166 Dr. Tycho E:son Auren on 



are independent of the wave-length, we have reason to 

 believe that the fact that the radiation has not been quite 

 homogeneous must be of less importance. Corrections can 

 also easily be made for the absorption produced by 

 scattering in the lighter elements, as will be shown below. 

 In determining the absolute absorption coefficients by means 

 of the method generally adopted, the irregularities in the 

 action of the generator and the variations of the gas 

 pressure in the bulb create appreciable difficulties by 

 altering the composition of the radiation. Using homo- 

 geneous radiation obtained by reflexion on a crystal and 

 using the ordinary method, Bragg and Peirce * have made 

 a series of experiments, but the authors represented them 

 as being merely preliminary and announce that they intend 

 to renew the experiments by the aid of a compensation 

 method. These experiments recently have, however, been 

 repeated by Owen f . In the following pages I shall revert 

 to an experiment performed by Hull and Rice \ with the 

 aid of very hard homogeneous rays, which research, how- 

 ever, only includes the elements Al, Cu, Pb. I shall also 

 review a research recently published by Barkla and 

 White §, which has also been worked out by using a 

 homogeneous radiation and which likewise includes a few 

 elements only. 



My preliminary experiments, published in my paper 

 above mentioned, have been continued, and I am now going 

 to give the results obtained. In these experiments 1 

 have particularly endeavoured to make determinations with 

 more accuracy and to perform them so as to make them 

 comparable as far as possible. In order to increase the 

 sensibility of the method I have in the later experiments 

 used a more sensitive electrometer — a string-electrometer 

 from A. B. Vetenskapliga Instrument, Lund, Sw r eden, which 

 gave an indication of about 12 divisions on the scale for 

 1 volt, and the current in the bulb had amounted to 2 milli- 

 amperes and in this way the intensity of radiation appreciably 

 increased. By way of obtaining fully comparable results 

 I have tried, in all my experiments, to keep the composition 

 of the radiation as constant as possible. The generator used, 

 a high-tension transformer of 4 kilowatts, has been working 

 very evenly. The generator was charged so as to keep a 

 constant potential difference between the electrodes in 



* Bragg and Peirce, Phil. Mag. xxviii. p. 626 (1914). 

 t Owen, Proc. Roy. Soc. xciv. no. A 664, p. 510 (1918). 

 X Hull and Rice, Pliys. Rev. ix. p. 326 (1916). 

 § Barkla and White, Phil. Mag. xxxiv. p. 270 (1917). 



