62 Drs. Ramsay and Young on the Influence 



that our work was merely qualitative ; on the contrary, it was 

 rigorously quantitative ; and we then showed that the num- 

 bers calculated for the vapour-pressure of ice, using as data 

 an extension of the vapour-pressures of water below 0°, ex- 

 trapolated from Regnault's measurements above 0°, his deter- 

 minations of the heat of vaporization of water and fusion 

 of ice, and of the specific heat of ice, agreed closely with 

 those found by us. As regards benzene, however, our work 

 had no pretence to be quantitative. 



Fischer's experiments were made by a process identical in 

 principle with that employed by Regnault. The water, or 

 benzene, on which he experimented, was introduced, by a pro- 

 cess well devised for excluding air, into a vacuum connected 

 with a gauge, and on alterating the temperature of the liquid, 

 alteration of pressure was noted and registered. His experi- 

 mental results are very regular, and, so far as water and ice 

 are concerned, confirm ours., and agree well with theory. 



But although Fischer's experimental results with benzene 

 are equally regular, yet they present certain anomalies which 

 are difficult to explain. From his results he calculated con- 

 stants — one series to represent the relations of the pressure of 

 vapour in contact with liquid, and the other to represent 

 similar relations for vapour in contact with solid, employing 

 formulas of the general form p = a + bt + ct 2 . From the num- 

 bers calculated by means of these constants, he concluded that 

 the vapour-pressure of liquid benzene is not identical with that 

 of solid benzene at the melting-point of the solid. This conclu- 

 sion is evidently opposed to the second law of thermodynamics ; 

 and, if it had not been apparently supported by Fischer's really 

 excellent experimental measurements, might have been dis- 

 missed at once as absurd. But on revising Fischer's results, 

 we find that the constants employed by him, if used to calcu- 

 late the vapour-pressures of the solid at low temperatures, 

 give results which are by no means in accordance with his 

 measurements. Indeed at —8° the calculated pressure is 

 13*51 millim., whereas Fischer found 14'2 millim. ; and it is 

 evident, from a graphic representation of his results, that the 

 divergence would increase at lower temperatures. Now it is 

 known that the relations of pressure to temperature are better 

 expressed by means of a formula of the type suggested by 

 Biot, p = a + bat + c/3* ; or, for a small range of pressure, by 

 the simpler form p = a + ba t . On calculating constants from 

 Fischer's results by means of this formula, we found that while 

 a curve was obtained agreeing better with his experimental 

 results, the anomaly which he supposed (viz. want of coinci- 



