396 Prof. Thomson, On the presence of 



ionization, but if instead of the lead screen the tank was filled with 

 water, I could detect no diminution in the ionization ; the sand and 

 sawdust likewise produced no effect, while the bricks produced a 

 slight increase in the ionization. To compare the effects of these 

 screens on radium radiation, I brought near the tank 15 mg. of 

 radium bromide enclosed in a layer of lead about 4 cm. thick 

 and placed it at such a distance from the testing vessel that the 

 ionization was about doubled ; in this case the introduction of 

 water in the tank produced a most marked diminution, amounting 

 to about 40 per cent, of the ionization due to the radium. 



Patterson found that the ionization inside a closed vessel was 

 largely due to easily absorbed radiation proceeding from the walls 

 of the vessel. This result has recently been confirmed by an 

 entirely different method by the experiments made by Campbell 

 at the Cavendish Laboratory. Wood working at the Cavendish 

 Laboratory has shown that the diminution in the ionization 

 produced in a vessel of given volume by a given screen depends 

 on the material of which the walls of the vessel are made, the 

 absolute diminution in a tin vessel being twice as great as for a 

 zinc one of the same size. These results are both explained if we 

 suppose that the radiation which ionizes the gas in the vessel is 

 partly due to secondary radiation proceeding from the walls of 

 the vessel, this secondary radiation being excited by radiation 

 which comes from outside and penetrates the walls of the vessel. 



Source of the Radiation causing the ionization of gas in a closed 

 vessel. We have seen that the laws governing the ionization inside 

 a closed vessel are such as to make it improbable that the radiation 

 causing this ionization comes from a radio-active impurity present 

 in the neighbourhood of the vessel. If, however, the radiation is 

 not due to an impurity it must come from the non-radio-active 

 substances around the vessel, for the radiation can not be derived 

 entirely from ultra-mundane sources, for as we have seen, radiation 

 of this kind comes upwards from the earth — it cannot however 

 have penetrated the earth, for it is stopped by a few inches of lead. 



The study of the ionization of gases in closed vessels seems to 

 me to point most strongly to the conclusion that all substances emit 

 a kind of radiation which, like Rontgen rays, is able to ionize a gas 

 through which it passes ; they also absorb such radiation so that 

 the stream of radiation inside an unlimited extent of radiating 

 substance is not infinite, but is such that the amount of energy 

 crossing in one direction unit area in the substance in unit time 

 is q/Q\ where q is the energy radiated from unit volume of the 

 substance in unit time, and \ is the coefficient of absorption, 

 i.e. when energy of radiation I travels through a distance dx, an 

 amount of energy I \dx is absorbed. We see from this that the 



