MOUNT WILSON OBSERVATORY. 211 



OTHER INVESTIGATIONS. 

 THE GREEN AURORAL LINE. 



The preliminary observations of this interesting line by Babcock have 

 been considerably extended during the past year, and the interference method 

 has been applied to a precise measurement of its wave-length and to a study 

 of its width and brightness. The wave-length is found to be 5577.350 i. a. 

 with an uncertainty estimated to be not greater than 1 part in 1,000,000. 

 The width of the line is shown to be less than 0.035 a, a value which may be 

 even further reduced by subsequent observations with a large interferometer 

 recently constructed. Finally, the order of magnitude of the surface bright- 

 ness of the diffuse green auroral light in the background of the sky has been 

 found to be 10 ~^ times that of the green radiation from a mercury lamp. 



For the purpose of correlating the brightness of the persistent aurora 

 with the occurrence of certain solar and terrestrial phenomena, a small etalon 

 has been constructed and provided with a camera and automatic shutter. 

 This etalon is made of very thin plane-parallel glass, with cover-glasses 

 cemented over films of gold, and so has constant absorptive power. Since 

 it remains permanently in adjustment it is adapted to the use of observers 

 unfamiliar with the technique of interferometry. 



A consideration of the probable temperature of the upper levels of the 

 earth's atmosphere in connection with the limiting value of the line-width 

 indicates for the origin of the green line a gas whose atomic weight is at 

 least 4. 



THE VELOCITY OF LIGHT. 



Two alternative plans have been under consideration by Dr. Michelson 

 in connection with this investigation. In the first, which is a combination 

 of the methods of Fizeau and Foucault, the image of the slit is focused on a 

 concave mirror from which the light returns to the revolving octagonal 

 mirror, thence to the concave reflector of 30-foot focal length, and from this 

 to the distant station on the San Antonio ridge. In front of the first concave 

 mirror is placed a grid which corresponds to the teeth of the Fizeau wheel. 

 The chief difficulty encountered in this arrangement is the scattered light from 

 the mirror and grid. This may be overcome to a large extent by using the 

 sputter process on both mirror and grid (the bars in the latter being inclined 

 to avoid direct reflection). It is estimated that this form of apparatus will 

 furnish results considerably more accurate than those now in use if the 

 second arrangement about to be described does not furnish sufficient light. 

 The accuracy will depend upon the spacing of the grid, increasing in propor- 

 tion to the total number of intervals. If it is possible to make this as large 

 as 500, the order of accuracy may be estimated at 1 in 25,000. In view 

 of the faintness of the return image, however, it is doubtful if the light scat- 

 tered from a grid of this character would permit of sufficient accuracy in the 

 estimate of brightness. 



The second method, which is a modification of that of Foucault, seems on 

 the whole to be more promising and is certainly much simpler. In this 

 arrangement the return light is very much fainter than in the method already 

 described, but it is possible to eliminate the scattered light entirely. More- 

 over, the determination by this method depends upon observations of the 

 position of the return image and not upon estimates of brightness. The 

 expression for the intensity of the return image is 



/ = 6(f/Dr2 



