MOUNT WILSON SOLAR OBSERVATORY. 275 



INVESTIGATION OF THE IRON ARC. 



In the investigation of the relation between the iron spectrum and 

 the corresponding solar lines, and in the determination of standards 

 of wave-length, it was found that a detailed study of the iron spectrum 

 was required as a preliminary. One phase has been the examination 

 by Mr. St. John and Mr. Babcock of the changes in wave-length in 

 the arc at atmospheric pressure that depend upon the portion of the 

 arc used and upon the current strength. This involved a comparison 

 of the spectrum near the pole and at the center of the arc. In the 

 effort to carry the comparison to the third decimal, great difficulty 

 was encountered in obtaining plates free from instrumental shifts of 

 very small but determinable amounts. These were finally eliminated 

 by making the exposures rigorously simultaneous. The exposures thus 

 began and ended at the same moment, one being continuous and the 

 other rapidly intermittent. A similar arrangement was used in com- 

 paring arcs with different current densities. 



Comparisons have been made between the center and negative pole 

 which involve about 1,600 lines between X 2900 and X 6700. About 

 250 lines were found for which the wave-lengths at the negative pole 

 are greater by amounts exceeding 0.015 a than those at the center, 

 while for about 50 lines the wave-lengths at the negative pole are 

 less by amounts exceeding 0.015 a. The differences in wave-length 

 between the positive pole and the center are of the same sign but less 

 in magnitude. 



In an arc 6 mm. long, carrying a current of 6 amperes— that recom- 

 mended by the Wave-length Committee of the International Union 

 for Cooperation in Solar Research — the differences are of the order of 

 0.01 A between the center and a point 1 mm. from the negative pole. 

 In a 12-ampere arc the wave-lengths of the two classes of lines are 

 respectively longer by 0.007 1, and shorter by 0.012 a, than in a 

 4-ampere arc. Between arcs carrying 5 and 7 amperes, the displace- 

 ments of the lines showing positive shifts do not exceed 0.001 a, but 

 those with negative displacements still give easily measurable shifts. 



The reality of such displacements being open to question because 

 of the unsymmetrical widening of the lines, the point has received 

 particular attention. The exposure time at the pole was decreased 

 until the lines were weaker and narrower at the pole than at the center 

 of the arc. The displacements were still large. Measurements with 

 the Hartmann micro-photometer gave displacements of the maxima 

 of the lines of the same order of magnitude as those found with the 

 measuring microscope. 



The displacements are indirectly related to pressure shift. The 

 lines showing a longer wave-length at the negative pole are those 

 belonging to groups giving very large positive pressure shifts; those 

 showing a shorter wave-length at the negative pole are lines giving 



