26 



Mr. W. Crookes. On the [Jan. 13, 



large spectroscope of very accurate construction, and the results, I 

 think, are sufficiently important to be worth bringing before the notice 

 of the Royal Society. 



The spectrum consists firstly of an exceedingly faint and hazy pair 

 of bands ; these are too faint to measure, but they appear to be in 

 about the same position as the bands 1986 — 2009, and 2031 — 2075 in 

 the spinel spectrum given further on. Next is seen the characteristic 

 crimson line of the alumina spectrum ; this, when examined with a 

 fine slit and high power eyepiece, is seen to be double, the distance 

 apart of the components being about half the distance separating the 

 D lines. Then come a pair of fainter and rather nebulous orange 

 lines ; beyond them is a dark space followed by a continuous spectrum 

 extending to the green. The following are the measurements of the 

 spectrum : — 



Scale of 



X. 



1 



Remarks. 



spectroscope. 



A"" 



10-550° 



6942 



2075 



The first component of the double crimson 

 line. 



10-548 



6937 



2078 



The second component of the double crim- 

 son line. These lines are nearly as sharp 

 as the components of D. 



10 -450 



6707 



2223 



Approximate centre of a narrow band, 

 shading off at each side. 



10 -400 



6598 



2297 



Approximate centre of a narrow band, 

 shading off at each side. This band is 

 somewhat sharper and brighter than the 

 one at 2223. 



10-360 



6514 



2357 



Approximate commencement of the con- 

 tinuous spectrum which extends into the 

 green, shading off too indefinitely to 

 admit of measurement. 



The accompanying cut (fig. 1) gives the spectrum drawn to the 

 -i scale. 



A* 



Fig. 1. 



150 170 190 210 230 250 2 70 2 90 310 3 30 350 370 



The Alumina Spectrum. 



