POLTTECHNIG ASSOCIATION PROCEEDINOS. 947 



That terrestrial magnetism varies in intensity is shown by the 

 declinations of the magnetic needle; that these variations are 

 effected by the sun, is also quite generally conceded; but no satis- 

 factory explanation of the phenomena has yet been given. M. Aime 

 attributed these effects to thermo-electric currents produced by the 

 unequal heating of the earth in its diurnal rotation; but far the 

 largest portion being covered by water, which is not susceptible to 

 thermo-electricity, we find in this fact a suificient objection to the 

 theory. Faraday conceived the idea that magnetic variations were 

 due to changes in the magnetic properties of oxygen. This gaseous 

 element, he proved, was strongly magnetic when cold, but corre- 

 spondingly less so when heated. To the calorific eflect of the sun 

 upon the atmosphere, therefore, he attributed magnetic perturba- 

 tions. The prominent objection to this view is, that the magnetic 

 movements are quite regular, while the heating effects of the sun 

 are variable. 



Father Secchi, on reducing the observations made at various 

 observatories, and constructing curves representing the motions of 

 the needle, arrives at the conclusion that the sun acts upon the 

 magnetic needle as if it were itself a large magnet placed at a great 

 distance from the earth, and having its poles, of the same name as 

 those of the earth, turned to the same side of the heavens. Both 

 the sun and earth, we may add, must be regarded as magnets. 

 Minor disturbances in the declination of the needle have been 

 ascribed, by M. Reslhuber, to changes in the humidity of the 

 atmosphere. With regard to the aurora, De la Rive presents the 

 following summary: 



" 1st. All observations concur in demonstrating that the aurora 

 borealis is a phenomenon having its seat in the atmosphere, and 

 which consists in' the production of a luminous ring having for its 

 center the magnetic pole, and of a greater or less diameter. 



" 2d. Experiment demonstrates that, on bringing about in rarified 

 •air a reunion of two electricities near the pole of a powerful mag- 

 net, a small luminous ring is produced similar to that which consti- 

 tutes the aurora borealis, and animated by a similar movement of 

 rotation. 



" 3d. The aurora borealis would be due, consequently, to electric 

 discharges taking place in the polar regions, between the positive 

 electricity of the atmosphere aiid the negative of the terrestrial globe. 



" 4th. The electric discharges taking place constantly, but with 

 intensities vai-iable, according to the state of the atmosphere, the 



