Age of North Island giant moa (Dinornis novaezealandiae) bones found on the forest floor in the Ruahine Range

Abstract

During the nineteenth century, multiple discoveries of moa bones lying on the ground surface in exposed situations generated debate about the recentness of moa extinction. Subsequently, clearance of land for agricultural purposes has probably resulted in the destruction of many surface moa bones and such finds are now rare. The recent discovery of a pair of tibiotarsi from a North Island giant moa (Dinornis novaezealandiae) lying on the forest floor at Aorangi Awarua, Ruahine Range, has allowed the age of two surface moa bones to be tested. Both tibiotarsi were AMS radiocarbon dated and returned ages of 681±28 and 721±29 14C years BP. The 2σ calibrated ranges are statistically the same age and suggest that the bones are from one bird that died sometime between AD 1286 and 1390. Although the bones post-date human settlement of New Zealand, they are not anomalously recent. The results show that, given favourable environmental conditions, moa bones can persist in exposed situations for long periods, and do not necessarily reflect survival of moa into relatively recent times.

Keywords:

• Dinornithiformes
• Dinornis novaezealandiae
• extinction
• moa
• New Zealand
• radiocarbon

Introduction

Moa bones are commonly preserved in four main types of natural (non-archaeological) deposit: caves/tomos/rockshelters, alkaline mires, carbonate-rich sand dunes, and loess (Worthy & Holdaway 2002). Each of these deposits has physical and/or chemical characteristics that favour bone preservation and, as a result, bones can remain in such sites for thousands of years. Occasionally, moa bones have been discovered on the surface of the ground in situations where they appear never to have been buried, are fully exposed to the weather, and are not associated with any of the deposits mentioned above. Although such discoveries are exceedingly rare today, it appears that finds of surface moa bones were a more common occurrence during the nineteenth century.

In 1856, the first government expedition into Central Otago found abundant moa bones scattered across the ground throughout the region (Hector 1873). Field (1894, pp. 562–563) noted:

the immense number of moa-bones scattered on the surface of the ground in the South Island when the settlers first arrived there, and their rapid disappearance afterwards

and that in 1852–53 his cousin had found:

apparently, the whole of the bones of a medium-sized bird [moa], where it had seemingly lain down and died, beside a large flax-bush, at St Albans, near Christchurch.

In the 1890s, on Maungaraki near Westmere in the Wairarapa, moa bones were found:

lying on the surface on the top of the mountain, and free from bush. Some were exposed to the air, but others were covered by a heavy growth of thistles. None were covered by earth … there can be little doubt that the bones have been lying undisturbed on the surface since the bird died. (Hector 1895, p. 655)

Another find of surface moa bones was reported by White (1893, p. 504), who wrote:

When engaged in sowing grass-seed lately, on a broad clearing, I was greatly surprised to notice a collection of broken bones, intermixed with a number of highly-polished stones, which evidently were the gizzard-stones from a large bird. It is worthy of special notice that this collection was simply on the surface of the clay soil and not in the least buried.

The exposed position in which surface moa bones were found caused naturalists to question their antiquity, and ultimately the recentness of moa extinction. For example, in writing about moa bones from Puketoi Creek, Central Otago, Gillies (1875, p. 545) stated that they were:

found on the surface … in a tolerably perfect state of preservation; and when it is borne in mind how rapidly the bones of stock on the runs become decayed through the action of the atmosphere, this circumstance of the presence of the moa bones on the surface goes a long way towards proving that the bird existed in recent times.

White (1893, p. 504) also remarked that a find of surface moa bones:

had the same appearance as if they had not been there more than some fifteen years or thereabout.

Haast (1872, p. 71) concluded that:

The testimony that Moa bones have been found lying loose amongst the grass on the shingle of the plains … where probably a bird has died and decayed, is too strong to be set aside altogether, or to be explained by the assumption that the bones became exposed … We are, therefore, almost compelled to conclude that the bones have, in some instances, never been buried under the soil, but remained lying on the surface where the birds died. I can, however, not conceive that Moa bones could have lain in such exposed positions for hundreds, if not thousands, of years without decaying entirely.

However, Haast (1872, p. 72) noted also that the absence of Māori traditions relating to moa placed:

an almost insurmountable obstacle in the way of our supposing that the Moa bones found lying on the plains or hill-sides are of such recent origin as their position might at first suggest.

Despite these early reports of surface moa bones, few are now held in museum collections, and to our knowledge none have been radiocarbon dated. Therefore, several questions remain. Can bones survive in exposed locations for hundreds of years? Were surface moa bones once shallowly buried but have since eroded from sources that may not always be obvious? Or do surface moa bones represent the remains of the last surviving birds? Recently, a new discovery of surface moa bones has allowed the question of their antiquity to be examined for the first time using radiocarbon dating.

Study site and specimens

On 4 April 2012, JRW visited a site at Aorangi Awarua, in the western Ruahine Range near Taihape, where two moa tibiotarsi were lying exposed on the forest floor. Although the bones were weathered, and the ends were missing, the tibiotarsi were identified as being from a North Island giant moa (Dinornis novaezealandiae) based on their length (at least 74 cm). The bones were located on a ridge (NZTM E1862205/N5610605) at approximately 820 m above sea level and within a sparse understorey of mountain beech (Nothofagus solandri) forest (Fig. 1). The site receives a mean annual rainfall of c. 1100 mm and has a mean annual temperature of 9 °C (Leathwick 2003). The bones were on flat to gently sloping ground, with no obvious banks from which they might have been eroded. The left and right elements were found lying in the correct orientation and superficially embedded on the forest floor. A thorough search of the ground surface within 25 m of the tibiotarsi, and the top few centimetres of soil immediately around the tibiotarsi, failed to find any other bones. This raises the possibility that the tibiotarsi may have been removed from another site nearby and placed on the forest floor relatively recently. However, there was no further evidence to substantiate this possibility, and the most plausible explanation is that the bones had been lying in situ from the time of death and only the most robust elements were still present. The bones were removed from the site and are being temporarily housed at Landcare Research, Lincoln (specimen numbers X12/14a and X12/14b), but in the future will be permanently stored in the collections of Whanganui Museum (specimen numbers 2012.61.1 and 2012.61.2).

Figure 1 Location of moa bones on Aorangi Awarua, Ruahine Range. A, View north from grassy clearing at NZTM E1861800/N5609600 showing approximate location of moa bones (arrowed). B, Two North Island giant moa (Dinornis novaezealandiae) tibiotarsi (foreground), photographed in situ, and a view of the immediate environment around the bones. C, Closer view of the two in situ D. novaezealandiae tibiotarsi with pen (145 mm) for scale.

Sampling and analysis

A sample of bone was taken from each tibiotarsus for radiocarbon dating. The external surfaces of the samples were stained brown, probably from tannins in the leaf litter, and hence were a potential contaminant for radiocarbon dating. Therefore we ground off the external surfaces using a Dremel tool until the lighter, relatively well-preserved interior bone was exposed. The samples were submitted to Waikato Radiocarbon Dating Laboratory for analysis using accelerator mass spectrometry, and prepared using the ultrafiltration method (Higham et al. 2006) to further eliminate any residual contaminants in the bone gelatin.

Results

There was no significant difference between the two radiocarbon dates (Chi square test: t = 0.9846; 5% confidence limit 3.84), supporting the premise that the left and right tibiotarsi were from the same bird (Fig. 2). Calibration of the combined radiocarbon dates (using the ShCal04 calibration curve [McCormac et al. 2004] and R-combine function in OxCal 4.1 [Bronk Ramsey 2001]) for the bone samples indicated that the bird had died between AD 1286 and 1390 (Table 1; Fig. 2).

Figure 2 Radiocarbon dating results for North Island giant moa (Dinornis novaezealandiae) tibiotarsi from Aorangi Awarua, Ruahine Range. A, Comparison of the calibrated age ranges for the two radiocarbon dates. B, Calibrated age range of combined radiocarbon dates. White circles are median ages, and bars represent 95.4% confidence ranges. The Tūrangi moa was described by Worthy (2002).

Table 1 Radiocarbon measurements of North Island giant moa (Dinornis novaezealandiae) tibiotarsi from the surface of the forest floor at Aorangi Awarua, Ruahine Range. Due to the small size of the samples, delta 13C was measured on prepared graphite using the AMS spectrometer.

Lab no.	Tibiotarsus	d^15N vs air (‰)	Total N (%)	d^13C vs PDB (‰)	Total C (%)	C:N	Radiocarbon age (years before present)	SD	Calibrated age AD and confidence
Wk-33997	Right	3.47	15.37	−22.67	43.94	3.34	681	28	1300–1389 (68.2 %) 1295–1392 (95.4 %)
Wk-33998	Left	3.88	15.40	−22.74	45.32	3.43	721	29	1286–1380 (68.2 %) 1276–1389 (95.4 %)

Discussion

It is now widely accepted that moa became extinct shortly after human colonization due largely to hunting pressures (Holdaway & Jacomb 2000). The age of the Ruahine moa's death is approximately within the century after the initial human settlement of New Zealand (Wilmshurst et al. 2008). Worthy (2002) described another recent skeleton of D. novaezealandiae from near Tūrangi which dated to 715±60 14C years BP (recalibrated here to 95.4% confidence range of AD 1230–1403). This date overlaps with the entire age range of the specimen described in this paper (Fig. 2). The Ruahine bones therefore are relatively young for a natural site but not anomalously recent. Several younger or contemporaneous radiocarbon dates have been obtained on moa remains from both natural deposits (e.g. Wood et al. 2012a,b; Rawlence & Cooper 2012) and archaeological middens (e.g. Petchey 1999). Therefore, it seems likely that in some situations moa bones can persist for long periods of time in relatively exposed situations. Local calcareous geology, such as the marine Tertiary sedimentary sequence at Aorangi Awarua, may play an important role in buffering soil pH and increasing the preservation potential of bone. Surface moa bones therefore do not demonstrate a persistence of moa into relatively recent times. An analogous situation exists with the many desiccated soft tissue remains of moa which were discovered during the nineteenth century, but which are rare today. It was suggested that these, as with surface moa bones, may represent relatively recent specimens (e.g. Travers 1876); however, radiocarbon dating has since shown that while many are <1000 years old, none are anomalously recent (Rawlence et al. 2012).

JRW was also shown bones in private collections from two little bush moa (Anomalopteryx didiformis) that had previously been found by locals at different sites around Aorangi Awarua (one tibiotarsus, one complete and one partial tarsometatarsus from a tomo; and one tarsometatarsus found in a creek bed, likely to have been washed out of a cave or stream bank). Therefore, bones of both moa taxa that were characteristic of dense North Island forests (Dinornis and Anomalopteryx) (Worthy & Holdaway 2002) have now been found around Aorangi Awarua.

Extensive land clearance for agriculture, often involving fire, during the nineteenth and early twentieth centuries probably caused the loss of most surface moa bones throughout New Zealand. Such bones are now likely to exist only in remote, undisturbed and rarely visited areas.

Acknowledgements

We are grateful to Richard Steedman for reporting the discovery of these bones and to Trevor Worthy for passing this report on and for subsequent discussion about the bones. Nic Rawlence provided comments on the manuscript. The Aorangi Awarua Trust kindly allowed the bones to be collected and studied. JRW was shown to the site by Richard Steedman and Tama Wi Paki (Trustees of the Aorangi Awarua Trust), the Wi Paki family (Sabrina, Kawhia, Te Iringa and Maatakotahi), and the Sage family (Richard, Max and Jonny).