12 



DISCUSSION OF THE DISTURBANCES 



Owing to partial incompleteness in the number of observing months in some 

 years, it became necessary to fill out the number for the annual inequality from 

 the results of the complete years. Their number for each month in the complete 

 years is given in the following table, the numbers for 1844 having first been divided 

 by two, in order to make the hourly observations comparable with the bi-hourly in 

 the years 1841 and 1842: 



The last column contains the ratio of the mean monthly value to the mean 

 annual value. By means of these ratios, and using the observed monthly values 

 in each defective year, the numbers in the following table were filled up, all the 

 deduced values being indicated by brackets. As in the preceding table, the values 

 refer or were made to refer to bi-hourly observations : 



The ratios in the last column show the annual inequality in the distribution of 

 the disturbances. The principal maximum occurs in October, 1 the secondary in 

 April ; the two minima, nearly of equal amount, occur in the months of February 

 and June. The progression of the numbers is regular. 



1 At Toronto this maximum occurred in September ; the first minimum is likewise one month earlier 

 at this station than at Philadelphia. 



