PROF. C. G. BARKLA ON X-RAYS AND THE THEORY OF RADIATION. 
357 
or rise in the absorption in A1 for a wave-length of 0'5xl0 ~ 8 cm. approximately, 
which again shows that A1 has a characteristic radiation of about the same penetrating 
power as that of the K radiation from tin. 
Thus all the methods, direct and indirect, show that under the usual conditions the 
light elements emit characteristic radiations of moderate penetrating power, and that 
the higher the atomic weight of an element, the more penetrating is its radiation. 
As these radiations are very much harder (of shorter wave-length) than those of the 
K series in the lighter elements, it is evident that they constitute a new series of 
characteristic radiations; and as there is no evidence of an intermediate series, this 
series should be called the J series. The mass absorption coefficients in A1 (x/p) A1 for 
the radiations from N to S vary from about 2‘5 to 175, or the wave-lengths from 
O' 56 to 0’5 x 10 " 8 cm. approximately. # 
PRIMARY X-RADIATION. 
The theory of emission of characteristic X-rays outlined above should also be 
extended to the process of production of X-rays by the incidence of cathode rays on 
any material, as in the ordinary X-ray tube. For it has been shown by Beatty! 
that the generation of a radiation of given penetrating power or frequency necessi¬ 
tates the incidence of electrons with energy greater than that of a quantum of that 
radiation. This applies not only to the production of radiation which is evidently 
characteristic of the particular anti-cathode employed, but to that heterogeneous 
radiation which becomes more and more penetrating as the speed of the exciting 
cathode rays increases. Such a law indicates a similarity in the process of production 
of all the X-rays generated in the anti-cathode. If mere collision between an electron 
of the cathode stream and an atom or a component of an atom in the anti-cathode 
resulted in the generation of X-rays, the resultant radiation consisting of all kinds 
of pulses produced by random collisions would possess the properties of a mixture of 
homogeneous radiations but without such a definite superior limit to the frequency of 
its components as has been actually observed. The definite law indicates a definite 
process, a process similar in essentials to that described in connection with the 
production of the characteristic X-radiations. 
There is no reason to deny that what has usually been regarded as the method of 
production of X-rays is inoperative ; the probability is rather that the electromagnetic 
radiation generated in an anti-cathode by the random collision of the exciting cathode 
particles is of small intensity; it does not constitute the bulk of the radiation 
recognised as X-radiation. (Yet a complete theory must take account of the 
experimental fact of a decided polarization in the primary beam—indicating an 
* [These values are probably rather high, as they were obtained by comparison of the wave-length of the 
a line with the absorbability of the whole K-radiation.— August, 1917.] 
t ‘Roy. Soc. Proc.,’ 1912. 
3 D 2 
