VOL. LXXXVII.] PHILOSOPHICAL TRANSACTIONS. J 77 



XIII. A Third Catalogue of the Comparative Brightness of the Stars; with an 

 Introductory Account of an Index to Mr. Flamsteed's Observations of the Fixed 

 Stars contained in the id Volume of the Historia Ccelestis. To which are added, 

 several Useful Results derived from that Index. By Win. Herschel, LL. D., 

 F. R. S. p. 293. 



In my earliest reviews of the heavens, I was much surprized to find many of 

 the stars of the British catalogue missing. Taking it for granted that this cata- 

 logue was faultless, I supposed them to be lost. The deviation of many stars from 

 the magnitude assigned to them in that catalogue, for the same reason, I con- 

 sidered as changes in the lustre of the stars. Soon after however I perceived that 

 these conclusions had been premature, and wished it were possible to find some 

 method that might serve to direct us from the stars in the British catalogue, to the 

 original observations which have served as a foundation to it. The labour and 

 time required for making a proper index withheld me continually from undertaking 

 the construction of it: but when I began to put the method of comparative bright- 

 ness in practice, with a view to form a generated catalogue, I found the indispen- 

 sible necessity of having this index recur so forcibly, that I recommended it to my 

 sister to undertake the arduous task. At my request, and according to a plan 

 which I laid down, she began the work about 20 months ago, and has lately 

 finished it. 



The index has been made in the following manner. Every observation on the 

 fixed stars contained in the 2d volume of the Historia Ccelestis was examined first, 

 by casting up again all the numbers of the screws, in order to detect any error 

 that might have been committed in reading off the zenith-distance by diagonal 

 lines. The result of the computation being then corrected by the quantity given 

 at the head of the column, and, refraction being allowed for, was next compared 

 with the column of the correct zenith-distance as a check. Every star was now 

 computed by a known preceding or following star; and its place according to the 

 result of the computation laid down in the Atlas Ccelestis, by means of proportional 

 compasses. This was necessary, in order to ascertain the observed star: for the 

 observations contain but little information on the subject; most of the small stars 

 being without names, letters, or descriptions. The many errors in the names of 

 the constellations affixed to the stars, and in the letters by which they are denoted, 

 also demanded a more scrupulous attention; so that only their relative situation* 

 examined by calculation, could ascertain what the stars really were which had been 

 observed. 



Every observed star being now ascertained, its number in the British catalogue 

 was added in the margin at the end of the line of the observation ; and a book with 



and as we could not always agree in his deductions from them, especially in the most important particu- 

 lars ; we have found it our duty to the public to show, by our observations, our conviction that he has 

 completely failed in the most important circumstances of the inquiry. 

 VOL. XVIII. A A 



