136 REPORTS ON THE STATE OF SCIENCE. 
origin of colour in carbon-compounds. Broadly speaking, two alternative 
conceptions have been advocated. On the one hand, it has been suggested 
that the selective absorption of light by coloured compounds is due to a 
peculiar structure of the molecule and that certain types of structure 
in which ethenoid linkages and other unsaturated centres are present 
are specially susceptible to the ‘ optical resonance ’ which is universally 
recognised as the cause of the absorption of light by vapours and is 
probably also the cause of the less abrupt absorption of light by liquids 
and solids. On the other hand, it has been suggested that colour is due 
not so much to molecular structure as to change of structure, and that 
only those molecules which are capable of existing in isomeric forms, 
and may, therefore, be supposed to be in a state of continued oscillatory 
isomeric change, are capable of resonating to light of definite periodicity. 
This theory was introduced by Hewitt* to account for fluorescence; it 
was applied by Armstrong and Lowry ? in explanation of the storage of 
energy in phosphorescent and in triboluminescent substances, and has 
been put forward by Baly and Desch? and by Baly and Stewart‘ as 
an explanation of colour in organic compounds; it has also been used 
by Baeyer * and by Green ® to account for the specially intense colour 
of some of the derivatives of triphenylmethane. 
Two different types of isomeric change have been specially considered, 
namely— 
(1) Isomeric changes involving the oscillatory transference of an 
atom of hydrogen from carbon to oxygen as in ethylic acetoacetate, from 
nitrogen to oxygen as in isatin, and from oxygen to oxygen as in p- 
nitrosophenol. 
(2) Changes involving only a rearrangement of the bonds in the 
molecule without any substantial alteration in the relative positions of 
the atoms, as, for instance, in Kekulé’s well-known hypothesis, in which 
an oscillation of the linkages in benzene was assumed to take place in 
order to account for the identity of the 1:2 and 1:6 diderivatives. 
Changes of the former kind have been carefully investigated in 
several typical compounds and have been found (in direct contradiction 
to Laar’s hypothesis of tautomerism) to be subject to the ordinary laws 
governing chemical change. The occurrence of change of the latter 
type is at present purely speculative, as no case is known in which the 
occurrence of such a change has been demonstrated; but, in view of the 
extreme stability of the linkages in compounds such as the sugars, it 
is exceedingly probable that if ever an example of the second type of 
change be discovered, it will be found to obey the same laws, and to 
be governed by the same conditions as those which obtain in isomeric 
changes of the first type. 
The most important consequences of the application to luminous 
phenomena of considerations based upon the ordinary laws of chemical 
change are perhaps those concerning the effects produced by the pre- 
sence of foreign substances, and by the change from the fluid to the 
solid state. 
' Proc. Chem. Soc., 1900, 16,3; Zit. phys. Chem., 1900, 24, 1. 
* Proc. Roy. Soe., 1908, 72, 258-264. 
’ Trans. Chem. Soc , 1904, 85, 1029; 1905. 87, 766. 
‘ Thid., 1906, 89, 502. 5 Annalen£1907, 354, 152, 
5 Proc. Chem. Soe,, 1908, 24, 206. 
