Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 40, 1907
This text is also available in PDF
(1 MB) Opens in new window
– 79 –

Art. V.—Geology of Centre and North of North Island.

[Read before the Otago Institute, 10th September, 1907.]

Plate XIII.

A Great deal of interest is attached to the northern part of the North Island from a geological point of view. This interest is partly a result of the direction of the trend of the land, which, somewhat to the west of north, offers a striking contrast to that of the rest of the Dominion, which is directed north-east and south-west. It is of some importance to know whether this direction of the northern portion indicates a new structural direction, or whether the land is composed of broken fragments of mountain-ranges parallel to the great structural feature of the North Island—the Tararua-Ruahine-Kaimanawa-Raukumara chain.

Additional interest attaches to the extreme north, because here Mr. McKay has mentioned the occurrence of intrusive masses and “sills” of crystalline rocks of plutonic character, which he has classified with the syenites. Except for the occurrence of tonalites and other dioritic rocks from the Cape Colville Peninsula, and of granites as boulders in conglomerates at Alexandra and at Gisborne, plutonic rocks are unknown in the North Island. From a popular point of view, the greatest interest attaches to this part of New Zealand because volcanic action has been more pronounced here than elsewhere, and is still maintained spasmodically. No comprehensive attempt has been made to deal with these volcanic areas since Hochstetter's time, though much information has been gained by several investigators in various parts of the district.

– 80 –

In a short paper of this kind it will be impossible to attempt anything more than a general discussion of these three matters.

A reference to any map which shows the contours of the western Pacific at once makes it clear that the northern peninsula is not continued far as submarine ridge below the waters of the Pacific. Still, there are submarine ridges parallel to it. A small ridge of this nature lies relatively close to the land, but does not extend far to the north. Over a portion of it the water is less than 500 fathoms in depth. A second ridge, of much greater importance, lies five hundred miles to the west. This, like the other, has a fairly large portion which is less than 500 fathoms below the surface of the water. The ridge continues as far north as New Caledonia without in any place dipping below the 1,000-fathom level.

There is, however, another submarine ridge of some importance north of New Zealand. Commencing about three hundred miles north of the Bay of Plenty, this ridge, less than 1,000 fathoms below the sea-level, extends continuously nearly as far north as Samoa. In ordinary maps it is not indicated as continuous, but as divided into two portions between the Kermadecs and the Tonga Islands. There does not appear to be any reason to divide the ridge into two parts in this manner. It is true that those soundings that have been made between these groups of islands indicate rather deeper water, but none of the soundings are in the direct line of the ridge, and all parts of it are extremely narrow. The evidence that is available seems to point to the continuous nature of the ridge rather than to its separation into two parts. The ridge appears to be a continuation of the trend-line of the main structural features of New Zealand. Wherever the ridge rises to the surface it displays volcanic rocks, as at the Kermadecs and at Tongatabu, though it must be remembered that Professor Thomas has obtained specimens of syenite from the former group. To the east of this ridge there is a deep rift in the bed of the Pacific. In places it is 5,000 fathoms in depth, and there appears to be definite evidence that it is 4,000 fathoms and more in depth throughout a distance as great as the length of the ridge that borders it so closely on the west. The evidence in favour of the continuity of the rift is similar to that given above—viz., in those places where discontinuity is generally represented in maps no soundings have been made in the direct line of the rift.

So far as submarine contours go, it appears from the foregoing statements that there is no definite evidence as to the nature of the northern peninsula. Trend-lines there undoubtedly are, and some of these are parallel to but not continuous with

– 81 –

the peninsula. On the other hand, there is strong evidence of pronounced structural lines in the bed of the Pacific in the same direction as the mountain-ranges, and, if an intermediate depth of 1,600 fathoms be disregarded, actually continuous with the dominant structural features of the North Island.

If we turn to the rocks of this portion of the land we find three main types—(1) volcanic rocks of many kinds; (2) Cainozoic sediments, probably of Miocene age; and (3) intensely folded and often contorted sandstones and shales, which have been classed as Carboniferous, though there is no definite evidence that they are older than the Mesozoic. With these are associated in many northern localities the plutonic rocks previously noted.

Of these three rock-groups, the first two are not folded, and therefore afford no evidence as to the structural features at present being considered. The slates and sandstones have, however, been subjected to earth-pressure of an intense nature, and it is from them that information is to be expected. Though the whole of the area has been geologically examined, it is a remarkable fact that there is in the reports that describe the country practically no statement as to the direction of the strike and dip of these older sediments. I was able to make a few observations last summer in the Bay of Islands, and here the beds are much contorted, and are often so changed that the stratigraphic planes are completely obscured. However, from the observations that could be made, there appeared to be a north or north-east strike, and the same direction appeared to be represented in the hills between Mangonui and the Oruru Valley, and in the shales that are occasionally displayed in the range extending from Reef Point to Raetea. This statement is very general, but it remains the only indication of the structural lines of the country. So far as it goes, it indicates that the trend of the land is not a result of structural characters, but, as it were, accidental, because here it happened that fragments of mountains with a northerly strike in many ranges were left in such large numbers as to constitute an apparent north-westerly trend. This view is in accord with that expressed by Suess.

At present a portion of the district is being examined in detail by the reorganized Geological Survey, so definite information will shortly be forthcoming.

The second matter of special interest is the occurrence of plutonic rocks at various places, which has been noted by McKay, who referred them to syenites or diorites. At Mangonui Township McKay states that these rocks are interbedded with sandstones and shales. of this no evidence could be seen. Certainly the character of the rocks varied somewhat: the colour is darker, and they are more compact in some places than in others. When

– 82 –

examined under the microscope it was seen that the differences were due to unimportant variations in a diorite rock. The rock is not coarse-grained, and the feldspar is nearly all triclinic, andesine and oligoclase being chiefly present. All the ferro-magnesian mineral is hornblende, but it is much decomposed into serpentinous and chloritic substances. There is some magnetite. In the absence of analyses, the rock appears to approach the syenites somewhat closely. At Ahipara other specimens were obtained that appear to represent the mass that extends from that locality to Reef Point, though the actual specimens were obtained from Ahipara only. The rock, again, is not particularly coarse-grained, and in hand-specimens is less grey than the diorite mentioned above. The separate minerals are clearly seen in hand-specimens, and in addition to feldspar and a dark ferro-magnesian mineral, olivine was evidently present.

When seen in thin sections the rock is at once identified as an olivine norite. The feldspar is a basic variety of labradorite. Augite is plentiful, and but slightly schillerised. The hypersthene is not abundant, and is generally associated with olivine, which is rather frequent. This appears to be the only olivine norite recorded from New Zealand, though it is probable that similar rocks exist in the Darran Mountains, near Milford Sound. Other specimens of plutonic rocks were obtained from the Raetea Saddle. They were almost entirely norites, but were wanting in olivine.

In those sections that were seen the relations between the plutonic rocks and the Mesozoic shales were not clearly displayed. No actual contact was observed, but from the irregular succession of the rocks on the road to the Raetea Saddle it was evident either that the Mesozoic sediments rested on a highly eroded surface of plutonic rock, or that the plutonic material was intruded into the sediments, and is therefore of Post-Mesozoic age. This is the view taken by McKay, and, although it is impossible to mention any section that negatives it, there are a few facts which suggest that more vigorous investigation is yet required. It is obvious that the intrusion of such large masses of plutonic rock would be likely to induce much contact action, yet when search was made in the sediments no evidence of contact action could be found, even when the outcrop of norite was close at hand. The slight schillerisation of the pyroxene also indicates that the plutonic matter has been subject to much dynamic action since its formation. Since there is no evidence of earth-movements in this district since the period of folding of the Mesozoic sediments, it would appear that the norite received its character of schillerisation at a period not later than that of the folding of the Mesozoic

– 83 –

sediments. As this folding probably took place immediately after their deposition, it appears that the norite can hardly be of Post-Mesozoic age.

The volcanic rocks extending from Mount Egmont, Ruapehu, and the Bay of Plenty northward have not received much attention, so far as general statements are concerned, since Hochstetter's time. Apparently he made extensive collections, but only a few of his specimens appear to have been submitted to microscopical or chemical examination. A few of them, however, were described by Zirkel.* All of them are classified as rhyolites, though with very different structures in the different specimens. of those examined, the majority were collected near Taupo and near Rotorua, though there were examples of obsidian from Tuhua as well. Mica was identified in many examples, but no rhombic pyroxene.

As a result of his observations, Hochstetter** classed the whole of the volcanic rocks of New Zealand in two divisions, called an older and a younger series. The different occurrences in the region under discussion were classed as follows:—


Older volcanic rocks. Tertiary and older Quaternary (Plutovolcanic).


Northward of Auckland Harbour, on the west. Andesite and dolerite breccias, with dykes of basalt.


South of Manukau, and thence to Aotea Harbour. Basalt conglomerates and basalts without distinct cones.


Volcanic table-land between upper and middle Waikato. Pumice and trachyte tuffs, with old extinct craters of trachytic, andestic, and doleritic rocks.


Younger volcanic formation. Acid and basic products. Cones with distinct or stuffed-up craters.


Taupo zone. Rhyolitic and trachytic lavas. Obsidian and pumice important. Includes the large volcanoes around Taupo.


Mount Egmont. This may belong to the older period.


Auckland zone. Sixty-three eruption-points, with distinct craters and lava-flows.


Bay of Islands. Between Hokianga and Bay of Islands. Similar to Auckland zone.

Since Hochstetter's time important work has been done by Professor Thomas. The results of his first paper may be thus summarised: Augite-andesites were found at Mount Edgecumbe; Ngauruhoe; Ruapehu, west side; Wanganui River,

[Footnote] * “Rense der ‘Novara’ : Geologie,” vol. i, p. 109.

[Footnote] ** “Reise der ‘Novara’ : Geologie,” vol. i, p. 200.

[Footnote] ‡ Thomas, Trans. N.Z. Inst., vol. xx, p.306 et seq.

– 84 –

on west side of Taupo; Whangamata Bay, West Taupo; Titiraupenga. Rhyolites were found in several places. Some contained quartz, brown hornblende, and augite. Spherulitic and axiolitic types were mentioned, and a banded type from Motutaiko Island, in Lake Taupo.

In a second paper* the rocks of Tongariro are described as typical augite-andesites, but in a few instances there was a little olivine—e.g., the summit ot Tongariro, at the red crater, and at Otukou. It is noticeable that in all of Professor Thomas's descriptions there is no mention of the occurrence of rhombic pyroxene.

Captain Hutton described many rocks from this district. Rhyolites are recorded from Taupo; hornblende-rhyolite, from Lake Tarawera and Lake Rotorua; augite-rhyolite, from Ateamuri; enstatite-rhyolite, from Lake Taupo; chlorite-rhyolite, from Okaro; pitchstone, from Maketu, Tauranga, and Mayor Island; spherulitic pitchstone, from Rotorua; obsidian, from Mayor Island, Taupo, and Lake Rotoiti. Trachyte is recorded from the Sugarloaves, Taranaki, but subsequently Hutton classified this rock as an andesite; from Whangarei, based on an identification of Cox§; from Runanga, Napier—Taupo Road, based on an identification of Hector Hornblende - andesites are recorded from Sugarloaves, Taranaki; Mount Egmont; eastern base of Mount Ruapehu; Tokatoka; Kaipara. Augite - andesites, from Mount Egmount; Mount Pirongia; Okaro; Mount Tarawera, eruption of 1886. Enstatite-andesite, from Sugarloaves, Taranaki; Ruapehu; Horohoro; White Island; Puponga, in Manukau Harbour; Helensville; Kamiti, in Kaipara Harbour; Whangarei Heads. Olivine-andesite, from Mount Egmont. A dolerite is recorded from Kakepuku; and basalt from Mount Eden and Rangitoto, near Auckland.

Thomas, in a report on the Tarawera eruption, published by the New Zealand Government in 1888, has mentioned the rocks of Mount Edgecumbe as augite-andesite, and the lava emitted as bombs from Tarawera in 1886 is also described under the same name.

Hill** has described the rocks of Ruapehu as basalt, trachyte, and andesite.

[Footnote] * Thomas, Trans. N.Z. Inst., vol. xxi, p. 349 et seq.

[Footnote] † Hutton, Royal Soc. N.S.W., 1889, p. 102 et seq.

[Footnote] ‡ Hutton, Trans. N.Z. Inst., vol. xxxi, p. 483.

[Footnote] § Cox, Geological Reports, 1876–77, p. 95.

[Footnote] ∥ Hector, Geological Reports, 1870–71.

[Footnote] ¶ Thomas, “Report on Eruption of Tarawera,” pp.13 and 58.

[Footnote] ** Hill, Trans. N.Z. Inst., vol. xxiv, p. 617; also Trans. Aust. Assoc. Adv. Sci., vol. iii, p. 170.

– 85 –

Park* has mentioned dolerite, phonolites, porphyritic trachytes, and pitchstone as occurring on Ruapehu; but neither of these last two authors appears to have made anything more than a field examination of the rocks.

Rutley has described a large number of rhyolites from the Rotorua area. Several of these suffered from geyser-action and have become more or less silicified. Descriptions of rocks from Tuhua (Mayor Island), in the Bay of Plenty, are included in this paper.

A different type of rock, a pantellaritic liparite lava, has recently been described by F. von Wolff from Mayor Island. This is the only mention of soda-rich types from the district—at any rate, in technical descriptions.

During the presence of the “Discovery” expedition in New Zealand, rocks were collected by Ferrar near the Aratiatia Rapids, on the Waikato River. They have been described as rhyolites and andesites by Rastall.§ Reference is here made to a peculiar reddish pyroxene that it is stated may be strongly soda-bearing.

The Auckland rocks have also been described by Shrewsbury, who classed them all as basalts.

The literature referring to the Thames rocks and those of the Cape Colville Peninsula is quite extensive, but there is a very general agreement as to the rock-types and the succession of lavas. The most recent publication on the district appeared in 1905, from the pen of Professor Sollas, with descriptive notes by A. McKay. Photographs of many of the rock-types appear in this work.

It is recognised by all workers in this field that the andesites are very varied in type and structure. They range from dacites to hypersthene andesites, with some olivine. Augite and hornblende types occur as well, but there are no unusual minerals present. Sollas speaks in several places of the micropœcillitic structure as peculiar. The mineral with which this structure is most commonly associated he has identified as quartz. Coarsely spherulitic rhyolities from this locality have received considerable attention from Rutley** and Sollas.†† The spheru-

[Footnote] * Park, Geological Reports, 1886, p. 70.

[Footnote] † Rutley, Quart. Journ. Geol. Soc., vol. lvi, p. 493. et seq.

[Footnote] ‡ F. von Wolff, “Centralblatt für Mineralogie, &c., 1904,” p.208 et seq.

[Footnote] § Geological Mag., Decade v, vol. ii, p. 403 et seq.

[Footnote] ∥ Shrwsbury, Trans. N.Z. Inst., vol. xxiv, p. 366.

[Footnote] ¶ Sollas, “Rocks of Cape Colville Peninsula,” 2 vols.; Government Printer, Wellington.

[Footnote] ** Rutley, Quart. Journ. Geol. Soc., vol. 1v, p. 449 et seq., particularly p. 466; also vol. 1vi, p. 509.

[Footnote] †† Sollas, “Rocks of Cape Colville Peninsula,” vol. i, pp. 120, 121.

– 86 –

lites are nearly an inch in diameter at times, and have irregularly curved radiating arms. Rutley regards the objects as a result of refusion of the rhyolite. Sollas rejects this explanation, and states that these features, as well as certain isotropic feldspars, have resulted from processes of decomposition. This explanation he afterwards withdrew, but did not substitute another.

A peculiar type of rock with a semi-brecciated appearance is called by Professor Sollas “wilsonite.” He suggests that its peculiar structure is due to the association of fragments of lava ejected during an eruption which retained their viscosity until they reached the ground.

A very complete bibliography of the literature of Cape Collville geology is given in the introduction to Professor Sollas's work.* Unfortunately, it is impossible to represent the results of different authors here. This is less regrettable because they are in essential agreement as to all the main features. Reference, however, should be made to the geological map of the district in the second volume, and a similar map by Professor Park.

In the second volume of Professor Sollas's report there are descriptions of rocks collected by McKay on the western spurs of the Kaimanawa Range. Some of these are probably material ejected by Ruapehu and its neighbours, for the rocks agree with those of the volcanoes in all essential particulars. Others agree with rocks near Lake Taupo. Other descriptions are given of rocks from the Sugarloaves, Taranaki. The only special feature to notice is the occurrence of hypersthene in one example as a core of a hornblende crystal.

Fox,§ in a paper on the volcanic rocks near Auckland, has described certain tuff-beds as being formed of fragmentary matter ejected by the Cape Colville eruptions, and others as formed during the eruptions of the Waitakerei volcanoes.

The physiography of this region has been referred to by many authors. Hill, in particular, and Park have described the physiography of the Ruapehu region, and further descriptions have been added by Von Friedlander, who visited the district after the eruption of Te Mari in 1896. Marshall and Alison have also written on the subject in the volumes of the “New Zealand Alpine Journal.” Thomas, in papers quoted above, has dealt fully with Tongariro. An accurate map of Tongariro has been drawn by Cussen.

[Footnote] * Sollas, “Rocks of Cape Colville Peninsula,” vol. i, p. 124.

[Footnote] † Park, “Geology and Veins of Hauraki Goldfields,” N.Z. Inst. Min. Eng., 1897.

[Footnote] ‡ Sollas, “Rocks of Cape Colville Peninsula,” vol. ii, pp. 160–65.

[Footnote] ‡ Sollas, “Rocks of Cape Colville Peninsula,” vol. ii, pp. 160–65.

[Footnote] § Fox, Trans. N.Z. Inst., vol. xxxiii, p. 452 et seq.

[Footnote] ∥ Friedlander, Trans. N.Z. Inst., vol. xxxi, p. 498.

– 87 –

The general features of the physiography of the whole district were, of course, fully outlined by Hochstetter,* who travelled throughout the district in 1864. The general results of his work require no very great modification, though, of course, there has been much change in the Tarawera and Rotomahana district as a result of the eruption of 1886.

Another description has been given by Marshall and by Gregory. Cussen§ has written papers on the changes in the course of the Waikato River, as well as a paper on the country to the west of Taupo, that is still very imperfectly known.

McKay has lately discussed the locality of the eruption, from which all the pumice was dispersed.

In regard to the age of the outburst of volcanic action in this part of New Zealand, we have Hill's statement that there is a pumice-bed interstratified with Miocene (Cretaceo-tertiary) clays at Tolaga Bay. Park** states that the activity of Ruapehu and Egmont began in the newer Pliocene. In the Thames district Park†† gives the Upper Eocene age for the commencement of volcanic action. Hector‡‡ states that the Thames andesites are of Cretaceo-tertiary age. Hutton§§ places the Thames andesites doubtfully in the Oligocene, and the volcanic rocks of the central region in the Pliocene. Afterwards∥∥ he states that the eruptions began in the Miocene.

In this paper an attempt will be made to combine the results obtained by the various authors named above with the observations made by the author of this paper.


A recent paper by Professor Park¶¶ has revised the classification of the Cainozoic rocks of New Zealand. Nearly all the Cretaceo-tertiary rocks of Hector, as well as his Eocene rocks, are referred to the Miocene as a result of a fresh examination of typical sections. If this reasonable conclusion is adopted, an

[Footnote] * Hochstetter, “New Zealand,” 1867.

[Footnote] † Marshall, “Geography of New Zealand,” p. 73 et seq.; Whitcombe and Tombs, 1905.

[Footnote] ‡ Gregory, “Australasia,” vol. i, pp. 577–82; Stanford.

[Footnote] § Cussen, Trans. N.Z. Inst., vol. xx, p. 316; vol. xxvi, p. 398.

[Footnote] ∥ McKay, Mines Reports, 1899, p. 16.

[Footnote] ¶ Hill, Trans. N.Z. Inst., vol. xx, p. 304.

[Footnote] ** Park, Geological Reports, 1886, p. 71.

[Footnote] ††Park, “Hauraki Goldfields,” p. 13.

[Footnote] ‡‡Hector “Outline of New Zealand Geology,” p. 87.

[Footnote] §§Hutton, “Geology of New Zealand,” Quart. Journ. Geol. Soc., 1885, p. 192.

[Footnote] ∥∥Trans. N.Z. Inst., vol. xxxii, p. 172.

[Footnote] ¶¶Park, Trans. N.Z. Inst., vol. xxxvii, p. 491.

– 88 –

important change must be made in the age of the Thames andesites, which rest on rocks that have hitherto been classed as Cretaceo-tertiary. They must be accepted as of Upper Miocene age at the earliest.

Hill's observations prove the Miocene age of some acid eruptions, probably that of the interior region near Taupo; so it appears that volcanic action commenced in the Thames and Taupo regions almost simultaneously towards the end of the Miocene period.

At Auckland, Fox has shown that the scoria-beds in the Waitemata series are of the same nature as the Waitakerei rocks, and, as the Waitemata beds are Upper Miocene, there can be no doubt that the great series of Waitakerei andesites are of Upper Miocene age. The main features of the Waitakerei rocks, stratigraphical, petrographical, and physiographical, are repeated at many points further north, notably at Kamiti, Kaipara Harbour; Manaia Peaks, Whangarei; the entrance to Hokianga; St. Paul, and the surrounding district, Whangaroa; south of Mangonui; North Cape district. It therefore seems reasonable to refer all these areas to eruptions of Upper Miocene age. In making this correlation, it must be remembered that the rocks have most striking characteristics in common, and that in several cases actual stratigraphical evidence that warrants such a correlation is to be found.

There is little evidence as to the age of the rocks of Karioi and Pirongia. Stratigraphically they rest upon Miocene limestones, and are possibly of late Miocene age. The rocks are dolerites, and differ markedly from all other volcanic material of the North Island, so far as my experience goes.

Another group of rocks about which there is at present but little information is that of the older basalts between Kerikeri and Orotere, and, further on, between Mangonui and Ahipara. I know of no stratigraphical evidence as to their age, and they are here termed “older” merely because of the mineralogical changes of serpentinisation that they have undergone, and because of the extensive weathering changes by which their surface has been altered. At Kerikeri they rest on Miocene rocks.

There appear, then, good evidences of great volcanic activity towards the close of the Miocene; but this activity was more pronounced in the northern part of the district than in the southern, for in all the extensive Miocene rocks near Wanganui there are no pumice or fragments of volcanic rock to be found, even in the upper rocks of the series. That this period of activity extended into the Pliocene is possible, though, owing to the general absence of Pliocene deposits, there is no absolute

– 89 –

proof of the statement. Such Pliocene deposits are, however, found in the Hawke's Bay and Wanganui districts. The former district has been described by Mr. Hill,* who mentions pumice and volcanic material in the Upper Pliocene only. From this it appears that the volcanic action which distributed pumice in the Miocene became dormant in the Upper Pliocene, or became extinct, and a new district became active in its place.

At Wanganui, Park has stated that volcanic material is found in the Upper Pliocene only. This agrees with Hutton's statement and with that in his geological history of New Zealand.§ With these statements my observations entirely agree, and I would add that the lower gravels of volcanic material in the Upper Pliocene at Wanganui contain a much larger quantity of pebbles of Mesozoic sediments and of rhyolites than the higher strata of gravels, which consist almost entirely of andesitic pebbles. This suggests that in the early Pliocene the sediments of the range west of Taupo had not become so nearly obliterated by volcanic ejecta as now, and that the Ongaruhe was then cutting its gorge vigorously through the white rhyolite, while the Wanganui did not have its headwaters obstructed, in bringing gravels from the Kaimanawas, by the huge andesitic masses of Ruapehu and his fellows. Later on, as Ruapehu grew, the source of sedimentary pebbles was cut off, and the steep slopes of Ruapehu yielded more and more material to the streams that coursed down its sides.

Further north the volcanic cones at Auckland are of extremely recent age. Their lava-streams flowed down valleys that still exist. So recent are the lavas that streams still flow beneath them through the loose scoriaceous matter of their lower surface. In no instance has a stream cut a higher-level channel on the surface of the lava. The same remarks apply to the volcanic matter at Whangarei and at the Bay of Islands. This volcanic action, however, appears to have lasted a considerable time. The rocks of the plateau of the lower Waikato are similar to those of the Auckland caves. Though still quite fresh at a little distance from the surface, there is a deep and fertile soil formed from the lava, and streams have cut deep channels through it. The same remarks apply equally to the Bay of Islands. It appears, then, that though the present cones and their lava-flows are of extremely recent age, they represent only the final effort of a long period of activity, which may have commenced in the Pliocene.

[Footnote] * Hill, Trans. N.Z. Inst., vol. xx, p. 301.

[Footnote] † Park, Geological Reports, 1886, p. 71.

[Footnote] ‡ Hutton, Trans. N.Z. Inst., vol. xix, p. 339.

[Footnote] § Hutton, Trans. N.Z. Inst., vol. xxxii, p. 173.

– 90 –

Physiography of the District.

So many writers have already discussed this aspect of the subject that little need be added here. There are, however, a few matters that seem to have, in part, escaped attention previously, and others which allow of very different interpretations.

The actual craters of the large volcanoes have often been described. It will, perhaps, be interesting to make a few remarks on the crater of Ngauruhoe, which I have visited six times since 1891.

The two earliest accounts, by Dyson and Bidwill, represent the crater as a profound abyss which could not be descended on any side, nor could the bottom be seen.

In December, 1890, it was possible to walk all over the bottom of the main crater, though steam-jets of some size were to be found in many places. Round each steam-jet there was a small cone of sulphur. The small scoria cone on the north rim of the main crater was then much more active than the main crater itself.

A year later the crater had completely changed, and there was a large pit near the centre of the main crater. This pit was the scene of rather violent activity, and it was impossible to see to the bottom of it.

But little change has taken place since that time, though the pit has become larger, and has changed its position rather to the west, so that in December, 1906, its western side coincided with the western flank of the mountain. At this time the mountain was rather inactive, and it was possible to see to the bottom of the pit. It was about 250 ft. in depth, with nearly vertical sides, which were encrusted with sulphur, and from which steam issued in clouds. At the bottom of the crater was a pond of water of bluish-green colour. There was a scum, apparently of sulphur, and the water was in ebullition. Sulphuretted hydrogen was being emitted in small quantity, but sulphur dioxide was in far larger amount. The small-rim crater to the north was nearly quiescent.

In February, 1907, the mountain became rather violent, and emitted large quantities of dust, which fell over the country to leeward. The mountain was ascended in March, during the continuance of the active phase. The crater appeared to have undergone no material change, but the shower of acid rain and mud prevented me from making more exact observations. The mud was six inches deep on the rim of the crater, and extended 2000 feet down the side of the cone.

The craters on Ruapehu and Tongariro have undergone

– 91 –

no material change within the period of my observations, and they have been accurately described by many observers.

The violent eruptions of Tarawera in June, 1886, have been so fully described by Hector, Hutton, Percy Smith, and Thomas, as well as a host of other writers, that it is unnecessary to refer further to them here. The features of this volcanic area have lately been examined by Bell,* and the changes that have occurred since the eruption are described by him.

An important feature of the physiography of the district is described by Cussen. This is the range of old folded sediments here referred to as Mesozoic, though stated by Cussen, in conformity with the usual custom, to be Carboniferous. The range commences ten miles to the west of Tongariro, and extends throughout the country to the west of Taupo. There is little doubt that, though the old sediments have actually been found only in few places on this range, it is really an old denuded range which has been smothered beneath the accumulation of volcanic material. This range was first discovered by Hochstetter.

The deep dry valleys found at intervals in the pumice country are deserving of some notice. They are especially frequent on the north of Taupo. In many of them no water has ever been known to flow, yet they are 150 ft. to 200 ft. in depth, with nearly vertical sides, and 30 or 40 yards wide, and often of great length. Even if heavier rainfall is assumed to have taken place in the past, it is hard to account for these. The eruption of Tarawera afforded a clue to their origin, for the torrential downpours of condensed steam and mud which succeeded the eruption caused the erosion of such channels in several places, notably near the road between Rotorua and Wairoa. It seems reasonable to suppose that the dry channels have generally been formed in this way.

Another physiographical feature which is most striking is the steep, straight-sided form of many of the hills in this region. Horohoro is a well-known example. The straight sides are formed of rhyolitic lava in most cases, though Cussen states that Titirangenga, in which straight sides are noticeable, is formed of augite-andesite. These remarkable hill-forms have been described as fault-lines along which the surrounding land has fallen in. Hochstetter first held this view, and more recently Gregory has adopted it, and the theory was mentioned by Marshall.§ Gregory describes one fault-plane along the

[Footnote] * J. M. Bell, Geograph. Journal, 1906, p. 369.

[Footnote] † Cussen, Trans. N.Z. Inst., vol. xx, p. 320.

[Footnote] ‡ Gregory, “Australasia,” vol. 1, p. 582; Stanford.

[Footnote] § Marshall, “Geography of New Zealand,” 1905, p. 183.

– 92 –

flank of the Paeroa Mountains parallel to the Tarawera fissure. These vertical scarps are general in the whole district. They are noticeable at Ngatira, on the Rotorua line, where the railway enters the plateau. They are prominent on the Rotorua side of this plateau and on the flanks of Ngongotaha, on Tarawera itself, and in the southern portion of the district such scrps are very prominent on the sides of all the streams that cross the railway-line between Mokau and Porootarao. It is evident that these features are most general, and, as in the southern district there can be no doubt that they are due to the resistant nature of the rhyolite, there is no reason why the same explanation should not be accepted for Horohoro and its fellows. If these features are due to faulting, it is remarkable that the eruption of Tarawera should have occurred in solid rock, midway between two profound adjacent faults parallel to it, for the sides of Tarawera have notably this scarped form.

The distribution of pumice has long attracted attention. Cussen has suggested that it was derived from the Taupo basin, for he noticed that the pumice on the west of the lake became coarser as the lake was approached. McKay has, for reasons of a similar nature, stated that eruptions probably took place somewhat to the east of Taupo. He rightly states that the distribution of the pumice is so great that it is almost impossible that it should have been the product of a single volcano. He supposes that many of the vents have afterwards been smothered in the products of other volcanoes. This statement of McKay probably represents as near an approach to exactitude as can at present be made. At the same time, it is reasonable to regard the lake-basins of the volcanic region as areas that have been affected by violent explosions, possibly of a hydrothermal or perhaps of a truly volcanic nature. That lake-basins can be formed by such explosions we have good evidence in Lake Rotomahna, and its contours are not strikingly different from those of the other lakes. If the explosion were accompanied with volcanic action and emission of acidic tuff, we have in the present depressions of the volcanic plateau sufficient points of emission to account for the distribution of pumice. The form of Lake Taupo is particularly suggestive of an explosive origin, though its present dimensions do not probably represent merely the area of the exploded depression. Such a cataclysm causes the outlet to be stopped up, and the gathered waters gradually spread over the adjacent lowlands.

It is noticeable that though the actual melted rock at Tarawera was andesitic, yet pumice of an acidic nature was more widely dispersed than the andesitic tuff. If this view is correct, the lakes of the volcanic country must be regarded as filling

– 93 –

explosion cavities, as Lake Rotomahana actually does. It is perhaps advisable in connection with this part of the subject to state that there is every reason against the supposition that the pumice was derived from any of the present volcanic cones. Without any known exception, all the cones of the district are formed of andesitic rocks from top to base.

So far as the nature of the rocks is concerned, I am able to make a few additions to the descriptions given, and, in view of the large amount of literature now available, to generalise rather more widely as to the distribution of various rock-types.

Rhyolites of many types are found throughout the district. The purely glassy type, obsidian, is found at Mayor Island and near Tarawera; spherulitic obsidians are common at Rotorua and near Wairakei. The glassy base is usually trichitic. Spherulitic rhyolites are very abundant. The coarse types, from the Cape Colville Peninsula, contain nests of angular quartz grains and some tridymite. I am quite unable to agree with either Rutley or Sollas as to the origin of the spherulites. While being somewhat diffident in this matter, I cannot regard them as either due to refusion or to decomposition. They appear to be essentially original, though the exact conditions necessary to their formation cannot at present be defined. They are the last objects to form during consolidation of the rock. At Lake Taupo and in many other places there is a banded rhyolite. When examined microscopically the darker bands are found to consist of axiolitic structures of indefinite length, and the other portion consists chiefly of microscopic spherulites, and sometimes the micropœcillitic structure of Sollas is distinct. The rhyolites in the eastern portion of the district, in the valley of the Ongaruhe, have a groundmass in which there is little individualisation of minerals, and the rock has markings that somewhat resemble the damascened patterns on a gun-barrel. Tridymite is common in this type of rock, but quartz is absent. The minerals which have crystallized out are not very numerous. Quartz occurs quite infrequently, but its place is generally taken by tridymite in very small aggregates. In the spherulitic rhyolites of Tairua quartz is found in nest-like aggregates, and distinct grains are found in some Taupo rhyolites and in the silicified tuff of the Huka Falls. Feldspar is found in all but the more glassy varieties. Often it is confined to minute radially arranged microlites in the spherulitic types, but distinct crystals are found in the rocks that are not particularly glassy. It is most abundant from rocks in the south and west of the district. Sanidine is relatively infrequent, for nearly all the crystals belong to triclinic forms, apparently between albite and oligoclase.

– 94 –

Of other minerals, hornblende is sometimes found, but is not very frequent. Biotite is still more uncommon. Hypersthene is by far the most usual of all ferro-magnesian minerals, especially in the southern portion of the district, though further north its place is taken by hornblende in some measure. Augite is uncommon. The pumice offers no special peculiarities, for it is merely vesicular scoria of the rhyolites.

Few analyses of the rhyolites have been published. Hochstetter* quotes some analyses of hot-spring deposits near Rotorua. Some of these appear to be silicified rhyolites. Maclaurin and Pond† give analyses of pumice. The percentages of lime and magnesia are somewhat higher than is usual in this type of rock. Determinations of silica are given in “Rocks of Cape Colville Peninsula.” The percentage is rather more than 70.

There appears to be no record of rhyolites occurring anywhere to the north of Cape Colville, except in the Great Barrier Island. The only example known to me is a dyke penetrating. the Manukau breccias at Karekare: it resembles those from the Ongaruhe Valley.

Trachytes: The only example of this group of rocks that I have had was taken from one of the small hills near the Kaipara. It is composed almost entirely of feldspar microlites, but there is also a little biotite.

Andesites: These rocks have a wider occurrence than the rhyolites, and differ among themselves more in mineralogical composition, but less in structure.

Dacites have a considerable distribution in the Cape Colville area, and many of them are coarsely porphyritic. Sollas has described them under several names. Hornblende, pyroxene, and hypersthene dacites all occur. The last are least frequent. The minerals occasionally occur together, though hypersthene and hornblende are not associated in more than two or three specimens of dacites.

Outside of the Cape Colville area dacites have not been recorded, so far as I know. I have, however, had specimens of hornblende-dacite from the Hen and Chickens Islands, and in the main volcanic area Tauhara is formed of a hornblendehypersthene-dacite. The hornblende has a peculiar reddish colour.

Of other andesites there is a great variety. The Cape Colville Peninsula has numerous representatives of almost every

[Footnote] * Hochstetter, “New Zealand,” p. 435.

[Footnote] †Pond and Maclaurin, Trans. N.Z. Inst., vol. xxxii, p. 233 et seq.

[Footnote] ‡ Vol. ii, pp. 303, 304.

– 95 –

type, though I do not know of descriptions of any mica-andesite. Hornblende-andesites are less usual than hypersthene-bearing varieties, and augite-andesites are not very common. Two or more of these minerals may occur together. The structures, too, are many. Besides the ordinary structures of andesitic rocks, Sollas has described the micropœcillitic, in which quartz forms grains of relatively large size, with highly irregular boundaries, and in the grains are included the constituents of the groundmass. Spherulitic varieties are also described in some number.

Mount Egmont consists entirely, so far as my researches go, of andesitic rocks. The usual type is a hornblende-augiteandesite, in which the augite is a pale green. The hornblende is sometimes completely resorbed, and an augite-andesite results. Occasionally a little olivine is found. This description agrees with that of other workers, though Hutton first described the Sugarloaf rocks as trachytes, and he has also mentioned a hypersthene-augite-andesite from this locality. Sollas mentions a little hypersthene in one type: I have found none in any of my sections. Mr. R. Browne sent me some fine lamellar specimens of hæmatite which were obtained from a tuff-bed on the lower slopes of Mount Egmont.

Ruapehu and its neighbours are entirely formed of hypersthene-augite-andesite, so far as I know. Specimens have been collected all over the east and south sides of the mountain, and from the west and north sides collections have been made from streams. The augite is pale brown, and the hypersthene is strongly pleochroic. There is no hypersthene in the groundmass, which is usually hyalopilitic, though sometimes pilotaxitic. A little olivine is occasionally found. It is usually surrounded by numerous hypersthene crystals. I have found no hornblende in any of my numerous specimens, and no examples of phonolites, basalts, or trachytes, mentioned by Park and Hill. In Thomas's descriptions of the rocks of these mountains there is no mention of hypersthene. This must be regarded as an oversight, for the mineral occurs so invariably in my specimens that I cannot fail to think that some, at any rate, of his must have contained it.

Hypersthene-andesites are recorded by Hutton from many other localities, and augite-andesites from many by Thomas. The latter mentions this rock as the product of the eruption of Tarawera in 1886. This statement has been confirmed by Hutton and Rutley. The specimens I have gathered from this volcano are hypersthene-augite-andesites again, but the rock is very fine-grained, and identification of the minerals is not easy, but there is no doubt that hypersthene occurs.

– 96 –

In the present state of our knowledge one appears justified in making the statement that nearly all the cones that rise above the rhyolite plateau are formed of hypersthene-augite-andesites. Tauhara appears to be the only exception recorded.

It may here be stated that Hochstetter referred to many of these rocks as trachydolerites, and that this name has been widely adopted in the reports of the Geological Survey of the past.

Hypersthene-andesites have a considerable development further north. All the specimens that I have gathered from the Waitakerei region belong to this type, and from the Little Barrier Island a pure hypersthene-andesite was given me by Mr. Cheeseman, F.L.S. At Whangarei Heads, Parua Bay, a similar rock was found. At Whangaroa hypersthene-andesites and hornblende-hypersthene-andesites were obtained from St. Paul's Dome. One is probably justified in assuming that these rocks occur in the other regions where the typical Manukau breccia occurs—viz., at Hokianga and at the North Cape.

In the central region it can be clearly seen that distribution of pumice succeeded earlier eruptions of the Ruapehu region, for at Waiouru and in the Onetapu Plains pumice rests on the surface of andesitic rocks. That the distribution of pumice was succeeded by eruption of andesitic matter is shown by the andesite tuff that rests on the pumice in the same localities.

Basaltic rocks show less variation, and have a wider occurrence. Pirongia and Karioi appear, from the specimens that I have collected, to be formed entirely of a porphyritic rock of this class, which is perhaps best called a dolerite. The olivine is much serpentinised, augite in large crystals is plentiful, and andesine-labradorite feldspar as well. The groundmass is augite feldspar and magnetite. Amongst New Zealand rocks this type resembles some of the dolerites of Dunedin more closely than any others that I have seen.

The older basalts which occur north of Kerikeri, and between Kaitaia and Ahipara, are very fine-grained; olivine much serpentinised, and fine; feldspar very plentiful, as well as augite and magnetite. I do not know the localities from which the eruption of these took place.

The rocks of the cones at Auckland and of the Waikato plateau, as well as those of the Bay of Islands, have always been classed as basalts. All that I have examined prove to be basanites. The nepheline is not present in any quantity, but it can be detected by gelatinisation and staining, as well as by the cubes of salt obtained when the solution derived from treatment of the rock-powder with hydrochloric acid is evaporated. These

– 97 –

basanites are usually fine-grained, though this character is far less noticeable in the specimens from the Waikato area, which are relatively coarse but even-grained, and thus different from the Karioi-Pirongia rocks.

A consideration of these statements will show that our knowledge at present allows us to classify the products of volcanic action as follows:—


Later Miocene,—


Andesites of Cape Colville Peninsula.


Andesites of Manukau breccias in their many occurrences.


Rhyolites of north of Taupo.


Dolerites of Pirongia and Karioi.*


Older basalts of Kerikeri.*


Later Pliocene,—


Hornblende-andesite of Mount Egmont.


Augite-hypersthene-andesite of Ruapehu and other cones of the plateau.


Hypersthene-dacite of Tauhara.


Basanites of lower Waikato.




Andesites of Ngauruhoe and Tongariro.


Basanites of Auckland and Bay of Islands.


Andesite of Tarawera.

A very interesting type of basanite is found in the Domain volcano, Auckland, in the form of ejected blocks only. The iron-ore is ilmenite; feldspar is oligoclase-andesine; olivine in elongated crystals; augite is violet, and shows strong pleochroism, and sometimes has a fringe of ægerine; nepheline idiomorphic and small; ophitic and micrographic structures are well shown, the latter as typically as in the celebrated type from the Labauer Berg.


There is little evidence in regard to the structural meaning of the direction of the North of Auckland Peninsula.

That the plutonic rocks of Mangonui and Ahipara are diorites and norites, but no evidence is available as to whether they are intrusive or older than the Mesozoic sediments.

Volcanic rocks are chiefly rhyolitic in the central region, but the rhyolites are penetrated by andesitic pipes, over which large cones have been built up.

The lake-basins are probably areas of violent hydrothermal explosions, and from these explosions pumice was distributed.

[Footnote] * Perhaps early Pliocene.

– 98 –

The sharp scarps of many of the rhyolite hills do not indicate the action of faults, but are due to erosion.

The sequence of eruptive rocks is suggested.

Note.—Specimens lately collected by Mr. R. Speight show that hornblende-andesite with much hypersthene occurs on the north slope of Ruapehu, and also on A Tama. This confirms Hutton's statement. The rock resembles that of Egmont in some respects, but must be scantily distributed on Ruapehu.

Since the above was in type I have received specimens of rock from the Patua Range, north of Mount Egmont, from Mr. N. Cochrane, and others from near Albatross Head, Kawhia, from Mr. R. Browne. In both instances the rocks are similar to those of Mount Egmont, except that pyroxene is entirely absent.

Explanation of Plate XIII.

Recent and Pleistocene. Sands, gravels, and pumice.


Cainozoic. Chiefly Miocene limestones and marls.


Mesozoic. Chiefly Triassic shales and sandstones.


Rhyolite. Eruption began in Miocene.


Hornblende-andesite, Mount Egmont; dacite, Tauhara.


Andesites of Cape Colville. Eruption in Miocene.


Manukau breccia. Hypersthene-andesites, Miocene.


Volcanoes of rhyolite plateau. Hypersthene-andesites, Upper Pliocene to Recent.


Dolerite of Pirongia and Karioi (Miocene ?).


Basanites. Waikato, Auckland, &c.


Older basalts of Kerikeri.


Diorites and gabbros. Age uncertain.

Note.—The map, Plate XIII, is largely based on the work of McKay, Park, and Cox so far as the boundaries of the sedimentary and volcanic rocks are concerned. The author alone is responsible for the boundaries of the different divisions of volcanic rocks.