The Latest Phase in Vulcanicity of the Lyttelton Volcano
[Read before the Philosophical Institute of Canterbury, October 2, 1935; received by the Editor, November 9, 1935; issued separately, December, 1936.]
The latest phase of vulcanicity in connection with the Lyttelton volcano presents certain interesting analogies. In a recent paper on the Differentiation of Hawaiian Lavas Howard A. Powers (1935) draws attention to an alkaline phase which marks a late period in the active life of these volcanoes. He says (p. 62), “The nephelite, basalts, nephelite-melilite basalts, and trachyte described by Cross and Sidney Powers occur only among those parasitic ‘cinder cone’ eruptions on the surface of the dormant volcanoes.” Before reading this paper the present author had been impressed with the fact that similar alkaline basalts mark what are apparently the latest developments of the Lyttelton volcano. The ordinary lavas poured out therefrom are basic in composition, with basic andesites containing olivine as a dominant facies. Such rocks would no doubt be called feldspar basalts according to British usage, but their normative composition shows a feldspar percentage well above the 62.5 per cent, fixed by Washington as dividing basic andesites from true basalts after a consideration of similar basic andesites in the Hawaiian Islands. There is no doubt whatsoever as to the classification of the later extrusions of Lyttelton; they are definitely basalts.
Special attention was directed to them by numerous examinations of the Halswell Quarry, which consists of a great mass of basalt over 180 feet thick, capping the lower portion and the northern flank of a spur stretching out from the Lyttelton volcano in a westerly direction towards the plains. When it was first quarried the top consisted of a high pile of rocks forming a definite landmark, but all this and much more of the rock has now been removed for road metal, and the chief supply is at present obtained from the middle slopes on the northern side of the spur. Over a considerable area the underlying floor of scoria and agglomerate has been exposed, and the fragments contained in this show an entirely different facies from the main stone of the quarry, which is remarkably uniform in character for a desposit of such thickness. It has usually been supposed that this is a flow which came from the direction of Lyttelton Harbour; but careful examination of the slopes above the quarry right to the edge of the crater-ring discloses no occurrence of similar rock. If it had occurred there it is inconceivable that a stone of such resistant weathering properties should have left no trace, and the most satisfactory explanation of its mode of formation is that it was the output of a parasitic cone formed during the latest stages of vulcanicity.
One special feature of the quarry is to be had in the jointing of the basalt. This is of three types: (1) a platy jointing with division-planes parallel to the surface on which the lava consolidated, the plates being thin, from 1 to 2 inches in thickness near this under
surface, and usually from 3 to 8 inches away from it; (2) a spheroidal parting under the influence of which the rock breaks up into onion-like shells, usually about 12 inches thick, the spheroidal masses being at times over a chain in diameter; (3) a parting crossing this at high angle in curving lines, almost as if the divisions were formed by the intersection of two spheroids of the (2) type. The regularity of the last two types and their freedom from disturbance indicates the slow cooling of a stationary mass of rock of great thickness, which probably welled up from below and filled a crater-like hollow in which it consolidated. No conduit for this supply has been exposed up to the present. This rock is not decidedly alkaline, but the analysis shows that it contains 1.70 per cent. of normative nepheline.
The conclusion as to the mode of formation of this mass of rock is not based solely on the evidence furnished by the quarry itself. On the hills further south-east in the vicinity of Tai Tapu and behind Sir Heaton Rhodes' house there are two knobs of similar basalt whose form is in no way related to that of the lavas which have been or could have been poured out from the Lyttelton centre. The first of these is on a spur leading up to Coopers Knobs. It forms a prominent mass on the crest of the spur and it rests on fragmentary material of character similar to that of the mass itself. Extending down the spur, but not upward as far as I can see, there is a large dyke, and it seems very probable that the knob has been formed from a local centre on the line of this dyke, that is, it is a parasite cone formed on the dyke.
On the opposite side of the valley to this a sugarloaf-shaped mass, elliptical in cross section, with the longer axis of the ellipse radial to the Lyttelton centre, whose form has no apparent relation to that of the ordinary eroded external slope of the volcano. It forms a definite excrescence on a landscape eroded from the quaquaversal dipping lavas and fragmentary beds derived from the Lyttelton volcano. The basalts of this knob show the platy jointing which is such a striking feature of the Halswell Quarry; the planes of division are inclined at high angles and are generally at right angles to the long axis of the ellipse which the knob forms. It is possible that this cone has been constructed on a short dyke with the radial orientation characteristic of the district. There is also something abnormal in the form of the land-surface in this locality which suggests that some incidents in its volcanic history are different from those which occurred elsewhere. The most striking of these is the plateau form of the spur reaching away from the knob, and the fact that the dominant lava capping it is a basalt.
I have not got analyses of either of the rocks just mentioned, but they show a definite though variable reaction to staining after treatment with hydrochloric acid, and therefore they no doubt contain traces of a feldspathoid, probably nepheline, since the staining affects at times small quadrangular particles wedged in between the feldspar-laths and with low polarisation colours. This remark also applies to other outcrops in the vicinity of this knob, so that it is probable that the outpouring of alkaline basalt was on a fairly wide scale in this area and not restricted to small centres. This is to be borne in mind when reference is made to the definite alkaline
flow exposed on the point near the Ahuriri Bird Sanctuary, which is after all not so far from the occurrences just referred to, but situated at the end of the spur leading therefrom. This rock contains 5.25 per cent. of normative nepheline, and Dr P. Marshall in a private communication states that the groundmass contains a considerable quantity of his recently discovered alkaline mineral ameletite. This occurrence is not obviously derived from a parasitic cone, but the lava cannot be traced up the slope, so it is apparently in the same position as that at the Halswell Quarry. It may be urged that in these cases, owing to the inclination of the beds being steeper than the external slope of the volcano, the upper portion of the flows which could have yielded these deposits, has been removed by erosion; but on the other hand some trace of them should have remained since they are rocks with more powers of resistance than the ordinary andesites of the area, and these less resistant andesites do form slopes which are completely accordant at times for considerable distances with the angle of the flows which were poured out. In this Ahuriri case the alkaline basalts may be connected in some way with the occurrences some two or three miles away at a higher level, and in both cases the flows represent the last flows poured out.
There is, besides, the case of Quail Island, which is possibly a parasite cone on the floor of the caldera. This shows two or more definite flows, alkaline in composition, with phenocrysts of olivine standing out in places on the weathered surface. These beds also belong to a late stage of vulcanicity. The composition of the Halswell Quarry rock, and those at Ahuriri and on Quail Island, are cited in a paper by the present author (1924, pp. 252–3).
In addition to these flows there are also dykes of alkaline character which belong to a late phase of the activity of the volcano. They are mostly trachytic, but basalts with alkaline relationships occur occasionally, and these latter may be connected genetically with the outpourings of basalt. The trachyte dykes are definitely alkaline in character, and are frequently phonolitic, that is, they are trachytoid phonolites, phonolitic trachytes, and trachy-andesites, even occasionally trachy-basalts. In those cases where the silica percentage is high and they are related chemically to rhyolites their alkaline character is very marked. Anorthoclase is an important feldspar and the femic minerals are aegirine-augite and a blue-green hornblende; olivine, too, is of frequent occurrence even in the trachytes. Some of these, however, are an almost pure potash-soda feldspar (bostonite). These dykes penetrate right to the summit of the craterring, and, while it cannot be maintained that all belong to a late phase of the vulcanicity, it is certain that many do, and even those which never reached the surface may date from the same epoch as those that did. One of the most striking of the latter is the big dyke on the western side of Heathcote Valley. This is very variable in composition and texture, but it is definitely trachytoid; in places it is a trachy-andesite and then again almost tinguaitic. It has flowed out of a fissure and welled over the sides like a mushroom, suggesting that when it was intruded it overflowed the adjacent surface, that this was approximate in form to that now existing, and also that the
contact underneath this overflowing dyke is a part of the pre-existing landscape. This is the nearest approach to a definite trachytic lava-flow that I know of, though some of the other large trachytic dykes may have had a similar form originally, erosion having removed the mushroom-like portion. The composition of the Lyttelton rocks is given by the present author in a paper on the Intrusive Rocks of Banks Peninsula (1923, pp. 130–1).
When the circumstances of the Banks Peninsula rocks are compared with those of the Hawaiian Islands the resemblance is seen to be very striking. In one case cited by Cross from Hawaii Is. (1915, pp. 35–6) trachyte is evidently of a date intermediate between those of basaltic eruptions; but on other islands of the group as cited by Sidney Powers (1920, pp. 269–73) the correspondence is complete. He says, “On Maui, the trachyte occurs as flows, cones, and in one instance probably as a volcanic neck. In every case observed the trachyte is younger than the main mountain and probably appeared when the dissection of the latter was well under weigh” (cf. the dyke near Heathcote Valley). He mentions a trachyte flow evidently derived from a fissure eruption. With regard to the nepheline basalts, which like those of Banks Peninsula contain only a small amount of nepheline, he says (loc. cit., p. 273), “In each instance of a local centre the eruption has been long after the main volcano became quiescent.” And then again (loc. cit., p. 280) he says, “Towards the close of the evolution of the volcanic edifice another change appears to take place occasionally—the appearance of unusual rock types. Trachy-andesite, in contrast to basalt, is found on the summits of Haleakala and Mauna Kea; trachyte forms a veneer over portions of west Maui, etc.” However, Stearns (1935, p. 181), when referring to the trachyte occurring on Oahu, says, “Because the trachyte on Oahu is exposed at what appears to be the lowest stratigraphic position in the Waianae series, and because elsewhere it marks the dying phase of a basaltic volcano, it may indicate the summit of an older volcano buried beneath the Waianae lavas.” The consensus of opinion is therefore that trachyte and a slightly alkaline basalt mark the end-phase of Hawaiian vulcanicity, and in this respect the case of the Banks Peninsula volcanoes is analogous.
It is not maintained, however, that trachyte is always an end-phase, for numerous instances can be cited where it is definitely intermediate, and others where it is probably intermediate. In his account of the Tertiary volcanic rocks of Skye, when referring to the trachyte dykes, Harker says, “We assign all the rocks, on such evidence as is attainable, to some of the very latest stages of igneous activity of the region.” All the same he admits that an earlier series of trachytes is definitely sandwiched in between two series of basalts. In the Auckland Islands trachytes occur as sills, dykes, and flows, antedating the latest phase of basaltic eruptions, and may be earlier than even an older phase (Speight, 1909, p. 723). This corresponds to what Marshall found on Campbell Island (1909, p. 691). Richards (1916) mentions that trachytes occupy an intermediate stratigraphic position among the volcanic rocks of south-eastern Queensland. In South Victoria Land kenytes form a late phase, and definite trachytes
the latest stage of this phase as shown in tuff cones on the summit and round the base of Mount Erebus (Jensen, 1916, p. 94, seq.), although it must be admitted that Prior (1907, p. 123), somewhat discounting the opinion of Ferrar, regards the basalts as younger than the kenytes and trachytes.
Numerous cases could be cited of the association of trachytes with basalts on oceanic islands, but their true stratigraphic position may be uncertain. It is clear, however, that trachytes and alkaline basalts do mark an end-phase, and even in those cases where they are apparently intermediate they may mark an end-phase of what has been a definite period of intermediate vulcanicity as is suggested by Stearns.
In his work on the Evolution of Igneous Rocks Bowen (1928, pp. 236–240) refers to the association of trachyte and basalt specially in oceanic islands, the former being a differentiate from small bodies of basaltic magma. Although he mentions the occurrence of a phonolite or a phonolitic trachyte as a possible end-type, he does not mention an alkaline basalt, except that rocks of the plateau-basalt type, when cooled rapidly, as they might be in a dyke or in a small intrusion, tend to develop analcite. Instances of this could be cited from the igneous intrusions of the Malvern Hills area.
One further point may be mentioned in connection with the relations of the trachytes to the alkaline basalts of the Lyttelton region. In no case has a trachyte or other dyke been observed to cut alkaline basalts either on the flanks of the Lyttelton volcano or on Quail Island—and the same is true of the rocks of Mount Herbert—although trachyte dykes in some cases cut the underlying rocks and show an erosion-surface where they terminate in an upward direction. It cannot be definitely stated, but it is reasonably certain, that the alkaline basalts represent the ultimate phase of vulcanicity, even post-dating some, if not all, of the trachyte intrusions.
In these remarks concerning the Lyttelton volcano I have hardly referred to Mount Herbert. The slightly alkaline basalts of this mountain are also an end-phase, whether they be regarded as coming from the Akaroa centre or from some concealed orifice on the southeastern flank of the Lyttelton volcano. Up to the present I have seen no parasitic cones on the flanks of the Akaroa volcano analogous to those of Lyttelton, though it is possible that one or two may occur. Also, if these cones were low down on the flanks of the volcano, as they are in other cases, they might have been removed as a result of the eroding action of the sea on the external coastal margin of the peninsula.
The occurrence of an alkaline facies as an end-product of a basic volcanic episode can no doubt be plausibly explained by some method of differentiation. There are some indications of an alkaline phase in the final stages of consolidation of the common type of Lyttelton lava, for the feldspar of the groundmass is frequently of oligoclase, even in the case of those which show dominant phenocrysts of labradorite, augite, and olivine; and it is conceivable that the differentiation of large masses of this rock might produce a sodic residual, and that this differentiate might again yield the trachytic
product and in the final phase an alkaline basalt. It is not proposed to advance any hypothesis to account for such a differentiation, except to suggest that separation by gravitative flotation or settling of crystals does not account for it entirely. The dominant facies of the lava is the one which contains a notable percentage of the phenocrysts mentioned above, and the latest flows exposed on the edge of the crater-ring and also the dominant flows on the surface of the flanks of the cone are of this type, the percentage of labradorite in some cases being very high, certainly over 50 per cent. of the rock. Since labradorite has a specific gravity of between 2.71 and 2.72, while that of the rock in the solid state is about 2.68, and in the liquid state was probably less, it is conceivable that the phenocrysts of feldspar might settle down along with those of the augite and olivine, and this might account for the association of these phenocrysts with the penultimate stage of extrusion. But there is a further stage, viz., that of the trachytes and trachy-andesites with lighter phenocrysts—sanidine, anorthoclase, oligoclase—which would float in the unconsolidated magma; but they are associated with augite, hornblende, and with olivine which would sink in such a magma, if the labradorite sank in the magma associated with it, and it appears improbable therefore that this end-phase can be accounted for entirely by gravitative concentration. The occurrence of olivine as phenocrysts in these rocks is an interesting example of its early formation being associated with acid characters in the groundmass. There is, besides, the basic ultimate phase which has yielded rocks of higher specific gravity. This could easily be explained if it were granted that both types came from some side pocket of the main magma in which differentiation took place partly by gravitative settling, the more acid differentiate being erupted first and the more basic last.
In his paper on Pacific Lavas Barth (1931, p. 527) concludes that crystal settling is adequate to explain the differentiation of these lavas, but Howard A. Powers in his paper on the Differentiation of Hawaiian Lavas (1935, p. 69) states, “The theory of crystal settling does not provide for the development of an unsaturated liquid fraction from the original sub-basalt.” His statement of the probable course of differentiation is interesting since he says (loc. cit., p. 70), “After slight differentiation of the primary Hawaiian magma possibly by separation of olivine and calcic plagioclase crystals it yields soda-basalts and soda-sub-basalts. Significant differentiation of the magma yields mostly rocks of the alkaline division-pacificite, tephrite, basanite. Extreme differentiation yields trachyte and nephelite-melilite basalt.” He sums up against such a course of differentiation being explained by fractional crystallisation. The first of these stages is admirably represented in the case of the Lyttelton volcano, and the last is also well represented by the trachytes, but his intermediate stage, viz., that of pacificite, appears to belong to the ultimate and not to the intermediate stage in the case of the Banks Peninsula volcano; and there is no occurrence as far as is known of tephrite or of true basanite.
References to Literature.
Barth, T. F. W., 1931. Petrography of Pacific Lavas, Am. Jour. Sci., 5th series, vol. 21, pp. 377–405, 491–530.
Cross, W., 1915. Lavas of Hawaii, U.S. Geol. Surv., Prof. Paper, 88.
Harker, A., 1904. Tertiary Igneous Rocks of Skye, Mem. Geol. Surv. U.K.
Jensen, H. I., 1916. Rep. on the Petrology of the Alkaline Rocks of Mount Erebus, Antarctica, Brit. Antarctic Expedition, 1907–9, Rep. Sci. Invest.
Marshall, P., 1909. Geology of Campbell Island and the Snares.
Powers, H., 1935. Differentiation of Hawaiian Lavas, Am. Jour. Sci., 5th series, vol. 30, pp. 57–71.
Powers, S., 1920. Notes on Hawaiian Petrology, Am. Jour. Sci., 4th series, vol. 50, pp. 256–280.
Prior, G. T., 1907. Nat. Ant. Expedition 1901–4, Natural History, i, Petrography.
Richards, H. C., 1916. Volcanic Rocks of South Eastern Queensland, Proc. Roy. Soc. Queensland, vol. 37.
Speight, R., 1909. Physiography and Geology of the Auckland, Bounty, and Antipodes Islands, Subantarctic Islands of New Zealand.
——1923. Intrusive Rocks of Banks Peninsula. Rec. Cant. Mus., vol. 2, no. 3, pp. 121–150.
——1924. Basic Volcanic Rocks of Banks Peninsula, ibid., vol. 2, no. 4, pp. 239–267.
Stearns, H. T., and Vaksvik, K. N., 1935. Geology and Ground-water Resources of the Island of Oahu, Hawaii, Territory of Hawaii, Division of Hydrography, Bulletin 1.