Volume 69 1960 > Volume 69, No. 3 > The cultural succession and ethnographic features of D'Urville Island, by I. W. Keyes, p 239-265
                                                                                             Previous | Next   

- 239
THE CULTURAL SUCCESSION AND ETHNOGRAPHIC FEATURES OF D'URVILLE ISLAND

In this comprehensive survey of D'Urville Island during pre-European and early European times, Mr. Keyes, a geologist with Geological Survey, Lower Hutt, adopts the scheme of New Zealand prehistory proposed by Mr. G. L. Adkin elsewhere in this issue of the Journal.

INTRODUCTION

THE MARLBOROUGH SOUNDS, of which D'Urville Island forms a part, provide one of the most spectacularly distinct and intricately beautiful areas of coastline in New Zealand. In spite of the apparent steepness of the terrain, the 600 miles of coastline with the innumerable bays, secluded coves and extensive inland waterways, have revealed considerable traces of a once numerous prehistoric population. This area has been occupied extensively by successive peoples who in turn have all made their mark on the landscape, but of these it was the race who scarped the steep spurs with small flights of terraces, “giving a profile like some gigantic staircase” 1 and covered the hilltops with small rectangular pits, who have aroused the most speculation and interest in the past. There are numerous traditions relating to a large variety of peoples who supposedly occupied the territory in past times, but some of them, due to complete lack of tangible evidence, can be dismissed as doubtful. However, as revealed particularly well on D'Urville Island, four distinct superimposed phases of settlement or four main cultural influences have been successively dominant over the area. These cultural divisions, based on sound evidence, fit in with the recognized successive cultural occupation pattern of the New Zealand area. 2

Historically, too, D'Urville Island has claims to be an interesting area, for it has been connected with both the early European discovery and settlement of the country. The first European to visit the area was Tasman in 1642, who seeking shelter from adverse weather, anchored between the Rangitoto Islands and Pelorus Sound entrance off the east coast of D'Urville Island, from the 21st to 26th December. Cook, during each of his three voyages to New Zealand, selected in this region the sheltered waters of Pelorus Sound for his anchorage. On one occasion 3 he landed on the eastern side of D'Urville Island, “in the second cove down from Stephens Island”, for water, and while here, he noted the excellent fishing to be had in these waters. Climbing Remarkable Cone, - 240 and looking down into Catherine Cove, he observed signs of a deserted native village on the shore. Stephens Island and Cape Stephens retain in their names the memory of his visit to the Island, and were named by Cook in honour of one of the then British Admiralty Board Secretaries. It was D'Urville, the French explorer, after whom the island (Rangitoto) was named, who in April of 1827 really observed the eastern coast in detail, being particularly interested in the vegetation, and delighting in the collection of the many botanical species that were completely unknown to him. While on this coast, he too, like Cook previously, observed on the shores of Catherine Cove the presence of a native village. Today many of the topographic features along the coast still bear the French names he originally assigned to them, and remain a vivid testimony to his heroic endeavours against the turbulence of the Pass, which he finally overcame after a three day struggle. On the 18th September, 1834, on the shores of Port Hardy, the first contingent of British soldiers ever to set foot on New Zealand soil were landed. 4 Piloted by the whaler John Guard who established the first whaling station in the Marlborough Sounds, H.M.S. Alligator in company with the schooner Isabella from New South Wales, while heading for the Taranaki coast to rescue Guard's wife and children 5 as well as the crew of his wrecked vessel Harriett who were in the hands of the local Maoris, was blown southward off course, and sought refuge in Port Hardy. The soldiers on board (about 60 men of the 50th or Queen's Regiment) were taken ashore, and the opportunity used to occupy them with target practice. They remained here until the 20th. D'Urville Island, too, had in the whaler James McKenzie McLaren the first permanent white settler in the Nelson district. 6 With his Maori wife, he lived in the village of the chief Te Whetu at Oterawa on the east side of the Island, up to 1846. Before finally departing from this locality, he built with local timbers a 25-ton coaster, the Ocean Queen. Colonel Wakefield made use of his experience and local knowledge and engaged him on the 14th of January 1840 to watch for, and act as pilot to guide, the New Zealand Company's immigrant ships, which had failed to rendezvous at Port Hardy at the appointed time, 7 through the uncertain waters of Cook Strait, to the then newly established settlement at Port Nicholson.

In the place-names allocated to the various geographic features, there is preserved a wealth of historical associations. Kupe the early navigator, during his voyage down the east coast of the North Island, before continuing on to the west coast of the South Island, sailed close to the northern end of D'Urville Island, and today the rocks off Cape - 241 Stephens that he named bear record of his visit—Nga Tamahine-a-Kupe (The daughters of Kupe). 8 Traditionally, too, the naming of French Pass is attributed to the shag Te Kawau-a-toru Kupe took with him on the voyage, whose mission it was to search for strong currents. 9 Moving south Te Kawau-a-toru encountered French Pass (plate I), and while endeavouring to pit his strength against the surging currents, was overcome, breaking his wing in the struggle. The broken wing became the navigable gap that exists on the eastern side, and his body formed the reef across the Pass. In commemoration of his gallant effort, the Pass was named in his memory Au-miro-o-te-kawau-a-Toru (the swishing current of the shag Toru). 10 This name was later abbreviated and the Pass became known simply as Te Aumiti. Traditionally, the naming of D'Urville Island as Rangitoto is attributed to Nuku-tama-rore, 11 who, journeying in the vicinity of French Pass, noticed smoke rising from the eastern side and immediately recognized a similarity of this island to Rangitoto at Ahu (apparently an active volcano) in the Hawaiian group, and thus named it accordingly. 12 These three place names are amongst the oldest in the district. The other names around the island owe their origin to both pre-Maori, Maori, and very early European occupation. Of the pre-Maori and Maori place-names, correct interpretation is particularly difficult, as many have become shortened and the original context in which they were employed is no longer known. Although it is usually unwise to attempt to give the meanings of Maori place-names when the history is not known, likely explanations for several, based on a good local knowledge of the site, have been offered.

Of the European names, those given by Cook have been mentioned previously and the names on the east coast allotted by D'Urville can be easily recognized. The naming of the two famous harbours during the period of early European systematic exploration is of considerable interest, for the association of names given to various landmarks, particularly in Port Hardy, has often been commented on. The naming of Port Hardy took place during the visit of H.M.S. Alligator in September 1834 (with John Guard as mentioned previously) when Captain Lambert named the area in honour of the Vice-Admiral Sir Thomas Hardy, at that time First Sea Lord at the Admiralty, but more famous as Lord Nelson's Flag-Captain at the battle of Trafalgar. 13 The following month (2nd to 5th October), the Alligator returned to the area, and under the command of Lieut. Woore a survey of the harbour was carried out for the Admiralty. 14 Following the tradition - 242 established by Captain Lambert, Lieut. Woore continued to name various features in honour of Lord Nelson and his campaigns (refer map). Today Mt. Woore, overlooking Port Hardy, preserves the memory of the first surveyor.

Greville Harbour or, as it has been known, Brooke Harbour, 15 is the large southern harbour on D'Urville Island, well known for its famous boulder bank. This remarkable feature, composed mainly of local rock from the cliffs on the north side of the harbour, has formed in an area where the incoming seas from the north-west have been deflected southwards by the northern coastline, and the bank represents the area where the force of the sea is counteracted by the outflowing waters from the interior causing a pronounced “rip” which results in the tidal load being deposited. The original survey of the harbour was carried out by Captain Stokes of H.M. Surveying Ship Acheron in 1849, and it is to him that the name Greville Harbour is due. It has been thought likely 16 that it was named after the Duke of Wellington's aide-de-camp at the battle of Waterloo, Algernon Frederick Greville, who was well known for his descent from a most notable scholastic family and from Lord Brooke, several generations previously. Evidently the choosing of this name for the harbour was felt to be in keeping with the tradition established for Port Hardy and the associated landmarks.

Today to many people D'Urville Island still remains a terra incognita. It is known mainly by its good fishing, its two extensive inland harbours which have offered shelter to many yachtsmen in bad weather, and in connection with that narrow turbulent stretch of water that provides the gate-way to Nelson from the north for steamer traffic (plate I). Others remember it, too, for the prolific supply of stone artefacts that were collected from its beaches in the early part of this century, and later during the depression years of the 1930's by a few who endeavoured to make a little extra money in those difficult years by sale of items to private collectors. Systematic digging on many of the beaches 17 yielded much material, 18 and today many large long-established private collections contain numerous articles from the D'Urville Island beaches. The best-known of all the sites for yielding artefacts was Moawhitu beach 19 (locally known as Long Beach) at Greville Harbour, and today the Otago Museum houses many items that have been found there. As well as artefacts from settlement areas, many adzes from workshop sites in both rough, semi-finished and broken states comprise a considerable bulk of the material collected. This area, as has been discussed previously, 20 provided the ancient neolithic New Zealander with an extremely workable rock in large quantities and of unsurpassed excellence. This metamorphosed argillite, associated with the intrusive rocks on the eastern side of the island, was quarried from - 243 boulders and outcrops and worked into suitable shapes in numerous sheltered bays around the coast. Flakes and chips today, however, provide the only evidence left of workshop sites, where in the distant past adzes were roughed out by the hundred before being taken away for final finishing. Artefacts of this material have been found widely distributed throughout both islands of New Zealand, and it is amongst those items representative of New Zealand's earliest culture that this stone is predominant.

PRE-HUMAN HISTORY

The Sounds area portrays, with its main north-south trending ridges and strong relief, the remnants of mature dissected river-valley systems, 21 which, due to local marginal crustal downwarping, have produced a complex pattern of drowned valleys or “rias”. The submergence of this coastal block along with post-glacial redrowning has allowed the sea to inundate the original valleys to an estimated depth of 300 feet. 22 It has been suggested that in past geological time a parallel system of faults governed the formation of the long north-south valleys, 23 but although rejected by a subsequent author, 24 it should not be disregarded as a factor in their origin.

Although today D'Urville Island is separated by a narrow and deep stretch of water (plate I), it nevertheless retains a remnant of its past connections to the mainland at French Pass. The reef which crosses between the two land masses (only 120 yards wide) is the last surviving trace of a land connection to this now insular block. It is only in very recent geological times that D'Urville Island has become separated from the rest of the Marlborough land-block. In the past history of this area, during early Pleistocene times, it is evident that the Marlborough block was connected to the north-west Wellington region by a land bridge, 25 which permitted a free exchange of floral and faunal elements between the North and South Islands. 26 This land connection, considered to have lasted until about 11,000 years ago, gradually disappeared owing to the tectonic submergence of the Marlborough block, until finally it disappeared with the post-glacial rise in sea level and the resultant intensified wave-action on the considerably reduced remaining isthmus. French Pass until this time provided a barrier to the sea by a steep ridge running obliquely from the main “backbone” of D'Urville Island, through Reef Point, and across to the mainland slopes of Mt. Pleasant. Hydrographic charts today indicate a rapid steepening on the under-sea contour on both sides of the reef. With the gradual submergence of the isthmian link across Cook Strait, this ridge finally remained the only land bridge across to D'Urville Island which faunal movements could follow. Other areas of the Sounds like Stephens Island, and the Brothers Islands, were by now completely segregated, and by their - 244 isolation preserved their prehistoric fauna, like tuatara (Sphenodon punctatus) and native frog (Liopelma hamiltoni). By sudden or by gradual movement the connecting ridge, due to local downfaulting along faults present on both sides of the Pass, became depressed in relation to the surrounding land-surface and settled as a mobile block. This remnant ridge, considerably reduced in height above sea level, still remained a connecting link, but gradually, with the increasing rise in the post-glacial sea level, 27 wave attack on this low ridge both from the north and south slowly eroded the remaining surface until finally, close to the mainland, the sea breached this barrier and poured through the gap. This widened and others formed, and finally, as can be seen today, a submerged reef traverses the gap. The final disappearance of this “bridge” naturally prevented the movement of animals, and thus on D'Urville Island representatives of many native creatures were retained. Proof that the French Pass route was the last connection to the mainland can be validated by the presence of the sub-species Paryphanta hochstetteri obscura (native land snail) which today only occurs on D'Urville Island and in the western sector of the Marlborough Sounds. Sufficient time, however, has elapsed since the island became completely isolated for the sub-species of Wainuia urnula nasuta to evolve from the essentially North Island species Wainuia urnula, and become a totally restricted community. It is interesting to note also that the native frog, Liopelma hamiltoni, only to be found on Stephens Island, is also a different species from that found elsewhere in New Zealand, Liopelma hochstetteri, and has developed as such due to its completely geographically isolated habitat, with its specialized environment.

With the arrival of the Polynesian explorers to this coast, moas, tuatara and other native creatures were rapidly exterminated. Today, the only living faunal evidence of a past connection to the mainland is to be seen in the occasional native land snail living in the bushed central portion of the island, and a few surviving cedar trees which escaped the bushman's axe during early European settlement.

GEOLOGICAL IMPORTANCE

The present land surface of the Sounds, denuded of its cover of Tertiary rocks, exposes in its deeply dissected surface an undermass of hard resistant rocks, which withstand fairly well the natural weathering processes. Although a unified geomorphic province, geologically the rocks of D'Urville and Stephens Island are distinct from those comprising the northern Marlborough district. The rocks of D'Urville Island, which have been mapped as part Te Anau Series and predominantly Maitai Series, 28 form the northern extremity of the famous “mineral belt” series of Permian-age rocks, stretching from Stephens Island in the north to Top House on the Wairau river in the south, some 60 miles in total extent. Because some of these rock units are particularly well exposed on D'Urville Island, their type localities have been established - 245 here, and the formation names which they bear 29 have been taken from the local names and are applied to all the Permian stratigraphic units in New Zealand. The rocks on the island can be broadly classified into three distinct types: sandstones and mudstones, volcanic rocks, and the mineral belt intrusives with the associated serpentinous masses. This last belt of rocks which is to be found along the eastern side of the island, easily distinguishable by the paucity of vegetation 30 due to the thin magnesia-rich soil, is of particular importance to New Zealand ethnologists, for it is within this formation that the notable metamorphosed argillite (or “baked argillite”) occurs. 31 The outcrops of metamorphosed argillites 32 in this zone are Te Anau series sediments preserved in small areas on the extreme coastal margin, which have been affected by the presence of heated intrusive bodies along their contact areas. 33 The presence of mineral ores (particularly copper) associated with this belt was recognized at an early date, 34 and on D'Urville Island drives were sunk at a locality on the south-east side behind Copper Mine Bay, in a zone where the oxidised ore outcropped on the surface. In 1896-7 Dr. Schauinsland of Vienna collected samples of serpentine along the east coast, which were studied by the notable German geologist Dieseldorff and formed part of the material used in his doctoral dissertation. 35 The great economic potential of the serpentine along this coast is well known, but the difficulties of sea transport involved make it an unattractive proposition at present.

In spite of the geologically stable substratum, the steepness of the area as a whole, combined with the early European destruction of bush, the introduction of stock, and more recently the prolonged spells of dryness, have brought considerable changes to the original landscape. Ancient earthwork sites are thus affected and are rapidly tending to lose their original well-defined features and blend into the stock-scarped hill-sides. Undoubtedly, because of this, numerous sites of former habitation will not be able to be recognized with absolute certainty, and many, perhaps large important sites, will be indicated only by the smallest seemingly insignificant surface trace.

THE OCCUPATIONAL PATTERN FOR THE SOUNDS AREA

The Marlborough Sounds area has been transiently occupied in the past by representatives of all New Zealand's early native cultures which, each in their turn have yielded to, and been driven south by the pressure of more powerful peoples from the North Island. This constant drift - 246 southward through the decades of prehistoric time by pressure of newcomers to the North Island from the Pacific area, is briefly the most acceptable occupational pattern for the New Zealand area. 36 Not only does tradition relate this, but archaeology helps provide the important additional substantiating evidence. 37 Archaeology has however recently indicated, particularly in the Auckland area, 38 that these southward movements did not necessarily indicate complete vacation of certain areas by a culture, but rather that isolated pockets resisted interference from new arrivals and preserved their cultural elements in relative isolation in the more favourable northern climatic area, until a comparatively recent date.

The earliest cultural group for which there is both a good traditional as well as an archaeological record, was the ancient Waitaha, the first prehistoric arrivals to the land. 39 This group of people are represented by several migrations, at separated intervals, who established themselves at different parts of the country. The first initial influx settled the North Auckland area, but the second, finding their kinsmen already settled in the North Island, proceeded on to the South Island. A third section missed New Zealand completely, and made their landfall at the Chatham Islands, 40 where because of the greatly differing environment they tended to develop local cultural traits, differing in many ways from their relatives in New Zealand, but basically retaining their related affinities. 41 In the Sounds area there is as yet little trace of the occupation sites of this people, but numerous places have yielded their typical triangular cross-sectioned, tanged, hog-backed adze-form artefacts, ranging from heavy narrow-bladed adzes to elongated slender chisels, 42 as well as other associated items of material culture. On D'Urville Island too, the lowest occupational levels, rich in moa remains, synonymous culturally and possibly in time with the Wairau remains, 43 is at present under investigation. 44 There were several mythical peoples said to have inhabited this area in past times, 45 but either because of fanciful attributions to these people or complete lack of any substantiating evidence, they cannot be assigned to any particular culture. They are thought to be tales of remnant groups of ancient Waitaha who inhabited isolated pockets, and were only to be seen occasionally. Amongst these folk there was a more definite group which has received - 247 tribal status—the Rapuwai, 46but as there is little supporting evidence of their individuality, they are regarded on affinities 47 as being either a related group, or a nomenclatural synonym of Waitaha. 48

The next cultural group to arrive in New Zealand are considered to be the Ngati-Mamoe. 49 Unfortunately, culturally little is yet known about them, either through their not being a particularly numerous people, or, what is more likely, because their cultural peculiarities were so subtle as to disallow complete segregation of their material remains from those of their predecessors or successors. However, apart from this, they are definitely considered to have had a considerable Melanesian admixture. 50 Often bearing the totally misleading and now discredited name Moriori, 51 these are the people who scarped the steep hill slopes of the Sounds area with their “staircase” earthworks, dwelling pits, and mounds covering their cremated dead. 52 Tradition mentions the wide-spread coverage of New Zealand by this culture, but to this area alone, these unusual features appear totally restricted. Because of the fact that their culture assumed this unusual form within this area, it can only be regarded as an influence of the exceptional local environment. On D'Urville Island too (as will be described later), these people also made their characteristic landmarks. Their arrival in New Zealand influenced the already established Waitaha, but the actual contact between them at first is regarded as friendly rather than warlike. To some degree the arrival of this new culture heralded the gradual retreat southward of the Waitaha people.

The next arrivals to the country were Toi and his followers in search of his grandson Whatonga, about 1150 A.D., who settled in the Bay of Plenty area, taking wives from the tangata whenua (Ngati-Mamoe). As a result various local sub-tribes gradually sprang up. These tribes, which are all related to the common ancestral stock, gradually asserted their independence and extended their influence over the lands they occupied and exerted pressure on the Ngati-Mamoe and remnant Waitaha peoples in their vicinity. By this time, with the expanded pressure in the north, a large section of North Island Waitaha crossed Cook Strait (about 1477 53) and joined their kinsmen in the South Island. Apart possibly from D'Urville Island, the Waitaha seem not to have occupied the Sounds area to any great extent. With the arrival of the Fleet-Maori, who found the Bay of Plenty area particularly well populated, the descendant tribes of Toi and Whatonga began to look southwards for new territories. With the spread of this group south-ward, the Ngati-Mamoe were affected and sought new lands across Cook Strait, away from their northern pursuers. They were, however, apparently restricted to the Sounds vicinity by the presence of the more - 248 numerous Waitaha peoples further south, with whom they did not wish to conflict, and were thus forced to occupy the extremely hilly terrain, which necessitated the establishment of new settlement systems to suit most satisfactorily the requirements of the challenging landscape. The off-shoot tribes of the Toi-Whatonga group in their turn cast favourable glances at the land across the Strait, and finally took up residence in the Sounds, driving the Ngati-Mamoe southward with depleted numbers. The Ngati-Mamoe prior to the arrival of this later cultural group had on numerous occasions fought with the Waitaha and had pushed them southward. The first established of the descendant Toi tribes, the Ngati-Kuia, 54 is credited with driving the Ngati-Mamoe out of the district. 55 The other sections, now established as separate distinct tribes (although closely related) in the Wairarapa and Wellington districts, quickly followed. Rapidly in succession the Ngati-Kuia, Rangitane, Ngati-Kuri, Ngai-Tara, Ngapuhi, Ngati- Tumatokokiri, 56 Ngai-Tahu and Ngati-Apa, established themselves in the area and waged minor inter-tribal warfare amongst themselves. The Ngati-Kuia and Rangitane (the canal builders) were the main tribes, and occupied Pelorus Sound and Queen Charlotte respectively, the others occupying small pockets in the surrounding areas. With this large inter-related tribal assemblage in the northern part of the South Island, sections of the Waitaha and Ngati-Mamoe moved further southward, finally during early European times occupying the Southland area and intermarrying extensively. 57 This then was essentially the pattern of tribal occupation in the area when the influence of New Zealand's fourth cultural group, the Fleet-Maori, under the destructive influence of Te Rauparaha, made its appearance in the South Island.

Commencing in 1828 to harass the occupants of the South Island, in revenge for their combined part against him in the battle of Wai-o-rua and for other insults, Te Rauparaha gradually reduced many of the tribes to mere remnants of their former strength, and these retreated into isolated aeas, deserting their tribal lands. Finally, being virtually in complete dominance of this area by his conquests, Te Rauparaha reallotted these lands to his allied tribes and offered the rest for sale to the Crown (in 1853), thus imposing the final Fleet-Maori cultural phase of occupation upon the area, which survives in certain areas even today.

THE CULTURAL OCCUPATION PATTERN FOR D'URVILLE ISLAND

From the foregoing it can be realized that D'Urville Island is in itself a complete ethnographic unit, which contains in its relatively small area evidence for a successive cultural occupation pattern which follows and substantiates that already known for the rest of the Marlborough Sounds.

- i
Plate I
Photo C.A.A.—N.Z.G.S. (E. J. Thornley), Au-miro-te-kawau-a-Toru: The swishing current of the shag Toru. French Pass. D'Urville Island in background.
- ii
Plate II, figure 1
The lagoon at Otu Bay, one of the most famous sites on D'Urville Island.
Plate II, figure 2
A flight of small terraces marking the Ngati-Mamoe occupation of Greville Harbour.
- iii
Plate III, figure 1
A large takuahi (hearth) and scattered chips from the workshop site at Ragged Point Beach, Greville Harbour.
Plate III, figure 2
Photo C.A.A.—N.Z.G.S. (S. N. Beatus), A small prominent pa at Ohaua bay, south coast of D'Urville Island.
- iv
Plate IV
Photo C.A.A.—N.Z.G.S. (S. N. Beatus), South side of Greville Harbour—Puketutu Bay. Lettered localities referred to in text under site no. 25, p. 261.
- 249

The Waitaha

It has been realized for some time that the presence of very old middens on D'Urville Island and in the Sounds, about 2 feet below the present ground surface and containing numerous moa bones, 58 belonged to the earliest people to occupy New Zealand. In itself the presence of moa bones in an archaeological site cannot be taken as a decisive cultural criterion, for both the Waitaha and Ngati-Mamoe peoples are known to have come in contact with the moa. However, as recently revealed, 59 these stratigraphically lowest archaeological horizons on the island, occupying in all cases extremely favourable positions, contain very abundant flakes and chips of metamorphosed argillite from the quarry sites along the main eastern ridge of the island. Because of the abundance of these flakes which are not present in later middens, it is evident that all the quarry sites on the island are linked to this earliest period of settlement. The artefacts of metamorphosed argillite which have been found on the island are both in type and material reminiscent of those unearthed at the two relatively closely situated and most important Waitaha archaeological sites in New Zealand—Horowhenua and Wairau Bar. It is with certainty, then, that these older middens and quarry sites can be correctly associated with the Waitaha culture, and in time may be found to date slightly earlier than the Wairau Bar settlement.

Ngati-Mamoe

Although early accounts mention the presence of numerous “pit-dwelling sites”, 60 similar to those in the Marlborough district, on D'Urville Island, their presence today is not discernible. There does exist, however, belonging to this culture, a splendid example of a flight or tier of undefended habitation terraces on the south side of Greville Harbour (plate II, fig. 2, No. 24 on the locality map). This monument in itself serves particularly well to mark the presence of this little-known people on the island.

Pre-Fleet Toi-Whatonga Descendant Tribes

From the traditional account of the occupation of D'Urville Island by the tribes of this group, it is rather difficult to decide whether the settlement of each successive tribe in this group represented a complete territorial domination or, as more suggestive owing to the time factor involved, a mutual occupation pattern. The presence of three pa sites on the island is attributed to this occupational period, as the final Fleet-Maori domination came at a much later date, when the musket had already established its superiority over a pa site defended in the traditional manner. These pa sites, discussed under location numbers 10, 33 and 34, unfortunately cannot be separated from each other in age on a basis of degree of obliteration of earthworks, as they appear roughly in similar states of preservation. The first of these tribes to take up lands on D'Urville Island appears to be the Rangitane and - 250 Ngati-Kuri, who remained undisputed holders of the territory for 250 years, 61 up until the early period of European exploration. The next tribe to settle in the area was the Ngai-Tara-pounamu. 62 These people came from the vicinity of Taranaki and were at first represented by the survivors from a fishing expedition, which was blown south, and finally reached Greville Harbour, where they landed at Moawhitu beach. Being impressed by the apparently abundant food supply and the kind assistance the owners of the land gave them, 63 they decided to bring their relations from Taranaki to settle in the area permanently. While living at Otu, they were attacked by the Ngati-Kahungunu (from the Wairarapa) in revenge for a misdeed they had committed, and many were slain. 64

The next arrivals to D'Urville Island were the Ngati-Kuia, who were already well established in the Sounds, their ancestors having contacted the early “pit-dwellers”. They fought the resident Ngai-Tara and drove them from the island and adjacent areas to the Pelorus Valley. 65 A section of the closely related tribe, the Ngati-Apa, joined the Ngati-Kuia towards the end of the 18th century and took up lands in the vicinity of French Pass. 66 About this period also, a small section of the Ngati-Tumatatakokiri, who had originally occupied Arapawa Island from an early date, 67 were driven to D'Urville Island. At this period, 68 when the presence of Te Rauparaha across Cook Strait began to be felt in the district, the combined tribes of the area, including the sections on D'Urville Island, sent contingents in 1824 to what was to become the battle of Wai-o-rua. The failure of this venture to unseat the Ngati-Toa from Kapiti, with its disastrous consequences for all those who took part, is well known.

Fleet-Maori Domination

Te Rauparaha in 1828 began his campaign of utu for the part the South Island tribes had played in the attack on Kapiti, and landed first on the east side of D'Urville Island. 69 After concentrating his vengeance on the Ngati-Kuia of Pelorus, where they were almost exterminated, he shifted to the section living on D'Urville Island 70 and at Otu in 1829, Te Waihaere, 71 the paramount chief of the Ngati-Kuia, along with many - 251 others of the tribe were killed, many of the remainder being taken as slaves to Kapiti. After his vengeance had been satisfied, Te Rauparaha divided up his conquered lands amongst his various friendly and allied tribes. D'Urville Island was allocated to the Ngati-Koata, Ngati-Haumia and Ngati-Tumania 72 and various villages were built at suitable sites. A small section of the Ngati-Raukawa 73 were in occupation of the east coast. Even today, there are to be found the remains of the Ngati-Koata village at Haukawakawa or Madsen Bay, inhabited by descendants of this tribe.

IMPORTANT ETHNOLOGICAL FEATURES ON D'URVILLE ISLAND

The following sites, most of which are being placed on record for the first time, are discussed under successive numbers, which by reference to the accompanying maps can be found referring to the relevant site under discussion. The observations should be regarded as a preliminary survey of surface sites as most of the features discussed, particularly the major ones which are quite extensive, would require considerable detailed work. This remains a task for the future, and undoubtedly also further intensive investigation could yield small additional sites which are not covered in this article.

Site No. 1. Washley Peninsula.

On the south side of Washley Peninsula, principally situated on the upper areas of the two major south-trending spurs (S5&6/254942) 74 are a series of approximately twenty extensive terraces. 75

Nos. 2, 3 & 4.

South of Patuki, on the ridges above the three main valleys, extending down to Garden Bay, numerous pits and terraces above the valley mouths have been observed. 76 It is probably in this general area that the sealer Samuel 77 anchored in 1824 and was attacked (at “Tapapekohia”) by members of the Ngati-Kuia tribe, for an accidental insult the chief, Te Waihaere, had suffered.

No. 5. Te Marua.

“A cape close to Tinui Island.” 78 This cape, now bearing the name Kidnap Point, is possibly the locality where the whaler Mary & Elizabeth was plundered on 10th August, 1834, 79 by the chief Tomawk (possibly of Ngati-Raukawa) who had a grudge against the previous captain, who earlier had visited the island.

- 252

No. 6. Skull Bay.

At the northern end of the bay, on the north-west-facing spur (S5&6/241921) above the prominent stack rocks, are two large rectangular pits. The lower measures approximately 14 feet by 6 feet, and is about 5 feet deep. Above this at a higher level, the second is of slightly smaller dimensions. Above these pits are a series of natural slump features, giving the erroneous impression of artificial earthworks. In the small gully (S5&6/238917) at the centre of the bay, behind the beach ridge, broken oven stones, midden material, and chips of metamorphosed agrillite lie scattered over the wind-scoured ground surface and on the southern lower hill slopes.

No. 7. Otu Bay.

This locality can be regarded as one of the most important sites on the island. Although the meaning of the name is unknown, a first-hand knowledge of the area gives a most likely clue to the possible interpretation of its original implication. The word otu is generally applied to the wash boards connected to the prow of a canoe, which prevent waves which hit the bow from splashing in-board. There are two topographic features in the bay which could lend themselves to this figurative interpretation. The first and possibly the most likely is the long peninsula of land which runs out to Bottle Point, on the south side of the bay. This landmass forms a barrier to the south and south-west winds and shelters the bay from heavy seas derived from that quarter, preventing them racing into the bay and breaking over the beach. The second is the boulder beach of the bay, where heavy seas have built a large boulder storm-beach ridge which because of its size has formed a natural dam, preventing drainage from the hinterland (which has formed a large lagoon) flowing directly into the sea.

For the earliest inhabitants of the island, dependent on a foraging existence, the area in the vicinity of Otu Bay must have indeed been the most favourable in the district. Barred in the north, east, and south by steep hills, the hinterland forms a naturally sheltered and secluded ampitheatre. Today, stretching in long fingers, the inland gullies are filled with swamp vegetation which thrives on the permanent entrapped moisture from this natural catchment area. West of the swamp regions, the valley broadens out and across its floor stretches the large deep brackish swamp, which even today around the margins forms the home for a prolific avian population (plate II, fig. 1). The area which contains traces of past occupation is restricted to a large deflated and scoured region of sand dunes, covering several acres, beneath the hills on the south side of the lagoon behind the beach area (plate II, fig. 1, lower right corner; S10/159895). All that is visible today are thousands of flakes of black and grey metamorphosed argillite with the occasional partly worked adze, covering the area. This is the remaining proof of habitation of the area by a section of the Waitaha. Near Bottle Point further evidence of the presence of this culture has been unearthed in the form of broken moa bones. 80 Mollusc shells in the drift-sand un- - 253 doubtedly recall the presence of the Ngai-Tara-pounamu and the Ngati-Kuia in the locality also. The small open bay south of Otu preserves in its name an early traditional incident which took place at Otu. The Ngai-Tara-pounamu at Otu were attacked by a chief of the Ngati-Kahungunu, Tawake, in revenge for the killing of two boys of their tribe, one being eaten at Otu: hence Te Puna-a-Tawake, The oven of Tawake. In 1829, in revenge for the part the Ngati-Kuia took in the battle of Wai-o-rua, a relative of Te Rauparaha crossed to Otu and killed the chief Te Waihaere and took many of the tribe back to Kapiti as prisoners.

No. 8. Port Hardy Station.

The homestead, which takes its name from the harbour, occupies a small valley which was undoubtedly an important agricultural and settlement site. The area is extremely sheltered in aspect, and the presence of a stream as well as being a source of fresh water has been responsible for the building up of a thick deposit of fertile alluvium, washed off the surrounding hills and carried down in times of heavy rain. The soil in several areas has a distinctly blackened appearance, and midden material is scattered throughout most of it.

No. 9. Waiua.

Immediately behind the beach ridge, scattered burnt oven stones and charcoal-blackened soil indicate the use of the area as a site of semi-permanent occupation at some period. The depressed area behind the beach, occupied by the Waiua sheep station homestead, is well sheltered from the prevailing north-west wind, and although occasionally subject to flooding, the situation can be regarded as most favourable.

No. 10. Pa Site.

This small pa site, on the north side of Port Hardy, is situated on a peninsula (S10/234894), formed at the lower end of a prominent ridge by the post-glacial rise in sea level. This narrow promontory pa is surrounded on three sides by steep cliffs, and the landward access is cut off by a fosse cutting completely through the narrow neck, with raised earthworks on either side. The peninsula, approximately 300 feet long by 100 feet wide, bears the remnants of two long terraces on the western side, and three on the southern end. It is obvious, judged by the angular debris at the base of the cliffs, that over the years the faces have continually been eroding back, exposing today an area considerably less than existed during the period of its occupation. Mixed with fallen debris, numerous flakes of black metamorphosed argillite lie on the beach, but there is no evidence of an association of this material with the pa above.

No. 11. Quarry Site.

This site, although only of limited extent, represents the first of three additional sites assigned to the earliest culture in New Zealand, from which the well-known metamorphosed argillite used for adze manufacture was obtained. This site, situated on the north side of the East Arm of Port Hardy (S5&6/240905), about 50 yards

- 254
MAP I
The northern half of D'Urville Island
- 255
MAP II
The southern half of D'Urville Island (continued from Map I)
- 256

below the bulldozed road on the ridge, occurs on the extreme western margin of the “intrusive belt” and is in immediate association with large outcropping blocks of serpentine and Te Anau Series argillite. The quarry is distinguished by a single block of dark grey material about 3 feet square, surrounded by a scattering of large flakes covering the surrounding area for about half a chain. Although numerous blocks of the material occur in the region, this appears to be the only outcrop which has been worked.

No. 12. Midden.

A small lens of midden material, predominantly composed of the locally occurring mollusc Chione stutchburyi, extending for about 6 feet in length and 4 inches in thickness, is to be seen in the bank above the beach. This occurrence is most likely to be related to the period in which the adjacent pa was occupied.

No. 13. Camp Bay.

At the head of Camp Bay (S10/244886), directly above the beach level, several horizons of midden material are exposed in the low bank. The lowest layer, the base of which corresponds with the present high-tide level, consists entirely of pieces of charcoal and blackened soil, while at the northern end burnt oven stone are numerous. This lowest horizon reaches its maximum thickness at the northern end (about 6 inches) but is considerably disturbed by a small stream, which has cut down through it washing out and re-covering exposed material with fresh silt. The band thins out to only about 2 inches towards the southern end. This horizon is overlain by a southward tapering layer of silt about 3 inches deep, upon which rests a thin lens of more recent midden material composed entirely of fresh-looking shells of Chione stutchburyi, the commonest intertidal mollusc in the bay. This upper layer, about 1 inch below the present topsoil, is concentrated at the northern and southern ends of the beach where it reaches a thickness of 2 inches and 8 inches respectively, tending to lens-out and disappear in the central region.

It is obvious that the lower horizon represents a period of habitation by early arrivals to the island, but a broken, though distinctly tanged adze of black metamorphosed argillite and several flakes obtained from the northern end of the deposit suggest the presence of representatives of the Waitaha culture. During this period of occupation the small stream at the northern end provided a water-supply and close to its banks the umu were situated. There are no traces of shell in this layer. This corresponds with what Dr. Wellman has found at other sites, where middens of the lowest occupational horizon are represented predominantly by moa bones and stone flakes, and the latter by a mollusc and fish dietary pattern. The sterile layer which formed above this site after it was deserted apparently accumulated during periods of heavy rainfall, representing top soil washed off the surrounding steep hills and carried down in the creek, as indicated by the greatest thickening of this layer at the northern end of the beach. The upper horizon at a very much later date, reflecting a different dietary habit, probably - 257 relates itself to the phase of occupation responsible also for two adjacent sites discussed below.

Directly above and overlooking Camp Bay is a large semi-circular depression (S10/243884) facing directly north-east. It is situated on a spur approximately 200 feet above sea level, immediately behind a small bushed gully leading down towards Camp Bay. The two outer “wings” of the feature appear to have slumped naturally downwards on either side of the spur, while the frontal concave area slopes downward towards the valley. In spite of the natural slumping effects, the site is distinctly artificial and extends for over a chain. It is unlikely that this feature formed part of the “Maori Village” recorded by Dr. Schauinsland in 1896-7 81 on the main ridge directly above Camp Bay and Waitai, as its marked degree of obliteration suggests greater antiquity.

A small midden of Chione stutchburyi is situated at the base of the hill at the eastern end of a small indentation at the south side of the entrance to Camp Bay (S10/241889).

No. 14. Oterawa.

The village site of Te Whetu and his followers—Ngati- Raukawa. It was here that Thomas McLaren lived for a considerable time and built his coaster, the Ocean Queen. This was also the village that Wakefield had noticed from Mt. Pasco, observing that the eastern hillslopes had been cleared for potatoes.

No. 15. Waitai Beach.

Midden traces on the south bank of a small creek at the northern end. Shell material protrudes through the present soil cover.

No. 16. Black Beach.

At the north side of this gully at beach-level is exposed an outcrop of black Te Anau argillite which, in contact with a serpentine mass, has become indurated “to a very hard rock of flinty texture”, 82 for a foot or so around the contact zone. It is mentioned 83 that this hardened material was used for the manufacture of implements, but it was not specifically stated that this outcrop was an actual quarry site. It has been assumed by another author 84 that this is a definite prehistoric quarry, but in fact there is little evidence to support this. The indicative chips at the base of the outcrop and in the immediate vicinity are lacking. There are however numerous flakes to be found along the beach, particularly towards the southern end, but it is more likely that these came from the Mt. Ears locality on the main ridge directly above the beach.

No. 17. South end of Black Beach.

On the northern side of the stream mouth at the southern end of Black Beach (S11/250867), midden debris, oven stones, and broken - 258 reject flakes are most numerous. A large midden area is rapidly being eroded away by the undercutting action of the stream.

No. 18. Quarry Site.

This quarry (S10/235877), situated immediately behind a small sheltered beach, consists of an outcrop and a large boulder on the western bank of a small stream. Very large flakes of black metamorphosed argillite lie scattered in the immediate area, and extend along the surrounding beach. It is evident by the number of partly worked and broken adzes to be seen at low tide on the beach, that this bay was also an extensive workshop site. The most noticeable feature on the beach is the presence of several large (up to 18 inches in diameter) boulders of hard green sandstone, probably connected with the Te Anau Series formations, that have been carried to the site for use as hammer stones for breaking down the exposed rock at the quarry face. In some cases, judged by the chipped and broken appearance of many of the naturally occurring beach rocks, these have played their part as anvils.

No. 19. Quarry Site.

This site extending from above Black Beach to the summit of Mt. Ears, on the main backbone ridge of the island, can be regarded as the largest and most important quarry site for metamorphosed argillite in the area. The continuous outcropping blocks of this material associated with serpentine boulders have been worked on the summit of the ridge and can be seen particularly well where the scrub has been cleared prior to the erection of a boundary fence. The chips of dark grey material cover the ridge for almost three-quarters of a mile and are particularly numerous around each outcrop of worked metamorphosed argillite.

No. 20. Moawhitu Beach.

This beach has been one of the most notable features on the island for yielding artefacts in the past. The gravel beach-ridge which effectively enclosed the low lying region behind, has allowed the formation of an extensive swamp and a large lagoon which is a common occurrence in numerous places around the coasts. This phenomenon is a characteristic feature of all drowned land areas, where original river valleys being submerged have allowed the sea to penetrate to the headwaters of the rivers, ensuring that any debris carried by the river is rapidly deposited close in shore where the stream comes in contact with the sea. This rapidly develops a bar, and storm-piled debris soon accumulates as well. On Moawhitu beach it is apparent that the shingle beach was occupied extensively. Over recent years attempts have been made to lower the level of the lagoon and thus drain the adjacent swamp area. In the course of the work several wooden artefacts 85 have been recovered, and artificially cut tree-stumps have become visible.

- 259

No. 21. O-Ihenga.

It is possible that this name applies to the rocky area east of Moawhitu beach, below the steep cliffs of Mt. Jackson, to which a section of the Ngai-Tara-pounamu on D'Urville Island were enticed to hapuku (groper) fishing by the Ngati-Kahungunu. 86 This embayment has numerous rocks and weed-covered underwater reefs which provide a suitable habitat for such aquatic life.

No. 22. Ragged Point Beach.

This small bay immediately south of Ragged Point is one of the few areas overwhelmed by wind-blown sand. The most noticeable ethnologic feature of this beach is the prolific abundance of chips and flakes of black and grey metamorphosed argillite, the remains of artefact-working activities, and broken adzes in all stages of manufacture littering the area (plate III, fig. 1) in a belt for a distance of about 100 yards. It is obvious, judging by the dunes that have built up and constant clouds of sand that are blown away during high winds, that this area has suffered from severe deflation. Deflation is the process whereby the topsoil is constantly being removed by wind action allowing heavy items such as rocks, or in this case artefacts, to be deposited at a lower horizon than that in which they originally occurred. This naturally means that there is complete mixing of artefactual remains, which were originally separated into distinct cultural horizons with a marked stratigraphic interval between them, now accumulated at a common level. Most of the material exhibits on the upper surfaces a pronounced dulling and smoothing due to the constant abrasive action of the sand. The recognizable typology of the majority of adzes, in many cases in a finished state though having been subsequently broken during use, definitely relates them to the Waitaha period of occupation. There could be a mixing of material, however, due to the disturbed nature of the area. Midden remains are lacking and obsidian is rare, but the presence of a large well-preserved hearth or takuahi (plate III, fig. 1) towards the south end of the beach, near an old stream channel, possibly preserved in the lowest horizon, undoubtedly indicates a period of semi-permanent occupation at some time for the working and manufacture of stone tools at this site. The takuahi is composed of four large elongated water-worn boulders, set up on edge, enclosing an area of 33 by 20 inches. Of particular interest was the discovery of three large flat smooth stones, adjacent to the takuahi, which when removed revealed a cache of broken and partly finished adzes in a small cavity underneath. Included with them was a honing stone, 7 inches long and one inch in diameter, of schist, which is to be found nearby in the Marlborough Sounds area. Deep longitudinal grooves on two sides show its extensive use. It has been assumed 87 that the presence of the vast amount of material in this area indicated a close proximity to a quarry site “near the south entrance to - 260 Greville Harbour”. There is however no evidence for any in the immediate vicinity, and all the rock material observed is considered to have come from the quarry sites on the east side of the island.

No. 23. Owhai Bay.

This beach composed of a distinct boulder bar at the northern end and covered on the inland side with sand dunes banked up against a large grove of trees, has formed an effective barrier to the stream in the valley to the south. This damming of the valley has resulted in the development of a large area of swamp land behind (plate IV, lower right corner). At the northern end, large areas of charcoal-blackened soil overlie the boulder area, and broken disturbed oven stones are prevalent. Scattered midden material, disturbed by the wind and stock, lie amongst the drift sand at the southern end, and an abundance of chips and flakes with numerous broken adzes of local stone lie in a disturbed confusion.

No. 24. Puketutu (plate IV).

At the northern end of the beach (east of Mr. B. L. Woodman's homestead) on the inner sharp-fronted, steep eastern spur which forms the lower part of the main northern ridge originating on the summit to the south-east (2,100 feet), a series of small terraces (plate II, fig. 2) mark the past habitation of a section of the Ngati-Mamoe culture, who because of the challenging topography of the Sounds area established a most distinct series of unique earthwork forms. This site, which has its counterpart in the adjacent Marlborough Sounds, consists of 15 small slightly crescentric terraces each of an area of approximately 8 by 5 feet, with a small scarp no more than 5 feet high above which the next and each successive terrace occurs. Although, because of their age they appear badly slumped (in illustration) they nevertheless still remain relatively clearly defined. Several of the terraces have a slightly raised outer rim, but in some cases this original edge has slumped. Directly above this “gigantic staircase”, the spur flattens out on the ridge, and several adjoining rectangular depressions can be easily distinguished. Several of the pits are only separated from the neighbouring ones by a very small wall about 2 feet wide, and these are tentatively correlated with the “double pits” of the Marlborough Sounds, 88 also a distinguishing criterion of the Ngati-Mamoe, and in this case linking with the terraces below. Although situated on the ridge top, they are markedly infilled, a further suggestion of their considerable age, and thus cannot be associated with the comparatively well-preserved pits of the later people at the centre and west side of the bay.

Beyond the last of the pits, and slightly further up the ridge, a deep, steep-sided fosse cuts completely through the ridge. The presence of this undoubtedly more recent feature should in no way be associated with the adjacent ancient earthworks, for the lower section of this ridge, as well as commanding a fine view of the entrance to Greville - 261 Harbour, is naturally protected by the steep hill-slopes, and the cutting of an artificial fosse completely segregates this area from the main ridge, thus making defence of this feature a simple task. It is probable that this site was selected on its defensive merits, by a section of the Ngati-Kuia or Rangitane people occupying the bay, but it is apparent that for some reason the site was not fully utilized. On the northern side of the lower end of the ridge, an artificial terrace which links two small additional spurs could either be assigned to the fortifying activity at the later period of occupation, or to the systematic deposition of soil, excavated during the establishment of the rectangular pits during the earlier Ngati-Mamoe period of occupation.

No. 25.

On the lower end of the narrow spur, west of the bushed area and the stream behind Mr. Woodman's homestead (plate IV, position “a”), there are four approximately square storage pits, with raised outer rims situated at successive levels above each other. The lowest pit has slumped badly, and has lost its original form, but the upper three are in a good state of preservation, being about 4 feet deep and 6 feet square. Further west beyond this spur, and on the lower slopes of the hill, behind the woolshed, are a further series of pits of similar dimensions (plate IV, position “b”). On two small natural slump-mounds, above the woolshed, several circular depressions are noticeable. Immediately behind the shed, on another larger mound, are a series of seven grouped pits. Apart from the six on the south-west side which are obliterated to a great extent, the major pit at the northern end of this group is quite significant. It is rectangular in plan, measuring about 20 feet by 10 feet, and is about 8 feet deep. All these pits in this area, and undoubtedly more may yet be discovered, are regarded as kumara storage pits belonging to the late occupants of the bay. Their situation on either side of the broad, gently-sloping north-facing valley suggests that this flat area was an important agricultural site which supported a large population.

No. 26.

A small lens of midden material, predominantly Chione stutchburyi, occurs about 6 feet above sea level, at the south-west side of the entrance to the arm.

No. 27.

A large area of midden material and blackened soil occurs at the secluded head of the arm, just above high-tide level, and extends into the bush on the south-west side.

No. 28. Opotiki.

On the large gently sloping naturally slumped terrace, 20 feet above beach level, are signs that the area was used extensively for agricultural purposes in the past. Comparatively sheltered at the northern head of Manawakupakupa (plate III, fig. 2, top right), long low mounds running obliquely or parallel to the beach imply - 262 the clearing of the fertile well watered flat of stones prior to cultivation. At the eastern end of this extensive area of flat which covers several acres, on the banks of a small but deeply incised stream, three or four south-east-facing terraces are situated slightly below the main terrace level. Close to the edge directly above and parallel to the beach, several small circular pits or depressions continue in a line across the area. On the west side of the terrace directly below the last depression, midden material lies scattered over a wide area. None of these artificial features are particularly well defined, merging completely under certain lighting conditions with the landscape. It is apparent that this noticeable degree of natural obliteration indicates that this site was probably occupied at a reasonably early date, possibly by the Ngati-Kuia or Rangitane.

No. 29.

Blackened soil, charcoal, broken oven stones, and midden material is noticeable in the stream banks east of the homestead, below the bridge crossing the stream.

No. 30.

It is possibly in this bay north of Haukawakawa that the village site noticed by Cook and later by D'Urville originally occurred. Possibly either Ngati-Apa or Rangitane.

No. 31. Madsen Bay or Haukawakawa.

The name Haukawakawa (kau—wind; kawakawa—bitter or severe), dates from the period of the Ngati-Kuia occupation. 89 This settlement under the head man, Mr. Turi Ruruku Elkington, is occupied by descendants of the Ngati-Koata people, a hapu of Ngati-Toa. They date their occupation in the area from the late 1800's, and like many associated tribes took up residence on the lands conquered and controlled by Te Rauparaha, presented to them in recognition of their support of the Ngati-Toa during past battles.

No. 32. Quarry Site.

This site, previously described, 90 is situated “on a spur between Woodman's Bay and the small bay nearer the French Pass”. It occurs just below the summit of the spur, on the north side. On the south side of this spur, slightly above sea level, a large boulder has been worked extensively.

No. 33. Ohaua Bay pa.

This hill pa is the dominant feature at the south end of D'Urville Island (plate III, fig. 2). It faces directly south, towards the pa on Hautai (Burial Island), and is protected on this side by low cliffs. Five terraces encircle this site, which rises to about 100 feet above sea level. Beyond this, running up the western slopes of the main south valley of the island, large rectangular plots of land, several - 263 chains square, are clearly defined by long parallel mounds, indicating past agricultural areas. On the ridge above this area and a little to the north, several rectangular pits overlook Te Puna lagoon. At Ohaua bay, the presence of a quarry site has been recorded. 91 There is little now to indicate this, but it could have occurred on the western ridge opposite the pa, where large serpentine masses outcrop above the beach.

No. 34. Hautai Island.

On the gently sloping northern face of this island, composed entirely of serpentinous rock, eight large distinct terraces (along with several minor features on the top of the island) traverse the slope to the extreme eastern and western margins for a distance of about 80 yards. Apart from the northern side, the remainder of the island rises sheer from the water, and the terraced defences protect the only approach, which at low tide is joined to the mainland by a shingle spit. These terraces directly face (plate III, fig. 2) the small conical hill pa on the mainland at Ohaua Bay.

ACKNOWLEDGEMENTS

I wish to express my appreciation to Dr. C. A. Fleming and Dr. J. B. Waterhouse for the opportunity of visiting D'Urville Island in the process of geological mapping work. To the Moleta families at Waitai, Mr. Gilbert Leov of Port Hardy, Mr. David Leov at Patuki, and Mr. and Mrs. Woodman at Greville Harbour, sincere gratitude is due for their splendid hospitality. Thanks are also due to Mr. T. K. Elkington for his comments on the early history, and to Dr. H. W. Wellman for discussions on his recent work.

REFERENCES
  • ADKIN, G. L., 1948. Horowhenua. Wellington, Polynesian Society, Memoir Vol. 26.
  • — — 1952. “The Antiquity of Man in the New Zealand Area.” N.Z. Science Review, 4:41-45.
  • — — 1959. “An Unrecorded Type of Patu.” Journal of the Polynesian Society, 68:86-92.
  • ANDERSEN, J. C., 1942. Maori Place Names. Wellington, Polynesian Society, Memoir Vol. 20.
  • BEATTIE, H., 1920. “The Southern Maori and Greenstone.” Transactions of the N.Z. Institute, 52:45-52.
  • BROTHERS, R. N., and GOLSON, J., 1959. “Geological and Archaeological Interpretation of a Section in Rangitoto Ash on Motutapu Island, Auckland.” N.Z Journal of Geology and Geophysics, 2:569-577.
  • BUICK, T. L., 1900. Old Marlborough. Palmerston North, Hart and Keeling.
  • COTTON, C. A., 1913. “The Tuamarina Valley: A Note on the Quaternary History of the Marlborough Sounds District.” Transactions of the N.Z. Institute, 45:316-322.
  • — — 1942. Geomorphology. Christchurch, Whitcombe & Tombs Ltd.
- 264
  • DIESELDORFF, A., 1901. Beiträge zur Kenntniss der Gesteine und Fossilien der Chathaminseln sowie einiger Gesteine und neuer Nephritifundorte Neu-Seelands. Marburg, Universitäts-Buchdruckerei (R. Friedrich).
  • DUFF, R. S., 1946. “Native Quarries of Baked Argillite.” Records of the Canterbury Museum, 5:115-124.
  • — — 1950. The Moa-Hunter Period of Maori Culture. Wellington, Dept. of Internal Affairs.
  • — — 1956. “The Evolution of Polynesian Culture in New Zealand.” N.Z. Science Review, 14:147-151.
  • ELVY, W. J., 1949. Kaikoura Coast: Maori History, Traditions and Place-names. Christchurch, Whitcombe & Tombs Ltd.
  • — — 1957. Kei Puta te Wairau: A History of Marlborough in Maori Times. Christchurch, Whitcombe & Tombs Ltd.
  • FIELD, A. N., 1942. Nelson Province 1642-1842. Nelson, A. G. Betts.
  • FREEMAN, J. D., 1959. “Henry Devenish Skinner: A Memoir.” pp. 9-28 of Freeman and Geddes (eds.) Anthropology in the South Seas. New Plymouth, Thomas Avery & Sons Ltd.
  • GERARD, S., 1938. Strait of Adventure. Wellington, A. H. & A. W. Reed.
  • GRACE, Archdeacon, 1901. “Maori Traditions from D'Urville Island.” Journal of the Polynesian Society, 10:65-71.
  • HENDERSON, J., 1924. “The Post Tertiary History of New Zealand.” Transactions of the N.Z. Institute, 55:580-599.
  • HOCHSTETTER, F. von, 1959. Geology of New Zealand, (translated and edited by C. A. Fleming). Wellington, Government Printer.
  • INSULL, H. A. H., 1952. Marlborough Place Names. Wellington, A. H. & A. W. Reed.
  • JOBBERNS, G., 1935. “The Physiography of Northern Marlborough.” N.Z. Journal of Science and Technology, 16:349-359.
  • JONES, W. M., 1939. “Magnetic Observations of Nelson Serpentinites, with Remarks on the Geology of the Asbestos Deposits.” N.Z. Journal of Science and Technology, 21:128B-142B.
  • KEYES, I. W., 1958. “Blackstone.” Journal of the Polynesian Society, 67:158-161.
  • McINTOSH, A. D. (editor), 1940. Marlborough, a Provincial History. Blenheim, Marlborough Provincial Historical Committee.
  • OLIVER, W. R. B., 1942-4. “The Vegetation and Flora of D'Urville Island and Stephen Islands.” Records of the Dominion Museum, 1:193-227.
  • PEART, J. D., 1937. Old Tasman Bay. Nelson, R. Lucas & Son Ltd.
  • POWELL, A. W. B., 1930. “The Paryphantidae of New Zealand: their Hypothetical Ancestry.” Records of the Auckland Institute and Museum, 1:17-56.
  • PYBUS, T. A., 1954. The Maoris of the South Island. Wellington, A. H. & A. W. Reed.
  • REED, J. J., 1959. “Soda-Metasomatised Argillites Associated with the Nelson Ultramafic Belt.” N.Z. Journal of Geology and Geophysics, 2:905-915.
  • RIGG, T., CHITTENDEN, E., and HODGSON, L., 1957. A Survey of Soils, Vegetation and Agriculture of the Eastern Hills, Waimea County, Nelson. Nelson, Cawthron Institute.
  • RUTLAND, J., 1894. “Traces of Ancient Human Occupation in the Pelorus District, Middle Island, New Zealand.” Journal of the Polynesian Society, 3:220-232.
  • — — 1897. “On the Ancient Pit Dwellings of the Pelorus District, South Island, New Zealand.” Journal of the Polynesian Society, 6:77-84.
- 265
  • SKINNER, H. D., 1958. “David Teviotdale.” Journal of the Polynesian Society, 67:321-323.
  • TAYLOR, W. A., 1950. Lore and History of the South Island Maori. Christchurch, Bascands Ltd.
  • TE PUNGA, M. T., 1953a. “The Paryphantidae and a Cook Strait Land Bridge.” N.Z. Journal of Science and Technology, 35(B):51-63.
  • — — 1953b. “A Late Pleistocene Land Bridge Across Cook Strait, New Zealand.” N.Z. Journal of Science and Technology, 35(B):161-192.
  • — — 1958. “Evidence for a Low Sea-Level 9,900 Years Ago.” N.Z. Journal of Geology and Geophysics, 1:92-94.
  • THOMSON, J. A., 1918. “Maori Rock-Quarries on D'Urville Island.” N.Z. Journal of Science and Technology, 1:321-322.
  • WAKEFIELD, E. Jermingham, 1955. Adventure in New Zealand. Christchurch, Whitcombe & Tombs Ltd.
1   Buick 1900:51.
2   Adkin 1952:44.
3   Field 1942:18.
4   Field 1942:65.
5   Guard's wife is credited with being the first white woman to live in the South Island, and her son and daughter became the first European children to be born there.
6   Field 1942:78.
7   The ships were due to arrive on the 10th of January, but although Wakefield lit a beacon on the highest hill (Mt. Pasco), they did not appear. After contacting McLaren, he departed for Port Nicholson, only to find the vessels he was awaiting had already arrived here.
8   Peart 1937:110.
9   Peart 1937:111-112.
10   Andersen 1942:249.
11   Andersen 1942:249.
12   Wakefield page 21 in Adventure in New Zealand interprets the meaning as “blood-coloured sky” which is the common translation, but this is in no way applicable here. Rangitoto as a place name has been applied at least six times in the North Island but they each have their own special interpretation. Another meaning of rangitoto can be black lava or scoria, probably reminiscent of the Hawaiian Islands.
13   Field 1942:66.
14   Insull 1952:52.
15   This latter name appears on the map prepared by Dr. H. Schauinsland in 1896-7 and published by Dieseldorff in 1901.
16   Field 1942:67.
17   Skinner 1958:322.
18   Freeman 1959:11.
19   Freeman 1959:11.
20   Keyes 1958.
21   Cotton 1942:442.
22   Cotton 1913:320.
23   Henderson 1924:592.
24   Jobberns 1935:358-359.
25   Te Punga 1953b.
26   Powell 1930 and Te Punga 1953a.
27   Te Punga 1958:93.
28   Rigg et al. 1957:8-9.
29   Greville, Rangitoto, Waiua and Stephens formations.
30   Oliver 1942-4:201.
31   Keyes 1958.
32   More correctly metasomatised argillite; refer Reed 1959:911-914, for a petrological description of this material.
33   Although having displaced the “mineral belt” to the mouth of Pelorus Sound, Elvy 1957:2 refers to the importance of the argillite or pakohi (pakohe) that this area has yielded for the production of high-quality artefacts in the past.
34   It was noted by Hochstetter in 1859. Hochstetter 1959:241.
35   Dieseldorff 1901.
36   Adkin 1952:44-45.
37   There are other schemes demanding a cultural evolution of the present Maori race from within the country itself. These are based mainly on restricted interpretation of artefacts and do not take into account or include the rich heritage of tradition, which combined with material evidence provides the basic clues to the past settlement pattern within New Zealand.
38   Brothers and Golson 1959:575-576.
39   Adkin 1948:112-122.
40   This view has been recorded by several authoritative early writers as has recently been emphasized by Adkin 1959:91.
41   Duff 1956.
42   Seen in both private collections and in museum accessions, e.g., the A. C. O'Connor Collection in the Dominion Museum, Wellington.
43   Duff 1950.
44   Dr. H. W. Wellman, personal communication.
45   Elvy 1949:4.
46   Peart 1937:7.
47   Pybus 1954:35.
48   Pybus 1954:34.
49   Adkin 1948:122-124.
50   Buick 1900:57.
51   Duff 1956.
52   Buick 1900:69.
53   Taylor 1950:22.
54   Elvy 1957:29.
55   Elvy 1957:21.
56   From the Wanganui district, and settled in Queen Charlotte Sound, but were fought by the Rangitane, Ngati-Kuia and Ngati-Apa and later died out. Peart 1937:13, 16.
57   Beattie 1920:47.
58   Rutland 1897:82.
59   Dr. H. W. Wellman, personal communication.
60   Rutland 1897:83.
61   Gerard 1938:147.
62   Grace 1901:68.
63   Elvy 1957:28.
64   Peart 1937:117. Another pa at Tawera Point in Pelorus Sound was simultaneously attacked also.
65   Elvy in McIntosh 1940:390.
66   Peart 1937:17.
67   Elvy in McIntosh 1940:389.
68   Gerard 1938:147 mentions final occupation of the island by Ngai-Tahu, but there seems to be very little evidence to support this; more likely a small group of this tribe was in occupation of a bay or peninsula for only a short period.
69   Buick 1900:188. During one of his several visits to the island, Te Rauparaha cremated the body of one of his wives to prevent it falling into the hands of the vengeful local tribes.
70   Peart 1937:36.
71   Peart 1937:21.
72   Buick 1900:209.
73   Field 1942:74.
74   All grid references quoted refer to the Lands and Survey one inch to one mile topographic map series.
75   Dr. J. B. Waterhouse, personal communication.
76   Dr. J. B. Waterhouse, personal communication.
77   Peart 1937:107.
78   Peart 1937:105.
79   Peart 1937:105.
80   Freeman 1959:11.
81   Dieseldorff 1901:plate 3.
82   Jones 1939:137B.
83   Jones 1939:137B.
84   Reed 1959:907.
85   Mr. Gilbert Leov, personal communication.
86   Peart 1937:117.
87   Duff 1946:124.
88   Rutland 1894:221.
89   Mr. T. R. Elkington, personal communication.
90   Thomson 1918:321.
91   Duff 1946:124.