118 THE ROYAL SOCIETY OF CANADA 



zinc. It, too, was carefully dried but it is probable that it 

 contained some air, The methyl iodide was introduced by 

 evaporation into the electrometer by placing a small bottle of it 

 with cork removed in the small chamber attached to the instrument 

 which usually contained the metallic sodium. The instrument 

 was filled with air when this was done and the pressure of the methyl 

 iodide vapour was probably about one half a centimetre of mercury. 

 These readings served to bring out the fact that the ionisation 

 in air on the ocean was about one half what it was on land, that with 

 methyl vapour in air the ionisation in the latter was nearly doubled, 

 that the ionisation in hydrogen was very low and was about the same 

 on land as on water, and that the so-called natural ionisation of acety- 

 lene was extremely high. With the tube SiSi down the loss of charge 

 from the charged electrical system was about the same with acetylene 

 as with air which shewed that the high value obtained for "q" with 

 acetylene was not due to any loss of charge over the supports arising 

 from any action of the gas on the pieces of amber which carried the 

 fibres. 



IV SECOND INVESTIGATION. 



In the second investigation the readings were taken with a number 

 of different gases, first, in a room free from radio-active contamination 

 in the Physical Laboratory at Toronto and then out on the ice on 

 Lake Ontario at a point about half a mile from the shore where the 

 water was about 20 feet deep. The temperature in the laboratory 

 when the readings were made was about 21°C. and the temperature 

 on the Lake from 0°C. to — 10°C. The manner in which the electro- 

 meter behaved is illustrated by the diagram shewn in Fig. 3 which 

 represents the readings in volts taken with air on a particular day 

 from about 9 o'clock a.m. until after 6 o'clock in the evening. 



After this portion of the investigation had been completed a 

 set of readings was taken in the Laboratory with each of the gases 

 under natural ionisation and then with them traversed consecutively 

 by streams of alpha, beta and gamma rays of moderate intensity. 

 Measurements with the beta rays were made by placing in the chamber 

 which usually contained the metallic sodium a small capsule containing 

 some uranium oxide. The back and walls of this capsule were made 

 of thick zinc and the face consisted of a sheet of aluminium thick 

 enough to cut ofif all alpha radiation. 



The arrangement is shewn in Fig. 4 and from the diagram it will 

 be seen that the effect of the rays on the gas in the electrometer was 

 the same as it would have been if the radiation had been emitted from 

 a portion of the walls of the instrument. In taking the readings 



