146 Mr. J. A. M'Clelland. 



If the Lorenz expression for S, namely, ^ - -=-, be preferred to 



* -\-2 a 



ifc may be substituted in either of the above formulae. 



In either case the actual numbers will, of course, be changed more 

 or less, but the relation above pointed out will still hold good. 

 The discrepancies will, however, be somewhat exaggerated by the 

 change. 



This is suggested as a first approximation to a new law. It may be 

 useful in both chemical and physical science. It holds good, however, 

 only for the metallic elements. 



" Selective Absorption of Rontgen Rays." By J. A. M'CLELLAND, 

 M.A., Fellow of the Royal University of Ireland. Com- 

 municated by Professor J. J. THOMSON, F.R.S. Received 

 June 11, Read June 18, 1896. 



(From the Cavendish Laboratory.) 



The experiments described in this paper were made to determine 

 whether or not the Rontgen rays given off by a vacuum bulb were of 

 a homogeneous nature, by examining the manner in which they are 

 absorbed by different substances. The induction coil and vacuum 

 balb for producing the rays were enclosed in a wooden box thickly 

 lined with metal, with a small hole in the top, directly beneath which 

 and close up to it the vacuum bulb was placed. Over the hole a 

 well-insulated metal disk was placed and connected to one pair of 

 quadrants of an electrometer. The two pairs of quadrants are first 

 connected together and with one terminal of a battery of small 

 storage cells, the other terminal being connected to earth. 



The quadrants of the electrometer are then separated from each 

 other and from the storage cells, and the induction coil turned on. 

 The Rontgen rays passing through the hole in the box and falling 011 

 the charged disk discharges it, and the intensity of the radiation is 

 measured by the rate at which the spot of light from the electro- 

 meter needle moves across the scale. The metal lining of the box is 

 connected to earth, and the small hole covered with a single sheet of 

 tinfoil to screen the electrometer from direct electrical disturbances. 



The substance whose absorptive power is to be examined is placed 

 over the hole so fchat the rays traverse it before falling on the charged 

 disk. 



Evidences of selective absorption were sought for in the following 

 manner. The rate of leakage was accurately determined when the 

 rays were passing through one of the substances used, say a plate of 

 glass. Sheets of tinfoil were then substituted for the glass and the 



