VOL. LIII.] PHILOSOPHICAL TRANSACTIONS. '2Q 



This must certainly have caused some disturbance to the other observers, and 

 might possibly influence their judgment: and the fact it seems was, that each 

 observer had a second watch in liis hand, and they instantly stopped their 

 watches, each having his hand at his watch ready to stop. This problem there- 

 fore is easily solved, and the surprize at the coincidence entirely vanishes; so that 

 this observation can be looked on as no more than the observation of one person, 

 and he too not much practised in observing. It is proper further to observe, 

 that another person was present at this observation, who confirmed the above 

 account. 



The very near coincidence of the three observers at Greenwich, in the time of 

 the external contact, remains now to be accounted for. Mr. Green did not call 

 out at this time, because he was forbid by Dr. Bradley, who was present, though 

 not in a condition to observe because of his bad state of health. This problem 

 therefore may be solved in the following manner. The observation of the ex- 

 ternal contact was undoubtedly more uncertain than the former, and yet we find 

 two of the observers agreeing to the same second, and the third differing only 

 one second from them. If we attend to the following circumstances, we shall 

 be immediately satisfied by them. Each observer had a second-watch in his 

 hand ; the 3 observers were at the same window of the same room, one of them 

 on the leads immediately without the window, and the other two within the win- 

 dow; therefore each observer was within hearing and seeing of each other; con- 

 sequently the instant one of the observers stopped his watch, may it not be pre- 

 sumed that the noise of the nicking of it might be heard by the rest } especially 

 as there was a profound silence during the time of the observation. Mr. S. has 

 thought proper to take notice of these facts, because several persons both at 

 home and abroad have expressed their surprize at this coincidence, and that such 

 an exactness may not be established as a precedent in these sorts of observations; 

 and because he thought it essentially necessary, in all sorts of observations, espe- 

 cially in one of so much importance in astronomy as this, that every the minutest 

 circumstance should be particularly related. 



We are now to find the limits of the error arising from the difference of longi- 

 tude between Tobolsk and the Cape. Mr. S. found that an error of I"' IC in 

 time, in the difference of longitude between these two places, will cause an error 

 of l" in the sun's parallax. But as we are certain that this error in longitude does 

 not take place ; therefore we are certain that the error in the parallax is within 

 one second of the truth. The difference of longitude between the Cape of Good 

 Hope and Paris is determined, both by the observations of M. de la Caille and 

 Mr. Mason ; the difference of longitude between Paris and Upsal in Sweden is 

 settled by the observations of Jupiter's first satellite, and the difference of longi- 

 tude between Upsal and Tobolsk is settled, by the observations of the contact at 



