142 The Rev. Dr. Robinson on the Difference of Longitude 



May 21, . . 0.94414 0.44375 k 



Same, 0.38710 e 



0.74356 0.17845 f 



The final means are, therefore, 



R = 1". 14'. + ^'•^^^^'^ = Im. 14'. 439 

 ^ 2.62159 



k' = 1 14 + 1:9^^ = 1 14 423 

 ^ 2.52437 



E = 1 14 + 2:21^ =1 14 427 

 ^ 2.38505 



F=l 14 +-5i^ = l 14 270 

 ^ 1.58180 



The result F has obviously far less weight than the other two, which must 

 be attributed not merely to Mr. Finlay's total want of practice in such obser- 

 vations, but also to the small optical power of his telescope. Though it differs 

 but little from the others, I think it best to omit it, and consider the mean of r' 

 and E as the definitive result 



l™. 14'. 425. 



But had I used it and retained the two omitted on May 20th, this would be 

 only 0'. 03 less, and identical with the result given by Mr. Dent's chronometers. 



These, however, require a correction for what is called the Personal Equation 

 of the transit observers. It may appear strange that two practised observers 

 should not observe the passage of a star over a spider's line at the same instant, 

 but the fact is undoubted, and the difference is not of a decimal or two, but in 

 the case of perhaps the first of European astronomers, it exceeds a second. The 

 cause is unknown, but as from its being almost invariably independent of the 

 declination, it appears not to originate in the eye, the probability is, that it is 

 caused by some exercise of thought in associating the indications of the ear to 

 those of the eye. In most cases it is constant for many years in the same indi- 

 vidual ; in some, probably by carelessness, it goes on increasing. 



The usual method of determining its amount is thus : the observer, e, ob- 



