THE TRANSATLANTIC LONGITUDE. 73 



Boutelle— Goodfellow = — 0M32 

 Boutelle— Chandler = — 0.223 

 Goodfellow— Chandler = — 0.090 



or reducing to Mr. Goodfellow, as the standard of comparison, 



Goodfellow— Gould = — 0'.304 



Goodfellow — Mosman = + 0.150 



Goodfellow— Dean = + 0.029 



Goodfellow— Chandler = — 0.090 



Goodfellow— Boutelle = + 0.132 



In considering these quantities, the attention is at once attracted by the unusual 

 magnitude of some of them, by the excessive tardiness of my own signals as com- 

 pared with those of the other five observers, and by the fact that the personal 

 differences in ordinary time-determinations had not been comparable with those 

 here deduced. For example, although my own observations have usually been 

 somewhat later than those of the many others with whom I have measured personal 

 equations on past occasions, there is no room for the hypothesis that my difference 

 from Mr. Mosman can have reached the enormous value of nearly half a second for 

 chronographic observations. Indeed, a very thorough study of our observations at 

 Valencia established the fact, that it must certainly have been less than 0'.05 upon 

 those occasions when observations were made by both of us during the same night. 



A similar inference is deducible from a comparison of the longitude-results 

 themselves. Thus, the time being determined by myself alone for the first series 

 of exchanges, the resultant value for the longitude between Foilhommerum and 

 Heart's Content is 56". 477;. for the second series, where the clock-correction is 

 derived from interpolation between one determination by myself alone, and one made 

 by Mr. Mosman and myself jointly, the deduced value is 56.487; while the mean 

 01 the other three series, all which depend upon time determined by Mr. Mosman 

 alone, gives 56.465, and one of these three gives 56.481. Since the observer at 

 Newfoundland was the same for all five series, it is very evident that no decided 

 personal difierence existed between Mr. Mosman and myself. That it could have 

 amounted to one-tenth part of the value deduced on the 23d and 28th of May at 

 Cambridge, is totally out of the question. 



So too with Mr. Chandler's comparisons, which indicate for him a habit of 

 observing nearly a quarter of a second later than Mr. Mosman, although more than 

 two-tenths of a second earlier than myself. Until he went to Calais, he had ob- 

 served exclusively with the same signal-key which I have employed at Cambridge ; 

 and at Calais his key was similarly adjusted. And during a very considerable 

 series of observations with a large transit-instrument during the last two years, in 

 which Mr. Chandler took part, I had convinced myself that so large a difi'erence 

 as one tenth of a second between our observations was out of the question. Yet in 

 the present comparisons my observations were recorded later than his, by more than 

 two-tenths of a second. 



The difference between Messrs. Chandler and Boutelle seems, from examination 

 of the Calais record, likewise to have been by no means so large as these special 

 observations would indicate. A series of similar observations with the large transit 



10 July, 1869. 



