in the Solar Spectrum. 501 



substituted for that of the clouds or of the sun, nor when co- 

 loured media are interposed in the passage of the rays before 

 or after their dispersion. In short, they may be produced at 

 •will by placing one or two plain mirrors near the extremities 

 of the aperture made in the shutter, and by giving them a 

 horizontal inclination of variable extent, such that the brilliant 

 image which they receive produces a spectrum which is partly 

 superposed upon that of the direct rays. This well deserves 

 to be made a class experiment, as a variety of that of the two 

 mirrors of Fresnel. It is much easier, does not require any 

 dehcate or costly apparatus, and extends simultaneously to all 

 the monochromatic zones of the spectrum. 



"As the band of light which is admitted into the prism 

 proceeds from different parts of the sky, it is easy to account 

 for the changes in the appearance of the longitudinal lines 

 according to the degree of transparency of the atmosphere. 

 The modifications which this system undergoes with the 

 changes of exterior illumination are explained by the unequal 

 distribution of diurnal light, which, during a state of greater 

 or less complete serenity of the air, varies with the distance of 

 the sun from the meridian. 



" These lines of interference have not been mentioned by 

 any author, at least as far as I can learn." 



In 1844', M. Ad. Erman seems to have had a glimpse of 

 the phajnomenon which I had discovered, but attributed it to 

 the accidental defects of the vessels which he employed, and did 

 not pay any attention to it*. It is but somewhat recently that 

 the horizontal bands have become the subject of investigation 

 by two experimenters, well-located for optical researches, 

 M. Zantedeschi at Venice, and M. Ragona-Scina in Sicily. 



M. Zantedeschi has published the results of his experi- 

 ments in a separate volume, which has not yet been reviewed 

 in France. C)nly a hundred copies of this work having been 

 printed t, it will not be out of place here to give its conclu- 

 sions, which confirm mine on almost every point. 



" The appearance of the lines of the solar spectrum is 

 in a necessary relation with the smallness of the aperture, 

 and with the interval which separates the prism from 

 it. At a constant distance the transversal lines appear with 



* On the Law of Absorption of Light by the Vapours of Iodine and Bro- 

 mine. — Comptes Rcndus,t. ii\\.p.8'^2. 



t Riccrchc Jisico-chimicofisiologiche sulla Luce; in 4to. Venice, 1846. 

 In this the author treats of the inHuence of solar light on the germination 

 of seeds, on tiie changes of colour |)roduced in organic and mineral sub- 

 stances by the action of light alone, on a new analysis of the solar spectrum; 

 in short, on the conversion (passage) of ponderable matter to the radiant 

 state (?) 



