154 Mr. W. Sutherland on Molecular Refraction. 



Values of ^^^2.^. 



Substance ...CH^ C,R,, C^Ru G^K,, CgHig CgH^o OioH^a 



Q-6 24-8 29-5 34 38-7 43-3 47-9 

 DifFerences...4x4-55 4-7 4-5 4*7 4-6 4-6 



Substance ...CnH24 C12H26 C13H28 C14H30 C15H32 C16H34 



52-4 57-1 61-6 66-1 70-7 75*2 

 DiiFerences 4-5 4-7 4-5 4-5 4-6 4-5 



This is the most extended series of homologous compounds 

 yet studied, and it is seen that from the first to the sixteenth 

 member the difference in molecular refraction corresponding 

 to a difference of CH2 is constant ; the mean value is 4"57, a 

 number almost identical with 4'56, the value deduced by 

 Landolt as the general mean given by the alcohol, aldehyde, 

 fatty acid, and sether series. This fact and the fact that 

 isomeric bodies of similar structure have the same refraction- 

 equivalent show that structure does not produce a pro- 

 nounced effect under some circumstances ; but when we 

 come to consider the effect produced by the double binding 

 of two carbon atoms or by the change of valency in a poly- 

 valent atom we are face to face with a most important fact 

 which we can explain as due either to the structural term b 

 in our formula, or to a distinct change in the physical 

 properties of the atom when it changes its valency (the 

 double-binding of a carbon atom is equivalent to the change 

 of tetrad to triad). But in some cases the difference between 

 the values of the same atom with different valencies, as for 

 example 12 and 20 for iron in the ferrous and ferric states, 

 seems too great to ascribe to structure and makes probable 

 the idea that the actual physical properties of the matter of 

 the atom are different. 



For the full discussion of this interesting question we 

 should require a large number of careful measurements on 

 gases and vapours. Those that we possess are too discordant 

 in certain cases amongst themselves to admit of any great ac- 

 curacy in their discussion. Bleekrode (Proc. Roy. Soc. xxxvii., 

 1884) gives indexes and density for a number of hquefied 

 gases, Mascart for a number of gases and vapours (Compt. 

 Rend. Ixxviii. and Ixxxvi.), and Dulong for a number of gases ; 

 but nothing very definite can be deduced from their results 

 except that in the simpler compounds the atoms have some- 

 times different refraction-equivalents in different compounds. 



