radiocarbono

Oceania Publications, University of Sydney

Radiocarbon Dating Fish Bone from the Houhora Archaeological Site, New ZealandAuthor(s): Fiona PetcheySource: Archaeology in Oceania, Vol. 35, No. 3 (Oct., 2000), pp. 104-114Published by: Wiley on behalf of Oceania Publications, University of SydneyStable URL: http://www.jstor.org/stable/40387161 .Accessed: 16/10/2014 20:38

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Wiley and Oceania Publications, University of Sydney are collaborating with JSTOR to digitize, preserve andextend access to Archaeology in Oceania.

http://www.jstor.org

This content downloaded from 201.148.81.39 on Thu, 16 Oct 2014 20:38:38 PMAll use subject to JSTOR Terms and Conditions

ArchaeoL Oceania 35 (2000) 104-15

Radiocarbon dating fish bone from the Houhora archaeological site, New Zealand

FIONA PETCHEY

Abstract

This paper presents preliminary radiocarbon results of snapper bone gelatin from the Archaic occupation at Houhora, Northland, New Zealand. It is part of a larger investigation into dating snapper (Pagrus auratus) and barracouta (Thyrsites atun) bone from New Zealand archaeological sites (Petchey 1998). A range of ana-

lytical techniques is applied and it is concluded that the Houhora

snapper bone is suitable for radiocarbon dating as the bone is well-

preserved with low levels of contamination. Four bone gelatin 14C determinations of snapper are compared to determinations of other materials from the same context, including three previously unpublished results of shell and charcoal. The results suggest that

snapper bone gelatin has a minimal inbuilt age, a reservoir effect which is comparable to shell, and will yield reliable 14C determinations when collagen degradation and contamination is minimal. A total of twelve acceptable radiocarbon results places the Archaic occupation at Houhora during the early to mid 14th century AD.

Bone has played an important role in the development of radiocarbon analysis in New Zealand and until recently, moa bone was the preferred sample type when dating moa hunting contexts (e.g. Trotter and McCulloch 1984; Caughley 1988). Discrepancies in bone determinations have, however, resulted in considerable uncertainty regarding their accuracy. A range of explanations have been put forward to account for this variation, including carbon fractionation due to dietary preferences, inade- quate sample pretreatment, varied radiocarbon standards, diagenetic effects and/or contamination (e.g. Rafter 1978:138; Anderson 1991:777, 779). Recently, bone "collagen" (specifically that fraction remaining following dcalcification in acid) has been considered to be of low priority as a datable material (Higham 1993:97). Gelatinisation (purification of the sample by denaturing the collagen molecule in weakly acidic hot water) appears, however, to be an acceptable pretreatment improving on the old collagen determinations (Petchey 1999).

A range of bone types have been radiocarbon dated in New Zealand with varying degrees of success, including dog, rat, human, seal, fish, and bird. Unfortunately, the reliability of these different species for 14C analysis is largely unknown. In particular, fish and seal bone deter-

Radiocarbon Dating Laboratory, School of Science and Technology, University of Waikato, Private Bag 3105, Hamilton, New Zealand.

minations were excluded from Anderson's (1991:768) list of suitable radiocarbon sample types due to uncer- tainties with the marine reservoir effect and uptake of 14C (see also Grant-Taylor (1974:160) and Law (1981:234)). The ability to date fish bone has, however, a number of advantages for archaeologists. First, fish bone is found widely in archaeological deposits throughout the prehistoric sequence, in both the North and South Islands of New Zealand. Second, it is directly correlated with an identifiable archaeological event, in this case fishing.

In this paper, radiocarbon results from the site of Houhora (New Zealand Archaeological Association site number N03/59) are presented, and the reliability of

Figure 1 : North Island of New Zealand showing the location of Houhora.

104


Figure 2: Plan of Houhora showing location of radiocarbon samples from Archaic layers (adapted from Shawcross 1972:604, figure 14.1).

snapper bone as a dating medium evaluated by comparison with new and existing radiocarbon determinations. Particular attention is given to the 14C content, inbuilt age, contamination and degradation of the fish bone in question.

Houhora

Houhora is located at the entrance of Houhora Harbour, to the north of the North Island (Figure 1). The site of Houhora consists of a large coastal midden containing remains of moa, seal and fish, and associated artefacts typical of early tropical East Polynesian assemblages (Anderson and Wallace 1993:5).

Two major excavations were undertaken at the site. The main excavation (Figure 2) took place between November 1965 and January 1966, while squares A6, A7, half of A8 and B6 were excavated in 1972 (L. Furey, pers. comm. May 1997). The most detailed published plan of the Houhora excavations is given in Shawcross (1972). Shawcross' plan (Figure 2) reveals a concentration of occupation debris at the centre of the excavation. Ovens surround the perimeter with large numbers to the

south west. To the north west a high concentration of fish cranial elements were interpreted as debris from food processing (Shawcross 1972:606, see also Nichol 1988). The only contemporary account of stratigraphy is given in Roe (1969:figure 3). Nichol (1988:201) reanalysed the stratigraphy of Houhora using bag labels from samples held in storage and identified four main layers (2a, 2b, 2c and 3). The widespread distribution of these layers gen- erally matched Roe's (1969:figure 3) account of stratigraphy, except that Layer 2c was missing. The lower Archaic layers (Layer 2b and Layers 3a, b and c) con- tained remains of cooking and other subsistence activi- ties. A later agricultural layer (Layer 2a) cut into Layer 2b (Roe 1969:14-20).

Three charcoal radiocarbon determinations (NZ-914, NZ-915 and NZ-916) were obtained during excavations in 1965-1966. These samples were not identified to species and are, therefore considered unreliable (McFadgen 1982:384). Two moa bone samples (NZ- 5007 and NZ-5008) were also collected from Layers 2c and 3b, but submitted for dating several years later (Millener 1981). These and other moa bone "collagen" determinations have also since been considered to be of questionable reliability (Anderson 1991:779).

105


Consequently, in order to evaluate these original determinations Anderson and Wallace (1993:10, table 1) submitted more samples from the Archaic layers, including four charcoal samples, two of which were removed from latex pulls taken at the time of the 1965-1966 excavation, and two from bags collected in 1965 and 1972. A single shell determination (NZ-7920) was also collected from Pull 2.

On the basis of ten radiocarbon results (including the "questionable" moa bone collagen and unidentified charcoal results), Anderson and Wallace (1993:14) suggested that occupation began some time in the 13th century AD. This was consistent with a previous estimate by Roe (1969:36) based on artefact forms. Anderson and Wallace (1993:14) also suggested that occupation was short lived, with no substantial period of time elapsing between the lowest layers (layers 2b and 3).

Experimental

Fish bone samples

Fish bone samples stored in bags labeled Houhora C8, "Fossick A" were obtained from the Auckland University Anthropology Department (see Petchey 1998). The specific context of this material is unknown (L. Furey, pers. comm. May 1997), though it seems likely that the fish bone came from the two lower occupation layers (2b and 3) which are indistinguishable in age (Anderson and Wallace 1993). This is supported by identification of a square number (C8) which strongly suggests removal from an in situ deposit, and it is proba- ble, as suggested by Nichol (1988:195-196), that this fish bone sample was removed from baulks by archaeologists following completion of the excavation. Moreover, the Archaic layers near square C8 were rich in fish bone, with a particularly high frequency of snapper head parts per square metre (Shawcross 1972:604, figure 14.1). Finally, in situ middens from the more recent layer (Layer 2a) did not have the bone typically associated with the older middens (Roe 1969:78).

Closer investigation of snapper diet and habitat as well as examination of the preservation of the archaeological fish bone were undertaken prior to radiocarbon analysis in order to evaluate the impact of depleted 14C, inbuilt age, contamination and degradation on the radiocarbon determination.

Depleted 14C Concentration

A depleted 14C concentration may be introduced into fish by contact with different water masses, areas of upwelled deep ocean water, or by feeding at depth. None of these factors should, however, significantly influence the Houhora snapper bone. First, a distinct snapper popu- lation has been identified in the East Cape/Bay of Plenty region (Paul 1992:11) where there is no significant upwelling (Heath 1985:87-88). In addition, snapper

occupy and feed in surface waters above 200 m (Paul 1992:5). They should not, therefore, be seriously affected by the depleted 14C gradient which occurs below ca. 200 m around New Zealand (Lassey, Manning and O'Brien 1990:124). These hypotheses are supported by radiocarbon results of otoliths from modern and historic snapper collected in the East Cape/Bay of Plenty region which compare favourably with surface water 14C values, modern shellfish samples and historic shell data from around New Zealand (see Higham and Hogg 1995).

Inbuilt Age

Snapper may reach ages in excess of 60 years (Paul 1992:7). The age range of the Houhora snapper sample was, therefore, determined in order to estimate the degree of inbuilt error introduced into the radiocarbon determinations. First, measurements were taken of specific cranial bones in the archaeological collection (Appendix 1), and their fork length calculated using the technique outlined by Leach and Boocock (1995). The values obtained (Appendix 2) were then substituted into average age-length (fork length) growth curves which have been developed for snapper from the east coast of the North Island of New Zealand (Paul 1992:10). From this assessment it was determined that snapper from Houhora, square C8, {n = 171) range between approximately 280 to 819 mm in length, with a mode of 490 mm and a mean of 493 mm. This equates to an age range of ca. 4 to >40 years, a mode of 27 years and a mean of 28 years. Because complete replacement of mature collagen takes around 10 years (Chisholm 1989:20-21), the influence of this inbuilt age is significantly reduced, and is less than the standard error inherent in charcoal radiocarbon determinations (McFadgen, Knox and Cole 1994).

Preservation and Contamination

Collagen is subject to alteration following the death of the animal via a combination of chemical, physical and biological processes. The decay process involves the breakdown of soft tissue, which is manifest as a loss of collagen and a corresponding increase in susceptibility to humic and non-humic contaminants which may alter the apparent age of the sample. It is recommended, therefore, that prior to 14C analysis, contamination levels and the amount of remaining collagen are assessed. For this research physical characteristics, percent nitrogen in whole bone, isotopie analysis, infrared analysis and yield information have been collected. The methods used are outlined in Petchey (1998).

A. Physical Appearance. Bone from Houhora had a soft exterior indicative of a moderate degree of alteration. Slight discolouration to the bone resulted in a dull orange colour (7.5YR 7/4). A few burnt fish bone elements were noted, but were not selected for dating.

B. Percent Nitrogen: Total Collagen Content. Protein

106


Sample no. 613C %C 15 % Atomic C/N

Snapper modern gel 4 -11.1 to -12.2 36.5 to 41.5 12.8 to 14.2 13.8 to 15.4 2.96 to 2.97 HC8 gelatin 1 -13.6 42.4 13.3 15.4 2.96 HC8 residue 1 -13.6 37.1 14.1 10.8 2.96 HC8humics 1 -17.3 41.8 14.4 8.4 2.94 Table 1 : Stable carbon and nitrogen measurementst and atomic C/N ratios for gelatin, humic and residue fractions. Instrument error for nitrogen analyses is 1. Sample error for carbon (combination of machine error plus sample het- erogeneity) is 0.5. 613C and 15 values are given in %c versus VPDB and AIR, respectively.

is the only component in modern bone that contains nitrogen, and collagen makes up 90% of the bone nitrogen (Garlick 1969:503-504). The amount of collagen remaining in the archaeological samples can, therefore be estimated by comparing the percent nitrogen (%N) in the archaeological bone to modern snapper bone values.

The bulk %N value for archaeological fish bone is 2.11%. This suggests that just over half of the original protein remains compared to modern fish bone which has a %N value of around 4.12 (n=2). These bulk N% estimates assign the collection to the Class II (very well to well-preserved) preservation state, as defined by Stafford, Brendel and Duhamel (1988:2258, table 1) (i.e. 0.6 to 3.5% N).

C. Pretreatment. Petchey (1998) has demonstrated that well-preserved "Class bones, less than 1000 years old with greater than 40% extractable collagen, should give reliable 14C determinations when gelatinised after a NaOH wash. The following pretreatment was therefore adopted: First, the sample was cleaned and ground, decal- cified in 2% w/v HC1 at 4C for 4 days, then rinsed with distilled water ("acid insoluble collagen"). This removes bone hydroxyapatite, secondary carbonates, some acid- insoluble fractions such as collagen breakdown products, and some humic contaminants (Hedges and van Klinken 1992:286). This "acid insoluble collagen" was treated with 2% w/v NaOH for 1 hour to remove humic acids, washed with distilled water, rinsed with HC1, and washed again. The sample was then gelatinised by heating in weakly acidic water (pH = 3, 90C for 4 hours). Gelatinisation denatures the triple-helical collagen molecule and removes acid-insoluble impurities, insoluble collagen and degraded collagen gelatin residue (Law and Hedges 1989:250). Lastly, the gelatin was freeze-dried.

Van Klinken and Hedges (1995:268) have noted that around 8% contamination can remain following the gelatinisation procedure. In a sample of 900 BP this equates to a possible error of 80 years (an age of 833 BP) if contaminated by modern carbon. The influence of old carbon would have a greater impact, but the chances of contamination by depleted 14C are considered to be minimal (cf. Higham 1993). Combining the gelatin pretreatment with a NaOH wash should, however, remove greater than 92% of any contaminants. Moreover, contamination at the 5 to 10% level can often be recognised analytically when the

sample is of "good preservation" (i.e. >20% original collagen remaining) (Hedges and van Klinken 1992:284). Houhora fishbone preservation was therefore assessed using a combination of isotopie, yield information and infrared analysis of collagen and gelatin fractions.

D. Pretreatment Yield. Crude gelatin yields were measured as a percentage with respect to bone. Interpretation of the preservation state of the bone (i.e. "good", "poor" or "non-collagenous") was based on recommendations by Hedges and van Klinken (1992:284) with the excep- tion that the amount of extractable protein was adapted according to the yields obtained for modern fish bone following the pretreatment described above. Using this method it was calculated that 18 g of protein could be extracted per 100 g of modern bone (i.e. 100% extractable protein). To avoid variations caused by incomplete dryness of the freeze-dried gelatin, this value ("extractable protein") was corrected according to the yield of CO2 obtained following combustion of the sample. Losses incurred during pretreatment are assumed to be constant.

The crude gelatin yields obtained were uniform at 6.71 to 6.94 wt%. This equates to approximately 39% extractable protein remaining, and indicates that the snapper bone is of "good preservation" (i.e. >20% of the original collagen extracted). The yields of insoluble residue after each gelatinisation were also low (0.96-1.29%).

E. Isotopie Analysis. Stable isotope values (i.e. 813C and 15) were obtained for gelatin, humic acid and gelain residue fractions. These results were compared to modern snapper standards (Table 1). Stable isotope results on the gelatin (813C = -13.6%o; 15 = 13.3%o) are equivalent to modern snapper gelatin. Similarly, there is no noticeable contamination present in the isotopie analysis of the residue (813C = -13.6%o), though the humic acid 813C is depleted by comparison (613C = -17.3%o) (Table 1).

A C/N atomic ratio was also used to detect contamination in bone. Values of 2.9-3.6 for gelatinous extracts of bone are thought to indicate "good prehistoric collagen" (DeNiro and Weiner 1988). Unfortunately, C/N atomic ratios have been shown to only be successful when the bone is severely degraded (van Klinken and Hedges, 1995) and some variation in these acceptable limits have

107


Figure 3: Infrared spectrograms of acid-insoluble collagen (HC8HC1) and gelatin (HC8gel) from Houhora snapper. Individual spectra were obtained using 16 acquisitions before Fourier transform, at a spectral

resolution of 4 cm"1, using the empty chamber as the background reference spectrum.

been reported (e.g. van Klinken 1999). For the archaeological snapper samples, both the gelatin and contami- nant fractions shown in Table 1, fall within "good prehistoric collagen" levels and are comparable to modern snapper bone gelatin values.

Similarly, the percent carbon (9cC) or nitrogen (%N) in gelatin can indicate gross contamination by either car-

Lab No. Species Lifespan (McFadgen, Knox and Cole 1994:224).

NZ-914 cf. Vitexlucens Metrosideros excelsa A , ^ All A , long: >300 ^ yrs Lagarostrobus colensoi Prumnopitys taxifolia

NZ-916 cf. Vitexlucens Metrosideros excelsa Lagarostrobus colensoi , _ All , long: >300

_ yrs Prumnopitys taxifolia

Prumnopitys ferruginea Agathis australis

My r sine australis Short:

Figure 4: Bell plots of calibrated pooled radiocarbon ages for Archaic layers,

Houhora. Error bars denote 1 and 2 deviations.

Charcoal and Shell Samples

Two additional shell samples (Wk-5034 and Wk-5035) and a charcoal sample (Wk-5485) were selected for comparison with the fish bone determinations. Charcoal was obtained from archived material belonging to two of the original radiocarbon determinations (NZ-914 and NZ- 916). Two pieces of the short lived My r sine australis were removed from NZ-916 for radiocarbon analysis (Wk-5485) (Table 2). Shell came from two bags; the first (Wk-5035), a sample of Lunella smaragda labeled Houhora E7, L2b, SE corner, shell; and the second (Wk- 5034), a sample of Austrovenus stutchburyi, labeled Houhora D6, L2b, 2nd Hangi. Following Roe's (1969) account, the second of these two samples can be prove- nanced to a large oven in squares C6, C7, D6 and D7 (Figure 2).

Radiocarbon Analysis

Radiocarbon determinations were measured at the University of Waikato Radiocarbon Dating Laboratory according to the procedures outlined by Higham and Hogg (1997). All age estimates were calibrated using the program OxCal v3.3 (Bronk-Ramsey 1999). Because the production and movement of 14C between the atmos- phere and ocean is variable the conventional radiocarbon age (CRA) for each sample needs to be calibrated in order to derive a calendar age. Terrestrial samples were, therefore calibrated using the terrestrial calibration curve

of Stuiver et al (1998), with 275 years subtracted from the CRA to account for the southern hemisphere offset in 14C (McCormac et al 1998). Both the shell and fish results were calibrated using the marine curve of Stuiver, Reimer and Braziunas (1998) with the reservoir correc- tion for New Zealand (AR) set at -2515 years in order to adjust for regional oceanic variation in 14C (Higham and Hogg 1995).

The calibrated 14C results are shown in Figure 4 and Table 3. The prefixes 'NZ' and 'NZA' refer to ages calculated at the Rafter Radiocarbon Laboratory, Institute for Geological and Nuclear Sciences Ltd. (IGNS), and 'Wk' to those measured at the University of Waikato Radiocarbon Dating Laboratory. These results were evaluated using OxCal combine probabilities calculations (Bronk-Ramsey 1999). Using this method, dates from the same sample are combined prior to calibration using the R Combine method (chi square test). All other 14C determinations are calibrated, combined, and then assessed in the light of the combined data. The agreement (A) between these two distributions (the "agreement index") should not fall below 60.0% (

Lab No. Provenance Material Species 813C%0 CRA Cal 68% (Batch) (BP) (AD)

NZ-914 Layer 2b, Square G6 Charcoal Not identified -25+ 69749 -

NZ-915 Layer 3b, Square E4 Charcoal Not identified -251" 56361 -

NZ-916 Layer 3b, Square E3 Charcoal Not identified -25 t 77561 -

NZA-2436 Base of Layer 2b, Charcoal Pittosporum sp. (20%) Square E6. Pull 2. Dodonaea viscosa (40%)

Leptospermwn scoparium (20%) Beilschmiedia taraire twig (20%) -26.2 63286 1293-1417

NZA-2437 Layer 2b, Square C8. Charcoal Olearia sp. (25%) Pull 4. Beilschmiedia taraire twig (25%)

Pseudopanax sp. (25%) Leptospermum scoparium (25%) -26.3 77487 1217-1299

NZA-2438 Layer 3, Square A7. Charcoal Coprosma sp. (36%) -25.2 72786 1261-1322 Bagged sample from Pittosporum sp. (43%) 1350-1390 1972 excavation. Brachy glottis repanda (7%)

Hebe sp. (24%) Wk-5485 Layer 3b, Square E3. Charcoal Myrsine australis (100%) -25.5 64040 1300-1334

1338-1373 1378-1400

NZ-5007 Layer 2c, Square D9. "Collagen" Anomalopteryx didiformis -21.1 56356 1328-1344 Euryapteryx curtus 1394-1435 Pachyornis mappini

NZ-5008 Layer 3b, Square DIO. "Collagen" Dinornis struthoides -22.5 58546 1326-1348 Anomalopteryx didiformis 1391-1421 Euryapteryx curtus Pachyornis mappini

NZ-7920 Layer 2b, Square E6. Shell Austrovenus stutchburyi 0.9 812+37 1451-1502 Pull 2 Paphies australis

Wk-5034 Layer 2b, Square D6. Shell Lunella smaragda (100%) 1.8 96040 1332-1420 Wk-5035 Layer 2b, Square E7. Shell Austrovenus stutchburyi (100%) 0.7 106045 1281-1327

Wk-4920 Layer 2b/3, Square C8. Gelatin Pagrus auratus (100%) -13.6 101040 1306-1393 Wk-4921 Layer 2b/3, Square C8. Gelatin Pagrus auratus (100%) -14.2 100040 1310-1399 Wk-4968 Layer 2b/3, Square C8. Gelatin Pagrus auratus (100%) -14.4 95040 1337-1426 Wk-4969 Layer 2b/3, Square C8. Gelatin Pagrus auratus (100%) -14.8 105040 1289-1330

Table 3: Radiocarbon determinations from Houhora: Archaic layers. See Anderson and Wallace (1993). Pull loca- tions after Anderson and Wallace (1993:11) and L. Furey (pers. comm. May 1997). t 813C assumed, not measured.

Results

A total of three shell determinations have been obtained from the Archaic layers at Houhora. NZ-7920 was anom- alous and was considered, at the time, to be incorrect for some "sample constituent or technical reasons which are not yet understood" (Anderson and Wallace 1993:12). Recent research (Higham and Hogg 1995) indicates,

however, that Austrovenus stutchburyi and Paphies australis are reliable species for radiocarbon analysis. It is, therefore, suggested that "retouching" of the pulls may have contaminated this sample from Pull 2, as indicated by Coster in correspondence with Anderson and Wallace (1993:11). This places doubt on other samples removed from the latex pulls (see below). NZ-7920 is excluded from further analysis. Two additional shell 14C determi-

110


nations were measured (Wk-5034 and Wk-5035). These have an overall agreement index of 72.4% (An = 50.0%, = 2) and a reservoir corrected calibrated age of AD 1317-1381 at 1.

All three original charcoal samples (NZ-914, NZ-915 and NZ-916) were unidentified and are likely to have been subject to in-built age, which can result in errors of 300 years or more (McFadgen, Knox and Cole 1994:224). Analysis of charcoal from NZ-914 and NZ- 916 (Table 2) confirms that they are almost completely composed of long-lived species (R. Wallace, pers. comm. May 1997). NZ-914, NZ-915 and NZ-916 are, therefore excluded from the final analysis. Of Anderson and Wallace's (1993) four charcoal samples, one of the bagged charcoal samples (NZ-7921), has since been shown to have come from a nearby site called Houhora Terrace excavated in 1972 (L. Furey, pers. comm. May 1997) and is not included in this discussion. Two charcoal samples were removed from latex pulls and may be suspect (NZA-2437 and NZA-2436). They are, however, in agreement with the remaining charcoal radiocarbon determinations, NZA-2438 and Wk-5485, and have an overall agreement of 98.1% (An = 35.4%, = 4). These four charcoal determinations give a calibrated age of AD 1298-1318 and 1352-1388 at 1.

Although the moa bone "collagen" samples are in agreement [Aoverall = 121.3% (An = 50.0%, = 2), giv- ing a calibrated age of AD 1328-1345 and 1394-1424 at 1, the combined moa "collagen" determinations are in poor agreement with other sample types [A = 56.9% (

locating stored radiocarbon samples at IGNS. Dr. Tom Higham (Waikato Radiocarbon Dating Laboratory, University of Waikato) commented on draft copies of this paper. Louise Furey (Department of Anthropology, University of Auckland) provided information on sample provenance. Dr. A. Hogg (Waikato Radiocarbon Dating Laboratory) and Dr. F. Leach (Archaeozoology Laboratory, Museum of New Zealand, Te Papa Tongarewa) provided valuable discussion. A University of Waikato Doctoral Scholarship funded this research.

References

Anderson, A.J. 1991. The chronology of colonization in New Zealand. Antiquity 65 (249): 767-95

Anderson, A. & Wallace R. 1993. The chronology of Mount Camel Archaic site, Northland, New Zealand. New Zealand Journal of Archaeology 15:5-16.

Bronk-Ramsey, C. 1999. OxCal program v3.3. Oxford Radiocarbon Accelerator Unit.

Caughley, G. 1988. The colonisation of New Zealand by the Polynesians. Journal of the Royal Society of New Zealand 18:245-270.

Chisholm, B.S. 1989. Variation in diet reconstructions based on stable carbon isotopie evidence. In Price T.D. (ed.), The Chemistry of Prehistoric Human Bone, pp. 10-37. Cambridge University Press, Cambridge.

DeNiro, M.J. & Weiner S. 1988. Chemical, enzymatic and spectroscopic characterisaton of "collagen" and other organic fractions from prehistoric bones. Geochimica et Cosmochimica Acta 52:2197-2206.

Garlick, J.D. 1969. Buried bone. In Brothwell D., & Higgs E. (eds), Science in Archaeology: a Survey of Progress and Research (2nd edn), pp. 503-513. Thames and Hudson, London.

Grant-Taylor, T.L. 1974. Radiocarbon dating. New Zealand Archaeological Association Newsletter 17:157-162.

Heath, R.A. 1985. A review of the physical oceanography of the seas around New Zealand - 1982. New Zealand Journal of Marine and Freshwater Research 19:79-124.

Hedges, R.E.M. & van Klinken G.J. 1992. A review of current approaches in the pretreatment of bone for radiocarbon dating by AMS. Radiocarbon 34:279-291.

Higham, T.F.G. 1993. Radiocarbon Dating the Prehistory of New Zealand. Unpublished PhD Thesis, University of Waikato, Hamilton.

Higham, T.F.G. & Hogg A.G. 1995. Radiocarbon dating of prehistoric shell from New Zealand and calculation of the DR value using fish otoliths. Radiocarbon 37:409^116.

Higham, T.F.G. & Hogg A.G. 1997. Evidence for late Polynesian colonisation of New Zealand: University of Waikato radiocarbon measurements. Radiocarbon 39:149-192.

Jansen, H. S. 1984. Radiocarbon dating for Contributors. Institute of Nuclear Sciences INS-R-328. Institute of Nuclear Sciences, Department of Scientific and Industrial Research.

Lassey, K.R., Manning M.R. & O'Brien B.J. 1990. An overview of oceanic radiocarbon: its inventory and dynam- ics. Reviews in Aquatic Sciences 3(2 & 3): 1 17-146.

Law, G. 1981. Radiocarbon dating, some guidance for users. New Zealand Archaeological Association

ewsletter 24:228-236.

Law, I.A. & Hedges R.E.M. 1989. A semi-automated bone pretreatment system and the pretreatment of older and contaminated samples. Radiocarbon 31:247-253.

Leach, B.F. & Boocock A.S. 1995. Estimating live fish catches from archaeological bone fragments of snapper, Pagrus auratus. Tuhinga, Records of the Museum of New Zealand Te Papa Tongarewa 3:1-28.

McCormac, F.G, Hogg A.G., Higham T.F.G., Lynch-Steiglitz J., Broecker W.S., Baillie M.G.L., Palmer J.G., Xiong L., Pilcher J.R., Brown D. & Hoper S.T. 1998. Temporal variation in the interhemispheric 14C offset. Geophysical Research Letters 25(9): 1321^1.

McFadgen, B.G. 1982. Dating New Zealand archaeology by radiocarbon. New Zealand Journal of Science 25:379-392.

McFadgen, B.G., Knox F.B. & Cole T.R.L. 1994. Radiocarbon calibration curve variations and their implications for the interpretation of New Zealand prehistory. Radiocarbon 36:221-236.

Millener, PR. 1981. The Quaternary Avifauna of the North Island of New Zealand. Unpublished PhD thesis, University of Auckland.

Nichol, R. 1988. Tipping the feather against a scale: archaeozoology from the tail of the fish. Unpublished PhD. thesis, University of Auckland.

Paul, L.J. 1992. The biology of the snapper. In Mossman S. (ed.), All You Want to Know About Snapper, pp. 3-15. New Zealand Fishing News, Auckland, New Zealand.

Petchey, F.J. 1998. Radiocarbon Analysis of a Novel Bone Sample Type: Snapper and Barracouta Bone from New Zealand Archaeological Sites. Unpublished PhD. thesis, University of Waikato.

Petchey, F.J. 1999. New Zealand bone dating revisited: A radiocarbon discard protocol for bone. New Zealand Journal of Archaeology (1997):81-124.

Petchey, F.J. & Higham T.F.G., 2000. Bone diagenesis and radiocarbon dating of fish bones at the Shag River Mouth site, New Zealand. Journal of Archaeological Science 27:135-50.

Rafter, T.A. 1978. Report on the Work of the Institute of Nuclear Sciences, Vol 28. Department of Scientific and Industrial Research, Lower Hutt, New Zealand, Jan-Dec 1977.

Roe, N. 1969. An Archaeological Assemblage from Houhora. Unpublished MA thesis, Department of Anthropology, University of Auckland.

Shawcross, F.W. 1972. Energy and ecology: Thermodynamic models in archaeology. In D.L. Clarke (ed.), Models in Archaeology, pp. 577-622. Methuen, London.

Stafford, T.W. Jr., Brendel . & Duhamel R.C. 1988. Radiocarbon 13C and 15N analysis of fossil bone: removal of humtes with XAD-2 resin. Geochimica et Cosmochimica Acta 52:2257-2267.

Stiner, M.C., Kuhn S.L., Weiner S. & Bar-Yosef O. 1995. Differential burning, recrystallisation, and fragmentation of archaeological bone. Journal of Archaeological Science 22:223-237.

Stuiver, M., Reimer P.J. & Braziunas S. 1998. High precision radiocarbon age calibration for terrestrial and marine samples. Radiocarbon 40:1127-1151.

Stuiver, M., Reimer P.J., Bard E., Beck J.W., Burr G.S., Hughen K.A., Kromer B., McCormac FG., van der Plicht J., & Spurk M. 1998. INTCAL98 Radiocarbon age calibration, 24 000-0 Cal AD. Radiocarbon 40:1041-1083.

Trotter, M.M. & McCulloch B. 1984. Moas, men and middens. In P.S. Martin, and R.G. Klein (eds). Quaternary

112


Extinctions: A Prehistoric Revolution, pp. 708-727. The University of Arizona Press, Tucson, Arizona.

van Klinken, G.J. 1999. Bone collagen quality indicators for palaeodietary and radiocarbon measurement. Journal of Archaeological Science 26: 687-695.

van Klinken, G.J. & Hedges R.E.M. 1995. Experiments on col-

lagen-humic interactions: speed of humic uptake, and effects of diverse chemical treatments. Journal of Archaeological Science 22:263-270.

Weiner, S. & Bar-Yosef O. 1990. States of preservation of bones from prehistoric sites in the Near East: a survey. Journal of Archaeological Science 17:187-196.

Appendix 1: Houhora: Measurements (mm) of five paired cranial bones (optimal estimator given in bold).

Right Dentary LeftDentary Right Left Right Left Premaxilla Premaxilla Maxilla Maxilla

RDI* RD2 RD3 LD1 LD2 LD3 RP1 RP2 LP1 LP2 RM1 RM2 LM1 LM2 - - 45.70 31.17 11.75 41.56 15.32 ______

38.42 26.70 14.36 42.77 30.15 13.11 39.20 13.18 - 16.40 - 17.05 24.79 - 35.62 14.03 45.96 15.46 - 18.36

- - 47.75 33.15 - 35.89 13.45 38.62 14.71 - 19.92 33.49 14.29 - 36.55 14.15 32.41 12.71 - 16.04

42.57 28.21 - 54.94 39.33 - 39.79 14.35 44.81 15.75 - 18.12 - 21.66 ------ 15.68 42.14 16.85 - 17.02 41.24 17.38

32.10 - 39.90 14.63 - 20.79 - 23.36 47.93 - 13.31 - - 10.41 - 15.30 - 15.36 - 24.05

_ 40.94 26.84 11.24 - 19.02 44.30 15.77 - 18.90 37.37 - 47.21 - 19.89 - 14.21 - 16.93 -

39.44 - 11.85 - 32.11 - 37.85 13.57 - 15.86 - - - - 13.34 46.34 16.82 ______

23.72 _____ 14.28 - 18.59 42.84 - - 18.36 34.88 - 10.24 46.06 _ _ _ 18.27 - 15.03

28.43 - - 29.24 - 16.80 41.88 - 11.99 - - - 44.48 13.86 38.73 15.63

- 11.53 - 40.92 15.35 - 15.44 33.08 14.72 38.56 - 34.57 12.41 58.44 22.21

- 15.38 - - 12.57 51.29 19.08 - 13.57 53.91 44.10 30.37 12.69 - 43.89 16.51 28.38 10.45 - 18.57

- - 37.37 26.18 12.95 43.06 14.23 - 16.99 40.39 - - - 21.34 - - 14.00 50.07 16.70

------- 13.33 ----- 44.70 16.53 - 12.26 -------- 11.15

29.15 18.23 8.18 42.11 - 14.73 - 18.88 36.64 12.95 - 11.98 41.77 23.40 13.47 30.80 - - 52.67 20.69 - 18.21 50.92 - 13.90 37.47 25.74 - - 15.53 48.07 30.75 13.50 - 16.31

- 13.77 _ - _ - 18.14 ----- 35.91 13.14

23.56 - - 28.56 - - 16.14 - 15.48 40.62 ______-- 10.85 47.05 30.98 - 15.42 - 13.24 - 13.19

------- 37.70 13.54 28.18 - 36.22 24.78 - - 15.56 26.85 - 41.55 31.50 14.59 48.16 17.51 - 15.74

31.47 21.23 9.02 - - 10.94 - 12.79 29.10 __-_--- 11.78

113


33.89 24.44 _______ 15.79 ------ 17.88

53.61 35.71 - 11.60 - 13.40 - 13.94 ----- 48.69 17.07 - 14.34 - - 40.16 _ _ _ 13.22

35.69 - 13.47 - 16.99 - 11.20 - - 9.81 34.56 11.83 - 11.26

28.31 - 7.58 32.87 22.74 - - 19.58 - 8.88 - 11.71 - - 10.27 39.42 13.35 - 11.09 - - 13.48 - 15.25

- 19.08 - 16.78

- 15.79 29.23 10.07

- 13.14 - 16.25

Total no. of elements = 249 *See Leach and Boocock 1995:2, figure 1 for measurement parameters.

Appendix 2: Houhora: Estimated Fork Length (mm) from the largest measured dimension ("optimum estimator").

RDI RD2 RD3 LDI LD2 LD3 RP1 RP2 LP1 LP2 RM1 RM2 LM1 LM2

641 620 632 656 543 819 610 619 581 586 620 667 437 641 610 559 565 573 498 635 595 603 536 574 572 616 595 576 553 481 567 487 607 567 601 523 533 455 576 519 577 538 473 554 371 555 561 596 518 529 552 504 565 491 460 550 555 541 577 494 517 514 440 531 480 455 516 549 523 573 456 503 508 413 514 476 508 518 520 565 455 496 502 506 455 501 478 514 537 446 488 472 504 423 494 451 504 530 445 486 471 491 406 485 429 488 515 433 484 470 488 403 467 423 481 513 410 455 439 477 454 404 469 510 385 452 331 466 453 468 499 340 439 434 439 464 495 427 424 400 462 492 420 413 377 432 491 416 385 425 483 403 357 425 482 386 347 422 451 371

410 449 351 350 442 342

423 280 418 418

415

114


Article Contentsp. 104p. 105p. 106p. 107p. 108p. 109p. 110p. 111p. 112p. 113p. 114

Issue Table of ContentsArchaeology in Oceania, Vol. 35, No. 3 (Oct., 2000), pp. 97-128Front MatterAssemblage Variability in the Willandra Lakes [pp. 97-103]Radiocarbon Dating Fish Bone from the Houhora Archaeological Site, New Zealand [pp. 104-114]Research ReportsCoastal and Inland Lapita Sites in Vanua Levu, Fiji [pp. 116-119]Crossing the Pwanmwou: Preliminary Report on Recent Excavations Adjacent to and South West of Mangaasi, Efate, Vanuatu [pp. 120-126]

Book Notices [pp. 127-127]Back Matter

radiocarbono

Documents

bone collagen

bone determinations

moa bone

houhora snapper bone

barracouta thyrsites

acceptable radiocarbon

new zealand fiona petchey

new zealandauthors