[ s6i 3 
In the middle of February, when the firft change 
was, the froft was intenfe ; and the pendulum did 
not, for fome days, throw-out fo far by about 7' as 
it generally did ; which was about 1° 37' on one 
fide, and i® 40' on the other. At the change in 
Auguft, I obferved no difference. It appears by 
thefe trials as if the clock gained in warm and loft 
in cooler weather : but this is not clear. It began 
to gain before the weather grew warm. Whether 
this be owing to damp, or any other caufes i longer 
experience and abler obfervers may difcover. 
[ VoL. LXI. 4 C' 1770 
