OF SUNLIGHT THROUGH THE EARTH’S ATMOSPHERE. 
23 
be wondered at. For it must be recollected that the absorption by aqueous vapour 
takes place principally in the infra-red region of the spectrum, and, if it were entirely 
confined to that region, the presence or absence of the aqueous vapour would have no 
practical effect on the luminosity of the visible rays transmitted.The absorption by 
this vapour where it takes place is large and follows the ordinary laws, as already 
stated, being principally in the iufra-red of the spectrum, the heating effect will 
therefore diminish much more rapidly than the total illuminating value of sunlight as 
increased thickness of atmosphere is penetrated. In other words, the amount of light 
transmitted from the sun bears no comparison with that of its total heating effect. 
It should be remarked that the portion of the spectrum used in these observations 
is almost free from any absorption by aqueous vapour, and, consequently, the scattering 
effect of the small particles in the atmosphere is probably almost entirely the cause 
of the loss of light from it, and enables a factor for such scattering to be arrived at 
without much difficulty. The measurement of the amount of light transmitted through 
different thicknesses of atmosphere has hitherto been almost entirely made on star and 
moon light, and, for reasons given in Part I. of this paper, the coefficients arrived at 
are probably slightly too large for the transmission of sunlight. 
The results of these observations made at different altitudes, combined with others 
made in the laboratory, point to a probability that the particles which selectively 
scatter light are due to water. Their dimensions are probably very closely, if not 
exactly, the same, and the “ mist” particles are large compared with them. 
* See a paper in No. 224 of tlie ‘ Proceedings of the Royal Society,’ 1883, by General Fbsting 
and the Author. 
