Vertical Component of Magnetic Foece. 389 



about 6 h p.m. ; the values, however, do not differ greatly from noon till 8 h p.m. ; they are a maximum about 

 2 h a.m. These epochs differ little from those obtained for the horizontal component. 



Diurnal Variation of the Mean Disturbance. — The following are the mean disturbances in micrometer 

 divisions for each hour, without regard to sign : — 



12h lhA.M. 21' 3 h 4 h 5 h 6 h 7 h 8h 9h 10 h ll h h 1> p.m. 2 h 3 h 4 h 5 h 6 h 7 h 8 b 9 h 10 h ll' 1 

 22-2 222 22-1 218 21-5 210 18-9 15-3 13-4 125 11-8 130 15-0 15-1 16-7 20-2 22-4 24-9 24-9 .20-7 16-5 14-7 15-9 17-3 



These quantities give the same law as that already obtained for the mean differences, p. 387. The mean 

 disturbance of the vertical component is a minimum at 10 h a.m. ; it is a maximum at 5 h 40 m p.m. ; a secondary 

 minimum occurs at 9 h 10 m p.m., and a secondary maximum about 2 h a.m., — the value of the disturbance 

 varying little from midnight till 5 h a.m. 



Annual Variation of the number of Positive Disturbances. — The numbers per cent, of hourly observations 

 in each month, which were greater than the normal means for the corresponding hours, are as follow : — 



Jan. 



Feb. 



March. 



April. 



May. 



June. 



July. 



Aug. 



Sept. 



Oct. 



Nov. 



Dec. 



54-5 



41-7 



7T0 



50-0 



52-3 



55-8 



55-4 



46-3 



62-3 



57-3 



40-7 



35-4 



Taking the mean of each three months as the mean for the middle month — 

 43-9 55-7 54-2 57"8 52-7 54-5 52-5 54-7 55-3 53-4 44-5 43-5 



The number of positive disturbances of the vertical component is a minimum at the winter solstice, it is 

 a maximum at the equinoxes ; but the numbers differ little, in the means of the three months, from February 

 till October. In the year 1844, there were, in 100 observations, 52 greater and 48 less than the normal 

 means at the corresponding hours. 



Diurnal Variation of the number of Positive Disturbances. — The numbers per cent, of hourly observations 

 which were greater than the monthly means at the corresponding hours, as deduced from the 5-day series, are 

 as follow : — 



12 h l h A.M. 2 h 3 h 4 h 5 h 6' 1 7h 8 h 9 h 10 h ll h h l h p.m. 2 h 3 h 4 h 5 h 6 h 7 h 8 h 9>> 10 1 ' llh 



38-3 37-7 390 40-3 39-3 40-6 40-6 42-8 40-3 40-6 50-8 55-3 57-2 617 64-5 63-9 67-1 70-6 687 69-3 67'4 60-1 47-9 41-5 



The number of positive disturbances is a maximum about 5 h p.m., and a minimum about l h a.m. The 

 reverse, of course, holds for the numbers of negative disturbances ; they are a maximum about l h a.m., and a 

 minimum about 5 h p.m. Since the numbers of positive and negative disturbances have the same epochs of 

 maximum and minimum as their sums, it may be still a question whether the mean positive and negative 

 disturbance obeys the same law. 



Diurnal Variation of the Positive and Negative Means of Disturbance. — Dividing the sums of positive and 

 negative disturbances by their respective numbers, we have the following means : — 



12h l h A.M. 2 h 3 h 4h 5 h 6 h 7 h 8 h 9h l()h llh 12 h lh P . M . 2h 3h 4 h 5h 6 h 7h 8h 9h 10 h nh 



+ 97 11-3 9-3 9-5 10-0 10-4 10-3 10-4 10-0 12-1 11-6 14-8 19-1 18-4 19-8 26-3 287 31-6 32-2 26-1 19-6 14-3 10-9 10-0 

 -299 28-8 30-2 30-1 29-0 28-3 247 18-9 157 127 11-9 10'8 9-6 97 11-0 9-3 9-8 8-8 8-8 85 10-0 15-4 20-5 22-5 



These quantities give nearly the same results as have already been obtained for the sums of disturbances. 

 The maximum of the mean positive disturbance occurs about 6 h p.m. 



The minimum between ll h p.m. and 8 h a.m. 



The maximum negative about 2 h a.m. 



The minimum about 6 h p.m. 



The mean positive disturbance is about three times as great at 5 h and 6 h p.m. as at the hours from 

 10 h p.m. till 8 h a.m. ; and the mean negative disturbance is three-and-a-half times as great at the hours from 

 midnight till 4 h a.m., as at the hours 5, 6, or 7 p.m. 



Note on the least probable error of an Observation of the Vertical Component. — Investigations similar to those 

 for the probable error of magnetic declination have not been made for this component. The mean difference, 

 however, for an observation of the vertical component is least for the months of December and January, when 

 it is 12-3 micrometer divisions, and for the months of June and July, when it is 14 - micrometer divisions. 



mag. and met. obs. 1844. 5 f 



