576 SCIENCE PROGRESS. 



The intimate connection of crystallographic phenomena 

 with many physico-chemical problems relating to solids 

 having been thus pointed out it is of some interest now to 

 consider what views respecting the chemical constitution of 

 crystalline bodies are possible and reasonable, and how these 

 may be expected to affect physical chemistry ; the chemical 

 constitutions of crystalline substances have long been a 

 favourite theme for discussion, but owing to the very 

 gradual accumulation of facts the last word on the subject 

 will long remain unsaid. 



A considerable proportion of crystalline substances exist 

 in several forms which differ crystallographically amongst 

 themselves and are consequently termed polymorphous 

 modifications of the same compound ; thus the a- and ^- 

 forms of chloral hydrate are polymorphs. Since polymorphs 

 are in general very readily converted one into the other by 

 change of temperature, and their properties would seem to 

 differ, only in physical and not in chemical respects, it is 

 generally considered that the chemical molecules in the 

 various polymorphous modifications of a compound are 

 identical ; and inasmuch as the physical properties differ 

 markedly it has, until recently at least, been customary to 

 presume the existence, in polymorphous modifications, of 

 physical molecules composed of aggregates of chemical 

 molecules, the number of the latter constituting these 

 aggregates being different in the various polymorphs. 

 This distinction, being very vague and perhaps somewhat 

 metaphysical in tendency, has like many of the household 

 terms of physics and chemistry, proved invaluable as a 

 disguise to our ignorance ; although from its very nature 

 somewhat difficult to attack, the notion of a distinction 

 between physical and chemical molecules has a weak point 

 in that it is at present quite unnecessary. Tutton, from 

 a consideration of the results of much experimental work 

 in the light of Barlow's investigations into crystalline 

 structure, has demonstrated the probability that, at all 

 events in the cases which he has examined, a crystal is 

 built up merely of chemical molecules, and that these are 

 not necessarily aggregated together in any such way as is 



