278 On Irregularities observed in 



binnacles continually to assist their motion. The cards of these* 

 had a metal rim round their circumference, weighing one ounce 

 eleven drachms avoirdupois, which, as the directive power of 

 magnetisoi diminished, became too heavy for the needle to carry 

 round: tliey were also frequently found to disagree with each other 

 from I to ^ of a point ; the consequence, most probably, of the 

 different local attraction to wliich they were exposed. These com- 

 passes ceased therefore to be attended to, except as an occasional 

 assistance to the helmsman, and a position was selected in each 

 ship, in which a compass on a more suitalde construction was 

 permanently fixed ; by this the ship's course was directed, azi- 

 muths taken, and bearings of land, &c. noted, during the voy- 

 age. 



This standard compass, as it may be called, was placed in the 

 Isabella exactly amidships, between the main and mizen mast, on 

 a stout cross beam elevated nine or ten feet above the deck ; this 

 beam wa5 the usual walk of the Greenland pilot, or of the quar- 

 ter master, as afifording a better view of the ice among which 

 the ships were frequently steered, than from the deck. The ele- 

 vation was an advantage to the compass in s\ich high magnetic 

 latitudes, by rendering it less liable to accidental disturbance, otj 

 the removal of such implements of iron as were required to be 

 kept on deck for use. The Alexander not having a similar cross 

 beam, her compass was fixed amidships on a box of sand placed 

 on th,e companion, betv/een live and six feet above the deck. 



Secondly; Captain Flinders found that in his compass per- 

 manently fixed as described, no error took place when the ship's 

 bead was on the magnetic north or south points ; showing, that 

 at such times the attraction of the ship, and of magnetism, were 

 in the same line of direction. The maximum of error also took 

 place when the ship's head was at right-angles to these points; 

 namely, at east or west ; being however in opposite directions, 

 in excess of tlie true variation on the one side, and in defect on 

 the other ; so that the extreme difference occasioned by altering 

 the course from 6ast to west, or the reverse, woidd be twice the 

 error at either. 



On the intermediate points, the ratio of the error to its maxi- 

 mum, was as tl e " sine of the angle between the ship's head and 

 the magnetic meridian to the sine of eight points, or radius," or 

 sufficiently near to admit of corrections being calculated for every 

 course, when the error on a single one was known by observa- 

 tion. 



Thus far the experiments wliich Captain Flinders tried in every 

 ship corresponded, excepting only that the maximum of error in 

 different ships at the same place would differ materially. 



* Burt's patent binnacle compasses. 



The 



