Meteorological Obfervations. 567 
\ diculatly to the folar rays, and had a metal- 
_~ 
‘lic fcale. 
Jan. 12th. 1792, at 8 A. M. - - - 19. 
Aug. 8th. ce ars UL - - - 74. 
1793. Lowett - - - - 28. 
Higheft - ? - - 78. 
As the preceding obfervations were made, 
I believe, with common mercurial thermo- 
meters, the mean annual heat of each place 
cannot be determined accurately from them. 
As the greateft degree of cold within the twenty- 
four hours, which occurs about half an hour 
before fun-rife, has feldom been obferved, 
the mean heat deduced from thefe obferva- 
tions, will be fome degrees higher than 
the true mean.— The beft thermometers for 
obfervations of this nature, are thofe invented 
by Mr. Six, which point out the greateft and 
leaft degrees of heat in the obferver’s abfence. 
The following account of very great degrees 
of cold, which were obferved at Chatham in 
January 1776, with the height of the baro- 
meter and direction of the winds at that time, 
with fome other ,obfervations, was communi- 
cated to me by Dr. Percival. 
Height 
