On the reliability of archaeological rat bone for radiocarbon dating in New Zealand

Holdaway & Beavan (1999) discussed the radiocarbon dating of bone of various species from the site of Hukanui Pool, Hawkes Bay. We question their conclusion that two apparently reliable rat bone gelatin determinations from the Hukanui Pool site provide support for the entire suite of rat determinations from previously dated 'natural' sites. We present evidence that contradicts their conclusion that bone material from the broad range of archaeological midden sites is generally less well-preserved than bone from 'natural' caves in New Zealand such as Hukanui Pool. We show that when dates from archaeological bone from Pleasant River and Shag River Mouth are evaluated, the state of preservation is comparable with material from the 'natural' site of Hukanui Pool, and should provide accurate and reproducible radiocarbon determinations. Our conclusion has serious implications for the acceptance of the model proposed by Holdaway (1999), because if archaeological bone is well-preserved but yields unreliable and unreproducible results, it is likely that well-preserved 'natural' bone is similarly affected.