ATMOSPHERIC MAGNETISM — DIURNAL VARIATION — AMOUNT. 
71 
declination, according to the usual mode of marking this effect by the north end of the 
magnet, then the variations already described should come on as the sun passes mid- 
way between them, but there should be a difference in time. As the luminary appears 
and approaches, the needles a and h (fig. 16 ) will most 
probably be affected together; but, as he draws nigh, if 
the places have eastern declination, the one that is south 
will be soonest affected, and for the time most strongly, 
but will in a period more or less extended, be followed 
by the corresponding action at the other place. For as 
each needle will have returned from the first half of its 
series of changes to 0° by the time the sun is on its mag- 
netic meridian, and as it will arrive at this meridian, as 
regards the south needle, before it does so for the north 
needle, so the south magnet should precede the other in 
its changes. If the declination of both were westerly, 
then the north needle would precede the south. s 
2943 . The hypothesis advanced, besides agreeing with the facts regarding the 
direction of the needle’s motions, as is the case generally, and if my hopes are well- 
founded, will be the case also in more careful comparisons ; should also agree in the 
amount of force required for the observed declinations at given hours. I have 
endeavoured to obtain experimental evidence of the difference of action of oxygen 
and nitrogen on needles subjected to the earth’s power, but have not yet succeeded. 
This however is not surprising, since a saturated solution of protosulphate of iron 
has failed under the same circumstances. More delicate apparatus may perhaps 
yield a positive result. 
2944 . That small masses of oxygen should not give an indication of that which is 
shown by the atmosphere as a whole is not surprising, if we consider that the mass 
of air is exceeding great, and includes a vast extent of the curves on which it, by the 
hypothesis, acts ; and yet that the effect to be accounted for is exceeding small. 
The extreme declination at Greenwich is 12 ', equal to about 4 ' 24 " of east and west 
alteration on the free needle, so that that is the whole of what has to be accounted 
for. One could scarcely expect such an effect to be shown by small masses of 
oxygen and nitrogen acting on only a few inches in length of the magnetic curves 
passing through them, unless one could use an apparatus of extreme and almost 
infinite sensibility ; but from what I have seen of oxygen when compared at different 
degrees of dilution ( 2780 .), or at different temperatures ( 2861 .), I am led to believe 
that the effects on it produced by the sun in the atmosphere will ultimately be found 
competent to produce these variations. 
2945 . Where the air is changed in temperature or volume, there it acts and there 
it alters the directions of the lines of force ; and these by their tension carry on the 
effect to more distant lines ( 2934 .), whose needles are accordingly affected. The trans- 
ferred effect will be greater or less according as the distances are less or greater, and 
Fig. 16. 
