430 



Mr. J. E. Marsh on 



the case is different, we have no longer a simple repetition of 

 the dehydrogenizing process. It is not a mere change from a 

 single to a double linking which gives us terephthalic from 

 dihydroterephthalic acid. The change is much more pro- 

 found. Hence we are unable to assert that in terephthalic 

 acid or in benzene we have carbon atoms united alternately 

 by double and single linkings. 



The constitution of benzene, to which v. Baeyer is led, is 

 essentially the same as that previously proposed by Prof. 

 Armstrong (this Journal, 1887, vol. xxiii. p. 108), namely, 

 that the six carbon atoms are singly linked, and that the six 

 free affinities react each upon all the rest, and are directed 

 * towards the centre of the system. 



CH 



HC(1)CH 



00 



HC(|)CH 

 CH 



Tig. 7. 



Now, can we interpret such a constitution as this on the 

 hypothesis of the tetrahedric grouping ? Or, to put it in a 

 more concrete form, can we represent the six carbon atoms 

 of benzene and their mode of union with one another by 

 means of regular tetrahedrons which shall represent the 

 carbon atoms and have their affinities situated at the angles ? 

 There can in fact be constructed eight different geometrical 

 formulae for benzene satisfying the above conditions, of which 

 two are represented. 



Fig. 8. Fig. 9. 



The others being constructed by supposing the carbon atoms 

 to occupy different positions with regard to the plane passing 

 through their points of union: thus 5 on one side of the plane 

 and 1 on the other, or 4 on one and 2 on the other, and also 

 by different orientation. Now we know no isomers of benzene, 

 and it may be that benzene fulfils the conditions of all the 

 eight formulae or of only one of them. Let us consider the two 

 given as being perhaps the most probable. Of these, the first 



