Hitherto, during the whole time we sailed along the coast, we had scarcely met with any fields of drift-ice but such as were formed of rotten, even, thin and scattered pieces of ice, in many places almost converted into ice-sludge, without an "ice-foot" and often dirty on the surface. No iceberg had been seen, nor any large glacier ice-blocks, such as on the coasts of Spitzbergen replace the Greenland icebergs. But east of Svjatoinos the ice began to increase in size and assume the same appearance as the ice north of Spitzbergen. It was here, besides, less dirty, and rested on a hard ice-foot projecting deep under water and treacherous for the navigator.
The ice of the Polar Sea may be divided into the following varieties:—
1. Icebergs. The true icebergs have a height above the surface of the water rising to 100 metres. They often ground in a depth of 200 to 300 metres, and have thus sometimes a cross section of up to 400, perhaps 500 metres. Their area may amount to several square kilometres. Such enormous blocks of ice are projected into the North Polar Sea only from the glaciers of Greenland, and according to Payer's statement, from those of Franz-Josef Land also; but not, as some authors (GEIKIE, BROWN, and others) appear to assume and have shown by incorrect ideal drawings, from glaciers which project into the sea and there terminate with a perpendicular evenly-cut border, but from very uneven glaciers which always enter the sea in the bottoms of deep fjords, and are split up into icebergs long before they reach it. It is desirable that those who write on the origin of icebergs, should take into consideration the fact that icebergs are only formed at places where a violent motion takes place in the mass of the ice, which again within a comparatively short time results in the excavation of the deep ice-fjord. The largest iceberg, which, so far as I know, has been measured in that part of the Polar Sea which lies between Spitzbergen and Wrangel Land, is one which Barents saw at Cape Nassau on the 17/7th August 1596. It was sixteen fathoms high, and had grounded in a depth of thirty-six fathoms. In the South Polar Sea icebergs occur in great numbers and of enormous size. If we may assume that they have an origin similar to those of Greenland, it is probable that round the South Pole there is an extensive continent indented by deep fjords.
2. Glacier Ice-blocks. These, which indeed have often been called icebergs, are distinguished from true icebergs not only by their size, but also by the way in which they are formed. They have seldom a cross section of more than thirty or forty metres, and it is only exceptionally that they are more than ten metres high above the surface of the water. They originate from the "calving" of glaciers which project into the sea with a straight and evenly high precipitous border. Such glaciers occur in large numbers on the coasts of Spitzbergen, and they are there of the same height as similar evenly-cut glaciers on Greenland. According to the statement of the Dane PETERSEN, who took part both in KANE'S expedition in 1853-55 and in Torell's in 1861, the glaciers, for instance, at Hinloopen Strait in Spitzbergen, are fully equal, with respect to their size and the height of their borders above the sea-level, to the enormous and much bewritten Humboldt glacier in Greenland. In Spitzbergen too we find at two places miniatures of the Greenland ice-currents, for instance the glacier which filled the North Haven in Bell Sound, another glacier which filled an old Dutch whaling haven between Recherche Bay and Van Keulen Bay, a glacier on the north side of Wablenberg Bay and perhaps at that part of the inland ice marked in my map of the expedition of 1872 as a bay on the east coast of North-east Land. It is even possible that small icebergs may be projected from the last-mentioned place, and thence drift out into the sea on the east coast of Spitzbergen.
Glacier-ice shows a great disposition to fall asunder into smaller pieces without any perceptible cause. It is full of cavities, containing compressed air, which, when the ice melts, bursts its attenuated envelope with a crackling sound like that of the electric spark. It thus behaves in this respect in the same way as some mineral salts which dissolve in water with slight explosions. Barents relates that on the 20/10th August 1596 he anchored his vessel to a block of ice which was aground on the coast of Novaya Zemlya. Suddenly, and without any perceptible cause, the rock of ice burst asunder into hundreds of smaller pieces with a tremendous noise, and to the great terror of all the men on board. Similar occurrences on a smaller scale I have myself witnessed. The cause to which they are due appears to me to be the following. The ice-block while part of the glacier is exposed to very severe pressure, which ceases when it falls into the sea. The pressure now in most cases equalises itself without any bursting asunder, but it sometimes happens that the inner strongly compressed portions of the ice-block cannot, although the pressure has ceased, expand freely in consequence of the continuous ice-envelope by which they are still surrounded. A powerful internal tension must thereby arise in the whole mass, which finally leads to its bursting into a thousand pieces. We have here a Prince Rupert's drop, but one whose diameter may rise to fifty metres, and which consists not of glass but of ice.
Glacier ice-blocks occur abundantly on the coasts of Spitzbergen and north Novaya Zemlya, but appear to be wanting or exceedingly rare along the whole north coast of Asia, between Yugor Schar and Wrangel Land. East of this they again occur, but not in any great numbers. This appears to show that the Western Siberian Polar Sea is not surrounded by any glacial lands. The glacier ice is commonly of a blue colour. When melted it yields a pure water, free of salt. Sometimes however it gives traces of salt, which are derived from the spray which the storms have carried high up on the surface of the glacier.
3. Pieces of ice from the ice-foot formed along the sea beach or the banks of rivers. They rise sometimes five or six metres above the surface of the water. They consist commonly of dirty ice, mixed with earth.
4. River Ice, level, comparatively small ice fields, which, when they reach the sea, are already so rotten that they soon melt away and disappear.
5. The walrus-hunters' Bay Ice; by which we understand level ice-fields formed in fjords and bays along the coast, and which have there been exposed to a comparatively early summer heat. The bay ice therefore melts away completely during summer, and it is not commonly much pressed together. When all the snow upon it has disappeared, there is to be seen above the surface of the water a little ice of the same colour as the water, while under water very considerable portions of unmelted hard ice are still remaining. This has given rise to the walrus-hunters' statement, which has been warmly maintained, that the ice in autumn finally disappears by sinking. Nearly all the ice we met with in the course of our voyage belonged to this variety.
6. Sea Ice, or heavy ice, which often exhibits traces of having been much pressed together, but has not been exposed to any early summer heat. The walrus-hunters call it sea ice, wishing, I imagine, to indicate thereby that it is formed in the sea farther up towards the north. That it has drifted down from the north is indeed correct, but that it has been formed far from land over a considerable depth in the open sea is perhaps uncertain, as the ice that is formed there cannot, we think, be very thick. It has rather perhaps drifted down from the neighbourhood of some yet unknown Polar continent. Of this ice are formed most of the ice-fields in the seas east of Greenland, north of Spitzbergen, between Spitzbergen and the north island of Novaya Zemlya, and north of Behring's Straits. In the northern seas it does not melt completely during the summer, and remains of sea ice therefore often enter as component parts into the bay ice formed during the following winter. The latter then becomes rough and uneven, from remnants of old sea ice being frozen into the newly formed ice. Sea ice is often pressed together so as to form great torosses or ice-casts, formed of pieces of ice which at first are angular and piled loose on each other, but gradually become rounded, and freeze together into enormous blocks of ice, which, together with the glacier ice-blocks, form the principal mass of the ground ice found on the coasts of the Polar lands. The water which is obtained by melting sea-ice is not completely free from salt, but the older it is the less salt does it contain.