L. A. Bauer— Variation of Terrestrial Magnetism. 198 
lars give us a clue as to the location of the operating cause. 
Sometimes the secondary inclination would be greater, some- 
times less than the primary. Some of the differences would be 
plus andsome minus. Hence, on certain parts of the earth’s sur- 
face the perpendiculars will point above the surface, at others 
below. ‘The result will be that the enveloping surface will lie 
partly above the earth, partly below. The difference between 
this surface and the earth’s will everywhere give us a measure 
of the deflecting forces, the maximum departure of the sur- 
faces occurring over the seat of the forces. The curve of 
intersection of the surface with the earth will pass through 
all the points on the earth where the deflecting forces have neu- 
tralized each other and where in consequence, the particular 
element, under consideration, has suffered no change. 
This is precisely the method we would adopt in making a 
thermic survey of a region in order to locate the cause of dis- 
turbance in a uniform heat distribution. We observe the tem- 
perature at various points, imagine perpendiculars erected, lay 
off on them proportionately the temperature observed and 
imagine a surface passing through the ends. The seat of heat 
disturbance would then be indicated on this model by a hillock 
or peak. 
Instead of considering one element alone it would be prefer- 
able to include all three at the same time, viz: declination, 
inclination and intensity, and combine them into vector quanti- 
ties. These latter could in turn be broken up into three rec- 
tangular components, or if the earth’s magnetic force possesses 
a potential, as in fact the greater part seems to do, be treated 
as functions of a potential. For my purposes, however, it 
seems preferable to consider the observed elements separately, 
but simultaneously, and to begin with the dip. The special 
reason why this path appears better is that I shall have to carry 
out the idea referred to in the foregoing for various dates, 
some of them anterior to intensity observations. When con- 
sidering the secular shift of the disturbing forces, we are, 
namely, restricted to angular quantities, declination and incli- 
nation. In order then to have all my results strictly compar- 
able, I shall have to deal almost entirely with these two 
elements. 
On the supposition that the earth is uniformly magnetized 
about the geographical axis, we shall get the following value of 
the potential at any external point P if the X axis of a rec- 
tangular system of codrdinates is taken coincident with the 
magnetic or geographical axis, and the origin at the earth’s 
center :* 
* See Encyclopedia Britannica, Article Magnetism, p. 232, eq. xxx. 
