of the ol Particles from Radium C and Thorium C. 515 



curves agree within the experimental error, it is clear that 

 the ratio of numbers of ions found by Geiger is approxi- 

 mately the same as the corresponding ratio obtained from 

 a curve taken wholly in air. This doubt is thus unfounded, 

 and Geiger's result (2'37 x 10 5 ) for the total number of 

 ions produced by one RaC ol particle does not seem to be 

 in error, as far as it is affected by the doubt just considered. 



The ratio between the maxima of ThO, and ThC 2 is *564. 

 It may be assumed that the ionization curve for ThCi, of 

 range R, is similar to the last R cms. of the curve for ThC 2 . 

 Hence the ratio between the maxima is the ratio between the 

 numbers of the two types of a particles given out in the 

 complex transformation of thorium 0. These results give 

 36 per cent, of ThCj and 64 per cent, of ThC 2 « particles, 

 in good agreement with the values 35 per cent, and 65 per 

 cent, found by Marsden and Barratt * by direct counting. 



This apparatus was not designed to detect the ionization 

 due to very small numbers of a particles. It was thus not 

 sensitive enough to detect the rare a particles of very long 

 range from thorium discovered by Rutherford and Wood f , 

 which are present in the proportion of 1 to 10,000. In the 

 present investigation the ionization current (corrected for 

 natural leak) fell to zero, within the experimental error, 

 outside the ranges of the different a particle?. 



§ 5. Possible Anomalies in the Ionization Curve near the 

 End of the Range. 



It is now desired to draw attention to certain anomalies in 

 the ionization and scintillation curves near the end of the 

 range, which have recently been discussed theoretically and 

 experimentally by several writers. 



In 1912, Herzfeld % pointed out that probability variations 

 in the number of encounters of the a particles with gas 

 molecules should cause variations in the ranges of the indi- 

 vidual a particles. Herzfeld assumed that a definite number 

 of collisions was necessary to stop an a particle, and took this 

 number to be equal to the total number of ions produced by 

 the particle. He gave a theoretical scintillation curve for a 

 parallel beam of a particles, which showed that their number 

 should remain constant almost up to the end of the range and 

 then fall off gradually to zero. In the case of polonium 



* Marsden and Barratt, Proc. Phys. Soc. xxiv. p. 50 (1911). 

 t Rutherford and Wood, Phil. Mao-, xxxi. p. 379 (1910) ; xli. p. 570 

 (1921). 



| Herzfeld, Phys. Zeits. xiii. p. 547 (1912). 



