EXPEEIMENTAL KNOWLEDaE OF THE PEOPEETIES OF MATTER. 115 



firming this result of theory that the ice- vapour and water- vapour curves 

 are distinct and meet at an angle. 



Effect of Pressure on Melting-point. 



Professor J. Thomson had proved that the melting-point of ice must 

 be lowered by pressure, and had calculated the amount of this lowering 

 of the freezing-point by a given pressure ; his result was subsequently 

 experimentally verified by Sir William Thomson. The amount of the 

 lowering is ■0073?i° for n atmospheres of pressure.' 



Mousson 2 made experiments with a very powerful hydraulic press 

 with a view to keep ice liquid at a temperature much below zero, or to 

 lower the melting-point of ice many degrees by immense pressure ; these 

 experiments suggested themselves to him in consequence of Sir W. 

 Thomson's experiment in which by 17 atmospheres' pressure he lowered 

 the melting-point of ice more than one-tenth of a degree. Mousson ob- 

 tained the following results : — first, he succeeded by great pressure on 

 water in preventing the solidification of it till its temperature was lowered 

 to —5°; second, he lowered the temperature of a piece of ice to —18°, 

 and liquefied it by a pressure which he calculated to have been not less 

 than 13,000 atmospheres, and the diminution of volume he estimated at 

 13 per cent. 



Bunsen^ obtained results for the raising of melting-points of some 

 substances which expand during fusion ; thus spermaceti at 1 atmo fuses 

 at 4!7'7°, but at 156 atmos at 50'9°. So paraffin melting at one atrao at 

 46'3° melts at 100 atmos at 49"9°. And Hopkins, with spermaceti, wax, 

 sulphur, and stearine, using pressures up to 800 atmos, obtained a rise of 

 melting-point with increased pressure."* 



Professor Dewar has quite recently ^ made a series of experiments by 

 a Cailletet apparatus on the relation between the temperature at which ice 

 melts under different pressures. The temperatures were measured by a 

 thermo-electric arrangement ; a thermo-junction was frozen in a test-tube 

 placed inside .the iron bottle of the apparatus, whilst another thermo- 

 junction outside was kept at the constant temperature of ice melting at 

 atmospheric pressui'C ; the two thermo-junctions were connected with a 

 galvanometer, by means of the deflections of the needle of which the difl^er- 

 once of temperature of the junctions was deduced. 



The freezing-point was lowered 0-18° for 25 atmos, and 2-1° for 300 

 atmos, giving a mean i-eduction of 0-0072 for 1 atmo. Similar results up 

 to 700 atmos agreed in giving the same reduction per atmosphere. By 

 this method, therefore, it is possible always to graduate the pressure scale 

 of the Cailletet apparatus or to correct the graduations. 



Definitions of Boiling-points. 



The view generally adopted with reference to the boiling-point of a 

 liquid is that it is the temperature at which the vapour given off" from its 



■ Phil. Mag. (3) xxxvii. p. 123. 



^ Annales de Chimie et de Physique, 1859 (.3), Ivi. p. 252 ; Poggendorff's Annalen 

 1858, t. cv. p. 161. 



' Poggendorff's An7ialen, Ixxxi. p. 562. 

 * Report Brit. A.ssoo. 1854, p. 57. 

 ' Proc. Roy. Soc. xxx. p. 533. 



I2 



