492 SCIENCE PROGRESS. 



a difference in the arrangement of the atoms. But it does 

 not seem clear that Gay-Lussac referred to anything more 

 than what we should now call a difference of linking or group- 

 ing of the atoms. Thus in a compound (RCNO) n , there 

 may be special linking between R and N, and C and O, 

 or between R and O, and N and C, and in either case the 

 atoms may have the same position, thus : — 



r; o r o 



n! c n c 



The dotted lines indicate the division into groups. This 

 isomerism is realised in the case where R isan alkyl (methyl 

 or ethyl). For in the decomposition products of an isocy- 

 anurate the alkyl appears linked with N, and C with O, 

 whereas when the cyanurate is broken up by the same 

 reagents the alkyl clings persistently to O, C retaining N. 

 Thus the groups pass unchanged from molecule to molecule. 



And here we come upon a class of facts which led 

 chemists to suspect, or perhaps unconsciously to assume, 

 the definite as opposed to the chaotic state for the atoms 

 in a molecule. It was the persistence of types which 

 brought about this result, — the habit which bodies have of 

 preserving, despite substitutions, their properties and their 

 chemical functions. "Thus," says Dumas, "in a building, 

 stone is replaced by stone but the form is preserved." 

 Here the term "building" implies fixity of the relative 

 positions of the atoms. The fact that aromatic radicals, as 

 C 6 H 5 , and C I0 H 7 , introduced into fatty molecules preserve 

 their nature, that propyl remains different from isopropyl, 

 shows that there are certain groups of atoms which are 

 separated from the rest, not chaotically jumbled with them. 

 But, within these groups, is the position of the atoms, or of 

 their orbits, a constant or an ever-changing one ? It re- 

 mained for Biot and Pasteur to supply the answer to this 

 question. 



Biot showed as early as 1817 that a substance (oil of 

 turpentine) may even in the form of vapour rotate the plane 

 of polarised light, and concluded that this power depends 

 on the internal structure of the molecule. 



