MARYLAND GEOLOGICAL SURVEY 423

treatment of each element, has been clearly demonstrated by recent
researches. We have in this way obtained a grasp of the secular
variation phenomenon such as but a very few years ago we had no hope
of possessing for many centuries to come. And the only reason why
the grasp is not still more powerful is because of the very fact that
we do not possess sufficient inclination and intensity data to go with
the early observations of the magnetic declination.

If we next turn to the consideration of the so-called local disturb-
ances we shall be most, thoroughly convinced of the absolute necessity
of observing the effect of the earth's magnetism in its totality. And
this question of local disturbances is of no less interest to the practical
needs of the land surveyor than that of the actual value of the mag-
netic declination or that of the secular change of declination.

The outcome of all detailed magnetic surveys has been that the
lines of equal magnetic declination are very far from being smooth
and beautifully curved lines. The more numerous the observations
on which the chart is based, the more sinuous the magnetic lines.
Compare the successive isogonic charts of the Coast and Geodetic
Survey and note how the sinuosities have multiplied with increase of
data. Or behold the extremely devious isogonic lines which resulted
from the most thorough state declination survey thus far made in this
country—that of New Jersey. There was a time when these sinuosi-
ties were looked upon as unnatural, due doubtless to defective data
and what not, and hence to be cut out and eliminated. But the mag-
netician has begun to recognize that these sinuosities are the very
things he wants and that they represent the normal state of things,
and the smooth curves, the abnormal. To cut out, therefore, entirely,
or at best smooth down " local disturbance " phenomena, does not
satisfy even the practical needs, much less the demands of science.

A glance over the values of the magnetic declination data, given
on the chart in a later chapter, shows how impractical and misleading
it is to give an average value, or an estimated value, from an isogonic
chart for a county on the Piedmont Plateau, for example, a value
which the surveyor would naturally think could be applied over his
entire county if given by an authoritative publication. If there are