and fill it with gas ; 

 to the other metal tube, which is also furnished with a safety- 

 stopcock, a curved glass tube dipping into the mercurial trough is 

 adapted. By means of this arrangement the pressure of the gas 

 maj^ be ascertained, or it may be collected for analysis. With gases, 

 the apparatus, with the exception of the bellows, is immersed in a 

 trough full of water, and with vapours, the whole apparatus, inclu- 

 ding the bellows, is placed in a large zinc stove, where it is heated 

 by the vapour of boiling water. For operating under low pressures, 

 the bellows is surrounded I)y a box in which a counter pressure is 

 estabhshed. This box and the rod of the bellows are then united 

 by a caoutchouc tube, the elasticity of which leaves the rod sufficient 

 play. 



In a new apparatus with which I have not yet experimented, I 

 have placed the bellows in the neck of the balloon, in order to avoid 

 opposite pressures. 



The apparatus being filled with gas or vapours, the sound is pro- 

 duced very pure, and as often as majf be desired, by the action of 

 the bellows ; the harmonics issue readily, and it is necessary to take 

 several in order to be certain of the fundamental sound. 



The formulse 



v^sj^k, c\(k,-V) = o\k-\), 



and 



c,(A:,-l ) ^ c{k-Y) 

 k, k 



show by the velocities of the sound the ratios ^i and k of the spe- 

 cific heats of a gas and of the air ; then the same specific heats at 

 a constant volume c\, and at a constant pressure c,. 

 The formula 



c\(],,-\) = c,{k-\) 



gives c\ h—\ 



c. A:, — l' 

 the values of this proportion derived from the rapidity of the sound, 

 or from the specific heats at a constant pressure, are equal to the 

 number of simple atoms which enter into the composite atom, or in 

 a simple relation with this number. 



Conclusions. 



1 . All gases resounding in the same pipe present the same nodal 

 surfaces for the harmonics of the same order. 



y. The formula by which Laplace expresses the velocity of sound 

 is confirmed by experiment. 



3. 'I'he law of Dulong and Carnot upon the specific heats of gases 

 is conformable to the mechanical theory of heat and to experiment. 



4. The velocity of sound in a gas is independent of the jiressure 

 and of the state of saturation : it depends entirely on the temperature. 



7). The velocity of sound and direct experiment give the same 

 values for the specific heats of gases at constant pressure. 



