PRESSURE IN FORMATION OF ROCKS AND MINERALS 733 



words, the fact that a rock has flowed is no criterion of its strength, 

 nor does it indicate that at the time of flow the rock was necessarily 

 liquid as a whole. 



Let us now consider briefly the effects of uniform pressure upon 

 some of the properties of a pure crystalline substance, taking up 

 first its melting-curve. For a pure crystalline substance there is, 

 at a definite pressure, a single temperature — the melting-point — ^at 

 which its solid and hquid forms can coexist indefinitely in equilib- 

 rium, this being the temperature at which the vapor pressure of 

 solid and that of the liquid become identical. The melting-point 

 is influenced by change of uniform pressure — such as we may con- 

 ceive to be exerted by either a gas or a mobile liquid (e.g., oil) which 

 is insoluble in the substance. By joining up the melting-points at 

 various pressures we obtain the melting-curve, the slope of which 

 (dT/dP) at any point can easily be shown to be 



dT^TAV 

 dP AH 



where T, AV, and AH are respectively the melting-point, the change 

 of volume, and the heat change (latent heat of fusion) at that par- 

 ticular point on the curve. From this equation it is obvious that 

 the general shape of the curve is determined by the way in which 

 AV and AH vary with the pressure; that the condition for a 

 maximum melting-point is that AV should vanish while AH remains 

 finite, and for a critical end-point solid-liquid that AV and AH 

 should vanish simultaneously. But we cannot predict from theo- 

 retical principles how AH and AV will vary — in other words, the 

 shape of the curve can be determined only by experiment. In this 

 connection it may be noted that there is no general relation between 

 AH and AV, for we may have a large latent heat associated with 

 a small volume change and vice versa; indeed, in two particular 

 cases we have a negative volume change with a positive heat effect, 

 these two exceptional cases — ice I and bismuth — being the only 

 pure substances whose melting-point is known to be lowered by 

 pressure. There is little doubt that the silicates are normal in this 

 respect, showing a rise of melting-point with increase of uniform 

 pressure, because everything indicates that they expand on melting 



