![]()
Part I
The Nature and Study of Paleoethnobotanical Remains
![]()
1
Paleoethnobotanical Remains
Glossary
dichotomous key a tool that helps the user identify an unknown organism or other entity, such as a pollen grain or wood specimen. A dichotomous key is composed of a series of statements consisting of two choices (different characteristics of the entity) that lead the user to the correct identification.
Introduction: The Paleoethnobotanical Approach
Paleoethnobotany is the study of the interrelationships between humans and plants through the archaeological record. In this chapter I describe the paleoethnobotanical approach and introduce the major kinds of archaeobotanical remains—macroremains (seeds, fruits, underground storage organs, wood), pollen, phytoliths, and starch grains—and discuss how they are studied.
Paleoethnobotany is essentially an archaeological approach, and many paleoethnobotanists are trained as archaeologists. Plant remains, the primary research materials of paleoethnobotany, must be recovered from sites and identified. Human interactions with the plant world—for example, cultivation and consumption of domesticated plants—result in the deposition and preservation of four primary kinds of plant remains. Macroremains are larger plant tissues that are visible to the naked eye. Often, plant tissues preserve by becoming charred, for instance, in a cooking accident. Food preparation may also result in the deposition of starch on grinding stones and in cooking pots, where it may be preserved in crevices or food crusts. Decay or burning of plant tissues such as leaves, stems, and fruits releases phytoliths (plant opal silica bodies) into site sediments. Certain kinds of cooking techniques may also result in phytolith deposition in vessels or ovens. In the process of creating gardens, people cut down and weed out some plants and cultivate and encourage others. This changes the mixture of pollen that is deposited over the landscape. Pollen deposited and preserved in lake sediments records the history of human–plant interrelationships on local and regional scales; phytoliths that wash into lakes with eroding soil also reflect past vegetation. Pollen may also be ingested, for example, in a tea made from flowers, and preserved in human fecal remains (coprolites) with other plant and animal tissues. It is not uncommon for plants to be represented in the archaeological record in multiple ways.
The nature of these diverse archaeobotanical materials also demands expertise in botany. Archaeologically trained paleoethnobotanists must learn plant taxonomy, anatomy, and laboratory skills necessary to recover and identify plant remains. Even paleoethnobotanists whose primary training is in botany must adapt their skills to deal with fragmentary materials and the incomplete archaeological record.
As a field, paleoethnobotany has its roots in the early to mid-nineteenth century, when well-preserved plant remains from Egyptian tombs, lakeside Swiss villages, and Peruvian mummy bundles were studied by botanists. Study of macroremains continued in the late nineteenth and first half of the twentieth centuries in Europe, the Mediterranean region, and coastal Peru.
Pollen and phytoliths also began to be studied systematically at that time. The real potential of fossil pollen studies was realized in 1916 when Lennart von Post presented the first modern percentage pollen analysis. This approach permitted direct quantitative description of past vegetation composition. Another early application of pollen analysis was determining relative dating of archaeological sites by comparing pollen assemblages from sites to a sequence of vegetation changes during the late Pleistocene and Holocene epochs established from studies of peat bog and lake deposits. Early twentieth-century applications of phytolith analysis focused on identifying cultivated grasses of the Near East.
In spite of early studies of well-preserved plant remains from Peru by European and Latin American scholars, American archaeologists and botanists did not show much interest in archaeological plant remains until after 1930. A similar situation existed for pollen analysis, with early research largely limited to the arid southwestern United States. In North America phytoliths were mostly of interest to botanists and soil scientists, and starch to food scientists.
Analyses by Gilmore (1931) and Jones (1936) of desiccated plant remains from rock shelters in the United States helped fuel interest in paleoethnobotany in North America, and publication of Excavations at Star Carr by British archaeologist J.G.D. Clark (1954) convinced many of the importance of biological remains for archaeological interpretation. With increased interest in American archaeology on reconstructing subsistence and paleoenvironment, greater emphasis was put during the late 1950s and 1960s on recovering and analyzing macroremains and pollen. After Struever (1968) described flotation, archaeologists began to recover macroremains from a diversity of sites, not just those in which dry or waterlogged conditions preserved quantities of material. By the 1970s, pollen analysis was incorporated into archaeological investigations in many regions of the United States, Canada, and Latin America.
The development of phytolith analysis in archaeology was tied to two research foci of the 1970s and 1980s: the origins and intensification of agriculture, and reconstructing past environments. The potential of starch analysis was first recognized through study of organic residues on stone tools, which began in the late 1970s and early 1980s in Oceania and North America. These studies revealed that organic residues, including identifiable starch, survived more often than expected (i.e., not just in arid sites). By 2006, when Ancient Starch Research (Torrence and Barton, eds.) was published, it had become apparent that starch was surprisingly common in the archaeological record.
Paleoethnobotanical research has diversified and grown exponentially in the last three decades. Among the research topics that have been addressed through the study of archaeobotanical remains: identification of domesticated plants, agricultural practices, and agricultural intensification; reconstructing diet and subsistence; prehistory of beer and wine making; use of wild foods by agriculturalists; wood acquisition, use, and environmental impact; plants and political complexity, social status, and social change; plants as indicators of ethnicity; identification of fiber and wooden artifacts; prehistory of spices; plants in ritual and medicine; plants in economic systems; human environmental impacts, management, and land use; study of living conditions of prehistoric populations; tracing the settlement by humans of remote regions, such as Polynesia; investigating diet and health through study of coprolites; study of agricultural fields and gardens; plants in human evolution; spatial and activity analysis; and diet of living and extinct primates.
Paleoethnobotanical Data and Their Study
Macroremains
Macroremains are larger plant tissues visible to the naked eye. They include seeds, fruits, nuts (all plant reproductive structures), wood, roots and tubers, and other vegetative materials. The last named will not be covered here. Table 1.1 lists the scientific names of plants mentioned in the text.
Seeds
A seed is a reproductive structure composed of a protected embryonic plant (Shackley 1981). In the major group of seed-bearing plants, the angiosperms, seeds develop in the ovary, the basal portion of the female part of the flower. The ovary and seeds develop into the fruit. Angiosperm seeds have three major parts: embryo, endosperm, and seed coat (testa) (Figure 1.1). The embryo is composed of the first “true” leaves, the primary root, and the cotyledon or “seed” leaf (Evert 2006). Dicotyledons have two “seed” leaves, monocotyledons one. In some plants, such as legumes, cotyledons are large and contain stored nutrients to nourish the seedling. Seeds of other plants contain specialized tissue, the endosperm, which provides these nutrients. The amount of endosperm and its position in the seed relative to the embryo differ among plant groups and are important traits in seed identification. Embryo and endosperm, if present, are protected by the seed coat, a one- or two-layer outer covering. Seeds coats may have distinctive coloration, texture, or surface features that can be important distinguishing characteristics. The point where the seed was attached to the ovary is the hilum, or attachment scar.
Table 1.1 Scientific names of plants mentioned in text | arrowroot family | Marantaceae |
| banana family | Musaceae |
| bird-of-paradise family | Heliconiaceae |
| canna family | Cannaceae |
| carrot | Daucus carota |
| cotton | Gossypium |
| daisy family | Asteraceae |
| grass family | Poaceae or Gramineae |
| ginger family | Zingiberaceae |
| gourd | Lagenaria siceraria |
| hickory family | Juglandaceae |
| legume or bean family | Fabaceae or Leguminosae |
| maize | Zea mays |
| palm family | Arecaceae or Palmae |
| peach | Prunus persica |
| sedge family | Cyperaceae |
| Squash | Cucurbita |
| sweet potato | Ipomoea batatas |
Figure 1.1 Examples of angiosperm seeds, illustrating characteristics that aid identification: (a) Asteraceae (daisy family); seeds are often oblong and sometimes ridged; a pappus (ring of hairs or scales) may be present; an attachment stalk (remains of funiculus) is often prominent; (b) Poaceae (grass family) seeds have a lateral or basal-lateral embryo; endosperm tends to be abundant; some grasses have elongated, grooved seeds (e.g., ryegrass); others have seeds that are flattened and lightly ridged (e.g., crabgrass); (c) Fabaceae (bean family); many legumes are beanlike, with a notched attachment area or hilum; the seed coat is usually smooth; seed interior consists mostly of two broad, thick cotyledons, with little or no endosperm. Not to scale.
Not all distinguishing characteristics may be preserved in archaeological seeds. Size and shape may be distorted by charring, for instance, or seeds may be broken. The seed coat is an important source of diagnostic characteristics, including color for uncharred seeds, texture (whether the testa is smooth or has a raised pattern), attachments, and scars like the hilum. If the seed coat is lost during charring or badly eroded in desiccated materials, identifiability decreases markedly. Waterlogged seeds may also lose their outer layers and suffer cell distortion.
Accurate identification of archaeological seeds to botanical taxon is largely a matter of practice, and is grounded in comparing archaeological specimens to known kinds of seeds. Photographs and line drawings are helpful visual aids, but are no substitute for identified examples. Study of comparative seed specimens helps the novice learn the diagnostic features of each kind of seed, how it appears when broken or charred, and how it can be distinguished from similar seeds.
Fruits and Nuts
A fruit is a ripened ovary and its adhering parts—the seed-bearing organ of the plant. The ovary wall, or pericarp, can be soft and fleshy, leathery, hard, or thin and paperlike, depending on the form of the fruit (Figure 1.2). Internal arrangement of seeds in the fruit is also variable. Some fruits have only one seed, which is fused to the ovary wall; other fruits have several seeds. Although whole fruits are sometimes found archaeologically, it is more common to find fragments of inedible fruit parts. Very large seeds, such as palm and other tropical fruit seeds, often produce macroremains that are more similar in size and appearance to fruit fragments than to small seeds.
Nuts are indehiscent (not splitting open to release seeds when ripe), usually one-seeded, hard and bony fruits. Some nuts, such as those in the hickory family, are covered with a leathery husk. Fragments of husk, nut shell (hard pericarp), and nut meats (embryo) may all be found archaeologically. There are two major types of succulent fruits: drupes and berries. Drupes are fleshy fruits with one or more seeds enclosed in a hard, stony portion of endocarp, such as the peach pit illustrated in Figure 1.2. Berries are similar to drupes, but their seeds are surrounded only by a hardened seed coat. Legumes are dry fruits that open along two sutures, releasing the mature seeds. The cotton fruit shown in Figure 1.2 is an example of a capsule, a multichambered fruit that opens along one or more sutures.
Identifying fragmented fruit and nut remains presents the analyst with a set of problems somewhat different from those encountered in seed identification. Published manuals or drawings are less useful as identification aids, because fragmented remains do not resemble illustrated whol...
