in the length of the pendulum vibrating seconds. 341 
The pendulum was composed of a bar of plate brass 1,6 
inches wide, and rather less than the eighth of an inch thick. 
These dimensions were chosen that the pendulum and the 
thermometer placed near it, might be affected with equal 
readiness by any change of temperature. A flat circular 
weight nicely turned, and pierced in the direction of its dia- 
meter to receive the bar, was slid upon it, and fastened with 
screws and rivets at such a distance from the knife edge which 
served as the point of suspension, and which will presently 
be described, as that the pendulum made two vibrations less 
than the pendulum of the clock, in eight or nine minutes. 
The inside of the weight having been previously tinned, it 
was exposed to a sufficient degree of heat to solder it to 
the bar. 
That part of the bar which was below the weight, was 
reduced to the width of 0,7 inch, and covered with black 
varnish, in order to enable me the better to observe its coin- 
cidence with the pendulum of the clock, in the manner which 
has been fully described in the Philosophical Transactions 
for 1818, in an “ Account of Experiments on the length of 
the Pendulum vibrating seconds in the latitude of London/' 
With the contents of this Paper I shall suppose a previous 
acquaintance, as an occasional reference to it will save much 
repetition. 
To the top of the bar, a strong cross piece of brass was 
firmly rivetted and soldered, and a triangular hole having 
been made in the bar, a knife edge was passed through it, 
and a perfect contact between rhe back of the knife edge and 
the cross piece was insured by grinding them together. It 
was then secured in its place by two screws, the heads of 
