448 



SCIENCE. 



[N. S. Vol. XXIII. No. 



It includes about 13,000 double stars north 

 of 31° south declination. 



Part I. consists of a tabular catalogue, in 

 quarto form, giving in order of right as- 

 cension the position, first measures, ob- 

 server and other data, in eleven columns. 



Part II. contains notes on the stars, 

 selected measures (up to date of print- 

 ing), showing motion or otherwise, present 

 relation of the components, and references 

 to all published measures. 



This part will include some 10,000 un- 

 published observations made at the Yerkes 

 Observatory in the last five years, and 

 about as many more furnished principally 

 by Messrs. Aitkin and Hussey of the Lick 

 Observatory and Eric Doolittle of the 

 Flower Observatory. 



It is hoped that the work may be pub- 

 lished during the year 1906. 



Photographic Photometry of Rapidly 



Changing Variable Stars: J. A. Paek- 



HURST and F. C. Jordan. 



The light-curves of certain rapidly 

 changing variables were determined from 

 a series of short exposures on a single plate, 

 taken with the twenty-four-inch Yerkes re- 

 flecting telescope. In order to utilize fully 

 the advantages of the method: (1) the ex- 

 posures were made short, so as not to 

 smooth out the light-curve; (2) a number 

 of comparison stars were used; (3) diam- 

 eters of the focal images were measured to 

 0.001 mm. under the microscope. 



The variable stars investigated by this 

 method were: 



star. 

 D Oephei, 

 W Ursw Majoris, 

 RW Tauri, 

 Hi- IdO.'t Cygni, 

 82 Cassiopeice, 



Results.— (1) ,U Cephei. The well- 

 known curve was reproduced, with average 

 residuals of ± 0.04 magnitude. .in( 2) W 



Ursw Majoris. The curve is similar to that 

 drawn by Miiller and Kempf ; with correc- 

 tions to their ephemeris for epoch 1777, 

 + 2-"; epoch 1945, -{-13™ (3) RW 

 Tauri. The range is 3^ magnitudes, the 

 most rapid change at the rate of three 

 magnitudes per hour; the correction to 

 Pickering's ephemeris at epoch 2598 is 

 -f 38°^; epoch 2602, -f 37™. (4) 14. 1904 

 Cygni. The range is 0.7 magnitude, the 

 period is i^ 1™ 26^.4 ; the curve resembles 

 the 'cluster type.' (5) 32 Cassiopeim 

 (suspected variable). No variation was 

 found beyond accidental errors of meas- 

 urement, the mean residuals for the four 

 plates, each covering the greater part of the 

 suspected period, being ± 0.05, ± 0.05, 

 ± 0.06 and ± 0.04 magnitudes. 



Spectrograms taken at Daroca, Spain, 

 August 30, 1905 (U. S. Eclipse Expedi- 

 tion) : S. A. Mitchell. 

 Five spectrographs were employed ; three 

 gratings and the other two alike having a 

 dispersion of one weak prism. Weather 

 conditions were perfect. Results with the 

 two larger instruments are as follows: 



1. Parabolic grating, diameter four 

 inches, with 14,438 lines to the inch and 

 a focal length of five feet. The dispersion 

 of this instrument is about the same as the 

 Bruce three-prism spectrograph of the 

 Yerkes Observatory and the Mills spectro- 

 graph of the Lick Observatory. The dis- 

 tance from -Dj to H is almost exactly seven 

 inches, the total length of the photographed 

 spectrum being 9.5 inches. The definition 

 is excellent throughout the whole length of 

 the flash, which extends from D^ to A3300, 

 and shows a very great number of lines. 



The spectrum taken near mid-totality 

 shows some interesting coronal rings. The 

 green 'eoronium' ring appears very plainly 

 and two rings in the extreme ultra-violet 

 are just as prominent on the photograph 

 as the green ring. As the plate used has 



