634 



SCIENCE. 



[N. S. Vol. XX. No. 515. 



and will continue to be, the chiefly impor- 

 tant method of sociology ; and assuredly, in 

 the course of time, it will bring our knowl- 

 edge of society up to standards of thor- 

 oughness and precision comparable with 

 the results attained by any natural science. 

 Feaniclin H. Giddings. 

 Columbia University. 



REGENT ADVANCES IN THE ANALYSIS OP 

 TEE EARTH'S PERMANENT MAG- 

 NETIC FIELD* 



The 'earth is a great magnet' and as 

 such is subject to the same laws which per- 

 tain to any other magnet— these are facts 

 established by the experience of over four 

 centuries. How and whence the earth has 

 received its magnetism are questions we 

 can not as yet answer, nor, in my opinion, 

 shall we be able to answer them definitely 

 until we have solved the problems as to 

 the causes of the variations of the earth's 

 magnetism. I firmly believe that when we 

 have discovered the causes of the periodic 

 _and aperiodic variations, such as the di- 

 urnal variation, annual variation, secular 

 variation and magnetic perturbations, we 

 shall have strong hints given us as to the 

 07-igin of the earth's magnetism. It is 

 through the study of the variatiom, then, 

 that we hope some day to be able to attack 

 the problem as to the origin with some 

 degree of success. Until this study has 

 been completed, it is not believed that any- 

 thing more than mere surmises, such as the 

 magnetic literature contains in quanta can 

 be given. 



Whether the earth is a magnet like a 

 lodestone or an electromagnet, is another 

 question which can not as yet be definitely 

 answered, though there are various indica- 

 tions that the earth's magnetization par- 

 takes of the character of both. Here again 



* Presented before the Joint session of Section 

 A of the International Electrical Congress and 

 the American Physical Society, at St. Louis, on 

 September 23, 1904. 



the definitive answer depends upon the suc- 

 cessful solution of the questions as to the 

 variations of the earth's magnetism both 

 as to time and space. 



These introductory paragraphs are in- 

 tended to emphasize the proposition that 

 if progress is to be made in the subject of 

 the earth's magnetism, we must first make 

 a careful and exhaustive study of the facts 

 which are daily experiences, before at- 

 tempting broad, theoretical generalizations 

 based on more or less inadequate data per- 

 mitting at the most mere qualitative tests 

 of the deductions of theory. What are 

 needed are the facts for quantitative tests. 

 Even then, it will be found, in some in- 

 stances, that more than one theory will 

 satisfactorily explain the same facts and 

 that a final decision must be left to future 

 generations. However, the facts will re- 

 main as a permanent acquisition. The ac- 

 cumulation of clean-cut facts regarding the 

 earth's magnetism is the great task of the 

 present generation. 



In the hope of enlisting interest in this 

 comparatively unexplored field of scientific 

 inquiry, it will be my endeavor to reveal 

 some of the gaps to be filled as well as to 

 exhibit those facts considered as safely es- 

 tablished. It must be remembered that we 

 are working in a field bordering on several 

 other sciences, such as astrophysics, geo- 

 physics, geology and meteorology, so that 

 he who wishes to become an expert must 

 have at his command the ability to make 

 the best and most intelligent use of the 

 experimental facts of several of the older, 

 recognized sciences. The physicist now-a- 

 days has no time to attempt to master so 

 special and comprehensive a subject as 

 that of the earth's magnetism, with its 

 manifold ramifications into cognate sci- 

 ences, for he finds it sufficiently difficult to 

 keep in touch with the rapid advances in 

 his own subject. However, if the physi- 



