336 NINETEENTH CENTURY. FT. IIL 



a dark ground ; each strip will represent an image of the 

 slit, and the whole will be a continuous spectrum as before. 

 But now suppose one set of waves to be wanting ; take out 

 one of your strips and you will have a dark space. This 

 represents one of the black lines in the spectrum where a 

 dark image of the slit is thrown, and if you take out those 

 which correspond to the lines in the sun spectrum No. 2, 

 you will have an illustration of ' Fraunhofer's lines.' 



Fraunhofer measured these black lines with the greatest 

 care, and he found that in every ray of sunlight they came 

 exactly in the same places. Then he tried the light of the 

 moon and Venus ; still the black lines were the same, for 

 these planets, as you know, only shine by the light of the 

 sun. But, when he turned his telescope to the stars and 

 caught their light, he found a difference. There were dark 

 lines in the star-spectrum, but they were not all in the same 

 place as those in the sun-spectrum, as you will see if you 

 compare No. 2, Plate I., with the star-spectrum, No. 5, in 

 which the lines seen on the right-hand side of the solar 

 spectrum are entirely wanting. 



Fraunhofer, therefore, argued in this way : If the black 

 spaces were caused by some of the waves being stopped in 

 coming through our own atmosphere, they would be the same 

 in any spectrum wherever the light came from. But as 

 these dark spaces are different in the starlight from what 

 they are in the sunlight, there must be some real difference 

 between the light oj the sun. and the light of the stars before 

 it comes to us. This was the first step in the study of the 

 heavenly bodies by means of spectrum analysis. 



' Experiments on the Spectra of different Flames, 

 1822. For more than forty-five years these black lines re- 

 mained a complete puzzle to all who studied the spectrum, 

 but in the meantime Six John Herschel, Mr; Fox Talbot, 



