and portions wore rent off and floated away. If this be the true history of the formation of these icebergs, the absence of all land débris in the portion exposed above the surface of the sea is readily understood. If any such exist, it must be confined to the lower part of the berg, to that part which has at one time or other moved on the floor of the ice-cap.
"The icebergs, when they are first dispersed, float in from 200 to 250 fathoms. When, therefore, they have been drifted to latitudes of 65° or 64° south, the bottom of the berg just reaches the layer at which the temperature of the water is distinctly rising, and it is rapidly melted, and the mud and the pebbles with which it is more or less charged are precipitated. That this precipitation takes place all over the area where the icebergs are breaking up, constantly, and to a considerable extent, is evident from the fact of the soundings being entirely composed of such deposits; for the diatoms, Globigerinæ, and radiolarians, are present on the surface in large numbers; and unless the deposit from the ice were abundant it would soon be covered and masked by a layer of the exuvia of surface organisms."
The observations which have been detailed leave no doubt that the antarctic sea-bottom, from a little to the south of the fiftieth parallel, as far as 80 south, is being covered by a fine deposit of silicious mud, more or less mixed, in some parts, with the ice-borne débris of polar lands and with the ejections of volcanoes. The silicious particles which constitute this mud are derived, in part, from the diatomaceous plants and radiolarian animals which throng the surface, and, in part, from the spicula of sponges which live at the bottom. The evidence respecting the corresponding arctic area is less complete, but it is sufficient to justify the conclusion that an essentially similar silicious cap is being formed around the northern pole.
There is no doubt that the constituent particles of this mud may agglomerate into a dense rock, such as that formed at Oran, on the shores of the Mediterranean, which is made up of similar materials. Moreover, in the case of fresh-water deposits of this kind, it is certain that the action of percolating water may convert the originally soft and friable, fine-grained sandstone into a dense semi-transparent opaline stone, the silicious organized skeletons' being dissolved, and the silex redeposited in an amorphous state. Whether such a metamorphosis as this occurs in submarine deposits, as well as in those formed in fresh water, does not appear; but there seems no reason to doubt that it may. And hence it may not be hazardous to conclude that very ordinary metamorphic agencies may convert these polar caps into a form of quartzite.
In the great intermediate zone, occupying some 110 of latitude, which separates the circurapolar arctic and antarctic areas of silicious deposit, the diatoms and Radiolaria of the surface-water and the sponges of the bottom do not die out, and, so far as some forms are concerned, do not even appear to diminish in total number; though, on a rough estimate, it would appear that the proportion of Radiola-
