In "Ancient Starch Research" (R. Torrence & H. Barton eds.), "How to take samples from residues of potshrds?" presents two methods: direct application of ultrapure water to the surface with a pipette, and scraping a small area (3-5mm) with a fresh and sterile scalpel blade (p.198).
Indeed, the study by A. Crowther applied these two methods to take residue samples, and its result present residues on undecorated Lapita pottery contained taro starch, Colocasia esculenta.
- Crowther, A. 2005. Starch residues on undecorated Lapita pottery from Anir, New Ireland. Archaeology in Oceania 40: 62-66.
The problem is whether to applly these methods to ancient starch analysis of charred resudues on potttery found from Japanese archaeological sites frequently.
The charred residues are various, such as boiled over remains, burnt remains, and charred seeds. Charred seeds can be identified from their morphology. However, boiled and burnt remains were made by cooking, and thus, their identifications present addditional difficulty.
Can ancient starch analysis help to identify these residues?
If their contents were from plants, starch granules of cooked materials could be gelatinised by heating, that is to say, starch granules could be degraded completely. If so, these residues cannot be identified.
The identification of potsherd residues with starch analysis can be dependent on their conditions or preservations.
At present, as the initial work, residue samples on potsherds found from a Jomon site, taken by spot sampling with water and scraping residues, are examined under transmitted light microscopy.
The results are not present yet. As well as this approach, a further research will be conducted: whether to extract starch from potsherd residues made by cooking experiments, and to identify their starch granules.