64 sosman: problems of the oxides of iron 



similar in the form of their temperature-magnetism curves. 

 In each case the effect of rising temperature appears some dis- 

 tance below the inversion point, and increases more and more 

 rapidly as the inversion is approached. One cannot avoid the 

 conclusion that these inversions all have the same fundamental 

 basis. When that basis is found, we will have considerably 

 advanced our knowledge both of polymorphism and of magnetism. 



So much has been said about polymorphism (including under 

 this term allotropy, polymerism, and all the other names given 

 to the phenomena of sudden change of physical and chemical 

 properties without change of percentage composition), and so 

 much that has been said is mere quibbling over the meanings 

 of words and phrases, that I shall not undertake any general 

 review of the questions that are raised by the inversions of iron 

 and its oxides. But there is one problem to which considerable 

 attention has lately been given, namely, the problem of the two 

 types of polymorphism in crystalline substances. 



Polymorphic changes in crystalline solids can be divided 

 roughly, as is well known, into two groups. On the one hand, 

 there are those changes which carry with them rather profound 

 changes in physical and chemical properties and which often 

 require considerable time to bring about. On the other hand 

 are those reversible changes which occur without delay at a 

 given temperature and which involve relatively small changes 

 in crystal form and in physical properties. An excellent example 

 of a substance showing both types distinctly is silica. Si0 2 

 occurs in three principal forms, quartz, tridymite, and cristob- 

 alite, which are only slowly transformed one into another, 

 and which have certain ranges of stability outside of which they 

 frequently occur in an unstable but very persistent state. Each 

 of these principal forms, on the other hand, possesses one or two 

 inversion points, at which reversible crystalline changes occur: 

 quartz at 575°; tridymite at 117° and 163°; cristobalite at a 

 temperature varying from 200° to 275°, depending upon the 

 previous history of the crystal. 18 



11 Fenner, C. N. Am. Jour. Sci., 36: 331-384. 1913. 



