118 



ROYAL SOCIETY OF CANADA 



The first observations were taken during the evening of Nov. 25, 

 1908, and are recorded in Table III. During the readings the baro- 

 meter vt-as practically stationary and stood at 753 mm. The curve in 

 Fig. Ill illustrates the readings taken. The mean value of the readings 

 it will be seen was 15.95 ions. In no case was there a deviation from 

 the mean value of greater than 3%, and the curve shows no evidence of 

 any marked variations which might be ascribed to changes in the 

 radiations which contributed to the conductivity of the enclosed air. 



-«"Eâfïiïïlïï; 





■t F J>r J 



Fig. 2. 



Fig. 3. 



Similar sets of observations were taken on the afternoon and 

 evening of Nov. 27, 1908. These are recorded in Tables IV and V, and 

 illustrated by the curves shewn in Figs. Ill and IV. Here again it 

 will be seen that the extreme deviation from the mean value did not 

 exceed 3.5%. Moreover, the variations which did occur represent 

 practically the limits of accuracy with which the readings could be 

 taken b}^ the measuring instrument. 



S$;:;te;C 'I:; ■l:;:il:;;;(;;;'t:'n-:::| = :'|::: I ■•|-^t-!.- I- -r-i-- :-i:=:t ?» h=Ht=:;r,:-i:. ;|:;- 1; )■,-:) .:!;,;|:-- I' -1 :' I • 1: I" " i -tr-l-'H^H:: 



'T-;^- 



.:ji 



i'T 



1 IG. 4. 



Fig 5. 



Again, sets of observations were taken on the afternoons of 

 December 3, 4 and 7. The results are recorded in Table VI, VII and 

 VIII, and the corresponding curves are given in Figs. V, VI, and VII. 

 From these it will be seen that in one case the deviations from the 



