1891.] on Liquids and Gases. 867 



goes by the name of Gay-Lussac's law. Now, if we do not allow the 

 volume of the gas to increase, we shall find that the pressure will 

 increase in the same proportion that the volume would have increased 

 had the gas been allowed to expand, the pressure having been kept 

 constant. To decrease the volume of the gas, then, according to 

 Boyle's law, will require a higher initial pressure, and if we were to 

 represent the results by a curve we should get a hyperbola, as before, 

 but one lying higher as regards pressures. And so we should get a 

 set of hyperbolas for higher and higher temperatures. 



We have experimented up to the present with air, a mixture of 

 two gases, oxygen and nitrogen; and the boiling-point of both of 

 these elements lies at very low temperatures, — 184° and — 193*1° 

 respectively. The ordinary atmospheric temperature lies a long way 

 above the boiling-points of liquid oxygen and liquid nitrogen at the 

 ordinary atmospheric pressure. But it is open to us to study a gas, 

 which, at the ordinary atmosf)heric temperature exists in the liquid 

 state ; and for this purpose I shall choose water-gas ; in order that it 

 may be a gas at ordinary atmospheric pressure, however, we must 

 heat it to a temperature above 100° C, its boiling-point. This tube 

 contains water-gas at a temperature of 105° C. ; it is under ordinary 

 pressure, for the mercury columns are at the same level in both the 

 tube and in this reservoir, which communicates with the lower end 

 of the tube by means of the india-rubber tubing. The temperature, 

 105°, is maintained by the vapour of chlorobenzene, boiling in the 

 bulb sealed to the jacket, at a pressure lower than that of the 

 atmosphere. 



Let us now examine the effect of increasing pressure. On raising 

 the reservoir, the volume of the gas is diminished, as usual, and 

 nearly in the ratio given by Boyle's law ; that is, the volume 

 decreases in the same proportion as the pressure increases. But a 

 change is soon observed ; the pressure soon ceases to rise ; the distance 

 between the mercury in the reservoir and that in the tube remains 

 constant, and the gas is now condensing to liquid. The pressure 

 continues constant during this change, and it is only when all the 

 water-gas has condensed to liquid water that the pressure again rises. 

 After all gas is condensed, an enormous increase of pressure is 

 necessary to cause any measurable decrease in volume, for liquid 

 water scarcely yields to pressure, and in such a tube as this no 

 measurements could be attempted with success. 



Representing this diagrammatically, the right-hand part of the 

 curve represents the compression of the gas, and the curve is, as 

 before, nearly a hyperbola. Then comes a break, and great increase 

 in volume occurs without rise of pressure, represented by a horizontal 

 line. The substance in the tube here consists of water-gas in 

 presence of water ; the vertical, or nearly vertical, line represents 

 the sudden and great rise of pressure, where liquid water is being 

 slightly compressed. The pressure registered by the horizontal line 



