64 



EIGHTH REPORT — 1838. 



Lane, ."xnd one of mine, on the 13th of October, 1838, in the 

 gardens of the Palace at Kew. These are collected in the fol- 

 lowing table. 



Table X. 



Observations of the Dip in London in 1837 a^nd 1838, 

 with approved Needles. 



We have therefore 70°02'-9 in August 1821, and 69° 17*3 in 

 May 1838 J or a diminution of 45''6 in 16*7 years, equivalent 

 to a mean annual decrease of 2'* 73, corresponding to the middle 

 of the interval, or to the beginning of the year 1830. The 



* This is the mean of fourteen results, extremely accordant with each other, 

 obtained in difTerent azimuths ; (see Table V.). It will be remarked that it is 

 decidedly the highest of the results from which the mean dip in London has 

 been derived. The observations with the same instrument at Kew, as well as 

 every comparison between this and other instruments, give reason to believe 

 that the high dip in the Regent's Park, in November 1837, is not attributable 

 to any instrumental error. It may then have arisen either from the dip on 

 those days being actually greater by three or four minutes than its general 

 average, or from some local disturbing influence. The locality is the same in 

 which the observations in 1821 were made, and the result in question may 

 on that account appear more strictly comparable with them; but though 

 the locality is the same, it is not one in which we can feel confident that no 

 change may have occurred in regard to magnetic influence. The Regent's Park 

 is certainly not so eligible a situation 9iow for magnetic experiments as it was 

 in 1821. These considerations have induced me to derive the London Dip in 

 1838 for the purpose in the text, from the mean of the observations and local- 

 ities in Table X, rather than from those in the Regent's Park alone ; and not 

 to give to the latter result that additional weight in comparison with the 

 others to which it would seem entitled as derived from observations in so many 

 azimuths. 



