THE RETURN OF DAYLIGHT. 301 



ably the result of the fresh E. and S. E. winds we have 

 liad for the past two days. In a few days I shall re- 

 lieve Chipp of the extra duty of taking sights, and 

 shall make daily observations, when possible, for posi- 

 tion to determine the extent and character of our drift, 

 and its connection with the direction and force of the 

 wind. 



March 2Sih, Sunday. — Pumping and distilling as 

 usual. A singular circumstance occurred during the 

 past week in connection with the leak. On Wednesday 

 and Thursday the amount of water coming into the 

 fire-room was about eight times as much as before and 

 since. No greater amount of water seemed to come in 

 forward, but yet it was necessary to keep the steam- 

 cutter's engine going nearly all the time aft. During 

 these two days the forge was lighted in the fire-room, 

 for some machinist's Avork, and as the heat from it was 

 sent up the smoke-stack we supposed that it might have 

 caused a melting of the ice therein, and thus added 

 to the amount of water in the fire-room bilge. To 

 strengthen this supposition, when the fire was extin- 

 guished the water went back to its old condition. But 

 then, to our surprise, the fire-hole, about thirty feet 

 from the mizzen chains, was found closed up, a slab of 

 ice having^ shoved in under. Now what chanii-e could 

 have gone on under the ship to affect the leak ? At no 

 time was any sound heard by us in the cabin, the men 

 in the forecastle, or the watch in the fire-room. The 

 whole thing is a puzzle to us yet. 



At midnight a faint dawn light coald be seen on the 

 northern horizon. At eleven a. m. I inspected the ship. 

 .The berth deck has been somewhat drier during the 

 past week, but the deck-house is and has been in a 

 fearful slop. This is unavoidable, because, however un- 



