216 CARNEGIE INSTITUTION OF WASHINGTON. 



parison with terrestrial spectra. As is well known, the spectral lines at the 

 sun's limb, when compared with those at the center, are affected in two ways : 

 first, they are in nearly all cases shifted toward the red by slight amounts 

 which vary from line to line; second, they are in general wider and more diffuse 

 in character and change their appearance in a variety of ways. No adequate 

 explanation of the cause of these effects has as yet been found. 



Mr. St. John and Mr. Babcock have undertaken the investigation of this 

 subject by several different methods, using both grating spectrographs and 

 an interferometer. The instrumental difficulties encountered in obtaining 

 thoroughly reliable measurements of the small displacements involved make 

 such independent observations especially desirable, the range in the differ- 

 ences between different spectrograms being sometimes four or five times as 

 great at the limb as at the center of the sun. The 150-foot tower telescope 

 and 75-foot spectrograph, and the Snow telescope and 30-foot spectrograph 

 have been used for the grating spectra, while the interferometer work has been 

 carried on wholly with the Snow telescope. 



With the latter instrument two series of photographs are now being secured. 

 A new set of totally reflecting prisms has been mounted on the 30-foot spectro- 

 graph, and these make it possible to photograph strips of spectrum, two 

 from the center of the sun, two from one limb, and one from the opposite 

 limb. A rotating sector reduces the intensity of the light from the center, so 

 that all exposures are simultaneous. Observations are made at varying dis- 

 tances from the sun's limb along both the equatorial and the polar diameters. 

 To eliminate so far as possible instrumental errors, the spectrograph and prism 

 system are reversed with respect to the sun's image between exposures, the 

 prism system is reversed with respect to the spectrograph, and atmospheric 

 lines are used to check the results. In this way 30 spectrograms, each con- 

 taining 4 or 5 exposures, have been taken, covering the region X X4900-6400. 



The results of the measures, which are as yet incomplete, indicate fairly 

 close agreement with those obtained from photographs taken with the 150- 

 foot tower telescope and 75-foot spectrograph, primarily for the study of the 

 solar rotation. 



A second series of observations on the limb effect is being made by Mr. 

 Babcock with the aid of an interferometer, the small diaphragm of the etalon 

 being placed at various positions along the sun's polar radius. Differential 

 measurements are made between the spectrum from the center and from each 

 of these points along the radius. The atmospheric lines are used to eliminate 

 instrumental changes which may occur during the successive exposures. 

 While this method is not very well adapted to points very near the limb, 

 it is extremely useful along the greater part of the sun's radius. The few 

 measurements so far obtained are in fair agreement with those from the 

 method in which a grating is used. 



The question has been raised by Evershed whether the amount of the dis- 

 placement is not different at the north and south limbs of the sun. From 

 the extensive results obtained as a by-product of the work on solar rota- 

 tion, Mr. St. John concludes that the difference, limb minus center, is the 

 same, within the limits of error, at corresponding points of the two hemi- 

 spheres, as well as within the same hemisphere. The table on the following 

 page shows his results for the differences in the limb displacement, southern 

 minus northern hemisphere. 



