176 ANNUAL EEPOKT SMITHSONIAN INSTITUTION, 1913. 



two dark lines of absorption, where light of the arc spectrum was 

 taken aAvay. Similarly, in the spectrum of iron, a great number of 

 bright lines are found in the green ; and if iron vaj^or is interposed 

 between an electric arc light and the slit of the spectroscope, a great 

 number of absorption lines will be found at the corresponding places. 

 Also in the spectra of the sun and of many of the stars there occur 

 dark lines corresponding exactly in place to the bright lines of the 

 spectra of the chemical elements found upon the earth's surface. 

 From these indications it is clear that these chemical elements exist 

 as vapors in the substance of the sun and stars. The number of 

 chemical elements in the sun and stars is so considerable and the num- 

 ber of their spectrum lines is so great, that the solar and stellar 

 spectra are thronged with dark lines, so that it takes the most exact 

 knowledge of the positions of the lines to insure for them a correct 

 interpretation. 



But in recent years a great deal more has been learned by the aid of 

 the spectroscope in regard to the sim and stars than of their mere con- 

 stitution, for it is found that although the spectrum lines occur almost 

 exactly in the same position in the spectra of the heavenly bodies 

 that they do in the spectra of the laboratory, yet there are slight and 

 very significant deviations of position which are attributable to the 

 motion of the heavenly bodies to or from the earth. For, just as in 

 the whistle of a locomotive, there is a sharping or flatting of the pitch, 

 depending upon whether the locomotive is coming toward the ob- 

 server or going away from him, so in the light of the stars there is a 

 displacement of the spectrum lines toward the violet or toward the 

 red, according as the star is approaching toward or receding from the 

 earth. One may go even farther, and say that there is a difference in 

 the position of the spectrum lines of the sun according as we take 

 the light from one edge of the sun or the other. For one edge is ap- 

 proaching the earth by virtue of the rotation of the sun while the 

 other is receding. It is also shown that the position of the spectrum 

 lines depends upon the pressure of the gases in which they are pro- 

 duced, so that it is possible to determine by exact measurements the 

 pressures under which the gases lie in the sun and stars, although 

 these are so extraordinarily remote that it takes light minutes or 

 years to reach the earth from them. Finally, it has been shown by 

 Zeeman that the form of the spectrum lines of the chemical elements 

 differs according to whether the light is produced in a magnetic field 

 or not. Accordingly it is possible to determine from measurements 

 of the solar spectrum whether magnetic fields exist in the sun, and, 

 if so, to what intensity they rise. 



All these kinds of measurement, which depend upon extremely 

 slight displacements of the spectrum lines, evidently require that 

 great accuracy shall be obtained in the determinations of the positions 



