Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 53, 1921
This text is also available in PDF
(2 MB) Opens in new window
– 157 –

Art. XVIII.—An Account of the Geology of Green Island Coalfield.

[Read before the Wellington Philosophical Society, 13th October, 1920; received by Editor, 31st December, 1920; issued separately, 4th July, 1921.]

Plate XXXVI.

Introduction 157
Outline of Geology 157
Previous Observers 160
Topography 160
General Geology 161
   Triassic Formation 161
   Notocene Sedimentaries 162
      Piripauan 162
         Coal-measures 162
         Quartz Sand 162
         Shelly Limestone 163
      Glauconitic Mudstone 164
      Sandstone 165
      Marl 165
      Greensand 165
      Caversham Sandstone 166
      Fauna and Age of Beds above the Piripauan 166
   Volcanic Rocks 167
      Basalt No. 1 167
      Basalt No. 2 168
      Basalt No. 3 168
      Dolerite 169
      Trachydolerite 169
      Basalt Dyke 170
   Notopleistocene Deposits 170
Economic Geology 171
   Coal 171
      History and Mining 171
      Thickness of Seams, Faults, &c 171
      Coal mined and available 172
   Gold 172
   Sand 172
   Clay 172
   Other Materials of Economic Value 174
Bibliography 174


The area described is the Green Island coalfield, which lies at its nearest point four miles south-west from Dunedin. The Kaikorai Stream forms the southern boundary from the Burnside marl-pit to the sea, whence the border follows the coast to Brighton Creamery. A line due north from the marl-pit to Abbott's Hill Road makes the eastern border, the western being formed by a line north from Brighton Creamery to the Chain Hills. The northern boundary is the Chain Hills and Abbott's Hill Road. The area includes Freeman's, Fernhill, Green Island, and Jubilee Collieries, as well as the Brighton Mine.

To Mr. p. G. Morgan, Director of the Geological Survey, the writer is indebted for allowing Mr. G. E. Harris to draw the accompanying map.

Outline of Geology.

The oldest rocks of the district are the schists which form the Chain Hills. Overlying the basement rock with great unconformity are the Notocene sedimentaries. The lowest beds are the coal-measures and quartz sands which outcrop in Fernhill and Christie's Creeks, and at Brighton; at the last-mentioned locality a shelly limestone overlies the terrestrial beds. Resting with disconfomity on the quartz sand or shelly limestone, and comprising the greater part of the area mapped, is a series of marine beds of which the upward succession is glaueonitic mudstone, sandstone, marl, greensand, and Caversham sandstone. There is evidence of a local unconformity between the marl and greensand. All the Notocene sedimentary

– 158 –
– 159 –
– 160 –

rocks have a dip approximating 1 in 8 (7°) in a direction varying between 20° and 55° south of east. In the north-eastern part volcanic rocks rest on an eroded surface of the marine beds and form the caps of the ridges. The upper part of Saddle Hill, which is just outside the western border, is composed of basalt. Beds of clay represent the principal deposit formed since the cessation of volcanic activity.

Previous Observers.

The earliest reference to the Green Island area may be quoted from Hutton (1875, p. 13): “In January, 1862, Dr. W. Lauder Lindsay delivered a lecture in Dunedin on ‘The Place and Power of Natural History in Colonization,’ in which he mentions the volcanic rocks of Dunedin, Saddle Hill, &c…. He considered the sandstones of Green Island and Caversham to be either Tertiary or Upper Cretaceous.”

Hector first reported on the field in his “Departmental Report of the Geological Survey of Otago” of 1864, and subsequently in the Reports of Geological Explorations during 1871–72.

Hutton (Hutton and Ulrich, 1875, p. 47) gave a brief account of the geology, in which he placed the sedimentary rocks and coal-measures of Green Island in his Oamaru formation. In his “Geology of New Zealand” (1885, p. 206) he refers to the belemnites from Green Island, but maintains that other palaeontological evidence argues for a Tertiary age for the rock in which it occurs.

McKay (1877, pp. 59–60) searched unsuccessfully in the vicinity of the Chain Hills and Abbott's Creek for fossils. At Green Island he obtained a number of fossils from a shaft sunk on Mrs. Shand's property. The Brighton calcareous beds were also visited, and a collection made therefrom.

Marshall (1906) describes the volcanic rocks of the area, and mentions the Brighton beds as being Tertiary in age.

Park (1910B, pp. 90, 120–35) includes the belemnite bed in his Waipara series, of Cretaceous age He describes and gives sections of the marine and terrestrial rock exposed between Saddle Hill and Dunedin, classing them in his Oamaru series, the coal-measures corresponding to the lowest beds at Oamaru

Morgan (1916, p. 14) argues that the coals of Brighton and Green Island are of the same age—i.e., Cretaceous—and that in some upper horizon an unconformity exists.


The chief point of interest is the evidence of elevation and depression of the land in post-Tertiary times. The Kaikorai Stream indicates depression of the land, for it has a wide flood-plain and is tidal several chains up-stream from its junction with Abbott's Creek. From Abbott's Hill and Kaikorai Hill the slopes to the middle of the valley are in places fairly steep: waterfalls and rapids are found in the upper parts of the creeks. Kaikorai Stream and the creeks to the west of Stony Hill show a mature form in their lower reaches. Abbott's Creek is at sea-level where it crosses the Main South Road, Christie's Creek reaches the level of the sea on crossing the old Brighton Road, and the Kaikorai is at sea-level 10 chains above the junction with Abbott's Creek From the Main South Road to the sea the flood-plain of the Kaikorai has an average width of about 60 chains. Plains composed principally of gravel stretch far up the creeks. Obviously, the physiography points to a depression of the land.

Terraces about a chain wide and 15 ft. above sea-level occur on the right bank of the Kaikorai Stream, about one mile and a half from the

– 161 –

sea. They have been cut out of the glauconitic mudstone, and are in an excellent state of preservation. A few chains to the east of the Saddle Hill railway-bridge over Abbott's Creek a cutting exposes slightly consolidated sand, as much as 15 ft. above sea-level, resting on a green-coloured mud which contains a few fresh-looking sea-shells. At about the 200 ft. contour-line in the Abbotsford Valley, and to the north of Green Island Station, the country takes on a mature form that is lost at a lower level.

This break in topography is independent of structure. Evidently, then, the sea at one time stood 15 ft., and at another 100 ft. above its present level, but good evidence is wanting to determine the sequence of the diastrophic events. The freshness of the 15 ft. terraces suggest that an elevation of that amount took place after the depression.

General Geology.

The following table shows the classification adopted:—

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

Rocks Probable Age.
River gravel, mud, sand, and clay Notopleistocene.
Basalt No. 3 Post-Awamoan.
Basalt No. 2
Basalt No. 1
Caversham sandstone Awamoan, Hutchinsonian, and Ototaran.
Marl Waiarekan.
Sandstone Paparoan and Kaitangatan.
Glauconitic mudstone.
Shelly limestone Piripauan.
Quartz sand and coal-measures.
(Great unconformity.)
Schist Triassic.

Triassic Formation.

This rock forms the Chain Hills. The junction of schist and Tertiary rocks near Freeman's colliery occurs along Fernhill Creek. Farther south a bluff of rock extends from the Chain Hills into the valley in which are situated the Saddle Hill mines. This mass of metamorphic rock extends south along the hills forming the southern border of Taieri Plains to Otakaia.

The schist is not as completely metamorphosed as that of Central Otago. There is not the frequent lamination of quartz and mica. Thin sections show muscovite, feldspar (albite), and quartz. Dolomite is occasionally found occupying small cavities. The dip of the formation varies. In the railway-cuttings on the east side of the tunnel the following dips were obtained: 20° to the south-east, 40° to the south-east, and 70° to the south-west. Near the lower entrance to Freeman's mine the beds are almost horizontal. A quartz reef 7 ft. wide, and striking 14° south-east, outcrops on the schist spur projecting towards Christie's Creek.

Diverse opinions have been held as to the age of the schists of Otago: Marshall (1918, p. 29): Trechmann (1918, p. 171). Provisionally the rocks of Chain Hills which show a similarity to those of Lawrence (Marshall, 1918, p. 29) may on the same grounds be. placed in the Triassic.

– 162 –

Notocene Sedimentaries.

Rocks of the Piripauan group crop out along a narrow strip which, from near the junction of North Taieri and Abbott's Hill Road, follows Fernhill Creek, then Abbott's Creek to Samson's and Loudon's mines, crosses the Main South Road, and bends round the bluff of schist near Christie's Creek. Isolated outcrops occur on either side of the valley leading north from the Brighton Creamery. The downward succession is: Shelly limestone (absent from Fernhill outcrop), quartz sands, coal-measures.


The coal-measures are composed of quartz sands, fireclay, shales, and seams of coal. Generally six seams of coal are present. The seam above the lowest is the thickest, and is the only one worked.

The coal is close to the schist, so that it is only natural that there will be considerable differences in the thickness of strata below the main seam.

The thickness of the coal-measures is between 100 ft. and 140 ft. The dip is 1 in 8, and is fairly constant over the whole area, but the direction varies within 35°. At Brighton the coal dips about 55°, at Fairfield 20°, and at Abbotsford 32°, all south of east. The main seam averages 16 ft. in thickness, reaching a maximum of 30 ft. near Christie's upper-mine mouth. To the west of Christie's Creek the bed worked splits into two seams, each with a thickness of 2 ft.

The coal of the main seam may be classed as a good lignite, taking a lignite as a coal that contains over 20 per cent. of water. Analyses of the coal are given in the Dominion Laboratory Reports, 1907, p. 59, and Gordon Macdonald's “Brown Coal of Otago,” Colliery Guardian, 22nd November, 1912.

The upper seams in places reach a thickness of 6 ft., but everywhere are inferior to the main seam in quality. Very frequently pebbles of quartz and bands of iron sulphide are met with in the coal. The iron sulphide quickly oxidizes, and is probably chiefly marcasite.

Quartz Sand.

The quartz sand, which is well exposed in five quarries, has a thickness of 50 ft., consists of loose well-rounded quartz-grams and rarely small fragments of schist. The rock-fragments from which the sands were derived must have suffered a great deal of attrition before only the quartz remamed.

In all exposures may be seen two textures of sand—fine and coarse—generally showing current-bedding. Near the top is usually a band of gravel or conglomerate, 1 ft. thick. Clay lenses occur in most of the outcrops.

The lower half of the sand at Gray's pit consists of coarse sand containing pebbles ranging up to ½ in. diameter. Above is 4 ft. 6 in. of fine sand, succeeded by a cemented band of fine sand 1 in. thick and 1 in. of clay. The clay is followed by 6 ft. of coarse sand, and then by fine sand up to the band of conglomerate near the top.

The pit owned by Freemans is rather spoilt by fireclay, which occupies fissures that have been formed in the sand. Mr. E. R. Green informs the writer that 5 chains to the north-west of this pit the fireclay was found to have been intruded vertically through the coal-seams. A fissure had been formed, the fireclay softened by water and squeezed up into the opening by the pressure of the overlying rocks.

– 163 –

The lower 20 ft. of Loudon's pit is composed of alternating bands of coarse and fine sand; above rests a coarse sand with pebbles up to 1 in. diameter, followed by a cemented band 1 in. thick, which in turn is overlain by a layer of fine material.

The old sand-pit near the Southern Trunk Railway consists of bands of coarse and fine sand alternating in a regular manner.

The Jubilee sand-pit has a bed of white well-rounded sand. A layer of gravel forms the uppermost portion of this deposit at Brighton.

Shelly Limestone.

This formation outcrops in vertical cliffs in the valley above Brighton Creamery, where it has a thickness of 50 ft., but no outcrop of the limestone was found at any other locality.

McKay (1877, pp. 59–60) in his report for 1873 says that his Green Island fossils were obtained from a concretionary greensand.

The collection contains, besides fossils from the greensand, pieces of shelly limestone similar to that occurring at Brighton. The shaft to which the collector refers was in all probability Clarkson's, which was situated from 10 to 15 chains east of Walton Park Colliery.

The limestone, which is the only bed giving an indication of the age of the series, has yielded few good specimens. The following have been obtained: Pecten n. sp., Belemnites sp., Ostrea sp., Venericardia sp.

The Pecten the writer obtained has a strong midrib, and is unlike any New Zealand species hitherto described. The bad state of preservation of the Belemnites makes its determination difficult. Hector (1874, p. 356), taking as his holotype the dibranchiate cephalopod from the greensand of Amuri Bluff, described the form in the Brighton and Amuri beds as Belemnitella lindsayi, but in his Progress Reports of 1873–74 (p. xii) expressed the opinion that the fusiform bodies were true Belemnites, and termed them Belemnites lindsayi. Later (1887, p. xxix) he again identified the Brighton fossil as a Belemnitella. Park (1910B, p. 90) sent specimens to Dr. Bather, who pronounced them to be true Belemnites. Marshall (1917, pp. 459–60) sent them to authorities on Belemnites. Professor Stolley, of Brunswick, stated they belong to Hibolites; Professor Steinmann, of Bonn, and Professor Holzapfel, of Strassburg, regarded them as similar to Belemnites minimus of the Chalk; and Lissajous reported that they belong to the genus Neohibolites Stolley. The ages given by these authorities range from Upper Jurassic to Cretaceous. The determinations make the fossil a belemnite or a subgenus of Belemnites that does not reach the Eocene. Trechmann (1917, pp. 338–39) compares the Brighton form with the belemnites in the Upper Senonian of Selwyn Rapids.

Tentatively the shelly limestone and the coal-measures that underlie it may be placed in the Piripauan.

By Park (1910B, pp. 90, 112) and Hutton and Ulrich (1875, pp. 46–48) the coal-measures of Green Island and Abbotsford were thought to be Tertiary in age. The first-named writer placed the Brighton coals in the Cretaceous, but Hutton (1885, p. 266), considering the identification of the fusiform bodies of the shelly limestone doubtful, uses other palaeontological evidence and finds no Cretaceous rocks in the area.

Marshall (1906, p. 389) gives a Tertiary age to all the coal-measures of the Green Island coalfield, on the ground that the cephalopod is Actinocamax, but later (1916, pp. 117–19) accepted Hector's determination, Belemnites lindsayi, and correlated the Brighton beds with the Wangaloa beds.

– 164 –

Morgan (1916, p. 14) advocates a Cretaceous age for the coal of Brighton and Green Island.

Unfortunately, McKay's fossils from Green Island do not settle the question as to whether the coals of the whole area are of the same age, for doubt might well be expressed whether the shelly limestone of his collection really came from Green Island. In his report there is no mention of this rock, but only of the concretionary greensand. All the specimens show some weathering on one side, as if they had been collected from an outcrop. This would indicate that he did not separate the fossils obtained from the greensand from those obtained at Brighton. On the other hand, it is difficult to think such a mistake has been made, when it is known that McKay himself drew up the list of fossils for Green Island and included an it the belemnite. The weathering observed on the limestone could have been produced while the specimens were exposed on the surface” a few years. Hector's Progress Report (1877, p. xiii) on McKay's journey reads as if the belemnite was collected from Green Island. The uncertainty may be dispelled if Christies sink a shaft in the neighbourhood of Walton Park.

Other considerations will show that the coal-measures of Green Island and Brighton are of the same age. There is a lithological similarity between the coal-measures and quartz sands of the two places. The main seam, containing the best coal, is the second seam from the base. Also, the quartz sands are everywhere about 50 ft. thick, and have a band of conglomerate or gravel in the uppermost portion. The sequence as interpreted by Park finds no parallel in Otago. If the coals were of different ages the Brighton beds would be limited to that locality, for the shaft put down in the Saddle Hill Colliery workings did not encounter another set of coal-bearing beds.

Glauconitic Mudstone.

This formation, which is 800 ft. thick, has a fairly wide extent. It forms the main part of the valley of Abbott's and Waterfall Creeks, the hill above Loudon's mine, and the land around Stony Hill and above Brighton.

The dip, determined from the outcrop as a whole and single exposures, is found to be the same as that of the underlying coal. The basal part of the formation is more sandy than the average mudstone. High up in the beds—e.g, by Saddle Hill and in Abbott's Creek—there is a close resemblance to a marl. In Waterfall Creek, 7 chains below the waterfall, the uppermost portion is represented by the following: Cemented glauconitic sandstone, 1 ft. 6 in.: micaceous marl, 9 ft.: cemented glauconitic sandstone, 2 ft.: glauconitic sandstone, 80 ft.

Thin sections of the cemented greensand show angular quartz-grains 0.4 mm. by 0.4 mm., muscovite, green glauconite exhibiting pleoehroism, and oligoclase twinned on the albite law. The position and shape of some of the grains of glauconite indicate that the mineral was deposited in the chambers of Foraminifera, and that subsequently the shell was dissolved away (see Plate XXXVI, fig. 1). The oligoclase could not have been derived from the schist, because the latter has its plagioclase feldspar twinned on Carlsbad but never on the albite law.

On the under-surface of a projecting ledge of cemented glauconite sandstone in Waterfall Creek crystals of gypsum about 2.5 mm. in length were obtained. They can be recognized by the pinacoidal cleavage (010) and the typical minus-unit pyramids (111). Sulphuric acid, which is produced during the weathering of the iron sulphide contained in the sandstone, has reacted on the calcite of the shells to form calcium sulphate.

Picture icon

Fig. 1.—Section of cemented greensand of Waterfall Creek. Shows glauconite in the chambers of a shell. × 45.
Fig. 2.—Crystal of sodalite with inclusions. From the trachydolerite.

– 165 –

The glauconitic mudstone outcropping on the bank of Abbott's Creek near its junction with Waterfall Creek contains septarian nodules. The boulders are perfectly spherical to the eye, and are about 4 ft. in diameter. They resemble the well-known Moeraki boulders.

An inspection of the contact of the glauconitic mudstone with the underlying beds shows it to be disconformable. The shelly limestone, taken as a whole, has a fairly even surface, but in places an irregular contact has been observed. At the point marked x on the map, 13 chains north-east of the Brighton sand-pit, a glauconitic sandstone lies fully 5 ft. below the level of the limestone at 10 ft. on either side; and 4 chains east of this junction a quartz sand, consisting at its base principally of clay, rests on an irregular surface of the Cretaceous strata. The sand grades into a glauconitic sand-stone, which higher up is replaced by a glauconitic mudstone. At Green Island, in exposures at the Jubilee sand-pit and 30 chains to the north-west of Fernhill sand-pit, the sandy base of the glauconitic mudstone contains quartz pebbles up to ½ in. in diameter, and a lens of rock of the nature of a greensand, 30 ft. long and 2 ft. thick, containing quartz pebbles, occurs in the upper part of the quartz sand in the Jubilee pit. Gray's pit shows a similar body, but it is not accessible. Since glauconite is not likely to form where pebbles ½ in. in diameter are deposited, the lower portion of the glauconitic mudstone has most probably been rewashed. The evidence indicates a small erosion interval.


Sandstone beds stretch from the source of Waterfall Creek to Abbott's Creek, round the lower slopes of Kaikorai Hill, and down to Green Island Station. In many places they are exposed in cliffs owing to slipping. The sandstone, which varies from 170 ft. to 200 ft. in thickness, rests conformably on the underlying beds. It is composed of quartz-grains and scales of muscovite, slightly compacted. An excavation made close to one of a line of springs marking the junction of sandstone and mudstone, near Samson's house, showed a sand made up of well-rounded quartz-grains, similar to that occurring above the coal-measures.


The marl occupies a strip from the southern slope of Kaikorai Hill to the railway-line near Green Island Station. It attains a thickness of 170 ft. and has a dip of 1 in 8 40° south of east. The mud has a light-blue colour when freshly exposed, but weathering soon produces a cream tint. The lower part of the bed is somewhat sandy, and contains glauconite. Pyritic concretions occur in the Burnside pit.


A greensand 2 ft. thick rests on the marl at the Burnside pit. About 15 chains to the south-east, in the Kaikorai Stream, a more accessible outcrop of the bed is visible. Here the base of the greensand, which is 6 ft. thick, contains flakes and small pieces of marl up to 1 in. in greatest diameter. Some 9 in. above the contact in a layer 2 in. thick of phosphatic nodules, succeeded 10 in. above by another band with a similar thickness. The nodules appear to be quite similar to those dredged by the “Challenger” Expedition (Deep Sea Deposits, p. 396) south of the Cape of Good Hope. An analysis of a nodule by the Dominion Analyst showed it to contain: Insoluble in acid, 16.68 per cent.; lime (CaO), 38.38 per cent.; carbonic anhydride (CO2), 12.55 per. cent.; water and organic matter, 3.83 per cent.

– 166 –

Under the microscope the greensand is seen to be made up of grains of glaueonite and calcite, with occasional crystals of quartz. The glauconitegrains, which are the most abundant, are rounded, and up to 0.2 mm. in diameter. The Dominion Analyst supplies the following partial analysis of the greensand: Potash (K2O), 2.72 per cent.; phosphorus pentoxide (P2O5), 4.32 per cent.; calcium carbonate (CaCO3), 39.87 per cent. Considering the amount of glauconite present, the percentage of potash is low. The only other analysis of New Zealand greensand the writer can find is that of a sample from Iron Creek, Broken River, Canterbury (52nd Ann. Rep. Dom. Lab., p. 28), which also contains a low percentage of potash—namely, 1.35 per cent.

After the deposition of the marl, elevation caused its erosion, as is shown by the presence of pieces of the marl in the greensand. The bands of phosphatic nodules point to elevation to allow some of the glauconite to be washed away and the concretions concentrated into layers.

Caversham Sandstone.

The Caversham sandstone outcrops about 25 chains to the east of the Burnside marl-pit, whence it extends to Burnside and Caversham beyond the area mapped. The rock which is exposed near the Burnside pit is a compact calcareous sandstone. Outside the area the formation is fully 300 ft. thick.

Fauna and Age of the Beds above the Piripauan.

Mr. p. G. Morgan, after inspecting the fossils obtained by McKay from the glauconitic mudstone, informed the writer that the collection as a whole is unlike any from Oamaru. Mr. J. Marwick, of the Geological Survey, who examined the same collection, reports that the specimens, with the exception of a Dentalium, which is nearest D. pareorense Pilsbry and Sharp, are casts for the most part indeterminable, and comments in the same manner as Mr. Morgan.

The fauna comes from the base of the bed. Odontaspis elegans Agassiz occurs in the mudstone at Abbotsford Station. The cemented greensand at the top of the glauconitic mudstone yielded Dentalium solidium Hutton and a leaf which resembles the dicotyledon Daphnophyllum australe von Ettingshausen (1891, p. 275).

The marl of the Burnside pit contained the following teeth: Isurus retroflexus Agassiz, Notedanus marginales Davis, Odontaspis elegans Agassiz, and O. attenuata Davis.

The Geological Survey forwarded a collection of Foraminifera from the Burnside pit to Mr. F. Chapman, of Melbourne, who determined the following*: Haplophragmium latidorsatum Born, H. emaciatum Brady, Cyclammina incisa Stache, Gaudryina reussi Stache, G. pupoides d'Orb., Nodosaria radicula (L.), N. raphana (L.), Marginulina asprocostulata Stache, Cristellaria rotulata (Lam.), Truncatulina ungeriana (d'Orb.), and Rotalia soldanii var. nitida Reuss. Commenting on their age, Chapman states: “In regard to the Foraminifera of the Burnside marls, they show some affinity with the fauna described by Stache (Whaingaroa), and also with our [Victorian] Balcombian beds. They are therefore low in the series.”

The following is a list of fossils from the Caversham sandstone taken from Hutton (1875, pp. 51–52) and given modern names: Atrina distans (Hutt.), Amusium zitteli (Hutt.), Cucullaea attenuata Hutt., Fulgoraria arabica Hutt. (Mart.), Galeodea senex (Hutt.), Glycymeris laticostata (Q. & G.),

[Footnote] * Communication to Mr. p. G. Morgan, received September, 1920.

[Footnote] † Still living.

– 167 –

Lima paucisulcata Hutt., Pecten beethami var. B Hutt., P. huttoni (Park), Turritella cavershamensis Harris, Pericosmus compressus Tate.

Mr. J. Marwick looked over Hutton's specimens and obtained as well: Chione chiloensis truncata, Sut., Chione sp. ind., Malletia australis (Q. & G.), and Venericardia pseutes Sut.

Mr. Morgan collected Pachymagas parki (Hutt.) from the base of the Caversham sandstone near Green Island Cemetery.

All the mollusca occur in the Awamoan, and Pachymagas parki ranges from the Ototaran to the Awamoan at Oamaru.

There can be little doubt that the Waikouaiti and the Caversham sandstone are the same horizon. Marshall (1906, pl. xxxvi) maps the sandstone of the latter locality continuously to the north of Blueskin Harbour, where volcanic rocks make a separation from the sandstone of the former locality. J. A. Thomson (1918, pp. 196–97) collected from the Waikouaiti sandstone Pachymagas abnormis Thomson, which he states does not occur below the Hutehinsonian at Oamaru. The Caversham sandstone with the greensand appears to represent the Awamoan, Hutchinsonian, and Ototaran stages.

The Balcombian beds are regarded by Chapman as Oligoeene (1914, p. 46), an age that J. A. Thomson (1920, p. 323) thinks corresponds with the Waiarekan. The marl, then, may be placed in the Waiarekan stage, leaving the Paparoan and Kaitangatan for the underlying beds.

Volcanic Rocks.

The following igneous rocks are described below: Basalt No. 1, basalt No. 2, basalt No. 3, dolerite, trachydolerite, basalt dyke.

Basalt No. 1.

This rock occurs on the slopes of Abbott's and Kaikorai Hills. It is the lowest volcanic outpouring in the Abbotsford Valley, and rests on an eroded surface of the marine sedimentaries. There are two good outcrops—one in a cliff below Kaikorai Trig., and the other at the back of Mr. Meechan's house. At the former place the basalt, which is traversed by vertical and horizontal joints, is well weathered. At the latter locality the upper portion is so greatly weathered that it would be difficult to recognize it were it not for the more solid rock below.

Macroscopically the rock is dark grey in colour. Where weathered the augite crystals can be seen, but they do not protrude. This basalt may be microscopically described thus:—

Phenocrysts.—Labradorite (2.6 mm. × 0.66 mm.) is twinned on the albite law, but at times there is a combination of the albite and pericline twinning. Frequently it has magnetite inclusions in the centre. Augite is in large idiomorphic crystals (3.3 mm. × 1.5 mm.), and of a very light brown colour. It exhibits slight pleoehroism. Often this mineral has orthopinacoidal twinning. Some crystals show zoning, the outer border being a little more inclined to a violet colour, owing probably to the presence of titanium. Olivine (1.5 mm. × 1.2 mm.) is allotriomorphic. Along fracture-lines the mineral has changed to serpentine, and some crystals have been entirely altered to this mineral. One crystal has inclusions of microlites of feldspar. Magnetite occurs in grains. The ferro-magnesian minerals are not plentiful, and there is a greater amount of augite than olivine. Spherosiderite, with its irregular pleochroism, is in patches, and in one section is seen as a narrow vein.

– 168 –

Groundmass.—Feldspar laths (0.24 mm. × 0.88 mm.) are of labradorite twinned once or twice. They show a flow-structure. A few grains of olivine and magnetite are also present.

The following is an analysis of the basalt:—

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

Silica (SiO2 50.20
Alumina (Al2O3) 19.70
Ferrous oxide (FeO) 7.35
Ferric oxide (Fe2O3) 6.80
Lime (CaO) 8.72
Magnesia (MgO) 3.71
Soda (Na2O) 2.97
Potash (K2O) 1.47
Water (H2O) Moisture 1.08
Combined 2.65

Basalt No. 2.

Basalt No. 2 occurs on the north-east of the Chain Hills, where it rests on the schist. In the water-race to the west of Abbott's Hill Road the rock, which is in the form of hexagonal columns, has at its base a bed of agglomerate. Owing to the absence of outcrops of basalts Nos. 1 and 2 to the west of Abbott's Hill, it has not been found possible to ascertain the position of No. 2 in the sequence of eruption. Thin sections show—

Phenocrysts.—Occasional crystals of olivine, and more rarely of augite. The former is idiomorphic, and usually has a border of iddingsite.

Groundmass.—The groundmass contain a great abundance of violet-coloured irregular grains of augite. The feldspar is labradorite which approaches a lath shape. Olivine is not common.

Basalt No. 3.

Basalt No. 3 is found above basalt No. 1 on Kaikorai and Abbott's Hills. There are no outcrops—its position and extent have been determined from boulders. The weathered rock has a distinctive steel-grey colour. Under the microscope, shoes show—

Phenocrysts.—Labradorite (1.6 mm. × 0.8 mm.) often containing magnetite in the centre. Olivine allotriomorphic, and changed along fracturelines to serpentine. Not infrequently a border of iddingsite is present. There are a few idiomorphic crystals of augite of a light-brown colour. A cross-section of a crystal showing prisms and both pinacoids has inclusions of microlites of feldspar.

Groundmass.—The groundmass has feldspar laths (labradorite) (0.7 mm. × 0.24 mm.) set in a mass of very fine-grained feldspar. Grains of olivine and augite are not common.

The composition of the rock 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.]

Silica (SiO2) 48.36
Alumina (Al2O3) 15.85
Ferrous oxide (FeO) 8.46
Ferric oxide (Fe2O3) 4.90
Lime (CaO) 10.80
Magnesia (MgO) 6.18
Potash (K2O) 1.37
Soda (Na2O) 1.18
Combined water (H2O) 2.13
– 169 –

The word “dolerite” is used in the sense in which Marshall (1906, p. 410) employs it—namely, “as a term covering all the types of coarse basic rocks, irrespective of age, if they are of effusive character.” Dolerite caps Kaikorai and Abbott's Hills. The rock weathers to spheroids, leaving in the spaces white bands of sepiolite and reddish-black bands of iron oxide. The dolerite can easily be distinguished in the field, since the augite projects on the weathered surface. In fresh samples the augite, olivine, and feldspar can be distinguished. Steam-pores are abundant, and often filled with amygdules of secondary calcite.

Phenocrysts.—Labradorite (1.05 mm. × 0.3 mm.) is not plentiful. Augite occurs in large idiomorphic crystals (3.1 mm. × 1.25 mm.), and is of a violet colour showing strong pleochroism. Twinning is frequent. One crystal has zonal banding, successive layers differing in tint indicating differences in the titanium content; another shows hour-glass structure. The augite commonly encloses olivine in a poecilitic fashion. Olivine (2.1 mm. × 0.48 mm.) is sharply idiomorphic, and is altered to serpentine along practice-lines, but not to the same extent as in the basalts. Of the olivine and augite, the former was the first to crystallize.

Matrix.—The matrix is coarse, being made up of feldspar laths, (labradorite), augite, magnetite, and a little olivine.

There is little distinction between the groundmass and the phenocrysts

The following analysis shows the composition of the rock:—

Silica (SiO2) 44.86
Alumina (Al2O3) 13.05
Ferric oxide (Fe2O3) 8.82
Ferrous oxide (FeO) 6.90
Lime (CaO) 9.94
Magnesia (MgO) 11.60
Potash (K2O) 1.46
Soda (Na2O) 2.10
Combined water (H2O) 1.49


The trachydolerite of this area is found on the northern slope of Abbott's Hill, whence it stretches for many miles to the north towards Flagstaff. A contact with the other lava-flows could not be found. Trachydolerite in the Dunedin district occurs both as a hypabyssal and as an effusive rock, with no distinctive character that would serve to discriminate them, under the microscope. The wide extent of the rock above Abbotsford suggests that it is a lava-flow. In no place in Dunedin is the trachydolerite intercalated with other lavas (Marshall, 1906, p. 407). It probably followed the eruption of the dolerite.

In hand-specimens the rock is readily distinguished by the large crystals of feldspar that project on the weathered surface. Microscopically it shows the following:—

Phenocrysts.—Large idiomorphic crystals of oligoclase (3.2 mm. × 5.8 mm.) with extinction angle of 10. It is twinned on the Carlsbad law. The nepkeline (0.2mm. × 0.8 mm.) is clear and often shows hexagonal outline. Rarely light-blue sodalite occurs. A special feature of this mineral is the great number of inclusions, most of which are arranged in lines with two directions crossing at an angle of 60. The inner part of the crystals

– 170 –

contains more inclusions than the outer (Plate XXXVI, fig. 2). Owing to the small size of the bubbles, it is difficult to determine whether they are filled with gas or liquid. Olivine (0.2 mm. × 0.2 mm.) is not common. It contains inclusions similar to those in the sodalite, but they are not arranged in lines. Augite (0.4 mm. × 0.4 mm.) is of a light-violet colour exhibiting pleochroism. Usually the olivine and augite are mantled with aegirine. The olivine is frequently bunched, as also is the augite. Iron-ore grains are present.

Matrix.—The feldspar is not resolvable. Aegirine occurs in grains. At times it is decomposed, leaving a pseudomorph of magnetite.

The composition of a sample of the rock was as follows:—

Silica (SiO2) 51.85
Alumina (Al2O3) 15.72
Ferrous oxide (FeO) 5.30
Ferric oxide (Fe2O3) 6.21
Lime (CaO) 4.36
Magnesia (MgO) 3.06
Potash (K2O) 4.75
Soda (Na2O) 7.06
Combined water (H2O) 2.64

Basalt Dyke.

Stony Hill is formed of a basalt dyke elliptical in horizontal and wedge-shaped in vertical section. The basalt has been intruded through the glaueonitie mudstone formation, which here is of a sandy nature. The joints of the igneous rocks near the contacts are parallel to the plane of cooling, and only about 2 in. apart A band 6 in. thick of the mudstone has been baked to a cream-coloured rock showing feldspar and augite. Thin sections of the basalt may be described thus:—

Phenocrysts.—Large crystals of labradorite, in places containing inclusions of magnetite. There are smaller crystals of olıvıne and augite. The former is distinctive in that, besides containing a great number of small round grains of magnetite, it shows under crossed nicols a mosaic of colours.

Groundmass.—The groundmass consists of sharply outlined labradorite laths, augite, olivine, and magnetite grains.

The baked mudstone, under the microscope, is seen to consist of large idiomorphic crystals of feldspar (orthoclase and andesme) and augite set in a groundmass of well-outlined feldspars of the same composition as the larger ones.

Ntotopleistocene Deposits.

This includes the clays, muds, and gravels deposited after the close of the Notocene. The clay deposits, owing to their wide extent in the lowlands, have not been mapped. For the greater part the clays are mixed with volcanic boulders, which occur either as a band near the base or throughout the bed. At Abbotsford Station 20 ft. of clay with a band 2 ft. thick of water-worn pebbles 3 ft. from the base, rests unconformably on the glauconitic mudstone. Similar successions occur in a cutting near the Saddle Hill railway-siding, and in an exposure on the Fernhill Coal Company's railway near the crossing of Waterfall Creek. These clay deposits all occur at a height of from 120 ft. to 160 ft. above sea-level.

– 171 –

The Abbotsford Tilery Company work a deposit of pure clay that reaches a thickness of 35 ft. Towards Abbott's and Kaikorai Hills the number of boulders in the formation increases. In a railway-cutting near Freeman's mine large boulders are mixed with no order throughout its depth of 15 ft. Park (1910b, p. 200) states that moa-bones have been discovered at the bottom of the clay at Abbotsford.

A difference of opinion exists as to the origin of the deposit. Beal (1871, p. 276) and p. Thomson (1874, p. 312), mistaking structural marks brought out by weathering on a basalt for glacial striae, conceived a glacial origin for the clays containing boulders at Green Island and Dunedin. Hutton and Ulrich (1875, pp. 69–70) considered that the deposit had been formed by the ordinary weathering of the volcanic rocks. Park (1910a, pp. 593–94) classed this formation as a boulder-clay. Marshall (1910, p. 337) says, “In the clays about Dunedin the only recognizable mineral grains that they contain are those of the most resistant minerals contained in the underlying rock—a matter that at once suggests it has been formed in situ…. There are a few places at relatively low levels where there is a layer of well-rounded pebbles and boulders beneath the clay. These mark old shore-lines…. The clay that covers the boulders in the localities referred to has been washed down the hillside on to them.”

The presence at Abbotsford of well-rounded boulders consisting of dolerite, basalt, trachydolerite, and rarely phonolite rocks at the base, all of local origin., is very damaging to the idea of classing the deposit as a boulder-clay. Had the beds a glacial origin boulders of schist would naturally occur, since that rock outcrops at no great distance from the clay. The bed of gravel from 120 ft. to 160 ft. above sea-level, together with the break in topography at about the 200 ft. contour, signify that the clay accumulated when the land was about 100 ft. lower than it is at present. The mud of the raised beach on the Saddle Hill Railway contains Chione stutchburyi (Gray), a common form on the present-day beach.

The formation of the gravel forming the flood-plains of the creeks, and the sand and mud of the lower part of the Kaikorai Stream, commenced when the land was depressed.

Economic Geology.

History and Mining.

The history of coal-mining in the Green Island coalfields may be found in Hutton and Ulrich's Geology of Otago, Gordon's Handbook of New Zealand Mines, the New Zealand Mines Reports, and Denniston's “Report on the Green Island Collieries, Otago” (1877, pp. 143–73).

Thickness of Seam, Faults, &c.

The greatest thickness of coal is met with in the Saddle Hill mine, where it averages 20 ft. At Freeman's it averages 14 ft., while at Brighton and Green Ialand mines the general thickness is 10 ft.

The main seam is practically free from fireclay. Lenses about 4 in. long are in places met with in the upper part of the seam. Two bands of pyrites ¾ in. thick, one near the floor and the other 5 ft. 3 in. from the floor, run through the coal of the Saddle Hill mine.

– 172 –

The faults which have been encountered in the workings have the same strike as the coal, and usually a throw of less than 6 ft. A fault striking east 65° south, with a throw to the west of about 100 ft., separates the Jubilee and Walton Park Collieries.

Coal mined and available.

Most of the coal within a few hundred feet of the surface has been extracted. The Saddle Hill and Jubilee mines have taken practically all the coal from the outcrop in Christie's Creek to the line indicated on the map. The distance to the west that the coal could be worked was found to be limited because of the splitting of the seam.

The Jubilee Company has now turned its attention to an outcrop about 20 chains south of the railway-siding. Here the amount of coal is small, as the boundary to the dip is the Old Brighton Road. Christie Bros, intend drawing the pillars in the Walton Park area and mining the ground to the south of it. In Freeman's mine most of the pillars from a strip about 20 chains wide back from the outcrop in Fernhill Creek have been extracted. The Fernhill people have worked a block about 4 ½ acres in extent to the north of their entrance. The greater part of the working still contains pillars. The other companies have taken small amounts from areas indicated on the map.

To the end of December, 1919, 2,438,453 tons had been won from the Green Island coalfield, but about 48,000,000 tons remain in this area mapped.

The life of Freeman's mine in its present position is limited. Later a shaft with a depth of at least 300 ft. will have to be sunk in the neighbourhood of Abbott's Creek to reach the dip coal. Christies have many years ahead of them in the field previously mentioned. The Jubilee has at present the triangular area to the south of the railway-siding. The Green Island Company has yet about 900,000 tons to extract. Much coal probably lies to the rise in the Fernhill and Brighton mines.

An unprospected field lies between Brighton and the area wrought alongside Christie's Creek. Bores put down at A and B would reach the coal-measures at about 300 ft. and prove the area.


Hutton and Ulrich (1875, pp. 141–43) give an account of the Saddle Hill reef near Christie's Creek. About 2,000 tons of stone were crushed for an average yield of 5 dwt. of gold per ton—a yield that is unpayable.


Good outcrops of sand free of the overlying formation for a few chains back, are found in many parts. The conglomerate band and lenses of clay give little trouble. In all the pits two classes of sand occur—a sharp sand used for cement-work, and a clean, white, rounded sand used by plasterers. At present each, coal company works the deposit on its ground.


The clay of this area is one of the best of the province. The deposit near Abbotsford Station owned by the Abbotsford Tilery Company is 35 ft. thick, is singularly free from boulders, has a fine texture, and makes an excellent tile. A detailed examination of the clay was made by the Dominion Analyst, who reports as follows:—

– 173 –

“Only one sample was forwarded by the Inspector of Mines, but as this appeared to have been taken from two different seams it was divided into two samples.

Analysis (1.) (2.)
Silica (SiO2) 66.80 66.10
Alumina (Al2O3) 16.16 16.40
Iron oxide (Fe2O3) 4.20 3.48
Titanium dioxide (TiO2) 0.94 0.92
Lime (CaO) 2.10 1.80
Magnesia (MgO) 0.97 1.15
Soda (Na2O) 2.26 1.60
Potash (K2O) 1.74 2.02
Water at 100. C. (H2O) 0.90 2.10
Combined water and organic matter 4.25 4.10
100.32 99.67
Rational Analysis (1.) (2.)
Feldspar 33.16 26.99
Quartz 27.10 27.89
Clay substance 39.79 45.17

“Small briquettes and tiles were made from the samples, and the physical tests on these showed that good bricks and tiles could be obtained between the temperatures 1050. and 1100° C. At 1140° C. the bricks and tiles showed distinct signs of overbuming. The porosity of small tiles made at 1060° C. appeared quite satisfactory.

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

Temp. C. Air Shrinkage. Total Shrinkage. Water-absorption, Three Days. Colour. Hardness. Remarks.
No. 1.
100 6.25 Easily moulded.
970 6.25 13.22 Red Scratch with file Good brick.
1060 6.25 15.88 " " "
1140 15.62 1.78 Dark red " Slightly over burned.
1270 12.50 Very dark red Completely vitrified
No. 2.
100 6.87 Easily moulded.
970 7.81 9.77 Red Scratch with file Good brick.
1060 7.81 11.56 " " "
1140 12.50 1.92 Dark red Not scratched Slightly overburned
1270 12.50 Very dark red Completely vitrified Decidedly overburned

“Microscopically the samples showed the following composition:—

No. 1.—Large irregular crystals of iron-stained quartz. No free crystals of magnetite or rutile, but the quartz in some cases penetrated by capillary or acicular crystals of what appears to be rutile. Feldspar plentiful; some fairly elongated crystals.

No. 2.—Not quite so coarse in texture as No. 1, and not so much ironstained quartz. A little magnetite and a few crystals of rutile. Feldspar plentiful and elongated.”

A large brick-kiln was erected by Gray, of the Fernhill Colliery; but the clay, besides containing a great number of boulders, was found to be unsuitable for brickmaking. Bricks were made some years ago from a bed, about 15 ft. thick, near the Walton Park shaft.

– 174 –

The clays associated with the coal-seams were tested at the Abbotsford Tileries, and found not to withstand a sufficiently high temperature to be called fireclay.

Other Materials of Economic Value.

The Burnside marl-pit was opened in 1903, and works were erected near by. An attempt to produce an hydraulic cement proved unsuccessful. The Milburn Lime and Cement Company now use the marl from this pit to mix with the lime from Milburn to form a cement.

The limestone of Brighton was burnt many years ago for its lime, but the rock proved to be of too low a grade.

The greensand contains constituents that have value as fertilizers (Ries, 1905, pp. 155–56), but the thickness of the deposit near Green Island is not sufficiently great to warrant mining.


Beal, L. O., 1871. On the Disposition of Alluvial Deposits in the Otago Goldfields, Trans. N.Z. Inst., vol. 3, pp. 270–78

Chapman, F., 1919. Australian Fossils. Robertson, Melbourne.

Denniston, R. B., 1877. Report on Green Island Collieries, Otago, Rep. Geol. Explor. during 1876–77 [No. 10].

Ettingshausen, C. von, 1891. Contributions to the Knowledge of the Fossil Flora of New Zealand, Trans. N.Z. Inst., vol. 23, pp. 237–310.

Gordon, H. A., 1887. The Handbook of New Zealand Mines, pt. ii.

Hector, J., 1872. Report on the Clutha and Green Island Coalfields, Rep. Geol. Explor. during 1871–72 [No. 7], p. 165–72.

—– 1874. On the Fossil Reptilia of New Zealand, Trans. N.Z. Inst., vol. 6, pp. 333–58.

—– 1877. Progress Reports, Rep. Explor. during 1873–74 [No. 8], pp. i–xx.

—– 1887. Progress Reports. Rep. Geol. Explor. during 1886–87, No. 18, pp. ix–li.

Hutton, F. W., 1885. Geology of New Zealand, Quart. Jour. Geol. Soc., vol. 41, pp. 191–220.

Hutton, F. W., and Ulrich, A. H. F., 1875. Geology of Otago. Mills, Dick, and Co, Dunedin.

Macdonald, G., 1912. Brown Coal of Otago, Colliery Guardian, Nov. 22.

Mckay, A., 1877. Reports relative to Collections of Fossils in the South-east District of the Province of Otago, Rep. Geol. Explor. during 1873–74 [No. 8], pp. 59–60.

Marshall, P., 1906. The Geology of Dunedin (New Zealand), Quart. Jour. Geol. Soc., vol. 62, pp. 381–424.

—– 1910. Glaciation of New Zealand, Trans. N.Z. Inst., vol. 42, pp. 334–48.

—– 1916. Relation between Cretaceous and Tertiary Rocks, Trans. N.Z. Inst., vol. 48, pp. 100–19.

—– 1917. The Wangaloa Beds, Trans. N.Z. Inst., vol. 49, pp. 450–60.

—– 1918. The Geology of the Tuapeka District, N.Z. Geol. Surv. Bull. No. 19 (n.s.).

Morgan, p. G., 1916. Records of Unconformities from Late Cretaceous to Early Miocene in New Zealand, Trans. N.Z. Inst., vol. 48, pp. 1–18.

Park, J., 1910A. The Great Ice Age of New Zealand, Trans. N.Z. Inst, vol. 42, pp. 580–612.

—– 1910B. Geology of New Zealand. Whitcombe and Tombs, Christchurch.

—– 1912. The Supposed Cretaceo-Tertiary Succession of New Zealand, Geol. May. dec. v, vol. 9, p. 496.

Ries, H., 1905. Economic Geology of the United States, pp. 155–56. Macmillan Co., New York.

Thomson, J. A., 1918. On the Age of the Waikouaiti Sandstone, Otago, New Zealand, Trans. N.Z. Inst., vol. 50, pp. 196–97.

—– 1920. The Notocene Geology of the Middle Waipara and Weka Pass District, North Canterbury, New Zealand, Trans. N.Z. Inst., vol. 52, pp. 322–415.

Thomson, P., 1871. On the Sand Hills, or Dunes, in the Neighbourhood of Dunedin Trans. N.Z. Inst., vol. 3, pp. 309–32.

—– 1874. Glacial Action in Otago, Trans. N.Z. Inst., vol. 6, p. 312.

Trechmann, C. T., 1917. Cretaceous Mollusca, Geol. Mag., vol. 4, pp 337–42.

—– 1918. Triassic of New Zealand, Quart. Jour. Geol. Soc., vol. 73, pp. 165–246.

Picture icon

Fig. 1.—General view of striated rocks at Circle Cove.
Fig. 2.—Near view of striated rocks.