248 CARNEGIE INSTITUTION OF WASHINGTON. 



single scale i)late for each area will be corrected and strengthened 

 by those for the adjacent areas of the same zone, which can be made 

 available for the purpose by the zonal comparison plates. When the 

 scales for each group of three areas have been combined into a final 

 mean for the middle area of the group, the two plates of long and short 

 exposure for that area will be used to extend the mean scale to the 

 fainter stars. By this means it should be possible to reach a hmiting 

 magnitude of 16 or 17, which is but Uttle short of that for the photo- 

 graphic investigation. The status of the observations is indicated by 

 the last two columns of the table on page 246. 



A Mkthod foh Determining the Color of a Star. 



An expeditious and convenient method of determining the color of a 

 star has been much desired. Estimates based upon spectral classifica- 

 tion do not include, to an appreciable extent, the influence of intrinsic 

 luminosity nor of any color-change produced by a possible scattering 

 of fight during its passage through space. Both color-index and 

 effective wave-length include these factors; but color-index requires 

 a knowledge of both photographic and photovisual magnitudes on the 

 international scale, and hence a good deal of observational labor, while 

 effective wave-length involves the use of an objective grating w^hich, 

 for a large instrument at least, is cumbersome and inconvenient. 



Preliminary experiments by Mr. Seares indicate that the ratio of the 

 exposure-times necessary for the blue and yellow light of a star to pro- 

 duce images of the same size can be used as a measure of its color and 

 that the requisite observations are very quickly made. An isochro- 

 matic plate is used, with a yellow filter for the yellow image, and with- 

 out filter for the "blue" image, all exposures being on the same plate. 

 With suitable precautions for the elimination of various systematic 

 errors, the uncertainty in the result from a single plate including two 

 values corresponds to about 0.06 mag. in the color-index. The 

 greater redness of stars of high luminosity, spectral type remaining the 

 same, is easily seen in the results from even a single plate. 



Distribution of Color in Nebula. 



Experience in various fields of investigation would lead us to antici- 

 pate a lack of homogeneity in the distribution of the glowing gases of a 

 nebula, and in many instances differences of distribution have actually 

 been detected. For example, it has been kno\^Ti for some years that 

 the distribution of the violet radiation at X3727, the blue-green radia- 

 tion of Hj3 and the chief nebular lines, and the red radiation of the 

 Ha is very different in different parts of the Orion nebula. Many 

 objects, however, are too faint for spectroscopic examination and 

 special methods of investigation are required. Some preliminary 

 experiments by Mr. Seares with color photographs promise interesting 

 results. Color-sensitive and ordinary plates combined with filters 



