Natural Philosophy. — Meteorology , 397 
at 96° and 45°, as deduced from several years observations. 
Mr Colebrook has given the following additional results : 
Mean. Temp. Extremes. 
Stellenbosch, - 66°.33 87° and 50° 
Zwartland, - 66 . 5 85 and 54* 
Tulbugh, 66 .66 95 and 52 
The observed mean temperature of Cape Town agrees in a very 
remarkable manner with Dr Brewster’s Formula of Mean Tem- 
perature — 81 1 cos lat. ; for in the present case we have 
81* cos 33° 55' 1 5* - - - - = 67°.6 
Observed mean temperature, - - 67 *3 
Difference 0°.3 
12. Variation in the Bulbs of Thermometers. — In mercurial 
thermometers with a perfect vacuum above the mercury, M, 
Flauguergues has observed, that the freezing point has gradu- 
ally risen nine-tenths of a degree, and has gone on increasing 
for years. He attributes this to a permanent change of form,, 
produced by the constant pressure of the atmosphere on the 
bulb. He therefore recommends, that thermometers should be 
made with open terminations. The same fact had been long 
before observed by M. Angelo Bellani of Milan, who mentions 
a very pretty experiment, for shewing that it arises from the 
glass. Take a mercurial thermometer, which has not been ex- 
posed for some months to temperatures near that of boiling 
water, whose scale rises to the boiling point or higher, and whose 
degrees are at least a line long, so that tenths of a degree can be 
easily seen. Having carefully marked the freezing point, plunge 
it in boiling water, and, upon replacing it in melting ice, it will 
be found, that the freezing point has sunk ^th of a degree, in 
consequence of the expanded glass not having resumed accurate- 
ly its original form. Hence, it follows, that the new barometer 
of the Rev. Mr Wollaston, must be liable to error from this 
cause. — See Bibl. Univers vol. xx. p. 117., and vol. xxi. p. 252. 
13. On the cause of the Ascent of Clouds in the Atmosphere . — ■ 
M. Fresnel has ascribed the ascent of clouds in the atmosphere 
principally to the following cause. Air and all colourless gases 
allow the solar rays, and even radiant heat, to pass through them, 
without heating them sensibly. When a cloud, therefore, is com- 
posed of small globules of water, or minute crystals of snow, the 
