134 Transactions of the American Institute. 



surface, but which, reflecting the popular idea that the sun must be 

 habitable, maintained that his true nucleus must be cold and dark, 

 and was surrounded, at a distance of 4,000 miles, by a cloudy stratum 

 from which the light emanated, and in which were the cavities through 

 which the dark nucleus became visible. But this theory, based as it 

 was on insufficient data, though accepted by astronomers as the best 

 attainable by the telescope, was evidently inadequate to explain the 

 facts, and a little more than fifty years later it gave way to newer and 

 more probable theories, founded on facts gathered by a more wonder- 

 ful instrument than the telescope. It is my purpose to spend the hour 

 which you so kindly accord to me in an exposition of the principles 

 upon which this new spectroscopic method is based, and a rapid 

 review of the results which have been reached by it in solar physics. 



The spectroscope was first brought into notice as an instrument of 

 scientific research by Professors Bunsen and Kirchkoff, in 1860. As 

 its name indicates, it is an instrument for viewing spectra. We can 

 best elucidate its character by attempting to answer three questions : 

 1st. "What is a spectrum ? 2d. By what means may spectra be best 

 produced and observed ? and 3d. To the solution of what problem 

 may the spectroscopic method be applied ? 



In the first place, then, what is a spectrum ? The statement of a 

 few elementary optical principles will enable us to answer the ques- 

 tion. Light, from whatsoever source, which, so long as it traverses 

 the same medium, passes in a straight line, is bent or diverted from 

 its course, i. e., is refracted, whenever the medium changes. But 

 not only this; so long ago as 1672, Newton observed that an ordin- 

 ary light-ray was not only bent from its course on entering a denser 

 medium, but was also broken up into several other rays, all of them 

 colored. Seven of these colors were distinguished by him, viz., red, 

 orange, yellow, green, blue, indigo and violet ; and he demonstrated 

 that, since the refrangibility of each color was different — that of 

 violet being greatest, and of red least — these colors must be separated 

 from each other by refraction. If, therefore, a beam of light from a 

 round opening be refracted by passing it through a prism, it will be 

 not only bent out of its course, but will be decomposed, and will 

 produce an elongated, rainbow-like image upon the opposite wall, 

 made up of the seven colors above named, in the order indicated. 

 This long and brilliantly-colored image is called spectru?n. [The 

 lecturer here illustrated these statements experimentally by means of 

 an electric light from a Bunsen battery of 100 cells. The intensely 

 heated carbon points were first projected on the screen. A lens 



