No, I am not going to discuss the starch you eat at meals or the starch that converts to storage cells that begin to settle around our waist as we grow older! What I am talking about is starch that can be found on artefacts or in soils at archaeological sites along with pollen and phytoliths. Right away I see some of you rolling your eyes and saying that you are a "paleopalynologist" and don’t work in that geologic zone known as the "overburden!" However, if you are the least bit curious, if you want to broaden your horizons, or if you are one of those palynologists who really do work with archaeologists, then you might want to read on a bit further.

During the mid 1990s, Bonnie Williamson of The University of Witwatersrand in South Africa was a graduate student working at the nearby Rose Cottage Cave site where she was asked to examine the surfaces of hundreds of stone tools for attached residues. The archaeologist hoped that some of those Middle and Late Stone Age (60,000-25,000 years ago) tools might still have residues stuck to their cutting edges; residues that could be tested for traces of animal blood, amino acids, DNA, and even pollen. Because of the age of the site and prevailing assumptions about our ancient ancestors, most believed the stone tools would reveal that these early cultures primarily hunted and butchered a variety of large game animals, but, when the tests were completed, the researchers were shocked. Williamson found that just over 50% of all the residues on the stone tools came from plant, not animal, origins. There were no pollen grains in the residues, but some of the residue consisted of plant fibers. Nevertheless, the strongest cases for plant usage came from preserved starch grains in the residues along the cutting and grinding edges. Her study confirmed that some cutting tools were multi-purposed (another shock for the archaeologist) because they had both starch and blood residues on them.

There is no question about the importance of starch research or about the startling new insights that starch grains are bringing to the field of archaeology. Earlier this year botanist Linda Perry and her co-authors (2007, Science 315:986-988) made international headlines when they reported their discovery of starch grains from chilli peppers in a number of archaeological sites in the Caribbean, and Central and South America. For decades, scientists had wondered about the antiquity of this New World plant and how early it had been domesticated and then distributed into Central America and the Caribbean islands. For years archaeological sites throughout the semi-tropical and tropical regions of the New World revealed evidence of early farming activities and occasionally provided hints about which plants were being grown. However, clear fossil evidence for chilli peppers was lacking. Frequent cycles of wetting and drying, warm temperatures, high levels of microbial activity, and rapid organic reduction in the upper layers of tropical soils ensured the destruction of most forms of organic remains in archaeological soils. However, starch grains are preserved in many tropical soils and this team of researchers, lead by Linda Perry, used starch evidence to confirm that chilli peppers originated in Bolivia, were domesticated before 6,000 BP, and were then spread rapidly as a cultigen first to Central America and then throughout the Caribbean.

Before her chilli starch discoveries, Perry and others (2006, Nature 440:76-79) used records of starch grains and phytoliths in archaeological soils to show that the spread and use of maize in the southern highlands of Peru occurred before 4,000 BP; nearly a millennium earlier than previously suspected. Why couldn’t they do this with fossil pollen evidence? They looked, but because soil conditions were so destructive there were no usable pollen records. In other regions of the world, starch researchers are also rewriting the history books about how and when various cultigens and plants were being used or grown. Two recent studies reported in the Journal of Archaeological Science by Mark Horrocks focus on the discovery of starch grains in archaeological soils in the South Pacific. Horrocks’ discovery of 3,000 year-old taro and yam starch grains in archaeological soils on the Fiji Islands proved that early Lipita cultures living on those islands practiced a horticultural economy based on plants that they had brought with them to those islands. Earlier reports claimed that early Fiji Islanders had been strictly foragers. At Pitcairn Island, Horrocks and Weisler recovered starch grains of both taro and sweet potatoes in archaeological deposits and believe that their starch discovery confirms that early Polynesians brought cultigens with them when they first occupied Pitcairn Island around AD 1400. As palynologists you might again wonder why these and other researchers didn’t rely on fossil pollen records. The answer is that for decades archaeologists have searched in vain for good pollen records from numerous island sites throughout Oceania. With rare exceptions, those efforts have proven futile. I know first hand about such problems because I have processed and examined archaeological soils from half a dozen sites on various islands in Oceania searching in vain for preserved pollen. The few pollen records that do exist for that region of the world come mostly from lake sediments in extinct volcanic craters.

During the past decade there has been a number of published articles and submitted manuscripts discussing the search for, and recovery, of starch grains in archaeological sites all over the world. Starch grains are now being searched for in soils, in the residue stuck to ancient pot sherds, and stone and bone tools, inside amphorae, on stone artefacts, in bedrock mortars, on grinding stone surfaces, in coprolites left behind by ancient cultures, in the dental calculus found on the teeth of ancient burials, and from acetate peels taken from the surfaces of items suspected of containing starch grain residues. When soil and environmental conditions are favourable, many of those same sources also provide fossil pollen evidence. Together, fossil pollen and starch grains form convincing records about plant usage by early cultures.

Starch grains from Nothoscordum bivalve (Liliaceae). Photo: V. M. Bryant	Nothoscordum bivalve starch grains under polarized light. Photo: V. M. Bryant

I am primarily a palynologist so why should I be so excited about trying to search for starch grains? When you work with archaeologists and when you want to know as much as possible about our earliest ancestors, then you don’t want to miss any opportunity to gain more information. When I first entered the field of pollen studies during the early 1960s, few archaeologists were searching for fossil pollen in archaeological soils. Sites were being excavated, burials were removed, coprolites were discarded, amphorae from underwater shipwrecks were emptied on the sea floor to lighten them for transport to the surface, and pueblo sites were excavated without ever searching for pollen remains. Fortunately, most of that has changed during the last half-century. Today, sampling for pollen at archaeological sites is common practice in most regions of the world. In temperate and Arctic regions, fossil pollen records are now providing archaeological insights about ancient diets, origins of agriculture, functional use of various types of artefacts, the cargo contents of sunken shipwrecks, how paleoenvironmental changes affected early cultures, the medicinal use of plants, burial practices, room usage in pueblos and other early structures, and the use of pollen as paints and dyes. Nevertheless, hundreds of studies from semi-tropical and tropical regions describe how pollen studies were dutifully attempted, but yielded either no pollen or such meagre amounts that the data were unreliable. However, in many of those same regions where pollen and other forms of plant preservation are absent, starch grains are well preserved.

I would be the last person to suggest that archaeologists should abandon the search for fossil pollen and instead focus only on fossil starch grains! Instead, I propose that archaeologists and palynologists should both be thinking about the potential of finding starch grains in the sediments they examine. Pollen and starch samples should be collected from storage pits, burial features, floor surfaces, hearths, and from the cultural levels exposed during excavations. Soil samples used for starch studies must be collected using sterile tools that have been washed and cleaned with de-ionized water, then dried. When collecting dirt for starch studies one should wear a clean pair of sterile, latex gloves (be certain they are not powdered in starch!) during each sampling to prevent possible contamination. Individual soil samples do not need to be larger than about 20 grams and each should be placed in a sterile, plastic or glass bag or container that can be tightly sealed. Sampling using metal tools or contact of samples with metal should be avoided because metallic residues may be transferred to the sample and this can complicate later starch interpretations. If these careful precautions are taken, and if one were to collect twice as much soil as needed for each sample, then the same samples could serve dual purposes for both pollen and starch studies. Like pollen, starch grains can be identified to family, genus, and sometimes species. And, like pollen, the most essential need is for a good, modern, starch reference collection!

---

This article first appeared in CAP Newsletter 30(1):7-9, 2007. It is included here with permission of Vaughn Bryant.