444 Transactions. — Chemistry and Physics. 



(3.) Major Palmer's approximate absolute, * 11 h. 39 m. 

 4-81 s. E. 



Major Palmer's result, it will be seen, is identical to 

 within a hundredth of a second of time with the value 

 which has been obtained from Captain Stokes's chronometric 

 work when the most recently accepted longitude for Sydney 

 Observatory is used. On the other hand, the results deduced 

 from Captain Carkeek's and Messrs. Thomson and Jackson's 

 observations seem at first sight a good deal out of line with all 

 the other determinations. They were consequently not taken 

 into account in deciding upon the longitude to be used for the 

 purposes of the time-service. But Major Palmer showed 

 in his report some four years later that Messrs. Thomson and 

 Jackson's determination was susceptible of treatment which 

 placed it in a different light. Messrs. Thomson and Jackson, 

 in reducing their observations, had not taken into account the 

 errors of the moon's tabular place. Major Palmer pointed 

 out that the average of these errors for the days on which the 

 moon was observed at Eockyside and the Hutt was about 

 025 s., which would probably cause an error of between + 6 s. 

 and -f 7 s. in the resulting longitude ; t therefore Messrs. 

 Thomson and Jackson's corrected longitude of the Observatory 

 might be taken approximately as 11 h. 39 m. 9 s. E. This 

 differs from the ultimately accepted longitude by less than 

 4 s., very little more than the error (3'29 s.) which shortly 

 before this had had to be recognised in the absolutely deter- 

 mined longitude of Sydney Observatory. The problem of 

 exactly ascertaining a longitude by observation is notoriously 

 one of extreme practical difficulty ; and Messrs. Thomson and 

 Jackson's result, when subjected to this revision by Major 

 Palmer, showed that their long and patient series of observa- 

 tions had been carried out with much skill and care, and was 

 an honourable and worthy piece of work. The details of 

 Captain Carkeek's calculations are not available, as they were 

 accidentally destroyed many years ago by fire. 



To sum up, it will be seen that all the foregoing deter- 

 minations may be arranged in two groups — one with a value 

 of about 11 h. 39 m. 9 s., and the other with a value of about 

 11 h. 39 m. 5 s. The former of these approximate values was 

 practically known as long ago as 1874 to be erroneous ; the 

 latter by the same year was believed to be correct, and two 

 years later was known to be correct, on the assumption that 

 Sydney's longitude was reliable. Sydney's 1903 value differs 

 by only about 1 s. from its 1874 value ; so that Dr. Hector 



* For details of Major Palmer's work see his report to Government, 

 Appendix to Journals of House of Eepresentatives, H. No. 6, 1876. 

 f See Loomis's " Practical Astromony," p. 316 (seventh edition). 



