The Stratigraphy and Structure of the Waipapa Group on the Islands of Motutapu, Rakino, and the Noisies Group near Auckland, New Zealand

Transactions of the Royal Society of New Zealand : Geology, Volume 5, Issue 9, 30 January 1968, Page 215

The Stratigraphy and Structure of the Waipapa Group on the Islands of Motutapu, Rakino, and the Noisies Group near Auckland, New Zealand

W. Mayer

By

University of Auckland

[Received by the Editor, April 7, 1967.]

Abstract

The stratigraphy in an area of complexly deformed geosynclinal sediments of doubtful, but possible, Jurassic age is based on the mapping of five units of differing lithologies and stratum thicknesses which are described in detail.

Two periods of compression, with the principal horizontal stress acting in a constant ENE—WSW direction, are recognised. The first period of tectonism, during the Early Cretaceous, resulted in megascopic concentric folding with an average trend of the fold axes to 350° and a small plunge to NNW. Folding was accompanied and followed by the formation of shear and tension joints. The second and milder deformative phase, which probably followed the first period closely, led to mesoscopic folding of first-period joints. Predominantly normal faulting along two major directions commenced prior to the Lower Miocene, but movement along some faults may have recurred at a later time. The NW-SE and NE-SW aligned fractures divide the area into a number of fault blocks, some of which have been tilted and rotated about a vertical axis.

An investigation of the joint pattern revealed a close and constant relationship of tension joints and certain sets of shears to the trend of the fold axis at any particular locality. The value of the beta-plot and pi-plot methods in the structural analysis of these sediments was investigated. They were found to be of very limited use and subject to possible errors of interpretation if used to the exclusion of conventional methods.

The history and age of tectonic events is discussed with reference to other areas in Auckland and Northland.

Introduction and Previous Work

The area under discussion consists of a group of small islands in the Hauraki Gulf near Auckland city and is shown on the locality map, Fig. 1. The islands rise to a maximum height of 392 ft at Motutapu and show a rolling mature topography. They are generally steeply cliffed on their seaward sides. No detailed description of the geology of this area has previously been published. Hochstetter (1864), who did not visit the area, mapped the southern part of Motutapu Island as Tertiary in age. Park (1887) made a short visit to Motutapu and described some of the Tertiary sediments exposed in a cliff section along the

western coastline. He also collected and identified a number of macrofossils. Brothers and Colson (1959) have described a Recent dune section from Motutapu Island. A very brief outline of the geology of this area was given by Grant-Mackie (1960).

Brief Outline of the General Geology

As shown in the geological maps (Figs. 2 and 3), the basement rocks of the Waipapa Group outcrop over ithe greater part of these islands and reach a possible maximum thickness in excess of 11,000 ft. They consist of apparently unfossiliferous greywackes and argillites with smaller amounts of cherts, volcanic argillites, and spilitic lava. A Jurassic age is tentatively assigned to them.

The clastic sediments were deposited in part of the extensive Upper Paleozoic to Mesozoic New Zealand geosyncline, predominantly by the action of turbidity currents. The writer proposes to give a detailed description of their sedimentation, provenance, and petrology in a later paper. These rocks have been highly deformed and disrupted by tectonic folding and faulting and by jointing. The strike of the strata indicates a major structural trend NNW to SSE.

The basement rocks are overlain with a pronounced angular unconformity by sediments of the Waitemata Group, of Otaian (Lower Miocene) age.

Method of Study

No detailed stratigraphic description has been made in the past of sediments of the Waipapa Group. In this small area of islands, where good exposures occur along shore platforms, an attempt is here made to erect a stratigraphy based on

lithological units and bedding thicknesses. The clastic sedimentary rocks can be divided into four mappable units:

1. Massive, thick beds of greywacke. 2. Alternating beds of greywacke and argillites, the coarser grained beds being generally more than lin thick. 3. Thinly bedded greywackes and argillites, where the greywacke beds seldom exceed lin in thickness. 4. Argillites, usually thinly bedded.

A fifth unit mapped consists of cherts, jaspers, spilitic lava and volcanic argillites. The rocks of this unit, however, often occur as discontinuous lenticular bodies and cannot generally be used as marker beds.

In order to map these units in detail, it was often necessary to collect specimens at short intervals across the strike for cutting to determine the unit they belonged to. Most specimens were oriented in the field to establish the right way up of the strata from sedimentary structures displayed on polished surfaces.

Lithological Units

1, Massive thick beds of greywacke. The terra greywacke is used in this paper for a texturally and/or compositionally immature sandstone with a high degree of induration. The term argillite refers to highly indurated sediments where the constituent particles are of silt or clay size. All size terms used are those given by Williams, Turner, and Gilbert (1958: 279).

The rocks of this unit are usually greenish to grey in colour; they are very hard and have a massive appearance. They form prominent outcrops, reaching in one locality a possible maximum thickness of 1,500 ft. Apart from rare, thin argillite partings that indicate the strike and dip of these rocks no signs of bedding are apparent (PI. 1, Fig. 1). However, specimens taken at intervals from this thick bed showed it to be composite. Its grain size varies between that of a coarse pebbly greywacke to a fine-grained rock only just coarser than an argillite. The finer grained horizons never persist for long; most of the massive greywackes are mediumand coarse-grained.

No sedimentary structures have been seen in these rocks, but inclusions of argillaceous rock fragments, which may be up to sin long, are frequently present. The massive greywackes have, as a rule, a very high feldspar content. Veining is never conspicuous; where present, veins are thin and spaced widely apart.

2. Alternating thick beds of greywackes and argillites. These sediments, which are marked by a constant alternation of greywacke and argillite layers (PI. 1, Fig. 2), represent the flysch deposits typical of geosynclinal successions. They show various shades of grey when fresh, but are brownish, buff, and even creamy-grey where weathering has been intense. The greywacke beds average about Ift in thickness, but they range from lin up to 6ft. The argillites are seldom more than 6in thick and mostly only between 2 and Sin. They are often laminated, being made up of several thin layers 2-smm thick of alternating coarser and finer grained argillaceous material.

Most of the sedimentary structures, such as sole markings, graded bedding, convolute bedding, and small scale cross-bedding, which are commonly associated with turbidite sequences, are found. They are not confined to the coarser grained greywacke beds, but occur also in some of the unlaminated argillites.

The greywackes are generally of medium-sand-grain size, but may be coarser grained in the thicker beds. These sediments have a higher quartz content than the massive greywackes described above. The argillites are, on the whole, coarse- to medium-grained siltstones; the fine-grained types are generally lacking. Veining is much more common in these rocks than in the massive greywackes.

3. Alternating thin beds of greywacke and argillite. The individual beds are generally not more than lin thick (PI. 1, Fig. 3) and may be as thin as O.lin, with an average thickness of Normally, the greywacke layers are thicker than the argillites. The latter are darker grey in colour. The greywackes are usually of fine-sand-grain size, while the argillites are mostly of coarse to medium silt-grain size. Graded bedding and occasional small scale cross-bedding are found in both the greywackes and the argillites. Veins are extremely common in these rocks, often forming a dense network.

4. Argillites. These are usually dark grey or almost black in colour. Thick sequences of argillites usually appear massive, showing few signs of bedding when examined in the field (PI. 1, Fig. 4). However, almost all samples revealed more or less well defined bedding when cut and sectioned. Most layers or laminae in the argillites are less than thick. Usually, the bedding is very fine with successive laminae differing slightly from each other in grain size. The argillites are fine- to coarse-grained siltstones; finer grained mudstones and claystones are almost completely absent. Both grading and small scale cross-bedding may be encountered. A dense network of veins traverses the rocks, some up to 3in in thickness.

5. Cherts and associated rocks. The cherts form the most prominent members in an association that also includes considerable thicknesses of red, green, and grey volcanic argillites, irregular patches of manganese, and some interbedded spilitic lava.

The term volcanic argillite is adopted from Reed (1957: 42) Who concludes that, as the chemical composition of the very similar rocks at Wellington is intermediate between that of “ normal ” argillites and that of pillow lavas, they represent a mixture of volcanic material and fine-grained sediment.

(a) The cherts. The cherts in this area show a great variety of colours, including white, cream, grey, green, red, buff, and black, which are due mainly to the presence in the cherts of iron and manganese in varying concentrations and also to the degree of oxidation of the iron. All the cherts are extremely hard and break with a conchoidal fracture. Their lustre may be dull, vitreous, or waxy. Most conspicuous in the field and also the most common types are the red cherts or jaspers; white cherts are rare.

Lenticular masses of banded chert occur, up to 850 yards long and 280 yards wide. The individual bands are from to Ift thick with an average of 2-3 in. They are separated by thin shaly partings generally less than lin in thickness (PI. 2, Fig. 1). The banding in the cherts is often very irregular: the bands may pinch and swell and lens out completely or two separate bands may join into a single one.

The chert bands are usually massive, showing no internal structures, but in one outcrop they show well-developed laminations. The individual laminae are from o.2mm to 2cm thick and average 2-3 mm. The colour of these laminae alternates between cream, buff, and black. In some laminated cherts all the laminae are dark grey or black. The occurrence of bedded cherts as lenticular bodies has been described by various authors from New Zealand and elsewhere (Halcrow, 1956; Davis, 1918; Bryan and Jones, 1962). The cherts also occur as single bands interbedded with grey, green, or red volcanic argillites at intervals of generally 6in to Ift. The bands may be up to 2in thick, but are usually much thinner (PI. 2, Fig. 2).

( h) The volcanic argillites. These may be green, red, black, or a distinctive dovegrey in colour. Generally, they have the same colour as that of the chert bands with which they are interbedded (PL 2, Fig. 2). They are extremely fine-grained, so that mineral grains can only rarely be identified. The volcanic argillites occur in beds up to 20ft thick. Usually, however, thin chert bands which intervene confine them to thinner layers. The overall thickness of volcanic argillite plus chert bands may reach up to 200 ft,

All the cherts examined contain radiolarian casts. These are most common in the laminated cherts (PI. 2, Fig. 3), being so plentiful in some laminae that the spherical casts almost touch. Red, green, and grey volcanic argillites also contain occasional radiolarian casts.

(c) The spilitic lavas. Spilitic lavas, which outcrop in only one locality in this area, are dark red in colour and very hard. The rock resembles an outcrop of jasper in the field, but lacks the characteristic banding (PI. 2, Fig. 2). The outcrop is about 20ft wide at its widest and can be followed over a distance of 50 yards. The spilitic lavas are interbedded with a sequence of dove-grey volcanic argillites containing thin chert bands.

( d) Manganese. Manganese ore occurs as small elongated masses a few feet long and no more than Ift wide interbedded with jaspers. Some cherts have a very high content of manganese, which is also found as irregular streaks and layers interbedded with the coloured volcanic argillites.

Stratigraphic Succession and Correlation of Section

The geological map (Fig. 2) shows the distribution of the lithological units just described. The extent of the Tertiary covering strata is also shown. As very few inland outcrops occur in this area, the drawing of the lithological boundaries across the islands is to some extent hypothetical.

The thicknesses of the various stratigraphic units shown may in some cases have been overestimated owing to the possible presence of more folds and faults for which, because of the high degree of deformation in these rocks, no satisfactory evidence has been found. Strata from one island may also be repeated in an adjacent one within this area. The cross-sections shown in Fig. 4 are to some extent diagramatic and do not show up all the structural complexities in the rocks.

The total possible maximum thickness of strata in this area amounts to 11,370 ft. The maximum thicknesses estimated for the various islands are as follows: Motutapu Island, 5,440 ft; Rakino Island, 2,280 ft; Otata Island, 1,950 ft; Motohoropapa Island, 1,200 ft; Maria Island, 500 ft.

The basement rocks on Motutapu Island become progressively younger towards the north-east. In the lower part of the sequence coarser grained sediments predominate, while thick outcrops of argillites occur towards the top. Rakino and the other outlying islands in this area mainly consist of argillites.

Apart from the cherts and associated rocks and the lowermost I,oooft of strata in this sequence outcropping on the south-eastern coastline of Motutapu, all lithological units to the north-east of the Motutapu Fault can be traced across the island. The massive and coarse-grained sediments, which outcrop very prominently on the south-eastern coastline of Motutapu, are mapped on the evidence of large boulders on the north-western shore platform, where Tertiary sediments form the cliffs. The stratigraphic position of a small outcrop of massive sediments on the downthrow side of the Motutapu Fault is not known. The rock is quite distinct mineralogically and is not found anywhere else in this area.

As can be seen from the map (Fig. 2), the cherts and associated rocks may be interbedded with any of the other four lithological units mapped. Every outcrop of this association of rocks that was mapped differs in its make-up from all the others. While in one outcrop chert is dominant, in others volcanic argillites are more common. Fig. 5 shows cross-sections through some exposures of such rocks as seen on the Shore platform.

The sections show the general stratigraphic conformity of these rocks with the surrounding sediments. In the section Fig. 5, b, the chert ribs, here interbedded with dove-grey volcanic argillites, show a dip towards the lava in the centre of a synclinal fold.

It has been observed that where chert is the dominant lithology the outcrop is markedly lenticular. This is best shown at Administration Bay, on Motutapu (Figs. 2, sa). On the other hand, where chert occurs as ribs only within volcanic argillites the outcrop is of a more continuous nature, e.g., in the northern part of Motutapu (Fig. 2). Here, too, however, the extent of the outcrop is limited along the strike. Furthermore, the lithologies are changing along the length of the outcrop. While dove-grey volcanic argillites and spilitic lava are present where these rocks are exposed along the north coast of Motutapu (Fig. sb), thick green chert bands interbedded with green Volcanic argillites occur in the southern extension of this unit at Station Bay. This shows that such lithologies cannot be used as stratigraphic marker horizons.

Correlation and Age

Very little is known about stratigraphic relationships and ages in the Waipapa Group, which outcrops extensively over part of the Auckland and Northland areas (Kear and Hay, 1961; Thompson, 1961) (Fig. 1). Both lithologically and petrologically the rocks in this area resemble closely those described from the Waipapa Group in other localities (Halcrow, 1956; Brothers, 1956; Hopgood, 1960), but the apparent absence of fossils in this area and their scarcity elsewhere make age correlations within the group very difficult.

Fossils discovered so far in the Waipapa Group range from Permian in Northland (Homibrook, 1951; Leed, 1951, 1956; Waterhouse, 1964) to Triassic and Jurassic in the Hunua Ranges and on Ponui Island (Milligan, 1961). Ponui Island is the closest known fossil locality to this area (Fig. 1). The sediments there are continuous along the strike with strata in the eastern part of Waiheke Island, where the rocks may also be presumed to be Jurassic in age. Considering the amount of repetition of strata by folding and faulting that has occurred in this area and assuming that the same holds for Waiheke, the thickness of strata separating the two areas may not be very great. A tentative Jurassic age is therefore assigned to the Waipapa Group outcropping in this area.

Structure of the Waipapa Group

Structural analysis of the rocks of the Waipapa Group is complicated by their high degree of deformation, by the presence of numerous planes of shear and fracture and by the rare occurrence of well-defined bedding planes. The presence of many faults adds further to the difficulties. Actual fold hinges could not generally be observed in the field in this area, so that the presence of most folds had to be inferred from reversals of dip. The determination of the right way up of beds from various sedimentary structures has been helpful and has enabled the writer to map fold axes with some degree of confidence.

Major Structural Trends

The dominant trend of the fold axes in this area (Figs. 2,7) is between north and north-west, with the strata most commonly dipping towards the west. This major strike trend is in general agreement with that found in the sediments of the Waipapa Group outcropping from the Hunua Ranges in the south to the northern part of the Northland peninsula (Fig. 1) and is reflected by their outcrop pattern.

The average strike on Motutapu Island is 350°, with a slight plunge to the north. This is shown on a pi-diagram on a lower hemisphere equal-area projection prepared from sixty-one bedding plane measurements (Fig. 6a). Locally, the fold axes may show a variation in strike ranging from 310° to 015°. The more common occurrence of westerly-dipping beds is clearly borne out by this diagram.

On Rakino Island, the strike of the sediments varies between NNE and NW. A pi-plot, containing all the bedding planes measured on Rakino, shows an average trend of 356° with a northerly plunge of 12°. Local divergences from the mean strike can in some parts of the island be attributed to the rotation of small fault blocks.

The rocks on Motohoropapa Island in the Noisies Group strike between 320° and 330°, and on neighbouring Otata Island between 350° and 360°. All the sediments present on these two islands dip in a westerly direction. The greatest deviation from the regional trend occurs on the Four Islands, where the strike is between 300° and 320°. On Maria Island, the sediments trend 005°. This corresponds well with the average trend, just east of north, of the major fold axes on Waiheke Island immediately to the south (Halcrow, 1956).

Two prominent fault systems, striking approximately north-west and north-east respectively, have divided the area into a number of small fault blocks.

Folding

Deformation in this area has resulted in the folding of sediments on both a megascopic and mesoscopic scale. The folds are of the concentric or parallel type, in which the major internal movement during folding has occurred parallel to the bedding planes. Marked development of slickensides along some bedding planes, which are produced by such bedding-plane shear movements, has been observed.

A very weakly developed fracture cleavage has been noted in some outcrops of volcanic argillites. The cleavage planes dip very steeply and are nearly parallel to the axial plane of the fold. These sediments appear to be the least competent in this area and have responded to the deformative stresses by movement along parallel planes normal to the principal horizontal stress. The occurrence of weak cleavage in volcanic argillites has also been observed by Reed (1957) in the Wellington area.

1. Megascopic folding. While on Motutapu, the coarser grained sediments have been folded into relatively large and open fold structures which may be traced right across the island, finer grained sediments, particularly in the core of a large fold, have responded by the formation of a number of smaller, tighter folds which in some localities have a very short axial length and plunge steeply at both ends.

The area on Motutapu Island bounded by the Motutapu and Home Bay Faults (Fig. 2), where the outcropping sediments are mainly coarse-grained, is dominated structurally by a major syncline, the axis of which has been displaced repeatedly by cross faults. The anticlinal folds flanking the syncline on either side have been cut by the two major strike faults present and cannot be traced across the island.

As seen in the cross-sections (Figs. Lb, 4c), the axial planes of the folds show an overturn towards the north-east on the southern coastline of Motutapu, but have been overturned towards the south-west on the northern coastline. These opposed directions of overturn are in part at least due to a tilting in the opposite direction of fault blocks in the north and the south of the island.

East of the Home Bay Fault, the folding in the finer-grained sediments has been tighter and on a smaller scale (Fig. 4b). Some of these folds appear to have a short axial length and plunge at up to 55°, while the average plunge of the major fold is constant at 10-15° towards the north.

On Rakino Island, where most of the sediments are fine-grained, a syncline is the most prominent structural feature (Figs. 2,4 a). It is accompanied by several tight folds of smaller dimension. Here, too, faulting has displaced fold axes, generally towards the west. The axial planes of folds on Rakino Island do not show any marked overturning.

With the exception of The Four Islands (Fig. 3), through which passes an anticlinal fold axis, no fold structures have been observed in the islands of the Noisies Group, where the strata dip mainly to the south-west.

2. Mesoscopic folds. Small-scale folds have been observed in several parts of this area, the exact localities being shown on the map (Fig. 2). These folds always strike parallel to the trend of the major folds in any given locality. In some cases the axial planes in the small folds dip in accordance with the dip in the respective megascopic fold structures, but in most cases the small-scale folds do not show any overturn. The plunge of these folds is generally towards the north at between 10° and 55°, but in two places a southerly plunge has been observed.

The best examples of small-scale folds are seen in outcrops of cherts at Administration Bay on Motutapu Island (PI. 2, Fig. 1). The folds have an amplitude of up to Ift, with a varying wavelength of up to 2ft, The small-scale folding at this locality is confined to the cherts; the adjacent sediments have been completely unaffected. The writer does not believe that these fold structures have formed as a result of slumping of the chert mass, however, as all the folds have a uniform trend and conform perfectly to the regional strike of the enclosing sediments. Folding of the cherts seems to have occurred by tectonic deformation due to a more plastic behaviour of the cherts. The preferential development of folds of this nature in chert masses enclosed in thick sedimentary sequences has been described from other parts of the world, e.g., by Davis (1918) and Bryan and Jones (1962). Tectonism is in most cases believed to be the cause of the folding. Some slumping has, however, occurred in volcanic argillites immediately adjacent to the folded cherts at Administration Bay. The outcrop there gives a chaotic appearance with no preferred alignment of fold axes.

Small-amplitude folds in outcrops of fine-grained sediments, which have resulted from deformation of earlier-formed low-angle shear joints, have been observed in two localities. They plunge at between 10° and 20° in a northerly and southerly direction respectively, and the trend of their fold axes is parallel to the regional strike in this area. Near the north-eastern corner of Motutapu small-scale folds have been produced by the deformation of high-angle cross-joints. The folds have an amplitude of about 9in and a wavelength of 2ft . They plunge NNW at 55°.

Jointing

Jointing is the most conspicuous feature in the rocks in this area. Joints transect the strata in many directions and at varying angles, ranging from megascopic wideclefted fracture planes to minute features only visible under the microscope. A distinction is made in this paper between two genetically different types of joints: shear joints, which are planes of rupture along which shearing and some slip has occurred due to compression, and tension joints, which may form during compression and parallel to the direction of main stress or following the release of compression.

1. Shear Joints. These dip generally at a lower angle than tension joints. The joint planes usually outcrop as straight lines and are frequently slickensided. The fissures are often filled by veins of quartz, prehnite, and chlorite, sometimes containing angular fragments of the country rock which were incorporated during fracturing and movement along the joint planes. Displacement of strata along such joints of up ito several feet was often noticed in the field. Finer-grained sediments have been subjected to more intense shearing than the coarser-grained more competent rocks.

Shear joints dissect the rocks in this area in many directions and with varying degrees of intensity. Fig. Bis a strike-frequency diagram of all shear joints recorded from Motutapu Island. Two systems of conjugate shear joints well developed in the

field, one dipping at angles of 30—35° and the second at between 45° and 70°, have produced the maxima in the directions 005° and 330°. The acute angle of about 35° between the strike of the planes of these systems is bisected by the major fold direction. A second broad double maximum is present in Fig. 8 showing directions to 045° and 300° approximately. These maxima were also produced by a double system of joints dipping at similar angles to those mentioned above. In this case the acute angle formed by the joints is bisected by the principal stress axis. Both systems of joints show a definite relationship to the principal direction of compressional stress and to the axis of folding. Many other joint directions that are present cannot, however, be directly related to these.

The close relationship of particular sets of joints to a fold axis has often been observed in the past, and for this reason shear joints have frequently been used as an aid to the structural analysis of highly deformed rocks. The intersections of shear planes are plotted as beta-points on a stereographic projection and are then contoured according to their density. Any density maximum thus obtained, theoretically represents a fold axis. If several maxima result, two or more periods of folding are sometimes postulated.

Both Brothers (1956) and Hopgood (1960) have used the beta-plot to analyse the structure of sediments of the Waipapa Group in small areas of North Auckland. In this area beta-plots prepared from measurements in several localities by plotting beta points in similar numbers to those used by the above-mentioned authors, about 300 per plot, have produced maxima which vary greatly within the vector 300° to 060°. Plunges ranged from 5° to 45° in either a northerly or southerly direction. Several diagrams showed up two or even three maxima, although none of these was marked.

These obviously unsatisfactory results led the writer to attempt to examine closely practically every outcrop in the area to measure vaguely preserved bedding planes and to produce the litho-stratigraphic map shown on Fig. 2. The results showed that the strike of the strata in several localities diverged from a mean direction owing to local curvature of a fold axis or to rotation of fault-bounded blocks or to movement along individual faults. Since, as mentioned earlier, some sets of shear joints bear a very close relationship to the axis of a fold at a particular locality, a change in the strike of fold axes, such as occurs frequently on the eastern coastline of Motutapu, where the strike ranges from NW to NNE, will introduce several maxima into a beta-plot.

The results show that fallacious conclusions may be arrived at when employing the beta-plot in an area of deformed sediments where field relations are not easily apparent. The non-recognition of a fault which has locally rotated strata may result in two beta maxima appearing on a plot if shear joint measurements from both sides of the fault are used, and an unrecognised local swing in the fold axis will produce equally misleading results. The use of only about 300 beta points in a plot is quite insufficient and may be a further reason for the appearance of double maxima, even in areas with a constant strike. This number of beta points may be obtained from only about 25 field measurements, and unless particular joint sets are selected this is barely enough to record every shear joint direction and dip in an area such as the present one (Fig. 8).

In order to test the basic accuracy and validity of the beta-plot method in a case like the present one a locality was selected at the northern end of Motutapu where the structural trend was constant. Seventy-five field measurements were made of all the shear joints present in this small area. This was found to be the maximum number of measurements possible for plotting without obliterating the location of intersections, and it resulted in 2,760 beta points.

The result is shown in Fig. 6b. A small maximum of a little over 4 per cent strikes to approximately 343° and plunges to NNW at about 18°. This agrees closely with the result shown on Fig. 6a based on bedding-plane measurements and

it bears out that the concept of the beta-plot is valid in principle, though its usefulness in areas of deformed rocks such as the present one appears to be very limited

The maximum in Fig. 6b is only about 2 per cent above the density recorded over much of the remainder of the diagram. This is no doubt due in part to a large number of spurious intersections which have no value but which inevitably occur in this type of plot. Ramsay (1964) has drawn attention to this shortcoming of the beta-plot method and has suggested that it be replaced by the pi-plot. Since in pi-diagrams the poles of planes are plotted, such records represent actual observations in the field and eliminate all spurious data. Furthermore, larger numbers of points can be plotted, giving the diagram a greater statistical accuracy, as Ramsay has pointed out.

For the purpose of comparison with the results obtained from the beta-plot (Fig. 6b), a pi-diagram was prepared from 386 shear joint measurements, about five times the number used in Fig. 6b, from the northern part of Motutapu. The diagram (Fig. 6c) shows two girdles. The normal to the dominant girdle trends 345°-350° and plunges NNW at approximately 22°. This again compares well with the results of Figs. 6a and 6b. The maximum in this girdle is far more substantial than that recorded in the beta-plot. The diagram also bears out the more frequent occurrence in this area of easterly-dipping low-angle joints. A second, less well defined, girdle indicates a fold axis aligned 175° and plunging south at about 20°. The pi-plot thus reflects clearly the field evidence of the presence of both northward and southward plunging folds in the fine-grained sediments in the northern part of Motutapu. This would not be easily apparent on a beta-plot, where because of the presence of many spurious intersections maxima tend to be smaller.

While these results support Ramsay’s view that the pi-plot is superior to the beta-plot method, the objections put forward earlier with regard to the use of the beta-plot in structural analysis of Waipapa Group sediments also apply to the pi-plot.

2. Tension joints. These are the most prominent joints in well exposed outcrops. They can be clearly followed over distances which are limited only by the extent of the outcrop. The joint planes are often curved over short distances and commonly dip at high angles of between 70° and 90°. Two prominent sets occur with an intervening angle of 90° between them (Fig. 7). The parallel planes of each set are usually spaced at regular intervals ranging from lin to Ift (PL 1, Fig. 1; PI. 2, Fig. 4). Tension joints may be present only as hair-line fractures in these rocks, but commonly they have formed clefts which may be up to several inches wide. Such fissures are sometimes filled with secondary quartz. Slickensides are absent from the surfaces of the joint planes.

Because of their prominent development and high angle of dip, tension joints can be clearly seen on air photographs where shear joints are invisible or only imperfectly visible. Fig. 7 shows the alignment of tension joints in various parts of Motutapu and Rakino Islands. It makes no attempt to record the frequency of joint sets in one particular area. In some localities one set of joints is better developed than the other. The joints follow two main directions, one of which is always parallel to the axis of folding, even to the extent of following a local curvature in the strike, as can be seen in the sediments just south of Mullet Bay, Motutapu Island. The second direction of jointing cuts across the strike or b-c joints at approximately right angles, forming cross, or a-c joints. In a few localities, joints were noted cutting obliquely across the trend of the fold axis. Apart from the close relationship of fold axes and tension-joint directions, shown in Fig. 7, the two dominant joint sets bear also a relationship to the two major trends of faulting in this area. Both the joints and the faults are predominantly tensional features, but the faults are of later formation.

The diagram in Fig. 6d incorporates 553 field measurements of tension joints from Motutapu Island. It shows a strong concentration in the vector 350° to 010°

and reflects the close adherence of this particular joint set to the dominant trend of the fold axes. The set of cross joints is represented by several maxima ranging between WNW and WSW. The diagram suggests a greater abundance of oblique joints than were actually observed in the field. The densities in the respective parts of the plot are mainly due to the incorporation of a number of strike joints from strata deviating in strike from the main NNW trend (Fig. 7).

The presence of similar tension-joint systems has been reported in sediments of the Waipapa Group from the Auckland and Northland areas (Brothers, 1956; Haler ow, 1964; Hopgood, 1960). Both Brothers (1956: 468) and Hopgood (1960: 196) observed in their respective areas of study that tension joints are well developed in thick sandstone sequences, but to a lesser extent in argillites. In this area it was found that, while this observation is correct where beds of argillites are interbedded with standstones, in thick sequences of argillites as, for example, on the northern coastline of Motutapu (PI. 2, Fig. 4), on Rakino, and on the islands of the Noisies Group, tension joints are at least as well developed as in coarse-grained sediments.

Faulting

Faults have been recognised in the field by their topographic expression, by narrow zones of extensive shearing, and by the displacement of lithological boundaries. It is likely that more minor faults are present in the area than have been detected. As Shown on the map (Figs. 2,7), faulting has occurred mainly along two dominant directions. One set of faults is aligned NW—SE and roughly parallels the trend of the major fold axes. A second set strikes approximately NE-SW and cuts the first set almost at right angles, thus forming a number of rectangular faultbounded blocks. On Rakino Island the cross faults strike almost east—west. Block faulting of this nature, but on a larger scale, is a common feature in sediments of the Waipapa Group elsewhere in the Auckland and Northland areas (Firth, 1930; Kear and Hay, 1961; Thompson, 1961).

The nature of the faulting has been predominantly normal. Although one fault plane only was observed in the field with a dip of 80°, it is believed that a high angle of dip characterises most of the faults. Some of the individual fault blocks have been subjected to some rotation about a vertical axis and have also been tilted, sometimes in opposite directions, which in some localities has contributed to an overturn of fold axes both to the north-east and to the south-west. Small transcurrent movement, a necessary accompaniment of fault-block rotation, is evident on the Mullet Bay Fault on Motutapu (Fig. 2). The strata on either side of the fault have been deflected so as to indicate movement in a sinistral sense.

The major fault in the area, the Motutapu Fault, has a well-marked topographic expression which indicates a throw towards south-west of at least 300 ft. The actual displacement is likely to be much larger, but the strata south-west of the fault, which strikes there at right angles to the major trend, cannot be correlated across.

The throw on the Home Bay Fault is difficult to determine accurately as it displaces very steeply dipping sediments along their strike (Fig. 4b). The displacement is, however, likely to reach I,oooft. Displacements on most other faults in the area are smaller.

Sequence and Age of Tectonic Events

Compressive stresses acting nearly horizontally in an ENE—WSW direction induced a period of major folding along a NNW-SSE axis plunging at a low angle to NNW. Folding occurred predominantly by means of bedding-plane slip (concentric or parallel type of folding) until, further lateral shortening by folding becoming impossible, the rocks responded to the continued deformational stresses by rupture. Sets of intersecting planes developed along which shearing and some slip movement occurred. The stresses involved do not seem to have been of sufficient magnitude to form thrust faults on a large scale. It is possible that the initial formation of shear joints commenced right at the onset of deformation and that shearing continued and was intensified while folding was in progress.

While some sets of shear joints can be directly related to the stress direction and the axis of folding, others do not seem to bear any such relationship and it is not possible to explain their origin in terms of a unified stress pattern.

Simultaneously with folding, tension joints developed in at least one direction —i.e., parallel to the main compression axis. The formation of tension joints perpendicular to the principal stress, which is believed to occur mainly during elastic release of compression (de Sitter, 1964: 106), may have taken place immediately following the first phase of deformation. Their regular spacing and close relationship to sometimes sinuous fold axes, a feature due in part at least to a second phase of tectonism, strongly suggests their formation at this stage.

A second period of compression, of milder force but with the principal horizontal stress acting in an unchanged direction, produced the small-scale folds in some low-angle shears and high-angle tension joints. Both Brothers (1956) and Hopgood (1960), who have investigated the structure of the Waipapa Group in small localities of Northland, mention two phases of deformation. Hopgood (1960: 202) is also of the opinion that the principal stress direction remained unchanged, but suggests an anticlockwise rotation of 30° following the first period of folding in order to explain the differing trends of the two generations of folds in that area.

As mentioned above, this phase of compression may have been partly responsible for producing local swings in the strike of some fold axes, and also of joint planes within the fold structures, by means of an uneven application of stress. In many localities the deviation from the major strike direction is due, in addition to this or wholly, to rotation of fault blocks. Renewed shearing and slip along established joint planes probably occurred during the second deformative period, but it is not known whether new shear planes were produced. The final cessation of compression most likely led to a more pronounced development of the two major sets of tension joints established earlier.

During uplift of the area, and with the principal stress acting in a vertical direction, faulting along two major directions formed fault blocks which, owing to differential movement along their predominantly normal faults, have in some cases been tilted or rotated. Faulting of this type is commonly found in areas of concentric folding (de Sitter, 1964: 187).

Most of the faulting seems to be unconnected in time with the development of tension joints, although the major fault and joint directions are near parallel and owe their origin largely to tensional stresses. The rotation of some fault blocks, which has also rotated the joints, makes it clear that the faulting postdates the formation of tension joints.

The age of the major folding of the Waipapa Group can be ascribed to the close of the Jurassic or the beginning of the Cretaceous. At this time the sediments along the whole length of the New Zealand geosyncline, of which this area forms a part, were deformed in the Rangitata Orogeny. The second period of deformation must have occurred between the Lower Cretaceous and the Lower Miocene, as rocks of the latter age have suffered no folding in this area. Dips of up to 5° in the sediments of the Waitemata Group are due partly to an initial slope and partly to compaction over the very irregular basement of Waipapa rocks.

Brothers (1956; 473) has given a summary of fold trends and ages of folding for the Auckland and Northland areas recorded up to that date. It appears from this that while Early Cretaceous fold axes generally follow a NW-SE trend, later folding has been reported from several localities in Northland along different directions. Both Battey (1950) and Mason (1953) have described folding of Lower Tertiary age along an east-west direction, which was followed in the Hokianga area (Mason, 1953: 368) by Upper Tertiary folding along a NW-SE axis.

The Lower Tertiary east-west folding in Northland deviates considerably from the trend in this area and, unless this orientation was brought about by a later largescale rotation of fault blocks, the stresses which produced the folding there are unlikely to have been responsible for the second period of deformation in this area, where the axis of compression remained unchanged. This would indicate that renewed deformation in this area followed closely the first period during the Early or Middle Cretaceous. The NW-SE folding of Upper Tertiary age reported by Mason (1953: 368) did not affect this area.

Block faulting on a large scale has affected many parts of New Zealand during the Kaikoura Orogeny, which appears to have commenced as early as the Miocene in Northland and Auckland and was of progressively later age as it moved southwards. The Motutapu Fault in this area does not displace sediments of the Waitemata Group and must, therefore, be pre-Lower Miocene in age. In the Hunua Ranges, however, the Wairoa Fault mapped by Firth (1930) clearly displaces sediments similar in age and lithology to the Waitemata Group in this area. It is likely, therefore, that movement along some faults in this area continued right through the Miocene. This is also suggested by the presence of faults within the Waitemata Group on Motutapu which follow the established trend of the Waipapa Group faults and have throws of up to 100 ft.

Acknowledgments

The author wishes to thank Dr P. F. Ballance and Dr R. N. Brothers of the University of Auckland for helpful comments and for reading part or all of the manuscript of a thesis for M.Sc. of which the subject matter of this paper formed a part.

References

Battey, M. H., 1950. The geology of Rangiawhia Peninsula, Doubtless Bay, North Auckland. Rec. Auck. Mus. 4: 35-59. Brothers, R. N., 1956. The structure and petrography of greywackes near Auckland, New Zealand. Trans, roy. Soc. N.Z. 83: 465-482. Brothers, R. N.j Golson, J., 1959. Geological and archaeological interpretation of a section in Rangitoto ash on Motutapu Island, Auckland. N.Z. 11 Geol. Geophys. 2: 569577.

Bryan, W. H.; Jones, O. A., 1962. Contributions to the geology of Brisbane, 3: The bedded cherts of the Neranleigh-Fernvale Group of south-eastern Queensland. Proc. roy. Soc. Qd 73: 17-36. Davis, E. F., 1918. The radiolarian cherts of the Franciscan Group. Univ. Cal. Pub. Dep. Geol. 11: 235-432.

de Sitter, L. U., 1964. Structural Geology. New York: McGraw Hill. Firth, C. W., 1930. The geology of the north-west portion of Manukau County, Auckland. Trans. N.Z. Inst. 61: 85-137. Grant-Mackie, J. A., 1960. Outline geology of the inner islands of the Hauraki Gulf. Proc. N.Z. ecol. Soc. 7: 23-27.

Halcrow., H. M., 1956. The geology of Waiheke Island. Trans, roy. Soc. N.Z. 84: 51-69. Hochstetter, F. voNj 1864. Geology of New Zealand, trains. by C. A. Fleming. Wellington: Government Printer.

Hopgood, A. M., 1960. Notes on the structure of greywackes and argillites at Tawharanui Peninsula, Auckland. N.Z. Jl Geol. Geophys. 3: 192-202. Hornibrook, N. de 8., 1951. Permian fuisilinid Formaminifera from North Auckland Peninsula, New Zealand. Trans, roy. Soc. N.Z. 79: 319-321.

Rear, D.; Hay, R. F., 1961. Sheet INorth Cape, Geological Map of New Zealand, 1:250,000. Wellington: D.S.I.R. Leed, H., 1951. Permian reef-building corals from North Auckland Peninsula, New Zealand. N.Z. Jl Sci. Tech. 833: 126-128. N.Z. geol. Surv. paleont. Bull. 25: 15-24.

Mason, A. P,, 1953, The geology of the central portion of Hokianga County, North Auckland. Trans, roy. Soc. N.Z. 81: 349-374. Milligan, E. N., 1959. Some fossils from hitherto undifferentiated Permian-Triassic-Jurassic rocks near Auckland. N.Z. Jl Geol. Geophys. 2: 195-198. Park, J., 1887. Kaipara and Wade districts, Auckland (Rodney, Waitemata, and Eden counties). N.Z. geol. Surv. Rep. geol. Explor. 1886-7: 219-229.

Ramsay, J. G., 1964. The vises and limitations of beta-diagrams and pi-diagrams in the geometrical analysis of folds. Q. Jl geol. Soc. Land. 120: 435-454. Reed, J. J., 1957. Petrology of the Lower Mesozoic rocks of the Wellington district. N.Z. geol. Survey Bull. 57: 1-60.

Thompson, B. N., 1961. Sheet 2A — Whangarei, Geological Map of New Zealand, 1:250,000. Wellington: D.S.I.R. Waterhouse, J. 8., 1963. The Permian faunal succession in New Zealand. ]. geol. Soc. Aust. 10: 165-176. Williams, H.; Turner, F. J.; Gilbert, G. M., 1958. Petrography. San Francisco: W. H. Freeman and Company.

W. Mayer, Geology Department, University of New England, Armidale, N.S.W. Australia.