50 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXVI.) 
through them. The difference in air under different conditions would be very small, 
still it is that difference which concerns us in atmospheric magnetism ; and it is very 
important to know, whether, when the magnet indicates an increased intensity of 
force, it is altogether due to a real increase of the amount of the power at its source 
as it comes to us from the earth, or in part to a change in the magnetic constitution 
of the space around the magnet hitherto unknown to us. 
2870. If what is now often indifferently called magnetic force or intensity have its 
results distinguished as of two kinds, namely, those of quantity and those of tension, 
then we shall more readily comprehend this matter. At present a needle shows both 
these as magnetic force, making no distinction between them, yet they produce effects 
on it often in opposite directions; for as they increase or diminish they both affect 
the needle alike ; but as it is assumed that the tension can change whilst the quan- 
tity remains the same, and the quantity can be altered, yet the tension remain un- 
affected, the result by the needle will then be uncertain. If the tension in a given 
region be increased by diminishing the conducting power, the needle will show in- 
creased force if it be increased by an increase of magnetic power in the earth from 
some internal action, the needle will still show increased force, and will not distin- 
guish the one effect from the other. If the quantity in a region be increased by in- 
creasing the conducting power, the needle will show no such increase ; on the con- 
trary, it will indicate diminution of force, because the tension is diminished ; or if the 
quantity be diminished by diminishing the conducting power, it will show increased 
force. The force might even lose in quantity and gain in tension in such proportions 
that the needle should show no change ; or it might gain in quantity and lose in 
tension, and the needle still be entirely indifferent to the whole result. 
2871. If my view be correct, then the magnet is not, as at present applied, a per- 
fect measure of the earth’s magnetic force ; for that may not change when the magnet 
by the influenee of the different conditions of day and night, or of summer and winter, 
may show a difference. How far these uncertainties in its indication may affect the 
value of the observations made on the horizontal and vertical components of the 
earth’s magnetic force as indications of that which they are expected to tell us, I do not 
know ; but involving, as the effects do, two very different conditions, namely, variation 
of the conducting power and variation of the amount of force at its source, the one 
of whieh is chiefly in the atmosphere and the other in the earth, it seems to me to 
be of great consequence to the development of the theory of terrestrial magnetism, 
to have some method, if possible, of distinguishing these two points or effects from 
each other. 
2872. Referring again to the model globe, fig. 7 (2874.), it appears to me, that if a 
magnet be used as the intensity test, it will indicate a less intensity at P rather than 
a greater one, for the very reason that the conducting power of the whole globe has 
been increased ; and also, though the apparent diminution of intensity will probably 
be greater there than elsewhere, that the effeet will occur in other parts, espe- 
