274 Prof. Norton on Molecular Physics. 



action of these increases, as we have seen, from the vernal to the 

 autumnal equinox, and an action to augment the horizontal 

 force each day is followed by one to diminish it, the tendency 

 of the daily accumulation of residual currents should be to di- 

 minish its mean daily value from the vernal to the autumnal 

 equinox. Such a tendency does in fact manifest itself. As the 

 result of the observations at Philadelphia already referred to, 

 the mean monthly value of the horizontal force was 0*0018 of 

 its absolute value less in September than in March*. 



In the former memoir it was maintained that the Irregular Dis- 

 turbances of the magnetic needle might be satisfactorily explained 

 if we admit the existence of occasional photospheric currents 

 proceeding from various points over the preceding and following 

 hemispheres of the earth, and that the ordinary region of maxi- 

 mum excitation lies in the plane of the ecliptic from 60° to 90° 

 to the west of the point of the earth's surface that has the sun 

 in the zenith, and that the region diametrically opposite to this 

 is a secondary region of special excitation. In special instances 

 the point of maximum excitation may have other positions, 

 nearer the meridian in which the sun lies. An adequate cause 

 for such occasional currents may be found if we conceive that 

 they result from the penetration from time to time into the 

 earth's photosphere of bodies of auroral or vaporous matter, 

 expelled from the sun, and arriving with absolute velocities 

 ordinarily less than that of the earth in its orbit. The pho- 

 tospheric currents may be conceived to result either directly, 

 from the impact of the auroral matter, or indirectly, from elec- 

 tric discharges at special localities within the photosphere, con- 

 sequent upon the reception and distribution of such bodies of 

 matter. 



* There are still other operative causes that tend to produce annual va- 

 riations of horizontal force, viz. all the changes that occur annually in the 

 effective action of the ecliptic currents, whether developed in the crust of 

 the earth or its photosphere. The general causes of change are (1) a varia- 

 tion in the velocity of the earth in its orbit ; (2) a variation in the direction 

 of the currents excited; (3) a change in the extent of the portion of each 

 parallel of latitude that is exposed to the impinging action of the aether or 

 auroral matter'; (4) a change in the direction of the progressive motion of 

 the solar system as compared with the direction of the orbital motion of the 

 earth. The epochs of maximum and minimum dependent upon the first 

 cause should fall near the solstices ; and those dependent upon the third 

 and fourth causes should fall near the equinoxes. The effects of the second 

 cause will vary with the locality. The currents due to the general motion 

 of the solar system are most intense just before the vernal equinox, and least 

 intense just before the autumnal equinox. 



The conjoint action of the two systems of currents that have been under 

 consideration, in determining the annual variations, might be strikingly ex- 

 emplified by considering those which occur at the intertropical station of 

 St. Helena. 



