F. W. Very— On the Solar Constant. 205 



the outer limit to a quantity which might approximate to the 

 value of I near the Earth's surface, though there is no reason 

 why B should exactly equal I even then. But actually Bigelow 

 makes his B diminish from 2 at the top to 1*27 at the bottom 

 of the air, 1*27 being the value of I=B at sea-level under a 

 vertical sun. Hence his B can not represent loss by reflection. 

 Moreover, everywhere and at all times B=I, becoming zero 

 after sunset, the whole relating to a purely diurnal pheno- 

 menon, although the thermodynamic facts on which it is 

 supposed to be based are true of the night as well as the day. 



In connecting the thermodynamic and radiant quantities for 

 the entire atmospheric column, the solar rays have been treated 

 in the paper under discussion as if they belonged to the atmos- 

 phere, and the atmospheric heat as if it were directly and 

 immediately produced by the solar rays, while I and B appear 

 to be treated in the radiant equation as if they pertained to 

 two equal quantities relating to two halves of an atmospheric 

 process. In reality they relate to the erroneous theory of an 

 instrument, as has been shown above. 



A relation of approximate equality is also assumed between 

 the heat produced by the solar radiation at a surface and the 

 heat simultaneously exhibited throughout an air column. But 

 it is not correct to equate the solar radiation received by an 

 air column in one minute, even though it be absorbed by the 

 air, to the thermal energy maintained in that column during 

 the same interval of time and deduced from thermodynamic 

 data, for the latter represents the accumulated radiant energy 

 from many hours of sunshine ; and no such relation would 

 have been found if the computations had been made correctly. 

 The solar rays from a zenithal sun, passing through a vertical 

 column of 1 sq. cm. section for 1 min., and varying, according 

 to Bigelow, between a thermal equivalent of 4*00 and 1*27 

 gram cal. at top and bottom of the column, have suffered 

 losses which are largely reflective, and the atmospheric heat 

 derived from them after the elimination of all purely reflective 

 losses must suffice for maintaining the temperature of the air 

 at night as well as by day, so that the effective heating of the 

 entire air column by the radiation absorbed in the interval of 

 a single minute is only a fraction of a calorie. 



Let us take the height of a homogeneous atmosphere at 

 0° C. and 760 mm pressure, as 7'991 km. Then, neglecting 

 latent heat of evaporation of air, since it is very small at the 

 mean temperature of the air, and assuming that the specific 

 heat of the atmosphere averages 0*238, the mass of a column of 

 1 sq. cm. section at an average temperature of 2/oT/2/?=255° 

 Abs. C. is 



7-991 X 10 5 X 0-001276 = 1020 grams, 



