of the current which produces motion must on the whole be opposed to the direction of the current produced by this motion. Thus in each armature wire the A.C. and D.C. flowing simultaneously neutralise each other to some extent. This neutralisation can obviously not be absolute, for one current is alternating, that is continually changing, and the other is continuous, that is of constant value. At times the one and at other times the other current predominates, but on the whole the difference between the two currents is smaller than either, and that is the reason why a given D.C. machine, if used as a converter, will give on its D.C. side nearly twice the output that can be taken from the same machine if worked as an ordinary D.C. generator.
The connections for the supply of the A.C. are tappings of the armature winding brought out to so-called "slip rings" S on which brushes bear. Since the same armature winding serves both the A.C. and the D.C. side of the machine, and since the e.m.f. induced in both is produced by the wires of the armature cutting through the same magnetic flux, it will be obvious that there must be a definite relation between the e.m.f. of the alternating and that of the continuous
