PHYSICAL PROPERTIES OF WATER, ETC. 49 



an ingenious process he subjected water to external tension, and found that it could 

 support at least fifty atmospheres. The calculation was made on the hypothesis that a 

 moderate negative pressure increases the volume of water as much as an ecpjal positive 

 pressure diminishes it. 



I was led to the conclusion that the internal pressure of a liquid must be greatly 

 superior to the external, as a consequence of the remarkable results of Andrews' 

 experiments on carbonic acid, and of the comments made on them by J. Thomson and 

 Clerk-Maxwell. 1 It was Prof. E. Wiedemann who, while making an abstract of my 

 paper [Appendix E) for the Beiblatter zu den Ann. d. Physih, first called my attention 

 to Berthelot's experiment. 



In Appendix F a short account of Laplace's calculations is given, and it is shown 

 that the work required to carry unit volume of water, from the interior to a distance 

 from the surface greater than the range of molecular forces, is 



2 K x 1 cub. inch, 



where K is the internal molecular pressure per square inch. The speculation above 

 would make this work 



72 inch-tons. 



But, in work units, the heat required to vaporize 1 cub. inch of water at 0°C. is 



62 - 5 



606 x 1390 foot pounds, or 



1/28 



163 inch-tons. 



The two quantities' are at least of the same order of magnitude, and it is to be 

 remembered that what has been taken out in the one case is very small particles of 

 water; in the other, particles of vapour. This raises another extremely difficult 

 question, viz., — What fraction of the whole latent heat is required to convert water, in 

 excessively small drops, into vapour ? 



The comparison above, if it be well founded, would seem to show that the utmost 

 reduction of volume which water at 0° C. can suffer by increase of pressure is - 283 ; i.e. 

 that water can be compressed to somewhat less than 3/4ths of its original bulk, but 

 not further. 



Of course the whole of this speculation is of the roughest character, for two reasons. 

 The Kinetic gas formula has been proved only for cases in which the whole volume of 

 the particles is small compared with the space they occupy. The compression formula 

 is only an approximation, and was obtained for the range of pressures from 150 to 

 450 atmospheres ; while we have extended its application to much higher pressures. 



1 Theory of Heat, chap, vi., London, 1871. 

 (PHYS. CHEM. CHALL. EXP. — PART IV. — 1888.) 7 



