726 



SCIENCE 



[N. S. Vol. XXX. No. 777 



Precautions Necessary in Photographic Photom- 

 etry: J. A. Pabkhubst. 



The results of experiments with stellar and 

 laboratory plates, both focal and extra-focal, were 

 shown by curves thrown on the screen; and the 

 possible errors in stellar magnitude arising from 

 each source were indicated. ( 1 ) Comparison of 

 developers, gradation very different. (2) Time of 

 development, possible error exceeding one magni- 

 tude. (3) Temperature of development, error 

 nearly one magnitude for range of ten degrees 

 Fahrenheit. (4) Effect of sky fog, possible error 

 of half a magnitude for fog of 0.03 of a density 

 unit. (5) Temperature of exposure, difference of 

 0.1 magnitude for plates at + 17° and — 2° C. 

 (6) Reduction formulae for disk diameters of focal 

 images. Error of half a magnitude possible if 

 log diameter were used in the formula instead of 

 square root of diameter. ( 7 ) Atmospheric absorp- 

 tion. Difference of 0.17 magnitude between the 

 visual and photographic absorptions at 60° zenith 

 distance. (8) Curvature of commercial plates. 

 Error negligible for focal images but might 

 amount to half a magnitude for extra-focal 

 images. (9) Corrections for distance from axis 

 for plates taken 7 mm. from the focus of the Zeiss 

 doublet, amounts to 0.33 magnitude at 3° from 

 the axis. (10) Correction for distance from axis 

 of focal plates might differ by 0.8 magnitude 

 between summer and winter temperatures. (11) 

 Correction for color of stars of solar type is one 

 magnitude. (Details of the work will be pub- 

 lished in the Astrophysical Journal.) 



Standard Photographic Magnitudes: Henbietta 



S. Leavitt. 



Observations for the purpose of determining the 

 absolute photographic magnitudes of a sequence 

 of forty-seven stars near the North Pole have 

 recently been in progress, as described in Harvard 

 College Observatory Circular 150. Sequences of 

 forty stars in the Pleiades and twenty-six stars 

 in PrcEsepe have also been measured, and the 

 results compared with those obtained for the 

 polar stars. 



About one hundred and fifty plates were used, 

 taken with eight telescopes. Several methods 

 were employed for determining absolute magni- 

 tudes, independent of the visual scale, but all may 

 be grouped in the three following classes: 



1. Photographs were taken, diminishing the 

 light by means of screens, or by reducing the 

 aperture of the telescope, and superposing a 

 second exposure of the same length with full 

 light. A similar effect was obtained by attaching 



an auxiliary prism of very small angle to the 

 object glass; this deflected a part of the light, 

 forming secondary images of the brighter stars. 



2. The light was divided by interposing two 

 thin plates of Iceland spar. The positions of the 

 four images of each star furnished the means of 

 determining the relative amount of light in each 

 image. 



3. Photographs were taken, having several ex- 

 posures on the Pole Star, and on the star to be 

 observed, the images being out of focus by varying 

 amounts. 



The results obtained by these radically different 

 methods are accordant with each other in the 

 great majority of cases. They also agree closely 

 with the Harvard photometric scale as far as the 

 magnitude 13.2, after allowing for difference of 

 color. We apparently have a satisfactory work- 

 ing basis for determining the magnitudes of stars 

 in all parts of the sky, on an approximately cor- 

 rect scale. 



In the discussion that followed Mr. Stebbins's 

 and Miss Leavitt's papers, Messrs. Parkhurst and 

 Humphreys called attention to the large error 

 which might sometimes be incurred by assuming 

 that the absorption of our atmosphere is the 

 same in different azimuths. Professor Pickering 

 remarked that it was the practise to guard against 

 this source of error at Harvard by establishing 

 an arbitrary limit for the residuals obtained with 

 the meridian photometer and similar instruments, 

 and rejecting all observations in which this limit 

 was exceeded. 

 A Variable Star whose Light Curve Resembles 



that of R GororuB Borealis: Annie J. Cannon. 



This star, like R Goronce Borealis and RY 

 Sagittarii, has long periods of normal brightness 

 followed by sudden fluctuations of large range 

 at irregular intervals. Its position for 1900 is 

 R. A., SMS"- 12"; DecL, + 19° 02'.0. It follows 

 the Durchmusterung star +19° 1081 about 3' 

 and is south 0'.2. 

 The Pivots of the Nine-inch Transit Gircle of the 



U. S. Naval Observatory: F. B. Littell. 



This paper, which was read by Professor Eichel- 

 berger in the author's absence, gave the results of 

 several determinations of these errors. The in- 

 vestigation is valuable not only for its application 

 to the work with this instrument, but is of general 

 interest on account of the experience with various 

 methods for determining inequalities of pivots. 

 The Algol System, Z Draconis : R. S. DuGAN. 



The material for this paper consists in 18,384 

 settings made with the Pickering sliding prism 



