NO. 3 RADIATION OF THE ATMOSPHERE — ANGSTROM 2>7 



It is therefore evident that our formula can satisfy the conditions 

 only between the limits within which the observations are made, 

 and that in particular an extrapolation below 4 mm. water- vapor pres- 

 sure is not admissible without further investigations. These condi- 

 tions will be more closely considered in connection with the observa- 

 tions made on Mount Whitney, where the absolute humidity reached 

 very low values. 



For the case where p approaches very high values, the formula 

 seems to indicate that the radiation approaches a value of about 0.11 

 cal., which may show that the water vapor, even in very thick layers, 

 is almost perfectly transparent for certain wave lengths. This is 

 probably only approximately true, and the apparent transparency 

 would probably vanish totally if we could produce vapor layers great 

 enough in density or thickness. In a subsequent chapter I shall dis- 

 cuss some observations that indicate that this is the case, and also that 

 the formula given above must prove inadmissible for very great 

 densities. 



2. RESULTS OF THE CALIFORNIA EXPEDITION 

 The observations were taken simultaneously at different altitudes : 

 (a) At Claremont (125 m.) and on the top of Mount San Antonio 

 (3,000 m.) ; (b) at Indio in the Salton Sea Desert (o m.) and on 

 the top of Mount San Gofgonio (3,500 m.) ; and (c) at Lone Pine 

 (1,150 m.), at Lone Pine Canyon (2,500 m.) and on the summit of 

 Mount Whitney (4,420 m.) . 



A. INFLUENCE OF TEMPERATURE UPON ATMOSPHERIC RADIATION 



Among the observations taken by this expedition I will first dis- 

 cuss some observations at Indio and Lone Pine separately, because 

 they indicate in a very marked and evident way the effect upon the 

 radiation of a very important variable, the temperature. The Indio 

 observations of the effective radiation are given in table III and are 

 graphically plotted in figures 17 and 18, where the radiation and the 

 temperature during the night are plotted as functions of time. As 

 will be seen from the tables, the humidity varied very little during 

 these two nights. 



As long as the temperature during the night is constant or almost 

 constant, which is the case in mountain regions and at places near 

 the sea, the effective radiation to the sky will not vary much, a fact 

 that has been pointed out by several observers: Pernter, Exner, 

 Homen, and others. But as soon as we have to deal with climatic 

 conditions favorable for large temperature variations, the effective 



