458 Rey. Mr. Luoyp on a New Method of Observation 
Hence, if » denote, as before, the total intensity of the magnetic force at Dublin, 
that at London being unity, we have 
om 
$= .9380 x ee = 1.0208.* 
The mean results of the two methods, then, agree ina very remarkable manner ; the 
agreement extending to the fourth place of decimals inclusive. But the differences 
between the partial results and the mean (by which we are accustomed to judge of the 
value of observations) are very different in the two cases. The greatest of these dif- 
ferences, in the method which forms the subject of this paper, is only .0012; while the 
greatest difference, in the three comparisons of the horizontal intensity, amounts to 
.0061, and the corresponding difference in the value of the total force is .0066. 
The difference .006, though it does not appear to be greater than that commonly 
met with in different comparisons of the horizontal force at two places, is yet much 
beyond the limits of the errors of observation ; and, to account for it, we must sup- 
pose the horizontal force to have varied at one or both of the places of observation. 
The existence of such variations seems to be well established. Besides the regular pe- 
riodical changes dependent on the hour and on the season, the horizontal force appears 
to be liable also to accidental fluctuations, or irregular oscillations round its mean 
state ; and the variations of the latter kind (like those of the barometer in our cli- 
mates) are probably more considerable than those that are periodic and regular. These 
variations are, in all probability, the effects of changes both in the intensity and direc- 
tion of the magnetic force; but the latter appear to be (in these high magnetic lati- 
tudes) the predominating cause. The relation which subsists among these changes is 
obtained by differentiating, the equation h—@ cos 8, considering h, ¢, and 8, as all 
variable ; dividing the result by the equation itself, we find 
. = = @ — tan 8 sin Vd; 
the change of dip, dé, being expressed in minutes. When the dip is 71°, the last term 
of this equation becomes — .00084 x d8; so that considering the change of dip as the 
sole cause of the effect observed, a variation of .006 in the amount of the forizontal 
force will be produced by a variation of 7' in the dip ; and this is, probably, within the 
limits of the irregular changes to which that element is subject. 
* The values of the dip employed in the preceding calculation are the apparent values, reduced to 
Needle I, as the standard. When the correction due to the latter needle (see note, p. 451) is applied, 
there will be a small alteration, amounting to +.0007, in the computed value of the relative intensity. 
