IN THh; PRESENCE OE DUST-EREE AIR AND OTHER GASES. 
•200 
compared wibii what condenses in the form of drops. From the cpiantity of water 
which separates, and the size of the drops, we may calculate the number, assuming 
the water to be equally divided among them. 
It is assumed here that the cloud-particles are actually liquid drops and not ice- 
crystals, in spite of the fact that the condensation begins at temperatures much below 
the freezing point, and that the temperature when the particles are full grown is, as 
we shall see, also slightly below the freezing point. 
Let us consider first the quantity of water which separates out in consec[uence of 
an expansion of a given amount. Let us suppose the expansion to be completed 
before the drops have grown to more than a very small fraction of their final size, so 
that the theoretical lowering of temperature results. Let q be the temperature 
Centigrade before expansion, the lowest temperature reached. 
In consequence of the condensation of the water, heat is set free, and the tempera¬ 
ture of the moist adr rises. A stationary state is reached at a temperature Cj when 
just so much water has separated that the vapour remaining is in equilibrium in 
contact with the drops. The subsequent changes will be slow, being due to the 
inflow of heat and vapour from the walls. They appear to have little effect upon the 
size of the drops, for the colour changes very little, and only gradually fades away ; 
evidently through the drops becoming unequal in size. This is not difficult to 
understand, for the air which comes in contact with the walls of the tube, since these 
are covered with water, must remain saturated in spite of its rise in temperature. 
If we consider 1 cub. centlm. of the moist gas, we have the following equation 
connecting the temperature t reached at any moment with the quantity of water q 
which has condensed, 
l^q = CM {t - t,), 
where M is the mass of unit volume of the gas and C its specific heat at constant 
volume. It will not introduce any serious error, for the present purpose, if we 
neglect the heat spent in raising the temperature of the small quantity of vapour 
present. L is the latent heat of vaporisation, which changes slightly as the 
temperature changes during the process, but may be considered with sufficient 
exactness as equal to 606, its value at 0° C. 
Now,- 
q = Pi - p> 
where is the density of the vapour just befoi’e condensation begins, and p the mean 
density of the vapour remaining uncondensed at any moment. 
