This paper examines the circulation of cassava germplasm (Manihot esculenta) as a problem in biocultural history. It reviews how the plant was introduced into the Moluccan islands as part of the ‘Columbian exchange’ and how it has subsequently moved around through the distribution of stem cuttings. Using a combination of lexical, ethnographic, botanical and genetic data, including material from ecologically-contrasting field sites, it evaluates the evidence for tracking germplasm movement at the local and regional level, considers the extent to which this contributes to local agrobiological diversity, and reflects upon some methodological problems of integrating sociocultural and scientific evidence.