148 Mr. F. Osmond and Prof. W. C. Roberts- Austen. 



Of the substances used, the above showed the effect best, but with 

 wood, paraffin, and water, although small, it could always be 

 detected. We conclude from the above results that the Rontgen 

 rays are of different kinds, and that the substances given in the 

 table differ very much from tinfoil in their selective absorption. 

 After the rays have been screened by passing through some tinfoil 

 layers additional layers are much less absorbent, while the absorption 

 produced by other substances is not so much diminished. 



Of the substances tried, those which are fluorescent gave the most 

 marked difference as compared with tinfoil. 



The above results were all obtained with one vacuum tube, which 

 was working extremely well. It produced a very rapid leak from the 

 charged disk, and the pressure of its residual air was very small. 

 In fact, after working for a time it became too strong for the coil 

 that was being used to work it. Another vacuum tube, in which the 

 pressure of the residual air was greater and which was not so 

 efficient in producing leakages, was then used, and several of the sub- 

 stances used before were again tested, but in no case was any evidence 

 of selective absorption obtained. As far as the test was efficient, the 

 radiation from this bulb was homogeneous. 



A third tube was then used, more efficient than the last in produc- 

 ing leakage, but not so good as the first used. With this tube experi- 

 ments made in the same way as before gave evidence of selective 

 absorption, but not so marked as with the first tube. 



It seems therefore that as a tube becomes more efficient the 

 character of the rays given off becomes less homogeneous. 



" On the Structure of Metals, its Origin and Changes." By 

 F. OSMOND and W. C. ROBERTS-AUSTEN, F.R.S., Professor of 

 Metallurgy, Royal College of Science. Received June 10, 

 -Read June 18, 1896. 



(Abstract.) 



The authors begin their paper by stating that it has been shown 

 by Herbert Tomlinson that the atomic volume of metals is intimately 

 connected wi^h their thermal capacity* and with Young's modulus. f 

 He considers, in view of the work of Wertheim,J of Maxwell, and 

 of Heen,(| and as the result of his own experiments, that the value of 



* ' E,oy. Soc. Proc.,' vol. 38 (1884-85), p. 488. 



f ' Phil. Trans.,' Part I, 1883, p. 32. 



J ' Ann. de Chim. et de Phys.,' vol. 12, 1844. 



' Phil. Trans.,' vol. 156, 1866, p. 249. 



|| 'Bull, de 1'Acad. Eoy. de Belgique,' vol. 4 (1882). 



