xl GENERAL RESULTS OF THE MAKERSTOUN OBSERVATIONS. 
TABLE 33.—Diurnal Variations of the Horizontal Component of Magnetic Force for Different 
Periods, deduced from Days selected as free from Irregular Disturbances, in the Years 1844 
and 1845. 
Six Months. 
March. ‘ —________—_—|| Twelve 
April. I , Sept. to March Months. 
Feb. to Aug. 
0-00 
+0046 
-— 0018 
— 0010 
+0057 
+0139 
+0227 
+0178 
+ 0036 
— 0338 
— 0679 
— 0907 
— 0764 
— 0478 
— 0056 
+0136 
+0209 
+0291 
+0360 
+0409 
+0346 
+0285 
+0217 
+0185 
+0130 
RKB OOCODNANRWNHK OS 
— 
71. The following are the approximate epochs in apparent time deduced from Table 33. (See also the 
dotted curves, Plate IT.) 
Dec. Jan. Feb. March, April. May, June. July, Aug. Sept. Oct. Nov. 
Minimum, 105 45™ a.m. 102 25™ a.m. 9 55™ am. 9) 50™ a.m. 105 10™ a.m. 
Maximum, 5h 30™ p.m. 7h O™ pM. 6" 30™ p.m. 6» 55™ pM. 6» 35™ p.m. 
Minimum, 1» 30™ a.m. SQM cc. eadeesttetee 1> 55™ a.m, 
Maximum, 6> 35™ a.m. DEAS OM A Mem: 5. | aeRanc ae 55 30™ a.m. 
A comparison of these epochs with those deduced from Table 32 will shew, that the effect of disturbance is 
to accelerate the epochs of the forenoon minimum and afternoon maximum, those of the latter being most 
affected. In the undisturbed diurnal variation the afternoon maximum occurs latest in March and April, and 
in July and August. 
72. Diurnal Variation of the Effect of Disturbance on the Horizontal Component.—The following result is 
obtained upon the assumption, that intermittent disturbance which affects the hourly mean position does not 
affect the monthly mean of the 24 hours; or, that the differences found No. 57, between the monthly means of 
the undisturbed days, and of all the days, is due to continuous and regular laws, which have little effect on the 
relative hourly positions ; it appears very probable from No. 58, that this assumption is not quite accurate, but 
that the negative quantities in the following Table are too small, and the positive ones too large; those 
for the summer months, however, must be near the truth, as disturbance had little or no effect on the mean for 
that group: the error in the values for the other periods cannot affect the epochs of positive and negative 
maxima. 
