Sutron—Influence of Water- Vapour upon Nocturnal Radiation. 29 
and at great altitudes, the air is relatively cold at night; but in 
the middle reaches, it is relatively warm. We may assume that 
velocities up to five or six miles per hour can do no more than 
bring down the air from the middle reaches to mix with the lower 
air, but that greater velocities may bring down the air from 
considerable heights. Heating due to adiabatic contraction would 
doubtless largely neutralize this process. 
Some results obtained in 1896 and 1897 gave a somewhat 
higher wind-velocity corresponding to the slowest fall of tempera- 
ture, namely, a fall of 3° in two hours, from 8-10 p.m., fora 
velocity lying somewhere between 64 and 8 miles per hour; but 
otherwise the results agree quite well with those of the later period 
1900 to 1903. I attempted to use these wind-results for the 
purpose of applying corrections to the observed fall of temperature 
so as to obtain better values for the dew-point and humidity factors, 
but without any particular improvement worth notice. The 
following are the results :-— 
Dew-Point Fall of Number Relative Fall of Number 
at Temperature, of Humidity at |Temperature, of 
8 p.m. 8-10 p.m. | Clear Nights. 8 p.m. 8-10 p.m. | Clear Nights. 
Under 30° 3°8 68 Under 35 °/, 5-1 41 
30°-34°-9 3°6 98 Ba 9/39... 45 31 
35°-39°-9 3-8 68 A044. 4:3 37 
40°-44°-9 3°8 33 45 ,,-49 ,, 3°7 46 
45° and over. 3°3 29 50 ,,-54 ,, 374 51 
55 ,,-09 ,, 2°8 36 
60 ,,-64 ,, 2°6 28 
65 ,, & over. 2°3 26 
Therefore, even after correcting for wind-velocity no essential 
difference is introduced; and, as before, the dew-points are 
uninfluential. 
It is important to determine whether, for a given humidity- 
percentage, the fall of temperature depends at all upon the 
temperature itself. The following table shows that it does. It 
SCIENT., PROC. R.D.S8., VOL. XI., NO. III. aD} 
