Holocene of the North Island of New Zealand: a Coastal Reconnaissance
H. W. Wellman,
By
Victoria University of Wellington
[Received by the Editor, June 21, 1961.] Abstract Fifty Holocene stratigraphic sections, well distributed around the coast of the North Island, are described. Layers of volcanic ash or water-borne pumice make close correlation possible. Several previously undescribed ash showers and pumice layers are named. Leigh ( c . A.D. 1), Taupo (A.D. 150), and Loisels ( c . A.D. 700) pumices, and the Kaharoa Ash (A.D. 1300) are the most widely distributed. Provisional dates given above are based on stratigraphic interpolation from the radio-carbon-dated Taupo Pumice and will be more closely defined when critical radio-carbon samples have been processed. Petrology of the pumices and ash showers, described in an appendix by Miss G A. Ghallis, proved essential for accurate correlation. Sea-level changes are inferred from a comparison of the Holocene deposits with those now accumulating. Drowned forests at the base of many sections indicate the lower sea level at the end of the Pleistocene. Sea level was higher than at present at about 2000 8.C., and very slightly lower at A.D. 150. Since A.D. 700 it has differed little from that of the present day. Maori occupation soils, an important part of the top of most sections north of New Plymouth and Napier, are used to infer Maori population and distribution in the past. A rapid and simultaneous population increase from a value too small to be recorded to about 150,000 at A.D. 700 is the most conclusive result of the study. Eastland population probably decreased to a minimum at about A.D. 1300 and then increased again. Coromandel and Northland populations reached their maximum after A.D. 1300. The large pre-A.D. 1300 population makes it likely that present day Maoris are the descendants of the earliest inhabitants, and it is suggested that the traditional “Fleet” canoes, supposedly of A.D. 1350, may have been those of the first people to arrive here. Charcoal is the most critical indicator of Maori occupation. Charcoaldarkened soils extend inland at those places that were scrub-covered in 1780 and probably represent forests that were burnt by the Maori. Introduction The Holocene Period extends back for seven thousand or so years from the present day and largely corresponds to the historical period of ancient civilizations. It
followed the last glacial epoch of the Pleistocene, climate (Brooks, 1950) and geography of the late Holocene being not very different from that of the present day. New Zealand has a written history that extends back for a mere two hundred years and for its earlier history is dependent on Maori tradition and on the record of the stratified rocks. The significant Maori traditions are those that were held on the arrival of the Pakeha (Sharp, 1957). They have been rationalized according to nineteenth century Pakeha opinion and are now taught as our earliest history (Sinclair, 1959). Like the early tradition of all races, Maori traditions are partly historical and partly mythical; and served to explain the past as well as to record it. Stratigraphy may help to separate the historical from the mythical part of the tradition. The events of the Holocene as recorded by stratigraphy are thus of considerable importance as pre-history. They are also a guide to the geological events that are likely to take place in the immediate future. In this last respect the study of the Holocene is of greater importance than the study of any other equal interval of time in the past. Only at particularly favourable localities have sediments accumulated sufficiently rapidly and uniformly to provide a clear record of Holocene events. Coastal deposits are the most easily correlated because they record the events of the sea as well as those of the land, besides being well exposed by marine erosion at many places. Favourable accumulations consist of interbedded layers of blown sand, beach deposits, ash, water-borne pumice, soil-wash, and midden material, the optimum thickness being from four to ten feet. Thicker deposits are usually irregular, and thinner ones are difficult to interpret. The following factors have controlled the nature of the sediments and the rate at which they have accumulated at particular sites; (1) Local climate, as reflected by storms and wind direction and intensity; (2) tsunamis (tidal waves); (3) human population; (4) volcanic eruptions; (5) vegetation; (6) eustatic changes in sea-level (controlled by world climate) ; (7) tectonic elevation or depression of the land. The stratigraphic record is confused and not easy to interpret. In particular, tectonic elevation and depression cannot be distinguished from eustatic sea-level changes in individual sections. In the following account sea-level changes are expressed throughout as though entirely eustatic, the land being considered stationary.
The sections described are those that contain the most complete geological record of the Holocene, but they are not average sections. In a sense they are geological curiosities, for they are restricted to a few places where rapid and uniform accumulation has been followed by equally rapid erosion. Unless erosion is checked many of these deposits will be destroyed in a relatively short time. Where land values are high because of settlement erosion is causing economic loss, and sea walls are being built to prevent it. Present-day settlement favours the same sheltered localities as those favoured by the earliest settlers, and many sections of archaeological interest are likely to be hidden by sea walls in the near future. After sea-walls are built coastal deposits are protected and can be examined at leisure if excavated. But excavations have to be unusually extensive to rival natural coastal sections, and the time and cost is prohibitive for all except the most important localities. It is thus important that those sections now exposed be described in detail as soon as possible.
To contain a worth-while record most Holocene sections have to be several hundred feet long and at least four feet high. They are consequently quickly found, the major problem in a reconnaissance being transport. Most of the North Island beaches are only a short distance from roads and can be reached without
much difficulty. Those most difficult of access are on the eastern side of Northland near the Bay of Islands, where many possibly important sections have not been examined.
The North Island coastal sections were examined at various times during the last four years. The east coast from Mahia to Reporoa was studied in the greatest detail, about a month being spent in the field, several of the more important sections being examined several times. About two weeks were spent examining the sections in the Bay of Plenty, Eight weeks were spent in a systematic examination of the beaches in Coromandel, Northland, and the west coast. The parts of the North Island coast examined are shown by heavy lines in Fig. 1.
About fifty useful sections were measured and about a hundred examined, but not measured. The measured sections are well distributed over the North Island coast and give a fair idea of the regional stratigraphy. Because Holocene deposits are mostly built from locally derived material lithologic correlation of the kind used in older sediments is impossible. The common molluscan fossils are also useless for correlation, because the time is far too short for significant evolution to have taken place and the differences caused by climate changes have been negligible. Correlation has to be based on short and dramatic events, and in this respect we are fortunate that several large volcanic eruptions took place in New Zealand during the Holocene. The products of the eruptions were rapidly distributed by wind as ash showers, and even more widely as water-borne pumice by rivers and by the sea (cf. Richards, 1957, for rate of oceanic pumice transport). Where ash showers or water-borne pumice are present, correlation is possible, provided that the products of individual eruptions can be recognized.
In the following account the term “ ash ” is used for air-transported volcanic deposits that have not been significantly reworked, and “pumice” for material sufficiently buoyant to float in rivers or the sea, and “ tuffaceous ” for fine-grained volcanic material that has been reworked and mixed with sediments. Most tuffaceous sediments are not sufficiently distinctive to be correlated directly with particular eruptions, and they may be considerably younger than the volcanic material they contain.
The earliest eruption at present known to be useful for North Island correlation is that from near Lake Taupo and dated at about A.D. 150 by many C l 4 samples. It covered 5,000 square miles with over a foot of ash, and at least twice that area with over an inch. It is the youngest of several showers that occurred during a short interval of time and was described by Baumgart (1954) as the Taupo Lapilli shower. It is distinguished from other showers by a large proportion of pumice lumps, ' which are particularly conspicuous at its margin. At most places the pumice, referred to here as Taupo Pumice, can be distinguished from other pumices without much difficulty. Enormous quantities were washed into the rivers, carried out to sea, ‘ widely distributed by wind and currents, and finally thrown up on the beaches, to be covered by sand and soil. Seaborne Taupo Pumice directly overlies Taupo ash at several sections, and the first pumice probably reached the beaches shortly after the eruption. Rounded lumps of charcoal from forests destroyed by the eruption were washed up with the Taupo Pumice at a few beaches. Taupo Pumice is still being thrown up on many beaches, as it was frequently in the past, but the “ primary deposits that immediately followed the eruption are generally much thicker and coarser than the younger secondary 5 deposits. The area covered by the Taupo Lapilli and the position of primary deposits at present known on the coast are shown by Fig. 2. Pumice about the same age as Taupo Pumice, occurs on several beaches in eastern Northland; it is named Leigh Pumice,
The other important pumice is darker, more basic in composition, and stronger than the Taupo Pumice. It came from a volcano in the Bay of Plenty, most probably from near Mayor Island, and is referred to as Loisels Pumice (Wellman, 1960). The pumice has not yet been dated by G l4 , but from its average stratigraphic position relative to .the Taupo Pumice it is thought to have been erupted about A.D. 700. Loisels Pumice is widely distributed on the beaches on the east side of the North Island, but is absent from most of the beaches on the west side. Its distribution is shown by Fig, 3.
The Kahoroa Eruption from Mt Tarawera, near Rotorua, may be important for dating. No pumice was erupted, but its ash is probably present in a few coastal sections near Gisborne and in the Bay of Plenty. A single C l 4 sample gives a date of about A.D. 1050 (Golson, 1957). Other Holocene eruptions are known from Mt Egmont and Mt Rangitoto, but the ash and pumice from these were not widely distributed.
Human occupation is recorded by shells of edible molluscs, by burnt stones, by bones of fish, birds, and seals, by flakes of obsidian and chert, and exceptionally by other more finished artifacts. Charcoal, either as small scattered fragments or as a darkening of the soil, has proved to be the most sensitive indicator of human occupation. Stratigraphic levels that carried a relatively dense population are dark, seemingly greasy, and distinctive (Taylor, 1958). Where the population was less dense they are somewhat lighter in colour but darker than most soils or sediments. Rare angular fragments of charcoal a few inches below the other definite signs are considered as being the earliest evidence of human occupation. The fragments of occupation charcoal are easily distinguished by their angular shape and higher stratigraphic position from the water-borne, rounded charcoal lumps of the Taupo Eruption.
There are no traces of human occupation below or immediately above the Taupo Pumice in any section. Rare fragments of charcoal first appear a few inches below the Loisels Pumice and become more abundant at the base of the pumice layer. The “ lower occupation layer ”, with abundant charcoal and other signs of human occupation, lies just above the Loisels Pumice, and is well defined in Eastland; and uniformly present as a weak occupation layer in Coromandel and Northland. An overlying layer with few traces of occupation contains the ash that is tentatively correlated with the Kaharoa Ash of A.D. 1300. An “upper occupation layer ”, somewhat less well defined than the lower in Eastland but well defined in Coromandel and Northland, is overlain in many sections by blown sand or soil-wash that is still accumulating.
Estuarine clay and soil with standing stumps occur in several sections, a few feet or more below the Taupo Pumice. The beds indicate a sea level at least ten feet lower than that of the present day and are thought to have been deposited at the close of the Pleistocene, when the sea is known to have been lower than at present (Godwin et al, 1958). Because of tectonic movements the upper part of these deposits is likely to be of different ages at different places.
It has recently become possible to date sediments that contain carbon (as charcoal, as wood, as carbonate in shells, or as charcoal from bones) by the percentage of the radio-active isotope C l 4 they contain, provided the carbon was formed from carbon dioxide at the time the sediments accumulated. Two C l 4 dates have already been given, but only a few Holocene (as distinct from Pleistocene) datings are made in New Zealand each year, and if full advantage is to be made of the method it has to be restricted to the dating of samples that mark events and not merely used for dating objects that are of no importance for correlation (Fleming, 1953: 122). Consequently correlation should precede and not follow G l4 dating. When sufficient useful G l4 determinations have been made,
TT , , . mi i 0 i n terms of the conventional B.G. and A.D. dates, butmdl te ta beendone the purely sequential terminology of the stratigrapher is needed.
Divisions of the Holocene Six stratigraphic divisions are defined on the basis of the ash showers and occupation layers mentioned above. A seventh and lowest division is used for beds that indicate a significantly lower sea level and are considered to be Pleistocene in age. The divisions, their approximate age-ranges in years, and the information used in defining them are outlined in Fig. 4. They are described in greater detail below. Ages in years are given mainly for their archaeological interest; they are certain to be modified in the future when further C l 4 datings become available. It should also be noted that the C l 4 dates themselves are subject to modification because of the variations in the percentage of C l 4 in the atmosphere and ocean at different times in the past, but the corrections are probably less than 200 years for most of the Holocene. The Holocene divisions are not cultural but stratigraphic, and except that they are much shorter in duration, are similar in nature to the stages of the New Zealand Mesozoic, Tertiary and Pleistocene. Because it is likely that C l 4 determinations will make it possible to date the divisions with reasonable certainty in the near future, formal names are hardly required and the divisions are therefore numbered, but type localities are desirable. The section on the south side of Gove (No. 15) is an admirable type locality for all six of the Holocene divisions.
Division (1) ? A.D. 1800 to present day. Deposits above the upper occupation layer. Generally either wind-blown sand or soil and mud flows, but absent from most localities where sediments are not now accumulating. Age of base of layer is estimated and is doubtless somewhat different at different places. Division(2) c. A.D. 1300 to ? A.D. 1800. Upper occupation layer and, on east coast, upper part of relatively barren layer below. Base taken at base of thin subaerial ash present at several east-coast localities, which is tentatively correlated with the Kaharoa Ash of c. A.D. 1300.
Division (3) ? A.D. 700 to c . A.D. 1300. On east coast, lower occupation layer and lower part of overlying relatively barren layer. At Coromandel and Northland deposits with moderately abundant charcoal. Base clearly defined in most sections by lower limit of Loisels Pumice.
Division (4) ? A.D. 550 to ? A.D. 700. Oldest layer with evidence of human occupation. Base defined by lower limit of charcoal fragments. Within most of the area studied the base of this layer is only a few inches below the base of the Loisels Pumice and the interval of time represented by the division is small.
Division (5) A.D. 200 to ? A.D. 550. Deposits between the lowest charcoal fragments and the lowest Taupo (or Leigh) Pumice. It includes one or more thin subaerial ash showers.
Division (6). Several thousand years B.G. to A.D. 200. Deposits between the base of the Taupo (or Leigh) Pumice and the highest horizon that formed when sea level was significantly lower. At North Taranaki and elsewhere this layer includes several ash showers. It represents a relatively long time interval, including a short interval when sea level was probably a few feet higher than now, and will doubtless be subdivided in the future.
Division (7). Deposits formed during the last cool period, when sea level was significantly lower than at present. Marked by drowned forests, and peat below present mean high-water mark. Base not well defined. It represents a long interval of time and because of sediment compaction and tectonic movements its end may not be accurately dated solely from the rise in sea-level.
The sections on which these divisions have been based are described below. The most reliable are those that extend for several hundred feet and are well exposed throughout. The main advantage of an extensive section is that any particular bed can usually be examined at different heights above present sea level, allowing deposits that accumulated simultaneously but in different environments to be related. For instance, if possible Holocene sea-level changes are neglected, a particular bed is most likely to contain sea-borne pumice near the upper limit of waves, and subaerial ash and evidence of human occupation a few feet higher. From mean high-water mark to a level five or so feet higher is the most important vertical height range, and the longer the section the greater is the opportunity of finding any particular bed extending through this range. A few extensive sections (Figs. 5,7, 9, 12, and 13) illustrated by longitudinal sections drawn with a highly exaggerated vertical scale, show the considerable height range of individual beds. These sections, and several that are not figured, are based on a series of closely spaced vertical measurements.
Most measured sections are recorded as “ text sections ” that generalize the stratigraphy from several points along a beach and express It in terms of the average or the best exposed vertical section. This method is essential in order to save space, but gives a wrong impression of the height of some of the deposits relative to sea level. Loisels Pumice is younger than Taupo Pumice and is shown at a higher stratigraphic position in all text sections in which both occur. From the text sections alone it would be reasonable to infer a rise of a few feet in sea
, , j • +Q „ ro i u tW PPn the eruption of the two pumices, but at most X dg Jo" are actually most abundant at almost the same height above sea level. The height difference is merely apparent and is the result of the method of representation adopted.
East Coast Sections Sections are described in a counter-clockwise direction around the North Island coast (Fig. 1). Approximate positions are defined by sheet district and grid reference. Section 1 is exposed two miles west of Gape Palliser lighthouse at the point where the Tertiary sediments of Kupe’s Sail reach the coast. It shows on both sides of a small bay but is well exposed for about twenty feet only. Section I.— Kupe’s Sail, near Cape Palliser ( Nl6B, 798827). Soil of blown sand, oven stones at base ' ’ I.oft Sandy soil, moa bone at top I.oft Sand and beach boulders, boulders die out inland 2.oft Beach sand, with scattered Taupo Pumice 2.oft Beach sand with Taupo Pumice lumps up to 2ft 2.oft Grey beach sand 2.oft Tertiary sandstone, top about 10ft above M.H.W.M, at least 20.0 ft T aupo Pumice is abundant and was seen in similar sand 500 ft north-west between Kupe Stream and the west side of the Tertiary sediments. It lies at the same height above sea level as present-day drift wood and drift pumice, and the difference between sea level during the Taupo Eruption and that at the present day is unlikely to be more than 2ft or 3ft. This is rather surprising because the coast is fringed by a series of beach ridges that appear to be Holocene or late Pleistocene in age, and indicates a relatively rapid lowering of sea level relative to the land during some part of the Holocene. The moa bone near the top of the section was identified as Euryapteryx geranoides (Owen, 1848) by Dr J. G. Yaldwyn,
of the Dominion Museum (pers. comm. 23/9/1959). The Holocene sediments probably accumulated gradually, and the bone from its stratigraphic position would appear to be no more than a few hundred years old. The oven {umu) stones are the only evidence for human occupation and are scattered through the base of the layer immediately above the bone. The coast is extremely windswept and so inhospitable that the small population it would have supported are unlikely to have left a continuous record of their presence, and the moa bone is probably younger than the first period of human occupation recorded from the more favourable localities to the north. The coast was examined near Flat Point, from Castle Point to Owahanga River, and to the north at Porangahau and Blackhead Beach, but no satisfactory sections were seen south of Waimarama Beach. Shell middens are not uncommon in the more sheltered parts of the coast, but they are small and patchy compared with those farther north, and indicate that on this cold and windswept part of the North Island coast the total population was small. At several places, particularly at Blackhead Beach, north of Porangahau, the middens are in heaps, not evenly spread as is usual to the north. The difference is attributed to a difference in living habits caused by the colder climate. Shelters would be important and the middens were probably heaped up near them. Thick charcoal layers composed of the stems and leaves of bracken-fern lie near the base of several of the Blackhead Beach middens. Bracken is no longer abundant at this place, but was probably the dominant plant after the original scrub and forest had been destroyed. Section 2.— Waimarama Beach ( N 14 2, 442985) Blown sand and beach boulders , 2.oft Occupation layer with shells and charcoal I.oft Blown sand o.2ft Dark clay, charcoal, and moderately abundant Loisels Pumice (H 6) o.3ft Tuffaceous sand 2.oft Estuarine mud with wood and stumps c. s.oft The section at Waimarama Beach (Section 2) extends for about 300 ft from the cliffs at the south end of the beach. It is fairly well exposed and uniform for most of its length. The two layers with charcoal are correlated with the two occupation layers to the north. Loisels Pumice is moderately abundant, but was not seen on beaches to the south except as small fragments of drift pumice. No Taupo Pumice was seen, but the tuffaceous sand below the Loisels pumice probably contains Taupo Pumice sand. The estuarine mud crops out only at the north end of the section, and its relation to the overlying beds is not clear. It probably represents a lower sea level, but no stumps that are certainly in place were seen. This is probably the layer from which Hamilton (1888: 313) reported moa bones. Most of the beaches between Waimarama and Cape Kidnappers were examined, but no clear sections were seen. Cape Kidnappers to Mahia Peninsula From Cape Kidnappers around the coast of Hawke’s Bay to the Wairoa River the outer coast is not favourable for occupation sites, being either rapidly eroding cliffs or exposed gravel beaches. The more favourable localities around the inner harbour at Napier were not examined. At the coast on the north side of Wairoa River a thin section shows about a foot of occupation soil, underlain by Taupo Pumice that rests on a dark clay that appears to have formed when sea level was lower than now. Thicker sections were found at the mouth of Tahaenui Stream, 12 miles east of Wairoa, and in gravel pits near Nuhaka. The Tahaenui Stream section
(Section 3) is exposed on the right bank of the stream about 300 ft from the coast and extends for 50ft. Two ash bands can be identified with reasonable certainty, the Taupo Ash because of the large size of the fragments and the Waimihia Ash by the central dark band (Baumgart, 1954: 465). The load casts at the base of the clay between the two ashes and the evenness of the Waimihia Ash indicate that the lower part of the clay and the Waimihia Ash were deposited in water. The Taupo Ash was deposited on a soil and shower bedding is less well preserved than in the Waimihia Ash. Section 3.— Mouth of Tahaenui Stream ( Nil 6, 055915) Non-pumiceous blown sand o’nf^ Pumiceous blown sand ;• •••• 2.oft Ash with i-inch pumice lumps (Taupo) o.lft Sandy pumiceous clay with load casts at base - I.oft Granule-grade ash with light top and dark base (Waimihia) o.3ft Light granule-grade ash _ (Waimihia) ...... • o.2ft Dense clay silt, no remains of vegetation 2.5 ft Greywacke gravels, top about sft above M.H.W.M. 3.oft Section 4.— Nuhaka Gravel Pit, south of road entrance ( Nll6, c. 085912 Dark grey blown sand with ovens and shells near top. Up to ...... 6.oft Pumiceous blown sand I.oft Pumiceous earthy silt with soil at top •••••• I.oft Granule-grade light ash (? Waimihia Ash) o.7ft Blue Lake silts _ I.oft Old beach gravels, seaward dip of 3°, primary , s.oft The Nuhaka gravel-pit, Section 4, is clearly exposed down to the permanent water .table which is probably not far above M.H.W.M. The gravels at the base of the section are similar to those on the present-day beach, but slightly lower. The beach has prograded, and sea level has probably risen a few feet since the gravels formed. The section and similar ones at Tahaenui Stream and near Gisborne are discussed later in the description of sea-level changes. Dunes composed of pumiceous sand similar to that in the upper part of the last two sections are best developed on the isthmus that connects Mahia Peninsula with the mainland, and are overlain and underlain by non-pumiceous sand. The pumiceous sand can be traced north along the east coast for many miles and represents an enormous volume of pumiceous ash. At the isthmus .the pumice part of the sand is being winnowed by the wind to accumulate in pure layers that are easily confused with directly deposited ash. Farther north the pumice sand is finer, has fewer air cavities, and does not concentrate as readily. A similar dunesand sequence is present on the western side of the North Island at Wanganui, Taupo Pumice sand being present in the younger dunes and absent from the older ones (Fleming, 1953: 117). Most of the likely bays between Nuhaka and Waikokupu and those to the south-east on the western side of Mahia Peninsula were examined, but no clear sections were seen. The southern point of Mahia Peninsula was not examined. The only good section on the western side of the peninsula (Section 5) is about a mile south of the Post Office and store. It forms the upper part of a low cliff and extends for over a thousand feet, the lower 4ft of the cliff being composed of Tertiary rock. The terrace at the top of the section is about 9ft above M.H.W.M. and extends inland for about a hundred feet. It would be considered pre-Holocene and possibly interglacial if the sediments were not clearly dated as Holocene by water-borne Taupo and Loisels pumice. The full section, well exposed for its whole length, is illustrated in a longitudinal section by Fig. 5. The column (Section 5) generalizes the main features. Except at the west end, the base of the section is now sft above M.H.W.S.T. and the two pumice layers are
considerably higher. It seems unlikely that they could have been deposited with sea level at its present position, and recent tectonic uplift is the most probable explanation for their anomalous height. Section s. —Mahia Peninsula, Generalized Section ( Nll7, 290869 ) Moderately developed soil with grass ...... o.2ft Shelly sand o.sft Sandy soil with scattered charcoal I.sft Local occupation layer o.4ft Sandy soil with charcoal o.sft Continuous occupation layer, Loisels Pumice in layer and at base o.3ft Wind blown sand, Cookia sulcata common 2.oft As above with Taupo Pumice as scattered lumps o.sft Black sand and angular fragments Tertiary sandstone o.sft Shell sand and gravel of Tertiary mudstone resting on top of Tertiary mudstone, c. M.H.W.S.T. + sft o.6ft Two miles to the west and opposite the entrance to the lagoon that lies on the north side of Mahia Isthmus, there is a thick Holocene section at the base of a steep slope that rises to hills of Tertiary rock. The sea has repeatedly advanced by eroding the mudflows from the cliff and has repeatedly been driven back by further mudflows. Because of the rapidity of accumulation the stratigraphic record is unusually complete, but large thickness variations and poor exposures make the section difficult to interpret. More detailed examination supplemented by some excavation is required. Because of the variation in the thickness of individual layers the position of M.H.W.M. shown on the accompanying column (Section 6) applies to the lower part of the section only. Section 6. —Mahia Lagoon Section (Oraka Beach) West End. Generalised. ( NII7, 275882) Soil mudflow with scrub on poorly developed soil 2.oft Occupation layer (local only) o.sft Soil mudflow 2.5 ft Tuffaceous sand (? Kaharoa) ...... o.lft Soil with scattered charcoal grading down into , I.oft Fine conglomerate of mudstone with abundant charcoal I.oft Dark sand and Loisels Pumice o.lft White pumiceous sand, possible ash shower o.lft Tertiary mudstone conglomerate 2.oft Very white silty ash overlying charred leaves (? Ohui Ash) 0.02 ft Tuffaceous soil o.2ft Very white silty ash (? Ohui Ash) 0.05 ft Tuffaceous swamp soil with wood fragments o.Bft Water-borne Taupo Pumice, Ift lumps, carbonized wood o.Bft Scree of Tertiary mudstone o.2ft Conglomerate of Tertiary sandstone, shells ...... 2.oft Estimated position of M.H.W.S.T. The thin, very white, silty ash layers immediately above the water-borne Taupo Pumice have not been appreciably disturbed since they were deposited and are clearly subaerial ash. The layer two feet above and that immediately below the Loisels Pumice are less clearly defined, but are probably ash showers. The two occupation layers are distinct, but not very dark in colour, probably because the site was too steep for continuous occupation. From Mahia Peninsula northwards the coast is steeply cliffed in Tertiary rocks until just south of Young Nick’s Head, where two good coastal sections are exposed. The southern section at Maraetaha River is two miles south of the headland and ten miles south-west of Gisborne Town (Section 7). The section is exposed for 70ft on the north-east (left) bank of Maraetaha River and is 600 ft from the coast. Inland the ground surface rises gently for about 200 ft to the base of Tertiary mudstone hills. Except that the occupation layers die out upstream the section is remarkably uniform and complete for its whole length.
Section 7. Maraetaha River Section ( NIO7, 309232 Immature soil with grass Occupation layer, umu (oven) at east end of section ...... ...... o.sft Mudflow of soil and Tertiary mudstone 2.oft Soil layer o.2ft Continuous layer of fine ash (shower or rewash) (M 4) 0.02 ft Soil grades laterally into occupation layer, charcoal at top o.lft Shelly sand, ? blown o.3ft Rain-wash clay, soil at top I.oft Conglomerate of Tertiary mudstone, mudflow _ I.oft Occupation layer, umu (oven) at base, grades down into o.Bft Loisels Pumice, well defined layer. Fragments up to lin o.2ft Tuffaceous soil, rare charcoal ...... o.sft As above, no charcoal o.sft Taupo Lapilli, coarse-sand-grade brown pumice. Jin lumps at top o.2ft Rain-wash clay with mud-flow conglomerate lenses 3.oft Fine-sand-grade tuff (Waimihia Ash) _ _ _ o.sft Coarse-sand-grade even white tuff (Waimihia Ash) o.3ft Base about 2ft above M.H.W.S.T. Although the central dark band was not seen, the identification of the basal ash as Waimihia is reasonably certain. The Taupo Ash contains the usual large lumps of pumice. The absence of water-borne Taupo Pumice and the presence of water-borne Loisels Pumice a foot higher in the section suggests that sea level rose in the interval between the two eruptions. Loisels Pumice is restricted to a thin layer and is not abundant. The upper ash has been partly reworked and is mixed with soil, but is so continuous as to be probably a subaerial ash deposit; it is tentatively correlated with the Kaharoa Eruption. Conspicuous Maori fortification embankments that have not been appreciably eroded and cannot be more than a few hundred years old are built on both banks of Maraetaha River about 600 ft inland. Section 8 is well exposed on the left bank of the river at the centre of the fortifications, and extends for 50ft. Although the river is tidal, the section is too far inland to be reached by seaborne pumice, and so the Loisels Pumice is absent. Deposition has been slow but, judging from the position of the Taupo Ash, probably fairly uniform, and it is reasonably certain that the ill-defined dark band and the underlying silt with very rare charcoal and a few burnt stones represent the lower occupation layer on the coast. Most sites of archaeological interest are too far inland to receive water-borne pumice, and this section is given as an example of how correlation may be effected at a favourable locality. Section B. —Maraetaha River, Diggings at Maori Earthworks ( NlO7 300230) River silt deposited in c. 1930 o.sft Occupation layer, at places 3ft thick, average I.oft River silt o.sft 111-defined dark band, probably lower occupation layer I.oft Silt with very rare charcoal and a few burnt stones 2.oft Silt, no evidence for human occupation I.oft Taupo Ash, coarse-sand-grade brown pumice (M 17) o.2ft Silt, no evidence of human occupation I.oft Ash, medium-sand-grade, lenses only (? Waimihia) o.2ft Silt exposed down to M.H.W.S.T. 9.oft The northern coastal section (9) is at Orongo Beach, a mile south of Young Nick s Head, and extends from the rock cliffs at the south end of the beach, where it is thin, for 300 ft north to disappear under the modern beach gravels. The identity of the two lowest ash showers Waimihia and Taupo — reasonably certain. Loisels Pumice is rare. Its presence above the subaerial Taupo Ash and the absence of water-borne Taupo Pumice suggests a rise in sea level in the interval between the two eruptions. That the Taupo Ash is now being eroded by waves is a further indication of a post-Taupo sea-level rise. The origin and
identity of the two tuffaceous layers in the upper part of the section are uncertain. Mr Leo Fowler, of Gisborne, found glass and pig bones at the base of the upper occupation layer. The upper tuffaceous layer is thus less than 200 years old and, if actually a subaerial ash, must be correlated with the Tarawera Eruption of 1888. Section 9. —Orongo Beach Section ( NlO7, 329254) Soil from drain, variable up to 2.oft Continuous light-grey tuffaceous layer 0.02 ft Sandy soil o.sft Occupation layer with glass and pig bones at base I.oft Tuffaceous layer, ? Kaharoa (M 3) o.lft Occupation layer with many shells I.oft Tuffaceous soil o.sft Tuffaceous soil with scattered charcoal I.oft As above with rare Loisels Pumice o.lft As above with no charcoal I.oft Ash with lin pumice fragments at top (Taupo Lapilli) o.sft Soil, tuffaceous 3.6 ft Ash, fine white above, coarser below (Waimihia) o.sft Sequence uncertain ? 10.0 ft Old soil with tree stumps 3.oft The wide alluvial flats of the Waipaoa Valley lie to the north of Young Nick’s Head and meet the sea at Poverty Bay. The bay is fringed by very recent sandhills that are building seaward. Mr W. A. Pullar, of the Soil Bureau at Gisborne, has made a detailed study of the valley soils and kindly supplied the following information on the distribution of the Taupo Ash and sea-borne Taupo Pumice. The ash covers the higher part of the plain inland but is covered by the Waipaoa River alluvium in the lower part of the valley near the river. The sea-borne pumice forms a well-defined bank that lies about half a mile inland and decreases in height from 10f,t above sea level at the north side of the valley to a few feet above sea level at the south side, where it is overlain by river silt. Loisels Pumice was found only near the junction of the old railway line to Motuhora with the main line from Gisborne to Napier. It directly overlies the toe of the Taupo Pumice bank at about sft above sea level. Sections 10, 11 and 12, which are important for indicating a Holocene high sea-level, are exposed on Tuaheni Point on the east side of Poverty Bay. Section 10. Tuamotu Island, south end ( N9B 320425 ) Light coloured sand o.sft Occupation layer, gradational contact at base I.oft Shelly pumiceous sand with water borne Taupo Pumice (M 18) l.Ott As above, but without pumice ■ y| . .... I.oft Ash, silt grade (Waimihia) (M 23) ...... ...... -. o.sft Ash, coarse sand grade (Waimihia) (M 24) o.lft Clay with brackish-water mollusca o.sft Tertiary rock ' The section extends for about 300 ft and is well exposed. The base is 15ft above sea level. Section 11. —Coast opposite Tuamotu Island ( N9B, 426327) Dark soil U.oxt Pumiceous soil ...... I.oft Ash, fragments up to 1 inch (Taupo) ' o.2ft Tuffaceous soil ' * I.sft Laminated fine tuffaceous silt o.sft Ash, fine lapilli (? Waimihia) o.lft Laminated green silt o.3ft Clay, gravel and mudflows, base at M.H.W.M ..... 6.oft
The section matches that at Tuamotu Island, except that the Waimihia Ash is less certainly identifiable. It extends for 200 ft but has been disturbed by mudflows. About a mile east of Section 11, Section 12 is exposed at the low cliff at the end of the Sponge Bay Road from Gisborne. The section lies between headlands of Tertiary rock and extends for about 50ft. It is fairly well exposed. This section, the two just described, and the Nuhaka Gravel Pit and Tahaenui Stream sections are compared and discussed later in the description of sea level changes (Fig. 18). Section 12. —Sponge Bay Road Section ( N9B 433329 ) Gravelly soil, glass fragments o.3ft Ash, lapilli grade (Taupo) o.2ft Greenish-white clay, gradational contact at base I.oft Ash, medium sand grade, partly reworked (Waimihia) : I.oft Greenish-grey swamp clay 6.oft Estuarine clay with woody layers and estuarine shells c. 2.oft Dark carbonaceous siltstone with 3ft logs. Old soil 3.oft M.H.W.M. The estuarine beds with brackish water mollusca (Section 12) extend beneath alluvial flats to crop out north of Gisborne at the south end of Wainui Beach, where they are less well exposed than to the south. Section 13 is exposed at the south end of Tatapouri Beach, six miles northeast of Gisborne, not far from the East Coast Road. The section extends for 500 ft and lies to the east of the hotel on the coast road north-east of Gisborne. It is being gradually eroded back but is mostly hidden by loose sand during the summer. Taupo Pumice is absent, probably because the base of the section is younger than the eruption, but the band of Loisels Pumice is well defined, with some charcoal and burnt stones below. The occupation layers are not well defined, possibly because the site is inhospitable. Section 13. -Tatapouri Beach Section ( N9B , 519389 ) Light blown sand with grass on immature soil I.oft Soil layer, poorly defined . o.3ft Sand, probably blown, grades laterally into occupation layer I.sft Soil layer, poorly defined o.2ft Sand, probably blown, umu 150 ft east near base 2.5 ft Soil layer, discontinuous o.3ft Sand with flat cobbles of Tertiary sandstone at base o.sft Sand with lin lumps of Loisels Pumice concentrated at base o.sft Sand with charcoal and burnt stones ...... o.sft Sand with pebbles of Tertiary sandstone and shells o.sft Base of exposed section at M.H.W.S.T. Loisels Beach (Section 14), the type locality for Loisels Pumice, is 20 miles north of Gisborne, and can be reached by a branch road that leaves the East Coast road 18 miles from Gisborne. The beach is about a mile long and for most of its length is backed by dunes that are being rapidly eroded and now form cliffs up to 12ft high, Loisels Pumice is abundant, the largest lumps being up to 6in across. It was first recognized by Mr W. Waye, of Gisborne, who noticed that it is darker and more dense than Taupo Pumice. The name “ Loisels ” is chosen for the pumice because of its abundance and first recognition on this beach. For most of the length of the beach it is abundant as a band at the base of the lower occupation layer. The clearest section, described in the text column (Section 14), is at a small stream I,oooft south of the point where the road reaches the beach. The change in the thickness of the layers when traced inland is represented by Fig. 6.
Section 14. —Loisels Beach, I,oooft South of road entrance. Blown sand with immature soil and grass o.sft Occupation layer, obsidian common, grading down into I.oft Blown sand, light coloured I.oft Occupation layer, shells, charcoal, fish bones, dog dung, large bone fragments, and abundant obsidian I.oft Sand, blown or beach, worm borings at top o.7ft Loisels Pumice lumps up to 6in. Scattered lumps above o.3ft Light coloured sand, rat bones o.sft Sand as above, no evidence of human occupation 3.oft Cemented shelly sand on Tertiary sandstone at about M.H.W.S.T o.sft Loisels Pumice forms a well defined band six inches below the base of the lower occupation layer. Rat bones immediately below the Loisels Pumice are of interest in view of the suggestion by Sharp (1956) that rats may have arrived earlier than man in some islands, but the bones are not low enough stratigraphically to verify this interesting possibility. Taupo Pumice was seen as a primary water-borne deposit at the north side of the second stream north of the road entrance, where it is separated from a well-defined band of Loisels Pumice by Gin of dark sand. The two occupation layers are distinct for almost the whole length of the beach. Cooks Cove is a small sheltered inlet on the south side of Tolaga Bay, 25 miles north-east of Gisborne. It is separated from Tolaga Bay by a narrow ridge of Tertiary sandstone that has been cut through by the lateral erosion of a small stream which now flows directly into the Bay through a natural tunnel, and has abandoned its former outlet at Cooks Cove. The cove itself was sketched by Cook (Beaglehole, 1956), the amount of vegetation then being similar to that at the present day. Although no permanent habitations were shown, the area seems to have been cultivated. A photograph taken in about 1900 (Duncan, 1903; pi. 7) shows that a thick cover of tall scrub had grown up after Cook’s visit. This advance of scrub is probably related to the decline in Maori population, which
reached its lowest level at about 1900. Section 15, based on continuous exposures on the south side of the cove (Fig. 7), is one of the best in the North Island, and is made the type locality for the Holocene divisions.
Section 15. — Cooks Cove, South Side ( N9O ). Occupation layer, many Haliotis iris o.sft Shelly loam with charcoal fragments o.9ft Tuffaceous horizon, probably Kaharoa Ash shower, extends for 50ft (M 5) _ o.lft Shelly sand with many whole shells, not midden material o.9ft Occupation layer, charcoal, burnt stones, shells, fish bones, rare obsidian and flint o.9ft Loisels Pumice, lumps up to 2in, base well defined o.2ft Shelly beach-sand with rare bones and charcoal at top I.oft Tuffaceous silty sand I.oft Taupo Ash: yellow, medium sand grade tuff, coarse top o.sft Sandy silt with up to 2ft, no standing trunks 3.oft Base 2ft below M.H.W.S.T. Two distinct occupation layers that are correlated with those to the south are separated by shelly sand with a discontinuous light-brown tuffaceous layer, probably subaerial ash, in the centre. A well developed soil at the top of the section makes it probable that the ash is more than 200 years old, and the layer is tentatively correlated with the Kaharoa Ash. The shelly sand is not wind-blown, but appears to be wave-deposited; it is well above present-day average storm waves and suggests either a higher sea level or, more probably, a tsunami or a period of extremely severe storms. Loisels Pumice is abundant and lies in its normal position—at the base of the lower occupation layer and a few inches above the lowest charcoal. The Taupo Eruption is represented by an ash band six inches thick at the head of the inlet but thinner and less distinct seaward, where wave action was stronger, and by sea-borne lumps of white pumice up to Sin across at the top of the ash. Estuarine
sand with 3ft totara logs, but with no stumps in position of growth, underlies the Taupo Ash at the head of the inlet, the lowest logs being several feet lower than timber would accumulate at the present day. Five samples taken from three horizons have been forwarded for G l4 dating: (1) wood from below the Taupo Pumice; (2) charcoal; and (3) shells from immediately above the Loisels Pumice; (4) charcoal; and (5) shells from the upper occupation layer.
No worthwhile sections were seen within the cove on the north side, but the thick section described below (Section 16) has accumulated at the foot of sand* stone cliffs 200 ft north of the entrance. The section extends for 300 ft, but is irregular and not fully exposed for its whole length. The two well defined occupation layers are correlated with those on the south side of the cove, but the base of the lower layer is four feet above the reach of present waves and no Loisels Pumice was seen. The ornament found near the base of the lower occupation layer is composed of coarsely crystallized calcite probably from one of the veins that cut Tertiary sandstone near the cove. The underlying fish-scale layer represents a single catch of several hundred fish and, judging from the size and uniformity of the scales, all snapper. Cooks Cove is a suitable locality for detailed archaeological exploration, particularly of the lower occupation layer. Fig. 8, a sketch from air photographs, shows the position of the two sections 15 and 16. Section 16. —North side of entrance to Cooks Cove ( N9O ) Scree of Tertiary rocks, immature soil above 2.oft. Occupation layer, charcoal, and many shells I.oft Soil and scree, few shells or charcoal 3.oft Occupation layer, calcite ornament 2.oft Layer of fish scales up to iin thick, extends for 6ft 0.03 ft Soil and scree with rare shells and charcoal I.ooft Soil with no charcoal or shells. M.H.W.S.T. at base 3.oft Tolaga Bay is not being eroded and no Holocene sections are exposed. Kaiaua Bay, the next large bay, five miles north of Tolaga Bay, is about two miles long and is backed by sandhills for most of its length. The two occupation layers and the band of Loisels Pumice show at several places, Section 17 being exposed on the north side of the first stream north of the road from the beach. The blown sand with layers of pumice sand at the base of the section is thought to represent the Taupo Eruption. The lens of tuffaceous sand a foot higher in the section is in the same stratigraphic position as the better-defined band at Anaura Bay, six miles north, and is correlated with it. The ash band is tentatively correlated with the Ohui Ash of Coromandel. Section 17. Kaiaua Bay, north side first stream north road entrance. Blown sand, variable up to 3.oft Occupation layer with abundant charcoal 2.oft Blown sand o.sft Occupation layer: charcoal, obsidian, flint, shells, rat bones, large bones of (? seal) at base I.oft Loisels Pumice, scattered layer o.2ft Sand, medium brown I.oft Tuffaceous lens in sand, not well defined (? Ohui Ash) o.lft Sand, medium brown I.oft Blown sand with layers of pumice-sand (? Taupo Ash) 3.oft Anaura Bay, reached by a branch road from the main East Coast Road, is about a mile long. Except at the south end where thickened by soil and mudflows from the hills of Tertiary rock, Holocene sediments are only a few feet thick. A longitudinal section along the south end of the beach is shown by Fig. 9. A representative section showing the lithology in greater detail is given by Section 18. The two occupation layers are well defined. The Loisels Pumice at the base of the lower layer is rare and the fragments smaller than to the south. A thin, inconspicuous tuffaceous band below the Loicels Pumice extends for 400 ft and is so continuous and uniform in thickness that it can hardly be explained other than as an ash shower. Taupo Pumice was not seen, but ,the Taupo Ash is probably represented by the white pumiceous sand exposed near the base of the section at the north end,
Section 18. —Anaura Bay, south end. Mud flow, south end of section only o.sft Occupation layer, continuous and well defined o.sft Clay, mud flow or soil creep o.Bft Lens of shelly sand grading north to occupation layer o.3ft Clay, as above I.4ft Lens of shelly sand, at south end only o.2ft Occupation layer, well defined and continuous, rare Loisels Pumice at base o.sft Clay, as above o.sft Light-grey very fine ash (? Ohui Ash) extends for 400 ft o.lft Clay as above 3.oft Sand, white and tuffaceous (? Taupo Ash) I.oft Beach sand with shells on Tertiary sandstone at about M.H.W.S.T I.oft At Tokomaru Bay, nine miles north of Anaura Bay, a thick Holocene section at the north end of the bay is mostly hidden by a retaining wall built to protect the road from marine erosion. No worth-while sections were seen to the north at Waipiro Bay or at Teporoa Bay. The section at Reporua (19) is the most northern seen on the East Coast and is 60 miles north-east of Gisborne. The section is well exposed but, although it contains two occupation layers that appear to correspond with those at the beaches to the south, it is not extensive enough to be satisfactory. The sediments were mostly deposited by the stream that flows along the foot of the section. Section 19. — Reporua, south side stream near road. Soil with iron and glass I.oft Sandy occupation soil o.sft Soil of clay and fine gravel 2.oft Occupation layer with abundant shells and charcoal o.sft Loisels Pumice, 2in lumps o.lft Sandy soil with scattered charcoal o.7ft As above, no charcoal V o.7ft Blown sand, no charcoal I.oft Clay and fine gravel, stream deposit 2.oft
From Reporua north to Waiapu River the coast consists of either steep, rapidly eroding coast, from which any sections that existed have been eroded, or of sand dunes that are building seaward and appears to be devoid of worthwhile sections. The six miles of coast between the mouth of Waiapu River and East Cape was not
examined. The five Eastland sections that record the lower part of the Holocene are shown together in Fig. 18 and are discussed later under sea-level changes. The remaining eleven are shown, together with the Coromandel and Northland sections, in Fig. 14, the lowest Loisels Pumice being used as the datum horizon. The columns are on the same scale and are arranged from south to north. The lowest Division, 7, is either not exposed or absent because of unconformity in most of the sections, but is probably represented in the Waimarama, Orongo, and Cooks Cove sections. The horizon of the Taupo Eruption, the base of Division 5, is defined by ash or by water-borne pumice in five sections and is considered to be represented by tuffaceous sand in two. Ash that is younger than the Taupo Eruption and slightly older than the oldest charcoal occurs at Mahia Lagoon, Kaiaua Bay, and Anaura Bay. It is thought to represent a short-lived but violent eruption, and is correlated with the Ohui Ash of Coromandel. Division 4, beds with charcoal or other evidence of human occupation below the Loisels Pumice, is thin and remarkably uniform in thickness in all sections except the two at Mahia, where it is absent. The Loisels Pumice is present in all sections and defines the boundary between Divisions 3 and 4. The two occupation layers are reasonably well defined in all the localities that were suitable for human occupation. The boundary between Divisions 2 and 3 lies between the occupation layers and in the Mahia Lagoon, Maraetaha, and Cooks Cove sections is defined by the ash band tentatively correlated with the Kaharoa Eruption. Division 1, the relatively barren layer above the upper occupation layer is present in all sections but varies considerably in thickness and composition. Bay of Plenty Sections The eastern shore of the Bay of Plenty was examined at many places between East Cape and Te Araroa, but although occupation layers are well developed, the coast is not being eroded, and no good Holocene sections were seen. About a mile west of East Cape a marine bench cut in Tertiary mudstone with large boring molluscs still in place, extends for about a hundred feet at 6ft above M.H.W.M. and is directly overlain by 4ft of sandy soil with evidence of human occupation in a conglomerate at the base. The marine bench is probably Holocene in age but cannot be closely dated. A thicker but equally unsatisfactory section is exposed in dunes a mile east of Te Araroa. Below a growing pohutukawa tree with a 2ft trunk ten feet of blown sand contains three distinct occupation layers. At the base, gravel with charcoal and Loisels Pumice rests directly on Tertiary mudstone. Most of the rocky coast from Matakaoa Point to Gape Runaway, and all the accessible bays between Cape Runaway and Opotiki were examined, but the only diagnostic section seen is at Opape (Section 20). Opape is at the last greywacke headland before the sandy beach that extends west for ten miles to Opotiki. The pohutukawa tree is at least a hundred years old, and the ash below the upper layer of scree is tentatively correlated with the Kaharoa Ash. The upper occupation layer is not represented, probably because the site became less favourable as the scree advanced and buried the narrow strip of flat ground that had been occupied earlier. Loisels Pumice is abundant but no Taupo Pumice was seen. Section 20.— Opape, west side of Headland. Scree of greywacke, with 3ft-diameter pohutukawa tree at top 4.oft Dark grey fine volcanic ash, extends for at least 200 ft o.2ft Scree of greywacke, as above 2.oft
Beach sand with scattered charcoal and Loisels Pumice, grades into occupation layer to north ! o.2ft Beach sand with Ift log and umu stones I.oft Beach sand, no charcoal. M.H.W.S.T. I.oft From Opotiki west to Waihopai River the main road follows the coast. No Holocene sections were seen. Section 21, exposed at the west end of Waiotahi Estuary, is reached by a branch road that leaves the main road on the west side of Waiotahi River, crosses the hills to the east side of Ohiwa Estuary, and then strikes east at the foot of coastal cliffs to the headland at the west end of Waiotahi Estuary. The section is about 600 ft west of the end of the road, and extends for 200 ft. Section 21. —Waiotahi Estuary , east end headland. Mudflow and gravel, 6ft pohutukawa growing above I.sft Soil o.sft Peat o.lft Mudflow, obsidian flake o.7ft Light grey fine ash (? Kaharoa) o.2ft Soil with blown sand at base o.sft Mudflow with Sin laminated silt at base I.oft Occupation layer, rare Loisels Pumice o.sft Beach sand, with thin occupation layer at base I.oft Beach sand with rare Loisels Pumice I.oft Loisels Pumice band, 2in lumps, some light-coloured pumice (M2l) o.3ft Sand, with large umu stones o.4ft Beach sand, medium grade. No Loisels Pumice or charcoal 3.oft Greywacke gravels, top at about M.H.W.S.T.
The ash layer near the top is well defined for the whole length of the section. Its position, over two feet below the large pohutukawa tree, makes it certain that it is older than the Tarawera Eruption of 1888, and it is tentatively correlated with the Kaharoa Ash. Loisels Pumice is abundant as a primary deposit, the pumice lumps being up to Sin through. Pumice which is lighter in colour than Loisels Pumice first appears with Loisels Pumice in this section. Judging from its stratigraphic position in sections farther west, it probably came from an early phase of the same eruption. No Taupo Pumice was seen in this section, but what may be Taupo Ash shows in a swampy flat north of the main road at the east end of the Waiotahi Lagoon. The ash, of coarse-sand to lapilli grade, rests on peat. It is overlain by Sin of peat, by Sin of steel-grey ash that is probably Kaharoa, and by bin of peat. A thin but more complete section (22) shows on the seaward side of Wainui Road in a swampy flat near Wainui at about M.H.W.M. The section extends for 300 ft, but the upper part is largely eroded. The pumice immediately below the Loisels Pumice is similar to the light pumice described from Waiotahi Lagoon. From Its stratigraphic position it would appear to immediately pre-date the eruption of the typical dark Loisels Pumice, the interval of time between the two eruptions being sufficient to enable the lighter pumice to be washed ashore before the arrival of the darker material. The underlying very fine white ash is correlated with the Ohui Ash and with that below the lowest charcoal in the three east coast sections. The light pumice near the base of the section is similar to and is correlated with the Taupo Pumice in other sections. Section 22. — Wainui Road Section ( N7B ) Dark sandy soil, mostly eroded I.oft Tough black soil o.2ft Loisels Pumice, up to 2in o.2ft Pumice, ? Loisels, lighter than above, up to 4in o.lft Brown sand, pumiceous o.sft Very fine white pumiceous ash (? Ohui Ash) o.lft Brown sand, as above o.2ft Dark soil and peat o.3ft Lumps of light pumice up to l£in (? Taupo) o.2ft Peat as above o.2ft Brown sand, pumiceous I.oft A fairly detailed examination was made of the spit on the north side of the entrance to Ohiwa Estuary. The outside of the spit is a low shelving beach backed by growing sandhills, and no sections are exposed. The inside of the spit is being eroded, but exposures are not good. The floor of the estuary from H.W.M. down to L.W.M. at least is dark sand with standing stumps up to 2ft through, some of which are probably totara. Judging from the present lower limit of large trees, other than pohutakawa, the stumps definitely indicate a sea level at least 10ft lower than at present. A subaerial ash that is now preserved only as lenses directly overlies the stumps and associated soil. The ash can be traced up into the base of the sandhills, where it overlies dune sand, the stumps and soil being absent inland above the water-table. The best section through the ash showed lin of pumice lapilli, with lumps up to an inch, overlain by half an inch of fine ash, and that in turn by lin of coarse-sand grade ash. The lapilli layer is correlated with the Taupo Pumice, and if the correlation is correct the level of the sea must have risen considerably since the Taupo Eruption. There is no evidence for a corresponding rise at the Wainui Road section, and the rise is probably local and due either to compaction of the soft sediments below the spit or to seaward slumping, possibly caused by earthquakes. A 2ft occupation layer with water-borne Loisels Pumice directly overlies Taupo Ash. The Loisels Pumice is at the same level as drift pumice is being deposited today, and indicates that the rise in sea level took place
before it was deposited. That the occupation layer once extended at least 6ft farther towards the sea is shown by oven stones that have been eroded from it and now litter the beach. Two points are of interest. The preservation of the stumps and old soil only below present permanent water-table makes it almost certain that preservation of wood is largely dependent on absence of oxidising conditions. The only wood that was found preserved in deposits older than the Loisels Pumice in any of the sections examined is either below present-day permanent water-table or within impervious and usually highly carbonaceous clay. The inferred slumping or compaction of the sediments beneath the lagoon suggests that definite evidence for either eustatic or tectonic sea-level changes have to be related fairly directly to compact rock. This may be particularly important in New Zealand. Historic earthquakes are known to have caused slumping and compaction of several feet (Henderson, 1944). Fortunately most of the sections described are not far above compact rock and are unlikely to have compacted or slumped appreciably. From Ohiwa west to Ohope Beach the coast is backed by prograding dunes and no Holocene sections are exposed. Ohope Beach itself, the sheltered bay east of the greywacke cliff's of Whakatane Point, is protected by sea-walls and the base of the thick occupation layer that shows in road cuttings is not exposed. No sections were seen at Whakatane town on the west side of Whakatane Point. From Whakatane a sandy beach extends for thirty miles to the promontory at Maketu. At the several places where examined the beach is prograding, no cliff sections being exposed. Section 23 is exposed at Bledisloe Park, on the east side of Maketu Point, within Waihi Estuary, and is at the foot of a 100 ft cliff. A Maori fort with well-preserved earthworks, known to have been occupied during early European settlement, stands on the top of the cliff above the section. The base of the section is about 20ft above M.H.W.M. and is too high to contain water-borne pumice. The single ash shower is clearly pre-Tarawera and is tentatively correlated with the Kaharoa Ash, but no sample was taken for mineralogical study. The underlying occupation layer is thin but definite. The occupation layers and shell-scree suggest two phases of Maori occupation, the occupation layers indicating occupation of a convenient site at the coast, and the thick scree of shells the later period when defensive positions became important. If the correlation of the ash with that at Kaharoa is correct, then the occupation of the hill fort is an extremely late phase in the Maori history of this district. The shell scree extends beyond the section examined and represents several hundred tons of material that had been carried up the 100 ft from the beach to the top of the hill. Only shells large enough to be useful as food would be expected to have been carried up this height, but this is not the case, about a tenth of the material being too small to be useful. Abundant slave labour is the most likely reason why the shells were not more carefully sorted on the beach. Section 23. —Bledisloe Park, Waihi Estuary ( N 5 9, 944503) Scree of shells from Maori fort at top of hill, many species of shells but little charcoal 6.oft Scree of old volcanic rocks, no charcoal, up to 12.0 ft Occupation layer, many shells and much charcoal, umu I.sft Light-grey glassy ash, discontinuous (? Kaharoa) o.lft Scree, as above o.6ft Occupation layer with shells and charcoal o.lft Gravel from volcanic rocks, at least 6.oft No worthwhile sections are exposed at Maketu, on the west side of the headland, or at Mt Maunganui, 16 miles north-west. The favourable inlets in the Tauranga Estuary were not examined in the detail they deserve. No sections are exposed at Bowen town or at Waihi Beach.
Correlation The Bay of Plenty sections are correlated in Fig. 10. They differ considerably in .thickness and location, the most constant feature being the ash shower that is correlated with the Kaharoa Eruption. Coromandel Sections The most southerly beach examined at Coromandel Peninsula is at Whiritoa, about ten miles north of Waihi. Section 24 is exposed at the south end of the beach. Section 24. —Whiritoa Beach South end ( N 5 3, 393079) Dark blown sand ■■ | 'lnoccupation layer *•“- Blown sand, Loisels Pumice (M 8) at base o nr Blown sand, variable thickness up to * 2.oft Very fine light grey ash (Ohui) (Ml 3), up to .... o.3ft Blown sand, base not seen. Top sft above M.H.W.S.T 3.oft The ash band is well defined and can be correlated with reasonable certainty with that exposed in Section 25, which shows near the middle of the beach. The occupation layer below the Loisels Pumice is thicker than usual and the pumice layer may not be primary. The section is not exposed for a sufficient length to check this point. Section 25. Whiritoa Beach, near middle ( N 5 3, 391085 Blown sand 4.oft Occupation layer LOft Blown sand LOft Occupation layer 2.oft Blown sand LOft Occupation layer LOft Blown sand o.sft Occupation layer with Loisels Pumice o.3ft Occupation layer, up to I.7ft Very fine light-grey ash o.2ft Blown sand 2.oft To the north the coast was next examined at Whangamata but no worth-while sections were seen. Section 26 is exposed at Opoutere Inlet, seven miles north of Whangamata. The section is exposed on the west side of a small tidal stream immediately downstream from the road to Ohui Beach. It extends for 20ft and thins to the south. Similar middens show on the road half a mile west and at several places nearby. The section differs from most in that there is no evidence for human occupation before the Loisels Eruption. Section 26. —Opoutere Inlet (Wharekawa Har.) ( N 49 361290) Good soil on shell midden, mostly pipi o.Bft Sandy occupation layer o.3ft Midden, shells mostly pipi . I.sft Sandy occupation layer . o.2ft Midden, shells mostly pipi o.3ft Sandy occupation layer, Loisels Pumice at base o.2ft Tuffaceous sandy clay LOft Tuffaceous clay, M.H.W.M. at base I.2ft The most southerly large primary accumulation of Loisels Pumice is at Ohui Beach, from one to two miles north of Opoutere Estuary. The pumice is in lumps up to a foot through and forms a bank up to three feet thick that extends for the whole length of the beach from 10ft to 15ft above M.H.W.M. The pumice directly overlies an inch to three inches of light ash that extends inland at the base of an ill-defined occupation layer.
To the north the coast was examined at Tairua and at Hot Water Beach, but no worthwhile sections were seen. At Cooks Beach, on the south side of Mercury Bay, Loisels Pumice forms an extensive bank that is somewhat thinner than that at Ohui. The basal tuff band is absent. Dunes that rest on beach sand with sawn kauri logs, lie between the bank of Loisels Pum'ce and the beach. They accumulated rapidly about a hundred years ago, when the kauri forest was felled, and are now being eroded back. The uppermost Holocene division is probably related to this period of coastal aggradation. At Cooks Beach township a thin section exposed near the corner of Esplanade Road, shows blown sand overlying a thin occupation layer with Loisels Pumice at the base resting on brown blown sand. The Loisels Pumice is about 2ft below the top of the section and about 3ft above M.H.W.M. At Riverside Drive, at the extreme south-west end of the beach, a section similar to the one at Esplanade Road rests on 3ft of shelly sand, the base of which is about 4ft above M.H.W.M. The shelly sand appears to be a beach deposit and possibly represents deposits that formed when sea level was several feet higher. No pumice or ash deposits were seen that could be used for dating. No worth-while sections were seen on the road from Goroglen to Whitianga, but to the north two clear sections are exposed on the northern side of the peninsula north of Mercury Bay. The eastern (Section 27) at Mahinapua, about a mile west of Oplto Point, is exposed near the middle of the bay and extends for about 100 ft. It terminates downward in a layer of black sand that rests directly on sandstone. Section 27. —Mahinapua ( N4O, 295767) Moderately well developed soil under scrub o.2ft Light coloured blown sand I.oft Well defined occupation layer 2.oft Yellow gritty sand, fragments charcoal and shells 6.oft As above, with Loisels Pumice o.2ft Shell sand with charcoal fragments I.oft Shell sand, no charcoal ... o.sft Black sand resting on hard sandstone at M.H.W.M. o.2ft Section 28. —Black Jacks Bay, Otama Beach ( N4O, 242783) Moderately developed soil under grass o.2ft Blown sand o.Bft Well defined occupation layer 2.oft Medium brown sand, contains two poor occupation layers 2.oft Medium brown sand with Loisels Pumice I.oft White sand with fragments of charcoal I.oft As above, no charcoal -2. oft Section 28 is exposed at the coast 500f,t east of the stream at the west end of the beach. The top is about 10ft above M.H.W.M. and is at the level of the coast road. Correlation is reasonably certain down to the base of the charcoal fragments, but Taupo Pumice is absent. The northern tip of Coromandel Peninsula was not examined, but a section remarkably similar to the last two described was found at Oamaru Bay, five miles north of Coromandel Town and on the west side of the Peninsula (Section 29). Section 29. —Oamaru Bay Section ( N 3 9, 973752) Fairly well developed soil o.2ft Dark blown sand o.7ft Occupation layer o.7ft Dark shelly earthy sand with scattered charcoal and Loisels Pumice lumps that become common at base 2.oft Shelly sand with rare charcoal, M.H.W.S.T. at base o.Bft Shelly sand and sand (dug out), 3.oft
The section is near the middle of the bay on the north side of a small stream and was traced for 20ft. The top of the section is at road level, the road being 50ft inland and 150 ft from the base of low cliffs. As in the two previous sections charcoal extends down for a few inches below the Loisels Pumice, the pumice being overlain by a weak occupation layer that indicates a smaller population density in this area immediately after the Loisels Eruption than in Eastland. From Coromandel to Thames sea-walls have been built to protect the coast road, and although abundant traces of occupation layers were seen no worth-while Holocene sections remain exposed. At Miranda, ten miles west of Thames, extensive shell beds appear to be partly midden material. Several sections are exposed, but neither Taupo nor Loisels Pumice was found. No worth-while sections were found to the north on the western side of the Firth of Thames. Even drift pumice appears to be absent. The shell beds record a high sea level during the Holocene, and have been described by Schofield (1960). Auckland and Northland Sections The Auckland Isthmus was an extremely favourable area for Maori occupation and supported a large population. Occupation soils are thick and extensive, but buildings, roads, and sea-walls now hide coastal sections. The first section seen on the east coast is at Hatfield Beach, 20 miles north of Auckland City (Section 30). The section is well exposed, and extends for several hundred feet along the central part of the beach. Progressive increase in degree of weathering with age is conspicuous. Loisels Pumice is absent in the section and as drift material from the beach. The brown pumice is correlated with that at the Leigh Section described below. Section 30.— Hatfield Beach, central part ( N3B, 213955) Blown sand 2.oft Occupation layer k o.sft Blown sand, charcoal fragments I.oft Blown sandy soal, charcoal fragments o.Bft Brown sandy soil, no charcoal o.sft Brown pumice (Leigh) in lin lumps o.lft Yellow brown sandy soil, possible break at base o.sft Bright yellow sandy soil 3 Oft Soft sand (at M.H.W.M.) o.sft Yellow sand 4.oft Eighteen miles north, Section 31 is exposed at Ti Point, near Leigh, The pumice although slightly browner in colour was first thought to be Taupo Pumice, but it is mineralogically distinct (see Appendix), with an entirely different distribution (Fig. 2), and is named Leigh Pumice, the type locality being the section described below. The site of the volcano of origin is uncertain. Section 31. —Ti Point, near Leigh ( N 3 4, 300250) Soil, moderately developed . o.3ft Occupation layer, shells and charcoal I.2ft Earthy pebbly soil, rare charcoal o.sft Pebbly soil, no charcoal, with light brown pumice (Leigh Pumice) (M2O) at base . . . i.Qft Hard iron-stained grit, M.H.W.S.T. at base I.oft Pea-sized gravel, similar to that on beach today I.oft The coast was next visited six miles north-west of Leigh, but no clear section is exposed. Section 32 shows at Mangawhai Estuary, 16 miles further north-west. The section is fairly well exposed beneath the roots of macrocarpa trees and is being slowly eroded back by the sea. It is about 400 ft from the main east coast road, from which it is separated by a narrow tidal channel. Loisels Pumice is
extremely rare, and the horizon of the lowest fragment is ill defined. It is not known farther south in Northland. Section 32.— Mangawhai Estuary ( N2B, 098478) Moderately developed soil, 3ft macrocarpa trees, iron wire I.oft Light sand o.3ft Sandy soil, charcoal o.sft Clean sand I.oft Occupation layer 2.oft As above, with Loisels Pumice, base not well-defined o.2ft Occupation layer, grades down into ...... I.oft Sand, no charcoal, rests on Tertiary mudstone o.sft
Three miles north a road has recently been constructed from Molesworth to the headland on the north side of the Mangawhai Estuary. Section 33 was exposed in the road cutting at the headland, but will soon be hidden by drifting sand. The column is generalized from a longitudinal section (Fig. 11). It is an example of a dune-sand section that can be dated reasonably accurately because it is unusually well exposed. The base of the section is 50ft above sea level and the water-borne pumice fragments are small and extremely rare and were probably blown from the beach below. The back of the section slopes up at 20°. The part on the seaward side of the road has been concentrated by the wind so that obsidian flakes and other artifacts now rest on the brown dune-sand at the base of the section. In this section and in most of the others in Northland weathering is more noticeable than to the south. Section 33.— Molesworth Headland (N 2 9, 108539) Loose blown sand 2.oft Occupation layer, weak, grading into dark soil o.7ft
Light yellow sand with shells and charcoal 2.Oft Loose sand with rare Loisels Pumice at base ...... •••••• . •••••. ••••••. • o*2ft Earthy sand, rare charcoal, and shells •;•••■ As above, no shells or charcoal. Very rare Leigh pumice o.4ft Brown dune-sand. Road level 50ft above sea level 4.oft To the north the coast was examined at Waipu Cove and at several places on the south side of Whangarei Harbour, but no worth-while sections were seen. Section 34 shows at Ngunguru Beach, ten miles north-east of Whangarei. The section extends for almost half a mile but is hidden for much of its length by retaining-walls. For the whole of its length it is remarkably uniform but somewhat thinner than desirable. The lower pumice is similar to that at Leigh and is correlated with it. Only a few small fragments of Loisels Pumice were found, but they occur in their correct stratigraphic position and are probably primary. The occupation layer is well-defined and continuous, in keeping with the favourable nature of the site. Section 34. —Ngunguru Beach ( N2O , 015085) Moderately good soil o.3ft Occupation layer with shells and charcoal I.oft Earthy sand, with few shells and little charcoal, very rare small fragments of Loisels Pumice at base . o.sft As above, but without Loisels Pumice o.sft Light yellow sand, no charcoal o.sft Loose sand with Leigh Pumice (M2O) o.sft Light yellow sand, with M.H.W.S.T. at base 2.oft As above, dug out 2.oft Section 35 is exposed about two miles to the east and extends for I,oooft from Ngunguru School to the headland at the entrance to Ngunguru Harbour. A longitudinal section drawn with a highly exaggerated vertical scale is shown by Fig. 12, Section 35 below being generalized from the longitudinal section.
Section 35. —Generalized Section, Ngunguru School ( N2O, 025089) Thin soil ...... ....:, o.2ft Blown sand ...... ...... I.sft Occupation layer with many shells ...... o.2ft Dark sandy soil ...„; o.4ft Occupation layer with shells and charcoal o.sft Earthy sand with rare charcoal I.oft Rusty sand with greywacke pebbles, Leigh Pumice at base I.oft Rusty sand with iron nodules, M.H.W.S.T. at base I.sft Well sorted sand, probably wind blown. Base not seen I.oft Soil with tree roots and wood, lower sea level 2.oft The basal beds with tree roots and wood that indicate a lower sea level are exposed only at the east end of the section. The overlying white sand is probably wind-blown; it thickens westward. Above comes rusty sand with iron nodules that is overlain by a thin layer of light-brown pumice. The pumice is correlated with Leigh Pumice. It is most abundant at about two feet above M.H.W.S.T. and cuts out when the layer is traced to higher levels to the west or to lower levels to the east, and suggests a sea level close to that of the present day. The progressive downward increase in the degree of weathering of that part of the section that accumulated slowly is much more conspicuous than in the southern part of the North Island, and could probably be used to trace the boundary between the preand post-Leigh deposits inland, the sand above the Leigh Pumice being faintly rusty and without the iron nodules that occur below. The upper part of the section is similar to that at Ngunguru except for the absence of Loisels Pumice. No sections were seen at Tutukaka or Matapouri, on the road north from Ngunguru. The 80 miles from Matapouri to Waitangi were not examined, but it is likely that many good sections are exposed in the numerous bays of this hospitable coast. Section 36 is exposed about two miles north of Waitangi in the first bay south of Onewhero Bay. The basalt bombs are the main point of interest in this section. They are not well exposed but appear to directly underlie Loisels Pumice and to directly overlie yellow soil that from its degree of weathering appears to be about the same age as Leigh Pumice. Their age is thus between A.D. 200 and A.D. 700. They were not seen elsewhere and a local origin is likely. The volcanic cones of Te Puke lie two miles south-west of the bay and in the direction of the prevailing winds. Mr. David Rear, of the New Zealand Geological Survey, who has made a comparative study of the degree of dissection of volcanic cones, considers the cones of Te Puke to be the youngest volcanoes in Northland (pers. comm. 4/4/60). Section 36. —First largish bay south of Onewhero Bay {Nil, 577558) Moderately well developed soil with buffalo grass o.sft Yellow soil with oven stones o.sft Pea-sized gravel with Loisels Pumice at base f o.sft Basalt bombs up to Sin o.3ft Yellow soil ...... o.sft Greywacke beach gravels, resting on greywacke at M.H.W.M. I.oft The section at Onewhero Bay is one of the best in Northland. It is exposed for 700 ft at the south end of the beach, and for a few hundred feet on both sides of the pohutakawa-clad point at the centre of the beach, A cross section of the south end with an exaggerated vertical scale is shown by Fig. 13, and a generalized column that includes information from the north end of the beach is given in Section 37. The basal layer, with trees in position of growth within leached greywacke gravels resting on greywacke, is exposed at the pohutukawa-covered headland at the north end of the beach and appears to extend out below M.H.W.M. to the south. Its relation to the overlying layer of sandy soil with lumps of slightly rusty
Section 37.— Generalized Section, Onewhero Beach {Nil, 565570 ) Poor soil with buffalo grass n '2ft Shelly pea-sized gravel Pea-sized gravel with abundant charcoal ncf Rusty pea-sized gravel with traces of charcoal ...... o.sft Well defined continuous charcoal layer with Loisels Pumice o.lft Rusty pea-sized gravel with traces of charcoal 0-2 ft Light-yellow fine ash (Ohui Ash) i Sandy soil with gravel layers ..... ••••" ••■ Pumice in one inch lumps (Leigh Pumice) (Ml 5) U.lit Rusty gravels and fragments of leached greywacke with trees in position of growth, exposed 2ft below M.H.W.S.T. 2.oft pumice that are correlated with the Leigh Pumice is not clear. A well-defined layer of acidic ash overlies the rusty soil, extends for 200 ft, and is correlated with the ash at Ohui Bay, 130 miles south-east. The overlying section is normal, the Loisels Pumice layer being well defined and lying a few inches above the lowest charcoal. The most striking feature is a continuous charcoal layer that extends for at least 200 ft. It is slightly below the Loisels Pumice. The charcoal is too continuous to be attributed to cooking fires and almost certainly records a major scrub probably the first clearing of the beach prior to permanent settlement by the Maoris. No worth-while sections were seen on the south side of Kerikeri Inlet. At Kerikeri settlement only the top of a section is exposed, the base being hidden by retaining walls. The coast was not examined from Kerikeri settlement to Whangaroa, or from Whangaroa to Mangonui. At Whangaroa the only likely area is hidden by retaining-walls. Section 38 is exposed at Cable Bay, three miles west of Mangonui. The section is typical but somewhat thinner than normal. Loisels Pumice is abundant about 4in above the base of dark soil that is clearly occupational material. The old soil at the base of the section appears to have once extended down below M.H.W.M. and is good evidence for a lower sea level prior to the Leigh Eruption. Section 38. —Cable Bay (Harris’s), South Side Doubtless Bay {N7, 989846 ) Poor soil o.lft Dark soil (occupation layer) with abundant Loisels Pumice at base o.6ft Dark soil as above o.3ft Silty soil, greywacke derived, with Leigh Pumice at base I.7ft Soil as above, M.H.W.S.T. at base I.sft Gritty greywacke residual o.sft Old soil resting on greywacke, indicating low sea level o.2ft About a mile west, a long but poorly exposed section shows below sand dunes on the west side of Oruru Estuary. A fairly well-defined occupation layer with Leigh Pumice overlies Gin of clean sand with Leigh Pumice at its base, which in turn overlies sand with shells but no charcoal. The layer at which Loisels Pumice should occur is well defined, but the pumice was not found. From Oruru Estuary the western shore of Doubtless Bay extends north in a sweeping curve to Karikari Peninsula. South of the peninsula the coast is flanked by dunes that are slowly building out, and the base of the Holocene is not exposed. A well defined bank of Loisels Pumice was traced for I,oooft north of Tokerau at the north end of the beach and is likely to continue for most of the length of the beach. The bank is six inches thick and was traced inland under a sandy occupation layer, and extends seawards until it thins out at about three feet above M.H.W.S.T. Several fairly good sections are exposed in the bays on the south side of Karikari Peninsula, the best being that at Whatuwhiwhi given in Section 39. Whatuwhiwhi is three miles east of Tokerau, and the section is east of the stream in the bay at the end of the road on the south side of the Peninsula, The section extends
for about 20ft, and was partly exposed by excavation. The pumice near the base of the section, named Whatuwhiwhi Pumice, is the only unusual feature. It occurs in sediments that appear to have been deposited when sea level was several feet lower, and underlies brown sand with pumice that is thought to be Leigh. In general appearance the Whatuwhiwhi Pumice is intermediate between the Loisels and Leigh pumices and, if widespread, would be confusing, but it has been found only at Whatuwhiwhi, where it can be shown to be much older than Loisels Pumice and considerably older than the probable Leigh Pumice.
Section 39.— Whatuwhiwhi, north side Doubtless Bay ( N 7, 912977) Poor soil o.2ft Blown sand I.oft Occupation layer, with abundant Loisels Pumice (Mil) at base I.oft Occupation layer ...... o.2ft Shelly sand I.oft Brown sand with ? Leigh Pumice (M 22) I.oft Rusty sand with boulders of Cretaceous sandstone and hard grey (Whatuwhiwhi) pumice (M 25), M.H.W.S.T. at base I.oft Soft sticky clay with sub-fossil pohutukawa roots in place, rests on weathered ? Cretaceous rocks o.sft The exposed east-facing coast of Karikari Peninsula, between Whatuwhiwhi and Matai Bay, four miles north, was not examined. Drift Loisels Pumice is so abundant at Matai Bay as to attract public interest, but the beaches are not being eroded and Section 40, on the north side of the southern of the two Matai bays, was the only one seen. The section extends for 10ft only and consequently cannot be considered reliable. Except that Leigh Pumice is absent, and the occupation layer below the Loisels Pumice is slightly thicker than usual, the section is normal. Section 40.— Matai Bay, Merita, North Side of Southern Bay ( N 7, 910040 ) Good soil o.2ft Soil wash ...... I.sft
Occupation layer 0-2 ft Glean blown sand - o.2ft Occupation layer - Glean sand, as above, with one-inch lumps of Loisels Pumice (Ml2) o.3ft Occupation layer with abundant charcoal I.oft Glean white sand, M.H.W.S.T. at base 2.oft Peaty soil, lower sea level? 2.oft From Cape Karikari the coast was examined for ten miles south-west along Karikari Bay to Blackney Point on the eastern entrance of Rangaunu Harbour and then for ten miles to the south on the eastern side of the harbour. Middens are extensive at Karikari Bay, but there is no coastal erosion and no good sections are exposed. Oven stones are the most conspicuous feature of the middens, and are so abundant as to suggest that there was a taboo on the use of those previously used (see Best, 1903: 97). North of Rangaunu Harbour the coast was examined at Pukenui and Houhora, within Houhoro Harbour, along the centre of Great Exhibition Bay, and at Parengarenga Harbour, but the only worth-while sections are at Te Pua and at Te Hapua, within Parengarenga Harbour. Te Pua is two miles south of Paua and lies on the south side of the northeast projecting peninsula on the west side of the harbour where Section 41 was traced for about 200 ft. The basal part of the section is probably pre-Holocene and appears to be separated by an unconformity from the charcoal-bearing upper part. The most conspicuous feature is the charcoal layer, which is well defined and similar to the layer at Onewhero Beach but slightly older as judged from its stratigraphic position relative to the Loisels Pumice. Section 41. —Te Pua, half a mile south-east of Paua, Parengarenga Harbour (. N4, 422369) Fairly well developed soil on Chione layer ...... ...... o.3ft Carbonaceous soil with rare Loisels Pumice at base o.2ft Sandy soil with definite charcoal fragments o.6ft Charcoal layer, on ? unconformity o.lft Rusty sandy soil with kauri roots, M.H.W.S.T. at base 2.5 ft Te Hapua is only two miles north-west of Paua but lies on the northern side of the harbour and is 16 miles distant by road. Section 42 is exposed about I,oooft north of Te Hapua Wharf. The section closely resembles that at Te Pua and is separated into two parts by an unconformity at the base of the charcoal-bearing layers. The lowest part is pre-Holocene and actually outside the scope of this paper. The well-defined old soils with kauri stumps are separated by even-bedded, well-sorted sands that are almost certainly beach sands. The old soils are thought to have formed during minor marine regressions within a general period of marine transgression, probably during the Last Interglacial. The Holocene deposits are thin and show no features of unusual interest. Section 42. —Te Hapua, I,oooft north of Wharf (N2, 413410) Soil, moderately developed o.2ft Shell layer with abundant Loisels Pumice o.sft Occupation layer with Loisels Pumice at base o.sft Occupation layer, unconformity at base o.sft Old soil with kauri stumps in place o.sft Even bedded beach sand, old higher sea level 3.oft Old soil with kauri stumps as above I.oft Even bedded beach sand as above, old higher sea level with M.H.W.S.T. at base 4.oft The coast east and north of Parengarenga Harbour was not examined. Section 43 is poorly exposed at the south end of Twilight Beach, about four miles south of Cape Maria van Dieman and on the west coast of the North Island. The basal part of the section is well exposed for several hundred feet at the south
end of the bay and appears to extend down below M.H.W.M. The upper part is less well exposed and extends for 50ft only. The bay is exposed to westerly gales and the site is inhospitable. The occupation layer is consequently not well developed. Sand-hills slope up gradually at the back of the beach and half-inch lumps of Loisels Pumice have been blown up by the wind to a height of 100 ft. Section 43 — South end Twilight Bay, South of Cape Maria van Diemen ( Nl, 190411) Poor soil on clean sand and shells • o.lft Darkish sandy soil, with umu stones at base I.oft Earthy sand with Loisels Pumice at base o.sft Earthy sand with charcoal fragments I.oft Light yellow sand f .sft Clay with wood, top 3ft above M.H.W.S.T., base not seen I.oft Ninety Mile Beach was examined at several places. Occupation layers, common several hundred feet inland, are mostly wind-eroded or covered by active dunes, but rest on a well-developed soil that indicates a long period of vegetation cover prior to Maori occupation. West-coast sections were examined to the south of Ahipara, Herekino, Whangape, Oponui, Omapere, Maunganui Bluff, and Dargaville, but no primary pumice deposits were found and the sections could not be stratigraphically dated. Drift Loisels Pumice was found as far south as Maunganui Bluff, and some primary deposits were probably missed. The coast was not examined from Dargaville south to the Waikato River. Correlation The nine most complete sections from Coromandel and Northland are correlated in Fig. 14. Loisels Pumice occurs in seven and is used as a basis for correlation. Leigh Pumice occurs in the six Northland sections but is absent from the three Coromandel sections. Division 7, beds with drowned forest that indicate a considerably lower sea level, and Division 6, the sediments above the drowned forest and below Leigh Pumice, occur in two sections only and are considerably more weathered than to the south. Division 4, the beds with charcoal below the Loisels Pumice are thin but extremely uniform in thickness in all sections. Division 3, is represented by sediments with scattered charcoal above Loisels Pumice and below a well-defined occupation layer. On the east coast this division includes the “ lower occupation layer” and its poor development at Northland and Coromandel indicates a lower population density than on the east coast. Division 2, a well-defined occupation layer, overlies in all sections, and is correlated with the upper occupation layer of the east-coast sections. Division 1, the youngest, is represented in most sections by dune-sand that is still accumulating. West Auckland Sections The coast was examined for ten miles south of Waikato River, the only worthwhile section seen being two miles south of the river mouth (Section 44). The section is exposed for about 100 ft on the south side of a small stream. The pumice is almost certainly from the Taupo Eruption and was probably brought to the coast by the Waikato River. The associated rounded lumps of charcoal are similar to those in Hawke’s Bay and the Gisborne district and are doubtless from trees that were charred during the eruption. No Loisels Pumice was seen, but the relation of the occupation layer to the Taupo Pumice is the same as in other sections, the first evidence of human occupation being considerably later than the Taupo Eruption.
Section 44. —Coast two miles south of Waikato River ( Nsl, 227913 ) Poorly developed soil o.2ft Loose sand o.2ft Occupation layer with shells, abundant charcoal, and rare obsidian I.3ft Clay soil, rare charcoal in top few inches 3.oft Rounded pebbles of Jurassic sandstone o.sft Water-borne Taupo Pumice, lumps up to Ift, and water-borne rounded charcoal up to 2in 2.oft Magnetite sand o.sft Tuffaceous soil, M.H.W.S.T. o.sft The eight miles of coast south to Raglan Harbour, and the north shore of the harbour were not examined. Raglan township is a favourable site with an almost continuous layer of occupation material extending for several miles on each side of the township. A well exposed and remarkably uniform eight-foot-high cliff with few depressions that extend to sea level forms the coast, and occupation material is thin and mostly well above the level of accumulation of seaborne pumice. Section 45 is exposed at Lorenzen Bay, two miles west of Raglan township. Section 45. —Lorenzen Bay, Raglan Harbour {N64, 412453) Dark soil with scattered charcoal o.2ft Brown peaty soil o.2ft Water-borne Taupo Pumice, lumps up to 6 inches o.3ft Peaty soil, no charcoal or Taupo Pumice I.oft Gravel with large range of rock types up to 4in I.oft White ash in hollows, deeply weathered, up to o.2ft Deep soil weathered in Tertiary mudstone at M.H.W.M 2.oft Drift Taupo Pumice is now accumulating at the same level as the primary deposit, and sea level during the Taupo Eruption must have been the same (within one or two feet) as at present. The deeply weathered white ash near the base of the section is probably late Pleistocene in age. The deep weathering must have taken place when sea level was several feet lower than at present, and the section is considered to record the rise in sea level at the end of the Last Glaciation. Older Pleistocene deposits are well exposed on the point north-east of Lorenzen Bay, They include gravels composed of acid volcanics, with tree-stumps in position of growth, that extend out below mean sea level and are capped by thick, deeply weathered andesitic ash. A well defined unconformity separates them from the younger deposits just described. At Raglan township a Holocene section is well exposed on the eastern side of the car bridge to the motor camp, and at the township itself, where it extends for a mile (Section 46). The Taupo Pumice is not represented, but from degree of weathering its horizon is estimated as being at the top of the red sandy soil. The lower part of the section indicates a higher sea level than at present and has the greatest interest. It underlies the upper part conformably and is Holocene in age. The estuarine clay is similar to that in the harbour today and contains the same estuarine molluscan species, Chione stutchburyi being dominant. The top of the clay is at least two feet higher than clay being deposited today, and indicates that sea level was at least two feet higher at some time since the end of the Last Glaciation. The pebbles and boulders above the clay probably accumulated near M.H.W.M. and can be correlated with reasonable certainty with the gravels at Lorenzen Bay. Section 46.— Raglan, east side of bridge to Motor Camp ( N 6 4, 311456) Pipi midden I.oft Red sandy soil, probably ash, iron nodules I.oft Beach sand 2.oft Pebbles and boulders, various rocks I.oft Sand passing down into estuarine clay with leaves and shells 2.5 ft Beach sand, M.H.W.M. at top, at least 2.oft
Several extremely favourable sites lie on the north-facing coast outside the Raglan Harbour bar, but occupation material is thinner than would be expected, and no worth-while sections were seen. To the south, the eight miles of westfacing coast between Woody Head and Aotea Harbour is extremely wind-swept and without sheltered inlets. Cliffs from 100 ft to 200 ft high are separated from the coast by a belt of low prograding dunes several hundred feet wide. Because of the abundant and well distributed rainfall the sand has not been eroded down and only surface occupation layers show, and these are neither extensive nor thick. Primary deposits of sea-borne Taupo Pumice behind the coastal dunes at Ruapuke and Toropaurau streams show that sea level during the Taupo Eruption did not differ from that of the present day by more than two feet, but they cannot be related to occupation layers. Good evidence for Maori occupation is found on the Pleistocene dunes that cover the hills behind the cliffs, but no stratigraphic succession is evident. Many acres of flats have been terraced and several thousands of tons of sand from the old dunes dug out and mixed with the soil on the flats. The only obvious occupation sites are hill forts on the highest points with a clear view of the coast. Oblong pits, probably hut sites, are well preserved, with occupation material nearby, Shell-fish remains are remarkably rare, in contrast with their abundance on the hill sites of about the same age at Aotea and Kawhia harbours. A rectangular pillar of limestone that stands six feet out of the ground and weighs about half a ton is a prominent feature at the hill fort north of Toropauru River. The block was transported along the coast for at least two miles and then dragged up to its present position about 200 ft above sea level. The north side of Aotea Harbour was not examined. Two sections that record an early period of Maori occupation show on the southern headland (Sections 47, 48) . The ground rises gently at the back of Section 47, and the rate of accumulation was probably slow and fairly uniform. Section 47. —South side Aotea Harbour. 2 miles south-east of bar ( N 7 3, 362186). Well-developed soil ...... o.2ft Soil-wash with pipi layer near top 2.oft Midden, mostly pipi £ Q£ t Sandy soil with iron nodules 2.oft Iron sand, rusty with granules of water-borne Taupo Pumice I.oft As above, but more rusty and without Taupo Pumice ’ I.oft A similar section (48) shows a mile to the west, at the first point east of the bar at the mouth of Aotea Harbour. It shows two distinct occupation layers, which from their position relative to the Taupo Pumice probably correspond with those in Eastland. Section 48. — South side of Aotea Harbour, I mile south of bar ( N 7 3, 351192) Occupation layer, many pipi £ Q£ Blown sand ™ 0 3ft Sandy soil with fragments of charcoal ...... ", ””” £ - Q£ t Blown sand with rusty bands 0 3ft Light peaty soil with sea-borne Taupo Pumice i nft Tuffaceous soil, M.H.W.M. at base """ 40 £ Andesitic tuffs and gravels with standing stumps that are probably older than the Last Glaciation underlie the Holocene section and extend down below mean sea leve! Many of the stumps have thick resin surrounding the bark and are probably kauri. Aotea Harbour is now near the southern limit of kauri, and the stumps suggest a past climate no colder than at present Sand dunes rise to 200 ft above sea level on the headland between Aotea and Kawhia harbours. There is little vegetation, and sand drifts inland on most days,
making the area extremely unfavourable for habitation. Several hundred tons of shell middens are dotted over the higher parts of the dunes. Most rest on soil up to a foot thick, which indicates that the dunes were covered by vegetation for a long period prior to Maori occupation. The heaped middens, in contrast with the evenly spread middens in warmer sites, indicate that most meals were eaten in shelters. The middens are about 200 ft above sea level, several miles from the shellfish estuaries and at least a mile from the nearest water supply. A clear view of the coast is the only obvious advantage that the site offers. The deposits are probably those of a late period, when warfare was general, but there is no definite stratigraphic evidence for dating. The destruction of the vegetation that once covered the dunes was probably a result of Maori occupation.
South of the dune-covered headland a continuous and fairly thick layer of occupation material covers several hundred acres near Kawhia township. It extends for about a quarter of a mile inland and is thickest at the existing Maori settlement, which lies towards the harbour entrance just inside the dunes. The locality was extremely favourable for Maori settlement, having a large sheltered harbour and an extensive area of soil suitable for growing kumara. Occupation was probably continuous from the earliest times, but unfortunately no good sections are exposed. An excavation on the coast near the centre of the Maori settlement penetrated 9ft of occupation material, the thickest deposit in New
Zealand known to the writer. The deposit thins out to a few feet 20ft on either side of the excavation, and is divided into two parts by a relatively barren layer near the middle. The horizon of the Taupo Pumice was not reached, and correlation of the upper and lower parts of the deposit with the two occupation layers at Aotea Harbour is consequently uncertain. The south side of Kawhia Harbour was examined at a few places near the road. Occupation material is far less abundant than near Kawhia township and no worth-while sections were seen. The large area of dune sand near Taharoa Lake and the nine miles of coast to the south were not examined. The coast was next examined at Marakopa, 12 miles south of Kawhia Harbour. Heaped shell middens on sites that command a good view of the coast are scattered over dunes north of Marakopa River and, like those on the dunes between Aotea and Kawhia harbours, they mostly rest on well-developed soil. At most places gravel is mixed with the upper part of the soil, which contains traces of charcoal, and is absent from the lower part which is without charcoal. The gravel was probably added to the soil by the Maoris .to improve its texture for growing kumara (Taylor, 1958), and it is likely that continued Maori agriculture eventually caused the dunes to be eroded. Advance of dunes from the north side of Marakopa River is forcing the river south, and river erosion has exposed a continuous 500 ft-long section at the township (Section 49). The two occupation layers are well developed and are correlated in Fig. 15 with the two layers at Aotea Harbour. The gravel in the soil below the lower suggests agriculture during the first period of occupation. Over most of the length of the section the horizon of Taupo Pumice is above the reach of waves and only a few small fragments of pumice were found. Section — Marakopa Township, south bank river , Loose sand with poorly developed soil o.sft Occupation layer, up to I.oft Magnetite sand, up to 2.oft Occupation layer . o.2ft Light sandy soil with gravel and charcoal 2.oft Dune sand with fragments of water-borne Taupo Pumice at base 4.oft Rusty soil with pebbles I.sft Pebbly sandy clay o.sft Clay, M.H.W.M. at base 2.oft River gravels and old soil extending below M.S.L. 2.oft Kiritehere Beach, two miles south of Marakopa, is the next sheltered bay. At least two occupation layers show in the sand-hills on the north side of the stream, but they are not well exposed. On the south side wind erosion has penetrated deeper than on most west-coast localities, and a veneer of midden material has been concentrated. In contrast with most of the other North Island sites bones are relatively abundant. The 30 miles of coast south to Awakino River was not examined, and no worth-while sections were seen between Awakino River and Waiiti Beach. Correlation The four most complete west-Auckland sections are correlated in Fig. 15The absence of the Loisels Pumice and of any other well-defined horizon above the Taupo Pumice makes correlation of the occupation layers uncertain. Taranaki Sections Section 50 is built up from exposures on Waiiti Beach between the road and the headland 300 ft north. Taupo Pumice is a foot or so below present drift pumice and indicates a sea level that may have been slightly lower. The overlying andesitic ash is tentatively correlated with the Newall Ash (Taylor, et al, 1954: 142).
Section 50. —Waiiti Beach, North-east end (N 9 9, 028088) Black beach sand with coke in top Sin I.oft Rusty black beach sand I.oft Grey beach sand I.oft Andesitic fine-sand-grade ash (Newall), worm-bored o.2ft Black beach sand with charcoal o.lft Sequence uncertain, about M.H.W.M. 2.oft Yellow-grey clay soil with rare Taupo Pumice lumps at top 3.oft Sticky clay with roots in place, lower sea level 3.oft No sections are exposed at Urenui Beach, which is situated at the mouth of Urenui River, but an excellent section (51) shows two miles to the west at the mouth of Onaero Stream. The section is well exposed for several hundred feet between the mouth of the stream and the headland to the east. The three thin upper ashes are closely spaced and probably represent three outbursts during a single eruptive period. They are correlated with the latest Egmont eruption, the Burrell Shower of Oliver (1931), with reasonable certainty. The underlying ash is thicker, but otherwise similar to those above. From its stratigraphic position it appears to be considerably older, and is correlated with the Newall Ash. Both showers post-date human occupation. Water-borne Taupo Pumice is absent. The Stent Ash is distinctive and useful for correlating Holocene beds on the North Taranaki coast. Clay with stumps in position of growth shows at Urenui Stream Section and at most North Taranaki beaches. The clay extends below mean high-water mark, and was deposited when sea level was several feet lower than at present. It does not appear to have been appreciably eroded and is considerably younger than the immediately underlying thick weathered ash (? Egmont), which also extends below sea level but is not so closely related to present-day topography, and records an earlier period of low sea level. Section 51. —Beach at Mouth of Onaero River, East end ( N 9 9, 925002) Grey blown beach sand with moderately developed soil I.sft Yellow-brown fine-sand-grade andesitic ash (Burrell) (M 26) o.lft Black beach sand 0.05 ft Andesitic ash as above (Burrell) 0.02 ft Black beach sand o.lft Andesitic ash as above (Burrell) 0.02 ft Black beach sand I.oft Yellow-brown fine-sand-grade andesitic ash (Newall) (M 27) o.2ft Black beach sand with abundant charcoal, burnt stones I.sft Brown clay with scattered andesite pebbles I.oft Red clay I.oft As above with rusty root marks o.sft Granule-grade light ash (Stent Ash, upper member) o.lft Clay o.2ft Laminated silt-grade putty-coloured ash (Stent ash) o.lft Clay as above o.lft Coarse tuffaceous ash, much darker mineral (Stent ash, lower member) o.lft Clay with rusty root marks, and stumps up to 3ft in position of growth (lower sea level) 3.oft Weathered brown clay, old ash (? Egmont), top at M.H.W.M 20.0 ft Section 52 is exposed at Onaero Beach, one mile west of Onaero River. The section is well exposed in a small bay between headlands of Tertiary sandstone about half a mile west of the end of the road to the beach. The upper Holocene is poorly represented by two feet of dark sandy soil at the top of the section. No trace was seen of the Burrell or Newall showers or of water-borne Taupo Pumice. The Stent Ash is represented by the three bands seen in the Urunui Stream section. Pumice below the Stent Ash is the most important feature of the section. The pumice is much softer than Taupo Pumice and has a much greater proportion of heavy minerals. It is named the Onaero Pumice, the type locality being the Onaero Beach section here described.
Section 52. —Onaero Beach, west end (N lO9, 903999) Sandy dark soil 2.oft Sand-grade light ash (Stent Ash, upper member) (M2B) 0.05 ft Sandy clay o.3ft Laminated silt-grade putty-coloured ash (Stent Ash mid. mem.) (M2B) o.2ft Sandy clay o.2ft Sand grade light ash (Stent Ash, lower member) (M2B) ..... ...... o.2ft Peaty sand o.6ft Pumice lumps (Onaero Pumice) o.lft Peaty sand I o.sft Pumice lumps as above (Onaero Pumice) (M 29) o.lft Peat with standing logs M.H.W.M. (lower sea level) 6.oft Many sections were examined on the 30 miles of coast to the west, but all are incomplete and disappointing. The most interesting feature is a progressive change in the height of a Pleistocene sea level defined by the contact between dune and underlying beach sand. The contact lies below a thick weathered ash that is tentatively correlated with the Egmont Shower. At Urenui Stream the contact is 30ft above M.H.W.M. It slopes gradually to reach M.H.W.M. 20 miles west, a few miles beyond New Plymouth. The height difference indicates a slight regional tilt of the North Taranaki coast, but the age of the deposits is uncertain and the duration and the rate of tilting are unknown. Two sections are well exposed near Cape Egmont, at the end of Stent Road, a mile north of the Cape, and at the Cape itself (Sections 53, 54). Section 53 is composite, being based on two sections 300 ft apart. Section 53. —Stent Road ( Pungarehui) Beach, north end ( Nllß, 353657) Black sandy soil o.sft Trace andesitic ash o.lft Black sandy soil o.lft Traces of several minor ash showers o.2ft Rusty black sand with Taupo Pumice and andesite boulders o.3ft Reworked ash and andesite pebbles I.sft Andesitic coarse-sand-grade ash (Stratford) o.sft Soil with iron nodules I.oft Gross-bedded andesitic coarse sand and pebbles o.6ft Laminated silt-grade putty-coloured ash (Stent Ash mid. mem.) o.3ft Soft andesite agglomerate 3.oft Peat at about M.H.W.M. 3.oft Hard old agglomerate The thin andesitic ashes at the top probably represent the Newall and Burrell showers. The locality is windswept, with little evidence of human occupation. Taupo Pumice is represented by Ift boulders about 6ft above M.H.W.M., just within reach of present-day storm waves. The underlying o.sft of andesitic ash is correlated with the Stratford Ash, shown by Taylor et al. (1954, Fig. 1) as the youngest of the “ older ” ashes of Taranaki. Stent Ash is represented by its middle member, which is about as thick as at the Onaero Section 35 miles north-east. Peat, above considerably older agglomerate, was formed when the sea was at least 10ft lower than now. Section 54. —Cape Egmont, opposite Lighthouse ( Nllß, 351649) Dark soil, reworked I.oft Andesite ash (Stratford) I.oft Sand with iron nodules o.2ft Rusty sand o.3ft Cross-bedded andesite sand and pebbles I.oft Laminated silt-grade putty-coloured ash (Stent Ash) o.lft Light-coloured tuffaceous silt I.sft Soil of coarse andesitic sand o.sft Old soil with roots (lower sea level) o.2ft Clay, old soil M.H.W.M 2.oft Hard agglomerate, more than ...... 10.0 ft
Section 54 extends for 200 ft between two small headlands of hard agglomerate. The late Holocene is represented by the foot of dark soil at the top of the section. Taupo Pumice was not seen. The lower part of the section is similar to that at Stent Road Beach, which is only a mile to the north.
Correlation The five North Taranaki sections are correlated in Fig. 16. The Urenui Stream Beach and the Onaero Beach sections are combined in one column and the Stent Ash is used as a datum horizon. The upper Holocene is best represented in the beaches to the north-east and the lower Holocene in those to the south-west. West Wellington Sections The coast was examined at several places south-west of Cape Egmont, but no good Holocene sections were seen north-west of Wanganui River (Section 55). The Wanganui River section is composite, the lower part being exposed in the Wanganui River cliff at Aramoho and the upper part in the top of the cliff on the opposite side of the river. A pre-Taupo ash and subaerial Taupo ash are probably represented in the lower part of the section. The height reached by the top of the river-borne bed of Taupo Pumice is the most interesting feature. The top of the pumice is 26ft above normal river level and only six miles from
the coast. The gradient of over 4ft per mile to the river mouth is so steep that it is likely that the river built out several miles seaward immediately after the eruption. Section 55. —Wanganui River , six miles from mouth ( Nl3B, c. 580880 ) Pakeha midden material -Oft Maori occupation layer o.3ft Soil on Taupo Pumice .Oft River-borne Taupo Pumice, irregular bedding with much sand and scattered rounded lumps of charcoal 6.oft Loose pebbles of Taupo Pumice • o.sft Clay of Taupo Pumice 1 .Oft Coarse-sand-grade Taupo Pumice o.lft Light-grey tuffaceous silt (? subaerial Taupo Ash) o.lft Well-developed soil 2.oft Sandy river silt 12.0 ft Light-coloured tough silty clay, probably ash 0.02 ft River-deposited sand, base at river level at about M.H.W.M 4.oft Thirty miles to the south, 10ft of river-borne Taupo Pumice forms a terrace on the north bank of Rangitikei River about a mile from its mouth. The top of the deposit is only a few feet above present flood level and the coast line was probably close to its present position after the aggradation caused by the Taupo eruption. Section 56. —Manawatu River, two miles d.s. Main Road, Diversion Cut ( N 15 2, 770180). Poorly developed soil - o.2ft Medium sand, blown, with numerous granules of Taupo Pumice I.oft Occupation material, much charcoal, few stones and shellfish o.3ft Rusty sand LOft Sand, probably blown LOft Extremely well bedded river sand with horizontal logs near base and scattered lin lumps Taupo Pumice 10. Oft Dark peaty sand base about M.H.W.M. at river level LOft Not seen ? ft Estuarine clay with Chione stutchburyi from excavation ? s.oft The Manawatu River section (56) is extremely well exposed for a length of one mile in an artificial channel that cuts off the large Foxton meander. Flood plain deposits make up most of the exposed part of the section and show remarkably little variation. The estuarine clay at the base is represented by dredged material from below M.H.W.M. that is not exposed. The whole sequence was probably deposited with sea level at about its present position. The Taupo Ash shower just reached the drainage basin of the Manawatu River and no welldefined primary layer of river-borne pumice is present. The lowest lumps of Taupo Pumice probably represent the horizon of the Taupo Eruption, the 10ft of river sand and the overlying beds being post-Taupo in age. As the river sands are flood-plain deposits of wide extent and not merely valley fill they represent a greater volume of post-Taupo detritus than is known elsewhere on the North Island coast, and indicate more than a mile of coastal progradation in the last 2,000 years. The absence of any trace of old soils in the 10ft of river sand indicates unusually rapid deposition. Aggradation on the coast to the north from the flood of pumice washed down the Rangitikei and Wanganui rivers would have allowed the Manawatu River sand to accumulate in the embayments formed to the south, and the river sand probably accumulated shortly after the eruption. A well-defined layer of occupation material extends for a few hundred feet onlyIt is older than a belt of sand dunes with a moderately well developed soil, and far older than the bare dunes of the coastal belt.
Stratigraphic Dates Most longitudinal sections along individual beaches (Figs. 5,7, 9, 12) show a proportional change in the thickness of the Holocene stratigraphic division when traced laterally. Correlation charts (Figs. 14 A and B) show this relation to be general on the eastern side of the North Island for those divisions between the Taupo Pumice and the top of the upper occupation layer. As the simplest explanation, it is assumed that the rate of accumulation remained constant at each section, although more rapid in some than in others, and proportional thicknesses are used to determine stratigraphic dates. This uniformity does not extend to Division 1, the relatively barren layer of blown sand or soil-wash above the upper occupation layer, which at most places accumulated more rapidly than the layers below. Widespread erosion caused by the extensive felling and burning that followed the introduction of steel axes would cause the beaches to build out and more sand to blow inland, and is the most likely, although not a fully satisfactory, explanation of this extremely recent rapid increase in rate of accumulation. A tentative date of A.D. 1800 is adopted for the base of Division 1.
If A.D. 150 is accepted for the Taupo Eruption and A.D. 1800 for the base of Division 1, approximate dates can be estimated from stratigraphy _ for events recorded in the sections that contain Taupo Pumice. Loisels Pumice is the most widely distributed key deposit, and the dates of later and some earlier events are based on the date adopted for it.
Six usefully-thick measured sections contain both Taupo and Loisels Pumice. The stratigraphic date of the Loisels Eruption ranges from A.D. 500 to A.D, 1000. The average, A.D. 700, is adopted. Five sections contain ash that is tentatively correlated with that at Kaharoa. Three sections indicate A.D, 1350 for the date of Kaharoa Eruption, in fair agreement with the A.D. 1050 date given by C l 4. The base of the upper occupation layer probably differs in age from place to place. Average stratigraphic values give A.D. 1450 at Coromandel and Northland and A.D. 1550 on the east coast. Division 4, sediments with charcoal fragments below the Loisels Pumice, is well defined in most sections. Its base, stratigraphically about 200 years older than the pumice, is placed at A.D. 500. The data used to determine these stratigraphic dates are shown in Fig. 17. Dates are classed in one hundred year intervals, the number of sections that indicate each interval being shown by numbers and curves. The accuracy of the stratigraphic dates will be tested and improved as further C l 4 dates become available. Sea Level Changes Except in regions of extremely high tectonic activity present sea-level does not differ from that of the last 5,000 years by more than 15ft and criteria for usefully determining Holocene sea levels have to be accurate within a few feet. The most accurate method is by direct comparison of old deposits with those accumulating now. At most places the coast has not appreciably advanced or retreated during the last few thousand years, and the sediments of successive layers can be directly compared with those being deposited at the same height today. But past variations in wind direction and intensity are unknown, and on exposed beaches slight changes in these would have a pronounced effect. Abnormally high tides and tsunamis (tidal waves) may have been important, but the tsunami that washed over the Gisborne coast in 1947 (Laing, 1954) did not leave any recognizable deposits or cause appreciable erosion. Most of the measured sections have been referred to the upper limit of storm waves, which is defined by the lower limit of terrestrial vegetation, by .the base of cliffs that are being eroded, by drift pumice and wood fragments, and less clearly by the contact between beach and dune sands. In general, the more exposed the beach the more difficult is it to estimate past sea levels, the most accurate determinations being those from sheltered inlets, if they can be related to compact rock. Unless so related, sea-level changes may be apparent only, and due to the compaction of underlying unconsolidated sediments. In a basin in which sediments continuously accumulate to keep pace with a slowly rising sea level the apparent downward movement of a bed at the surface will be greater than that of one at the bottom of the basin, the difference depending on the nature of the sediments and being as much as twice the true rate if the sediments are fine-grained with a high initial water content. A more complex relation is likely after the extremely rapid rise in sea level that followed the Last Glaciation. Sediments would have accumulated rapidly in inlets and bays as sea level rose and compaction may well have been delayed until triggered off by earthquakes well after the rise in sea level. The apparent effect in any particular section will be a late rise in sea level, greatest where the thickness of fine-grained sediments is greatest. The apparent post-Taupo rise in sea level at the Ohiwa Estuary and the drowned forest at Wanganui described by Fleming (1957) and dated by C l 4 as about A.D. 940 may be attributed to this effect. Most of the sections described in the text were deposited on benches cut in solid
rock after the sea level rise at the end of the Last Glaciation, and they cannot have been appreciably affected by compaction. The height of deposits of sea-borne pumice may be used to indicate past sea levels, provided that wave intensity has remained constant. In most sections Loisels Pumice is at the same height as drift pumice is accumulating today, and sea level at the time of the eruptions was the same as now within a foot or two. The average height of sea-borne Taupo Pumice (and the almost coeval Leigh Pumice) is about a foot lower than Loisels Pumice, as determined from all sections in which both pumices occur and sea level during the Taupo Eruption was probably slightly lower than it has been since. Anomalous sections at Mahia Peninsula that may have been tectonically elevated are not included in the average. At many sections soils associated with sea-borne Taupo Pumice are a few feet
below the lowest soil forming today and are overlain by beach sand and gravel. The height difference is additional evidence that sea level during the Taupo Eruption was slightly lower than now. Conclusive evidence for a significant rise and fall in the level of the sea is provided by the Tahuenui Stream and Nuhaka Gravel Pits sections near Nuhaka and by the Tuamotu Island and the two Sponge Bay sections near Gisborne! The five sections are shown together on Fig. 18, and are correlated by the Waimihia and Taupo ash showers. The Waimihia Ash is the most conspicuous surface shower forty miles south of Nuhaka and contains a diagnostic central layer darkened by basic minerals (Baumgart, 1954). The central layer is well developed at the mouth of Tahuenui Stream but was not seen in the thinner deposits farther north. The Taupo Eruption is represented by ash in four of the five sections and by water-borne pumice at Tuamotu Island. Its identity is reasonably certain in all five sections. The lowest Holocene beds exposed consist of beach gravels in the two southerly sections, and of gravels and mudflows and of an old soil with logs and standing stumps in the two northerly. The old soil extends down below M.H.W.M. and is conclusive evidence for a lower sea level than that at the present day. It overlies Tertiary sandstone and the change cannot be due to compaction. The evidence for a lower sea level is not as conclusive in the two southerly sections, but is suggested by the beach gravels being at a lower level than those now forming.
In all these sections the lowest beds are overlain by silts. At Sponge Bay tuamotu Island the silts contain molluscs that have not been transported and show that the silt _ was deposited in sheltered shallow water, either marine or more probably brackish. The silts in the southern sections are similar to those bemg deposited at present in brackish-water lagoons south of Nuhaka, and different from those being deposited in New Zealand fresh-water lakes in that
they contain no plant fragments. The silts were probably deposited below M.H.W.M. and, by comparison with the beds below, provide good evidence for a rise in sea level. The zone within which the rise is considered to have taken place is shown on each of the five columns in Fig. 18 by an upward-directed arrow. The silts are now from 2ft to 17ft above M.H.W.M., and sea level has fallen at some time since they were deposited. The time of fall of sea-level is judged from the change from water-deposited silt to overlying sediments that are interpreted as being old soils, and is indicated on the five columns by a downwarddirected arrow. It is slightly above the Waimihia Ash in some sections and slightly below in others, and was doubtless controlled by the level of sedimentation, the effects of a lower sea level showing first in the sediments that were highest above sea level.
Land was warped appreciably during the Hawke’s Bay earthquake of 1931 (Marshall, 1933), and the total tectonic movement during the Holocene is doubtless significant, at least in the more mobile parts of New Zealand. The changes that for simplicity of description are expressed as solely eustatic are doubtless partly tectonic. The sections extend over 30 miles. The changes are a rise followed by a fall in sea level and not a movement in one direction. This makes it probable that eustatic movements are represented and that sea level stood several feet higher than now during or slightly before the Waimihia Eruption some 4,000 years ago. The high sea level at that time must be recorded by sediments in many Holocene sections, but good evidence was seen at a few places only. At East Cape the beach with boring molluscs may well represent this period of higher sea level, as may the band of shelly sand at the north end of Cook Beach at Coromandel. The shell banks at Miranda, on the western side of the Firth of Thames ( Schofield, 1960), are the best evidence for a high sea level probably at this time. On the west coast, estuarine mud above M.H.W.M. at Raglan Harbour is conclusive evidence, and the beach sand interbedded with the Stent Ash on the north Taranaki coast is somewhat less conclusive. The inferred eustatic sea level changes for the Holocene are shown in Fig. 19.
Volcanic Eruptions
Ash showers and water-borne pumice are invaluable for correlating Holocene deposits in New Zealand. Of the several hundred Holocene volcanic eruptions only the pumice eruptions and the largest of the others have more than local importance for correlation. Ash showers are important as soil-formers, their present-day importance depending directly on thickness and inversely on age. Consequently all thick Holocene ashes are likely to be shown on the ash-shower map by Taylor et al. (1954) of the New Zealand Soil Bureau. The Soil Bureau names are used wherever correlation is reasonably certain. Type localities were not designated for all sections, and the writer wishes to thank Messrs N. S. Taylor and H. S. Gibbs for assistance in correlating the ashes of the coastal sections with those already described inland by the Soil Bureau. Coastal sections have been measured only where they record rapid accumulation, and the ash they contain is generally less eroded and much thicker than that in soil sections. Several previously unrecorded pumice deposits and thin ash showers have already been mentioned and named. The name is that of the locality, where stratigraphic position is well defined, and not that of the eruptive site, which may be uncertain or unknown. The mineralogy of selected ashes and pumices is given in an appendix by Miss G. A. Challis, results essential for correlation being mentioned in the following descriptions, in which the individual Holocene ash deposits and pumice deposits are described in approximate chronological order.
Burrell Ash The name Burrell Ash is used by Taylor et al. (1954) for a post-Taupo ash that covers about 50 square miles to a depth sufficient to control soil formation around Mt Egmont. The ash is named after Mr A. W. Burrell, who discovered a Maori oven ( umu ) beneath it on the slopes of Mt Egmont, and who considered it to be about 500 years old. The oven and ash were first described by Oliver (1931) and samples from the charcoal in the oven taken by Mr H. S. Gibbs below 15 inches of Burrell Ash have an age of about 400 years (Rear, 1957: 79). The date of the ash shower is thus about A.D. 1600. Ash that is tentatively correlated with the Burrell Ash shower occurs in the Urenui Stream Beach section a foot above the ? Newall Ash and 2ft Gin above the lowest charcoal. From its stratigraphic position the date is estimated as A.D. 1500. The ash is in thin layers interbedded with magnetite sand, and indicates at least three eruptions within a period of a few tens or hundreds of years. The total thickness is l|in. From its distribution and andesitic composition a Mt Egmont origin is certain. Kaharoa Ash The Kaharoa Ash is mapped by Taylor et al. (1954) as covering about 2,000 square miles on the south-west side of the Bay of Plenty with a centre near Mt Tarawera. The ash overlies the Taupo Lapilli and is overlain by the Tarawera Ash. According to C l 4 determinations from a single locality the date is between A.D. 1074 and A.D. 1214 (Golson, 1957). Thin ash layers tentatively correlated with the Kaharoa Ash occur in coastal sections between Mahia Peninsula and East Cape and at most sections in the Bay of Plenty. The shower needs to be traced to the margin in order to check these correlations. The Kaharoa Ash at Lake Rerewhakaitu, near the centre of the eruption, does not contain any pumice (pumice being defined here as material that floats in water), and there is no evidence that pumice was erupted. Newall Ash The name Newall is used by Taylor et al. (1954) for the parent material of the Newall gravelly and bouldery sand soil which is mapped on the flanks of Mt Egmont. According to Mr H. S. Gibbs the Newall Ash is younger than the Stratford Ash and older than the Burrell Ash. The name is taken from Newall Road, on the slopes of Mt Egmont. Ash that is tentatively correlated with the Newall Ash occurs at Waiiti Beach, at Urenui Stream Beach and, some 40 miles west, at Stent Beach. At Urenui Stream Beach the ash is 2in thick, andesitic in composition, and of fine-silt grade. It overlies 18in of magnetite sand with abundant charcoal, and clearly post-dates human occupation. From its position relative to Taupo Pumice and to the lowest fragments of charcoal the Newall Ash is thought to have been erupted at about A.D. 1000. Loisels Pumice The Loisels Pumice Eruption took place at a most convenient time for dating archaeological sections in New Zealand-just after the population of the North Island coast had become large enough to leave well-defined traces. The primary sea-borne deposits that mark the time of the eruption are clearly indicated in many east coast sections by the incoming of pumice in abundance. In most sections the pumice becomes progressively less abundant above the primary deposit, but a few sections contain layers of abundant pumice that can .be misidentified as primary. Deposits are accepted as primary only where the incoming of pumice is sudden, and where sediments without pumice but otherwise similar extend down for an appreciable distance below.
Experience has shown that unless Loisels Pumice is found as drift, primary deposits are unlikely to be found exposed nearby. The most southerly primary deposit known is at Waimarama Beach, in Hawke’s Bay. Primary deposits occur in northern Hawke’s Bay at the north side of Mahia Peninsula, and then on almost all the east coast beaches as far north-east as East Cape. On the eastern shore of the Bay of Plenty drift Loisels Pumice is fairly abundant, but no primary deposits are known. In the central part of the bay the pumice is abundant and in larger lumps than on the east coast, but only four primary deposits are known: at Opape, at the Waihopai Estuary, and at two places in Ohiwa Estuary, At Coromandel Peninsula several primary deposits are known on the east coast, and one on the west coast at Oamaru Bay. The pumice lumps at the latter locality are larger and the deposits thicker than those in the Bay of Plenty, and at Whiritoa and Ohui Bays closely overlie a layer of subaerial ash —the Ohui Ash— that may have come from the same volcano as the pumice.
Loisels Pumice was not found either as primary deposits, or as drift in the Firth of Thames south of Oamaru Bay, or on the Auckland east coast south of Leigh. The most southerly primary deposit in Northland is at Mangawai, 30 miles north of Leigh, where the pumice is extremely rare. The most southerly well-defined primary deposit is at Onewhero Bay, in the Bay of Islands. It becomes more common to the north-west, primary deposits being conspicuous within Doubtless Bay, at Gable Bay, Tokerau, and Whatuhiwi. At Matai Bay, on the headland between Doubtless and Great Exhibition bays, drift pumice is so abundant as to attract public curiosity, but the primary deposit is thinner than would be expected from the abundance of the drift material. At the far north primary deposits occur within Parengarenga Harbour at Paua and Te Hapua, but the lumps are smaller and the deposits thinner than to the south. The north coast from North Cape was not visited. On the west coast the only primary deposit seen was at the south end of Twilight Bay, the first bay south of Gape Reinga. Drift pumice becomes progressively less abundant along the west coast farther south, the most southerly lumps seen being at Maunganui Bluff, 120 miles south of Cape Reinga.
The largest primary deposits are at Ohui and Cooks beaches on the west side of Coromandel Peninsula and at Tokerau —the main eastward-facing beach within Doubtless Bay. At Ohui the bank of pumice is up to 3ft thick and extends for at least I,oooft, being covered by blown sand at the south end. The bank is wedge shaped, thickest at about present storm-beach level and thinning out seawards. At Cooks Beach the bank extends for the whole length of the beach. The lumps are smaller and the bank considerably thinner than at Ohui. At Tokerau the bank is about a foot thick; it was traced for 300 ft but is probably much more extensive.
Typical Loisels Pumice is dark-grey when wet and medium-grey when dry. The pumice is so recent that at only a few places has it been stained by weathering. From its colour alone it can be distinguished from Taupo and most other pumices. It cannot be scratched by the finger-nail and lumps less than an inch through are broken with difficulty by the fingers. It is thus stronger than most other pumices and would appear to be extremely suitable for light-weight concrete aggregate if it could be found at a convenient site in sufficient quantity. Being strong, lumps over three inches through are usually not well rounded even after long transport. It is usually streaky, sharply defined grey and light bands extending through about a quarter of the lumps. The texture is fine, gas cavities being smaller, shorter, and less continuous than in Taupo Pumice. It floats well and seldom becomes water-logged, only a small proportion of the pumice in watersaturated primary deposits sinking in water. Small inclusions of dark minerals and volcanic rock are common.
At Cooks Beach, and particularly at Ohui Beach, the pumice is less uniform than to the north or south, there being a downward gradation from dark pumice like that at the type locality, to abundant lighter coloured pumice with large gas cavities and silky threads. The absence of the lighter coloured pumice from the more distanct beaches is probably due to its becoming waterlogged and sinking.
The distribution of Loisels Pumice outlined above and shown by Fig. 3 i nc p. cates an east coast origin. The size distribution makes it certain that the origin is near the southern part of the east coast of the Coromandel Peninsula and it is significant that only here is there a closely associated subaerial ash.
Because of their importance in soil formation Holocene ash showers have been closely studied by the New Zealand Soil Bureau and all the important inland showers are probably known. Loisels Pumice does not match material from any of the known showers, and an inland origin is unlikely. Of the islands in the Bay of Plenty, White Island is an active volcano and Whale Island is thermally active, but they both lie too far south-east to be likely sources. Mayor Island, 16 miles east of Ohui, is in a more likely position, but according to Brothers (1957) there is no evidence for volcanic activity during the last two thousand years. On the other hand dark pumice is reported by Mr Selway Hovell (pers. comm.) to be abundant on the sea bed at 91 fathoms three miles north-east of Mayor Island. It would be interesting to know if large lumps of pumice are abundant at Mayor Island or in the other Bay of Plenty islands and if they are underlain by a related ash deposit.
The Loisels Pumice provides a key horizon for correlating archaeological and other Holocene sections on the east coast of the North Island, and its age is important. From its most frequent stratigraphic position relative to the Taupo Pumice of A.D. 150 and to the base of superficial dunes (Fig. 17) its age is estimated as being about A.D. 700. A more certain age should be established from the C l 4 samples now being processed.
Ohui Ash A well defined ash layer that was traced for I.oooft at Ohui Beach, on the western side of Coromandel Peninsula, is named the Ohui Ash. It is about Sin thick, is light brown-grey in colour, and ranges in grain size from fine silt to lapilli. It directly underlies a thick layer of sea-borne Loisels Pumice that includes light as well as the normal dark pumice. The lapilli fragments in the ash closely match the lighter-coloured pumice fragments above. At Whitiroa Beach, 16 miles south, the ash up to Sin thick, and finer grained than at Ohui, but with the same mineral content. Also with the same mineral content is an ash layer at Onewhero Beach, Northland, 120 miles north of Ohui. It is about an inch thick, light brown-grey in colour, of silt grade, and was traced for 100 ft only. The ash at the three localities lies at the base of the lowest charcoal fragments and underlies Loisels Pumice, From the Onewhero and Whitiroa section the ash is estimated to be about 200 years older than the Loisels Pumice and is tentatively dated A.D. 500.
Ash in about the same stratigraphic position that may represent the southern part of the Ohui Shower has already been mentioned in the description of sections in the Bay of Plenty and on the east coast. No samples were taken and mineral content is uncertain. The Bay of Plenty deposit at Wainui Road is an inch thick and midway between Taupo and Loisels Pumice. The east coast deposits at Anaura and Kaiaua bays, north of Gisborne, and at Mahia Lagoon, to the south are, on the average, stratigraphically closer to Loisels than to Taupo Pumice. Distribution and mineral content suggest that the Ohui Ash came from the same
eruptive centre as the Loisels Pumice. In spite of the small thickness of Ohui Ash at all known localities its wide extent indicates that a large volume of material was erupted.
Taupo Pumice Much of the 1 aupo Lapilli shower is pumice, and an enormous amount was transported out to sea shortly after the eruption and washed up on distant beaches to provide a useful stratigraphic key bed. On its way to the sea the pumice dammed the rivers, leaving pumice terraces that are conspicuous in the Wanganui (Fleming, 1957), Waikato, Wairoa, Mohaka, and Rangitikei rivers. Drift Taupo Pumice is common on west coast beaches of the South Island, as far as Greymouth, and on east coast beaches as far south as Kaikoura, but no definite primary deposits are known except at D’Urville Island, at the north end of the South Island. The “pumice beach” mentioned by Fleming (1933; 119) at Wairau Bay, near Blenheim, is no more than extremely rare scattered lumps which may not be primary and which cannot be related to the “ Moa Hunter ” deposits of Duff (1956).
Primary deposits of sea-borne Taupo Pumice occur on the western and southeastern coast of the North Island but are not known with certainty between North Cape and East Cape. They have been described from Wanganui by Fleming (1957), and from between the Manawatu and Otaki rivers by Adkin (1952). The most southerly North Island deposit is at Gape Palliser. By far the largest deposit is that on the Gisborne flats, a few miles south of Gisborne city. Small deposits occur at Mahia Peninsula, Loisels Beach, and Cooks Cove. What may be Taupo Pumice occurs at Wainui Road, in the Bay of Plenty, but no deposits are known at Coromandel, in the Firth of Thames, or on the west coast of Northland peninsula.
Thick deposits occur a few miles south of Waikato River, and thinner ones farther south at Raglan and Aotea harbours, and on the coast between Raglan and Kawhia harbours. Thin deposits occur on the North Taranaki coast at Waiiti Beach and near Cape Egmont. They probably occur, but were not seen, on the South Taranaki coast.
The Taupo Eruption was much larger than the Loisels, but it took place considerably earlier, and many primary sea-borne deposits have been eroded or buried. The distribution of the pumice is shown by Fig. 2. The pumice Itself is that commonly sold in New Zealand as an abrasive and is so well known as to be regarded as typical pumice. It is light yellow-brown in colour but is often superficially stained and has been confused with Loisels Pumice for this reason. Gas cavities are irregular and usually strongly elongated. Most worn pebbles are ellipsoidal, the long axis of the pebbles being in the direction of cavity elongation. In addition to the water-borne pumice, much volcanic sand from the eruption was transported to the beaches and mixed with ground-up pumice to produce pumiceous sand that is conspicuous in sand dunes at Wanganui (Fleming, 1953) and southwards on the west coast, and from Mahia northwards on the east coast.
Taupo Ash The Taupo Ash is shown by Taylor et al. (1954) as covering 8,000 square miles in the centre of the North Island. It js by far the largest ash shower known in New Zealand and ranks with the largest historical eruptions. The most detailed study is by Baumgart (1954), who named several component members and defined their thickness by isopachs in the central area near Lake Taupo. The upper member—the Taupo Lapilli—is by far the most important, extending farthest from the centre and having a volume greater than that of the other members combined. Away from the centre the Rotongaio Ash is most useful
for correlation. It forms a resistant dark band that is conspicuous when only an eighth of an inch thick, being one of the thinnest extensive formations known. It is well shown in ash sections between Te Whaiti and Gisborne (Fig. 20). At the shower margin the Lapilli Member can be recognized by its large pumice fragments. At Gisborne, almost 100 miles from the eruptive centre, they are up to in long, providing a striking indication of the intense atmospheric turbulence caused by the eruption.
Taupo Ash extends farther east than is shown by Taylor et al, in their soil map and makes up part of their Gisborne Ash. It reaches the east coast and is interbedded in coastal sections at Nuhaka, Mahia Peninsula, Maraetaha Beach, Orongo Beach, and Cooks Cove. It probably occurs at the Bay of Plenty at Waihopai Lagoon and Wainui Road, but was not seen farther north. It may have extended north-west to Hamilton, where thin lapilli ash was seen at a few localities favourable for preservation, and south-west to Wanganui, where tuffaceous sii underlies water-deposited Taupo pumice at Aramoho. The progressive thinning of the Taupo Ash and its relation to older and younger ashes when traced from Te Whaiti to near Gisborne is shown by Fig. 20. The distributions of the subaerial ash and the water-borne deposits in the main river valleys and on the coast are shown by Fig. 2. About twenty samples of wood and charcoal that are closely associated either with the Taupo Ash or with water-borne Taupo Pumice have been dated by C Except for two mutually consistent but anomalous dates of B.C. 450 iron l Wanganui (Fleming, 1957, corrected for industrial effect) the dates range l rorn
A.D. 50 to A.D. 250 and are fairly evenly distributed around the average of A.D. 150, which is adopted here as the most probable date of the eruption.
Leigh Pumice
Pumice that because of its stratigraphic position and appearance was first correlated with Taupo Pumice is widely distributed as scattered sea-borne lumps up to lin through, defining a reference horizon from Hadfields Beach to Parengarenga Harbour. The pumice is slightly yellower than Taupo Pumice and has a consistently higher refractive index. The name is taken from Leigh, 40 miles north of Auckland, where the pumice is well exposed at Ti Point. Stratigraphic dates, based on its stratigraphic interval below Loisels Pumice, range from 300 B.C. (one section) to A.D. 1 to A.D. 200 (three sections). A.D. 1, 200 years older than Taupo Pumice, is adopted as the most probable stratigraphic date, but until a section is found that contains both Taupo and Leigh pumice, most probably in the Bay of Plenty, the relative ages of the two pumices will remain uncertain. The origin of the pumice is unknown. Its western-Northland distribution makes the centre of the North Island unlikely, and the Bay of Plenty or an origin to the east of New Zealand is more probable.
Stratford Ash
Taylor et al. (1954) show the Stratford Ash covering an area of about 500 square miles with a centre near Mt Egmont. No type locality is given, but Sin to 9in of Stratford Ash is described (p. 138) as resting on the considerably older Egmont Ash. In the map legend the Stratford Ash is shown as being older than the Taupo Ash. Ash that is tentatively correlated with the Stratford Ash was first seen a few miles south-west of New Plymouth in the Cape Egmont and Stent Beach sections. The shower is mapped by Taylor et al. as extending to within a few miles of Urenui, but the ash was not seen in the coastal section on the eastern side of the North Taranaki coast. At Cape Egmont the ash is Ift thick, andesitic in composition, and of coarse-sand grade. A few miles to the north, at Stent Beach, the ash is similar in composition and grain-size but only half as thick. At Stent Beach the Stratford Ash overlies the Stent Ash and underlies Taupo Pumice. From its stratigraphic position the ash is considered to date about 1,000 B.C. The composition of the ash and its coarse grain size make an origin from Mt Egmont reasonably certain.
Stent Ash
The Stent Ash has been traced for 50 miles along the North Taranaki coast and is used as the datum for section correlation (Fig. 16). At the type locality, Stent Beach, near Gape Egmont, the ash overlies Sin of old soil which rests on soft agglomerate, which in turn rests on peat that formed when sea level was several feet lower. It is overlain by a distinctive 2ft layer of cross-bedded andesitic ash with water-worn pebbles of andesite —a wave-sorted ash that may indicate a sea level higher than that of the present day.
At Urenui Stream Beach and at Onaero Beach bands of coarse-sand grade andesitic ash 2in to Sin thick overlie and underlie the Stent Ash and are separated from it by 2in to Sin of clay. To the west, at Stent Beach and Gape Egmont, the upper ash layer is represented by the wave-sorted material but the lower layer is absent. Close stratigraphic association indicates that the three ash layers are of about the same age and from the same source, and they are named the Stent Ash Group.
The Stent Ash is of silt grade at the type locality and becomes slightly finergrained to the east. The colour, a light brown-grey, is the same in all sections. The ash is easily identified because of its close resemblance to putty in colour and
texture. Judging from its stratigraphic position relative to the Taupo Pumice the ash was probably erupted about 2,000 B.C. A Mt Egmont origin is likely from mineralogy.
Onaero Pumice
The Onaero Pumice is known only on the north Taranaki Coast from New Plymouth and at the type locality at Onaero Beach 20 miles north-west, where it occurs in two thin layers Gin apart, within peaty sand, that rest on peat with standing logs, and is overlain by sandy clay that contains the Stent Ash.
The pumice fragments range from |in to lin in diameter and are sub-angular and slightly rusty. When first collected they were softened by weathering and could be crushed easily between the fingers, but hardened on drying. The subangular shape is probably due to weathering. Many evenly distributed dark crystals about a millimetre long give the pumice a distinctive speckled appearance. The pumice is not streaky and does not contain any obvious rock inclusions.
From its stratigraphic position the pumice is judged to be about a thousand years older than the Stent Ash, and its date is estimated to be about 3,000 B.C. A Mt Egmont origin is likely from mineralogy.
Whatuwhiwhi Pumice
The Whatuwhiwhi Pumice is known only from the type section at Whatuwhiwhi Beach on the south-east side of Karekare Peninsula, Northland. It is mixed with an old soil directly overlying clay with sub-fossil pohutukawa roots in position of growth, and is about a foot below brown sand with pumice that is correlated with the Leigh Pumice.
The pumice lumps are moderately well rounded and from |in to 2in in diameter. They are rusty but not appreciably weathered, the original colour being a uniform medium grey. The pumice is not streaky and has no visible mineral or rock inclusions. In general its appearance is intermediate between that of the Taupo and the Loisels pumices. From stratigraphic position alone its date is estimated as being about 1,000 8.C., but its association with deposits that formed when sea level was several feet lower make a considerably greater age probable.
Magnitude of Eruptions and “Active Volcanoes ” of New Zealand
According to the limited data available (Baumgart, 1954, map), within the central part of most New Zealand ash showers the ash thickness decreases by half for equal distance intervals from the centre. If the central thickness is taken as a and measured in feet, and the distance in which the thickness decreases by half taken as b and measured in miles, the volume of the shower in cubic miles can be shown to be approximately 1/500 ab 2 . The value of bis a measure of the “ explosiveness of the eruption and is the most critical parameter in determining the volume of the central part. Marginal thicknesses are considerably greater than indicated by this relation and decrease more nearly in proportion to the square of the distance from the centre. Volumes cannot be determined from this relation, but the average value of the product of the square of the distance in miles by the thickness in inches is probably proportional to volume and is used to estimate the magnitude of the showers for which no isopachs are availableThe estimated volume, probable volcano of origin, and characteristic mineralogical features of the main Holocene ash and pumice deposits are given in Table I*
Gutenberg and Richter (1949: 253) listed as active volcanoes all vents that, according to available historical records, have emitted lava, ash, or smoke. Historical records extend back less than two hundred years in countries like New Zealand, but more than two thousand years in countries of old civilizations, and the number of volcanoes listed as active in any particular volcanic region depends directly on the length of its recorded history. These biased records can now be rectified by C l 4 dating of volcanic deposits and a more precise definition of “ active ” is now required. It is suggested that all vents that have emitted lava or ash since the beginning of the year A.D. 1 be classed as active volcanoes. According to this definition the active volcanoes of New Zealand include Mt Egmont, or more probably the parasitic cone of Fanthams Peak, Taupo, Rainbow Mountain, possibly Mt Edgecumbe, a vent in the Bay of Plenty probably near Mayor Island, Mt. Rangitoto at Auckland, and Te Puke, in Northland, as well as the vents of historical eruptions already listed by Gutenburg and Richter (1949) at Ruapehu, Ngauruhoe, Tongariro (Te Mari), Tarawera, and White Island. The position of the New Zealand active volcanoes and an indication of the volume of material erupted since x\.D. 1 is shown in Fig. 21.
Fossil Soils
The greater part of the most coastal sections accumulated well above M.H.W.M., the rate of accumulation being gradual and from a tenth to a hundredth of an inch per year. During the Holocene, climate was much as now, and vegetation would have grown on the sediments as they accumulated. Thus a large part of the sections are soils of accumulation, but worm borings and poorly defined root impressions are the only direct evidence that they were soils. An even greater part of inland deposits is likely to be old soil, and in general soils must be considered to be the rule rather than the exception in terrestrial sediments that accumulated slowly. The most common features of such old soils are a tendency to cuboidal fracture and a lack of distinct fine bedding, the two features being related and probably largely caused by the activity of earthworms. Such old soils are often recorded as loess or soil-wash by geologists, the inferred mode of origin being correct but the conditions of accumulation not fully expressed. The complete absence of traces of vegetation is due to oxidization, which is pronounced above water-table except in arctic and desert regions and in areas of extremely high rainfall. The effectiveness of oxidization is most clearly shown by the way in which the stumps in old soils disappear in almost all sections when traced above the water-table of the present day, and by the absence of timber in archaeological sites except in peat, airtight clay, or desert sites (Wheeler, 1956: 99).
Layers that are commonly regarded as old soils occur in many sections. They are of two kinds: (1) peaty or weathered layers interbedded in well-bedded and more rapidly deposited layers; (2) dark layers that look similar to the humus layer of some present-day soils. The first kind represent minor unconformities or periods of still-stand and are more important in Pleistocene than in Holocene sections. The second kind, with more Holocene interest, have been partly misinterpreted. Charcoal is extremely resistant to weathering, and if mixed with soil permanently darkens it. Such charcoal-darkened soils have been misinterpreted as restricted soil horizons when they are actually only the charcoal-bearing parts of thick soil sections. Such layers are thickest at sites that were favourable for Maori occupation, and from their stratigraphic position are assumed to be an indication of human occupation. Darkened soil is thickest on the coast but extends inland for several miles in those parts of New Zealand that were scrub-covered at the time of Pakeha settlement (see McLintock, maps 14, 15) where it probably records the burning
of the virgin forest by the Maori. A similar but thinner and more superficial layer of darkened soil in the inland part of the North Island and in the South Island probably represents the burning of the forest and scrub by Pakeha settlers. Charcoal-darkened layers provide useful reference surfaces for determining rates of soil accumulation and erosion.
Maori Occupation Layers and Inferred Population Trends
The following features, given in order of decreasing abundance and increasing diagnostic value, are used as evidence for human occupation; (1) darkening of soils by finely divided charcoal; (2) visible charcoal fragments; (3) shells of edible molluscs; (4) scattered burnt stones; (5) ovens ( umu) in place, usually associated with charcoal and shells; (6) fragments of flint and obsidian; (7) bones of mammals and birds, usually burnt and broken. The significance of dark soils and soils with charcoal fragments is established in long, well-exposed sections by lateral passage into deposits with clear evidence of human occupation. Other than the rare lumps of Taupo charcoal, no charcoal was found more than a few inches below definite evidence for human occupation, and charcoal by itself is considered significant. Occupation layers usually extend laterally for several hundred feet or more at most beaches. They do not represent habitation sites, but correspond to the charcoal, sea shells, and litter left on beaches by campers and Maoris at the present time. As might be expected, the deposits are thickest at the best sites—namely, sheltered localities, near streams, on what was grass rather than on bare soil or sand, and usually not far in from the edge of the beach. They thicken and become lighter in colour where sand or silt accumulated rapidly, and thin out towards hills where erosion exceeds deposition (Fig. 6). Only clear occupation layers are recorded as such in the sections.
According .to maps presented by McLintock (1959: 7, 73), in 1959 about 50 per cent of Maoris lived within ten miles of the coast, in 1926 about 60 per cent, and in 1780 about 75 per cent. Maoris being essentially a coastal people, the intensity and extent of the coastal occupation layers gives a good indication of the total New Zealand population at different times in the past. That this is true for the upper layer is shown by the relation between its thickness and the estimated population density in 1780. The agreement is close except for the coasts of Taranaki and western Wellington, where the 1780 population was considerably larger than that inferred from the upper occupation layer. The 1780 population map was compiled partly from Cook’s coastal observations and partly from estimates made later by Buck, and may include some populations that grew later than 1780. The population of western Wellington is known to have increased considerably when it was conquered and settled by northern invaders early in the nineteenth century (McLintock, 1959: 2). Estimates for the 1780 population range from 100,000 to 500,000, the commonly accepted figure being 250,000. The population of the lower occupation period can be estimated from its intensity and extent relative to the upper. In Eastland the lower occupation layer is somewhat thicker and better defined than the upper, but in Coromandel and Northland it is uniformly less well-defined in all sections. On balancing these changes and assuming a population of 250,000 for the upper layer, a total population of 150,000 is estimated for the lower occupation layer. The reasons for the changing density of population are uncertain but might be interpreted if the relative importance of agriculture to hunting and fishing during hungry parts of the year were better known. The density of the 1780 population is clearly related to agriculture. Kumara was by far the most important food plant, and the population was largest where the most kumara was grown.
The date of the introduction of kumara is uncertain, but even if it were introduced at an early period, many hundred years would have been needed to develop strains and methods of agriculture suited to the New Zealand climate. It is relevant that Maori strains of kumara are still the best for New Zealand conditions. If this interpretation is correct, and the kumara became more important with time, then the decrease in the Eastland population that is inferred from the relatively barren zone between the lower and upper occupation layers may be due to overhunting and to a possible inland migration prior to the growth of the latter population that was based essentially on kumara. The simultaneous and rapid increase of the earliest population over an area that extended from Hawke’s Bay to North Gape is the most important and most definite inference to be drawn from the coastal occupation layers. Charcoal and other evidence for human occupation, completely absent a few inches below, progressively increases in abundance up to the horizon of the Loisels Pumice at the base of the lower occupation layer. As a reference horizon, the Loisels Pumice could hardly have been more happily placed. The slight difference in depths at which charcoal first appears suggests that a population large enough to leave recognizable traces first built up in the extreme north—the warmest part—but the depth difference is slight, and probably represents no more than a hundred years.
The population for the first few hundred years after New Zealand became inhabited was doubtless small and the date of the first human arrival is unlikely to be established from direct evidence, but it can be inferred from the time required for the population ,to build up to the estimated 150,000 at the time of the Loisels Eruption. Until each favourable site was completely exploited without increasing individual effort population increase was probably logarithmic, the population doubling during an interval that was probably more than 20 and less than 50 years.
If the Loisels Eruption occurred at A.D. 700, and the population was 150,000, the time of arrival of the first hundred or so people would have been between A.D. 200 and A.D. 500 A better estimate will be possible when the Loisels Eruption has been dated by C l 4. Meanwhile the complete absence of occupation charcoal or other signs of human occupation for a considerable interval above the Taupo Pumice makes it likely that New Zealand was uninhabited in A.D. 200, the time of the Taupo Eruption. The estimated changes in the Maori population are shown by Fig. 22.
Adkin in 1952 reviewed the published geological evidence for man being in New Zealand well before the supposed arrival of the Fleet in A.D. 1350. Adkin thought that the southern part of the west coast of the North Island at Horowhenua was first occupied some 2,000 years ago. The time may well be not far in error but the evidence is inconclusive. He assumed that a line of middens now some 4,000 ft inland were built on the actual coast line. He stated that the coast is now prograding about two feet per year, and by assuming a constant rate of progradation he estimated the middens to be some 2,000 years old. He also recorded a bank of sea-borne Taupo Pumice 300 ft inland. At the time of his account the Taupo Pumice had not been dated, but it is now known that the pumice bank, if Taupo, formed 1,800 years ago and that the coast has since prograded with an average rate of only a seventh of a foot a year, 14 times as slow as the rate used by Adkin. The middens cannot be 14 times as old as suggested by Adkin, and they must have been built well back from the coast. Doubtless drifting sand made the coast unfavourable, permanent occupation being within the belt of well-established vegetation a mile or so back from the coast, and not at the coast itself as assumed by Adkin.
Two contrasting views are held about the arrival of the Maori in New Zealand. Tradition (Archie, 1957) tells how Kupe set out from Hawaiiki, discovered the two islands of New Zealand, and returned home in A.D. 950, to be followed by Toi, who made the same journey in A.D. 1150; and how in the midsummer of A.D, 1350 or thereabouts the nine fleet canoes set out with the ancestors of the present Maoris. The dates are based on genealogies. The two early ones are uncertain, but the date of the Fleet is stated to be remarkably consistent among different tribes (Archie, 1957: 62). Tradition does not make it clear whether New Zealand was inhabited prior to the arrival of the Fleet, but prior occupation is now generally accepted.
The opposed view, based essentially on the difficulty of planned Polynesian navigation to New Zealand and back, was first clearly stated by Sharp in 1956, and has been given wider publicity in his Pelican Book. Sharp considered that New Zealand was populated by chance arrivals in canoes that had been driven south by storms. He demonstrated the possibility of long drift voyages and emphasized the impossibility of planned ocean voyages without instruments. The contrast lies entirely with the planned Fleet migration of tradition. Tradition does not conflict with the possibility of pre-Fleet people having drifted here.
If the population at A.D. 700, the inferred date of the Loisels Eruption, was 150,000 or of that order, then the few that tradition tells of arriving in the nine Fleet canoes of A.D. 1350 are unlikely to have dominated or impressed their culture on the many who were well established here. The present Maori population must have descended basically from the first people to arrive here and the tradition of the first canoes to arrive, whether by drift or by planned voyages, could well have persisted. It is suggested that it is these first canoes that are recorded by the Fleet tradition, their arrival being some thousand years earlier than indicated by the genealogies.
Acknowledgments This study originated in work on the beaches of Gisborne with Mr W. Wave and Mr Allen Pullar. Mr Waye was the first to recognize the distinctive nature of the Loisels Pumice and traced it from Poverty Bay to Coromandel. Land Rover transport provided by the British Petroleum Company greatly speeded up the examination of the Gisborne and Bay of Plenty beaches. Miss Gwenyth Challis has resolved several correlation problems by petrological examination of critical samples, and has kindly provided the Appendix to this account. The cost of field work in Northland and western Auckland was partly covered by a research grant from the University of New Zealand. Mr Paul Vella has read the preliminary draft and has made many improvements to it. The maps and diagrams were drawn by Mr E. Hardy, of Victoria University. APPENDIX Mineralogy of Selected North Island Pumices and Ash Showers By Gwenyth A. Challis, Victoria University of Wellington This study was made at the suggestion of Dr H. W. Wellman to assist correlation of the coastal Holocene deposits of New Zealand. Fourteen pumice and twenty ash samples were examined. They are mostly those that proved difficult to correlate in the field. Except for one sample of Taupo Pumice from D’Urville Island, at the north end of the South Island, all are from the North Island. The heavy mineral fraction of one of the first samples examined—Loisels Pumice from Waimarama Beach found to contain appreciable garnet. Contamination was suspected, and it was then found that the sand at this beach contains much garnet. The pumice samples were then cleaned with compressed air to remove contaminating sand. Ash samples cannot be cleaned, and contamination with beach sand is suspected for some samples. The few harder pumices were crushed between glazed paper with a wooden roller; the remainder were broken down by rubbing between the fingers, the intention being to separate the grains with a minimum of grinding. Heavy minerals were separated with bromoform (S.G. 2.85). The refractive index of the glass was determined for all samples by standard methods, using liquids of known refractive index. The probable error of all refractive-index measurements is 0.003 and is about the same as the range of variation within individual ash showers and pumice deposits with refractive index less than 1.515. Individual variation is greater than the probable error when refractive index is greater than 1,515. Critical samples were re-examined with almost identical results. Heavy minerals are given in all descriptions in order of decreasing abundance, and are followed by their approximate weight percentage of the sample. The descriptions of the Taranaki samples are given at the end. The remainder are grouped under individual eruptions given in approximate chronological order. Samples of the Taupo, Kaharoa, and Tarawera ashes from Lake Rerewhakaitu were kindly supplied by Mr C. G. Vucetich, of the N.Z. Soil Bureau. The Taupo Ash sequence was collected from near the type locality at Taupo by Mr J. Lewis. The remaining samples are from coastal sections, and their stratigraphic position is indicated by the numbers “Ml ”, etc., in the text sections in the body of the text. The number of the text section is given in the following list.
Ml, Tarawera Ash, Lake Rerewhakaitu Strong, very hard, dark-grey lapilli. Composition is mostly glass with a refractive index of 1.520. Large feldspar crystals (R.I. > 1.54) and quartz common. Magnetite, augite, and hypersthene total about 1 per cent. M2, Kaharoa Ash, Lake Rerewhakaitu Very light-grey ash ranging from fme-sand-grade to lapilli up to fin. Composition is mostly colourless volcanic glass with a refractive index of 1.495. Small feldspar fragments common, large angular quartz fragments abundant. Magnetite, hypersthene, and rare augite total only 0.2 per cent. M 3, ? Kaharoa Ash, Orongo, Section 9 Very light grey fine-sand-grade ash. Complete absence of crystal rounding suggests a primary ash deposit. Composition is mostly colourless glass with a refractive index of 1.501. Angular quartz fragments common, altered feldspar abundant. Hypersthene, magnetite, and basaltic hornblende total less than 0.2 per cent. M 4, ? Kaharoa Ash, Maraetaha, Section 7 Sample contains much tuffaceous material but cannot be considered a subaerial ash. M 5, ? Kaharoa Ash, Cooks Cove, Section 15 Light-coloured sand-grade ash. Composition is mainly glass with a refractive index of 1.503. Small fragmented crystals of quartz and feldspar make up most of the remainder. Hypersthene, magnetite, basaltic hornblende, and augite total about 0.5 per cent. Hypersthene and hornblende in 2mm-long prisms with some glass adhering. The adhesions and lack of crystal-rounding suggest a primary ash deposit. M 6, Loisels Pumice, Waimarama, Section 2 Dark-grey highly vesicular pumice with small circular vesicles. No visible ferromagnesian crystals. Very strong, cannot be broken with fingers. Composition is predominately glass with a refractive index of 1.512. Small angular crystals of quartz and feldspar fairly common, feldspar sanidine and oligoclase. Augite, hypersthene, magnetite, and zircon total less than 1 per cent. M 7, Loisels Pumice, Waiotahi, Section 21 Very dark-grey pumice with abundant tiny circular vesicles. Very hard and cannot be broken with fingers. Composition almost entirely colourless glass with a refractive index of 1.517. Augite, hypersthene, and magnetite total 1 per cent. MB, Loisels Pumice, Whiritoa, Section 24 Dark-grey pumice. Some lumps streaked with light-grey. Many small circular vesicles. Floats well in water. Hard and not easily broken. Composition is mostly colourless glass with a refractive index of 1.518. Quartz as rather large angular crystals, feldspar as microlites. Hypersthene, magnetite, augite, and basaltic hornblende total less than 1 per cent. M 9, Loisels Pumice, Ohui. Above Type Ohui Ash. Dark and light-grey banded pumice. Very hard, cannot be broken by fingers. Composition mostly colourless glass with a refractive index of 1.512. Quartz and feldspar crystals fairly abundant. Magnetite, hypersthene, greenish-brown hornblende, augite, and zircon total 1 per cent.
MlO, Loisels Pumice, above type Ohui Ash Dark- and medium-grey banded pumice. Hard and highly vesicular, with small round vesicles. Composition mostly colourless glass with a refractive 5 index of 1.512. Highly pleochroic hypersthene, augite with large magnetite inclusions basaltic hornblende, magnetite, and zircon total 1.5 per cent. Zircon particularly plentiful. Mil, Loisels Pumice, Whatuwhiwhi, Section 39 Dark-grey very hard pumice. Some lumps with light- and dark-grey bands. Composition mostly colourless glass with a refractive index of 1.512. Feldspar crystals fairly common, quartz rare. Hypersthene, magnetite, augite, basaltic hornblende, and zircon total 0.5 per cent. Ml2, Loisels Pumice, Merita, Section 40 Dark and medium-grey banded pumice. Strong, cannot be broken with fingers. Composition mostly colourless glass with a refractive index of 1.510. A few small angular quartz and feldspar crystals. Hypersthene, magnetite, basaltic hornblende, and augite total less than 0.2 per cent. Ml 3, Ohui Ash, Whiritoa, Section 24 Light-coloured sand-grade ash. Composition predominately curved shards of very pale glass with numerous vesicles. Refractive index of glass 1.503. Quartz and feldspar abundant. Hypersthene, magnetite, basaltic hornblende, green hornblende, augite, and zircon total about 1.5 per cent. Ml 4, Ohui Ash, Type, Ohui Beach Light-grey fine-sand-grade ash with visible ferromagnesian crystals. Composition 90 per cent colourless glass with a refractive index of 1.502, in rather large curved shards. Quartz plentiful in large angular pieces up to 1 mm. Feldspar rarer and mostly kaolinized. Magnetite, hypersthene, basaltic hornblende, green hornblende, and zircon total about 2 per cent. Green hornblende has Z A c 19°. Ml 5, Ohui Ash, Onewhero, Section 37 Light-grey ash with average grain size of 0.1 mm. Few visible ferromagnesian crystals. Over 95 per cent curved shards of colourless glass with a refractive index of 1.501. Small, very angular pieces of quartz plentiful, feldspar weathered and common. Hypersthene, augite, zircon, and green hornblende total less than 0.1 per cent. Lack of crystal-rounding indicates a primary ash deposit. Ml 6, Taupo Lapilli (Mostly Pumice), Lake Rerewhakaitu Very poorly sorted pumice from fine-sand-grade to lumps up to several inches through. Light-grey with large elongated vesicles. Easily broken with fingers. Crystals of quartz, no visible ferromagnesian crystals. Composition mostly colourless glass with a refractive index of 1.502, constant for different parts of sample. Magnetite, hypersthene, and augite total less than 0.1 per cent. Ml 7, Taupo Lapilli, Maori Earthworks, Maraetaha River, Section 8 Light-grey lumps up to lin. Mostly colourless glass with a refractive index of 1.497. Large angular quartz crystals particularly abundant, feldspar rare. Magnetite and hypersthene total about 1 per cent. Hypersthene in large prisms up to 2mm long, highly pleochroic in green and golden brown. MlB, Taupo Pumice, Tuamotu Island, Section 10 Light-grey highly vesicular pumice with large elongated vesicles. No visible crystals. Easily broken in fingers. Composition predominately glass with a re-
fractive index of 1.502. Large kaolinized feldspar crystals corarrion, large angular quartz crystals rarer. Magnetite, hypers.thene, and basaltic hornblende total less than 0.5 per cent. Hypersthene vividly pleochroic in green and deep golden brown with numerous large inclusions of magnetite. Basaltic hornblende scarce, pleochroic in dark and light brown with Z A c near o°. Ml 9, Taupo Pumice, D’Urville Island, Greville Harbour Light-grey, highly vesicular pumice with large, elongated vesicles. Lumps up to 3 inches. Can be crushed with fingers. Composition is mostly colourless glass with a refractive index of 1.500. Quartz and feldspar crystals fairly common. Feldspar mainly oligoclase. Magnetite and hypersthene total less than 0.1 per cent. M2O, Leigh Pumice {Type), Ti Point, near Leigh, Section 31 Medium-grey pumice with weathered rusty surfaces. Composition mostly colourless glass with a refractive index of 1.516. Quartz and feldspar crystals fairly plentiful. A little of the less altered feldspar has a refractive index greater than Canada balsam and from extinction angle appears to be andesine. Magnetite, hypersthene and augite total 1.5 per cent. M2l, Leigh Pumice, Ngunguru, Section 34 Light-grey, slightly banded pumice with visible quartz and ferromagnesian crystals. Easily broken with fingers. Composition mostly colourless glass with a refractive index of 1.517. Small crystals of quartz and feldspar moderately abundant. Magnetite, hypersthene, augite and zircon total about 1 per cent. Hypersthene with numerous inclusions of magnetite and basaltic hornblende. M 22, Leigh Pumice, Whatuwhiwhi, Section 39 Grey pumice weathered brown on surface. Few visible crystals Occasional bands of darker grey. Easily broken with fingers. Composition mostly colourless and pale-brown glass. Refractive index of colourless glass 1.521, of brown glass 1.548. Few feldspar microlites, no quartz. Magnetite, hypersthene, augite, basaltic hornblende, and zircon total 0.2 per cent. M 23, ? Waimihia Ash , Tuamotu Island, Section 10 Light-grey very-fine-sand-grade ash. No ferromagnesian crystals visible. Average grain size 0.1 to 0.2 mm. Composition is over 95 per cent colourless glass with a refractive index of 1.498. A little altered feldspar, some quartz. Magnetite, hypersthene, and augite total less than 0.1 per cent. Lack of crystal-rounding and adhesions indicate a primary ash. M 24, ? Waimihia Ash, Tuamotu Island, Section 10 Light-grey, highly vesicular lapilli with long elongated vesicles. Easily crushed in fingers. Composition is predominately glass with a refractive index of 1.498. Large crystals of altered feldspar (mainly sanidine) and angular fragments of quartz fairly plentiful. Magnetite, hypers.thene, and basaltic hornblende total about 1 per cent. Hypersthene in slender prisms up to 1.5 mm long. M 25, Whatuwhiwhi Pumice {Type) Whatuwhiwhi, Section 39 Medium-grey pumice, weathered rusty on surface. Visible ferromagnesian crystals. Breaks easily in fingers. Composition is mostly colourless glass with a refractive index of 1.521. Feldspar crystals and microlites plentiful, no quartz. Augite, magnetite, hypersthene, and zircon total nearly 2 per cent.
M 26, ? Burrell Ash, Urenui Stream Beach, Section 51 Light-brown medium-siand-grade ash with abundant dark crystals mainly magnetite. Composition is mainly colourless glass with a refractive index of 1.518. Feldspar (andesine) common. Magnetite (dominant), augite, basaltic hornblende, and very rare hypersthene total 35 per cent. Augite has numerous large magnetite inclusions. The unusually high heavy-mineral content may be due to contamination by dark beach sand in which the ash is interbedded. M 2 7, ? Newall Ash, Urenui Stream Beach , Section 51 Similar to Burrell Ash but with fewer dark minerals. Composition mostly palebrown glass with a refractive index of 1.523. Rather calcic andesine crystals common, quartz very rare. , Magnetite (dominant), augite, basaltic hornblende, and very rare hypersthene total 18 per cent, but may have been increased by contamination with beach sand. M2B, Stent Ash, Onaero Beach, Section 52 Coarse-sand-grade and silt grade ash in separate layers. Both layers light in colour but with visible ferromagnesian crystals. Grain size determined solely by size of glass fragments and feldspar grains. Composition mostly colourless glass with a refractive index of 1.508 and less abundant brown glass with a refractive index of 1.527. Some glass shards curved, some long and fibrous. Zoned feldspar (andesine) common, quartz very rare, augite and basaltic hornblende in stout prisms up to 2 mm long. Augite, 2V 55°-60° (+) and Z A c 43°, is faintly pleochroic, with X light greenish-yellow and Y or Z green. Magnetite (dominant), augite, and basaltic hornblende total 9 per cent, but may have been increased by contamination with beach sand. M 29, Onaero Pumice {Type), Onaero Beach, Section 52 Medium-grey pumice with abundant large crystals of augite and hornblende. Abundant dark crystals up to 3 mm long distinguish this pumice. Composition largely colourless volcanic glass with a refractive index of 1.513. Zoned feldspar of intermediate composition uncommon, quartz rare. Augite, basaltic hornblende, and magnetite are unusually abundant for a pumice and total 6 per cent. Augite distinctly green but non-pleochroic. Hornblende vividly pleochroic in dark and light brown, some crystals pleochroic in deep reddish-brown to straw-yellow. Z A c is near 0° in all crystals. Remarks The significant comparative features— refractive index, percentage, and order of abundance of heavy minerals —are set out in Table I (Order of Abundance: O absent; 1, 2, 3, etc., increasing abundance). Taranaki samples can be easily distinguished from the others, having over 6 per cent heavy minerals and the others 1 per cent or less. Also, in the four Taranaki samples augite is more common than hypersthene, whereas the reverse is true for all except three of the other 23 samples. The importance of augite in Taranaki ashes was first noted by Taylor (1933). Taranaki samples are given in chronological order at the end of the list. The others are grouped under individual eruptions and arranged in chronological order, the grouping being based on stratigraphy, distribution, and mineralogy. Colour and appearance are useful diagnostic features that were used during preliminary field correlation. Except for the Taranaki samples, which are probably dark because of their extremely high heavy-mineral content, no direct relation exists between colour and either heavy-mineral content or refractive index, Leigh Pumice, with a consistently higher refractive index, being only
slightly darker than Taupo Pumice and considerably lighter in colour than Loisels Pumice, with an intermediate refractive index. Mineralogical features are thus largely unrelated to appearance and provide extra information critical for distinguishing eruptive material that is similar but of different ages. Refractive indices are controlled by the chemical composition of the glasses, andesitic samples having a high and rhyolitic samples a low refractive index. Kaharoa, Ohui, Taupo, and Waimihia are rhyolitic, with a refractive index = or < 1.503. Tarawera, Loisels, Leigh and all the Taranaki deposits are andesitic, with a refractive index = or > than 1.510. The clear separation makes division into two groups easy, but the reason for the gap between 1.503 and 1.510 is unknown. Refractive index is most reliable for the rhyolitic samples, in which all the glass has the same refractive index. The average refractive index of andesitic samples that contain brown as well as colourless glass cannot be reliably determined without melting them; the values given in the table are for the colourless glass and are lower than the average values. In general the percentage of heavy minerals increases with increasing refractive index and is about ten times as great in ashes as in pumices of the same refractive index. The ratio of augite to hypersthene tends to increase with increasing refractive index, being closest to unity in samples with a refractive index of about 1.516. One sample of Tarawera Ash was examined. Its refractive index is one of the highest and well within the andesite range, but the sample is from near the vent and may not be representative of the shower margin. Kaharoa Ash from the same section, and probably from Mt Tarawera also, has by contrast the lowest refractive indices of any of the samples. The ashes from the coastal sections at Orongo and Cooks Cove, which have been tentatively correlated with Kaharoa ash, are similar to each other. They have a slightly higher refractive index than Kaharoa and a not dissimilar heavy mineral content. The two ashes overlie Loisels Pumice and must correlate with Tarawera, Kaharoa, or an unnamed ash on refractive index. Correlation with Kaharoa is much more likely than with Tarawera. Loisels Pumice samples were taken from localities at the extreme north and south and from anomalous-looking lumps, but all seven samples are consistent. Refractive index ranges from 1.510 to 1.519, with an average of 1.513. On the whole hypersthene, magnetite, and augite are about equally abundant. Basaltic hornblende occurs in five samples and zircon in four. Magnetite is less common relative to total heavy minerals than in almost all other samples. Ohui Ash is represented by two samples from Coromandel and by one from the Bay of Islands. The three samples are remarkably similar and distinct from other samples. As they are all at about the same stratigraphic horizon they are considered to be part of the same shower. Refractive index ranges from 1.501 to 1.503. Hypersthene and magnetite are generally most abundant. Augite, zircon, and basaltic hornblende are generally present but less abundant. Green hornblende was seen in these three samples only. The ash has a similar refractive index to Taupo Ash and Pumice but can be distinguished by the green hornblende and zircon and by the considerably larger content of heavy minerals. No samples are available of east coast ash, thought to be the same age on stratigraphy. Taupo Pumice is uniform in appearance at _ all localities and has an equally uniform mineralogical composition. The refractive index has the range of 1.497 to 1.505, with an average of 1.502. Heavy minerals are not abundant, the most common being hypersthene and magnetite. The five members of the Taupo Ash sequence (Baumgart, 1954) that underlie the Taupo Lapilli Pumice were found to have refractive index and mineral content similar to that of the Taupo Pumice.
Leigh Pumice is represented by three samples, which have similar composition. Refractive index ranges from 1.516. to 1.521. Heavy minerals, magnetite, hypersthene, augite, basaltic hornblende, and zircon in decreasing order of abundance total about 1 per cent. The pumice has about the same stratigraphic position as Taupo Pumice, and was confused with it in the field, but is slightly darker in colour, has a considerably higher refractive index, generally contains zircon as a minor constituent, and must be considered distinct. The presence of zircon suggests magmatic relationship with the Loisels Pumice and Ohui Ash, and the Bay of Plenty as a possible source. Waimihia Ash is not represented by any definite samples, but two samples from Tuamotu Island are considered from their stratigraphic position to be of this shower. Refractive index is slightly, but probably not significantly, lower than that of Taupo Pumice, but the heavy minerals have almost the same order of abundance and there is no mineralogical reason for or against correlation. Whatuwhiwhi Pumice is represented by one sample from the type locality. Its refractive index is not significantly higher than that of overlying Leigh Pumice, but augite is more abundant than hypersthene. Taranaki samples include probable correlatives of the Burrell and Newall ashes, a definite correlative of the Stent Ash, and Onaero Pumice from the type locality. The ashes have a similar heavy-mineral content; magnetite, augite, basaltic hornblende, and hypersthene in decreasing abundance; but differ appreciably in refractive indices. Magnetite is dominant, being about 10 per cent by weight of the ashes, and these Holocene ashes have probably made a significant contribution to the black sand on the west coast beaches of the North Island. The titanium content of the magnetite was not determined.
Onaero Pumice is distinctive in appearance and composition, A speckling of dark minerals makes up 6 per cent by weight, four times the heavy mineral content of any of the other pumices. References Adkin, G. L., 1952. Geological Evidence for the Antiquity of Man in the New Zealand Area, N.Z. Sci. Rev. 4: 41-45. Archey, G., 1957. Maori and Polynesian: An Outline Sketch of History and Research. Science in New Zealand. A. W. & A. H. Reed, Wellington. Baumgart, I. L., 1954. Some Ash Showers of the Central North Island. N.Z. J. Sci. Tech. 358 (6): 456-67. Best, Elsdon, 1903. Food Products of Tuhoeland. Trans. N.Z. Inst. 35: 45-111. Brooks, G. E. P., 1950. Climate through the Ages. London; Benn. Brothers, R. N., 1957. The Volcanic Domes at Mayor Island, N.Z. Trans, roy. Soc. N.Z. 84 (3): 549-560. Fleming, C. A., 1953. Material for a Recent Geochronology of N.Z. Roy. Soc. N.Z., Rep. 7th Sci. Cong.: 114-23. N.Z. J. Sci. Tech. 388 (7); 726-731. Godwin, H., Suggate, R. P., Willis, E. H., 1958. Radiocarbon Dating of the Eustatic Rise in Ocean Level. Nature 181: 1518-19. Golson, J., 1957. The Contributions of Natural Science to Archaeological Research. N.Z. Sci. Rev. 15 (7-8); 56-60. Gutenburg, 8., Richter, C. F., 1949. Seismicity of the Earth. Princeton: University Press. Hamilton, A., 1888. Notes on a Deposit of Moa Bones in Te Aute Swamp, Hawke’s Bay. Trans. N.Z. Inst. 21: 311-318. Henderson, J., 1944. Earthquake Risk in New Zealand. N.Z. J. Sci. Tech. 5B: 195-219.
Laing, A. C. M., 1954. Notes on Tsunamis Reaching the New Zealand Coast. NZ. J Sci. Tech., 358: 470-472. Marshall, P., 1933. Effects of [1931] Earthquake on Coastline near Napier. N.Z. J. Sci. Tech. 15: 79-92. MgLintock, A. H. (ed.), 1959. A Descriptive Atlas of New Zealand. Wellington: Govt. Printer. Oliver, W. R. 8., 1931. An Ancient Maori Oven on Mt Egmont. J. Polynes. Soc. 40: 73-80. Richards, A. F., 1957. Identification of Isla San Benedicto Pumice on the Islands of the Central Pacific Ocean. Geol. Soc. America Bull. 68: 1785. Schofield, J. C., 1960. Sea Level Fluctuations During the Last 4,000 years as Recorded by a Chenier Plain, Firth of Thames, N.Z. N.Z. J. Geol. Geophys. 3: 467-85. Sharp, A., 1956. Ancient Voyagers of the Pacific, Polynes. Soc. Memoir. 32. Sinclair, K., 1959. A History of New Zealand. Pelican Book A 344. Taylor, N. H., et al. 1954. Soils of the North Island of N.Z. (with map of ash showers). N.Z. Soil Bureau Bull, (n.s.) 5. Wheeler, M., 1956. Archaeology from the Earth. Pelican Book A 356. Wellman, H. W., 1960. The Stratigraphy of Human Occupation Layers in North Island Coastal Sections, N.Z. Asian Perspectives 3 (1); 91-94. Dr H. W. Welllman, Geology Department, Victoria University of Wellington, P.O. Box 196, Wellington.
AshName Pumice Place Origin Date or Estimate Nature 2 ab Cubic Miles H/A Index No. Mean Range Ash Mt 1886 And. ? 50 0.5 0.5 1 1.520 — 1 And. ? 50 0.5 0.5 1 1.520 —. AshBurrell Mt Egmont 1650 And. 90 0.2 0.02 (1) 1.518 — Ash Mt Egmont 1350 c. And. 200 0.5 0.02 (1) 1.523 — Ash Mt 1300 c. Rhy. 150 3 2 1 1.502 — Pumice of Bay c. And. — ? 1 7 1.514 8 AshOhui Plenty c. And. ? 1000 ? 3 4 3 1.502 2 And. ? 1000 ? 3 4 3 1.502 2 Lap. Taupo nr. 150 Rhy. 5000 14 8 4 1.500 5 pre-Lap. Taupo nr. 150 Rhy. 700 2 8 5 1.500 5 Pumice of ? 1 And. — p 2 3 1.518 5 Stratford Ash Mt ? And. 200 0.2 — f — — — Ash Mt Egmont ? And. 400 0.5 0.03 1 1.508 — Ash Taupo nr. 2000 ? Rhy. ? 6000 5 4 (2) 1.498 0 Rhy. ? 6000 5 4 (2) 1.498 0 Pumice Mt Egmont pre-3000 And. ? 0.03 1 1.513 — — p 0.03 1 1.513 — Whatuwhiwhi Pum. ? Plenty pre-3000 And. — p 0.06 1 1.521 — (1), Lap. distance by rhyolitic. And. refractive one H/A halved which ; inferred; = ratio ab 2 = product of central of hypersthene to augite; index sample, correlation thickness is halved; H/A inferred; = ratio ab 2 = product of central of hypersthene to augite;
Table I.—Summary of Data on Composition and Volume of Holocene Ash and Pumice Deposits.
of No. Sample Name of Ash, Locality and Section No. IndexRefractive colourless of heavy weight by Hypersthene pi Q. Magnetite hornblende Zircon CT Green DP Augite g re 1 Tarawera Ash, Lake Rerewhakaitu 1.520 1 1 3 0 0 0 2 o Kaharoa Ash, Lake Rerewhakaitu 4. 1.495 Kaharoa Ash, Lake Rerewhakaitu 0.2 1.495 2 0.2 3 2 0 3 0 0 0 0 1 0 1 3 ? Kaharoa Ash, Orongo (9) 1.501 < 0.2 3 2 1 0 0 0 4 ? Kaharoa Ash, Maraetaha (17) tuffaceous only 5 ? Kaharoa Ash, Maraetaha (17) tuffaceous only ? Kaharoa Ash, Cooks Cove (15) 1.503 0.5 4 3 2 0 0 1 6 Loisels Pumice, Waimarama (2) 1.512 < 1 3 2 0 1 0 4 7 Loisels Pumice, Waiotahi (21) 1.517 1 2 1 0 0 0 3 8 Loisels Pumice, Whiritoa (24) 1.518 < 1 4 3 1 0 0 2 ' 9 Loisels Pumice, Ohui (dark and light grey) 1.512 < 1 4 5 3 1 0 2 10 Loisels Pumice, Ohui (dark and medium grey) 1.512 1.5 5 2 3 1 0 4 11 Loisels Pumice, Whatuwhiwhi (39) 1.512 0.5 5 4 2 1 0 3 12 Loisels Pumice, Merita (40) 1.510 0.2 4 3 2 0 0 1 13 Ohui Ash, Whiritoa (24) 1.503 1 6 5 4 1 3 2 14 Ohui Ash (Type), Ohui Beach 1.502 2 5 4 3 1 2 0 15 Ohui Ash, Onewhero (37) 1.501 <0.1 4 0 0 2 1 3 16 Taupo Lapilli Ash, Lake Rerewhakaitu 1.502 <0.1 2 3 0 0 0 1 17 Taupo Lapilli Ash, Maraetaha River (8) 1.497 1 1 2 0 0 0 0 , 18 Taupo Pumice, Tuamotu Island (10) 1.502 < 0.5 3 2 1 0 0 0 19 Taupo Pumice, D’Urville Island 1.500 < 0.1 1 2 0 0 0 0 20 Leigh Pumice (Type), Ti Point, near Leigh 1.516 1.5 2 3 0 0 0 1 21 Leigh Pumice, Ngunguru (35) 1.517 1 3 4 0 1 0 2 22 Leigh Pumice, Whatuwhiwhi (39) 1.521 0.2 4 5 2 1 0 3 23 ? Waimihia Ash, Tuamotu Island (10) 1.498 <0.1 2 3 0 0 0 1 24 ? Waimihia Ash, Tuamotu Island (10) 1.498 1.0 2 3 1 0 0 0 25 Whatuwhiwhi Pumice (Type) (39) 1.521 c. 2 2 3 0 1 0 4 26 ? Burrell Ash, Urenui Stream Beach (51) 1.518 35 1 4 2 0 0 3 27 ? Newall Ash, Urenui Stream Beach (51) 1.523 18 1 4 2 0 0 3 28 Stent Ash, Onaero Beach (52) 1.508 9 0 3 1 0 0 2 29 Onaero Pumice (Type), Onaero Beach (52) 1.513 6 0 1 % 0 Q §
Table ll.—Comparative Mineralogical Features of New Zealand Holocene Ash and Pumice Deposits.
Permanent link to this item
https://paperspast.natlib.govt.nz/periodicals/TRSGEO19620727.2.2
Bibliographic details
Transactions of the Royal Society of New Zealand : Geology, Volume 1, Issue 5, 27 July 1962, Page 29
Word Count
32,283Holocene of the North Island of New Zealand: a Coastal Reconnaissance Transactions of the Royal Society of New Zealand : Geology, Volume 1, Issue 5, 27 July 1962, Page 29
Using This Item
In-Copyright Materials
In-copyright materials are made available under a Creative Commons Attribution 4.0 International licence. This means that you may copy, adapt and republish this material, as long as you attribute both the author and the Royal Society of New Zealand.
In-copyright taxonomic materials are made available under a Creative Commons Attribution No-Derivatives 4.0 International licence. This means that you may copy and republish this material, as long as you attribute both the author and the Royal Society of New Zealand.
For advice on reproduction of out-of-copyright material from this periodical, please refer to the Copyright guide.
Log in