April 19, 1883] 
NATURE 
593 
reverse, it was possible to obtain a current constant always in 
direction ; but such a current would be far from constant in 
intensity, and would certainly not accomplish all the results ob- 
tained in modern continuous-current machines. This irregularity 
might, however, be reduced to any extent by multiplying the 
wires of the armature, giving each its own connection to the 
outer circuit, and so placing them that the electromotive force 
attained a maximum successively in the several coils. A prac- 
tically uniform electric current was first commercially produced 
with the ring armature of Pacinotti, as perfected by Gramme. 
A dynamo-machine was not a perfect instrument for converting 
mechanical energy into the energy of electric current. Certain 
losses inevitably occurred. There was the loss due to friction 
of bearings, and of the collecting-brushes upon the commutator ; 
there was also the loss due to the production of electric currents 
in the iron of the machine. When these were accounted for, 
there remained the actual electrical effect of the machine in the 
conducting wire; but all of this was not available for external 
work. The current had to circulate through the armature, which 
inevitably had electrical resistance ; electrical energy must there- 
fore be converted into heat in the armature of the machine. 
Energy must also be expended in the wire of the electromagnet 
which produced the field, as the resistance of this also could not 
be reduced beyond a certain limit, The loss by the resistance of 
the wires of the armature and of the magnets greatly depended 
on the dimensions of the machine. To know the properties of 
any machine thoroughly, it was not enough to know its efficiency 
and the amount of work it was capable of doing; it was neces- 
sary to know what it would do under all circumstances of varying 
resi tance or varying electromotive force ; and, under any given 
conditions, what would be the electromotive force of the arma- 
ture? Now this electromotive force depended on the intensity 
of the magnetic field, and the intensity of the magnetic field 
depended on the current passing round th: electro nagnet and 
the current in the armature. The current then in the machine 
was the proper independent variable in terms of which to express 
the electromotive force, The simplest case was that of the 
series-dynamo, in which the current in the electromagnet and in 
the ar uature was the same, for then there was only one inde- 
pendent variable. The relation between electromotive force and 
current might be most conveniently expressed by a curve. 
Whe i four years ago the lecturer first used such a curve (since 
named by Deprez the ‘‘ characteristic curve”) for the purpose of 
expressing the results of his experiments on the Siemens 
dynamo-machine, he pointed out that it was capable of solving 
almost any problem relating to a particular machine, and that it 
was also capable of giving good indications of the results of 
changes in the winding of the magnets, or of the armatures of 
such machines The use of the characteristic curve wa; illus- 
trated with reference to charging accumulators and Jac »bi’s law 
of electric transmission of power. : 
When the dynamo-machine was not a series-dynamo, but the 
current in the armature and in the electromagnet, though pos- 
sibly dependent upon each other were not necessarily equal, the 
problem was not so simple. In that case there were two vari- 
ables, the current in the electromagnet and the current in the 
armature ; and the proper representation of the properties of 
the machine would be by a characteristic surface, of which a 
model was exhibited. By the aid of such a surface any problem 
relating to a dynamo-machine could be dealt with, no matter 
how its electromagnets and its armature were connected to- 
gether. Of course in actual practice the model of the surface 
would not be used, but the projections of its sections. 
The properties of a machine depended much upon its dimen- 
sions. Suppose two machines alike in every particular, except- 
ing that the one had all its linear dimensions double that of the 
other. The electrical resistances in the larger machine would be 
one-half those of the smaller. The current required to produce 
a given intensity of magnetic field would be twice as great in the 
larger machine as in the smaller. The comparative characteristic 
curves of the two machines when driven at the same speed 
were shown ina diagram. ‘The two curves were one the pro- 
jection of the other, having corresponding points with abscisses 
in the ratio of one to two, and the ordinates in the ratio of one 
to four. At first sight it would seem that the work done by 
the larger machine should be thirty-two times as much as that 
which would be done by the smaller, Practically, however, 
no such result could possibly be attained for many reasons. 
First, the iron of the magnets became saturated, and conse- 
quently, instead of eight times the electromotive force, there 
would only be four times the electromotive force. Secondly, the 
current which the armature could carry was limited by the rate 
at which the heat generated in the armature could escape. 
Again, the larger machine could not run at so great an angular 
velocity as the smaller one. And lastly, since in the larger 
machine the current in the armature was greater in proportion 
to the saturated magnetic field than in the smaller one, the dis- 
placement of the point of contact of the brushes with the com- 
mutator would be greater. Shortly, the capacity of similar 
dynamo-machines was pretty nearly proportionate to their weight, 
that was to the cube of their linear dimensions ; the work wasted 
in producing the magnetic field was directly as the linear dimen- 
sions; and the work wasted in heatinz the wires of the armature 
was as the square of the linear dimensions. 
A consideration of the properties of similar machines had 
another important practical use. Mr. Froude was able to con- 
trol the design of ironclad ships by experiments upo models 
made in paraffin wax. It was a much easier thing t> predict 
what the performance of a large dynamo-machine would be, 
from laboratory experiments made upon a model of a very small 
fraction of its dimensions. As a proof of the practical utility of 
such methods, the lecturer stated that by laboratory experi- 
ments he had succeeded in grea'ly increasing the capacity of the 
Edison machines without increasing their cost, and with a small 
increase of their percentage of efficiency, remarkably high as 
that efficiency already was. 
The electric properties of the electric arc were experimentally 
illustrated ; in particular it was shown that the difference of 
potential between the carbons was nearly indepentent of the 
current, 
When a current of electricity passed through a continuous 
conductor it encountered resistance, and heat was generated, as 
shown by Joule, at a rate represented by the resistance multiplied 
by the square of the current. If the current was sufficiently 
great, heat would be generated at such a rate that the conductor 
would become incandesgent and radiate light. Attempts had 
been made to use platinum and platinum iridium as the incan- 
descent conductor. But these bodies were too expensive for 
general use, and besides that, refractory thouzh they were, they 
were not refractory enough to stand the high temperature 
required for incandescent lighting, which should be economical 
of power. Commercial success was not realised until very thin 
and very uniform threads or filaments of carbon were produced 
and inclosed in reservoirs of glass, from which the air was 
exhausted to the ntmost possible limit. Such were the lamps 
made by Mr. Edison with which the Institution was temporarily 
lighted. The electrical properties of such a lamp were exa- 
mined, and in particular it was shown that its efficiency 
increased and its resistance diminished with increase of current. 
The building was lighted by about 230 lamps, each giving 
sixteen candles light, produced each by 75 Watts of power 
developed in the lamp. To produce the same sixteen candles’ 
light in ordinary good flat-flame gas-burners, would require — 
between 7 and 8 cubic feet of gas per hour, contributing heat 
to the atmosphere at the rate of 3,409,090 foot-pounds per 
hour, equivalent to 1250 Watts, or nearly seventeen times as 
much heat as the incandescence lamp of equal power. 
At the present time, lighting by electricity in London must 
cost something more than lighting by gas. What were the 
prospects of reduction of this cost? Beginning with the engine 
and boiler, the electrician had no right to look forward to any 
marked and exceptional advance in their economy. Next came 
the dynamo, the best of these were so good that there was little 
room for econo ny in the conversion of mechanical into electrical 
energy; but the prime cost of the dynam -machine was sure 
to be greatly reduced. Hope of considerably increased economy 
must be maiily based upon probable improvements in the 
incandescence lamp, and to this the greatest attention ought to 
be directed. It had been shown that marked economy of power 
could be obtained by working the lamps at high pressure, but 
then they soon broke down. In ordinary practice, from 140 to 
200 candles were obtained from 1 horse-power, developed in the 
lamps, but for a short time he had seen over 1000 candles per 
horse-power from incandescence lamps. The problem, then, was 
so to improve the lamp in details, that it would last a reasonable 
time when pressed to that degree of efficiency, There was no 
theoretical bar to such improvements, and it must be remem- 
bered that incandescence lamps had only been articles of com- 
merce for about three years, and already much had been done. 
If such an improvement were realised, it would mean that it 
