Horizontal Component of Magnetic Force. 371 



P.M., and at lO*" 10™ a.m. ; minima at G^ 30" a.m., and at I'' 40"> p.m. Although the greatest value occurs 

 about 4"^ 40™ p.m., the mean difference varies little from 4'' 10™ p.m. till 2'' 10™ a.m. The following are the 

 approximate epochs of maxima and minima of the diurnal variation in each of the astronomical quarters of the 

 year ; — 



Max. lO^A.M. Min. 2" p.m. Max. +6'' 30™ p.m. 



W a.m. 2^ P.M. + l*- A.M. 

 + 3^ P.M. 



+ gl" a.m. l*" P.M. 6'^ P.M. 



10" A.M. 2" P.M. S"" P.M. 



Where the principal maximum or minimum is distinctly marked the former is indicated above by + and 

 the latter by — 



These epochs indicate that the diurnal law of variation of disturbances was different in different quarters 

 of 1844. The law,s of variation in three quarters bear a considerable resemblance to each other, and to that for 

 the year, namely, winter, spring, and autumn ; in these quarters the minima occur about 5 a.m. and 2 p.m. ; 

 one maximum occurs in all about 10" a.m., the other occurs in two about 6" p.m., but in spring it is about I'' 

 A.M. The variation for summer differs most from the others ; there is but one maximum and one minimum. 

 Much of these differences may be due to the fewness of the observations upon which the law depends. 



Annual Variation of the number of Positive Differences. — -The following are the numbers of differences in 

 100 which were positive : — 



Jan. Teb. .March. A\ii:il. .May. June. July. Aug. Sept. Oct. Nov. Dec. 



531 54-8 60-1 52-1 52-5 45-7 47-1 49-8 54-3 61-2 55-6 59-6 

 Whence the horizontal component was oftenest in excess of its monthly mean for the corresponding hours 

 about October and March, when there were about 60 positive to 40 negative observations ; it was least often in 

 excess about June, when there were about 46 positive to 54 negative observations. The number of positive 

 differences was less than the number of negative differences in the months of June, July, and August, but greater 

 in the other nine months ; and in the year there were nearly 54 positive to 46 negative observations. 



Diurnal Variation of the number of Positive Differences. — The number of positive differences in 100 for 

 each hour of Malierstoun mean time in 1844 are as follow : — 



12li lliA-M. S' 8'' 4l> 5l> 611 71" SI" S' 101" 11'' Ol" l' P.M. 21' S" 4l> S^ (!>' 7'' i^ 9'' lO' ll^ 

 5io o6T .38-6 o3-0 57-0 58-9 510 53-2 .51-6 55-4 o9-6 56'1 S2-2 50-3 51-6 48-1 47-1 49-4 48-7 .54-8 49-4 57-0 55-4 .57-3 



The number of positive differences is a ma.\-imum about 4'' a.m. and lO*" a.m. ; it is a principal minimum 

 about 4 P.M. and a secondary minimum about 7'' a.m. The number of positive differences is greater than the 

 number of negative differences in each horn-, excepting those from 3" till 7"- 



Mean Disturbance. 

 Making nse of the means deduced from the five days least disturbed (Table XXIX.) as appro.ximate nor- 

 mal means, as has been already done for the magnetic declination, pages 349 and 350, we obtain the following 

 results. 



Annual Variation of the Mean Disturbance. — The following are the mean disturbances for a single obser- 

 vation in parts of the horizontal component in each month of 1844 : — 



Jan. Feb. .March. April. May. June. July. .Aug. Sept. Oct. Nov. Dec. 



0000 I 358 560 865 774 584 431 490 643 602 700 599 521 



These values differ little from those for the mean difference ; the law of variation is therefore the same, 

 namely, maxima near the equinoxes and minima near the solstices. 



Diurnal Variation of the Positive and Negative Sums of Disturbances. — The following are the sums of dis- 

 turbances in scale divisions for each hour in 1844. 



12b 1 A.M. 2^ 3b i^ 5b $b 7h 8b 9h IQh nh Qh Jb p.m. oh 3h ^h 5h Qh 7h gh gb lO"" 11^ 

 + 495 424 415 450 482 661 665 586 636 625 619 737 879 759 829 951 1090 1005 841 726 685 488 504 425 

 -843 846 924 749 633 729 521 618 694 742 775 612 521 476 426 612 414 478 638 632 636 921 808 940 



The sums of positive disturbances — those which increase the value of the horizontal component — are a 

 maximum at 4" 10™ p.m., and they are a minimum about 1" 30™ a.m. ; there is the appearance of a secondary 



