210 ANNUAL EEPOET SMITHSONIAN INSTITUTION, 1913. 



not be able to solve the riddle until the physicist answers for us the 

 questions: What is a magnet? What is magnetism in general? 



In the Devil's Dictionary by Ambrose Bierce, published in 1911, 

 t he following definitions are given : " Magnet, n. — Something acted 

 upon by magnetism. Magnetism, n. — Something acting upon a 

 magnet." In explanation the author cynically remarks : " The two 

 definitions immediately foregoing are condensed from the works of 

 1,000 eminent scientists, who have illuminated the subject with a 

 great white light, to the inexpressible advancement of human 

 knowledge." ^ 



A line of thought first suggested by Schuster and Lord Kelvin, 

 that every large rotating mass, due to an As yet undiscovered cause, 

 may be a magnet, should be considered in conclusion, though we 

 may do so but briefly. If this be true, then magnetism is not confined 

 to our planet alone, but all celestial bodies are surrounded by mag- 

 netic fields. Thus far no laboratory experiment, possibly owing to 

 lack of required sensitiveness in the measuring instruments, has 

 detected any magnetic field arising solely from rotation. Schuster 

 and Swann have recently discussed the character and magnitude of 

 the effects from the possible causes which may operate if the earth's 

 magnetic field be related in some manner to its rotation. 



In 1900-1903 Sutherland propounded a theory for the origin of 

 the earth's magnetism, which, briefly stated, is this: We know that 

 electricity is an essential constituent of matter, and that in every 

 atom, if it be electrically neutral, there are equal amounts of nega- 

 tive and positive electricity. So with the whole earth. Since it is 

 almost electrically neutral, suppose that the total negative charge, 

 while practically equal to the total positive one, occupies a slightly 

 different volume from that of the positive charge, or, in brief, that 

 the volume densities of the two body charges differ slightly, then, 

 because of the rotation of the electric charges with the earth, a mag- 

 netic field arises. I have recently repeated Sutherland's calcula- 

 tions and, as I had previously found that the earth's intensity of 

 magnetization increased systematically toward the Equator, I have 

 included a term to represent such a possible effect. The computa- 

 tions show that to satisfy the known phenomena of the earth's mag- 

 netism, the volume density of the negative charge must be smaller 



1 These definitions and accompanying remarks may have had their origin in the follow- 

 ing interesting anecdote told in the American Review of Reviews for August, 1909, of the 

 late Prof. Simon jsjewcomb, by Mr. A. E. Bostwick, associate editor of the Standard Dic- 

 tionary. Of the definitions in physical science for this dictionary Newcomb had general 

 oversight, and on one occasion he took exception to the definitions framed for the words 

 " magnet " and " magnetism " as based, in the absence of authoritative knowledge of 

 the causes, simply upon the properties manifested by the things. After writing and 

 erasing alternately for an hour or more, he finally confessed, however, with a hearty 

 laugh, that he himself could offer nothing better than the following pair of definitions : 

 " Magnet, a body capaple of exerting magnetic force ; and magnetic force, the force 

 exerted by a magnet." 



