PROF. C. G. BARKLA ON X-RAYS AND THE THEORY OF RADIATION. 
347 
expect that such an L radiation would at most account for only 7 per cent, of the 
K absorption, for n L is only about }n K . If no such additional L radiation is 
discovered, the natural conclusion is that the displaced K electrons are replaced by 
electrons of still lower frequency or possibly, though highly improbably, from outside 
the atom. 
It is possible that there are not electrons of every series in every atom of the same 
substance, or, if there are, they may not be in the best position to enable them to 
replace an electron of an adjacent higher series. If, for instance, a K electron were 
ejected from an atom which did not'contain an L electron, or in which the L electron 
was situated on the other side of the centre, its place might be taken by an M 
electron, with the result that the energy emitted as characteristic radiation would be 
equal to that of one-quantum of K radiation + one quantum of L radiation. This 
would be equal to one quantum of radiation of slightly greater frequency than what 
we have previously regarded as the K radiation, and might account for a neighbouring 
spectral line in the K radiation. 
The origin of various spectral lines in a particular series (say the K series) was 
first explained by Kossel as the result of electrons falling from the various outer 
rings into the K ring, and was not given in the early communications of the writer. # 
According to Kossel, the K a radiation is that emitted when an electron falls from 
the L ring into the K ring ; that resulting from the fall from the M ring into the 
K ring, and so on. In support of this he shows that the necessary energy relations 
hold. Thus the energy emitted when an electron falls from the M ring into the K 
ring is the sum of the energies of fall from the M ring into the L ring, and from the 
L ring into the K ring; what Kossel shows is that hn Kp = hn La + hn Ka , that is, one 
quantum of Kp radiation possesses the sum of the energies of one quantum of L a 
radiation and one quantum of K a radiation. This indicates that, while the fall of 
electrons takes place between consecutive rings, it sometimes takes place across an 
intermediate ring position. 
These observed relations thus confirm the conclusion to which the writer was led 
in a most direct manner from measurements of the energies of corpuscular and 
fluorescent (characteristic) radiations and their relation with the associated absorption 
of primary radiation. 
The experiments described above lead to the conclusion that a characteristic 
radiation (of series K, say) is probably emitted when an outer electron falls into the 
position of a displaced K electron, the energy previously absorbed in displacing the 
K electron in excess of that carried away as 'kinetic energy being now re-emitted as 
a quantum of K radiation, or, what seems more probable, as a quantum of K radiation 
accompanied by quanta of lower frequency radiations. 
* Sir Ernest Rutherford kindly sent a copy of Kossel’s paper to the writer a few weeks after the 
publication of the preliminary notes in ‘ Nature,’ drawing attention to some similarity of conclusion. This 
paper was evidently published some time previous to the writer’s notes. 
VOL. CCXVJI.—A, 3 0 
