692 PROGRESS OF ASTRONOMY FOR 1891 AND 1892. 



ligbt of the stars were measured. In the hitter case, tlie results will 

 differ with the color of the star, according' to the method of measure- 

 ment employed. This is a serious defect in the measures of the bright- 

 ness of the stars in catalogues hitherto published. Since the present 

 measures relate to rays of a single wave-length, the same result should 

 be obtained whether the method of comi)arison was by the photographic 

 plate, the eye, or the thermopile." 



Tlie Draper catalogue gives the ai)proxinui,te positions of the stai-s for 

 the year 11)00, with their reference numbers in the Bonn Dnrchmusterung- 

 and the Harvard Photometry; their class or' spectrum by letters; their 

 photographic magnitudes and the differences of these from the magni- 

 tudes of the Durchmusterung, the Argentine General Catalogue and 

 the Harvard Photometry. A well arranged table gives the details of 

 the measures of magnitude on the various j)lates on which each star 

 ai)pears. The whole sky to 25° south declination was photographed 

 twice with plates overlapping. 



Volume 26, part 1, of the Harvard Annals ^ives additional details 

 respecting the photogTa])hs, their measurement and reduction not con- 

 veniently included in the catalogue volume itself— a complete history 

 of the Draper Memorial. A point brought out in the various matters 

 discussed in this volume is the predominance of the first type spectra 

 in the Milky Way elsewhere referred to, and tlu^ systematic underval- 

 uing' of the brigiitness of the Galactic stars by about one-tifth of a mag- 

 nitude, by the ''Durchmusterung" and '• Uranometria Argentina" as 

 compared with the Harvard photometric and photographic magnitudes. 



A third volume is to follow devoted to the work of the 8-iiich Draper 

 telescope during the years 1889 and 1892 and to the discussion of stars 

 of i^eculiar spectra. 



A fifth ti/pe of stellar spectra. — Prof. E. ( '. Pickering has proposed to 

 class in a "fifth type" stars whose spectra resemble those of the stars 

 discovei'ed by Wolf and Kayet. In general, his photographic survey 

 has confirmed Secchi's fourfold division of stellar spectra, but many 

 stars in Orion and the neighboihood differ considerably from the 

 ordinary first-type stars, the additional lines, instead of being- faint as 

 in Vega, being nearly as intense as the hydrogen lines, while two 

 classes of objects, the i)lanetary nebuhe and the stars, the spectra of 

 Avhicli consist chiefly of bright lines, are left unprovided for. Prof. 

 Pickering points out the close similarity of the grouping of the lines in 

 these three classes and also the striking character of their distribution. 

 While stars of the second and third types are about equally divided 

 between the Milky Way and the regions remote from it, two-thirds of 

 the first-type star lie in or near the Milky Way and of the Orion stars 

 four-fifths are found in the Milky Way. 



A similar distribution of the planetary nebula* has long been recog- 

 nized, and Prof, Pickering shows that, of thirty-three stars known as 



