abbot: solar constant of radiation 109 



servers at high altitudes, and these also are found to agree closely 

 with the theory of Rayleigh, and with our own observations. 



Not less interesting is the determination of the distribution of 

 energy in the sun's spectrum, and thereby of the probable tem- 

 peratures existing in the sun. The solar temperatures may be in- 

 ferred also from the value of the solar constant of radiation itself, 

 and the two methods agree substantially in giving the probable 

 solar temperatures as between 6000° and 7000° absolute Centi- 

 grade. 



Recent halloon experiments. Notwithstanding the satisfactory 

 state of the theory of solar constant measurements by the meth- 

 od of Langley, depending upon spectro-bolometric observations 

 at high and low sun combined with measurements by the pyr- 

 heliometer, and notwithstanding the close agreement between 

 results obtained by this method for many years at stations of 

 differing altitude from sea-level to 4420 meters elevation, there 

 still exists the possibility that if we could, indeed, go outside the 

 atmosphere altogether, we should obtain values differing mate- 

 rially from those above given. So long as we observe at the 

 earth's surface, no matter how high the mountain top on which 

 we stand, the atmosphere remains above us, and some estimate 

 must be made of its transmission before the solar constant can be 

 determined. Different persons will differ in the degree of con- 

 fidence which they will ascribe to measurements of the atmos- 

 pheric transmission such as have been considered, and there are 

 still some who totally disbelieve in the accuracy of the results 

 thus far obtained, even though they be confirmed by observa- 

 tions at such differing altitudes. Accordingly it has seemed 

 highly desirable to check the results by a method of direct ob- 

 servation by the pyrheliometer, attaching the instrument for 

 this purpose to a balloon and sending it to the very highest pos- 

 sible altitudes. By a cooperation between the Smithsonian In- 

 stitution and the United States Weather Bureau, experiments for 

 this purpose were made in July and August of the year 1913. 



The instruments employed were modified in form from the 

 silver disk pyrheliometer, which has been described above. As 

 the apparatus could not be pointed at the sun, the disk was 



