938 Dr. R. D. Kleeman on Disintegration of an Ion 



the clusters in these cases, we have seen, is principally due to 

 the action of the electric field. Since the mean free path at 

 constant pressure passes through a maximum as the electric 

 field is increased from zero upwards, these mean free paths 

 lie in the region corresponding to the fields which are greater 

 than that corresponding to the maximum path. 



The kink in curve C in fig. 5 for C0 2 at a pressure of 

 5 mm. of mercury is one of the first order. For since the 

 ■curve D lies above the curve C the disintegration of a cluster 

 is principally due to bombardment by neutral molecules. 

 The mean free path of a cluster under these conditions is thus 



X 



"5 cm. when — has the value 180. The period of life of a 

 p 



cluster in C0 2 at a pressure of 5 mm. of mercury under 



ordinary conditions is thus equal to the time it takes to pass 



over its mean free path. This time is equal to 



DX '0 



760 x -85x900 



= 4-3xl0- 6 sec, 



and the period of life of an ion cluster at atmospheric 

 pressures thus equal to 2*83 X 10" 8 sec. since it varies in- 

 versely as the pressure. But the period of life is less than 

 this value, since we have seen that the velocity of an ion 

 cluster is not greater than proportional to the electric field for 

 fields of the strength used. It seems thus to be considerably 

 less than the period of life of an ion cluster in air at a standard 

 pressure which was found to be of the order 10 ~ 7 sec. by a 

 different method in a previous part of the paper. 



The Effect of Light- Waves on the Stability of a Negative 

 Ion Cluster. 



An experiment was carried out to test whether the negative 

 ion clusters are dissociated by light-waves of much greater 

 wave-length than those waves which constitute ultra-violet 

 light. The light from a Nernst lamp was allowed to pass 

 through a mica window into the gas between gauze and 

 plate and could be cut off at will. But no appreciable 

 difference was observed in the collision current between the 

 light on and off when air was in the chamber. However, 

 other gases will be tried some of which may show the effect 

 looked for. The method is extremely sensitive, thus a change 

 of 1 per cent, of the clusters would about double the collision 

 current. 



