Variation Phenomenon of Terrestrial Magnetism. 505 



tion of the inner sphere were unknown and proceed as we did 

 at London to find them. It took 96° of differential motion to 

 produce the maximum at St. Helena, which we have found 

 from observation to have occurred about 1890, covering thus a 

 period from zero of about 208 years instead of 256 as Wilde 

 has it. Hence, instead of 6° being equal to 16 years as at 

 London and Good Hope, we find 6°= 13 years and the total 

 period 60x13 or 780 years instead of 960 and the annual 

 amount of retardation 27 - 8 7 instead of 22*5'! In other words, 

 for London and Cape of Good Hope the inner electro-dynamic 

 sphere must make one complete differential revolution in 960 

 years, but to accommodate St. Helena it must go at a more 

 rapid gait and perform the complete revolution in 780 years ! 

 The table before us is the one Wilde presents at the end of 

 his second paper, "with the view," as he says, "that they may 

 be of some service to science in the distant future." 



Declination periods at London, St. Helena, and the Gape of 

 Good Hope for the cycle of secular changes of the earth's mag- 

 netism. 



London. St. Helena. C. of Good Hope. 



Declination, Epoch. Interval. Epoch. Interval. Epoch. Interval. 



0° outward march W. 1657-1817 160 1683-1939 256 1609-1881 272 



0° return " W. 1817-2137 320 1939-2163 224 1881-2089 208 



0° outward " E. 2137-2457 320 2163-2387 224 2089-2297 208 



0° return " E. 2457-2617 160 2387-2643 256 2297-2569 272 



960 960 960 



It will be seen from this table that the consequences of 

 Wilde's theory are : 



(1). That the period of a complete secular change is the 

 same all over the globe — this period being approximately 960 

 years. 



(2). That the declination at every point on the earth, at 

 some time during the cycle, passes through zero. 



(3). That the total amounts of easterly and westerly decli- 

 nations are approximately equal. 



How utterly contradictory these three conclusions are to 

 facts and what little reliance can be put upon Wilde's table, 

 will be seen from the next diagram (B) giving Secular Varia- 

 tion Types of the Magnetic Declination. At no station have 

 we record of a complete period. There are a number, how- 

 ever, where the series of observations will warrant our fore- 

 telling the length of the period certainly within 50 years. 

 The next half century will witness the completion of several 

 of the American stations. The curves on the diagram, with 

 the exception of Christiania, have all been plotted from the. 



