318 
electric circuit. As soon as the poise com- 
mences to move along the beam, this wheel, with 
its insulated spaces, commences to turn under 
its commutator-strip ; and with every passage 
of a tooth under the strip a current passes into 
the cylinder. Inside of the cylinder are two 
toothed wheels, mounted on the central shaft, 
and capable of being ratcheted round by means 
of a little lever arm and pawl, operated by a 
magnet placed directly under each of the wheels. 
SCIENCE. 
[Vou IIL, No. 68 
locked up in a way to be absolutely exterior to 
any control on the part of the operator. 
An old proverb has said that ‘the proof of 
the pudding is in the eating.’ In fig. 5 may 
be seen a half-dozen curves corresponding to 
as many test-pieces. The vertical scale gives 
the stresses, while the horizontal scale gives the 
normal stretches, of the pieces under exami- 
nation. ‘The two steel curves bear to each 
other a strong resemblance. Each commences 
. al i 
Att 
Fig. 4. 
One of these wheels is intended to drive the 
cylinder in one direction, and the other in a con- 
trary. One electro-magnet is connected with 
the mercury on the bottom of the beam, and 
the other in the mercury-cup on the top. As 
a consequence, as soon as the beam makes con- 
nection with either cup, the poise commences 
to travel: the corresponding electro-magnet 
acts, and rotates the cylinder in one way or the 
other. The cylinder might be placed in New 
York, and the registering-cylinder in Cincin- 
nati, and the two work absolutely in harmony 
with each other: so, should it ever be deemed 
expedient, the cylinder may be enclosed or 
with a line slightly inclined to the axis of 
stress at a constant tangent. As soon as 
the elastic limit is reached, a sudden point of 
inflection occurs. Very soon, however, there 
is a second point of inflection, and the curve 
takes on a parabolic form. The steel curves, 
as well as that given by the specimen of Ulster 
iron, may be taken to be typical forms obtained 
from the material which is nearly homogeneous. 
The lines are quite true, and without any spe- 
cial irregularities, until the maximum stress is 
reached, shortly before the specimen breaks. 
The stress then commences to decrease, owing 
to the rapid reduction of area of the piece; 
