ATMOSPHERIC MAGNETISM — EFFECT OF COOLED AIR. 
49 
dip is 70 ° nearly, and at the equator, where it is 0°, would be a very different matter, 
and the necessary natural results of such a difference ought to appear hereafter. 
2868. But a magnetic needle or bar is not a test of such intensity, i. e. it will not 
tell these differences, or it may tell them in a contrary direction. To understand this 
point, we have to consider that a needle vibrates by gathering upon itself, because of 
its magnetic condition and polarity, a certain amount of the lines of force, which 
would otherwise traverse the space about it ; and assuming that it undervvent no 
change by change of temperature, it would be affected in proportion to any variations 
in the intensity of these lines, provided everything else remained the same. But being 
under natural circumstances surrounded by the atmosphere, which is a medium liable 
to variation in its magnetic condition, both by heat and rarefaction, and by these 
variations affects the intensity or quantity of the force, it will vary in its indications 
by variations in these conditions. Thus, for instance, if it were in a large sphere of 
oxygen, I expect that it would, by its number of vibrations or otherwise, indicate a 
certain intensity ; if the oxygen were expanded, that it would indicate a higher in- 
tensity, although the same amount of lines of force and magnetic energy were passing 
through the oxygen as before. If the oxygen were made dense, then becoming a 
better conductor, I presume it would convey onwards more of the force and the mag- 
net less, for the power would be partly transferred from the unchanging magnet to 
the improving conductor around it. 
2869. These experiments can hardly be made with oxygen except by means of ex- 
tremely delicate apparatus, but like effects are easily shown experimentally in selected 
analogous cases. Thus let a thin small tube of flint-glass, about 1 inch long and ^ an 
inch in diameter, be filled with a saturated solution of protosulphate of iron, and 
suspended horizontally by cocoon-silk (2279.) between the poles of the electro-magnet, 
in a vessel which may either contain air or water, or other media (2406.). In air it 
will point axially, and will be analogous to a needle under the earth’s influence, and 
it will point with a certain amount of force. Fill the vessel with water, and now it 
will point with more force than before, though the water is a worse magnetic con- 
ductor than the air which was previously there ; and it is precisely because the water 
is a worse conductor that the liquid magnet or test indicates more power. Increase 
the conducting power of the surrounding medium by adding sulphate of iron to it, 
and the indication of strength by the tube goes on diminishing, first returning to the 
degree of power it had in air, and then descending to lower gradations, for it returns 
with less and less force to its axial position when disturbed from it. So the magnetic 
needle employed for measuring intensity or magnetic force (for the same meaning is 
at present understood by the two terms), indicates, in a certain manner, the power 
thrown upon itself, and, I conclude, accurately, provided the condition of the sur- 
rounding medium remains magnetically unchanged ; but if it be placed in different 
media or in an altering medium, I expect that it will not measure accurately the in- 
tensity in them, i. e. it will not measure directly the amount of force passing relatively 
MDCCCLI. H 
