ome very small and very
hard, so hard that they will scratch the runners. The friction set up by
runners in such temperatures may be known as _rolling_ friction, and the
effect, as experienced by us during the Winter Journey and elsewhere, is
much like pulling a sledge over sand. This rolling friction is that of
snow crystal against snow crystal.

If the barometer is rising you get flat crystals on the ice, if it is
falling you get mirage and a blizzard. When you get mirage the air is
actually coming out of the Barrier. Thus far Wright's lecture.

Since we returned I have had a talk with Nansen about the sledge-runners
which he recommends to the future explorer. The ideal sledge-runner
combines lightness and strength. He tells me that he would always have
metal runners in high temperatures in which they will run better than
wood. In cold temperatures wood is necessary. Metal is stronger than wood
with same weight. He has never used, but he suggests the possible use of,
aluminium or magnesium for the metal. And he would also have wooden
runners with metal runners attached, to be used alternately, if needed.

The Discovery Expedition used German silver, and it failed: Nansen
suggests that the failure was due to the fact that these runners were
fitted at home. The effect of this is that the wood shrinks and the
German silver is not quite flat: the fitting should be done on the spot.
Nansen did this himself on the Fram, and the result was excellent. [I
believe that these Discovery runners were not a continuous strip of metal
but were built up in strips, which tore at the points of junction.]
Before it is fitted, German silver should be heated red hot and allowed
to cool. This makes it more ductile, like lead, and therefore less
springy: the metal should be as thin as possible.

As runners melt the crystals and so run on water, metal is unsuitable for
cold snow. For low temperatures, therefore, Nansen would have wooden
runners under the metal, the metal being taken off when cold conditions
obtained. He would choose such wood as is the best conductor of heat. He
tried birch wood in the first crossing of Greenland, but would not
recommend it as being too easily broken. In the use of oak, ash, maple,
and doubtless also hickory, for runners, the rings of growth of the tree
should be as far apart as possible: that is to say, they should be fast
growing. Ash with narrow rings breaks. There is ash and ash: American ash
is no goo