Geology of the Southern Waitakere Hills Region West of Auckland City.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 74, 1944-45, Page 49

Geology of the Southern Waitakere Hills Region West of Auckland City. By E. J. Searle, Auckland University College. [Read before Auckland Institute, September 15, 1943; received by the Editor, September 27, 1943; issued separately, June, 1944.] Abstract.—A short account is given of the geology of a region west of Auckland composed mainly of the Tertiary volcanic Manukau Breccias of the southern half of Waitakere Range. These are underlain conformably on their east by sediments of the Tertiary Waitemata Series which are covered locally by Pleistocene clays. The writer deals also with the petrography and origin of the Manukau Breccias and of clastic dykes that they include. Introduction. This outline of the geology of the southern portion of the Waitakere Hills, west of Auckland City, is based on reconnaissance work carried out 10 years ago under the guidance of Professor J. A. Bartrum and submitted as a thesis for the Master's degree. No survey of the geology of the whole area had previously been attempted, although from the earliest days of New Zealand geology parts of the region, notably between Onehunga and Manukau North Head, have received attention, particularly from Hochstetter (1864) and Hutton (1870). Other writers on limited aspects of the geology have been Smith (1881), Cox (1881, 1884), Park (1886, 1889), Fox (1902), Mulgan (1902) and Bartrum (1923, 1926, 1937). General Description of Area. The Waitakere Range, which constitutes most of the region described, is a plateau composed essentially of fragmental volcanic rocks, which has a general elevation of about 1100 feet, but is surmounted by occasional peaks of which Te Torokawharu (1506 feet), north-west of Huia Bay, Manukau Harbour, is the highest according to Mr. A. D. Mead, Waterworks Engineer, Auckland City Council* Personal communication.. Vigorous short streams have carved characteristic precipice-walled deep gorges, causing very strong relief, and have very steep courses, often with falls as much as 80 feet in height. A canyon in the lower course of Piha Stream on the west coast is a remarkable cleft, maintaining a width of only 11 feet for a height of 200 feet above the stream bed before widening. The larger streams wander on wide alluvial flats between steep valley walls near their mouths; this clearly is a result of the recent submergence that is so well exemplified throughout most parts of Auckland Province. Dykes and inextensive thin flows vary the general fragmental material, but rarely affect the longitudinal profiles of the streams; thick pillow lavas furnish an exception, however, and give rise to falls at the head of the lake impounded by Nihotupu Dam. The east face of Waitakere Range descends very abruptly to lowlands along a line running approximately north from Puponga Point, a tongue-like promontory on the north coast of Manukau Harbour; this scarp appears not to be due to faulting but to sudden change in the resistance to erosion of the rocks concerned, for east

of the line mentioned the resistant volcanic rocks are replaced by soft sediments of Tertiary and later age. The drainage is insequent both in the western highlands and the eastern lowland region, for in this latter the structure is confused, so that, although there is some variation of hardness in the Tertiary rocks, no directional control of stream direction has been possible. The streams of this eastern region are gently graded and notably in the case of Whau Stream end in lengthy tidal creeks in consequence of the recent submergence noted above. Along the west coast the powerful waves of Tasman Sea have eroded seacliffs usually many hundreds of feet in height, although over a considerable portion of this length of coastline the bases of these cliffs are protected by beach and wind-blown sands. A remarkable feature of the local geology is the abrupt southern termination of Waitakere Range and its replacement by the waters of Manukau Harbour and its fringing low flats; these latter are underlain by Pleistocene silts capped locally by the products of basaltic eruptions and pass west into a highly elevated range of poorly consolidated Pleistocene dune and other sands (Gilbert, 1921). Hochstetter (1867) ascribes this abrupt transition to the presence of a north-east—south-west fault margining the northern shore of Manukau Harbour and its existence is also accepted on considerations of broad regional structure by Bartrum (1937); it cannot be traced, however, beyond the narrow isthmus that separates west from east coast sea waters about three miles east of Onehunga. It had long been believed that the volcanic rocks of Waitakere Range terminate at this fault, but Mr. C. W. Firth† Personal communication to Professor Bartrum. recently has found that they outcrop at sea level at Clark's Beach and Te Toro on the southern shores of Manukau Harbour, 10 miles south of their nearest northern outcrop. Topography. As will be shown later, the deposition of the volcanic mass of the Waitakere Hills followed closely upon that of the Tertiary (Waitemata Series) sediments further east. Broadly considered these hills are a one-cycle range that owes its elevation above adjacent lowlands to the superior hardness of its materials. Nothing need be added to what already has been stated about the topography of this highland, but it is different for the eastern lowlands, for these show erosion levels comparable with those of the Silverdale-Takapuna area described by Turner and Bartrum (1929) and indicative of discontinuous uplift. The highest of these erosion levels is a much dissected surface about 600 feet above sea level near Titirangi and north from there fringing the steep face of Waitakere Range; it is part of the dissected peneplain of Auckland recognised by Cotton (1922) and Bartrum (1923) and is a product of the widespread late-Tertiary peneplanation that has been recorded by various writers from many parts of North Island, as, for example, Bell and Clarke (Whangaroa, 1909), Grange (North Taranaki-South Auckland, 1927), Bartrum and Turner (North Cape, 1928) and Ongley and Macpherson (near East

Cape, 1928). Benson (1935) has discussed the relation of the peneplanation of this date to that of Cretaceous age in southern New Zealand. If, as appears to be the case, Pliocene rocks of the Wanganui-Rangitikei basins have been truncated during this period of erosion, its date cannot be earlier than later Pliocene. Other erosion levels include one at 350 feet above sea level which is well displayed near Oratia, two miles west of Glen Eden, and another at about 120 feet which is prominent on the shores of the harbour at Auckland. The most striking erosion level of the present area, however, is in the New Lynn-Point Chevalier district and varies in elevation from 40 feet to 60 feet above sea level. It has its maximum extension in a peninsula 3 ½ miles long on the east bank of Whau Stream. Locally, as at Whau Creek near New Lynn, it has been carved into a stream terrace about 20 feet above sea level. It is underlain either by Tertiary strata or by silts and lignitic shales of Pleistocene age. A minor feature of topographic interest is the existence of a relatively broad floodplain above a temporary baselevel afforded by a lava barrier just upstream from Western Springs in the valley of Motion's Creek, which enters Waitemata Harbour immediately east of Point Chevalier. Oakley Creek shows an exactly similar floodplain. Mention may finally be made of two circular shallow hollows about 2 chains across on the east bank of Whau Creek near its mouth; they appear to represent sinkholes comparable with those described by Henderson and Grange (1926) from near Ngaruawahia where the surface collapse that has occurred is the result of the removal of incoherent sands below the surface by groundwater streams. Coast. Recent submergence of North Auckland and Auckland regions is so well known that there is no need to detail the evidence of its occurrence. Turner and Bartrum (1929) show that it followed a sharp uplift of the same order of magnitude as itself at a date subsequent to the outbreak of early members of the cluster of basaltic volcanoes characterising the Auckland isthmus. A bore sunk to a depth of 125 feet below mean high water level in Waitemata Harbour south of Northcote Peninsula failed to pass through the Recent silts that have filled an ancient channel excavated during the phase of uplift referred to.* Information supplied by Mr N. L. Vickerman, Assistant Engineer, Auckland Harbour Board. This implies submergence of the order of not less than 150 feet. At first sight the relatively straight coastline on the west of the area now described offers no evidence of submergence, but on close survey the evidence required is supplied by the drowned mouths of some of the larger streams, such as Piha Stream. Presuming that initially this shoreline was deeply embayed, its present extremely simple nature, which has almost reached maturity in the shoreline cycle, shows that there has been very considerable retrogression of the coastal cliffs since submergence was completed. Between North Head and Karekare these cliffs are now protected by a belt of sand 5 miles long and nearly three-quarters of a mile in maximum width (near Ohaka Head). Locally there were once lagoons behind barrier

beaches or spits, but these have now been filled by swamp or windblown material. This recent progradation is due to abundant supply of sand from both north and south of the area as a result of wave erosion of a coast formed of soft poorly consolidated sands, largely aëolian in origin, which build plateau-like ranges as much as 600 feet and more in height. From Karekare north there are precipitous sea-cliffs (see Pl. 12, Fig. 8) sometimes as much as 600 feet in height interspersed with sandy bays where the story of fairly recent progradation may again be read, as, for example, in a long, forested foredune at Piha and a small more recent one at White's Beach south of the mouth of Anawhata Stream. Minor shoreline features include the usual stacks and reefs, while “tunnels” and caves are frequent both in ancient and modern sea-cliffs and usually have been excavated along fault or major joint fractures, though occasionally along dykes with transverse columnar jointing which have proved a ready prey to the waves. There is an excellent group of six of these “tunnels” and caves separated by a wide belt of sand from the sea near Windy Point, 2 miles north of Manukau North Head, which were regarded by Smith (1881) as indicative of recent uplift of the shore relative to sea level; there is no evidence, however, that their floors, now buried beneath sand, are above sea level. Bartrum (1937) has described from the sea-cliffs at Mercer's Bay north of Karekare a vast natural shaft descending 300 feet to sea level from its mouth high up on the cliffs and with a width of from 75 feet to 90 feet. A tunnel 200 yards long was carved along a zone of close-spaced sheeted fractures until it intersected a transverse dyke 12 feet wide with prominent horizontal columns which have collapsed above the tunnel and thus caused the shaft. At the base of most of the headands of the coast-line there are horizontal platforms eroded from the resistant conglomerates even where they are bedded in inclined sheets and normally from 2 feet to 4 feet above mean high water level; they descend steeply at their outer edges into water seldom less than from 15 feet to 20 feet in depth (see Pl. 11, Fig. 4). Landwards they commonly are overhung by the sea-cliffs, which are seen to be suffering excavation at their bases. Their surface is sometimes rough and irregular, but on wider platforms in any but the extremely coarse fragmental beds it normally is level and covered by shallow pools of water. Hutton (1870) regarded them as proof of recent shoreline emergence, but Bartrum (1923, 1926, 1935) has claimed that they are the product of storm waves of the sea at its present stance, while he has ascribed the frequent planation of their surfaces to subaërial processes which have reduced elevations to the levels of adjacent pools of water. Wentworth (1938) at Hawaii and Jutson (1940) from Port Philip Bay, Victoria, have described shore platforms of closely similar type; neither of these writers has made reference to the prior work of Bartrum. The latter's theory of origin appears to receive support from the fact that the only platform of this type in Manukau Harbour, apart from one at North Head, is eroded in volcanic conglomerate at Puponga Point where it meets waves driven through the harbour entrance by westerly gales.

Stratigraphy. Synopsis. The oldest rocks actually exposed in the area are the Lower Miocene sediments of the Waitemata Series, but the existence of a buried older basement is afforded by pebbles in bands of conglomerate present in these sediments; this older basement is constituted by Mesozoic greywackes and Onerahi rocks (? Mid-Eocene). The Waitemata strata are exposed only in the eastern lowlands and westwards pass under the volcanic rocks of the Manukau Breccia Series. Although the two series are conformable one with the other, occasional disconformities indicate some variation of conditions of deposition whilst the complete sequence was accumulating. Near Avondale, New Lynn and Point Chevalier, Pleistocene clays, silts and occasional lignitic layers, rest unconformably on the Waitemata beds. They are shallow-water floodplain deposits accumulated in deep valleys exeavated in the Tertiary terrain. They have been affected by later members of the compressional movements that have caused severe local folding and thrusting of the underlying Waitemata beds, for near New Lynn they are gently and occasionally even fairly steeply inclined. In addition, at Cornwallis, on the east coast of Puponga Point, Manukau Harbour, there are poorly consolidated dune sands of Pleistocene age rich in titaniferous magnetite and referable to the “Lignite Formation” of Hochstetter (1867) which is prominently developed in a high sand range which runs south-south-east for over 25 miles from Manukau South Head to the mouth of Waikato River (Gilbert, 1921). Buried Mesozoic and Early Tertiary Basement. Greywacke pebbles up to 1 ½ inches in diameter occur in a thin band of conglomerate of the Waitemata Series (Lower Miocene) at Blockhouse Bay, Manukau Harbour. Greywackes outcrop freely about 20 miles to the north-east and east in the islands of Hauraki Gulf and on the mainland at Maraetai (Firth, 1930); Ferrar (1934) has referred these rocks in the adjacent northern Rodney Subdivision to the Waipapa Series of probable Jurassic age. Angular blocks of poorly foraminiferal argillaceous limestone ranging up to 3 ½ feet in diameter, though usually much smaller, and closely similar to the limestone of the Onerahi Series (see Ferrar, loc. cit.), which Finlay and Marwick (1940) have assigned to the Bortonian (Middle Eocene), occur in Waitemata sediments at various places on the Manukau north coast, as at Hillsborough Bay and a little east of Blockhouse Bay. These blocks clearly have not been transported far, for their material is very soft; they indicate, therefore, the existence near at hand in Waitemata times of a moderately elevated coast composed of this limestone, although it is not known to-day to outcrop nearer to Auckland than Dairy Flat and Okuru, about 15 miles north of Auckland City. These included fragments represent part of the evidence used by Park (1886), Bartrum (1924) and Bartrum and Turner (1929) in support of erosional unconformity between the Waitemata sediments and those of the preceding Onerahi Series. The extent of erosion must have been considerable for ammonite-bearing concretions derived from the Danian Otamatea Series (Ferrar, 1934), which underlies the Onerahi Series, occur in Waitematan conglomerates (Albany conglomerates of Bartrum,loc.

cit.) at Hawke's Quarry near Kaukapakapa, 20 miles north-west of Auckland.* Oral communication from Professor J. A. Bartrum. Waitemata Series—Hutchinsonian (Lower Miocene). As near Auckland, the typical rocks are friable yellowish-brown feldspathic sandstones interbedded with light-grey to bluish thinner mudstones often well laminated in character. They contain few determinable fossils other than Foraminifera. There are, however, variants from this general type of sediment; for example, at Waikowhai and Blockhouse Bays on the Manukau Harbour there is a 12-feet band of intraformational conglomerate with boulders as much as 1 foot and more in diameter and composed of sandstone indistinguishable from that of adjacent Waitemata beds. Then on the western side of Puponga Point boulder-like masses as much as 1 foot across of white marly mudstone encased in a thin skin of volcanic tuff occur in false-bedded sediments which here include lenses of coarse tuff. These “boulders” evidently are mud balls rolled by currents in shallow water. About half a mile west of this occurrence there is distinct discontinuity of bedding indicated by the superposition of a gently inclined series of argillaceous beds upon a steep, undulating surface of gritty strata. Shallow-water conditions of deposition are demonstrated again by pebble bands in various places such as Blockhouse Bay, Shag Point and Titirangi Bay, while at the eastern head of Onehunga Beach on the same northern coast of Manukau Harbour as the last localities there is a thin bed (noted by Park, 1886) which contains rounded pebbles of pumice up to ½ an inch in diameter, as well as worm tubes and carbonaceous matter. Evidence of fairly general shallow-water deposition is thus common, yet Mr. W. J. Parr, of Melbourne, considers that the Foraminifera of some of the finer-grained beds suggest deeper waters than obtained for any others of the numerous foraminiferal horizons near Auckland that he has examined.† Personal communication. Some of the pebble bands mentioned deserve further consideration. At Blockhouse Bay, about 200 yards east of Duck Creek, a 5 foot bed outcrops, dipping at 50° to the north-north-west; it has pebbles as much as 2 ½ inches in diameter at its base, but becomes progressively finer upwards until it grades into foraminiferal sandstone. In it are greywacke of the Waipapa Series (Ferrar, 1925), amygdaloidal basalt and biotite dacite or rhyolite; this last probably is the same rock as the rhyolite that Bartrum (1924) records as fragments in Waitemata beds at Riverhead 14 miles north of Blockhouse Bay. No similar rock is known in situ within at least 20 miles of Auckland. At Shag Point, about 300 yards south-west of Titirangi Beach, several pebble bands from 1 foot to 4 feet in depth alternate with sandstone in a series dipping at 40° to the east-north-east. Mulgan (1902) noted one of the bands and regarded it as a phase of the Parnell grit which is described below. The presence of gneissic diorite and epidiorite amongst the pebbles, however, indicates that correlation is with the Albany conglomerates (Bartrum, 1920, 1924), in which gneissic diorites are most characteristic included rocks. Typical Parnell grit is indeed present, for the pebble series has been over-thrust upon it along a plane that dips at 30° to the north-west.

The Parnell grit has been described so often by various writers and its origin so freely discussed (see, for example, Turner and Bartrum, 1929) that no good purpose can be served by considering it in detail in this paper, particularly as no fresh facts have come to light. The main occurrences are a 10 foot bed dipping to the south-east half a mile west of White Bluff and two yellowish bands 13 feet thick separated by 25 feet of normal Waitemata sandstone between Waikowhai and Mission Bays, half a mile west of the previously-mentioned outcrop. There are also five exposures of the same type of rock on the shore between Titirangi and Little Muddy Creek and just west of Shag Point there is a further outcrop in which a 3 foot bed with large blocks of white foraminiferal marl separates basal coarse grit from an upper stratum of coarser nature. In addition Mr. C. W. Firth has informed the writer that no fewer than four beds of this characteristic rock were encountered in a drainage tunnel 1 ½ miles long between New Lynn and Karaka Bay, south of there on the Manukau coast. In typical Parnell Grit there are variably sized fragments of volcanic rocks (mainly andesites) and of black shale and sandstone and, as a rule, numerous broken polyzoans. These last are particularly abundant in the beds at Mission Bay and include both slender and massive cup-shaped forms, these latter often over 2 inches in diameter; in addition there are fragments of Pecten and other molluscs and of corals, echinoids, worm tubes and brachiopods. Such fossils as these, which rarely are specifically identifiable, are usually the only larger forms present in the Waitemata beds throughout the whole of their extent and the only satisfactory macrofauna yet derived from them near Auckland is that at Waiheke Island, which Powell and Bartrum (1929) have referred to the Hutchinsonian (Lower Miocene). Foraminiferal faunas are, however, obtainable from a number of localities; samples from 13 of these were submitted to Mr. W. J. Parr, of Melbourne, who kindly made determinations of species. Dr. H. J. Finlay's recent intensive work on New Zealand Foraminifera, however, has made revision of the determinations necessary, and the writer is greatly indebted to Dr. Finlay for undertaking this onerous task and furnishing the check list of nearly 160 species that is published as an appendix to this paper. The Manukau Breccia Series. General Description. Beds of this volcanic series have been noted and discussed, particularly as regards their relation to the Waitemata Series, by Hochstetter (1864), Hutton (1870), Cox (1881, 1882) and Park (1886, 1889). Cox and similarly Park in his later report (1889) believed that they are an upper conformable continuation of the Waitemata Series due to culmination of the same series of volcanic eruptions as gave rise to the Parnell Grit. Bartrum (1924) found no indication of unconformity between the breccias and the Waitemata sediments near Kaukapakapa, north of the present area, but later Turner and he (1929) noted that in a water-supply tunnel at Huia, on the Manukau Harbour, the volcanic beds rest on an uneven surface eroded from Waitemata sediments. The conclusions reached during the present study are given in a later section of this paper.

The rocks of the Waitakere Range, which commonly are referred to as the Manukau Breccia Series, are andesitic fragmental beds with numerous intrusive dykes and minor interbedded flows. They cover nearly 90 square miles in the area now described and continue to the southern arms of Kaipara Harbour, about 35 miles north-west of Auckland. Rock masses similar in petrographic and other characters are extensively developed in Coromandel Peninsula and its northern extension in Great and Little Barrier Islands and in the northern portions of Auckland Province, as at Whangarei (Ferrar, 1925), Hokianga Heads, Whangaroa and North Cape (McKay, 1894). In grade the various beds of this series vary in the Waitakere Ranges from fine grained, well-bedded tuffs to coarse breccias with blocks as much as 5 feet across (see Pl. 11, Fig. 1; Pl. 12, Fig. 8). The constituent fragments usually are angular, but nevertheless there frequently are beds of well-rounded boulders and finer material which usually show good assorting and large-scale false bedding. These latter characters, with common sub-horizontal bedding, suggest that they are beach deposits; further evidence in favour of marine deposition of the series will be given later. Petrographically there is remarkable uniformity of rock-type; andesites alone are represented and pyroxene varieties of these rocks preponderate. The beds are considerably broken by joints and other fracture planes which have a general east-west direction in the southern portion of the coast, but movement has seldom taken place along these fractures for the dissevered portions of boulders on opposite sides of them are rarely displaced. The joint crevices often are filled by calcite; opal occurs as the filling at Whatipu and Puponga with the addition of zeolites at this last locality. Faults are not entirely absent and at Paratutai, Manukau North Head, an example occurs where they are referable to two periods, for two small dykes have been cut by the same normal fault, which has displaced the upper dyke little more than 3 feet although the lower and evidently the earlier dyke has been moved 10 feet. Because of the highly competent nature of the beds, folds occur very rarely in the breccia series, although local belts of acute folding and thrusting are common in the weaker underlying Waitemata sediments in most regions of their occurrence (see Pl. 11, Fig. 2). In ancient sea cliffs about a mile north of Pararaha Stream, on the Tasman coast, however, fine-grained bedded tuffs with occasional lenses of conglomerate have been crumpled into what is largely a confused jumble of relatively small blocks of strata (see Pl. 11, Fig. 3). In the centre there is a series of almost vertical S-shaped folds separated by a fault from a triangular mass which appears to represent a large S-shaped fold of which the crest has been out-squeezed with the production of minor drag folds. As the section is followed north, close-spaced variously-inclined thrust planes dipping to the north-west succeed the folds and in their direction parallel the chief structural direction noted by Turner and Bartrum (1929) in Waitemata beds near Auckland. The finer-grained tuffs usually constitute relatively thin, finely laminated strata interbedded with beds of coarse débris which predominate amongst the fragmental beds of the present district in

contrast with the relation that obtains in the northern half of the Waitakere mass, where massive tuffaceous sandstones are more abundant than coarse clastics. A particularly good exposure of these fine-grained tuffs occurs on the West Coast Road not far east of its junction with Piha Road. They contain material petrographically identical with that of the coarser clastics and include Foraminifera near Nihotupu Dam and on the West Coast Road 1 mile east of Waiatarua, though by no means so freely as the massive tuffaceous sandstones of the northern half of Waitakere Range beyond the present area. Dr. H. J. Finlay reports that tuffs collected by the present writer from the Scenic Drive about 1 mile north of Titirangi contain indeterminate Globigerinidae with siliceous forms which apparently have come from Cretaceous strata disrupted by the volcanic eruptions. Bedding discontinuities often appear in the tuffs and at Henderson Valley there is a decided unconformity comparable with one figured by Laws (1931) from the Tertiary sediments near Papakura, for upper beds of fine tuff dipping gently to the south-west, rest on the truncated inclined edges of lower beds which dip more steeply to the north-west. Clastic Dykes. A number of interesting clastic dykes intersect the breccias at various points on the Tasman coastal section, though concentrated mainly about three-quarters of a mile north of Manukau North Head. They appear to occur only in ancient or modern sea-cliffs and have not been observed in the many excellent sections afforded by stream gorges further inland The light colour of many of them contrasts with the sombre tone of the breccias and they were duly noted by Hutton (1870). More recently Taylor (1930) has made a detailed study of seven of them and has stated that they are “… clastic dykes the material of which has been forced up from below during a state of tension in the over-lying breccia.” As Taylor's view of their origin is not accepted by the writer it will be necessary to describe individual dykes in some detail. The most prominent of the dykes (A of Taylor) appears in abandoned sea cliffs about three-quarters of a mile north of the Heads as a steep, Y-shaped, light-coloured mass 8 feet wide at the fork of the Y 60 feet above the base of the cliff, but only three feet in width at the base (see Pl. 12, Fig. 5). The northern fork of the Y continues upwards for a further 40 feet or more to the top of the cliff, but the other soon pinches out. The filling is soft, fine-textured whitish tuff with numerous fragments of acidic pumice which usually are small, but occasionally may be as much as 2 inches in diameter. Dyke B of Taylor is a few yards further north, and is an irregular sheet, 18 feet in width below, which rises vertically for 70 feet. It is filled for 30 feet from ground level by material indistinguishable from that of the breccia of its walls and includes blocks of andesite as much as 18 inches across. There is then a horizontal joint plane and the filling thereafter becomes progressively finer in grain. The basal 10 feet of this finer material is andesitic and shows faint but distinct bedding, the lower layers being sub-horizontal and gently convex viewed from above, but the higher steeply inclined as if an open fissure had been filled by material rilling from above perhaps with

the assistance of rain waters. Above this purely andesitic material much of the white pumiceous filling noted in Dyke A enters as a constituent. At a point 40 feet above the base of the cliff, the dyke divides into two branches which pass around a narrow horse of breccia 30 feet in length and then reunite, while the dyke flattens to dip at 30° to the south-south-west as it passes to the top of the cliff. A few yards further north the next dyke (C of Taylor), which is only a few inches in width, contains fine-textured white fragmental material in its lower portions, but as it gains height the size of fragment becomes larger and the white constituents become eliminated from the general andesitic filling. Dyke D of Taylor is 3 feet wide where it rises steeply from floor level at the south-west corner of the “Big Cave,” a familiar landmark of the district, but narrows rapidly to 6 inches within the height of 3 feet; it then as rapidly expands to 30 inches and maintains this width to where it outcrops 30 feet up on the top of a small promontory. Similar irregularities of width characterise it along its strike in the cave. The walls of the fissure that it occupies show vertical slckensides and 2 feet vertical displacement of one wall with respect to the other is demonstrated by the displaced halves of a dissevered boulder. Both in this and the next dyke (E), which rises ill-defined from the roof of the Big Cave, the filling includes white pumiceous fragments in the lower portions, although higher it is indistinguishable from adjacent breccia. Another dyke 2 chains south of A has not been described by Taylor (loc. cit.); it is a regular sheet 6 inches in thickness which extends 30 feet to the top of a low cliff. It differs from previous dykes described in that the lowest 8 feet of its filling is of andesitic fragments although higher up white pumiceous material dominates. In the low cliffs at the northern head of the sand bight at Whatipu there is a further infilled fissure 10 feet in width which was noted by Taylor; its filling is fairly fine andesitic tuff without any of the white material. A few of these clastic dykes also occur in sea cliffs north of the present area, while at Windy Point, about half a mile north of the main group of dykes, there is an example which differs markedly from all others in its characters, though regarded by Taylor as analogous to the others (see Pl. 12, Fig. 7). It is a wedge-shaped sheet which is 5 feet in thickness at its upper limit, where it dips east-south-east at 20°, but decreases in width downwards with accompanying flattening of dip until it pinches out at the base of the cliffs. Northwards it makes contact with a mass of highly angular material which Taylor, following Hutton (1870), regarded as denoting a fault zone. This is not so, for the material is agglomerate which marks the site of an early eruptive vent and includes coarse fragments of glassy slaggy lava, others of oxidised greatly vesiculated rock and angular blocks of poorly vesicular andesite identical with that in the breccias near at hand and sometimes as much as 30 inches by 10 inches in exposed dimensions. In addition there are fragments of brown andesitic pumice and of highly pyritised andesites, while tuff occurs both as scanty matrix and as large blocks. Perfectly shaped crystals of augite reaching ½ an inch in length abound in the tuff, while a thin-section shows in additiongreen-

ish hornblende and hypersthene. Probably the most instructive constituents of this agglomerate, however, are spirally twisted fusiform lava bombs; one that was measured was 1 foot in length and weighed 22lbs. The filling of the adjacent dyke is identical in character with the material of the agglomerate, except that it is not so coarse, and contrasts with that of all other dykes. It clearly represents superficial ejecta which have gravitated into an open fissure produced by magmatic heaving, and it is unfortunate that Taylor (loc. cit.) should have regarded this particular dyke as similar to the others. As already mentioned, Taylor believed that the dykes are tension fissures filled from below by injection. The source of the filling in his view was crystal tuff which underlay the breccias that are the country rock of the dykes and was exposed just north of Windy Point, where upthrown by the fault that he and Hutton (1870) postulated at this place. Tuffs certainly occur locally, but do not contain the white material characteristic of the dykes which was believed by Taylor to be andesite selectively kaolinised because of its pumiceous nature. Microscope study shows that this is not so and that the material is acidic (see Pl. 12, Fig. 6). The writer considers that Taylor's hypothesis of origin of these dykes is unsatisfactory for several reasons:— 1. Apart from their occurrence in the dykes, the white pumiceous constituents of the filling are absent from the great mass of fragmental rocks that constitutes Waitakere Range. 2. The mode of filling of dyke B seems incompatible with injection from below on account of bedding layers which suggest strongly that the material rilled in from above. 3. Unless its fragments were “rafted” by muds of which no trace now exists, it is difficult to understand how the very coarse filling at the base of dyke B can have been emplaced by injection from below. 4. In most of the dykes the pumiceous filling is in the basal portions and is replaced at higher levels by andesitic material. In one instance this relation is reversed. This fact is more readily explained by a theory that invokes filling from above than by Taylor's hypothesis. 5. The dyke at Windy Point has been shown almost unquestionably to represent a fissure filled from above. The writer believes that fissures developed probably as a consequence of broad compressional arching and that these were filled by material washed in by rain waters and fallen from the upper weathering edges of the fissures. Dyke B is strong evidence in support of this belief, while that at Windy Point has an analogous origin. Open fissures appear occasionally on the tops of some of the sea cliffs of the area where erosion has freed the rock from its cover of soil. Under the writer's hypothesis the explanation of the pumiceous filling of the dykes presents no difficulty. Taylor (1927) records a deposit of similar pumice near Te Henga, 10 miles north of the main group of dykes, and Professor Bartrum has informed the writer that, a mile south of Te Henga, a subaërial deposit that includes rhyolitic pumice outcrops at high levels above the sea cliffs and that pumice also enters into the constitution of the elevated range

of Pleistocene sands between Manukau North Head and Port Waikato. It is evident, therefore, that pumice was freely distributed in the area during the Pleistocene; it could readily have been washed into any open fissures by rain waters. Huge Blocks of Tuff in Breccias at Whatipu. Hutton (1870) remarked on colossal blocks of fine-grained sediment (which he believed to represent sandstone of the Waitemata Series) in the normal breccias at Manukau North Head. One such block measures 40 feet by 30 feet by 20 feet, while another larger one nearby is over 2 chains across its bedding and has even greater length. They lie in various attitudes and are not of Waitemata sandstone but of andesitic tuff identical with that of similar beds of the breccia series. The great number of dykes in the cliffs of the locality suggests that the site of the vent that upheaved these blocks lay close at hand. Dykes and Sills. Several of the earlier writers on the area noted the numerous dykes in the Manukau breccias at Paratutai, the knob-like island at North Head, and two parallel sheet-like masses were regarded by Hutton (1870) as interbedded flows; they have glassy selvedges on both upper and lower surfaces and are clearly transgressive on the opposite side of the island from that where they accord with the bedding. They are thus dykes continued as sills. Occasional thin sills were noted by the writer, but the vast majority of the very numerous intrusions are dykes; most of them are less than 6 feet across and very few exceed 20 feet. They may occur wherever there are coarse fragmental beds, but have a distinct tendency to cluster in groups as if close to eruptive centres. In spite of the coarseness of the breccias which they invade, dykes less than 1 inch in width may be followed for several yards, and one or two less than 3 feet thick were traced for nearly a quarter of a mile. The dyke rocks are bluish-black pyroxene andesites which vary little in mineralogical constitution but range in texture from finegrained aphyric types to others with phenocrysts of feldspar or pyroxene as much as ½ an inch in length. Two dykes which are only 12 feet apart in the south branch of Piha Stream, three-quarters of a mile above its confluence, illustrate these extremes of texture; they probably represent injections that differ in date, but other explanations are possible. Some of the dyke rocks are moderately, though not highly, vesicular and do not appear to have occluded any noteworthy quantities of volatile constituents, for wall rocks are unaltered in all cases. The vesicles of a dyke ½ a mile north of Lion Rock, Piha, have a lining of oxide or hydroxide of manganese. In other cases there may be a pellicle of a bright-blue material which was early noted by Hochstetter (1864). It is also common in the vesicles of lavas but is in so thin a film that its composition could not be determined, although it was found to contain iron. Flows. Earlier writers have not recorded the occurrence of flows, although Bartrum (1930) evidently was aware of their existence. They are not as conspicuous as the dykes, but are by no means infrequent, particularly at Karekare and Piha and along West Coast Road; most of them appear to be relatively thin and inextensive. A narrow deep flow associated with an interesting breccia beneath

it outcrops in a promontory called “The Watchman,” which terminates in lofty bluffs on the north side of Karekare Stream. It is limited on either side by what is taken to be a friction breccia, for it is composed wholly of fragments identical in character with the rock of the flow, which is a characteristic vitrophyric type with flow banding made prominent by parallel lenticular streaks of whitish glass. The breccia beneath the flow is unlike the usual breccias, for it lacks bedding and consists of angular fragments of the characteristic overlying lava set in a matrix of red oxidised ash. It is believed to be an agglomerate deposited very close to an earlier centre of eruption. One of the best exposures of flows is a series of six, some as much as 50 feet in depth, in sea cliffs 500 feet in height 1 ½ miles north of Piha Stream. Although Bartrum (1930) has described pillow lavas from Muriwai and informs the writer that another thick mass occurs near Te Henga north of the present area, none of the western flows observed by the writer were of this type. He noted them only at the head of the lake impounded by Nihotupu Dam in a mass about 50 feet in depth. Only the upper part of the exposure shows pillows and the rock below this appears to have columnar jointing. Three fan-like columnar structures are exposed in a quarry near at hand and it is possible that they, in common with the columnar lava beneath the pillows, have been formed by fairly slow upwelling of lava to form dome-like masses in the manner postulated by Bartrum (loc. cit.) for comparable structures at Muriwai. The interstices between pillows of the Nihotupu flow are occupied by tuff, while large vesicles in the lava contain opal or calcite, or, more rarely, chabazite or aragonite. Foraminiferal tuff lies beneath the lava, so that this latter probably represents a submarine flow. Other flows of the area possibly also were submarine but escaped too rapidly to permit the formation of pillows. Origin and Mode of Accumulation of Manukau Breccias. Park (1889) stated that the fragmental volcanic rocks of Waitakere Range were the product of submarine eruptions in shallow seas; Bartrum (1929) came to the same conclusion and believed that most of the parent volcanoes were located not far west of the present coast. One of the most puzzling features of the rocks concerned is that they do not include in more than very minor amount the highly scoriaceous material that is so characteristic of normal volcanoes; their constituent blocks appear to be fragments disrupted either by explosions or by a vigorous eroding agent, such as waves, from earlier poorly vesicular lava. Limitations of space do not permit discussion of this major problem. The writer believes that transport and deposition were effected by the sea, for the rocks are usually in sub-horizontal well-bedded sheets which lack the foreset bedding that is characteristic of stream fans or deltas, although at times bedding layers of any kind are absent from coarse breccias and fine tuffs alike. Confirmation of this belief is given by the fact that marine fossils have been found in the series in a number of widely separated localities and include Foraminifera at Nihotupu, Waiatarua and near Titirangi, as well as a few species of molluscs in moderately fine breccia at Henderson Valley.

The blocks of the breccias are often so angular that they cannot have been transported far; the areas of provenance, therefore, must have been fairly near at hand. Support is given to this conclusion by the discovery in a number of localities of agglomerates which, in some cases, are clearly occupying the actual vents of extinct volcanoes. One such example already has been described from Windy Point, between Manukau North Head and Karekare; another is afforded by a funnel-shaped mass of agglomerate 80 yards in width which penetrates well-bedded breccia-conglomerate associated with a number of dykes immediately south of the Blow Hole, Piha. In its material there are red oxidised blocks of scoriaceous lava as much as 4 feet in diameter. A similar pipe outcrops at White's Bay, 2 miles north of Piha Stream; it is 33 feet wide at beach level but enlarges upwards as it passes through well-bedded conglomerate which dips at 40° to the south-west. It is bordered on the north by highly irregular intrusions and has given rise at a height of 100 feet above the beach to a columnar sub-horizontal lava flow. Additional evidence of the proximity of an important eruptive vent continues for some distance north in this bay, for discontinuous beds composed of scoriaceous fragments and lenses of red oxidised ash are common along with many irregular small flows of lava. Reasons already have been given for the belief that other centres of eruption were located close to Manukau North Head and to the Watchman at Karekare. Park (1886) believed that the topography at Puponga Point, Manukau Harbour, was a reflection of an ancient crater located there; the rocks of the locality are wholly coarse conglomerates inter-bedded with thick beds of tuff and the topography to which Park drew attention is merely a cuesta developed on a bed of conglomerate folded into an anticline or dome that has been faulted on its southern flank. Age of Manukau Breccias and their Relation to the Waitemata Series. Hochstetter (1864), Cox (1881) and Park (1889) have agreed that the Waitemata beds and the volcanic strata above them form a conformable succession, although Park had denied this in earlier writings. Bartrum (1924) reached the same conclusion with regard to the two series of beds in the Kaukapakapa region, although later Turner and he (1929) noted that in a water supply tunnel at Huia, on the north coast of Manukau Harbour, the volcanic series appeared to rest on a surface eroded upon Waitemata sediments. Professor Bartrum informs the writer, however, that this appearance of unconformity may be the result of the slipping of the massive breccias down the slope of the spur that is traversed by the tunnel. The writer's study of various contacts has convinced him that conformity is the rule. As already described in an early section, the upper beds of the Waitemata Series near Puponga Point have bedding discontinuities and include rolled mud balls which are also the product of shallow water deposition. These upper beds are mainly tuffaceous sandstones with lenses of coarse tuff and are followed with complete conformity by typical coarse volcanic conglomerate.

Along the greater part of the eastern slopes of the Waitakere Hills there is no marked change in type of sediment as one passes from Waitemata beds to those of the Manukau breccias, for the lowest strata of this latter series are fine-grained and weather to whitish or mottled clays which obscure the actual contacts. In Opanuku Stream, Henderson Valley, however, fine-grained horizontally bedded sandstones of Waitemata type show gradual passage into fine volcanic breccias which undoubtedly are members of the volcanic series. From these latter Mr. A. W. B. Powell obtained Lima colorata (Hutton) and Opella cf. subfimbriata (Suter) which enabled him to correlate the containing strata with the Hutchinsonian (Lower Miocene) fossiliferous beds at Pakaurangi Point, Kaipara Harbour. A little south of Motutara, north of the area now described, there is a small molluscan faunule in sandstones of the Manukau Breccia Series; it has been described by Powell (1935) and referred to the Awamoan (Middle Miocene). The Manukau breccias of Motutara thus are younger than the Hutchinsonian (Lower Miocene) beds of the Waitemata Series at Oneroa, Waiheke Is. (Powell and Bartrum, 1929). Pleistocene and Recent Formations. Subaërial Pleistocene Sands. Near Cornwallis on the south-east side of Puponga Point, Manukau Harbour, there is a bed exposed for a height of 10 feet of soft partially consolidated sands rich in oxidised titaniferous magnetite. As shown by his section, Hochstetter (1864, pp. 17, 18; 1867) placed these sands in the same series as Waitemata beds exposed beneath them; they differ in character, however, from any known Waitemata strata, and in addition exemplify the large scale false bedding that is characteristic of dune sands. Their exact counterparts appear in the sands of the Pleistocene range between Manukau South Head and Waikato River (Gilbert, 1921). Hochstetter (loc. cit.) placed these latter in his Pleistocene “Lignite Formation”; Ferrar (1934) more recently has referred them to his Kaihu Sands (Pleistocene). They attain an elevation of over 600 feet in higher parts of the plateau-like range mentioned above and have wide distribution north of the present area beginning at Anawhata, 9 miles north of Manukau North Head, where they occur above the sea cliffs in remnants of a dissected plateau, roughly 500 feet in average height, which broadens rapidly beyond Te Henga to form a conspicuous feature of the landscape. The occurrence of these wind-bedded sands at Puponga Point is of interest in demonstrating that when they were deposited the entrance to Manukau Harbour was not in its present location but further south. Pleistocene Clays, Silts and Lignites. Cox (1884) early examined clays with seams of impure lignite near Whau Creek, Waitemata Harbour, and correlated them with Hochstetter's Lignite Formation of the Manukau lowlands. Turner and Bartrum (1929) have dealt with them more recently and note that the 40 feet to 60 feet erosion level that is so prominent around the shores of Waitemata Harbour is underlain in part by these beds as well as by Waitemata strata and occasional tuffs that resulted from early eruptions of the basaltic volcanoes of the Auckland isthmus.

Beds of this Pleistocene formation cover about 4 square miles of the present area, with their greatest development near Avondale and there is good reason to believe that they once continued thence to Hobsonville, 7 miles north, as a broad sheet which has subsequently been removed by erosion. In the main they consist of white or grey plastic clays with not infrequent pumice and other sands and silts and many thin lensoid beds of lignitic mudstone which are especially freely developed at Point Chevalier Beach. Gardner's clay pit, near New Lynn Railway Station, affords an excellent section of the formation; in several of the beds there are woody stems in the position of growth with roots attached. Pumice pebbles as much as 3 inches in diameter occur in some of the layers, and at Hobsonville there are thin beds of small pebbles of greywacke and other rocks (Turner and Bartrum, 1929). The contact of this Pleistocene clay formation with underlying Tertiary strata is seldom visible, but Mr. C. W. Firth states that excavations for a sewage tunnel between New Lynn and Karaka Bay, south on the Manukau coast, showed clearly that the younger beds occupy steep-walled trenches eroded from Waitemata strata.* Personal communication. Similar relations are suggested near Point Chevalier, while the fact that at Whau Creek, near Avondale, excavations for the foundations of a bridge reached 50 feet below sea level without passing below the Pleistocene clays shows that the trenches occupied by these latter were carved deep below modern sea level; this trenching indicates that sharp uplift (or lowering of sea level) preceded the deposition of the beds concerned. The observations just recorded also give information as to the depth of this Pleistocene series, for, while descending to more than 50 feet below sea level at Whau Bridge, its beds are found at 100 feet above that level south of New Lynn. Although generally bedded horizontally, as noted by Turner and Bartrum (loc. cit.) they may locally be gently inclined; south of New Lynn their dip in one clay pit is 30°, so that they shared in the later of the compressional movements that have affected the underlying Waitemata sediments. The writer agrees with the conclusion of Turner and Bartrum (loc. cit.) that this series of Pleistocene beds was deposited “… in lakes or on the swampy floors of the valleys of sluggish streams during progress of very slow subsidence.” Recent depoits of the present area include marine sands and muds which occupy drowned valleys in the submerged regions of Waitemata and Manukau harbours; extensive mudflats dotted here and there by groves of mangroves are ubiquitous around the shores and extend as tongues up most of the numerous tidal creeks. In addition, there are occasional floodplain deposits and, as already recorded, on the Tasman coast there has been recent progradation of the shore line by sand beaches followed by a relatively narrow belt of windblown sand which extends to the base of the former sea cliffs. It is only at Muriwai, well north of the present area, where the resistant rocks of the Manukau Breccia Series turn obliquely inland, that the

Fig. 1.—Intraformational drag folding due to delta slumping; tuffs of Manukau Biecias, 1 mile east of Big Muddy Creek, Manukau llarbour. Fig. 2.—Deformation of Waitemata Series, shora at east end of Blockhouse Bay, Manukau Ilarbour Fig. 3.—Fault along crest of anticline; tuffs of Manukau Breccias near Pararaha Stream, Tasman Coast. Fig 4.—Short platform about 4 feet above high water level and about 40 feet in width, Manukau Brecias, Manukau North Head

Fig. 5.—Clastie dykes A (Y-shaped) and B of Taylor (1930) in Manukau Bieccias three-quarters of a mil north of Manukau North Head Fig. 6.—Matenal of pumee in elastic dykes of Fig 5, the phenocryst is hypersthene. Fig. 7.—Clastie dyke filled by volcame ejecta, Windy Point, two miles north of Manukau North Head. Note the bedding of adjacent bieccias Fig 8.—Differential weathering in 200 feet bluff of Manukau Bieccias near Parataha Stream, Tasman Coast.

Geological Sketch Map of Southern Waitakere Area

belt of modern wind-blown sands materially increases its width and threatens danger to adjacent farm lands. Geological History of the Area. The most interesting phases of the geological history of the area are those that concern late Tertiary and post-Tertiary time. These have been adequately dealt with by Turner and Bartrum (loc. cit.) and make so long a story that recapitulation of it is impracticable in this short paper. In any case no facts that materially affect the conclusions of these writers have been brought to light during the present study. Petrography of the Manukau Breccias. Nearly 120 thin-sections of rocks of the area were examined and included many made by the writer and others lent by Professor J. A. Bartrum. They indicate that there is little petrographical variation in the rocks of the volcanic series, as was indeed to be expected, for, although they had not undertaken systematic microscopic investigation, both Marshall (1925) and Bartrum (1925) have recorded that the rocks concerned essentially are andesites. The typical rock, whether in dyke, flow or fragmental masses, is an hyalopilitic pyroxene andesite with augite and hypersthene in subequal proportions and commonly, though not always, as phenocrysts as well as in the groundmass. Plagioclase near andesine-labradorite in composition is the most persistent and abundant phenocryst; its crystals generally show zonal growth and both Albite and Carlsbad twins. Frequently they are pitted by zonally-arranged inclusions of glass and, in a few rare instances (e.g., S.62), include tiny needles of rutile. The groundmass typically consists of laths of feldspar with pyroxene and a good deal of glass; magnetite and ilmenite in small needles and dendritic growths are unusually prominent in the rocks of dykes near Whatipu, Manukau North Head (S.8; S.16) and cluster about vesicles filled with chlorite. This latter secondary mineral is derived from the pyroxenes and shows all stages in its replacement of those minerals: it has been dispersed freely throughout the fabric of the rocks. In the rocks of widespread localities there are what appear to be chloritic replacements of original olivine, for they have the characteristic fractures, form and, more rarely, resorption borders, while a few unreplaced cores of olivine appear in the rock (S.23) of a 30-feet dyke in a former stack a quarter of a mile south of Ohaka Head. A green pleochroic mineral taken to be bowlingite is enclosed by chlorite in the rock of a large dyke (S. 34) 100 yards north of White's Bay; it has eminent cleavage, moderately high refractive index and birefringence about that of tale and is uniaxial. Hornblende is rare, although it was found in andesitic fragments present in some of the clastic dykes, including that at Windy Point, and as small, rare crystals in other rocks, such as that of a dyke at Fishing Rock, Karekare. The two pyroxenes present in the rocks occasionally show a mutual reaction that has been noted by Bartrum (1917) from hypersthene-olivine basalt at Ruatangata, near Whangarei, and from a quartz norite at the Cleddau-Hollyford Saddle, Fiordland (Bartrum, 1920). The result of this reaction is

that earlier-formed hypersthene is rimmed by augite; Bowen (1928, pp. 58–59) has shown that reaction rims of this kind can only occur when there has been rapid cooling of the magma. Marshall (1908) recorded a rhyolite intrusion at Karekare. The writer made an intensive unsuccessful search for this rock, and can only conclude that a superficially light-coloured weathered andesite has been mistaken macroscopically for rhyolite. Chemical analyses of the rocks under description are not available, but Professor Bartrum has kindly allowed the publication of the following analysis of a pillow lava present in this same Manukau Breccia Series at Muriwai, 7 miles north of the area dealt with in this paper. The rock is similar microscopically to the more basic of the andesites described above. Analysis and calculation of the norm and classification under the C.I.P.W. system are the work of Mr. F. T. Seelye, of the Dominion Laboratory, Wellington. Analysis of Composite Sample from Main Flow at Muriwai. Analysis Mol Prop. Norm. SiO2 53.16 .880 Al2O3 16.69 .164 Q. 8.6 Fe2O3 3.38 .021 or 6.12 FeO 4.88 .068 ab 27.25 MgO 3.51 .088 an 28.08 CaO 8.32 .148 di 8.99 Na2O 3.17 .052 hy 8.94 K2O 1.02 .011 mt 4.87 H2O+ 1.58 il 2.28 H2O- 2.79 ap 0.67 CO2 0.02 TiO2 1.15 .015 ZrO2 none P2O5 0.26 S 0.02 Cr2O3 . none NiO trace MnO 0.14 SrO ?trace BaO 0.03 Symbol: II. (4) 5. (3) 4. 4.” Andose. 100.12 Acknowledgements. The writer wishes to thank Professor J. A. Bartrum for valuable help given during the preparation of this paper, and Mr. W. J. Parr and Dr. H. J. Finlay for undertaking investigation of the Foraminifera. Literature Cited. Bartrum, J. A., 1917. Additional Facts Concerning the Distribution of Igneous Rocks in New Zealand. Trans. N.Z. Inst., vol. xlix, pp. 418–424. — 1920. Additional Facts Concerning the Distribution of Igneous Rocks in New Zealand, No. 2. Ibid., vol. lii, pp. 416–422. — 1923. Physiographic Notes on Auckland. N.Z. Nature Notes for Austr. Ass. Adv. Sc., Wellington, 1923, pp. 19–20. — 1924. The Geology of the Riverhead-Kaukapakapa District. Trans. N.Z. Inst, vol. lv, pp. 139–153. — 1925. Igneous Rocks of North Auckland, N.Z. Gedenkboek Verbeek, Verhand. v.h. Geol.-Mijn. v. Nederl. en Kol., Geol. Ser., vol. viii, pp. 1–16. — 1926. Abnormal Shore Platforms. Journ. of Geol., vol. xxxiv, pp. 793–806.

— 1929. Notes on the Geology of Auckland. Municipal Record, Auckland, Jan., 1929, pp. 2–5. — 1930. Pillow Lavas and Columnar Fan-Structures at Muriwai, Auckland, N.Z. Journ. of Geol., vol. xxxviii, pp. 447–455. — 1935. Shore Platforms. Rep. Austr. and N.Z. Ass. Adv. Sc., Melbourne, 1935, pp. 135–143. — 1937. Geography and Geology of Auckland and Environs. Auckland, a Handbook for Austr. and N.Z. Ass. Adv. Sc., Auckland Meeting, 1937, pp. 7–13. — and Turner, F. J., 1928. Pillow Lavas, Peridotites and Associated Rocks from Northernmost New Zealand. Trans. N.Z. Inst., vol. lix, pp. 98–138. Bell, J. M., and Clarke, E. de C., 1909. The Geology of the Whangaroa Subdivision. N.Z. Geol. Surv. Bull. no. 8. Benson, W. N., 1935. Some Land Forms in Southern New Zealand. Austr. Geogr., vol. ii, no. 7, pp. 3–22. Bowen, N. L., 1935. The Evolution of the Igneous Rocks. Princeton Univ. Press, Princeton. Cotton, C. A., 1922. Geomorphology of New Zealand, Part I., Wellington, N.Z. Cox, S. H., 1881. Geology of Rodney and Marsden Counties. Rep. Geol. Explor. during 1879–80, pp. 13–39. — 1884. Prospects of Coal Occurring at Whau. Rep. Geol. Explor. During 1883–84, pp. 10–11. Ferrar, H. T., 1925. Whangarei-Bay of Islands Subdivision. N.Z. Geol. Surv. Bull. no. 27. — 1934. The Geology of the Dargaville-Rodney Subdivision. N.Z. Geol. Surv. Bull. no. 34. Finlay, H. J., and Marwick, J., 1940. The Divisions of the Upper Cretaceous and Tertiary in New Zealand. Trans. Roy. Soc. N.Z., vol. lxx, pp. 77–135. Firth, C. W., 1930. The Geology of the North-west Portion of Manukau County. Trans. N.Z. Inst., vol. lxi, pp. 85–137. Fox, C. E., 1902. The Volcanic Beds of the Waitemata Series. Ibid., vol. xxxiv, pp. 452–493. Gilbert, M. J., 1921. Geology of the Waikato Heads District. Ibid., vol. liii, pp. 115–127. Grange, L. I., 1927. The Geology of the Tongaporutu-Ohura Subdivision. N.Z. Geol. Surv. Bull., no. 31. Henderson, J. H., and Grange, L. I. The Geology of the Kawhia-Huntly Subdivision. N.Z. Geol. Surv. Bull., no. 28. Hochstetter, F. Von, 1864. Geologie von Neu-Seeland. Reise der Novara, 1 Bd, 1 Abt., Cotta, Stuttgart. — 1867. New Zealand. Cotta, Stuttgart. Hutton, F. W., 1870. Geology of the North Head of Manukau Harbour. Trans. N.Z. Inst., vol. ii, p. 161. Jutson, J. T., 1940. The Shore Platforms of Mt. Martha, Port Philip Bay, Victoria, Australia. Proc. Roy. Soc. Vic., vol. lii, pp, 164–174. Laws, C. R., 1931. Geology of the Papakuia-Hunua District, Franklin County, Auckland. Trans. N.Z. Inst., vol. lxii, pp. 37–66. McKay, A., 1894. On the Geology of Hokianga and Mongonui Counties, Northern Auckland. Rep. Geol. Explor. during 1892–93, pp. 70–90. Marshall, P., 1908. Geology of Centre and North of North Island. Trans. N.Z. Inst., vol. xl, pp. 79–98. — 1925. Igneous Rocks of New Zealand. Gedenkboek Verbeek, Verhand. v.h. Geol.-Mijn. v. Nederl. en Kol., Geol. Ser., vol. viii, pp. 357–369. Mulgan, E. K., 1902. On the Volcanic Grits and Ash Beds of the Waitemata Series. Trans. N.Z. Inst., vol. xxxiv, pp. 414–435. Park, J., 1886. Waitemata, Eden and Manukau Counties. Rep. Geol. Explor. during 1885, pp. 147–164. — 1889. On the Waitemata Series. Trans. N.Z. Inst., vol. xxii, pp. 391–399. Powell, A. W. B., 1935. Tertiary Mollusca from Motutara, West Coast, Auckland. Rec. Auck. Mus., vol. i, pp. 327–340. — and Bartrum, J. A., 1929. The Tertiary (Waitemata) Molluscan Fauna of Oneroa, Waiheke Island. Trans. N.Z. Inst., vol. lx, pp. 394–447. Smith, S. P., 1881. On Some Indications of Changes of Level of the Coastline in the Northern Part of the North Island. Ibid., vol. xiii, pp. 398–410.

Taylor, N. H., 1927. Pumiecous Silts in the Waitakerei Ranges, Auckland. N.Z. Journ. Sc. and Techn., vol. ix, pp. 166–167. — 1930. Clastic Dykes in Volcanic Breccia, Manukau North Head. Ibid., vol. xi, pp. 305–309. Turner, F. J., and Bartrum, J. A., 1929. The Geology of the Takapuna-Silverdale District, Waitemata County. Trans. N.Z. Inst., vol. lix, pp. 864–902. Wentworth, C. K., 1938. Marine Bench-forming Processes: Water Level Weathering. Journ. of Geomorph., vol. i, pp. 6–32. Appendix. List of Foraminifera from the Waitemata Series, Manukau North Coast, determined by Dr. H. J. Finlay. Localities. 1. Eastern promontory of Mission Bay. 2. West face of Cape Horn, 3-foot bed in centre of anticline (interpolated from Parr's list). 3. 1 ½ to 2-inch bed in lower part of same anticline. 4. An upper bed of same anticline, near small stream. 5. 25 yards west of first small stream from east end of Hillsborough Bay. 6. The same; softer sandstones west of small stream. 7. East side of same stream, Hillsborough Bay. 8. Below Parnell Grit, Blockhouse Bay. 9. 10 chains east of Duck Creek. 10. At base of point, western side of White's Bluff. 11. Five chains west of point, White's Bluff. 12. About 15 chains west of Locality 11. 13. First large bay west of White's Bluff. Localities. 1 2 3 4 5 6 7 8 9 10 11 12 13 1. Rhabdammina c.f. abyssorum Brady 6 6 c c 2. Psammosphaera aff. fusca (Schultze) 4 c 3. Ammodiscus incertus Stache 1 1 4. Haplophragmoides sp. (crushed) 2 1 4 5 5. Trochammina? sp. (crushed) 1 1 6. Bolivinopsis cubensis C. & B. 4 3 4 3 2 1 7. Spiroplectammina sp. 2 . 1 8. Siphotextularia cf. heterostoma (Forn.) 1 9. Textularia sp. 2 10. Haueslerella pukeuriensis Pari 7 11. Karieriella bradyi Cush. 1 1 12. Multinottiella cf. communis (d'Orb.) 1 13. Gaudiyina cf. crespinae Cush. 1 14. Massilina tenuis (Czj.) 1 15. Sigmoilina n.sp. aff. celata (Costa) 1 16. Quinqueloculina spp. 1 17. Robulus sp. 3 1 18. “Planularia” (auct.) n.sp., (costatc) 1 19. Saiacenaria italica (Defr.) 2 20. Astacolus sp. 1 21. Vaginulinopsis n.sp. 1 1 22. Dentalina aff. consobrina (d'Orb.) 4 1 2 3 1 3 5 1 3 23. Nodosaria longiscata d'Orb. 2 1 2 c 2 4 1 2 24. N. pyrula d'Orb. 3 1 2 25. N. holoserica Schwager 5 1 2 26. Lagenonodosaria cf. separans (Brady) 1 1 27. L. n.sp. 2 28. Chrysalogonium n.sp. (tiny) 1 1 4 3 4 2 29. C. n.sp. 1 2 4 2 30. C. polystomum (Schwag.) c c c 4 3 3 c 5 c 31. C. (?) cf. challengeriana (Thal.) 1 2 2 32. C. (?) cf. fistuca (Schwag.) 2 4 33. C. (?) n.sp. 1

1 2 3 4 5 6 7 8 9 10 11 12 13 34. Siphonodosaria aff. adolphina (d'Orb.) 2 2 35. S. aff. jaivisi (Cush.) c c 8 6 3 36. S. aff. alexanderi (Cush.) 4 3 2 4 2 37. S. n.sp. (tiny, smooth) 5 2 1 5 6 5 1 2 38. S. globulifera (Kreuz.) 1 1 39. S. cf. bradyi (Cush.) 3 1 1 1 c 1 1 1 4 40. S. sp. 3 41. Glandulina radicula (L.) 1 42. Lagena acuticosta Reuss 1 43. L. hexagona Will. 1 44. L. hispida Reuss 1 45. L. hispidula Cush. 1 1 46. L. striata d'Orb. 1 1 47. L. orbignyana Seguenza var. A. 1 1 48. L. orbignyana Seguenza var. B. 1 49. L. orbignyana Seguenza var. C. 2 50. L. marginata (W. & B.) var. 2 1 51. L. cf. meridionalis Wiesner 1 52. L. cf. fasciata Egger 2 53. L. sp. 2 54. L. sp. 2 55. L. sp. (2 forms) 6 56. L. elongata Ehrenb. 1 57. L. semistriata Will. 1 58. Frondicularia n.sp. 2 59. F. n.sp. aff. multicostales Fin. 2 60. Plectofrondicularia awamoana Fin. 1 61. P. parri Fin. 1 1 62. P. australis H.A. & E. 1 63. Amphimorphina. aff. crassa C. & B. 2 64. Bolivina anastomosa Fin. c 5 65. B. n.sp. aff. anastomosa Fin. 2 c 6 1 1 66. B. lapsus Fin. c 4 1 67. B. aff. beyrichi Reuss 2 3 7 3 5 3 68. B. aff. semialata Bagg 3 c 69. B. aff. ouachitaensis H. & W. 5 2 1 70. B. aff. danvillensis H. & W. c c c c c 4 c c 71. B. n.sp. 2 1 72. Loxostomum n.sp. (common in Mahoenui) 5 c c 73. L. hentyanum (Chap.) 6 1 74. L. aff. delicatulum Cush. c c c 1 75. Rectobolivina maoriella Fin. 1 2 1 4 c 2 76. Virgulina n.sp. 1 1 77. V. aff. vicksburgensis Cush. 7 1 1 1 5 c 1 78. Virgulopsis n.sp. 1 1 79. V. pustulata Fin. 3 80. Uvigerina miozea Fin. 8 2 81. U. aff. auberiana d'Orb. 4 2 2 4 3 c 1 2 82. Angulogerina aff. oligocenica (Andreae) 2 1 1 83. A. n.sp. 1 2 2 c 84. A. australis H.A. & E. c 4 85. Trifarina bradyi Cush. c 7 1 86. T. bradyi var. 7 4 2 87. Siphogenerina rerensis Fin. 5 1 88. S. n.sp. (tiny) 1 c 1 5 89. Bulimina truncanella Fin. 1 1 1 90. B. aff. striata d'Orb. 1 91. B. aff. pupula Stache. c 4 2 6 1 1 92. Cerobertina bartrumi Fin. 1 93. Cancris lateralis Fin. 1 1 94. Chilostomella sp. 2 1 95. Chilostomelloides n.sp 3 3 2 c 5 4 3 1 c 4 1 1

1 2 3 4 5 6 7 8 9 10 11 12 13 96. Pleurostomella sp. A. 1 1 1 3 97. P. sp. B. 6 2 7 1 1 1 3 3 1 98. Nodosarella sp. 1 99. Ellipsoglandulina sp. 2 100. Ellipsolagena sp. 1 1 1 101. E. sp. 1 102. Cassidulina subglobosa Brady. 3 1 3 103. C. arata Fin. c 104. C. aff. sicula Seg. 4 1 105. C. carinata Cush. 4 1 3 106. C. sp. 1 107. Cassidulinoides orientalis (Cush.) 1 1 1 2 2 108. Ehrenbergina n.sp. aff. mestayeri Cush. 2 1 109. E. n.sp. aff. osbornei Fin. 2 110. Pullenia sphaeroides (d'Orb.) 1 c 1 1 111. P. subcarinata (d'Orb.) 1 1 112. Astrononion aff. australe C. & E. 1 4 c 113. A. n.sp. aff. novozealandicum C. & E. c 114. Nonionella zenitens Fin. 3 1 6 115. N. novozealandica Cush. 1 8 6 1 116. Nonion n.sp. aff. marginatum C. & E. 5 117. N. aff. soldani (d'Orb.) 1 2 118. Spirillina sp. 1 119. Anomalina subnonionoides Fin. 4 2 120. A. parvumbilia Fin. 2 2 1 121. Gyroidina n.sp 2 1 4 122. G. aff. zelandica Fin. c 3 1 123. Eponides ecuadorensis (G. & M.) 1 1 2 2 2 1 124. Epistomina elegans d'Orb. 3 125. Parrella n.sp. 2 126. P. bengalensis (Schwag.) 3 2 127. Alabamina tenuimarginata (C. P. & C.) 1 1 c 1 2 128. Pulvinulinella n.sp. 7 129. Amphistegina lessoni d'Orb. 6 130. Elphidium cf. ornatissimum (Karrer) 1 1 131. E. cf. advenum Cush. 1 3 132. E. n.sp 1 133. Notorotalia spinosa (Chap.) 1 134. Cibicides cf. perforatus (Karrer) 4 135. C. cf. refulgens Mont. 2 136. C. mediocris Fin. c 3 1 137. C. n.sp. A. 5 2 138. C. n.sp. B. c 2 139. C. aff. robertsonianus (Brady) 2 140. Dyocibicides biserialis C. & V. c 4 2 141. Vagocibicides maoria Fin. 1 2 142. Discorbis baconica Hant. 6 5 143. D. bertheloti (d'Orb.) 1 1 144. D. scopos Fin. 1 1 145. D. sp. 1 146. D. aff. saulcii d'Orb. 1 147. Buningia creeki Fin. 5 148. Parvicarinina altocamerata (H.A. & E.) 3 1 149. Heronallenia wilsoni (H.A. & E.) 1 150. Globorotalia dehiscens C.P. & C. c c c c c 5 1 c 3 6 c 151. G. sp. A. c 4 6 c c c c 1 1 1 1 c 152. G. sp. B. c 5 3 c 5 3 153. Globigerina n.sp. A. 6 3 c c c c c c c c c 154. G. n.sp. B. (tiny) 5 2 1 1 3 6 155. G. bulloides d'Orb. 3 6 c c 3 c 3 3 2 3 156. Sphaeroidina bulloides d'Orb. 2 2 3 157. Radiolaria 2 c 1 158. Pteropods 1 The numbers in the columns are the number of specimens mounted on the slides.