March 17, 1005.] 



SCIENCE. 



423 



For these reasons, the writer has pre- 

 pared the present modification of his 

 method, and has added tables suitable for 

 the rednction of isolated groups of plates 

 made at any declination distant more than 

 15° from the pole. 



The tables will appear in a short time as 

 one of the Columbia 'Contributions.' 

 Beccnt Researches of the Henry Draper 



Memorial: Edward C. Pickering. 



A photograph was shown of the spectrum 

 of A Cephei, Avhich has a spectrum closely 

 resembling that of ^ Puppis and contains 

 the second series of lines probably due to 

 hydrogen. A method of observing occulta- 

 tions photographically was explained and 

 a printed enlargement of a photograph of 

 the emersion of -q Virginis, on December 

 28, 1904, was shown, which had been taken 

 by Mr. Edward S. King. A rotary motion 

 was given to the plate-holder so that the 

 star gave a continuous trail, the time being 

 indicated by a motion given to the plate at 

 regular intervals. It appeared that the 

 star increased in light during emersion for 

 about a third of a second. The principal 

 portion of the paper was devoted to the 

 study of the distribution of the stars ac- 

 cording to their spectra. The results were 

 based upon an examination of the Draper 

 plates by Mrs. Fleming. About 6,000 

 plates have been inspected, each showing 

 on the average the spectra of a thousand 

 stars, with small dispersion, and on these 

 all that were peculiar were noted. Using 

 a larger dispersion, about thirty thousand 

 stars have been classified and catalogued. 

 Visual counts of the number of stars in 

 different parts of the sky have little value 

 owing to the uncertainty of the magnitudes. 

 The same might be said of a classification 

 of photometric magnitudes of the stars 

 tal <en as a whole. It was shown that stars 

 of each class of spectrum should be con- 

 sidered by themselves, as the distribution 

 differs widely. Thus, the Milky Way ap- 



pears to consist wholly of stars of the first 

 type. The helium or Orion stars have a 

 different distribution, forming a Milky 

 Way of their own, mainly in Orion and 

 Argus. It was shown that the classifica- 

 tion of the variable stars, proposed by the 

 writer in 1880, was confirmed by their 

 spectra, and that the latter formed a means 

 of determining the class, in some cases, 

 even better than the light curve. 



Note on Two Variable Star Catalogues: 

 J. G. Hagen. 



Father Hagen presented to the meeting 

 some specimen pages of two Catalogues of 

 Variable Stars, now in preparation, one by 

 the Astronomische Gesellschaft and the 

 other by Professor E. C. Pickering. Of 

 the former catalogue seven pages had been 

 printed for presentation at the astronom- 

 ical congress in Lund, last September. A 

 copy had been sent to Father Hagen in time 

 for the Philadelphia meeting, but Dr. 

 Mueller's report at Lund came, unfortu- 

 nately, too late. For this reason only those 

 features of the catalogue could be men- 

 tioned that presented themselves to the 

 reader of these seven pages. 



More definite explanations could be given 

 on the other catalogue, since Professor 

 Pickering was himself present, and had 

 shown one specimen page previously to 

 several friends. His catalogue will be a 

 'Bibliography of the Variable Stars,' with 

 the lists of the known maxima and minima, 

 and the sources from which they were 

 taken. Father Hagen brought out the 

 fact that the two catalogues will supple- 

 ment one another. The catalogue of the 

 Astronomische Gesellschaft will give exact 

 positions and elements of light variations, 

 with very condensed references to all ac- 

 cessible publications, on each variable star. 

 Professor Pickering's 'Bibliography' will 

 give fuller details of the spectra of the 

 variables stars from the rich material at 



