and the Line-spectra of Hydrogen and Oxygen. 361 



it yields another strong ray (\ = 3080) of the H 2 0-spectrum. 

 The ray 4158 occurs, namely, amongst a group (3986, 4005, 

 to 4645) of rays which H. W. Vogel has observed in the 

 spectrum of rarefied hydrogen (whether obtained electrolyti- 

 cally or from potassium formiate), the derivation of which 

 from hydrogen was, however, very doubtful. In view of the 

 fact that in these methods of preparation of hydrogen it is 

 difficult to avoid the presence of traces of water-vapour, it 

 was a probable assumption that at least some of the doubtful 

 rays, and amongst them in particular the ray 4158, might 

 arise from the traces of water-vapour contained in this hy- 

 drogen, and might be rays in themselves very weak in presence 

 of hydrogen, but which, being reinforced by harmonic vibra- 

 tions of the latter, become visible. If this surmise were cor- 

 rect, it might be expected that the wave-lengths of those of 

 VogePs rays which belong to the same group of H 2 rays as 

 4158 might also be reduced to the H 2 0-spectrum by multi- 

 plication with the factor f^. The supposition was verified in 

 a remarkable manner, in particular for the rays 3986, 4005, 

 4047, 4065, 4067, 4078, 4152, 4158, 4168, 4193, 4201, &c, 

 so that these belong to the H 2 0-spectrum ; and by division 

 by 0*6336 are converted into very feeble rays of hydrogen, 

 hitherto unknown, with the exception of H a , which on account 

 of their extreme feebleness have escaped observation. Their 

 approximate wave-lengths are placed together with the nearest 



lines of the solar spectrum, as given by Angstrom. 



H-rays. 



6291 

 6321 

 6387-3 

 6415-7 



6418-9 

 6436-2 

 6553-0 

 H a 6562 

 6578-3 



6617-7 

 6630-3 



Solar spectrum 

 (Angstrom). 



6291-4 

 6321-5 



? 



6415-6 

 (Chromosphere, 

 Young) . 

 6418-7 

 (6438? Ca) 

 ? 



6562 

 6576-9? 

 (Chromosphere, 

 Young) . 



? 



With a possible error of 1*6 of Angstrom's unit. 

 Phil. Mag. S. 5. Vol. 24. No. 149. Oct. 1887. 2 B 



