878 



SCIENCE 



[N. S. Vol. XXXII. No. 833 



the selenium photometer has been perfected so 

 that bright stars can be measured more accurately 

 than by either visual or photographic methods. 

 With a selenium cell attached to the 12-inch re- 

 fractor, an exhaustive study of the light-curve of 

 Algol has been made, with special attention to the 

 constancy of the light at maximum. It has been 

 found that Algol is a star of continuous variation, 

 the light-cur^'e showing both a secondary mini- 

 mum and changing light between minima. From 

 these observations new elements have been de- 

 rived, which indicate that the companion, which 

 has often been considered a dark body, in reality 

 gives more light than our own sun; and in addi- 

 tion is much brighter on the side which is turned 

 toward Algol, due presumably to the heating effect 

 of the intense radiation which is received from the 

 primary. A complete account of this work is to 

 be published in the Astrophysical Journal. 



Note On the System of Algol: R. H. Cuetiss. 



(Read by Professor Stebbins. ) 



Observations employed in a paper by the author, 

 published in the Astrophysical Journal, 28, 150, 

 suggest a variation in the period of revolution of 

 the center of mass of the eclipsing pair of Algol, 

 about the center of mass of the system. Tisser- 

 and's hypothesis as to the rotation of the line of 

 apsides of the orbit of the eclipsing pair is still 

 consistent with the facts. On the basis of this 

 hypothesis and Chandler's elements with Vogel's 

 well-known data, the period of rotation of Algol 

 must lie between days and 4.23 days with a 

 probable value equal to the light period. Reject- 

 ing the assumptions of Vogel and Tisserand, 

 limiting values of the oblateness of figure of the 

 eclipsing stars are computed, the identity of the 

 rotation and light periods being assumed. Cer- 

 tain constants of the system are computed on the 

 basis of Vogel's data. 



A neio Sixteen-and-one-fourth-inch Doublet: Joel 



H. Metcalf. 



This doublet has been constructed by the writer 

 and is now mounted at the Harvard Observatory. 

 The glass is of the general form of the Petzval- 

 Voigtlander type working with an aperture of 

 f 5.5. Its special features are ( 1 ) the use of 

 dense flint (Series O of Parra Mantois) instead 

 of the light flint usually employed in these con- 

 structions. This reduces the steepness of the 

 curves, the thickness ol the lenses and the re- 

 sultant curvature errors. The glass is very trans- 

 parent and free from veins as well as finely an- 

 nealed. (2) Astigmatism has been almost entirely 



eliminated so as to secure round star images suit- 

 able for measurement all over the field. This has 

 been done even at the expense of some flatness of 

 the field which has been corrected in another way. 

 (3) The careful elimination of coma. This has 

 been done by theory and the residual amount neu- 

 tralized by the spacing of the front and back 

 combinations by actual experiment. (4) The 

 spherical aberration of the second and fourth 

 orders has been eliminated by theory and the 

 careful spacing of the lenses of the front com- 

 bination. This has been done for the spectrum in 

 the region of the G line by the use of nearly 

 monochromatic light in testing. The fourth order 

 aberration has been eliminated by local polishing 

 or parabolizing the front surface of the first 

 crown lens. The errors remaining are a slight 

 curvature of the field which has been practically 

 eliminated by Professor Pickering by the use of 

 plates mechanically bent to the proper curvature. 

 The one remaining appreciable error in a field of 

 eight or ten degrees is the secondary color correc- 

 tion which is inherent in all lenses made of the 

 Oidinary crown and flint glass. In a lens of such 

 great absolute and relative aperture this error is 

 quite marked, as the enlargement of the images of 

 bright stars shows. Unless the use of other glasses 

 will inaterially reduce this secondary spectrum it 

 would seem useless to construct doublets of 

 greater size with such large relative aperture, for 

 the effect would be to increase the scale of the 

 picture without giving greater space-penetrating 

 power. 



Solar Prominences Photographed xoith the Rum- 

 ford Spectroheliograph of the Yerkes Observa- 

 tory: Frederick Slocum. (Read by Mr. Park- 

 hurst. ) 



This paper consisted of: (1) A series of pho- 

 tographs of a large quiescent prominence observed 

 from March 4 to April 28, 1910 (two rotations 

 of the sun). Maximum height 106" = 77,000 km. 

 Maximum lateral extent, 47° of the sun's limb = 

 57,000 km. (2) A series of photographs of a very 

 active prominence observed on March 25, 1910. 

 Beginning as a small insignificant cone, this prom- 

 inence developed rapidly, and in a few hours van- 

 ished at a height of 7'.4 = 320,000 km., after 

 having passed through a variety of fantastic 

 shapes. (3) Miscellaneous photographs of prom- 

 inences obtained at the Yerkes Observatory dur- 

 ing the past few years. 



An Instrument for Rapidly Solving Spherical Tri- 

 angles: F. W. Dyson. 



