38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, 68 



by Rubens/ In view of these figures it seems possible that the diver- 

 gence of the observed Nernst lamp spectrum curve from the com- 

 puted black-body curve may be partly due to the increasingly incom- 

 plete absorption of the rays by the bolometer strip. A part of the 

 divergence is surely due to carbon-dioxide absorption. Although the 

 experiments do not indicate any absorption by water-vapor less than 

 0.03 cm. ppt. HgO, they are not quite conclusive. Owing to the 

 impurity of the spectrum, fine line absorption, like that which Abney 

 photographed in par, may occur here undetected, and it may possibly 

 produce its complete effects with very minute quantities of water 

 vapor. If so, a part of the discrepancy may be due to water-vapor 

 absorption. Finally, it is not certain that the two energy curves 

 should coincide, for the Nernst lamp may depart widely from being 

 a perfect radiator. 



It will be noted that when observed through the spectroscope alone 

 energy from the lamp is found beyond 16 /x even after allowance is 

 made for the field light. The lamp curve is the mean of many 

 observations and no doubt is felt that there is some kind of energy 

 here. Although there may be in it some field light a return of energy 

 is to be expected in this spectrum region beyond the carbon dioxide 

 found where the work of Rubens indicates increased transparency 

 of water vapor. 



ATMOSPHERIC TRANSMISSION OF SOLAR ENERGY BETWEEN 



9 AND 14 n 



Even with the maximum amount of water vapor possible in the 

 tube work (0.26 cm. ppt. H2O) the transparency observed between 

 9 and 14 |U, had appeared to be very great. It was thought worth 

 while to attempt a measure of this transparency between 9 and 14 /u, 

 by observing the amount of energy transmitted from the sun through 

 the earth's atmosphere for various air masses. In this way there 

 would be brought about a several-fold increase of the amount of 

 absorbing vapor in the path. Indeed at the maximum this amounted 

 to 3 cm. ppt. HoO. 



Reference has already been made," in discussing the transparency 

 of this wave-length region to the remarkable early work of Mr. Lang- 

 ley. As far as air-mass 3.76 practically no absorption was found by 



^ Verhandlungen Deutsche Physikalische Gesellschaft, 13, p. 88, 1911. His 

 values are at 2 n, 0.5 per cent transmission; 41.1, 8.6; 61.1, 16.0; 12 [a, 37.6; 

 26 m., 76.7; 52 M-, 91-3; 108 M, 9I-5- 



' S. P. Langley, The Solar and the Lunar Spectrum. Memoirs National 

 Academy of Sciences, IV, p. 159, 1886. 



