Delta Rays from Gases. 83 



granite is so much greater than that of basalt. It is just 

 possible that frequent heating and cooling of some specimens 

 of basalt might cause the development of the more minute 

 system of cracks to such an extent as to cause a considerable 

 permanent lowering of the conductivity accompanied by a 

 change in the effect produced by heating, so that the con- 

 ductivity would fall like that of granite. This is just what 

 was noticed in the first basalt determinations, but these, as 

 has been shown, were not of a sufficiently reliable nature for 

 much stress to be laid on their results. 



It would seem as a result of all the determinations that 

 for temperatures up to 500° or 600° C. the conductivity of 

 the earth's crust may be taken as about 4x 10" 3 without risk 

 of serious error, unless the conductivity is sensibly alFected 

 by the large pressures involved. 



In conclusion I wish to express my sincere gratitude to 

 Dr. Joly for his kind interest and many valuable suggestions 

 during the progress of the work. 



Physical Laboratory, 



Trinity College, Dublin, 

 October 6th, 1913. 



VI. Delta Rays from Gases. 

 By Norman Campbell, Sc.D* 



1, TT is generally believed that when the molecules of a 

 JL gas are ionized the electrons are expelled from them 

 with velocities similar to those of the 8 rays from metals ; 

 the belief is based upon the exact parallelism of the ionizing 

 powers of rays and their power of exciting 8 rays from 

 metals. There appears to be no direct experimental support 

 for the belief, no observations on the ionization of gases 

 which make it necessary to believe that the electrons ejected 

 in ionization have an average initial velocity of several volts f . 

 The absence of such observations appears to be due to the 

 fact that measurements of the ionization of gases have been 

 made at too great pressures or in too great electric lields. 



* Communicated by the Author. 



t Kleeman (Proc. Roy. Soc. lxxxiii. p. 195 (1910)) produced evidence 

 that the electrons liberated in ionization possess a finite average 

 momentum in the direction of that of the ionizing rays ; but, since it 

 appears that the d rays from metals do not possess such a momentum, 

 his experiments, whatever their interpretation, do not support the belief 

 mentioned. 



G-2 



