1895-96.] Dr X Halm on the Temperature of the Air. 263 
certain distance from the soil, the daily fluctuation of temperature 
disappears. Assuming the upper limit of that part of the atmos- 
phere which receives heat from the soil to extend as far as this 
critical distance, we naturally have the condition — - 
T - n A = u — const. 
So far, we have the following result : The total amount of heat 
given by the soil to the whole atmosphere by radiation (as well as 
by conduction ) can be considered as directed against two masses 
of air with the same coefficient of radiation , one of which masses 
has the temperature of the lowest layer , t, the other a constant 
temperature , u. 
It is important to state that this result is obtained without any 
assumption as to the distribution of temperatures within the 
atmosphere. Certainly, a heated element of air cannot remain in 
its position ; we know that heat is carried upwards by continual 
convection currents, which produce, on the other hand, a horizontal 
flow of air near the earth’s surface. But it seems quite evident 
from what we deduced before that this perpetual circulation of 
heat can only affect our result so far as it influences the tem- 
perature of the upper and lower limits of the part of the atmos- 
phere under consideration, — that is to say, the temperatures t and u. 
Now, however considerable the influence of the upward directed 
currents be on the temperatures of the single layers, there can be 
no doubt that at a comparatively small height they fade away, 
and that therefore we have not the slightest reason for altering 
our supposition as to the constancy of the temperature u. How, 
on the other hand, the horizontal currents affect the changes of 
the temperature t from one moment to another must be considered 
at a later part of this paper. 
At first sight, it would appear rather unreasonable to have two 
masses of air with different temperatures, without supposing that 
there exists radiation between these masses themselves. No doubt 
there will be. Let us suppose the mass of air at temperature t 
radiating heat against the mass of air at temperature u , — that is 
to say, contributing one part of that heat which is necessary to 
maintain the constancy of the temperature u. Consequently, the 
whole quantity of the required heat being necessarily the same in 
