Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 64, 1935
This text is also available in PDF
(4 MB) Opens in new window
– 275 –

The Geology of the Glenomaru Survey District, Otago, New Zealand.

[Read before the Otago Institute, November 14th, 1933; received by the Editor, May 30th, 1934; issued separately March, 1935.]


  • Introduction

  • Previous Work.

  • Summary of General Geology.

  • Summary of Stratigraphy.

  • Detailed Stratigraphy.

  • A. The Coastal Succession.

  • (i) The Middle Triassic System.

  • (ii) The Ladino-Carnic, Carnic, and Rhaetic Series.

  • (iii) The Jurassic System.

  • B. Notes on the Inland Occurrences.

  • Petrology.

  • A. Detailed Description of Pebbles from Conglomerates.

  • B. General Remarks on the Nature of the Pebbles.

  • Geomorphogeny.

  • General.

  • Coastal Features.

  • Acknowledgment.

  • Bibliography.


The district described comprises an area of some 40 square miles, lying on the east coast of the South Island of New Zealand, about 80 miles south of Dunedin, Otago. The township of Owaka lies in the south-western portion of the area, in a fairly extensive and low-lying basin through which meanders the Owaka River. Until comparatively recent years the whole district, except the lower-lying parts, was covered with dense beech and pine forests, and although most of these have now been removed by the local sawmillers, the slopes of the higher ridges and some of the more isolated areas still remain in their natural wooded state.

The field work was carried out in the earlier part of 1932, and comprised both the geological mapping and the contouring of the whole area. Some good sections were obtained both in the railway cuttings and on the coast, the one in Roaring (Shaw) Bay being especially noteworthy, as it is the classical exposure of the Triassic sediments in New Zealand.

The petrographic section of this paper deals almost exclusively with the pebbles from the Triassic and Jurassic conglomerates, but the writer hopes to submit for publication shortly an account, with several analyses, of the greywackes and slates occurring in the area.

– 276 –

Previous Work.

W. L. Lindsay, as early as 1862 (p 30), examined the rocks immediately south-west of Nugget Point, and recognised their Triassic age, while Alexander McKay, in 1873, also collected fossils at this and other localities throughout the area. Two years later F. W. Hutton placed the beds at Roaring Bay in the Maitai formation, which he considered to be of Triassic age; the rocks between Roaring Bay and Cannibal Bay he included in the Putataka formation, to which he assigned a Lower Jurassic age. In 1903 Professor Park and Mr A. Hamilton visited the coastal section, and the former investigator in the following year published a detailed section of the Roaring Bay sequence, and a generalised section of the coast from Kaka Point (5 miles north of Nugget Point) to the mouth of the Catlins River (5 miles south-west of Nugget Point). He also listed and described the fossils which he collected during this visit.

Professor R. Speight, in 1904, described the dyke which cuts across the neck of land between Nugget Point promontory and the mainland.

The most recent work carried out in this district was that of Dr C. T. Trechmann, who collected and gave detailed descriptions of fossils from Roaring Bay (Trechmann, 1917). As a result of his palaeontological work he was able to subdivide the Triassic sequence, and to indicate that the Noric beds were missing at this locality, though he remarked that Dr Marshall had found the fossil characteristic of this series (Pseudomonotis) at Glenomaru, 7.½ miles from the coast. Dr Otto Wilckens later (1927) amplified Trechmann's work to some extent by detailed descriptions of the fossil collections from Roaring Bay made by Park and Hamilton in 1903 for the Geological Survey. In 1923 Dr Trechmann also described a few Jurassic fossils collected by Dr Marshall from the beds between Roaring and Cannibal Bays. Some fossil plants from near Owaka, which had been obtained partly by Hector in 1865 and partly by A. McKay in 1873, were described by Dr E. A. N. Arber (1917), who, with hesitation, assigned to them a Rhaetic (?) or Lower Jurassic (?) age.

Marshall (1912) and Cotton (1922) both comment upon the curious pattern of the drainage system of the area.

It will be seen, therefore, that the greater part of the work carried out in the Glenomaru Survey District has been confined to the more exposed coastal regions. The discovery of the plant fossils, and of the Pseudomonotis beds at Glenomaru is all that is available from the inland area.

Summary of General Geology.

The majority of the rocks occurring throughout the area are of Triassic and Jurassic age, although locally these are covered by a superficial veneer of Recent dune sands and alluvium. In the north-eastern half are steeply dipping Triassic and Jurassic members, and in the south-western, gently flexed Jurassic strata.

– 277 –

A survey of the coastline, which in the main cuts directly across the general direction of strike, reveals beds of age decreasing south-wards from Campbell's Point. The Triassic system comes to an end near the south head of Roaring Bay, and all the contorted strata south-west of this point are Jurassic. The Recent deposits occur in the form of patches of alluvium and sand which make up river and coastal flats.

The Mesozoic rocks consist, for the most part, of a thick series of greywackes, felspathic sandstones, claystones, conglomerates, etc., whose hardness and compactness indicate that they have suffered considerable pressure. These rocks were deposited probably under deltaic, estuarine, or near-shore conditions. During the great Hokonui (Closing Jurassic-Lower Cretaceous) Orogeny they were strongly folded and were later reduced by erosion to a peneplain state before the close of Cretaceous times. It is probable that, in common with the rest of the East-coastal region of the South Island of New Zealand, the present district was covered by marine sediments during the late Cretaceous and Tertiary Eras. All traces of these younger beds have, however, been removed by subsequent erosion from the area here described, though they remain to the north (Wangaloa), north-west (Chatton), and west (Southland).

The present topography doubtless evolved subsequently to the uplift experienced in Pliocene and early Pleistocene times during the Kaikoura Orogeny of Cotton (1916, p. 248). Still later warping probably accounts for the Owaka “basin” or “depression,” and the formation of the swamps connected therewith.

The most recent movement was a minor uplift, evidence of which exists in the form of raised beaches, much as in other parts of the East Coast district of Otago.

Summary of Stratigraphy.

The oldest rocks exposed in the area are in the north-eastern portion. Although fossils occur in them, they are so poorly preserved and fragmentary that they give no reliable indication of age, but apparently they belong to the Middle Triassic. The rocks of Nugget Point itself are placed by Dr Trechmann (1917) in the Middle Triassic (?), so that in the absence of faulting, those to the northeast must therefore be older, and may thus possibly extend into the Lower Triassic. They consist for the most part of hard, fairly coarse-grained greywackes, with interbedded finer-grained fossiliferous claystones and a conglomerate.

The beds recognised by Trechmann (1917, p. 180) as Ladino-Carnic, comprising greywackes, sandstones, and claystones, are the oldest beds whose age has been definitely recognised, and they are observed to best advantage on the coast in the extreme north of Roaring Bay, and in the Nugget Point Peninsula. They are followed by some 2000 feet of greywackes, conglomerates, sandstones, and claystones, whose fairly abundant fossil fauna establishes their age as Carnic. These are most clearly exposed in the cliffs about Roaring Bay, and are seen at various points further inland, where they are recognisable particularly by their massive conglomerates.

– 278 –

The marine Noric beds, which appear to thin out and disappear towards the coast, are represented near Glenomaru Railway Station by gritty, pebbly sandstones and interbedded mudstones, both of which are abundantly fossiliferous in places. It is probable that they are represented on the coast by a disconformity.

The Triassic deposits are brought to a close by about 200 feet of fossiliferous, pebbly sandstones and indurated claystones, the latter bearing indeterminate plant remains. With no sign of unconformity, these Rhaetic beds pass up into the Jurassic sequence which continues in an unbroken succession of greywackes, sandstones, mudstones, and conglomerates—all with an extremely high dip—for about 14,000 feet. Fossils, which are fairly plentiful at several stages, allow no doubt of their Jurassic age.

There is then a great gap in the succession, for the next deposits are of Recent age.

It is interesting to note that at no stage in this extensive sequence of strata is there any sign of contemporaneous igneous activity. The only outcrop of massive igneous rock in the whole area is that of the dyke of porphyry at Nugget Point, which is intrusive into the Triassic rocks. The conglomerates, however, include numerous igneous pebbles—mostly of intrusive rock types, but also including volcanic forms—indicating the existence in Triassic and Jurassic times of a batholithic mass uncovered by erosion.

Detailed Stratigraphy.

Although outcrops are fairly plentiful throughout the area, the general stratigraphical succession is revealed to best advantage on the coast. Therefore, in the following pages the coastal section will be reviewed in detail, and supplemented by information gained inland.

A.—The Coastal Succession.

1. The Middle Triassic System.—The lowest member of this series forms conspicuous outcrops in an anticline exposed on the coast at Campbell's Point, extending seaward as an expanse of low-lying, jagged reefs. Their strike is well-marked, but variable, and it can be seen to veer through twenty or thirty degrees within a distance as small as twenty yards. At this point the rocks are banded, consisting of alternating layers of coarse and fine-grained greywackegrits. The coarser rock often contains tiny pebbles of greywacke, claystone, and a reddish, jasperoid material. Sea-erosion has removed some of the less-resistant rock in many places so that the harder layers only now remain.

These pass up into a softer, fine-grained claystone. Outcrops of this are sparse, and the beach is more or less clear of reefs for some 600ft. (indicating a thickness of 560ft.) until we come upon another outcrop of hard, coarse-grained greywacke-grit, 140ft. thick.

The next member of the series is a comparatively soft, fine-grained, dark-grey claystone, which is, in general, traversed by numerous small joints. Outcrops occur at various points on the

– 279 –

beach, followed southward to within about 5 chains of Hay's Gap. Occasionally there are traces of fossils in this rock and among these a form like Daonella seems to be predominant. Other forms, if present, are indeterminate. At the point where Nugget Creek crosses the beach these fossiliferous claystones are interrupted by a resistant conglomerate 16ft. wide, which weathers to a rusty-red colour and forms a well-marked outcrop. The largest pebble observed was about 11 inches in diameter. The cementing medium is exceedingly hard; the pebbles include both igneous and sedimentary rocks, the latter being mainly greywacke, claystone, or silicified mudstone.

Five chains north of Hay's Gap are further hard and relatively coarse-grained greywacke-grits, which, followed through the Gap itself, become laminated, and are interbedded with thin bands of softer rock. The bands are often cemented with ferruginous material which stands out in thin layers from the rest of the rock as was noted by Park (1904, p. 380).

Following on these immediately to the south of Hay's Gap promontory there is a thickness of about 200ft. of jointed dark-grey fossiliferous claystone containing a Daonella (?), and similar to that mentioned above. This gives way, higher in the sequence, to more coarse-grained indurated sandstones which are often interbedded with softer rock. Fourteen chains south of Hay's Gap is the centre of a syncline occurring in the harder beds, followed by an anticline five chains further south. From this point onwards the Middle Triassic Beds dip to the south-west at angles varying from 45 to 90 degrees, with some variation in strike, which is here nearly parallel to the coast. Professor Park (1904, p. 380) comments concerning this part of the sequence: “Towards the Nuggets the sandstones become coarser in texture, and in places assume the character of greywacke. They form high, rocky points and numerous isolated flat reefs on the beach below high-water mark, separated by stretches of sand.” Seventeen chains north-west from the Fishing Station is a layer of dark-grey claystone, about 7–8 feet thick, in which was found a Spiriferina.

The highest member of the Middle Triassic is a greyish claystone containing abundant plant remains, none of which, unfortunately, are determinable. This bed, which is about 200ft. thick, forms the core of the Nugget Point peninsula, and the lighthouse itself stands upon it.

At the point where the track leading to the lighthouse leaves the beach near Boatlanding Bay, there is a thin bed of fossiliferous claystone. The only definite form recognised here by the writer was a badly preserved Spiriferina. Professor Park, however, notes the following from this locality: Three species of Spiriferina, Epithyris, Rhynchonella, Pleurotomaria, and fragments of a bivalve shell resembling Modiolopsis (1904, p. 381). Indefinite fossils also occur in the sandstones underlying this bed.

The most noteworthy feature of the Middle Triassic beds is their marked variation in dip and strike. An attempt was made to trace the Campbell's Bay conglomerate inland, but after about a

– 280 –

quarter of a mile it was obscured by the dense bush. The section along Nugget Creek road is a disappointing one, for there are only a few poor outcrops of weathered sandstone, making any correlation with the coastal beds impossible. From the variation in the strike and dip it appears certain that some wedging of the strata occurs.

The total estimated thickness of Middle Trias deposits in this locality is 5200ft.

2. The Ladino-Carnic, Carnic, and Rhaetic Series.—These three series are best considered under one heading, since they all outcrop in a compact sequence making up Roaring Bay. The section across the bay was described in 1904 by Professor Park, and the present writer can do no more than quote his sequence and measurements (also confirmed by Dr. C. T. Trechmann in 1917).

Following conformably upon the Middle Triassic plant-bearing claystones we have, proceeding southwards*:—

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

17–18 a. Indurated sandstone and greywacke, 150ft.
b. Spiriferina beds—dark-blue claystones, 100ft.
16 c. Porphyry dyke.
d. Greywacke.
15–14 e. Fossiliferous, crumbling claystones with Halobia zitteli. About 210ft. thick. Seen in the road cutting descending to the bay.
f. Sandstones and claystones, 250ft., not exposed on beach.
13 g. Breccia-conglomerate, angle of dip 77deg., 20ft.
12 h. Highly indurated claystones, 55ft.
11 i. Hard greywacke, 15ft.
10 j. Thinly laminated claystones and sandstones, 24ft. k. Breccia-conglomerate, slaty and granitic, 27ft.
9 Breccia-conglomerate, stay and granite, 27ft.
l. Gritty sandstone, 15ft.
8 m. Indurated sandstones with occasional beds of claystone, 200ft., nearly vertical.
7 n. Granitic conglomerate, 10 inches thick; angle of dip 86 deg.
o. Indurated claystone, 50ft.
6 p. Myophoria bed—claystones, 10ft. Contains an ammonoid shell, Myophoria nuggetensis, Pleurotomaria, and other Carnic fossils.
5 q. Sandstones and claystones with subordinate bands of gritty sandstone, 450ft.
4 r. Mytilus problematicus bed, 29ft.; angle of dip 78 deg.
3 s. Coarse sandstones with occasional gravel layers, 600ft.
2 t. Clavigera beds, 10ft.—sandstone containing Clavigera bisulcata and Spiriferina diomedea.
1 u. Plant beds—sandstones and claystones.

[Footnote] * The numbering corresponds as far as possible with that on Trechmann's section; the lettering with Park's Divisions. For diagrammatic section see Trechmann, 1917. p. 180.

– 281 –

There are three layers containing Clavigera bisulcata, separated by beds of barren sandstone, the sequence in ascending order being:—

  • Dark-blue indurated claystones—at the base.

  • Bed of granitic conglomerate 4in thick.

  • Clavigera bed, 40in., very fossiliferous.

  • Sandstone, 4ft.

  • Clavigera bed, 2ft.

  • Sandstone, 4ft.

  • Clavigera bed, 18in.

  • Indurated sandstones and claystones.

The upper Clavigera bed is a coarse, gritty, often pebbly, sandstone containing such an abundance of Clavigera and Spiriferina valves that the rock is in places moderately calcareous. At the south head of Roaring Bay, the strata are nearly vertical, and the lower Clavigera bed stands up like a wall, presenting an even surface 70ft. high and 100ft. long, thickly encrusted with Clavigera and Spiriferina shells which have weathered out of the matrix.

In the above list of beds a. and b. (18–17) have been referred to the Ladino-Carnic age. The beds d. to s. (15–3) are considered to be Carnic, while the remainder are classed as Rhaetic. Nowhere is there any indication in this succession of the Noric beds characterised by Pseudomonotis, which are well developed at Glenomaru. It is therefore to be inferred that a non-sequence or disconformity exists between beds s. and t. (3 and 2).

The lists of fossils determining the ages of the series are given below:—

Ladino-Carnic (Beds a. and b. or 18–17).

  • Lamellibranchiata:—

  • * Daonella indica Bittner.

  • Brachiopoda:—

  • *Halorella zealandica Trechm.

  • *Dielasma cf. himalayana Bittner.?

  • Dielasma cf. zealandica Trechm.

  • *Spiriferina fragilis Schlotheim.

  • *Mentzeliopsis spinosa Trechm.?

Carnic (Beds d. to s. or 15–3).

  • Cephalopoda:—

  • *Grypoceras cf. mesodiscum Hauer.

  • *Proclydonautilus mandevillei (Marshall).

  • *Discophyllites cf. obneri Mojsisovics.

  • Gasteropoda:—

  • *Pleurotomaria (Sisenna) hectori Trechm.

  • Trochus (Tectus) marshalli Trechm.

  • *Bourguetia (?) arata Trechm.

[Footnote] * Species noted by Trechmann.

[Footnote] † Additional species noted by Wilckens.

– 282 –
  • Lamellibranchiata:—

  • * Halobia zitteli Lindstrom var. zealandica.

  • *Myalina (?) (Maoria) problematica (Zitt.) (“Mytilus problematicus”).

  • *Hokonuia rotundata Trechm.

  • *Pecten sp.

  • Nuggetia morganiana Wilck.

  • *Anodontophora edmondiiformis Trechm.

  • *Myophoria nuggetensis Trechm.

  • *Palaeocardita quadrata Trechm.

  • Gonodon mellingi Hauer.

  • Brachiopoda:—

  • Spirigera cf. wreyi.

  • Rhynchonella (Halorella) cf. griesbachi Bittner.

Only a few of these species could be obtained from the rocks at present exposed in Roaring Bay. According to the writer's collecting, they were zonally distributed throughout the Carnic as follows:—

Beds:— d e p r
Pleurotomaria (Sisenna) hectori X
Halobia zitteli X
Myalina? problematica X
Myophoria nuggetensis X
Gonodon mellingi X
Rhynchonella cf. griesbachi X

Rhaetic (Beds t. and u. or 2–1).

  • Brachiopoda:—

  • Clavigera (Hectoria) bisulcata Trechm.

  • Spiriferina diomedea Trechm.

  • Spiriferina sp.

The grounds on which Trechmann assigns a Rhaetic age to these beds are the similarity of Clavigera bisulcata to Spirigera oxycolpos of the Alpine Trias., but more especially the fact that elsewhere in New Zealand, viz., near Kawhia, in the North Island, C. bisulcata and S. diomedea are associated with a form compared with the Alpine Rhaetic fossil Arcestes rhaeticus Clark and lie upon the Noric beds with Pseudomonotis.

In addition to the above lists Wilckens notes the presence of:—

  • Rhynchonella nuggetensis

  • Rhynchonella maorica

  • Spiriferina parki

in the Halobia and Spiriferina beds (AA) of Park, which includes both the Ladino-Carnic and Lower Carnic.

The total estimated thickness of Ladino-Carnic, Carnic, and Rhaetic is 3200ft.

[Footnote] * Species noted by Trechmann.

[Footnote] † Additional species noted by Wilckens.

– 283 –

3. The Jurassic System.—The Jurassic system is assumed to commence at the southern side of the south head of Roaring Bay in a vertical, coarse, felspathic sandstone (J1) with pebbles of both igneous and sedimentary rocks sometimes up to nearly 2in, diameter. As it increases in thickness, this sandstone becomes less pebbly and more finely grained, frequently weathering to a rust-red colour, due to a ferruginous cement. The total thickness is about 250–260ft.

The next 1050ft. of strata (J2) consist of claystones and slaty mudstones interbedded near the base with fine-grained ferruginous sandstone, and near the top with layers of hard greywacke. Five chains north-east of the mouth of the creek flowing into the northern side of Sandy Bay is a hard, indurated sandstone (J3) 460ft. thick, which at the creek-mouth itself becomes concretionary in places, and also shows a well-developed series of joints. The concretions measure up to 6in. or 7in. in diameter.

Above this comes an extensive thickness—some 2000ft.—of claystones and indurated mudstones (J4). About 330ft. from their base there is an extremely fossiliferous band extending, with inter-bedded barren claystones, through 30ft., and consisting in its richest parts, almost entirely of the compressed and distorted valves of the lamellibranch Pseudaucella marshalli (first obtained here by Dr P. Marshall). Solution of the carbonate of the shells has caused both the bed and the neighbouring rock to be traversed by numerous calcite veins. One chain south of the mouth of the main Sandy Bay Creek the claystone becomes laminated and interbedded with occasional layers of hard, crystalline, fine-grained greywacke. Bands of sandstone with indeterminate plant (?) remains come in higher in the sequence, while the uppermost 260ft. has numerous greywacke layers. Slight bending of the strata is also in evidence in this upper part.

Concerning the age of this Pseudaucella claystone (J4), it may be noted that the fossil originally doubtfully described by Trechmann (1923, p. 269) as Aucella marshalli, was also found by Marshall near Kawhia harbour (North Island, New Zealand) in a similarly crowded shell-band 1000–1500ft. above the top of the Triassic Sequence as recognised in that district, a view since confirmed by the Geological Survey. Dr. Marwick (1924, p. 305) created for the species a new genus Pseudaucella, and considered it marked a zone of middle Liassic age (Henderson and Grange, 1926, p. 37).

The next member (J5) is a hard, relatively coarse, indurated sandstone, concretionary in places, which forms the south side of Sandy Bay. One hundred and ten feet from its base is a thin layer of conglomerate containing a variety of pebbles both of igneous and sedimentary rocks, ranging in size up to 8in. long by 2.½in. wide. The pebbles of the latter class are usually fine-grained—argillite, greywacke, indurated mudstone, etc. The greatest width of the conglomerate is about 11in., but it is very variable, thinning out altogether at points within a few yards of one another.

The first little bay to the south of Sandy Bay (Bay A) is excavated in a rather soft, banded, and more or less concretionary

– 284 –

sandstone with occasional pebbles (J6). On the south side of this bay hard sandstone (J7) is again encountered which shows well-developed honeycomb-weathering. Between Bay A and Boatlanding Bay the sandstone is fine-grained and banded, being composed of alternating thin layers differing slightly in colour and grain size.

Extending across Boatlanding Bay (330ft.) is an indurated and jointed, banded claystone (J8), in which the sea, on the south side of the bay, has worn flat pavements. There are traces of fossils here, but nothing definite was recognised. McKay (1873) collected fossils at this locality, but unfortunately he did not give their names. Near the southern shore of Boatlanding Bay the claystone becomes sandy and concretionary, finally passing up into indurated banded felspathic sandstone (J9) which shows honeycomb weathering. This sandstone forms the headland between Boatlanding Bay and Bay B, but on the northern side of the latter it passes into a conglomerate which is possibly 6ft. thick, and which contains igneous and sedimentary pebbles up to 2in. diameter. The rock within Bay B itself is an indurated greenish sandy mudstone (J10) 120ft. thick. It is interbedded with hard, sometimes pebbly sandstone, and is succeeded by 60–70ft. of pebbly felspathic sandstone (J11) forming the small promontory between Bay B and Tuck's Bay.

On the northern side of Tuck's Bay a greenish-grey, slightly indurated mudstone of small thickness comes in, followed by a darkgrey friable claystone (J12) which continues to the south side of the bay, where it gives way to a hard, jointed sandstone (J13) interbedded with crystalline greywacke. These rocks, with occasional pebby and felspathic layers, continue to the northern shore of Cannibal Bay, where there is a fossiliferous claystone (J14) containing Trigonia, Ostrea, Anomia, Inoceramus, Pholadomya, Venus (Park 1904, p. 386), and Ammonites and Belemnites (McKay, 1873, p. 72). In spite of careful searching the writer was unable to obtain all the above forms, but succeeded in finding a number of slightly distorted shells resembling very closely the Aucella spitiensis cf. var. extensa Holdhaus described and figured by Trechmann (1923, p. 267) from the Upper Jurassic of Waikato, South Heads, North Island, New Zealand.

The claystones at the north shore of Cannibal Bay are only about 150ft. thick (dip 87° S.), and they are the last or uppermost of the continuous coastal sequence. The structure and lithology of the remaining beds (Jurassic) are well revealed by inland outcrops, and at one other point on the coast, viz., False Islet.

The above-mentioned claystones (with Aucella cf. spitiensis) are replaced within a few chains by a greenish, banded mudstone which continues to the southern margin of Cannibal Bay. The bed is inter-stratified near the middle with harder, impersistent layers of pebbly felspathic grit, causing it to stand out as a low ridge throughout its whole extent. (N.B.—These Cannibal Bay mudstones and claystones, together with the grit, are included under J14).

At the south side of Cannibal Bay, on False Islet, an indurated sandstone (J15) forms a steep slope just over the summit of which are bands of felspathic sandstone containing many pebbles of quartz-

– 285 –

ite, sandstone, jasperoid slate, silicified mudstone, and various igneous rocks. The dip at this point is 85° south. The pebbles are present through a thickness of about 145ft., and in some places are in such abundance as to warrant the term conglomerate. About the centre of the Islet the dip changes to the northward, whereby the sandstones and pebbly layers are repeated to the south (see Section E-F). The southern end of the Islet is traversed by a fault which repeats the pebbly layers and brings into view the underlying sandstones. The dip here is 45° north.

The total estimated thickness of Jurassic sequence is 13,800ft.

B.—Notes on the Inland Occurrences.

The geological map (Plate 51) attached hereto indicates the general distribution of the various members in the sequence. The accompanying block diagram (Text Fig. II) was drawn in an attempt to demonstrate how the more resistent strata can be followed inland as well-marked ridges separated by valleys excavated in the softer rocks.

With regard to the Noric series, it may be noted that although no bed to which a definite Noric age can be assigned occurs in the Roaring Bay section, coarse, pebbly, felspathic grits, with thin mudstone layers, both, especially the latter, containing thickly crowded specimens of Pseudomonotis richmondiana Trechm., are found about 6.½ miles inland, directly along the line of strike from the southern part of Roaring Bay. The fossils were found on a ridge, 35 chains south-east of Trig. E, and according to Dr. Marshall (personal communication to Dr. Benson) they also occur on the steep banks of the Glenomaru Stream, 30 chains north-east of Glenomaru Railway Station. Local residents state that they may be found between these two points.

It is not clear what thickness may be assigned to the Noric beds near Glenomaru; the fossiliferous band itself is only a few feet thick, but the pebbly felspathic grits extend through 1550–1650ft. Whether or not the latter represent the whole of the Noric sequence here cannot definitely be stated. The pebbly felspathic grits have been traced to within 2.½ miles of the coast, but then are obscured by bush. Nearer the coast no exactly similar beds could be recognised.

Fifty-two chains east of Trig. E, on Wright's “Harakeke” Farm, outcrops of fossiliferous mudstone contain the following characteristic New Zealand Carnic forms in a fair state of preservation:—

  • Gasteropod:—

  • Pleurotomaria (?)

  • Lamellibranchiata:—

  • Anodontophora edmondiiformis

  • Anodontophora ovalis

  • Gonodon cf. mellingi Hauer as of Wilck.

  • Brachiopod:—

  • Spirigera wreyi

– 286 –

Careful searching revealed no rocks containing the characteristic Roaring Bay Rhaetic fossils, viz., Clavigera bisulcata and Spiriferina diomedea, but beds of similar lithology could be traced at isolated localities for about two miles inland.

Fossiliferous Localities in the Jurassic Sequence:—

  • 1. Sandy Bay. Bed J4.

  • Pseudaucella marshalli……Mid. Lias.

  • 2. Seventy-six chains north-west of locality 1. above. Bed J4.

  • Pseudaucella marshalli……Mid. Lias.

  • 3. Fifty-four chains south-west of Trig. II, on hillside. Base J9.

  • Aucella (?)

  • 4. Three-quarters of a mile north-west of Tuck's Bay in roadside quarry. Top J9.

  • Pseudomonotis cf. echinata Sowerby….Lower Oolite (Trechm., 1923, p. 271)

  • Inoceramus haasti {Park, 1904, p. 385.

  • Inoceramus labiatus Schloth.

  • Pecten

  • 5. Two miles thirty-two chains north-west of Boatlanding Bay (between Sandy Bay and Tuck's Bay) in quarry reserve. J8.

  • An ammonite {Park, 1904, p. 385.

  • Pinna

  • Arca

  • Panopaea

  • Pholadomya

  • 6. Boatlanding Bay (between Sandy Bay and Tuck's Bay). J8. No forms recognisable. McKay's collection (1873) unnamed.

  • 7. North side of Cannibal Bay. J14.

  • Aucella cf. spitiensis……Upper Jurassic (?)

  • Trigonia Park, 1904, p. 386.

  • Ostrea

  • Anomia

  • Inoceramus

  • Pholadomya

  • Venus

  • Ammonites {McKay, 1873, p. 72.

  • Belemnites

  • 8. North end of Bridge over Catlins River, in quarry. J14. No recognisable forms.

  • 9. In Owaka River bank at main-road bridge. Upper J14. Plants obtained by McKay (1873, p. 61) and Hector (1865) and described by Arber (1917, p. 10).

– 287 –

With some hesitation Arber assigns the age of these beds to the Rhaetie (?), noting that they may be either Rhaetic (?) or Lower Jurassic (?). He states that he is inclined to favour a Rhaetic age on account of the abundance of Thinnfeldia (1917, p. 11).

On the evidence of the faunas obtained in lower Jurassic strata, this plant bed must at least be younger than Middle Liassic, or than Lower Oolite if the Pseudomonotis cf. echinata gives a reliable age indication. The presence of Aucella cf. spitiensis in the lower part of this plant-bearing band, indicating an Upper Jurassic age, is an apparent contradiction to the evidence offered by the plants, since Thinnfeldia appears to be very scarce even in the Middle Jurassic of New Zealand (Arber, 1917).

At present, on account of the weighty evidence afforded by the plants, the writer must place the J14 mudstones in the Lower Oolite.

The beds J1 to J4 are included in the Liassic Series, while J4 to J15 are assigned to the Lower Oolite.

Total Thickness of Triassic and Jurassic Systems.—The total thickness of the Mid. Trias, Ladino-Carnic, Carnic, Noric, Rhaetic, and Jurassic beds (allowing 1000ft. for the Noric) is 23,000ft. This appears an abnormal thickness, but there are no major faults to account for it by repetition of beds. Henderson and Grange (1926, p. 31) note that the thickness of Triassic and Jurassic strata exposed in the Huntly-Kawhia district is 28,000ft. (the rocks described by them as Lower Cretaceous have now been proved to be Upper Jurassic).

Recent Deposits.—These are purely superficial and occur locally as thin layers of sand and alluvium as well as sand dunes. McKay (1873, p. 61) states that he collected moa bones from the sandspit at the mouth of the Catlins River, while local residents informed the writer that quite a number of Maori skeletons had been found in the sand at the back of False Islet. Recent raised beach sands and conglomerates are described later among the coastal features.

N.B.—Ongley (1933) has mapped and described the continuation of the Trias-Jura rocks to the north-west of the area at present under consideration.


A.—Detailed Descriptions of Pebbles from Conglomerates.

All percentage estimations and measurements in the following descriptions have been made by eye only, except where otherwise noted. The determinations of felspar compositions are according to the methods and charts of Winchell (1927, pp. 277–341).

1. Middle (?) Triassic Conglomerate at Mouth of Nugget Creek.

Nos. 2119, 2121. Quartz-mica-diorites. Hypidiomorphic; average grain-size, 1–1.¼ mm. The felspar (70%) is usually more or less prismatic, with albite twinning general. Combinations of carlsbad and albite twins are occasionally seen. Almost every crystal is zoned, the composition of the outermost layers being about Ab88An12 and that of the kernels about Ab60An40. There is very little orthoclase

– 288 –

present. Decomposition to kaolin and sericite is taking place more in the cores of the felspars than at the edges. Quartz (15–20%) forms clear, irregular (interstitial) crystals which show very slight straining. The biotite (5%) is pleochroic from pale golden yellow to fairly deep reddish-brown, and occurs as ragged aggregates and little flakes which frequently show incipient chloritisation. Accessories include a few grains of magnetite and plentiful needles and little prisms of apatite.

Nos. 2120, 2122, 2123. Quartz-porphyrites (perhaps with trondhjemitic affinities). Holocrystalline; porphyritic. The phenocrysts (which range in length up to 2.½ mm.) consist of quartz, felspar, and chlorite, the latter being pseudomorphous after some original ferromagnesian mineral. A number of the quartz crystals have been corroded by the magma and show rounded outlines and inlets of the groundmass. There are also perfectly idiomorphic quartz crystals which have not suffered corrosion. The felspar, being fairly decomposed, was difficult to determine, and gave a variety of compositions where determinations were possible. However, from the presence of zoning in some of the crystals it was concluded that the composition varied from Ab70An30 to about Ab84An16. The felspars, as well as the quartz, exhibit corrosion phenomena. Albite twinning is fairly common, and the decomposition product is kaolin. Chlorite is now all that remains of former ferromagnesians. It is usually a flbrous variety with negative elongation and a double refraction of 0.004 to 0.005. The pleochroic scheme is: Y = Z = green; X = light yellowish green; absorption is Y = Z>X. The clear octagonal and hexagonal outlines of some of the pseudomorphs, however, reveal that the original mafic minerals were probably augite and hornblende. The percentage of the ferromagnesians was not great (about 5%). Biotite may have been present. Accessories include sphene, apatite, and iron ore. The apatite is sometimes distinctly dichroic from pale brownish-purple to pale bluish-grey. The groundmass is very finely crystalline and indeterminate.

No. 2124. Very fine-grained, indeterminate base in which are set a few small angular quartz and decomposed felspar fragments and grains of iron ore. The hand-specimen is greenish yellow and very hard; it is probably a silicified mudstone.

No. 2125. Conglomerate matrix. The section shows very decomposed angular and broken fragments of felspar and quartz measuring up to 1 mm. in diameter, set in a fine quartzose and muddy ground. One felspar indicated a composition about oligoclase-andesine, while another showed distinct zoning. Embedded here and there in the section are more or less rounded pieces of exceedingly fine-grained sedimentary rocks, and one small mass of carbonaceous matter.

2. Lowest Carnic Conglomerate of Roaring Bay.

No. 2106. Granite Pegmatite. Micropegmatitic; the plagioclase (15–20%) is allotriomorphic and frequently shows zoning (Ab88An12 to Ab70An30). The crystals reach 3.½ mm. in diameter and are all twinned on the albite law, a few exhibiting pericline or carlsbad twinning in addition. Decomposition is progressing more rapidly in

– 289 –

the centres of the plagioclases. Orthoclase (50–55%) occurs both as large crystals up to 2.½ mm. diameter and in micrographic intergrowth with quartz. The larger crystals sometimes show subidiomorphic outlines, and in one or two cases the margins of these can be seen passing into the intergrowth. The orthoclase is decomposing generally to sericite and kaolin, but granules of epidote as well were observed in the plagioclase. Quartz, usually somewhat strained, reaches 20–25%, while biotite, commonly altered to masses of chlorite, quartz, epidote, and magnetite makes up about 5% of the section. Iron ore (1%) and apatite are accessories. About 40% of the rock consists of a coarse micrographic intergrowth of quartz and orthoclase.

No. 2107. Mica porphyrite. Macroscopically the rock is finegrained, brown in colour, with green felspar phenocrysts. The thin section is holocrystalline and porphyritic, the phenocrysts reaching 2.½ mm. in diameter. The felspar phenocrysts are usually decomposed, and often dotted with bright green spots and clusters of chloritic matter. The general decomposition product appears to be kaolin. Albite lamellation is almost universal in the phenocrysts, carlsbad twinning being also occasionally seen. In one part of the section there is a glomeroporphyritic mass of felspar crystals up to 2 mm. in diameter with an occasional chloritised biotite crystal, and interstitial groundmass. The felspars in this patch are subidiomorphic tabular to more or less elongated. Long needles of apatite are scattered abundantly throughout the felspars in the cluster. As a rule the felspars are somewhat zoned, the kernels being medium andesine and the outer layer about Ab75An25. The biotite (5%) occurs in idiomorphic to subidiomorphic flakes up to 1 mm. long. Pleochroism is from light golden-brown to very dark brown. In every case the crystal is decomposing round the edges to a mixture of little chlorite flakes and grains of magnetite. In some cases the biotite has gone over completely to chlorite and magnetite. Magnetite occurs in irregular and rounded grains up to 0.4 mm. They are fairly abundant throughout the section. Idiomorphic crystals of apatite and zircon also occur. The groundmass is microcrystalline and indeterminate; it apparently consists of felspar and quartz, chloritic and ferritic matter, the latter giving the whole a brownish tinge.

No. 2108. Albitised (?) quartz dolerite. Holocrystalline. The section is composed mainly of felspar laths and tables set at all angles, the interstices being filled by smaller felspar laths and tables together with quartz, grains of epidote, sphene, and shapeless pieces of chlorite. There are occasional grains of magnetite also. The felspar (60%) has an average composition about Ab91An9. The lathy form predominates over the tabular. Albite twinning is common, but combinations of albite and carlsbad twins are also frequently seen. Some of the crystals have started to decompose to kaolin dust, while others appear to have been more or less epidotised. The maximum length of the felspar laths is 1.25 mm. The quartz (10%) fills up the spaces between the felspar laths, where it frequently forms a kind of poecilitic background in which are set the smaller felspar laths and tables, epidote grains, etc. The chlorite (10–15%)

– 290 –

is usually the light green variety, although occasionally yellowish green also, this latter type appearing fibrous between crossed nicols. The double refraction of the epidote varies even in a single crystal from very weak to strong, thus indicating a variation in the Fe2O3 content. Probably all members of the range clinozoisite to epidote are present. The colour varies from colourless to very pale yellowish-green. Plentiful small irregular grains of sphene can be distinguished from the epidote by their higher refractive index and double refraction. The magnetite occurs as small broken fragments invariably surrounded by chlorite.

Nos. 2109, 2110. Partially crushed hornblende-granite. Average grain-size in the less crushed parts is about 1–1.¼ mm., but much finer in the more crushed portions. Some of the plagioclase is zoned slightly, the average composition being about medium oligoclase. The orthoclase, (15%) is generally untwinned. Quartz makes up about 20% of the section, but most of it has been crushed and is very often streaked out into bands, giving the rock a somewhat gneissic appearance, while porphyroclastic structure is well demonstrated in places. The mafic minerals include streaked-out aggregates of little twisted biotites (12–15%), and small, broken hornblendes (3%). A litle chlorite is being produced by decomposition of the biotite and hornblende. Accessories include ilmenite in small grains and aggregates, sometimes in radiating clusters of little needles; sphene in grains up to 0.3 mm.; and small prisms and needles of apatite.

No. 2111. Granophyre. The specimen was very decomposed, and the section consists mostly of a fine micrographic intergrowth of quartz and very much altered orthoclase, in which are set occasional twinned crystals of plagioclase (oligoclase?), clusters of iron ore (ilmenite?), and patches of green chloritic matter. In places the intergrowth is very regular.

3. Second Carnic Conglomerate, Roaring Bay.

This conglomerate is made up of much smaller pebbles than the other conglomerates of the area, with the result that each of the chips collected for sectioning contained a variety of rock types.

No. 2113. Rounded to subangular fragments of the following are all present in this section.


Vesicular basaltic glass containing a few “swallow-tailed” felspar microlites, with green chloritic matter and quartz in the vesicles.


Coarse dolerite made up of felspar laths with plentiful chloritic matter in the interstices.


Andesitic or trachytic fragments; felspar laths show fluxional arrangement.


Rhyolite; groundmass appears to be devitrified, from its extremely microcrystalline state.






Granitoid rock (Trondhjemite). Cf. No. 2094.

– 291 –

2114. Pebbles of:—


Carbonaceous shale.


Fine-grained slaty material.


Banded rhyolite (?).


Vesicular basalt glass.


Fine-grained greywacke.

The matrix is similar to that of No. 2113.

Nos. 2115, 2116, 2117, contain pebbles similar to Nos. 2113 and 2114, and No. 2118 is composed almost entirely of a pebble which resembles very closely that of (g), No. 2113, except that the porphyro-clasts do not reach such a large size.

4. “Granite” Conglomerate, Roaring Bay.

No. 2094. Soda-rich grqnite, or Trondhjemite. Hypidiomorphic; average grain-size about 1.½–2 mm. The felspar, which makes up the bulk of the rock, is generally very decomposed, and often kaolinisation renders determination difficult. By far the most abundant felspar is oligoclase (Ab80An20). One small patch of a vermicular intergrowth of plagioclase and quartz was noted. There is about 5–10% of orthoclase, the exact amount being indefinite on account of decomposition products, but there can be no doubt that it is very subordinate. Quartz (20–25%) occurs in comparatively clear grains up to 2 mm. diameter. It is typically without crystal boundaries and usually has lines of minute inclusions which are often curved. It frequently shows undulose extinction. Small crystals of apatite and biotite are sometimes enclosed. The biotite, of which there is only about 5%, occurs as irregular flakes, and sometimes as subidiomorphic plates. The mineral is strongly pleoehroic, the colour ranging from a very pale straw-yellow to a deep reddish-brown. A peculiar feature is an intergrowth with quartz (Text Fig. I). The latter mineral seems mainly to be in lines parallel to the cleavages of the biotite. Magnetite dust is often present along the cleavage cracks in the mica. Zircon can be distinguished by the pleochroic haloes surrounding it, in the biotite; one of these was 0.053 mm. in radius. Apatite, and magnetite with a limonitic decomposition product, are common accessories. Calcite forms isolated and irregular patches, and may either be derived from decomposition of the felspars or introduced by percolating solutions.

No. 2095. Trondhjemite. Cf. No. 2094.

No. 2098. Trondhjemite. Coarse grained variety (crystals up to 3.½ mm.). The percentage estimations in this section were made with the aid of a Shand's recording micrometer. See table, p. … The plagioclase is more sodie than in the previous two rocks (Ab91An9 to Ab89An11).

Four other trondhjemitic rocks were described from this conglomerate, the mineral composition in each case being essentially the same as in the preceding descriptions. No. 2097, however, differs in that its characteristic feature is the recrystallisation of the quartz which now takes the form of a granular mass of interlocking allotrio-

– 292 –

morphic crystals, demonstrating well the “sutured” structure so common in some hornfelses. The felspar in this rock is oligoclase (Ab84An16), and accessories comprise zircon, sphene, and iron ore. Some parts of the section are almost gneissic in appearance. No. 2105 is a trondhjemitic pegmatite whose felspar, albite-oligoclase, often forms large crystals up to 6 mm. diameter.

Other pebbles sectioned include two porphyrites Nos. 2096 and 2103, the latter being very decomposed, but appearing to have dacitic affinities.

No. 2096. Holocrystalline; porphyritic.

Phenocrysts.—The only phenocrysts are of plagioclase. These reach 1.25 mm. in diameter and are very numerous. Usually sharply idiomorphic in outline, they are, as a rule, strongly zoned. Their composition in the centres is An45Ab55, but on the borders it is about An24Ab76.

Groundmass.—In some parts of the section this might be described as having the orthophyric structure, since there are often numerous litle stout felspar prisms arranged haphazardly amongst finer felspar needles. However, the structure appears to vary in different parts. It usually consists of small felspar needles without any definite arrangement, interspersed with which are little crystals of iron ore and patches of indefinite chloritic substance which is probably the result of alteration of former ferromagnesians.

A greywacke, No. 2104, also listed from this conglomerate, contained shell fragments resembling polyzoons, while the matrix in which the pebbles were bedded (No. 2100) revealed small sections of crinoid (?) stems and a few fragments of felspar with a composition as basic as andesine-labradorite.

5. Triassic Conglomerate near Clinton, Otago.

Professor W. N. Benson kindly allowed the writer to section and describe a number of rocks which were collected by him from the Triassic (Carnic) conglomerates near Clinton, 74 miles south-west from Dunedin. The collection includes No. 2126, a potash-granite or pegmatite; Nos. 2127 and 2128, quartz-biotite-felspar-gneisses; and Nos. 2129, 2130, 2131, 2132, and 2133, quartz-mica-diorites. Descriptions of Nos. 2127 and 2129 are appended, each being fully representative of its group. The potash-granite is quite a normal member of that family.

No. 2127. Quartz-biotite-felspar gneiss. Allotriomorphic; average grain-size = about .7 mm. The felspar, which forms 50% of the rock, has a composition about An25Ab75. There is probably 10–15% of orthoclase present. Twinning is not a very common feature of the felspars, but the crystals frequently show alteration to kaolin, and very occasionally to tiny flakes of sericite. The crystal boundaries are not well defined, the quartz and felspar forming a more or less coarse-grained mosaic in which are set numerous biotite flakes. Slight zoning is shown by some of the plagioclase crystals, while others exhibit undulose extinction. Quartz (25–30%) occurs as clear

– 293 –

allotriomorphic grains which commonly show undulose extinction. The biotite, which is very abundant in this rock (20%), usually has irregular outlines and occurs in crystals up to 1 mm. long. The pleochroism is from a pale yellow to a deep reddish-brown, one or two of the crystals showing lamellar twinning. Small inclusions of zircon must be fairly plentiful, as indicated by the numerous pleochroic haloes, the largest of which was 0.04 mm. in radius. A small amount of greenish chloritic matter is present, and accessories include muscovite, magnetite, and zircon; the muscovite appears to be primary—fringing the biotites in places; the magnetite forms a few little isolated grains.

Both the hand specimen and section indicate that the rock is a recrystallisation product, the former being a comparatively finegrained, light-coloured rock whose surface sparkles due to the numerous little biotite flakes. Banding is not very obvious in the hand-specimen.

No. 2129. Quartz-mica-diorite. Hypidiomorphic; average grainsize = 1.½–2 mm. The felspar is seen often in fairly large, subidiomorphic crystals which are sometimes tabular and sometimes prismatic. Quite often they exhibit zoning, the interior being oligoclase-andesine or acid andesine, and the outermost layer having the composition Ab73An27. Albite-lamellation is almost universal, but twins of the carlsbad type are seen only in a few cases. Decomposition products include both kaolin and sericite. A small amount (1–2%) orthoclase was observed. The quartz (15%) frequently shows undulose extinction; it is interstitial and usually has numbers of small inclusions and bubbles. Hornblende, though present to the extent of only 5% in the section, is shown by the hand-specimen to be much more plentiful (20–15%). The crystals do not generally show good outlines, and the size (in the hand specimen) ranges up to 5 mm. in length. The pleochroism is: Z = deep green. Y = dark olive green. X = light yellowish green. Absorption Z > Y > X.

Occurring as inclusions in the hornblende are small round spots of quartz, and some litle grains of magnetite. The biotite, of which there is some 5–10% in the hand specimen is pleochroic from light yellow to dark brown, and sometimes shows incipient chloritisation. Magnetite dust is occasionally to be seen along the cleavage cracks, and sometimes the cleavages themselves show slight twisting. The iron-ore occurs as little grains of magnetite. Apatite occurs sparsely in little grains and prisms. The biotite and hornblende are occasionally intergrown. The hand-specimen is a light-coloured rock showing quartz, felspar, and black ferromagnesians. It is rather coarsegrained.

6. Conglomerate on South Side of Sandy Bay (Jurassic).

No. 2141. Aplite. (Allotriomorphic). The rock consists of an equigranular (average grain-size about 0.2 mm.) aggregate of quartz and felspar (orthoclase) scattered throughout which are larger quartz grains (up to 3 mm.) and subidiomorphic prismatic crystals of felspar

– 294 –

(up to 2.½ mm.). These latter include plagioclase, commonly zoned from basic albite Ab91An9 to basic oligoclase Ab73An27, a few crystals of microperthite, and one or two of orthoclase. The larger felspars are decomposing to kaolin dust and sericite flakes, being generally quite clouded, while the alteration of the felspar in the granular aggregate has not advanced so far. The majority of the quartz shows slight straining. There is about 10% of green chloritic material occurring as wisps and subidiomorphic flakes (with negative elongation) replacing original biotite, some of which has not entirely disappeared. Other minerals present are magnetite grains, epidote grains (often associated with the chlorite), a few little apatite needles and prisms, and a number of small zircons, which, when they occur in the biotite, are surrounded by the characteristic pleochroic haloes. There is probably 30% of felspar (plagioclase predominating over orthoclase) and 5% of quartz occurring as larger crystals, the granular aggregate making up about 55% of the section. Of this aggregate 45% is orthoclase, about 10–20% albite, and the remainder quartz; the magnetite, apatite, epidote, and zircon make up perhaps 2%.

The hand specimen is a fine-grained granitic-looking rock, carrying occasional large pinkish crystals of non-striated felspar in a greyish crystalline matrix speckled with darker (chloritised) biotites.

No. 2150. Aplite. The same in structure and minerals as No. 2141. The grain-size, however, is smaller in general, that of the granular aggregate being about 0.075 mm., while there appears to be more orthoclase among the larger crystals, and less biotite. The largest felspar is 4 mm. long and 0.3 mm. wide.

No. 2144. Albitised quartz-dolerite. This rock, though a good deal finer grained, is very similar in structure and mineral content to section No. 2108 of Conglomerate No. 2 above.

No. 2145. Ceratophyre. Porphyritic. Phenocrysts: A few idiomorphic crystals (up to 1 mm.) of felspar, most of which is twinned albite (Ab95An5), but some of which is orthoclase. These are all dusted with fine decomposition products, and sometimes show chlorite as an alteration. Other phenocrysts include a small number of fairly large crystals of apatite (up to 0.3 mm. long by 0.04 mm. wide) which all show dichroism from brown to dark brown. Irregular patches of epidote, chlorite, and calcite are scattered throughout the section. An outstanding feature is the presence of vesicles, filled with secondary material, the largest of which is about 5.½ mm. long. The shape is generally irregularly ovoid to rounded, and the filling is mostly epidote in groups of radiating crystals with the centres of radiation situated at the margin of the vesicle. Each vesicle has a narrow border of chlorite, and the epidote is sometimes associated with grains of quartz. The groundmass, throughout which are interspersed small pieces of epidote, sphene, and chloritic matter, with perhaps a little opaque iron ore, is fine grained and is composed of little lath-like felspars showing a frequent tendency to trachytic structure.

– 295 –

7. Conglomerate on North Side of Bay. B (Jurassic).

No. 2140. Trondhjemite.

No. 2139. Granite. Allotriomorphic; average grain-size about 1.½ mm.

  • Felspar Acid oligoclase (Ab89An11) 20–25%

  • Orthoclase 50%

The felspar is all more or less decomposed—the orthoclase perhaps more so than the plagioclase. The latter generally has albite twin lamellation, whereas the orthoclase shows only an occasional carlsbad twin. Quartz is fairly abundant (25%) and has very slight undulose extinction. Accessories: A few small scraps of chloritised biotite, and a grain or two of magnetite.

8. Pebbly Sandstone on North Side of False Islet (Jurassic).

No. 2134. Sediment. The section contains:—

  • (a) A fragment of very decomposed rock made up of epidotised felspar prisms whose average size is 1.½ mm. long by 0.3 mm. broad set in an indeterminate base.

  • (b) A piece of much-altered volcanic rock composed of little felspar laths (acid andesine?) with fluxional arrangement, plentiful magnetite grains, and chloritic matter with an occasional larger, zoned felspar (andesite?).

The rest of the rock contains abundant felspar fragments of all sizes and shapes up to 1.½ mm., but usually a good deal smaller than this. The most basic felspar seen had the composition of basic oligoclase. Quartz does not seem to be very plentiful. Magnetite occurs as grains and streaks of dust, epidote as moderately abundant grains, and chlorite as little shreds.

All the above are set in a fine-grained, muddy, indeterminate base.

No. 2135. Granite Pegmatite. Cf. No. 2106.

No. 2137. Porphyrite. Sections consist of phenocrysts of quartz and oligoclase (probably Ab88An12) set in a very fine-grained quartzofelspathic groundmass in which are little imperfectly radiate aggregates of felspar resembling spherulites. The felspar phenocrysts are nearly all twinned on the albite law, and are very much decomposed to sericite and kaolin. The quartz shows rounded outlines and inlets of the groundmass, and remains clear. There are pseudomorphs of chloritic matter after some original ferromagnesian mineral (hornblende and perhaps biotite).

Note.—A number of pebbles of granite, diorite, porphyrite, and porphyry, apparently very similar to some of the types noted above, have been described by Dr. P. Marshall (1903) from the Triassic conglomerates of Nelson, New Zealand.

No. 2160. Dyke, Nugget Point. Although a number of sections from this dyke were cut and examined, the writer has nothing to add to the very complete description given by Professor Speight in 1904. The rock is a felspar porphyry, the hand specimen being a

– 296 –

dark grey, fine-grained rock containing numerous conspicuous phenocrysts of white felspar.

B.—General Remarks on the Nature of the Pebbles.

Several outstanding features are revealed in the detailed petrographic investigation of the Mesozoic conglomerates:—

  • (a) The majority of the igneous rocks examined are of acid composition, and include granites, trondhjemites, pegmatites, granophyres, quartz- and mica-porphyrites, quartz-mica-diorites, rhyolitic and andesitic rocks, albite dolerites, and a ceratophyre.

  • (b) Many of these rocks are highly sodic, e.g., trondhjemites, albite dolerites, and ceratophyre.*

  • Amongst the most interesting of the above rock-types are the trondhjemites or sodic granites. Goldschmidt, who proposed the name for the group, gives a definition (1916, p. 76) which may be translated as follows: “I define the trondhjemite as a leucrocratic acid plutonic rock whose essential light-coloured constituents are a sodium-rich plagioclase (of the oligoclase or andesine series) and quartz, while potassic felspar is almost entirely absent, or plays a very subordinate role. Among the most sparing, often very sparing, dark minerals, biotite is the most important, though in a smaller division its place is sometimes taken by amphibole (rarely) or even more rarely by a diopside pyroxene.”

  • All the trondhjemites described from the Mesozoic conglomerates conform to this definition, and, as shown in the following table, resemble very closely in mineral composition the rocks obtained by Goldschmidt himself.

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

No. 5 No. 1 No. 45 Dragaasen Trondhjemite Frenstad Skavlien
Quartz 24 25 30 23 21 31
Muscovite 1.5 2 4
Potash Felspar (8) (7) 7 (4) (9) 7
Albite 57 50 47 49 56 46
Anorthite 6 13 13 15 8 8
Biotite 2 5 3 8 5 5
Augite 1
Ore and Apatite 0.4 1 1 0.2 0.1 0.4
  • The mineral content of No. 5 in the table above was measured accurately with a Shand's recording micrometer, but those of Nos. 1 and 45 are estimations by eye only. The three right-hand columns show Goldschmidt's figures (1922, p. 10).

  • As far as the writer is aware, quartz-biotite and quartz-biotite-felspar intergrowths similar to those found in sections No. 2094 and No. 2095 (Text Fig. I) have rarely been described. In the absence of any evidence of metamorphism

[Footnote] * Compare sodic igneous rocks described by Bartrum from Jurassic conglomerates in the Kawhia district, New Zealand.

– 297 –
  • it would appear that the intergrowth is probably the result of partial replacement of the biotite during late stages of crystallisation of the rock under the action of its own highly siliceous residual liquid.

Picture icon

Text Fig. I.—Quartz-Biotite Intergrowth (No. 2095). Q.: quartz; Fel.: felspar; Bi.: biotite; Mg.: magnetite; Ap.: apatite; Ca.: calcite; De.: decomposing biotite.

  • Rocks similar to the trondhjemites and the quartz-micadiorites have lately been described by Bartrum and Benson (1932) from the Fiord region near Preservation Inlet in the south-west of the South Island of New Zealand.

  • The fact that pebbles of albite-dolerite, ceratophyre, and trondhjemite occur together in the conglomerates suggests, from their common richness in sodic felspar, that they may be consanguineous.

  • With reference to the albite-dolerites, the writer has been able to compare them with very similar rocks described

– 298 –
  • by Dr. Benson (1915) from the Devonian System of New South Wales.

  • (c) Basic rocks (basalt glass) are also present to a minor extent.

  • (d) A great number of the plutonic rocks show strain structures, while some of them have been partially crushed.

  • (e) Two gneisses (Nos. 2127 and 2128) occur as pebbles in the Clinton Triassic conglomerate. The problem of their origin is a difficult one, and nothing definite can be stated here except, perhaps, that the abundance of biotite in them suggests a sedimentary or composite rather than an igneous nature for the original rock.

  • (f) The crystals of biotite in some of the rocks show large, dark, pleochroic haloes surrounding small zircons. In recent years it has been established (Holmes, 1927, p. 67) that the largest haloes produced by elements of the uranium family should have a radius of 0.016 mm., and those produced by elements of the thorium family a radius of 0.02 mm. The largest halo, however, measured in the rocks at present under description was 0.053 mm., an unusually large dimension. Similar large haloes have recently been described by Iyer (1932, p. 500) from Indian granites (0.030 mm. radius), and by Wiman (1930)* from Swedish granites and gneisses (0.057 mm. radius).

  • (g) Pebbles of greywacke, mudstone, carbonaceous shale, and slate are present, but there is nothing resembling the schists of Central Otago.

  • (h) In conclusion, it may be stated that the terrain from which the Mesozoic conglomerates, sandstones, mudstones, etc., were derived, consisted in part of greywacke, mudstone, shale, etc., and partly of igneous rocks including plutonic, hypabyssal, and effusive types, the former perhaps predominating. In what direction this terrain lay from the present area is a matter of uncertainty, although the evidence afforded by the presence of pebbles similar to the granites and diorites of the Fiordland district indicates that it may have been somewhere to the south-west.



The orogenic movement which followed upon the deposition of the long series of Triassic and Jurassic sediments reached its maximum in the early Cretaceous. Northwards of the area the Mesozoic strata were flung into broken or overturned folds, while further south, where the movement and stress were not so intense, a number of shallow anticlines and synclines were produced. Succeeding this orogeny

[Footnote] * Abstract in Am. Jour. Sci., Volume XXIV, Sept., 1932, p. 248.

– 299 –

there was probably a period of erosion and perhaps peneplanation. Submergence then took place, and Late Cretaceous (?) and Tertiary (i.e., Notocene) sediments were laid down. Emergence followed, and warping of the new land surface took place in the following manner. (See Text. Fig. II.)

Picture icon

Text Fig. II.—Block Diagram showing positions of consequent streams upon warped surface of Notocene sediments (dotted). The arrows show the directions of the warp axes (N.W.-S.E. and N.E.-S.W.).

The land surface was first warped on north-west south-east axes pitching gently to the south-east, and as a result of this the courses of the Owaka and Catlins Rivers were determined by a broad pitching syncline. The whole region was then affected by a north-east south-west crosswarping movement which continued to operate for a long period. Hence, by this latter movement, the position of the Glenomaru Creek as a tributary of the Clutha River was doubly determined. The Glenomaru Creek, then, drained the northern flank of the broad anticline, and was localised thereon at an early stage by the synclinal axis of the crosswarp. By the crosswarp also was probably formed the shallow basin-like depression now containing the Catlins Lake and the alluvial flat around Owaka township. The effect of continued or renewed crosswarping movement at a late date is indicated by the formation of the swamp near Owaka (perhaps formerly a lake) and its outlet gorge. A certain amount of regional depression probably took place due to the formation of the Catlins Lake.

In time the whole of the overlying Notocene sediments were eroded away, and the previously determined courses of the Glenomaru and Nugget Creeks were superposed upon the underlying Mesozoic rocks in such a way that they flowed in a direction mainly transverse to the strike. Subsequent tributaries were then eroded in the softer of the Mesozoic strata, and the topography gradually assumed its present form.

The above hypothesis as to why the Glenomaru and Nugget Creeks should pursue courses independent of the general strike of the region

– 300 –

was adapted from the explanation by Douglas Johnson of Appalachian Geomorphic Evolution (1931). The application of the theory is not to be confined only to the area under discussion, for by a glance at Text Fig. III it can be seen that the general direction of drainage of adjoining areas is north-west south-easterly, while the north-east south-west crosswarping has been active in the formation of inland plains and lakes, and the cutting of gorges by the Taieri and Tokomairiro Rivers.

Picture icon

Text Fig. III.—Block Diagram (by Dr W. N. Benson) to show the position of the N.E.-S.W. elevation axis along which the crust has been warped upward, causing ponding of the rivers and the formation of gorges where the ponded rivers have cut their way out to the coast. Note the decided difference in topography between the schist (north of Clutha River) and greywacke (south of Clutha River) areas.

The surface forms of the greater part of the area are quite characteristic of a region where a steeply dipping series of alternately hard and soft beds is encountered. The softer layers, consisting of claystone and mudstone, weather more rapidly than the interstratified greywackes and sandstones, due to their finer grain-size and abundance of minute joints. The result is a series of long parallel sandstone and greywacke ridges which are separated by deep valleys excavated in the mudstone and claystone. Where exceptionally wide bands of the less resistant rock occur (e.g., extending back from Sandy and Cannibal Bays, and to the south-east and north-west of Owaka), the topography is notably subdued. (See Plate 51.)

The main creeks (Glenomaru Stream and Nugget Creek), as explained, are superposed consequent streams. In general, they flow across the direction of strike, and their tributaries, which have cut back along the softer bands in the direction of strike, are of the subsequent type. The long sandstone ridges, then, for the most part,

– 301 –

may be termed “subsequent divides.” The subsequent tributaries, in their turn, have smaller insequent tributaries which drain the slopes of the subsequent divides. The resultant stream-network thus has a typical trellised pattern (Marshall, 1912, p. 40; Cotton, 1922, p. 83). The texture of dissection is fine, and the cycle of erosion has now reached the mature stage. A certain amount of rejuvenation has followed upon a recent uplift of about 20 feet, causing many of the smaller creeks to cut little gorges in their older valleys.

Coastal Features.

From Campbell's Point to Nugget Point the coastline consists of open sandy stretches separated by small headlands which continue seawards for some distance as reefs exposed at low tide. From Nugget Point to Cannibal Bay the coast becomes precipitous and inaccessible except where it is broken by a few of the larger bays. Some of the cliffs between these two points plunge sheer into the sea from over 300ft. and are very dangerous at their bases on account of the slippery nature of the boulders and the rapid tidal rise. Between Cannibal Bay and the Catlins River mouth the sandy shore-line is broken only by the perpendicular cliffed embayments of False Islet.

The most recent land movement, viz., an uplift of from 10 to 20 feet, is clearly demonstrated by raised beach remnants and wavecut platforms at various points, particularly between Roaring Bay and Campbell's Point. The small, boggy, coastal plain lying behind False Islet is the result of this uplift also, for it is backed by ancient cliff remnants at several points where it joins the eastern margin of Jacob's Hill.

The sandspit joining False Islet to the mainland is a tombolo, and was probably formed shortly after the uplift.


The writer is indebted to Dr. W. N. Benson and Dr. F. J. Turner, Otago University, for helpful criticism and assistance rendered during the compilation of this paper. To Mrs A. Campbell, of the Nuggets, thanks are also due for the loan of a small cottage in that locality.


Arber, E. A. N., 1917. The earlier Mesozoic Floras of New Zealand. Pal. Bull. Geol. Surv. N.Z., No. 6.

Bartrum, J. A., and Benson, W. N., 1932. The Geology of the District around Preservation and Chalky Inlets, South-West Fiordland, N.Z., Part III. MS. at Otago University.

Benson, W. N., 1915. Geology and Petrology of the Great Serpentine Belt of New South Wales. Part IV, The Dolerites, Spilites, and Keratophyres of the Nundle District. Proc. Linn. Soc. N.S.W., vol. xl, pt. i.

——, 1921. Recent Advances in New Zealand Geology. Pres. Address, Fifteenth Meeting A.A.A.S.

Bittner, A., 1899. Himalayan Fossils: Trias. Brachiopods and Lamellibranchs. Pal. Ind., ser. 15. vol. iii, pt. ii.

Cotton, C. A., 1916. The Structure and Later Geological History of New Zealand. Geol. Mag., dec. vi, vol. iii, pp. 243–9.

– 302 –

——, 1922. Geomorphology of New Zealand, Part I. New Zealand Board of Science and Art. man. no. iii.

Goldschmidt, V. M., 1916. Geologisch-Petrographische Studien im Hochgebirge des Sudlichen Norwegens. IV, Ubersicht der Eruptivgesteine im Kaledonischen gebirge zwischen Stavanger und Trondhjem. Videnskapsselskapets. Schrifter I, Mat.-naturv. Klasse 1916, no. ii.

——, 1922. Stammestypen der Eruptivgesteine. Videnskapsselskapets. Schrifter I, Mat.-naturv. Klasse 1922, no. x.

Henderson, J., and Grange, L. I., 1926. The Geology of the Huntly-Kawhia Subdivision. Bull. N.Z. Geol. Surv., no. xxviii.

Holmes, A., 1927. The Age of the Earth. London, Ernest Benn.

Hutton, F. W., and Ulrich, G. H. F., 1875. Report on the Geology and Goldfields of Otago. Dunedin, Mills, Dick & Co.

Iyer, L. A. N., 1932. A Study of the Granitic Intrusions with their Associated Rocks in Ranchi and Singhbhum Districts, Bihar and Orissa. Rec. Geol. Surv. Ind., vol. lxv, pt. iv, pp. 490–533.

Johnson, D., 1931. A Theory of Appalachian Geomorphic Evolution. Jour. Geol., vol. xxxix, no. vi, pp. 497–508.

Lindsay, W. L., 1862. The Place and Power of Natural History in Colonisation, with Special Reference to Otago. Published as pamphlet, Dunedin, Jan., 1862.

Marshall, P., 1903. Boulders in Triassic Conglomerate, Nelson. Trans. N.Z. Inst., vol. xxxvi, pp. 467–470.

——, 1912. Geology of New Zealand. Wellington, John Mackay.

Marwick, J., 1924. Myalinidae from the Jurassic of New Zealand. Trans. N.Z. Inst., vol. lvi, pp. 304–306.

McKay, A., 1873. Reports Relative to Collections of Fossils in the South-East District of the Province of Otago. Repts. Geol. Expl. N.Z., 1873–74, pp. 59–73.

Ongley, M., 1933. Kaitangata-Green Island Subdivision. Ext. Ann. Rept. Dept. Sc. and Ind. Res., 1932–33, 27th. Ann. Rept. Geol. Surv. Br., pp. 12–18.

Park, J., 1904. On the Subdivision of the Lower Mesozoic Rocks of New Zealand. Trans. N.Z. Inst., vol. xxxvi, pp. 373–404.

Speight, R., 1904. Note on a Dyke at Nugget Point. Trans. N.Z. Inst., vol. xxxvi, pp. 477–479.

Trechmann, C. T., 1917. The Trias. of New Zealand. Q.J.G.S., vol. lxxiii, pp. 165–246.

——, 1923. Jurassic Rocks of New Zealand. Q.J.G.S., vol. lxxix, pt. iii, pp. 246–286.

Wilckens, O., 1927. The Fossils from Shaw Bay, near Nugget Point. East Coast of Otago, Palaeontology of New Zealand Trias. Pal. Bull. N.Z. Geol. Surv., no. xii, pp. 20–31.

Wiman, E., 1930. Studies of some Archean Rocks in the Neighbourhood of Upsala, Sweden, and of their Geological Position. Bull. Geol. Inst., Upsala, vol. xxiii, pp. 1–170.

Winchell, A. N., 1927. Elements of Optical Mineralogy, Part II. New York. John Wiley and Sons.

Picture icon

Plate 51. To face page 302

Picture icon

Block Diagram of Part of the Glenomaru Survey District.