8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 69 



air masses in Table I. This is of importance in deducing a value of 

 the albedo of the earth from these results, for it tends to show not 

 only that fog- layers near the boundaries of the earth's surface differ 

 little in reflecting power from those' directly under the sun, but also 

 that rough clouds do not differ very much from smooth ones in re- 

 flecting power. This latter point of course should not be urged too 

 far, for it is obvious that clouds with very deep holes and furrows 

 must reflect less than smooth ones. 



Referring to the discussion of cloud reflecting power in Volume II, 

 Annals of the Smithsonian Astrophysical Observatory, page 145, we 

 find that using 65 per cent as the reflecting power of a cloud surface 

 a value of 33.7 per cent is obtained as the total amount of the incom- 

 ing solar radiation over the whole earth reflected to space by clouds. 

 Substituting 78 per cent for 65 per cent this value becomes 40.4 per 

 cent. It seems probable that the low cloud reflection value of the 

 early Mt. Wilson work (65 per cent for cloud reflecting power) can 

 be attributed largely to the uncertainty of the extrapolations neces- 

 sary, since the observations were limited to a small range of nadir 

 distance. Moreover, the contribution from the very bright area near 

 the angle of specular reflection was perhaps minimized. 



Following the method of pages 162 and 163 (Annals, Vol. II), a 

 new value of the albedo of the earth is derived. Using 78 per cent 

 as the cloud reflecting power, the albedo of the earth (as defined by 

 Bond, see article by Russell, Astrophysical Journal, 43, p. 175) 

 becomes 43 per cent. Russell (Astrophysical Journal, 43, p. 190) 

 derives for it a value of 45 per cent from a consideration of Very's 

 visual observations on Venus and the moon. 



It will be clear that the method here adopted to get the cloud 

 reflecting power (i. e., taking the ratio of the total radiation received 

 by the pyranometer per square centimeter of horizontal surface from 

 the cloud, to the total radiation received from sky and sun by a square 

 centimeter of horizontal cloud surface) may give different results 

 from measurements by visual or photographic methods as employed 

 in photometry. Although even in the present work part of the solar 

 rays is missing, owing to water vapor absorption, the results are more 

 clearly applicable to considerations of the earth's temperature than 

 photometric results would be. Still it is probable that the difference 

 is small. 



The planet Venus according to Russell's discussion of Miiller's 

 observations, has a Bond albedo of 59 per cent for visual rays. 

 Because of its high reflecting power and the absence of telescopic 



