2.6: Ancient Foodways- Zooarchaeology and Paleoethnobotany

Steven R. Kuehn, MS, Illinois State Archaeological Survey

Rebecca K. Albert, MA, University of California, Santa Barbara

Emily Johnson, MA, University of California, Santa Barbara

Learning Objectives

• Define and describe zooarchaeology and paleoethnobotany
• Outline the primary methods and techniques used in faunal and floral analysis
• Discuss how zooarchaeological and paleoethnobotanical data provide information about past human behavior and the interrelationship between people and their environment

When studying the past, often the most basic questions about daily life prove to be the most challenging. What did people eat? Did they grow their own food or did they procure animals and plants from the environment around them? What sort of faunal and floral resources were available from the surrounding landscape? Did the plants and animals available vary seasonally and did they change over time? In addition to food, what other resources (e.g., hides, antler, wood, bone, medicine) did local plants and animals provide? What else can the animal and plant remains found at archaeological sites tell us about the people who lived there? And how do archaeologists analyze and interpret the data recorded in plant and animal remains in order to answer these questions?

Zooarchaeology and Paleoethnobotany

Animal and plant remains found at archaeological sites provide important insights about past human behavior and how humans interacted with their environment. The evidence obtained can tell us about past dietary practices, what environments were like in the past, how and when plants and animals were first domesticated, foraging strategies, and how plant and animal resources were used for medicines, tools, and ritual items. Analysis of past animal and plant remains are two specialized fields in archaeological research.

Zooarchaeology is the study of animal (zoology) remains from prehistoric and historic archaeological sites. The overall goal of zooarchaeological research is to develop a better understanding of how past peoples interacted with their environment, specifically with animals (fauna). Zooarchaeology is an interdisciplinary field of study that combines zoology and archaeology, with an emphasis on the latter. Fields such as ecology, biology, osteology, taphonomy, animal husbandry, and various others also contribute to zooarchaeological studies. Zooarchaeological or faunal remains are a reflection of past human behavior and living strategies. Analyzing animal remains from archaeological deposits can provide a wealth of information about how past peoples lived (e.g., Russell, 2012)—not just their behavior but also their impact on the environment they lived in.

Paleoethnobotany is the study of how people (ethno) used plants (botany) in the past (paleo), or how people interacted with and impacted their environment through the use, manipulation, and consumption of plants. Paleoethnobotany can be used to understand many things: what plants people ate, how people acquired plant foods, how people prepared plants for consumption, the origins of agriculture, what the environment was like at the site, and so on. Fields such as botany, ecology, agronomy, and soil science also greatly inform paleoethnobotanical studies.

Special Topics: Taxonomic Classification

All biological organisms are scientifically ranked and placed in defined categories in order to group together those sharing similar traits or common ancestry. These taxonomic ranks are hierarchical in structure, meaning that those higher up in the rank are similar in general ways, while those lower in rank are more closely related. The scientific names used in taxonomic classification are agreed upon by the scientific community (the International Commission on Zoological Nomenclature) so that there is mutual understanding between researchers and guidelines for creating and refining taxonomic ranks. The two most common terms used, genus and species, are written in italics and represent closely related species.

This table displays the taxonomic classification for the coyote and common sunflower:

Rank	Animal example	Plant example
Kingdom	Animalia	Plantae
Phylum (division in plant)	Chordata	Tracheophyta
Class	Mammalia	Magnoliopsida
Order	Carnivora	Asterales
Family	Canidae	Asteraceae
Genus	Canis	Helianthus
Species	latrans	annuus
Common name	coyote	common sunflower

An animal closely related to the coyote, such as a red fox, would be in the same ranks for the categories of kingdom through family but would differ by genus and species (red fox is Vulpes vulpes). In scientific research and reporting, authors regularly use the scientific names of animals and plants because common names can vary considerably, causing confusion. The Latin names used in taxonomic classification are accepted and understood worldwide.

At its core, archaeology consists of the analysis and interpretation of material culture left behind by past peoples. Faunal and floral remains utilized by prehistoric and historic peoples represent specialized material culture obtained from the living biotic environment (often termed ecofacts) and as such are often subsumed under the category of environmental archaeology. Obtaining data during floral and faunal analysis is only the initial step in the process of addressing research questions in archaeological studies. In order to gather significant information from faunal and floral remains, it is necessary to examine in some detail the methods and techniques employed.

Zooarchaeology Faunal Analysis

Zooarchaeological faunal analysis focuses on the study of preserved faunal or animal remains recovered from archaeological sites. Typically, the preserved remains consist of the hard parts of an animal such as bones, teeth, antler, shell, and scales. But also keep in mind that the category “faunal” includes more than the bones of mammals and other vertebrates; it also includes other parts of these animals such as scales, eggshells, rattles, and so on. Invertebrate animal parts may also survive in the archaeological record, such as remains of mollusks, echinoderms, and even arthropods. (That said, our main focus in this chapter will be bones from vertebrates.) Depending on various taphonomic factors, a representative sample of animal remains may preserve over time, forming a zooarchaeological or faunal assemblage for a particular site. Soft items (e.g., hair, skin, muscle, and marrow) only rarely preserve in the archaeological record, in settings such as unusually anoxic wet (e.g., peat bog) or arid (e.g., dry cave) conditions. During analysis, archaeologists examine each faunal remain or specimen individually and record a series of characteristic traits and conditions. This initial step focuses on identifying elements and taxonomy: essentially, identifying the part of the animal as well as the type of animal represented by each specimen. (See Special Topics: Taxonomic Classification, above.) Archaeologists use the information they obtain in this process to develop a zooarchaeological profile of the assemblage, which they can then use to address research questions regarding the human behavior that created the assemblage.

Animal Remains and Basic Osteology

In order to recognize the variety of faunal remains present in an archaeological assemblage, it is necessary to have a basic familiarity with a diverse array of animal skeletons. The animal kingdom can be divided into vertebrates and invertebrates. Vertebrates have backbones or spinal columns and at least some internal skeletal elements. They include mammals, birds, reptiles, amphibians, and fish. Invertebrates, in contrast, lack a spinal or vertebral column. Common invertebrates include mollusks (bivalves, gastropods, squids, octopi), arthropods (insects, arachnids, crustaceans), annelids (earthworms, leeches), and cnidarians (jellyfish, corals, sea anemones). Other than some mollusk and arthropod remains, remains of invertebrates rarely occur in the zooarchaeological record, so the primary focus in this chapter is on the analysis of vertebrate skeletal remains.

Figure 1. Three representative animal skeletons with the major elements labeled. A is a mammal (a dog), B is a bird (a generic chicken or duck), and C is a fish (a generic bass-like fish).

The typical vertebrate skeleton consists of axial and appendicular elements. The axial skeleton consists of the skull (cranium, mandibles, teeth, and hyoid bone) and the rib cage (ribs, sternum, and vertebrae). The appendicular skeleton includes the bones of the forelimbs and hindlimbs, the shoulder or pectoral girdle, and the hip or pelvic girdle. The number and character of the different axial and appendicular bones vary between taxa, so zooarchaeologists need to be familiar with the general osteological structures of and variations between mammals, birds, reptiles, amphibians, and fish (Figure 1). Each element differs in some respect between animals, and the differences are more pronounced between animals that are not closely related. For example, forelimb bones are similar between dogs and foxes but very different between dogs and beavers. Some animals, such as turtles and fish, have very distinctive skeletal elements that are easily recognized, but differences are still found between the same bones of different species (e.g., catfish bones are easily distinguished from those of freshwater bass). These differences can be used to make classifications between specimens at the class, genus, and species levels.

Mollusk or shellfish remains are some of the few invertebrate remains typically encountered in a zooarchaeological assemblage. Both bivalves and gastropods (univalves) have hard shells with distinct characteristics (e.g., shape, thickness, color, size and orientation of teeth, and presence of exterior bumps or ridges) that can be used for identification. Even when working with fragmentary pieces of shell, examination of multiple characteristics can often be used to distinguish between different species. Whether an archaeologist is working with vertebrate or invertebrate remains, however, the poor condition of the specimens often prevents identification to a specific element or distinct taxon. Faunal remains should never be assigned a level of identification that is not wholly supported by the available evidence.

Comparative Specimens and Reference Guides

The analysis of zooarchaeological remains begins with recognizing and identifying specimens. Each piece of bone or shell is examined individually and compared to known specimens, either in a comparative collection or in reference manuals. A faunal comparative collection consists of elements from individual animals, all properly cleaned and labeled, that can be used to help an archaeologist identify typically fragmented or incomplete zooarchaeological remains. Reference manuals consist of books, journal articles, and similar publications that provide detailed illustrations (drawings or photographs) of skeletal elements that can be used to identify a specimen. In general, comparative collections are preferable to reference manuals. It is often difficult to obtain the precision necessary to accurately recognize small, distinct traits visible on skeletal elements through drawings or photographs. In addition, many skeletal remains can be quite small, making them difficult to illustrate. Finally, no reference manuals exist that contain detailed illustrations of every bone from every animal found in a given region.

Identifying and Analyzing Specimens

As part of the analysis, archaeologists record the following data for each piece of bone: element, side of the body (when applicable), portion or section of the element present, the age of the animal, and taxonomic classification, to the most precise level possible. For age, an approximation generally equivalent to juvenile or adult is used, based on epiphyseal closure of the long bones, tooth eruption, occlusal wear on the chewing surfaces of the teeth, or similar traits (O’Connor, 2000, pp. 80–97). (See Special Topics: Aging and Sexing Skeletal Elements.) Mollusk shells are treated in a similar manner; the element, portion, and taxon is recorded for each specimen.

Special Topics: Aging and Sexing Skeletal Elements

Animal skeletons grow and change throughout the life of the animal. With careful observation, it is possible to recognize and note these changes and use this information to determine the approximate age of the animal at the time of death. The most common of these is tooth eruption. In mammals, for example, deciduous or milk teeth erupt from the gumline first, then are pushed out and replaced by the permanent or adult teeth. Because this process occurs at a regular rate, it is possible to roughly determine an individual’s age based on the eruption and growth of the teeth present. Once adult teeth are fully erupted, they remain with the animal through the rest of their life, but they wear down through use over time. As teeth grind together and chew up food items, the bumps and ridges on the occlusal (chewing) surface of the teeth wear down. With some variation, this occlusal wear can be used to determine age. Diet can affect the results (e.g., eating coarse foodstuffs like grasses can wear teeth down faster), but in general, greater occlusal wear is found in older individuals.

Some skeletal elements are made up of smaller pieces of bones, and the fusion of these pieces can also reflect age. As juveniles, mammal skulls consist of a series of smaller bones that fuse together when the animal is an adult. The sutures, where the smaller bones fuse together, can provide evidence of age. Appendicular limb or long bones also change during development. In juveniles, long bones consist of a bony diaphysis and two epiphyses (one on each end) that are separated by cartilage. As the bones grow, the cartilage disappears and the ends of the diaphysis fuse with the two epiphyses, forming a complete, intact bone. Depending on how much fusion has occurred, this can be used to age the element. Each long bone element fuses at a different rate, so each bone can be used to determine specific ages until the animal reaches adulthood. With experience, an analyst can also recognize muscle-attachment wear, arthritis, and other physical signs typically associated with aged individuals.

Skeletal elements are constant within a given species, but for some animals there are differences between males and females. This is an example of sexual dimorphism, where two sexes of the same species display different characteristics. In certain animals, for example, males have a baculum or penis bone, an element not found in females of that species. In some cervid (deer and elk) taxa, males grow antlers while females do not. Thus, if you find a deer cranium with antlers attached, you can recognize it as a male. In other animals such as caribou, however, both males and females grow antlers, although those of the males tend to be larger and more robust. In other animals, the size and shape of horns and antlers can be used to differentiate males from females.

Other skeletal elements may also differ between the sexes, but these differences are not always mutually exclusive. Elements in males are typically larger and more robust than the same elements in females. For example, canine teeth in male carnivores are usually larger than those in female carnivores, but a large female could have elements larger than those of a small male. This problem is compounded when examining fragmentary specimens, which are most common in an archaeological faunal assemblage. Identifying these traits or finding a baculum, for instance, would allow the analyst to recognize male or female elements in a zooarchaeological assemblage. In general, relatively few zooarchaeological remains can be securely identified to sex.

Each faunal specimen is also examined for evidence of noncultural and cultural modification. Noncultural modification—that is, alterations caused by nonhuman agents—includes such things as weathering, rodent and carnivore gnawing, and similar taphonomic effects (see the chapter on Ancient Environments; Lyman, 1994). These modifications can impact bone preservation and identifiability, so understanding the nature of these impacts aids in accurately documenting and interpreting faunal assemblages. Cultural modifications include those resulting from direct and indirect human activity. Possible cultural modifications include whether or not a specimen shows evidence of burning, the presence and types of butchery marks, and modification relating to the manufacture of bone or shell tools, ornaments, and similar items. Butchery marks include cut marks, fractures, splinters, and other damage caused by carcass processing (e.g., Binford, 1981, pp. 87–181). In the past, people used bones, antler, shell, and other animal parts as raw materials for a wide array of implements and items used in daily and ceremonial activities. Modified faunal remains can be recognized by cut, scrape, and/or polish marks that resulted from impacts that changed the original shape of the element.

Paleopathologies are a unique situation, as they can result from internal (e.g., disease) or external (e.g., trauma) factors. External factors can be caused by human behavior (Baker and Brothwell, 1980; Bartosiewicz, 2013). Animal husbandry, or the domestication and breeding of certain taxa, can also result in changes to animal bone that can be recognized in the zooarchaeological record (O’Connor, 2000, pp. 147–159). For example, evidence at ancient sites in South America show that domesticated camelids, such as alpacas and llamas, showed changes in dental morphology (tooth shape) after domestication (Wheeler, 1995).

Special Topics: Identifying Microfauna

Most skeletal remains are visible to the naked eye and can be observed and studied simply by looking at them. Some smaller remains, such as fish scales or rodent bones, can be examined more easily using a magnifying lens or loupe. Certain faunal remains, such as insect parts, are difficult if not impossible to distinguish without visual enhancement. Magnification can also make it easier to identify butchery marks, evidence of tool manufacture, and pathologies.

If identification of microfaunal remains is so challenging, then is it necessary to do it? In truth, many small fragments or crumbs of bone are often unidentifiable to element or taxon and provide little if any substantial information. Other remains, however, can provide a wealth of knowledge for the observant analyst. Small rodent teeth, for example, can be difficult to identify unless they are examined under magnification, but they are exceptional indicators of the environment and the climate in which the animal lived. Rodents represent approximately 40% of all mammalian species on Earth and occur naturally on every continent except Antarctica. They have relatively short life spans, occur in a wide range of biomes, and are sensitive to changes in climate and vegetation. This information can be particularly useful if a scientist wishes to know more about paleoenvironmental conditions and changes over time. Because of this, putting extra effort into the study and analysis of rodent teeth can yield great rewards.

Recovery Methods, Sampling, and Context

A faunal assemblage is not a perfect representation of all of the animal remains present at an archaeological site. Rather, it is a biased sample. Not all faunal material preserves over time. Acidic soil conditions, for example, can result in the destruction of more fragile animal remains. In addition, recovery techniques (e.g., hand collection, screening, or flotation) can impact the composition of an assemblage. Flotation through fine-grained mesh will undoubtedly recover more microfaunal remains than screening through one-quarter-inch mesh, for example. Finally, it is rare that an entire archaeological site is completely excavated. Because of this, only the faunal material present in the areas investigated will ever make it to the zooarchaeologist for analysis.

Context is likewise an important consideration. It is typically assumed that the majority of faunal material recovered from an archaeological site represents the remains of animals used by the site inhabitants. Some bone and shell recovered may be intrusive, meaning that it was deposited before or after human habitation of the site. In addition, some animal remains are commensal; that is, they represent animals that live in or near human habitations but are not directly associated with human activity. For example, in any given backyard there may be mice, voles, snakes, frogs, snails, and other animals that are not noticed by the people living in the house associated with that yard.

Special Topics: Indirect Evidence of Human-Animal Interaction

In addition to bone, shell, teeth, and other faunal remains, indirect evidence can be used to recognize the presence of different animals at prehistoric and historic archaeological sites. Rock art paintings and carvings, for example, often portray various animals that were important to the human inhabitants. Gnaw marks found on recovered bones can indicate the presence of dogs or rodents even if their physical remains do not occur in a zooarchaeological assemblage. Animal tracks, although rare, are on occasion found at archaeological sites and can likewise demonstrate the presence of certain animals. The paw prints of dogs and cats, for example, have been found impressed on roofing tiles at sites from the Roman Empire. One clay roof tile in a Roman-era site near a highway in Lincolnshire, UK, shows evidence of a cat paw print. The tile suggests that the owners of the house were wealthy, since only wealthier people could afford to use that kind of material (Daley, 2017). Material culture relating to animal domestication, such as bridle and tack hardware from horses, can be used to recognize the presence of horses at a site even if no horse bones are recovered. Finding a pet collar or a discarded bird cage in a historic privy could likewise indicate that the residents of the adjacent home once kept a variety of pets.

Insect nests, such as those from mud daubers, are occasionally found among the remnants of burned prehistoric structures in the southeastern United States and demonstrate the presence of insect pests in human settlements. Paleofeces (e.g., dehydrated fecal remains, typically found in arid or anaerobic conditions) are one uncommon type of indirect evidence found at archaeological sites. Both human and dog paleofeces were found at several sites in the American Bottom flood plain region of Illinois. Analysis of the faunal remains contained in the dog paleofeces provided direct evidence for dog diet and showed that the dogs ate various fish, amphibians, and small mammals. These examples represent only a small sample of the instances in which indirect evidence provide insight into human-animal interaction in the archaeological record.

Paleoethnobotany and Floral Analysis

Paleoethnobotany is usually split into two methods of analysis based on the size and type of plant remains being studied: macrobotanical analysis and microbotanical analysis. Macrobotanicalanalysis deals with plant remains such as carbonized seeds and plant parts that you can see with your eyes or with the aid of a low-powered microscope. Microbotanical analysis deals with plant remains that you can see only with the aid of a high-powered microscope, such as pollen, starch, and phytoliths. Starch , a semicrystalline carbohydrate produced by plants, is helpful for plant identification because of morphological (shape and size) variation between different families and genera of plants (Messner, 2011). Phytoliths are made of silica (glass, or sand) and help give plants structure by forming in between cell walls when plants take up silica from groundwater. As with starch, phytoliths can be identified to plant family or genera based on their unique shape and size. Starches are found in carbohydrate-rich plants such as beans, tubers, maize, and cereal grains, while phytoliths are found in greater quantities in the grass family (Piperno, 2006). In fact, many grasses feel rough to the touch because of phytoliths. Pollen is a fine powder that plants make when they reproduce. Plant and tree pollen is dispersed by the wind and settles into soil or on top of bodies of water, such as lakes, where it falls to the bottom and accumulates in the sediment over time. Archaeologists can take core samples of the sediment to investigate changes in pollen type and abundance throughout time. Pollen morphology is also unique to families and genera.

This section will broadly cover a few of the sampling, processing, and analytical methods paleoethnobotanists use to learn about plant use in the past. While macrobotanical and microbotanical data can be used together to answer similar social and environmental questions, the ways that each type of botanical data is recovered and processed can be very different.

Taking into consideration the environment and conditions at a site can help archaeologists determine the suitability of certain sampling strategies. As with zooarchaeological remains, sampling strategies are often determined or guided by the unique taphonomy at an archaeological site. For example, botanical remains in archaeological contexts can be affected by the pH of the soil and whether the botanical sample has been carbonized, waterlogged, or desiccated (completely dried out). Botanical remains that have been carbonized or desiccated tend to preserve in archaeological soils much better than samples that have not. On the opposite side of the spectrum, samples that have been waterlogged—that is, that have continually been in water protected from the effects of oxygen—also tend to preserve very well. However, once such samples are removed from water, they can deteriorate quickly. Botanical remains in more acidic soils risk being dissolved at a faster rate, especially if they are not carbonized or desiccated. All of these factors guide what method archaeologists choose to recover botanical remains.

Sampling Methods

Paleoethnobotanical analysis begins with sampling from a field site. Depending on whether an analyst will be doing macrobotanical or microbotanical recovery, these samples can range from soil to residues on artifacts to dental calculus. Macrobotanical data are derived almost entirely from plant remains found in soil (except for the cases where botanical remains are desiccated or preserved in waterlogged contexts), while microbotanical data can be derived from remains isolated from both microbotanical and macrobotanical sampling methods.

Soil Sampling

Macrobotanical analysis begins with sampling soil both across an archaeological site and within individual features or units. Important factors to consider when collecting soil samples include 1) what questions researchers want to ask from the data; 2) the context of the site and its features; and 3) the archaeologist’s time, resources, or personnel constraints. But how much soil or residue do you need for a sample and how does sample size differ based on the analyses someone wants to make?

Sampling across a site generally takes the form of one or a combination of four methods: blanket sampling, probabilistic sampling, random sampling, and systematic sampling (Guedes and Spengler 2014). Generally, blanket sampling (also called full coverage or total sampling) is the preferred sampling method because it includes soil samples taken from all units and individual features. This reduces the possible bias introduced by other sampling methods (Pearsall, 2000). For a further discussion of how factors like time, personnel, person power, and site size might dictate one of the other sampling strategies, see Guedes and Spengler (2014).

In addition, archaeologists must also consider how best to sample within a feature or a stratigraphic level. The three most common strategies include bulk sampling (collecting soil from a single layer within a single context); pinch sampling (collecting a small amount of soil from various spots and combining these samples into one larger sample, usually done at sites that consist of large cultural layers or levels of construction fill); and column sampling (taking soil in a line down a single column at each unique stratigraphic level). Unlike site-wide sampling methods, in which only one strategy is used, it is common to use all three feature-sampling methods listed above for features and contexts.

For macrobotanical (e.g, wood, charcoal, seeds) analysis, the standard practice is to take 10 L of soil for each context sampled (D’Alpoim et al. 2014). This amount of soil is large enough to be representative of the context as a whole. If a researcher wants to answer questions that also require microbotanical analysis, a subsample of soil from the 10 L should be taken at the same time so that both the macrobotanical and microbotanical data from the site can be fully integrated. Microbotanical analysis only requires a small amount of soil—roughly 100 mL to 300 mL (or half a cup).

Other Sampling Methods

Other common sources for collecting microbotanical data include dental calculus and residues on artifacts and dental calculus. Dental calculus is mineralized plaque that accumulates in layers on teeth, collected from both animal and human remains using a sanitized dental pick or sonicating toothbrush in a lab setting. Microbotanical analysis of dental calculus can give paleoethnobotanists direct information about what a person was eating, unlike the previous methods mentioned, which are only secondary sources of information about diet or how people were using plants in their day-to-day life. Residues include the remains of carbonized food from the inside of ceramic sherds, absorbed material from a ceramic sherd, or material from the cutting edges of lithics. Residues from ceramic sherds can tell archaeologists what people were cooking or storing in ceramic vessels, while residues from lithics can tell archaeologists what people were cutting with these tools. These residues are typically collected in the lab using a sanitized sonicating toothbrush. Pollen data are recovered by taking a soil core from lake sediment and sampling each stratigraphic level. Pollen is used to reconstruct ancient environments.

Processing Soil for Macrobotanical Remains

The process for recovering macrobotanical remains from soil is called flotation. While there are several different methods of flotation, the goal of the process is to separate the dense material in the sediment (the soil matrix itself, rocks, bone, stone tool flakes, etc.) from the light material (wood charcoal, carbonized and modern plant seeds, plant parts, etc.). These light and heavy fractions are separated by adding the soil sample to water and agitating to dissolve the soil into solution, then letting the heavy fraction sink and letting the light fraction float. The light fraction is skimmed off of the top and collected in a small piece of chiffon gauze that allows only water to pass through (Figure 2). The remaining heavy fraction is collected in a 2-millimeter mesh screen. Both the light and heavy fractions are dried away from the sun to avoid heat-induced popping or stress on botanical material; this may take two to three days. After drying, the heavy fraction is sorted to ensure that no plant material remains. Any remaining botanical material is added to the dried light fraction and labeled and sealed in plastic bags for future analysis.

Special Topics: Different Types of Flotation and How Archaeologists Use Them

Flotation is an important step in macrobotanical analysis that paleoethnobotanists use to separate macrobotanical remains such as seeds and wood charcoal from other soil components. While the principle behind flotation is simple—botanical remains will float because they are lighter than water and bone, stone, pottery, and lithics will sink because they are heavier—there are several ways to go about floating soil samples depending on the size of the samples, access to water, the budget for the project, and other factors.

Figure 2. Flotation process using a SMAP-style machine. A) View of the machine. B) Putting soil on the window mesh in the tank. C) Water comes up from the bottom of the tank and soil is manually agitated by hand to help it break up and go into solution. D) Light-fraction botanical material that was skimmed off the top. Photos courtesy of the authors.

Bucket flotationthe simplest method of flotation, can be done just about anywhere. As the name suggests, with bucket flotation an archaeologist puts a small portion of the soil sample in a bucket, adds water, and gently stirs with their hands to break up the dirt and encourage it to go into solution. Next, they slowly decant the water and botanical remains floating on top into another bucket lined with chiffon to catch all of the botanical material. More water is added to the bucket to dissolve more soil and force more botanical remains to float and the process is repeated until no more botanical remains float to the surface. The heavy fraction left in the bucket is then saved in window-screen mesh to look at later for other archaeological material. The pros of bucket flotation are that it requires minimal equipment that is readily available in the field and uses much less water than other methods of flotation. However, it can be a very tedious process if soil samples are large in volume or there are hundreds of samples to process, and it can be difficult to break up soils with a heavy clay content.

SMAP flotation is the next most popular method. It is named after the project that created it, the Shell Mound Archaeological Project. This style of flotation requires the construction of a simple machine (Figure 2) that forces water to rise from below into a large tank or bin and pushes the light botanical material to the top of the tank, where it can be skimmed off the top in the same manner as in bucket flotation. SMAP machines allow archaeologists to process large volumes of soil and can be inexpensive to make. However, they require some expertise to build and cannot recycle water in areas where it is scarce. Lastly, a Flote-Tech machine runs in the same way as a SMAP machine. It can also process large volumes of soil, but the difference is that it allows water to be recycled. However, it is the most expensive method to build.

Each method of flotation has its benefits and drawbacks, and the machine that works best for a project will ultimately be decided by the specific context of the site, the soil, the budget, and available materials.

Analysis of Macrobotanical Remains

In the laboratory, analysis of macrobotanical remains requires only a few simple tools. A geological sieve is necessary to separate the light fraction into different sizes of classes, both to simplify the analysis and to ensure that small seeds are not lost within large pieces of grass or fine silt sediment. The most common sieve sizes are 2 mm, 1 mm, 0.5 mm, and finally the pan, which catches the remaining material, but the size that is used varies by region and by researcher. For example, some projects use 2 mm, 1.4 mm, 0.7 mm, and pan, while some regions without many small seeds skip the 1.4 mm fraction entirely.

A scale is used to weigh the entire light fraction as well as each size class from the geological sieves. This allows the researcher to calculate statistics such as densities and ratios. This information is valuable for answering archaeological questions (see the section titled Paleoethnobotanical Data below for more information on common statistical approaches to macrobotanical data). Finally, a low-powered stereo microscope (also called a dissecting microscope) allows analysts to better see small seeds and unique morphologies that need to be identified for later analysis.

Wood is one of the most ubiquitous botanical remains a paleoethnobotanist will come across in both the field and the laboratory due to its abundant use as a source of fuel and shelter. Wood is primarily encountered as wood charcoal that has been preserved by fire through incomplete burning, either accidentally or intentionally. In rarer cases, at sites with extreme weather conditions a paleoethnobotanist may encounter wood samples that are waterlogged or desiccated. Because wood charcoal is often highly fragmented and it would be impossible to know whether the wood in your sample is from one branch or ten, it is only ever weighed rather than counted. In addition, data for wood charcoal is only recorded at the 2 mm fraction and above since these data are representative of the entire sample. The small amounts of charcoal remaining in the other, smaller fractions are negligible.

Seeds are the primary material paleoethnobotanists use to answer anthropological questions regarding past peoples’ foodways, environment, economy, and more. All seeds 1 mm and larger, both whole and broken, are counted and weighed. Generally, seeds below 1 mm in size are counted only if they are whole, although counting practices may vary by project. Seeds smaller than 1 mm are not weighed because they are too light to register on the scale; for these seeds, archaeologists record only count data.

When identifying seeds, it is important to identify distinct morphological features such as size, general shape, surface texture, appearance of the fractured seed compared to the appearance of the whole seed, and more. Much of this knowledge will come from analyzing samples with a skilled graduate student or professor and by comparing the appearance of the sample to online, printed, and physical reference collections. Popular online reference manuals include the Digital Plant Atlas, the website for the Ohio State University Department of Horticulture and Crop Science, and the Searchable Seed Image and Illustration Gallery. Popular books that have useful images and descriptions of key family characteristics include the Seed Identification Manual (Martin and Barkley, 2004) and Botanical Macro-remains (Schoch et al., 1988).

Analysis of Microbotanical Remains

After samples are collected, they are chemically processed. The process that is used is based on the type of analysis the researcher is doing. For example, starch remains are more delicate than phytoliths or pollen. The differences in the robustness or fragility of samples is why the same method cannot be used to extract all types of microremains. Chemical processing often involves the separation of materials, often by using specific chemicals to dissolve parts of the sample while leaving others intact. Chemical floatation can also be used. Just like standard soil flotation, chemical flotation uses differences in material densities to separate the microbotanical remains from their matrix. This process involves chemicals other than water so that specific densities can be achieved to separate the microremains. Methods for processing these samples can vary greatly, and each lab that analyzes microbotanical samples has a standardized method for each sample type.

Figure 3. Photomicrograph of maize starch granules in nonpolarized light (left) and polarized light (right). Scale bar is 100 microns (0.1 mm). Photo courtesy of the authors.

After samples are chemically processed, drops of the isolated microbotanical sample are transferred to a slide using a pipette and mixed with a mounting medium. These slides are then analyzed under a high-powered microscope with filters using polarizing or colored light. Polarizing filters make it possible to identify starches because they have a distinctive cross shape under polarized light (Figure 3). Colored filters can increase the contrast between the background of the field of view and the microbotanical remain being viewed.

Figure 4. Photomicrograph of a maize phytolith. A maize (Zea mays spp. mays) rondel phytolith is circled in red. Source: Albert et al. (2018).

[/caption]Usually, the analyst needs to be able to recognize unique combinations of shape, features, and size to identify a sample to the level of species or genus. For phytoliths (Figure 4), these shapes include rondel, cross, bilobate, polylobate, bulliform, and hair cells. Sometimes different shapes and sizes of phytoliths come from different parts of the plant (Madella et al., 2005; Piperno, 2006). The shapes of starches can include lenticular, bell, elliptical, or oval, and their surface features can include fissures, margins, lamellae, or hila (Messner, 2011). As is the case with macrobotanical and zooarchaeological analysis, a comparative collection, reference manuals, or online reference websites can help an archaeologist identify these diagnostic features, and the constraints of these different tools for macrobotanical analysis are the same for microbotanical analysis. For commonly utilized reference manuals and online databases for microbotanical analysis, please see the For Further Information section below.

Zooarchaeological Data

Quantitative Data, NISP, and MNI

After specimens have been identified, the faunal remains need to be quantified (or counted) in such a way that useful, meaningful data are collected. Quantitative data are used to measure the relative abundance of each taxa so that zooarchaeologists can determine what animals humans were using, how important each animal was in comparison to other animals used, and what the presence of animal remains might reveal about past environmental conditions. The information recorded can be of varying usefulness when analyzing a faunal assemblage. Certain basic data are typically recorded for all faunal specimens because other types of data can be calculated based on that information. The basic data that is always recorded is referred to as primary data and the data that is calculated based on primary data is called secondary data (see Grayson, 1984; Lyman, 2008; Reitz and Wing, 1999, pp. 142–238).

The most common primary data that is recorded is the Number of Identified Specimens per taxon or NISP. The NISP documents the number of specimens identified to each element for a given taxon. For example, if an assemblage contains three rabbit molars, two rabbit vertebrae, nine raccoon rib fragments, and seven raccoon humerus pieces, the NISP for rabbit would be five and the NISP for raccoon would be 16. The taxon does not have to be specific, however; if six fragments of long-bone shafts are classified as unidentifiable large mammal, then the NISP for large-sized mammal would be six. It should also be noted that NISP counts recognize partial elements as well as side of the body the element came from, the age of the animal, the size of the element, and other factors. Suppose that the same assemblage contains the following deer bones: one left femur, proximal end only; one left femur, distal end only; two right femora, both complete; and one femur, mid-shaft portion only. The deer NISP is five because each specimen is counted separately, even though some of the femur fragments may be from the same element. The NISP is the basic unit of data that most other quantitative measures in faunal analysis are derived from.

Figure 5. A sampling of tarsometatarsus elements from a domestic chicken (Gallus gallus). The top row has seven right tarsometatarsi (all adult) and the lower level has five left tarsometatarsi (three to the left are adult, the two on the right are juvenile). The specimen on the far left is a right tarsometatarsus from a comparative specimen. Based on these elements, the NISP=12. Based on sided elements only, you get a conservative MNI=7 based on the presence of seven right tarsometatarsi. However, the actual MNI for these elements is nine because there are seven adult right tarsometatarsi and two juvenile left tarsometatarsi..

One drawback of the NISP is that on the surface, it can overestimate the presence and therefore the relative importance of a particular taxon. Using the above example, there are more raccoon bones (NISP=16) than rabbit (NISP=5) or deer (NISP=5) bones in the assemblage, which would seem to suggest that raccoons were used to a greater extent than rabbits or deer. The NISP does not reflect the number of individuals represented in the assemblage, only the number of specimens or pieces of bone. To compare individuals, the analyst must calculate the Minimum Number of Individuals per taxon, or MNI. MNI calculations are based on repeating elements from the same taxon (Figure 5). If an assemblage contained four right ulnae and two left ulnae identifiable as muskrat, the MNI for muskrat would be four. Minimally, at least four individual muskrats must be represented, as each animal has only one right ulna. There could be as many as six muskrats represented if the two lefts were from totally separate animals, but this cannot be determined with certainty. Having multiple skeletal elements from animals of different ages (e.g., both adults and juveniles) can also affect MNI counts, as adult and juvenile ulnae, for example, would represent separate animals. In general, paired elements are the most useful in determining MNI; bones that are similar and/or occur in relatively large numbers, such as ribs, are less suitable and less likely to provide significant results.

The MNI is usually calculated only for specimens that can be identified to the genus and/or species level, but some exceptions are sometimes made (see Reitz and Wing, 1999, pp. 198–199). This means that MNI estimates are not made for general classification categories (e.g., “indeterminate fish” or “indeterminate mammal”). As a result, potentially significant information on faunal representation may be lost if only the MNI is used. If, for example, an assemblage contains 500 fish bones but none are identifiable to the genus level, then none will be included if the analyst focuses only on MNI results. MNI can also be impacted by the state of preservation to a greater extent than NISP. When bones are poorly preserved, it is less likely that they can be classified to the genus/species level. In most faunal analyses, both NISP counts and MNI determinations are included, so that complementary lines of information are available.

Secondary Data

In addition to NISP and MNI, an array of other data can be gathered from a faunal assemblage. Most of these, however, are drawn from or derive from NISP or MNI calculations or do not provide information that reflects past human behavior. Specimen weight (used to calculate biomass) may be recorded, for example, but the results have more to do with the initial size and robustness of the bone and do not necessarily reflect relative dietary importance. For example, one deer bone might weigh as much as several hundred small fish bones. That does not mean, however, that deer were necessarily a more important food item, although the raw weight values would seem to support such an argument. In contrast, the NISP and MNI counts would suggest that fish were of greater dietary value than deer. Lyman (2008, pp. 84–113) and Reitz and Wing (1999, 222–231) provide excellent summaries of the pros and cons of various types of primary and secondary zooarchaeological data (see also Grayson, 1984).

Paleoethnobotanical Data

After paleoethnobotanical data are quantified (counted and/or weighed), additional statistical methods can be used to examine patterns within the data (Marston, 2014). The simplest way to report botanical data is through absolute counts, which are the raw counts for each individual taxon identified. Other descriptive statistics frequently used are:

1. Density calculations: Seed or charcoal counts/weights are compared against the volume of the original soil sample taken in the field to provide comparability among all samples taken at a site.
2. Proportions/percentages: These calculations are used to standardize the count of one taxon compared to another category of taxon. For example, a comparison of the ratio between maize kernels and maize cupules (the cup-like structure on the cob that holds the maize kernel) at a site can tell you where ancient people were processing dried maize (as a high cupule ratio would indicate) and where they were using the kernels for cooking (as a high kernel ratio would indicate).
3. Ubiquity: Data that records presence or absence are used to calculate the percentage of samples in which a given taxon appears. For example, if you have 10 samples and maize is found in 6 of them, the ubiquity value of maize is 0.60 and tells us maize is present in 60% of samples.

Other types of statistics in paleoethnobotanical reports often include Z-scores (data that is transformed into standard deviations around a population mean), diversity indices (calculated based on the relative values of taxa to indicate homogeneity or heterogeneity of a sample), and more complicated multivariate calculations such as cluster analysis, principal components analysis, and correspondence analysis (see Smith, 2014 for a discussion and examples of multivariate statistics in paleoethnobotany). These statistics can be presented in a number of ways to highlight the patterns or trends analyses reveal.

Ancient Foodways: What Can Zooarchaeology and Paleoethnobotany Reveal?

Faunal and floral remains from archaeological sites can provide a wealth of information about past human behavior and interaction with an environment. Some of the more common areas of study are addressed here, but ongoing research is discovering ever more applications for zooarchaeological and paleoethnobotanical data.

Dietary Patterns

Zooarchaeological and paleoethnobotanical material provide direct evidence for the diet of prehistoric and historic people. Although faunal and floral remains that are recovered do not reflect all of the food refuse people discarded in the past, they provide a sample of what people were eating. This information provides important insight into how animals and plants were processed for consumption and how human dietary strategies changed over time. Not all faunal and floral remains recovered, however, represent food refuse; this issue is addressed in detail below.

Recovered floral remains demonstrate what sorts of wild plants people gathered, what domesticated crops they grew, and how important various taxa were to the site inhabitants. Similarly, the faunal remains recognized reveal what wild animals inhabitants procured, what domesticated animals they used, and the importance of different animals to the inhabitants. The proportional amounts of animal and plant remains illustrates their relative significance in the diet. By examining assemblages from different sites and different time periods, it is possible to see how dietary strategies varied between regions and changed over time.

In addition, recognizing patterns in what people were eating and how they were processing those types of foods can inform archaeologists about cuisine and the foodways (cultures that surrounded food) at archaeological sites. Combining dietary analysis with ceramic and lithic use-wear analyses, for example, enables archaeologists to use multiple lines of evidence to acquire a more complete understanding of food consumption as well as the social ramifications of dining, feasting, and gender roles (e.g., Kooiman, 2018; Twiss, 2012).

Foraging Strategies

The types of animals and plants recovered at an archaeological site can provide insight into the hunting and foraging practices a site’s inhabitants used. Certain faunal and floral species are available or abundant only at certain times of the year and human foragers likely optimized their efforts by focusing on those taxa only when they were likely to yield high returns. For example, nut trees such as oaks produce ripe acorns only in the fall, so foraging groups likely scheduled their seasonal travels so that they were close to oak forests during the autumn months. Likewise, these same mobile foragers might have planned to camp near large lakes or rivers in the spring, when spawning fish were in the shallows and could be caught more easily. By recognizing the time of year when different taxa are available, archaeologists can often determine the seasonality (season of occupation) for a site.

The relative abundance of different animal and plant remains also provides information about foraging behavior. Some hunter-gatherer groups (e.g., big-game hunters) focused their efforts on a select range of large animals, although most groups employed a broad-based subsistence strategy in which they procured a wide array of animals and plants to fulfill their dietary needs. Other groups may have adjusted their foraging strategies on a seasonal basis in order to take advantage of available abundant resources. Agricultural communities, for example, put considerable effort into farming during the spring planting and fall harvest seasons, but during the other seasons they supplemented their diets through hunting, fishing, and foraging.

Paleoenvironmental Reconstruction

Different types of plants prefer different environmental conditions, so the presence, absence, disappearance, or appearance of plant species through time at a single site may indicate changes in an environment such as drought, desertification, increased precipitation, or changing temperatures. These changes in local habitat can also be indicative of human actions that impact the environment such as overgrazing, tree harvesting, agriculture, and environmental management (e.g., Asouti and Austin, 2005). One method of recognizing these changes is through analysis of pollen cores, which can reveal distinct changes in environmental conditions over long periods of time. (See Special Topics: Pollen Coring and Environmental Interactions, below.)

Similar information can be derived from analysis of the different types of fauna represented at a site. Depending on the sensitivity of the taxa, it is possible to reconstruct what a habitat was like during the period a site was occupied. Many animal species flourish only within restricted habitats, and these taxa can be particularly useful when examining environmental changes over time.

Special Topics: Pollen Coring and Environmental Interactions

How can archaeologists, environmental scientists, and geographers learn more about the environment surrounding a lake and the actions of the people occupying the areas around that lake? The analysis of pollen, known as palynology, can help us understand environmental changes on a local level. Palynologists collect sediment core samples from the floors of bodies of water and analyze the pollen, the male gamete of plants, within those samples. Unlike other microbotanical remains, pollen can usually be identified to the species level and therefore can give palynologists a detailed picture of the surrounding plant environment. Pollen also collects in layers of sediments at the bottom of bodies of water chronologically. The pollen at the bottom of a core sample will be the oldest and the pollen at the top of a core sample will be the youngest (MacDonald and Edwards, 1991). Cores collected from these bodies of water can vary in length depending on what periods of time the analysts are interested in and how fast these layers accumulate locally.

After these cores are collected, palynologists sample these cores for radiocarbon dating, also called Accelerated Mass Spectrometry dating, and process the remaining samples to dissolve the sediment, usually with a chemical like hydrofluoric acid. After processing, palynologists identify and quantify all the pollen from the samples in order to understand the surrounding plant community. From these dated samples, palynologists can model the changes in the plant environment surrounding the body of water throughout time and what factors may have been at play. For example, if there was an increase in pollen from agricultural crops, such as maize, in tandem with a decrease in different types of tree pollens, that would indicate that there was deforestation for agricultural purposes (Alenius et al., 2012).

Plant and Animal Domestication

Tracking changes in seed size (from smaller to larger) across time is one indication that a plant may be undergoing domestication. The recovery of evidence of nonlocal plants in archaeological contexts, such as maize in eastern North America (e.g., Larson et al., 2014), can be evidence of domestication (and of interregional exchange, as discussed below). Similar changes are often seen among domesticated animals, which typically exhibit physical traits that reflect breeding efforts to increase meat production or reduce aggression. It can take many generations for these changes to manifest themselves skeletally, but they can be recognized in the archaeological record over time.

Butchery Practices

Close examination of animal bones may reveal evidence of butchery, such as cut and chop marks, but the bones themselves also provide insight on how carcasses were butchered and processed. First, the location and types of marks identified are direct evidence of the specific manner in which animal carcasses were reduced in size and portioned for further processing. Butchery marks are commonly found near limb joints and muscle attachment points, evidence of the removal of meat-rich limb portions. Other types of cuts, such as those found in areas with little muscle, may represent marks associated with hide or pelt removal.

The bones themselves—and the absence of some bones—are likewise indicative of choices made during the butchering process. If an animal is killed at a great distance from the habitation site, for instance, hunters may choose to leave low-value lower limb or cranial portions at the kill site. As a result, lower leg and skull bones may be underrepresented in the zooarchaeological record of the habitation but may be much more common at kill locations. A large percentage of high-value bones (those supporting large amounts of meat or choice cuts) in an assemblage may indicate the differential distribution of meat resources, possibly reflecting tribute, feasting, or allocation of foodstuffs based on status. Finally, heavily fragmented appendicular bones can also be indicative of marrow extraction and the manufacture of bone grease.

Trade and Resource Movement

The recovery of faunal or floral remains from species not typically found in an area can provide evidence of trade, population mobility, and similar movement of people and resources. For example, marine whelk shells and shark teeth are occasionally found at prehistoric sites across the eastern United States, far inland from coastal areas. These items were likely transported, either directly or indirectly, by traders and travelers moving between these areas. The recovery of tropical plant species in temperate zones (e.g., maize, as noted above) reflects not only interregional trade but also evidence of domestication. Exotic items were undoubtedly highly valued and likely encouraged further contact and commerce. It is likely that many other goods, most of which were perishable and are thus not recognized in the archaeological record, were exchanged between neighboring groups.

Nondietary Use of Faunal and Floral Material

The nondietary use of faunal and floral resources is often overlooked. Most research is focused on dietary strategies and reconstructions of a paleo-environment. A wide range of nondietary uses is recognized (e.g., Russell, 2012), many more than can be addressed in this chapter. A sample of these uses is presented here.

Plant resources were widely used for nondietary purposes but are often difficult to identify because they typically do not preserve well in the archaeological record. Wood was used extensively for construction projects, tools, bowls, utensils, weapons, canoes, and numerous other items in addition to being burned for fuel. Grass, reeds, plant fiber, and pine needles were used to create cordage, woven basketry, mats, and rugs (e.g., Nichols et al., 2000). Storage containers were manufactured from dried gourds such as the bottle gourd (Lagenaria siceraria).

The ritual and medicinal use of plants was similarly significant. Tobacco (Nicotiana tabacum), peyote (Lophophora williamsii), ayahuasca (Banisteriopsis caapi), and other mind-altering substances such as opium poppy (Papaver somniferum) are derived from plants. Various forms of alcohol likewise come from plant products and have mind-altering effects. These items were widely employed by shamans, healers, and the ruling elite in a variety of ceremonial and ritual activities (e.g., Morehart et al. 2005; Pauketat et al. 2002).

Many faunal remains similarly represent resources used in nondietary ways. Certain items, such as animal hides, sinew, and feathers (for fletching arrows and spears) were likely used for clothing, shoes, and other items but rarely preserve archaeologically. Bone, antler, shell, and other faunal material were fashioned into a variety of items, including perforating implements (awls, needles, pins, chisels), flintknapping tools (flakers, billets), weaving and hide-processing tools (shuttles, hide scrapers, beamers), gardening instruments (hoes made from scapulas, digging sticks, rakes made from antlers), gaming pieces (dice, gaming sticks), hunting items (spearpoints, fishhooks, harpoon heads), and decorations or ornaments (hairpins, beads, combs, pendants). Mussel shell was carved into beads, buttons, gorgets (ornamental collars), spoons, and scoops, and large marine whelks were transformed into bowls and containers. Turtle shells were made into bowls and rattles.

Faunal remains, both modified and unmodified, were also important as ritual and ceremonial paraphernalia. Small carnivore jaws were often cut and polished and perhaps functioned as items of sympathetic magic. Bone and shell effigies have also been recovered and may have been used in a similar fashion. Numerous other examples exist, but sometimes it is difficult to determine the exact function of these items. Bone, antler, shell, and other faunal material were modified to make a wide array of tools, ornaments, decorations, and ritual items.

Summary: Case Studies

Zooarchaeological and paleoethnobotanical remains can provide considerable insight about past human behavior at prehistoric and historic archaeological sites. Much of this information relates to dietary strategies and paleoenvironmental reconstruction, but faunal and floral remains also provide significant information about the nondietary uses of these resources. We provide two case studies about how zooarchaeological and paleoethnobotanical data are employed in archaeological research.

Special Topics: Case Study, Early Twentieth-Century Dining in East St. Louis, Illinois

Archaeological investigations in the Goose Hill neighborhood, a turn of the century working-class section of East St. Louis, recovered substantial zooarchaeological remains from several different households that allowed for the investigation of differential dietary patterns in an urban residential neighborhood. The information summarized here is taken from Dappert-Coonrod and Kuehn (2017).

The Goose Lake neighborhood in East St. Louis was established shortly after the Civil War. By the turn of the twentieth century, it was inhabited primarily by working-class individuals and families, many of whom were employed at the nearby National Stockyards and associated meat-packing plants. The Goose Lake neighborhood also contained numerous commercial shops and services that supported these residents. Archaeological investigations were conducted in this location as part of the New Mississippi River Bridge Project. All of the work was overseen by the Illinois State Archaeological Survey of the Prairie Research Institute of the University of Illinois at Urbana-Champaign. In addition to the archaeological field and laboratory work, extensive historical and archival research was conducted on this property so researchers could better understand the social and cultural context of this historic community.

Zooarchaeological material chosen for analysis came from three households: 1) the Mead House, also known as the Maple Hotel or the Davis Boardinghouse, a boardinghouse establishment occupied by a variety of tenants who likely lived in the establishment for a few months or years; 2) the Benner household, a single family long-term residence owned by Elizabeth Benner, a widow, and her daughter Victoria; and 3) a rental property at 819 Rear Bowman Avenue, which was characterized as an informal worker’s cottage and was occupied by an ever-changing series of tenants, usually couples or small families. Multiple privy deposits were associated with each household that contained relatively large amounts of faunal material discarded by the site inhabitants. These deposits allowed for a detailed comparison of dietary practices among the occupants of the Mead House, the Benner household, and the tenant (or multi-tenant) household.

Cattle, swine, and chicken remains occur in all three assemblages and predominate among specifically identifiable fauna. Mutton (sheep) remains were found in the Mead House assemblage and eggshell fragments were recovered in the Mead House and Benner assemblages. Fish remains (e.g., carp, brown bullhead) were recognized in all three assemblages and cottontail rabbit bones were found in the Mead House and Benner assemblages. Other taxa that were recognized include double-crested cormorant in the Benner household assemblage, turkey in the Mead House assemblage, and graylag (domestic) goose in the tenant household assemblage. Overall, the composition of the assemblages indicate that the various site inhabitants relied on beef, pork, and chicken (and eggs) as dietary mainstays but also included other domesticates (e.g., sheep, turkey, goose) and some wild fauna (e.g., rabbit, various fish) in the diet. Dog bones from at least two dogs were found in the Mead House assemblage, suggesting that pets were kept by some of the residents. Rodent bones were also recognized, indicating that urban pests like mice and rats were a problem.

The cuts of meat, particularly beef and pork, represented by the elements that were recovered provide a closer look at dining in the different households. The Mead House inhabitants consumed more steaks, roasts, and hams (as well as chicken and egg dishes) and less soups and stews. At the Benner house, they ate a variety of steaks, roasts, hams, stews, and soups, making use of lesser-quality cuts of meat. The occupants of the tenant house, however, ate more soups and stews than the residents of the Mead House or the Benner household. Most of the assemblage bones from the tenant household were from low-value cuts, particularly beef, that were suitable mainly for soups and stews. By economizing on cheaper cuts of beef, the tenant household residents were able to buy at least some moderately more expensive pork hams. This indicates that dietary choices were affected, at least to some extent, by the economic means of each household. The residents of the Mead House boardinghouse received at least some higher-cost beef and pork cuts as part of their room and board. The widow Mrs. Benner was moderately well off financially, it would seem, but still needed to be frugal in her spending. The tenant household occupants, however, were renters with limited funds, and this is reflected in their dietary choices.

Interestingly, the serving dishes and other tableware recovered during the archaeological investigations at each property support the patterns observed in the faunal remains. The dishes from the Mead House boardinghouse included large serving platters for group meals and dishware associated with the serving of steaks and roasts. The Benner tableware, in contrast, contained more fashionable dishes, perhaps reflecting the income and status associated with single-family home ownership. The dishes found in the tenant household privies were of somewhat lower value, but there were also some fancier porcelain dishes that were comparable to those found at the nearby Benner house. While no single line of evidence is sufficient to answer all questions regarding dietary choices, purchasing power, and socioeconomic status, looking at faunal remains and ceramic dishware together can provide important insight into these research issues.

Special Topics: Synthesis of Macrobotanical and Microbotanical Data at the Site of Los Naranjos, Honduras

Los Naranjos is a major multicomponent archaeological site in Honduras that lies on the northern border of Lake Yojoya. The site has monumental architecture and sculpture that date from 1000 to 500 BCE and evidence of jade and pottery styles that link it to the Gulf Coast Olmec. The inhabitants of the site were thought to have been dependent on maize agriculture as their primary food source by 800 BCE, an assumption that has been challenged by recent paleoethnobotanical work that combines both macrobotanical and microbotanical analyses. This case study, from an article by Morell-Hart et al. (2014), will reveal why it is important to use multiple lines of evidence when considering human relationships with plants in the past.

A lake pollen core taken during previous excavations in the 1980s suggested a meaningful variation in plant pollen over time. Levels from the core that corresponded to earlier time periods revealed a high percentage of tree species and no crops such as maize or beans. Researchers interpreted a rise in plants such as Chenopods (the family that contains the popular grain quinoa), maize, and other grasses over time as evidence of early agriculture. Archaeologists interpreted the evidence to show that agricultural intensification occurred later in time based on an increasing percentage of plants from the sunflower family, ragweed, and other plants that flourish in habitats that have been disturbed by human activity. However, the pollen from this core comes primarily from wind-pollinated plants and therefore does not provide a complete picture of plant use and does not show direct evidence of plant use at the site of Los Naranjos itself.

To fill in the gaps in information left by the existing pollen record, Morell-Hart and colleagues collected and analyzed plant data from six kinds of contexts, including macrobotanical samples from light fractions and microbotanical samples taken from the edges of obsidian artifacts used as tools. The samples came from interior surfaces, matrices between surfaces, architectural fill and pit fill, and a midden, or trash dump.

The analysts recovered maize in several different forms, including phytoliths, starch granules, and a charred kernel. However, they did not find other parts of maize that are typically ubiquitous (cupule and cob fragments). They also found amaranth and chia seeds and starch granules of manioc and sweet potato that they identified from residues on obsidian tools. Tree crops were important for early agriculturalists, and botanical evidence from the site suggests that people were using palm species for a variety of culinary and non-edible uses. Interestingly, species that are commonly thought of as belonging to the Mesoamerican agricultural suite (beans and squash, primarily) were not evident at the site. Plants with medicinal or cultural uses were identified through both macrobotanical and microbotanical analyses, including tobacco, various grass species, and sunflower species.

The analysis of the pollen core that was done in the 1980s led researchers to believe that maize agriculture and its two sisters, beans and squash, were present in the region by 1000 BCE. However, microbotanical analysis of starch granules and phytoliths and macrobotanical recovery of whole seeds paints a different picture. The more recent paleoethnobotanical interpretation reveals that instead of engaging in widespread intensive agriculture, the residents of Los Naranjos were likely exploiting a wide variety of plants, including seed, root, and tree crops. This new analysis also illustrates the use of plants for nonculinary purposes, notably the use of tobacco, which may have been used ritually as well.

While the pollen data were not incorrect, the interpretation of human activity without considering other lines of evidence led to a conclusion that did not show the whole picture of what was happening at the site. Flotation is important for recovering macrobotanical material but is not as directly associated with human activity as microbotanical residues sampled directly from artifacts. Synthesizing a variety of botanical data enriches our understanding of human practices, including nonculinary uses of plants, and complicates our view of the past.

Discussion Questions

1. What can animal and plant remains from archaeological sites tell us about past peoples and how they lived?
2. Aside from diet, what information about past lifeways and environmental conditions can be determined from animal and plant remains from archaeological sites?
3. One important part of archaeological research is the capability to examine how human behavior changed over time. Provide one example of how changes in daily life might be recognized by comparing different faunal and floral assemblages from two sites from different time periods.
4. Zooarchaeological and paleoethnobotanical remains do not always preserve over time. Describe how the absence of certain animal and plant remains might impact our view of life in the past.
5. Following up on Question 4, are there ways for archaeologists to fill in knowledge gaps in the archaeological record when studying the past? What types of indirect evidence might archaeologists use to obtain more detailed insight into the past? For example, even if no horse bones were recovered, the presence of horseshoes and metal bridle pieces would indicate that horses were present at a site. What other examples of indirect evidence can you think of?
6. Faunal and floral resources were often used in ritual and ceremonial activities as well as for utilitarian or everyday activities. Discuss two examples each of animal and plant resources used in ritual activities. Can you provide a similar example from modern-day life?
7. Archaeologists studying animal and plant remains specialize in the identification and analysis of faunal and floral remains. Discuss how knowledge of ecology, zoology, botany, and other fields can provide important insight for interpreting the remains recovered.

About the Authors

Steven Kuehn is a zooarchaeologist with the Illinois State Archaeological Survey, a division of the Prairie Research Institute of the University of Illinois at Urbana-Champaign. He has more than 30 years of professional experience as an archaeologist, working primarily in the Upper Midwest and Great Lakes region. Steven has analyzed zooarchaeological remains from Paleoindian through Historic Euroamerican sites in the Midwest, the Great Plains, the Southeastern United States, and Alaska. Some of his recent research focuses on Paleoindian subsistence behavior, Archaic bison hunting in Minnesota, and the archaeology of domestic dogs in the Midwest. In his free time, Steven enjoys reading, hiking, and playing board games with his family. Steven also writes short stories and mystery fiction, including novels in his Jacob Caine Archaeology Mystery series

Rebecca Albert earned her MA in Archaeology at the University of California, Santa Barbara in 2021. She was previously the assistant coordinator for the Central Coast Information Center, one of nine archaeological and built environment data clearinghouses in the California Historic Resources Information System. Her undergraduate and graduate research focused on microbotanical analysis of carbonized residues from the Upper Great Lakes in order to gain a deeper understanding of the history of plant domestication in North America and how these plants may have informed prehistoric foodways. In her free time, Rebecca enjoys quilting, cooking, and playing with her cat, Juliet.

Emily Johnson is a paleoethnobotanist pursuing her PhD in Anthropology at the University of California, Santa Barbara. She has conducted excavations and/or analyses of archaeological sites in Turkey, Israel, Guatemala, Mexico, and throughout the United States. Her current research is focused on investigating the timing, spread, and development of the nixtamalization process throughout ancient Mesoamerica. When not reading about all things corn, Emily enjoys crafting and playing with her cat, Scout.

Further Exploration

Microbotanical reference manuals:

PhytCore DB (a free website with registration): online database of photomicrographs of phytoliths, searchable by shape, type of sample, site, etc.

Starch Grain Database: a free online database of starch grains collected from modern samples

Cooked Starch Database: a free image database of starches from plants that have been cooked in various ways. From Henry, Hudson, and Piperno (2009).

References

Albert, R.K., Kooiman, S.M., Clark, C.C., and Lovis, W.A. (2018). Earliest microbotanical evidence for maize in the northern Lake Michigan Basin. American Antiquity, 83(2), 345–355.

Alenius, T., Mökkönen, T., and Lahelma, A. (2012). Early farming in the northern boreal zone: Reassessing the history of land use in southeastern Finland through high-resolution pollen analysis. Geoarchaeology, 28(1), 1–24.

Asouti, E., and Austin, P. (2005). Reconstructing woodland vegetation and its exploitation by past societies, based on the analysis and interpretation of archaeological wood charcoal macro-remains. Environmental Archaeology, 10, 1–18.

Baker, J., and Brothwell, D. (1980). Animal diseases in archaeology. Academic Press.

Bartosiewicz, L. (2013). Shuffling nags, lame ducks: The archaeology of animal disease. Oxbow Books.

Binford, L.R. (1981). Bones: Ancient men and modern myths. Academic Press.

Daley, J. (2017, June 2). Cat left a pawprint in a 2,000-year-old Roman roof tile. Smithsonian Magazine.

Dappert-Coonrod, C.P., and Kuehn, S.R. (2017). A boarder, a widow, and a tenant sit down for dinner: Foodway comparisons in the Goose Hill neighborhood, East St. Louis, Illinois. Midcontinental Journal of Archaeology, 42(1), 4–36.

Grayson, D.K. (1984). Quantitative zooarchaeology: Topics in the analysis of archaeological faunas. Academic Press.

Guedes, J.D., and Spengler, R. (2014). Sampling strategies in paleoethnobotanical analysis. In J.M. Marston, J.D. Guedes, and C. Warinner (Eds.), Methods and theory in paleoethnobotany (pp. 77–94). University Press of Colorado.

Henry, A., Hudson, H., and Piperno, D. (2009). Changes in starch grain morphologies from cooking. Journal of Archaeological Science, 36, 915–922.

Kooiman, S.M. (2018) A multiproxy analysis of culinary, technological, and environmental interactions in the northern Great Lakes region. [Doctoral dissertation, Michigan State University, East Lansing, Michigan].

Larson, G., et al. (2014) Current perspectives and the future of domestication studies. Proceedings of the National Academy of Sciences, 111(17), 6139–6146.

Lyman, R.L. (1994) Vertebrate taphonomy. Cambridge Manuals in Archaeology. Cambridge University Press.

———. (2008). Quantitative paleozoology. Cambridge University Press.

MacDonald, G.M., and Edwards, K.J. (1991) Holocene palynology: Principles, population and community ecology, palaeoclimatology. Progress in Physical Geography: Earth and Environment, 15(3): 261–289.

Madella, M., Alexandre, A., and Ball, T. (2005) International code for phytolith nomenclature 1.0. Annals of Botany, 96(2): 253–260.

Marston, J.M. (2014) Ratios and simple statistics in paleoethnobotanical analysis: Data exploration and hypothesis testing. In J.M. Marston, J.D. Guedes, and C. Warinner (Eds.), Methods and theory in paleoethnobotany (pp. 163–179). University Press of Colorado.

Martin, A.C., and Barkley, W.D. (2004). Seed identification manual. The Blackburn Press.

Messner, T.C. (2011). Acorns and bitter roots: Starch grain research in the prehistoric eastern woodlands. University of Alabama Press.

Morehart, C.T., Lentz, D.L., and Prufer, K.M. (2005) Wood of the gods: The ritual use of pine (Pinus spp.) by the ancient Lowland Maya. Latin American Antiquity, 16(3), 255–274.

Morell-Hart, S., Joyce, R.A., and Henderson, J.A. (2014) Multi-proxy analysis of plant use at Formative Period Los Naranjos, Honduras. Latin American Antiquity, 25(1): 65–81.

Nichols, D.L., McLaughlin, M.J., and Benton, M. (2000) Production intensification and regional specialization: Maguey fibers and textiles in the Aztec city-state of Otumba. Ancient Mesoamerica, 11(2), 267–291.

O’Connor, T. (2000) The archaeology of animal bones. Sutton Publishing.

Pauketat, T.R., Kelly, L.S.., Fritz, G.J., Lopinot, N.H., Elias, S., and Hargrave, E. (2002) The residues of feasting and public ritual at early Cahokia. American Antiquity, 67(2): 257–279.

Pearsall, D.M. (2015). Paleoethnobotany: A handbook of procedures (3rd ed.). Routledge.

Piperno, D.R. (2006). Phytoliths: A comprehensive guide for archaeologists and paleoecologists. AltaMira Press.

Reitz, E.J., and Wing, E.S. (1999) Zooarchaeology. Cambridge Manuals in Archaeology. Cambridge University Press.

Russell, N. (2012) Social zooarchaeology: Humans and animals in prehistory. Cambridge University Press.

Schoch, W.H., Barbara Pawlik, and Fritz H. Schweingruber (1988). Botanical macro-remains. Haupt.

Smith, A. (2014) The use of multivariate statistics within archaeobotany. In J.M. Marston, J.D. Guedes, and C. Warriner (Eds.), Method and theory in paleoethnobotany (pp. 181–204). University Press of Colorado.

Twiss, K. (2012) The archaeology of food and social diversity. Journal of Archaeological Research, 20(4): 357–395.

Wheeler, J.C. (1995) Evolution and present situation of the South American camelidae. Biological Journal of the Linnean Society, 54(3): 271–295.