184) Mr. S. Besw'ick' sfw'iher Illustrations of a Method 



tion of any Jcind is required. The two accompanying illustra- 

 tions are intended as proof of its advantageous superiority in 

 this particular. Colonel Sabine has kindly reminded me of 

 the grand desideratum in this department of science, for the 

 accomplishment of which enormous labours have been under- 

 taken, and considerable sums of money annually expended, — 

 that " the chief difficulty in any empirical formula would be 

 to adapt it to different epochs." I have thought it necessary, 

 in consequence of this remindal, — and with a view to the com- 

 plete exposition of the method I propose, and to the clear 

 understanding of the important practical advantages it offers 

 over all others, — to present a few general explanatory remarks, 

 accompanied with two distinct proofs of its application to dif- 

 ferent epochs. Which proofs are intended to show, that this 

 * chief difficulty ' is completely removed; indeed so completely, 

 that not a single item would be added to the process, nor even 

 altered in its form, whether the epoch was three years in the 

 past, or three hundred years in the future. 



I am desirous of recording a matter of considerable import- 

 ance in this paper, to which I solicit the kind attention of 

 the gentlemen forming the Committee for Simultaneous Mag- 

 netical and Meteorological Observations referred to above. 

 The obvious failure of all previous methods, after the interval 

 of a few years, mainly results from a quarter which, up to the 

 present time, is not even suspected. It is this — the southern 

 magnetic point of convergence of the horizontal force, or pole, 

 does not revolve from east to west, as hitherto supposed ; on the 

 contrary, both poles move in the same direction from west to 

 east. Hence, if the numerical coefficients of Prof. Gauss's 

 general theory be reconstructed — the necessity of which is 

 suggested by Colonel Sabine*, — it can never become available 

 for practical purposes, except for the time being, in conse- 

 quence of this fundamental error. Every formula should be 

 required to prove its utility, by its application to successive 

 epochs of 50 or 100 years interval for 300 years past, before 

 it be relied on in its applications to the future. Should this 

 test be admitted, then my investigations into this matter enable 

 me to state, that the general formula of Gauss, and of all 

 others which suppose a revolution of the southern magnetic 

 pole from E. to W., will prove, during the test, an entire 

 failure. I would respectfully suggest to the Committee of 

 Observation, that e\Qry method be required to fulfill the de- 

 mands of this test. 



We will now try the merits of our own method by this rule; 

 and, in so doing, we wish to be followed by all others pro- 

 • Phil. Trans., part 2, 1849. 



