of' all kiu)wn Substances to the Magnetic Influence. 123 



cond 2 . 3 4- 2 . 6 = 4 feet 9 inches; the third 4.9 + 2.6 = 

 7 feet 3 inches, &c. Find now the mean deviations, by obser- 

 vation, of the two poles of the magnet at these several distances, 

 and multiply the tangent of the deviations by the reciprocals of 

 the relative forces before determined (Tables at pp. 99 and 115) ; 

 then, if the assumed position of the foci be correct, the numbers 

 thus obtained will correspond very nearly with the tangent of 

 the angle of the greatest deviation, as well as with each other. 

 If, however, the tangent of the greatest deviation be less than 

 these remoter products, then the assumed position of the foci 

 will be too near the extremity of the magnet ; but if the nearest 

 tangent be very much greater than the rest (we speak here of a 

 3-feet magnet), then the position assumed will be too far from 

 the pole. As, from principles already explained, the first tan- 

 gent ought to be somewhat greater than the products of the 

 other tangents*, it will be desirable, when a tolerably close ap- 

 proximation has been obtained, to verify the result by trying 

 the deviations at greater distances. 



What has been now suggested for the determination of the 

 foci of attraction, is equally applicable to two or more bars in 

 fixed combination, or at a given distance, as to single bars. A 

 convenient arrangement, represented in Fig. 6, designed for the 

 safe keeping of the magnets and their employment in practical 

 operations, will be well adapted for giving fixedness of position 

 during the determination of their powers. Here, the magnets 

 A 1, A 2, are laid in parallel grooves in a case about four inches 

 apart. C, C, are the neutralizing conductors of soft iron, which, 

 uniting the opposite poles of the magnets when not in use, serve 

 to preserve their energy uninjured, a a are two pieces of wood 

 fitted into grooves in the centre of the case, of the same size as 

 the conductors. Such is the arrangement when the bars are 

 required to be neutralized, and always when unemployed. 

 When required in the mine, it is only necessary to change the 

 position of one of the bars end for end, so that the two similar 

 poles may lie contiguously, at the same time removing the con- 



* On account of this difference being so great, when short magnets, or very 

 long compass needles, are employed, the first length must be rejected from the 

 results, and the approximation for the focal position be determined by the 

 observations at the more remote distances. 



