Physical and Chemical Behavior of Solids. 229 



wider apart. Consequently, as a first approximation, we 

 should expect the existence of the parallelism noted above. 



It may be noted, in passing, that the possession of really 

 comparable quantitative data on the actual flow pressure of a 

 series of substances in combination with data (densities and 

 melting-points) which are either already known or easily 

 obtainable, would probably enable us to predict the magnitude 

 of the effect of pressure on the melting point. For by col- 

 lating the experimentally determined flow pressure with the 

 value of cf) calculated for those substances for which the neces- 

 sary data exist, we could determine the relation between (f> and 

 the experimental value, and then by means of this relationship 

 compute, for other substances, <f> and the effect of pressure — 

 either uniform or unequal — upon the melting point ; moreover, 

 we should then be able to compute the latent heat of melting 

 as soon as we knew the density and melting point, both of 

 which quantities can be determined with ease and accuracy. 

 A further point is this, that there is a pronounced parallelism 

 between hardness (as ordinarily measured i and the <f> values, 

 hence between hardness and melting-point ; it might, therefore, 

 be possible to devise a method of measuring hardness from 

 which approximate knowledge of the change of melting-point 

 with pressure — or even the latent heat of melting — could be 

 derived. 



Closely related to the question of the flow of solids pro- 

 duced by compression is that of the effect of pressure on vis- 

 cosity : this we shall consider very briefly. 



The Effect of Pressure on Viscosity. 



The effect of pressure on viscosity has up to the present 

 been studied to but a slight extent; and the investigations 

 recorded are concerned with a very few liquids, the viscosity 

 being measured by the usual transpiration method, in a slightly 

 modified form of apparatus. Warburg and Babo* found that 

 an increase of pressure from 70 to 105 at m. caused an increase 

 of about 30 per cent in the viscosity of liquid carbon dioxide. 

 Later Warburg and Sachs, f working at pressures up to 150 

 atm., found that while pressure increased the viscosity of ether, 

 benzene and liquid carbon dioxide, it decreased that of water, — 

 an unexpected result which had previously been obtained by 

 W. C. Rontgen.J R- Cohen§ investigated the effect on water 

 and certain aqueous solutions of sodium chloride, and on tur- 

 pentine, over a range of pressures extending up to 600 atm., 

 and at temperatures between 0° and 30° ; and found an increase 

 of viscosity with pressure in the case of the more concentrated 



*Ann. Phys., xvii, 390, 1882. f Ibid., xxii, 510, 1884. 



{Ibid., xxii, 518, 1884. 8 Ibid,, xlv. 666, 1892. 



