[ 99 ] 
VI. On the probable Electric Origin of all the Phenomena of Terrestrial Mag- 
netism ; with an illustrative Experiment. By Peter Barlow, Esq. F.R.S. Cor. 
Mem. Inst, of France, of the Imperial Academy of St. Petersburgh, 8$c. 8$c. 
Read January 27, 1831. 
In order to show the incompatibility of the observed laws of terrestrial mag- 
netism with the supposition of the earth itself being a magnet, and at the same 
time their accordance with the laws which appertain to a body whose mag- 
netism is induced by electricity, it will be necessary to take a retrospective 
view of the several discoveries which have been made, connected with these 
subjects, since the commencement of the present century, and particularly 
within the last ten or twelve years. Up to the period of the scientific travels 
of M. Humboldt, it must be admitted that all the facts with which we were 
acquainted, relative to terrestrial magnetism, were a mere collection of obser- 
vations and phenomena, uncertain in themselves, unconnected with each other, 
and irreducible to any specific laws ; but the confidence inspired by the high 
character of this distinguished traveller, his accuracy of observation, and the 
perfection of his instruments, gave a new feature to the inquiry, and laid the 
foundation of our present knowledge in this science. 
M. Biot was the first to undertake the difficult task of reducing these results 
to some principle of calculation. Considering the earth as a magnet, he 
assumed an indeterminate distance to represent the distance of its two poles ; 
and then, supposing their power to vary inversely as the square of their distance 
from the point on which they acted, (a law which had been already established,) 
he obtained a general expression for the direction of a magnetic needle. He 
then made his indeterminate distance vary ; and comparing at every step his 
results with those observed, it was found that the nearer the poles were made 
to approach, the nearer the computed and observed results corresponded ; and 
finally, that the errors were reduced to a minimum when the two poles were 
coincident, or indefinitely near to each other. 
o 2 
