50 THE ROYAL SOCIETY OF CANADA 



that of the positive ion, but his results indicate that while the mobility 

 of the positive ion varied inversely with the pressure up to 100 atmos- 

 pheres, the mobility of the negative ion decreased less rapidh^ as the 

 pressure was increased than it should if it followed the inverse pressure 

 law. 



With moist carbon dioxide on the other hand, Kovarik found that 

 the mobilities of positive and negative ions were apjjroximately equal 

 to each other for pressure up to 57 atmospheres and that the mobilities 

 of both ions decreased with increase of pressure much more rapidly 

 than would be warranted by supposing the inverse pressure law to hold 

 As the air in the present experiments was obtained from a liquid air 

 compressor and no special pains were taken to dry it, other than passing 

 it through a chamber filled with potassium hydrate, it is possible that 

 the equality in conductivitj^ indicated by the results in Table IV 

 might, in small part at least, be traceable to the presence of moisture 

 in the air. 



It would appear, however, from some work which was done later, 

 and which is described in section VI, that the equality in the conduc- 

 tivity obtained with positive and negative fields in the experiments 

 just described was due in great measure to the particular form given 

 to the apparatus used in making the measurements. For it was found 

 when the apparatus was considerably modified, that results were ob- 

 tained which clearly pointed to the mobility of the negative ions pro- 

 duced in air at high pressurer by alpha rays being considerably greater 

 than that of the positive. In taking the readings given in Table IV 

 the capacity of the system was 3635 e.s. units and as the capacity of 

 the system in the liquid air measurements was only 200 it would follow 

 that the conductivity obtained in the air at 101 atmospheres with the 

 highest fields used was about 36 times that obtained with the same 

 field when the ionisation was produced in liquid air. This difference 

 was no doubt due to the rapid recombination of the ions which must 

 have taken place in the liquid air. The layer of ionisation in the liquid 

 air would be considerably narrower than that in the air at the 

 high pressure and consequently the tendency to recombination would 

 be very much greater in the case of the former than in that of the 

 latter. 



V. lonisalion in Air at Atmospheric Pressure by Alpha Rays. 



In another set of measurements the polonium coated plate used in 

 the expeiiments with the liquid air and with the air at the high pressure 

 described in sections III and IV, was attached to the end of an electrode 

 provided with a guard tube and inserted at the centre of a cylindrical 



