l8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 7I 



By omitting the two fog observations (Oct. 8 and Feb. 2) and 

 taking a general average of the twenty-four remaining Hump Moun- 

 tain observations, the following values are obtained: .^^=151.5 per 

 cent, 3; = 89.5 per cent, and ^=69.7 per cent, or approximately 

 .ar=i50 per cent, 3' = 90 per cent, and z=yo per cent. In Calama, 

 omitting the observations of July 21 and October 28, which differ 

 much from the others, the averages of the five remaining values 

 are .*-= 149.5 per cent, 31=85.2 per cent, and ^ = 80.2 per cent. Thus 

 it appears that in clear or somewhat hazy skies the radiation actually 

 received on a horizontal surface, as compared with what would be 

 received on a horizontal surface from an " equal " sky, shows the 

 following percentages. First 30° zone above the horizon 150 per 

 cent. Middle 30° zone 85 per cent to 90 per cent. Upper 30° zone 

 70 per cent to 80 per cent. From the two sets of observations taken 

 in low and apparently even fog at Hump Mountain it appears that 

 the middle zone has about a normal percentage but that the upper 

 zone is much increased and the lower zone correspondingly decreased. 

 This might be expected, however, since the fog particles are so dense 

 in a horizontal line, while vertically the thickness is much less, and 

 more radiation finds its way through the fog. The above ratios 

 are undoubtedly affected very considerably by the sun's being in a 

 particular zone at the time of observation, as will be shown under 

 the discussions of the observations taken at different azimuths. 



While the above ratios hold for the total amount of radiation re- 

 ceived from the three 30° zones, it must not be inferred that equal 

 areas of these zones in the various azimuths of the sky send out equal 

 intensities. In order to investigate this phase of the subject another 

 set of observations which we have termed "vertical zones" was 

 carried on at Hump Mountain. To do this, the pyranometer was 

 mounted bodily on an equatorial mounting such as is used for 

 pyrheliometers. The threads where the equatorial mounting screws 

 into its base, were left a little loose, and angles were marked off so 

 that the whole apparatus could be revolved about a vertical axis 

 through a known angle. The plane of the pyranometer strips was 

 pointed exactly at right angles to the sun's beams, as determined by 

 a spot of light. The metal thimble limiting exposure to a cone of 

 light of 60° arc was placed over the glass hemisphere. In all cases 

 the small disc to shade the direct rays of the sun was in place ver- 

 tically above the strips, even where the instrument pointed to quarters 

 of the sky away from the sun. 



