120 EEPORT— 1886. 



of air or moisture, the tube being heated to the required temperatures by 

 the vapour of aniline (or methyl beuzoate) at dift'ei'ent pressures. But 

 the method of work and the apparatus devised were varied to suit the 

 requirements of each case. 



Thus for benzene (melting 3'3°) a modification described in § 17 of 

 the memoir cited on page 19 ' is used, the bulb of the thermometer being 

 covered with cotton-wool which is soaked with benzene ; the benzene 

 persisted in solidifying just below fi'eezing-point 3'3°. 



As in the previous case of camphor, so in this case of benzene many 

 experiments were made near the melting-point above and below it, and 

 several with the solid at long intervals below and with the liquid at long 

 intervals above. The lines for solid and liquid, which had slight curva- 

 ture, met at a re-entering angle at a point where the temperature was 

 between 3-0° and 3-6°. 



With acetic acid it was found possible to cool it below the freezing- 

 point 16'4° and keep it liquid, and a large number of good results was 

 obtained; the curves meeting at about 16'3°, the two curves helo^r 

 the melting-point being very obvious, and each being the result of 

 numerous observations. Attempts made with the greatest care by the baro- 

 meter (statical) method gave, as with Regnault, no satisfactory results. 



The observed difference between the solid-vapour pressure and the 

 liquid- vapour pressure for the same temperature was nowhere much more 

 than 1 mm. This gives some idea of the accuracy required in this kind 

 of work. 



The next — and last — case taken in this paper is ice and water. Com- 

 parative results, i.e. results at identical temperatures for ice- vapour 

 and water- vapour were obtained from 0° to —5° ; tables are given of 

 observations of pressure for ice down to —16°; and these results when 

 compared with the results which Professor James Thomson obtained, as 

 mentioned, by recalculation of Regnault's data, are found to give 

 differences of vapour-pressures of ice and water gi-eater than his. 

 But when the observed pressures for ice, for temperatures below 0^, were 

 compared with the pressures calculated from a theoretical formula of 

 Professor Thomson, the authors found that their observed results agreed 

 more nearly with those so calculated than with those calculated fi-om 

 Regnault's results. 



Thus Drs. Ramsay and Young have shown that curves for pressure^ 

 from a liquid and a solid state of the same substance, are not continuous 

 in the cases of camphor, benzene, acetic acid, and water. 



The process in which the thermometer. bulb is covered with cotton- 

 wool (or asbestos fibre), and this soaked with the substance the boil- 

 ing-points of which at different pressures are required, gives results^ 

 according to Ramsay and Young, in Avhich the error due to overheating 

 of the vapour is got rid of, for the substance adhering to the cotton- 

 wool has so much free surface that it will, whether solid or liquid, 

 evaporate freely at the temperature corresponding to the pressure to 

 which it is subjected. The cotton-wool can be re-moistened continually 

 by an arrangement described in this paper and in ' J.C.S.' January 1885. 



In the last-mentioned paper they further describe their apparatus, and 

 show how it is used for solids as well as liquids, and apply it in particular 

 to the case of acetic acid. Regnault had obtained discordant results 



' FhiJ. Trans. Part I. 1884,"?. 47. 



