WheelocTc — Nature of Ionization Produced hy a Rays. 251 



Again, the effect of initial recombination, if it exists, should 

 be the same whether tlie field is ai)plied along the path of the 

 a particle or at right angles to it. Experiment shows that the 

 saturation current is obtained with a potential gradient of about 

 200 volts/cm. between the plates of the ionization chamber, 

 while from 1200 to 1500 volts/cm. are necessary for saturation 

 with the parallel field. 



The results of the experiments with the ionization ])roduced 

 at different parts of the range of the a particle seem to show 

 that neither initial recombination, nor the hypothesis of ioni- 

 zation in columns along the path of the a particle, completely 

 accounts for the lack of saturation obtained by experiment. 

 Initial recombination fails when the heki is at right angles to 

 the path, while columnar ionization does not entirely account 

 for the recombination obtained with the parallel field. 



Here again it is seen that both hypotheses may be needed to 

 completely explain the lack of saturation experimentally found. 



VI. Comparison of Curves obtained at different Pressures. 



Bragg and Kleeman* found that for a gas ionized by a rays, 

 saturation was obtained at a lower potential gradient when 

 the pressure of the gas was decreased. This is as would be 

 expected if initial i-ecombination takes place. At higher pres- 

 sures the ejected electron might not get out of the sphere of 

 influence of the parent molecule before it is stopped by encoun- 

 tering another molecule ; but as the pressure diminishes the 

 mean free path of the electron is increased, and the ejected 

 electron is then able to get farther away from the positive ion 

 and hence initial i-ecombination would occur less fre(5^uently. 



According to the hypothesis of columnar ionization, as the 

 pressure is diminished the density of ionization in the columns 

 would be diminished and the amount of recombination would 

 be less. Thus it is seen that either hypothesis might explain 

 the results obtained by experiment. 



Data obtained at different pressures : 



Parallel Field. 



Source 4*3<^™ from the middle of the chamber. 



Pressure 





Volts /cm. 





Current 



% lack sat. 



mm. mercury 





5 





10-4 



19-1 







10 





10-9 



15.2 







20 





11-45 



10-9 



544-7 





40 

 100 





11-83 

 12-5 



7-9 



2-7 







200 





12-75 



0-8 







400 





12-77 



0-6 







800 





12-85 



0-0 



*Bra 



'gg 



and Kleeman, 



Phil. 



Mag., xi, 1906. 





