POLYMORPHISM AT HIGH PRESSURES. 149 



Chemically Related Groups. 



(1) Nitrates. — KXO3*, I trigonal, II rhombic; NH4NO3*, I cubic, 

 II tetragonal, III monoclinic, IV rhombic, V tetragonal; CsNOs*, 



I cubic, II trigonal; RbNOs*, I trigonal, II cubic, III trigonal; 

 TINO3*, I cubic, II trigonal, III rhombic; AgNOs*, I trigonal, II 

 rhombic; NaXOa, trigonal; LiNOs, trigonal, rhombic, cubic; HgNOs; 

 Hg(N03)2; Pb(N03)2 cubic, monoclinic; A1(N03)3. This group has 

 already been sufficiently commented on in a previous paper. The 

 results are given again here for the sake of completeness and conven- 

 ience of reference. 



Halogen Compounds. — CCI4*, CBr/, I cubic, II monoclinic; 

 SiCb; CoCle*, I cubic, II triclinic. III rhombic; Hgl2*, I rhombic, 



II tetragonal; Hg2l2; AgCl, cubic; AgBr, cubic; Agl*, I cubic, II 

 cubic cjuasi-hexagonal ; NH4CI*, I cubic, II cubic; NH4Br*, I cubic, 

 II cubic; NH4I*, I cubic, II cubic; Til*, I cubic, II rhombic; KI, 

 cubic; Nal, cubic; Lil; CU2I2*, cubic; Sbis, trigonal. It seems to be 

 a general rule in the case of halogen compounds that compounds 

 formed by the substitution of one halogen for another crystallize 

 isomorphously, although there are exceptions. The effect of the sub- 

 stitution, in case the substance is polymorphic, is to displace the 

 temperature of transition. 



CCI4 and CBr4 each have three forms and might at first glance seem 

 to be a very good example of isopolymorphism. But this relation is 

 probably illusory, because the general character of the phase dia- 

 grams is so different, one being the inverse of the other. The crystal- 

 line form of ordinar\- CCI4 seems not to be known, so that we cannot 

 say whether the ordinary modifications of CCI4 and CBr4 correspond 

 or not. 



Possibly SiCb may show other forms at lower temperatures and 

 higher pressures than those reached here, in analogy with CCI4, 

 because in many compounds an atom of silicon is capable of replacing 

 an atom of carbon. The expectation need not be very strong, how- 

 ever, because the physical properties of corresponding Si and C com- 

 pounds are often very different, Si02 and COo for example. 



Other compounds analogous to C2CI6 are known to Ije polymorphic, 

 and it would be most interesting to examine them under pressure. 

 These are C2Cl4Br2, C2Cl2Br4, C2Br6. Preparation of them was beyond 

 my chemical resources. 



Hgl2 is unique among all substances so far examined in that its 



