TERKEsritlAL MAGNETISM HAZARD 255 



is SO small that it can not be definitely recognized when ordinary 

 magnetic disturbances are in progress. 



Much study has been given by Chree, Chapman, and others to 

 the details of the diurnal variation, particularly to a comparison 

 between conditions on quiet days and disturbed days and between 

 conditions in high and low magnetic latitudes. Material has been 

 provided for a more accurate study of those features which are not 

 purely local, such as the effect of change of geographic position 

 on the form and amplitude of the curves, through the publication 

 by many observatories of diurnal variation tables based on five 

 internationally selected quiet days, thus eliminating to a large ex- 

 tent the lack of homogeneit}^ which magnetic distui-bances might 

 introduce. 



Still another problem is presented by the secular change of the 

 earth's magnetism, which requires centuries for its development. 

 Time enough has not yet elapsed since the beginning of measure- 

 ments of the earth's magnetism to fully determine the character- 

 istics of this change, but we do know that it does not go on indefi- 

 nitely in one direction; eventually there is a reversal. 



In the case of declination, the records at some European stations 

 go back far enough to show two reversals, with a range, in the case 

 of London, of 35° in a little over 200 years. This suggested the 

 idea of periodicity and a motion of the magnetic pole as the cause 

 of the secular change, as it seemed possible to follow the occurrence 

 of a particular phase from east to west around the earth. 



Data accumulated during the past 25 3'ears sliow that the phenom- 

 enon is very complex. The change is by no means regular; the 

 periods indicated for different stations differ widely; unexpected 

 reversals occur and there are undoubtedly waves of shorter ]>eriod 

 superimposed upon the primary wave, if indeed there is a primary 

 wave. It is impossible to predict with accurac}' what conditions 

 will be 10 years from now. 



Bauer has made a mathematical analysis of the secular change of 

 the earth's field as a whole, and has concluded that the system of 

 forces involved is partly witliin the earth and partly outside and 

 that the strengtli of the field is changing as well as its direction. 

 It is not easy to conceive of a systeui of forces acting for a long 

 term of years to produce such a great change in the direction of the 

 earth's magnetic field. A loss of magnetism would not be surpris- 

 ing from our experience with artificial nuignets, but sut-h a rapid 

 loss as one part in 1,500 per year for SO years or more requires 

 verification on the basis of more accurate secular change data for 

 the whole earth. 



Variation of solar activity has been suggested as one cause of the 

 secular change of the earth's magnetism. The periodic change in 



