will be so favourable as to allow this "land floe" to survive a second summer. But, on the other hand, the first year's floe may rot away entirely during the first summer after its formation. I saw this happen in Franz Josef Land during the summer of 1897 to a land floe that I had watched from its birth to its disintegration, from the time the first "bay ice" was forming on the calm surface, through the period when the ice was thick and solid, until it had rotted entirely away.
The surface of such a primitive floe is as level as the surface of the sea, and before the winter snows and the drift from the land or other parts of the same floe cover it, it has the texture of a good Brussels carpet on the surface. It is never smooth or glassy like the ice formed on the surface of fresh water. One cannot curl, slide, or skate upon it. Ski stick on it and sledges will not glide over it. The surface is sticky, and even at low temperatures it wets through the thin soles of fur boots and proves very destructive to them. It has a sort of efflorescent appearance and a saltish taste. It is, in fact, the saltest layer of the floe, which may freeze to a thickness of 5 or 6 feet; sometimes less thick when strong currents flow under its surface, sometimes of greater thickness in sheltered lochs and bays. But while the surface of the floe is very salt, if a piece of
