370 MAGNETIC METHOD [Chap. 8 



to be a relation between fluctuations of the atmospheric electric potential 

 gradient and magnetic variations, it would seem reasonable that local 

 meteorologic factors can be of influence. Vacquier^*' has shown that, what- 

 ever their cause, such local differences exist not only in the variations 

 following local mean time but also in astronomical (nonperiodic, see 

 2, below) variations as well (see Fig. 8-42). 



It is also possible that local geologic conditions (rocks of high magnetic 

 susceptibility) give rise to induction damping and smoothing of the curves. 

 While Soske could not find differences between variations recorded in 

 Tucson and basalt areas in California, with local anomalies up to 3,5007) 

 Koenigsberger concluded from the rock susceptibilities in two Alpine 

 valleys that variation corrections may be in error by as much as lO-lSy. 

 In such areas, however, his investigations were concerned primarily with 

 nonperiodic variations. 



Because of possible discrepancies between observatory records and local 

 magnetic variations, the only procedure which satisfies all accuracy re- 

 quirements is the local determination of magnetic variations. This 

 requires a second instrument which may be read by another observer at 

 intervals of from 10 to 30 minutes (depending on magnetic activity) or 

 may be provided with a recording attachment as previously described (see 

 also page 356 and footnote 62a). The recording instrument should be 

 free from all magnetic disturbances caused by electric power lines, tram- 

 wa3''s, automobiles, and the like, and be protected against temperature 

 fluctuations. A recording hut or tent may be used for this purpose, or the 

 instruments may be set up in mine tunnels where available. 



2. Magnetic storms occur approximately at the same (or astronomic) 

 time all over the earth. Because of their irregular character, rapid time 

 change, and large amplitude, it is difficult to correct for them. To enable 

 observers to distinguish them from possible local effects, announcements 

 of magnetic storms having occurred the week before are publislied regu- 

 larly in the Oil and Gas Journal.^^ Arrangements ma}^ be made with 

 most observatories for collect telegraph report service. During a storm 

 it is advisable to suspend operations, since stations occupied at that time 

 would have to be i-echecked in any event. 



3. Secular variation is the slow change of magnetic elements in the 

 course of centuries. It is not the same all over the earth. Charts of 

 lines of equal secular variation are published b}^ the U. S. Coast and 



66 Terr. Mag., 42(1), 17-28 (March, 1937). 

 *^ Loc. cit. 



"Zeit. Geophys., 6(2), 74-78 (1930). 



6^ For an illustration of a magnetogram of a magnetic storm, see Oil and Gas 

 Journal, 28(41), 41 (Feb., 1930). 



