239 



differences of longitudes by signals, geodetic measures, oc- 

 cultations, and transits, the author passes to the most direct 

 and natural method, by transport of chronometers. He 

 says, " the determination by chronometers depends on the 

 perfection of these machines, and in particular on their rate 

 being unchanged by the agitation of a long journey. This, 

 strictly speaking, is never the case, though it is sometimes 

 very nearly accomplished, and its effect will disappear from 

 the mean of the results obtained in going and returning, 

 if the circumstances of the two journeys are nearly similar. 

 Unfortunately it rarely happens, that an astronomer has 

 the power of making these experiments on a sufficient 

 scale : but such an opportunity seemed to Sir William 

 Hamilton, and myself, to present itself in consequence of 

 Mr. Dent's chronometric visit to Paris, and the yet more re- 

 markable notice, read at the Newcastle Meeting of the 

 British Association, of the chronometric longitude of Sir 

 Thomas Brisbane's Observatory. Mr. Dent not merely pro- 

 mised us every assistance, but when, having obtained the 

 consent of the authorities of our respective observatories, 

 we proceeded to make the necessary pecuniary arrange- 

 ments, he treated the matter as one of science, not of com- 

 merce, and not only took on himself the expense and risk of 

 the journey, but came in person." 



Dr. Robinson proceeds to mention the particulars of the 

 journey, and of the comparisons which were made with the 

 fifteen chronometers which Mr. Dent brought with him. 

 Their rates and errors, as compared with Greenwich time, 

 had been determined, before starting, for the epoch of the 

 20th of September, 1888; they were compared with Dublin 

 time, in the observatory of Trinity College, on the 22nd of 

 that month, and with Armagh time on the 23rd and 24th; were 

 again examined in Dublin, on the 25th, and in Greenwich, 

 on the 27th of September. 



Y 



