TRANSACTIONS OF THE SECTIONS. 
19 
ing the brass and iron turnings, &c., and when the magnets have be¬ 
come loaded with iron it is swept off from them by frequent strokes 
of a brush. This is an exceedingly troublesome and inefficacious 
process. 
It appeared to the author that a temporary magnet of great power, 
formed by the circulation of an electric current round a bar of iron, 
might be substituted advantageously. The following is the arrange¬ 
ment which he has adopted. Several large round bars of iron are 
bent into the form of the capital letter U, each leg being about six 
inches long. They are all coated with coils of silk-covered wire, 
in the usual way of forming electro-magnets of such bars, and are 
then arranged vertically, at the interval of five or six inches from 
each other. 
All the wires from these coils are collected into one bundle at 
* 
their respective poles, and there joined into one by soldering, a large 
wire being placed in the midst of them and amalgamated. A gal¬ 
vanic battery is provided, which, if care be taken in making the 
junctions at the poles, &c., need not exceed four or at most six 
pairs of plates, of from twenty inches to two feet square. The poles 
of this terminate in cups of mercury, which are so placed that the 
large terminal wires of all the coils can be dipped into them, or 
withdrawn easily. 
The rest of the arrangement is purely mechanical. The required 
motions are taken from any first mover, usually a steam engine. 
The previously described arrangement being complete, a chain of 
buckets is so contrived as to carry up and discharge over the top of 
the magnets a quantity of the mixed metallic particles : most of the 
iron adheres to the magnets, while the so far purified brass falls into 
a dish or tray placed beneath to receive it. This latter is also one 
of a chain of dishes, the horizontal motion of which is so regulated 
that the interval between two dishes is immediately under the mag¬ 
nets, in the interval of time between two successive discharges of 
the mixed particles on the bars. 
At this juncture the communication between the galvanic battery 
and the magnets is interrupted by withdrawing the wires from the 
cups of mercury, and the result is, that the greatest part of the ad¬ 
hering iron drops off and falls in the space between the two dishes. 
The next dish now comes under the magnets, the communication is 
restored, and a fresh discharge from the buckets takes place, and 
so the process is continued. 
Some iron constantly adheres to the magnets, but this is found of 
no inconvenience as it bears but a small proportion to the total 
quantity separated. 
The author has had an imperfect apparatus of the sort above de¬ 
scribed at work for some time, and has found it to answer ; and 
suggests the application of electro-magnets for somewhat analogous 
objects in various manufactures. He particularly mentions needle 
and other dry grinding. 
