1813.] Specific Heat of the different Gases. 215 



two similar gazometers we may make a certain quantity of gas 

 pass from the one to the other as often as we please, without in- 

 terruption. 



When thegas which fills the reservoir, B, is not atmospherical 

 air, when it is hydrogen gas, for example, if we pass it often 

 from one gazometer to another, it will disengage the air con- 

 tained in 'the water, and take its place, so that the purity of the 

 gas will be injured. It would have been impossible to have pro- 

 duced a current of carbonic acid gas, nitrous oxide gas, or 

 oleriant gas, in this manner, though they are but imperfectly 

 soluble in water. This difficulty would have obliged us to re- 

 nounce this kind of gazometer, if we had not found a method of 

 obviating it. This method consists in putting only atmospherical 

 air into the reservoir, B, and in introducing the gas which we 

 wash to circulate into a bladder, V, (Fig. 3,) inclosed in a glo- 

 bular vessel, M, communicating by the tube, C, with the reser- 

 voir, B. If in this state of things we suppose a regular current 

 of atmospherical air proceeding from the reservoir, B, to the 

 vessel, M, through the tube, C, as the vessel, M, is accurately 

 shut, the air will press uniformly on the bladder, and there will 

 issue out through the tube, D, a regular current of the gas con- 

 tained in the bladder. 



Jf, on the other hand, we suppose that the constant current 

 issuing from the bladder, V, enters into another bladder, V, 

 (fig. 4.) which is empty, and placed, in the same manner as the 

 first, in another globular vessel, M', which is full of air, and 

 communicates with the reservoir, B', of the other gazometer, 

 by the tul>e, C, which reaches down to its bottom. The reser- 

 voir, B', being full of water, and its stop-cock, P', open, the 

 bladder, V, will become gradually filled with gas, and will drive 

 the air of the globular vessel, M', into the reservoir, B', in an 

 uniform current. It is now easy to form an idea of the appara- 

 tus which we employed, and of which we have given a vertical 

 projection in fig. />. 



Band B' are the two lower reservoirs of the two gazometers. 

 The reservoir, B, is supposed full of air, and B' full of water. Vis 

 a bladder filled with gas, whose specific heat is to be determined, 

 hydrogen gas, for example. The corresponding bladder, V, is 

 empty; a, /', r, d, e,J\ g, h, are the stop-cocks. Let us suppose 

 a, CtJ s h, alone open, if we make the gazometer, B, act, a regu- 

 lar current of common air will flow out of B, and pass into the 

 globular vessel, M. It will compress the bladder, V, and force 

 out a current of hydrogen gas. This gas will pass through the 

 tube C D E, about 40 inches in length, and which is surrounded 

 by a la rife r tube, FG. This last tube is kept filled with the 

 ileam Of boiling water by means of a small boiler, K, filled with 

 water, and kept constantly boiling. The steam passes through 



