300 TRANSACTIONS OF SECTION A. 
those points on which my opinion has been invited by Professor Hicks. 
Firstly, with regard to Béhr’s theory, such criticisms as I have made are in the 
main mathematical, and therefore unsuitable for a joint discussion between 
physicists and chemists. But I can give a statement of the present position 
of the theory as it appears to me. When Bohr’s theory is applied to a single 
nucleus of strength e or 2e, with a single rotating electron, it is remarkably 
successful in its deduction of the hydrogen series spectrum and of the Pickering 
series which it ascribes to helium. Its most striking success is, I think, not 
the very accurate deduction of the universal constant of spectra, but its appli- 
cation by Professor Fowler in his Bakerian lecture to a determination of the 
mass of an electron, on the supposition that the Pickering series comes from 
helium. The accuracy of this value cannot be ignored. But analysis shows 
that it is quite impossible to go further, and to derive the usual helium 
spectrum. I mean that in order to do so we must abandon at least one of 
Bohr’s premises which is vital to the deduction of the hydrogen formula. 
This fact is capable of rigorous demonstration, as is also the fact that, under 
the inverse square law, which Sir Ernest Rutherford has shown experimentally 
to be valid, Rydberg’s constant is not a feature of more complex atoms on this 
theory. 
There is also an experimental difficulty. Whatever its origin, the Pickering 
series should be accompanied by an ultra-violet one in the Schumann region. 
This series has been found by Professor Lyman in the hydrogen spectrum, 
whereas helium appears to have no Schumann spectrum. Professor Lyman is 
repeating these experiments, in view of their importance, but the balance of 
experimental evidence is against Bohr’s theory at present. 
I am inclined to agree with Mr. Moseley that my nebular and coronal 
elements may not be chemical elements in the ordinary sense. This opinion, 
that they are sub-elements, or bases of ordinary elements, will be found in my 
papers. Bourget, Buisson, and Fabry’s experiments, described in the 
“Comptes Rendus,’ show that these substances have the atomic weights which 
I calculated theoretically from their spectra, so that their existence appears 
to be real. Moreover, as in my papers, ordinary elements with series spectra 
can apparently only be formed from them by an alteration in the nucleus which 
does not affect its total charge. Evidence is accumulating to show that the 
nuclear structure may play an important part in series spectra, and therefore 
I am not inclined to agree with Professor Rutherford that the nucleus of a 
hydrogen atom is necessarily the positive electron. It seems to be more 
complicated. But with everything else in his admirable opening address I 
must express a general agreement. I must finally agree with Mr. Moseley that 
any ultimate atomic theory must involve Planck’s 4. In my own papers this 
was regarded as an angular momentum, as subsequently also by Bohr. The 
necessity for it is easily seen. For we only have one dynamical relation between 
the radius of the atom and the angular velocity of its electrons. Without 
the introduction of some new universal constant such as / no atom has any- 
thing in its nature which compels a definite size, and definite unchanging 
properties. 
Professor H. Basserr said that, as the number of elements which came 
before neon seemed of considerable importance in connection with the 
theoretical treatment of the constitution of the atom, it might be worth while 
considering whether the periodic law gave any hints on the matter. It was 
well known that Lothar Meyer’s atomic volume curve clearly demonstrated 
that, although the properties of the elements were periodic functions of their 
atomic weights, the periodicity was not of such a simple character as at first 
supposed by Newlands. Leaving out hydrogen for the moment, it was found 
that there were two short periods of eight elements each, beginning with neon 
and argon, and ending with fluorine and chlorine respectively, followed by two 
long periods of 18 elements—that was to say, of (2X8)+2 elements. These two 
long periods were followed by one very much longer period and a portion of a 
second. Unfortunately this very long period was so far incompletely known, 
and it was not certain how many elements it contained; but this much could 
be said, namely, that it contained approximately twice as many elements as 
one of the long periods, and possibly 38 elements, which would be (2x18)+2 
