The multi-disciplinary methods used to investigate early agriculture at Kuk Swamp in the Highlands of New Guinea are outlined. Methods adopted during the original investigations in the 1970s (directed by Jack Golson, with Philip Hughes from 1974), as well as during renewed investigations at the site (directed by Tim Denham from 1997 under Golson's supervision) are considered. Three methodological contributions to the study of early agriculture and plant exploitation are highlighted: X-radiography, biostratigraphic markers, and the integration of macrofossil and microfossil analyses. One outstanding methodological problem is considered, the representativeness of different components of a feature fill for reconstructions of past cultivation practices. Thoughts on an integrated method (macrofossil, microfossil and molecular) for future research on early agriculture and arboriculture in the region are presented.