in the Ultra-red Region of the Spectrum of the Sun. 243 



irradiation of the part of the glass bulb of the torsion- 

 apparatus nearer to the visible spectrum, which might pos- 

 sibly have produced a motion, was cut off by the sheet-metal 

 case surrounding the apparatus. 



To give the grating its third position, it was first placed so 

 that the motion of the torsion-apparatus was very slight; 

 then the finer adjustment was made by means of the micro- 

 meter-screw attached to the turning-arm of the stand. Yery 

 slight deflections of the radiometer-vane could be easily and 

 rapidly multiplied by measured interruption and restoration 

 of the irradiation. In the first position of the grating, in 

 which no action was shown, it could with certainty be as- 

 serted that the extreme effective rays fell upon the boundary 

 edge of the vane. In this position the reading c was made. 



IV. Results. 



Unfortunately, on account of unfavourable weather, but 

 very few measurements could be effected. These gave the 

 following results on starting from the red end of the first 

 spectrum and advancing steadily to the place where the action 

 of the apparatus ceased : — 





a. 



b. 



c. 



X. 



July 25, with iodine solution. 



139 40 0-0 



159 00 



130 21 55-5 



milliin. 

 0-0013281 



Aug. 7, „ „ „ 



167 30 00 



183 57 47-0 



154 57 194 



00013658 





165 46 39-9 



184 27 45-8 



154 38 56-6 



0-0013834 





The situation of the extreme ultra-red rays fell here through- 

 out in the red of the second spectrum, as one could easily be 

 convinced by inserting a red glass plate, since this absorbed 

 the blue end of the third spectrum, which coincided with the 

 red end of the second. 



When the grating was rotated still further, there came first 

 a narrow strip of insensitivity; very soon, however, action of 

 the dark rays upon the torsion-apparatus made its appearance. 

 It commenced within the luminous spectrum of the second 

 order, and extended to the extreme end of this spectrum, 

 where the transition to the dark part of the spectrum of the 

 second order rendered separation of the two spectra no longer 

 possible. 



It is therefore proved that rays occur in the sun's spectrum 

 of which the wave-length is double that of the extreme visible 



