are a great many people who apparently have never learnt to shut a door after them; where two or three are gathered together, you generally find at least one who suffers from this defect.  How many would there be among us, who numbered nine?  It is no use asking a victim of this complaint to shut the door after him; he is simply incapable of doing it.  I was not yet well enough acquainted with my companions as regards the door-shutting question, and in order to be on the safe side we might just as well put up a self-closing door.  This was done by Stubberud, by fixing the door-frame into the wall in an oblique position just like a cellar-door at home.  Now the door could not stay open; it had to fall to.  I was glad when I saw it finished; we were secured against an invasion of dogs.  Four snow steps covered with boards led from the door down into the passage.  In addition to all these new rooms, we had thus gained an extra protection for our house.

While this work was in progress, our instrument-maker had his hands full; the clockwork mechanism of the thermograph had gone wrong:  the spindle was broken, I believe.  This was particularly annoying, because this thermograph had been working so well in low temperatures.  The other thermograph had evidently been constructed with a view to the tropics; at any rate, it would not go in the cold.  Our instrument-maker has one method of dealing with all instruments —­ almost without exception.  He puts them in the oven, and stokes up the fire.  This time it worked remarkably well, since it enabled him to ascertain beyond a doubt that the thing was useless.  The thermograph would not work in the cold.  Meanwhile he got it cleared of all the old oil that stuck to it everywhere, on wheels and pins, like fish-glue; then it was hung up to the kitchen ceiling.  The temperature there may possibly revive it, and make it think it is in the tropics.  In this way we shall have the temperature of the “galley” registered, and later on we shall probably be able to reckon up what we have had for dinner in the course of the week.  Whether Professor Mohn will be overjoyed with this result is another question, which the instrument-maker and director did not care to go into.  Besides these instruments we have a hygrograph —­ we are well supplied; but this takes one of us out of doors once in the twenty-four hours.  Lindstrom has cleaned it and oiled it and set it going.  In spite of this, at three in the morning it comes to a stop.  But I have never seen Lindstrom beaten yet.  After many consultations he was given the task of trying to construct a thermograph out of the hygrograph and the disabled thermograph; this was just the job for him.  The production he showed me a few hours later made my hair stand on end.  What would Steen say?  Do you know what it was?  Well, it was an old meat-tin circulating inside the thermograph case.  Heavens! what an insult to the self-registering meteorological instruments!  I was thunderstruck, thinking, of course, that the