NATtTRAL PHILOSOPHY. 123 



yellow, it will appear white or black, according as the body has, 

 or has not, the power of reflection ; which is equivalent to saying 

 that this flame alters the nature of colors, deepening the hues of 

 some, and extinguishing others. 



Take a spirit-lamp and put into it a piece of common salt ; the 

 wick will soon become saturated with sodium in solution; the 

 flame, in consequence, will be yellow, and all colors will assume 

 a monotonous white, black, or gray. It is only when this sub- 

 stance is in excess that we have the total extinction of colors, but 

 a flame less rich will produce a partial extinction, and this is the 

 reason why coloi's are at all visible by gas-light. It may be asked, 

 whencedoes illuminating gas derive this sodium? From the coal; 

 from the water with which the gas was washed ; it comes also 

 from matters employed in its purification, and probably even from 

 the atmosphere. 



The only hues which resist only slightly the yellow flame, are 

 furnished by the blue ; all the other colors are profoundly modi- 

 fied. Fortunately, the flames which serve as sources of light are 

 never saturated with sodium, hence the effects are greatly mod- 

 ified. 



The light from the burning of magnesium alone brings out the 

 various colors, both natural and aitificial, in the same hues as they 

 appear by daylight. The services of chemistry render, then, to 

 painting, not only colors more or less rich, but also it has endowed 

 it with a mode of lighting, whereby the painter may be able to 

 work at night without incurring mistakes or illusions. — Scientific 

 American. 



SPECTRUM OF AQUEOUS VAPORS. 



M. J. Jannsen has just communicated to the Academy of Sci- 

 ences a memoir " On the Spectrum of Aqueous Vapor." His 

 observations Avere made with an iron tube thirty-seven metres 

 long, filled with steam, under a pressure of seven atmospheres ; 

 the light was furnished by sixteen gas jets. The spectrum showed 

 five dark bands, of which two, well marked, answered to D and 

 A (Fraunhofer) , and reminded the observer of the solar spectrum 

 seen in the same instrument toward sunset. According to the 

 first comparisons made between the spectrum of steam and that 

 of solar light, it appeared that the group A, B (in great part, at 

 least), C, two groups between C and D, are due to the aqueous 

 vapor in the atmosphere. Another interesting result was given by 

 the spectrum. The spectrum was very dark at the violet end, 

 and brilliant in the red and yellow, showing that aqueous vapor is 

 very transparent to the latter rays, and suggesting that it will 

 appear orange-red by transmission, and redder, according to the 

 thickness of the layer. This result requires to be carefully veri- 

 fied, and, if established, will explain the redness always observed 

 at sunrise and sunset. He hopes soon to be able to pronounce 

 upon the existence or non-existence of aqueous vapor in the at- 

 mosphere of the planets and other stars : at present he can only 

 say, that it does not exist in the atmosphere of the sun. 



