482 ‘ON IMPRESSIONS OF COLD 
instrument might amount to twenty degrees ; but when the 
congregated vapours sunk so low as to hover on the hilly tracts, 
the impression would frequently not exceed five degrees. It 
was evident, therefore, that the effect depends on the altitude 
of the lowest range of clouds, and might seem to result wholly 
from the difference of temperature which prevails there, com- 
pared with that of the surface. 
But the same conclusion was drawn from another set of ob- 
servations. In a calm day, when a mass of dark clouds was 
spread at no great elevation above the surface of the ground, 
the spheroid indicated only five millesimal degrees in a verti- 
cal position, and yet marked still the same quantity, when de- 
pressed to an angle of thirty degrees above the horizon. But 
had this impression of five degrees penetrated directly through 
the clouds, from the higher regions of the atmosphere, scarce- 
ly one-half of a degree could have escaped through the mass 
of vapours by the oblique passage. A range of clouds hence 
acts as a complete screen, absorbing and extinguishing all the 
hot or cold pulses received on it. 
Since clouds consist merely of dispersed aqueous globules, 
their influence, I conceive, may be safely inferred from that of 
water itself. I therefore inclosed an inverted pyroscope in a 
spheroidal cup, and suspended it a few feet above the ground, 
while the sky appeared clear and blue, (see fig. 7. Pl. XL.) ; 
then passing a silver tray under it, the reflected impression of 
cold amounted to twenty-five degrees: on interposing a plate 
of glass, this was reduced to two degrees ; but on removing 
it, and pouring a sheet of water over the silver, the effect was 
absolutely extinguished. 
Most of these observations and experiments were made du- 
ring the months of September and October 1817 on the top of 
the Tower at Raith,—a spot very favourable for such researches, 
and 
