272 Messrs. F. Soddy and T. D. Mackenzie on the 



spherical; one surface a b [as in trie annexed diagram] being 

 convex, and the other b c, concave : so as to obviate any re- 

 fraction of the incident and emergent conical pencils, with 

 (any) attendant dispersion into colour within the substance 

 of the prism. 



=>-f 



We may observe that these are not " lens-worked surfaces" 

 in the usual acceptation of the term " lens " — which deviates 

 or refracts rays. For these prism-surfaces are for the pur- 

 pose of completely eliminating any deviation of entering and 

 -emerging rays : — securing total reflexion at 90° (as desired) 

 without dispersion. 



Gross Flottbek, bei Altona, 

 March 17, 1907. 



XXII. The Relation between Uranium and Radium. By 

 Frederick Soddy, M.A., Lecturer in Physical Chemistry 

 and Radioactivity, and Thomas D. Mackenzie, B.Sc, 



Carnegie Research Scholar in the University of Glasgow* . 



IX a previous paper (Phil. Mag. June 1905, p. 769) one of 

 us described an experiment with a kilogram of uranyl 

 nitrate which had been purified initially from radium by 

 repeated precipitation of barium as sulphate in the solution, 

 and then left for a period of 550 days to ascertain whether 

 fresh radium was being generated in the solution. It was 

 found at the end of the period that the amount of radium in 

 the uranium solution had increased over a hundredfold. On 

 the other hand, if a direct change of uranium, or of uranium 

 X the first disintegration product, into radium took place, 

 the rate of production of radium in the uranium solution 

 should have been very much greater, it was stated about a 

 thousand times greater, than the rate actually observed. 



* Communicated by the Authors. 



