in the Ultra-red Region of the Spectrum of the Sun. 237 



the mirror. This slit served as the luminous object ; and an 

 image of it was projected by a second silver mirror of the 

 same dimensions, at the distance of about 1700 millim. In 

 the path of the rays proceeding from this second mirror the 

 diffraction-grating was inserted so that the cone of light 

 exactly covered its entire aperture. Hence the concave mir- 

 ror did not project a mere image of the slit, but the grating 

 produced on both sides of the image a series of diffraction- 

 spectra. One of these spectra, of the first order, was used for 

 the measurement of the wave-lengths. 



The silver mirrors above mentioned were made by silvering 

 and polishing the concave side of plano-concave glasses ; the 

 slit was formed by two strips of sheet metal 2 millim. broad 

 connected by cross strips. 



In order to render the extreme ultra-red rays perceptible I 

 did not, like Abney, employ their chemical effect, but their 

 thermal action was made visible by a radiometric torsion- 

 apparatus constructed by me. The arrangement of this appa- 

 ratus I have described in a previous paper*; here let the 

 statement suffice, that the rays falling on the lampblacked 

 side of a thin mica plate suspended in a vacuum produced a 

 motion of this disk. A small mirror joined to the disk rotated 

 simultaneously with it, by which a reflection of light was 

 thrown upon a scale. The motion of this reflection served as 

 an indication and a measure of the deflection of the vane. 

 The apparatus was connected with a mercury air-pump ; yet 

 it would have sufficed for the present experiment if it had 

 been made in a somewhat simplified form so that it could 

 have been hermetically closed after one exhaustion. 



The course of the rays in the experiment is clearly shown 

 by the diagram, fig. 1. 



The rays emanating from the heliostat H, set up outside 

 the window, pass first through an aperture A in the closed 

 shutter into the almost completely darkened room, then 

 through a pasteboard tube Pi about 500 millim. long, directly 

 attached to the shutter, and fall upon the first silver mirror S T 

 at a distance of 2510 millim. from the heliostat. By this 

 mirror they are concentrated upon the slit s, then arrive at 

 the second mirror S 2 , pass from this to the grating G, and are 

 here in part directly reflected so as to give rise to an image in 

 B of the slit s, in part dispersed. The extreme ultra-red rays 

 of the spectrum of the first order can be collected in T, where 

 they are incident upon the mica plate. 



For the sake of clearness, the angles of incidence of the 

 rays upon the mirrors Si and S 2 are drawn much too large in 

 * Phil. Mag. February 1883, p. 101. 



