FEBRUARY 15, 1884.] 
the couple, got by multiplying the weight so 
applied by its distance from the axis aa as 
read upon the scale-beam, is evidently equal 
and opposite to the moment of the couple with 
which the armature is turned. Thus the mo- 
ment of the force applied to run the armature 
was measured in foot-pounds. This moment, 
multiplied by the number of revolutions per 
minute and by 27, is the work expended in 
driving the armature, in foot-pounds per min- 
ute. 
It was found convenient to use a weight hav- 
ing a moment somewhat less than sufficient to 
bring the scale-beam to the horizontal, and 
employ a Chattillon spring-balance with a dial- 
face reading to 2 oz. to furnish the remaining 
part of the couple. This balance was fastened 
to a vertical cord passing around a small winch, 
which enabled the observer to bring the scale- 
beam to the horizontal with facility. 
The principal difficulty to be apprehended 
in testing with this dynamometer was a pos- 
sible tendency to oscillation, which might be 
caused by running the belt. Had this existed, 
it might have been checked by a dash-pot; 
but the weight of the platform and dynamo 
was sufficient to almost entirely obviate any 
such difficulty, and give very considerable 
steadiness of position. Indeed, the jarring 
seemed to increase its sensitiveness, and evi- 
dently enabled the platform to come to rest in 
its position of equilibrium by overcoming any 
initial friction existing. 
The structure was mostly built of three-inch 
plank, ten or twelve inches wide. The plat- 
form was designed to safely carry five tons at 
its centre. 
For permission to make use of the cradle 
dynamometer, the jury is indebted to the kind- 
ness of Professor Brackett. 
In addition to the use of the two dyna- 
mometers described, indicator-diagrams were 
taken from the steam-engine furnishing the 
power for driving the dynamo machines. 
Throughout the dynamometric tests the ma- 
chines were driven by a Cummer engine of 
about a hundred horse-power, which furnished 
power for other dynamos than that upon the 
cradle, as well as for two or three pieces of 
machinery on exhibition in Power Hall. It 
was thus impossible to make the taking of 
indicator-diagrams contemporaneous with the 
regular tests, owing to the large load which 
the engine carried; and they were generally 
taken after ten o’clock at night, at which hour 
the remainder of the load was thrown off. 
Although these indicator-cards were not used 
in the final computations, they furnished valu- 
SCIENCE. 
a 
able checks upon the performance of the dyna- 
mometers. 
The electrical measurements. 
For the purpose of making the electrical 
measurements as free from disturbance as pos- 
sible, a small room, about twenty feet long and 
ten feet wide, was fitted up in the basement of 
the central part of the large exposition build- 
ing. In this, three brick piers were built upon 
solid foundations, and two or three wooden 
brackets were firmly secured to the walls, so 
as to furnish firm resting-places for the galva- 
nometers. The main lines of the arc-lighting 
systems were run through this room, and very 
heavy copper conductors connected the room 
with the space in which the dynamos were ex- 
hibited ; so that the entire current from the in- 
candescent machines could be introduced when 
desired. 
The electric measurements consisted in the 
determination of the strength of the current, 
and the electromotive force between two points 
in the circuit. For this purpose several galva- 
nometers of different kinds were employed. 
For the measurement of current strength the 
principal instrument used was one of Sir Wil- 
liam Thomson’s current galvanometers, made 
by White of Glasgow. Although of recent 
invention and construction, the instrument is 
probably so well known as not to require any 
detailed description. It consists essentially 
of a magnetometer, and a coil of very low re- 
sistance. In the magnetometer four short 
magnets are combined to form a needle, the 
position of which is indicated by a very light 
yet very rigid aluminum index. A steel mag- 
net, bent in the shape of a semicircle, is placed 
in a vertical plane over the needle, so that 
the latter is approximately at the centre of the 
circle of which the magnet forms a part. One 
end of the magnet is furnished with a cross- 
piece of brass, from one extremity of which 
projects a pin which rests in a conical hole, 
and upon the other extremity is a ‘button ;’ so 
that freedom of motion around the pin as an 
axis is allowed. The opposite end of the 
magnet rests in a groove cut around the end 
of a screw, by the movement of which the 
plane of the magnet can be shifted towards 
the east or towards the west. 
The coil is fixed in a vertical plane at one 
end of a wooden table whose length is about 
one foot, and breadth about five inches. The 
table is furnished with levelling-screws, and a 
V-groove is cut lengthwise through the centre, 
at right angles to the plane of the coil. In 
