the Vicinity of the Cape of Good Hope. 337 



for local attraction if made on board ship, or for the difference 

 of epoch if any, were carefully stated ; and a table was also 

 given referring to different latitudes and longitudes, by which the 

 variation in any particular locahty within the limits of the map 

 might be approximately corrected for any other epoch (not very 

 distant from 1840), confonnably to the secular change which 

 had taken place in the previous half century, or more exactly 

 between 1787 and 1840. 



Dr. Buys Ballot has kindly engaged to furnish me with a full 

 account, both of Captain Klein's observations and of others made 

 in the Dutch frigate 'De Ruyter,' in a similar voyage between Ba- 

 tavia and Amsterdam. It appears that a large portion of these 

 observations apply to other parts of the ocean than those com- 

 prised in my map ; but the vicinity of the Cape of Good Hope, 

 which is included in that map, is specially pointed out by Dr. 

 Buys Ballot as a locality in which the Dutch observations are at 

 variance with the maps. Dr. Buys Ballot writes as follows : — 

 " It is true that the corrections for local attraction are not applied 

 (in the Dutch observations) in strictly due manner ; but I have 

 persuaded myself that since the local attraction was very small 

 in those two vessels, and did not exceed 4° in any direction, this 

 error could not affect the conclusions in any appreciable manner : 

 therefore I think that really the maps give in the vicinity of the 

 Cape a too great westerly variation." 



If I could entertain the slightest doubt about the accuracy of 

 the variation as given in my map for the vicinity of the Cape of 

 Good Hope, I should most readily wait for the particulars which 

 Dr. Buys Ballot is good enough to promise to furnish me ; but 

 I think that I am able to show that the variation in that locality 

 as given in the map, whether it be taken for 1840 or for the 

 present time, rests on too solid a foundation to be shaken by 

 observations, however carefully made on shipboard, but which 

 are uncorrected for the ship's attraction. 



On reference to the memoir accompanying the map (Phil. 

 Trans. 1849, part 2. art. xii.), it will be seen that the first in im- 

 portance amongst the authorities for the variation at the Cape 

 of Good Hope are the observations at the magnetic observatory 

 at that station, giving as a mean result 29° 7'. The only doubt 

 that could attach to a determination made with every care and 

 precaution, and with the excellent instrumental means of a fixed 

 magnetic observatory, resting also in this case on years of obser- 

 vation, would be the possibility of the existence of station error, 

 viz. possible local deviation occasioned by rocks of particular 

 character in the vicinity of the observatory affecting the needle ; 

 but this is a matter which has been long since set at rest by 

 the strict accordance which has been found between the results 



