152 
DYNAMO-ELECTRIC GENERATORS. 
deuly turned off without altering the light emitted by the 
remaining one. 
These characteristic curves vary in form almost indefinitely, 
according to the machine used. 
Professor Thompson then pointed out that it was of great 
importance that the distribution of potential in the commutator 
should be known, and that the current should be led off at the 
point of maximum potential, returning to the minimum point. 
In the older machines it was constantly found that this could 
not be done, since the maximum shifted when the resistance of 
the circuit was altered. By means of a small coil moveable in 
a circle between magnet poles, Isenbeck has recently been able 
to plot out curves showing how the electromotive force produced 
by moving the coil varies in different parts of the field. In 
this way a knowledge of the phenomena which take place in 
such a case has been obtained, and this has been extended by 
Professor Thompson by integrating the curves over a series of 
coils such as we have in an ordinaiy armature. The influence 
of the material of the core, and of the form of the magnet pole, 
on the electromotive force has thus been tested, and it is found 
to be advantageous to have plenty of iron in the armature and 
to reduce the size and alter the shape of the cheeks of the 
inducing poles. Experiments similar to those of Isenbeck were 
performed by the lecturer, and applied by him to illustrate the 
advantage of iron in the armature. 
When external conditions are so arranged that the electro- 
motive force increases regularly in any given coil as it approaches 
the magnet pole, it is found that the distribution of potential in 
the commutator also becomes symmetrical, increasing regularly 
from a minimum to a maximum value, and the current can be 
led off at these points without difficulty. 
Professor Thompson then showed that these points did not 
occur at the points on the commutator nearest the ynagnet poles, 
