370 PROCEEDINGS OF THE AMERICAN ACADEMY. 



asbestos paper and wrapped with a helix of iron wire, through which a 

 heavy current was passed. The ends of the tube were closed with plates 

 of glass, which were of such a thickness that they served in place of the 

 compensating plate, which is usually placed parallel to the transparent 

 mirror. The tube was exhausted by means of a mercurial pump, with 

 which it was kept in connection throughout the experiments in order to 

 remove the occluded hydrogen which was given off as soon as the sodium 

 was heated. This occluded gas gave a good deal of trouble, and the 

 behavior of the fringes seemed to indicate that a small amount of the gas 

 became entangled in the sodium vapor, no amount of pumping appearing 

 to remove it entirely. Its effect on the fringe shifts will be discussed 

 presently. 



The Dg line of helium was so strong in the tube used that no color 

 screens were necessary, the fringes being as distinct as those obtained 

 with sodium light. Owing to its proximity to the absorption band, very 

 large fringe displacements were obtained with a small amount of com- 

 paratively rare sodium vapor. 



In the electrically heated tubes the sodium vapor showed no tendency 

 to form the non-homogeneous equivalent of a prism, and the fringes con- 

 sequently remained straight, even after very great displacements. Two 

 observers were necessary, one counting the helium fringes as they passed 

 the cross-hair of the telescope, and the other noting the transit of the 

 fringes formed by the light from the monochromatic illuminator. If 

 the light used was on the same side of the D lines as the helium line, 

 the two sets of fringes were shifted in the same direction ; if on the 

 other side, the shifts were in opposite directions. In addition to com- 

 paring light from the monochromatic illuminator with the helium light, 

 a very careful comparison was made with the green mercury line, since 

 this light was to be used in the absolute determinations. 



At first very discordant results were obtained, but the source of error 

 was soon discovered to lie in the heating, and consequent ex2:)ansion of, the 

 base bar of the interferometer by the radiation from the sodium tube. The 

 instrument was then covered with layers of cotton and asbestos board, and 

 consistent series were at once obtained. There was still a source of 

 trouble which for a long time remained undiscovered. On heating the 

 tube it was found that the relative shifts changed as the temperature 

 rose, becoming constant after a shift of about a dozen helium fringes. 

 Repeated heating of the sodium to a high temperature combined with 

 pumping lessened the discrepancies, but the trouble was never completely 

 eliminated. It is now believed that a small amount of hydrogen is held 



