59 
1918--19.] The Origin of Anticyclones and Depressions. 
Thus variations of diathermancy, like adiabatic changes of volume, have a 
double effect, altering both the rate of absorption and the rate of radiation 
in such a way that the two results co-operate in increasing or diminishing 
the entropy of the mass. 
The rate of loss of heat by radiation is affected also by the temperature 
of the mass, and by adjacent clouds or other reflecting screens. 
The effects of* conduction between the lowest layers of the atmosphere 
and the underlying terrestrial surfaces have already been discussed 
ad nauseam. The result of steady secular heating from below, such as 
occurs in the tropics, and of steady cooling from below, as in glacial 
“ anticyclones,” will be dealt with later. Such alternate heating and cooling 
as occurs elsewhere is principally productive of fogs or of convection eddies ; 
the effect on the wind systems may be neglected. As regards the effects of 
conduction in portions of the atmosphere at higher altitudes, thpy are slow, 
but cumulative. Aided by all the small turbulent movements of the air, 
from the Brownian motion to the eddies of a waterspout, conduction tends, 
where vertical churning is absent, to the production of isothermal conditions, 
and where it is present, to the production of isentropic conditions. 
Considerable additions or subtractions of heat may occur within a mass 
of air, in virtue of the condensation or evaporation of moisture or ice. The 
effect of cloud in a fairly calm layer of air may be roughly compared to the 
effect of ice particles scattered through water. It tends to diminish the 
temperature changes which accompany changes of entropy. But whereas 
the presence of ice in water absolutely fixes the temperature (in the ideal 
case of perfect stirring), cloud only diminishes changes of temperature. It 
cannot annihilate them. For if the whole of the absorbed heat could be 
employed in evaporating a portion of the cloud, then this evaporation would 
increase the pressure of the aqueous vapour in the clouded mass. But the 
saturation pressure was already reached, or there would not have been cloud 
to begin with. Hence only part of the added heat can be used in the 
evaporation of the cloud, and part must go to raise the temperature of 
the air and so enable it to retain the additional vapour. 
There are a few other sources from which heat can be produced within 
a given mass of air, such as viscous friction, slow chemical action, electrical 
discharges, and the friction of meteors. The first of these is effective in 
diminishing eddy motion, and produces a permanent tendency to stagnation ; 
the others are occasional and random actions, and may be here ignored. 
Where so many activities are concerned a perfect balance at all times 
cannot be expected. Such of the actions as are of a fluctuating, or 
essentially local, nature may be dismissed from consideration in view of the 
