406 J. SAKURAI 



Meyer's calculation. Lösclimidt's calculation. observed. 



II. = G X 



96 95.9 



Now, Meyer's and LJlsclimidt's ponstants are only applicable to 

 benzene, and cannot be regarded as giving results agreeing with the 

 observed values, ^^•hen applied to the homologues of benzene. For, 



( 1 ) If we regard these constants to be applied to the side-chains 

 as well, then the calculated values of the homolofrues of benzene 

 would be less than the observed values ; for, while benzene possesses 

 abnormally low specific volume, its homologues show the constant 

 increase of 22 in the volume for an increment of C H„. The hvdroofen 

 in the side-chain cannot, therefoi'e, possess the value 0, and still less 

 the value 3. .5. 



(2) If we regard the above constants as applicalilc to the 

 benzene nucleus only, and the carbon and hydrogen in the side-chain 

 as possessing their normal ^alues, then the use of JMever's constants 

 means an advantage of 2.5 units over Kopp's A'alues for one atom 

 of hydrogen replaced by (' IT;, (^ H.,, &c ; fir Cr. Il5=91 instead of 

 93.5. For di-substitution products the advantage is 2.(1, ior tri- 

 snbstitution products 1.5, and so on, until we shall find in hexamethvl 

 benzene, fn* example, n body possessing normal specific \olume. 

 This, however, does not seem to be the case. The use of Löschniidt's 

 constants for onlv the benzene-nucleus leads to a sinijular result. l'V,r 

 mono-substitution products the advantage over Kopp's constants is 

 1 unit, as C\ 1L,=92.5 instead of 93.5 ; but for di-substitution pro- 

 ducts tlie disadvantage is 1 unit, for tri-substitution proilucts the 

 disadcantage is 2 units, and so on. 



