The Secondary Spectrum of Hydrogen. 581 



at 10,000 amperes, which he considers a low estimate. The 

 effect of the lightning may be partly a heating one, the rock 

 being heated so far as to transform a non-magnetic into a 

 magnetic constituent, which is then magnetized by the earth's 

 field. A second possibility is to attribute the chemical change 

 to the sun's heat, and the magnetization to the lightning 

 flash, since the experiments of Folgheraiter and Messrs. 

 Platania* have proved that both the earth's field and 

 lightning are operative in producing magnetization in rock 

 material. 



Experiments on haematite ores are in progress, but sufficient 

 has been done to find that the second allotropic form of 

 magnetite, into which they change on being heated at 100°, 

 behaves similarly to the more common form as regards loss 

 of magnetic power about 550° C. 



The experiments described in this paper were made in the 

 Natural Philosophy Buildings of Glasgow University, and I 

 have to thank Professor Gray for valuable suggestions made 

 during their progress. 



Glasgow, 30 Nov. 1908. 



LII. The Secondary Spectrum of Hydrogen, 

 By C. F. Hogley, 'A.R.C.Sc, B.Scj 



IT is well known that vacuum-tubes filled with hydrogen 

 emit two spectra. One of these consists of the well- 

 known lines which are seen reversed in the spectrum of the 

 sun and in the spectra of some stars. The other spectrum is 

 known as the secondary spectrum, and consists of a multitude 

 of fine lines crowding the visible region, being especially 

 conspicuous in the neighbourhood of the yellow. These 

 lines have not been observed in stellar or solar spectra, and 

 it has been regarded as doubtful whether they are due to 

 hydrogen at all. 



It seems almost certain that if this spectrum is not due to 

 hydrogen, it is due to some more easily condensible substance. 

 Sir James Dewar has shown that all gases except hydrogen, 

 neon, and helium, are readily condensed by charcoal at the 

 temperature of liquid air. This method of condensation 

 seems to afford an opportunity of testing whether or not the 

 secondary spectrum is in reality due to hydrogen. 



* G. & G. Platania, Comptes JRendus, cxli. p. 974 (1905). 

 t Communicated by the Hon. R. J. Strutt, F.R.S. 



