496 HISTORY OF SCIENCE. 



a white heat. Our terrestrial atmosphere, in which now so few ele- 

 ments are found, must have possessed, when the earth was in a state 

 of fusion, a much more complicated composition, as it then contained 

 all those substances which are volatile at a white heat. The solar 

 atmosphere at this time possesses a similar constitution."* 



When the significance of the spectra of the heavenly bodies as re- 

 vealing their chemical constitution had been established by the classical 

 researches of Kirchhoff, the study of the spectra was entered upon 

 with great eagerness by several devoted and enthusiastic observers, 

 both in England and upon the Continent. Our countryman, Dr. 

 Huggins, was one of the earliest and most successful labourers in the 

 new field of discovery. 



The spectroscopic observation of the stars, extending to the accu- 

 rate measurement and comparison of the positions of their light or dark 

 with those of known substances, may easily be understood to be a task 

 of greater difficulty than the study of the sun lines. The image of a 

 star being but a point, when such an image falls upon the slit of the 

 spectroscope, the spectrum it produces is without appreciable breadth, 

 that is, it is merely a variously-coloured line, in which the eye is un- 

 able to perceive the extremely minute breaks which would represent 

 dark lines. It is necessary to widen out the image of the star upon 

 the slit, so that instead of a mere point it shall form a little line of 

 light parallel to the slit, and this is accomplished by making use of a 

 cylindrical lens. Then the instrumental appliances require great 

 nicety of construction, for the telescope must follow the apparent 

 motions of the heavens in such a manner that the position of the image 

 shall not be changed by a hair's breadth. Further, the apparatus must 

 include the means of observing simultaneously the spark or flame- 

 spectra of known terrestrial elements produced by the same train of 

 prisms. 



The earlier researches of Mr. Huggins on stellar spectra were un- 

 dertaken in conjunction with Professor W. A. Miller, and a memoir 

 published under their names appears in the " Philosophical Transac- 

 tions " of 1 864. In this paper the authors remark that the success of 

 Kirchhoff in determining by spectrum analysis the nature of some of 

 the constituents of the sun suggested to them the extension of the 

 same method of investigation to the fixed stars. They began their 

 experiments in January, 1862, the investigation upon which they en- 

 tered being one that had never previously been attempted or even 

 imagined. Nothing, indeed, previous to Kirchhoff's researches, could 

 have appeared more impossible than that men should ever attain to 

 any knowledge of the chemical constitution of the stars. A philosopher 

 of the present century had even pronounced that other knowledge of 

 the fixed stars than coul 1 be gathered from their motions was never to 



* As quoted by Roscoe in "Spectrum Analysis." 



