80 Longitude of the Melbourne Observatory. 



errors, it may be inferred that our adopted longitude is 

 correct, or any difference that may be found could be con- 

 verted into a correction of our adopted longitude. On com- 

 paring the Greenwich and Melbourne lists it was found 

 that on fifteen days the moon had been observed at both 

 places, and on interpolating the Greenwich errors, to make 

 them correspond to the time of the Melbourne errors, and 

 taking their mean, it was found that the mean error of the 

 Nautical Almanac was + 0"58s. from the Greenwich observa- 

 tions, and + 0'57s. from the Melbourne ones. These results are 

 so nearly identical as to show that our adopted longitude is 

 quite as accurate as can be possibly obtained from the method 

 of moon culminations. A distinguished American mathe- 

 matician. Professor Peirce, of Harvard University, from 

 theoretical considerations, estimated one second of time as 

 the utmost limit of accuracy to be obtained by this method. 

 Professor Hall, however, of the Washington Observatory, has 

 recently discussed the longitude of his Observatory, as deter- 

 mined by means of the Atlantic cable, transportation of 

 chronometers, and moon observations; and assuming the 

 telegaphic result to be the correct one, he finds a difference 

 of rather more than two seconds to exist between the moon 

 and electric determination, while the chronometric and 

 electric results are nearly identical. Now, if we convert the 

 above difference between the errors of the moon's place, as 

 found at Greenwich and Melbourne into a correction of the 

 latter 's longitude, it will amount to only three-tenths of a 

 second ; combining this with a weight proportional to the 

 number of observations from which it is derived, it would 

 indicate an increase to our adopted longitude of only three- 

 hundredths of a second of time. Having thus reached the 

 limit of accuracy of which the method of moon culminations 

 is capable, any other determination of our longitude would 

 have to be made either by transmission of large numbers of 

 chronometers — a very expensive and troublesome process — or 

 by means of the electric telegraph. In conclusion, I will 

 state that I consider the longitude of Melbourne to be as 

 well determined as that of any other place in the Southern 

 hemisphere, and better than that of any other place in Aus- 

 tralasia. The only other places in Australia where long-con- 

 tinued observations of moon culminations have been made 

 for finding the longitude are Parramatta and Sydney ; at 

 both of these places, however, very inferior instruments were 

 used. For the latter place, however, a fine transit circle, of 



