22 



DISCUSSION OF THE VERTICAL FORCE. 



The last column shows that the magnitude of the disturbances is rather irregu- 

 larly distributed over the several months without following any apparent law. 



The magnitudes of the disturbances, as before, do not appear to follow any law. 



The disturbances which increase the force preponderate over those which decrease 

 it ; the ratio of the annual means is 1.3 to 1.0. At Toronto the reverse was found ; 

 the disturbances which decrease the force preponderate over those which increase 

 in the ratio of 1.4 to 1.0. 



Diurnal Inequality of tTie Disturbances. In the bi-hourly combination of the 

 disturbances we make use of the series of observations extending from .February, 



