372 Prof. Owens on Thorium Radiation. 



the more penetrating kind, though possibly remaining the 

 same percentage of the whole, will, in actual measure, be 

 too small to affect the electrometer. Consequently, the curve 

 for a thin layer should much more closely follow the usual 

 absorption law than for a thick one, which we see by an 

 inspection of curve, fig. 4, is actually the case. The 

 agreement of the first part of the curve is perfect. 



Consequently, the more absorbable kind may be assumed 

 to be homogeneous at least as far as absorption phenomena 

 indicate. 



The same general method was used to investigate the nature 

 of the more penetrating kind, but a comparatively thick 

 aluminium sheet about *13 millim. was used instead of 

 '0008 centim. foil. The results are given in Table IV., and 

 shown graphically on fig. 5. 



The curve, however, shows no approximation to represent- 

 ing a simple absorption law, indicating that the more pene- 

 trating radiation is complex, consisting probably of a number 

 of component parts. In this respect it is unlike the /3 radia- 

 tion described by Rutherford for uranium, which he considered 

 to be approximately homogeneous in character. 



In Table V. are given the results for thorium sulphate, 

 using thin aluminium foil to cut down the radiations. The 

 results are plotted on fig. 6. 



Table VI. and fig. 7 give the results for the nitrate. 



It will be noticed that the more penetrating radiation for 

 both is a smaller per cent, of the total than with the oxide, 

 and along the first part of both curves the current decreases 

 in geometrical progression very closely as the thickness of 

 foil increases in arithmetical progression, which is what 

 would be expected if the more penetrating kind is small in 

 amount and the major part homogeneous. 



When ordinary foolscap -paper was used instead of the 

 aluminium foil, very curious results were obtained. With a 

 thick la} r er of the oxide the first layer of paper cut down the 

 conduction-current about 50 per cent.; the next fifteen addi- 

 tional layers produced practically no further diminution in its 

 value. Fifteen more layers cut the current down to 33 per cent. 



With paper a very considerable time was required for the 

 current to come to a steady value as successive layers were 

 added. 



When a thin layer of oxide is used, the action is quite dif- 

 ferent, the absorption curve being quite regular and approxi- 

 mating to those obtained when aluminium foil was used. 



The explanation may possibly be that the penetrating 

 radiations from a thick layer of the oxide in passing through 



