MOUNT WILSON OBSERVATORY. 237 



(3) Estimates from the visibility curves point to the disappearance of the 

 fringes at between 30 and 40 feet in the case of y Andromedae, and between 

 40 and 50 feet in that of a Arietis. 



(4) The following stars have all shown high visibilities of the order of 50 



per cent or more. No correction has been made for the factor of seeing. 



a Canis Majoris. a Andromedae. y Orionis. 



a Canis Minoris. a Pegasi. 7 Cassiopeiae. 



o Geminorum. /3 Leonis. « Cygni. 

 a Ophiuchi. 



(5) The stars a Aquilae, a Leonis, and a Virginis show practically full visi- 

 bility under good conditions of seeing. For a time it was thought that the 

 readings on a Lyrse fell below those on a Aquilse, but full visibility has since 

 been observed on several occasions. 



The use of an auxiliary interferometer with one variable aperture for the 

 purpose of matching the visibilities of two sets of fringes and for measurement 

 of the seeing through disappearance of the comparison fringes has proved to 

 be successful when the seeing is good. The principal set of fringes can be 

 matched readily with the auxiliary fringes, and the readings showing the 

 vanishing-point of the latter remain nearly constant during the night. Under 

 conditions of bad seeing, however, the auxiliary and the main set of fringes 

 have been found to vary independently of one another, the change sometimes 

 amounting to as much as 50 per cent in a very short interval. On such nights 

 direct eye estimates of visibility as determined through experience are used, 

 and the seeing is judged from the character of the images themselves. 



An examination of the diffraction images of a star given by the beams of 

 light from the mirrors of the interferometer set from 8 to 20 feet apart, and a 

 comparison with the images formed when they are less than 4 feet apart, 

 yields some information on the question of seeing with the use of large aper- 

 tures. The central disks are about 20 per cent larger in the case of the 

 greater distance and in appearance they are somewhat more diffuse. On 

 a night of uniform seeing the movement of the image from its central position, 

 however, is very nearly the same for the range of from 8 to 20 feet in the 

 separation of the mirrors. 



RADIOMETRIC OBSERVATIONS OF STARS. 

 Measures of stellar radiation with thermo-couples were begun by Mr. 

 Pettit and Mr. Nicholson at the 60-inch telescope in October 1921, and at the 

 100-inch telescope in December. On 14 observing nights, 236 complete 

 observations were made on the total radiation from 64 stars ranging from 

 — 1.6 to 9.8 visual magnitude. In the preliminary work the galvanometer 

 deflections were read visually from a scale, but since March 1 the deflec- 

 tions have all been registered photographically on a moving plate. The 

 thermo-couple cell is mounted in the Newtonian plate-holder of the 100- 

 inch telescope, the observations being made by shifting the star from one 

 junction to another with one of the guiding-screws. Suitable contrivances 

 have been provided for placing a water-cell 1 cm. thick in the path of light 

 from the star, and for projecting an image of an electric light upon the thermo- 

 couple for testing purposes. A cable 135 feet long connects the thermo-couple 

 at the Newtonian focus with the D' Arson val galvanometer in the basement, 



