The Importance of Wilma Wetterstrom's Ethnobotanical Analyses of Plants from Arroyo Hondo, Richard I. Ford

Richard I. Ford

Richard Ford received his B.A. (1963) from Oberlin College and his M.A. (1965) and Ph.D. (1968) from the University of Michigan. He joined the University of Michigan Museum of Anthropology and Department of Anthropology in 1969 as Assistant Curator of Ethnology and Director of the Ethnobotany Laboratory and Assistant Professor of Anthropology. He became Curator of Ethnology and Professor of Anthropology in 1977. Ford served as Director of the Museum of Anthropology from 1972 to 1983 and again from 2002 to 2005. He also served as Chair of the Department of Anthropology (1989–1996) and as Associate Dean for Research and Computing in the College of LSA (1987–1989). He retired from the University in 2007 and is now Curator and the Arthur F. Thurnau Professor Emeritus of Anthropology and Botany, residing in Santa Fe, New Mexico, where he maintains his active commitment to teaching, research and public service.

The Importance of Wilma Wetterstrom's Ethnobotanical Analyses of Plants from Arroyo Hondo

February 21, 2016


Two external questions about Wilma have haunted me for years. Why did it take so long for her to complete her dissertation? It reality it didn’t. This was a very complicated dissertation for an individual to doalone as I will explain.

Wilma was part of Doug’s team working at Arroyo Hondo but that was the excavation phase.  Although she had assistance collecting the flotation samples from the hearths, floors, and trash dumps, she had to float each, dry them and package them with the appropriate levels and log numbers. When this was done, they had to be sent to the University of Michigan Ethnobotanical Laboratory for her alone to sort. She spent countless hours huddled over a microscope examining each float sample sorting it into constituent biological parts - charcoal, corn kernel fragments, charred seeds, rootlets, and non-plant items. Hundreds of charred plant part samples received this meticulous treatment. But this was just the start! Next she had to identify the content in each category by comparing them to reference specimens in an archaeobotanical herbarium. This allowed her identify specifically the plant part, the plants common name, and most importantly its botanical scientific binomial. All the sorted samples received this attention. In the end she had to tabulate the number of each category. What started as a vial of charred seeds could become ten of different botanically identified plant species, each in it own vial. Wilma’s first Arroyo Hondo publication was about that she did not find: chile! (Wetterstrom 1976a)

However, this research was a preamble to the most important part of the project: explaining why these plants were at Arroyo Hondo. This involved studying the autecology of each species to understand variability in plant edible part production. The phytogeography of each species had to be mapped and analyzed.

Wilma soon found that the nutritional and botanical literature did not provide information about the nutritional composition of all the edible plants she recovered with the food remains at Arroyo Hondo. To answer this problem for each edible species, Wilma had to enroll in the School of Public Health to learn the laboratory techniques to do a nutritional description of each of her wild plants. To complete this phase of her project she had to go into the field and collect sufficient quantities of each plant to note the variability in morphology and productivity.

Her flotation analysis only reinforced what she suspected: the dominant edible plant at Arroyo Hondo was maize. Now another problem emerged.  The  corn was a type of flint. There was ample literature about its yield but some other maize from earlier sites was a more primitive Chapalote that did not have good productivity literature. The crew at Arroyo Hondo worked very closely together and shared research problems and sought solutions. At Arroyo Hondo Mike Marshall and Ed Kelley planted an experimental garden field with Chapalote to measure its productivity and to determine the range of morphological variability from one cob to another. Although Wilma did not need these data for her immediate project, intellectual curiosity led her to participate.They recorded the growth of the other cultivated plants to note intraspecies variation and environmental challenges.

As she became more familiar with the plant assemblage, she recognized that the human population at Arroyo Hondo would face periods of food scarcity and undergo periods of nutritional stress. If this were true, evidence should be present in the skeletons and dentation of the inhabitants, especially the children, as she learned from Professor John Robeson.  She took a course on teeth from Professor Stanley Garn. She did this successfully and was ready to take these new skills to her colleague Ann Palkovich for her to examine the Arroyo Hondo skeletons and teeth.

Although mainframe computers were becoming an important part of archaeology and the Michigan archaeologists were leaders in this new development, as evidenced by the piles of IBM punch cards the professors and their students carried around, some calculations could still be done by hand on a Texas Instrument portable calculator that Wilma used. Computer simulations were still rare in archaeology. With the quantitative data Wilma was assembling, she felt confident that she could create sufficient algebraic models of maximum and minimum yields to predict the plant nutritional variability she suspected was limiting population growth and responsible for poor health.

Wilma had to have a better understanding of climatic factors that limited plant production for her model. Wilma was not a dendroclimatologist or a palynologist. However, on the Arroyo Hondo project you did not need these technical skills yourself. Doug Schwartz encouraged open and frequent communication among all his staff members. To comprehend the climate during the occupations of Arroyo Hondo, Wilma could talk to Martin Rose, who was staff dendroclimatologist. To examine changes in plant communities she could talk to Dr. Vorsila Bohrer, who was staff palynologist, and to Ed Kelley, who was the staff ecologist. Doug’s expectation for open communication among the research staff and the sharing of ideas and data assured the Arroyo Hondo Project as an innovative and original interdisciplinary archaeological project.

With this explanation for the complexity of Wilma’s dissertation research and the advanced multidisciplinary education her dissertation required, it is easy to understand why it required more time to complete than most would expect.

What ever happened to Wilma?

Since the Arroyo Hondo monograph was published, Wilma changed geographic research areas. She has conducted archaeobotanical research in Madagascar, Egypt, Syria, and Spain. She is an Associate in Botany at the Harvard University Herbaria where she continues her paleobotanical research to identify plants from abroad as she did with Arroyo Hondo botanical remains. Now, however, they come from areas in the Old World where she receives charred plant remains from several significant archaeological projects.  One of her Michigan professors, Dr. Henry Wright, was the first American to initiate a long-term archaeological project with Madagascar colleagues on the island. He was assembling a research team in the Old World tradition when he  invited Dr. Wetterstrom to join it as the staff ethnobotanist and she has identified his plants ever since (Wetterstrom 1992b).

In Egyptian archaeology Wilma is employed by Ancient Egypt Research Associates, Inc. in Cambridge where she is Science and Arts editor of their monographs and newsletter. In recent years, AERA has explored the development of urbanism, labor organization, and the elementary structures of ancient daily life at the once Lost City of the pyramid builders at Giza. She has been an important facilitator of archaeobotany in Egyptian archaeology. One of her major contributions to Egyptian archaeology is a detailed review of the transition from foraging to agriculture in the Nile valley (Wetterstrom 1993). She has worked extensively on plants in the daily lives of Old Kingdom residents and delineated the importance of barley and emmer wheat.

Her active paleoethnobotanical fieldwork procedures duplicate her earlier Arroyo Hondo research. In Madagascar she introduced large-scale, water flotation to separate charred plant parts from a soil matrix. In so doing she found the oldest rice in Madagascar and many other identifiable plant parts to reveal the weed flora in agricultural field and other edible plants from arboreal environmental zones.

In Italy, Kenya, and Jordan she again brought her botanical skills and models to interpret the plants recovered there. Unfortunately, we do not find Wilma in New Mexico any more but she has a very distinguished paleoethnobotanical career elsewhere in the world. She is well known in the paleoethnobotanical institutes in Europe and Israel, (e.g.,Wetterstrom 1992). Wilma considers major theoretical issues beyond the plants from individual sites. She is a major contributor to issues related to agricultural origins (Wetterstrom 2004)  in Egypt and the Near East (Miller and Wetterstrom 2000). I think this information is more than sufficient to answer the second question.

Many of the approaches Wilma used at Arroyo Hondo were precedent setting for paleoethnobotany in the Southwest. We will discuss each of them in sequence.

1.    Analyzing and interpreting the archaeology plants from a single site.

2.    Using water separation (flotation) to recover archaeological plants from a soil matrix.

3.    Analyzing the nutritional value of edible plants.

4.    Describing famine plants.

5.    Investigating how nutritional deficiencies adversely impact skeletal, dental, and population growth.

6.    Modeling prehistoric population, diets and climate.

Original Contributions to Southwestern Archaeology

Food, Diet, and Population at Prehistoric Arroyo Hondo

Pueblo, New Mexico (1988) was highly praised in reviews when it appeared. I have been able to find 12 book reviews and all are complementary without criticism. They appeared in local New Mexico newspapers where it was recommended to general readers.  Others were in professional journals in archaeology and botany. Again, it was reviewed for setting a high research standard. In fact Rege Wiseman called it a ”..veritable tour de force as a carrying capacity study” (Wiseman 1987:44).

Professionals working on the topics listed below frequently cite her and continue to do so almost 30 years after publication. Without listing all the citations (they can be found under Google Scholar), here is a brief summary of citation counts revealing the significance of her Arroyo Hondo research.

Single site paleoethnobotany=17 citations

Flotation=Southwest 11; Old World=24 citations. Wilma almost single handedly introduced flotation methods to African archaeology.

Famine foods=4

Nutritional deficiencies=7; all cite Wilma

Modeling nutrition and ecosystems=5

Analyzing and interpreting the archaeology plants from a single site

In the Southwest, as explained above, for large sites this was quite unusual. Many of us who analyzed a specific category of plant remains tired of being listed in site report as “an appendix.”  On the other hand, it is difficult to have the scientific and botanical training to identify and analyze all the remains. Having these skills made Wilma unusual.

The paleoethnobotany by one researcher of a single archaeological site was unusual. Wilma’s study of plants from Arroyo Hondo in many respects was unprecedented. Paleoethnobotanists were accustomed to divide the plant collection into botanical categories and then to send the divided assemblages, e.g., Zea mays, cucurbits, seeds, etc., to specialists. Wilma elected to identify and analyze  the  together herself by learning plant taxonomy and identification techniques. Beyond this decision, she used estimated human population models for the entire site as revealed as a complete settlement by the field archaeologists. In other words, she viewed Arroyo Hondo as a site occupied by people who used plants according to their needs and the number of remains and their distribution reflected how and where in Arroyo Hondo people used the different genera of plants.

Paleoethnobotanists often do identify and discuss all the plants recovered from small sites or those confined to a contracted excavation space, often only a portion of a site. Such reports are numerous.  However, large site excavations are quite rare in New Mexico and for its time Arroyo Hondo was one of the largest site excavations in the state and continues after 45 years to have that distinction.

Using water separation (flotation) to recover archaeological plants from  soil

In 1960 rarely did one observe a screen on an archaeological site in New Mexico. Recovery of most archaeological remains of any size was done through a visual inspection of a shovel full of dirt or the trowel scrapings on a unit or structure’s floor. Small artifacts and certainly individual seeds or charred corncob parts were simply missed. Three legged tripod suspended screens were a Midwestern tradition going back to the CCC excavations. When screens were introduced in the Southwest was when salvage archaeology became the mainstay of excavations and external reviewers required them. However, by 1960 another recovery revolution was becoming standard in the Midwest.  Initiated by Stuart Struever in Kampsville, IL., it was large scale water separation of plant parts and small artifacts, e.g. debitage, from the soil matrix soaked in water. This recovery method was primarily brought to the Southwest by doctoral students who were trained in the Midwest. Four Southwest archaeology projects were hubs for flotation recovery in the 1970s: Arroyo Hondo (Wetterstrom), Cibola near Zuni (Watson, LeBlanc, Redman), Mimbres Valley (Minnis), and Salmon Ruin (Doebley).  Wilma initiated an extensive flotation program at Arroyo Hondo. She was notable in her successful recovery of tiny annual plant seeds and charred corn parts that were traditionally missed. Contract projects in Northern New Mexico were soon imitating her with equal success. Today flotation of archaeological soil is expected of all open air archaeological projects, academic or contract.

Discovering the nutritional value of edible plants

We eat to live but to understand why we need the nutritional values of consumed plant foods, raw and cooked. The problem is: What are the food values of most wild and even tended edible plant parts? The wildlife literature, especially for ducks and geese, has some information for sportsmen hunted animals. But until Wilma started her research, the chemical nutrition constituents of tended vegetable plants were unknown in their raw state, no less after cooking. To rectify this situation for the Southwest and Arroyo Hondo in particular, Wilma found herself in Dr. John Robeson’s nutrition class and Walter Block’s laboratory in the School of Public Health at Michigan. Here she learned the techniques to chemically analyze edible plant parts and to understand their nutritional value for human populations. She did original nutritional analyses on banana yucca (Yucca baccata) and Indian rice grass (Achnatherum hymenoides). The laboratory results were included in her Arroyo Hondo monograph and  they were built into her population subsistence model.Information about other plants she identified came from published sources. Others working on wild plants in prehistoric southwestern diets have cited her work. Her research was very important to Huckell and Toll (2004).

Describing famine plants

In his review of Wilma’s monograph, the esteemed Harvard ethnobotanist, Richard Schultes, complimented her for using plant ecology to understand phytogeography and plant productivity (Schultes 1988). It was this basic ecological approach that allowed her to conclude that there would be periods of wild food scarcity and even times of agricultural shortfalls. Wilma welcomed  this type of discussion by showing what the causes of food shortage were and what Arroyo Hondo residents could do to correct it by acknowledging famine foods (Wetterstrom 1986: 87-124).

A close review of her famine plants shows that most have not always had that designation. Many in earlier times were themselves seasonal stables and maintained the good health of the inhabitants. Going from staple to starvation was a result of prolonged droughts or growing human populations that outstripped plant production. Minnis addressed Wilma’s alternative nutritional plants directly in a series of publications (Minnis 1991) and he recognized that these low yielding plants are rare.

Speth and Spielman (1983) developed the same logic into a model of variable energy contributions from animals at different seasons of the year. Clark (1988) considered the same topic but regarding animals in the pueblo Southwest.

Investigating how nutritional deficiencies adversely impact skeletal, dental, and population growth

Wilma recognized as a graduate student that she was contributing to a new archaeological subdiscipline: archaeonutrition. Once Wilma established that the Arroyo Hondo’s changeable and unpredictable environment would not sustain a healthy human population, she determined that nutritional deficiencies would leave signatures on the bones in the form of Harris Lines (growth arrest lines) and the teeth (enamel hypoplasia). These were investigated on the skeletons from Arroyo Hondo with convincing success by Ann Palkovich.

Modeling: Prehistoric diets, population, and climate

Since Wilma published her monograph about population size and diet, many others scholars have constructed simulation models to predict corn production, wild plant subsistence, animal calories, and human population growth and distribution. Wilma’s model was a “snapshot” in time and did not track either a precipitation trend line or a minimum-maximum plant production curve over a period of years. Hers was algebraic but it was successful to establish the nutritional stress the Arroyo Hondo population faced.Moe recent models w take several forms and all dependent upon the computer to run time sensitive, multiple iteration simulations. Most models are about maize production (Van West 1990). Others are quantitative isotopic models (e.g.,Decker and Tieszen 1989).  Then there are some that try to simulate entire ecosystems (Kohler et al. 2012).

Kohler and his colleagues at Crow Canyon have made Southwest Colorado into a laboratory for learning about climatic variability, differential crop plant yields, and correlations between contemporary climatic variation and prehistoric tree ring patterns. Most of our best maize production models come from work in Southwest Colorado (see Burns 1983, Kohler 2010, Van West 1990 and Varien et al. 2007). Wilma had hoped that her pioneering model would stimulate other archaeologists in the Rio Grande area to attempt to do the same thing for comparative purposes.

One of the best examples is Kohler and Varian (2012) with a series of essays that developed new algorithms to predict productivity of a regional subsistence pattern. Many of these excellent chapters cite Wilma’s earlier work because it set an early standard in the Southwest and it is nice to read a new generation of scholars are returning to her results for direction.

The most common models are those that simulate maize production in the areas surrounding particular archaeological provinces. When areas are too small, the results are meaningless.  When the area is too large, it is multivariate and difficult to control.

Unlike the more recent models listed above, Wilma did not develop an algorithm but she did design a very useful algebraic product that proved to be just as effective for predicting calories and population needs. All models do not have to be elaborate computer simulations. Spielman, Schoeninger and Moore (1990) in their bone isotopic study at Pecos Pueblo used Wilma’s work at Arroyo Hondo as a reference point and a comparative base without constructing a new simulation.


Wilma’s Arroyo Hondo research quickly evolved into a very complex dissertation with no precedents that she would undertake herself benefiting from communication with other scientists on the Arroyo Hondo staff and University of Michigan Professors, including  Walter Block, John Robeson, Stanley Garn, Volney Jones, and Richard Ford. The project was extremely complicated but it had to be to use archaeobotanical data in a scientifically convincing way to answer what started as a seemly simple question: What was life like for the inhabitants of Arroyo Hondo? In the end she answered it by showing that it was hard, unpleasant, uncomfortable, and not very fulfilling. In other words, she demonstrated that Thomas Hobbes was correct: At Arroyo Hondo people lived in ”… continual fear, and danger of violent death: and the life of man, solitary, poor, nasty, brutish and short." (Leviathan 1651, Part 1,Ch. 12)


Burns, Barney Tillman

1983 "Simulated Anasazi Storage Behavior Using Crop Yields Reconstructed From Tree Rings: AD 652-1968 (Colorado)."

Clark, T.

1998 Faunal Resource Depletion and Nutritional Stress in the Pueblo IV Period Migration and Reorganization: The Pueblo IV Period in  193-207.

Danforth, Marie Elaine

1999  Nutrition and Politics in Prehistory. Annual Review of Anthropology 28:l-25

 Decker, Kenneth W. and LL Tieszen

1989  Isotopic reconstruction of Mesa Verde diet from Basketmaker III to Pueblo III. The Kiva 55 ( 1): 33-46

Huckell, Lisa and Mollie S. Toll

            2004 Wild Plant Use in the North American Southwest. Editor, Paul Minnis People and Plants in Ancient Western North America, 37-114. Smithsonian

Kohler, Timothy A.

2010  A new paleoproductivity reconstruction for Southwestern Colorado, and its implications for understanding thirteenth-century depopulation." Leaving Mesa Verde: Peril and change in the thirteenth-century Southwest: 75-101.

Kohler, Timothy and Mark A. Varian, EDS.

2012  Emergence and Collapse of Early Villages: Models of Central Mesa Verde Archaeology. University of California Press.

Kohler, T. A., Bocinsky, R. K., Cockburn, D., Crabtree, S. A., Varien, M. D., Kolm, K. E., and Kobti, Z.

2012 Modelling prehispanic Pueblo societies in their ecosystems. Ecological Modelling, 241, 30-41.

 Miller, Naomi and Wilma Wetterstrom

2000 The Beginnings of Agriculture: The Ancient Near East. In, The Cambridge Historical, Geographical, and Cultural Encyclopedia of Human Nutrition,   ed. by Kenneth F. Kiple. Cambridge: Cambridge University Press, pp. 1123–1139.     

Minnis, Paul E.

1991 Famine foods of the northern American desert borderlands in historical context. Journal of Ethnobiology 11 ( 2): 231-257.

Speth, John D. and Katherine Spielman

1983  Energy source, protein metabolism, and hunter-gatherer subsistence strategies. Journal of Anthropological Archaeology   2( 1): 1–31.

Katherine A. Spielmann, Margaret J. Schoeninger, and Katherine Moore

1990  Plains-pueblo interdependence and human diet at Pecos Pueblo, New Mexico.  American Antiquity 55(4): 745-765.

Schultes, Richard

1988  Book Review: Food, Diet, and Population at Prehistoric Arroyo Hondo Pueblo, New Mexico. Wilma Wetterstrom. Arroyo Hondo Archaeological Series 6. School of American Research Press. Economic Botany 42 (1): 28.

Van West, Carla R.

1990  Modeling Prehistoric Climatic Variability and Agricultural Production in Southwestern Colorado: A GIS Approach. Unpublished Ph.D. dissertation,  Department of Anthropology, Washington State University, Pullman.

Varien, Mark D., Scott G. Ortman, Timothy A. Kohler, Donna M. Glowacki, and C. David Johnson. 2007 Historical ecology in the Mesa Verde region: results from the Village Ecodynamics Project. American Antiquity): 273-299.

Wetterstrom, Wilma E.

1976a They Never Savored Chili.” Exploration 1976. School of American Research, Santa Fe, New Mexico. pp. 2–7.

1976b The Effects of Population Size on the Nutrition of Arroyo Hondo Pueblo, New Mexico. Vol. 1 and 2. Ph.D Dissertation, Department of Anthropology, University of Michigan, Ann Arbor.

1988  Food, Diet, and Population at Prehistoric Arroyo Hondo Pueblo, New Mexico. Arroyo Hondo Archaeological Series 6. School of American Research Press. 

1992a  "Climate, diet and population at a prehistoric pueblo in New Mexico." Issues in Environmental Archaeology, Institute of Archaeology, London (1992): 35-61.

1992b “Une Contribution à la paléoethobotanique du plateau central de Madagascar,” by Wilma Wetterstrom and H. T. Wright. Taloha 11: 147–66. Programme Tri-Institutionnel, Musée d’art et d’archeologie (Antananarivo) SAREC (Uppsala) INALCO (Paris).

1993 “Foraging and Farming in Egypt: The Transition from Hunting and Gathering to Horticulture in the Nile Valley.” In, The Archaeology of Africa: Food, Metals, and Towns, ed. by T. Shaw, P. Sinclair, B. Andah, and A. Okpoko. London: Unwin Hyman, pp. 165–226.

2004 “The Broad spectrum revisited: Evidence from plant remains,” by Ehud Weiss, Wilma Wetterstrom, Dani Nadel, and Ofer Bar-Yosef.  Proceedings of the National Academy of Sciences.

Wiseman, Regge N.

1987 Book review: Food, Diet, and Population at Prehistoric Arroyo Hondo Pueblo, New Mexico. Arroyo Hondo Archaeological Series 6. School of American Research Press. El Palacio 44-45.