Chase.] 3g8 LApril. 



all portions of the earth, produce an atmospherical directive energy 

 towards the poles of maximum cold, which appear, according to Sir 

 David Rrewstei*, to coincide with the magnetic poles. 



Now, if we consider that in addition to these permanent currents, 

 there is a continual motion of silent convection, the warm air rising, 

 and the cold air descending in parallel columns, like the particles in 

 a vessel of boiling water,* and if we remember that the warm air is 

 charged with moisture which is condensed as it ascends, parting 

 thereby with much of its heat and electricity, we can hardly deem 

 it necessary to adopt Dr. Dalton's hypothesis that ferruginous matter 

 is the source of atmospheric magnetism. Still the existence of va- 

 porized iron in the air undoubtedly contributes an increased intensity 

 to the magnetic currents, and it may probably be an important agent 

 in the production of magnetic storms. 



The two vibratory systems represented in Figs. 1 and 2, are con- 

 joined during the hours when the sun is above the horizon, and the 

 laws of motion applicable to the first system correspond precisely, as 

 we shall see hereafter, with the laws of the solar-diurnal variation 

 deduced from General Sabine's admirable discussions of the St. 

 Helena observations. It is not so easy to explain in its minute 

 details the comparatively insignificant lunar-diurnal variation, but I 

 am convinced that the aerial currents produced by lunar attraction, 

 will sufficiently account for all the magnetic influence that is due to 

 the moon exclusively. The changing barometric pressure, and the 

 deposition of dew during the night, modify these currents in such a 

 way as to disguise the simple effect of any slight disturbing cause ; 

 nevertheless, there is a manifest tendency underneath all the disguise, 

 to maxima and minima at the precise hours when they ought to occur 

 in consequence of the moon's attraction. 



In the influence of the violet rays upon magnets, the connection of 

 the violet rays with the tension of brass in the polariscope, the excite- 

 ment of magnetic vibrations in iron by percussion and torsion, the 

 increase of magnetism by cold and its diminution by heat, and the 



all around the globe." I am not aware that any one has ever pointed out the 

 combined effects of convection, absorption of heat from the earth, and the daily 

 superposition of the currents represented in Figs. 1 and 2. 



* It is very probable that this motion of convection is a more important 

 agency than has generally been supposed. If we close the lower drafts of a com- 

 mon air-tight stove, and open a register immediately over the fire, the cold air 

 does not rush directly to the draft pipe ; but it falls with great velocity to the 

 surface of the fuel, as may be shown by dropping pieces of paper through the 

 register. 



