ever, as new deposits are thrown down, they keep near the surface, to be able to get their food ; so that, if to-day a catastrophe were to overwhelm the whole marine life of the Arctic regions, it would be found (supposing by upheaval or otherwise we were able to verify the fact) that the animals would only be imbedded in the upper strata of clay, and that the bottom one, with the exception of a few dead shells, would be azoic ; yet I need.mot say how erroneously we should argue if, from this, we drew the inference that, at the time the bottom layers or strata of this laminated clay were formed, there was no life in the Arctic waters, or that they were formed under circumstances which prevented their being fossiliferous. The bearing of this on the subject in question need scarcely be pointed out. It ought to be noted that, supposing we were able to examine the bottom of the Arctic sea (Davis Straits, for instance), it would be found that this clayey deposit would not be found over the whole surface of it, but only over patches. For instance, all of the ice-fjords would be found full of it to the depth of many feet, shoaling off at the seaward ends ; and certain other places on the coast would be also covered with it ; but the middle and mouth of Davis Straits and Baffin's Bay, and the wide intervals between the different ice-fjords, would either be bare or but slightly covered with small patches from local glaciers ; yet we should reason most grievously in error, did we conclude therefrom that the other portions of the bottom, covered with sand, gravel, or black mud, were laid down at a different period from the other, or under other different conditions than geographical position. These ice-rivers seem, in the first place, to have taken their direction according to the nature of the country over which the inland ice lies, and latterly according to the course of the glaciers. No doubt they branch over the whole country like a regular river-system When the glacier reaches the sea, the stream flows out under the water, and,
1 It may be somewhat superfluous for me to say that these subglacial streams are totally different in nature from the streams which flowed in the old water-courses found under the drift in various parts of the world. These were the beds of the preglacial rivers, and are known to miners as " sand-dykes," " washouts," &c. On the North Pacific slope of the Rocky Mountains they are very common, and are eagerly sought for by the gold-miners, the "old beds " generally yielding a considerable amount of gold. In California, so thoroughly have they been explored by the gold-diggers that, if proper records had been kept, a map of the preglacial rivers might now be drawn, almost as detailed as that of the postglacial or present river-system. The courses of these ancient rivers appear to have been generally in the same direction, and to have had their outlets in the valleys near about the same places as the present rivers. Sometimes these channels seem to cross nearly at right angles. The old Yuba channel, for instance, when its course was interrupted and diverted, ran through the site of the present village of " Timbuctoo," crossing the bed of the present river at Park's Ear; thence running in a north-westerly course, and falling into the Rio de las Plumas (Feather River), near Oroville, a considerable distance from its present junction with that river at Mary'sville. These old channels exhibit the same windings and precipitous falls as the present river; and they have been cut in various places by canons and ravines ; and portions of the older deposit, carried down, mingle with the loose gravel and sand detached by more recent aqueous action.
