Other formats

    TEI XML file   ePub eBook file  

Connect

    mail icontwitter iconBlogspot iconrss icon

Report on the Geology & Gold Fields of Otago

Section V. — Historical Geology

page 74

Section V.
Historical Geology.

Having described all the facts known to me relating to the geology of the Province, it remains now to collate them, and try to deduce from them the causes to which they owe their origin, or in other words the geological history of the Province. I have kept this section distinct from the last because it stands on a differeut footing. The last section simply described my observations, which are open to all to examine and verify or correct as the case may be; while the present section is an attempt to theorise on those observations, and to build up from them a history of the geological changes that have taken place in this portion of the earth’s surface. But as these observations are necessarily very incomplete, and indeed sometimes probably erroneous, it is evident that too much stress must not be laid on the accuracy of the conclusions drawn from them. Even if all the facts were as accurately known as possible, the interpretation of them would not be so easy as might be supposed, for sometimes the facts admit of more than one explanation. Nevertheless whatever may be the difficulties in the way, we must try our best to overcome them, for not only will a knowledge of the geological history of New Zealand be of the utmost importance to general theoretical geology, but it will also throw much light on the deposition of our auriferous alluvia, and it has therefore its immediate practical, as well as its future theoretical value. Consequently it is of no use saying that we ought to wait until our knowledge is more complete before attempting to theorise, for if we did so we might wait for ever, and still not be satisfied as to the fullness of our evidence; while if we carefully endeavour to make out from the observations before us the most probable history of the geological changes in Otago, we shall at any rate be in the best possible position that our present knowledge admits of for solving accurately the practical problems presented to us.

Eozoic and Palæozoic eras.—Our knowledge of the eozoic and palæozoic rocks of Otago is too limited to enable us to speculate much about them. It is, however, evident from the facts previously mentioned (page 40), that the metamorphism of the Manipori formation has no connection with the eruption of granite at Pre-page 75servation and Chalky Inlets, and that most probably the metamorphism of the Wanaka and Kakanui formations took place during or after the deposition of the rocks of the Kaikoura formation; and, as the Kaikoura formation is not more altered where it rests on the Manipori formation than it is at other places, it follows that the more intense metamorphism of the latter formation must have occurred at an earlier period than that of the Wanaka and Kakanui formations. This shows a considerable difference in age between the Manipori and Wanaka formations, which is further proved by the abundance of dykes in the former which do not appear to penetrate any higher.

The slight unconformity that exists between the Kakanui and Kaikoura formations, shows that, at an intermediate age (perhaps during the Devonian period), an extent of low flat land existed in Otago which underwent a slight denudation. This land, the earliest of which we have any evidence in New Zealand, sank during the deposition of the Kaikoura formation (perhaps the carboniferous period), and was afterwards elevated again into dry land in the Permian period.

This last elevation is proved by the unconformity that exists between the Maitai and Kaikoura formations, which, however, is not so great in Otago as it is further north. But even in the north there is perhaps no evidence that New Zealand, during the Permian period, was a mountainous region, for there are no signs of any great local denudations having taken place, such as would be produced by mountain streams, and it is more probable that New Zealand then formed a very subordinate part of a large continent, which, judging by the similarity of the shells and plants found in the following formations with those found in Australia, India, and Europe, probably stretched away far to the northward.

Mesozoic era.—At the commencement of the triassic period, this continent began in New Zealand to subside below the sea; but from the occurrence of plant remains throughout the Maitai formation, it is probable that this subsidence was not very great, and that the southern portion of the Permian continent was not yet quite submerged, when another slight elevation, accompanied by an outburst of volcanic action, took place, which caused the continental vegetation again to advance to the southern and western portion of New Zealand. This was followed by subsidence, during which the mixed estuarine and marine Putataka formation was deposited. The granite also, erupted into the palæozoic rocks during the deposition of the upper part of the Maitai formation, was exposed at the surface by denudation, and fragments of it were brought down by rivers running from west to east and were mixed with the gravel on the sea shore. At the close of this formation (that is to say towards perhaps the middle of the jurassic period), the whole of the country was again elevated, and the chain of the New Zealand Alps formed.

page 76

The proofs of this elevation are (1), That all the rocks of precretaceous age partake in the main foldings of the strata (see page 24); (2), The absence of all rocks belonging to the upper jurassic and lower cretaceous periods, and (3), The total unconformity between the Putataka and the Waipara formations, which is especially well seen in the Provinces of Canterbury and Marlborough, where rocks belonging to the Waipara formation are found filling up valleys of erosion in the Maitai and Putataka formations, proving conclusively that the country had previously undergone a considerable amount of subærial denudation. The foldings of all the rocks older than the Waipara formation, show that this upper secondary elevation was no local affair, but part of a large movement, which probably resulted in the upheaval of a continent. How far this antarctic continent extended we have as yet no means of knowing, but that it probably extended to South America is shown by all the formations later than this upheaval containing fossils related to, or identical with those of Patagonia and Chili. Since this upheaval the New Zealand Alps have never been submerged.

Subsequently (upper cretaceous period) this continent subsided, and at the close of the mesozoic era, the north part of this island was 6000 feet, and Otago was 3000 feet lower than at present. This is proved by the height to which the Waipara formation extends in Marlborough (Benmore) and Otago (Mount Hamilton) respectively. That the Alps were not entirely submerged, is shown by the Waipara formation, which was deposited during this depression, abutting against the older rocks, and occurring in such positions as to show that it could never have overspread the whole of the Alps. This, however, can be better proved in Canterbury and Marlborough than in Otago. It was at this time that the sea swarmed with huge saurians (plesiosaurus, &c.), and although none of their remains are as yet known to occur in Otago, they may still be found between Shag Point and the Otepopo River in the hills behind Hampden.

New Zealand was now reduced to a narrow chain of islands, and from that time to the present it has always remained isolated from any large continental area.

Cainozoic era.—At quite the commencement of the tertiary era, New Zealand was again upheaved. This upheaval was not so violent as the last, and probably therefore did not affect so large an area. Still in Otago, Marlborough, and Nelson, a certain amount of folding of the rocks took place, and the lately formed beds of the Waipara formation were caught up in them. In Canterbury, however, this was not the case, and the Waipara beds still continued in a horizontal position, although elevated above the sea*

* I mention these facts, although referring to districts outside of Otago, as we are by means of them enabled to roughly estimate the difference between this and the former elevation. They are all derived from personal observation of those districts.

page 77The proofs of this upheaval are—(1) The complete uniformity between the Waipara and Oamaru formations as seen at Mount Hamilton (see p. 51) and Shag Valley (see p. 50) in Otago, and in the Weka pass, district in Canterbury.* (2) The absence of rocks of lower eocene age, the fossils from the Oamaru formation being quite distinct from those of the Waipara formation, and containing some 13 or 14 per cent, of still living forms; and (3) the great amount of subærial denudation that had taken place between the two periods, such as the hollowing out of the Shag Valley (see p. 50), and the denudation of the north side of Mount Hamilton and Centre Hill (see p. 51).

It was this lower eocene upheaval that gave the final form to the internal geological structure of the Alps, for all the rocks of a later date are, with local exceptions, more or less horizontal, and circle round the bases of the hills formed by the mesozoic and palæozoic rocks. Thus showing that since the deposition of the tertiary strata no great internal changes in the position of the rock masses have taken place, although vast external changes, caused by denudation, have torn the tertiary rocks into fragments.

But not only was the last touch given at this time to the geological structure of the Alps, but the chief valleys were also marked out at this period. For example, we find our oldest tertiary rocks occupying the valley of the Waitaki as far as Big Gully Creek, and the Maruwhenua to beyond the township. In the valley of the Shag they extend up nearly to its source, stretching indeed once beyond that into the Maniototo Plain as far as the Kyeburn. We find them again at Waihola gorge, and at Lake Wakatipu, and in some of the branches of the Shotover, shewing that these even had been hollowed out during the eocene period. In the valley of the Mataura they occur in the Waimea Plains, and they stretch up the valley of the Waiau as far as Manipori Lake and Centre Hill. These facts prove convincingly that the Alps had undergone enormous denudation before the close of the eocene period, and that most of the present valleys in them had been hollowed out to a considerable extent at that time. But they also prove that at the close of the eocene period Otago was again depressed to about 1500 or 2000 feet below its present level, and that the sea then entered the long winding valleys that had been previously formed, and penetrated as far as Te Anau and Wakatipu Lakes. It was at this period of greatest depression that volcanic action burst out with such energy in the neighbourhood of Dunedin and other places; and during the middle miocene period Otago was again slightly raised above its present level, so that no rocks belonging to the Ahuriri formation could be deposited here. This elevation was probably not very great, for the Ahuriri formation occurs in

* Reports of Geological Explorations, 1872·3, p·44.

page 78the Waimakariri Valley, in Canterbury, at an elevation of about 3000 feet above the sea, and if the Province of Canterbury was so much depressed at that time, Otago is not likely to have been much elevated. It was, however, sufficient to cause the Oamaru formation to be considerably denuded, and the Pareora formation now lies, as in the Shag Valley, in the hollows that were then formed in it.

During the upper miocene period the whole of the South Island was considerably depressed, and, if I am right in supposing that Mount Prospect near Lake Te Anau is composed of rocks of that age, Otago must then have stood at least 3246 feet lower than at present.

No marine rocks of older Pliocene age are known in any part of New Zealand, and consequently I infer that at that time the whole country stood at a higher level than at present; but at Wanganui there is a bed of blue clay containing marine shells, of which about 24 per cent, are now extinct, which, therefore, we must refer to the newer Pliocene period. Also in many places round the coast of New Zealand, such as Oamaru, Mataura, Wanganui, Taranaki, Cape Kidnappers, &c., we find pleistocene deposits containing marine shells, in some of which all are still living, and in others there are from 5 to 10 per cent. of extinct forms. Consequently there can have been no general elevation of the land extending over a long period since Pliocene times.

That the subsidence which followed this last elevation was continued until Otago stood at a much lower level than at present, and that it was subsequently followed by an elevation, which may probably still be going on, is proved by many facts. The silt deposit at Oamaru, already described, proves conclusively a subsidence in pleistocene times to an extent of at least 500 feet, and consequently a subsequent elevation of a like amount. At Bushy Park, south of Shag River, a raised beach occurs at about 100 feet above the present sea level. On the seaward side of the Dunedin Peninsula, High Cliff, which is about 900 feet high, presents a remarkable series of five or six terraces (fig. 18) apparantly due to marine denudation at various levels, but I have not been able to examine it closely, and it is possible that in so hard a rock these terraces may have kept their shape for along time, and they may have been formed during pauses in the movement of a previous subsidence, or they may be caused by harder ledges of rock. For these terraces and even cliffs are not by any means of universal occurrence on the east coast of Otago, the hills sometimes sloping gradually down to the sea. The best example that I know of
Fig. 18. High Cliff, Dunedin.

Fig. 18. High Cliff, Dunedin.

page 79this is at Quoin Point, between the Taieri and the Tokomairiro (fig. 19), which exhibits a very characteristic plain of marine denudation. Akatore Hills.
Fig. 19.—Quoin Point.

Fig. 19.—Quoin Point.

The Southland plains are distinctly terraced, and as Dr. Hector remarks are "marked by ridges which doubtless indicate the successive channels which were formed by the Mataura and Oreti rivers at the time they poured their waters into an extension of the present bay, prior to the last elevation of the land."* The network of gravel beds which surround the Hokanui and Moonlight ranges also shew evidence of marine action, for it is impossible that the alluvia of the Mataura, the Oreti, and the Jacobs River should all join one another with gradual slopes behind the seaward range of hills, without the intervention of some uniform widely acting cause, such as the sea; and we have further evidence of this action in the flat topped beds of eocene limestone in the Waimea plains. This would show that the land here has been lately about 600 feet lower than at present, which corresponds very well with the height of the silt formation at Oamaru. The chain of small islands between Paterson’s Inlet and Ruapuke are also quite flat-topped, the result evidently of marine denudation. They are between 140 and 200 feet high.

In the West Coast Sounds nothing perhaps is more striking to the geological eye than the contrast between the immense, almost perpendicular, cliffs found up the Sounds and the absence of any high sea cliffs between them. This certainly implies that the land must be moving either up or down, and constantly therefore presenting a new line of action to the surf. For if the land had been stationary, the furious swell of the Pacific Ocean, impelled against the same line of coast by the almost constant succession of westerly gales, must have long ago eroded them Back so far that an iron bound coast line of high cliffs would have been formed all along that part of the coast where the mountains approach close to the sea. But this is by no means the case. At all the headlands between the Sounds the spurs of the mountains, covered with bush, slope at angles of between 25° and 50° to the sea, and only a line of low cliffs is seen at their base. (See fig. 21.) That this movement is one of elevation there are several proofs.

On Coal Island, just opposite Otago’s Retreat, there is a sea worn cave situated about 12 feet above the present sea level, and Dr. Hector mentions others at elevations of from 10 to 20 feet. Off Green Islets there is a remarkable rock, between 90 and 100

* Geological Exploration of the West Coast of Otago, p. 441.

page 80feet high, which at a height of about 40 feet is cut into a step-like terrace all round. (See fig. 20, a and b) On the east side this terrace is cut into two steps (fig. 20, a), and just on a level with the terrace the rock is pierced completely through in a north and south direction. It is quite evident that both the
Pierced Rocks of Green Islet.Fig. 20. Pierced Rock off Green Islets.—a, view looking north; b, view looking west; c, position of hole.

Pierced Rocks of Green Islet.
Fig. 20. Pierced Rock off Green Islets.—a, view looking north; b, view looking west; c, position of hole.

terrace and the hole were caused by marine denudation when the land stood 40 feet lower than at present. The low land east
Fig. 21, Entrance to Doubtful Sound from the north.

Fig. 21, Entrance to Doubtful Sound from the north.

of Windsor Point is also distinctly terraced. These terraces can be easily recognised, but I was not able to estimate their height.

The entrance to Doubtful Sound is also distinctly terraced on both sides, (fig. 21), and I estimated the highest to be about 800 feet above the sea level.

On the other hand Dr. Hector has brought forward the following evidence in favour of the view that these Sounds are at present undergoing subsidence. Speaking of the streams that run into Edwardson Sound, a branch of Chalky Inlet, he says "the flats along the lower parts of these streams are true valley deposits, such as may be seen in any mountain valley, formed by the gradual change of the water course from side to side; and as I did not discover any remains above high water mark of the brackish water deposit with estuarine shells, which is now slowly filling up the basin, or any trace of terraces round the mouths of the rivers, I conclude that the land at the head of this sound, unlike most parts of the New Zealand coast, is not rising; and the consideration of the page 81nature of the falls almost demonstrates that it is, on the other hand, an area that is being submerged. The examination of the walls of the chasm through which the first mentioned river—[the north river at the head of Edwardson Sound]—falls—[the falls are 25 feet high, and the river falls into a pool in the chasm 36 feet deep]—convinced me that it has been cut by the action of falling water, which may have followed a pre-existing fissure. The rock is a granular quartzite, compact and close grained in texture, but still such a rock that can be slowly worn away by the mechanical action of running water. The depth of the rocky chasm below the fall must, therefore, of itself prove that the fall was once higher, and as it now falls to the sea level, as a necessary consequence it follows that the land must then have been more elevated."*

Again, speaking of the low neck of land, some 400 yards in length, between Cunaris Sound in Chalky Inlet and Last Cove in Preservation Inlet, he says: "This neck of land is quite low, but bounded on both sides by lofty and precipitous mountains. The torrents which descend from these flow into Preservation Inlet, and accordingly add to the eastern side of the isthmus by the quantity of subangular fragments of rock that they bring down. ∗ ∗ ∗ The absence of the characteristic terraces which mark the existence of a ‘col,’ or ancient strait, was a further proof that the coast line here is not rising, as in a situation like this they could hardly fail to be present." (1. c. p. 452.)

This is all the evidence in favour of subsidence that I can find in Dr. Hector’s Report, and I do not think that it has anything like the weight of that previously adduced in favour of elevation. In such steep-sided ravines as are these sounds, no deposits form along the sides except at the mouths of streams, and this, together with the heavy rains that are constantly falling, are quite sufficient to account for the absence of any brackish water deposit with estuarine shells above high water mark. For if formed on the sides they would be all washed down, and if formed at the mouths of the streams they would be at once either washed away or covered up with debris. Terraces could not be expected close to the mouth of the rivers, but only high up them; because, as the laud rises, the mouth of the river is constantly advanced at the sea level, and the terraces are cut out higher up where the mouth used to be. Dr. Hector describes the flats along the lower parts of the Streams as "true valley deposits;" but these streams are constantly bringing down gravel and boulders of all sizes, which cover up the estuarine deposit. The flats at the head of Bradshaw Sound, Bligh Sound, and Milford Sound, which are all that I was able to examine, I found to consist of fine mud and sand which had evidently been deposited in quiet water, and it was quite different from the material now being brought down by the streams, and which covers up the

* Geological Exploration of the West Coast of Otago, p. 451.

page 82mud deposit higher up the stream. I certainly did not notice any terraces, but the flats at the mouth of these streams are covered with such a dense impenetrable vegetation, that exploration far from the river bank is almost impossible. The opinion that I formed after an examination of these flats was quite opposite to that of Dr. Hector, for I thought that they showed evidence of elevation.

The deep pool in the chasm, mentioned by Dr. Hector, has, no doubt, like the sounds themselves, been formed when the land stood at a higher elevation than at present, but I do not think that the hole necessarily implies recent subsidence, for its shape might have been preserved below the water for an immense period of time. With regard to the last evidence brought forward by Dr, Hector, viz., the absence of terraces on the low neck of land between Chalky and Preservation Inlets, even if we grant that terraces are characteristic of a col, which, however, I am not prepared to do, still, they could hardly be expected in this case, for the tidal current at the head of these sounds is very slight, and even if a terrace had been cut, it would have been at once obliterated by the immense quantity of debris that falls from the lofty and precipitous mountains on either side of the col.

The Rev. R. Taylor, in his "New Zealand and its Inhabitants" (London, 1855), states that earthquakes in 1826 and 1827 so altered the form of a small cove, called "The Jail" by the sealers, about 80 miles north of Dusky Sound, that it could hardly be recognised. The cove became dry land, and trees wore seen under water near the coast, having probably been carried down by landslips into what was previously deep water.

It has been already mentioned that all the river valleys in the interior exhibit a remarkable series of alluvial terraces. These terraces are generally supposed by geologists to be caused by the rivers rapidly cutting down their beds during elevation of the land, but as Dr. Haast has stated* that there is another agency by which alluvial terraces are formed, "namely the retreat of the river sources to higher and more distant regions," it is necessary for me to show that he is mistaken on this point.

When the land is stationary or sinking, the lateral denudation on the sides of the valley is as great as the vertical denudation, and consequently no terraces would be left. But if by any combination of causes a terrace should he left, it would by this process be a rock terrace, and not formed of alluvium. Dr. Haast’s theory also would not account for terraces in the lower portions of the rivers, for here,

* Report on the formation of the Canterbury plains, p. 14. The other reason given in this Report to account for terraces, viz., "The permanent diminution of their (the rivers) waters through changes at or near their sources," need not be discussed, for it would act in exactly the opposite direction to that supposed by Dr. Haast.

Trans. N. Z. Institute, vi., p423.

page 83when the land is stationary, the constant level of the sea prevents the rivers from deepening their channels, and when the land is sinking, it is obvious that all rivers must be filling up their old beds throughout the whole of their lower portions. If Dr. Haast was right in his surmise, we should see terraces-equally well developed in all mountainous countries, and those of Wales and Switzerland ought to be as remarkable as those of New Zealand, which is very far from being the case.

The evidence, therefore, appears to me to show that the latest movement in Otago has been one of upheaval, both on the east and and west coasts, and that the amount of this upheaval has been, at least, 500 feet.*

Glacier period.—In the last section I adduced abundant evidence to show that the Otago glaciers formerly extended much further than they do now, and this could only have been caused either by a much colder climate than at present, or by the land having stood at a much greater altitude. Now, although we have proofs of the former great extension of our glaciers, we have no proof whatever that they descended nearer to the sea level than at present, although they probably would do so when they were larger, and had larger supplies of snow. But there is no proof whatever that they reached the sea. There are no stratified moraines nor boulder clay to be seen; no marine shells have been found in any of the glacier deposits, even when, as in the case of the moraine near the mouth of the Taieri, they are now at the sea level. Also, there are no signs in any of the newer pliocene and pleistocene deposits of any change in the fauna that would indicate a colder climate. On the contrary, the pliocence and pleistocene beds at Wanganui contain recent species of shells that cannot now live south of Cook Strait, while the cold that would be necessary to bring Back our glaciers to their former dimensions would be sufficient to exterminate throughout New Zealand all those shells that cannot now live on the coasts of Otago. We find, however, that some of them have existed in New Zealand ever since the upper miocene period. We have, therefore,

* Captain Fairchild informs me that a Weka (Ocydromus heciori (?),) is found on the Solanders; consequently, this island cannot have been entirely submerged since it was separated from New Zealand. It is 1,100 feet high, and an elevation of 400 feet would connect it with New Zealand.

Dr. Hector and Dr. Haast have suggested that the mountains may have had a more plateau-like form, and consequently, collected more snow during our glacier period than now. This, no doubt, would tend to increase the size of the glaciers, but we must remember that the New Zealand Alps have been undergoing denudation ever since the jurassic period, and that in the upper eocene period, many of the valleys in them had been hollowed out nearly as deeply as now, consequently, during the comparatively short period that has elapsed since the date of our last great glacier period, denudation cannot have affected the shape of the mountains to such an extent as to make it worth while to take this cause into consideration.

page 84no trace of a cold or glacial epoch, and we are obliged to fall Back on elevation of the land as the main cause of the former extension of our glaciers. And, as we have already seen that New Zealand has not stood at a higher level than at present since older pliocene times it is to that date that we must refer our last glacier period.

If we now turn for confirmatory evidence to the glacier deposits themselves, and to the older alluvia, we find that the external forms of the older moraines are almost entirely obliterated, as for instance that at the junction of the Lindis with the Clutha, and that at Black Mount; while that near Taieri Mouth has not only lost its external shape, but is almost entirely covered on the seaward side by what appears to be marine gravel. Others of a rather later date have also been much modified as at the southern extremities of Lakes Wakatipu, Wanaka, and Hawea. We also find that the polished surfaces and striæ have completely, disappeared in places where we know that ice must once have passed over. In some places, as near Queenstown, the rounded, smooth surface still remains, but in others not only has the polishing and grooving been obliterated, but subsequent weathering has so altered the once "moutonneés" surfaces as to make them quite unrecognisable, as in the west coast sounds; and there are places in the Maniototo and Dunstan districts, where masses of rock ten or twelve feet in thickness have been removed since the ice passed over them.

This great amount of weathering may in part be due to the nature of the rocks, but the modifications of the moraines certainly indicate their great age, and a study of the older river alluvia leads us to the same conclusion. For instance, I have pointed out that in the Shotover there are two distinct river channels separated by a sharp ridge of rock, and that the older channel is filled with gravel (see fig 16). Now it is evident that no glacier can have come down the Shotover since the older channel was filled up, or it would have ploughed all the gravel out again and rounded the sharp ridge between the two. Indeed the V-like form of the older channel, as well as the gold found in the bottom of it, prove that it was cut out by the river, and not by a glacier, which are known not to cut sharp V-shaped but round bottomed U-shaped valleys. Remembering, therefore, that rivers deepen their beds when the land is rising and their velocity increasing, and fill them up when the land is sinking and their velocity consequently diminishing, we see that the Shotover supplies evidence that since any glacier came down it, the land has been elevated, then depressed, and subsequently elevated again. And if to this we add the first depression which caused the glacier to retire, we have two depressions and two elevations subsequent to the great extension of the glaciers. The same conclusion will be arrived at by a study either of Lammerlaw Creek, at Waipori, or Few’s Creek, at Lake Wakatipu.

page break
Geology of Otago. Hutton Plate II

Geology of Otago. Hutton Plate II

page 85

If also the old lake basins in the interior have been hollowed out by glaciers, * it is evident that an immense length of time must be allowed for filling them up.

Consequently we have proofs not only of the great length of time that has elapsed since the greatest extension of the glaciers took place, but also of a second elevation, and consequent second advance of the glaciers, which, however, was less than the first, as the ice did not a second time come down the Shotover. These two elevations we may with much probability place between the Pareora and Wanganui formations, and between the Wanganui and pleistocene deposits respectively; the earlier elevation being the greater and the longer continued. The New Zealand glacier period has therefore nothing in common with the glacial period of the northern hemisphere.

If now, in order to assist the memory, we throw the various oscillations in level that Otago has undergone into the form of a diagram, as in Plate II., where the central horizontal line which represents the present elevation of the Province, is divided into spaces proportional to the length of the different geological periods, we see that these oscillations have been more numerous and more extensive since the middle of the jurassic period than before it; and that since that period there have been three main upheavals, and two minor ones, or five in all; of which the first represents the period of the antarctic continent, and the last but one the period of the greatest extension of our glaciers. It is, however, probable that we had a glacier period during each of the main upheavals, and indeed we have proofs, as I shall show in the next section, that there were large glaciers in Otago during the middle, or eocene upheaval.

* The proofs of this will be given in the next section.