RECENT ADVANCES IN SCIENCE 285 



many cases orbits have been calculated both on the supposition 

 of uniform brightness of the discs and also on those of a darkening 

 towards the limb, such as is found in the case of the sun. The 

 evidence in favour of such a darkening does not appear very 

 convincing. In the final chapter is given a general discussion 

 of the results obtained. 



The Measurement of the Heat from the Stars. — In Pub. Ast. 

 Soc. Par. vol. 26, p. 169, 1914, a preliminary account was 

 given of some experiments on the measurement of radiation from 

 stars made with the Crossley reflector of the Lick Observatory, 

 by W. W. Coblenz, of the U.S.A. Bureau of Standards. Further 

 particulars of the methods employed, and of the various prob- 

 lems which the perfection of the method may be expected to 

 assist in solving, are given in Pop. Set. Monthly, vol. 86, p. 

 432, 191 5. The method adopted was to focus the rays from 

 the star on a specially constructed very delicate thermo- 

 element, placed in a glass cell with a fluorite window, at a 

 very low pressure. The deflection produced in a delicate 

 galvanometer connected with the cell gave a measure of the 

 total energy received. It was found that red stars gave a 

 much larger deflection than blue ones of the same visual mag- 

 nitude. Some idea as to the distribution of energy in the 

 spectrum was obtained b}^ measuring the deflection when the 

 radiation was passed through a cell of water. The results 

 obtained and the information which they provide as to the 

 temperatures of different types of stars have been inde- 

 pendently discussed by Keivin Burns and P. W. Merrill (Pub, 

 Ast. Soc. Pac. 27, p. 1 10 and p. 120, 191 5). It is found that 

 the results enable the temperatures of the later-type stars 

 (types G — M) to be assigned with considerable accuracy, but 

 that the temperatures of stars of types B and A are more un- 

 certain, owing to the fact that they are found not to radiate 

 as black bodies. The evidence, however, tends to indicate that 

 type A stars are of a higher temperature than type B. From 

 type A to type M the average temperature steadily decreases 

 from about 8,ooo° to about 2,500°. The temperature of the 

 Sun (type G) is generally supposed to about 6,ooo°, in fair 

 agreement with the preceding result. These first experiments 

 were admittedly experimental, and we may hope that, with 

 the further perfecting of the work, very valuable results will 

 be obtained. 



19 



