Mr. J. R. Napier on Ships' Compasses. 867 



perimental card newly magnetized, compared with another of Sir Snow 

 Harris's, also newly magnetized, gave the following results : — 



Single bar needle, ..,.932 grs. deflected 90° by 82° torsion of wire. 



Scoresby's principle,.. 838 grs. deflected 90° by 121° torsion of wire. 

 And the comparative powers found after dividing the torsion by the 

 weight are consequently 888 to 1453, or 100 : 1G4, so that the experi- 

 mental Scoresby needle card has here about 64 per cent, more power 

 than Sir Snow Harris's single bar needle. A comparison of other cards 

 on the apparatus which Mr. Walker (a member of this Society) has kindly 

 supplied me with, gives the following results : — 



Weight. Power. 



Grains. Torsion Weight. 



Experimental card, compound needle, 838 1330 



Captain Walker's card, T shaped needle, 989 1203 



Sir Snow Harris's azimuth card, edge bar needle, 820 1090 



Admiralty card, compound needles, 2023 1000 



Sir Snow Harris's steering card, edge bar, 932 794 



Gray's (Liverpool) card, two dipping needles,... 1708 515 



Keen's (Liver.) patent card, two edge bar needle, 2040 442 



Therefore, in regard to this particular card, at least sufficient reason can 

 be given for the employment of even eight magnetic bars, viz., with cards 

 of equal weight, the compound needle card has about one-half more 

 power ; or if, as Sir Snow Harris says, the extra power is not needed, 

 then a much lighter card with less pivot friction, and consequently more 

 durability, will be as powerful and sensitive as Sir Snow Harris's. If, 

 indeed, the cost of the latter article, as supplied to the public, is any in- 

 dication of the difficulty of its construction, it would appear that this 

 particular arrangement of Scoresby's needles is much less difficult. 



Sir Snow Harris, however, evidently here refers to the Admiralty 

 compass card, which certainly does not appear of very easy construction. 

 But, I am not aware that Dr. Scoresby in any of his works recommends 

 this particular arrangement ; and, when on a visit at Torquay about three 

 years ago, I called on him, and he showed me his compasses and magnetic 

 machines, and the needles of the former instruments were much more 

 simply arranged. 



The method he recommended to me was to bend the steel plates two 

 or more on each side round the cap, and bring them nearly together, 

 say within about ^ of an inch, at the extremities or poles. I had a card 

 or two fitted up on this plan for one of our steamers. The result of an 

 experiment on it, however, was not so satisfactory as anticipated. I 

 ascribed the failure to bad steel, or defective hardening, which Dr. 

 Scoresby shows must have been the case, besides the fittings being too 



