378 



PROCEEDINGS OF THE AMERICAN ACADEMY 



resulting time, or 



0.845 2R 



The last column but oue gives the 



-y/n — 1 2a 



average residuals, or the fifth column divided by the third. The last 

 column gives the average deviation of the three sets of four readings 

 each, of which the sets of twelve readings are composed. It serves to 

 show the accordance of the successive readings. 



The first seven lines give the results for the seven minima which 

 were observed. The next three lines group together all the observa- 

 tions of each observer. Lines 11 and 12 place together all those in 

 which the light is decreasing, and those in which it is increasing. The 

 results of all these sets is given in line 13. The rejected sets obtained 

 on Nov. 19 are given in line 14. Line 15 groups those in which the star 

 has its full brilliancy, or when the nearest minimum was more than 

 five hours distant. The last line gives the results of lines 13 and 15, 

 or the entire series, excepting those of Nov. 19. A set taken Oct. 7 

 is also included, which was taken so near the minimum that a was 

 sensibly equal to zero. For this reason line 16 is not exactly equal to 

 the sum of lines 13 and 15. 



TABLE IV. — CoMPARisox of Resclts. 



The observations of Nov. 22 show a large correction to the mini- 

 mum. This is not easily explained unless the deviation is real. The 

 measures before the mininuini give a correction of + 15 minutes ; 

 those after, of -f- 18 ; those of P. alone, + 26 ; of S., + 14 ; of W., 

 -|- 12. As the probable error of the mean result is only 2.6 minutes, 

 and a nearly equal number of measures were made on each side of 



