ATMOSPHERIC MAGNETISM — EXPERIMENTAL ILLUSTRy^TIONS. 
93 
daily variations of a magnet at the bottom of a deep mine, half-way up, and at the 
mouth of the shaft. The results might tell us much about the holding power of the 
earth and the depths to which the deflections of the magnetic lines of force pene- 
trate, and might even give us a rough expression of the changes of the internal 
power (or the absence of such changes) when freed from those dependent upon the 
atmosphere. 
2998. Another reason why the experimental results must not be applied too closely 
is as follows. If the lines of force of the earth were perfectly regular, then the 
change produced amongst them by the sun and air would be regular also. But as 
the natural system is not regular either between the tropics, as at Sister’s Walk and 
Longford in St. Helena, or in the higher latitudes, as at Hudson’s Bay, so apparent 
inconsistences may and must result. The probability is that the greatest irregularities 
in the arrangement of the earth’s magnetism are in and near the surface of the earth, 
and that above they tend to adjust with each other into a more regular order. Still 
the irregularities must extend their influence very far upwards, so that the contortions 
of the magnetic meridians or lines of force are not likely to be effaced, or much dimi- 
nished, at the region coinciding with the place of the atmosphere’s effect. 
2999. But though the lines are irregular in the large space affected by the sun, the 
result will be expansion of the whole as a system and diamagnetic polarity. The 
lines of force below will be affected by those above ; and so, though a perfect similarity 
between different places is not to be expected, still the kind of change at the earth’s 
surface is not likely to be so uncertain as it might at first appear. Therefore I 
believe the globe (2996.) will be found very useful in giving information regarding 
the probable effects at the magnetic meridian and equator, due to the place of the 
sun in tlie two chief quadrants for any given month of the year, or hour of the day. 
3000. The passage of the magnetic meridian is important, and appears far more so 
after the experiment described (2978.) than it did on a former occasion (2942.). Being 
very often inclined to the astronomical meridian, it must have great influence in de- 
ciding when the daily declination changes in its direction. The place of greatest action, 
and its travelling north or south along a line of magnetic force according as the decli- 
nation was west or east in relation to the helix as a sun, was confirmed by an experi- 
ment ; and the further observation (a consequence of the former), that when the sun 
was equidistant from a place more north or south than itself, its action was far stronger 
on that side at which its path and the declination direction made an acute angle than 
on the other side where it was obtuse, was also confirmed. Thus if the helix, moving 
from east to west, were passing a place north of it having western declination, then 
the action was stronger on the western side of the place than on the east for equal 
distances of the helix from the magnet. 
3001. The passage of the magnetic equator by the sun is also important, since the 
direction of the diurnal variation of the experimental needle is then altered; and this is 
of the more consequence, because by the great degree of natural declination in many 
