THE RETURN OF DAYLIGHT. 235 



for this was one of the water lanes opened in the smash 

 up at that time. 



January 2^th, Thursday. — I am able to record a 

 still further diminution of the leak. The work of fill- 

 ing in the sj)aces between frames, etc., has proceeded 

 all day, and we now find that the Sewell pump, run- 

 ning forty strokes a minute, has been able to hold the 

 water in check. The amount of water pumped out has 

 been 1,800 gallons per hour; and comparing this with 

 the 2,250 gallons per hour on the 27th, shows that two 

 days' work by Nindemann and Sweetman has dimin- 

 ished our leak 450 gallons per hour. The work is still 

 proceeding. In order still further to economize coal a 

 stove was started in the deck-house to-day instead of 

 continuing a fire in the Baxter. Heat is necessary to 

 save the spar deck bilge-pump from freezing, but when 

 we can save it by burning fifty pounds a day instead of 

 one hundred pounds, we are bound to save the fifty. 



January SOth, Friday. — Nindemann and Sweetman 

 continue their slow and tedious job of stuffing plaster 

 of Paris and ashes in the spaces between frames, etc. 

 The water, being unable to get abaft the fillings read- 

 ily, rises between the frames and the outside planking 

 and trickles out under the berth deck at the shelf. Still 

 we are gaining on the leak, and I hope that when we 

 get the spaces filled up inside to a level with the water 

 outside, and have choked up the limber holes in the 

 cant frames (for I believe they exist), so that we have 

 got a ready means of passage interrupted, we shall be 

 able to keep water out of her to a reasonable extent 

 by the use of the spar deck bilge-pump connected with 

 the Baxter boiler. 



Melville, with his never-failing readiness of resource, 

 has commenced a piece of work by which he will run a 



