318 Intelligence and Miscellaneous Articles, 



agents. He also showed that commonly the upper part of a valley, 

 where the erosive action is perhaps least, is very much the steepest, 

 and urged other objections to the great excavatory powers often 

 attributed to glaciers. He then described one or two cirques in 

 detail, and showed that they were worked out by the joint action of 

 many small streams, and of the usual meteoric agents working upon 

 strata whose configuration was favourable to the formation of cliffs. 



XXXIX. Intelligence and Miscellaneous Articles. 



ON THE SPECTRA OF SULPHUR. BY M. G. SALET. 



^PHE employment of instruments and methods the delicacy and per- 

 •*■ fection of which greatly surpass those of our organs has often 

 entailed difficulties and errors. When spectral analysis is applied to 

 the minute quantities of matter which fill the Geissler tubes, we 

 often find ourselves in the presence of impurities which can only be 

 revealed by the spectral method itself; hence uncertainties. These 

 impurities may proceed not merely from the primitive gas, but also 

 from that operated on previously with the mercurial air-pump, from 

 the mercury of the pump, from the grease of the stopcocks, from the 

 sulphuric acid used as desiccating agent, from matters deposited on 

 the surface of the glass, and from the metallic electrodes, which pos- 

 sess the property of absorbing and afterwards diffusing a certain 



o 



number of gases. One can easily conceive how M. Angstrom, by 

 indicating these impurities, recently arrived at the suppression, as 

 not belonging to the pure gas, of all the supplementary spectra of 

 hydrogen described by M. Wiillner. But it seems to me that the 

 discovery of M. Pliicker is not shaken by these facts, or that, at 

 least as regards sulphur, which I have studied, there really exist 

 two perfectly distinct spectra — one composed of lines, the other of 

 bands, and both characteristic to the same degree. The first is ob- 

 tained by the disruptive discharge ; the second can be produced by 

 discharges of less tension, by the incandescence of sulphur in the 

 hydrogen-flame, and, finally, though with less distinctness, by the 

 absorption of sulphur vapour alone, 



1. The Electric Spectrum. — I enclose the sulphur in a glass tube 

 similar to M. Pliicker's, but not presenting metallic electrodes. Each 

 extremity of the tube is surrounded by a brass sheath, which is 

 heated by means of a lamp in order to vaporize the sulphur. When 

 we wish the electricity to pass, we connect the sheaths with the 

 poles of a coil or a Holtz machine ; and by influence the tube be- 

 comes as intensely luminous as if the electrodes penetrated the inte- 

 rior. When an excellent vacuum has been made in the apparatus 

 during the vaporization of a large portion of the sulphur, which di- 

 stilled outside, there need be no fear of the presence of a foreign 

 gas ; besides, when such a tube is cold, the electricity no longer cir- 

 culates ; and, on employing metallic electrodes, even a Geissler's 

 tube in which a vacuum has been made over boiling sulphur arrests 

 the spark perfectly. The following are the wave-lengths of the 

 middle of each band observed in the spectrum obtained in this manner 

 by heating moderately, and employing electricity of feeble tension : — 



