75G Intelligence and Miscellaneous Articles. 



several mm. of mercury depends on the direction of the electric 

 field applied, the initial ions being produced by a beam of a particles 

 shot through the aluminium leaf and ending its course in the 

 aluminium plate. In a letter to the Philosophical Magazine of 

 July, 1912, I pointed out that this effect .would be produced if the 

 electrons liberated from the gas molecules are projected more or 

 less in the direction of propagation of the a particles, and an ion 

 undergoes a cycle of clustering in the gas during which for a time 

 it is an elementary ion. Dr. Campbell, in a footnote to a paper 

 published in the January number of the Phil. Mag., says that since 

 he finds no asymmetry in the number of electrons given off from 

 the opposite surfaces of the metal plate and leaf, the electrons are 

 not projected in the direction of motion of the ionizing « particle. 

 To obtain an insight into the reliability of such experiments, let us 

 consider the experiments that have been carried out with ultra- 

 violet light. The velocity of the ejected electrons is practically the 

 same in the two cases. It was found that the quantity and quality 

 of the electrons given off from a metal plate under the influence 

 of ultra-violet light depended greatly on the smoothness of the 

 surface of the plate, and even when polished depended greatly 

 on the nature of the process of polishing employed ; further, 

 the sensitiveness of the surface changed with time. The expe- 

 riments of various observers showed extraordinary discrepancies. 

 It was in time found that only tolerabl} r consistent results were 

 obtained when at a low pressure the surface of a plate was 

 removed by making it the anode in a discharge-tube, or depo- 

 siting a new surface, or removing the surface by evaporation 

 in the case of a liquid (such as mercury), and then taking the 

 measurements without allowing the surface to come into contact 

 with a gas at a high pressure. These processes remove the film of 

 gas and oxide from the surface, and the matter deposited through 

 polishing in the case of a plate. I therefore fail to see how much 

 reliance can be placed on experiments of the nature carried out 

 by Dr. Campbell unless the same precautions are taken. The 

 difference between the number of electrons emitted from the two 

 surfaces in the experiment is bound to be a small fraction of the 

 total number emitted, and could therefore be easily rendered 

 uncertain in sign by errors of the magnitude observed with ultra- 

 violet light. It is not surprising therefore that Dr. Campbell did 

 not obtain very consistent results. Moreover, in the interpre- 

 tation of such experiments we must also not ignore the possibility 

 that the process of ionization of a molecule in the solid state 

 by an a particle may be somewhat different from that when the 

 molecule is in the gaseous state. The fact that a molecule is 

 more easily ionized by ultra-violet light in the solid than in the 

 gaseous state renders such a thing highly probable. 



Tours faithfully, 



E. D. Klee^iax. 



