78 THE POPULAR SCIENCE MONTHLY. 



tions of incandescent hydrogen gas. It was also found by this ob- 

 server that the proper motions of some of the fixed stars in a direction 

 to or from the earth might be detected by means of the displacement 

 of their spectral lines, a method of research which was first enunciated 

 by Fizeau. Hitherto, in such applications of the spectroscope, the 

 body to be examined was viewed as a whole. It had not yet been at- 

 tempted to localize the use of this insti'ument so as to examine par- 

 ticular districts of the sun, as for instance a sun-spot, or the red flames 

 already proved by De la Rue to belong to our luminary. This appli- 

 cation was first made by Mr. Lockyer, who in the year 1865 examined 

 a sun-spot spectroscopically, and remarked tlie greater thickness of the 

 lines in the spectrum of the darker portion of the spot. 



Dr. Frankland had previously found that thick spectral lines cor- 

 respond to great jiressure, and hence the inference from the greater 

 thickness of lines in the umbra of a spot is that this umbra or dark 

 portion is subject to a greater pressure ; that is to say, it exists below 

 a greater depth of the solar atmosphere than the general surface of the 

 sun. Thus the results derived from the Kew i:hotoheliograph and 

 those derived from the spectroscope were found to confirm each other. 

 Mr. Lockyer next caused a powerful instrument to be constrixcted for 

 the purpose of viewing spectroscopically the red flames round the sun's 

 border, in the hope that if tliey consisted of ignited gas the spectro- 

 scope would disperse, and thus dilute and destroy the glare which pre- 

 vents them from being seen on ordinary occasions. 



Before this instrument was quite ready these flames had been an- 

 alyzed spectroscopically by Captain Herschel, M. Janssen, and others, 

 on the occasion of a total eclipse occurring in India, and they were 

 found to consist of incandescent gas, most probably hydrogen. But 

 the latter of these observers (M. Janssen) made the important obser- 

 vation that the bright lines in the spectrum of these flames remained 

 visible even after the sun had reappeared, from which he argued that 

 a solar ecli})se is not necessary for the examination of this region. 



Before information of the discovery made by Janssen had reached 

 this country, the instrument of Mr, Lockyer had been completed, and 

 he also found that by its means he was able to analyze at leisure the 

 composition of the red flames without the necessity of a total eclipse. 

 An atmosphere of incandescent hydrogen was found to surround oiir 

 luminarj', into whicli, during the greater solar storms, sundry metallic 

 vapors were injected sodium, magnesium, and iron, forming the three 

 that most frequently made their appearance. 



Here we come to an interesting chemical question. 



It had been remarked by Maxwell and by Pierce as the result of 

 the molecular theory of gases that the final distribution of any num- 

 ber of kinds of gas in a vertical direction under gravity is such that 

 the density of each gas at a given height is the same as if all the other 

 gases had been removed, leaving it alone. In our own atmosphere 



