46 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXVI.) 
melted a neat round portion of the wax, showing the place of the current under natu- 
ral circumstances ; but when the magnet was thrown into action, then the wax on 
the mica remained unchanged, the hot air being thrown so far away from the axial 
line, and so cooled by its forcible mixture with the neighbouring air, as to be unable 
to melt a spot of wax anywhere. The moment the magnetic power was suspended, 
the column of hot air rose vertically and regained its original position. 
2857- Carbonic acid gas was then sent into the shade, until twice as much as the 
contents of the shade had passed through the pipe (2855.) ; but as it was heavy and 
the common air could make its way out only at the bottom of the shade, there was 
no doubt air mixed with the carbonic acid, which at last remained about the poles. 
The platinum coil being now heated, the column of hot gas rose vertically, as before. 
On putting on the magnetic force it was deflected from the axial line, passing equa- 
torially, and melted the wax about half an inch off from the former place. Believing 
that even this effect might be due to the air mingled with the gas, other two volumes 
of carbonic acid gas were directed into and through the vessel. After this the mag- 
netic force caused much less deflection of the rising column. Two volumes more of 
carbonic acid were sent through, and now the hot current of gas rose so nearly ver- 
tical that there was scarcely any sensible difference of its place when the magnetic 
power was in full action, or when it was entirely absent. Hence I conclude that car- 
bonic acid gas is very little affected in its diamagnetic relations by a difference of 
temperature equal to that between natural temperatures and a full red heat. 
2858. Nitrogen . — This gas was prepared by passing common air slowly over burn- 
ing phosphorus, and after being washed for twelve or fourteen hours, was sent into 
the shade so as to displace the carbonic acid. As it was lighter than the latter, it per- 
formed that service very well, and the portion remaining in the vessel probably con- 
tained no other oxygen or air than that it carried in with it. This nitrogen being then 
heated by the platina coil, was almost as indifferent to the magnet as the carbonic 
acid. The heated column rose (nearly) to the same spot against the mica, whether 
the magnetic power was active or not. It went outwards or equatorially a very small 
degree when the magnet was active, but this I attributed to a little oxygen still left 
with the nitrogen ; and indeed nitric oxide gas shows oxygen in nitrogen so prepared. 
The platina coil was raised to as high a temperature as it could well support without 
fusion, and yet there was only this small effect sensible ; hence I conclude that hot 
nitrogen is not more diamagnetic than cold nitrogen, and that indeed its magnetic 
relation is noways affected by such change of temperature. 
2859. I raised the French shade (2855.) an inch for a moment, and then instantly 
placed it down again ; and now, on making the magnet active and the coil hot, there 
was so much effect of dispersion of the gas within, that the melted spot of wax ap- 
peared nearly an inch outside of the standard place, yet only a very small portion of 
air or of oxygen could have entered the vessel under these circumstances. 
2860. The nitrogen of the air is therefore, as regards the magnetic force, a very 
