38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



radiation to the sky must be subject to considerable changes also. 

 Such conditions are generally characteristic of inland climates and 

 are very marked in desert regions, where the humidity is low and the 

 balancing influence of the neighborhood of the sea is absent. Indio 

 is situated in a desert region. In the middle of the day the tempera- 

 ture reached a maximum value of 43 ° C. on the 23d and 46 ° C. on 

 the 24th of July. In the evenings at about 8 o'clock the temperature 

 was down to 30 C, falling continuously to values of 21 ° and 19 C, 

 respectively, in the mornings at 4 130, when the observations ceased. 

 From the curves it is obvious that there is a close relation between 

 the radiation and the temperature. Every variation in the tempera- 

 ture conditions is accompanied by a similar change in the radiation. 

 In fact a decrease in the temperature of the surrounding air causes 

 a decrease in the effective radiation to the sky. This is even more 

 obvious from the observations taken at Lone Pine on August 5 and 

 August 10, when very irregular temperature variations took place 

 during the nights. The humidity conditions appeared almost con- 

 stant. From the curves (figs. 19 to 21) can be seen how a change in 

 the one' function is almost invariably attended by a change in the other. 

 In regard to the radiating surfaces of the instrument, one is pretty 

 safe in assuming that the total radiation is proportional to the fourth 

 power of the temperature, an assumption that is based upon the con- 

 stancy of the reflective power of gold and of the absorption power of 

 platinum-black soot within the critical interval. The radiation of 

 these surfaces ought, therefore, to follow the Stefan-Boltzmann law 

 of radiation. For the radiation of the atmosphere we thus get : 



E a t = E s t — Rt 

 Knowing E s t and Rt, of which the first quantity is given by the 

 radiation law of Stefan, to which I have here applied the constant 

 of Kurlbaum ((7=7.68 • io -11 ), and the second quantity is the effec- 

 tive radiation measured, I can calculate the radiation of the atmos- 

 phere. We are led to try whether this radiation can be given as a 

 function of temperature by an expression 



E at = C-T« (1) 



similar in form to the Stefan-Boltzmann formula, and in which a 

 is an exponent to be determined from the observations. From (1) 

 we obtain : 



log £ a , = log C + a log T 



Now the observations of every night give us a series of correspond- 

 ing values of E at and T. For the test of the formula (1) I have 



