VARIATION OF THE COMPASST. JOJ 



Table V. contains the dip of the magnetic needle from the Dip of the nee- 

 years 1786 to 1805. For the first sixteen months, viz. from <^^^ ^''^j^ i^^S 

 September^ 1786, to December, 1787, both inclusive, the diurnal changci 

 dip was observed as freqnentiy as the variation, but as Ihero 

 does vol appear to ha am/ diurnal cdleraiion in the dip, to make 

 it at all interesting to communicate so many observations as 

 were made, the mean therefore for each month has been 

 thought sufficient for insertion. 



To explain the foregoing Table it must be observed, that 

 each of the numbers in the four first columns of the above 

 Table, are each of them the mean ot several means, as ex- 

 pressed in the line against those numbers ; and as each of 

 those jneans, are again the mean of five observations at least, 

 each of the numbers in the first line, said to be the mean of 

 jiine means, is therefore a mean of forty-five observations j 

 and so of all the rest. 



The numbers in the fifth column, entitled true dip, are the 

 means of the numbers contained in the four preceding columns 

 ^n the same line with it. 



The dipping needles used by Norman, the inventor of the ^^^ ^. ..gi^ 

 dipping needle, who observed the dip at London in the year tohaveformer, 



1576 to be* 71° 50'; and of Mr. Bond, who observed it in \f encreased; 



, , . , , , ,, but tor the last 



1676 to bet 73° 47'; not bemg so much to be depended century has 

 upon aS' the needles that have been in use for near a centnry increased. 

 past, render the progressive increase of the dip fronn Norman's 

 time, to the time of its maximum, somewhat doubtful. But 

 Mr. Whiston, whose needle there is reason to believe was 

 more to be relied upon, in the year 1720 determined the dip 

 tobei75° 10'; this, when compared with many, and very 

 accurate observations made by Mr. Cavendish with several 

 needles, in the year [| 1775, who found it to be 72"' 50', makes 

 the decrease in this period, of 55 years on a mean, 2' 9 per 

 mnum. And from a comparison of my own observations of 

 the dip in 1805, which was 70° 21', with the above of Mr, 

 pavendish in 1775, its annual decrease, on a mean, appears 

 to have been 4', 3 ; and its progressive annual decrease, on a 

 pean, in the above mentioned period of 30 years, 1 ', 4. 



I can- 



* New Attractive, c. 4. -f- Longitude found. 



+ Longitude and latitude found by dipping needle, p. 7 — 94» 



jl Phil. Trans. VoLLXVI, p. 400. 



