384 Results of Makeestoun Observations, 1844. 



Intermittent Distokbances. 



Effect of Intermittent Disturbances on the Yearly Mean of the Vertical Component. — Performing discussions 

 similar to those already made for the magnetic declination and horizontal component, we find the means of the 

 vertical component as deduced from different series of days as follow, the value at micrometer divisions being 

 unity : — 



Mean vertical component for 1844, as deduced from the 120 days, greater than that from the whole series by 0-000021 

 60 OOOOOl.'J 



The effect of disturbances, therefore, in 1844, was to diminish the mean value of the vertical component, 

 as well as of the horizontal component. 



Effect of Disturbances on the Monthly Means of the Vertical Component. — The corrections of the means 

 from the 10-day and 5-day series selected in each month, to those from the complete series, are, in micrometer 

 divisions, as foUow : — 



June. July. Aug. Sept. Oct. Nov. Dec. 



. 3-1 _31 -20 -4-8 -4-3 +0-6 +5-1 

 ■ 10-6 -3-1 -0-1 -2-6 -2-3 +0-8 +7-2 



These quantities oscillate considerably ; taking the mean of each three in order to eliminate the irregu- 

 larities, we have — 



lOdays, -I-3-6 -2-2 -08 -6-4 -3-3 -4-4 -2-8 -3-3 -37 -2-8 +0-5 +1-9 

 5 days +5-1 -1-3 -0-9 -5-5 -5-0 -5-7 -4-6 -2-0 -1-7 -1-4 -I-1-9 -|-1'9 



From whence it appears, that the maximum effect of disturbances, to diminish the monthly mean, occurs about 

 May, and the maximum effect, to increase the monthly mean, about January. The differences of the means 

 from the selected and complete series, are evidently partially due to the secular change, the means of the 

 selected series not corresponding to the middle of the month ; the error due to this cause, however, does not 

 affect the above result to any marked extent. 



Effect of Disturbances on the Hourly Means of the Vertical Component. — Considering, first, the differences 

 of the hourly means for the year, as obtained from a comparison of the whole series with the 60-day series of 

 observations (see last column of Table XLIX.), we obtain the following results : — 



The mean effect of disturbances upon the hourly means of the vertical component, is a positive maximum 

 at 5'' 30"' P.M.; it is a negative maximum about 2'' a.m., and it is zero at 10*" a.m., and near lO*" p.m. The 

 comparison of the whole series ^vith the 120-day series (see last column of Table XL VIII.) gives exactly 

 the same result. 



The times for the maximum effect of disturbance on the hourly means of the vertical component, are about 

 1| hours after those for the horizontal component. 



The greatest effects of disturbance in increasing and diminishing the hourly means of the vertical compo- 

 neut for the year 1844, as deduced from the comparisons with the 60- day series of observations, are as follow : — 



Maximum effect in increasing the vertical component (5'' 30"" p.m.) =0'000196 



diminishing (2t>A.M.) =0000148 



As the whole diurnal range in the mean for the year fi'om the whole series of observations is 0-000419, the 

 effect of disturbance is greater on this component than on either of the other elements already discussed, being 

 fi-om about + ^- to — ^ of the whole range. 



The differences for each month from botli series of comparisons, give, with slight variation, the same times 

 of maximum and minimum effect, the amount being greatest at the equinoxes. 



The following are the differences between the hourly means for the 120-day series, and those for the 60-day 

 series, or means for 120 days niimis means for 60 days. (See last columns of Tables XLIV. and XLV.) 



