MOUNT WILSON SOLAR OBSERVATORY. 187 



Some parallax plates and a few photographs of Mars and Saturn were 

 also taken with the same arrangement. The work of Professor Barnard on 

 these planets is described below. 



Forty-eight plates, completing the program of the Kapteyn selected areas, 

 were taken by Mr. Fath. An examination of these plates for nebulse leads 

 to the conclusion that, if the plates are fair samples, there are about 164,000 

 spiral nebulas in the entire sky which could be photographed with an expo- 

 sure of one hour with the 60-inch reflector, using Lumiere Sigma plates. 



A series of plates was taken by Mr. Fath through red and blue ray filters 

 of certain regions of the Milky Way. The purpose of the investigation was 

 to detect, if possible, any loss of light in space. The method adopted re- 

 quired the assumption that a large proportion of the stars should be of the 

 same spectral type. This seemed to be the case from an investigation by 

 Pickering, who found that the A-type stars condense toward the Milky Way 

 and that the proportion of A type to all other types increases as the stars 

 become fainter. 



The plates taken seem to show definitely that the fainter stars are redder 

 than the brighter ones. On the assumption that the type of spectrum was 

 the same throughout, this would indicate a loss of blue light of the fainter 

 stars due to their greater average distance. This explanation is not satis- 

 factory, owing to other results noted below. 



Photographs taken for the Harvard College Observatory include one with 

 equivalent exposures on the regions surrounding Nova Geminorum No. 2 

 and the region containing the Harvard Polar Sequence, and several of 

 clusters, to be studied by Professor Bailey. 



Photographic Photometry. 



Although the investigations in photographic photometry with the 60-inch 

 reflector were interrupted by the absence of Mr. Scares during the months 

 April to July, the following results may be reported: 231 plates have been 

 obtained, chiefly for further investigation of the correction to the observed 

 magnitudes depending upon the distances of the stars from the axis of the 

 instrument, and for an independent determination of the magnitude scale of 

 the Harvard Polar Sequence; 12 plates are of the Algol variable RR Dra- 

 conis, which is of unusual interest because of its range of variation, extend- 

 ing from magnitude 9.70 to 13.50. With but few exceptions the photographs 

 have been measured and fully reduced. 



The distance error varies proportionally with the distance from the axis. 

 For a given distance it also depends on the size of the photographic image. 

 As previously reported, its effect is such as to increase the brightness of the 

 brighter stars ; for fainter stars its value is zero. Further investigation has 

 shown that for the faintest images on the plate the effect is negative, that is, 

 the observed is less than the real brightness. 



The different values of the correction required for certain plates by stars 

 of the same brightness and at the same distance from the axis referred to in 



