CELESTIAL CHEMISTRY. 267 



or from radiating objects, a spectroscope of direct vision is prefer- 

 able. Mr. Browning effects this by placing a dense flint glass prism 

 of 00 between two prisms of light crown of 22. Other opticians 

 follow the same principle, with variations of detail. Such spectro- 

 scopes act clearly, but they lose much of the dispersive power of the 

 chief prism. In Browning's pattern the loss is about 1. Although 

 employed by Secchi and other observers, spectroscopes of this kind 

 are by no means the best for astronomical research ; though amateurs, 

 who merely desire to see the principal lines which bright celestial 

 objects afford, will probably find them adapted to their purpose. 



One of the most interesting branches of spectroscope inquiry is the 

 absorption of certain portions of the spectrum by solutions. The fluid 

 can be put in a test tube ; but for many of these experiments a pris- 

 VH/lic cell is better. This consists of a rectangular glass cell, one 

 side of which is composed of a prism. When the solution to be ex- 

 amined is put in the cell, it forms a fluid prism ; and if the glass 

 prism and the solution prism correspond pretty closely in refractive 

 power, one undoes the refractive work of the other, and thus light 

 comes straight through the combination to the slit of the collimator. 



To view the spectra of gases, narrow tubes are employed, in which 

 the gas, in a highly rarefied state, is rendered incandescent by the 

 discharge of a HuhmkoriFs coil. 



ON THE OCCULTATION OF THE SPECTRUM OF A FIXED STAR. 



Mr. Huggins, in a recent communication to the Royal (Eng.) As- 

 tronomical Society, remarked that while all observations hitherto 

 negative the existence of an atmosphere to the moon, one had 

 occurred to him, as not yet made, which might add to the evidence on 

 the question. This was the disappearance of the spectrum of a star 

 when occulted by the moon's limb. It appeared to him that if the 

 moon possessed an atmosphere, it should show itself by producing 

 absorption bands, and also by a gradual fading of the spectrum instead 

 of a sudden disappearance. He also thought that, supposing an 

 atmosphci e highly charged with aqueous vapor, the red rays would be 

 least affected, and that end of the spectrum would be last to go, while 

 a clear but dense atmosphere would, by refraction, render the blue 

 rays most persistent. He therefore adjusted his spectrum apparatus 

 on the star s Piscium about three minutes before its occultation. He 

 could not speak positively as to any extra lines being introduced by 

 the moon's contact, the air not being in good condition, but certainly 

 no part of the spectrum disappeared before the other, the obscuration 

 occurring not in the direction of the length of the spectrum, but of 

 its breadth, cut ting off all the colors at the same time, and not by a 

 process of fading, but as if an opaque screen had been slowly intro- 

 duced, about i 2 Q- of a second being occupied in the occultation. This 

 experiment, therefore, further corroborated the general impression of 

 the absence of any appreciable lunar atmosphere. 



