Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 38, 1905
This text is also available in PDF
(1 MB) Opens in new window
– 425 –

Art. LI.—The Occurrence of Gold at Harbour Cone.

[Read before the Otago Institute, 8th August, 1905.]

Plates IX-XIII.
Introduction.

The occurrence of gold in igneous rocks in New Zealand has been treated of in only one instance—that of the andesites of the Thames goldfields. In these the gold occurred in such small quantities as to require special means of determination, and existed only in the bands of rock immediately next to the walls of the lodes. On the Otago Peninsula, however, it occurs so plentifully in an alkaline syenite that the usual fire methods of assay can be used for its determination. This syenite occurs at the base of a mountain on the Peninsula called Harbour Cone.

Harbour Cone, as will be seen on the accompanying map, is situated about the centre of the Otago Peninsula, a volcanic peninsula which juts out from the coast of the South Island of New Zealand, here composed of sedimentary and ancient metamorphic rocks. The mountain lies at the back of the settlement of Portobello, and for the last fifty years its slopes have been farmed. It rises on one side from the Otago Harbour and on the other from Hooper's Inlet—a shallow sea-connected lagoon—at first in fairly gentle slopes, and from an elevation of 500 ft. rather steeply to its summit, 1,044 ft. high (trig.). The sides are for the most part grassy, but a portion of the central cone is covered with bush. The top is composed of a hard cap of solid rock. Its sloping sides and steep central cone give it the typical appearance of a volcano, and popular belief has always considered it an extinct one.

About 1874 the district was startled by the discovery of gold at Harbour Cone. Shafts were sunk and a drive made in a valley to the south-east of the summit, and a five-head battery was erected. However, after a short time it was seen that with the amalgamation process then in vogue profitable treatment of the ore was not possible, and operations were suspended. While the mine was being worked it was visited by Professor Ulrich, who reported very favourably on its prospects, but, despite this, work had to be abandoned. Since then nothing more has been done; the battery has been removed, and the shafts been allowed to fall in.

– 426 –

The mountain presents several difficulties for a geological examination. The surface of the rock composing it is everywhere covered with a thick mantle of decomposed rock and vegetable mould, which often renders the nature of the rock beneath a matter for conjecture. The evidence afforded by surface boulders has sometimes to be relied on, and, since they may have rolled some distance, deductions based upon them may not always be reliable.

A map showing any topographical features was not procurable, and in preparing a map of Harbour Cone and its environs it was necessary to first survey it. A contour map was made, the survey being made with aneroid barometer, prismatic compass, and Abney level. The position of all main points was fixed by means of cross-bearings, and contours then run at every 100 ft. from sea-level.

Two shafts existed in the valley where mining was originally carried on. The lower one was filled with fallen earth and logs, and hence was not accessible. The upper one was in a better state of preservation, but it was only possible to descend as far as the first cross-drive, as water had risen in the shaft below. It was thus possible to penetrate only about 5 ft. or 6 ft. into the auriferous rock, as will be seen in the section (Plate XIII).

It is hard to understand how the auriferous nature of the rock, or, indeed, how the rock itself, was ever discovered. The mountain-sides and all the valleys and stream-beds were carefully prospected, but no trace of the rock could be discovered on the surface, and the site of the mine renders it improbable that it was accidentally discovered in sinking for water. Panning-off of the creeks showed no colours of gold nor trace of pyrites, nor any other mineral that would point to the possible existence of gold.

The purpose of this paper is to describe the characteristics and occurrence of the auriferous syenite and associated rocks, and if possible to account for the occurrence of gold in the rock. The occurrence of gold in plutonic and volcanic rocks is a question bearing strongly upon its presence in lodes and allied bodies, and the occurrence of it in comparatively large quantities is a distinct peculiarity.

In rock-analysis the methods described by “Berringer's Assaving” were exclusively used, and in the assay of the syenite for gold and silver the usual fire methods were used. The assays were run with one or two check assays at the same time, and the assays and checks agreed almost exactly. The charge employed was,—

– 427 –
Grains.
Ore ground to pass 100-mesh sieve 1,000
Sodium-carbonate (Na2Co3) 1,500
Borax (Na2B4O7) 750
Litharge (PbO) 1,000
Powdered charcoal (C) 20

the silver value of the litharge being subtracted from the weight of the bullion.

With each three assays a blank was run. Broken glass was powdered in the same mortar and on the same buckling-plate, and put through the same sieve and then assayed with the same charge. The result was the silver value of the litharge only, thus proving that no gold or silver had been introduced through the agency of the apparatus used.

General Geology.

The rocks of the Otago Peninsula and of its neighbourhood are in three distinct groups: (1) The basement schists of Otago; (2) the Tertiary sandstones; (3) the volcanic rocks of the Peninsula.

The schists of Otago form the vast pene-plain constituting Central Otago. Round the edges of this plain they are overlapped by beds of various ages, in all classes unconformably. The schist country commences at Brighton, about eight miles to the south of Dunedin. The schists are micaceous and very rich in quartz foliæ and veins. In places mica-schist gives place to chlorite-schist, usually in bands. Immediately north of Brighton the highly denuded surface of the schists dips beneath the Tertiary sandstones, which are almost horizontal. The foliæ of the schist itself are as a whole nearly horizontal in this area, though often showing local plications. The surface of the schists is undoubtedly the basement on which the Tertiary sandstones and volcanic rocks of the Peninsula were laid down. Dr. Marshall has reported fragments of schist occurring in the Port Chalmers breccia thrown up at a late stage of the eruptive period of the Peninsula, which proves that these rocks exist somewhere beneath. Thus this schist clearly underlies the Tertiary sandstones and volcanic rocks of the Peninsula, and it is on their highly denuded surface that the next series, the Tertiary sandstones, occur. The age of these schists is not yet fixed certainly. The two authorities on the matter are Sir James Hector, F.R.S., Director of the Geological Survey of the colony, and Captain Hutton, formerly Provincial Geologist of Otago. The former, in his “Outline of New Zealand Geology,” which forms a summary of the work done by the Geological Survey Department, calls these rocks

– 428 –

“clay-schists” of Lower Silurian age. In his “Geology of Otago” (1875) Hutton agrees with Hector, and calls the schists (Wanaka series) “Lower Silurian,” but in his “Sketch of the Geology of New Zealand”* he gives their age as Ordovician. In a paper published in the “Transactions of the New Zealand Institute,” 1899, he classes the schists as Archæan.

The Tertiary sandstones overlying these schists form the surface of a large block of country to the south and west of the Peninsula. On the Peninsula itself it is almost everywhere covered by tremendous outflows of lava. However, along the shore of the Pacific Ocean on the south side of the Peninsula it outcrops, forming cliffs on the top of which are the lava-flows. It is for the most part composed of rounded grains of quartz and finer detritus, and is in many places stained with oxide of iron, doubtless in part derived from the decomposition of the lavas above. It outcrops underneath the basalt at Sandy Mount (see map) at about 100 ft. above sea-level. In the “Geology of Otago” Hutton says that above the coal-deposits which occur in the lower portions of the series are a series of conglomerates (p. 48). These occur as outcrops at Kaitangata, but do not outcrop at Dunedin. In the early part of the winter of 1904 the writer was present at boring operations conducted on the flat comprising the lower part of Dunedin. A steam percussive borer was used, which in most cases reduced the rock pierced to fine powder. After passing through about 50 ft. of harbour-silt and volcanic rock in a decomposed state, and 100 ft. of sandstone, a very hard material was encountered which the borer would scarcely touch. Rounded and broken fragments of quartz were brought up by the pump, which evidently came from a quartzose conglomerate exactly similar to that found above the Kaitangata coal. Thus it seems that this coal exists not far below 150 ft. from the surface.

The sandstone is undoubtedly derived from the denudation of the schistose country at the back. The quartz pebbles forming the conglomerate above referred to are undoubtedly schistose in their origin, showing in places foliation-planes in the separate pebbles. The age of these sandstones is given by Hector as Cretaceo-tertiary in his work above referred to. Hutton (“Geology of Otago”) classes them with the Oamaru series of Oligocene age. In the outcrop along the cliffs to the west of Dunedin fragmentary remains of fossil shells are to be obtained from the cliffs, notably Pecten hochstetteri, which

[Footnote] * Quart. Journ. Geol. Soc., May, 1885.

– 429 –

occurs in considerable abundance. This fossil has been identified by the Geological Survey and Hutton, but in a recent paper by Professor Park and Captain Hutton the identification is disputed, the name Pseudamusium huttoni being assigned to the fossil.

The remainder of the Peninsula is covered by a thick covering of volcanic rocks, forming lofty mountains and ranges. With the exception of basalt they are almost all very alkaline in character. Dr. Marshall in his section on geology in “Dunedin and its Neighbourhood” shows a map in which areas are covered by basalt, dolerite, trachyte, trachytoid phonolite, nepheline basanite, nepheline tinguaite, and kenyte, similar to that described by Gregory from Mount Kenya in East Africa. It is upon the very highly denuded surface of the Tertiary sandstones that these rocks outpoured, a fact shown by the variation in the level of the outcrops of the sandstones.

Harbour Cone.

A glance at the map accompanying will show that this peak and its immediate neighbourhood is on the surface almost entirely composed of bostonite. The surface of the land is precipitous, and rises to Harbour Cone to form in appearance almost a typical denuded volcano and its bared solid pipe. the bostonite has a very great extent in this neighbourhood. It forms cliffs along the shore of the harbour to the east of Port Chalmers, and also tops the small hills surrounding the central peak. The central portion of the peak itself is composed of solid basalt extending down to the 900 ft. contour-line. Below this down to 750 ft. boulders of basalt cover the ground thickly, having been wedged off by frost and other natural agencies and rolled into their present position. On the northern slopes of the mountain an irregular area is occupied by a coarse breccia, very indurated, containing fragments of trachytoid phonolite and other rocks in large and small angular fragments. This is the Port Chalmers breccia, the greatest occurrence of which is on the peninsula on which Port Chalmers is built, which consists almost entirely of the breccia. The breccia also forms the upper portions of the small hill to the north-east of the main peak, as indicated on the map. The most important inclusion in the rock is large and small masses of alkaline syenite, showing that this rock must have enormous extent under the surface at an unknown depth, as the focus of the explosive eruption producing the breccia is somewhere near Port Chalmers.

The flanks of the mountain are pierced by numerous dykes. The position and nature of those occurring in the neighbourhood

– 430 –

of the mine are shown on the accompanying map. They are very varying in their nature, in this small valley four varieties being met with—two varieties of trachytoid phonolite, one of bostonite, and one of tinguaite. The large auriferous mass of alkaline syenite rises to within a few feet of the surface here.

Along the hill-slopes to the west there are two outcrops of Tertiary sandstone indicated on the map. One of these has been quarried, so that the dip and nature of the rock is plainly visible. It is surrounded above and below by bostonite. The sandstone dips about 5° west, and where visible is in no way contorted or faulted. The end of the drive down the shaft shown on the map is walled with a sandstone crumbling in the hands and glittering with scales of a golden-yellow mica.

In the valley in which the mine is situated a chip was obtained from a boulder, which under the microscope exhibited peculiar characters. This probably was derived from a dyke of the rock, the dyke now being covered with loam, as no trace of it in situ could be discovered. A description of the petrographical characters of the rock will be found under “Petrography.”

The first outflow in this area was undoubtedly the bostonite. The absence of any distinct flow-structures in it prevents any conclusions being arrived at as to its probable vent. It seems to have welled up and covered the neighbourhood with a deposit of enormous thickness. It flowed over a very highly uneven sandstone surface, as the outcropping of that rock at between 200 ft. and 300 ft., and its complete absence from the river-bed below, proves. It would seem that an original sea-cliff existed along that portion of the sandstone now outcropping. The top of this cliff now shows where the bostonite has weathered away.

Some time after the flow of bostonite the intrusion of the auiferous syenite took place. This intrusion is of large extent, since it occurs in fragments in the breccia thrown up at Port Chalmers as mentioned above. It probably forced up the sandstone in places, assuming dome-like prominences. This is probably the origin of the sandstone found in the drive in the mine, it being merely a portion of the main beds carried mechanically upwards. It is undoubtedly after the flow and consolidation of the bostonite that this intrusion took place, firstly because the bostonite has been altered along its junction with the syenite, secondly because this alteration is of small extent owing to the solid state of the bostonite.

Dykes were formed first of bostonite (see “Petrography”),

– 431 –

which was probably intruded into the hot bostonite. The heat of the bostonite kept this dyke in a molten condition long enough for the magma to act on the first-formed crystals, as described under “Petrography.” Then followed dykes of tinguaite and phonolite, in what order it is almost impossible to say. They were probably connected with the large eruptions of alkaline rocks on other parts of the Peninsula.

At a later stage followed an eruption of an entirely different nature. The basalt cap now topping the mountain is certainly the upper portions of a basalt-filled volcanic pipe, which communicated with the vent of a basalt-emitting volcano on the surface of the bostonite. After that enormous denudation went on and entirely denuded away any outflows of basalt which may have occurred. Some time during this period the explosive eruptions having their focus near Port Chalmers formed a deposit upon the partially denuded bostonite, which must have had a form then approximating to its present one, as a deposit of the breccia is found on the mountain-slopes. Since then more denudation has taken place. The breccia has, with the exception of the deposits shown on the plan, been entirely removed. A river has cut its passage through the flow, leaving the projecting hummocks now forming Quarantine Islands, which are entirely composed of bostonite. The hard pipe of basalt has resisted denudation while the bostonite all round it has been denuded away, thus giving to the mountain its present form. Thus, though the mountain has the typical form of a volcano and its projecting neck, it is far from probable that the materials composing its slopes have been ejected from its summit, but rather the white bostonite once formed a high plateau on the top of which was once a basalt-emitting volcano.

Thus, in this particular area, the order of outflow seems to have been—(1) bostonite; (2) intrusion of bostonite dykes; (3) intrusion of syenite mass, intrusion of tinguaite and phonolite dykes in unknown order; (4) outflow of basalt; (5) explosive eruptions producing breccia.

In giving the order of flow for the whole Peninsula, Dr. Marshall, in the work above referred to, writes—(1) trachite (bostonite); (2) basalt and nepheline basanites; (3) green trachytoid phonolite; (4) kenyte; (5) Port Chalmers breccia.

In the neighbourhood of Harbour Cone it seems probable that the phonolite and tinguaite dykes preceded the basalt, since the latter is remarkably fresh, while the former is very decomposed. The phonolite, however, lends itself to decomposition very readily, so that arguments based on relative decomposition are perhaps not reliable.

– 432 –

Note on Pseudamusium Huttoni.

This is the most abundant and characteristic of the molluses of the Dunedin Tertiary sandstones. It was originally identified by Hutton as Pecten hochstetteri described by Zittel (“Reise der ‘Novara’—Palæontology,” tab. xi, fig. 5a and 5b), but, as Professor Park* has lately pointed out, the shell does not correspond with Zittel's Pecten hochstetteri, the latter having one valve ribbed, while the former has both valves smooth. He has therefore given it its present name. A sketch of this shell is shown in Plate IX, fig. 1.

Petrography.

The volcanic rocks of Harbour Cone, like almost all those of the volcanic portion of the Otago Peninsula, show by their constituents a clear derivation from an alkaline magma, with the exception of the dolerite occuring at the summit of the mountain, of which the minerals do not indicate any alkaline chemical composition. The characteristic minerals of all are in many cases very strongly alkaline, the persistent occurrence of ægerine being most striking. The rocks, though differing in their mineralogical characters, indicate by their mineral composition a very close relationship in their chemical composition. These general resemblances point to the fact that the rocks come undoubtedly from one common magma situated below the district, and any differences in mineral composition must be explained by one of the theories of the differentiation of magmas. It is more than likely that a differentiation or a very great local variation in magmatic composition will account for the presence of the dolerite in such close proximity to this collection of alkaline rocks, as the difference in mineral composition is no greater than has been recorded in other flows not only in the same neighbourhood but issuing from the same vent.

For determining the plagioclase feldspars the following method was employed: Crystals showing albite twinning were selected cut as nearly at right angles to the composition plane as possible, which was indicated by adjacent lamellæ extinguishing between crossed nicols at right angles on either side of the albite plane. Several crystals were measured, and the maximum value for the extinction was compared with the table given by M. Levy (“Étude sur la Détermination des Felspaths dans les Plagnes minces, 1894,” pt. ii, p. 29, et seq.).

Many of the rocks are nephelinitoid, but the nepheline is often ultra-microscopical. In cases where its presence was suspected a portion of the very finely ground rock was treated

[Footnote] * Paper read before the Otago Institute, 9th September, 1904.

– 433 –

with dilute hydrochloric acid on a microscope-slide, and then slowly evaporated to dryness. Under the high-power objective the presence of nepheline was indicated by the occurrence of small cubes of sodium-chloride.

Auriferous Syenite.

Two writers have mentioned this rock. Professor Ulrich, in the “Geology of Otago” (1875), by Hutton and Ulrich, describes it as “a coarsely crystalline rock composed of triclinic feldspar, hornblende, and some quartz, and being more or less densely impregnated with pyrites.” Professor Park* describes the rock as a “grey porphyritic rock of plagioclase feldspar and hornblende,” and names it a “porphyritic diorite.” As will be seen below, there is no visible quartz in the rock, and it is distinctly neither porphyritic nor yet a diorite. In hand specimens the rock is seen to be composed chiefly of feldspar and a dark mineral resembling hornblende in habit. It is seen to be impregnated sometimes rather freely with pyrites, showing on the fractured surface small flakes of the mineral. Some specimens, however, contain hardly any sign of pyrites. It is extremely hard and tough, being very difficult to fracture on account of the latter property. Under the microscope the rock is seen to be holocrystalline and of course texture. The bulk of the rock is composed of hypidiomorphic crystals of orthoclase, sometimes twinned on the Carlsbad law, and triclinic feldspar twinned rather coarsely on the albite law. These give a maximum extinction on the albite plane of 12°, being therefore oligoclase.

Throughout the body of the rock are very many small crystals, with irregular outline, of green transparent ægerine pleochroic, and showing the high birefringence characteristic of that mineral. Small rounded grains and needles of ægerine also occur abundantly.

Large and very decomposed crystals are present intergrown with the feldspar, and where these are not entirely decomposed they show the low birefringence and straight extinction of nepheline. In most cases, however, the mineral is entirely decomposed, and as it is not present in great abundance no absolute determination is possible.

Throughout the rock are large granular masses, in shape roughly that of typical hornblende. In most instances these are the same right through the mass, but in some cases a core of brown pleochroic hornblende is seen (Plate IX, fig. 2). It is possible to cut as many as ten sections without finding this feature. Where it occurs the mass presents the following characters:

[Footnote] * Rep. N.Z. Geological Survey, p. 34 (1888–89).

– 434 –

In the centre is the rounded core of horneblende, evidently from its pleochroic characters rich in iron. This is surrounded by a granular mass of opaque small rounded grams. Some of these grains are yellow in reflected light and others black, thus showing themselves to be pyrites and magnetite. They occur in about equal proportions. Between these grains small plates of green ægerine are to be seen, and needles and basal sections of clear apatite showing the typical uniaxial characters of that mineral. In places small pieces of the hornblende also occur interstitially. There is, besides, an isotiopic mineral whose characters show it to be analcite. It is a decomposition product such as would be expected in an alkaline rock.

These granular masses with their central core of hornblende form an interesting example of resorption zones. They are in no way the result of weathering, but result from the chemical effect of either the magma or superheated percolating solutions. The original hornblende has been acted upon by one of these agencies, with the resulting granular mass of ægerine, pyrites, and magnetite. Where no core of the hornblende remains the mass has a distinct hornblendic habit as a whole, being either a rounded mass such as would result from the cutting of an irregular crystal of hornblende parallel to the base, or else showing as a prism. Where interstitial hornblende remains, or where a core is left, its orientation is the same as that of the mass surrounding it. The apatite needles were originally in the hornblende, a feature very common in this mineral. The order of consolidation has been (1) apatite, (2) hornblende, (3) nepheline, (4) feldspar, (5) resorption action and consolidation of pyrites in the body of the rock.

Decomposition of the feldspars has produced kaolin and not serecite.

A chemical analysis of the rock gave as follows:—

Per Cent.
SiO2 50·8
Al2O3 22·5
Fe2O3 9·1
CaO 2·3
MgO 1·4
K2O 3·9
Na2O 5·7
99·6

From a comparison of this analysis with those given by Rosenbusch* the rock most closely resembles an amphibole

[Footnote] *“Elemente der Gesteinslehre” (ref. p 131, Analysis, p. 29, No 10) 1901.

– 435 –

foyaite from the Zwart Koppies (Transvaal. South Africa). The analysis of this is:—

Per Cent.
SiO2 53·73
(TiO2) 0·09
Al2O3 20·35
Fe2O3 3·74
FeO 1·13
(MnO) 0·51
MgO 0·47
CaO 2·72
Na2O 7·94
K2O 6·05
(P2O5 2·02
Ph2O 0·23

A comparison of these shows that this rock contains more silica and alkalis than the Portobello syenite and much less iron. It is to be remembered, however, firstly, that in the portion of the rock from which specimens were obtained chemical action has taken place with the formation of abundant magnetite and pyrites, which sufficiently accounts for the high percentage of iron. The rock also is in a decomposed state, and the alkalis have probably been removed to some extent. The lower percentage of silica, however, is only to be accounted for by an original difference in magmatic composition. The difference is so small that it does not preclude classing the rocks together. The total absence of all sphene, and consequently of T1O2, is noteworthy, as this is usually an accessory in alkaline syenites. The greater amount of water is also explainable by the weathered state of the Portobello rock. Rosenbusch shows graphically the percentage of silicon and the metals in certain rocks in his work above referred to. A graphic comparison is given here of the Portobello syenite and the amphibole foyaite, where the resemblance is at once seen. The differences noted above are also to be noticed.

Picture icon

Graphic representation of chemical composition of Porbello syenite compared wtoith foyaite (Brathagen). (After Brogger.)

– 436 –

The rock is also compared with a foyaite by the graphic method described by Brogger.* In this method the relative molecular proportions of the oxides were calculated from the analyses and then shown graphically, and the departures from a typical foyaite illustrated. The type taken is one given from Brathagen. The molecular proportions of the rocks are:—

Portobello Syenite. Foyaite [(Brathagen).
SiO2 0·8816 0·9250
Al2O3 0·2294 0·2201
Fe2O3 0·059 0·0064
FeO 0·0183
CaO 0·043 0·0286
MgO 0·035 0·0111
K2O 0·042 0·0583
Na2O 0·098 0·1732

The general similarity of the figures produced is to be noticed at once, and the high amount of iron, higher amount of potash, and lower of soda is to be seen.

Picture icon

Representation of amount of silicon and metals in Portobello syenite and foyaite. (After Rosenbusch.)

[Footnote] *“Die Eruptivgesteme des Kristianiagebietes,” vol. iii (1898), p. 248, et seq.

– 437 –

Bostonite.(Plate XII, fig. 8.)

This rock, which, as mentioned in “General Geology,” constitutes almost the whole of the mountain and neighbourhood, is, where not discoloured by iron-oxide, quite white. Two varieties exist. The first is a finely crystalline rock in which no crystals of feldspar can be seen with the unaided eye; the second a coarse variety in which large crystals of feldspar are seen, often ½ in. in length. This latter exists along the foot of the cliffs fronting the harbour, and it is doubtless to the deep burying and consequent slow cooling that this coarsely crystalline character is due. The remainder of the mountain is composed of the finer variety. Under the microscope the rock is seen to be composed of a number of polysynthetically twinned lathes of feldspar showing an evident flow structure. The rock is completely crystalline, the only other mineral present being magnetite in small granules and rounded masses. The rock is rather decomposed, the resulting product being white kaolin.

Ulrich in describing the district calls the rock a trachyte, to which family it undoubtedly belongs. Though the rock is a lava, its characteristic feldspathic mineralogical characters ally it to the hypabyssal bostonites. Harker (“Petrology for Students”) defines these rocks as consisting “of feldspar, quartz never being abundant, and ferro-magnesium minerals being typically absent. Phenocrysts may or may not be developed, the bulk of the rock being a groundmass of little feldspar rods often with partial flow disposition.” There evidently is a very close resemblance between this rock and the bostonites, except in their occurrence. The name has been applied to the rock even though it is a lava.

A determination of the alkalis gave soda (Na2O) = 5·7 per cent., and potash (K2O)=4·2 per cent. Since the rock consists almost entirely of feldspar, this shows that the feldspar is anorthoclase, rather rich in soda, thus allying the rock to the ceratophyres.

A variety of the rock occurring in a band surrounding the syenite is seen to be much harder than the mass of the bostonite. It is densely impregnated with iron-pyrites. Under the microscope its structure is that of the last variety, but it has large masses and small grains of pyrites in it, together with patches of red iron-oxide. This band is certainly due to the altering effect of the intrusive mass of syenite.

Bostonite Dyke.

This is a rock which occurs in a large dyke 8 ft. across, as indicated in the map. Its course can be clearly traced for 5 or 6 chains, as the solid rock shows above the surrounding loam and soil.

– 438 –

In hand-specimens the rock is very coarsely crystalline, large crystals of feldspar, both orthoclase and plagioclase, being visible. Microscopically the rock is holocrystalline and very coarse in structure. Plates of orthoclase, often with Carlsbad twinning and polysynthetically twinned plagioclase, make up the bulk of the rock. The plagioclase gives a mean extinction angle on the albite twinning plane of 10½°, and is therefore a variety of oligoclase. Dark patches of a granular nature consisting of magnetite grains occur in elongated shape throughout, often roughly assuming an amphibole form. These are penetrated by clear needles of apatite. These show another example of resorption zones. The outline of many of the granular masses is distinctly hornblendic, and the inclusion of apatite, a mineral often found piercing crystals of hornblende, points to the original existence of that mineral in the rock. Reaction of the magma has corroded the crystals, leaving only the magnetite and apatite, which evidently resisted the action. Differing from the syenite, there are no remaining cores of hornblende in the rock.

The naming of this rock is somewhat problematical. Originally possessing amphibole, orthoclase, and oligoclase, together with its non-porphyritic holocrystalline and coarse nature, its characters would class it with the Plauen'cher grund type of Brogger. Among the hypabyssal rocks to which it belongs there is no rock to which it corresponds. The entire absence of trachytic structure, and its coarse nature and evenly developed crystals, render it different from any of the types of porphyries. In view of its almost wholly feldspathic nature, and the mixture of feldspars which clearly places it in the porphyries, it seems to be a coarse example of a bostonite.

The occurrence of the resorption zones in this as in the syenite seems to point to its intrusion into a still-hot bostonite lava. This surrounding hot material kept the magma surrounding the first-formed crystals, the hornblende, in a liquid condition, giving it time to act on the crystals, producing the resorption of them.

Though the name “bostonite” has been given to the rock. It is only on account of its association of feldspars, almost entirely feldspathic composition, and occurrence in a dyke, and not on account of its resemblance to typical bostonites.

Tinguaite. (Plate XI, fig. 6; Plate XII, fig. 7)

A dyke of this occurs as indicated on the map. Its course can be traced by small outcrops for a distance of 3 or 4 chains, though a determination of the width and dip cannot be made In hand-specimens the rock is a dull-green, rather soft, and so finely crystalline that no minerals can be distinguished. Under

– 439 –

the microscope the rock is holocrystalline and porphyritic. It consists mainly of a groundmass of a number of small crystals, rounded and irregular grains of ægerine, together with an immense number of fine needles of this mineral. The ægerine is of a grass-green colour, pleochroic and very birefringent. Small needles and lathes of feldspar occur in such small proportions as to render any determination of their mineralogical properties impossible. Throughout this groundmass are numerous porphyritic, hypidiomorphic, and idiomorphic crystals of hornblende, of a brown colour, strongly pleochroic. In places these are larger, with typical hornblendic characters. In parts of the section they form nests, each nest being composed of long laths and basal sections of hornblende surrounded by smaller needles of hornblende and small masses of ægerine. These are comparable to the basic secretions found in platonic rocks.

The order of consolidation was evidently—(1) hornblende (crystals of this having been separated out, there was evidently a tendency for others to form in its vicinity); (2) ægerine; (3) feldspar.

The stages of crystallization seem to have to a great extent overlapped, the ægerine having been in some cases contemporaneously crystallized with the hornblende in the nests, and the needles of ægerine and feldspar being consolidated together.

Treatment with acid and evaporation discloses cubes of sodium-chloride under the microscope showing the presence of nepheline. The name “tinguaite” has therefore been given to the rock.

Trachytoid Phonolite.

Two varieties of this rock occur as dykes, possessing distinct characters.

No. 1.—This occurs, as shown, near the mouth of the valley containing the mine. The course of the dyke is shown by the outcrop of the rock in situ from place to place along the line shown on the map. It appears to be about 7 ft. broad, and is inclined at about 10° to the vertical. It is in a state of great decomposition, so that the question as to whether it was ever nepheline-bearing is not to be determined. Under the microscope it is holocrystalline and porphyritic. The phenocrysts consist of scattered crystals of hornblende of rather irregular outline, though sometimes rounded or elongated. Most of these are of a deep-brown colour, intensely pleochroic, while others show a dark-and light-green pleochroism. An interesting feature of the rock is the occurrence of crystals showing a gradual change from a brown-coloured core to a green margin. The interior, then, possesses a lower extinction angle than the exterior. The groundmass consists in part of granular calcite with brilliant

– 440 –

birefringence. This is derived from the decomposition of the feldspar—an intergrown mass of small plagioclase lathes and fine needles, irregular masses and grains of bright-green ægerine, showing brilliant birefringence. The feldspar is partially decomposed to calcite, and in places to calcite and kaolin, and is twinned polysynthetically on the albite law. It is of the variety oligoclase. No evident flow structure is present. Magnetite occurs in small rounded grains throughout. Though nepheline does not occur in the rock, it seems probable that when the rock was fresher it was present in small crystals; therefore the name “trachytoid phonolite” has been given to it.

No. 2 (Plate X, fig.4) occurs in the valley just above the minemouth. It is not to be seen in situ, but its course is marked by a line of surface boulders extending as shown on the plan. It seems to have formed a dyke along this line. It is holocrystalline and porphyritic. The phenocrysts consist of large square and rectangular plates of orthoclase and large columnar crystals of oligoclase very finely twinned (maximum extinction on albite plane = 10°). Calcite frequently occurs in granular masses, often taking the orientation, both externally and internally, of the finely twinned oligoclase, a fact apparent between crossed nicols. The groundmass consists of an aggregate of plagioclase laths and needles and small irregular masses of green ægerine. The feldspars show a pronounced trachytic structure, and are twinned on the albite law. Treatment of the powdered rock with hydrochloric acid, on evaporation on a microscope-slide, shows under the high power small cubes of sodium-chloride, thus proving the existence of nepheline in very small crystals. On account of this the name “trachytoid phonolite” has been given to the rock.

Basalt. (Plate X, fig. 3.)

As mentioned under “General Geology,” this rock forms the top of the mountain, and apparently extends down for an unknown depth, filling an old volcanic pipe. The rock in hand specimens is seen to be very undecomposed; it is coarsely crystalline, and phenocrysts of feldspar and augite can be distinguished, the latter often of large size. Olivine is also distinguishable, but in far less quantity than distinguishes the basalt family as a whole. Under the microscope the rock is seen to be composed chiefly of mass of feldspar laths twinned on the albite law. They are all plagioclase, with a maximum extinction of 32½°, thus being a variety of labradorite of medium composition. These laths show a general flow structure, and wrap round numerous small hypidiomorphic crystals of augite. These laths are of a clear pale variety, showing the typical non-pleochroic and highly birefringent characters of that mineral.

– 441 –

In this mass of augite and plagioclase are large phenocrysts of augite, often quite idiomorphic, and showing the usual cleavage and characters. Olivine is also present throughout the rock in rounded grains and large rounded plates, but is not remarkably abundant. It is quite undecomposed, and is distinguishable from the augite with difficulty with ordinary polarised light, but under crossed nicols it shows a much higher birefringence.

A residuum of glass is present interstitially with the plagioclase laths. It is not present in great quantity, and is filled with small colourless beads. In places these unite to form colourless needles (margirites), and elsewhere to form darkcoloured rods which appear as small dark needles piercing the glass under the low-power objective.

The coarse crystallization and almost entirely holocrystalline characters of this rock ally it closely to the dolerites. The feldspar laths are much larger than are usually found in basalts and more typical of dolerites. The residiuum of glass and flow structure, however, class it with the basalts. The coarse structure is doubtless due to the fact that it has solidified in a pipe, the exposed portion now, perhaps, being originally many hundred feet deep. The flows from this pipe were no doubt much finer in their crystallization.

Under “General Geology” reference was made to a boulder found in the mine valley with peculiar petrographical characters (Plate XI, fig.5). Its mineral constituents find no parallel among any rocks that are described in any of the standard works. In hand-specimens the rock is fresh and very dark, showing phenocrysts of augite in places, and small shining facets of a dark mineral under a magnifying-glass. Under the microscope it is seen to be holocrystalline and prophyritic. The phenocrysts consist of long polysynthetically twinned plagioclase with a maximum extinction on the albite plane of 19°, thus being andesine. Rounded plates of augite are scattered through the mass, and in places these are quite idiomorphic, and show unbroken crystalline outline. They are very pale green in colour, and non-pleochroic, with high birefringence.

Enclosing these phenocrysts are a mass of irregular grains of soda amphibole (probably hastingsite). They have a greenishbrown colour, and are very pleochroic. These have needles and small laths of andesine scattered throughout them, and form two-thirds of the bulk of the rock.

The order of crystallization has evidently been—(1) augite; (2) plagioclase phenocrysts; (3) plagioclase in the ground-mass; (4) amphibole. This order of crystallization is a variation from the normal, and somewhat suggests that of the diabases, though in them the last mineral to crystallize is the augite, and horn

– 442 –

blende is rarely present. The rock seems to be most closely related to the phonolites and tinguaites, the soda amphibole taking the place of the ægerine common to those rocks, and nepheline not being represented. The acid nature of the feldspar prevents its inclusion with the diabases or Hohenbegger's teschenites, to which it has some affinities.

Its occurrence is only another example of the numerous varieties in which the crystallization of the magma beneath has resulted, and the determination of the respective formation of ægerine, soda amphibole, or nepheline is doubtless to be sought in the varying chemical composition due to be sought in the varying chemical composition due to magmatic differentiation.

The Occurrence of Gold.

An examination of the valley containing the mine leads to no sign of the auriferous rock on the surface, and no sign of pyrites or other possible gold-bearing minerals in the streams. A drive indicated on the map pierces the solid bostonite on the upper side of the valley. Two shafts are sunk lower down, about 15 ft. above the level of the stream-bed. The only one accessible was in a fair state of preservation, but, owing to water, could only be descended a distance of 20 ft., to the first cross-drive.

The following materials wall the shaft and the drive (1) About 11 ft. of loam derived from decomposition of the bostonite, and from vegetable sources; (2) about 1 ft. of very decomposed bostonite; (3) cap of decomposed syenite; (4) solid syenite; (5) the last 18 ft. of the drive is walled with friable micaceous sandstone. (See Plate XIII, section AA, through drive.)

Along the border-line of the decomposed bostonite was to be seen a band about 3 in. in thickness of hard undecomposed mineralised bostonite, mention of which is made under “Petrography.” The syenite is in almost every specimen obtained pyrites-bearing, and the mineralised bostonite is very rich in that mineral.

Assays of the samples obtained gave results as follows:—

No. 1. —Syenite showing in hand - specimens no visible pyrites (mean of three assays): Gold = 0·000066 percent., silver = 0·00033 percent.; giving, per ton of 2,240 lb.—gold = 10·45 gr., silver = 2 dwt. 1·74 gr.

No. 2.—Sample showing a few small specks of pyrites on fractured surface (mean of three assays): Gold = 0·00021 percent., silver = 0·00099 percent.; giving, per ton—gold = 1 dwt. 8·92 gr., silver = 6 dwt. 12·8 gr.

No. 3.—Sample showing plentifully specks and small flakes of pyrites on fractured surface, the richest sample in pyrites obtainable (mean of three assays): Gold = 0·0013 percent.,

– 443 –

silver = 0·0056 percent.; giving, per ton—gold=8dwt. 11·8 gr., silver = 1 oz. 16 dwt. 14·1 gr.

No. 4.—Sample of mineralised bostonite. This rock is everywhere richly impregnated with pyrites, the fractured surface showing a brass-yellow colour. Mean of two assays: Gold = 0·0022 percent., silver = 0·00051 percent.; giving, per ton—gold = 14 dwt. 8·9 gr., silver = 3 dwt. 3·8 gr.

Assay of the micaceous sandstone gave no sign of gold or silver.

From the above it will be noticed that the amount of the precious metals in the rock becomes greater as the amount of pyrites increases, and therefore it does not seem altogether illogical to infer that the gold and silver exist in the pyrites, being the so-called “sulphide gold.” The proportion of gold to silver, about 1:5, in the syenite is almost constant, but in the mineralised bostonite the amount of gold suddenly increases, and the proportion of gold to silver becomes about 4:1.

The coarse crystallization of the syenite and its occurrence in the Port Chalmers breccia render it indisputable that this rock exists as a large intrusive mass under the district, and the only question to settle is whether the precious metals and pyrites were introduced into it subsequently to its consolidation or whether they were an original constituent of the liquid magma that was forced up. As mentioned under “General Geology,” the intrusion of the syenite was probably subsequent to the outflow of the bostonite, and the former carried mechanically upwards a portion of the sandstone beds. Another possible solution for the presence of the sandstone is that the syenite was originally a surface of the crust, and on it the Tertiary sandstone was deposited, the bostonite afterwards flowing out and covering the sandstone: but this was probably not the case. Supposing a pyrites-bearing alkaline rock such as the syenite was exposed on the surface. The action of atmospheric water would have weathered the rock to a great extent, and subjected it to the well-known process of secondary enrichment. The pyrites as well as most of the gold would have been dissolved and redeposited in the deeper portion of the deposit, as has occurred in almost every exposed lode. The pyrites, however, in this deposit is unaltered at the surface; and thus it is to be safely concluded that it has never been exposed to the direct action of a large amount of atmospheric water for any length of time.

From the presence of the resorption zones in the mass it is evident that chemical action on the hornblende crystals has taken place. This action may have occurred in three ways: (1) Action on infra-telluric crystals of hornblende by the magma

– 444 –

or other agency before or during ejection; (2) action of the magma on the first-formed crystals after it had attained its present position; (3) subsequent action either of percolating water from above or below, or of water which constituted a part of the magma at a high temperature and charged with sulphur or sulphides.

The first of these is out of the question, since the movement of the mass during ejection would have detached and scattered throughout the mass the granular particles surrounding the hornblende core. It is certain that the presence of the pyrites in large masses in the rock is closely connected with its presence in small granular masses in the resorption zones, and therefore the formation of the zones was contemporaneous with the introduction of the precious metals. Taking it as probable that the mass was ejected after the outflow of the bostonite, we have two facts which have bearing on the matter: one the presence of a small zone of pyrites bearing syenite, the other the total absence of pyrites and gold and silver from the micaceous sandstone. The enrichment of the syenite and bostonite by ordinary vadose circulation is not probable, firstly on account of the absence of any possible gold-bearing bodies except the deeply buried schist, and secondly on account of the limited enrichment of the bostonite, a rock quite as porous as the syenite. There is absolutely no reason why the enrichment of this should have ceased a few inches from the junction of the rocks. The subsequent penetration of the mass by highly heated and sulphide-charged water from below, if this penetration occurred after the consolidation of the syenite, is for the same reason untenable.

It is thus to the magma or to some portion of the magma itself that we must look as the mineralising agent. Since the syenite has only been assayed in samples taken from the surface, it is impossible to say whether the rock may or may not be gold-bearing throughout its mass, and this leaves the solution of the problem to theoretical reasoning. However, we have seen that in all probability the reaction did not occur in the magma before ejection, and can conclude that the particular constituents which finally caused this action were not then concentrated enough to bring this about. When, however, the magma was forced up nearer the surface the mass would commence to solidify, and the portions of it which remained liquid longest would rise to the place of least pressure—that is, the portion nearest the surface. They would do this without difficulty in the semi-solidified viscous mass. At the surface they were, however, retained by the solid bostonite, which, however, they penetrated for a few inches, and the action causing the

– 445 –

resorption zones took place. When this action ceased, partly from the disappearance of the hornblende crystals and partly from the formation of a surrounding shell of resulting granular products which prevented further action, the remainder of this liquid solidified in its turn, forming the pyrites visible to the unaided eye. Thus the action which could not proceed in the earlier stages of consolidation was rendered possible by the collection of certain constituents of the magma in the higher portions of the intrusion.

The complete absence of the pyrites from the sandstone is most striking, but when the great porosity of this rock is considered, and the consequent free circulation of water through it, a subsequent removal of the sulphide and gold and silver seems probable. This sandstone lying along the surface of the syenite as it was carried up would be the easiest channel for vadose water to circulate, and the passage of a large body of water for a comparatively short time has effected what the limited amount passing through the comparatively impermeable syenite and bostonite has not been able to do. Owing to this very porosity it seems probable that the sandstone was once mineralised to a much wider extent than the bostonite. The unstability of gold-bearing sulphides has been illustrated in every lode which carries sulphides, where the “iron cap” is often quite poor in precious metals the deeper portions of the reef being sometimes extremely rich.

The greater richness of the mineralised bostonite may be due to a differentiation of the reacting portion of the magma, comparable to Soret's classical experiment, when the more basic substances crystallized at the colder end of the tube, the “colder end” being the lower surface of the bostonite. A similar cause may have resulted in the differentiation of the gold and silver, the bostonite being so much richer in gold.

From theoretical grounds it would appear improbable that the auriferous belt in this rock extends to any great distance below the surface. Portions of the syenite ejected by the Port Chalmers breccia are in such a weathered state that all pyrites has been removed from them even if it was ever present. A general segregation of the aqueous solution of sulphides to the upper portions of the intrusion seems more probable than a general segregation throughout the mass, since no sign of any marked small segregations is visible on the small portion of the rock exposed. This aqueous solution was strongly alkaline as well as sulphide-bearing, as its reaction with the hornblende produced ægerine.

The occurrence of sulphides and gold in this syenite is a very striking feature. Its general alkaline properties render it more

– 446 –

than probable that it is derived from the magma whence flowed the tinguaites, phonolites, basanites, and kenytes of the remainder of the Peninsula, and in these rocks so far absolutely no sign of pyrites has been discovered. A differentiation of a magma so complete as to include all the pyrites in one outflow seems improbable. The fact that all the other rocks were either ejected through dykes or volcanoes seems rather to show that the free communication with the air enabled the sulphides to escape, being either carried up mechanically with the steam present, or else being vaporised and escaping with the reduction of pressure consequent upon entrance to a surface-connected fissure. Whether the precious metals have also escaped or not is to be decided only by careful assays by fine methods on large bodies of these rocks, and theoretical considerations seem to show that traces of them, at least, will be found

The district is comparable to that of the Thames and Coromandel goldfields. There the country rock is andesite, and the existence of gold in this country rock is a disputed point. In the Otago Peninsula, however, the gold occurs comparatively richly in one rock at least ejected from the magma beneath, and the formation of auriferous lodes in this district was prevented, either on the lateral - secretion or ascension theories, only by the absence of lode-forming fissures and the circulation of a suitable gold-dissolving solution.

Explanation Of Plates IX-XII.
Plate IX.

Fig. 1. Pseudamusium huttoni, reduced size, 4:3, with profile views.

Fig 2. Microphotograph of syenite, showing resorption zone round hornblende.

Plate X.

Fig. 3. Microphotograph of basalt, crossed nicols.

Fig. 4. " phonolite, crossed nicols.

Plate XI.

Fig. 5. Microphotograph of undetermined rock, crossed nicols, showing augite and soda amphibole.

Fig. 6. Microphotograph of tinguaite.

Plate XII.

Fig. 7. Microphotograph of tinguaite, crossed nicols. a, hornblende nest; b, ægerine twin.

Fig. 8. Microphotograph of bostonite, crossed nicols

Plate XIII.

Map of Harbour Cone.