president's ADDRKSS. 19 



minute amounts with precision and certaint}', amounts not only 

 infinitely heyond the range of chemical detection, but also quite 

 outside the powers of the spectroscope unless the material is first 

 specially concentrated. In a paper on " The Evolution of the 

 Elements, ' read before the British Association at its last meeting, 

 Mr. F. 8oddy remarked : — '''J'he smallest quantity of any element 

 that can be detected Ijy the spectroscope contains 10'" indi\idual 

 atoms, whereas the disintegi'ation of a single atom accompanied 

 with the expulsion of one a particle is not greatly, if at all, below 

 the limit of detection by present methods."* 



The radio-activity method is thus in this case something of the 

 order of 10,000,000,000 times more sensitive than the spectro- 

 scopic. The former method depends on the intense ionising power 

 of tlie emanation, whereby air submitted to its action is brought 

 into a state of partial disintegration known as ionisation, in which 

 condition it becomes an active conductor cf electricity. The em- 

 anation from a known quantit}^ of radium is collected in air during 

 a fixed period, and the conducting power of the air is then deter- 

 mined by suitable means. The same method applied to any 

 specimen under examination gives the relative value from which 

 its radium content can be readil}' calculated. 



Mr. Strutt considers that 5 x 10"^- gram radium per c.c. rock 

 may be taken as a reasonable average for the rocks constituting 

 the earth's crust. Taking the mean density of the rock at 2-7, 

 this would be equal to 1*85 x 10'^- gram radium per gram rock. 



Assuming the internal heat of the earth to be entirely derived 

 from the disintegration of radium uniformly distributed through- 

 out its mass, and taking Lord Kelvin's data for the thermal 

 conductivity of the rocks in sit a, Mr. Strutt calculates that the 

 amount of radium necessary to account for the observed lieat 

 gradient near the surface is about 0175 x 10"^^ gram per c.c, an 

 amount greatly less than the smallest proportion found in any 

 igneous rock examined. From this and other data he concludes 

 that radium does not exist in the earth's centre, but is confined 



* Chemical News xciv., 86, 24 Aug. 1906. 



