8 Prof. R . Bunsen on a Method of determining the 



following expression, supposing that p t (as is most convenient) 

 lies for this second weighing on. the balance : — 

 G=T y + T 5 +p t + m y + Sr y +p q , 

 and consequently, by subtraction from both equations, the 

 weight P^, viz. the weight of gas or vapour contained in the 

 vessel IV., 



¥.g=Pg + ™v— m g + lr y ~dr g —A iy — A 4 . 

 Just in the same way the weight of Pi is found, which is 

 that of the vessel III., containing dry air at equal temperature 

 and pressure : — 



Pi =Pl + m v — m i + % r v — Sri — A m — A 3 . 

 Now, since the weights of equal volumes of vapours or gases, 

 at a sufficient distance from their points of condensation, and 

 when both are at one and the same temperature and under one 

 and the same pressure, are proportional to their densities, it 

 follows that the desired specific gravity of the gas or vapour is 

 given by the equation 



When one does not possess the means of completely pumping 

 the air from the vessel V., it is necessary to calculate the weight 

 A of the residual air, which is deduced from the volume of the 

 vessel, the temperature, and the height of the column of mer- 

 cury in the manometer at the time of sealing up. The equation 

 iS then S . J^ + 7ft v -777 ff + 3r v -Sr,-A IV -A 4 + A _ 

 jt? 1 +m v — m 1 + ^r y — S^ — Ani— A 3 + X 

 When the vessel V., containing 176*6 cub. centims., was sealed 

 up, the temperature stood at 18° C, and the height of the 

 manometer m, 003 of mercury; therefore 

 A = 0-00085 grm. 



The connectors l k and V^ of the vessels I. and V. only play 

 the part of weights during the weighing ; so that it is convenient 

 to replace them from the very first by glass weights, which 

 are less changeable, and by means of which new connectors 

 can be adjusted to the proper weight for the vessels III. and 

 IV. It is especially advisable, when the connectors III& and 

 IVjt have been in contact with dry gases or vapours, to weigh 

 „ them before each new experiment, and, when necessary, to pro- 

 vide them with a new caoutchouc lining. 



The thermostat, or air-bath for constant temperature, in which 

 the vessels with dry air or gases or vapours can be heated to 

 one and the same temperature, is represented in fig. 3. 



A is a cylinder of sheet-copper 40 centimetres long, closed at 

 the lower end ; its section is an ellipse, having a horizontal major 



