HorizonTAL COMPONENT OF MAGNETIC Force. XXXI 
50. The following Tables have been deduced from the observations of the bifilar magnetometer in the same 
manner as the Tables already given for the magnetic declination. The variations are expressed in terms of the 
whole horizontal component, the latter being equal to unity. 
TABLE 22.—Monthly Means of the Variations of the Horizontal Component of Magnetic Force 
at Makerstoun. 
Month. | E | F 1846. 
| | 
January | -005055| -008747 | -012663 -014943 | -016211| -017925 -017805 
February -005230| -008826 | 012845) -015013) -016316| -017573 -017649 
March ‘005627 | -008584| -012661) -014988  -016354}| -017806| -018064 
April -005439 | -008760 | -012976| -014890 | -016354| -017731) -018406 
May -006492 | -009769 | -013679)| -015340, -016716| -017751| -019054 
June -006786 | -010233 | -014425)| -015645 | -016570} -018455) -019316 
July -006714| -010104| -014584| 015572) -016939} -018146) -019305 
August | 006796 | 010257 | -014376) -015407 | -016388 -018016| -019116 
September | .007054| -010542| -014360) -015078 -016233)| -017857| -018552 
October -007482| -010774| -014344| -015461 | -016480) -016981)| -019230 
November -007692| -011579| -014740) -015851 | -017161)| -017584| -019029 
December -008239 | -012065 | -015212) -015895 -017775)| -018591)| -019010 
| 
51. Monthly Mean Values of the Variations of the Horizontal Component.—The horizontal force has in- 
creased in the greater number of cases from month to month ; in March or April, and in August or September, 
the mean is generally less than in the immediately preceding months. The means for 1848 and 1849 are consi- 
derably less accurate than those for the preceding years, depending as they do on only two daily observations ; 
and the means for the end of 1846 and for 1847 are much affected by excessive disturbance. 
TABLE 23.—Yearly Means of the Variations of the Horizontal Component of Magnetic Force, 
' with the Secular Change. 
Secular Increase. 
Horizontal 
Component. Each Mean of 
Year. 4 Years, 
| 
0.006550 
| .010020 | -003470 
-013905 -003885 
-015340 -001435 
-016625 -001285 002519 
-017868 -001243 -001962 
018711 -000843 -001202 
-020611 -001900 001318 
52. Table 32 contains the yearly means of the quantities in Table 22, together with the resulting yearly 
values of the secular change. The secular change appears to have been very large in the years 1842 to 1844, 
and considerably smaller and more regular in the years from 1844 to 1847 ; the change from 1847 to 1848 is 
smaller, and that from 1848 to 1849 is greater, than for each of the preceding three years. It is not impro- 
bable that the change from 1842 to 1844 is increased by instrumental causes, such as stretching of the sus- 
peusion wires of the magnet, while the variation of the changes in 1847-9 is evidently connected with the great 
disturbances of the year 1847-8. If we take the mean yearly secular change from 1845 till 1849, as probably 
unaffected by instrumental error, we find it = 0:001318, the horizontal component being unity ; or, if we take 
the absolute value of the horizontal component, = 3°388, we find— 
MAG. AND MET. obs. 1845 anv 1846. h 
