POOR, THE FIGURE OF THE SUN 403 



data sufficient to find the scale value. The relative lengths of the various 

 diameters can, however, be most accurately measured, and the shape, and 

 variations in the shape, of the disk determined. 



RUTHERFURD PLATES. 



The Observatory of Columbia University has in its possession a series 

 of 139 solar photographs taken by Rutherfurd during the years 1860-74. 

 This series of plates may be divided into two groups; one group of plates 

 covering the years 1860-66, and a second group taken during the years 

 1870-74. The plates of the first series were made with a small lens; those 

 of the second group, with his thirteen-inch photographic objective, which 

 was completed in 1868. The earlier plates were simple photographs of 

 the sun without orientation-marks or data of any kind. In 1870 he began 

 to place orientation-marks on the plates, but even after that date fully half 

 of the plates lack this essential. In these four years, Rutherfurd took a 

 hundred plates, grouped as follows : 



1870, Feb. 16— Oct. 14 61 plates 



1871, April 17— Aug. 19 14 



1872, Jan. 2— Nov. 27 13 



1874, April 5— Dec. 9 12 



K 



(( 



Total 100 plates 



Of the sixty-one plates taken in 1870, only four were available for meas- 

 urement, the remaining plates not having sufficient data to orient them. 

 These four plates were rejected in the preliminary investigation, but were 

 afterv/ards measured, and found to give satisfactory results. 



Of the fourteen plates taken in 1871, eight were found to be measurable. 

 A ninth plate was measured, but the measures were so discordant that it 

 was rejected. The remaining five plates were either poorly developed or 

 did not have orientation-marks upon them. 



Of the thirteen plates taken in 1872, ten were found to be measurable. 

 An eleventh plate was discarded after measurement, the separate measures 

 being very discordant. 



Of the twelve plates taken in 1874, only one had on it the full data for 

 orientation. This was a very poor plate and was discarded after an at- 

 tempt to measure it. 



This left available for measurement a series of twenty-t^\'o plates, of 



