RADIATION AND ABSORPTION. 77 
thermometer faced successively extents corresponding to one 
quarter hemisphere, one-third hemisphere, and two-thirds hemi- 
sphere; in each position I awaited the equilibrium of tem- 
perature, and noted at the same time the temperature of the 
ambient air and that of the apparatus. Analogous experiments, 
repeated at the temperature of melting ice and at other inter- 
mediate temperatures, have led me to the following result :— 
If from the ambient temperature nine-fourths of the lowering 
of the actinometer be taken away, we always find the tempera- 
ture of the artificial sky. This result evidently applies to the 
celestial vault, or rather to the zenithal inclosure ; consequently, 
if we observe during the night the temperature ¢ of the ambient 
air, and the lowering d of the actinometer, we shall thence 
deduce the zenithal temperature by the formula 
2=t—9 . 
which is the result of the graduation. 
21. Further on will be found a table containing some of the 
series of experiments which have been made during very beau- 
tiful nights, and in calm weather, to determine the zenithal tem- 
perature. These experiments confirm the fact that the zenithal 
temperature is lowered during the night, nearly as the tempera- 
ture of the ambient air; this progressive lowering, from the 
setting to the rising of the sun, is an essential fact which leads 
immediately to an important consequence. 
In fact, we have seen that the zenithal temperature is ex- 
pressed by the sum of two terms; one depending on the mean 
temperature of the atmospheric column, which is variable, and 
the other depending on the temperature of space, which is 
fixed. Now since the zenithal temperature experiences, in a 
single night, considerable variations, this is an evident proof 
that the fixed term which enters into its expression has only a 
very small value with relation to the variable term, and conse- 
quently that, in the nocturnal radiation, the heat of space is very 
small in relation to the heat which is derived from the radiation 
of the atmosphere. 
This consequence can scarcely be reconciled with the opinions 
which attribute to space a temperature the value of which 
would not be lowered beneath zero a very great number of de- 
grees; but it is perfectly reconcileable with the known facts 
which already might have furnished indications in this direction, 
had they all been analysed with the attention which they merit. 
