1 72 S. P. Langley on the Selective Absorption of Solar Energy. 



(such as that included between 15' of deviation) is gone over 

 several times in the course of one day, the measurement being 

 repeated on every single minute of arc, with a separate open- 

 ing and closing of the slit, and a record made of the full swing 

 of the galvanometer-needle for each observation. 



These observations are entered numerically, and corre- 

 sponding charts made on large sheets of section-paper. The 

 same narrow region will thus be gone over also on different 

 days, and the different charts subjected to a very rigid exami- 

 nation ; so that every feature which is not common to them 

 all is rejected or reexamined ; and in this manner the whole 

 spectrum is studied. These original charts are on a scale four 

 times as large linearly as that the reader now sees (Plate III.). 



In addition to this, on some clear days, tracings have been 

 made upon the chart directly corresponding to the movements 

 of the galvanometer-needle : that is to say, the observer at the 

 spectro-bolometer has moved the bolometer through the whole 

 spectrum by means of the tangent-screw; the slit has been left 

 permanently open, so that the bolometer has been constantly 

 exposed ; and the observer at the galvanometer, seeing the 

 needle moving, as a hot or cold part of the spectrum passes 

 over, calls the deflection of the galvanometer corresponding to 

 each minute of arc, while a third person plots the same on 

 section-paper. In this way as many as eight curves, like 

 those here given of the prismatic spectrum, have been obtained 

 between noon and sunset on one day, giving a picture of the 

 action of the selective absorption of the atmosphere in every 

 part of the spectrum, as the rays of the sinking sun pass 

 through greater depths of air. This third method (very useful 

 when, as in this case, many observations have to be taken in 

 a short time) is nevertheless less accurate than those before 

 described. 



A careful bolometric and also optical setting is made on 

 some well-known line, usually C, at least once daily, to make 

 sure that the adjustment of the instrument is equally accurate 

 for the visible and invisible rays. 



Description of the Apparatus. 



The rays of the sun are reflected horizontally from the 

 mirror of the large Foucault siderostat through an aperture in 

 the north wall of the observatory, and received upon a plate with 

 a slit, whose jaws, moving each way from the centre by a mi- 

 crometer-screw, can be regulated so as to allow a beam of any 

 desired size to pass. 4-J- metres from this slit, at the distance 

 of its principal focus, is a collimating lens, L (fig. 4, p. 174), of a 



