102 THE ROYAL SOCIETY OF CANADA 



nuisk that coming from the earth. Flemming* found from balloon 

 observations that at an altitude of 3000 metres about the same 

 amount of active deposit could be collected as at the surface, while 

 Saakef and GockelJ found the amount collected at high altitudes greater 

 than at sea level. 



The explanation just put forward agrees with the trend of Hess's 

 observations: a diminution in the ionization due to penetrating 

 radiation was first observed followed by an increase proljably due 

 to the gradual accimiulation of active deposit. 



Sect. 4. — Note on the Origin of the Penetrating Radiation and Con- 

 ditions to he Fulfilled in the Correct Measurement of the 

 Gradient. 



Chauveau* concludes from a review of existing data that three 

 sources may contribute to the penetrating radiation: — - 

 (i) The effect of radioactive matter suspended in the atmosphere 

 contributing •! to -2 ions per cc. per sec. 



(ii) The contribution of a layer of active deposit carried down by the 

 electrostatic potential gradient of the atmosphere: contribution 1 to 

 2 ions and subject to large fluctuations depending on meteorological 

 conditions. 



(iii) The largest contribution of 4-5 ions due to the radium content 

 of the soil: we have already seen that the whole efïect is almost en- 

 tirely due to the surface layer of a few centimetres depth. 



In the light of the information brought forward in the present 

 paper the following conditions should obtain a measurement of the 

 gradient. 



(i) Observations should l)e taken over a level plain, consisting of 

 a homogeneous soil in the surface layer: the radium content of sample 

 taken at a few centimetres' depth but over as wide an area as possible 

 should be determined by a separate experiment, as well as their co- 

 efficients of absorption to j'-rays. 



(ii) The ionization in a cavity a few centimetres deep should be 

 measured: from a practical point of view this is most easily ac- 

 complished by placing a shallow wooden box (about 25 centimetres 

 deep) and as wide and Inroad as possible on two trestles and measuring 

 the ionization beneath the box as closely as possible to the lower sur- 

 face. By filling the box to various depths with soil typical of that 

 found in the locality measurenionts equivalent to the measurements 



*FlemminK, Phiis. Zeit. 9, p. 801, 1908. 

 tSaake, Phxjs. Zeit., 4, p. 62G, 190:^. 

 ÎGochcl, Phi/s. Zeit., 8, p. 701, 1907. 

 * Chauveau, loc. cit., p. 72. 



