492 Proceedings of the Royal Society 
in fig. 7) cut out of the same tube and kept from touching, by being 
fixed to a vulcanite ferrule ( v in fig. 6, which shows the inside of 
half the collar) placed inside and between them. The ends of the 
iron pieces slide on the iron ferrules beneath, and are in conducting 
connection with them. Electrical contact is made by springs press- 
ing on this composite collar, and which are metallically connected 
with the binding screws, the poles of the armature coils. The collar 
and springs at each end form the breaks or commutating arrange- 
ment of their respective coils. The cross line of separation (e efig. 
4) can he fixed in any position, and currents in one or different 
directions thereby obtained in the course of a revolution. The 
pressure of the springs against the collars is regulated by screws. 
Fig. 6. Fig. 7. 
When the machine is prepared for working, the cross lines of the 
commutating collar of the magnetic coil are placed at right angles 
to the plane of the coil, the position of maximum effect. If the 
handle of the machine be turned when the circuit of the electric 
coil is open, one or two turns bring the hand of the operator to 
something like a dead halt ; the resistance to further motion 
is so great as to challenge its continuance. If, now, the 
external circuit be closed, immediate relief is felt, as if part of 
the internal current had been diverted into the external circuit 
from the coils of the electro-magnet. The relief thus experienced, 
moreover, bears some proportion to the conductivity of the external 
circuit. With an easy circuit, the work expended in turning the 
handle is easy ; with a resisting circuit, the driving resistance 
becomes correspondingly great. The hand is thus made to sym- 
pathise with the nature of the external circuit, and the experi- 
menter feels as if he were charged mechanically with a resistance 
offered electrically. Suppose, for instance, we have a piece of thin 
wire to heat or melt ; at first little or no driving resistance is felt, 
but the moment that the wire begins to get hot, the arm becomes 
charged with a heavy resistance, which grows as the wire rises in 
temperature till it melts, and then suddenly the excessive no-circuit 
