JuLy 1, 1915] 

(1) A small part of the energy of kathode rays 
falling on a radiator is converted into X-rays, the 
average frequency of the latter increasing with the 
velocity of the kathode particle. : ; 
(2) X-rays in passing through matter give rise to a 
B radiation. The initial energy of the escape of the 
electrons increases with the frequency, and is probably 
proportional to it. : 
(3) Electrons or X-rays of appropriate energy are 
equally able to excite the characteristic radiations in 
an atom. 
The results which have been shown to hold for the 
X-rays hold equally for the 8 and y rays, which have 
much greater individual energies, e.g. Gray and 
Richardson have shown that the 8 rays from radio- 
active matter are able to excite the characteristic radia- 
tions of the elements in a number of substances, while 
y rays in passing through matter give rise to high- 
speed electrons. It was long ago suggested by Bragg 
that 8 rays and X-rays are mutually convertible forms 
of energy, e.g. a 8 particle falling on matter may be 
converted into an X-ray of the same energy, and the 
latter in passing through matter may in turn be 
converted into an electron of identical energy. This 
assumes that the energy of an X-ray and an electron 
are mutually convertible, and the energy may appear 
under suitable conditions in either of the two forms. 
While the general evidence indicates that this point of 
view may hold closely for the conversion of the energy 
of a single X-ray into that of a swift electron, it is very 
doubtful whether it holds for the converse case of the 
excitation of an X-ray into an electron. We shall see 
later from experimental evidence that in general the 
energy of the electron required to excite ‘an X-ray of 
definite frequency is always greater than the corre- 
sponding energy carried off in the form of an X-ray. 
It was early observed that there appeared to be a 
close connection between the emission of 8 and y rays 
from radio-active matter. In all cases, the two types 
of radiation appeared together. A closer examination, 
however, showed that there were very marked differ- 
ences between the relative energies of the 8 and y rays 
from different radio-active elements. For example, 
radium C emits intense 8 rays and also intense 
y rays; on the other hand, radium E. emits 
intense $B rays over a wide range of velocity, 
but exceedingly weak y rays. Differences of 
a similar kind were observed amongst a number 
of the radio-active elements. One striking distinction, 
however, was to be noted. All the radio-active sub- 
stances which give a marked line spectrum of 6 rays 
also emitted intense y rays. On the other hand, a 
substance like radium E, which gave scarcely any 
y rays at all, gave a continuous spectrum of £ rays in 
which no lines have so far been observed. It thus 
appeared probable that the line spectrum of the 8 rays 
was intimately connected with the emission of y rays, 
and this conclusion has been completely established by 
recent experiments. As we have seen, y rays in pass- 
ing through matter give rise to high-speed f rays. 
Using radium B and radium C as a source of y rays, 
the 6 radiation excited in a number of metals by the 
passage of y rays was analysed in a magnetic field 
by Messrs. Robinson and Rawlinson and the writer, 
and was found to consist in part of definite groups of 
8 rays. When lead was the absorbing material, the 
magnetic spectrum of the £ rays excited by the y rays 
was found to be nearly identical with the primary 
B-ray spectrum of radium B. This striking result 
shows that those 8 rays escaping from the radio-active 
atom which give rise to a line spectrum must result 
from the conversion of y rays into 8 rays in the radio- 
active atom. The slight differences observed in the 
spectrum for different metals is probably connected 
NO. 2383, VOL. 95| 
NATURE 


447 

with the energy required to excite one of the charac-- 
teristic radiations of the element used as absorber. 
An explanation of the marked differences in the 
character of the 8 and y radiation from different radio-- 
active atoms can, I think, be given on the following 
lines. Some of the y rays are broken up in their 
escape from the atoms, and the energy of each con- 
verted y ray is transferred to an electron which escapes. 
with a definite velocity dependent on the frequency of 
the y radiation. Taking into account a large collec- 
tion of disintegrating atoms, each of the possible 
modes of characteristic vibration of the atom gives 
rise to an electron of definite speed. In this general 
way we may account for the line spectrum of the 
8 rays which is so commonly observed. On this view, 
we should expect to obtain a well-marked line spectrum 
of B rays when a substance emits strong y rays—a 
result in accord with observation. 
In order to account for the marked differences in 
the types and intensity of y rays from different radio- 
active substances, it seems necessary to suppose in 
addition that the primary 8 particle always escapes 
from the nucleus in a fixed direction with regard to. 
the structure of the atoms under consideration. For 
example, we have already pointed out that radium E, 
although it emits intense 8 rays which give a con-- 
tinuous spectrum over a wide range of velocity, emits 
very weak y rays. Since there can be no doubt that 
the 6 rays have sufficient speed to excite the charac- 
teristic modes of vibration which must be present in 
the atom, we are driven to the conclusion that the 
8 particle escapes in such a direction that it does not 
pass through these vibrating centres. On this view, 
the type of characteristic y rays which are excited, and 
consequently also the corresponding speed of the B rays. 
which arise from the converted y rays, will depend 
entirely on the direction of escape of the primary 
B particle. The definite direction of escape of the 
primary £ particle, which varies for atoms of different 
substances, also suffices to explain a number of other 
differences observed in the mode of release of energy 
from various radio-active atoms. It is supported by 
many other observations which indicate that the atoms 
of a particular radio-active substance break up in an 
identical fashion. 
We have so far considered only in a qualitative way 
the relation between the groups of rays in a f-ray 
spectrum and the emission of characteristic y rays. 
During the last few years there has been a growing 
body of evidence that the energy E carried off in an 
X-ray of frequency v is proportional to this frequency, 
and is given by E=hv where h is Planck’s funda-- 
mental constant. If the whole of the energy of an 
X-ray can be given directly to an electron, the energy 
communicated to the latter should be hv. There is no 
doubt that in many cases this simple relation holds 
very approximately, but the measurements so far avail- 
able are not sufficiently precise to settle definitely 
whether a part of the energy may not appear in 
another form. : 
Assuming that the transfer of the energy from an 
X-ray to an electron is complete, we should expect to 
find ‘groups of B rays of energy corresponding to hv 
where v is the frequency of the y rays found experi- 
mentally. Such a relation is found to hold within the 
limit of experimental error for three marked groups 
of low velocity B rays emitted from radium B. On 
the other hand, it is found that many of the high-- 
velocity groups of B rays from both radium B and 
radium C have energies many times greater than 
correspond to any observed frequency. Not the 
slightest evidence, however, has been obtained that 
corresponding high frequencies of vibration exist in’ 
the radio-active atom; in fact, all the evidence points. 
