TRANSACTIONS OF THE SECTIONS. 33 
daily means were obtained for each week. It was not possible, however, for me to 
reduce and discuss the observations for a long period in this way ; I therefore limited 
myself, as regards the changes of mean daily force, to the six weeks commencing 
January 28, ending March 9, 1844, to which period belongs one of the most symme- 
trical curves projected in plate 4 already cited. I performed this discussion for six 
places included within the latitudes of 42° South and 55° North, and I found that 
the changes of daily mean force from day to day followed, on the whole, the same 
law at all the places. The differences and the exaggeration of some of the move- 
ments, especially in high latitudes, seem to depend on the different values of the dis- 
turbance at different places ; but so exactly do the curves in their general forms (and 
frequently in their more minute inflexions) follow each other, that we might, were 
the curves for each place presented separately, suppose that they were the same 
curve. The sudden changes, and, indeed, with a few exceptions, all the turning 
points, occur nearly simultaneously over the globe within the limits considered. 
There seems to be a period of about two days and a half, or of about sixty hours, 
for which I can offer no hypothesis. This period is exhibited more markedly some- 
times near the equator, sometimes in high north latitudes; but in general Hobarton 
has the movements slightly smaller than the other stations; but this may depend on 
the epoch of the year. I can see no connexion between these points and the posi- 
tion either of the sun or moon. I can form no other hypothesis than this, that they 
exhibit the variations of the intensity of a distant magnet. If we suppose a power- 
ful magnet to act at a distance by induction on a smaller magnet, we might expect 
that the daw of distribution of magnetism in the smaller magnet would be unaltered, 
and that the variation in the force of the inducing magnet would be shown on the 
induced magnet. If the induced force were small compared with the proper force 
of the induced magnet, the variation due to induction at any point might be taken 
as proportional to the force at that point. ‘This conclusion is not far from the result 
of these discussions, The whole range of the curve of February 1844, may be re- 
presented approximately by 11 at Hobarton, 123 at Trevandrum, 13 at the Cape of 
Good Hope, and 14 at Makerstoun, the unit in each case being 0°00014 of the whole 
horizontal force at the place. The values of these forces are approximately, in Eng- 
lish units, 4°5, 7°8,4°5,3°4. The total forces are approximately 13°6, 8°1, 7°6, 10°5. 
The comparative values of the range given above cannot be quite accurate; they 
depend on the unit coefficient of the instruments having been accurately determined, 
which is by no means certain in most cases. Even on the theory suggested, it should 
be remembered that the inductive capacity may vary in different parts of the terres- 
trial magnet. The differences of the mean movements are small, however, when we 
consider that the law and amount of disturbance vary with the latitude. Indeed, 
the general agreement of the curves at different places lead one to conclude that the 
vertical force, and therefore the total force as well as the magnetic dip, obey the 
same laws. The diurnal variations of the magnetic declination may be supposed due 
to two opposing forces acting at right angles to the magnetic meridian. From a 
comparison that I have made between the daily mean values of the magnetic decli- 
nation at Makerstoun and Hobarton, I conclude that the mean declination also fol- 
lows similar laws, the north end of the declination magnet moving generally in the 
same direction at both places. The variations of the daily mean force from hour to 
hour are felt simultaneously on all meridians, and, as I have said, they appear to be 
independent of the position of the sun or moon. Within the limits specified, then, 
we may say that the magnetic intensity of the earth increases as a whole or dimi- 
nishes as a whole: it does not increase at one place and diminish at another. ‘These 
results are quite different from those for the variations of hourly mean force. The 
latter varies with the latitude and the sun’s position (longitude and hour-angle). 
The horizontal force is greatest near the equator shortly before noon; it is least at 
the same time in high latitudes. If we suppose, then, that the variations of daily 
mean force are due (for instance) to the varying force of the solar magnet, we must 
conclude that it acts on the whole terrestrial magnet simultaneously, so that there is 
an increase or diminution of force everywhere at the same time. ‘This hypothesis, 
however, will not explain the diurnal variation, and we must suppose that it is pro- 
duced by a wholly different mode of action. I have suggested that these variations 
et to the inducing action of the sun shifting the isodynamic lines, the direction 
858. 3 
