F. W. Very — Sky Radiation and Isothermal Layer. 373 



abed. The area between the two curves represents heat 

 communicated to the air through the absorption of solar radi- 

 ation. The slight upward convexity at a is due partly to heat 

 from absorption of terrestrial radiation and partly to conden- 

 sation of moisture in the lower air. Local and temporary 

 irregularities, as at b and c, indicate upper regions of aqueous 

 condensation. In region a, the nuclei which serve as centers 

 of aqueous condensation are mainly dust particles carried up 

 from the surface in the local convection, as Aitken has demon- 

 strated. At the higher condensation levels, since the relative 

 humidity is not often above 55 per cent, Blair suggests that the 

 condensation may be produced around nuclei of carbonic acid 

 (H 2 C0 3 ) which, though unstable at ordinary temperatures, ap- 

 pear to exist at the temperature of the upper isothermal layer. 

 To some extent volcanic and cosmic dust furnishes condensa- 

 tion nuclei at very great altitudes, where luminous night-clouds 

 are sometimes seen. The heat communicated to the air by 

 aqueous condensation is greatest in the lower and moister 

 layers, yet even here this source is of small account compared 

 with the great accession of heat from the sun's rays in the 

 upper air. 



Professor W. J. Humphreys* has published his opinion that 

 my explanation of the isothermal layer is incorrect because 

 " the water vapor is not there to absorb." f He accepts the 

 conclusion that the heat comes from the absorption of radiation, 

 but assigns the absorption to u oxygen and ozone." X The first 

 proposition, that water vapor is lacking, is disproved by the 

 U. S. Weather Bureau observations already cited. By spectro- 

 bolometric observations it can be shown that the ozone bands 

 " in the infra-red, one of which, 8*5 n to 10*5 /*, is very strong, 

 and of course would absorb earth radiations," § occur actually 

 in that part of the infra-red spectrum which would otherwise 

 be most readily transmitted to space, and where the telluric 

 radiation should also attain its maximum strength according to 

 Wien's law. Knut Angstrom,] who first assigned their origin 

 to these bands, gives the limits of greatest absorption " from 

 X = 9"1 fx to X = 10*0 /*." I have observed them on several 

 occasions in the solar spectrum, but not always with certainty 

 in the lunar spectrum, where, however, some allowance must 

 be made for the difficulty of the observation. Ozone has always 

 been regarded as one of the most variable constituents of the 

 atmosphere. If its regular diurnal production in the upper air 

 by the action of the sun's ultra-violet rays is conceded, the 



* Astrophysical Journal, vol. xxix, p. 14, January, 1909. 



f Op. cit., p. 22. % °P- cit -> P- 29. § Op. cit., p. 28. 



\ "Die Ozon-bander des Sonnenspektrums und die Bedeutung derselben 

 fur die Ausstrahlung der Erde," Arkiv for Matematik, Astronomi och Fysik, 

 i, p. 395, 1904. 



