MOUNT WILSON OBSERVATORY. 239 



spectral type B2. The heat index was then 7.7 mag. and the water-cell 

 absorption 1.28 mag. It was observed shortly after maximum on June 27, 

 1922, when the visual magnitude was 4.5 and the deflection greater than that 

 of Vega. The heat index was then 4.6 mag., and the water-cell absorption 

 1.46 mag. Between these two dates the visual brightness had changed 4.4 

 mag., while the variation in total radiation was only 1.3 mag. Similar results 

 were obtained from observations of R Aquilae at minimum and maximum. 

 Ten stars of class Md have been studied, but all except o Ceti and R Aquilae 

 near minimum of light. Some giants and dwarfs of type K 1 have been 

 observed, but little if any difference in heat index has been noted. The M- 

 type dwarfs have not as yet been investigated. 



In addition to the variable stars of long period which are being observed 

 at different phases of light, Algol and the Cepheid variables 8 Cephei and 

 rj Aquilae are also under observation. 



The transmission of the atmosphere, which plays an important part in 

 comparisons of the total radiation of the stars, has been studied for stars of 

 different spectral classes. Observations on Vega, Ai'cturus, and a Herculis 

 between air-mass 3 and the meridian show that the corrections follow the 

 secant law, and that the constants are distributed according to spectral class, 

 being smaller for a Herculis and greater for Vega. 



BOLOMETRIC OBSERVATIONS OF STELLAR SPECTRA. 



As early as 1916 the Director suggested to Dr. Abbot, of the Smithsonian 

 Institution, the possibility of utilizing the great hght-gathering power of the 

 100-inch reflector for bolometric observations of stellar spectra similar to 

 those which for many years past he had been making of the solar spectrum. 

 The project seemed feasible to Dr. Abbot, in view of improvements which he 

 had in mind for increasing greatly the sensitiveness both of the bolometer and 

 the galvanometer. Accordingly, he undertook the preparation of the spectro- 

 bolometric apparatus which, after a long delay occasioned by the war, was 

 completed in the spring of 1922. The actual construction of the coils and 

 needle system of the galvanometer and of the sensitive parts of the galvan- 

 ometer was carried out by Dr. Abbot's colleague, Mr. L. B. Aldrich, who has 

 shared in all of the observational work. 



The galvanometer is in vacuum and is surrounded by a special magnetic 

 shield of 16 concentric cylinders of iron prepared by Dr. Elihu Thomson. This 

 shield has been found to be remarkably efficient in use. The control of the 

 needle was at first effected by the introduction of small wire magnets between 

 the first and second shields, but later the inner shield was removed and 5 large 

 bar magnets were used above and below the apparatus. These were found 

 to be just sufficient to exercise the necessary control. The needle system 

 consists of a single group of 8 thin tungsten-steel wires, each 0.9 mm. long, 

 fastened to a glass stem 13 mm. long, which carries a platinized mirror of the 

 thinnest microscope cover-glass 0.6 mm. square. The beam of light used to 

 illuminate the scale descends the axis of the galvanometer, is reflected upon 

 the small mirror by a prism, and returns vertically to a photographic recorder 

 at a distance of about 6 meters. A number of especially brilliant experimental 

 lamps prepared at the Nela Research Laboratory make it possible to use this 

 long scale-distance. 



