Hoeizontal Component of Magnetic Force. 365 



In the group for the year, 



The maximum occurs about 1£ hours after the moon's inferior transit, 

 The minimum 2 ...... after superior transit. 



There are appearances of minima before and after the latter epoch, but they are not distinct. The 

 variation for the year, therefore, has the same epochs as that for the winter group. 



As the range of this variation is so small, it was supposed that the large disturbances might have 

 considerable effect in destroying its regularity. In order to test this, all those observations which differed 

 more than 25 scale divisions (0-0035) from the monthly means for the corresponding hours, were rejected in 

 the summations, quantities interpolated from the preceding and succeeding observations having been substituted. 

 The following are the resulting variations at nearly two hourly intervals : — 



Oh om 2 h 25" 4h 20 m 6 h 15 m 8 h 10»> 10 h 5 m 12 h m 13 h 55" 15 h 50 m 17 h 45 m 19 h 40 m 21 h 35 m 



1844 ) °' 000 I ° 47 ° 53 101 ° 55 ° 76 ° 74 164 187 ° 59 ° 78 ° 00 ° 71 



These give the same time of maximum, but the minimum occurs four hours before the superior transit, 

 instead of after it; a secondary minimum also occurs about seven hours after the superior transit, — a 

 secondary maximum occurring between the two latter epochs. The elimination of those observations affected 

 by large irregularities renders the variation for the year analogous, in its singular points, to that for the 

 summer months given previously ; it is quite possible, therefore, that the differences between the law of 

 variation for the summer and winter groups may be due chiefly to intermittent disturbances. 



Intermittent Disturbances. 



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

 discussions for the horizontal component of magnetic force similar to those already made for the magnetic 

 decimation (p. 343), we obtain the following results : — 



Mean of the horizontal component for 1844, as deduced from 120-day series of| n-nnni RQ 

 hourly observations, greater than that deduced from the whole series by J 



Mean of the horizontal component for 1844, as deduced from 60-day series of "1 _ nnnom 

 hourly observations, greater than that deduced from the whole series by J 



Hence the effect of disturbances in 1844, was to diminish the mean value of the horizontal component of 

 force; the mean for the 60-day series of observations being greatest, 0-000062 greater than that of the 120- 

 day series, and 0*000251 greater than that for the complete series. 



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

 means from the 10 days and 5 days selected in each month to the means from the complete series, are, in scale 

 divisions, as follow : — 



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



10days,-0-78 -2-47 -2-85 -2-25 -0-21 +0-13 -015 -0-77 -0-44 -3-62 -1:17 -1-67 

 5 days, -1-06 -2-93 -3-75 -2-93 -0-96 +0-82 +0-01 -0-97 -I'll -4-03 -2-44 -2-11 



The corrections for both series give the same result ; but it is most marked for the 5-day series, that 

 least affected by disturbances. This result may be stated as follows : — The effect of disturbances is to 

 diminish the montldy mean of the horizontal component in all cases, with the exception of that for June, and, 

 perhaps, July ; the diminution is greatest near the equinoxes, and least at the solstices ; at the summer 

 solstice the effect is to increase the monthly mean. When we compare this result with that already given for 

 the annual period (see p. 356), we might be induced to conclude that the latter is due to disturbances : this, 

 however, is not the case. If we apply the above corrections reduced to parts of force, to the monthly means, 



MAG. AND MET. OBS. 1844. 4 Z 



