250 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1925 



When it became apparent that the conception of the earth as a 

 permanent maj^net could not be sustained, the idea was advanced 

 that the earth's magnetic field might be due to electric currents 

 flowing about the earth, either below the surface or in the atmos- 

 phere — the earth an electro-magnet. The mathematical analysis of 

 the earth's field according to the method devised by Gauss and 

 extended by Neumaj^er and Petersen (1891), and Schmidt (1896), 

 indicateil that a small portion of the earth's magnetism, perhaps one- 

 fortieth, could be referred to forces outside th eearth, another small 

 portion to vertical electric currents, but by far the larger pait to a 

 system of forces within the earth. A new analysis made by liauer 

 in 1922, using improved data based on modern observations, gave 

 approximately the same result. He reached the conclusion that for 

 a satisfactory representation of the observed data it is necessar}^ to 

 recognize the existence of an internal magnetic system constituting 

 about 94 per cent of the total field, and an external system and a 

 non-potential system about equal to each other in amount. 



A comparison of his results with those previously obtained for 

 the e]>ochs 1842 and 1885 indicated that the intensity of magneti- 

 zation of the earth had been decreasing during tlie 80-year period 

 at an average annual rate of 1 part in 1,500, a rate of loss which it 

 is hard to reconciU^ with the age of the earth and tlie present inten- 

 sity of magnetization unless we suppose that there have also been 

 periods of increasing intensity. 



Bauer also carried »out the harmonic analysis separately for paral- 

 lels of latitude at 5° intervals from 60° south to 60° noith and found 

 an apparent correlation between distribution o! land and water and 

 intensity of magnetization, the intensity being greater for the 

 parallels falling largely on the land. 



In an earlier (1911) discussion of the earth as an electromagnet, 

 Bauer took the position that the system of magnetic forces Avithin the 

 earth required by the Gaussian analysis might be the result of mag- 

 netizing currents outside the earth, namely, negative electric cur- 

 rents circulating from west to east, but he later withdrew this 

 hypothesis, after further study of the problem. As it seems to be 

 established that the diurnal variations of the earth's field and the 

 disturbances nuist be ascribed to outside currents, it would be much 

 simi)ler to account for those changes if the field itself is due princi- 

 pally to outside currents. 



The idea of the earth as an electro-magnet naturally suggests the 

 possibility that its magnetism may be caused by its rotation. This 

 possibility has been the subject of much study, particularly by 

 Barnett, who has shown experimentally that a piece of iron may be 

 magnetized by rotation; but the observed eti'ect was nuich too snuill 

 to account for the earth's magnetism. Swann attacked the problem 



