( 291 ) 



more accurately than before, from the relative proportion in which 

 the isomers C„H,.AC and CbH^BC are formed by the introduction of 

 C in CeHjA or CgHsB, in Avhat proportion the isomers CgHjABC 

 are formed by the introduction of in CgH^AB. He observes first 

 of all that in a substance C^H^A there are two ortho and twometa 

 positions against one para position so that if the relation of the 

 isomers is as CgH^AC p : q : r (ortho, meta, para) this relation for 

 each of the ortho and meta positions and for the [)ara position will 

 be V, p : V, q : r 



He further gives the preference to an addition of these figures of 

 proportion instead of a multiplication, which had been used by me up 

 to the present in the prediction of these isomers. He prefers the 

 addition because he considers the figures of proportion to be proportional 

 to the directing forces which are exercised by the groups A and B 

 on the other positions of the core and that the cooperation of such 

 directing forces on one H-atom should be represented by a sum. 

 But only the j)roportion of those directing forces are known and 

 not their absolute value; the force which, in the nitration of 

 nitrobenzene, pushes the NO^-gro.up towards the ?/i-position may be 

 of quite a different order than the force which in the |nitration 

 of benzoic acid directs the same group towards the m-position. 

 Therefore the figures which represent the directing forces (or are 

 proportionate to the same) of two different groups cannot always 

 be simply added together; this then will be permissible only when 

 the two substituents present are equal. 



As an example of his method of calculation the following may be 

 mentioned. As in the nitration of bromobenzene 37.67o ortho, 62.17o 

 para and 0.37o of meta nitrobromobenzene is formed, the substitution 

 in the different positions of the benzene core takes place in the 



Br 



proportion | | ; for the proportion in which the isomers are 



Ü.15\ /0.15 

 62.1 



formed in the nitration of o-dibromobenzene the calculation gives 



l8.8 + 0.15/^\Br 



I I or 62.25:18.95; or 76.7 "/„ asymmetric nitro- 



C2.I -f 0.15\ >^ 18.8 4-0.15 

 G2.1 -f 0.1.5 



o-dibromobenzene and 23.3 7o vicinal whilst the experiment gave 

 81,3 "/'o asymmetric and 18.3 " „ vicinal. 



It cannot be denied that in a number of cases this method of 

 calculation gives figures which approach to the experimental ones a 



