33 Dating New Zealand Obsidians by Resonant Nuclear Reactions Foss Le ach Department , University o fOtag o and Harry Naylor Phys ic s Department , University of Auckland ABSTR ACT Measurements of hydration depths of prehisto ri c obsidian artefacts by thin-section techniques have yielded dates of mixed accuracy in different parts of the world, but have been notably unsuccessful in New Zealand. Resonant nuclear reactions, widely employed to study surface diffusion in solids, were attempted on Mayor Island obsidian from New Zealand. It was fou nd that artefacts of varying ages ha ve hydration rinds of very si milar overall thickness a nd that the rela tionship between the theoretical and actual character of the diffusion front is more complex than is widely assumed . Keywor ds : RESONANT NUCLEAR REACTIONS, OBSIDIAN DATING, FLUORINE PROFILES, HYDROGEN PROFILES. I NTRODUCT ION In 1960, and Smith published a p ape r which estab li shed that the ma nufac- ture of prehistoric artefacts could be directly dated by th e depth ofhyd r alio n which had occ urred on the fresh surface cleaved by prehi lone man. Since that li me no fewer than four entirely different techni ques h ave been developed lo measure this microscopic band of hydration. Friedman's method relied on the preparation of thin sections, the identification of th e laye r by virtue of its birefringent properties under polarised lig ht , a nd its me asure ment with a special Vicker's image sp liuin g microscope. There have been many variations on this basic theine si nce. Measureme nt s to within ± 0.2μm ca n be achieved with this method , correspo ndin g to a typ i ca l dating er ror of± 15- ± 20% over the last mille nnium. An obvious di sadva nt age of the method is th at it is partly destructive of artefacts. In New Zealand , ob idian chips are very com- mon on arch aeo logical sites, and o this is nol necessarily a se ri ous drawback. New Zealand arc haeologist were quick lo take adva ntage oft hi n ew method , and Ambrose a nd Green r eport ed en cou r aging results on no fewer th an 1 00 ob idian speci- mens in 1962. Thi was followed in 1 964 by a second paper giving the detail of measure ments ra nging from 0.8- 2.0μm from several arch aeo logical sites (Gree n 1964). The relative dating of the se site by th e hydration depths ap pe ared to closely fo llow the order of age established by ot her archaeological mean . Thu , ea rl y indications were tha t this technique of datin g would have a ro y future in ew Zea land. Am bro e and Green ( 1962) recognised the imp o rtance of keeping a cl o e watch for c han ges in hydration rate from one source to an o ther and fordifferenl environmental tempera - ture throu ghout New Zealand . There are a bout 18 ource of obsidian in ew Zea- land ; one of these, a green variety from Mayor Island , is e pecially co mmon in archaeological ites. For this reason, it wa s chosen for thi earl y pilot stud y and a l o for the research reported in this pape r. After this brief initial flurry of research , very li llle was done in New Zealand until abo ut 1970. In that y ear , the nece ary thin- ectio ning eq ui pment and high qualit y Ne w Zealand Journal of Ar chaeology. 1981, Vol. 3, pp. 33-49