CHEMISTRY 745 chlorine; but that in the case of di-derivatives the average Mol. rot. Diff. amount of heat developed is far greater, viz., —16500 Nitrobenzene 9-361. 1units. As the value found for tri-derivatives is only Benzene 111 '314 Fiuorbenzene 9-9701 33,000+13,500, it is to be supposed that the extra-out2Chlorobenzene 12going of affinity is in some way due to an interaction of 1-996 Bromobenzene 14-506{ the two halogen atoms. Similar differences are apparent 4-602 lodobenzene 19-1081 in the case of other properties. Thus the refraction equi- The fact that the differences are negative in the case of valents of chlorine and bromine, calculated from observa- mtro- and fluorbenzene is in itself almost a proof of tions made with compounds such as ethylene chloride and co-operation. When the effects produced by the other bromide, are distinctly higher than those deduced from halogens m benzene are contrasted with those they produce observations made with monochlorides and bromides. m propylic compounds, the differences observed are— Again, Thorpe and Rodger have shown that the viscosity constants of dichlorides and dibromides are remarkably Chlorine -0-253. Bromine - 0-086. Iodine 0-321. different from those of the monochlorides and monobro- In benzene, therefore, chlorine has a slight negative influmides. But the effect is most obvious in the case of iodo- ence, and only iodine has a marked extra positive effect: compounds. Thus the refraction equivalent (58-2) of such a result is in harmony with the fact that in carbon methylene iodide, CH212, is far higher than that calculated compounds generally iodine alone has a marked tendency (52T) on the assumption that the value of each atom of to exhibit residual affinity. iodine is that of the iodine atom in methyl iodide. In all But by far the strongest argument in favour of the view the cases referred to the two atoms of halogen are asso- here advocated is afforded by the consideration of the conciated either with a single carbon atom or with contiguous ditions which determine the appearance of colour carbon atoms, and there is little doubt that this will be in carbon compounds. The majority of col- carhon"1 found to be necessary to the production of the effect under q oured compounds may be referred to comconsideration. One of the very few observations bearing one type, that of quinone (a). Speak- P°undson this point is that of Perkin, showing that the molecular in g generally, they contain two ethenoid groups rotatory power acquired in a magnetic field is considerably C : R", in ortho- or para-positions, in association higher in the case of propylene bromide, CH3.CHBr.CH9Br, with an unsaturated benzenoid nucleus. As (10 82), than in that of the isomeric trimethylene bromide ^ compounds such as diketohexamethylene (b), in u CH.2Br.CH2.CH2Br, (10.34). The “co-operative effect” which the nucleus is saturated, are colourless, is most obvious, however, in compounds in which several colour must be supposed to originate in the CO benzenoid and ethenoid systems are conjoined. Thus if co-operation of the two ethenoids with the unthe atomic refractive power of carbon be calculated from the saturated nucleus, and each must be regarded H. H molecular refractive power of diphenyl, C6H5.C6H5, its ap- as an independent light - absorbing centre. parent value is about 6‘4, whereas in benzene it is only Whether the nucleus be represented as a H92 H / 0 about 6. Stilbene, in which an ethenoid group is interposed centric or as an ethenoid structure matters CO between two phenyl groups, Ph.CH:CH.Ph, is still more R" little j suffice it to say that coloured comabnormal in its behaviour, its molecular refractive pounds are known of the dihydrobenzene H. power (113-4) exceeding the calculated value by 12 units. type (c), so that it is to be supposed that either The very high refractive power of cinnamic aldehyde, the centric or the ethenoid-system may act as H,2 which exceeds the calculated value by 10 units, may be the third centre. A colour-producing system / accounted for in a similar way, and even seems to R" such as is present in quinone may, therefore, su est gg that this compound is to be represented as ^ Ph.CH:C:CH.OH, rather than as Ph.CH:CH.COH, i.e., be represented by the symbol =II If this symbol be that it is the isodynamic form of the aldehyde. It is by no means improbable that the principle here adopted as a broad definition of quinonoid structure, it is developed may prove to be of wide application, and that probably safe to assert that all coloured substances are it will afford an explanation of the many “anomalies” quinonoids. Iodoform is a compound of particular interest which are apparent on contrasting the properties especially from this point of view. Attention has already been directed of benzenoid compounds with those of the paraffinoids ‘} to the exceptionally high refractive power of methylene it may also serve to explain the fact that they often differ iodide, CH I , and to the probability that this is due to the 2 from paraffinoids in physical properties to an extent alto- co-operation2 of the two iodine atoms. Such a compound may gether disproportionate to the extent to which they differ be regarded as the analogue of diketohexamethylene, each from them in their chemical properties. Perkin’s observa- iodine atom exercising an influence similar to that exercised tions on the rotation of polarized light in the magnetic by the CO group in the keto-compound. When a third field by benzenoid compounds afford many striking illus- atom of iodine is introduced, forming iodoform, CHR, trations of their peculiar behaviour (Trans. Chem. Soc. an effect is produced equivalent to that observed when 1896, p. 1026). Thus, on contrasting the rotatory power is converted into quinone and a of phenol with that of benzene, and that of heptylic diketohexamethylene third absorbing centre is developed—the highly refractive alcohol with that of heptane, the difference due to the di-iodide becoming intensely yellow when converted into displacement of H by OH is 0-856 in the case of phenol, the tri-iodo-derivative. The introduction of a fourth iodine but only 0-181 in the case of the alcohol. That the atom has the effect of greatly intensifying the colour. superior effect exercised by the OH group is at least From this point of view it is easy to understand that all mainly due to the co-operative action of the centric system tri- and tetriodides, but only polymerized di-iodides such as and the oxygen appears probable from the fact that the mercuric iodide (Hgl^, should be coloured. difference between toluene, Ph.CH3, and benzylic alcohol, The appearance of colour in an open or pseudo-quinonoid Ph.CH2.OH (0-243), is but slightly in excess of the paraf- system, as distinguished from a closed or true quinonoid finoid value. Co-operative action can only be exerted by system, is dependent on a certain intensity of effect being contiguous radicles. A comparison of the values for ben- developed. Chlorine and bromine do not act sufficiently zene, nitrobenzene, and the mono-derivatives containing powerfully to condition colour, neither does the presence halogens, leads to similar conclusions :— of three phenyl groups in association with a single carbon S. II. — 94
