58o SCIENCE PROGRESS. 



of the method ; it could not be expected to hold rigidly as 

 concentrated solutions such as are frequently employed, and 

 the electrolytic dissociation occuring in aqueous solutions 

 introduces a complication which cannot always be success- 

 fully eliminated. There is little doubt that Nernst's work 

 in this field opens the way to a most powerful method of 

 determining the molecular weights of crystalline substances/ 

 These and many similar considerations seem to indicate 

 that at present the only legitimate view concerning the 

 matter is that the molecular weio-ht is in g-eneral the same 

 in both the liquid and the solid state ; there is thus a com- 

 plete analogy not only between the molecular properties of 

 gases and liquids on the one hand but also between those 

 of liquids and solids on the other. But just as in cases of 

 the former kind exceptions arise, the molecular weight not 

 being the same in the gaseous as in the liquid state, so 

 doubtless exceptional cases will be found to exist amongst 

 the latter. Consider such cases of polymorphism as that of 

 the diamond and graphite or of the red and yellow crystal- 

 line modifications of phosphorus ; these are frequently 

 quoted cases of allotropy of which the generally recognised 

 meaning is that it is merely polymorphism occurring amongst 

 the elements. The differences between the diamond and 

 graphite or between red and yellow phosphorus are far 

 'greater than those observed between ordinary polymorphs 

 such as those quoted above ; it would therefore seem 

 only reasonable to accept the view that such substances are 

 not ordinary polymorphs but chemical polymerides just 

 as oxygen O,. and ozone O3, or acetylene, C2H0, and 

 benzene CgH^;, are polymeric substances. Again, it is 

 known that the molecular weight of sulphur only corre- 

 sponds to the formula So at temperatures high above 

 its boiling point whilst near the boiling point the 

 gaseous molecules are at least three times as large and 



^ Whilst this article was passing through the press, Dr. Andreas Fock 

 published a valuable paper in which he shows, by apjjlying the equilibrium 

 method briefly described above to many pairs of isomorphous salts, that all 

 the salts examined have the same molecular weight in the crystalline, as in 

 the dissolved condition. 



