178 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 1915. 



sure of the gas inside was adjusted to a few centimeters of mercury. 

 The tube B was connected with the negative pole of a battery of 

 about 1,500 volts, the other pole being earthed. The potential was 

 adjusted so that a spark was on the point of passing between A and 

 B. Under such conditions the ionization due to an alpha particle 

 passing along the detecting vessel is magnified several thousand times 

 by collision of the negative and positive ions with the neutral mole- 

 cules. 



The entrance of an alpha particle into the detecting vessel is then 

 signified by a sudden haUistic throw of the electrometer needle, 

 and the number of particles entering the vessel in a given time can 



Fig. 2.— Apparatus for counting alpha particles. 



be counted by observing the throws. The amount of active matter 

 and its distance from the opening were adjusted so that 3 to 5 alpha 

 particles entered the opening per minute. The following table illus- 

 trates the results obtained: 



Number 

 of throws. 



Magnitude of successive throws, 

 scale divisions. 



First minute 



Second minute.. 

 Thiid minute . . . 

 Fourth minute. . 



Fifth minute 



Sixth minute 



Seventh minute. 

 Eighth minute. . 

 Ninth minute. . . 

 Tenth minute... 



Average per minute 



Average throw (divisions). 



11,12,10,11. 

 10, 11, 8. 

 10, 9, 13, 8, 12. 

 18*, 8, 12. 

 10, 6, 10. 



9, 10, 12, 11. 



10, 11. 

 11,13,8. 

 8, 20*. 

 8,12,14,6. 



It will be seen that the number of throws varies from minute to 

 minute. This is to be expected, since the chance of an alpha particle 

 entering the opening is governed by the ordinary laws of proba- 

 bility. It will be seen that two throws, marked by asterisks, are 

 much larger than the others. These were due to the passage of two 



