172 SCIENTIFIC RECORD FOR 1884. 



each of which the instrumental conditions were different — accord gen- 

 erally within the limits of the probable errors, and that there is no sys- 

 tematic change from recession to approach, so that the presumption 

 against error arising from defective instrumental adjustment appears 

 to be strong." {Nature.) 



PHOTOIVIETEY OF STABS. 



The new volume of Annals of the Observatory of Harvard College is 

 in many respects the most interesting and important of any of them, 

 for it embodies the results of three years' work with an entirely new and 

 unique instrument in a field hitherto but little explored, which Profes- 

 sor Pickering, the director, has made almost entirely his own by the in- 

 vention of novel and efficient apparatus, and by the energetic accumu- 

 lation, discussion, and i)rompt publication of results. 



The observations described in this volume consist of 94,476 separate 

 measurements of the brightness of 4,260 stars, all those visible to the 

 naked eye from the North Pole to 30° of south declination, and many 

 somewhat fainter. They were made by Professor Pickering, the di- 

 rector, aided by Messrs. Searle and Wendell, assistants in the Harvard 

 College Observatory, using the new meridian photometer, devised by 

 the director. 



With this photometer each of the 4,260 stars was compared, when 

 near the meridian, with the pole star on at least three nights, and 

 many of them on six or more. In these comparisons both stars are 

 seen under the same magnifying power and on the same background, so 

 that the differential effect of moonlight or twilight is eliminated, and- 

 any error due to local cloudiness or haziness over one star is sufficiently 

 guarded against by the number of different nights and the repetition of 

 the observations where this is suspected. Two persons work at the 

 same time, the observer managing the pole star image in the field and 

 making the settings of the eye-piece Nicol for the four i)ositions of 

 equal brightness, and the other bringing the other star into the field 

 by the other i^rism and reading off and recording the observer's set- 

 tings of the eyepiece. At the beginning, middle, and end of each con- 

 tinuous series of observations the prism of the south objective was also 

 turned to the pole star, and the two images of this were compared, in 

 order to furnish the corrections necessary on account of unequal trans- 

 parency and reflecting power of the two objectives and prisms. 



In the preliminary chapters of the work are given the details of the 

 various methods employed to discover and eliminate all sources of sys- 

 tematic error, and their completeness and thoroughness assure the pro- 

 fessional astronomer of the high degree of confidence which can be 

 placed in the results. Only a few of the most interesting points can be 

 noted here, and those very briefly. The discussion of a long series of 

 observations upon a list of 100 circumpolar stars, at upper and lower 

 culminations, gives for the co-efficient of atmosi)heric absorj)tion : 



