Florence Lane, Te Puna, Tauranga

Archaeological Investigations, U14/1465, Florence Lane, Te Puna, Tauranga

report to The New Zealand Historic Places Trust and Deep Creek Developments Ltd

Louise Furey

CFG Heritage Ltd. P.O. Box 10 015 Dominion Road Auckland 1024 ph. (09) 638 6624 [email protected] Archaeological Investigations, U14/1465, Florence Lane, Te Puna, Tauranga

report to The New Zealand Historic Places Trust and Deep Creek Developments Ltd

Prepared by: Louise Furey

Reviewed by: Date: 29 November 2008 Matthew Campbell Reference: 2006/17

CFG Heritage Ltd. P.O. Box 10 015 Dominion Road Auckland 1024 ph. (09) 638 6624 [email protected] Archaeological Investigations, U14/1465, Florence Lane, Te Puna, Tauranga

Louise Furey

Introduction January 2006. Pirirakau, a hapu of Ngati Ringinui, whose Th is report describes the archaeological investigations rohe includes this area, gave their support for the project. carried out in Florence Lane, Te Puna, Tauranga. Two Th e development area, situated inland from the Wairoa midden/pit sites (U14/1465A & B) were investigated, and River and Te Puna (Figure 1), straddles several geological the surface was scraped in the vicinity of a third midden formations: the northeastern edge of the Minden rhyolitic site (U14/3239) to expose any features present at the inter- dome, the level fl uvial terrace or plateau which extends face of the topsoil and sub-soil. Th e excavations were car- from the base of the hills to the margins of the Tauranga ried out on behalf of Deep Creek Developments Ltd. Harbour, and on the northwest side of the unnamed creek An archaeological assessment (Furey 2005) prior to the is the Te Puna Ignimbrite. Th e coastal terraces of the subdivision of DPS68008 into fi ve lifestyle blocks identi- Tauranga Basin tend to be fl at to gently sloping land with fi ed one previously recorded site, U14/1465, and recorded steeper sided ridges trending N to NNE, ending in low three new sites, U14/3239, U14/3240 and U14/3241. All coastal cliff s adjacent to the harbour (Briggs et al. 1996: 6). were described as shell middens in the absence of any vis- Th e terraces or plateaus are intersected by shallow valleys. ible features. However, further inland, the streams have cut down form- Over most of the new lots there was to be minimal ing deep, steep sided and narrow valleys such as the one impact on the landscape. However the formation of the link present in the Florence Lane development area. road between the previously no-exit eastern and western Th e parent rock is mantled in a number of tephra, the parts of Florence Lane through a steep-sided, deep gully last of which was deposited approximately 800 years ago. required the ridge on which U14/1465 was situated to be Th e most relevant are the Rotoehu Ash, a white shower used for fi ll and as a base for the road. U14/3239 would also bedded tephra typically between 300 and 500 mm deep, be destroyed through being within the road alignment. Th e and the more recent post-Rotoehu tephras of which there New Zealand Historic Places Trust granted an authority to may be up to 10 present. Th e most recent are the Taupo modify (2006/120) under section 14 of the Historic Places and Kaharoa tephras. Due to bioturbation and mixing of Act 1993, and the archaeological excavations took place in the relatively thin Holocene deposits it is diffi cult to distin-

Archaeological sites Pa U14/1465U14/1465 Other U14/1465U14/1465 U14/885U14/885 U14/885U14/885 U14/3221U14/3221 Figure 1. Location of the study area in the catchment of the Wairoa River.

Louise Furey 1 CFG Heritage Ltd. guish individual tephras (Briggs et al. 1996: 44). Although east side of the gully but had previously been disturbed the underlying geology usually infl uences soil characteris- and was not investigated further. tics, in this case the deposition of tephras over all geologi- U14/3239 cal types has created similar conditions. Tephra soils are generally friable and fertile. Located on the edge of the steep sided gully, this midden Th e unnamed creek which fl ows into the Wairoa River site had been partially disturbed by vegetation removal has the upper limit of tidal infl uence immediately down- prior to the assessment. Th e grassed level land adjacent to stream from the property. Th e stream would have been the steep slope was stripped of topsoil with a mechanical navigable by canoe for part of the way up its course, but excavator. Th e topsoil was found to be deep with none of probably not as far up as this area. Th e Wairoa River would the shell visible lower on the slope present. Th e depth of be approximately a 2 km walk to the east across country, or topsoil indicated repeated ploughing which had broken up a 3 km route via the stream to where it enters the Wairoa and destroyed any shell present. No other archaeological River. features were present. Within the development block the rear of the ﬂ uvial U14/1465A terrace abutted the hill slopes of the Minden Rhyolitic Dome. A narrow steep-sided ridge, rising to 60 m contour, Th e fl at area at the highest point of the ridge was consid- formed the southern boundary of the property. The toe of ered to have the highest potential for having a residential the ridge extended out onto the terrace in a low broad pro- structure. Th e location of the site ensured that it had not tuberance (Figure 2). The remainder of the north facing been ploughed in the past, an unusual situation in the slopes in the block were of broken form and subjected to Tauranga area. Th erefore the fl at area was turfed by hand slumping and soil movement in the past. The ﬂ uvial ter- to ensure no features or artefacts were destroyed or dis- race on the southern side of the stream was a relatively placed during exposure of the underlying deposits. Th e narrow margin at the base of the slope. The stream gully remainder of the area was opened up for excavation with a was approximately 20 m deep, with very steep sides. mechanical back hoe using a 1 m weed bucket, to maxim- ise allocation of time and resources for excavation of fea- Archaeological sites tures and recording. U14/1465A was described in 1982 as an eroding midden on Th e extent of the shell midden was uncovered, and the the eastern slope of the narrow ridge. Th e site was recorded turf and recent topsoil stripped off to ensure all archaeo- as part of an archaeological site survey programme directed logical features were uncovered. Th e topsoil was a dark by Dr Bruce McFadgen, then archaeologist at New Zealand grey-brown friable soil 80–120 mm in depth over a yellow Historic Places Trust, to record archaeological sites in brown mottled tephra. Th is tephra layer was distinguished Tauranga County. Th e shell described was on the eastern by brown mottles and small pieces of a cream coloured slopes of the ridge below the highest point in the property. coarse tephra indicating disturbance of a deeper underly- Adjacent to the fenceline on the highest part of the ridge ing tephra layer. Features were cut into the mottled yellow- there was an artifi cially fl attened area measuring approxi- brown tephra which was between 50–80 mm deep. Th is mately 10 x 5 m, associated with shell midden. At a lower tephra layer can be interpreted as the culturally disturbed level on the ridge there was a possible terrace measuring 5 upper horizon of the natural, undiff erentiated, yellow- x 2 m with a possible pit depression. Probing with a stain- brown tephra which was over 800 mm deep. less steel probe indicated there were “soft patches” in the Features were allocated sequential numbers, and subsoil where the probe met with little resistance as it was dimensions, fi ll type, colour and relationship to other fea- inserted into the ground. Th ese were interpreted as likely tures were recorded for each excavated feature. All features to be the fi ll of kumara storage pits. Shell midden is also were also mapped by hand to a 1:50 scale map (Figure 3). present on the broad toe of the ridge projecting out into On removal of the remnant topsoil, shell midden and the rear of the fl uvial terrace. Th is area, designated U14/ several fi rescoops were visible, as were several postholes. 1465B, was also considered to have potential for investiga- Th e majority of the features were on the narrow crest of tion although a farm track and water trough had disturbed the ridge which had a slight slope from south to north. Th e surface deposits. midden layer, consisting of whole and fragmented shell and Th ree other midden sites were located (Figure 2). occasional fi re cracked rocks within a black loam matrix, Two (U14/3239 and U14/3240) were on the west side of extended over an area of 5.8 x 4.2 m and had several scoops the gully, with shell visible at the top of the slope to the dug into the surface of the shell. To the west of the midden stream. U14/3240 was to be protected and not excavated, layer was an area of black-stained, thin tephra soil which while U14/3239 was in an area to be cut down to allow road merged into the midden. It is likely that the black-stained access into the gully. Th e third site (U14/3241) was on the tephra pre-dated the shell midden. Th e previously identi- fi ed fl at area on the western side of the turfed area was not

2 Florence Lane Figure 2. Contour map of development area showing archaeological sites. Based on map supplied by Apex Consultants, Tauranga.

level when the turf was removed, and was in fact, sloping. source of the black discolouration in this area. Th ere was Several other individual fi rescoops and postholes were no shell in these fi rescoops, nor was there any in 9, 10 or 16, also apparent. Later, aft er rain and completion of the exca- also fi rescoops, further to the west. See Appendix I for size vation, four infi lled storage pits were visible on the eastern and description of features excavated. side of the excavated area that had not been visible when Within the black lens there were stakeholes and small the exposed subsoil surface was dry. Th e superimposition postholes containing small fragments of shell midden of features indicates changing use of the site over time, or (Features 12, 14, 15, 30 and 33). Th ese were probably cut more likely, that more than one occupation is represented. into the black lens. In contrast, Features 29, 20, 21, 32 and Firescoops 17, 18 and 51 were fi lled with a similar black 36, fi lled with black loam, were only visible once the black loam with charcoal fragments, and may well have been the lens had been removed. Similarly stakeholes 34 and 35

Louise Furey 3 CFG Heritage Ltd. Figure 3. Features excavated, U14/1465A. under the shell midden were fi lled with black loam. All Miscellaneous features included Feature 49, a small so- these stakeholes are probably contemporary but no pat- called bin pit measuring 420 x 370 x 460 mm deep. Th is terns can be made from their arrangement. Depths were was the only bin pit found in the site, an unusual result for shallow except for postholes 12 and 32 which were 480 a site in the western Bay of Plenty area where pits of a range and 500 mm respectively and would have been very stable of sizes are found in large numbers. (and have a greater degree of permanence than small shal- Feature 25 is not an archaeological feature but was a lowly inserted stakes). Postholes and stakeholes visible hard semi-rectangular outline of natural yellow-brown when the shell was removed were fi lled with black loam tephra. It stood proud of the black discolouration under and fragments of shell. At least some of these pre-dated the midden deposit. At the time of excavation its purpose the midden: stakehole 28 was cut by fi rescoop 27 and only couldn’t be explained as it appeared to be surrounded by visible in the base of the scoop once the contents had been a soft er textured tephra of the same colour which was removed. Similarly 20, 21 and 31 were cut by fi rescoop 13. present over much of the excavated area. Th is in itself was Within this area fi rescoop 39 was also before deposition of not considered unusual as there is oft en variability in the the midden and was only visible once the shell had been upper horizon of the tephra layers caused by past vegeta- excavated. tion disturbance. However on refl ection and based on sub- To the eastern side of the excavation area six fi rescoops sequent events, it is likely that this was the only remnant of have been dug in a small area. Each was fi lled with shell in the unmodifi ed tephra and there are several unexcavated a black matrix. Posthole 42 is the earliest feature, having and undetected pit features adjacent. been cut by fi rescoop 41, which was in turn cut by 44 and While the excavations were in progress in January 2006 45. Th e relationship of fi rescoop 4 to this cluster of scoops is the weather conditions were very hot and dry. Th ere had unknown, but it also is likely to be part of the same general been no rain for some time prior, possibly up to 2-3 weeks. activity of cooking shellfi sh and raking out the debris. As a result the tephra sub-soil, with no organic material, dried out to a uniform pale colour, and freshly excavated

4 Florence Lane surfaces immediately turned loose and dusty. All visible Th e fl attish surface of the spur was excavated. A mechan- features on this site had been excavated, the site scraped ical backhoe was used to remove the previously ploughed down on several occasions, and attention turned to U14/ topsoil to a depth of between 150–180 mm over a 16 x 8 1465B at the toe of the spur. Aft er several hours of rain m area. Regularly spaced plough lines, on three diff erent one evening U14/1465A was checked for any other features orientations, were evident as the topsoil was stripped off . which might have become visible with the moisture. Four A geotechnical test pit had disturbed part of the northern rectangular pits were observed. Given the time constraints end of the site. Archaeological features were exposed on the pits were only investigated to the extent of determin- the bright yellow-brown subsoil at the base of the plough ing size (where not under a spoil heap) and depth. None of zone but the actual surface on which people had lived the pits were fully or partially excavated to investigate the had been churned up into the undiff erentiated ploughed posthole pattern in the fl oor of the pit. soil horizon. Th ere was a confi ned area of shell midden Feature 53 was the later of two pits: it had a fi ll of and postholes and several storage pits. Th ere were also a yellow-brown tephra, and on the surface there were char- number of irregularly shaped dark brown patches which, coal fl ecks associated with the adjacent fi rescoops. Th is pit on excavation, did not turn into constructed features. It is cut through the fi ll and wall of Pit 52 which had been fi lled assumed these resulted from vegetation disturbance. with mottled yellow-brown tephra. Pit 52 was 6.55 x 2.75 Th e stakeholes were small and relatively shallow x .95 m deep. Th e smaller Pit 53, which could be termed a although the surface from which they had been dug was bin, was 750 x 900 x 530 mm. Two other pits, 54 and 55, within the disturbed plough zone. Several had shell frag- were further to the east. Th e dimensions of Pit 54 could not ments within the fi ll, even where there was no overlying be obtained as the spoil heap was situated over part of the shell deposit. Stakeholes 60, 62 and 68, with similar shell pit. Pit 55 was 1.80 x 1.45 x .68 m, and like the other pits fi ll, could be part of a structure as there is a linear arrange- was fi lled with a mottled yellow-brown, which with mois- ment. Other stakeholes in the same area had a grey yellow- ture was of a slightly darker colour than the surrounding brown, or soft brown fi ll. Th ere were several postholes to soil. the north of the storage pits and stakeholes 66 and 70 were Th ree obsidian fl akes were found on the surface of dug into the fi ll of storage pits 75 and 88. Th ere was how- the shell midden or the black layer, and three other fl akes ever an absence of stakeholes in the vicinity of the shell found in the wider area. midden. Nine storage pits of varying sizes were uncovered U14/1465B (Figure 4 and Appendix 1). Due to shortage of time, and Th is site was visible on the surface as fragmented shell wet weather towards the end of the excavation period, only midden exposed in a farm track which went across the toe some pits were emptied of fi ll, or partially emptied in order of the spur. A water trough was immediately upslope of the to uncover the posthole arrangement in the fl oor of the pit. track. Th ere was no shell midden visible in the pugged and Pits 86, 91 and 92 predated the shell midden which was disturbed ground around the trough. Th e toe of the knoll on the pit fi ll. Firescoops 80 and 85, and 74 were probably is approximately 5 metres above the fl uvial terrace abut- responsible for the bulk of the shell and rakeout debris. ting the hills (see Figure 2) Scoop 74, utilizing the depression in the northern end of the partly infi lled storage Pit 91, was the largest and con-

Figure 4. Features excavated, U14/1465B.

Louise Furey 5 CFG Heritage Ltd. tained fragmented shell and loam. Pit 91 was in turn cut into the southwest corner of Pit 92, only defi ned on the surface aft er rain as a darker rectangle of yellow-brown mottled fi ll against the yellow-brown subsoil. Pits 75, 86, 87 and 88 were fi lled with clean mottled fi ll, and backfi ll- ing probably occurred soon aft er the pits were no longer used for storage. In contrast Pit 67 was completely fi lled with clean shells with no soil matrix, and the fi ll of Pit 69 was a dark brown loam with several lenses of yellow-brown tephra, and lenses of shell near the pit fl oor (Figure 5). Pit 91 also had a lens of shell near the fl oor. Pit 69 was completely excavated. A sump, 300 mm wide and 120 mm deep, extended across the width of the pit at the southern end. Th ere were no postholes in the fl oor of the pit, nor were there any external to the pit which might have formed part of the roof structure. Pits 67 and 91 also had no postholes in the fl oor. Th e method of covering or Figure 5. Cross-section through fi ll of Pit 69, U14/1465B. Fill weatherproofi ng these small pits, or bins as they are com- layers are as described: A is grayish yellow-brown mottled; monly called, is unknown and ploughing has destroyed B yellow brown; c mottled yellow-brown; D dark brown the upper edges of the features. loose loam with shell fragments. There are lenses of shell near the base of the pit fi ll.

Figure 6. Plan of partly excavated storage pits showing the posthole pattern in the fl oor. Pit 88 is on the left, and 75 on the right.

6 Florence Lane Pits 75 and 88 were part excavated (Figure 6). Th e fl oor Cominella adspersa (speckled whelk), and Paphies subtri- of Pit 75 had four postholes in the one-third excavated. Th e angulata (tuatua). Other species such as Zethalia zelan- ridgepole of the roof was most likely held up by two or pos- dica (wheel shell) and Buccinulum vittatum (lined whelk), sibly three posts placed centrally through the pit. Posthole Maoricolpus roseus (turret shell) and Zeacumantus lutu- 100 would have been a roof support and was 450 mm deep. lentus (horn shell) are considered not to have been delib- Th e remaining postholes (Features 97–99) are shallow and erately gathered and eaten, but were gathered incidentally 100 mm or less deep. Th ese probably supported a frame on along with the main shellfi sh species of cockle and pipi. All which the kumara was placed. Pit 88 also had a centrally shellfi sh with the exception of tuatua are found in harbour placed posthole (Feature 94) near one end which would environments. Dead tuatua shells washed into the harbour have been part of the roof structure. Th is posthole was 170 from the ocean beach may have been collected from the mm deep. Th ree other stakeholes (Features 93, 95 and 96) pipi beds. between 5 and 14 mm deep were also probably part of a Th e shell midden at U14/1465A was generally highly frame or rack arrangement within the pit. fragmented and burnt, and analysis would have proved Only three fi rescoops were located for this site: 74, diffi cult. Th e quality of the shell midden also dictated the 80/85 and 90. Each had fragmented and burnt shell in a suitability of feature contents for radiocarbon dating. dark matrix, with a small quantity of oven stones present. Storage pits Th e fi rescoops were obviously used to cook the shellfi sh, and shells raked out and distributed around the area. It is In comparison to sites excavated adjacent to the coast, likely that the fi rescoops were contemporary within one these sites have an unusually low number of pits. Th ey are occupation of the site, and later than the pit storage. also not large in size (Figure 6 and Table 2), with none of the super-sized pits evident on other western Bay of Plenty Midden analysis sites present. Many sites seem to have one pit larger than Shell samples for analysis were taken from the fi rescoops all others and these two sites are no exception. Pit 92 on and from major midden deposits. Where possible, a 10 litre U14/1465B is longer than any other pits but narrow for its bucket of shell was collected to provide a consistent-sized length at a ratio of 1:3.7. Pit 52 is the largest on U14/1465A. sample for comparable results. Where there was a lesser Th e relatively shallow depth of the pits can be accounted quantity of shell available, approximately 2–3 litres of shell for by the plough zone: the original depth of the pits may was collected. have been up to 200 mm deeper. Unusually there are few Shells were washed, dried and sorted to species. small pits, commonly termed bin pits, which are under Minimum numbers of shellfi sh were calculated by count- 1 x 1 m. Pits of this size occur more frequently on some ing the number of bivalve hinges and dividing by two to sites than large pits. Th eir function is unknown but it is get number of individuals. Gastropod whorls were counted assumed they are for storage of a variety of food and non- to obtain minimum numbers. Austrovenus stutchburyi food items. (cockle) and Paphies australis (pipi) were the main species gathered, and are from a harbour mudfl at environment. In four of the six samples pipi were present in greater numbers than cockle (Table 1). Other species were collected U14/ Pit No. Length x width (m) Depth (m) in such small numbers that their presence could be acci- 1465A 49 0.42 x 0.37 0.46 dental. Th ese species include Turbo smaragdus (cat’s eye), 52 6.55 x 2.75 0.95 53 0.75 x 0.90 0.53 54 >3.35 x >2.10 0.70 55 >1.80 x 1.45 0.68 1465B 67 0.46 x 0.96 0.18 69 1.13 x 0.55 0.75

cockle pipi tuatua eyecat’s whelk other rock 75 3.35 x 1.60 0.54 U14/1465A F13 191 23 1 3 4 P 86 3.40 x 1.90 0.70 U14/1465B F67 686 2629 2 P 87 2.70 x 1.45 0.85 F74 154 336 3 3 P 88 3.60 x 1.25 0.40 F74 10 cm 164 75 P 89 ? x 1.05 F80 101 109 9 P 91 1.30 x 0.67 0.79 Area B 35 58 P 92 4.60 x 1.20

Table 1. Shellfi sh MNI for each feature sampled. Table 2. Pit dimensions U14/1465A and B

Louise Furey 7 CFG Heritage Ltd. 3 Lab. No. CRA BP cal AD 68.2% cal AD 95.4% Wk 20987 714 ± 44 1520–1650 1460–1690 2.5 Wk 20988 675 ± 41 1540–1680 1470–1760 2

1.5 Table 3. Summary of radiocarbon dates. width (m) 1 a major activity. Th e kumara would not have been carried 0.5 far so it can be assumed the gardens were on the fl at land in the vicinity of the sites. Th e limited evidence contrasts with 0 0123456 excavated sites in the western Bay of Plenty area: sites near length (m) the coast at Lynley Park at Omokoroa (excavated by Furey, CFG Heritage report in preparation), U14/3283 (Furey & Figure 7. Pit dimensions of U14/1465A shown as squares, Hudson 2008), and Carmichael Block at Bethlehem (exca- and U14/1465B shown as triangles. There is a spread of pit vated by Don Prince, report in preparation); and sites sizes in both sites, although neither site has a large number further inland including at Oropi (Campbell 2004a, b), of small bin pits with less than 1 m dimensions. Waimapu (Furey 2004), Ohauiti (Campbell 2004c, 2005, Campbell and Harris 2007) and Ruahihi Pa further up the Wairoa River (McFadgen and Sheppard 1984). Each Chronology of these sites were larger, had more storage pits of vary- Two shell samples were submitted to the Radiocarbon ing size and shape, and had a greater range of occupation Laboratory at the University of Waikato for dating. Pipi evidence. shells were used in each case. Th e sample from U14/1465A Recorded sites in the immediate area are few in number. was from Feature 13, a fi rescoop. Th e shell is considered Further up the ridge from U14/1465 there are four midden to be contemporary with the construction of the fi res- sites (U14/1461–4); the ridge to the east, intersected by coop, and the adjacent fi rescoops and cooking area. A Oliver Road, has fi ve recorded sites on the ridge crest result of 714 ± 44 BP was obtained (Wk20987). Pipi shell (U14/670–1, 885–7); the ridge to the west also has four sites from F74 on U14/1465B, also a fi rescoop, was also submit- (U14/1467–70). Th e majority are middens although sub- ted. A result of 675 ± 41 BP was returned (Wk20988). Th e surface features may include storage pits. Pits are present calibrated ages of both samples are set out in Table 3. Th e on the ridge to the east of the development area. One of Radiocarbon Dating Laboratory’s reports are appended to these sites (U14/885) has been investigated. Recorded as a this report. pit and terrace site on a ridge crest at Walden Lane, the Th e calibrated age ranges indicate the probability that remaining intact part of the site was excavated by Phil the shellfi sh sample died (i.e. was collected and eaten) Moore (2005). Seven storage pits, a terrace, cooking area within the calendar age range shown. Th e results cannot be and shell deposits were investigated. Two radiocarbon age interpreted that the site was occupied for the entire period: determinations were obtained on pipi shell. At 95% prob- in fact the archaeological evidence on both sites suggest ability, the sites were occupied sometime from mid-15th that each occupation was for a relatively short period of to early 17th. Another nearby investigation, U14/3221, was no more than a year, or two or three kumara harvesting also on the hills overlooking the Wairoa River (Campbell seasons at most. Broadly speaking, the sites were occupied 2008). Firescoops and shell midden were uncovered in var- sometime between the mid 15th to mid 17th centuries. ious places along the ridge crest, but no pits were observed. A single radiocarbon age determination on pipi shell indi- Discussion cates that like the other sites in this area, occupation was U14/1465A and B are typical of sites found in the Tauranga within the period mid 15th to mid 17th centuries century. Basin although they are smaller than those generally U14/1465A has at least two separate periods of activity. found closer to the coast. Given their inland location, at Th e storage pits and some of the postholes can be assigned the margin of diff erent geological zones, the expectation to the earlier occupation. Th e pits would have been in use was that the sites might refl ect this in some way. Both sites for at least one season, i.e., dug prior to kumara harvest in are small in size, having four and nine pits respectively and March/April, and the crop stored and used over the next an assortment of postholes and small cooking scoops. Th e fi ve to six months. Bin pit 53 which has been cut through limited range of features is a refl ection of the small size the wall and fi ll of Pit 52 must have been used at a later of the sites, and the limited use to which they were put. time, and the midden over both pits suggests the site was Storage of kumara (and possibly other crops) is obviously either occupied for several sequential seasons with shift - ing of activity areas on an annual basis, or three separate

8 Florence Lane occupations are represented. Th e main activity area is the References centre of the narrow ridgeline—the slope to the west had Campbell, M., 2004a. Excavation of site U14/1945, Oropi little evidence of occupation apart from several fi rescoops, Valley, Tauranga. Archaeology in New Zealand, 47(1): but the trench placed through the slope deposits indi- 50–61. cated that tephra soil, probably from pit digging, had been Campbell, M., 2004b. Archaeological investigations of site thrown out over the slope. Th ere was no cultural debris U14/3207, Richmond Park, Hollister Lane, Tauranga. such as shell or oven stones in these slope deposits which Unpublished eport to Hollister Lane Developments fi ts with the general evidence of the midden being later Ltd. than the pits. Campbell, M. 2004c. Archaeological investigations of site One of the diffi culties in archaeology is in determin- U14/3207, Richmond Park, Hollister Lane, Tauranga. ing contemporaneity of features, particularly in these situ- Unpublished report to Hollister Lane Developments ations where ploughing has destroyed the upper levels of Limited. stratigraphy. Aligned pits are generally taken to indicate Campbell, M. 2005. Archaeological investigations of contemporaneous use, or pits with distinctive similar fi lls sites U14/1972, U14/3218 and U14/3235, Rowesdale, are likely to have been fi lled in at the same time. Using Tauranga: fi nal report. Unpublished report to Connell these assumptions, pits 75 and 88 with the same align- Wagner Tauranga and Rowesdale Developments. ment are likely to be contemporary. Similarly pits 86 and Campbell, M. and J. Harris 2007. Archaeological inves- 87, while not aligned with one another (or with 75 and 88), tigations of site U14/3218, Rowesdale, Tauranga; have the same type of fi ll consisting of yellow-brown sub- Season II: fi nal report. Unpublished CFG Heritage Ltd soil and shell lenses. Features such as postholes and fi res- report to Connell Wagner Tauranga and Rowesdale coops, and midden, over both pairs of pits suggest there Developments. may have been two or three uses of the site, and certainly Campbell, M., 2008. Archaeological Investigation of Site over several seasons. Stakeholes are scattered to the east, U14/3221, 350 Wairoa Road, Te Puna: Final Report. on slightly sloping ground, separated from the midden Unpublished CFG Heritage Ltd report to the New deposit and fi rescoops which are likely to be contempo- Zealand Historic Places Trust and Murray and Margie rary with them. Although no patterns or alignments can Lloyd. be detected, these stakeholes probably anchored small Furey, L. 2004. Archaeological Excavations U14/1920 and temporary structures. U14/2193 Cheyne Rd, Tauranga. Unpublished report to Th e sites can be interpreted as places where harvested Victoria Tauranga Ltd. kumara, grown in the immediate area, were stored. Th ere Furey, L. and B. Hudson 2008. Archaeological is little evidence on U14/1465B of cooking – what is present Investigations at U14/3283 and U14/3284, Omokoroa is partly on the fi ll of storage pits but any earlier cooking Road, Omokoroa: Interim Report. Archaeology in New associated with the pits was not found, nor were there any Zealand, 51(4). dumps of shell midden. U14/1465A had a defi ned area of McFadgen, B.G. and R.A. Sheppard, 1984. Ruahihi Pa. cooking and midden rakeout, but again it was later than National Museum of New Zealand Bulletin, 22. the infi lled storage pits. Moore, P., 2005. Archaeological Investigation of Site Both of these sites present similar evidence to other U14/885, Te Puna, Tauranga. Report for Aldridge sites excavated in the Bay of Plenty, where disused storage Family Trust. Unpublished Peninsula Research report. pits are fi lled in relatively quickly aft er they ceased to be used for storage. Th ere is no evidence of organic staining Acknowledgements from collapsed or burnt roofs or roof support posts left in Th anks to excavators Ben Th orne, Paula Th orogood, the ground. Although the pits were not fully excavated, the Raylene Reihana-Ruka, Susan Strongman, Peter fi ll was generally indistinguishable in colour and texture Caldwell and Brian Sarsfi eld (Jacko). Th anks also to Alf from the surrounding natural subsoil, and without char- McCausland and the late Peter Rolleston of Pirirakau for coal inclusions or debris associated with the occupation. their cooperation. Th ese sites, together with sites excavated in the wider area, suggest gardening and occupation in this area from the mid-to-late 15th century. A similarly early chronology from Oropi area near Welcome Bay (Campbell 2004a,b; Furey 2004), suggests that forest clearance and gardening well inland from the coastal margins was widespread in the western Bay of Plenty from this time.

Louise Furey 9 CFG Heritage Ltd. Appendix 1. Features excavated on U14/1465A and U14/1465B lled lled x 280 x 120 100 18 by Cut ecks shell of throughout 600 x 600 x 100 ecks,soft . Circular, sloping sides 460 x 80 x 570 texture. x 90 130 x 160 lled with grey-brown ash. On sloping ground. x 80 130 x 130 . Fill darker than surrounding soil x 110 x 170 130 shellne fragments stone and oven one x 480 x 170 160 , with shell and ovenstones x 200 x 120 130 lled with conc. Burnt shell, merging in with deep nutty, charcoal.ne Concentration 460 stones oven of x 580 within 1465A 1465A 5 1465A 6 1465A 7 1465A Posthole 8 1465A Posthole 9 Depression Fill of shell midden and dark 10 loam y-b 1465A Posthole Fill of shell Fill fragments dark of loam y-b and fragment one and loam y-b obsidian of Firescoop 11 Firescoop Dark brown loam, soft Black loam, small fragments charcoal. of Sloping sides Dark loam, y-b fi with Posthole shell intersected the wall and earlier than scoop brown ash, Yellow soft shallow fi scoop x 750, 750 280 x 200 x 90 1465A 540 x 110 x 470 1465A 20 1465A 80 x 90 x 18 60 x 60 x 90 21 1465A 22 1465A Posthole x 350 x 190 190 23 Posthole 24 1465A Posthole Black loam Posthole Black loam 25 Firescoop loam, Grey y-b with shell, soft Dark grey soft y-b, Dark grey to black loam with burnt shell. Scoop within midden rakeout Posthole area. loam with Y-b shell fragments Classic saucer shaped scoop x 150 x 130 150 60 x 60 x 60 80 x 80 x 180 Cut by 13 Cut by 13 500 x 500 x 50 1465A 1465A 12 Posthole Dark loam y-b with fi Site U14/ U14/ Site Feature Feature type 1465A Description 1465A 1 2 1465A Layer Layer 3 1465A Dark grey brown topsoil with crumb texture. 80–120 mm in depth Layer brown mottled ash. Yellow Distinguished brown mottles by 4 and small pieces Firescoop Dark grey brown with shell surrounded dark by loam. Crushed cockle, pipi, Filled with pipi, cockle, ovenstone fragments. Sloping sides. Posthole 5 fi ‘cream Between of cake’. 50–80 mm depth, negligible in places. Features cut layer. this through fragment Firescoops ovenstone. into of dug midden and under midden. 1465A Depth mm. to 100 13 1465A 1465A 14 1465A Firescoop 15 1465A 16 Dark brownish Dimensions black loam mm with crushed shell midden and charcoal Posthole fragments. Association 17 1465A Posthole Firescoop Shell midden in dark matrix. y-b 18 1465A Firescoop Dark matrix y-b with shell fragments Dark brownish black loam, soft 19 Irregular scoop fi Firescoop Black nutty loam with shell some and small stones. oven Within Posthole area of midden rakeout. Dark grey with y-b small shell fl crumbly dark loam g-b which has fl Irregular shape and base 80 x 130 x 100 90 x 90 x 120 Cuts 20, 21 x 180 950 x 750 Cuts 19

10 Florence Lane related to 43, 44, 45, 46 39–46,53 x 700 rescoops 580 x 430 x 120 ll, hard in places over ash with whole and ecks surface on ecks surface on x 900 750 x 530 x >2100 >3350 Cuts 52 49 by Cut ll x 90 x 120 110 bottomedat x 200 x 70 210 fi ll x 200 x 170 160 90 x 90 x 110. rescoops 450 x 600 x 100 Cuts 42, ll, charcoalfl ll, charcoalfl ll, ll 50 x 50 x >100 ll, shell no x 950 x 2750 6550 4,5, by Cut x 110 x 120 140 41 by Cut ll x 680 x 1450 >1800 11 by Cut tephra with whole pipi and cockle 240 x 300 x 500 ll surface on to depth 200 of mm then black fi y-b ne soft ne 1465A 42 Posthole Black loam, soft loam, 42 1465A Posthole Black 1465A 1465A 43 1465A 44 1465A 45 1465A Firescoop 46 1465A Firescoop 47 Shell and black loam1465A stones. and a series of One oven intercutting of fi Firescoop 48 Shell and black loam, irregular base Firescoop 49 Shell and black loam.1465A Scoop Shell 1465A Firescoop 50 1465A Bin pit 51 Black loam. On slope Black loam with shell. Shallow, fl 52 1465A Posthole ash level g-b to depth of Upper 90 soft of mm, y-b 1465A over Firescoop 53 Pit loam Grey y-b 54 Black loam 1465A Pit 55 Pit Mottled dark fi y-b fragmented Pit shell scattered through Mottled yellow brown fi Mottled yellow brown fi 500 x 460 x 120 Mottled fi y-b Cuts 43 400 x 300 x 60 43 by Cut x 100 x 320 470 300 x 400 x 100 x 460 420 x 370 x 120 x 420 160 850 x 900 x 120 1465A 1465A 1465A 35 1465A 36 1465A 37 1465A Posthole 38 1465A Posthole 39 1465A Posthole Grey brown loose fi 40 Posthole Shell dark over grey brown fi 41 Firescoop Shell Natural Shell Black loam. Predates shell midden and midden by covered Firescoop Shell fragments. Probably a root hole Shell. Series 5 intercut fi of 780 x 430 x 60 overlain 3 by x 350 x 350 550 60 x 60 x 100 x 100 x 870 70 Site U14/ U14/ Site Feature Feature type 1465A Description 1465A 27 1465A 28 1465A fi 29 1465A Firescoop 30 32 1465A Posthole Posthole Shell 31 Dark with g-b burntfi shell. charcoal No Posthole 1465A Posthole Shell. Found in base of scoop 27 1465A Posthole Dark grey brown loam, soft 33 Shell 34 Shell fragments Posthole Posthole Shell Shell dark over grey brown soft 900 x 850 x 150 Cuts 28 Dimensions mm x 80 70 x 90 Association Cut by 27 20 x 30 x 90 on edge of 13 x 50 70 x 60 50 x 50 x 80

Louise Furey 11 CFG Heritage Ltd. 84 84 oor. oor. nite nite innd plan view but oor of pit. Top of ph wider ph of than Top pit. of oor cult to fi cultto compact, greyner then y-b, ll fi ll x 220 x 130 120 . Central alignment in fl shell. A few fragments the at base x 180 x 140 140 ll, soft ll, er looser mixed with y-b lighter mottles level hard onto fl ll 50 x 50 x 80 83, 81, 75, shell, a few fragments x 300 x 170 160 90 x 130 x 120 x 90 x 100 100 x 360 x 100 180 x 180 x 110 170 ne x 120 x 170 170 ll and shell fragments, post mould had small cockles in dark post g-b x 240 (mould) x 110 110 under outside oor or mottled edge vertical, Upper not layer. sloping in for original topsoil. U14/ Site Feature Feature type 1465A Description original 56 1465B 1465B Layer soft 601465B 63 61 1465B Posthole Brown, 62 Posthole Present western overlying site of side features, grey-brown ashy loam. Possibly Posthole Posthole Dark loam, y-b soft Grey y-b Greysoft y-b, Dimensions mm Association x 80 x 100 100 1465B 64 Posthole Brown, soft soft 641465B Posthole Brown, 65 1465B Posthole Brown, 1465B 66 Posthole Mixed brown and soft y-b, 1465B 1465B 1465B 67 1465B 68 69 fl Pit Posthole 1465B Pit 1465B Black with shell 70 1465B Whole shell, little Very soil matrix. burnt No shell, no ovenstones 71 1465B 72 1465B Posthole Black with charcoal, shell and ovenstones. Capping grey mottled of y-b very 73 Posthole 74 Posthole Shell Posthole Greysoft y-b, 460 x 960 x 180 Firescoop y-b Grey, mixed top of pit and layer over extending further grey out from with y-b, sites. posthole No shell fragments in Steep sided, fragmented midden and loam visible not surface on as a defi upper 30 cm. Sump mm 120 deep x 300 mm wide across pit. of end one x 750 x 550 1130 68 by Cut x 170 x 100 100 Cuts 69 200 x 200 x 360 x 120 x 100 100 90 x 90 x 100 1465B 75 Pit mottled, slightly Y-b darker than natural, diffi 1465B feature 82 Posthole same as pit fi Y-b 900 x 800 x 400 Cuts 91

underneath a soft a underneath evident 200 in fi section. mm pit of Top base, looks like post pulled out 250 x 250 x 460 82, 75, 83, fi 1465B 77 1465B Posthole mixed 76 1465B 1465B Posthole 78 81 Grey y-b Posthole Posthole Greyfi y-b, same as pit fi Y-b, excavated. Only 1/3 hole cut into yellow natural 40 x 40 x 95 x 540 x 1600 3350 250 x 200 (hole) x 270 84 83, 81,

12 Florence Lane 84 83 91 x 850 2700 x 1450 ll surface. excavated. Not x ? 1.05 ll 86 pit of 450 x 400 x 160 cuts 86 ll below. 1300 x 790 x 670 cut 74 by 88 pit oor of 88 pit oor of 88 pit oor of 88 pit oor of 75 pit oor of 75 pit oor of 75 pit oor of 75 pit oor of 4 x 5 x 17 x 13 13 4 x 9 6 x 14 Floor pit x 10 x 12 21 Floor pit 7 x 3 x 3 x 12 12 Floor pit Floor pit Floor pit x 23 45 21 Floor pit Floor pit Floor pit ll in fl ll in fl ll in fl ll in fl ll in fl ll in fl ll in fl ll in fl 50 x 50 x 90 82, 81, 75, x 300 x 350 350 82, 81, 75, fi y-b fi y-b fi y-b fi y-b fi y-b fi y-b fi y-b fi y-b oor, y-b clean y-b fi oor, Site U14/ U14/ Site soft Feature Feature type 83 1465B Posthole soft Y-b, Description 841465B Posthole y-b, 1465B 1465B 85 86 Firescoop 1465B Pit Shell fragments and black loam. Dug into fi 87 Pit Lenses shell, of mottled large lighter lumps Darker of y-b, colour. mottling to Mottled with y-b shell some lenses and black lenses, soft 30 cm scraped below surface. part Upper wall of sloping inwards western on small (only side test in found Floor trench so dug possibly representative). not test pit. Posthole arrangement not known Dimensions mm Association 3400 x 1900 x 700 Cut by 85 1465B 1465B 88 1465B 1465B 89 1465B Pit 90 fl 91 1465B Pit Firescoop 92 upper 30 cm Pit grey with y-b black some Y-b, splodges surface. on Shell also Black loam and concentrated crushed shell Pit Orange brown mottled. Adze fragment fi found on surface stained Y-b, mm black, thick) shell in lens (100 black 200 mm above surface on Mottled slightly y-b, darker than surrounding sub-soil 450 x 600 x 350 4650 x ? x 1250 Cuts 86 cut 74, by 3600 x 400 x 1250 1465B 93 1465B Posthole Soft 94 1465B Posthole Soft 95 1465B Posthole Soft 96 1465B Posthole Soft 97 1465B Posthole Soft 98 1465B Posthole Soft 99 1465B Posthole Soft Posthole 100 1465B Soft

Louise Furey 13 CFG Heritage Ltd. The University of Waikato Radiocarbon Dating Laboratory

Private Bag 3105 Hamilton, New Zealand. Fax +64 7 838 4192 Ph +64 7 838 4278 email [email protected] Head: Dr Alan Hogg

Report on Radiocarbon Age Determination for Wk- 20987

Submitter L Furey Submitter's Code U14/1465A F13 Site & Location Te Puna, Tauranga, New Zealand

Sample Material Austrovenus stutchburyi Physical Pretreatment Surfaces cleaned. Washed in an ultrasonic bath. Tested for recrystallization: aragonite.

Chemical Pretreatment Sample acid washed using 2 M dil. HCl for 100 seconds, rinsed and dried.

14 d C -35.2 ± 5.3 ‰ 13 δ C 1.2 ± 0.2 ‰ 14 D C -85.1 ± 5.0 ‰ % Modern 91.5 ± 0.5 % Result 714 ± 44 BP

Comments

14/5/07

• Result is Conventional Age or % Modern as per Stuiver and Polach, 1977, Radiocarbon 19, 355-363. This is based on the Libby half-life of 5568 yr with correction for isotopic fractionation applied. This age is normally quoted in publications and must include the appropriate error term and Wk number.

• Quoted errors are 1 standard deviation due to counting statistics multiplied by an experimentally determined Laboratory Error Multiplier of 1 .

• The isotopic fractionation, δ 13 C , is expressed as ‰ wrt PDB.

• Results are reported as % Modern when the conventional age is younger than 200 yr BP. The University of Waikato Radiocarbon Dating Laboratory

Private Bag 3105 Hamilton, New Zealand. Fax +64 7 838 4192 Ph +64 7 838 4278 email [email protected] Head: Dr Alan Hogg

Report on Radiocarbon Age Determination for Wk- 20988

Submitter L Furey Submitter's Code U14/1465B F74 Site & Location Te Puna, Tauranga, New Zealand

Sample Material Austrovenus stutchburyi Physical Pretreatment Surfaces cleaned. Washed in an ultrasonic bath. Tested for recrystallization: aragonite.

Chemical Pretreatment Sample acid washed using 2 M dil. HCl for 100 seconds, rinsed and dried.

14 d C -30.8 ± 4.9 ‰ 13 δ C 1.1 ± 0.2 ‰ 14 D C -80.6 ± 4.6 ‰ % Modern 91.9 ± 0.5 % Result 675 ± 41 BP

Comments

14/5/07

• Quoted errors are 1 standard deviation due to counting statistics multiplied by an experimentally determined Laboratory Error Multiplier of 1 .

• The isotopic fractionation, δ 13 C , is expressed as ‰ wrt PDB.

• Results are reported as % Modern when the conventional age is younger than 200 yr BP. Marine data from Hughen et al (2004);Delta_R -7±45;OxCal v3.10 Bronk Ramsey (2005); cub r:5 sd:12 prob usp[chron] 1200BP Wk20987 : 714±44BP 68.2% probability 1520AD (68.2%) 1650AD 1000BP 95.4% probability 1460AD (95.4%) 1690AD

800BP

600BP Radiocarbon determination

400BP

1200CalAD 1400CalAD 1600CalAD 1800CalAD 2000CalAD Calibrated date Marine data from Hughen et al (2004);Delta_R -7±45;OxCal v3.10 Bronk Ramsey (2005); cub r:5 sd:12 prob usp[chron] 1200BP Wk20988 : 675±41BP 68.2% probability 1540AD (68.2%) 1680AD 1000BP 95.4% probability 1470AD (95.4%) 1760AD

800BP

600BP Radiocarbon determination

400BP

1200CalAD 1400CalAD 1600CalAD 1800CalAD 2000CalAD Calibrated date