260 Mr. J. R. Capron on the Comparison of some Tube 



forcement tres-considerable de la bande rouge 8 97.03 (W. 

 L. 5994) qui devient vive de presque invisible quelle etait en 

 Pabsence du refroidissement artificiel de la flamme." May not 

 this explain the varying intensity of the red line (and perhaps the 

 green also) in the aurora ? 



Iron- Spectrum. 



A comparison of this spectrum suggested itself to me (as it 

 has previously done to others), partly from the suspected re- 

 lations between the aurora and the corona, and partly from a 

 consideration of the views lately expressed by M. Gronemann 

 and others in favour of the aurora having its origin in the fall of 

 an incandescent meteoric powder. 



The spectrum was obtained from a spark taken over a solution 

 of perchloride of iron, and was remarkable for its brightness in 

 and about the green region. The lines varied considerably in in- 

 tensity; and with a fine slit the principal ones were sharp, distinct, 

 and clear. A group of three lines (a) stood out boldly in the 

 green as the most marked, and next to these a group of three 

 others more towards the violet end of the spectrum (see Plate VI. 

 spectrum 4). By the side of my own phosphoretted-hydrogen 

 and iron spectra I have placed the principal lines as given in 

 Mons. Lecoq de Boisbaudran's same spectra (reduced of course 

 to my scale), and with figures of wave-lengths for comparison 

 with the aurora-spectrum ; and to these comparisons I shall refer 

 hereafter (see Plate VI. spectra 1 & 2). 



Spectrum of Mercury. 



This spectrum is only given as useful for comparing with the 

 bright and principal aurora-line. It is easy to obtain with a 

 small coil, the metal being used as one electrode. The yellow 

 lines are distinct and steady ; but the green, which is very bright, 

 is apt to flicker as the spark moves on the surface of the metal 

 (see Plate VI. spectrum 5). 



