122 Prof. Barlow’s Account of Experiments on the Magnetism 
change remained the same ; and that the effect was independent 
of the direction of the axis of motion, viz. whether the axis was 
east and west, north and south, or in any other azimuth ; but it 
required a certain velocity, not less than 600 revolutions per mi- 
nute, to produce the full effect. It is obvious, therefore, that 
the mere rotation of an iron-shell, impresses upon it, during the 
motion of the same, a temporary magnetic effect ; and that this 
effect ceases the moment the motion is discontinued. 
The above experiments were begun in December 1824 ; and 
it was not till April 1825, that Mr Barlow learned that M. 
Arago had been making similar experiments in France, on cop- 
per and other metals ; but I am not aware of their actual date. 
They were not known in England, till M. Gay Lussac’s visit to 
London at the time above stated. I am not aware of the precise 
nature of these experiments ; and shall, therefore, only endea- 
vour to describe those which I have assisted Mr Barlow in 
making, and which he founded on the description he had re- 
ceived. They may, therefore, be considered as the experiments 
of M. Arago repeated, and varied as different circumstances oc- 
curred to suggest new ideas. The account he had of M. Arago’s 
experiment, was that, by placing a copper plate upon a vertical 
spindle, the plate being horizontal, and then placing just above 
it a light compass needle, but independent, of course, of the 
plate ; on causing the spindle and plate to revolve, the needle 
was considerably deflected, and more and more as the velocity 
was increased ; so that, when the plate was put into rapid rota- 
tion, the needle also began, after a few vibrations, to revolve, and 
at length with considerable velocity. 
1. In order to repeat this experiment, 1 connected the 
wheel of my turning lathe with a vertical spindle, which I 
could make revolve forty-five times per second ; and on this 
I placed a thin copper plate, about six inches in diameter, 
and over this a needle about five inches long, shut up in 
a close box, about one inch, or rather less, above the plate. 
When putting the lathe in motion, I found it to deflect the 
needle about five points ; the deflection being always in the 
same direction as the motion of the plate, but we could not cause 
it to revolve. The needle was, therefore, partly neutralized by 
a bar magnet, and the experiment repeated. We, then, very 
soon obtained a considerable rotatory motion in the needle ; and. 
