242 M. E. Pringsheirn on a Measurement of Wave-lengths 



fixed, and the grating placed so that the image of the cross 

 reflected from the mirror Gr to the mirror S 2 , passing through 

 the slit 8 fell upon the mirror S l5 and, reflected by this, 

 appeared on the screen in A somewhat enlarged. The ap- 

 • paratus was then so arranged that the deviation in the hori- 

 zontal plane was corrected by rotating the stand on which 

 the grating stood, while half of the deviation in the vertical 

 was obviated by one of the adjusting-screws attached to 

 the base of the grating, the other half by varying the 

 altitude of the mirror S 2 . In this manner the setting-up was 

 regulated so that the centre of the wire cross came to lie 

 exactly in the centre of the rectangular aperture. In order 

 to maintain the plane of incidence invariable with any other 

 position of the grating, the position of the horizontal line 

 proceeding from the horizontal wire of the cross, which ran 

 through all the spectra, was fixed by a lead-pencil line on the 

 pasteboard disk in front of the torsion-apparatus; and then 

 the two other screws in the stand of the grating-mirror were 

 adjusted so that the horizontal line always kept the same 

 position with any rotation of the micrometer-stand. After 

 this the mirror was turned back again into its previous 

 orientation ; and the same manipulation continued till the 

 grating had the correct adjustment in all orientations. 



In the second position also, in which the directly reflected 

 ray fell on the torsion-apparatus, the entire aperture A was 

 not employed, but only the small cut-out aperture, in order 

 to make a more accurate adjustment possible. In doing this 

 it was of course requisite to take care that the image of the 

 aperture fell exactly on the centre of each of the mirrors. 



In order that in this position also the directly reflected 

 rays should actually take the same position as the extreme 

 operative rays with the third position of the grating, it was 

 necessary to determine the place in the radiometer-vane 

 which with the third position is hit by the extreme ultra-red 

 rays. As the intensity of the dark heat-rays diminishes as 

 their distance from the red end of the spectrum decreases, 

 and as the effectiveness of the rays falling upon the ra- 

 diometer-vane increases with the distance from the torsion- 

 axis of the vane, the torsion-apparatus was placed so that the 

 extreme most sensitive edge of the vane was turned towards the 

 red end of the investigated spectrum. If then the rays falling 

 upon the vane called forth any motion at all, it was certain that 

 effective rays fell also upon the extreme edge of the vane. On 

 this account, with the second position of the grating the directly 

 reflected image of the slit was adjusted so that the vertical 

 thread of the cross coincided with the edge of the vane. The 



