RECENT ADVANCES IN SCIENCE 377 



sight this might appear to be purely fanciful. Perrin's treat- 

 ment of it is such, however, as to bring it within the range 

 of serious scientific thought. 



The fundamental change introduced by this idea is that 

 radio-active processes have to be regarded no longer as typically 

 exothermic, but, on the contrary, as strongly endothermic, 

 in the sense that a much greater amount of energy is first of 

 all absorbed by the radio-active parent substance than is 

 emitted in the observed radio-active change. Perrin has even 

 succeeded in calculating roughly the wave-length of the 

 " ultra-X " rays, which would account for the radio-active 

 change, by taking into account the kinetic energy of an ex- 

 pelled a particle. The frequency of the " ultra-X " radiation 

 is of the order io 21 vibrations per second (whilst that of visible 

 light is of order 5 x io 14 ). 



The attempt is made frankly by Perrin to bring radio- 

 active changes into line with all other atomic and molecular 

 processes, thermal reactions, photo-chemical reactions, fluor- 

 escence, phosphorescence, and ionisation, all of which involve 

 radiation of certain wave-lengths. Ionisation of a gas by 

 X-rays represents, in fact, according to Perrin a group of reac- 

 tions intermediate between ordinary chemical reactions and 

 radio-active transformations. Following out his idea Perrin 

 proceeds to consider several astrophysical or rather astro- 

 chemical problems, the existence of proto-atoms, and the 

 stability of atoms of various types as a function of their en- 

 vironment. 



The most striking characteristic of radio-active change, 

 viewed in relation to other types of chemical change, is the 

 apparent absence of any influence exerted by altering the 

 temperature of the radio-active material. In the case of 

 ordinary thermal reactions temperature is a most important 

 factor, the rate of change increasing threefold for a rise in 

 temperature of ten degrees in the neighbourhood of room 

 temperature. 



As Perrin points out, however, the presence or absence of 

 a temperature effect upon a process depends upon the frequency 

 or wave-length of the radiation producing the change. If the 

 effective radiation frequency is small, say of the order io 14 

 vibrations per second, there is known to be a sensible amount 

 of this type of radiation present per unit volume of any material 

 at ordinary temperatures, and this so-called density of radia- 

 tion increases rapidly as the temperature of the material rises. 

 The result is that the rate of the chemical process (depending 

 upon this low-frequency radiation) increases rapidly with the 

 temperature. If, on the other hand, the process requires 

 radiation of frequency io 15 vibrations per second, such a 



25 



