Intelligence and Miscellaneous Articles. 395 



of nitrate of uranium, and possibly the fluorescent material in the 

 asphalt, which may be related to the solid hydrocarbon fluorescing 

 green which Becquerel mentions (La Lumiere, tome i. p. 382). 



In this general connexion let me mention that I have observed 

 that while the acid salts of quinine generally are fluorescent, the 

 chloride is not, and that hydrochloric acid will decompose the acid 

 sulphate so as to destroy its fluorescence. 



There are several other points in connexion with this and the 

 foregoing subject, which I must leave for a subsequent discussion. 



P.S. — August 1. I have just obtained results with turmeric 

 which seem to indicate that its fluorescence is due to the presence 

 of a substance not yet observed, soluble in water, and without any 

 colour. — Silliman's American Journal, September, 1871. 



ON THE SPECTRA OF THE SIMPLE GASES. 

 BY M. A.-J. ANGSTROM. 



In the Becherches sur le Spectre solaire which I published in 

 1868, I already announced that the spectrum-observations to which 

 I had devoted myself had not convinced me of the correctness of the 

 opinion of Pliicker that one and the same gas, in the state of incan- 

 descence, could give spectra varying with its temperature. I rather 

 believed that in the appearance of the spectra a modification may be 

 observed which consists in the elevation of the temperature bringing 

 about a greater abundance of lines, and that the relative luminous 

 intensity of these lines may also undergo some changes, but that 

 nevertheless the spectrum preserves its character unaltered. It is 

 true that in disruptive discharges it happens, when the tension of the 

 gas is increasing, that the spectral lines spread, and even end by 

 uniting so as to form a continuous spectrum ; but even then one 

 cannot say that the result is a new spectrum. 



Several distinguished physicists, however, are of the opposite 

 opinion ; and probably the researches of M. Wiillner (according to 

 which hydrogen would have no less than four spectra, oxygen three) 

 have strengthened this conviction in the minds of many savants. 

 M. Dubrunfaut has expressed his doubts of the correctness of these 

 results : he remarks that the multiple spectra of oxygen and hydro- 

 gen may be due to nitrogen or mercury-vapour introduced by the 

 pump into the tubes. M. Wiillner, however, has shown (Comptes 

 Rendus, Jan. 17, 1870) that this explanation is inadmissible. 



Yet, as the question of the multiple spectra of the gases is a vital 

 one for spectral analysis, and in this light M. Wiillner's observations 

 are truly important, perchance the following analysis of the pheno- 

 mena observed will not be without interest. Permit me to com- 

 mence with a preliminary remark. According to the experience ac- 

 quired, at least by me, the results obtained concerning the spectra 

 of the gases are not absolutely sure when the rarefaction is carried 

 to its utmost limits. In proof of this I cite the following fact : — 

 On one occasion, when I rarefied as much as possible, by means of 

 a mercurial pump, the atmospheric air in a Geissler's tube, at the 

 same time causing the discharge of a RuhmhorfT coil to pass in the 



