90 Mr. Ivory's Investigation of the Heat extricated 



the same condition to which it was before brought by the 

 agency of heat alone. For, in both cases, there is the same 

 volume and the same temperature, and consequently there 

 must be the same density and pressure. 



And, since the air is precisely in the same state, it must 

 have acquired the same quantity of heat in both processes. 

 It follows, therefore, that when air, under a constant pres- 

 sure, expands by the agency of heat, the absolute heat which 

 causes a given rise of temperature, or a given dilatation, is 

 resolvable into two distinct parts ; of which one is capable of 

 producing the given rise of temperature, when the volume of 

 the air remains constant ; and the other enters into the air, 

 and somehow unites with it while it is expanding. Of this 

 latter part there is no perceptible sign, except the cold, or 

 the heat, which appears at the instant of its entrance, or exit. 

 The two heats have no mutual dependence on one another, 

 since either of them may be varied in any manner while the 

 other remains unchanged. It is necessary to distinguish them 

 by appropriate names. The first may be called the heat of 

 tejnpet-attire ; and the second might very properly be named 

 the heat of expansion ; but I shall use the well known term, 

 latent heat, understanding by it the heat that accumulates in 

 a mass of air when the volume increases, and is again extri- 

 cated from it when the volume decreases. 



We must next inquire according to what law the latent 

 heat accumulates when air expands. When a mass of air, 

 under a constant pressure, varies by the application of heat, 

 I assume it as an acknowledged principle that equal quanti- 

 ties of absolute heat produce equal increments of volume. 

 It is evident that this principle cannot be deduced by rea- 

 soning : it must be established by experiment. It is true, so 

 long as an air-thermometer can be reckoned an exact mea- 

 sure of heat ; for, if it were not true, the indications of that 

 instrument would be irregular. But, what proof have we 

 that an air-thermometer measures heat exactly ? To this it 

 must be answered, that we an-ive at the conclusion indirectly, 

 and that there is no direct proof. If we suppose that a given 

 quantity of absolute heat applied to ail bodies caused an incite- 

 ment of volume, always the same in the same body although 

 different in different bodies, it is evident that all bodies would 

 indicate by their dilatations the same progression of tempera- 

 tures. Two thermometers, made of any materials, which 

 agreed in two points of their scales, would always mark the 

 same degrees of heat. Now if we compare two thermometers, 

 one of air and the other of mercury, which have their scales 

 adjusted to the fixt points at which water freezes and boils, 



and 



