Prof. NortoD on Molecular Physics. 279 



succeeding period. Since it appears, from Mr. Scott's discus- 

 sion of the secular variations (see Report of Coast Survey for 

 1855, p. 337), that the secular period is shortening on the west- 

 ern coast of the Atlantic, we have to infer that we are at present 

 in that magnetic phase in which the reinforcement of intensity 

 from the new currents is less than the annual diminution. In 

 this circumstance we have the probable explanation of the an- 

 nual diminution of the horizontal force in the United States and 

 Canada. An increase of the photospheric currents may cooperate. 



Another general principle should be had distinctly in mind 

 in this connexion : it is that the action of the auroral matter re- 

 ceived from the sun upon the photosphere of the earth developes 

 there a system of currents, the tendency of which should be the 

 reverse of that of the corresponding system continually developed 

 in the crust of the earth by the aether of space. The relative 

 direction in which the solar matter approaches the earth is also 

 approximately the same as that of the impulsive action of the 

 aether — only that, in proportion as the velocity of recess from the 

 sun is greater, the direction of approach is displaced toward the 

 sun. As already intimated, the impinging solar matter also de- 

 velopes radial currents by direct action propagated to the crust 

 of the earth. This effect we have recognized in the partial de- 

 pendence of the morning variation of the declination, and of the 

 horizontal force, upon the eleven-year period of the sun's spots. 

 It is also strikingly manifest in determining the principal de- 

 flections of the needle during an aurora borealis, at the same 

 time that the ecliptic and equatorial currents from east to west 

 developed in the photosphere have the effect of diminishing the 

 horizontal force. This supposed action of the solar matter upon 

 the crust of the earth may arise either from the direct propaga- 

 tion of the impulses, as already intimated, or, more probably, 

 from an increase in the density of the aether, resulting from the 

 acceleration of the fall of the matter in question, produced by the 

 earth's attraction. 



The secular variations should also be dependent in some de- 

 gree upon the electric currents due to the solar matter. In fact 

 there is abundant evidence of such dependence. The annual 

 rates of variation of all the magnetic elements vary during the 

 eleven-year period, as they should do upon this supposition. 

 Thus the tendency to a westerly deflection of the needle, and to 

 a diminution of the horizontal and vertical forces, is least in the 

 year of minimum spots and magnetic disturbances. It is inter- 

 esting to observe, in the Philadelphia Observations, how mani- 

 festly this minimum tendency existed in the case of each of the 

 three elements in the years 1842 and 1843. Another evidence 

 of the dependence in question is afforded by the fact that the 



U2 



