PENETRATION OF DIMENSIONS OF GASES. 1Q7 



of it cement a stopcock, a, of large bore; and, at a suitable distance from that centre, 

 make two boles, the one to have a piece of bent tube, b, cemented into it to serve as a 

 gauge, the other to have a piece of copper wire, c, bent into the shape (c, Jig. 54), 

 passed through, it air-tight, by means of a cork, x, imbued with tallow. The other ex- 

 tremity of the cylindrical glass is likewise to be closed by a flat piece of wood larger than 

 the former, for the purpose of bearing a little cup, d, containing coloured water, into 

 which the gauge tube may dip, and in its centre it is to be perforated to admit of an 

 arrangement, as in fig. 55, where d is a perforated cupping-glass, having a stopcock, 

 b, mounted on it, the farther extremity of which opens into a glass pipe, c, which termi- 

 nates in a hole, in the centre of a flat copper circle, a, three quarters of an inch in diame- 

 ter : this arrangement is to be cemented, air-tight, into the flat piece of wood that closes 

 the lower extremity of the cylindrical glass, as is seen in Jig. 53. Moreover, beneath 

 the cupping-glass there is a glass reservoir, g, of suitable dimensions, filled with water. 

 The object of this arrangement is to fill a soap bubble with any gas, to expose it to at- 

 mospheric air, to burst it at pleasure, and to mark any thermal expansion of the two 

 gases by the indications of the gauge ; the mode in which this is accomplished will be 

 described in the following illustration. 



417. The whole apparatus having stood for some time in a quiet room, along with 

 the gases to be tried, until they have all acquired a uniform temperature, close the 

 lower cock, fill the cupping-glass with hydrogen gas, and raise the reservoir, g, so that 

 the level of the water may be near the top of the cupping-glass. The upper cock, a, 

 being open, convey through its bore, by means of a glass tube of smaller diameter, a 

 little soap-water, which is to be deposited on the copper circle in its centre, over where 

 the glass pipe, e, opens; the tube is then withdrawn. Next open slowly the lower cock, 

 and as the gas is expelled from the cupping-glass by the pressure of the water in the res- 

 ervoir, it expands a bubble in the large cylinder, the displaced atmospheric air passing out 

 through the upper cock. When this bubble has attained the dimensions desired, close 

 both cocks, and observe if the liquid in the gauge be stationary; if so, turn the wire c 

 on its axis, so as to bring its crooked extremity, which is within the cylinder, in con- 

 tact with the bubble; it bursts, there is a thermal disturbance, and an expansion of the 

 two gases, for the fluid in the gauge instantly falls, and as the gases cool, it slowly re- 

 turns to its former position. If a bubble of atmospheric air be employed instead of a 

 bubble of hydrogen, these effects will not ensue. We therefore conclude, that when 

 hydrogen gas is mixed with atmospheric air, the temperature suddenly rises, and there- 

 fore that it is probable that the volume of the mixture is less than the sum of the volume 

 of its integrant constituents. 



418. If a soap bubble, filled with hydrogen, be burst in an atmosphere of nitrogen 

 gas, which may be effected by using a more complex arrangement than that indicated 

 in the preceding section, there is also a thermal expansion, indicating that the constit- 

 uents of ammoniacal gas, even without chemically uniting one with another, exercise 

 an attraction for, or a pressure on, each other, a kind of capillary action. These com- 

 pounds, for they form a distinct class of bodies, a class by no means of small extent, re- 

 quire a distinct name. I have suggested that of capillary compounds, because they 



