368 Capt. Graydon's Celestial Coinpass, 



zon of the sea, even when wholly free from ice, cannot be de- 

 pended on within two or three minutes. There is, however, 

 practically, little or nothing to regret on this account, from the 

 almost constant opportunities that occur in these seas of re- 

 sorting to the more accurate method of observation by artificial- 

 horizons. The weather, which had for several hours been 

 rainy and thick, cleared up about noon on the 4th. In con- 

 sequence of the wind shifting to the N. W. we made sail from 

 the floe, in order to look for the buoy and to continue our ob- 

 servations on the magnetic attraction in that neighbourhood. 

 After making several tacks as near the place as the bearings 

 of the land and the soundings could direct us, but without 

 discovering the buoy, we were obliged for the present to give 

 up the attempt ; having to our great satisfaction observed a 

 floe, at least three miles in length and two in breadth, just 

 detached from the fixed ice, and rendering it necessary for us 

 to work out of its way lest it should force us towards the 

 shore : we onlyt, therefore, waited to put down some nets to 

 ascertain the nature of the bottom, and then hauled round the 

 floe. A quantity of shells, among which were a few of the 

 new species of Anomia discovered on the last voyage, with 

 some shrimps and echini, were all that we could thus fish up. 

 Having cleared the end of the floe, which drifted rapidly away, 

 and, as usual here, never made its appearance afterwards, we 

 made the ships fast to the fixed ice P. M. having by the late 

 disruption made considerable progress in the direction of the 

 strait." 



Page 317. — "The wind gradually falling, was succeeded 

 by a hght easterly bi'eeze, with which, at daylight on the 26th, 

 we steered under all possible sail up the strait. The course 

 being shaped, and no ice in our way, I then went to bed ; but 

 was immediately after informed by Mr. Crozier that the com- 

 passes had shifted from N ^ E. (which was the course I left 

 them indicating) to E i N., being a change of seven points in 

 less than ten minutes. After running half a mile in a true W. 

 by N. direction, the needles began to return to their true posi- 

 tion ; in half a mile farther they had resumed their proper di- 

 rection, and agreed exactly at North. Having sent a boat to 

 the Hecla immediately on our noticing the first alteration, I 

 found from Captain Lyon that a similar phenomenon was ob- 

 served to take place on board that ship, which was following 

 in our wake. The breeze slowly increasing from the eastward, 

 and the weather happily remaining unusually clear for that 

 direction of the wind, we soon arrived off" the narrow part of 

 the strait ; immediately on opening which, we met a tide or 

 current running above two knots to the eastward, with nume- 



