for computing Magnetic Declination. 189 



Sides. Angles. 



(13.) 89° 44' 11° 17' 



129° 58' 16° 15' 



219° 42' : 27° 32' : : 89° 44' : 11° 14'. 



11° 17' 

 11° 14' 



3' W. D., 



which is the mean west declination on the Atlantic for the 

 year 1492. It may be objected — Columbus affirms there was 

 no declination, whilst your method gives 3 minutes. Suppose 

 we grant this, it is only an error of 3 minutes in 357 years, or 

 ^ a second annually ; an amount so small, that in ordinary 

 practice even the whole would be scarcely noticed. But, 

 reader, this error does not exist. These three minutes show 

 the mean declination for the whole year ; in other words, it 

 is the declination for about the middle of the year — say Jmie: 

 and as the diminution was then going on at the rate of sixteen 

 and a half minutes per year, the three minutes would Just have 

 vanished at the close of the month of September. Accordingly 

 Columbus discovered the line "dohen passing this spot on the 

 25th of September. Hence there is not even a second of dif- 

 ference between the observation and our calculation, though 

 it involves so great an interval as 357 years. The next illus- 

 tration is for the epoch of 1722, involving an interval of 127 

 years. In this instance we shall not go through the whole 

 computation, but merely state the result as given in item 13. 

 It is as follows : — 



Sides. Angles. 



52° 30' 26° 61' 



152° 41' 22° 7' 



205° 11' : 48° 58' ; : 52° 30' : 12° 31'. 



26° 51' 

 12° 31' 



14° 20' W. declination, 



which is the mean declination for London, 1722. 



In the observations of Graham, inserted in the Philosophical 

 Transactions, No. 383, p. 96, for 1724, the mean declination 

 is the same as given in our computation, namely 14° 20' west. 



As before observed, the epochs we have taken are as me- 

 morable as any in the history of terrestrial magnetism, and 

 are as widely distant from each other, and from the present, 



