changes of climate are due to changes in the distribution of sea and land, as advocated by Sir Charles Lyell. But these difficulties disappear if we adopt the views advocated in the foregoing chapters. As Captain Osborn has pointed out, however, Sir Roderick's hypothesis leaves the real difficulty untouched. "A very different climate," he says, "must then have existed in those regions to allow driftwood so perfect as to retain its bark to reach such great distances; and perhaps it may be argued that if that sea was sufficiently clear of ice to allow such timber to drift unscathed to Prince Patrick's Land, that that very absence of a frozen sea would allow fir-trees to grow in a soil naturally fertile." *

As has been already stated, all who have seen those trees in arctic regions agree in thinking that they grew in situ. And Professor Haughton, in his excellent account of the arctic archipelago appended to McClintock's "Narrative of Arctic Discoveries," after a careful examination of the entire evidence on the subject, is distinctly of the same opinion; while the recent researches of Professor Heer put it beyond doubt that the drift theory must be abandoned.

Undoubtedly the arctic archipelago was submerged to an extent that could have admitted of those trees being floated to their present positions. This, as we shall see, follows from theory; but submergence, without a warmer condition of climate, would not enable trees to reach those regions with their bark entire.

But in reality we are not left to theorise on the subject, for we have a well-authenticated case of one of those trees being got by Captain Belcher standing erect in the position in which it grew. It was found immediately to the northward of the narrow strait opening into Wellington Sound, in lat. 75° 32′ N. long. 92° W., and about a mile and a half inland. The tree was dug up out of the frozen ground, and along with it a portion of the soil which was immediately in contact with the roots. The whole was packed in canvas and brought to "McClure's North-West Passage," p. 214. Second Edition.

England. Near to the spot several knolls of peat mosses about nine inches in depth were found, containing the bones of the lemming in great numbers. The tree in question was examined by Sir William Hooker, who gave the following report concerning it, which bears out strongly the fact of its having grown

in situ.

"The piece of wood brought by Sir Edward Belcher from the shores of Wellington Channel belongs to a species of pine, probably to the Pinus (Abies) alba, the most northern conifer. The structure of the wood of the specimen brought home differs remarkably in its anatomical character from that of any other conifer with which I am acquainted. Each concentric ring (or annual growth) consists of two zones of tissue; one, the outer, that towards the circumference, is broader, of a pale colour, and consists of ordinary tubes of fibres of wood, marked with discs common to all coniferæ. These discs are usually opposite one another when more than one row of them occur in the direction of the length of the fibre; and, what is very unusual, present radiating lines from the central depression to the circumference. Secondly, the inner zone of each annual ring of wood is narrower, of a dark colour, and formed of more slender woody fibres, with thicker walls in proportion to their diameter. These tubes have few or no discs upon them, but are covered with spiral striæ, giving the appearance of each tube being formed of a twisted band. The above characters prevail in all parts of the wood, but are slightly modified in different rings. Thus the outer zone is broader in some than in others, the discbearing fibres of the outer zone are sometimes faintly marked with spiral striæ, and the spirally marked fibres of the inner zone sometimes bear discs. These appearances suggest the annual recurrence of some special cause that shall thus modify the first and last formed fibres of each year's deposit, so that that first formed may differ in amount as well as in kind from that last formed; and the peculiar conditions of an arctic climate appear to afford an adequate solution. The inner, or first-formed zone, must be regarded as imperfectly developed,

being deposited at a season when the functions of the plant are very intermittently exercised, and when a few short hours of sunshine are daily succeeded by many of extreme cold. As the season advances the sun's heat and light are continuous during the greater part of the twenty-four hours, and the newly formed wood fibres are hence more perfectly developed, they are much longer, present no signs of striæ, but are studded with discs of a more highly organized structure than are usual in the natural order to which this tree belongs."*

Another circumstance which shows that the tree had grown where it was found is the fact that in digging up the roots portions of the leaves were obtained. It may also be mentioned that near this place was found an old river channel cut deeply into the rock, which, at some remote period, when the climate must have been less rigorous than at present, had been occupied by a river of considerable size.

Now, it is evident that if a tree could have grown at Wellington Sound, there is no reason why one might not have grown at Banks's Land, or at Prince Patrick's Island. And, if the climatic condition of the country would allow one tree to grow, it would equally as well allow a hundred, a thousand, or a whole forest. If this, then, be the case, Sir Roderick's objection to the theory of growth in situ falls to the ground.

Another circumstance which favours the idea that those trees grew during the glacial epoch is the fact that although they are recent, geologically speaking, and belong to the drift series, yet they are, historically speaking, very old. The wood, though not fossilized, is so hardened and changed by age that it will scarcely burn.

"British Association Report for 1855," p. 381. "The Last of the Arctic Voyages," vol. i., p. 381.

CHAPTER XVII.

FORMER GLACIAL EPOCHS. REASON OF THE IMPERFECTION OF GEOLOGICAL RECORDS IN REFERENCE TO THEM.

Two Reasons why so little is known of Glacial Epochs.-Evidence of Glaciation to be found on Land-surfaces.--Where are all our ancient Land-surfaces ?The stratified Rocks consist of a Series of old Sea-bottoms.-Transformation of a Land-surface into a Sea-bottom obliterates all Traces of Glaciation.Why so little remains of the Boulder Clays of former Glacial Epochs.-Records of the Glacial Epoch are fast disappearing - Icebergs do not striate the Se-bottom.-Mr. Campbell's Observations on the Coast of Labrador.-Amount of Material transported by Icebergs much exaggerated.- Mr. Packard on the Glacial Phenomena of Labrador.-Boulder Clay the Product of Land-ice.-Paleontological Evidence.-Paucity of Life characteristic of a Glacial Period.-Warm Periods better represented by Organic Remains than cold.-Why the Climate of the Tertiary Period was supposed to be warmer than the present.--Mr. James Geikie on the Defects of Palæontological Evidence.-Conclusion.

Tro Reasons why so little is known of former Glacial Epochs.If the glacial epoch resulted from the causes discussed in the foregoing chapters, then such epochs must have frequently supervened. We may, therefore, now proceed to consider what evidence there is for the former occurrence of excessive condi

tions of climate during previous geological ages. When we begin our inquiry, however, we soon find that the facts which have been recorded as evidence in favour of the action of ice in former geological epochs are very scanty indeed. Two obvious reasons for this may be given, namely, (1) The imperfection of the geological records themselves, and (2) the little attention hitherto paid toward researches of this kind. The notion, once so prevalent, that the climate of our earth was much warmer in the earlier geological ages than it is now, and that it has ever since been gradually becoming cooler, was wholly at variance

with the idea of former ice-periods. And this conviction of the à priori improbability of cold periods having obtained during Paleozoic and Mesozoic ages tended to prevent due attention being paid to such facts as seemed to bear upon the subject. But our limited knowledge of former glacial epochs must no doubt be attributed chiefly to the actual imperfection of the geological records. So great is this imperfection that the mere absence of direct geological evidence cannot reasonably be regarded as sufficient proof that the conclusions derived from astronomical and physical considerations regarding former ice-periods are improbable. Nor is this all. The geological records of ancient glacial conditions are not only imperfect, but, as I shall endeavour to show, this imperfection follows as a natural consequence from the principles of geology itself. There are not merely so many blanks or gaps in the records, but a reason exists in the very nature of geological evidence why such breaks in the record might reasonably be expected to occur.

Evidence of Glaciation to be found chiefly on Land-surfaces.— It is on a land-surface that the principal traces of the action of ice during a glacial epoch are left, for it is there that the stones are chiefly striated, the rocks ground down, and the boulder clay formed. But where are all our ancient land-surfaces? They are not to be found. The total thickness of the stratified rocks of Great Britain is, according to Professor Ramsay, nearly fourteen miles. But from the top to the bottom of this enormous pile of deposits there is hardly a single land-surface to be detected. True patches of old land-surfaces of a local character exist, such, for example, as the dirt-beds of Portland; but, with the exception of coal-seams, every general formation from top to bottom has been accumulated under water, and none but the under-clays ever existed as a land-surface. And it is here, in such a formation, that the geologist has to collect all his information regarding the existence of former glacial epochs. The entire stratified rocks of the globe, with the exception of the coal-beds and under-clays (in neither of which would one expect to find traces of ice-action), consist almost entirely of a