298 Proceedings of the Royal Society. 



trical measurements on which they depend, (about 14,000 in num- 

 ber,) which would have swelled the paper to an enormous bulk, 

 only the mean results of each set of measures are given ; but to af- 

 ford every opportunity of forming an impartial judgment of their 

 validity, not only the number of measures on which it depends is 

 annexed to each mean result as stated, but also the difference be- 

 tween the greatest and least measure taken, or the limits within 

 which all the measures necessarily lie. 



The stars themselves are arranged in order of right ascension 

 for convenience of reference. After the statement of the mean 

 results of the several sets of observations both of angle and dis- 

 tance, a final mean, with a mean date for an epoch, is deduced. In 

 the case of Sir "Wm. Herschel's stars, a comparison of the mea- 

 sures now obtained with those given in his catalogues, or now for 

 the first time brought to light by a careful examination of his 

 manuscripts, is subjoined. By this comparison several fresh in- 

 stances have been found of double stars, in which the relative 

 motion of the individuals composing them is satisfactorily proved. 

 In one remarkable case, (that of the star £ Equulei,) this change 

 has gone to an enormous extent, and is satisfactorily referred to 

 proper motion in the large star. In another, not less singular, all 

 the three stars of a triple star (f Cancri) are ascertained to be re- 

 latively in motion, describing orbits about each other, and forming 

 probably a ternary system connected by the mutual gravitation of 

 its members ; thus completely justifying the views taken by Sir 

 Wm. Herschel of this subject in his papers published in the trans- 

 actions of this society in 1802 and 1804. 



Annexed as an appendix to these observations, is a re-examina- 

 tion of about 36 stars, measured in the former paper already al- 

 luded to, and which were considered as presenting peculiar inte- 

 rest from the evidence then obtained of their relative motion, and 

 of their connexion in binary systems. The results of this re-ex- 

 amination are in the highest degree satisfactory, as, with only two 

 or three exceptions, these stars have been found to continue their 

 motions in the directions, and in the greater number of cases, with 

 nearly the velocities redicted. In the most remarkable case, 



