44 SECTIONAL ADDRESSES 
depending presumably on the distribution of nuclei as well as on the 
distribution of electron-densities as studied by W. L. Bragg and others. 
It is, indeed, an interesting exercise to construct a model of the molecule 
of camphor and then to inquire to which other electrons the shared valency- 
electrons of the carbonyl group must be coupled, in order to develop 
the magnificent loop which appears in the curve of rotatory dispersion in 
the region of absorption. ‘The question answers itself by mere inspection 
of the model, since it is clear that al] the electrons are involved, and 
not merely one, two or four of them. ‘Thus, even when the carbonyl 
group has been linked to two dissimilar radicals, either in an open-chain 
ketone or in a cyclic ketone, no rotatory power at all is developed. The 
whole of the rotatory power of camphor therefore depends on the contrast 
between the two radicals —CH,.CH,— and —C(CH3;),— which lie on 
either side of the plane which contains the —CH,.CO— radical. These 
two chains are separated from the carbonyl-radical by an unbridged gap, 
since the route which leads to them through the bonds is long and tortuous. 
It therefore seems clear that we are dealing with an intramolecular field 
of force, acting across two empty spaces, which destroys the symmetry 
of the environment and thus brings out the latent possibility of dissym- 
metry in the highly-polarisable carbonyl group. 
The picture thus exhibited directs attention to the carbonyl group, 
rather than to the asymmetric carbon atoms, which in the acetate of 
u-arabinose make no direct contribution of any importance to the rotatory 
power of the molecule. In this respect it is indeed essentially identical 
with the conception of induced asymmetry (or better induced dissymmetry) 
put forward by Lowry and Walker in 1924 (64), according to which the 
carbonyl group itself becomes dissymmetric under the influence of the 
dissymmetric internal field of force of the molecule. It therefore con- 
tributes directly to the optical activity of the molecule, whereas less 
polarisable groups, such as +>CH, or >CMe,, contribute relatively 
little to the total rotation, even when they are exposed to a similar 
dissymmetric field. 
I have had the privilege of talking over this problem with Prof. 
Born. He insists that the rotatory power thus induced in the carbonyl 
group cannot be expressed in terms of single potential-gradients along 
and across the plane of the —CH,.CO— group, but must be a function of 
the frequencies of the electrons with which the carbonyl group is coupled, 
since this coupling affects the frequencies of both components. It is, 
however, possible that in a monoketone, such as camphor, the characteristic 
frequencies of the hydrocarbon radicals on either side of the median 
plane may be summed up in a weighed mean, depending but little on the 
structure or configuration of the carbon skeleton. In that case regularities 
and simplifications may perhaps be encountered, in studying different 
cyclic ketones, which could not have been foreseen from the complexities 
of pure theory. 
PREDICTION OF THE SIGN AND MAGNITUDE OF OpTICAL RoTATORY POWER. 
The electronic theories discussed above have not hitherto led to any 
prediction of the magnitude of the optical rotatory power of a dissym- 
