ob,’ ei 
TRANSACTIONS OF THE SECTIONS. 19 
to'the Academy of Sciences, of which the title alone is mentioned in the ‘ Comptes 
Reudus.’ ‘This is the induction which magnetization must produce at the moment 
of the closure of the circuit ; as soon as the current traverses the coils of a machine, 
the iron nucleus which they contain becomes magnetized, and this magnetization 
must produce an energetic current of induction in a direction opposed to that of 
the primitive current. When the circuit is interrupted, the demagnetization tends 
to produce a direct current of induction ; but it cannot propagate itself, because the 
circuit is interrupted. The two effects therefore do not compensate one another. 
However, I shall refer to these experiments, which show very clearly the necessity 
of taking this element into account. 
1. A small electro-magnetic machine made by Froment was employed. It con- 
sists of a wheel furnished with six pieces of soft iron passing successively before the 
poles of three horseshoe electro-magnets, ‘The axle which bears this wheel, also 
bears the commutator. The wheel and the pieces of soft iron were removed, 
leaving the axle andthe commutator in their places. Of course the machine 
could no longer progress of itself, but the axle could be made to rotate, by trans- 
mitting to it in a suitable manner the motion of a wheel turned by a handle. There 
were then produced in the electro-magnets, alternations of magnetization and de- 
Magnetization similar to those which take place when the machine is in motion 
under the influence of the current. When-the current is passed without the axle 
being set in rotation, the deviation of the galvanometer was 48°; then on setting the 
axle in motion so as to make 408 turns in a minute, the deviation fell to about 30°; 
when the rapidity was increased, the deviation diminished still more. 
Thus without the production of any mechanical work by the machine, and with- 
out the development of those counter-currents arising from the approximation of 
pieces by mutual attraction, we see that the current is considerably diminished. 
This result appears to me to be perfectly incompatible with the calculations of M. 
Jacobi. 
2. A circuit was formed, consisting of a battery, a watch-work interrupter, and 
a coil into which a cylinder of soft iron might be introduced at pleasure. A mag- 
netized needle placed below a portion of the current deviated 32° when the iron 
cylinder was not put into the interior of the coil; but as soon as it was introduced, 
the deviation fell to 25°. 
3. By means of a battery motion was given to an electro-magnetic machine, the 
movement of which was produced by the magnetization and demagnetization of a 
single electro-magnet, in such a way that the current was interrupted in the whole 
circuit during the period of demagnetization. The circuit also included a coil of con- 
siderable resistance ; the moment a cylinder of soft iron was introduced into this 
coil, the motion of the machine was notably retarded. 
These three experiments show in a striking manner the loss of force which is 
caused by the employment of non-continuous currents in electro-magnetic machines. 
_ There is in this case an analogy with ordinary steam engines, in which a great part 
of the force which might be produced by the heat is uselessly employed. Better 
results in this respect would doubtless be obtained, if we could succeed in con- 
_ structing machines of a certain force, founded upon the principle of the rotation of 
currents by currents, or vice versd. But hitherto these machines have had too little 
power to allow of their being employed as motors. 
Szconp Memoir.—On the heat evolved by the current in a conductor, arranged so as 
to produce exterior work. 
_ I have already stated that when a circuit produces exterior work, a change must 
_be produced in the interior work, that, for example, there must be a diminution in 
_ the heat evolved either in the part of the circuit which effects the exterior action by 
; induction, or in the battery itself. The former of these suppositions has been in- 
_ vestigated in this memoir. 
The question, then, is to know, if a coil, for example, when traversed by a current 
undergoes the same elevation of temperature when it exerts no exterior action, and 
oo when it exerts one, such as the alternations of magnetization and demagnetization 
_ Which it produces upon a nucleus of soft iron when the current is not continuous. 
The method adopted consists in arranging, in the same circuit, two coils, each 
* 
