'622 Dr. G. C. Simpson on the Wilson- Gerdien 



the earth's surface ; but long before reaching it they will 

 become attached to cloud-particles or to the dust-particles of 

 the lower layers of the atmosphere, where the positive charge 

 will accumulate." 



At the time when this theory was published it was generally 

 lield that Elster and Geitel had satisfactorily explained the 

 normal electrical condition of the atmosphere by their theory 

 >of ion absorption; but in 1903 it was shown* that this 

 theory could not be substantiated by experiment. Thus 

 meteorologists were left without any satisfactory theory for 

 the normal electrical phenomena of the atmosphere. It is 

 easy to understand that under these circumstances an attempt 

 should be made to press Wilson's theory into explaining both 

 normal and abnormal electrical phenomena. 



This was done with considerable thoroughness by Dr. H. 

 Gerdien in his article on " Der Electrizitatshaushalt der 

 Erde und der unteren Schichten der Atmosphare/" in the 

 Pliysikolisclie Zeitsclirift \ . Gerdien introduced no essential 

 differences into the theory, but developed it into greater 

 detail than Wilson had done, and added numerical calcula- 

 tions to show that it was capable of accounting for all the 

 electrical phenomena associated with thunderstorms. 



Turning now to our criticism, the theory will be examined 

 first from the electrical side of the problem and then from 

 the meteorological. In the first part of the discussion we 

 shall accept the meteorological conditions supposed by the 

 theory ; i. e. we will assume the possibility of the formation 

 of a supersaturated stratum above a lower cumulus cloud, 

 and of a condensation layer above in which water vapour is 

 condensed around the negative ions ; and then examine 

 whether the electrical effects observed can be supposed to 

 follow from these conditions. Afterwards we will examine 

 the meteorological conditions themselves to find if they are 

 likely to be present during thunderstorms. (1) The first 

 question we will ask is : — 



Could a sufficiently rapid separation of electricity take 

 place, under the conditions supposed by the theory, to 

 account for the numerous lightning discharges observed 

 during thunderstorms ? 



In order to obtain simplicity of treatment we will follow 

 Gerdien in assuming that the processes we are investigating 

 ftake place in horizontal layers sufficiently near to one another 

 in comparison with their lateral extent to be considered as 

 infinite planes. Under such conditions we may assume that 



* Simpson, Phil. Mag. |_6] vi. p. 589 (1903). 

 t H. Gerdien, Phys. Zeit. vi. pp. 647-66 (1905). 



