Dent's new Compensation Balance for Chronometers. 91 
piece 6 with the bar; and as the length of the dotted line /, 
drawn from the centre of gravity of the compensation-weight d, 
to the junction with the bar,—as the length of this line, I say, is 
augmented by increase of temperature, the compensation-weight 
d makes a quicker and nearer approach to the centre of motion 
than in the old compensation-balance ; whereas, on a decrease, 
the contrary takes place. 
Fig. 4, shows the plan of a balance, in which the primary and continuous secon- 
dary compensation is obtained by means of one curved piece on each side of the 
balance. 
g, the balance, made of brass or other 
non-magnetic metal. 
h, two blocks or studs raised above 
the face of the balance, to form the sup- 
ports of the compensation. 
k, l,m, two lamine, each curved in 
such a manner as to combine the joint 
effect produced by the primary and sec- 
ondary compensation-pieces shown at 
and ¢ of Fig. 3, the part from & to 1 
(Fig. 4) corresponding to the primary 
compensation-piece, and that from / to 
m corresponding to the secondary com- 
pensation-piece of Fig. 3. 
n, two prolongations from the ends of the compensation-pieces ; of steel only. 
p, two compensation-weights, screwed to the prolongations z. 
q, four timing-weights. 
Having thus fully explained the principle of my primary and 
secondary compensation, I would remark that my invention em- 
braces every modification of this principle, by which the com- 
pensation shall diminish the distance of the compensation-weights 
from the junction of the laminze with the arm upon a decrease 
of temperature, and produce the converse upon an increase ; which 
is the reverse of what has generally been done in the ordinary 
construction of the balance. 
In order to adjust this balance, as regards the secondary com- 
pensation, if the chronometer gains at the extremes of tempera- 
ture, compared with the mean, the secondary compensation-piece 
must be shortened and the time restored, by adding to the nuts 
at the ends of the bar; if the chronometer loses at the extremes 
of temperature, the reverse operations must be resorted to. ‘The 
ordinary adjustments for temperature are made by sliding the 
weights p along the prolongations 7. 
