170 S. P. Langley on the Selective Absorption of Solar Energy, 



mountain just referred to, and on nearly every good day 

 during the first six months of 1882. There were of such: — 

 4 days in January, 8 in February, 9 in March, 9 in April, 

 9 in May, 12 in June; in all, 51 days. 



Very early the observations with this efficient apparatus 

 (the result of improvements due to the previous two years' 

 practice) showed, by an accuracy not hitherto attained in such 

 measures, the possibility of mapping the regions of the infra- 

 red spectrum believed to have been first observed on Mount 

 Whitney, and which have remained hitherto unknown. The 

 extreme narrowness of the bolometer-thread (one fifth of a mil- 

 limetre) and the size of the prismatic spectrum employed made 

 it also possible, in spite of the condensation of the latter, to dis- 

 criminate lines and gaps in its continuity which had escaped 

 previous observation*. The spectrum, as the reader now sees 

 it in the charts, was mapped with the intention of noting 

 these interruptions of energy, which had been in the previous 

 research designedly neglected. The bolometer shows the cold 

 in the principal visible Fraunhofer lines readily; but as their 

 effect individually is slight, it has not been indicated in the 

 part of the visible spectrum above C. 



The map reached approximate completeness early in April 

 1882, after which date observations have been still continued 

 daily, whenever possible ; so that every portion of the curves 

 here given, to the smallest inflection, has been observed from 

 three to twenty times, and the accidental variations due to 

 momentary interruptions of solar heat by invisible clouds have 

 been, it is hoped, nearly eliminated. 



The bolometric work, represented by the preceding 51 days' 

 observations, may be here regarded as being divided into two 

 classes having distinct though related objects. 



1st. To determine the general selective absorption of the 

 earth's atmosphere throughout the entire spectrum in con- 

 nexion with the observations already made here and on Mount 

 Whitney. For this purpose measurements have been made at 

 the following deviations :— 44° 30', 45° 30', 45° 53', 46° 12', 

 46° 30', 46° 45', 47° 30', 48° 00', 49° 00', 50° 00', 51° 00'. 



All these points have been measured on, twice each, the 

 whole forming a " series ;" and these " series " are observed at 

 least twice daily — namely at meridian, and when the air-mass 



* Nevertheless, as the thread, however narrow, is not absolutely linear, 

 it feels the cold before its centre coincides with the centre of the line. 

 The interruptions of the energy-curve are thus in fact all in a slight 

 degree too wide, especially at the commencement of each depression ; and 

 it is probable that the bands we have marked are really due to an aggre- 

 gation of finer lines. 



