of the Lower Strata of the Atmosphere. 4] 
September, at seven o’clock in the evening, the increase of 
the temperature from the height of 5 feet to that of 108 feet, 
was 2°. At ten o'clock, the dew being very plentiful, the dif- 
ference did not exceed 1°6, even although the sky continued 
perfectly transparent. The increase of temperature as we 
ascend does not become more sensible at the time of the rising 
of the sun; on the contrary, most frequently the difference 
between the temperature of the stratum of air nearest the 
ground and of that situate at the height of 105 feet, appeared 
to me obviously less at the moment of the sun’s rising than at 
that of its setting. This is probably owing to the abundance 
of the dew which is known usually to increase at sunrise ; I 
have even remarked that a little before sunrise, and after a 
strong dew, the phenomenon was sometimes reversed ; that is 
to say, the temperature seemed to decrease in proportion to 
the elevation, particularly when the sky was suddenly over- 
east. In the latter case, the temperature of the earth is al- 
most always warmer than that of the ambient air. Such was 
the case on the 5th October at half past six o’clock in the 
morning, the weather being calm and cloudy, with much dew ; 
the thermometer on the grass indicated 12°31, at the height 
of 5 feet, 12°, and at 105 feet, 11°.7. The same thing took 
place on 7th October, at six o’clock in the morning, the ther- 
mometer at 5 feet was at 10°.1, and that at 1(5 feet, 9°.75, the 
temperature of the ground being about 10°.3. We shall after- 
wards see that, in the severe colds of winter, and when the 
weather is obscure, the temperature of the stratum of air 
next the ground is in general warmer than that of the at- 
mosphere at a height of 50 or 100 feet, both at the time of 
sunset and sunrise. 
3p Qurstion.—To determine the limit of elevation at which the in- 
crease of temperature ceases ; is this limit constant, or does it 
vary according to the state of the atmosphere ? 
When the sky was perfectly clear and serene at the time of 
sunset, it pretty often happened that the limit of the increase 
of temperature upwards was beyond the summit of the mast, 
that is to say, above the height of 108 feet. In general, how- 
ever, it appeared to me to be found between 90 and 105 feet, 
