30 



SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 'J'] 



It would be highly advantageous to have another pair of stations 

 located, if possible, in well separated cloudless regions where the 

 months December to March are more favorable. 



Variation or thl Sun I9I8toI924 



TEN- DAY MEAN VALUES 



Fig. 2>-/. 



On the whole, the solar heat appears to have been continuously 

 below the normal for the past 2\ years. Of late months the tendency 

 to rise has become apparent, so that it seems likely that higher values 



