POLYMORPHISM AT HIGH PRESSURES. 



175 



normal cases. As a rule, therefore, the phase of larger volume is the less 

 compressible. A possible way of understanding this is described later. 



In connection with the multiplicity of type of transition lines 

 Roozeboom's classification of triple points may be mentioned. He 

 recognized eight different varieties, according to the relative location 

 of the lines in different quadrants. Figure 32 shows the groups. 

 At the time that he wrote examples of only one group were known, his 

 group 4. The data above include examples of groups 1, 2, 3, 6, and 7. 

 The only cases now missing are 5 and 8. The number of cases in the 

 respective groups are 1, 1; 2, 4; 3, 5; 4, 8; 6, 2; and 7, 1. 



Although the transition lines present great complexity of shape, 

 there are two special types of which no examples have been found, 

 lines with a minimum temperature or a critical point. The minimum 

 temperature does not seem essentially different from a maximum, and 



Figure 32. Reproduction of Roozeboom's classification of triple points. 



there is no reason to suspect that such cases may not exist. But it is 

 different with a critical point. If such exists, it means that by going 

 around this point we can pass by continual gradations from one crystal- 

 line system to another. \Yhether such a possibility exists is open to 

 gra^-e question. It is at least significant that no case of it has been 

 found al)ove. The closest approach to it was on the IV-VI line of 

 camphor, which could not be studied much further because of the 

 extreme stickiness of the transition. Perhaps theoretically there is 

 no objection to such a transition; we can imagine for instance that 

 after a certain stage in the uniform compression of a cubic lattice one 

 of the dimensions begins to change differently from the others, and 

 the crystal becomes tetragonal. But although we can imagine the 

 kinetics of such a change of system, we cannot conceive any adequate 

 physical reason for such a change, and it is certain that any such change 

 of system would be fundamentally opposed to all our present experi- 



