Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 77, 1948-49
This text is also available in PDF
(1 MB) Opens in new window
– 456 –

Part I. Physiography And Field Occurrence Of The Rocks.

Kapiti Island lies about three and a half miles west of Paraparaumu Beach on the west coast of Wellington (Fig. 5). The reader is referred to Ferrar (1928) for a general description of the island.

The northern, western and southern coasts of Kapiti are exposed to strong marine erosion and are steeply cliffed. The divide separating drainage to the north-west from that to the south-east lies close to the western cliffs, and from its crest, regularly spaced ridges slope to the south-east, separated by deeply cut stream valleys. In plan, the south-eastern coast presents a gently arcuate front, convex towards the sea, with two minor salients, at Kurukohatu in the north, and at Rangatira, in the centre, which are cuspate forelands composed of

– 457 –
Picture icon

Fig. 1—View due north from Pukerua Bay, Wellington, shows the regular east-sloping ridges of the southern portion of Kapiti Island (left). The form of Kapiti is attributed to dissection of a tilted peneplain, and the strait between Kapiti and Paraparaumu to synclinal structure, Waikanae delta (right centre) and Tararua foothills (right). Drawn from a photograph.

Picture icon

Fig. 2—View of Kapiti Island from Paraparaumu Beach shows the regular pattern of eastward-flowing consequent streams dissecting the warped (monoclinal) surface represented by regularly sloping ridges and interfluve remnants. Rangatira Point is on the coast beneath the summit. Phyllonite occurs beneath the bench (light centre) and on the islets on left Drawn from photograph.

– 458 –

beach boulders. That at Rangatira is backed by a low bench, some 100 ft. high, which may be a raised beach, and the cuspate forelands themselves, which are some 10 ft. high, have been formed, in part, when sea-level stood higher than it does at present.

South of Rangatira a narrow bench, 5–12 ft. above high tide level, is cut in the sea cliffs. As this is in places protected by talus, extends into the bays, and bears on its surface a varying thickness of irregularly bedded and locally consolidated beach gravel, it represents a previous beach level. Ferrar (1928, p. 314) noted an abandoned sea cave containing beach deposits at 20 ft. above high water level. The lowest raised beach has been ascribed to the activity of the 1855 earthquake, but it is much older, for its deposits are consolidated and weathered.

Behind the abandoned beach, on the coast south-east of Rangatira, steep facets truncate the toes of the spurs from the western summit-ridge of the island. Ferrar interpreted the truncated spurs as due to faulting, but their greater height towards the exposed southern end of the island suggests that they are sea cliffs, developed by retreat of the coast through a fault zone of less resistant rock, i.e., that they are fault-line features.

Viewed from the Wellington coast, both from the south (Fig. 1) and from the north, the regular eastward-sloping profiles of the Kapiti ridges suggest derivation from a warped or tilted peneplain, by the activity of consequent streams.* The view of Kapiti from the mainland at Paraparaumu (Fig. 2) supports this hypothesis, for there seem to be residual, slightly dissected interfluve surfaces on some of the ridges. In the north, streams and ridges have an easterly direction, in the centre of the island the ridges slope south-eastward, and in the south they slope to the south-south-east, an approach to a radial pattern. Kapiti may thus be part of an elongated anticlinal dome developed by warping of a peneplain.

The two Fishermens Islets, off the south-east coast of Kapiti, are stacks some 50–75 ft. in height (Plate 49, Fig. 3), and Tokomapuna, abut a mile off-shore, is a low islet surrounded by an extensive shallow reef. Possibly the Fishermens Islets are residual stacks carved from a bench of the same height as that behind Rangatira, and Tokomapuna must have been wave-cut at times of higher sea-level.

There is strong evidence, from the nature of the schistose rocks, that a long-established fault zone coincides with the eastern coastline of Kapiti, but the writer found no certain evidence that late faulting had influenced the island's physiography. On the contrary, a hypothesis of minor sea cliffing, producing a nick in an east-sloping monocline, dissected by consequent streams, seems adequate to explain the observed features of the coast. Although the mainland coast opposite is in places strongly cliffed, and has been considered a fault coast (by Ferrar, loc. cit., for example), certain views of it hint at the possibility that it, too, may be a much modified monocline (Fig. 1). Whether or not this is so, it would seem reasonable to consider the sea-filled depression between Kapiti and the North Island a synclinal structure separating the great Tararua anticline, on the east, from a minor fold, on the west, of which Kapiti Island is the only emergent

[Footnote] * Later observation suggests that the regularity of ridge profiles has been exaggerated by the shading of Fig. 1,

– 459 –

portion. Since undeformed Pleistocene sediments (Otaki formation) occupy parts of the syncline a little north of Kapiti. the folding may be attributed to the late-Tertiary, Kaikoura movements, without implying thereby any great precision of geologic dating.

Occurrence and relations of the schistose rocks: Metamorphosed rock, superficially resembling the schists of Otago and Marlborough, crops out at Rangatira, where originally reported by Ferrar, and on the two Fishermens Islets.

At Rangatira, the southern part of the foreland is underlain by schistose rocks which outcrop on the beach some 3 to 35 chains south of the boatshed. Other exposures were found protruding through the boulder-strewn surface of the raised beach, between the strand line and the higher land to the west. Soil in cuttings on the path between the boatshed and the caretaker's house contains weathered schistose fragments, and seems to have been derived from schistose rock. A large angular block of schistose rock was mapped near the point where the low Rangatira promontory joins the coastal cliff (Plate 49, Fig. 3), but farther to the south-west the coast is composed entirely of unmetamorphosed indurated sandstone and argillites such as compose the greater part of the island.

The s-planes of the schistose rocks have an average strike of about 17° east of north, varying from 0° to 27°E. The dips of the planes are consistent within each outcrop, and suggest that the rocks are disposed in an anticlinal structure, with a western limb dipping towards the unaltered sediments to the west at 40° to 50°, although individual readings are steeper.* Such variations in dip are probably of little significance in what must be considered a broad zone of intensely sheared rock. The specimens of phyllonite from Rangatira (P.7633, 7634) are of finer texture than those from Fishermens Islets (P.7635, 7636).

On the coast south of the Rangatira foreland, greywackes and argillites are exposed, locally massive and irregularly jointed, elsewhere showing a regular and consistent interbedding of alternating dark and light, or fine and coarse, argillites and sandstones. Large beach boulders that cannot have been transported far consist of conglomerates (P.7624) and finely banded argillites (P.7625, 7627). Pug zones, in which the rocks are greatly sheared and brecciated, occur at several points, but these nowhere resemble the phyllonites. Most bedding planes and shear planes within pug zones strike parallel to the bedding of the phyllonites, i.e., 17°E., but the dip is steeper (70° or more) to the west.

The Fishermens Islets (Tahoramaurea and Motungarara) are composed entirely of phyllonite which outcrops on their wide shore platforms, displays even more distinct “schistosity” than at Rangatira, and includes phases in which light and dark bands alternate (P.7636), and others with regularly arranged lensoid “augen” giving a false impression of foliation (P.7635). The strike and dip of the s-planes is somewhat irregular, swinging within a short distance, but the general dip is 45° to 50° to the east. In all the Kapiti phyllonites the s-planes are clearly parallel to the original bedding.

[Footnote] * On a later visit (31/3/40) additional exposures from beneath the beach gravel failed to support the idea or anticlinal structure Further outcrops of phyllonite were found six chains from the coast up the stream south of the boatshed.

– 460 –
– 461 –

Tokomapuna Islet, and the reefs which surround it, are reported by Mr. Wilkinson to be composed of normal unmetamorphosed grey-wacke and argillite; their appearance from a distance (and photograps taken on the islet) support this.

Judged by the known distribution of the Kapiti phyllonites, they occur in a linear zone, some 30 chains in width, between the relatively unaltered rocks of Kapiti itself, and those of the outlying islet of Tokomapuna. No actual contact between altered and unaltered rocks was seen by the writer, but no transitional rocks were seen, and Mr. N. H. Taylor, Soil Bureau, reports that a contact examined by him in a stream bed, 2 chains west-south-west of the boatshed, is abrupt and steeply dipping if not vertical, and that the fault pug is uncemented. It is difficult to believe that the intensive shearing necessary for the formation of phyllonites could be localised to such an extent that sediments represented by P.7625, 7626 and 7629 could remain in a comparatively unaltered condition within a few chains of the greatly altered phyllonites P.7633 and 7634. For this reason it is necessary to infer that considerable movement, at a date later than the formation of the phyllonites, has brought the two sets of rocks of contrasting metamorphic rank, into close contact. The zone of phyllonite may perhaps best be considered a “horse” of altered rock formed in an early phase of disastrophic movement, brought to its present position in a later phase of faulting. The sediments of Kapiti are certainly of pre-Cretaceous, possibly of pre-Triassic age, and the formation of the phyllonites may date from the post-Hokonui (Lower Cretaceous) orogeny. The movements which faulted the phyllonites against unaltered sediments cannot be closely dated, but preceded the formation of the peneplain which truncates the structures in the old rocks at Kapiti. This peneplain was deformed before the deposition of the Pleistocene Otaki formation, and, in common with some other peneplains recognised in New Zealand, is believed to be later Tertiary in age.

Structural relations: In the past, few have speculated on the relationship between Kapiti Island and the structures of adjacent parts of Cook Strait. Among later reviewers, King (1939, p. 555) noted that “In the North Island, the only counterpart of the Marlborough schist is on Kapiti Island, some seven [actually 3½] miles from the mainland.” King considered this occurrence of great importance to his argument that Cook Strait arose by “the rupture of once-continuous geological and topographical features and the development of a definite off-setting between the North and South Islands.” (p. 547). Dr. C. O. Hutton's demonstration (in Part II. of this paper) that the Kapiti “schist” is not homologous with the quartz-feldspathic schists of Marlborough and Otago renders this item of evidence in King's hypothesis invalid. Further, evidence presented below for the continuity of structure from Kapiti across Cook Strait to Marlborough is at least as strong as the evidence for a late Pliocene dislocation of the mountain axes on either side of the Strait.

Macpherson (1946. maps 1, 2) has drawn a southern extension of the Pohangina syncline through the depression between Kapiti and the Tararua Range.* This correlation is adopted here, and Kapiti

[Footnote] * Adkin (1919) recognised a depressed aieu between Kapiti and the Tararua Range, but correlated it with the Port Nicholson-Pukerua Bay depression.

– 462 –

itself is considered comparable with the dome-shaped “basement folds, with thin crestal sections and thick adjacent synclinal sections.” (Macpherson, 1946, p. 8) of Pliocene covering beds, which, west of the Pohangina syncline, represent the projection beneath the Palmerston-Wanganui basin of the southward-plunging Kaimanawa anticline. Wells drilled in such folds as Mt. Stewart and Marton by the Superior Oil Company showed that they had grown intermittently throughout the Pliocene. Probably the structure of which Kapiti is a part had a similar intermittent growth, but there is no evidence whether the warped and dissected peneplain of Kapiti is a pre-Pliocene surface, or a surface developed during stillstand at a Pliocene or post-Pliocene date. The Mt. Stewart dome has an asymmetrical cross section, with steep eastern flank, perhaps faulted. The interpretation here offered of the structure of the emergent parts of Kapiti is based on a hypothesis of steep monoclinal warping without late faulting, but it must be admitted that the monocline may steepen into a fault beneath the sea east of Tokomapuna. The deepest part of the strait of Kapiti is a narrow channel kept free of sediment by the 1-to-3–knot tidal current that flows through it.

If Kapiti is part of a crestal swell on an anticlinal fold, continuous for 150 miles from the Kaimanawa Range, an extension of the structure farther south-west may be sought. Submarine conditions in Cook Strait are not favourable for the preservation of structural features of the sea bottom. Tidal currents passing through the strait at speeds of up to 3½ knots are periodically strengthened by southerly and northerly gales. On the sea bottom within the strait, foul bottom, “ledge rock,” coarse detritus, and even derived Pliocene mollusca (locally), attest the strength of the bottom currents. Under such conditions, destruction of structural irregularities, and their replacement by a topography related to bottom scour, are to be expected, and these processes must have been accentuated if, as is generally believed, sea level retreated about 300 ft. during the last glaciation. Admiralty Chart No. 695, despite its inadequate scatter of soundings, supports the inferred dominance of bottom features related to current scour through the strait. The northern part of the strait, between D'Urville Island and Wanganui, is nowhere deeper than 75 fathoms. Farther south, as the narrowest part of the strait is approached, a central channel of more than 75 fathoms is developed, deepening to 100 fathoms in Lat. 41°, and with three small “holes,” over 150 fathoms in depth, in the narrowest part, between Arapawa Island and the Porirua-Terawhiti coast (Fig. 5). Where the strait widens again, between the Wellington Peninsula and Cloudy Bay, the channel shallows to under 100 fathoms, but irregular “deeps” of over 100 fathoms persist in its central course for some distance past Port Nicholson, and are marked on the chart as related to “overfalls” at the surface.

Considering the extent to which bottom scour seems to have determined the bottom configuration of the strait, it is significant that, from its deepest part, there rises, to within 5 fathoms of the surface, a rock pinnacle, known as Fishermans Rock. This submerged peak is of small area, but has a relief of about 1,000 ft. from its summit to the deepest part of the adjacent scour channel (178 fathoms). Fishermans Rock must certainly be the remnant of a

Picture icon

Fig 3—The east coast of Kapiti Island. View south-west from Rangatira. Schistose rock cops out on the boulder beach in the foreground and on the two Fishermens Islets (left of centre, on skyline). Tokompuna Islet (extreme left) and the main mass of Kapiti (right) are composed of unmetamophosed greywacke and argillites, Raised beaches, on headland in mid-distance, and on right.

– 463 –
– 464 –

larger feature which the bottom currents have so far failed to remove. Its position, about 18 miles south-west of Kapiti is here interpreted as an indication of the extension of the Kaimanawa-Kapiti fold-axis across Cook Strait.

Between Kapiti and Fishermans Rock there are other shallow rocks, known locally, as fishing grounds, brought to the writer's attention by Mr. N. H. Taylor, to whose recognition of the significance of these submarine features this account owes not a little.

In September, 1929, a local fisherman told Mr. Taylor of several uncharted reefs in the Kapiti area. One, at a depth of 15 fathoms, was reported to run east-west for about a mile and to lie two miles south of Kapiti. Another, between Kapiti and Mana, and seven miles south of Kapiti, was stated to be a triangular reef, 10 chains wide across its base. In response to an inquiry, the Secretary, Marine Department, has replied that the existence of such rocks, in the vicinity of Kapiti, is known to his department, but that they cannot yet be located with accuracy.

Still farther south-west, the Kapiti Island-Fishermans Rock axis, or rather the south-east facing scarp of this axis, is directly co-linear with the scarp forming the south-eastern coast of the Marlborough Sounds block from Wellington Head to White's Bay, and with the Wairau scarp bounding the Wairau Plain. The continuity of this feature, a south-east-facing scarp from Kapiti to the Wairau, now interrupted, but still recognisable as a “single lineament” (Cotton) in the structural sense, is advanced as evidence for the continuity of some, at least, of the fold-axes across Cook strait, a hypothesis which manifestly needs further testing, but which should not be neglected in favour of one involving extensive lateral displacement in late geologic time.

Another feature of the sea bottom of probable structural significance is a shoal, at 40 fathoms, which lies near the centre of the strait immediately west of a line joining the south end of the Wellington Fault (upper Kaiwarra and Silver Stream valleys) with the north end of the Awatere Fault.

The Wairau scarp is currently considered to be a fault scarp, and the well-known Recent fault traces, parallel to its course, bear witness to the occurrence of late faulting movements, but the gently sloping regular facets between the transverse gullies draining the scarp are very similar to those of the dissected surface at Kapiti here interpreted as monoclinal. Cotton (1947) has recently illustrated the facets of this scarp near Tophouse (1947, Fig. 5 B). It is difficult to understand the development of such a regular east-dipping surface as that of which the facets are obvious remnants, by normal erosion of a presumably steep initial scarp, prior to the establishment of the transverse streams which are now destroying it. The Wairau fault trace affords indisputable evidence of late faulting, but one may nevertheless speculate whether the Wairau tectonic scarp cannot be considered partly monoclinal in origin. Wellman (1945) has shown that the Pikikiruna Scarp, long believed to be a fault scarp, can more reasonably be interpreted as monoclinal.

– 465 –

List Of Rock Specimens From Kapiti Island.

(Numbers refer to specimens in the collection of minerals and rocks in the New Zealand Geological Survey.)

P.7623—Sheared greywacke from pug zone. Mouth of Waterfall Stream.

P.7624—Indurated greywacke-conglomerate boulder. Mouth of Waterfall Stream.

P.7625 Indurated breccia of sandstone fragments in argillite matrix. Mouth

P.7626 of Waterfall Stream.

P.7627—Alternating black-and-grey argillite. 17 chains N.E. of mouth of Waterfall Stream.

P.7628—Weathered and sheared greywacke. 12 chains N.E. of mouth of Taepiro Stream

P.7629—Massive indurated sandstone. 22 chains N.E. of mouth of Taepiro Stream.

P.7630—Sheared and slickensided argillite from pug zone. 27 chains N.E. of mouth of Taepiro Stream.

P.7631—Indurated conglomerate; pebbles up to 1 ½ in, 32 chains N.E. of mouth of Taepiro Stream.

P.7632—Indurated sandstone. 44 chains N.E. of mouth of Taepiro Stream.

P.7633—Weathered phyllonite, coast. 13 chains S.E. of Boatshed.

P.7634—Fresh phyllonite, coast. 13 chains S.E. of Boatshed.

P.7635—Phyllonite, east side of Tahoramaurea Islet.

P.7636—Phyllonite, west side of Tahoramaurea Islet.


Adkin, G. L., 1919. Further Notes on the Horowhenua Coastal Plain and the Associated Physiographic Features. Trans. N.Z. Inst., vol. 51, pp. 108–118.

Cotton, C. A., 1914. Supplementary Notes on Wellington Physiography. Trans. N.Z. Inst., vol. 46, pp. 294–298.

—— 1947. Revival of Major Faulting in New Zealand. Geol. Mag., vol. 84, no. 2, pp. 79–88.

Ferrar, H. T., 1928. Geological Notes on Kapiti Island. N.Z. Journ. Sci. and Tech., vol. 9, pp. 312–315.

King, L. C., 1939. The Relation Between the Major Islands of New Zealand. Trans. Roy. Soc. N.Z., vol. 69, pp. 544–569.

Macpherson, E. O., 1946. An Outline of Late Cretaceous and Tertiary Diastrophism in New Zealand. N.Z. Dept Sci. and Ind. Res. Geol. Memoir, no. 6.

Wellman, H. W., 1945. Takaka Coalfield. N.Z. Journ. Sci. and Tech., vol. 27, no. 3 (B), pp. 189–198.