no Rev. Mr Scoresby on the Uniform Permeability of 



Reducing, now, the fraction | to the same denominator as 

 the above, we have tJ^t^tt as the resultant influence of the whole 



' 1411344 



bar in the assumed position w, s. Therefore the entire or resul- 

 tant force acting upon the needle, in the deflected position w', /, 

 is to the force in the assumed position w, *, as ^Jn^ ^^ lin^' ^"^ 

 as ^ to I nearly. If, then, we apply this proportion to the ob- 

 served deviation of No. 1. (series in page 108), 59° 23', the tan- 

 gent of which is 168979, we have ^ : | : : 168979 : 144839, 

 = ^^"^ 23', which, it is satisfactory to find, corresponds very 

 nearly with the deviation observed when a very small compass 

 was substituted for the large one ; in that case, the angle formed 

 by the needle, as near as could be observed, being 55° 45'. 



Still, however, the deviation thus reduced is found to be consi- 

 derably greater than that given by a mean proportional, namely, 

 54°. The cause of this difference is probably to be found in the 

 peculiar direction, F^, of the strongest force, which evidently 

 is not strictly tangential to the meridional position of the needle ; 

 but must operate more favourably for overcoming the directive 

 force of the earth, than if, acting in the direction ^'M, it were 

 precisely at right angles to the terrestrial magnetism. 



Since now the calculated deviations of the three feet magnet 

 (Table, p. 24), as obtained from the mean ratio 121600, are 

 all, except the first, within the limits of the possible error of ob- 

 servation ; and since the ratios obtained from experiments with 

 the twelve-inch magnet (Table at p. 108) are all, with the''excep- 

 tion of the first, uniform within the probable limits of error, 

 — whilst the discrepancy at the first focal length has been suffi- 

 ciently, T trust, accounted for, — the position of the foci in both 

 these magnets may be considered as rightly determined *. For 

 all practical purposes, therefore, connected with the proposed 



• Though I have hitherto spoken of a fixed and determinate focal position 

 representing the whole of the magnetic forces of either half a regularly mag- 

 netized bar, yet I am aware that that very focal position will be liable to a 

 small variation at very short distances, in such cases being nearer the extre- 

 mity than the calculated position. Nevertheless, at distances beyond the 

 length of the magnet, no alteration in the position of the foci, I apprehead, 

 will be discernible ii) practice. 



