The Basal Beds of the Akaroa Volcano.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 70, 1940-41, Page 60

The Basal Beds of the Akaroa Volcano. By R. Speight. [Read before the Canterbury Branch, September 6, 1939; received by the Editor. January 8, 1940; issued separately, June, 1940.] Table of contents. A. Introductory. B. Onawe peninsula, description of, and general stratigraphy. C. Features of the general shore-line— I. To the east of Onawe, including a. Lushington Bay, with trachyte and trachyte tuffs. II. To the west of Onawe, including a. Le Petit Carenage to Tikao Bay, with trachyte rocks. D. Chemical analyses of representative rocks. E. The barry's bay—tikao bay shelf. F. Summary. A. Introductory. (See Plate 8.) Akaroa Harbour has long been regarded as a typical caldera, which has been drowned either by lowering of the land or by the rise of the sea. Its floor gradually deepens from mud-flats at its head to 18 fathoms at the entrance, a figure which gives the minimum amount of drowning to have taken place. In the upper portion spurs stretch down the inner slopes with rough centripetal orientation, once having divided stream-valleys, and still dividing them in their upper portions, while the lower parts of the spurs now divide the bays which fringe that stretch of the shore of the harbour. The most striking of these spurs is that which ends in Onawe Peninsula. Taking the bays in clockwise direction round the harbour they are as follows:—Tikao Bay, Le Petit Carenage or Brough's Bay, French Farm, Barry's Bay, Duvauchelles and Head of the Bay, Robinson's Bay, Takamatua or German Bay, Lushington Bay, and lastly French Bay on the shore of which the township of Akaroa stands (see Plate 8). The entrance to the harbour is flanked by bold, vertical cliffs, which rise on the western entrance to a height of about 500 feet; on the eastern entrance they are lower. The harbour presents some resemblance to that of Lyttelton, differing chiefly in so far as it is of later date and therefore has experienced a less prolonged erosion by stream and sea. While the crater-ring of Akaroa is broken down in only one place, viz., the harbour entrance, that of Lyttelton is broken in two, there being no counterpart in Akaroa of the low ridge near Gebbies Pass, where the earliest rocks of the area are exposed as a result of the removal by erosion of the covering of the later andesites and basalts of the Lyttelton system. In both cases the formation of the caldera may be attributed chiefly to the erosion by streams converging on a great central hollow, originally an explosion crater of moderate size, and in both cases there is something of a break in the profile of the inner slopes of the hollow. In the case of Akaroa this lies at an elevation of from 300 to 600 feet above the sea, and is most definite

in the stretch from Tikao Bay to Le Petit Carenage or Brough's Bay, and between French Farm and Barry's Bay (see plates 10, 11, 12), photos 1, 4, and 5) where the level surface suggests a shore platform. In other parts of the periphery of the harbour where the break occurs, such as between Takamatua and Akaroa (see Plate 11, photo 3), the beds do not lie flat, but there is unfortunately no clear exposure, either in natural or artificial sections, of the contact of the rock at this break in profile with the rocks forming the steeper slopes above and having the quaquaversal outward dip characteristic of a volcanic cone. In the case of Lyttelton the break just referred to is associated with the presence of volcanics and sedimentaries which form a substratum of much earlier date than the upper part of the cone, and this suggested a closer examination of the lower slopes in the case of Akaroa in order to see if there is any analogy in structure and geological history to that of Lyttelton. The basal beds of the latter were dealt with by the present author in his account of the “Geology of Gebbies Pass and Neighbourhood” (1936), and some reference was made to the possible structure of the Akaroa volcano in an earlier paper entitled “The Intrusive Rocks of Banks Peninsula” (1923). When the latter was written the author had not the advantage of a knowledge of the complexity of the early geological history of Lyttelton revealed by a subsequent and more detailed examination of the surrounding area. Perhaps the most important locality in the Akaroa area is Onawe Peninsula, therefore a full account of its special features which bear on the question at issue will be given first. B. Onawe peninsula. (See Plates 9 and 10, Photos 1 and 2.) This peninsula lies in the upper reaches of the harbour and forms the distal end of the ridge which separates Barry's Bay on the west from Duvauchelles Bay on the east. It stretches from north to south for a length of 60 chains, and is about 20 chains wide at its greatest breadth, which is near the southern end; its area is approximately 56 acres. It is pear-shaped in outline and elevation, with its highest point (384 ft.) near the distal end. Towards the proximal end, where it is joined on to the mainland, the ridge which forms the backbone of the peninsula sinks in height as well as narrows, and during high tides and in heavy weather the sea breaks across the isthmus for nearly half a chain, and the peninsula becomes a temporary island. The upper surface is covered almost completely with loess and soil, so that there is no exposure of rock except toward the southern end. Even the steep sides are masked in the same way, and the only places where its structure and the relations of the beds composing it can be seen lie on the seashore and on the face of the cliffs adjacent thereto, and to add to the obscurity, considerable lengths of the shoreline show no clear-cut exposures. All the southern or distal end of the peninsula for a distance of about 10 chains is formed from sea-level right up to the highest point of a syenite of yellowish-brown colour, somewhat vesicular or drusy in texture, and showing at times the spheroidal weathering characteristic of granite, but in many places the depth of weathering

is small, and the rock rings when struck with the hammer. The mass appears to dip to the south at angles of approx. 45°, so that the actual thickness exposed is some 400 feet. No contacts are visible, and therefore one cannot state what are its relations to the underlying or overlying beds, but the nearest solid rock is a gabbro, which is exposed on the eastern side of the peninsula for about 50 yards on the shore and the low cliffs adjacent thereto, while on the western side it is visible for about 30 yards; of course, it may really extend further under the mask of surface debris and soil, but the total width of the occurrence near the shoreline cannot reach three chains. As just mentioned no contacts show towards the syenite, and the gabbro may either have been (1) part of a mass already in position into which the syenite has been intruded, or, (2) it may have been intruded along the lower surface of the syenite itself already in position, and perhaps, if so, it has been responsible for the tilt of the latter to its present inclination, or, (3) the contact may be a fault. It is unfortunate that the only exposures are near sea-level, and all the upper part of the ridges north of the crest of syenite is completely masked with soil, so that one cannot say how far up the sides of the peninsula the gabbro extends, if, indeed, it does extend for any distance above the 20 feet or so where it is actually visible. The relations, therefore, of the granitoid rocks must be left indefinite, and in no more satisfactory condition than that expressed in my former paper on petrology (loc. cit., 1923, pp. 124–7). Assuming, however, that the law of decreasing basicity holds for plutonic rocks, it would seem very probable that the gabbro is the earlier of the two, and that the syenite had been intruded into a mass of rocks of which the gabbro was a pre-existing member. This consideration has, all the same, little diagnostic value as regards the order when it stands alone. In their paper on the Cainozoic Igneous Rocks (1939, p. 56) Benson and Turner say that this syenite probably does not belong to the Banks Peninsula series of Late Cainozoic basic rocks, but cite no special reason for the statement. An analysis kindly carried out by Mr. Seelye, of the Dominion Laboratory, no doubt gives a truer account of the chemical composition of the gabbro than that cited in my former account (1923, p. 125), and I have nothing further to add concerning the mineral composition of the rock as given therein. The eastern side of the peninsula is more exposed to the action of the sea than the western, so the section along the shore-line on that side will in general be clearer and will therefore be considered first. For a chain or so north of the gabbro the beach is masked by blocks of syenite, but this cannot be taken to imply for certain that the latter rock is in position at this spot, since the blocks may have rolled down the hillside from outcrops at a higher level. The first rock in position exposed north of the gabbro is a basic andesite or acid basalt which is visible for just over 30 yards. When this is examined under the microscope it shows some evidence of metamorphism (loc. cit., 1923, pp. 127–8), and this alteration becomes more pronounced in the direction of the gabbro, whereas some 20 yards away there is little evidence of change. If it be

conceded that the gabbro is the metamorphosing agent, then a volcanic mass must have been in position before its intrusion, and, incidentally, it must have been in position before the construction of the Akaroa volcanic cone, unless the gabbro is merely a deep-seated facies of the basic rocks of which the cone is constructed. Though this is possible it does not appear probable, for there is considerable chemical and mineralogical difference between the gabbro and the Akaroa suite of rocks, though this is not a fatal objection and they may in fact date from the same epoch. The unmetamorphosed facies resembles closely the basic andesites that occur in places round the shore of the upper part of the harbour, and its composition is given by Analysis No. 2, p. 71, the result differing materially from that cited in my former account (1923, p. 125), and differing in some particulars from the analyses of ordinary Akaroa rocks from higher levels given in my article on the “Basic Rocks of Banks Peninsula” (1924, p. 260). The next exposure consists of a trachytic tuff, light in colour, which continues along the shore-platform for nearly 9 chains, and may extend further still along the southern wing of a shallow indentation on the shore-line, but does not show in position. This tuff is well stratified, fine-grained in texture where it outcrops, and with occasional small fragments of rhyolite; one piece of greywacke was noted. Its chemical composition is given in Analysis No. 3, p. 71, and this shows it to be an ordinary trachyte; it is remarkable how little it shows the effect of weathering; the small percentage of free chlorine noted by the analyst can easily be explained by the proximity of the rock to the sea. Near its southern boundary it dips to the east at an angle of 20 degrees, but further north it has a variable inclination at low angles, generally to the north, but occasionally to the west. Its relations to the basic rock just mentioned are obscure, but I think it possible that the tuff dips under it, though I formerly expressed an opinion to the contrary (1936, p. 127); however, at one spot about a chain from its southern boundary it is apparently covered on the extreme edge of the shore-platform by a somewhat acid dolerite, and about 7 chains from its southern end by a fine-grained basic rock, either an acid basalt or an augite andesite. This last no doubt extends along the side of the hill above the indentation of the shore-line just mentioned, and at one point in a shallow cave or overhang there is an occurrence of a coarse agglomerate including masses of volcanic rock up to 18 inches in diameter, among which are fragments of trachyte, augite-andesite, and in one case the latter contains amygdaloids of chalcedony. This last andesite is a type I have not encountered in the cone rock of Akaroa, and it no doubt belongs to a basement possibly related to the andesites of McQueen's Valley in the Lyttelton area, which contain masses of chalcedonic quartz (Speight, 1935), and this suggests the possible extension of these pre-Cretaceous volcanics into the Akaroa area. Fine-grained andesite, very deeply weathered (Analysis No. 4, p. 71), continues for several chains toward the neck, lying fairly flat, and with a scoriaceous layer underneath. This may extend

right to the neck, or another layer of basic rock may come in at a slightly higher horizon, with yet another layer of scoriaceous material interbedded. In section this rock appears to be composed of a felt of oligoclase laths, and small dark denticulate forms of iron ore, perhaps derived from augite laths, and in this base are very occasional slightly larger forms of andesine, with numerous rounded grains of quartz, and occasional small amygdaloids of chalcedony; there are as well numerous areas stained with brown iron ore, the source of which is uncertain; masses of iron ore also occur. Although this rock is much weathered, the feldspar laths of the base show little sign of alteration. Near the neck the beds dip to the south, and between the adjacent road cutting and the wharf the dip changes to the north, and basalt or basic andesite overlies the exposed scoriaceous layer. A similar dip involving solid flows and scoriaceous layers continues along the shore-line to the north of the wharf in the direction of the head of the bay. A reference to one of these beds with analysis is given by the present author (loc. cit., 1924, pp. 260–1); the rock is a basalt on the borderline of augite andesite. On the western side of the peninsula a similar sequence occurs, but is not so clearly exposed. North of the gabbro no clear section can be seen for over a chain, but a whitish tuff then appears dipping south under a basic tuff between it and the gabbro. This whitish tuff is similar to the trachyte tuff on the eastern side of the peninsula, and is in an analogous position. It continues for several chains along the shore to a point where a somewhat coarse basalt with large olivines and titaniferous augite forms the lowest bed to be seen. This is succeeded, as on the eastern side of the peninsula, by a flat-lying basalt, with fragmentary material just visible in one place. At the neck the scoriaceous layers dip to the south, but nearer the road they swing round, the pattern being clearly developed on the broad shore-platform (see Photo. No. 2) cut in these incompetent beds. This shows them to be dipping first of all to the west, and then the inclination becomes irregular and probably dominantly north, as on the eastern side of the peninsula. The only exposure of basalt, apart from those occurring on the shore-line, that I have observed lies on the northern side of the gully just north of the syenite on the western facing of the ridge; this seems to indicate that the main mass of the peninsula north of the syenite consists of basalt or basic andesite. All these beds, including the syenite and gabbro, are cut by dykes, generally trachytic, but occasionally basaltic, so that they post-date generally, if not entirely, both the plutonic and the volcanics of the peninsula. Near the neck there is a criss-cross of dykes, variously oriented, which form a striking pattern on the broad shore-platform there developed. Some of these are the white variety of trachyte, practically a pure feldspar rock, which occurs in analogous position near the centre of the Lyttelton caldera; this facies has not been observed on the outskirts of either cone. It might be noted that some of the projections on the shore-line are determined by the

resistant qualities of massive trachyte dykes which cross the peninsula. As these intrusions, interesting though they are, have little apparent bearing on the question under discussion, nothing further will be said concerning them. C. Features of the general shore-line. The shore-line round the upper part of the harbour was examined in order to see if it furnished evidence that might throw light on the problem. The parts specially investigated were the points that stretch out finger-wise towards the middle of the harbour, notably those between Duvauchelles and Robinson's bays, between the latter and Takamatua, and between the last-named and Akaroa, all of which lie to the east of Onawe, while to the west all the shore-line to Tikao Bay was examined. 1. To the East of Onawe. On the point that occurs in the middle of Duvauchelles Bay the rock has been quarried to widen the road, and also for metal, but the stone is much decomposed, and weathering has penetrated to the depth of the excavation. The excessive depth of decomposition suggests that the rock may belong to an earlier episode than that associated with the construction of the main mass of the cone, but it cannot be taken as definite proof that this is the case. All the same many of these lower beds show the same character, and as it is of wide incidence it may have some diagnostic value. On the other hand, some of the rocks near sea-level show no marked amount of decay. In this case the rock is an olivine-andesite or acid basalt; the base is composed of oligoclase laths and augite grains, and there is some olivine very much decayed and also chalcedonic amygdaloids. A stain of brownish iron ore masked a large part of the section. On the southern side of the point between this and Robinson's Bay the lowest bed exposed appears to be a basic andesite, while above it lies a layer of scoriaceous material and then a definite basalt containing numerous olivine crystals as shown in thin section. Some of these beds have a slight dip towards the centre of the harbour, and some in the contrary direction. On the north side of the point the lowest bed exposed is in all probability a whitish, very much decomposed tuff, apparently bent up into an anticline and flanked with basalt. Whether this whitish bed is the stratigraphical equivalent of the trachyte tuff on Onawe or a very deeply weathered and bleached basaltic tuff cannot be said for certain, but it is overlain in places by reddish basaltic tuff under similar conditions of exposure to weathering agents, so it may be the representative of the trachytic tuff in this locality. On the point between Robinson's Bay and Takamatua there is some interbedding of flattish lying lava flows and very decomposed tuffs, both with wavy inclination, and with occasional exposures which suggest faulting, but is must be stressed that the lateral discordance, which in sedimentaries postulates faulting, cannot be considered conclusive in the case of volcanics, unless there is some definite evidence of movement, such as slickensiding, development of pug, etc., and these were not seen. On the southern side of the

point there is a considerable extent of the whitish, decomposed tuff analogous in character and position with that in a similar situation on the eastern side of the point between Duvauchelles and Robinson's Bay. a. Lushington Bay. (See Plate 11, Photo. No. 3). The next point following round the harbour, viz., that between Takamatua and Akaroa, which rises to 682 feet in the hill known as Otipua, presents one of the most interesting developments of trachytic rocks occurring in the Akaroa area. They are clearly seen on the shore-platform and cliffs forming the two headlands on either side of Lushington Bay. On following the shore past the old wharf at Takamatua, there is first of all considerable obscurity, but the first beds encountered in position are basalts and decomposed basic tuffs, striking north-east to south-west in places and dipping north-west. Then the trachyte beds show and continue right to the first point and on round the shore of the bay to its southern side. At the northern end the beds consist of fine and coarse-grained breccias, the size of the fragments reaching eight inches in diameter, but are generally less than three inches; they consist chiefly of trachyte, generally whitish in colour, occasionally pinkish, but there is a notable content of pinkish rhyolite, with flow lines, bands of micro-spherulites and micro-spherulitic groundmass in places; numerous microlites of sanidine also occur in the base. These fragmentaries are capped with massive trachyte, which forms precipitous cliffs up to 50 feet in height, and this is succeeded by an olivine-andesite or very acid basalt—the phenocrysts are an acid labradorite, and occasional olivine, while the base is composed chiefly of laths of oligoclase. The strike of the trachyte on the point is approximately north–south, and the dip to the east at angles of about 15°, but more to the north in the direction of the old Takamatua wharf the dip is to the west, and the same may be said in connection with the southern side of the point, so that the trachytes are involved in deformational movements along a north–south axis, and the immediately overlying basalts appear to be involved in them as well, judging from exposures on both sides of the point. On the end of the spur which divides Lushington Bay into two minor indentations, trachytic breccia and solid trachyte also occur, and they are clearly developed again on the southern point of the bay, where the breccia is well stratified, strikes north–south and dips east at angles between 10° and 20°. It consists almost entirely of trachytic material with blocks up to 12 inches in diameter, and is capped by basic tuffs and than by an olivine-andesite or very acid basalt exactly like that referred to a little earlier as lying over the tuff near the northern horn of the bay. In the lower parts the tuffs include layers of material dominantly trachytic interbedded with basic material. The trachytic content of these layers gradually lessens as they are traced upward, and then the deposit passes into a pure basic tuff, which has the same strike and dip as the trachytic beds. Higher up the hill over the olivine-andesite normal basalts appear, and as far as can be seen they continue right to the summit, with the same strike and a dip to the east, which is what one would

expect if they formed an integral part of the Akaroa cone in this locality, and they should without doubt be assigned to it, although Otipua Hill (see Plate 11, Photo. 3.) forms a notable break in the internal profile of the slopes on the eastern side of the harbour. The usual assemblage of variously oriented trachytic and basaltic dykes penetrating all rocks occurs in this locality, and shows typically along the shore-platform round Lushington Bay. 2. To the West of Onawe (See Plates 11 and 12, Photos 4 and 5). To the west of Onawe the exposures along the shore-line are obscure and discontinuous since this part of the harbour is somewhat sheltered from active erosion by the sea. However, between Le Petit Carenage (Brough's Bay) and Tikao Bay the sea has operated strongly and a well-defined shore-platform backed by a marine cliff has been cut. On the point between Barry's Bay and French Farm the beds have a flattish dip towards Onawe, but at the road-metal quarry near French Farm they dip north at moderate angles, a direction which can hardly be said to be directly away from Onawe, as they should have were the centre of vulcanicity to lie in the vicinity of the peninsula. From the shore-line the hillside rises steeply to a flat shelf (see Plate 11, Photo. 4), and on this a normal basalt is exposed, which differs from the rock of the quarry, but it cannot be told from such solitary outcrops what the main mass of the shelf is composed of. The quarry rock shows two facies, the first, fine grained and composed of a felt of narrow oligoclase laths, augite grains, and much ilmenite in denticulate margined laths, through which are distributed very sparingly small phenocrysts of labradorite, and the whole is clouded by an alteration product resembling calcite. This facies is apparently andesitic, while the second is doleritic with large crystals of labradorite occasional serpentinised olivine; a purple titan-augite in a somewhat coarse holocrystalline matrix composed of feldspar, some of which is labradorite, but in other cases untwinned and with an index of refraction lower than that of balsam; titanaugite, like the phenocrysts; much ilmenite in grains and laths with broken-comb margins; a great amount of calcite or other carbonate occurs in patches wedged in between the feldspar and augite. The carbonate appears to be secondary as there is some amount of alteration of the feldspars. These two types differ from any I have come across in the Akaroa area, and it is unfortunate that the disturbing effect of the carbonate somewhat detracts from the value of the analysis as giving the true composition of the unaltered rock (see Analysis No. 5, p. 71). The analyst notes that the carbonate is not present as pure calcite since the powdered rock does not effervesce readily with acid in the cold. The presence of this carbonate also influences the value of the norm, since the plagioclase indicated therein is entirely albite, a conclusion not borne out by the microscopical examination. A pure albite feldspar seems to occur in general in plutonic and dyke rocks, but not as a rule in flows

except in the Keratophyres and those types related thereto and called by Johannsen Sodaclase andesites (1937, pp. 137–8). There does not appear also to be any evidence of albitisation as in the spilites. But the whole circumstances merit further examination. It must be noted, however, that these rocks differ from any yet encountered in the Akaroa area, and they differ markedly from the ordinary andesites and basalts of the cone; also, they are associated with the Barry's Bay-French Farm shelf referred to earlier. On the opposite side of the bay from the French Farm quarry there is another development of trachyte tuff or breccia exposed on the beach between tide-marks or adjacent thereto. Like the occurrence on Onawe it apparently underlies a basalt dipping towards the centre of the harbour; what it rests on cannot be determined, although there is some indication that it rests on a more basic facies. When followed towards the head of French Farm Bay, a coarse basalt or dolerite is exposed, and then lying on the beach is an assemblage of boulders up to two feet in diameter, chiefly of dolerite, but with occasional trachyte, like that of the dykes penetrating the syenite and gabbro on Onawe, and very coarse in texture; these are probably from a dyke. Round the point towards Le Petit Carenage occurs a coarse dolerite, in fact, all the rocks sectioned from this stretch of coast, with the exception of the trachyte, were doleritic in character. Owing to the cover of soil and slip material it was not found possible to determine the nature of the rock forming the seaward slope of the ridge and the narrow platform above it extending from French Farm to the Carenage, but it is probably of basalt or dolerite. a. Le Petit Carenage (Brough's Bay) and Tikao Bay. (See Plate 12, Photo. 5). Immediately behind the shore-line the land rises steeply to a shelf with a general height of approximately 350 to 400 feet above sea-level, and accordant with the similar but less extensive shelf between Barry's Bay and French Farm (see Photo. 1). Between Le Petit Carenage (Brough's Bay) and Tikao Bay it increases in width since the spur from which it has presumably been cut is broader and stretches further out into the harbour. Exposures of rock in situ are rare except along the sea-front, where there is an interesting development having an important bearing on the question at issue. Along the southern shore of Le Petit Carenage a trachyte of somewhat peculiar character is exposed. At sea-level it dips approximately N.W. at moderate angles, and it weathers whitish, suggesting that it is soft, but the fresh rock is very hard, with glistening surfaces when recently fractured, and dark grey in colour with a greenish tint when seen in mass on the shore platform. In this section it shows occasional micro-phenocrysts of sanidine and rare plagioclase feldspars and also probable anorthoclase; numerous greenish, pleochroic aegirine-augite; in a base composed essentially of twinned and untwinned feldspar, the

former irregular and flaky in shape, and the latter in lath-like form, both with index of refraction lower than that of balsam, and the twinned forms with the extinction angle of albite-oligoclase; many of the laths have the denticulate margin and extinction of anorthoclase. Included in the base are numerous grains of augite coloured like the phenocrysts, and grains of titaniferous magnetite Analysis No. 6, p. 71, which shows it to be a trachyte decidedly Analysis No. 6, p., which shows it to be a trachyte decidedly sodic in character. I did not recognise any carbonate as occurring in the section. Rock of this type occurs over a fairly wide area of the Carenage-Tikao shelf, for it outcrops round the head of the tributary valley on the south leading down to the Carenage; it is developed along a considerable length of the shore-platform round the south-eastern corner of the shelf, where it shows some variation; it occurrs in the typical facies near the top of the shelf overlooking the shore-line on the east, and also at the base of the slope facing south near the beach at the head of Tikao Bay. Between the two occurrences on the eastern shore another trachytoid rock appears. At its northern end it has a somewhat flaky appearance, and it dips N.W. at moderate angles. This facies continues for some 12 chains, when it changes in colour from a whitish to a pink or red, and where this is washed by the sea it has all the external appearance of a syenite; this facies continues for about 6 chains, and forms the north-western corner of the promontory between the two bays. Its relations to the aegirine-augite trachyte just mentioned are not clear, but I think it underlies this rock, an opinion given with much reserve. In micro-section the two facies of the rock present little variation except in coarseness of grain. The phenocrysts appear to be sanidine up to 2 mm. in length, with probable anorthoclase, and subordinate plagioclase; quartz also occurs occasionally. The matrix is of trachytic texture, in some cases moderately fine in grain, but still coarse for a normal trachyte, while in other cases it is very coarse with quadrate and stumpy laths up to 5 mm. in length, many of the laths showing the characters of anorthoclase. While the base shows little signs of decay, there are many aggregations and flecks of iron-ore, some of which is derived from small and irregular laths of aegirine-augite. A few grains of titaniferous magnetite or ilmenite also appear, and there are accumulations of leucoxene derived therefrom; some of the iron-ore may be derived from the former. Aggregations and masses of calcite up to 1.5 mm. in diameter also appear, and the rock reacts strongly to acids in places, and specially so with the coarser types, the finer giving little or no response; the source of this calcite was not discovered, but it may have been derived from the plagioclase feldspars. The analyst notes in calculating the norm that in addition to the calcite a considerable quantity of carbonate is present as MgCO3 and FeCO3, so these minerals may be present as well. The composition of the rock as a whole is given by Analysis No. 7, p. 71.

The question arises as to the name to be given to this rock, and it seems to me that it might be called a trachyte-prophyry or quartz-bostonite as defined by Johannsen in his “Descriptive Petrography of Ingneous Rocks” (1937, vol. 3, pp. 26–7). The rock is too coarse for a typical trachyte, but the rocks called by the name suggested are dykes, and the occurrence under consideration is apparently not one, but either a massive flow or the margin of a plutonic body. It is visible only near the shore-line, so that its field relations are difficult to determine, but it has evidently a close genetic connection with the aegirine-augite trachyte just referred to. The presence of so much carbonate in the prophyry as well as the lower percentage of silica, differentiates them, the latter feature being somewhat remarkable since the porphyry contains some free quartz. In addition to these major occurrences trachytes outcrop in at least three places on the summit of, the shelf, the first near its highest point, the second on the south-east corner, and the third near its western margin. The former two of these outcrops are certainly not dykes since they are exposed over too wide an area, but there may be some doubt concerning the last, a reference to which will be made later. Trachyte rocks also fringe the northern shore, and also the western half of the southern shore of Tikao Bay, but I have made only a cursory examination of parts of the former. The rocks I have seen consist of flows, tuffs, and coarse fragmentaries, the last containing blocks of trachyte as well as basic material. The trachytes, including the coarse fragmentaries, are exposed on the southern side of the bay, and extend as far as small beach; the rocks at the other end of the beach are basalt or basic andesite, and such probably extend towards Wainui. Crossing the head of Tikao Bay is a massive trachyte, over 30 feet thick, striking approximately north–south, and dipping west at an angle of about 35°. At first I considered this occurrence to be one of the later trachyte dykes, and this may be correct after all, but I am inclined to think that it may be a flow with a dip in agreement with that of the other trachytes exposed along the margin of the bay. It is oriented in such a direction that it might coincide with the mass referred to earlier as being exposed towards the western border of the summit of the shelf, and a similar rock is exposed on the south side of the head of a small valley leading down to the Carenage. These facts fit the conclusion that it is either a titled flow or a dyke. All the trachytoid rocks of the shelf are penetrated by a later series of trachyte dykes, and by occasional basalts, all oriented towards Onawe, and it is possible that this major occurrence as well as some of the minor outcrops belong to these later intrusions. The wide spacing of the major outcrops on the summit of the shelf and the continuity of the trachytoid rocks round the shore-line from the Carenage to the head of Tikao Bay and a little beyond indicate that the main mass of the shelf above sea-level is constructed of these rocks, and that it constitutes the greatest development of such rocks within the Akaroa area. Their presence materially

strengthens the contention that the cone of the Akaroa volcano has been built on a basement which includes volcanics of an earlier date, and of a different petrological facies. Certainly some doubt may be expressed in this particular in connection with the Carenage-Tikao trachytes since basic rocks have not been observed to overlie them definitely, but these occur at a higher level on the interior slopes of the caldera in the immediate background of the shelf, and it would not be reasonable to conclude that the trachytes represent a separate and subsequent development on the floor of an already formed caldera, especially since related rocks are clearly overlain by basic flows at French Farm, Onawe, and Lushington Bay. These are older than the Akaroa basic rocks, and it is therefore almost certain that the Carenage-Tikao trachytes are also older. Trachyte dykes penetrate both series, but I do not think that any close relationship can be inferred as existing between the later trachyte dykes and the older trachyte flows and clastics. It only means that a reversal to an earlier condition of alkalinity has occurred either accompanying the extrusion of basic material or more probably marking an end-phase of these extrusions. These remarks as to the relationship of the trachyte dykes to the older trachytes also apply to the occurrences elsewhere round the harbour. Also it must be noted that in no locality is there any clear-cut exposure of the basement on which these older trachytes lie. D. Analyses of representative rocks. The following are the results of analyses carried out in the Dominion Laboratory by F. T. Seelye. The substance of the analyst's comments is also given. 1. 2. 3. 4. 5. 6. 7. SiO2 42.41 47.90 63.88 52.09 44.97 60.90 56.49 Al2O3 16.92 18.58 19.01 17.03 14.63 16.66 16.33 Fe2O3 3.56 4.19 1.90 12.61 2.29 2.47 2.98 FeO 8.99 8.88 0.45 0.15 9.82 4.02 4.23 (1) FeS 1.29 FeS2 0.94 MgO 4.79 1.13 0.36 0.44 3.63 0.49 0.76 CaO 10.16 5.74 0.40 3.87 8.05 2.23 2.67 Na2O 2.81 3.84 5.81 5.05 3.55 6.54 5.90 K2O 0.49 1.52 4.93 2.36 1.21 4.63 3.86 H2O+ 0.76 1.06 1.48 1.60 0.60 0.40 0.68 H2O– 0.48 0.36 0.76 1.41 0.55 0.10 0.97 CO2 2.99 1.15 nt. fd. trace 6.59 0.71 4.00 TiO2 4.15 3.79 0.35 2.01 3.25 0.48 0.72 P2O5 0.27 0.77 0.16 1.22 0.63 0.16 0.27 ZrO3 nt. fd. nt. fd. 0.08 0.01 nt. fd. nt. fd. nt. fd. S 0.05 0.04 0.17 0.03 0.03 MnO 0.13 0.19 0.01 0.16 0.17 0.16 0.16 BaO 0.04 0.06 0.06 0.08 0.07 0.19 0.15 (2) SrO 0.03 0.06 0.002 0.05 0.03 0.027 0.03 Cr2O3 nt. fd. nt. fd. nt. fd. nt. fd. nt. fd. nt. fd. nt. fd. V2O3 0.05 0.02 nt. fd. nt. fd. 0.02 nt. fd. trace NiO trace nt. fd. nt. fd. nt. fd. nt. fd. nt. fd. nt. fd. (3) CI trace trace 0.37 trace trace trace 0.03 100.32 100.18 100.06 100.18 100.23 100.20 100.26

Note (1). The sulphide in No. 1 is definitely magnetic, and is attacked by acids other than nitric with evolution of H2S. It is therefore reported as pyrrhotite; the sulphide in No. 2, however, appears to be ordinary pyrites. Note (2) The amounts of SrO were determined by S. H. Wilson spectrographically in the lime precipitates obtained in the course of the analyses In Nos. 1 to 5 (S. H. W. 34: 84) and in Nos. 6 and 7 (S. H. W. 34: 97). Note (3) Practically the whole of the chlorine in sample No. 3 is in water-soluble form, which might suggest the influence of sea-spray. (Note by R. S. that all these rocks are in close proximity to the sea.) Symbols. Assuming that any carbonates are secondary, and rejecting these from the norms the rocks would have the following symbols:— No. 1. Gabbro, Onawe Peninsula, east side—II. (III). 5. ″4 (4)5.—Hessose. No. 2. Contact Rock, Onawe Peninsula, east side—II. (4)5. ″3. 4.—Andose. No. 3. Trachyte Tuff, Onawe Peninsula, east side—I 4(5). 1. (3)4.—Kallerudose. No. 4. Andesite, Onawe Peninsula, east side—II (4)5. 2. 4.—Akerose. No. 5. Dolerite, French Farm Quarry—II. (III). 4. 1. 4″.—Pantellerose. No. 6. Trachyte, Le Petit Carenage, west side Akaroa Harbour—(I) II 5. 1. ″4.—Umptekose. No. 7. Trachyte or Trachyte Porphyry, point between Le Petit Carenage and Tikao Bay, west side Akaroa Harbour—I″. ″5. 1. 4.—Nordmarkose. If the carbonates are included in the norms as femic minerals, the symbols require some slight alteration. These have been given by the analyst, but not included herewith. Also, if the carbonates are secondary, the normative compositions of the original rocks may be possibly approached by eliminating the carbonates when recalculating them, also a similar adjustment may be made by eliminating the soluble NaCl from No. 3. It may be of interest to note that although the powdered samples of Nos. 1 and 7 effervesce readily in the cold with HCl, samples Nos. 2 and 5 do not, indicating the possibility of the presence of MgCO3 and FeCO3, or of compounds of these with CaCO3. In the norm of No. 7 the analyst has definitely included all these carbonates as required by the results of the analysis. If the carbonates are rejected the symbols are the same as given. It should be noted that the authors of the C.I.P.W. system stress the point that the presence of carbonates, in cases where they are secondary, exerts a disturbing effect on the calculation of the norm, and therefore such results should be taken with reserve. In the cases under consideration some of the carbonates are secondary, though it is possible that some are primary, and I cannot discriminate between them.

Plate 8.—This gives the position of localities referred to in the text, as well as the general position of the syenite and trachytoid rocks referred to therein.

Plate. 9—This is a sketch map of Onawe Peninsula, showing the position of various outcrops along the shore-line referred to the text, as well as some of the more important dykes, but no attempt has been made to indicate all of these as the scale would not allow.

Plate 10, Photo 1 —General view of Onawe Peninsula and west side of the upper harbour from near the road over the Duvauchelles-Robinson's Bay hill. The shelt extending from Barry's Bay to Tikao Bay can be seen in the middle distance. Plate 10, Photo 2.—View of the west side of Onawe taken from the shore of Barry's Bay and looking south. The pattern of folded basic cuffs can be seen on the shore-platform in the foreground; the neck almost cutting off the peninsula is on the left middle distance; the top of the rounded hill in the distance is of syente. The shelf between French Farm and Tikao Bay can just be seen in the right distance.

Plate 11, Photo 3 View from the highest point of Onawe Peninsula, looking east: Otipma Hill in the middle distance. Trachyte beds occur on the points near sea-level in the middle of the picture. Plate 11, Photo 4 View of shelf between Barry's Bay and French Farm. That between French Farm and Le Petit Carenage is seen on the left.

Plate 12, Photo 5—View of shelf between French Farm and Tikao Bay taken from the summit of Onawe. Le Petit Carenage is the opening in the middle of the picture. Trachyte tuff occurs at sea-level on the right, while the most interesting occurrence of trachytic rocks occurs along the shore of the promontory between the Carenage and Tikao Bay on the left of the picture.

E. The barry's bay–tikao bay shelf. (See Plates 10, 11, 12, Photos 1, 3, 4, and 5.) Another point to be considered in this connection concerns the mode of formation of the shelf between Barry's Bay and Tikao Bay, which has been referred to earlier. It is not maintained that it is continuous between these two bays since it is very indefinitely developed in the middle section on the spur between French Farm and Le Petit Carenage. Above a somewhat narrow terrace (see Photo. No. 5) this spur rises in irregular steps to fall again in elevation in the rear where it has been attacked by the headward érosion of tributary streams operating at right angles to the spur from both flanks. The Barry's Bay section may be only another of the shelves seen at various levels on the inside of the caldera, which appear to be determined by flat-lying layers of harder rock, and are therefore accidental in origin, although it must be pointed out that it is partly composed of rocks of a type not otherwise recorded from the area. The Carenage-Tikao shelf cannot be explained on the hypothesis that its surface is determined by layers of harder rock, for it apparently truncates the edges of tilted flows differing from the general run of the flows of the cone, or it cuts across the edges of dykes. Although in photographs and when seen from a distance its summit appears flat, it is not really flat. It has been attacked by streams from both flanks, and at its proximal (i.e., western) end the level sinks somewhat, while the remaining portion which extends seaward is crossed by somewhat subdued ridges, determined by harder strata some of which do not lie flat. It cannot be maintained that this striking landscape feature is accidental, so three hypotheses are advanced to account for it; these are as follows:— 1. It is part of an old shore-platform, formed when the sea stood at a higher level. 2. It represents an old surface of stream planation, formed when the caldera was being eroded and enlarged by stream action, posterior to the dying out of volcanic action. 3. It is part of the old surface on which the cone was built and may represent a surface of planation, resulting from either stream, or marine action, analogous to those “stripped surfaces” described by J. A. Thomson and others which occur in various parts of Canterbury and elsewhere in New Zealand. Against the first hypothesis it may be urged that there are no other remnants at this level round the harbour where one would expect them—with the possible exception of the Barry's Bay-French Farm shelf—and there is no other evidence of the former presence of the sea at this level, and no remnant of an old sea-cliff at the back of the shelf. The second hypothesis would require such a great amount of excavation by streams since the close of the volcanic episode, and would also apparently demand that the tributary valleys, such as those heading Barry's Bay, Duvauchelles

Bay, or Robinson's Bay, should show a double slope in cross-section, an upper one consequent on the cycle which produced the terrace, and another consequent on the lowering of the floor of the harbour, which has left the shelf now high above base-level. Finally, the last suggestion would be reasonable if one could definitely demonstrate that the beds which were planed to form the shelf all belonged to an earlier episode. Although it is probable that this is the case, it cannot be taken for granted in view of the general absence of outcrops, and the uncertain relationship of the beds that are visible. If this could be proved I should be inclined to favour the last hypothesis. Thus it is that where the break in the internal profile of the caldera is most definite, it cannot be proved to be associated in its entirety with an earlier volcanic series. Summary. The foregoing gives an account of the field and other relations of the beds which may be concerned with this problem, and the following is a summary of the evidence in favour of there being an igneous substratum exposed to view on which the cone of Akaroa has been constructed. 1. There is first of all the presence of plutonic rocks on Onawe Peninsula. These are not definite proof of the existence of an older igneous substratum, since it is possible that the syenite may be connected in some way with the magma which produced the trachyte of the dykes, especially as there is some chemical affinity between the two. All the same, trachyte dykes do penetrate the syenite and in individual cases must therefore be of later date. The syenite may, however, be more closely related to the earlier trachyte flows and tuffs, and it would then indicate the presence of an igneous foundation to the cone. Also, there is little doubt that the gabbro has metamorphosed some of the basaltic volcanics and, if so, it proves that pre-gabbro volcanics were in position before its intrusion, and even if the gabbro be regarded as the deep-seated equivalent of some of the basic rocks of the cone, it is almost certain that the metamorphosed rocks belong to an earlier epoch. 2. The next circumstance of importance is the occurrence of trachytoid rocks over a wide area in the middle of the caldera (see Plate 8). These rocks, whether they occur as flows, or as tuffs, or as the manifestations between the Carenage and Tikao, are entirely different from the flows of the cone, which are basic in composition throughout, and they certainly suggest an earlier volcanic episode when the rocks were different. It must be noted, however, that rocks of varied composition have been known to be extruded from the same volcanic focus at approximately the same time. The trachyte tuff is well-bedded and has an inclination which changes rapidly in direction and amount, the angles ranging up to 20°. This may be taken as evidence of folding, especially as the overlying basic rocks show a sympathetic distortion. So the waviness in angle is not due to the deposition of material from the air or a pre-existing irregular surface. There does not appear

to be any indication that the present disposition of these rocks is due to deposition on the external surface of a volcanic cone, and certainly not of one with the form of the recent Akaroa volcano. It must be mentioned, too, that while the evidence of the earlier extrusion of the series between the Carenage and Tikao is not convincing, except in so far as it is the lowest rock exposed at the base of the hillside, the evidence in connection with the occurrences at French Farm, Onawe, and Lushington Bay is conclusive. 3. The basalts which lie near the shore are inclined at very low angles, and have a wavy inclination extending over a considerable area, an arrangement which differs materially from that at higher levels. It is probable that some change in their attitude has taken place since they were laid down, since it is conceivable that they could have advanced over an irregular surface with such a small general inclination as that disclosed in the sections on the shore-platform round the harbour. The fact that they are dipping towards the centre of the harbour and not away from it certainly negatives the suggestion that they were poured out from a centre lying in the middle of the harbour, say, just south of the point of Onawe. Thus they experienced deformational movements which did not affect the main mass of the cone, i.e., they were antecedent to its formation. It should be mentioned that these lavas are interstratified with scoriaceous beds and are extrusive and not intrusive. 4. Many of the basalts round the shore are very deeply weathered. This may, of course, be due to the fact that they are merely the older beds of the general Akaroa sequence and do not belong necessarily to a separate and earlier episode; but the Akaroa cone rocks rarely, if ever, shows signs of deep decay. There does not appear to be any specially significant mineralogical content of the shore-line rocks as disclosed by the analyses which might account for their decayed condition, and one must naturally conclude that it is to be attributed to their more prolonged exposure to weathering agents than the ordinary rocks of the cone. This argument is of little value if it stands alone, but the condition of the rocks is such as might be expected if the first three arguments are valid. All the same it must be admitted that rocks do occur near the shore-line which are comparatively fresh and unaltered. No special feature of acidity or basicity can be cited which serves to distinguish the rocks at higher level on the inside of the caldera from those at a lower level, for andesitic as well as basaltic rocks are interstratified in both cases, but it should be mentioned that those from near sea-level which have been analysed show a dominant amount of albite in the norm, and this prompts the suggestion that they may represent a gradual passage from the trachytes with high soda content forming the base of the visible sequence into the basic andesites and basalts at higher levels, which also have a slight preponderance of albite in the norm (see 1924, p. 260). The facts just listed are those on which a conclusion must be based, and I think they are sufficiently definite to establish a prima facie case for consideration that an older and distinct substratum of

rocks is visible in places along the lower slopes of the inside of the caldera, belonging to a separate and earlier epoch of vulcanicity, but the break in the internal profile of the caldera is perhaps not dependent on the presence of this earlier series. It may be so nevertheless. When this investigation was commenced it was regarded as largely stratigraphical, but it has become dominantly petrological, and as such will be regarded as very incomplete, and as indicative of work yet to be done. Although a definite conclusion as regards the influence of the stratigraphy on the internal form of the caldera has hardly been reached, yet I think the investigation has indicated further directions of research, and it may also serve to emphasise the historical complexity of what is apparently a simple volcanic cone, and the occurrence of two distinct trachyte episodes in the evolution of a volcano which is dominantly basaltic. The results of this work have been materially enhanced by the excellent series of analyses carried out by Mr. F. T. Seelye in the Dominion Laboratory, and for this help and also for the kindly re-commendation given by the Director of the Geological Survey, as well as for the ready acquiescence in his request by the Dominion Analyst, I must express my indebtedness and sincerest thanks. References. Benson, W. N., and Turner, F. J., 1939. Mineralogical Notes from the University of Otago, New Zealand—No. 2, Trans. Roy. Soc. N.Z., vol. 69, p. 56. Johannsen, A., 1937. A Descriptive Petrography of Igneous Rocks, vol. 3. Speight, R., 1923. The Intrusive Rocks of Banks Peninsula, Rec. Cant. Mus., vol. 2, pt. 3, pp. 121–146. — 1924. The Basic Volcanic Rocks of Banks Peninsula, Rec. Cant. Mus., vol. 2, pt. 4, pp. 239–267. — 1935. The Geology of Gebbies Pass, Banks Peninsula, Trans. Roy. Soc. N.Z., vol. 65, pp. 305–328.