Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 36, 1903
This text is also available in PDF
(868 KB) Opens in new window
– 418 –

Art. XXXII.—On the Geology of North Head, Waikouaiti, and its Relation to the Geological History of Dunedin.

[Read before the Otago Institute, 13th October, 1903.)

Plate XXXII.

THE hill forming Waikouaiti North Head, locally known as Mount Cronin, is terminated on its south side—that is, on the side facing the inlet—by a sharply defined escarpment of great height. At the foot of this escarpment there is an immense talus which descends by a series of steep wave-like undulations towards the sea, and terminates against the shoreline in a sea-wall of irregular height, broken at intervals by breaches over which it is possible to gain the rolling surface of the talus downs above.

The talus is principally composed of large and small masses of sandstone that have slipped down from the cliffs above. It seems to owe its origin to a series of landslips in places so extensive that masses of rocks many hundreds of square yards in superficial area have settled down in such a manner as to retain their original position relatively to the horizon.

The fallen masses of sandstone forming the cliffs at the end of the sandy beach are so disposed that the original planes of deposition are nearly horizontal, thereby giving the impression, at first examination, that they are in situ, a decep-

– 419 –

tion not readily dispelled until the face of the rocky escarpment, above is carefully examined. Proceeding, however, along the sea-rim of the talus from the end of the sandy beach the masses of rock forming the sea-wall are found to lie tumbled in all directions. These circumstances may be explained on the reasonable assumption that, the line of sliding being necessarily more or less parallel to the face of the escarpment, the descent of the moving masses on the slippery surface of the underlying blue clays next the inlet would be uniform and gradual, whereas at the north-east end of the escarpment, where the supporting clays would be subject to undercutting by the sea, the overhanging sandstone would be liable to break away in large blocks whenever the unsupported weight of the mass became greater than the transverse strength of the material.

Mount Cronin, when viewed from the railway-line on the south side of Waikouaiti Inlet, is seen to be dome-shaped, and to surmount the surrounding country by several hundred feet.

General Geological Structure.

A general view of Mount Cronin shows that it is composed of a series of marine sandy beds now consolidated into soft sandstones. These sandstones rest conformably on marine blue clay, and are overlain unconformably by gravel and silt beds, which in their turn are crowned by a flow of basalt.

The basalt occupies the summit of the hill, and there can be little doubt that this commanding eminence owes its present existence to this hard cap, which has protected, the underlying loosely compacted gravels and silts from destruction by the ever-active agencies of subaerial denudation.

Omitting the sand-dunes, slope deposits, and other recent accumulations, we have thus in this interesting section three distinct formations—one marine, one fresh-water or terrestrial, and one volcanic. The peculiar association of these formations and their relations to each other throw a strong light on the geological history of Dunedin, and enables us to read aright sections heretofore involved in obscurity.

Classification of Rock-Formations.

To facilitate description the rock-formations occurring at Mount Cronin may be classified as follows:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Sand-dunes and talus deposit Recent.
Mount Cronin basalt Newer Pliocene.
Waikouaiti leaf beds Older Pliocene.
Waikouaiti sandstone Oamaru series.
Foraminiferous clays
– 420 –

Foraminiferous clays

These clays form the basement rock of this place. They are exposed at the north end of Waikouaiti Beach, where they occur as a flat gently shelving plane lying below high-water mark. A few chains around the rocky beach they rise in two places into the sea-cliff, a height of some 30 ft., forming small exposures in the face of the sandstone talus by which they are surmounted and surrounded.

In most places the clays are somewhat sandy, and where exposed to the weather are friable and crumbling and of a pale-blue colour. At all the points where I examined them they were found to contain a large assemblage of Foraminifera, among which many genera were represented, embracing forms possessing elaborate and beautiful shell-structures. Besides Foraminifera, I found some fragments of minute, obscure bivalves, but the only genus I was able to identify was a small Pecten.

At the first cliff-exposure after leaving the end of the sandy beach the clays contain irregular bands of dark-green glauconitic sandstone varying from 1 in. to 4 in. or 5 in. thick, except at the top of the outcrop, where there is a deposit of greensand the actual extent of which is obscured by the talus. The strike of the glauconitic veins exposed at the top of high-water mark is east-and-west, and the dip south at an angle of 62°. Small nests, streaks, and spangles of ironpyrites, some of which look like replacements of organic remains, are not infrequent in the clays and glauconitic sandstone.

The clays are so obscured by the talus that their relationship to the overlying sandstones cannot be ascertained at this place. Mr. McKay, in his report on eastern Otago, * mentioned the occurrence of calcareous sandstones at Cornish Head, but seemed to be unaware of the presence of the blue clays. In his map, however, he shows a strip of what he calls “marly clays” running from Waikouaiti Township northward through the saddle leading to Pleasant River. He does not appear to be clear as to the position of these clays, as, speaking of them (page 6), he says, “These are the probable equivalents of the Maheno marls that on the left bank of the Kakanui are seen resting on the Waireka tufas, and at the same place replace or pass under the Ototara limestone.” In White Bluff, he continues, “they are overlain by a volcanic breccia.” Again, on page 237 of the same report, Mr. McKay says, “These beds as they underlie the calcareous sandstones” [Ototara stone], “already described, are supposed to form part of the sequence affected by the fault along

[Footnote] * Reps. Geol. Expl., 1886–87, pp. 6–8

– 421 –

Woolshed Creek.” He further mentions, on the same page, that in the clays in the gullies south-east of Janet's Peak Pecten zittelli, Hutton. Leda, and Foraminifera are not rare.

We have already seen that in the flat ledges on the beach near Cornish Head the foraminiferous clays are associated with veins of glaueonitic material which strike east-and-west and dip south at high angles, whereas the overlying sandstones have only a slight inclination to the southward at angles rarely exceeding 12°.

If it be held that the glauconitic veins coincide with the original bedding-planes of the clays, then we must conclude that a stratigraphical unconformity of a very marked character separates the clays and overlying sandstones. But the evidence obtainable elsewhere does not support this view. At Pleasant River and near Palmerston the calcareous sand-atones (Ototara stone) are underlain conformably by green-sands and blue foraminiferous clays that seem to be the stratigraphical equivalents of the Waikouanti clays.

At Whare Flat, in Silverstream Valley, calcareous sandstones are underlaid conformably by greensands and blue sandy clays, the latter containing a large number of Foraminifera, among which I recognised several of the genera found in the clays at Waikouaiti Beach.

Among the Foraminifera from the blue clays at the end of Waikouaiti Beach I distinguished, with the kind assistance of Mr. A. Hamilton, the following genera: Nodosaria, four sp., Cristellaria, Frondicularia, Rotalina, Vaginulina, Dentdina.

These protozoans have such a wide range in time that they throw no light upon this question. The deep surface-clays between Waikouaiti and Palmerston make it impossible to trace a direct stratigraphical connection between the foraminiferous clays at these places, and consequently the relations existing between the blue clays and calcareous sandstones at Waikouaiti must still remain an open question. I am, however, of the opinion that these rocks belong to the same stratigraphical succession, and believe that subsequent investigation will confirm this view.

Waikouaiti Sandstones.

These consist of a series of soft yellowish - brown sandstones, which show a thickness of 360 ft. in vertical section in the great escarpment of Mount Cronin facing Waikouaiti Inlet. They are streaked at intervals with harder layers of a slightly more ferruginous or calcareous material, which weathers out in the face of the cliff in irregular honeycombed ledges often resembling aggregations of fucoid stems. In the upper horizons there are fossiliferous layers of a more cal-

– 422 –

careous character, which form low rocky ledges on the grassy slopes on the north side of the hill.

The whole series is of marine origin, and contains five distinct fossiliferous zones, which enable the beds to be divided into two well-marked groups. The lower group, corresponding to the Oamaru stone, comprises a thickness of some 220 ft It contains four fossiliferous zones, at the following elevations: At a height of 165 ft, Waldheimia; 230 ft., Meoma crawfordi, Pecten hochstetteri; 270ft, Waldheimia; 295ft, Waldheimia

In the lower zone Waldheimias (Magellania, Bayle) are scattered sparingly through a thickness of some 10 ft, in the third zone through 4ft., and in the upper through 2ft. Fossils are generally more abundant in the upper than in the lower zones.

The upper group of beds contains a number of forms which are common in the upper part of the Hutchison Quarry or Mount Brown beds of the Oamaru series. Among the mollusca collected were recognised Cardita, Venus, Limopsis, Nucula, Struthioldria; and, among the corals, Flabellum radians. Besides these, there were found the casts of many minute univalves and bivalves too indistinct to be identified.

Waikouaiti Leaf-Bed Series.

This series consists of a succession of silts, sands, and gravels about 45ft. thick resting unconformably on a slightly denuded surface of the marine sandstones. The sediments of this series, from their fluviatile or lacustrine origin, are subject to great variation in thickness, and this is especially noticeable in the case of the gravel and sand beds, which present the arrangement of current-bedding characteristic of deposits of this kind. On the west side of the cliff, under the trig, station, they consist of the following members in descending order:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Ft.
1. Sands and silts, with ferruginous concretions 7
2. Gravels 2–6
3. Gritty sands 8
4. Coarse gravels 7

The leaf-bed series, at the seaward end of the escarpment, consists of the following sequence, which naturally shows a close correspondence with that below the trig, station:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Ft.
1. Sands and grits 12
2. Grit-bed 3
3. Grey silts and sands 8
4. Coarse gravels 18
5. Leaf bed-fine grey silts 6
– 423 –

The leaf bed thins out toward the westward, and disappears altogether some yards before a point below the trig. is reached. The material composing this bed consists of excessively fine grey pumice sand or silt, in which occur numerous inclusions of grey pumice, occurring in rounded fragments, seldom exceeding 2 in. in diameter, but generally under 1 in. This is interesting as the first discovery of bedded pumice in any part of the South Island of New Zealand.

The leaf-impressions are numerous and not well preserved. They include what appear to be representatives of beech, myrtle, oak, elm, and Borne ferns. The leaf bed is the lowest member of the series, and is overlain by a bed of coarse gravel of variable thickness, which can be traced continuously from one end of the escarpment to the other. The gravels consist principally of pebbles and boulders of basanite and phonolite.

The basanite is blackish-green in colour, fine-grained in texture, and shows conspicuous crystals and grains of nepheline. When examined in polarised light it is found to be principally an aggregate of plagioclase, augite, olivine, and nepheline. The base is greyish-blue in colour, not abundant, and crowded with small laths and microlites of feldspar. The augite and olivine occur in large well-developed phenocrysts. The olivine is generally much altered and serpentinised. The nepheline occurs in large grains and in crystals often imperfectly developed. A little sanidine, biotite, apatite, and magnetite are present. In many places the rock is tinged red with iron-peroxide, evidently due to the oxidation of magnetite.

About 15 yards south of the trig. station the leaf-bed series and the underlying sandstones are traversed by a fault which hades to the south at an angle of about 45°. The vertical displacement of the beds caused by this fault is about 40ft.

Two minor faults with a displacement under 10 ft. can be traced indistinctly on the seaward side of the escarpment, near the outer rim of the basalt-flow.

Basalt-flow.

The summit of the hill is occupied by a flow of basalt which is probably a remnant of the flow which crowns the summits of Mount Royal, Mount Mackenzie, Mount Watkins, Derdan Hill, and other prominent elevations in the neighbourhood. These isolated patches are all that now remain of the basalt plateau that formerly extended from the Shag Valley southward to Waitati, and from the schistose mountains on the west eastward to the sea.

– 424 –

The basalt is fine in texture. Under the microscope it is seen to consist principally of plagioclase, augite, and olivine. The base is bluish-grey in colour, but not abundant. It is crowded with microlites of feldspar and augite, and thickly dusted with grains of magnetite. The feldspars occur as long narrow laths. The phenocrysts show orientation in one general direction, which doubtless indicates the original plane of fluxion. A little sanidme and apátite are present.

The existence of this outlier of basalt proves that the sculpturing of the present contours has taken place since the close of the Pliocene—that is, after volcanic activity had ceased in this part of Otago.

General Conclusions.

The Tertiary marine sandstones, leaf-bed series, and overlying flow of basalt are exposed in vertical section in a face of steep escarpment nearly 200 yards long and 400 ft. high. The face is bare or covered only with a scanty vegetation, thereby enabling the actual junction of the different formations to be examined at many points of contact.

The geological relations are so clear that the Waikouaiti section thus becomes of almost classic value when considered in connection with the sequence of events that constitute the geological history of the neighbourhood of Dunedin and the region round about Mount Cargill and Waitati.

Unconformity of Dunedin Volcanic Rocks to Oamaru Series.—Within the watershed of the Waitati, Silverstream, and Kaikorai Streams the great pile of volcanic ejecta which forms the hills around Dunedin is found resting upon different members of the Oamaru series, of which the Waikouaiti sandstone forms a part.

An examination of the geological structure of the neighbourhood of Dunedin shows that the Caversham sandstone had been elevated and in some places highly denuded before the commencement of volcanic activity in that region. The unconformity between the marine series and volcanic rocks is so well marked as to indicate a considerable lapse of time between the formation of the one and the eruption of the other.

Unconformable Relations of Leaf-bed Series to Oamaru Series.—No one seems to doubt that the Waikouaiti sandstone is the horizontal equivalent of the Caversham sand-atone. When, therefore, we find terrestrial beds composed of material derived from volcanic rocks that lie unconformably upon the Caversham sandstone near Dunedin resting upon equivalent marine beds at Waikouaiti, we can only conclude that even a greater unconformity in time exists between the terrestrial beds and marine series at Waikouaiti than between

– 425 –

the parent volcanic rooks and marine sandstones near. Dunedin.

Source of Basanite and Phonolite in Gravels.—The gravels underlying the basalt-flow at Mount Cronin are, as we have seen, principally composed of basanite and phonolite.

In his excellent paper on the lithology of Dunedin, Professor Ulrich pointed out that the great mass of the volcanic material in that region belonged to the phonolite group of rocks, in which nepheline forms an essential component; while some, he said, through accession of plagioclase, and either absence or presence of olivine, graduate respectively towards tephrite and basanite. * Since no basanite or phonolite is known in the neighbourhood of Waikouaiti, we may conclude that the Mount Cronin gravels were derived from the erosion of the basanites and phonolites described by Professor Ulrich at Mount Cargill and Purakanui.

Age of Volcanic Bocks around Dunedin.—I have just shown that the materials forming the gravel and sand beds at Mount Cronin are principally volcanic, and in all probability derived from the denudation of the volcanic area lying between Waitati and Dunedin. Before, therefore, discussing the age of the leaf beds it will be necessary, in the first place, to consider the probable age of the volcanic rocks in that area.

The volcanic rocks around Dunedin rest on a deeply eroded surface of the Caversham sandstone and underlying quartz sands and grits.

The Oamaru series, to which these rocks belong, is considered by Captain Hutton to be of Oligocene age; but the proportion of fossils belonging to living species in that formation would indicate, from European standards, rather a Miocene age. If, however, we accept the Oligocene age of Captain Hutton, it is evident that the earliest volcanic outbursts could not have commenced until some time in the Miocene period, for we find that the Caversham sandstone, one of the higher members of the Oamaru series, was deposited, consolidated, elevated, and partly denuded before the first outbursts began.

There is no evidence of contemporary volcanic activity during the deposition of the Oamaru series anywhere south of Palmerston. There is, we know, abundant evidence that volcanic outbursts took place during the deposition of the middle members of the Oamaru series in the Oamaru basin. But these outbursts were purely local, and there is nothing to show that they reached north of the Waitaki or south of the Shag River. At any rate, south of Palmerston the

[Footnote] * Proo. Aust. Assoc. Adv. Science, vol. Iii., 1891, p. 145.

– 426 –

marine Tertiary series is quite free from volcanic material. It cannot therefore be said that the eruptions around Duqedin were a recrudescence of Oligocene activity.

It is pretty clear that volcanic activity commenced in the Dunedin area after the Caversham sandstone bad been subject to subaerial erosion, and for that reason I am disposed to place the first outbursts in the later period of the Miocene, or in the beginning of the Pliocene.

Age of Waikouaiti Leaf Beds.—I have already shown that the materials composing the leaf-bed series at Mount Cronin were in all probability derived from the south, principally from the watershed formed by Mount Cargill range. If the pile of volcanic ejectamenta forming that watershed and neighbourhood was the result of Miocene eruptions, and suffered subaerial erosion sufficiently prolonged to yield the materials to form fluviatile or lacustrine beds at a place over twenty miles distant, it is evident that the said beds cannot be older than Pliocene. This conclusion is drawn from the fact that the leaf beds are composed of materials derived and transported by fluviatile action from a region where the parent rocks rest on a highly denuded surface of a marine formation, of Oligocene or Lower Miocene age.

So far the fossil flora of these beds has not been submitted to a palæobotanist; but, in any case, such determinations, except they are supported by other data, must always be regarded with suspicion.

For the present, therefore, until better evidence is forthcoming, I am inclined to place the leaf beds in the older Pliocene—that is, contemporary with the Wanganui older marine series, and about synchronous with the lacustrine beds of central Otago.

Bearing of Waikouaiti Section on Geological History of Dunedin.—From this section we learn that the piles of volcanic rocks around Dunedin were the result of at least two distinct periods of volcanic activity. And the presence of pumice, phonolite, and basanite in the leaf-bed series would tend to indicate that the earlier eruptions were of an acidic or semi-basic type.

The materials composing the gravels of the leaf-bed series were, as we have seen, apparently derived from the erosion of the ejecta of the earlier eruptions, and we can only explain their presence by assuming that these earlier outbursts were succeeded by a period of volcanic quiescence during which the newly ejected rocks were subjected to subaerial denudation.

The basic eruptions of the second period of activity smothered the newly formed gravels in the protecting cap of basalt which now crowns Mount Cronin and the higher hills in the neighbourhood.

– 427 –

It is not assumed that the ejecta of each era was the result of one stupendous outburst On the contrary, the alternation of fragmental matter and solid lava so often seen in the district around Dunedin seems to justify the conclusion that in each era there were alternating periods of quiescence and more or less paroxysmal activity, each distinguished by some peculiarity in the composition of its lavas and fragmental material.

The conclusion that a cessation of volcanic activity occurred after the earlier outbursts does not depend alone upon the evidence furnished by the section at Waikouaiti North Head. At several places within the Dunedin area there is seen a series of aqueous deposits both underlain and overlain by volcanic rocks.

This sedimentary series consists of well-rounded gravels, sands, and silts containing in places leaf-impressions, and in others such an excess of carbonaceous matter as to pass into impure coal and oil-shale.

At the oil-shale outcrops between Burns's and Williams's Streams, at the source of the Waitati, the succession of rocks in descending order is as follows:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Ft.
Probably lacustrine 1. Basic lava-flow of great thickness.
2. Oil-shale graduating into carbonaceous sandstone 18
3. Volcanic tuff partially stratified 12
4. Impure brown coal 3
5. Volcanic tuff 12–15
Marine 6. Caversham sandstone.

The tuff underlying the oil-shale contains numerous fragments of grey vesicular material resembling the pumice in the leaf bed at Mount Cronin.

In the Kaikorai Valley the Caversham sandstone is overlain by a series of beds of lacustrine origin, consisting of fine grey silts, sands, clays, and gravels composed principally of volcanic material. The lacustrine beds are overlain by a great pile of volcanic rocks.

The grey silts contain a large number of finely preserved leaf-impressions, and in a collection of these made in 1901 Dr. Marshall identified species of oak, elm, birch or beech, Magnolia, Piper, and Metrosideros.*

The Kaikorai section has a striking resemblance to that at Waikouaiti North Head, and there seems to be no reason to doubt that the leaf-bed series at Mount Cronin-is the horizontal equivalent of the Kaikorai leaf beds.

In all that has been written about the geology of Dunedin it is singular that the significance of the section exposed at Te

[Footnote] * Trans. N. Z. Inst., vol. xxiv., 1901, p. 586.

– 428 –

Manu Hill, locally known as Tanna Hill, situated near the University Mining School, has been entirely overlooked. Here we have a succession of rocks almost identical with that at Waikouaiti North Head and at Kaikorai Valley.

A few yards below Union Street Bridge, in the steep bank of the Leith, is seen a bed of grey silts containing a few imperfect leaf-impressions, numerous broken plant-remains, and a few larger pieces of carbonised driftwood. The silts are not continuous for more than 25 yards, and, so far as they are exposed, they thin out in both directions against a coarse tuff-like sandstone. They exhibit characteristic current-bedding. Overlying the plant bed there is a deposit of gravels and sands composed principally of volcanic material. The gravels in their turn are capped by the flow of rudely columnar basalt which forms this well-known eminence. A deposit of volcanic-ash rock rests on the undulating surface of the basalt. In the bed of the stream, under the bridge, the plant silts are seen resting on coarse volcanic breccia or agglomerate.

The succession in this important and interesting section is therefore as follows:—

1.

Surface clays and soil.

2.

Volcanic ash from 8 ft. to 15 ft. thick.

3.

Basalt-flow, about 25 ft.

4.

Gravels, about 18 ft.

5.

Grey plant silts, 5 ft. to 7ft thick, resting at ends against tuff-like sandstone of variable thickness.

6.

Coarse volcanic breccia.

At Waikouaiti, Kaikorai, and Waitati the plant-bed series rests directly on the Caversham sandstone, and at Te Manu Hill on fragmentary volcanic ejecta. The lesson to be derived from this is that Dunedin was situated within the theatre of volcanic activity during the period of the earlier outbursts.

Te Manu Hill is a place of great geological value, and I trust that the exigencies of the University will not require its removal, but that it will be reserved for all time as a geological witness of the past history of Dunedin. Its destruction would be an irreparable loss to geologic science, and cause the obliteration of a geological record that could never be replaced.* Apart also from its geologic and scenic value, Te

[Footnote] * Mr. A. Hamilton informed me, after this paper was prepared, that no interest had previously been attached to Te Manu Hill by geologists because it was believed that the gravel and silt beds were of comparatively recent date, and abutted against the basalt instead of underlying it. In order to put this question beyond all doubt I caused a pit to be sunk in the floor of the old quarry facing the Mining School, and trenches to be dug at the foot of the cliff on the Leith Street side of the hill. In all places the gravels were found below the basalt-flow, as described above.—J. P.

– 429 –

Manu Hill is an ideal site for the University astronomical observatory.

Duration of Volcanic Cessation.—The means to determine this are only comparative. The leaf beds at Waikouaiti and Kaikorai and the shale-beds at Waitati rest directly on the Caversham sandstone, from which we gather that the earlier eruptions did not reach these areas.

On the other hand, the fine silts, muds, and ash at these places clearly indicate that showers of finer ejecta at times fell in the lake-basin in which these deposits were forming, and doubtless also on the surrounding country. The gravel-beds which close the leaf-bed series, moreover, prove that fluviatile forces were active agents of erosion immediately prior to the beginning of the second and final era of volcanic activity, which culminated in the emission of the lavas overlying the gravels.

The plant-remains in the muds and silts inform us that a varied forest vegetation grew on the shores of the Pliocene lake; while the oil-shale at Waitati may be, held to indicate the existence of a rich and long-continued growth of freshwater gelatinous algæ or related forms.

At Te Manu Hill, near the Mining School, the silt and gravel series rests on a surface of coarse volcanic breccia, a circumstance which clearly indicates that the eastern shores-of the Pliocene lake reached to the area affected by the earlier volcanic eruptions.

The gravels and tuffs which close the leaf-bed series seem to have suffered little or no erosion prior to the renewal of volcanic activity. They were apparently overwhelmed suddenly by stupendous outbursts, more violent and widespread than those of the earlier period of eruption. We may therefore conclude that the period of cessation was approximately equivalent to the time required for the deposition of the intercalate leaf-bed series.

After the period of cessation volcanic activity commenced anew, culminating in the more basic outbursts in newer Pliocene times.

Direction of Drainage in Pliocene Times.—A study of the section at Mount Cronin affords some interesting information respecting the conformation of the land daring the interval of volcanic cessation.

Professor Ulrich showed that the volcanic rooks around Dunedin graduated from trachytoid phonolites to basanite.* The preponderance of phonolite and basanite in the Mount Cronin gravels is therefore highly instructive. It shows, in

[Footnote] * Trans. Aust. Assoc. Adv. Science, vol. iii., 1891, p. 145.

– 430 –

the first place, the southern origin of the leaf-bed material; and, as this material is obviously water-borne, it establishes the former existence of a drainage system running from south to north—that is, from Dunedin towards Waikouaiti, in a direction more or less parallel with the general trend of the present coast-line.

Again, the physical conditions and distribution of the materials in the different members of the leaf-bed series tend to show that this drainage system comprised a river with one or more shallow lakes along its course.

This ancient watershed required for its outer or eastern rim the existence of a land-area to seaward of the present coast-line, and it just suggests itself that this rim was probably formed of volcanic ejecta piled up during the earlier period of eruption. At any rate, there is evidence that the earlier eruptions did not extend so far south, west, or north as the later outbursts, but were centred in the neighbourhood of Port Chalmers, Purakanui, and Mount Cargill.

The geological history of volcanic activity around Dunedin in Tertiary times bears a peculiar likeness to that of the Hauraki goldfields of Auckland. After a study of the latter extending over seven years, I wrote in 1897 as follows: “After the pent-up forces had spent themselves in the first great paroxysm there was a period of rest, during which vegetation established itself on the muds and ashes washed into the low ground by the streams draining the slopes of the newly formed volcanoes. The land, however, was in a continued state of tremor, and the oscillations were too frequent to permit the continued growth and accumulation of sufficient vegetation to form workable seams of coal. The cessation of volcanic activity was of short duration. The plutonic forces burst out with renewed energy. The forests were devastated and utterly destroyed, and covered with hundreds of feet of ashes and solid lava of semi-basic character.”*

Description Of Plate XXXII.
Section, Waikouaiti North Head.
1.

Talus deposit.

2.

Basalt-flow.

3.

Gravels and Bands.

4.

Gray sandstone.

5.

Calcareous sandstone.

6.

Blue clays, foraminiferous.

Section of Summit of Mount Cronin.
2.

Basalt-flow.

3a.

a. Grey sands and silts.

3b.

b. Coarse grits.

3c.

c. Grey silts and sands.

3d.

d. Coarse gravels.

3f.

Sandy grits.

4.

Grey sandstone.

[Footnote] * Park, “The Geology of the Hauraki Goldfields,” 1897, p. 43.