24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 76 



Now that five independent determinations are usually made daily 

 at each station, the mean results are very accurate. A comparison has 

 been made of the daily determinations at the two stations over the 

 period January to October, inclusive, 1923. It proves that a half of 

 one per cent is the average daily difference between the indications of 

 the solar heat as it is outside the atmosphere, determined at these two 

 stations many thousands of miles apart, one in the Northern, the other 

 in the Southern Hemisphere, one at an altitude of 5,000 feet, the other 

 at 9,000 feet. 



The two stations join in indicating the march of the solar heat up 

 and down, and within the past year the fluctuations have ranged over 

 about 4 per cent. During the years 1914 to 1921, the results had run 

 generally at a level of about 1.95 calories per square centimeter per 

 minute. Beginning in 1921, a notable downward march began, and by 

 September, 1922, the monthly mean values were ranging at about 1.91. 

 This lower level continued, with minor fluctuations, for a number of 

 months, and the lowest values were reached in February and March, 

 1922. After that, there was a gradual increase until in September 

 and early October, 1923, the values had come to an average level of 

 about 1.93. Still more recently, there has begun a slump, so that at 

 latest advices, up to February i, 1924, the solar heat outside the 

 atmosphere is running at approximately 1.92 calories. It will be of 

 great interest, after two or three years of this steady investigation of 

 the solar radiation, to compare the results with meteorological 

 conditions. 



The reader might think it obvious that if the solar radiation falls 

 the temperature would fall also. Nothing so simple as this occurs. For 

 the earth's surface is so complex that its deserts, its mountains, its 

 oceans, and other features, with the circulation of the atmosphere, 

 modify extremely the efifects of the solar heat. It is easy to see, for 

 instance, that inasmuch as a quarter of the sun's heat is absorbed in 

 the atmosphere itself, and as the atmosphere has but a trifling capacity 

 for heat compared with the solid earth or the ocean, that its tempera- 

 ture must be almost immediately afifected by solar variations, far more 

 directly than the temperature of the ocean or the temperature of the 

 land. But since the atmosphere is in some regions hazy, humid, and 

 cloudy, in other regions dry and transparent, the quantity of solar 

 heat absorbed must vary very much from place to place. So the 

 changes in the solar heat must produce very different temperature 

 effects in the atmosphere in a cloudless desert region at high altitude 



