MARYLAND GEOLOGICAL SURVEY 417

If now the earth were homogeneously magnetized throughout in-
stead of being heterogeneously magnetized, the line joining the equiva-
lent magnetic poles, if prolonged, would pass through the points on
the earth's surface where the dip is equal to 90°, and this line would
be the magnetic axis of the earth. Now only about 7/10 of the total
force of the earth's magnetism can be referred to a homogeneous
magnetization, the remainder being due to the inequalities in the
earth's magnetization. Hence we must neither expect the points
of vertical dip to lie diametrically opposite to each other nor the
magnetic axis of that uniform magnetization part to coincide with a
straight line passing through the earth and connecting the vertical
dip foci. The former line passes through the earth's centre and con-
nects the points on the surface, lying respectively in latitude 78°. 3 N.,
longitude 67°. 3 W., and in latitude 78°. 3 S., longitude 112°. 7 E.,
while the latter does not pass through the centre of the earth, but off
to one side.

In consequence of the heterogeneous magnetization of the earth,
a magnetic meridian line is not a straight line leading to the focus of
vertical dip, but a very devious line indeed. And thus a great circle
passed through the direction pointed out by a compass needle at any
given place will not pass through the dip focus or so-called magnetic
pole, and the intersection of two of such circles will not coincide with
the magnetic pole of Ross. The average position of the successive
intersections of the great circles thus drawn would coincide more nearly
with the points above given, where the magnetic axis penetrates the
earth's surface. ¹

¹ Before leaving the matter of the " magnetic poles " of the earth, it will
be well to call to mind the definition due to Gauss, viz., that the magnetic
poles are the points where the potential of the earth's magnetism has a
maximum and minimum value. Now if the entire force of the earth's mag-
netism could be referred to a potential; if, for example, no part of terres-
trial magnetism is due to electric currents which pass from the air into the
earth or vice versa, then the points of maximum and minimum potential
would coincide with those of the dip foci. But if a part of the earth's
magnetism cannot be referred to a potential, then the two sets of points
need not necessarily coincide. There would appear to be some indication
that a small part of the earth's magnetism is of this nature.