Journal of = Ethnobiology VOLUME 6, NUMBER 1 SUMMER 1986 | Journal and Society Organization EDITOR: Willard Van Asdall, Arizona State Museum, Building 26, University of Arizona, Tucson, Arizona 85721. ASSOCIATE EDITOR: Karen R. Adams, eons of Ecology & Evolutionary Biology, University of Arizona, Tucson, Arizona 8572 NEWS AND COMMENTS EDITOR: Eugene Hunn, a of Anthropology, DH-05, University of Washington, Seattle, Washington 98195. BOOK REVIEW EDITOR: Charles - ieee, Office of Arid Land Studies, University of Arizona, Tucson, Arizona 8 : Steven A. Wi Dep of Anthropology, University of Pennsylvania, Philadelphia, mcsieshit eae 19 SECRETARY/TREASURER: Steven D. Emslie, Department of Zoology, University of Florida, Gainesville, Florida 32611. CONFERENCE COORDINATOR: Jan Timbrook, Department of Anthropology a Museum of Natural History, 2559 Puesta Del Sol Road, Santa cee California 93105. EDITORIAL BOARD BRENT BERLIN, Department of Anthropology, eet of California, Berkeley, California 94720, ethnotaxonomies, linguisti A. BYE, JR., jciecas «teed hele n and Organismic Biology, University of Colorado, Boulder, Colorado 80309; ethnobotany, ethnoecology. RICHARD IL. FORD, Director, Museum of Anthropology, hen of Michigan, Ann Arbor, Michigan 48109; archaeobotany, cultural ecology. B. MILES GILBERT, Box 6030, Department of eee. Ra Te Arizona University, Journal of Ethnobiology Special Sssue VOLUME 6, NUMBER 1 SUMMER 1986 MISSOURI BOTANIK( DEC 92 19 Cn ee a ne ee. a Although I could write at length about the many factors that have gone into this special issue, I'll confine my comments largely to one aspect—networking—to use a trendy concept. It is because of networking, I should imagine, that Steve Weber and Steve Emslie, conceived the idea of an issue of the Journal devoted to ‘‘new directions in ethnobiology”, and no one knows better than Karen Adams and I that several of the papers in this issue would not have been submitted for inclusion if we had not had numerous suggestions and much interest from various ethnobiologists—some members of the Society, some not. It is a pleasure to acknowledge your splendid participation. Society officers Weber and Emslie drafted the initial letter explaining their concep- tion of the special issue and inviting scholars to participate through preparing a “state of the art’”’ paper in their individual areas of ethnobiological involvement. When they submitted this draft, to be sent under my signature, they included a list of scholars they wished to invite. I added a number of names and sent out invitations. Some failed to respond, others offered encouragement but were overly committed to other projects and so declined, and a very few accepted. I next sent a letter to the 15 or so scholars on my list of “Tucson ethnobiologists”, informing them of the special issue and requesting their help in assembling a cadre of authors. The response, although not overwhelming, was more than adequate to kindle the next stage. One of the Tucson ethnobiologists, Joe Laferriere, suggested Stephen Brush and Ben Orlove from Davis, California and they, in tum, recommended Constance McCorkle, so it was through this piece of networking that we have a paper on ethnoveterinary medicine—an area of study about which I had been abysmally ignorant, my only previous contact with it having been a number of anecdotes of practices in Saudi Arabia related by my friend and colleague, Ted Downing. I trust you will find, as I have, this contri- bution on ethnoveterinary medicine as fascinating as it is unusual. Here’s an example of networking operating in a different way. Over the years Bob Bye and I have noted (as, most likely, have all of you) that from time to time, all ethnobiologists work in areas where they have had little training. We have discussed the need for ethnobiologists, not trained in biology, to know the value of voucher specimens. Bob wrote such a paper and although it had been planned for Volume 5, Number 2, unanticipated delays prevented its appearance, fortunately, until this issue— exactly where it should be published. Amadeo Rea and Harriet Kuhnlein were then inspired to write somewhat similar articles in their individual areas of ethnobiological study, so now we have three educationally oriented papers. Through networking we found scholars with interesting ideas about ‘new directions in ethnobiology” who, at the time the invitation was issued, were not already so overly committed but that they could share these views with us. It will be interesting to see what networking will produce in the second special issue of the Journal, at some indefinite future time. —W.V. The Development of a Society: An introduction to the special issue. In partial response to increasing academic diversi there has been a veritable proliferation of specialist journals available to scholars. While this trend has not always been universally welcomed, there are arguably some cases in which such a course of action is wholly justified. The foundation of the Journal and Society of Ethnobiology should be regarded as just such a case. My involvement with the Journal and Society goes back to its inception, and for the purposes of this special issue, which appears on the eve of the tenth anniversary of the first ethnobiology conference, it seems appropriate to consider anew some of the questions and issues surrounding ethnobiology which the institution of the Journal and the Society were partially intended to address. After 10 years, it is still necessary to ask what is ethnobiology? Is it any more than a title or a cover term applied to various types of research which do not fit comfortably into other disciplines? Since ethnobiology has no unifying theory of its own, can it justifiably be considered as an interdisciplinary field, or would it be better to view it as a subfield of already long-established scientific realms? Finally, what common thread exists to link together the diverse researches which have appeared in this journal, or been discussed at the Ethnobiology Conferences? What is the rationale underlying the investment of time and energy in fostering the development of a conference, society and journal devoted to the integration of this research? Ethnobiological research can be described as work that draws on both biology and anthropology to make statements about the interrelationship between living organisms and human culture, whether prehistoric, historic, or contemporary. Interest in this interrelationship is not new. Ever since anthropologists’ initial emphasis on natural history, biology and anthropology have been intertwined. For example, there were the early attempts by anthropologists to classify societies on a scale of evolutionary develop- ment according to their mode of subsistence, where the appropriation of nature was regarded as a critical factor in determining the advancement of other aspects of culture. Later studies focused on the systematic relationship between a sociocultural entity and its environment, and stressed — and change in ecological systems. More recently, interest in th Itigens, domesticated animals and agricultural complexes, and their co-variation with social organization and population, have all helped to stimulate interest and research in the interrelationship between biology and anthropology. There was a time when zoologists and botanists would be asked to contribute their expertise to the identification of biological materials derived from ethnographic and archaeological contexts without being expected to have any great insight into the research orientation as a whole. Nowadays, it is more usual for specialists to take an active role in model and hypothesis building that blends methods, concepts and models drawn from both anthropology and biology. Ethnobiological research is conducted within the con- straints imposed upon it by elements of both biological and anthropological theory, and, in turn, its contributions are weighted according to the prevailing requirements and questions of these two super-disciplines. The first Ethnobiology Conference was held in 1978 in Prescott, Arizona, and was sponsored by the Prescott Center College and organized by Steven D. Emslie. The second, in 1979, took place in Flagstaff, Arizona, and commemorated the contributions to the early development of ethnobiology of both Hargrave and Alfred F. Whiting. The proceedings of this conference were published in 1980 in the first Journal of Ethnobiology, under the auspices of the Center for Western Studies, a private corporation manage by Steven Emslie and me. However, our intention in setting up the Journal was not simply to record the papers given at the Flagstaff conference, but also to provide a much needed forum in which articles of ethnobiological interest could be presented together. In the past, these had been scattered in specialist journals, archaeo-biological data appearing in archaeological journals, plant-human material in plant journals, and so forth, and we felt that this hampered their accessibility to all researchers in ethnobiology, and that the integration and fruitful development of the field could only take place once it was formally established in the framework of a journal. The design we decided to use as the symbol of our newly created Journal was that of a split-twig figurine. The figurine, which is featured on all the covers of the Journal, neatly encapsulated its critical elements—for the original objects are pieces of plant material manipulated by humans to represent an animal. Split-twig figurines date to around 2000 b.c., and are found in the American Southwest. They are thought to have had some magical or religious significance for the Archaic hunter-gatherer peoples who made them. When we chose to use the term ‘ethnobiology’ in the naming of the Journal (partly in deference to the title of the Conference}, we were aware of the way in which the word breaks down into two elements, namely ‘ethno’—(from the Greek ‘etnos’—literally, race or peoples), referring to the human aspects of biological relationships, and ‘biology’, or the study of the entire range of living organisms. This seemed to render the term particularly appropriate for our purposes, since we wanted a title broad enough to cover the range of possible research orientation which relied on the integration of biology and anthropology. Perhaps this appears somewhat self-explanatory. However, ‘ethnobiology’ is not always defined as broadly, nor with the same general outlook as we intended here. One definition I came across recently stated that ‘ethnobiology’ was ‘‘a branch of the study of ethnology which relates to the distinctive physical and racial characteristics of specific ethnic groups or population isolates.’ Initially, we decided to put out two issues a year, one devoted to conference papers, the other to independently submitted articles. We invited a number of scholars to serve on the board, who were known for their achievements and interest in ethnobiclogy, and whose diverse research objectives and accomplishments were essentia al if the journal was to reflect adequately and intelligently the rich variety of data and ideas which could be subsumed within ethnobiology. For the first few years, Steven Emslie and I alternated editorial and managerial responsibilities. Then in 1982, we were elected the first President and Secretary/Treasurer of the Society of Ethnobiology, which we had founded a short time before as a non-profit organization whose primary purpose was to oversee the publication of the Journal and the organization of the Conference. The editorial board then proceeded to find a new editor for the Journal, and selected Willard Van Asdall, who has done an excellent job and contributed considerably to the Journal’s success up to the present. Alongside the foundation and development of the Ethnobiology Conference, Journal and Society, there has been continued diversification and growth within the field they were designed to reflect. It is important, then, that the Society adapts to new trends, and remains sensitive to the changing needs of its membership. At the time of writing, the Society is considering a number of projects for the future, including a monograph series, converting the Journal from a bi-annual to a quarterly issue, and instituting prizes and awards for achievement and excellence in ethnobiological research. The concept of the special issue was conceived by Steven Emslie and me as a periodic means to chart and evaluate the research foci of a fast developing interdisciplinary field. We decided that this would be best achieved by inviting prominent writers and workers who site themselves within the broad realm of ethnobiology to submit articles with the assessment of current and future achievements in mind. We hope this will be the first of several such pia ‘ial which will assist in the refinement and enrichment of ethnobiology as a w This special issue covers a number of important topics and issues of ongoing concern in ethnobiological research. These range from the appraisal of well-established fields in the discipline to the evaluation of new topics and models. For example, Bohrer assesses the achievements of ethnobotany, and offers some suggestions for its future. Delving into a specific category of ethnobotany, Holloway and Bryant present a thorough and critical review of the uses and applications of pollen analyses. In the field of zooarcha- eology, Lyman picks up the theme of species lists and discusses their heuristic poten- tial. Among the relatively new subdisciplines of ethnobiology, a lesser known one is ethnoentomology, and in a useful article, Posey traces its history and, at the same time, considers the value of ethnobiology as a generative source of new ideas. Bye’s discussion of the use and role of voucher specimens represents an important statement on the documentation of biological material so that it is of maximum utility both to the researcher and to later users of the data. Rea is similarly concerned with documentation and verification and he makes a number of recommendations for the more stringent and fruitful ; f ] studies. Kuhnlein’s paper brings us to another immensely important field within ethnobiology, nutritional studies. She too considers problems of method, with specific reference to the collection and chemical analysis of food samples. Interest in subsistence and cultural ecology remains as keen as ever. Winterhalder’s paper on foraging models among hunters and gatherers, and Brush’s on change in farming systems through the perspective of the loss of genetic diversity, mark two of the possible directions in which subsistence studies can go. Orlove and Godoy’s article on Andean highland patterns of crop and pasture management elaborates on familiar cultural ecological themes in demonstrating the complex interaction between agricultural systems and social organization. Another area of | in ethnobiol revolves around the labelling and classification of elements of the natural world. This has developed into an important research orientation including, at one extreme, the simple elicitation of folk plant and animal names and, at another, the exploration of the ways in which people organize and think about their surroundings. Articles by Elisabetsky and McCorkle deal with new and intriguing trends in the study of folk models and categorizations, but these are especially interesting because they represent prescriptive systems that organize both thought and action. Ellen’s paper, on the other hand, provides a valuable critique of the logical and philosophical underpinnings of the p g folk The articles presented in this issue reflect only a small portion of the types of research now considered to be part of ethnobiology, as a glance at the list of contents of back issues of the Journal will confirm. Over the years, article have appeared on such diverse subjects as nutrition, domestication and subsistence, environmental reconstruction, folk classification, and questions of method and theory. While we initially intended the Journal to merely reflect the field of ethnobiology, and to encourage the kind of effective communication among scientists that we thought would be essential if ethnobiology were to become truly interdisciplinary, it has become apparent that through the selection, editing and publication processes, it has actually helped shape ethnobiology. Ethnobiology is growing, and more scientists are prepared to identify their work as such. The simple addition of the _ ‘ethno’ to a word associated with the natural sciences happens increasingly, and signals the inclusion of even more new and fruitful topics under the umbrella of ethnobiology. Perhaps this reflects a specific characteristic of scientific inquiry today, namely the division and recombination of elements from several fields. The richness and potential of blending anthropology and biology in this way will, I hope, continue to be domonstrated by ethnobiology. Already, as Bohrer tells us, ethnobotany is beginning to capture public interest. No doubt this is true as well for other aspects of ethnobiology. Moreover, as ethnobiological objectives focus on broader issues of the integration of cultural and natural systems, as well as change, ethnobiology as a whole may be in a better position to develop theory of its own, as well as to make a richer contribution to anthropological theory than in the past. The future of ethnobiology looks promising. Developments in techniques of data recovery, a growing body of linguistic, ethnographic, archaeological, historical and experimental data, and a better understanding of natural and cultural processes are, of course, critical. However, it is the continual creative integration of ideas and data brought into ethnobiology from many different sources that is most likely to ensure the con- tinued growth and improvement of this newly interdisciplinary field. ven A. Weber President, Society of Ethnobiology vi J. Ethnobiol. 6(1):1-8 Summer 1986 VOUCHER SPECIMENS IN ETHNOBIOLOGICAL STUDIES AND PUBLICATIONS ROBERT A. BYE, JR.! Department of Environmental, Population and Organismic Biology Campus Box 334 University of Colorado Boulder, CO 80309 an Jardin Botanico Universidad Nacional Aut6noma de México 04510 México, DF ABSTRACT.— Voucher specimens sank a a part of ' ethnobiological studies. They physically and permanently d asis for review or reassessment of the original study. An adequate cook specimen must have diagnostic characters, be preserved in the condition, be accompanied by atone ae field data, and be main- eel and teadily accessible in a suitable repository institution. Planning prior to the Because they differ from taxonomic samples, special attention must be given to the collection and maintenance of ethnobiological specimens. In ethnobiological studies, the scientific name is based upon the identification of the voucher specimen and serves as the crucial link between folk knowledge and Western science. INTRODUCTION The voucher specimen is a critical component of ethnobiological studies. It provides the documentation for the scientific identity of the biological material about which obser- vations and data are recorded. To date, voucher specimens have been accepted in some, but not all, disciplines of natural history. With the need to maximize the value and use of these collections, the Association of Systematics Collections has published a report on voucher specimen management (Lee et al. 1982). A voucher specimen is an organism or sample thereof ‘‘which physically and per- manently documents data in an archival report by: (1) verifying the identity of the organism(s) used in the study; and (2) by doing so, ensures that a study which otherwise could not be repeated can be accurately reviewed or reassessed”’ (Lee et al. 1982:5). In order for a voucher specimen to fulfill its function, it must: (1) Have recognized diagnostic characters that are appropriate to the level of identification in the report. Specific life Stages or body parts may be required. (2) Be preserved in good condition by the investi- gator/collector according to acceptable practice. (3) Be thoroughly documented with field and/or other relevant reports. (4) Be maintained in good condition and be readily acces- sible in suitable repository institution” (Lee et al. 1982:7). The following points need to be emphasized so that a voucher specimen meets the requirements. First, the material preserved is a sample of the organism or the population that is actually studied. Second, the sample is adequate for identification and is deposited in an institution where it is cared for and made available to researchers. Third, essential data collected with the sample are physically associated with the specimen. Fourth, the archival report (published or unpublished) makes reference to the voucher specimen. Further details on the practice and justification of voucher specimens are found in Lee et al. (1982). 2 BYE Vol. 6, No. 1 For the ethnobiologist, voucher specimens are essential to his/her work. The specimen is the basis for identifying the organism. Matching a common name with a scientific name is not identification (see Mead 1970). The identification to at least the species level provides western scientists a common basis for comparison of biological, ecological, and cultural data as well as for reevaluation of the information. Although occidental scientific identification, classification, and nomenclature have their limita- tions, they are more easily applied on a universal level than folk taxonomies. With an accurate identification, the scientific name becomes the crucial link between people with folk knowledge and people trained in western sciences. Researchers can correlate and compare information generated and modified by generations of humans with that derived from more recent occidental scientific pursuits. The scientific name is the primary structure for bridging two cultures and for facilitating the mutually beneficial exchange of information. The return of knowledge to communities from which it originated is a critical component of ethnobiology today(Gomez-Pompa 1982; Toledo 1982) and can take many forms (e.g., native author publications, cultural rescue programs, health improvement projects, etc.). The use and appreciation of voucher specimens have been greatest among biological taxonomists. Consequently, they have set the standards for the formation and manage- ment of these materials. As other disciplines such as ethnobiology, ecology, and environmental impact studies develop the role of voucher specimens expands—and so too do the potential uses and the problems of application. The verification of identification and the change of the name (because of more accurate identification, up-dating nomenclature, or revised classification) are most efficiently carried out when the specimen is available to taxonomic specialists through normal channels such as revision of the holdings of recognized repositories or museum collections. Consequently the value of the specimen constantly increases for the taxonomist as well as for the ethnobiologist who can periodically consult his/her specimens for current identifications or annotations. Also, the specimen may yield more data such as chemical, ecological, and structural information for subsequent studies that were not part of the original investigation (e.g., Condon and Whalen 1983; McCain and Hennen 1986). Other disciplines may also benefit from the specimens as sources of data not otherwise tapped in their work. For example, biogeography can be aided by distribu- tional data from collections of an ethnobiological study in an inaccessible area. The early developmental and pre-reproductive forms of plants and animals are frequently encountered in ethnobiological studies and are not usually acceptable as specimens by taxonomists. However, the plants and animals may be in a non-reproductive State at the stage of recognition and employment by native people. Consequently, taxonomists are often reluctant (and sometimes refuse} to identify and manage SU! ethnobiological material. This frustratring problem can actually be resolved to the benefit of both the ethnobiologist and the taxonomist. Often structural characteristics, such a8 juvenile leaves of edible herbs, have been ignored by taxonomists and are not considered to be important or diagnostic. But these same characters may be critical to traditional people who rely upon the tender plants for food. Rather than considering this situation as an impasse to multidisciplinary cooperation, it provides fertile ground for contribution of new data to both areas of investigation. Looking at plant population studies of seedlings as an example, ecologists needed an easy system for identifying in 4 NO destructive manner germinating seeds and seedlings (Duke 1969). Such characters and life stages were not routinely considered by taxonomists. When the forms of get™r — nation and the types of seedlings demonstrated repeatable patterns, taxonomists began to incorporate these characters into their descriptions and identification keys. With 4 ethnobiologists working with native people who are keenly aware of the characteristics of the various life cycle stages of plants and animals, sufficient stimulus should i x Summer 1986 JOURNAL OF ETHNOBIOLOGY 3 present to encourage our colleagues to critically examine underutilized chemical, struc- tural, ecological, and life history features. CATEGORIES OF VOUCHER SPECIMENS A voucher specimen for organisms that are studied or observed may consist of one or more of the following categories (Lee et al. 1982:6-7): (1) the actual organism (whole or part}; (2) a sample of one or more individuals (whole or part] from a population; (3) a representation of the organism or its characters (e.g., photographs, sound recordings, etc.), although it may not be adequate but may be the only alternative that is practical and/or legal; (4) an associated specimen that is biologically or functionally related (e.g., pollen preparations, fiber slides, stomach contents, pathogens, etc.) of the organism; and (5) a corroborative specimen that provides additional data or characters (e.g., from the same individual or population but at a different time or stage in the life cycle) to a previ- ously collected voucher specimen. For the ethnobiologist, the voucher specimen should be the organism actually studied or a sample that originated from the same population at the time of observation (categories 1 and 2). This specimen should reflect the characters, characteristics, and stage of life cycle about which the informant or collaborator provides information. Often the condi- tion of the specimen may not reveal the diagnostic features (usually reproductive parts) that are required by the taxonomist for identification. Therefore, every effort should be made to also obtain a corroborative specimen that is collected from the same organism or population from which the original voucher specimen originated. This can be done by marking and recollecting later the organism or population in order to obtain material with reproductive and other taxonomically important features. In some cases the organism can be cultured, grown or raised until it reaches maturity, such as the case with a fresh root or seeds which can be planted and later pressed with flowers, fruits, and leaves for an herbarium specimen. In all cases, cross reference should be made between the original voucher specimen and the corroborative specimen and their relationship clearly noted. Sometimes, different collection numbers for the two specimens may be made but the connection between them should be specifically stated. Other times, the same collec- tion number can be used for the two specimens but the different collection dates are noted as well as the relationship. The other category of voucher specimens that is useful to the ethnobiologist is the associated specimen. Often the material under study (e.g., sample, phytoliths, wood fiber, seeds, exudates, bone tissue, etc.) is not the main character studied by the taxonomist but is the basic evidence used by the ethnobiologist. In such cases, a voucher specimen of the organism(s} is made according to the standard accepted by the experts studying that taxonomic group. Then, the special preparations of the parts (e.g., pollen, tissue sections, chemical extracts, etc.) are made from the voucher plant or animal specimen and analyzed. The results (products and data) are deposited along with a cross reference to the voucher specimen. Such special preparations may require separate management in ancillary collections such as those for pollen (palynological collection), wood (xylarium), fruits and seeds, stomach contents, etc. These associated specimens are linked directly to regular museum collections available to taxonomists. Also, the associated specimen should be duplicated so that one remains at the designated repository with the original voucher specimen and the other forms part of the investigator’s comparative collection. Such collections with selective parts and data are then used to identify (by comparison and degree of similarity) aterials from ] or contemporary sources. These comparative collections should include the distinctive parts that would be encountered in pre-processed, processed, utilized, and disgarded conditions. 4 BYE Vol. 6, No. 1 Today most collections of organisms are made so as to reflect the range of variation of characters of taxonomic and evolutionary significance. When an ethnobiologist pro- duces voucher specimens, his/her collection should be a representative sample of the population. Also extra efforts should be made to obtain sufficient quantity of the parts required to make the associated specimens from each voucher specimen. Hence an adequate number of mice or herbs should be collected if associated collections of teeth or seeds are to reflect significantly the variation in these parts. When using the comparative collection to identify unknown materials, the degree of similarity, the identity confidence, and the level of identification should be determined by establishing a basic set of quantitative and qualitative characteristics for each known taxon. Thus the identification can be determined objectively to a specific degree of accuracy. When reporting the new identification reference should be made to the characters used and the source of the comparative collection, including the citation of its voucher specimens. When reporting the identification of a previously unknown organism, one may not be 100% confident of the determination. The degree of confidence may be expressed by listing only the name of the taxonomic level at which one is sure. For example, if you are confident that the fruits are those of the member of a family, you may cite Chenopodiaceae. If you believe that they are members of a given genus, can report Chenopodium sp., or the subgenus if possible (e.g., Chenopodium, subgen. Chenopodium). Should you be confident that the material is of the species, you can provide the name with the generic name, specific epithet and the author (e.g., Cheno- podium album L.). But what if you are not sure about this identification? If the original material is in good condition and is more similar to that species than others in the area (but the other species may have indistinguishable fruits), one can report it as Chenopodium aff. album L. where “aff.’” means affinity. Bohrer and Adams (1977) suggest the use of the term type {i.e., Chenopodium album type) but that may be confusing since the type concept refers to nomenclatural type in taxonomic studies where there are specific rules, procedures and categories of types (Ride et al. 1985, Voss et al. 1983). If the original material is of poor condition but appears similar to the verified sample in the comparative collection, one can cite the identification as Chenopodium cf. album L. where “cf.’’ means com- pare (Bohrer and Adams 1977). This approach using corroborative and associated specimens has been involved in the early development of ethnobotany in the United States and Mexico. During the 1870s and 1880s, Major J. W. Powell and Dr. Edward Palmer obtained seed collections of edible grasses and herbs of the Southern Paiute Indians (Bye 1972). The seeds were collected from the Indian gathering baskets as well as from the prepared foods. This material serves as the voucher specimens. Dr. A. Gray and §. Watson grew the seeds and obtained herbarium specimens from the reproductive plants. This material serves as the corroborative specimen, upon which the identifications were based. Dr. Palmer extended this approach in his Mexican studies (Bye 1979) by collecting plant products at various stages of preparation and in the forms of exchange and storage (e.g., market bundles) while also obtaining herbarium specimens of the plants in the natural condi- tion. In order to distinguish the former specimens from the herbarium collections, Palmer called them “case specimens”. PREPARATION OF VOUCHER SPECIMENS Each taxonomic group of plants and animals requires special preparation in order to make an acceptable voucher specimen. Many plants and animals have relatively simple methods of preparation in terms of sampling, selecting material, killing, fixi and preserving the items. Other groups require considerable effort, materials, and equip- Summer 1986 JOURNAL OF ETHNOBIOLOGY 5 ment. Lee et al. (1982: Appendix I) presents a list of references to preparation of major taxonomic groups. It is always best to consult prior to the collection of specimens a taxonomic specialist in order to obtain advice and training. It is preferable to seek directions from the repository of the specimens so that the material will meet their specifications. Apart from the actual preserved organism, the basic data accompanying it (i.e., physically attached to the specimen or to the collection archives) should include (Lee et al. 1982:15-16): (1) a unique sample designation such as a collection number for each sample collected at one place and time where each collector or project maintains a sequential, non-repeating numbering system; (2) the position of the sample collection (including: country, state and other political subdivisons; natural geographic location such as river system, sea, etc.; local place name(s}; international location designation such as latitude and longitude or Universal Transverse Mercator grid; altitude or depth; and habitat); (3) the date (and time when appropriate} of collection; (4) the name of the collector (and other appropriate donor identification including project, station, and field number); (5) the identity to the lowest taxonomic level, such as to species where feasible; and (6) methods of collection and preparation, where appropriate (e.g., sample as a result of traditional harvesting technique; from cooking pot; from market stand; etc.}. Ethnobiological collections require additional data. Vernacular names should be designated and the language and etymology noted when possible. Specific terms or cultural characteristics should be recorded in the language used to obtain the information; if possible, the original phraseology should be retained. A summary of these data can be provided in the investigator’s native tongue. Reference can be made to the name of the informant or collaborator, supplementary or confirming information provided by others, as well as other pertinent data sought in the study. DEPOSITION OF THE VOUCHER SPECIMENS A critical element of a voucher specimen is that it be deposited, maintained properly, and available to researchers. Consequently, an investigator should consult the primary repository and the collection curator prior to obtaining specimens. Such advance-planning will assure that the sample is adequate for the study and that the specimens will be accepted in the collection and will be useful to the investigator and other researchers. In some cases, an institution may require fulfillment of certain prerequisites such as condition of the specimens, proper collecting permits, payment of handling fee, minimum sample size, etc. Also the investigator will benefit by knowing specifically what and how to collect as well as by having the institution make the contacts with appropriate taxonomists or having the specimens made available to specialists. If the ethnobiologist makes personal contact with a specialist, he/she should be sure that the specimens will be deposited in an appropriate repository (and not in a private collection) or that duplicates of the voucher specimens will be deposited according to the criteria above. . . Many ethnobiologists have faced the difficulties of finding museum collections with adequate parts of material for study and comparison. Hence, many of us have generated specimens that would be considered associated specimens as part of individual com- parative collections. We have also faced the frustration of curators who do not appreciate these specimens or whose institutions cannot support adquate curation of specimens that do not fit their current practice. Until ethnobiological collections become more acceptable and supportable, what are we to do? First, follow the initial step above—make _— with the taxonomist or curator prior to the study, If part of your collections are ~~ they meet t torial practices, the curator aad institution may be be willing to ‘match your donation with an ‘ancillary collection support. If that fails, do not give up on the original voucher specimens—make and deposit 6 BYE Vol. 6, No. 1 them. Then with the duplicated associated specimens, seek other institutions where your duplicates will be incorporated into the appropriate ancillary collections. If your vouchered associated specimens do not find their way to an official repository, share them with a colleague. The main points are to: (1) deposit your original or corroborative voucher specimens in a proper repository, (2) link them with your associated specimens and (3) have duplicates of the specimens apart from those in your parati llection available to others. One should realize that curators of museum collections do not receive adequate support or recognition for what they have now. This problem was one of the primary reasons for the Association of Systematics Collections’ effort to publicize the importance of such materials (Lee et al. 1982). CITATION OF VOUCHER SPECIMENS The existence of voucher specimens and the publication of the results based upon the studied organisms are interdependent. Therefore ethnobiological reports in forms of articles, books, and contract reports should include: (1) the scientific name of the plants and animals, and (2) the citation of the voucher specimen(s). In the case of the scientific name, it should be reported at the most accurate taxonomic level possible. Usually the species is accepted as the most useful taxon. Infraspecific taxa (e.g., subspecies, variety, cultivar, etc.) should be used where possible. The specific and infraspecific epithets should be followed by the author in order to complete the scientific name. The voucher specimen must be cited by a distinct identifier which includes: (1) a unique number for the specific item (e.g., collector's name and collection number or accession number assigned by the repository], and (2) an identifier for the repository. This identifier may be the name of the institution and its subunit or collection or, as in the case of major institutions, a code designation or abbreviation. A list of references on repositories for various taxonomic groups and their abbreviations is listed in Lee et al. (1982: Appendix II). An example of a voucher specimen citation is: Datura inoxia Miller, R. Bye # E. Linares 13,101, MEXU. The herbarium specimen of jimsonweed (Datura inoxia Miller) was collected by R. Bye and E. Linares, has their collection number of 13,101, and is deposited in the Herbario Nacional of the Univer- sidad Nacional Auténoma de México. By convention (Council of Biology Editors Style Manual Committee 1983), the collector name(s) and number(s}, like the generic name and the specific and subspecific epithets, are pinted in italics or written underlined. If one’s study includes others’ collections, then the collector, collection number, and repository are cited in a similar manner with an exclamation mark (!) following the repository’s abbreviation (e.g., MEXU)). CONCLUSIONS In ethnobiological publications using names of organisms, the scientific name (tO the most accurate taxonomic level) should be reported in addition to the native name(s). The name should be based upon the identification which in turn is documented by 4 voucher specimen. The voucher specimen serves to verify the identification, to update the identification and nomenclature, and to provide additional data, especially as techniques advance or if the study is not repeatable. Various categories of voucher specimens exist. In addition to the original voucher specimen which may contain ethnobiologically important features but may not have taxonimically diagnostic traits, the corroborative specimen may be required to provide more critical characters. Associated specimens may be useful in forming comparative collections for identifying processed or archaeological biological materials. Summer 1986 JOURNAL OF ETHNOBIOLOGY 7 The preparation of voucher specimens requires consultation with taxonomic experts and curators in advance of collection. Ethnobiological specimens need basic museum data along with other label data which are physically associated with the specimen or the collection. In order to fulfill their function as voucher specimens, samples have to be deposited at a repository where they will be maintained and will be available to researchers. In archival reports, studies based upon biological organisms should include the scientific name (including the author) as well as the citation of the voucher specimen (which is indicated by a distinct identifier: a unique number of the specific item as well as the name or abbreviation of the repository). ACKNOWLEDGEMENTS I thank W. L. Lee, B. M. Bell and J. F. Sutton for the invitation to participate in the Conference on Voucher Specimen Management, the members of the Work Groups as well as other participants provided stimulating ideas. Financial support for the Conference was provided by the National Science Foundation (Grant No. DEB-8020909} to the Association of Systematics Collections. The Editor and members of the Editorial Board of the Journal of Ethnobiology provided encouragement. K. R. Adams and A. M. Rea kindly shared suggestions. NOTES IServed as chairman of Characterization of Voucher Specimens Committee for the Council of Curatorial Methods of the Association of systematics Collections. 2"Guidelines for Acquisition and Management of Biological Specimens” can be ordered from the Association of Systematics Collections, Museum of Natural History, University of Kansas, Lawrence, Kansas 66045. LITERATURE CITED BOHRER, V. L. and K. R. ADAMS. 1977. presentation of herbivore and Ethnobotanical Techniques and pathogen damaged plant materials. Approaches at Salmon Ruins, New Taxon 32:105-107. Mexico. Eastern New Mexico Univ., COUNCIL OF BIOLOGY EDITORS Contr. Anthropol. 8(1):xix + 1-214. STYLE MANUAL COMMITTEE. 1983. CBE Style Manual: A guide for BYE, R. A., JR. 1972. Ethnobotany of authors, editors, and publishers in the Southern Paiute Indians in the 1870s; biological sciences. 5th ed. rev. and with a note on the early ethno- expanded. Council of Biology Editors, botanical contributions of Dr. Edward Inc., Bethesda, Maryland. Palmer. Pp. 87-104 in Great Basin DUKE, J. A. 1969. On tropical tree Cultural Ecology, a Symposium a4 seedlings. 1. Seeds, seedlings, D. Fowler, ed.). Desert Research Insti- systems, and systematics. Ann. tute Publications in the Social Wicd Tee, Gand. 605-161. Sciences, No. 8. Reno, Nevada. GOMEZ-POMPA, A. 1982. La etno- rice cay Sy Re Le ee botanica en Mexico. Biotica 7(2): botanical collection from San Luis 151-161. Potosi: Dr. Edward Palmer’s first LEE, W. L., B. M. BELL, and J. F. SUTTON major Mexican collection. Econ. (eds.). 1982. Guidelines for acquisi- Botany 33:135-162. tion and management of biological CONDON, M. and M. D. WHALEN specimens. Association of System- 1983. A plea fines collection wna atics Collections, Lawrence, Kansas. 8 BYE Vol. 6, No. 1 LITERATURE CITED (continued) MEAD, G. R. 1970. On the improper Zoological Nomenclature. Third Edi- usage of common names when giving tion. International Trust for Zoologi- botanial data. Amer. Antig. 35(1): cal Nomenclature, London. 108-109. TOLEDO, V. M. 1982. La etnobotanica MCCAIN, J. W. and J. F. HENNEN. 1986. hoy. Reversion del conocimiento, “Big fleas have little fleas” (big plants lucha indigena, y proyecto nacional. have little plants] even in herbaria. Biotica 7(2}:141-150. ASC Newsletter 14(1):1-4. VOSS, E. G. ET AL. (eds.) 1983. Inter- RIDE, W. D. L., C. W. SARBOSKY, G. meted Code of Botanical Nomen- BERNARDI, and R. V. MELVILLE clature. Reg. Veg. 97. Bohn, Schel- (eds.). 1985. International Code of tma & Holkema, Utrecht. J. Ethnobiol. 6(1}:9-18 Summer 1986 VERIFICATION AND REVERIFICATION: PROBLEMS IN ARCHAEOFAUNAL STUDIES AMADEO M. REA Curator of Birds and Mammals San Diego, CA 92112 ABSTRACT.—Biological materials from archaeological excavations, particularly faunal remains, are of most lasting scientific value if they are properly — individually numbered, and adequately reported in published accounts. A gi re urged never to put bones in plastic bags. A ] accounts 1 reasons for their devermninaticits, to mention comparative 1 materials used, and to give individual subs with their p sequent study and reverification. Knowl dge of the local fauna enhances the validity es interpreta- tions. Better appreciation of the ‘culture and its particular technology would improve biologists’ accounts. Archaeological bones have been used in reconstruction a past diets “a environments. _ They are me use to both paleontologists — neontologists. A signifi h t modifica- tions (particularly insular extinctions and extirpations) in species flenniticn, Eslataie com- parative osteological collections are weak in quantity and sometimes poor in quality; 82% of the world’s ca. 9,000 bird species are represented by ten or fewer skeletons in museums worldwide INTRODUCTION The very basis of science is empirical evidence. The taxonomic disciplines are based on specimen documentation, publicly available, subject to reverification. Ethnobiological studies are no exception. The biological specimens that a native consultant associates with her or his ethnotaxonomic identifications (voucher specimens, see Bye 1986}, together with that person’s critical comments, should be available to future workers once the research is published. Seemingly insignificant or casual comments later prove not to be so, but rather psychological criteria by which a native speaker delimits some ethnotaxon. With the onsaught of cultural homogenization, often the ethnobiologist is doing salvage ethnography, dealing with a small remnant of knowledgable native speakers; there may be no future generation to ask. While ethnobiological voucher specimens are delib ly collected with supporting ethnographic information, the axchaeazoological ' specimen has been preserved fortui- tously, only to be recovered by modern excavators, now increasingly aware of the significance of biological remains. In an earlier stage of the evolution of archaeology as a science, only worked bones—artifactual materials—were considered important. Later, all archaeological bones were analyzed, but only artifacts were saved. With the expan- sion of contemporary research techniques, it is now realized that all biological materials collected from dated proveniences are of , research value. direct my remarks here primarily to vertebrat ins and my examples are mainly from bird bones, with which I work, but the peace apply more generally.! SOME METHODOLOGICAL CONSIDERATIONS Handling and Packaging. When I asked a number of colleagues what topics to include in this paper, the first and most important response invariably was a plea for the proper 10 REA Vol. 6, No. 1 handling of fragile bones by excavators. All other problems pale by comparison. In particular, bird, amphibian, and reptile bones are fragile and easily damaged, especially after years in the soil. More often than not, archaeologists bag in the field bones with lithic artifacts. After preliminary laboratory sorting, bones, often several together, are carelessly thrown into small plastic bags designed for sandwiches, then these “baggies” are all dumped into a paper sack. Finally, the loosely packaged materials are mailed to the faunal analyst. By this time fresh breaks in the bones are usually evident everywhere and often the diagnostic features have crumbled. Gilmore’s (1946, 1949) early pleas for better handling of materials are still apropos. A simple rule: never put an archaeological bone into a plastic bag. Each bone should be wrapped carefully in cotton or some fine paper. (A soft paper, 11.5 x 11.5 cm, excellent for this purpose, is commercially available, in rolls, plain or scented; it is available on virtually every archaeological dig.) The individual bones should be packed so one cannot damage another and put into sturdy containers such as small cardboard boxes or plastic vials that are sufficiently long that the well-padded articular ends will not be damaged. The ee of the bones being identified to species will be enhanced greatly by this simple procedur Each individual specimen should be marked permanently with a unique identi- fying number, letter, or combination of these. In this way the faunal analyst will recognize the individual specimen when it is laid out in comparative series during identification and can coordinate the specimen later with matching individual data sheets or slips pro- vided by the archaeological team. Any published report should include this individual identifying number as well as the element name so that future workers, wishing to verify a determination, will know specifically which fragment or element had been so identified. Editors overly concerned with space would make this concession to science. The old (but unfortunately not yet fully abandoned) practice of simply putting data- bearing bits of paper in plastic bags or vials with the specimens, or, still worse, attaching paper scraps directly to the bone with rubber bands that eventually disintegrate, is a disservice to science! If each artifact from a site merits an individual identifying number, certainly archaeological bones deserve no less. A word to the technician cleaning and numbering vertebrate remains: when possible, avoid marking over such critically diagnostic areas as the articulating ends, muscle scars, and tendinal tubercles; bare sections of shafts are the best for numbers. The archaeological literature is replete with tantalizing early reports of animal remains, often analyzed and reported by workers unfamiliar with the local fauna, unaccompanied by any critical comments or provenience details within a site, and usually lacking specimen numbers. The retrievability of individual specimens is critical to science. Many questions may arise not thought of by original workers. Some examples with turkeys, Meleagris spp., illustrate this problem. Steadman (1980) presented a comprehensive study of the subfamily Meleagridinae ginning in the Lower Miocene. The collection numbers and institutional depositions of all recovered and reidentified specimens were included. Ranges were delimited through geologic time to the historic era. Grayson (1977) questioned the enormous range extension implied by wel oly (197 1) report of “large quantities of turkey bones’ excavated from Connley Cave in the northern Great Basin, dated 3,000-11,200 years B.P. The bones had been eoiited by well-known ornithologists at a major museum. Because the bones had not been individ- ually numbered and details published, the entire avifauna had to be reassembled and reanalyzed in an attempt to determine which elements had been identified as turkey. 2 area Grayson wi was able to demonstrate that the reported number of Meleagris o the number of male Sage Grouse, Centrocercus ba sianus, and the general category fancies pengidialieev: a S o Summer 1986 JOURNAL OF ETHNOBIOLOGY 1] of female Sage Grouse in the sample. Sexual dimorphism had been misinterpreted as taxonomic differences. Rea (1980) showed that the reports turkeys in the Southwest U.S. applied to two different species, one a paleospecies, M. crassipes, the other a neospecies, M. gallopavo, appearing locally subsequent to the development of sedentary agriculturalism. However, in the hot low-desert sites occupied by the Hohokam culture, few turkey bones have been recovered (McKusick 1980; Rea 1983:140). Most of those reported (Fewkes 1912; Gladwin et al. 1937; Greenleaf 1975) cannot now be found among artifact collections. Apparently few workers have appreciated the potential confusion between turkey and fragmentary elements of Sandhill Crane, Grus canadensis (see Hargrave and Emslie 1979). McKusick (1980) identified three strains of turkey from southwestern archaeological sites, two domesticated, one presumed feral. In completing her monographic work on these forms, she has experienced repeated frustration in relocating reported archaeological bones, often not saved “because they were just turkey” (McKusick, pers. comm.). Humans, directly or indirectly, have been agents in the spread of vertebrates into new areas, as we know so well in our own time from the Common Starling, Sturnus vulgaris, House Sparrow, Passer domesticus, and House Mouse, Mus musculus. Polyne- sians were aided in their simplification of Pacific island biota by the dogs, pigs, and rats they introduced (Olson and James 1982a, 1982b, 1984). The role of Indians in the spread of animals outside their natural ranges has generally been underestimated; but the archaeo- zoologist must be alert to the possibility of finding imported species in archaeological sites Bird trade in the Southwest is now well known through Hargrave’s (1970) study of macaws and McKusick’s (1976) and DiPeso’s (1976) analyses of the Chichimeca trading center in Chihuahua, Mexico. Additional species are still to be reported. Haemig (1978) used ethnohistoric and documentary evidence to explain the presence of the Great-tailed Grackle, Quiscalus mexicanus, in the Aztec capital. (Haemig’s [1979] interesting sug- gestion of the jay Cyanocorax dickeyi being in Mexico as a descendent of a human- introduced population of the Ecuador-Peruvian Cyanocorax mystaclis needs to be checked by comparative osteology; Hardy [1983] noted differences in their vocal repertoire.) In Puerto Rico and the Virgin Islands Olson (1982) has found the extinct rodent Isolobodon Portoricensis only in the context of cultural deposits, never in Pleistocene cave deposits on these islands. Hamblin and Rea (1985) found rather extensive evidence of bird impor- tation to Isla Cozumel. The Common Turkey, Meleagris gallopavo, has been carried aboriginally both to the north and south of its natural range (Rea 1980 and unpublished notes). These are only a few examples of prehistoric peoples spreading animals to new areas, but they should serve to jar archaeozoologists out of provincialism. Contamination. Healthy debate has been a hallmark of archaeology and in archaeozoology is productive when specimens can be retrieved and the particulars of provenience and association known. An example will illustrate this. Old World chickens, Gallus gallus, became widespread in the New World so quickly that their Hispanic introduction has been questioned (see summary in Gilmore 1950:393-394). Carter (1971) proposed, primarily on linguistic evidence, that the chicken Preceded the Spaniards. In places the archaeozoological record appears to support this hypothesis, at least superficially. At the Las Colina site in Phoenix (A.D. 1100-1450, Hohokam), chicken bones were common, but accompanied with Muscovy Duck, Cairina moschata, domestic Mallard, Anas platyrhynchos, Chinese Goose, Anser cf. cygnoides, domestic cat, Felis cattus, and domestic pig, Sus scrofa (Johnson 1981; Rea 1981). At Pueblo Grande ruins, a similar site in Phoenix, I identified chicken, as well as the Rock Dove (domestic pigeon), Colmba livia, in the small archaeofauna. Wasley and Johnson (1965) recovered chicken bones in a Hohokam platform mound in westem izona, but noted they (as well as a shotgun shell base) were from the bottom of an 12 REA Vol. 6, No. 1 area disturbed by pot hunting. At Mayan sites on Isla Cozumel dating up to the historic period, chicken bones (1.1% of the avifauna] were recovered from the top levels of pits (Hamblin and Rea 1979, 1985; Hamblin 1984). In all these cases it is either certain or highly probable that the chicken bones were intrusive. (See Hargrave 1972:6-14 for suggested leads on early chicken research.} PUBLICATIONS AND INTERPRETATIONS While papers are often filled with seemingly endless extrapolations and inter- pretations (anthropologists are not noted for brevity, as a review of journals demonstrates], editors and manuscript referees should insist that faunal analysts mention the related species they considered but rejected in reaching some identification. For instance, does an identification of Buteo jamaicensis mean that that was the only species of large Buteo the analyst happened to have for comparison? Were other expected species eliminated? Was the analyst aware of the potential range of species that once may have occurred within hunting range of the site? Readers are entitled to this information. Much more convincing would be a statement such as: the tarsometatarsus conforms in characters to B. jamaicensis, a species common in the area; the similarly sized B. regalis, B. lagopus, and Parabuteo unicinctus were eliminated on characters of the trochleae; B. niditus and B. swainsoni were eliminated on size. No skeletons of B. albonotatus or Buteogallus were available for comparison. Certain species are of significance because of the temporal occurrence documented by the archaeological context. Many bird species have been extending their ranges northward during historic times (Phillips 1968; Rea 1983:87-90). The discovery of the Norther Cardinal, Cardinalis cardinalis, in several archaeological contexts in the Southwest merited careful analysis (Ferg and Rea 1983). The possibility of its being a trade item from the ancestral southern range could not be eliminated. A reader, coming upon such a report without comment and careful substantiation, should indeed be skeptical. Recently, after five attempts with an anthropological editor from a major U.S. univer- sity, I failed to reinstate an introductory sentence in my faunal report to indicate country, location, cultural horizon, and approximate date of the site. The reasons he gave for his veto were that the site was well known (it is on another continent) and if I included such an introduction, then others might want to do likewise, contributing to redundancy in the overall work. Why not? Individual chapters and appendices by faunal analysts are often reprinted and circulated to specialists in the area. Such reports should be comprehensive in themselves, without reference to the whole work. While archaeologists are often satisfied with a mere catalog of recovered taxa and perhaps a bit of “interpretation” (the amount usually inversely related to the quantity of bone recovered!}, biologists may sometimes be faulted for thinking in terms of generic “man” and “his” effects on environments, as if ‘“‘“man’”’ somehow came devoid of all com- plexities of culture. In the Southwest, for instance, one could not equate the types of habitat modifications resulting from the town-dwelling Pueblo Indians, the nomadic Athabascans, the riverine rancheria Pimas, and the two-village, maximally xeric-adapted Papago bands. Each has a different language and classification system, a different set of oos, a different relative reliance on hunting, gathering, and agriculture. Even in this arid region, cultures could afford to be selective (Rea 1981). Islanders, perhaps, could not. “Panhuman” explanations of faunal and floral changes and extinctions are too sweeping, particularly where a diversity of technologies existed. L. Marshall (1984:805) has cautioned overkill enthusiasts to view extinction on each landmass as a discrete pit Archaeologists may be able to fine-tune this picture even more precisely in time and space. Summer 1986 JOURNAL OF ETHNOBIOLOGY 13 Perhaps the most frequent use of vertebrate remains from sites, both paleontological and archaeological, has been in pal ] j While each method of interpreting data has its pitfalls (see the extensive discussion in Grayson 1981), the presence of indicator species in a site will always remain a primary value of archae- ozoological studies. The validity of any interpretation depends on two factors: the fidelity of an organism to a particular habitat or microhabitat and the interpreter’s familiarity with the behavior and ecology of that organism. Some species are such generalists or are so mobile that presence in a site yields little ecological information (e.g., Great Horned Owl, Bubo virginianus, Coyote, Canis latrans), The Red-winged Blackbird, Agelaius phoeniceus, and Sora, Porzana carolina, may indicate mesic habitats, but do not have high fidelity during migrations, for instance, But the Least Bittern, Ixobrychus exilis, and Water Shrew, Sorex palustris, are indicators of well-developed emergent or other low surface vegetation in riparian habitats. Suites of species with narrow preferences are ultimately the most informative. Sometimes two closely relates species replace each other in response to habitat alterations. In the Southwest, for instance, the Gambel’s Quail, Callipepla gambeli, and Scaled Quail, Callipepla squamata, are almost invariably allopatric. Degradation of desert grasslands by overgrazing and consequent invasion of thornscrub increases Gambel’s Quail habitat at the expense of the other (Phillips et al. 1964; Rea 1973). Archaeozoologists (and this point is not limited to them) may be quite adept at species identifications in laboratories, but may have very little experience with the species in life. Unless the interpreter has had actual field experience hunting, trapping, netting, or fishing in the appropriate types of habitats, the interpretations will likely be naive at best. (No examples need be cited.) Familiarity with the habitat and the species found there should be a prerequisite for interpretations. Additionally, it would be appropriate if archaeologists in charge of team projects would arrange visits to their site by the biologists who will analyze floral or faunal materials. More questionable, in my mind, are some of the attempts to extrapolate prehistoric diets from archaeofaunal samples, even though many elaborate models for analyzi have been devised. Often only a mere handful of vertebrate remains may be recovered from a site, sometimes because of poor recovery techniques (too coarse screening), more often because very little material was preserved. Without direct evidence of cooking, it is hazardous to equate presence in a site with food or even with human agency. Rodents and reptiles, for instance, find ruins excellent localities for burrows, where some may die. Pot hunters, picnickers, and ranchers may contribute to the faunal assemblage. In my own experiences with peoples on a subsistence economy in the tropics and deserts, dogs seem to be the primary filterers of discarded bones. It is only by accident that anything survives. Overinterpretation of faunal remains is a futile exercise. Excavated biological material may serve purposes not originally anticipated by the archaeologist. Stanton’s Cave in the Grand Canyon, Arizona, was originally excavated to learn something of the enigmatic people who cached there split-twig figurines (the Society of Ethnobiology’s logo). But the 43,000-year accumulation of materials recovered in the meticulous fine-screened excavation ultimately proved of greater paleobiological terest (Euler 1984). eontology and archaeozoology may overlap, even in the relatively younger contexts of the New World sites. In the West Indies, for instance, several extinct birds are known only or primarily from middens: a crow, Corvus pumilis, a macaw, Ara autochthones, and a rail, Nesotrochis debooyi. The unidentified Ara sp. and Aratinga sp. from Postclassic Mayan sites on Cozumel Island are either new species or long-distance imports (Hamblin and Rea 1985). On the Pacific coast Morejohn (1976) recovered the extinct Pleistocene flightless duck, Chendytes lawi, from early middens. Extinct species of Hawaiian birds are just being discovered and described, many overlapping with Polyne- 14 REA Vol. 6, No. 1 sian colonization. Olson and James (1982a, 1982b, 1984) so far have documented the extinction of 54% of the endemic species of Hawaiian land birds, extinctions attributed to pre-European habitat changes and hunting. Dated archaeological faunas have always been an important source of specimen documentation for the former distributions of species now extinct, endangered, or geographically restricted. Parmalee has been prominent in documenting former ranges of both birds and mammals in North America (Parmalee 1958, 1960, 1961, 1967, 1971, 1981, etc., Parmalee and Perino 1970). Perhaps the most significant theoretical implications have arisen from such analyses of vertebrates recovered from islands, particularly those faunas permitting pre-human and post-human comparison. Such work is generating responses that seriously question the baselines used in the voluminous biogeographical studies of supposed fauna equilibrium and “turnover” on islands (see Olson and James 1982, 1984:777-778, Stead- man and Olson 1985, Steadman 1986, ii, 88; see also Rea 1983: 14 on deserts). COMPARATIVE COLLECTIONS: THE WEAK POINT Growth. Curators and other managers of collections used in faunal analyses should continuously upgrade their osteological collections, both in species numbers and the number of individuals in each series. I know of some analysts who regularly work with only a small number of species represented from their geographic area. One must be skeptical of their scientific results. Collections must grow if they are to be useful research tools, and state and federal permit agencies as well as museum and university adminis- trators must continuously be educated to this fact. Most have never had the practical experience of having attempted identificati Archaeof. 1 analysts might help remedy such ignorance by regularly supplying reprints and copies of reports to agencies and administrators. ertain researchers, particularly some “field” ornithologists and theoretical ecologists whose studies are unverifiable, have proposed in various recent publications that museums have sufficient numbers of specimens already, particularly of birds. Any working archaeozoologist, paleontologist, or taxonomist knows the falsity of such statements (see Olson 1981). In a worldwide inventory of avian anatomical skeletons, Zusi and others (1982) discovered that about a third of the species of birds of the world (2,706 species) are not represented anywhere by even a single skeleton. Of the approx- imately 9,000 avian neospecies, over 7,400 (82.3%) are represented by ten or fewer skeletons worldwide. Only 2.1% of the species are represented by over 200 skeletons in the world’s museums. Quantitative studies of avian skeletons are virtually impossible. Those who appreciate the use of osteological collections should be aggressive in their conservation and expansion. Only rarely should a specimen be discarded. What may be superfluous in a local collection will likely be missing in some other. Frequently it is necessary to borrow skeletons to complete a faunal analysis. When possible it is best to request the specific element needed rather than trust an entire valuable specimen to the mails. An updated listing of avian skeletal holdings in the world’s Major museums is available (Wood and Schnell 1986). Quality. The mere existence of a skeleton does not assure its utility. The quality of extant osteological preparations in many university and museum collections is often sO poor that the materials are nearly useless to the archaeozoologist or paleontologist. Poor cleaning and grease saturation are two major problems. Shot or impact damage an loss of elements through misguided use further reduces the utility of specimens. Skeletons cleaned by dermestid beetles often have diagnostic features obscured by adhering bits of tendons and cartilage, and they are sometimes more greasy than skeletons Summer 1986 JOURNAL OF ETHNOBIOLOGY 15 cleaned by other methods. Bacterial maceration usually provides the cleanest skeleton. However, cranial elements usually become badly disarticulated or lost. Such parts as the quadrates, pterygoids, lacrimals (ossa prefrontalia), larynx, syrinx, and sclerotic rings are most useful to phylogenetics if preserved in situ. My personal choice of preparation methods is maceration for post-cranial skeletons and dermestids for all cranial parts (including trachea and hyoids). Fat birds are best cleaned entirely by dermestids because the lipids may cause bone disintegration during maceration. Grease, which may saturate the bone walls from the inside, often renders the normally opaque external surfaces translucent. Working collections must systematically be checked for grease. Degreasing is expensive, time consuming, and dangerous. Ammonia water will sometimes remove light saturations, but will usually take off the carbon ink specimen numbers as well. Heavier saturations usually require lengthy soaking in a solvent such as acetone or gasoline, with rinsings in several progressively cleaner baths. Bones of young animals occur in archaeofaunal assemblages. This is not surprising because they are seasonally more abundant and often more vulnerable to human preda- tion. The rearing of young birds and mammals is a widespread practice among many North and South American tribes, particularly tropical ones. Nestling birds and juvenals still growing are very rare in osteological collections. Those in charge of building com- parative collections should make a special effort to salvage juvenal and young. Hargrave (1970, 1972) was one of the few ornithologists who consistently collected skeletons of juvenal birds, illustrating their diagnostic specific and generic features. Using his materials for comparison we were able to distinguish two species of turkeys even in young not fully grown (Hamblin and Rea 1985). SUMMARY It may appear that I have belabored some points here that seem obvious. Would that this were so. The archaeologist who walks into my laboratory with a bunch of unpro- tected, unnumbered bird bones in a “baggie” is the rule rather than the exception. Reports that give no collection numbers and fail to mention how an identification was arrived at or what species were compared still continue to be published, but are becoming fewer. If archaeozoology is to maintain itself as a science, the specimens on which it is based must be permanently stored and retrievable for reverification. The scientific value of dated f | materials may far exceed that originally envisioned by the archaeologist. Biologists have much to learn from archaeologists to the betterment of their studies and vice versa. In general, a true synthesis of the disciplines is a goal still to be achieved. NOTE 'Drs. David W. Steadman and Storrs L. Olson suggested critical topics for discussion in this paper on the basis of their extensive experience studying biological materials from archaeological sites. David Steadman and Philip Unitt kindly reviewed the manuscript, making various suggestions for its improvement. Authors who have kept me supplied with reprints will find their papers cited most frequently here. LITERATURE CITED BEDWELL, STEPHEN F. 1971. New BYE, ROBERT A., JR. 1986. Voucher evidence for the presence of turkey in specimens in ethnobiological studies the early Postglacial Period of the and publications. J. Ethnobiol. 6:1-8. Northern Basin. Great Basin CARTER, GEORGE F. 1971. Pre-Colum- Naturalist 31:48-49. bian chickens in America. Pp. 178- Vol. 6, No. 1 LITERATURE CITED (continued) 218 in Man Across the Sea. (C. L. Riley, J. C. Kelley, C. W. Pennington, and R. Rands, eds.}. Univ. Texas Press, Austin DIPESO, CHARLES C. 1976. Medio Period commerce. Pp. 141-192 in Casas Grandes: a fallen trading center of the Gran Chichimeca, Vol. 8. (C. C. DiPeso, J. B. Rinaldo, and G. J. Fen- ner, eds.). Amerind Foundation, Inc., Dragoon, Arizona. EULER, ROBERT C. [Ed.]. 1984. The archaeology, geology and paleo- biology of Stanton’s Cave, Grand Canyon National Park, Arizona. Grand Canyon Natural History Assoc. Monogr. No. 6. FERG, ALAN and AMADEO M. REA. 1983. Prehistoric bird bone from the Big Ditch Site, Arizona. J. Ethnobiol. 3:99-108. FEWKES, J. W. 1912. Casa Grande, Arizona. 28th Ann. Rept. Bureau Amer. Ethnol.:25-179. GILMORE, RAYMOND M. 1946. To facilitate cooperation in the identi- fication of m bones from archaeological sites. Amer. Antiquity 12:49-50. 1949. The identifi- cation and value of mes from archaeologic excavations. J. Mammalogy 30:163-169. 1950. Fauna and Ethnozoology of South America. Pp. 345-464 in Handbook of South American Indians, Vol. 6. Bureau of American Ethnology Bull. 143. GLADWIN, HAROLD S., EMIL W. HAURY, E. B. SAYLES, and NORA GLADWIN. 1937. Excavations at Snaketown: material culture. Univ. Arizona Press, Tucson. GRAYSON, DONALD K. 1977. A review of the evidence for early Holocene turkeys in the northern Great Basin. Amer. Antiq. 42:110-114. nisi seatisnntirineterilenie Ls Oe CENCE Sheet of the uses of archaeological verte- brates in paleoenvironmental recon- struction. J. Ethnobiol. 1:28-38. GREENLEAF, J. CAMERON. 1975. The fortified Hill site near Gila Bend, Arizona. Kiva 40:213-281. HAEMIG, PAUL D. 1978. Aztec Emperor Auitzotl and the Great-tailed Gracle. Biotropica 10:11-17. . 1979. Secret of the Painted Jay. Biotropica 11:81-87. HAMBLIN, NANCY L. 1984. Animal use by the Cozumel Maya. Univ. Arizona Press, Tucson. and AMADEO M. REA. 1979. La avifauna arqueoldgica de Cozumel. boletin de la Escuela de Ciéncias Anthropologicas de la Universidad de Yucatan 37:21-49. HARDY, JOHN WILLIAM. 1983. Voices of the New World jays, crows, and their allies. Family Corvidae. ARA Records, Gainesville, Florida. HARGRAVE, LYNDON L. 1970. Mexi- can Macaws, comparative osteology and survey of remains from the Southwest. Anthropol. Pap. Univ. Arizona No. 20. 1972. Comparative osteology of the chicken an American grouse. Prescott College Studies in Biology No. 1. _ sand STEVEN D. EMSLIE. 1979. Osteological identification of Sandhill Crane vs. turkey. Amer. Antiquity 44:295-299, JOHNSON, PAUL C. 1981. Mammalian remains from Las Colinas. Pp. 269- 289 in The 1968 excavations at Mound 8 Las Colinas Ruins Group, Phoenix, Arizona. (L. C. Hammack and A. P. Sullivan, eds.). Arizona State Museum Archaeological Series No. 154. MARSHALL, LARRY G. 1984. Who killed Cock Robin? an investigation of the extinction controversy. PP. 785-806 in Quaternary extinctions: a prehistoric revolution. (P. S. Martin and R. G. Klein, eds.}. Univ. Arizona Press, Tucson. Summer 1986 MUKUSICK, CHARMION R. JOURNAL OF ETHNOBIOLOGY 17 LITERATURE CITED (continued) 1976. The Casas Grandes avian report. Pp. 273-284 in Casas Grandes: a fallen trading center of the Gran Chichimeca, Vol. 8. (C.C. DiPeso, J. B. Rinaldo, and G. J. Fenner, eds.). Amerind Foundation, Inc., Dragoon, Arizona. 1980. Three groups of turkeys from Southwestern archae- ological sites. Pp. 225-235 in Papers in avian paleontology honoring Hildegard Howard. (K. E. Campbell, Jr., ed.). Contrib. Sci. Natur. Hist. Mus. Los Angeles County, No. 330. MOREJOHN, G. VICTOR. 1976. Evi- dence of the survival to Recent Times of the extinct flightless duck, Chen- dytes lawi Miller. Pp. 207-211 in Col- lected papers in avian paleontology honoring the 90th birthday of Alex- ander Wetmore. (S. L. Olson, ed.). Smiths. Contrib. to Paleobiol. No. 27. OLSON, STORRS L. 1981. The museum traditions in thology—a response to Ricklefs. Auk 98: 193- 195. . 1982. Biological archae- ology in the West Indies. Florida Anthropologist 35:162-168. d HELEN F. JAMES. 1982a. Fossil birds from the Hawaiian Island; evidence for wholesale extinc- tion by man before Western contact. Science 217:633-635. 1982b. Prodromus of the fossil avifauna of the Hawaiian Islands. Smithson. Contrib. Zool. No. 365. . 1984. The role of Poly- nesians in the extinction of the avi- fauna of the Hawaiian Islands. Pp. 768-780 in Quaternary extinctions: a prehistoric revolution. (P. S. Martin R. G. Klein, eds.). Univ. Arizona Press, Tucson. EE, PAUL W. 1958. Remains of rare and extinct birds from Illinois Indian sites. Auk 75:169-175. —___________. 1960. Additional Fisher records from Illinois. Trans. II]. State Acad. Sci. 53:48-49 1961. A_ prehistoric record of the Trumpeter Swan from central Pennsylvania. Wilson Bull. 73:212-213. . 1967. Additional note- worthy records of birds from archae- ological sites. Wilson Bull. 79:55-162. . 1971. Thirteen-lined Ground Squirrel, Prairie Chickens and other vertebrates from an archae- ological site in northeastern Arkan- sas. Amer. Midland Naturalist 86:227-229. 1981 [=1982]. A pre- historic archaeological record of the Pomarine Jaeger, Stercorarius pomarinus (Temminck), from Cen- tral Illinois. Trans. Ill. State Acad. Sci. 74:29-32. and GREGORY PERINO. 1970. A prehistoric archaeological record of the Roseate Spponbill in Il- linois. Trans. Ill. State Acad. Sci. 63:254-258. PHILLIPS, ALLAN R. 1968. The insta- bility of the distribution of land birds in the Southwest. Pp. 129-164 in Col- lected papers in honor of Lyndon Lane Hargrave. (A. H. Schroeder, ed.). Pap. Archaeol. Soc. New Mexico, No. i. PHILLIPS, ALLAN, JOE MARSHALL REA and GALE MONSON. 1964. The birds cs Arizona. Univ. Arizona Press, Tucso AMADEO M. 1973. The Scaled (Callipella squamata) of the South- west: systematic and historical con- siderations. Condor 75:322-329. 1980. Late Pleistocene and Holocene turkeys in the South- west. Pp. 209-224 in Papers in avian paleontology honoring Hildegard Howard. (K. E. Campbell, Jr., ed.). Contrib. Sci. Natur. Hist. Mus. Los Angeles eae No. 330. 1981. Avian Remains Vol. 6, No. 1 LITERATURE CITED (continued) from Las Colinas. Pp. 297-302 in The 1968 excavations at Mound 8 Las Colinas Ruins Groups, Phoenix, Arizona. (L. C. Hammack and A. P. Sullivan, eds.). Arizona State Museum Archaeological Series No. 154. . 1981. Resource utiliza- tion and food taboos of Sonoran Desert Peoples. J. Ethnobiol. 1:69-83. 1983. Once a river: bird life and habitat changes on the Middle Gila. Univ. Arizona Press, Tucson. STEADMAN, DAVID W. 1980. A review of the osteology and paleontology of turkeys (Aves: Meleagridinae). Pp. 131-207 in Papers in avian paleon- tology honoring Hildegard Howard. (K. E. Campbell, Jr., ed.). Contrib. Sci. Natur. Hist. Mus. Los Angeles County No. 330. 1986. Holocene verte- brate fossils from Isla Floreana, Galapagos. Smithson. Contrib. Zool. No. 413. and STORRS L. OLSON. 1985. Bird remains from an archae- ological site on Henderson Island, South Pacific: man-caused extinc- tions on an “uninhabited” island. Proc. Natl. Acad. Sci. USA 82:6191- 6195. WASLEY, WILLIAM W. and ALFRED E. JOHNSON. 1965. Salvage archae- ology in Painted Rocks Reservoir Western Arizona. Anthro. Pap. Univ. . No. 9. WOOD, D. S. and G. D. SCHNELL. 1986. Revised world inventory of avian skeletal specimens, 1986. Amer. Omithologists’ Union and Oklahoma Biol. Survey, Norman, Oklahoma. ZUSI, RICHARD L., D. SCOTT WOOD, and MARION ANNE JENKINSON. 1981. Remarks on a world-wide inventory of avian anatomical specimens. Auk 99:740-757. J. Ethnobiol. 6(1):19-25 Summer 1986 FOOD SAMPLE COLLECTION FOR NUTRIENT ANALYSES IN ETHNOBIOLOGICAL STUDIES HARRIET V. KUHNLEIN School of Dietetics and Human Nutrition Macdonald College, McGill University 21,111 Lakeshore Rd., Ste. Anne de Bellevue Quebec, H9X 1C0O Canada ABSTRACT.—Many ethnonutritional studies have suffered from lack of precision and poorly developed skills in data collection, sample collection, labelling of containers, and advanced planning. This paper addresses some of the practical concerns facing ethno- nutritional field workers and suggests ways and means to enhance accurate transmittal of information to laboratory workers, many of whom often have minimal or no field exposure. PURPOSES OF NUTRIENT ANALYSES IN ETHNOBIOLOGICAL STUDIES If all of the foods in local indigenous food systems were represented in tables of food composition originating from national laboratories and data banks, it would be unneces- sary to conduct basic chemical work to gain insight and understanding into the contributions of such foods to the nutrition and well being of those consuming them; indeed, this paper would then be superfluous. This is not the case, and these studies, including the field work which must precede the chemical analyses, are needed if we are to gain a full understanding of the diets and nutritional status of contemporary, historical, and prehistorical people. Nutrient analyses are needed to complete food composition data bases which are then used with quantitative dietary data to assess individual or group nutritional health status. den Hartog and van Staveren (1979) have described the vari thods of assessing dietary intake and, except to emphasize that both careful planning and statistical treat- ment are essential in overcoming inherent difficulties in quantitative food intake assess- ment, these procedures will not be dealt with herein. In addition to its use in defining quantitative nutrient intake of people, food com- Position dxta is useful for its own sake when an ingested item has not been previously studied and it is known to be consumed frequently and/or in reasonable quantity by a 8roup of people. Nutrient analytical work on foods helps to identify particularly superior or inferior potential new foods for horticultural or genetic studies. It contributes to the body of knowledge of how indigenous people met their nutrient requirements, and helps to fill the gap in our understanding of the intrinsic worth of indigenous diets and the complexity of effects modern dietary adaptation has on the health of native people. The importance of expanding our collective knowledge of the use and chemical properties of indigenous foods has been noted many times (Behar 1976; Kuhnlein 1984; Tumer 1981). This knowledge is particularly valuable for native or cultural groups who wish to Promote cultural avenues for health improvement and self-care. Providing information for wildemess survival training programs, and stimulating markets for new ethnic foods native people may wish to promote as cottage industries (Kuhnlein 1985}, further justify recording nutrient data in indigenous foods. It is perhaps worth mentioning that the 20 KUHNLEIN Vol. 6, No. 1 techniques and methods presented herein can also be interpreted for work on the toxicological and medicinal properties of ingestants. SAMPLE IDENTIFICATION AND DEFINITION The hallmark of superior sample collection in ethnographic settings is the elici- tation of accurate knowledge about the identification and preparation of the food samples by the local people. This expertise needed for good data collection is often not appreciated and is perhaps the most underrated skill in the entire chain of research events—a pro- cess which begins by identifying the groups and foods of interest and ends with collaboration with laboratory scientists and publication. Native or ethnographic “consultants” or “informants” play a central role in generating accurate data—they must be well-informed of the variety of practices in the community and must be capable of trust to relate it accurately. Many native people are highly motivated to share their wealth of knowledge for the ultimate purposes of recording it in the written media for the posterity of their groups (Van Asdall 1985). They exchange the trust for accurate knowledge by the data collector with their own trust that the information they give will be recorded honestly and presented in the best interest of their people. Tales of mismotivated consultants or informants are legion among ethnographic researchers. Some will provide any type of information or samples they perceive as desirable for money, as long as the funds are available. Others are reluctant to share what they feel is private family or group information, and will deliberately side-track an investigation. Yet others will insist their particular method of preparation is superior, and may ignore the techniques of others in the community. Because of these potential deterrants to good data collection it is wise to gather information and samples from several reliable sources in the community. The involvement and training of native people to collect accurate data and samples is a key component of any health promotion effort in contemporary indigenous settings. If they know the rationale and procedures used, native people are highly motivated to report results back to the community, and thereby stimulate local interest and the impetus to use the results for health promoting activities. It is clearly understood that the investigator must develop a reputation for returning the results of all data collection and sample analysis to the community and that this is requisite to the research and publication process. As reported elsewhere in this issue (Bye 1986; Rea 1986), the identification of ethnobiological samples with voucher specimens needs to be thorough and as complete as possible. Ideally, taxonomic identifications are made in the field with the help of a local botanist or zoologist. In the absence of this assistance, dried samples (plants) or detailed photographs of an intact entire individual organism will assist identification, and should be accompanied with the local linguistic terms and any other common language names that are used. In the case of food samples, it may be pertinent to the investigation to sample the single Taw species, as well as the final prepared product, since multiple processes and ingredients obviously modify the original nutrient content. Assurance that the sample is representative is necessary. Depending on the product in question, this can be done in a variety of ways: picking berries from several bushes in an area; thoroughly mixing a large quantity of product (for example, a large bowl of ground corn meal) before extracting the sample size needed; if a representative animal tissue is desired, taking ee les that are commonly eaten. In the latter case, the desirability of single muscle or organ tissue sampling must be carefully considered in lieu of mixed amounts of the total edible portion of like tissues. The sampling of ethnobiological food Summer 1986 JOURNAL OF ETHNOBIOLOGY 21 species in the field is as hazardous as getting good ethnographic data from “consultants”, especially if the species are limited in quantity or labor intensive, and the “consultants” would prefer not to part with the amount needed for a thorough chemical analysis. Whenever possible, the sampling strategy should be thorough enough to permit the desired statistical treatment. However, it is easy to allow statistical considerations to escalate the sample number so high that the costs of analyses are prohibitive. Caution is therefore needed in the original design of the project so that questions that are practical and answerable are put forward. In addition to sampling the basic food species, complete information on other techniques or ingredients used by other cooks should be defined, if they are not a part of the original sampling strategy. For example, Hopi piki bread is a finished food product which utilizes only four ingredients: corn meal, a culinary ash, water, and a small amount of cooking fat. One could select the most common preparation from First Mesa, for example, and take 4-5 samples of (1) blue corn meal, fine ground, (2) ash prepared from chamisa, Atriplex canescens, and (3) the finished product. From these (and published tables of water minerals and marketed fat nutrients) a definition of nutrients contributed by the ingredients could be made, with appropriate statistical tests. Descriptive infor- mation to round out the data collection could be made by interviewing 8-10 different piki preparers for: (a) Other types of corn used for piki (white, red, etc.). (b) Other types of ash used (bean plant, juniper] or baking soda. (c) Different recipes or procedures evaluated with measures or balances. (d) Different grinding implements used for the corn which might contribute flavor or mineral additions. (e) Other types of cooking imple- ments: griddle vs. piki stove. (f) Alternate sources of cooking fat: oils, lard, seed oils, animal spine fats, etc. To thoroughly sample each of these variables would result in more than 100 analyses, a prohibitive amount. Even to sample just the finished piki resulting from several of these variations would be prohibitive, and the difficulty to evaluate just one variable ari ses. Although these iderations p ta dous det t to the field investigator sampling ethnobiological food samples, keep in mind that even rudimentory sampling of finished products may give important results that provide the impetus for further investigation. For example, if all piki bread samples made with ash have high levels of calcium, (and they do), and the local population consuming it did not traditionally use milk (they did not) this is a significant find in ethnonutritional research. Further to these ierations on sampling in food preparation, care should be taken to note precise geographic locations, the individuals involved in sampling and any unusual soil or climate conditions. Careful notes made on the day of collection and ordering of samples is indispensable, especially when multiple samples are taken. Containers are best labeled with permanent ink on the container itself (not the lid, if it is removable); if stick-on labels are used for samples to be frozen, insure that the labels adhere at the desired temperature. SAMPLE TREATMENT FOR SPECIFIC NUTRIENT ANALYSES Harris and Karmis (1975) presented a guide for understanding general nutrient stability after standard processes of food preparation and preservation (Table 1). Heat treatment during cooking of any form, canning, and in some instances drying, causes destruction of the greatest number of nutrients. Exposure to air or oxygen, as In drying, dry roasting, or fine dicing or pureeing of foods also presents several nutrients for destruc- tion. Maintaining food products at neutral pH and away from direct light is protective for several nutrients, especially the vitamins. Reducing the pH below 7, asin pickling, fermenting or acidification with vinegar-based dressings is destructive to the Vitamin A 22 KUHNLEIN Vol. 6, No. 1 and carotene complex, folate and pantothenic acid. Raising the pH above 7, as occurs with adding baking soda or ash products to grain foods is destructive to thiamin, riboflavin, ascorbic acid, vitamin D and essential fatty acids. A quick review of this summary table TABLE 1.—Stability of Nutrients. Effect of pH Neutral Acid Alkaline Air or Max. Nutrient pH7 pH7 Oxygen Light Heat Cooking Losses Vitamins % Vitamin A S U S U U U 40 Ascorbic acid (C} U S U U U U 100 Carotene (pro-A} S U S U U U 30 Cobalamin (B-12} S S S U U S 10 Vitamin D S s U U Mi U 40 Folic acid U U S U U U 100 Niacin (PP) S S S S S S 79 Pantothenic acid S U U S S U 50 Pyridoxine (B-6) S S S S U U 40 Riboflavin (B-2) S S U S U U 75 Thiamin (B-1) U S U U S U 80 Tocopherol (E) S S S U U U 55 Essential amino acids Isoleucine S S S S S S 10 Leucine S S S S — S 10 Lysine S S S S S U 40 Methionine S S S S . S 10 Phenylalanine S S S S S S 5 Threonine S U U S S U 20 Tryptophan S U S S U S 15 ine S S S S S S 10 Essential fatty acids S S U U U S 10 Minerals S S S S S S 3 ee S = stable (no important destruction). U = unstable (significant destruction). Adapted from: Harris and Karmis, 1975. Summer 1986 JOURNAL OF ETHNOBIOLOGY 23 will guide planning of which nutrients are most effectively determined in prepared or preserved food products. It can also guide sampling and storage procedures. important procedure in food sampling, often overlooked in field collections, is thorough cleaning of the foods before packaging for shipment to the laboratory. Washing for removal of soil particles is especially important if the food is destined for mineral analyses. Microbiological assay for vitamins are sometimes upset by heavy microbe contamination of foods, a problem minimized through cleaning. If possible, it is best to sample food items only for the edible portion (EP], which means removing of husks, shells, skins, etc. This is most important if samples are to be fresh frozen and then thawed in the laboratory, since nutrients in the non-edible portions may migrate to the EP during the freeze-thaw process, and give an erroneous picture of what is normally consumed in the home setting. Although washing of the food product is important, the drying of the product before storage is equally critical, since excess water clinging to frozen samples will be inseparable from the nutrient containing EP after thawing and will therefore dilute the nutrient content, giving unrealistically low values. This consideration is especially salient for foods with initial low values of the nutrients in question. Sampling food products in field settings and storage under field conditions is to be considered carefully before deciding what analyses are practical. If mineral analyses are the objective, simple drying procedures can be done in the field, using open air/sun drying or gentle oven temperatures (not more than 100°C}. Samples need to be protected from contamination by other minerals, as given by metal implements or storage containers, but otherwise packaging is simply done in clean (unused) plastic bags. Drying also prevents shipping excess weight in the form of water. Freezing at -10°C in airtight plastic containers is needed for storage of samples destined for analysis of most vitamins. For most food, samples should be taken from the EP of the fresh state or from a freshly prepared product ready for consumption. Vitamin E is the only nutrient for which freezing is known to cause significant losses. Otherwise, all nutrients including minerals can be assayed from freshly frozen food products. Bye (1986:1-8; this issue] reports on recommended, and, in some cases, standardized procedures for connecting field notes, voucher, corrororative, and associated specimens of plants and animals. Some or all of these may ultimately be deposited in dif- ferent institutions. Although record keeping in field ethnonutritional studies have not yet been formalized, field logs can serve many important functions, e.g., they help to orient the investigator when several shipments are sent from the field to the laboratory and, along with logs of sample containers and storage conditions both during transit and in the laboratory, are useful in sorting out problems of anomalous nutrient values in similar food products. Information of this sort can best be transmitted to the laboratory staff in a detailed “transfer letter’ enclosed in the storage container explaining what samples are arriving from the field and to what conditions they have been exposed before (and during) shipment. This document might also specify dates of sampling and ship- ment, cleaning process, treatments, as well as actual names of the foods with their code numbers so that information and interpretation can be related, if and when necessary, to field logs. Within the laboratory, records should be kept of the length of storage time, storage conditions and frequency of freeze-thaw of samples. Canned and frozen foods are known to lose value for several nutrients with duration of storage and fluctuations in storage temperature (Erdman and Erdman 1982). . e techniques of analysis of the food samples are best predetermined with the laboratory staff prior to field collections so as to avoid errors and unnecessary mis- handling of samples and potential nutrient destruction in the process. At the same time, the actual size of sample needed for the various analyses should be determined to avoid 24 KUHNLEIN Vol. 6, No. 1 undersampling and the consequent need to delete duplicate measurements which insure validity and reliability of the assays. For some nutrients, pretreatment in the field will enhance nutrient stability (for example, liquid food products destined for folate analysis could have a known quantity of crystalline ascorbate added to preserve the folate content before packaging). Methods outlined in the JOAC (Horwitz 1980) orin Southgate (1974) give general guidance to consideration of sample sizes for particular nutrient analyses. Greenfield and Southgate (1985) present a thorough review of record keeping and laboratory procedures needed for quality food data. . A note of caution is needed for the interpretation and comparison of nutrient results when methods of sampling and analyses are not consistent. The compilation of nutrient composition tables from multiple data sets are especially problematical when consistency has not been maintained. It takes an experienced chemist to decipher variance in data when analytical methods differ. NOTE lp the nutrient content of piki bread, see Calloway, Giauque, Costa (1974), Kuhnlein, rane and Harland (1979), Kuhnlein and Calloway (1978}, and Kuhnlein and Callo- 9). way (197 LITERATURE CITED BEHAR, M. 1976. European Diets vs. Traditional Foods. Food Policy, Nov. 432-435. BYE, ROBERT A., JR. 1986. Voucher specimens in ethnobiological studies and publications. J. Ethnobiol. 6:1-8. CALLOWAY, D. H., R. D. GIAUQUE and F. P. COSTA. 1974. The Superior Mineral Content of Some Indian Foods in Comparison to Federally Donated Counterpart Commodities. Ecol. of Food and Nutr. 3:203-218. DEN HARTOG, A. P. and W. A. VAN STAVEREN. 1979. Field Guide on Food Habits and Food Consumption. Wageningen ICFSN Nutrition Papers. ERDMAN, J. W., JR. and E. A. ERDMAN. 1982. Effect of Home Preparation Practices on Nutritive Value of Food in Handbook of Nutritive Value of Processed Food. (M. Rechcigl, ed.). CRC Press, Inc. GREENFIELD, H. and D. A. SOUTH- GATE. 1985. A Pragmatic Approach to the Production of Good Quality Food Composition Data. ASEAN Food J. 1:47-54. HARRIS, R. S. and E. KARMAS. 1975. Nutritional Evaluation of Food Pro- ene oe KUHNLEIN, H. V. 1984. Nutritional value of Traditional Food Practices in Food Science and Human Welfare. (J. V. McLoughlin and B. M. McKen- na, eds.|. Boole Press, Dublin. . 1985. Wild Foods for Moder Diets in Nutrition Update. - Weininger and G. Briggs, eds.). Vol. 2. Wiley and Sons, Inc. and D. H. CALLOWAY. 1979. Adventitious Mineral a in Hopi Indian Diets. J. Food $ 44:282-285. _ 1978. Contemporary Hopi Food Intake Patterns. Ecol. of Food and Nutr. 6:159-173. and B. HARLAND. 1979: Composition of Traditional Hopi Foods. J. Am. Dietet. Assoc. 75:37-41. REA, AMADEO M. 1986. Vertification and reverification: Problems i archaeofaunal studies. J. Ethnobiol. 6:9-18. a SOUTHGATE, D. A. 1974. Guidelines Summer 1986 JOURNAL OF ETHNOBIOLOGY Vi LITERATURE CITED (continued] for the Preparation of Tables of Food Native Peoples. Can. J. Botany 59: Composition. S. Karger, Basel. 2331-2357. TURNER, N. J. 1981. A Gift for the VAN ASDALL, W. 1985. Sketches in the Taking: the Untapped Potential of Sand. J. Ethnobiol. 5(1):i-ii. Some Food Plants of North American 26 BOOK REVIEW Vol. 6, No. 1 A World Directory of Ethnobotanists. S. K. Jain, P. Minnis and N. C. Shaw. Society of Ethnobotanists, Lucknow, India. 1986. Indian Rs. 15; U.S. $2.00; L 1.50. The Indian Society of Ethnobotanists, established in 1980, has just published a most useful directory, one that will greatly strengthen the links that bind the many specialists in interdisciplinary fields tangential to economic botany and the sundry aspects of ethnobotany. This valuable contribution, listing the names, addresses and usually the specific research interests of 489 ethnobotanists, can be purchased by writing to Dr. Ved Prakash Tripathi, B-20 CRIR Colony, Nirala Nagar, Lucknow, India. Richard Evans Schultes Professor Emeritus Botanical Museum of Harvard University Cambridge, Massachusetts 02138 ]. Ethnobiol. 6(1):27-43 Summer 1986 GUIDEPOSTS IN ETHNOBOTANY VORSILA L. BOHRER Southwest Ethnobotanical Enterprises 220 New Mexico Drive Portales, NM 88130 ABSTRACT.—A review of the history of a natural system of maize classification is provided. Detours in the archaeological analysis of maize abound. Tests suggest the extrapolation of cob row number from kernel angles lacks predictive value. Flotation analysis is forging ahead. Current trends in regional studies might be better integrated with anthropological model construction or destruction. Charcoal identification merits improvement through use of metallurgical microscopes coupled with more rigorous use of plant anatomy. Analysis of leaf epidermal fragments in coprolites broadens the range of information recovered. The biotic factor, including the anthropogenic one, needs to be considered as seriously as climate and soil in the formation of prehistoric and modern plant assemblages. Ethnobotanical interpretation of prehistoric plant remains benefits by insights into the kaleidescopic patterns formed by modern and historic interrelationships between plants and man. It is never too late for contemporary studies. Museums rather than colleges or universities continue to provide workspace for ethnobotanists. Sugges- t j j d | : a ve e ‘. | ee E | ] . bo ] Serine niviciar Gia 4c oe ae oks and PA = INTRODUCTION GUIDEPOSTS IN ETHNOBOTANY was published in 1981; the first general session on ethnobotany at the annual meeting of the Society of American Archaeology was in 1983, and in May, 1985, after settling into my seat on Southwest Airlines Flight 192, I was pleased to notice that the cover story of their flight magazine featured the contemporary ethnobotanical work of Dr. Richard Felger in northwestern Mexico and southern Arizona. That same August, Forbes magazine featured an article on the applications of pollen analysis (Teitelman 1985). Apparently ethnobotany has emerged from the low visibility of a pioneer discipline into a new phase of prominence. Because your good editor suggested a historical perspective might be valuable, I wrote the first two sections on Maize and Flotation chronologically, but my initial resolve to follow this approach lessened by the time I reached the next section on Charcoal and Fecal Analysis and evaporated by the time I wrote Changes in Plant Life. Do not be misled by these initial digressions into history, for the pages that follow have taken another more personal orientation. I have participated in ethnobotany from the time when it comprised a handful of people, and witnessed with mixed emotions a series of changes in the field. The following pages typify my rejoicing over examples of fine work and brooding about areas that seem deficient. I regret that I can not know and reference all the best efforts in ethnobotany. On the other hand, I find it judicious to avoid citing what is less than satisfactory. The recent increased prominence of ethnobotany carries an increased responsibility toward producing a quality product; some sort of guide seems especially necessary right now. Lest you misunderstand, be assured that I admire the achievements of those who have pioneered in learning about maize, flotation, charcoal, cal analysis, environmental reconstruction and ethnobotanical interpretation. Nor are Ethnobotany is flying high in the 1980s. The first issue of the Journal of Ethnobiology s 28 BOHRER Vol. 6, No. 1 you to assume that I am setting myself up as an oracle for ethnobotany. I expect some disagreement with my judgment. A discourse such as this will perhaps stimulate a dialogue for improvement within the profession so that we can avoid, in spite of the meteoric growth of our discipline, a hollow center leaving us too vulnerable to with- stand criticism from outside. MAIZE Southwestern Pueblo Indian maize was among the first to be carefully sampled and classified according to a natural system (Anderson and Cutler 1942). A natural system relies on the ability to establish genetic relationships and trace origins. In the course of Anderson’s self-directed study of maize he learned to depict variability important in the analysis of introgressive hybridization by means of a pictorialized scatter diagram (Anderson 1946:175). In the first version developed for southwestern maize (Carter and Anderson 1945) each point on the scatter diagram registered index numbers that summ- ed data rather than recorded the primary measurements on each plant. In Anderson's subsequent publication (1946) the system was perfected by employing row number on the x- and kernel width on the y axis. The small circle or dot representing the inter- section of values was proportionately elongated to illustrate the amount of kernel pointing, and shaded to correspond to the degree of kernel denting. Both row number and kernel width were used because they were end products of many genes acting in concert. For example high row number, narrow kernels and pointed kernels tended to be inherited together. By choosing single characteristics that were the expression of multiple genes, Anderson could sit in a farm yard and measure the salient features of a cob by making a single mark on the appropriate intersection of his scatter diagram which represented all three observations simultaneously. He could average the data for all maize ears in a village. Later, one could derive average values for all villages belonging to a tribe, and for all tribes belonging to a linguistic group. With such economy of effort not only Pueblo Indian maize but all maize of Mexico could be classified in the hope of tracing its northward migration. In the 1930s long before Anderson developed an interest in maize, Carl Sauer, Professor of Geography, University of California at Berkeley, became interested in documenting the varieties of corn grown by Indian tribes in remote locations in north- western Mexico. Isabel Kelly, one of Sauer’s students, provided Anderson with critical collections of chapalote corn from Culiacan and Sinaloa and of the related maize reventador (Anderson 1944) from Jalisco and Zacatecas. Both races proved an essential modern link in understanding prehistoric maize in the southwest. Anderson combined his keen perceptions of morphology with his special interest in multiple gene complexes to provide informed speculation that maize reventador represented an old and primitive race (Anderson 1944:309). Later, the related chapalote race was thought more important (Wellhausen et al. 1952:57). The magnitude of Anderson’s contribution to the under- standing of maize in Mexico is apparent when we realize that ten of the races he delineated were incorporated into a total of 25 races recognized for Mexico (Wellhausen 1952: 13). When the descriptions of Mexican maize races were published in 1952 their cot- relation with certain patterns of distribution of chromosome knobs were known. For example, all those called “ancient indigenous races” had low chromosome knob numbers. Since then an outstanding team of maize cytologists undertook a study of nearly twenty years in mapping the distribution of chromosome knobs of most maize races in the Western Hemisphere as well as the majority of races of teosinte, the closest relative of maize (McKlintock et al. 1981). They were able to demonstrate that some races con sidered relatively ancient from the Pacific coast are related to each other with the 7s large knob: nal-tel, zapalote chico, chapalote, reventador and harinoso de ocho (McKlin- Summer 1986 JOURNAL OF ETHNOBIOLOGY 29 tock et al. 1981:40). Through many pages of carefully documented analysis the team was able to deduce two routes of entry of maize into the Southwestern United States— one along the Pacific Coast and a second that apparently ran from central Mexico northward to the Rio Grande. Although Barbara McClintock won the 1983 Nobel Prize in medicine for discovering transposable elements based on the mode of inheritance of color in corn kernels and leaves, ethnobotanists will remember her leadership in indepen- dent cytological corroboration of the integrity of the races that Anderson first deduced from maize morphology. The finer relationships between races revealed in the study she headed should sustain an interest in racial classification for some time. Given the statistical advances applied to maize variability since Anderson’s initial work (eg., Goodman and Paterniani 1969, Bird and Goodman 1977), and given that cupule width represents the nearest measure of kernel width when the cob alone is considered, it is a tribute to Anderson’s keen perception for the authors of a recent technical report on maize from the Peruvian coast to remark that of the seven characters used, row number and cupule width were the two most consistently useful traits in distinguishing maize types (Bird and Bird 1980:330). Studies in the 1980s have corroborated the early methods used to identify races, singled out the most ancient races of maize, and given the scientific world a fresh explanation of the seemingly abrupt transition from teosinte to maize as a catastrophic sexual transmutation (Iltis 1983; Gould 1984). The theory has not gone unchallenged (see Galinat 1985) but the persistent storms over maize evolu- tion that have raged for decades have lessened in intensity. In the Southwestern United States the foot and leg work necessary to provide information on the prehistoric distribution of maize types was done primarily by Hugh C. Cutler in cooperation with field archaeologists. In Cutler's many appendices to archaeological site reports he provided data on cob row number and cupule width. What his colleague Anderson learned of modern races in Mexico and what Cutler himself leamed about southwestern corn could by synthesized to interpret prehistory. Two appendices are particularly rich in insights to the linkage of history and prehistory (Cutler 1964 and Cutler 1966}. In the earlier report Cutler relates the maize from Carter Ranch Pueblo to the Pima-Papago race that he and Anderson first defined, to more recent racial definitions, and to his observations on Hopi maize. In the later report he provided a greatly expanded overview of southwestern maize as part of his Glen Canyon, Arizona research (Cutler 1966:12-15}. He presents views on the progression of races of maize in the southwest and the role of Mexican races like onavefio and harinoso de ocho in prehistory. Maize has provided us with an incredible number of detours in our thinking before we have found the main road. Differences of opinion about which is actually the correct toad disappear only in retrospect. Cobs, kemels, pollen, phytoliths and radioactive isotopes have all been the focus of considerable debate, and not without reason. While cobs are the most efficient way to gather data about prehistoric com, the process is far from perfect. Bumed cobs seldom represent more than a random cross-section of a cob devoid of kernels. Although adjustments for cob shrinkage can be made, the lack of attached kemels deprives the investigator of ancillary criteria to assess the racial affiliations of the population formerly present. In addition, corn grown under stress may have reduced row number (Emerson and Smith 1950 in Cutler 1966:11). Recently Mackey (1985) has advocated the use of maize to demonstrate stress in the Gallina Llaves Valley, New Mexico archaeological sites. A necessary prerequisite of such a study is establishing the existence of a uniform race through time. Real problems exist in establishing the former existence of any single race (Benz 1985). ; A modem statistical treatment of the races of maize in Mexico and the southwest heed development to serve as a template against which prehistoric populations can be itted. The loss of a host of criteria in archaeological maize makes it more difficult to accurately fit the narrow racial categories formed from intact modern maize (Wellhausen 30 BOHRER Vol. 6, No. 1 et al. 1952) rather than to the more broadly defined races first delimited (Anderson and Cutler 1942, Anderson 1946). We still do not know to what degree changes in southwestern corn result from cultural preferences or selection and which are products of diffusion of new races of maize. Ford (1981:13) has argued that from the genetic variability present in southwestern com by 300 B.C. one could derive the morphological variability seen in subsequent strains through natural and cultural selection. Ford’s ideas in conjunction with the evidence for prehistoric diffusion of harinoso de ocho to the mid-west (Galinat and Gunnerson 1963) might well be considered once the boundaries of our racial taxonomy are in better order and the accuracy of our dating verified with direct radiocarbon dating. To further complicate matters the process of extrapolating row number from prehistoric maize kernels (Cutler 1956) appears to have serious problems, despite seemingly impeccable theoretical grounds. The sides of the kernel on a fully developed ear should be compressed by the limitations of the space available on the nearly circular cob (360°). The circle could be divided into 8, 10, 12 or 14 parts corresponding to the number of rows on the cob. A cob of 8 rows should have kernel side angles of 45°, 10 rows 36°, 12 rows 30°, etc. If these assumptions were correct and one had only kernels to measure, the angles of the kernel sides could give an estimate of the row number. The accuracy of the kernel angle method of determining row number was tested recently by Pearsall (1980) when a series of 25 kernels from cobs of known row number were measured. The tendency of kemels from 8 and 10 row cobs to be confused and the inaccuracies in the assessment of cobs of 12 and 14 rows were noted. In a 1984 workshop on ethnobotany in Tucson I directed a similar test on the predictive value of maize kernel angles for row number. Two modern kernel covered ears were selected, one of 12 rows called “A” and one of 14 rows called “B”. Each was checked to verify that the ear had the same number of rows at the base and the tip, and the kemels were then removed from the cob. Without knowing the original row number, each student took 30 grains, measured the angles formed by the kernel sides, then classified the kernels as to 8, 10, 12, or 14 rows. Our results (Table 1) show little relationship to actual row number. A class member more experienced in measuring maize kernel angles who used a metal template instead of ruled angles produced similar results. Our obvious conclusion was since we could not estimate the true row number when we could verify the accuracy of our work, there was no point in applying the method to archaeological kernels which were apt to have suffered additional modification of the kemel angle through carbonization. __ Current techniques on how to recognize maize pollen in the United States now see™ in need of revision. North of Mexico palynologists rely upon the large diameter of maize pollen as a trait sufficiently unique to separate it from other grass pollen. In contrast palynologists in Mexico use features other than size to distinguish maize from the closely related teosinte. In the late 1970s when little barley grass (Hordeum pusillum) caryops*s TABLE 1.—Row number determined from kernel angles on two ears of maize. F Row No. Total Ear 8 0 12 14 kernels A 36 76 54 6 172 B 13 72 69 11 165 weal vas a *A was 12 rowed and B 14 rowed. Summer 1986 JOURNAL OF ETHNOBIOLOGY 3] were identified in the Southwest (Gasser 1981), palynologists showed legitimate curiosity as to the diameter of this New World cereal pollen. Early literature (Jones and Newell 1948:141) and independent measurements by southwestern palynologists confirmed that the diameter of little barley grass resembled most grasses. Recently, however, while doing background research on native barley I noticed that Hordeum pusillum from Granite Reef Dam, Arizona, produces larger pollen (Covas 1949:14). In addition, H. arizonicum bears pollen grains with modal pollen diameters of 60-68 micrometers (Covas 1949:17), which are very similar to chapalote maize pollen (Irwin and Barghoorn 1965}. If the hexaploid nature of H. arizonicum (Rajhathy et al. 1964:196] gives it certain competitive advantages in growing in modern Papago fields as Nabhan has observed (1983:200) it may have done so in the past. Arizona barley grass has been reported in Maricopa, Pinal and Pima countries (Gould 1951:108) in the Hohokam culture area. We have chanced upon one species that locally negates the assumption that the size of maize pollen produces a unique signature that differentiates it from other grasses. Perhaps systematic research into the size of pollen of all polyploid grasses should be undertaken. The acquisition of the perception that there is no unique pollen size signature for maize except where local plant geography allows, lets me view with concern another struggle now underway to revise the unique morphological attributes by which maize phytoliths can be recognized (reviewed by Rovner 1983:249-251). As much as we might desire to establish a universal “truth” about maize phytoliths, it might be more instructive to concede that truth is relative. Like maize pollen identification, the validity of maize phytolith recognition may be relative to the amount of ambiguity provided by phytoliths of other plant species growing in the area. Schoenwetter (1974:298-301) has supplied some thought provoking remarks on how one grapples with uncertainties in the maize pollen record in the heart of Mexico. His discussion might as easily apply to related problems with phytoliths. Although radiocarbon dating began around 1949, it was not until 1967 that the specialists realized maize dated younger than predicted (Bender 1981) because it was one of a number of grasses (Walker and Lewis 1979) and herbs which had a photosynthetic path (C-4 instead of C-3) that accumulated radioactive isotopes differently. Since then the knowledge has been used to advantage in the analysis of the 13¢/14¢ ratios in human bone collagen. In animals the isotope values clearly reflect the isotope ratios in their diet, that is, the proportion of C-4 to C-3 plants. Where succulent plants like cholla cactus (Opuntia) are eaten, a crassulacean acid metabolism (CAM) mimics the C-4 isotope Tatios (Troughton et al. 1974). Where maize is likely to be the only C-4 plant consumed, the ratio provides an indication of the importance of maize in the diet, as in Hopewell agriculture (Bender 1981). In the southwest the consumption of C-4 plants such as Amaranthus, Panicum, Portulaca, members of the Capparidaceae (Cleome) and Chenopodiaceae (Chenopodium) contribute to obscuring the vexing question of the former i ance of maize agriculture. Studies of its antiquity will benefit from accelerator radiocarbon dating, recently used to advantage in dating individual kernels. FLOTATION 32 BOHRER Vol. 6, No. 1 demonstration had portent for future ethnobotanical research. Ten years later, in 1964, Dr. Haury creatively applied Cutler’s flotation to the second excavation of Snaketown. There flotation was metamorphosed from a dirt sample scattered on a bucket of water to a fully conceived sampling plan for trash mounds and pits to obtain a chronological series for flotation analysis. Haury developed and directed the method of concentrating the charcoal (Bohrer 1970}. As an ethnobotanist my job began with the receipt of the bags of charcoal and seed. Since the early work on flotation at Snaketown, our understanding of the formation of the carbonized vegetal record and the fraction that is apt to be recovered by flotation has undergone considerable evolution both in North America and abroad. Flotation techniques in vogue in the mid 1970s have been reviewed (Watson 1976) and numerous variants have been reported since then. Research designs compatible with research objectives have been promulgated by Adams and Gasser (1980), Bohrer and Adams (1977), and Toll (1984). Schaaf (1981) has advocated the use of a sample splitter among other techniques, and methods have evolved to check the accuracy of flotation (Wagner 1982). A concommitant development has been the acknowledgement that carbonized seeds differ in their buoyancy and that certain conditions or properties of seeds may cause them to sink (Pendleton 1983). Such problems may eventually lead ethnobotanists who haven't yet done so to the examination of the heavy fraction or to adopt a modified form of water screening, as has Schaaf. The recent publications of the results of two large archaeological projects at Dolores, Colorado and in Chaco Canyon, New Mexico incorporate several innovations helpful in flotation analysis. At Dolores, archaeologists have formulated criteria for field recognition of four site abandonment modes (Breternitz 1984:166) (1) leisurely abandon- ment implying minimal preservation of artifactual botanical material; (2) deliberate abandonment, burning within 5-10 years afterward, providing moderate preservation potential; (3) ritual abandonment; (4) catastrophic abandonment allowing optimal preservation of materials. A number of advantages are apparent from this procedure. The segregation of sites in terms of potential for plant preservation helps set priorities in the analysis of pollen and flotation samples. Furthermore the method provides a distinct aid to interpretation, for the manner in which case 2 and case 4 are treated, are vastly different (Halley 1981; Minnis 1981). Finally, in a project of long duration, the prospect remains of increasing the accuracy of judgments made in the field from the results in the laboratory. At Chaco Canyon, Mollie Toll (1984) undertook flotation from two sites using different sampling designs. With the help of a third site, she has been able to bring into focus underlying patterns obscured by different seeds densities in flotation samples. Her innovations in dealing with unburned seeds may assist those working in other arid areas or historic sites where uncarbonized seed is seemingly still preserved. Two were shallow village sites subject both to erosion and alluvial deposition. A third site (Pueblo Alto) had living surface over a meter deep. Pueblo Alto was sampled like one of the shallow village sites and used to evaluate the possibility that some of the unburned seed might be prehistoric. The pattern of seed recovery from habitation room to habitation room at Pueblo Alto varied little. Weedy Chenopodium, Amaranthus and Portulaca were cat bonized in heating features and uncarbonized on floors in decreasing density with increasing distance from food processing features. The distribution suggests prehistoric primary deposition of unburned seeds of Pueblo Alto. In the shallow sites the conclu- sions drawn by using only burned seeds or bumed plus unburned seeds are similar—a pattern of using immediately available disturbed ground species in company with agricultural products. There are several qualifiers in the above study. The first is that the difference between burned and unburned naturally black seeds can be subtle. Even slight parching Summer 1986 JOURNAL OF ETHNOBIOLOGY 33 of a seed coat may deter degradation. The second is that high standards of sample selection need to be exercised rigorously. Sample locations that were cleaned out prehistorically and filled with alluvium, that were disturbed by rodents, that were ambiguous in terms of the source of cultural versus natural deposition (like post holes] had to be eliminated from consideration (Toll 1981:2). Unless non-cultural factors active in archaeological deposits are recognized and systematically excluded before flotation samples are analyzed, Mollie (1984:249) warns that ‘‘Non-cultural factors may introduce more variability into the archaeological record than the ones that can be related to past behavior’. Cultural and non-cultural factors represents two sides of the same coin—have we always examined each side with equal care? When the ground rules are carefully set forth and the cultural and non-cultural facets of seed deposition vigilantly explored, there seems to be a possibility that shallow, open sites as well as those over a meter deep may preserve a prehistoric seed fraction that appears unburned. Regional analyses of subsistence, particularly prominent in the last five years, allow us to better visualize the utilization of cultivated and domesticated crops, the nature of the fields and the encouraged plant pioneers. Probably no editor had a tougher scramble to keep North American papers current than Richard Ford (1985). The emerging evidence for prehistoric cultivation of agave and barley in the west and the domesticated Chenopodium and Amaranthus in the Eastern United States tremen- dously complicated the publication of his timely volume. Regional summaries by persons with the knack of distinguishing true subsistence variability in site assemblages from differences induced by variation in methods of data collection, analysis or degree of preservation are needed continuously. Ethnobotanists have been passive bystanders while hunter gatherer foraging strategy proliferates along with numerous other models for the preconditions for the spread of agriculture. Often models are not stated in such a way that they can be tested with archaeological or ethnobotanical data. The missionary zeal of the theoretically oriented would benefit by struggling with the uncertainties of organic preservation and related realities. Ethnobotanists in turn, could assist more with theory and with its translation into practical, testable hypotheses. CHARCOAL AND FECAL ANALYSIS The questions asked in 1963 of wood charcoal analysis (Western 1963:151) resemble those considered in site reports today. What kinds of trees and shrubs have been used for firewood, in construction and in the making of objects for daily and ritual use? Does the analysis of wood types reveal anything in regard to: (1) climatic differences; (2) nature of local vegetation (forest, scrub or grassland); (3) the importation of timber for special purposes; (4) the extent to which the qualities of different woods were appreciated? However, if the question in regards to whether or not cultural selection of woods is Practiced fails to receive priority, then all three of the preceding questions prove particu- larly difficult to answer. If for example only ponderosa pine and fir are used as beams in a site which is today surrounded by pinyon and juniper, are we looking at a change in vegetation or the importation of timbers for structural purposes. If hickory charcoal predominates in hearths, can we conclude hickory was the dominant forest tree? To what €xtent is cultural selection active in each case? Noe Only when mild post-excavational curiosity about the charcoal content of a site is teplaced by problem oriented provenience sampling can we discard weak interpretations ased on often explicit but questionable assumptions. At some sites it would be possible to determine a lack of bias on the part of the inhabitants in choice of wood if the type used for beams is the same as used in a quick hot fire, in the baking pit, and in the smelting furnace. It would also be possible to compare floated charcoal splinters 34 BOHRER Vol. 6, No. 1 to larger pieces from the same provenience to investigate differential destruction of charcoal types Many ethnobotanists, instead of employing cross sectional views of charcoal as a preliminary step in identification, use it as the only source of evidence. Now if specialists who identify modern wood regard the study of a cross section as only an initial step and secure additional information from radial and tangential sections, I fail to see how those among us who do so much less than that with prehistoric charcoal can hope to achieve a comparable level of reliability. We are dealing with an undetermined range of anatomical patterning in wood created by juvenile twigs as well as some shrubs and trees whose anatomy remains undescribed in texts. Current work could strive to match the standards set over 40 years ago in the identification of fish weir stakes and wattles from Boylston Street, Boston Bailey and Barghoorn 1942). To recapture this standard eager both proper tools and trainin ing the right tool helps get any job done quickly and well. In 1972 the Forest ean Laboratory a microscope with incident light (vs. iano to examine wood. In today’s technology an ordinary dissection of 30x has been replaced by metalurgical microscopes that have magnifications between 50 and 150 power and a working distance of 3 or 4 cm. or better. Such microscopes may already be employed in anthropology departments for lithic wear and pottery temper analysis. In a freshly broken piece of charcoal, critical features like spiral thickening, pitting on individual cell walls, or sieve plate details are apparent with no further preparation. It is true the same job can be accomplished with a Scanning Electron Microscope, but critical details of dicotyledon anatomy can be obtained more efficiently with a metallurgical microscope. Courses in plant anatomy and the technology of microslide preparation are rarely offered these days, and unless they are offered by someone around long enough to reach the rank of full professor, teachers are apt to be under qualified. Ethnobotanists younger than age 40 have reduced access to formal training in plant anatomy and many are unaware of how beneficial the acquisition of an educated eye and the ability to read the literature with true understanding can be. From the 1970s onward cheba and ethnobotanists have good naturedly accepted studies of the seed and pollen fraction of human feces without questioning the possibility of directly documenting the inevitable greens that are part of omnivorous dining. I know of two exceptions: a thesis from the University of Colorado (Stiger 1977) that deals with coprolites from Mesa Verde National Park and one from Texas A&M. University (Williams-Dean 1978) that concerns archaic coprolites from SW Texas. Throughout the same period the Journal of Range Management has regularly carried articles on determining range herbivore diets through fecal and fistula analysis (reviewed by Holechek, et al. 1982). Many articles detail problems relating to the quantification of results and their validity, a matter of interest to all. Identification of plant epidermal fragments resembles pollen and phytolith analysis in that it is labor intensive, requires expertise and an extensive plant reference collection. CHANGES IN PLANT LIFE Finding a new species as a result of archaeological investigation would be an event equivalent to a paleontologist recovering a new fossil form. In as much as our modern flora has essentially been in place 8,000 years (VanDevender and Spaulding 1979) such a discovery is unlikely. Our nation has been so bombarded by introduced plants that what is truly native can remain unrecognized. At such times the archaeological record makes a genuine contribution to our understanding. The presence of bugseed (Coris- permum) in hearths, in a Mimbres jar from Swartz ruin, New Mexico and in human coprolites from Cowboy Cave, Utah can be helpful finds when their age has been con Summer 1986 JOURNAL OF ETHNOBIOLOGY 35 firmed by accelerator radiocarbon dating (Betancourt 1984). Furthermore, the “useful weed” status of some annuals now rare or locally extinct can only be appreciated from their broad distribution in archaeological contexts (Bohrer 1978). Modern more aggressive introduced plants have altered their role in disturbed habitats. Local changes in southwestern vegetation have been of such a magnitude that a contrast is apparent within the lifetime of one person. Historic changes in dominant vegetation have been acknowledged recently in what may become a standard reference of the biotic communities of the American Southwest (Brown 1982). Our semi-desert grassland is almost obscured by scrubby trees, brush and cacti. Two native shrubs, bur- roweed (Isocoma tenuisecta) and snakeweed (Gutierrezia sarothrae) have replaced grass understory over millions of acres and serve as indicators of former grasslands (Brown 1982:131)}. In west central Arizona the grassland has shifted from perennials to introduced annuals (Brown 1982:129). In conifer woodland of the Great Basin, junipers have invaded large areas of former grassland (Brown 1982:53). Arizona climax cienegas and marshlands have decreased greatly since 1890 (Hendrickson and Minckley 1984). The tendency has been to account for modern plant assemblages by considering insufficient limiting factors such as either soil or climate. All too seldom is the biotic factor brought to attention, e.g. in terms of competition from other plants, grazing or uman intervention. The grazing industry has vested interests in not recognizing the role of overgrazing in the loss of grassland, just as the lumber industry does not speak of changing the composition of the primieval forest. Two organizations in the Southwest have evolved different approaches to environ- mental reconstruction. SARG, the Southwest Anthropological Research Group, discovered one plant ecologist (Kiichler 1964) who takes the anthropogenic factor seriously into account when mapping vegetation (Effland 1978}. Kiichler predicts what a modern plant assemblage would become if suddenly today human influence were removed. Unfortunately for archaeologists, Kiichler’s potential vegetation maps represent predic- tions that lack the supposition that landscapes will revert to pre-conquest vegetation. The link to the past is missing. In contrast, environmental reconstruction on the Gila Salt Aqueduct Survey in central Arizona (Fish 1985) began by obtaining numerous (N =400 +) pollen samples from a variety f sites. The importance of redundant informa- tion and duplication of effort became apparent by the internal patterning between suites of pollen samples. A valuable data bank of cultural and ecological information was created Suzanne Fish could recognize anthropogenic pollen floras because of their characteristic Signature in archaeological sites (an abundance of insect pollinated Arizona poppy Kallstroemia, spiderling Boerhaavia, and globemallow Sphaeralcea) contrasted so sharply with pollen spectra from locations away from intense human activity. Modern analogs to the prehistoric disturbance flora proved difficult to locate. Biotic factors active in the creation of a disturbance flora have changed in intensity and direction. Suzanne warns us (1985:84) “If disturbance taxa which disperse relatively little pollen can influence archaeological spectra so greatly, interpretation of environment and climate from more prolific producers should proceed with great caution.’”’ Our limitations are increasingly apparent. No neat, easy formula helps one detour around the dangers in the reconstruction of vegetation. The uncertainty (nay, disagreement) in attributing vegetational change to cultural or natural causes in the archaeological record has been pointed out recently by several authors (Bryant and Holloway 1983; King and Graham 1981). Overly confi- dent attempts to quantify past food resources based on the presumed distribution of vegetation, carrying eA fi calories have been numerous i in recent years (see Moran 1984). A more conservative of is expressed by King and Graham (1981:139) .. . ‘nit cabal be possible to esa the knowledge of a few key taxa necessary to fairly accurately quantify those resources and to project variations 36 BOHRER Vol. 6, No. 1 in their relative abundance and distributions back through time as they may have responded to environmental change.” CONTEMPORARY STUDIES It is never too late to conduct contemporary ethnobiological studies. I think many modern studies are better than years ago because of the ecological orientation. Even though previous studies existed on southern California Cahuilla ethnobotany, the work of Bean and Saubel (1972) provided important insights into the manipulation of plant life. One can seek examples from other parts of the world, but those nearest one’s own territory seem to carry more impact. For example, when I read how fan palms are planted from oasis to oasis by the Cahuilla, it was easier to understand how agaves might have found their place on terraced slopes in southern Arizona (Fish et al. 1985). No matter what our task is in ethnobotany, no matter how supposedly tedious, it needs to be treated as important and worthy of our best efforts. The seemingly insignifi- cant record that Gary Nabhan made of Arizona barley grass in Papago fields took on a new dimension when viewed from the perspective of a palynolgist working with the prehistoric pollen rain. Research by David Clawsen (1985) inadvertently helped us understand the color coding of maize that corresponds to the symbolic colors of the sacred directions among the pueblos. Although ritual observances frequently reinforce prac- tices (once) vital to the economic survival of a people, the particular advantage bestowed by preservation of strains of colored maize has been obscure. David Clawsen tells us color coding of crops such as potatoes, yams and many legumes helps a cultivator plant according to given maturity periods. The farmer who maintains a greater diversity of colored varieties of any one crop carries a kind of physiological crop insurance. Despite how the rains distribute themselves, or whether or not the frost is early or late, the chances are good that there will be some crop to harvest. Clawsen has provided ethnobotanists working with the archaeological record a lost dimension, for by the time we examine the physical remains of any crop, all the varieties are the same color, black. ETHNOBOTANISTS AND ETHNOBOTANICAL INTERPRETATION Ethnobotanical interpretation of prehistoric plant remains should consider factors like the method of excavation, sampling design, context of recovery, and post-depositional factors until all pieces of the puzzle fit in harmony (Fig. 1). But there are aspects of interpretation that transcend site specific boundaries. One must provide perspective derived from a broad range of experience in dealing with the archaeological plant record (directly or via literature) and from the rich, deep kaleidescopic patterns formed by modern and historic interrelationships of plants and man. Archaeological botanical evidence can be likened to a crumb surviving a whole loaf of bread. Or envision it as a stanza of a longer ballad. One needs to provide the refrain that brings unity, lest such botanical stanzas land in the scrap heap of our intellect. A supra-tribal, supra-national viewpoint combined with a competent ecological perspective can supply vitality to any otherwise drab scholarly discussion. Am I asking too much? The way our society organizes itself makes preparation for the task more difficult. e compartmentalization of knowledge into fixed disciplines works against the creation of an ethnobotanist. The time needed to acquire a dual background often proves excessive unless one begins as a undergraduate. It is none-the-less essential to develop the interdisciplinary vocabulary and theoretical understanding that allows one to read the literature directly, for therein lies a rich reservoir of ideas and a bridge to alleviate misunderstanding. Summer 1986 JOURNAL OF ETHNOBIOLOGY 37 Harvest Thresh t Winnow \ Sun Dry | ee. sic a Parch, Boil Grind Store Hearth Metate, Mano Mats, baskets or assoc. bin pottery River Sheet trash Mound trash Pit trash Loss, addition and mixing by ants, earthworms, rodents, etc. Differential resistance to erosion y part; by species | Differential conditions of preservation within one site; between sites | Differential recovery by flotations/screens (slumpage, churning; additions from wind, excavator, flotation) Differential identification | Quantification Interpretation FIG. 1.—The formation of the archaeological seed record. Neophyte ethnobotanists greatly benefit by being shown how the abstractions of the textbook apply to the field. In subjects like plant anatomy, taxonomy, ecology, evolution and core courses in anthropology, content is commonly emphasized over developing skills in problem solving. To resolve a baffling question a more experienced researcher can supply the impetus for a younger colleague to try a new approach or te-evaluate an old one. Many practicing ethnobotanists have incorporated these skills (Toffler 1971:414) in their own approach to problem solving: (1) classifying and reclas- sifying information, (2) determining its veracity; (3) shifting categories when necessary; (4) moving from the concrete to the abstract and back again; (5) looking at a problem from a new direction; (6) shifting from the inductive to deductive methods and back again. As ethnobotanists exist in increasing numbers as independent consultants, a form of ancillary or off-campus research internship would seem desirable for students. If problem solving skills are not formally taught, opportunities should be available where 38 BOHRER Vol. 6, No. 1 skills can be developed. In addition, the research topics that intrigue professionals who are directly and heavily involved in research are apt to be more original than well-massaged topics like the origin of agriculture. In the past people with ethnobotanical interests have found positions in colleges, universities or museums. Formerly men with doctorates held academic positions exclusively whereas now they are joined by a small group of women. Traditionally (since the 1930s) the museum has nourished and protected ethnobotanical careers of men and women of all levels of education. I am reminded of the support of Melvin R. Gilmore and Volney H. Jones at the Museum of Anthropology at the University of Michigan, of Margaret Towle’s association with the Harvard Botanical Museum, of Albert Whiting with the Museum of Northern Arizona, of Hans Helbaek and the National Museum of Copenhagen, Jaques Barrau at the Museum National de Histore Naturelle in Paris, Edgar Anderson and Hugh Cutler of the Missouri Botanical Garden. The trend continues to expand as many additional museums (such as the Bernice P. Bishop Museum in Honolulu and the Arizona State Museum in Tucson) find places for people with ethnobotanical skills. RESERVOIRS OF IDEAS AND INSPIRATION Everyone, including the ethnobotanist to be, must ask, ‘‘What shall I read?” “What is worth reading again and again?” “What are the classics?” Sometimes we read to know the level of understanding achieved in times past, but the real value is in capturing “the visions and the styles of the scientists who preceded us” (Thompson 1984:187). Short articles seldom contain such an essence and short articles indeed predominate the ethnobotanical literature. Really good autobiographies or biographies of ethnobotanists are all too difficult to find: their lack impoverishes us. My own library contains some favorite books: Edgar Anderson’s (1954) Plants, Man, and Life, Frank Cushing's (1920) Zuni Breadstuff, M.P. Harrington's (1932) Tobacco Among the Karuk, W.W. Hill's (1938) The Agricultural and Hunting Methods of the Navajo Indians and Paul Weatherwaxe’s (1954) Indian Corn in Old America. More recently I have added Gary Nabhan’s (1985) Gathering the Desert. The above list contains no volumes, chapters or appendices in archaeological ane reports because so few are models of “how to know, how to think, and how to write! (Thomson 1984:187). Few contain the flavor of imaginative thought, clean, stylish prose and front-line science (Thomson 1984). But there are some. I have already dis- cussed one recent example (Toll 1984) under flotation. Toll has distilled voluminous observations into a single paper with breathtaking conciseness. The facts at her dispo do not overwhelm her. Another example is Goodyear’s (1975) Hecla II and Ill. Goodyear is an archaeologist that sees no barriers between anthropology and botany. The waters of each intermingle and sustain him as he develops detailed methods of pling ultural and botanical data and constructs models of prehistoric subsistence activity in remote areas of southwestern Arizona. The strength and weaknesses of the techniques employed are candidly presented while the author places his work in a broad regional perspective. His manner of investigation has been rightfully described as thorough, creative and efficient (Gumerman 1977:12). Several examples of how to know, think and write date from the 1950s. One of these is V.H. Jones and R.L. Fonner’s Appendix C Plant Materials from Sites in the Durange and La Plata ire Colorado. Every shred of information is given the courtesy of thorous! treatment and the line of reasoning clear] sented. Findi are placed agains background of knowledge of plant soiie cas carefully ROO that later finds of cultivated amaranth are anticipated. The discussion of maize is unusually complete id. The long period of evolution of the published report is traced for the reader. Summer 1986 JOURNAL OF ETHNOBIOLOGY 39 The other example is Hans Helbaek’s (1952) Early Crops in Southern England. His understanding of what constitutes crops is truly refreshing, for he includes far more than wheat and barley. Flax, rye and crabapples; chess and oats weave in and out of discus- sions. The role of the Brassicas, Chenopodium, Polygonum and Galeopsis as plants of cultivated fields and gathered food sources forms a balanced part of his concern for early crops. His tables tell of the increase in grain size with the continued cultivation of several crops and one learns of the waxing and waning of the prevalence of naked barley against the backdrop of other available choices, including rye. Never is southern England treated as an island whose plant history is devoid of connections. The context is European and the scope epic in its coverage of the neolithic, bronze and early iron ages. The keeness of his observation and his critical perception of events prevails in his writing. Would that we could achieve as much in the years ahead. REFLECTIONS The history of research on maize is one of looping back over territory once thought to be conquered. Cob and kernel measurements, pollen identification, phytolith recog- nition, carbon 14 dating and corn evolution—all have been the focus of controversy. Has maize been such a popular band wagon that some people have published prematurely? Or is it that newly developed techniques to deal with maize have been readily accepted without critical evaluation? As we gain enthusiasm for the human capacity to begin to tinker with the evolution of many other plants besides maize, can not this same thing happen with every new species proposed as a cultivated or domesticated plant? The above described trends in the study of maize are ones that I would rather not project into the future. Researchers who worked with ethnographic maize and those who dealt with prehistoric remains sometimes hummed little tunes the other did not hear. In similar cases the meetings of the Society for Ethnobiology can serve as a forum. It is the larger archaeological projects that seem to have made the most headway with methodological problems concerned with flotation and pollen analysis. This is troubling, for larger projects are dwindling and the burden should be shared more evenly. Enough site specific studies have accumulated that syntheses are being produced, an admirable trend that should continue. Presently we perceive common denominators in subsistence practices without daring to ask “Why?” A greater participation in the theoretical realm of model building and model demolition surely must be near. Attempts to reconstruct vegetation have varied in approach and results. We have had to learn a lot about our limitations. We may have to be content with estimating abundance of a few species rather than dealing comprehensively with absolute abun- dance. By analyzing a spectrum of contexts in which prehistoric plants occur can we discover hing of their former competiti lationships? The disturbed ground plants of today and of pre-contact times were not always one and the same. The role of soil and climate in influencing the formation of a given plant community has enjoyed exclusive popularity too long. The biotic factor, including the anthropogenic one, also needs careful consideration. Nevertheless, the path is still not all that clear. Our growing humility might be transmitted to archaeologists who still believe in miracles. Museums have served and probably will continue to serve as a focal point for ethnobotany. Apprenticeships to museums and other independent organizations active in sponsoring research may be the best way for a prospective student to develop skills and apply book learning to concrete issues. It is never too late to conduct contemporary ethnobotanical studies. Field work in the ethnographic and ethnobotanical realm coupled with omnivorous reading sharpens appreciation for the multifaceted patterning of plant-man interrelationships. While it still can be secured, formal training in plant anatomy can reverse a downward trend in the quality of reports concerning charcoal, 40 BOHRER Vol. 6, No. 1 coprolite content and related applications. If ethnobotanists would first dream that archaeobotanical site reports might serve as reservoirs of ideas and inspiration, it would not be long before we could read them. A viewpoint that transcends both tribe and nation blended with a competent ecological perspective can, with continual grooming, guide others to ‘New Directions in Ethnobiology”’. LITERATURE CITED ADAMS, KAREN R. and ROBERT E. cautionary note. Plains Anthropol. GASSER. 1980. Plant microfossils considerations, and sampling tech- niques and approaches. Kiva 45:293- ANDERSON, EDGAR. 1944. Maize reventador. Ann. Mo. Bot. Garden 31:301-314. 1946. Maize in Mexico a preliminary survey. Ann. Mo. Bot. Garden 33:147-247. 1967. Plants, Man and Life. Univ. California Press, Berkeley. and HUGH C. CUT- LER. 1942. Races of Zea mays: I. their recognition and classification. Ann. Mo. Bot. Garden 29:69-86. BAILEY, I. W. and ELSO S. BARG- HOORN, JR. 1942. Identification and physical condition of the stakes and wattles from the fishweir, Pp. 82-89 in The Boylston Street Fishweir by Archaeology 2, Phillips Academy, BARGHOORN, ELSO S., M. K. WOLFE and K. CLISBY. 1954. Fossil maize from the Valley of Mexico. Harvard Univ. Bot. Mus. Leaflets 16(9):229- 23 9. BEAN, LOWELL J. and KATHERINE S. SAUBEL. 1972. Temalpakh. Malki Museum Press, Morongo Ind. Reserv. California. BENDER, MARGARET M., DAVID A. BAERREIS and RAYMOND L. STEVENSON. 1981. Further light on isotopes and Hopewell agri- environmental reconstruction: a 30:145-147. BETANCOURT, JULIO L., A. LONG, D. J. DONAHUE, A. J. T. JULL and T. H. ZABEL. 1984. Pre-columbian age for North American Corisper- mum L. (Chenopodiaceae) confirmed by accelerator radiocarbon dating. Nature 311:653-655. BIRD, ROBERT McK. and MAJOR M. GOODMAN. 1977. The races of maize V: grouping maize races on the basis of ear morphology. Econ. Botany 31:471-481. and JUNIUS B. BIRD. 1980. Gallinazo maize from the Chicama Valley, Peru. Amer. Anti quity 45(2):325-332. BOHRER, VORSILA L. 1970. Ethno- botanical aspects of Snaketown, 4 hohokam village in southern Ari zona. Amer. Antiquity 35:413-430. and K. R. ADAMS. 1977. Ethnobotanical techniques and approaches at Salmon Ruin, NM. Eastern New Mexico. Univ. Con- tributions in Anthrop. 8(1):1-215. 1978. Plants that have become locally extinct in the South- west. New Mexico J. Sci. 18(2):10-19. BRETERNITZ, DAVID A. 1984. Dolores archaeological program: synthetic report 1978-1981. U.S. Dept. Int., Bur. Reclamation, Engineering and Research Center, Denver. BROWN, DAVID E., ed. 1982. Biotic communities of the American South- west United States and Mexico. Desert Plants 4:1-342. BRYANT, VAUGHN M. and RICHARD G. HOLLOWAY. 1983. The role of palynology in archaeology, Pp. 191- 224 in Advances in Archaeological Summer 1986 JOURNAL OF ETHNOBIOLOGY 4] LITERATURE CITED (continued) Method and Theory 6. (Michael B. Schiffer, ed.). Academic Press, NY. CARTER, GEORGE F. and EDGAR ANDERSON. 1945. A preliminary survey of maize in the Southwestern United States. Ann. Mo. Botany Garden 32:297-323. CLAWSEN, DAVID L. 1985. Harvest Security and Intraspecific Diversity in Traditional Tropical Agriculture. Econ. Botany 39(1):56-67. COVAS, GUILLERMO. 1949. Taxonomic observations on the North American species of Hordeum. Madrojio 10:1- | CUSHING, FRANK H. 1920. Zuni bread- stuff. Indian Notes and Monographs 8, Mus. Amer. Indian, New York. CUTLER, HUGH C. 1956. The Plant remains, Pp. 174-183 in Higgens Flat Pueblo Western New Mexico, by Paul S. Martin, et al. Fieldiana, Anthrop. 45 ey: LO, Tee Plant remains from the Carter Ranch Site, Pp. 227-234 in Chapters in the prehistory of Eastern Arizona II, by Paul Martin, et al. Fieldiana, Anthrop. 55. . 1966. Corn, cucurbits and cotton from Glen Canyon. Univ. Utah Anthropol. Papers 80:1-62. EFFLAND, RICHARD. 1978. Applica- tions of computer graphic techniques to SARG data, Pp. 149-167 in Investi- gations of the Southwestern Anthro- pological Research Group: an experi- ment in archaeological cooperation (Robert C. Euler and George J. Gum- erman, eds.) Mus. N. Ariz. Bull. 50. EMERSON, R. A. and H. H. SMIIF. 1950. Pp. 1-30 in Inheritance of ker- nal rows in maize. Cornell Univ. Agric. Experiment Sta. Memoir 296 (cited in Cutler 1966 FISH, SUZANNE K., PAUL R. FISH, CHARLES MIKSICEK and JOHN MADSEN. 1985. Prehistoric Agave cultivation in Southern Arizona. Desert Plants 7(2):107-112, 100. FISH, SUZANNE K. 1985. Prehistoric disturbance floras of the lower Sonoran Desert and their implica- tions. Amer. Assoc. Strat. Palynolo- gists Contribution Series No. 16: 77-88. FORD, RICHARD I. 1981. Gardening and farming before A.D. 1000: pat- terns of prehistoric cultivation north of Mexico. J. Ethnobiol. 1(1):6-27. FORD, RICHARD. 1985. Prehistoric food production in North America. Mus. of Anthrop. Univ. Michigan Anthropol. Papers 75. GALINAT, WALTON C. 1985. Domesti- cation and diffusion of maize. Pp. 245-278 in Prehistoric Food Produc- tion in North America. (Richard Ford, ed.) Mus. of Anthrop. Univ Michigan Anthropol. Papers 75. GALINAT, WALTON C. and J. H. GUN- NERSON. 1963. Spread of eight rowed maize from the prehistoric Southwest. Harvard Univ. Bot. Mus. Leaflets 20:117-160. GASSER, ROBERT E. 1981. Hohokam use of desert plant foods. Desert Plants 3:216-234. GOODMAN, MAJOR M. and E. PATER- NIANI. 1969. The races of maize: II. choice of appropriate characters for racial classification. Econ. Botany 23:265-273. GOODYEAR, ALBERT C. Ill. 1975. Hecla II and I an interpretive study of archaeological remains from the Lakeshore project, Papago Reserva- tion, south central Arizona. Arizona State Univ. Anthropol. Res. paper 9. Tempe. GOULD, FRANK W. 1951. Grasses of Southwestern United States. Univ. Arizona Biological Sci. Bull. 7. GOULD, STEPHEN J. 1984. A short way to corn. Natural History 93:12-20. GRIFFIN, JAMES B. 1978. Volney Hurt Jones, ethnobotanist: an appreciation, Pp. 3-19 in The nature and status of ethnobotany (Richard I. Ford, ed.) Mus. Anthrop. Anthropol. Paper 67. 42 BOHRER Vol. 6, No. 1 LITERATURE CITED (continued) GUMERMAN, GEORGE. 1977. Review: Hecla II and III by A.C. Goodyear. Amer. Antiquity 42:141-142. HALLY, DAVID J. 1981. Plant preser- vation and the content of paleo- botanical samples: a case study. Amer. Antiquity 46:723-742. HARRINGTON, JOHN P. 1932. Tobacco among the Karuk Indians of Cali- fornia. Bur. Amer. Ethn. Bull. 94. HELBAEK, HANS. 1952. Early crops in southern England. The Prehistoric Society 12:194-233. HENDRICKSON, DEAN A. and W. L. MINCKLEY. 1984. Cienegas-vanish- ing climax communities of the American Southwest. Desert Plants 6(3):131-175. HENDRY, G. W. and M. K. BELLUE. 1936. An approach to Southwestern agricultural history through adobe brick analysis, Anthropol. Ser. 1(5). HILL, W. W. 1938. The agricultural and hunting methods of the Navajo Indians. Yale Univ. Publ. Anthrop. 18:1-194. HOLECHEK, JERRY L., MARTIN VA- VRA and REX D. PIEPER. 1982. Botanical composition determination of range herbivore diets: a review. J. Range Mgmt. 35(3):309-315. ILTIS, HUGH. 1983. From teosinte to maize: the catastrophic sexual trans mutation. Science 222:886-893. IRWIN, HENRY and ELSO S. BARG- HOORN. 1965. Identification of the pollen of maize, teosinte and Trip- sacum by phase contrast microscopy. Harvard Univ. Bot. Mus. Leaflets 21:37-56. JONES, MELVIN D. and L. C. NEWELL. 1948. Size, variability and identifi- cation of grass pollen. J. Amer. Soc. Agronomy 40:136-143. JONES, VOLNEY H. and ROBERT L. FONNER. 1954. Appendix C Plant materials from sites in the Durango and La Plata areas, Colorado, Pp. 93- 115 in Basket Maker II sites near Durango, Colorado by Earl H. Morris and Robert F. Burgh, Carnegie Inst. Washington Publ. 604. KING, FRANCES B. and RUSSELL W. GRAHAM. 1981. Effects of ecological and paleoecological patterns on sub- sistence and paleoenvironmental reconstruction. Amer. Antiquity 46:128-142. KUCHLER, AUGUSTUS W. 1964. Poten- tial natural vegetation of the coter- minus United States. Amer. Geo- graphical Soc. Special Res. Publ. 36. New York. MACKEY, JAMES C. 1985. A thirteenth century A.D. example of the success- ful use of archaeological corn collec- tion for paleoenvironmental recon- struction: a reply to Benz. Plains Anthropol. 30:149-159. ee BARBARA, T. A. O, and A. BLUMENSCHEIN. oa. Chromosome constitution of races of maize. Colegio de Post Graduados, Chapingo, Mexico. MINNIS, PAUL. 1981. Seeds in archae- ological sites: sources and som interpretive problems. Amer. Anti- Loz. MORAN, EMILIO F., ed. 1984. The eco- system concept in anthropology. Amer. Assoc. Adv. Sci. Selected Sym- posia Series 92. NABHAN, GARY P. 1983. Papago fields: arid land ethnobotany and agricul- tural ecology. Unpubl. Ph.D. Dissert. (Arid Lands) Univ. Arizona, Tucson. _ 1985 Gathering the desert. Univ. Arizona Press, Tucson. PEARSALL, DEBORAH M. 1980. Anal- ysis of an archaeological maize kernel PENDLETON, MICHAEL W. 1983. 4 comment concerning “Testing flota- tion recovery rates.” Amer. Antiquity 48:615-616. ee ee Tae .—_ anti ee SS os f a Summer 1986 JOURNAL OF ETHNOBIOLOGY 43 LITERATURE CITED (continued) RAJHATHY, T., J. W. MORRISON and S. SYMKO. 1964. Interspecific and intergeneric hybrids in Hordeum. Pp. 195-212, gs Genetics I. PUDOC, Wagenin ROVNER, IRWIN. 1983. Plant opal phytolith analysis: major advances in archaeobotanical research, Pp. 225- 266 in Archaeological Method and Theory 6 (Michael B. Schiffer, ed.) Academic Press, New York. SCHAAF, JEANNE M. 1981. A method for reliable and quantifiable sub- sampling of archaeological features for flotation. Mid-Cont. J. of Archae. 6:220-248. SCHOENWETTER, JAMES. 1974. Pollen records of Guila Naquitz cave. Amer. Antiquity 39(2}:292-303. STIGER, MARK A. 1977. Anasazi diet: the coprolite evidence. Unpubl. thesis. Univ. Colorado, Boulder. THOMSON, KEITH S. 1984. The litera- ture of science. Amer. Scientist 72: 185-187. TOFFLER, ALVIN. 1971. Future Shock. Bantam ed., Random House, New York. TOLL, MOLLIE S. 1981. Flotation and Macro-botanical Analyses at 29S]629: A Pueblo I-II Village in Chaco Canyon. Castetter Laboratory for Ethnobotanical Studies, Technical Series No. 49. 1984. Taxonomic diversity in ‘flotation and macro- botanical assemblages from Chaco Canyon, Pp. 241-250 in Recent research on Chaco prehistory (W. James Judge and J. D. Schelberg, eds.) Reports of the Chaco Center 8 Albuquerque. TROUGHTON, JOHN H., P. V. WELLS, and H. A. MOONEY. 1974. Photo- synthetic mechanisms and paleo- ecology from carbon isotope ratios in ancient specimens of C-4 and CAM plants. Science 185:610-612. VAN DEVENDER, THOMAS R. and W. GEOFFREY SPAULDING. 1979. Development of vegetation and cli- mate in the Southwestern United States. Science 204:701-710. WAGNER, GAIL E. 1982. Testing flota- tion recovery rates. Amer. Antiquity 47:127-132. WALKER, S. S. and J. K. LEWIS. 1979. Occurrence of C-3 and C-4 photo- synthetic pathways in North Ameri- rary flotation obs Mid-Cont. J. Archae. 1:77- WEATHERWAX, oto 1954. Indian Corn in Old America. The Macmil- lan Co., New York. WELLHAUSEN, E. J., L. M. ROBERTS one gaa tage 1952. Races ize in Mexico their origin, ee roe lett and distribution. Bussey Inst., Harvard Univ., Cambridge WESTERN, A. CELIA. 1963. Wood and charcoal in archaeology, Pp. 151-158 in Science in archaeology (Don Broth- well and Eric Higgs, eds.) Thames and Hudson, London. WILLIAMS-DEAN, GLENNA. 1978. Ethnobotany and Cultural Ecology of Prehistoric Man in Southwest Texas. Unpubl. Ph.D. Dissert. Texas ARM Univ., College Station. pas BOOK REVIEW Vol. 6, No. 1 Prehistoric Food Production in North America. Richard Ford, editor. Museum of Anthropology, University of Michigan Anthropology Papers, No. 75. Ann Arbor: University of Michigan Press, 1985. This volume of eleven papers assesses plant domestication, diffusion, horticulture, and the effects of these processes on the prehistoric human cultures north of Mexico. Originally derived from presentations made at an advanced seminar at the School of American Research in March, 1980, these manuscripts reflect a fertile interaction between selected archaeologists, archaeobotanists and crop evolutionists. Both its successes and failures are consequences of its transition into book form during a water- shed in American ethnobotany. First, the outline of the structure of the book and its unifying themes. In an excellent preface by Richard Ford, written four years after the seminar on ‘/The Origins of Plant Husbandry in North America,” the following issues are outlined as the themes which participants were asked to address. The authors who contributed most to discussion of a particular issue are noted in parentheses. 1. Demography and subsistence patterns at the time of crop introduction from Mexico (Patty Jo Watson on Mississippi watershed cultural changes; Paul Minnis on archaic economies in the Southwest). 2. Modes and routes of crop diffusion (Dee Ann Story on the proposed east Texas route via the Gulf Coastal Plain; Walton Galinat on eight-rowed maize diffusion; Frances King on cucurbits). 3. Genetic characteristics of early crop introductions (Charles Heiser and Frances King on cucurbits; Galinat on maize; Asch, Asch and Heiser reviewing chenopods, bottlegourds and tobacco, and Ford’s summary). 4-5. Integration into existing economies, and resulting changes (Minnis on the Southwest, Watson on the Mississippi Valley; and Ford on regional crop complexes). 6-7. T logical i ification of food production and constraints to intensifica- tion and expansion (Ford and Minnis on processing technologies; Watson on materi correlates of agriculture; and Cowan and Minnis on environmental management through burning, irrigation and transplants). 8. Effects on sedentarism (Watson and Minnis on respective regions). 9-10. Plant domestication, breeding or varietal evolution north of Mexico, and its recognition in the archaeological record (Heiser on sunflowers, sumpweed, and minor cultigens; Cowan on the same, as well as on trees; Asch and Asch on knotweed, maygrass and little barley; and Galinat, Minnis and Berry on changes in maize; Berry and Minnis on early dates of domesticates). These annotations suggest that there was quite a bit of overlap in coverage of certain items, such as squash, sunflower, maize and chenpod prehistory, and difficulties with dating early maize and cucurbits. While this redundancy may make some readers skip over reiterated details from one paper to the next, the authors must nevertheless be commended for tackling common problems from different perspectives. In this manner, we learn that acceptance of sumpweed, giant ragweed, devil’s claw, jerusalem artichokes, knotweed, groundnuts and panicgrass as true domesticates was open to debate at the time of the seminar. Early introduction dates for Cucurbita pepo, Lagenaria sicerana and more conservative dates for Zea mays were being accepted. The Eastern cultivated chenopod was being more confidently identified as Chenopodium berlandieri vat nuttalliae, and Cucurbita pepo var. texana as a wild progenitor or as a feral (regressive) form was being seriously reconsidered. Most importantly, this book, and in particular, Richard Ford’s synthetic essays, allow us to better compare and contrast the development of horticulture in the central Summer 1986 JOURNAL OF ETHNOBIOLOGY 45 Mississippi watershed and the U.S. Southwest. Although the book does not truly cover North America—it excludes the Upper Missouri, Florida, the eastern seaboard, the Great Basin, the western Plains, and northern Mexico—it is a valid advance toward a continental synthesis on horticultural evolution. Since North America (north of Mesoamerica} is still regarded by Europeans, the USDA and many Latin American ethnobotanists as altogether lacking an agricultural history which includes indigenous domestication and diversification, this volume proves that North Americans need not travel to other continents in order to study crop evolution. Yet Richard Ford’s Preface admits the volume’s unfortunate flaw—the years following the conference witnessed “the most productive period in the history of American ethnobotany”. Ford notes that “Numerous cultural resource management projects have yielded an unprecendented quantity of plant remains from archaeological sites throughout the United States”. The efforts to integrate these results into various chapters in the four years of manuscript revision prior to publication were uneven, and generally inadequate, since subsequent advances radically changed the picture presented at the 1980 conference. By the time the volume was distributed, new dissertations, journal articles and books had clarified the status of chenopods (Wilson), barley (Adams, Bohrer and Gasser}, agaves (Miksicek, Fish and Bohrer), pepos (Wilson and Conard], devil’s claw (Bretting, Nabhan, Whiting, Dobyns and Euler), panicgrass (deWet, Kaemlein and Nabhan), mustards (Bye), amaranths (Miksicek), and saltgrass (Felger and Yensen) and numerous “wild” plants in the East (Munsun, 1973). Maize diffusion and classification had been rewritten by Doebley and Goodman, and Merrick had reported on new insights into Cucurbita mixta and C. moschata at national meetings. Miksicek, Winter, Bohrer, Gasser and others have found earlier dates of crop introductions (tobacco, amaranth, etc.) into the Great Basin and Southwest. More disconcerting, turkey prehistory has lately been rewritten by Rea, McKusick and Steadman, but animal domesticates are hardly noted in the volume at all. Zooarchaeology by Szuter, Bayham, White, Emslie, and Fritz fails to merit any mention in the transition from gathering and hunting to agriculture plus hunting and gathering. New hypotheses on the origins of agriculture by Rindos (1984) and Johns supercede many of the ones presented here, although they remain controversial. There is no fault in the editor or authors for having some of their interpretations become obsolete due to new advances. That portions of this well-written book are already outmoded signifies great growth in interdisciplinary efforts, to which these scientists have all contributed. This obsolescence was not planned, nor was the seeming exclu- sion of other work. In short, Prehistoric Food Production offers a snapshot of a landscape that has already changed since this book’s conception. All North American ethnobiologists are indebted to those included in this volume who have laid such a ground- work for advancing a synthetic understanding of American agricultural origins. Selected References: Munsun, Patrick J., ed. 1984. Experiments and Observations on Aboriginal Wild Plant ood Utilization in Eastern North America. Indiana Historical Society, Prehistoric Research series, Indianapolis. . Rindos, David. 1984. The Origins of Agriculture: An Evolutionary Perspective. Academic Press, Orlando. For other notes on recent developments in North American plant domestication, readers are referred to articles which have appeared in Journal of Ethnobiology, Economic Botany, American Journal of Botany, American Antiquity and Desert Plants. —Gary Nabhan, Assistant Director Desert Botanical Garden Phoenix, AZ 85008, U.S.A. 46 BOOK REVIEW Vol. 6, No. 1 Of Plants and People. C. B. Heiser, Jr. faunas of Oklahoma Press, Norman, Okla- homa. 1985. Pp. xiii, 237; 65 figs Professor Heiser, who has published several books on economic plants of the Andean region, has a facility in sharing his enthusiasm and interdisciplinary outlook red the study of aoe of the poorly known economic plants of the Andes with his readers. 1 field botanist, his writings encompass botany, anthro- series and eee and bring out the often major roles that these usually minor crop plants play in the life of native peoples in South America. This new contribution is, in the words of the author, “unlike my other works .. . not confined to one subject or to a single group of plants . . . and all of the plants have interesting interactions with people.” In addition to its value to botanists and anthro- pologists, he has successfully striven to “write for the student and amateur in mind.” The book is divided into 13 chapters: Pepos and people; Totora and Thor; Little Orange of Quito; Chenopods; Sangorache (Amaranthus); Trip to Tulcan; Lupines; Green ‘“‘Tomates” (Physalis) and purple “Cucumbers” (Solanum); Peppers, Peperomias; Sump- weed (Iva); a Plague of locusts, Seeds, sex and sacrifice—religion and the origin of agriculture. The extensive bibliography is divided into sections, each chapter with its own list of items and often analytic notes. It is to be hoped that the appearance of this type of book may encourage the publica- tion of others in the field of ethnobiology. Ethnobotanists will want to have Of Plants and People at hand on their shelves. Richard Evans Schultes Professor Emeritus Botanical Museum of Harvard University Cambridge, Massachusetts 02138 J. Ethnobiol. 6(1):47-65 Summer 1986 NEW DIRECTIONS OF PALYNOLOGY IN ETHNOBIOLOGY RICHARD G. HOLLOWAY Palynology Laboratory Anthropology Department Texas A®M University College Station, TX 77843 VAUGHN M. BRYANT, JR. Anthropology Department Texas A®M University College Station, TX 77843 STRACT. oo palynology, fairly new as a discipline, originated in studies conducted less than 50 y ars ago. During this early developmental stage, it suffered of: (1) inadequate communication between botanically-oriented palynologists and field archaeologists, (2) differences in conceptual ar aber between archaeologists and archaeological palynologists, (3) fossil pollen stt ical sites often had been incorporated into final reports only as ae (4) a growing number of studies on the il pollen remains of archaeological sites were published almost exclusively in “contract bution is restricted, and (5} aa papaiine oi statistical] techniques had not been ‘applied to fossil pollen data from archaeological sites. Our article palynology, discusses some of the applications of ional pollen data in ‘archaeology, ‘and offers suggestions and concerns about the course of the discipline in the years to come. INTRODUCTION Palynology, broadly defined as the study of pollen and spores from both living and fossil seed plants (Hyde and Williams 1944), includes research in such various areas as pollen production and dispersal, composition and morphology of the exine (outer pollen wall), applications in stratigraphy and paleoecology, preservation, and in explaining man- plant interactions. Even though the field has a history of over 70 years, only during the Past several decades has significant emphasis been placed on analysis of fossil pollen from archaeological materials. Although most archaeologists know of the basics of Palynology, many of them are still unaware of the full range of data which palynology can provide. As a discipline, palynology is relatively young and its applications to archaeology are still being developed and refined. Thus, we feel that the full comprehension and appreciation of the impact and importance of archaeological palynology, requires a review of its historical development, an examination of its data base and specific techniques, a survey of the range of research questions which are currently arising, and finally, an exploration of the potentials of this discipline for future research needs in archaeology. HISTORICAL REVIEW Pollen analysis, initially defined as the calculation of pollen percentages, appears first to have been attempted on Swedish peat deposits by Gustav Lagerheim (in Witte 1905). In 1916, at a convention of Scandinavian naturalists (Davis and Faegri 1967), von 48 HOLLOWAY & BRYANT Vol. 6, No. 1 Post explained the potential of this technique by presenting the first percentage calcula- tions of fossil pollen. As Faegri (1981) observed, von Post regarded pollen analysis primarily as a method for dating Quaternary-aged sediments from northern Europe, although later he acknowledged the technique also could be a useful tool in paleoecology. The initial development of pollen analysis can therefore be viewed as a continuation or extension of the original palaeophysiognomic school which was exemplified by Blytt (1876). According to Faegri (1981:45) von Post’s pollen analysis was, “pure palaeofloristics: the regist of the occurrence and (what was new) relative frequency of taxa . The results and interpretation of von Post’s analyses were not based on pollen aualveis per se but on the techniques and data base associated with the palaeophysiognomic school (Faegri 1981). The influences of the palaeophysiognomic school can be clearly seen in the con- struction of the early pollen diagrams. The palaeophysiognomic school accentuated its emphasis upon the arboreal (tree) component of the forests. Thus, pollen from plants confined to the understory levels in the environment (non-arboreal plants) often were ignored or were calculated only as a function of arboreal pollen. Unfortunately, this limited approach to a i of fossil pollen became the standard during the early part of the twentieth cen During the ae "1900s, the field of pollen analysis did not produce the types of comprehensive fossil pollen studies which would become standard in later years. This problem was created initially by three main factors: (1) a poor understanding of pollen morphology; (2) the use, not the lack of availability, of poor quality optical equipment and (3) the use of inadequate laboratory extraction techniques for separating fossil pollen from matrix materials such as peat, soil, sand, and coprolites (Faegri 1981). The concept of using an ecological approach in pollen analysis was first introduced by Auer (1927) and later by Sears (1930a, 1930b) in their studies of North American iments. In Europe, similar concepts were first introduced at the 1933 meeting of the Baltic Committee (Faegri 1981). Once the ecological approach was introduced as a proper avenue of investigation, the directions and potential benefits derived from the field of pollen analysis were c forever. However, in spite of the demonstration of the new ecological approach to solve problems in fossil pollen studies, the influence of the older, synecological (palaeophysiognomic) school, was still evident in North American fossil pollen analyses up through the 1950s. This is evidenced by the reliance of palynologists such as Hansen and Potzger, upon arboreal pollen counts as the primary basis for their interpretations and constructing of pollen diagrams. Thus, as in the initial studies by von Post and others, these later North American palynologists tended to consider non-arboreal fossil pollen counts as representing unimportant information in the interpretation of paleoenviron- mental data. Almost immediately after the acceptance of an ecological perspective emphasizing ecological questions, palynology expanded to incorporate information related to the influence that man may have had upon prehistoric plant communities. In what seems to be the first pollen analysis of an archaeological site, Jessen (1935) reported on the fossil pollen analysis of sediments in North Jutland, Denmark. Later, Iversen (1941) again demonstrated the importance of using fossil pollen data in archaeology when he suc- cessfully dated the beginning of the Neolithic period in Denmark. He based his findings on the decline of local Ulmus (elm) pollen in sediments, which showed the appearance of herbs and weedy plant pollen normally associated with human disturbance. The first appearance of agriculture in southern Denmark was again documented by the later fossil pollen studies of Faegri (1944). Iversen’s and Faegri’s studies not only defined with fair precision the first arrival of agriculture into northern Europe and evidence that cereals of Middle Eastern origin et I 5 Eg Summer 1986 JOURNAL OF ETHNOBIOLOGY 49 were planted, but they also noted how cultures had altered the environment by clearing the forested areas using slash and burn techniques. Some of the later fossil pollen studies related to origins of agriculture in other regions of the world include Donner (1962), Duro (1965), Godwin (1944a, 1944b, and 1956}, Iversen (1949), Mitchell (1951, 1956), and Morrison (1959). Following Iversen’s (1944) example, other palynologists began using fossil pollen to clarify and interpret archaeological data. Troels-Smith (1960) examined pollen and seed remains from the Muldbjerg archaeological site located in West Zealand, Denmark. Based on pollen and plant macrofossil analysis of deposits from that Neolithic-age site, he reconstructed aspects of the past environment and showed how the overgrazing of domestic animals in that region of Denmark had created dramatic changes in the local plant composition. Dimbleby (1960) showed how the fossil pollen record from a Mesolithic site in England demonstrated changes in prehistoric forest composition caused by early man’s shift from a hunting and gathering economy to one which focused upon agriculture and pastoralism. Dimbleby (1963) conducted another fossil pollen study near Addington, England where he introduced a technique for calculating the concentration of fossil pollen in soils. He then used those pollen concentration values to show how Mesolithic cultures had modified the local vegetation. Applications of pollen analysis in North American archaeological sites was also pro- gressing. Paul B. Sears was one of the first researchers in North America to apply pollen analytical techniques to solving archaeological problems. In an early article, Sears (1932) suggested that a possible cause for the Hopewell expansion into areas of the eastern United States may have been a climatic shift that favored the growth of maize. Later, Sears (1937) conducted some of the pioneering pollen investigations of archaeological and non- archaeological sediments in the American Southwest. Although Sears’ initial attempts in the Southwest (Sears 1937) resulted in only limited success, he and others (Sears and Clisby 1952; Clisby and Sears 1956) continued examining open sites in the American Southwest while Roger Anderson (1955) began trying to recover fossil pollen from cave sediments. The work of these early palynologists demonstrated that fossil pollen was often difficult to recover but that it was present in many Southwestern sediments. In the late 1950s Paul S. Martin and his graduate students began a detailed look at the fossil pollen in lakes, peat bogs, and archaeological sites in the Southwest. Their studies convinced others that even though some deposits in the Southwestern United States were barren of fossil pollen, other areas yielded good records (Martin 1963). | The next two decades saw new developments in archaeological palynology. First, during the 1960s fossil pollen studies of archaeological sites became more common (Cox and Lewis 1965; Godwin 1967; Tsukada and Deevey 1967; van Ziest 1967); and secondly, the 1970s saw a dramatic rise in contract archaeology. The development of contract archaeology programs throughout the United States was caused by increased federal ding and resulted in a large quantity of archaeological fossil pollen studies, especially in the American Southwest. Unfortunately, most of these reports (both palynological and archaeological) were never published, and the data recovered lie buried in the offices of many federal and state agencies. Since there is no central clearing house for Palynological contract reports, for all practical purposes, the majority of these reports are not available as reference sources in most major libraries. Hall (1985} has provided some help by compiling a bibliography of southwestern palynology which includes many of the contract studies. VALIDITY OF POLLEN ANALYSIS Pollen analysis is a useful research technique in paleoenvironmental reconstruction and archaeology because of two important criteria: (1) preservation and (2) recognition. 50 HOLLOWAY & BRYANT Vol. 6, No. 1 Most pollen grains and spores produced by terrestrial plants have a chemically stable outer wall, called the exine which preserves well in many sediments because it is com- posed of a mixture of cellulose and a more durable compound called sporopollenin. Sporopollenin is composed of oxidative co-polymers of carotenoid and carotenoid esters (Shaw 1971) which are highly resistant to decay and have enabled ancient pollen and spores to remain preserved for millions of years. In addition to its preservation quality, extensive research has demonstrated that pollen grain characteristics and wall morphology are consistent within a plant species, but vary from the pollen morphology of other plant taxa. These differences permit the pollen of one genus, or species, to be recognized from types produced by other plant taxa. The utility of pollen analysis is based upon two major assumptions: (1) any particular fossil pollen assemblage is indicative of the original pollen contributing flora; and (2) the recontructed fossil vegetational communities can be used to infer paleoecologic and paleoclimatic parameters. The first assumption appears warranted subject to certain limitations. Plants produce varying amounts of pollen or spores which are dispersed and are carried by wind currents (anemophilous}, water currents (hydrophyllous), or by dif- ferent types of animals (zoophilous). Several researchers (Andersen, 1967; Erdtman, 1969) have estimated relative pollen productivity rates and attempted to correlate these with the mode of pollination, ie. anemophily, etc. After being dispersed, pollen falls to the earth’s surface in the form of a pollen rain which is influenced by a number of factors such as the rate of fall (Wright 1953); the method of transport (Janssen 1970); the effects of atmospheric and climatic conditions (Tauber 1965, 1967); and the effects of pollen recruitment into lake-type sediments (Bonny and Allen 1985; Davis et al. 1985). Once deposited, pollen and spores are subject to various types of degregation which selectively destroy certain pollen types and leave other types uneffected (Bryant and Holloway 1983). Numerous studies have also been conducted which address the presumed relation- ship between the recovered fossil pollen assemblage and the plant community from which it was derived (King and Kapp 1964, Andrews 1967; O‘Sullivan 1973; Adam and Mehringer 1975; Birks et al. 1975; Webb and McAndrews 1976; Webb et al. 1981; Heide and Bradshaw 1982; Holloway 1984). Recent numerical techniques, which are presented in a later section, have been very successful in elucidating this relationship. The suc- cessful application of these techniques notwithstanding, the lack of discrimination of exine morphological characters which would permit taxonomic identification of these pollen grains to the species level, has been a major obstacle in defining the precise rela- tionship between the paleovegetational community and the fossil pollen assemblage. The second major assumption used by palynologists involves the interpretation of the accumulated pollen data. In each study, the ultimate goal is to infer paleoecologic and ultimately, paleoclimatic conditions; yet to do this we just assume a valid relation- +> 1 +... j : 1 2 a ey & 1 a Mm ent ship between the communit p biotic and abiotic factors influence the distribution sh weeeatins plants. These pollen and abiotic factors influence the distribution of certain plants. These pollen and spore Pro ducing plants are restricted (today) to specific ecological habitats in which they surv1v¢, and in their optimum environments, are most abundant. Drawing on modern ecologi studies as analogues to paleovegetational communities, enables the palynologist t© effectively test this second assumption. From the information gained by modem ecological studies, the palynologist can beg!" to assign ecological parameters to the paleovegetational communities based on the presence of a few characteristic pollen types. This “indicator species” approach (Birks and Birks 1982) is usually successful enough to infer paleoenvironmental conditions. This procedure cannot rely solely on the modern ecological data as many other factors are involved in determining plant biogeographical distributions. As an alternative to the i ——— Summer 1986 JOURNAL OF ETHNOBIOLOGY 51 indicator species approach, Conolly (1961) stressed the importance of including all taxa present, rather than a few “indicator species”, prior to interpreting the pollen assemblages paleoecologically. This procedure may be cumbersome, especially with long temporal records consisting of many levels, but the numerical treatment of these data in terms of recognizing co-occurring and co-varying groups of taxa may provide the necessary information to demonstrate the validity of the relationship between the paleovegeta- tional community and the paleoenvironment. DISCUSSION During the past several decades, pollen analysis has become a standard procedure during many archaeological excavations. The major problem remaining is that much of the palynological and botanical data has not been completely synthesized with the archaeological data for the purpose of interpretation. Most often, palynological and other types of botanical data are included at the end of an archaeological report in appendix form. There may be an historical precedent for this, yet we suspect that a major cause may lie in the philosophical development of the two disciplines. Quaternary palynology, as an historically based descriptive science, has a long history of strict inductionist thought. Palynology, as an empirical science depends upon induc- tive reasoning which proceeds from one observation to the next in an attempt to provide generalities about the environment (Birks and Birks 1981). Edwards (1983) has sug- gested that this philosophical position may not be intellectually stimulating for some researchers. Although this may, or may not, be the case, the history and tradition of inductive reasoning in this field is strongly entrenched due, in large part, to the influence of historical geology on the field of palynology. Archaeological palynology is not only tied to both the historically based geological precepts of biostratigraphy and i ] tion, but also to those concepts inherent to the fields of anthropology and archaeology. During the past two decades, much of the emphasis in archaeology has been away from the descriptive techniques of the earlier ‘culture history” approach (Flannery 1968). thus, the “new” archaeology, which emerged in the late 1960s, concentrated upon analysis, synthesis, and hypothesis testing, emphasizing instead of descriptive work, a processual approach (Binford 1971, 1977; Watson et al. 1971; Morgan 1973, Schiffer 1981). This, however, was by no means a univer- sal shift in theoretical orientation, for much archaeological activity during this period was directed, as in palynology, toward data collection using primarily inductive approaches. But we do see among many archaeologists a shift towards incorporation of deductive principles within the archaeologists’ research design. As Binford {1983} notes, this does not imply the rejection of an inductive approach, but rather the integration of alternative research strategies in order to arrive at explanation. We see current archaeology as moving in a direction of scientifically testing explicitly stated hypotheses. Most palynologists, on the other hand, have not yet made this shift and thus continue in their collection of new data (Edwards 1983). This aspect, however, may not be all that unfortunate since many geographical areas are still lacking an adequate, comprehensive data base. For example, much of Texas (Bryant and Holloway 1985), California (Adams 1985), and the Great Plains (Baker and Waln 1985) are known palynologically only from a limited number of sites which in many cases lack adequate geological chronologies. A secondary problem in archaeological palynology has stemmed from the develop- ment of contract archaeology during the last several decades. Often there is too little time alloted for botanical analysis and interpretation, yet many federal and state regula- tions proscribe paleoenvironmental testing in the initial scope of work. Thus, because of time constraints, the vegetational and environmental data often are not effectively 52, HOLLOWAY & BRYANT Vol. 6, No. 1 incorporated into the archaeological analysis. With careful planning, however, this need not be the case. The paleoenvironmental data provided by palynology, even when presented in appendix form, can be incorporated within the interpretation of the archaeological data as in the case along the Yazoo River drainage system in central Mississippi (Thorne and Curry 1983). More of this type of integration is desperately needed in the field of archaeology. As a first step toward integration between palynological and archaeological data, palynologists need to utilize a rigorous scientific approach. As Edwards (1983) observed, only one paper (Garbett 1981) published in six major journals during 1981, dealt with testing an explicitly stated hypothesis and then discussed the level at which the hypothesis would be rejected. Although in many cases, the acquisition of new palynological data is necessary, in many geographical regions such as the Northeastem U.S. (Gaudreau and Webb 1985), the Great Lakes Region (Holloway and Bryant 1985), the Southeastern U.S. (Delcourt and Delcourt 1985), and the American Southwest (Hall 1985a, 1985b) palynological data are available to afford researchers the opportunity to test models and hypotheses of climatic and vegetational dynamics (Delcourt and Delcourt 1984). It is in this direction that palynology will have its most beneficial effect upon the understanding of the human impact on vegetation and the full utilization of the botanical and archaeological data. Without this, palynologically data will be forever relegated to the appendix. ANTHROPOGENIC STUDIES Human cultures modify the natural environment in which they live and thus often alter the ecological balance of the plant taxa which, in turn, are reflected in the local pollen rain. Learning to recognize ancient man’s alterations of the environment through analysis of the fossil pollen record is one of the important goals of the discipline of palynology. There are a number of studies which have focused upon anthropogenic factors as reflected by pollen analysis. Dyakowska (1958) was able to correlate two decreases of forest tree pollen in the fossil records of Poland to known historic events. Studies of cultivated plants, on the other hand, are often the best indicators of human plant modifica: tion (Behre 1981) and have generated the majority of archaeological interest (Martin 1963; Hill and Hevly 1968; Leroi-Gourhan 1969; Lytle-Webb 1978). One of the biggest problems with these types of studies has been the lack of adequate pollen morphological characters with which to separate the cultivated from the wild forms of the plant (Behre 1981). Much effort has been spent on the analysis and identification of these crop plant remains at archaeological sites and these studies d that the best evidence of prehistoric cultivation of plants still comes from plant macrofossils rather than from the fossil pollen record. in question (Behre 1981). Because of the differences between prehistoric and modem agricultural practices, modern analogues of pollen recovered from agricultural fields are not necessarily applicable to these types of studies. ee —— Summer 1986 JOURNAL OF ETHNOBIOLOGY 53 Additionally, pollen analysis is useful in the investigation of economic land use criteria. Edwards (1982) has discussed several methodologies for investigating the economic use of ancient ecotone regions between forest and “prairie” areas. However, as Edwards (1982) cautions, using anthropogenic interpretations of vegetational change during the Early Postglacial times often cannot be documented with any degree of certainty. In fact, during these periods, a natural rather than a cultural explanation is more likely to be correct. The fossil pollen analysis of ancient floor surfaces in architectural dwellings is yet another way that palynologists can help the archaeologist to understand past environ- mental and cultural phenomena. One of the first applications of this technique was demonstrated by Schoenwetter (1962) when he used soil scraping from peublo dwelling in eastern Arizona to date the periods of site occupation and infer past climatic condi- tions. Hevly (1964) attempted a similar study of ancient floor surfaces in abandoned pueblo wellings located in the Colorado Plateau area of the American Southwest. Hill and Hevly (1968) later attempted to determine room function through the application of fossil pollen analysis in their investigation of the sediments recovered from Broken K Pueblo, Arizona. Unfortunately, these earlier studies did not employ multiple pinch sampling as later sug- gested by Cully (1979). In addition to occupational floor levels found in structures, agricul- tural fields are also being examined. Dimbleby (1985) discussed the fossil pollen evidence of Goodburn, at Winterton, Hamberside, England, in which he used palynology to identify possible associated agricultural fields. Wiseman (1983) likewise has recovered fossil pollen data from possible agricultural fields in Lowland Central America. Under ideal circumstances pollen analyses from archaeological sites can also be used to determine a wide range of other cultural phenomena including: prehistoric diets, possible graveside rituals, subsistence patterns, use of native or cultivated plants, use of certain types of artifacts (e.g. grinding stones, pottery, and baskets), probable use of areas within architectural structures, intersite and intrasite dating, and preexcavation recognition of potentially important archaeological sites. Pollen analyses of archaeological sites are now serving as a useful technique for deter- mining the probable function of certain types of baskets, ceramic vessels, bedrock mortars, and milling stones. Experiments have demonstrated (Bohrer 1968) that pollen can be inadvetently included during the gathering and later storage of certain types of foods such as Zea (maize), Typha (cattail), Cleome (beeweed), Chenopodium (goosefoot), and Amaranthus (amaranth). Pollen from these plants often adheres to the inside sur- faces of baskets and ceramic vessels. Later, when recovered by the archaeologist, these same vessels can be analyzed for their fossil pollen contents and the resulting data can be used to determine probable functional use. Others have noted that the pollen con- tents on the surfaces of grinding stones often are reflective of the plants utilized (Hevly 1964; Bryant and Morris n.d.). Under certain circumstances, bedrock mortars can also be examined for their fossil pollen contents. Careful removal of dirt and the subsequent analysis of the fossil pollen trapped at the bottom of these mortars sometimes indicates which plant foods were ground in these mortars. On the other hand, an unfavorable aspect of this type of analysis is that bedrock mortars also may contain modern pollen that was deposited after the Mortars were actually in use. Therefore, it is rare that a palynologist is able to deter- mine with certainty the precise food plants which may have been originally processed in bedrock mortars. Basketry and other woven artifacts sometimes provides important clues to the types of plant material which they carried. Often the dirt trapped between the weaves con- tains fossil pollen grains that became embedded while the basket was still in use. Like the pollen recovered from the inside surfaces of ceramic vessels, these fossil pollen grains can be used to determine which plant materials were collected and/or stored in basket 54 HOLLOWAY & BRYANT Vol. 6, No. 1 containers. Many ceramic vessels made during aboriginal times were used to store food, as cooking pots, and as food containers for meals. If the foods in the vessels contained pollen, then often some of that pollen became trapped along the inside surfaces of these vessels during their use and can be utilized as clues to prehistoric uses of the vessels. Unfortunately, the archaeologist is not always able to test this sampling technique at archaeological sites since many ceramic vessels are recovered only as broken sherds. Hevly (1970), for example, recovered fossil pollen from a mid-Pueblo III age, sealed, ceramic storage jar excavated from a site in northern Arizona. His analysis revealed very high percentages of fossil pine pollen and fungal spores, which he believed represented a post- depositional phenomenon rather than the jar’s original juse. In a related study by Bryant and Morris (n.d.) the inside bottom portions of 42 complete ceramic vessels recovered from rooms and burials at the Antelope House Pueblo were scraped carefully to recover fossil pollen that was deposited when the vessels were in use. The fossil pollen spec- trum recovered from the bottommost vessel scrapings were statistically compared with the fossil pollen spectrum in the dirt matrix of each vessel. In cases where chi-square tests revealed that the two pollen spectra from the same vessel were of different origin, Bryant and Morris (n.d.) were able to assign probable functional use categories to over 30 corrugated, wide-mouthed vessels dating from the Late Pueblo III period as well as one Late Pueblo III plainware open bowl. Pollen analysis of soils recovered directly underneath a Neanderthal burial (Leroi- Gourhan 1975) at Shanidar Cave, Iraq, revealed unusually high percentages of pollen from tiny, insect-pollinated alpine flowers. Because of the low pollen productivity of zoophilous plants, it was assumed that the high percentages of those pollen types in the burial soils must have resulted from cultural, not natural introduction. Therefore, Leroi-Gorhan (1975) concluded that a large number of small, alpine flowers must have been collected from the nearby hillsides and then carefully placed in the Neanderthal grave at the time of internment. The significance of this research has had a profound effect upon our understanding of Neanderthal’s cultural activities and the possibility that they were the first group to adopt a basic philosophy about religion and the afterlife. At Broken K Pueblo in Arizona, Hill and Hevly (1968) noted the probable ceremonial use of Sphaeralcea (mallow) and pine pollen in an infant’s burial. In a nearby region of Arizona, Bryant and Morris (n.d.) noted what they suspected was a similar use of maize pollen as part of a graveside ritual at Antelope House Pueblo during the internment of a young child. These two studies, each from different pueblo sites, suggest that in some Southwestern groups flowers and pollen may have been thrown into the graves of the dead as some type of mortuary ritual accompanying internment. More importantly, at each of these burial sites, it was the fossil pollen data that helped establish new cultural insights into our understanding of prehistoric rites. The pollen analysis of preserved human coprolites is one of the most useful methods for determining prehistoric diet and certain types of cultural patterns. Pollen data recovered from human coprolites can, under ideal circumstances yield information 4s to: (1) the use and source of economic and background pollen types; (2) the seasonality of site usage; (3) diet; and (4) certain aspects concerning paleoenvironmental conditions (Bryant 1974). However, like other forms of pollen data, the best and most reliable records come from studies at a given site in which the data base consists of many coprolites all dating to the same time period, rather than from only a few or even a single specimen representing one stratum or one time period. Single coprolite specimens, like a single lithic artifact, may or may not be an accurate reflection of the entire time period represented by the stratum. Pollen studies of coprolites have shown that not all of the pollen found in coprolites reflect aspects of human diet. Within each human coprolite there are two distinct groups of pollen: economic pollen and background pollen. Economic pollen, as a specific category, TT —— a. —e Summer 1986 JOURNAL OF ETHNOBIOLOGY 55 is recognized by palynologists to include those pollen types that were probably ingested directly as part of the diet. Most of these economic pollen types are easily recognized because they come from plant species having pollen that is rarely, if ever, found as part of the normal atmospheric pollen rain. Economic pollen grains from zoophilous plant such as Cleome, Yucca (yucca), Opuntia (cactus), Agave (agave), Dasylirion (sotoll), Cucurbita (squash), and Prosopis (mesquite) are transported from flower to flower by insects and thus rarely become airborne in the atmosphere as do the wind pollinated background grains of Ambrosia (ragweed), Pinus (pine), or Quercus (oak). Therefore, when significant quantities of zoophilous economic pollen types are found in human coprolite specimens, they are generally interpreted to reflect the direct consumption of the plant, flowers, or honey made from those plants rather than representing the accidental inges- tion of these pollen grains from atmospheric sources. Some wind pollinated types of pollen such as Chenopodium, Zea mays, Iva (marsh elder), Ephedra (mormon tea), Typha and Helianthus (sunflower) may or may not be included as economic types depending upon the circumstances under which they are found. Since these types come from edible plants whose pollen often adheres to the collected seeds of these plants or to other plant parts which are known ethnographically to have been utilized, the presence of these pollen types can often be associated with economic use rather than representing deposition from strictly natural pollen rain sources. The preserved pollen in human coprolites sometimes may offer clues as to the probable seasonality of site occupation and insights concerning the paleoenvironment at the time when the coprolite was produced. When human coprolites contain high percentages of economic pollen from one or more plants that generally bloom during the spring and/or summer, then that information can be used to offer tentative evidence for seasonal usage of the site. Ethnobotanical reports of many subsistence-level proto- historic cultures in North America have noted that in most instances, flowers were eaten fresh and that honey was rarely, if ever, found or eaten (Barrows 1900; Castettler and Bell 1941; Curtin 1949; Newberry 1887; Palmer 1887) thereby confirming that flower pollen can generally be attributed to lity when recovered in coprolites. In a similar way, the presence of coprolitic pollen from plants which no longer inhabit the region may suggest that environmental changes have occurred which caused the extinction or out migration of certain plant taxa. From this somewhat cursory overview, it is evident that palynological investiga: tions of anthropogenic issues require asking specific types of questions dealing with both cultural and botanical science and the understanding of both anthropological and botanical research methodologies (King 1985). Furthermore, we believe that one of the major stumbling blocks that discourages the incorporation of palynological data into archaeo- logical research has been the use of techniques which are generally employed in strati- graphic palynology to answer non-stratigraphic types of culturally important questions. Many times, this type of palynological approach is just not applicable to solving anthropogenic issues. In attempting to formulate testable hypotheses concerning anthropological questions, palynologists must be prepared to discard precepts which are almost inviolate in biostratigraphy. For example, in some cases valuable information can be obtained solely by the use of presence/absence criteria in archaeological palynology. In such instances, these arguments may provide a better basis for data explanation and synthesis within the framework of archaeological concepts. DATA ANLYSIS AND STATISTICS Data analysis and interpretation are two aspects that have become rapidly more sophisticated during the past two decades. More than 60 years ago, von Post {Davis and Faegri 1967) based his initial pollen analysis on calculated, relative pollen frequencies. 56 HOLLOWAY & BRYANT Vol. 6, No. 1 Although in many cases this type of statistic is still useful, it lacks the needed precision for most of today’s studies. Throughout the first half of the twentieth century, relative pollen frequencies were used exclusviely by palynologists both in the reconstruction of paleoenvironmental conditions and in the interpretation of palynological data from archaeological sites. After the introduction of radiocarbon dating and the availability of tightly controlled temporal chronologies, Benninghoff (1962), and later Davis (1963, 1966, 1969), Jorgensen (1967), and Matthews (1969) helped to develop new methods for the calculation of fossil pollen ratios through the use of pollen concentration and pollen influx values. Although pollen influx data provide needed additional information used in paleoenvironmental reconstruc- tions, the lack of precise stratigraphic and temporal control within an archaeological site often precludes the use of pollen influx values in these sediments. Therefore, in order to provide more meaningful results from the analysis of pollen data recovered from archaeological sites, palynologists have begun to use pollen concentration values which can be expressed in terms of the number of fossil grains per unit weight or volume, and then those ratios can be compared between levels of the same site or between similar cultural strata from different sites. This type of comparison has proven useful for demonstrating smilarities and/or differences in the fossil pollen record. Although con- centration values can be computed using either sample weight or volume, we have found that the volume method is more reliable for comparing both intrasite and intersite pollen variability (Bryant and Holloway 1983). While multivariate statistical analysis of palynological data from archaeological sites can provide a wealth of information, the major utility of statistical analyses has been realized within the framework of biostratigraphic and paleoecological problems. Initially, much of the work was, and still is, concerned with statistically analyzing modern pollen assemblages and using these data to infer paleoecological changes. Initially, multi- variate statistics were utilized quite effectively to quantify the relationship between surface pollen and vegetation (Webb and Bryson 1972; O’Sullivan and Riley 1974; Birks et al. 1975; Webb and McAndrews 1976; Bernabo and Webb 1977; among others). Much of this research was aimed at obtaining modern analogues to fossil pollen assemblages. However, as many of these authors noted, some assemblages, especially those from late-glacial/Holocene times, lacked modern analogues and thus were outside the purvue of these techniques. Recently, Liu and Lam (1985) using discriminat analysis were able to statistically quantify these anomalous assemblages. In an attempt to reconcile surface pollen spectra with the extant vegetation, Davis and Goodlett (1960), earlier had devised the concept of R-values. Although a valid attempt at solving these problems, the model was never universally accepted primarily because, as noted by Livingstone (1968, 1969), of the model’s failure to accurately reflect over-representation of arboreal pollen types. Recently, Parsons and Prentice (1981:127) demonstrated the utility of Davis and Goodlett’s (1960) model “subject to certain caveats.” Parsons and Prentice (1981) developed three mathematical models to account for the observed discrepancies. These mathematical models involved the use of principal com- ponents analysis and regression equations which provide a good basis for inferring vegeta tional compositon from surface pollen data. Regression analysis has likewise been utilized recently (Webb et al. 1981; Heide and Bradshaw 1982) to estimate directly the plant abundance of vegetational stands based on their surface pollen rain. Delcourt and Delcourt (1984) investigated the distribution of 24 major tree taxa and produced paired maps showing the percentage of growing stoc volume (isophyte maps) and arboreal pollen percentages (isopool maps}. These data were produced using a geometric-mean linear regression and have proved valuable in the quantitative reconstruction of vegetational change in eastern North America. Summer 1986 JOURNAL OF ETHNOBIOLOGY a7 These attempts to correlate vegetational composition with surface pollen rain have been supported largely by the impetus of research projects such as CLIMAP (1976, 1981) which have utilized paleoenvironmental data to reconstruct ancient landscapes. Almost from the beginning of these studies, palynologists recognized the importance of ulti- mately relating the pollen/plant relationships to the interpretation of climatic data (Webb and Bryson 1972; Webb and Clark 1977; Kay 1979; Andrews et al. 1980; Webb 1980; Heusser and Streeter 1980). More recently, Howe and Webb (1983) have effectively discussed the methodologies employed in calibrating palynological data in terms of quantitatively derived climate estimates. These procedures can be combined with cer- in mapping techniques (Webb 1983) to provide measurements of changes in paleovegeta- tional communities. These types of paleoenvironmental data, although derived from differing sources, have been utilized by Delcourt and Delcourt for the production of statistical models reflecting biotic responses to climatic patterns occurring throughout the Holocene. Drawing upon the established data base, Delcourt and Delcourt (1983) have additionally provided a model designed to test vegetational responses to climate changes occurring at various periods within the Holocene. The model is exactly what is needed as it pro- vides mechanisms to predict future modifications of the biota in response to those climate changes induced by man (i.e., the estimated global warming trend caused by increased CO concentrations in the atmosphere). These types of quantitative models serve as generalized hypotheses for empirical testing. These techniques and approaches cannot, at this stage, be applied universally. The quantitative measurements have been derived primarily from sites located within the Upper Midwest, or southeastern U.S. The work in these geographical areas has been conducted for several decades. Not only are a large number of palynological sites available for study (Holloway and Bryant 1985; Delcourt and Delcourt 1985} from these areas, but extremely large numbers of surface samples and the corresponding vegetational data have been collected. It is only in those regions where a large accumulated data base is available that multivariate statistical analyses of data can provide both the models and the empirical tests for vegetational responses to climate. The ultimate goal of paleoenvironmental reconstruction is, of course, to draw on as much data as possible and then use those data for determining comparisons between adjacent geographical areas. Gordon and Birks (1974) attempted to statistically delimit Pollen zones (i.e. a series of adjacent levels with similar pollen assemblages] and then use those zones for chronological purposes within a given geographical region. The advantage of using such pollen zones is that they permit comparisons of two or more pollen records from different locations within a given geographical locale. The main disadvantage of using pollen zones is that their selection must be unbiased which argues for a numerical approach rather than an intuitive approach to the problem. Numerous statistical techniques are presently available for numerical analysis and zonation of pollen data. Birks and Gordon (1985) recently have summarized many of €se techniques and have provided computer programs for general dissemination. While these analytical techniques are readily available, few palynologists routinely use them. t is important to remember is that these numerical techniques were not designed to replace the interpretation of the palynologist, but rather to insure that the observed changes are real (not biased), and that a significant amount of variation is present before interpreting drastic changes. ; : Although as a general rule paleoenvironmental reconstructions from archaeological Sites pose severe conceptual restrictions due to the biased nature of the data base, a numerical approach has been demonstrated which may alleviate this problem. Fall et al. (1981) have utilized principal components analysis to extract paleoenvironmental data from Antelope House in Arizona. Based on over 100 pollen samples, the numerical method 58 HOLLOWAY & BRYANT Vol. 6, No. 1 reflected changes in vegetation patterns through time which were correlated with environmental indices reflecting temperature and moisture. These were likewise inter- preted in light of known cultural activity patterns and population movements (Fall et al. 1981). As Fall et al. (1981) correctly observed, in many archaeological sites, especially in the American Southwest, most palynologists have difficulty in distinguishing which pollen represents the naturally occurring regional pollen rain and which types reflect the cultural activities of man. In their study, the authors, using multi-variate statistical techniques, were able to successfully separate these two components. Once assured of the composition of the regional pollen rain, they demonstrated the effectiveness of these numerical techniques for the interpretation of paleoenvironmental conditions and in some cases, the human response to these conditions. SUMMARY During the past fifty years the fields of archaeology and palynology have begun to establish a working relationship in which the scientists of each discipline have tried what we mean. During the recent excavation and fossil pollen study of an archaeological site, 4 palynologist presented a wide array of interpretations as to what the prehistoric inhabti- tants may have been using as food, the types of plant materials they carried into the site and later used in the making of matting, clothing, and other items of their material culture. It was a carefully researched study and presented the types of information that most archaeologists would be pleased to obtain from fossil pollen studies of their site sediments. However, at this particular site, the sediment samples which had been examined and interpreted by the palynologist all came from culturally sterile strata which contained no artifacts or any other evidence suggesting that the site was even occupied by prehistoric groups during the deposition of those strata. . Another area in which communications between palynologists and archaeologists have been lacking is in the application and use of statistics. During the past several decades archaeologists have been quick to apply the use of statistical methods to studies of the cultural remains from their sites; however, until very recently most fossil pollen studies f archaeological sediments have not made effective use of new methodology in statistics. Hopefully, in the decades to come we will see an increased emphasis placed upon the use of statistical methods such as multivariate statistical-, principal component, and regression-analysis when working with the pollen data from arcl gical sites. In the past these statistical methods have been used in the analysis of pollen data from a few archaeological sites and have proved to be an effective method of interpreting inform tion. These methods of data analysis and statistical approaches represent some of the currently utilized techniques in studies of archaeological palynology. However, there are many other alternative approaches from the wider field of palynology which cur Summer 1986 JOURNAL OF ETHNOBIOLOGY a9 rently have limited application. Perhaps, as archaeological palynologists become more comfortable with the use of statistics as they apply directly to problems in archaeological palynology, then new avenues of data integration will develop which will provide a wide range of new information on which to base interpretations. Also, further studies are needed in the area of pollen degradation and in the better understanding of pollen sources which become trapped in archaeological sites. Unlike most peat bogs, lakes, and other terrestrial deposits where the activities of mankind are not a factor, the pollen in archaeological sites comes from both natural sources and from the activities of cultural groups. As shown earlier, there are many ways in which the activities of cultural groups can alter the natural vegetation through burning, agricultural or pastoral practices, through the selective use of certain firewoods or plants while others are not disturbed, and through the collection and use of certain plants to make their shelters or clothing. Degradation of fossil pollen in archaeological deposits is also an important issue. It is well to remember there are many factors which will lead to the destruction of pollen. In addition, experiments have shown that not all types of pollen grains degrade at the same rate, therefore, selective destruction of fossil pollen types can occur. That, in turn, will create a preserved pollen record that may be quite different from the actual pollen rain that was originally deposited. Knowledge of these differences are critical to the preparation of meaningful pollen analyses pertaining to archaeological sediments. Finally, one aspect which we hope will be corrected in the years to come is the wide variety and range of archaeological fossil pollen records which currently are being buried in various ‘‘contract-type” reports. What makes this problem especially troublesome is that all too often they are printed in very limited numbers and are not widely distributed. If this procedure continues during the next decades, it will contribute to the uninten- tional duplication of fossil pollen research and will hinder researchers in a given area from being able to utilize the full range of fossil pollen data which may be available only in numerous contract-type reports. What is desperately needed in the field of archaeo- logical palynology is a national or regional clearing house for these types of reports. In this way, the results of completed research will be available to the entire field. union of archaeology and palynology has been accomplished and now most archaeologists consider fossil pollen data from their sites as being an important aspect for them to consider. Furthermore, during the past several decades more and more palynologists with strong backgrounds in both botany and archaeology are entering the profession and are now in a position to communicate effectively with archaeologists. In addition, more and more pollen analyses are now becoming available from archaeo- logical deposits in a wide variety of geographical areas. In other words, the data base is expanding rapidly. Soon, it is hoped that the archaeological palynologist will be able to turn his or her attention more toward research involving the testing of specific hypotheses, and away from the simple data collection. LITERATURE CITED ADAM, D. P. 1985. Quaternary pollen Jutland (Denmark) Rev. Paleobot. records from California. Pp. 125-141 Paly. 3:267-275. in Pollen records of late-Quatemary ANDERSON, R. Y. 1955. Pollen analysis, North American sediments. (Bryant, a research tool for the study of cave i 84-85. V. M. Jr., Holloway, R. G., eds.) deposits. Amer. Ant. 21:8 A.A.S.P., pp. esas ANDREWS, J. T. and H. F. DIAZ. 1981. ANDERSEN, §S. T. 1967. Tree pollen rain Eigenvector analysis of reconstructed in a mixed deciduous forest in South Holocene July temperature depar- 60 HOLLOWAY & BRYANT Vol. 6, No. 1 LITERATURE CITED (continued) tures over Northern Canada. Quat. Res. 16:373-389. AUER, V. 1927. Botany of the interglacial peat beds of Moose River Basin. Geol. Sur. Can. Sum. Rpt. for 1926, Part C:45-47. BAKER, R. G. and K. WALN. 1985. Quaternary pollen records from the Great Plains and central United States. Pp. 191-205 in Pollen records of late-Quaternary North American sediments (Bryant, V. M. Jr. and Holloway, R. G., eds.) A.A.S.P., s. BARKLEY, F. A. 1934. The statistical theory of pollen analysis. Ecol. 15: 83-289. BARROWS, D. P. 1900. The ethnobotany of the Cahuilla Indians of Southern California. Ph.D. dissert., Univ. Chicago. BEHRE, K. E. 1981. The interpretation of anthropogenic indicators in pollen diagrams. Pollen et Spores 23:225- 245. BENNINGHOFF, W. S. 1962. Calculation of pollen and spore density in sedi- ments by addition of exotic pollen in known quantities. Pollen et Spores 4:332. BERNABO, J. C. and T. WEBB III. 1977. Cc ing patterns in the Holocene pollen record of northeastern North America: a mapped summary. Quat. Res. 8:64-96. BINFORD, L. R. 1971. Model building paradigms and the current state of paleolithic research. an archaeo- logical perspective (Binford, L. R., ed.), Academic Press, New York. 1977. For theory build- ing in archaeology. Academic Press, New York. 419 p. 1983. Working at Archaeology. Academic Press, Orlando, Florida. 489 p. BIRKS, H. H., M. C. WHITESIDE, D. M. STARK and R. C. BRIGHT. 1976. Recent paleolimnology of three lakes in Minnesota. Quat. Res. 6:249-272. BIRKS, H. J. B. and H. H. BIRKS. 1982. Quaternary palaeoecology. Univ. Park Press, Baltimore. 289 p. BIRKS, H. J. B., T. WEBB III and A. A. BERTI. 1975. Numerical analysis of pollen samples from central Canada: a comparison of methods. Rev. Paleobot. Paly. 20:133-169. BLYTT, A. 1876. Essay on the immigra- tion of the Norwegian flora during the alternating rainy and dry periods. Christiania, Nyt. Mag. f. Natur- vidensk 21. BOHRER, V. L. 1968. Paleoecology of an archaeological site near Snowflake, Arizona. D. dissert., Univ. Arizona, Tucso BONNY ALLEN, P. V. "1984, Pollen recruitment to the sediments of an enclosed lake in Shropshire, England. Pp. 231-261 in Lake sediments and environmental history (Haworth, E. Y. and Lund, W. G., eds.) Univ Min- nesota Press. BROOKES, D. and K. W. THOMAS. 1967. The distribution of pollen grains n microscope slides. Part 1. The non- randomness of the distribution. Pollen et Spores 3:621-629 BRYANT, V. M. JR. 1974. The role of coprolite analysis in archaeology. TAS Bull. 45:1-28. . 1974. Prehistoric diet in southwest Texas: the coprolite evidence Amer. Ant. 39:407-420. BRYANT, V. M. and R. G. HOLLOWAY. 1983. The role of palynology in archaeology. In Advances in archaeo- logical method and theory (Schiffer, M. D., ed.). 6:191-224, Academic Press, New York. . 1985. A late-Quater- nary paleoenvironmental record O Texas: an overview of the pollen evi dence. Pp. 39-71 in Pollen records of late-Quaternary North American sediments. (Bryant, V. M. Jr. am Holloway, R. G., eds.). A-A-S.P. as. BRYANT, V. M. JR. and D. P. MORRIS. Summer 1986 JOURNAL OF ETHNOBIOLOGY 61 LITERATURE CITED (continued) (n.d.) Uses of ceramic vessels from Antelope House: the pollen evidence. Unpubl. manu. on file at the National Park Service offices, Tucson, Arizona. BRYANT, V. M. JR. and G. DEAN- WILLIAMS. 1975. The coprolites of man. Sci. Amer. 232:100-109. BURRICHTER, E. 1969. Das Zwillbro- ecker Venn Westmunsterland in oor- und vegetationskundlicher Sicht Abhandl. Landesmuseum Naturkunde Munster Jg. 31, H.1, -60. CASTETTLER, E. F. and W. H. BELL. 1942. Pima and Papago Indian Agriculture. Univ. New Mexico Press, Albuquerque. CLIMAP PROJ. MEMBERS. 1976. The surface of the ice-age earth. Science 191:1131-1137. . 1981. Seasonal recon- structions of the earth’s surface at the last glacial maximum. Geol. Soc. Amer. Map and Chart Ser. 36. CLISBY, K. H. and P. B. SEARS. 1956. San Augustine plains-Pleistocene oo changes. Science 124:537- comers A. P. 1961. Some climatic and edaphic indications from the late- glacial flora. Proc. Linnaean Society London 172:56-62. COX, D. R. and D. M. LEWIS. 1965. Pollen studies in Crusoe lake area of prehistoric indian occupation. New York State Mus. Bull. 397:1-29. CULLY, A. C. 1979. Some aspects of pollen analysis in relation to archae- ology. The Kiva 44:95-100. CURTIN, L. S. N. 1949. By the Prophet of the ea San Vincente Found. Pub. Sante Fe. DAVIS, M. : 1963. On the theory of Sm — Amer. J. Sci. 261: 297-9 1966. Determination of absolute > pollen frequency. Ecol. 47:310-311 —_—__________. 1969. Palynology and environmental history during the aa period. Amer. Sci. 57:317- fai M. B. and K. FAEGRI. 1967. Forest tree pollen in south Swedish peat bog deposits (translation ana of the 1916 presentation by L. von ane oe et Spores 9:375- 401. DAVIS, J. C. GOODLETT. ta’ eran of the present vegetation with pollen-spectra in sur- face samples from Brownington Pond Vermont. Ecol. 41:346-357. DAVIS, M. B., R. E. MOELLER and J. FORD. 1984. Sediment focusing and pollen influx. In Lake sediments and environmental history. (Haworth, E. Y. and Lund, J. W. G., eds.) Univ. Minnesota Press, pp. 261-195. DELCOURT, H. BR. and P. A, DEL- COURT. 1985. Quaternary paly- nology and vegetational history of the southeastern United States. Pp. 1-39 in Pollen records of late-Quaternary North American sediments. (Bryant, V. M. Jr. and Holloway, R. G., eds.) A.A.S.P. Dallas. DELCOURT, P. A. and H. R. DEL- COURT. 1984. Late Quaternary pale- oclimates and biotic responses in Palaeogeogr. Palaeoclim. Palaeoecol. 48:263-284. DELCOURT, ?. A. H. EB. DEL- COURT and T. WEBB III. 1984. Atlas of mapped distributins of dominance and modern pollen percentages for important tree taxa of eastern North America. A.A.S.P. Cont. Ser. Num. 14, 131 p. Dallas. DIMBLEBY, G. W. 1960. Pollen analysis of a mesolithic site at addington, Kent. Grana Paly. 4:140-148. 1963. Pollen analyses from two Cornish barrows. J. Roy. Inst. Comwall, new series 4:364-375. 1985. The palynology of archaeological sites. Academic Press, New York. 176 p. DONNER, J. J. 1962. On the postglacial 62 HOLLOWAY & BRYANT Vol. 6, No. 1 LITERATURE CITED (continued) history of the Grompian highlands of Scotland. Soc. Sci. Fen 24:6. DURNO, S. E. 1965. Pollen analytical evidence of ‘Landnam’ from two Scottish sites. Trans. Bot. Soc. Edin- burg. 40:13-19. DKAKOWSKA, J. 1958. An example of the influence of man on the pollen diagram. Verhandlugen der vierten internationalen Tegung der Quater- botaniker in der Schwiez vom 6-16 August. Veroffenlichungen des Geo- botanischen institues 34:39-41. EDWARDS, K. V. 1982. Man, space, and woodland edge- speculations on the detection and interpretation of human impact in pollen profiles. Archaeological aspects of Woodland ecology. (Bell, M. and Limbrey, S., eds.). BAR international series 146:5-22. 1983. Quaternary palynology: consideration of a dis- ’ cipline. Prog. in Phys. Geogr. TOLES-195. RDTMAN, G. 1960. The acetolysis method: a revised description. Svensk. Bot. Tdsk. Bd 54:11.4. 1969. Handbook of Palynology. Munksgaard. FAEGRI, K. 1944. On the introduction of agriculture in western Norway. Geol. foren. Stokh. forh. 66:449-462. 1981. Some pages of the history of pollen analysis. Striae 14:42-47. FALL, P. L., G. KELSO and V. MARK- GRAF. 1981. Paleoenvironmental reconstruction at Canyon del Muer- to, Arizona, based on principal- component analysis. J. Arch. Sci. 8:297-307. FLANNERY, K. 1968. Culture history vs. culture process. Sci. Amer. 217: 119-122. GARBETT, G. G. 1981. The elm decline: the depletion of a resource. New Phytol. 88:573-585. GAUDREAU, D. C., T. WEBB III. 1985. Late-Quaternary pollen stratigraphy and isochrone maps for the north- eastern United States. Pp. 245-281 in Pollen records of late-Quaternary North American sediments. (Bryant, V. M. Jr. and Holloway, R. G., eds.) A.A.S.P., Dallas. GODWIN, H. 1944. Neolithic forest clearance. Nature 153:511. . 1944. Age and origin of the Breckland Heaths at East Anglia. Nature 154:6. _ 1956. The History of the British Flora. Cambridge Univ. Press 1967. Pollen analytic evidence for the cultivation of Can- nabis in England. Rev. Paleobot. Paly. 4:71-80. HALL, S. A. 1985a. Quaternary pollen analysis and vegetational history of the southwest. Pp. 95-125 in Pollen American sediments. (Bryant, V. M. Jr. and Holloway, R. G., eds] A.A.S.P., Dallas. _ 1985b. Bibliography of Quaternary palynology in Arizona, Colorado, New Mexico, and Utah. Pp. 407-427 in Pollen records of late- Quaternary North American sedi ments. (Bryant, V. M. Jr. and Holloway, R. G., eds.) A.A.S.P., Dall as. HEIDE, K. M. and R. BRADSHAW. 1982. The pollen-tree relationship within forests of Wisconsin and Uppét Michigan, USA Rev. Paleobot. Paly. 36:1-23. . HEVLY, R. H. 1964. Pollen analysis of Quaternary archaeological and lacus: trine sediments from the Colorado Plateau. Unpubl. Ph.D. dissert., Univ: Arizona, Tucson. . 1970. Botanical studies of sealed storage jar cached néat Grand Falls, Arizona. Plateau 42:150- 155. i HILL, J. N. and R. H. HEVLY. 196°. Pollen at Broken K pueblo: some new Summer 1986 JOURNAL OF ETHNOBIOLOGY 63 LITERATURE CITED (continued} interpretations. Amer. Ant. 33:200- 210. HOLLOWAY, R. G. 1981. Preservation and experimental diagenesis of the pollen exine. Unpubl. Ph.D. dissert., Texas A&M Univ., 317 p. 1984. Analysis of surface pollen spectra from Alberta, Canada by polar ordination. Northw. Sci. 58:18-28. HOLLOWAY, R. G. and V. M. BRYANT, JR. 1985. Late-Quaternary pollen records and vegetational history of Pollen records of late-Quaternary North American sediments. (Bryant, V. M. Jr. and Holloway, R. G., eds.) A.A.S.P., Dallas. HOWE, S. and T. WEBB III. 1983. Cali- brating pollen data in climatic terms: improving the method. Quat. Sci. Rev. 2:17-51. HYDE, H. A. and D. A. WILLIAMS. 1944. Right Word. Pollen analysis cir- cular 8:6. IVERSEN, J. 1941. Landnam i Danmarks stenalder. Dan. Geol. Unders. IV Rk. 66. 68 p. . 1949. The influence of prehistoric man on vegetation, Dan. geol. unders. 4 Rk. 3, 6. 25 p. JANSSEN, C. R. 1970. Problems in the recognition of plant communities in pollen diagrams. Vegetatio 20:187- 195. JESSEN, K. 1935. Archaeological dating in the history of North Jutland’s vege- tation. Acta Arch. 5:185-214. JORGENSEN, S. 1967. A method of absolute pollen counting. New Phyt. 66:489-493, KAY, P. A. 1979. Multivariate statistical estimates of Holocene vegetation and climate change, Forest-tundra transi- tion zone, NWT, Canada. Quat. Res. 11:125-140. KING, J. E. 1985. Palynological appli- Cations to archaeology: an overview. logical Geology. (Rapp, G. Jr. and Gifford, J. D., eds.). Yale Univ. Press, pp. 135- KING, J. E. and E. H. ‘LINDSAY. 1976. Late Quaternary biotic records from (Wood, W. R. and McMillan, R. B., eds.). Prehistoric man and his en- vironments, a case study in the Ozark highlands. Academic Press, New York. LEROI-GOURHAN, A. 1969. Pollen grains of Gramineae and Cerealia from Shanidar and Zawi Cheni. The Domestication and Exploitation of Plants and Animals. (Ucko, P. J. and Dimbleby, G. W., eds.}. pp. 143-148. 1975. The flowers found with Shanidar IV, a Neander- thal burial in Iraq. Science 190:562- 564. LIU, K. B. and N. S. N. LAM. 1985. Paleo- vegetational reconstruction based on modern and fossil data: an appli- cation of discriminant analysis. Ann. Assoc. Amer. Geogr. 75:115-130. IVINGSTONE, D. A. 1968. Some inter- stadial and postglacial pollen dia- grams from eastern Canada. Ecol. Mono. 38:87-125. . 1969. Communities in the past. Pp. 83-104 in Essays in plant age and ecol. (Greenridge, K. .). Nova Scotia Museum, Halifax LYTLE-WEBB, J. 1978. Pollen analysis in southwestern archaeology. Dis- covering past behavior. Pp. 13-28 in Experiments in the archaeology of the southwest. (Grebinger, P., ed.}. Gordon and Breach Press. MARTIN, P. S. 1963. The last ten thousand years. Univ. Arizona Press, Tucson. MATTHEWS, J. 1969. The assessment of a method for the determination of absolute pollen frequencies. New Phytol. 68:161-166. MCMILLAN, R. B. and W. R. WOOD. 1976. A summary of environmental and cultural change in the western 64 HOLLOWAY & BRYANT Vol. 6, No. 1 LITERATURE CITED (continued) Missouri Ozarks. Pp. 235-241 in Prehistoric man and his environ- ments. (Wood, W. r. and McMillan, R. B., eds.). Academic Press, New York. MITCHELL, G. F. 1951. Studies in Irish Quaternary deposits No. 7. Proc. Royal Irish Acad. 53 B: 14. 1956. Post boreal pol- len diagrams from Irish raised bogs. Proc. Royal Irish Acad. 57 B14, 185. MORGAN, C. G. 1973. Archaeology. and explanation. World Arch. 4:259- 276. MORRISON, M. E. S. 1959. Evidence and interpretation of Landnam in the northeast of Ireland. Surtryck ur Postaniska Natiser 112: Fax 2. Lund. NEWBERRY, J. F. 1887. Food and fiber plants of the North American Indians. Pop. Sci. Mon. 32:31-46. O’SULLIVAN, P. E. and D. H. RILEY. 1974. Multivariate numerical analy- sis of surface pollen spectra from a native scots pine forest. Pollen et Spores 16:239-264. PALMER. E. 1887. Plants used by Indians of the United States. Amer. Nat. 12:593-606. PARSONS, R. W. and I. C. PRENTICE. 1981. Statistical approaches to R- values and the pollen-vegetation rela- tionship. Rev. Paleobot. Paly. 32:127-152. SCHIFFER, M. B. 1981. Some issues in the philosophy of archaeology. Amer. Ant. 46:899-908. SCHOENWETTER, J. 1962. The pollen analysis of eighteen archaeological sites in Arizona and New Mexico. In Chapters in the prehistory of Eastern Arizona. (P. S$. Martin, ed.). Chicago Natural History Museum Fieldiana Anth. 53:168-209. SEARS, P. B. 1932. The archaeology of environment in eastern North America. Amer. Anth. 34:610-622. 1930. Common fos- sil pollens of the Erie Basin. Bot. Gaz. 89:95-106. . 1930. A record of post- glacial climate in northern Ohio. Ohio J. Sci. 30:205-217. . 1935. Types of North American pollen profiles. Ecol. 16: 488-499. 1937. Pollen analysis as an aid in dating cultural deposits in the United States. Early Man. (G. C. MacCurdy, ed.). London, Lippin- cott. 1942. Postglacial migration of five forest genera. Amer. J. Bot. 29:684-691. SEARS, P. B. and K. H. CLISBY. 1952. Two long climatic records. Science 116:176-178. SHAW, G. 1971. The chemistry of sporo- pollenin. Pp. 305-351. Sporopollenin, Academic Press, New York. TAUBER, H. 1965. Differential pollen deposition and the interpretation of pollen diagrams. Danm. Geol. Unders. Ser. II. 89:69 p. . 967. Investigations of the mode of pollen transfer in forested areas. Rev. Paleobot. Paly. 3:277-286. THORNE, R. M. and H. K. CURRY. 1983. Cultural resources survey Yazoo River, Items 3 & 4 and a paleo- environmental model of the lower Basin. Arch. Pap. Cen. Arch. Res. Number 3, Univ., MississipP1- TROELS-SMITH, J. 1960. The Muldbjerg Dwelling place: an early Neolithic archaeological site in the Aamosen bog, West-Zealand, Denarmk. Smith. Inst. Rpt. 1959:577-601. TSUKADA, M. and E. S. DEEVEY. 1967. Pollen analyses from four lakes in the southern Maya area of Guatemala and El Salvador. Pp. 303-333 in Quaternary Paleoecology, (Cushing, E. J. and Wright, H. E., eds') Yale Univ. Press, New Haven. VAN ZIEST, W. 1967. Archaeology and palynology in the Netherlands. Rev. Paleobot. Paly. 4:45-65. Summer 1986 JOURNAL OF ETHNOBIOLOGY 65 LITERATURE CITED (continued} WATSON, P. J., S. A. LEBLANC and C. L. REDMAN. 1971. Explanation in archaeology, an explicitly scien- tific approach. Columbia Univ. Press, New York. WEBB, T. III. 1983. Calibration of Holo- cene pollen data in climatic terms. Quat. stud. Poland 4:107-113. WEBB, T. III and R. A. BRYSON. 1972. Late- and postglacial climatic change in the northern Midwest, USA: quan- titative estimates derived from fossil pollen spectra by multivariate statis- tical analysis. Quat. Res. WEBB, T. Il and D. R. CLARK. 1977. Calibrating micropaleontological ata in climatic terms: a critical review. Ann. N.Y. Acad. Sci. 288:93- WEBB, T. Ill, S. E. HOWE, R. H. W. BRADSHAE and K. M. HEIDE. 1981. Estimating plant abundances from pollen percentages: the use of regres- sion on ae Rev. Paleobot. Paly. 34:269-300 WEBB, T. Hl and J. H. MCANDREWS. 1976. Corresponding patterns of con- temporary pollen and vegetation in Central North America. Geol. Soc. Amer. Mem. 145:267-299. WILLIAMS-DEAN, G. 1978. Ethnobotany and cultural ecology of prehistoric man in southwest Texas. Unpubl. Ph.D. dissert., Texas A&M Univ., College Station, TX. WISEMAN, F. M. 1983. Analysis of pol- len from the fields of Pultrouser Swamp. Pp. 106-119 in Pultrouser Swamp. (Tumer, B. L. I and n, P. D., eds.). Univ. Texas Press. WITTE, H. 1905. Stratiotes aloides L. finnen i Sveriges postglaciala afla- gringer. Geol. foren stockh. forh 27:432. WRIGHT, J. W. 1953. Pollen dispersion studies: some practical applications. J. Forestry 51:114-118. a ate _ an aed oy. Dis = aes ee — inte = ved a J. Ethnobiol. 6(1):67-81 Summer 1986 ON THE ANALYSIS AND INTERPRETATION OF SPECIES LIST DATA IN ZOOARCHAEOLOGY R. LEE LYMAN Department of Anthropology Oregon State University Corvallis, Oregon 97331 ABSTRACT. —Z haeologi gularly di thods of q ifying fi | remains, but seldom explore the inf ion potential found in species lists. Late Pleistocene and early Holocene mammalian species lists derived from sites in eastern Washington dntn oe ee 9 i ee tS i. So . } a indicate that species list / and paleoenvironmental conditions. The presence of particular taxa and the climatic regime in a region are both factors to which prehistoric people adapted, and thus play critical roles in building models of human settlement, subsistence, and land use systems. INTRODUCTION Discussions of how to quantify vertebrate faunal remains recovered from archae- ological sites have reached a point where the specialist and novi haeologist alil must read and digest a tremendous volume of literature (see Allen and Guy 1984; Binford 1984; Grayson 1984; Horton 1984; Klein and Cruz-Uribe 1984; Lyman 1984b; Nichol and Wild 1984 for a single year of publications}. Upon cursory study of this literature it might appear that in order to do zooarchaeological research, the analyst must not only decide how to count the faunal remains, but must also have a large bone sample. Statements that species lists are of little analytic and interpretive value (e.g., Smith 1976) only serve to strengthen this notion. In this paper I illustrate that such impressions are far from true by analyzing and discussing the implications of late Pleistocene/early Hol lian species lists. While the intent of my discussion is to show that large samples and taxonomic abundance data are not always necessary in zooarchaeology, it is not my intent to show the converse—that large samples and abundance data are never necessary. It should be clear at the outset that the amount as well as kind of data required t a particular h question depend, of course, on the question asked. DISCUSSION The potential for deriving significant information from small faunas, rare taxa, and species lists has been recognized before. In one of the more important discussions of that potential, Grayson (1981a} argued that because of the taphonomical difficulties inherent in ascertaining the paleoenvironmental meaning of the relative abundance of taxa, the analyst may choose to simply use the ecological attributes of the taxa represented in a fauna as the basis for interpretations of paleoenvironments. The power of such interpretations lies in their parsimony; there are fewer assumptions about the taphonomic history of the fauna when taxa are treated as attributes of a fauna that are either present or absent in contrast to treating those taxa as variables whose abundances are the basis of interpretations (see also Grayson 1984). 68 LYMAN Vol. 6, No. 1 Grayson (1981b) illustrates the elegance of the ‘taxa as attributes” approach by using two archaeological specimens (probably from the same individual animal] of the heather vole (Phenacomys cf. intermedius) from a single stratum sample of 406 specimens to corroborate an historic biogeographical model proposed for Great Basin mammals (Brown 1971, 1978). Those same two heather vole specimens, along with other faunal and floral taxa treated as attributes are used by Thompson and Mead (1982) as bases for their inferences regarding late Pleistocene climatic conditions in the Great Basin. Clarifying the zoogeographic history of particular taxa (e.g., Lyman 1983) can have important zoological implications. It is seldom explicit, however, in studies of prehistoric human subsistence systems, that taxa must be present in an area to be exploited by humans (see Lyman 1984a for an example where this point is explicitly made). Once this fact is recognized, the utility of understanding the zoogeographic histories of poten- tially exploitable prey animals to studies of human subsistence becomes obvious. As well, when those zoogeographic histories have environmental and climatic implications (e.g., Lyman and Livingston 1983), their significance for gaining a fuller understanding of human settlement systems and land use practices increases. Studies by Binford (1980, 1982) and Kelly (1983), for example, indicate that the degree of mobility in a human land use system may be closely tied to resource availability and climatic factors. Simply put, then, faunal data of even statistically low-resolution (nominal) scale can be important to archaeological research. EXAMPLES In order to elaborate on and illustrate the above points, I have chosen late Pleistocene and early Holocene faunas recovered from sites in eastern Washington (Fig. 1). The obvious reason for this choice is that this is the area with which I am most familiar (Lyman and Livingston 1983). The second and more important reason to use examples from this area and time period is based on issues of sample size and what we know of the faunal history of the area. Quite literally, we know very little of the latter, so this paper becomes a substantive contribution. Concerning the sample size issue, only five mammalian faunal samples that are early Holocene in age are available in this area (Table 1). As well, only three late Pleistocene mammalian faunas have been described for eastern Washington. In contrast, over a dozen middle Holocene and more than fifty late Holocene mammalian archaeofaunas have been reported. Further, less than 2000 identified specimens make up six of the late Pleistocene/early Holocene samples, while several late Holocene faunas have thousands of specimens, and several middle Holocene faunas are larger than most early Holocene faunas (Table 1). Clearly, the rule of thumb that the closer to the present we are in time the more we know holds here. Can we learn anything about late Pleistocene/early Holocene faunas and the early ecology of eastern Washington from the available data? I think we can, given the taxa represented in each of the faunas I have chosen to examine (Table 2). For clarity, I have divided the following into two sections. First, I explore the implications of the historical aphy of several taxa for human subsistence practices. Then, I discuss the climatic implications of two taxa, and how the inferred environment might have affected human and use practices. Historical zoogeography.—The species lists in Table 2 illuminate certain aspects of the late Quaternary historical zoogeography of eastern Washington. For example, the Jeppson Locality fauna, the Kennewick Roadcut fauna, and the Umatilla Mammoth Site fauna provide an unprecedented view of the Wisconsinan-age mammalian biota of the area (cf. Kurten and Anderson 1980:42-43). With these faunas we find that the mammoth Summer 1986 JOURNAL OF ETHNOBIOLOGY 69 121° 120° lig? 118° i?” of GA NADAS” nal 3, , « = ss alee we i < = ce re ee sei Sa 3 ~ Ep WALIDG © 50 100 KM 2g 63 RBS Pots ee) 5 SARS fe ut Pe ee oe eee Onychomys leucogaster + Perognathus Parvus Peromyscus os maniculatus + + 7 Eutamias sp. ~ ie ia +++ + eas americana Ursus sp. ™ Ursus arctos ba Vulpes vulpes = ? Antilocapra americana + Bison sp. : Cervus elaphus + Bootherium sp. + 72 LYMAN Vol. 6, No. 1 The presence of 9000 year old grizzly bear (Ursus arctos) remains in northeastem Washington conforms to Dalquest’s (1948) conjecture that this taxon entered the area in post-Wisconsinan times. There is limited evidence that grizzlies occupied a larger range in eastern Washington during the Holocene than this taxon did in the historic period (Lyman 1986), but additional details on the zoogeographic history of this taxon are unknown. The above examples indicate that any model of human adaptations during the latest Pleistocene and earliest Holocene cannot simply be based on modern faunal data. Mammoths, musk ox, bison, antelope, and grizzly bears are all locally extinct at present, and may never have been very abundant in eastern Washington. And yet, because these taxa constitute some of the largest mammals exploited by prehistoric North American peoples, their presence alone warrants careful modeling of early Holocene human adap- tive strategies. Hayden (1981) has argued that human subsistence can be precarious in areas where the richness (number) of taxa present is low because overexploitation of one taxon may stress the entire ecosystem, including the human element, and natural fluctuations in the abundance of one taxon may also cause stress because of the limited number of alter- native prey species. Are the known faunas dating to the late Pleistocene and early Holocene more or less taxonomically rich than the middle or late Holocene faunas of eastern Washington? While there is no statistically significant correlation of sample size (NISP) and taxonomic richness for the faunas listed in Table 1 (Kendall’s tau = 0.119; p > 0.2; two-tailed test}, given the available data it is difficult to answer this question. It appears that Gustafson’s (1972) suggestion of over a decade ago that the mammalian taxonomic composition of eastern Washington during the earliest Holocene was not appreciably different from the late Holocene fauna of that area cannot be seriously challenged (see also Lyman and Livington 1983). The taxonomic composition of eastern Washington’s mammalian fauna apparently changed during the late Pleistocene to early Holocene transition (e.g., loss of Mammuthus sp., Bootherium sp., and Alopex lagopus, with coincident addition of Antilocapra americana, Onychomys leucogaste!, and Spermophilus townsendii), but species richness may not have changed significantly. The focus of research on human subsistence should then change from monitoring simple species richness to taxonomic composition; that is, determining which taxa were present, and ascertaining how those taxa might have most efficiently been exploited. For example, successful hunting techniques for taking mammoths may not be appropriate for efficient exploitation of pronghorn antelope (see for instance the discussion 1n Frison 1978). Ecological zoogeography.—Late Quaternary environmental conditions in eastem Washington have been postulated on the basis of palynological data (Mehringet 1985 and references therein). Essentially, the late Pleistocene to early Holocene transition 1S reflected by climatic warming in conjunction with a gradual decrease in effective precipita tion. Nonetheless, the environment about 10,000 years ago seems to have been generat ly cooler and moister than at present. ; Interpreting the mammalian zoogeographic data for eastern Washington in climatic terms, Lyman and Livingston (1983) simply tracked the distributions of selected taxa across environmental and faunal zones defined by Dalquest (1948). They suggested the zoogeographic data broadly reflected the environmental history suggested by the palynological data for this area (see also Lyman 1980, 1984a, 1986). There is what might be thought of as a more direct technique for measuring the environmental significance of zoogeographic data, and it is that technique I will now illustrate. Summer 1986 JOURNAL OF ETHNOBIOLOGY 73 A climatograph (Graham 1984) can be constructed for any taxon with a limited distribution by recording the temperature and precipitation regimes at a series of points plotted at the edge of the taxon’s modern range. While it is clear that many variables control a taxon’s distribution, such as interspecific competition and historical events {see reviews in Brown and Gibson 1983 and Pielou 1979), ‘‘many of these variables are dependent on or interface with climatic parameters” (Graham 1984:111). As well, regardless of how we measure it, the present distribution and resultant apparent ecology and environmental tolerances of taxa constitute the basis of all interpretations of past environments derived from prehistoric faunal remains. Two taxa in the early Holocene faunas of eastern Washington have modern distribu- tions that are particularly conducive to constructing climatographs. These are the 10,000 year old arctic fox (Alopex lagopus) from 45FR50, and the 8700 year old Columbian ground squirrel (Spermophilus columbianus) from 45GR97. To construct climatographs for these taxa, I plotted ten points around the modern range of each taxon (Figs. 2 and 3). Climatic data were derived for each plotted point on each taxon’s range and both archaeological sites by consulting Hare and Thomas (1979) and Visher (1954) (Tables 3 and 4). The climatic data were then plotted on graphs with temperature on the vertical axis and Precipitation on the horizontal axis (Fig. 4 and 5). 20° w 50°N KS oow ia ° / ° 120 w Wo FIG. 2.—Modern distribution of arctic fox (Alopex lagopus) in North America (cross- hatched) (after Hall 1981), points where climatic data were recorded (dots and capital letters; see Table 3}, and location of 45FR50 (open circle in southeastern Washington} where remains of this taxon have been found in 10,000 year old sediments: 74 LYMAN Vol. 6, No. 1 50°N~ | [-50°N 125°W 40°N~ a eae 2 40°N 115° FIG. 3.—Modern distribution of Columbian ground squirrel (Spermophilus columbianus) in North America (cross-hatched] (after Hall 1981}, points where climatic data were recorded (dots and capital letters; see Table 4), and location of 45GR97 (open circle in central Washington) where remains of this taxon have been found in 8700 yeat old sediments. The type of climatic data available for the range of the arctic fox dictated the resulting climatograph; specifically, no seasonal precipitation data were available, and only seaso temperature data were available (Table 3). The climatograph is nonetheless informative, and suggests summer temperatures would have had to be about 1°C cooler than at present for arctic foxes to have lived in southeastern Washington (Fig. 4). Underwo and Mosher (1982) suggest that the population size of this taxon seems to be dictated by the size of the rodent populations upon which arctic foxes prey, and that arctic foxes are opportunistic feeders and climatic factors may only secondarily control their abun dance and distribution as some races can tolerate relatively warm, mild winters. © climatograph and the observations on the modern distribution of arctic foxes tend to Summer 1986 JOURNAL OF ETHNOBIOLOGY i> TABLE 3.—Climatic data a ti distributional limits of the arctic fox (Alopex lagopus) and 45FRS5O. See Figs. 2 and January July Ann Point Temperature Temperature Precipitation (°C) (C°) rs -30 1a 203 B -30 15 305 » -25 iS 305 D 20 17 406 E 20 iy 508 F 12.5 i746 813 G 20 17 813 H 12.5 15 1000 I ps: 15 1000 J 35 2 102 45FR50 LS 18 381 TABLE 4.—Climatic data for the distributional limits of the Columbian ground squirrel (Spermophilus columbianus) and 45GR97. See Figs. 3 and 5 Temperature (°C) Precipitation (mm) Point January July Annual January July Annual A -I 21 7 152 101 508 B al 21 10 101 101 508 C 1.5 21 10 76 101 508 D 4 18.5 7 76 101 508 E -4 a 7 101 101 508 F -6.5 21 4.5 101 101 508 G -6.5 15.5 1.5 101 101 508 H 4 18.5 4.5 101 127 508 I 9.5 15.5 4.5 101 127 508 J -15 15 ND 100° 100° 508 45GR98 -1.5 18 10 45 0 254 ae ah sat la ND; no data available. values are estimates. 76 LYMAN Vol. 6, No. 1 (FRS5O ae! JANUARY 1 <(04 TEMPERATURE (°C) -20— -304 T 100 200 300 400 ae 600 #00 800 900 1000 PRECIPITATION (mm) FIG. 4.—Climatograph for arctic fox (Alopex lagopus) derived from data in Table 3. The open circles with letter labels are the geographic points (Fig. 2). The upper part of the graph represents July temperatures plotted against annual precipitation; the lower part of the graph represents January temperatures plotted against annual precipitation. The upper dot is the July temperature plotted against annual precipitation for 45FR50; the lower dot is the January temperature plotted against annual precipitation for 45FR50. corroborate Gustafson’s (1972) original interpretations of earliest Holocene environments of eastern Washington as involving milder winters and cooler summers than at present. The type of climatic data available for the distribution of the Columbian ground squirrel (Table 4) also dictated the nature of the resulting climatograph (Fig. 5) for this taxon. Clearly, it is much too xeric at 45GR97 today for the Columbian ground squirrel, particularly during the summer months when about 100 mm of additional precipitation would be needed to make the site area suitable habitat for this taxon today. Even the winter months at the site are too dry for this ground squirrel. This taxon today prefers relatively mesic habitats, usually in grasslands because the Columbian ground squirre is a grazer (Tyser and Moermond 1983). This species is never far from water, whether free-flowing or in the form of moist sub-surface sediments (Turner 1972). In fact, seasonal sediment moisture, because it determines the vegetation cover, may be a factor which limits aboveground activity and dictates summer aestivation periods of this taxon (Tum¢t 1972). The climatograph thus matches and corroborates the inferences derived from palynological data that central Washington was moister during the early Holocene than at present. The significance of the climatographs for understanding prehistoric human land use is reflected in a Holocene sequence of cultural adaptations postulated by Galm et al. (1981). They suggest that between 11,000 and 8000 B.P. people hunted large and small Summer 1986 JOURNAL OF ETHNOBIOLOGY 77 20 GR97 e > JULY a ee | ar 15 NNUAL lo- ae ; GR9O7 : = 2 = = Ww $ [aa 4 —_ = 9 = o- ul oO. = Li] — -5> > ie ¢ 4 > 4 a 1K 3 a T a T T se) 100 200 300 400 500 PRECIPITATION (mm) BIG. 5. —Climatograph for Columbian ground squirrel (Spermophilus siren derived from data in Table 4. The open circles with letter labels are t (Fig. 3). The graphs on the left represent the July precipitation plotted against ae temperature, the January precipitation plotted against January temperature, and the dots represent the July and January climate at 45GR97 today. The lines on the right repre- sent the July, Annual, and January temperature ranges for all geographic points plotted against annual precipitation (508 mm) for all geographic points. The dot in the center tepresents the annual climate for 45GR97. 78 LYMAN Vol. 6, No. 1 mammals, fished for non-salmonids, and followed a nomadic but seasonally scheduled pattern of movements between winter camps located in the major river canyons and summer camps and resource extraction loci in the uplands. With warming and drying of environments between 8000 and 4500 B.P., people intensified fishing activities and the exploitation of seeds and roots, and spent more time in river canyons and less time in the increasingly xeric uplands than previously (see also Chatters 1982). In Binford (1980) and Kelly’s (1983) terms, it appears that a shift from a residential mobility strategy to a logistical mobility strategy took place as environments became progressively warmer and drier. The typical explanation for such a shift in land use strategies concerns changes in resource availability, especially in this case expressed as decreases in (animal) biomass (e.g., Kelly 1983 and references therein). The faunal data on which the climatographs are based tend to conform to this kind of explanation: the disappearance of arctic foxes and Columbian ground squirrels from increasingly xeric habitats in eastern Washington represents a decrease in exploitable faunal biomass in those habitats. SUMMARY AND CONCLUSIONS The increasingly numerous discussions of quantitative methods in zooarchaeology present a bewildering set of alternatives and opinions on how to best quantify a sample of faunal remains. Comparable discussions of taxa as attributes of a fauna and techni- ques for analyzing and deriving inferences from species lists are, in contrast, rare. Perhaps this is due to the relative parsimony of deriving inferences based on nominal scale faunal data and/or the suspected greater potential resolution afforded by ordinal and interval scale faunal data. Regardless of the reasons for the paucity of literature on nominal scale faunal data (a large portion of that literature is found in paleontological journals; see for example Raup and Crick 1979 and references therein), such data are potentially useful and informative. Using examples from the late Pleistocene and early Holocene of eastern Washington, I have shown here that nominal scale faunal data may (1) clarify the zoographic history of a taxon, (2) indicate fruitful analytical pathways towards modeling human subsistence by providing information on the taxonomic composition of faunas of particular time periods, (3) provide data indicative of past climatic conditions, and (4) suggest causes of changes in human land use practices. ACKNOWLEDGEMENTS I thank Donald K. Grayson and Stephanie D. Livingston for their comments on is earlier version of this paper. While they do not always agree with my methods an conclusions, they always make sure that I understand what I am doing. LITERATURE CITED AGENBROAD, LARRY D. 1984. New How minimal is the MNI? Archae- world mammoth distribution. ology in Oceania 19:41-47. Pp. 90-108. In Quaternary extinc- BINFORD, LEWIS R. 1980. Willow tions: A prehistoric revolution, P.S. smoke and dogs’ tails: hunter Martin and R.G. Klein, eds. Univ. gatherer settlement systems and Arizona Press, Tucson. archaeological site formation. Amer. ALLEN, JIM and J.B.M. GUY. 1984. Antig. 45:1-17. Optimal estimations of individuals i 1982. The archaeology of archaeological faunal assemblages: place. J. Anthro. Archaeology 1:5-31. Summer 1986 JOURNAL OF ETHNOBIOLOGY be LITERATURE CITED (continued) . 1984, Faunal remains from Klasies river mouth. Academic Press, New York. BROWN, JAMES H. and ARTHUR C. GIBSON. 1983. Biogeography. C.V. Mosby Co., St. Louis. BRYAN, ALAN. 1955. Excavations at Meyer Caves in east central Wash- ington. Davidson J. of Anthro. 1(1): 11-20. Seattle. CHANCE, DAVID H. and JENNIFER V. CHANCE. 1982. Kettle Falls: 1971 and 1974: Salvage Archaeology in Lake Roosevelt. Univ. Idaho Anthro. Research Manuscripts Series No. 69. Moscow. CHATTERS, JAMES C. 1982. Prehistoric settlement and land use in the dry Columbia Basin. Northwest Anthro. Res. Notes 16:125-147. COLE, DAVID L. 1968. Report on archaeological research in the John Day Dam Reservoir Area—1967. Univ. Oregon Museum of Natural History, Eugene. DALQUEST, WALTER W. 1948. Mam- mals of Washington. Univ. Kansas Publ. aaa Museum of Natural History No. Deaven, KEN and GLEN S. GREENE. 1978. Faunal utilization at 45AD2: A prehistoric archaeological site in the channeled scablands of eastern Washington. Tebiwa, Misc. Papers of oe Idaho State Univ. Museum of Natural History, No. FRISON, GEORGE C. 1978. Prehistoric hunters of the high plains. Academic Press, New York. FRY, WILLIS E. 1969. A Paleontological site survey conducted in the Horse Heaven Hills of South-Central Wash- ington. Northwest Sci. 43:156-161. GALM, JERRY R. and GLENN D. HART- UTH A. MASTEN and GARRY O. STEPHENSON. 1981. A cultural resources overview Bonneville power administration’s Mid-Columbia Project, Central Washington. Eastern Washington Univ. Reports in Arch. and Hist. 100-116. Cheney. GILBOW, DELBERT W. 1981. Inference of human activity from faunal remains. Unpubl. thesis, (Anth.), Washington State Univ., Pullman. GRAHAM, RUSSELL W. 1984. Paleo- environmental implications of the Central Texas. Quaternary Res. 21:111-114. GRAYSON, DONALD K. 198la. A critical view of the use of archae- ological vertebrates in paleoenviron- mental reconstruction. J. Ethnobiol. 1:28-38 . 1981b. A Mid-Holocene record for the heather vole, Phena- comys cf. intermedius, in the Central Great Basin and its Biogeographic Significance. J. Mammalogy 62:115- LAL, . 1984. Quantitative Zoo- archaeology. Academic Press, New York. GUSTAFSON, CARL E. 1972. Faunal remains m the marmes rock shelter and related archaeological sites in the Columbia Basin. Unpubl. dissert. (Zoology), Washington State Univ., Pullman. HALL, E. RAYMOND. 1981. The Mam- mals of North America, second edi- tion, 2 vols. Wiley, New York. HARE, F. KENNETH and MORLEY K. THOMAS. 1979. Climate Canada, second edition. Wiley, Toronto. HAYDEN, BRIAN. 1981. Subsistence and ecological adaptations of modem hunter/gatherers. Pp. 344-421. In Omnivorous Primates, R.S.O. Hard- ing and G. Teleki, eds. Columbia Univ. Press, New York. IRWIN, ANN M. and ULA MOODY. 1978. The Lind Coulee Site (45GR97). 80 LYMAN Vol. 6, No. 1 LITERATURE CITED (continued) Washington Archaeol. Res. Center Proj. Rep. No. 56. Pullman. KELLY, ROBERT L. 1983. Hunter- gatherer mobility strategies. J. Anthro. Res. 39:277-306. KLEIN, RICHARD G. and KATHRYN CRUZ-URIBE. 1984. The Analysis of Animal Bones from archaeological sites. Univ. Chicago Press, Chicago. KURTEN, BJORN and ELAINE ANDER- SON. 1980. Pleistocene Mammals of North America. Columbia Univ. Press, New York. HORTON, D.R. 1984. Minimum num- bers: a consideration. J. Archaeol. Sei. 11:255-271. LYMAN, R. LEE. 1980. Bivalve mol- of three genera. Northwest Sci. 54: 121-136. 1983. Prehistoric extra- limital records for Pappogeomys castoanops (Geomyidae) in North- western New Mexico. J. Mammalogy 64:502-505. 1984a. A model of large freshwater clam exploitation in the prehistoric southern Columbia Plateau culture area. Northwest Anthro. Res. Notes 18:97-107. 1984b. Bone density and differential survivorship of fossil classes. J. Anthro. Arch. 3:259-299. 1986. On the Holocene history of Ursus in eastern Washing- ton. Northwest Sci. 60, in press. ui... and STEPHANIE 1. LIVv- INGSTON. 1983. Late quatern. ian Zoogeography of eastern Washington. Quaternary Res. 20: 360-373. MARTIN, J.E., ADD. BARNOSKY and C.W. BARNOSKY. 1982. Fauna and flora associated with the west richland mammoth from the pleis- tocene touchet formation in south- central Washington. Thomas Burke Memorial Museum Res. Rep. No. 3. Seattle. MEAD, JIM I., EMILEE MEAD and D.W. STEADMAN. 1985. Faunal studies. Pp. 35-42. In Paleoenvironmental investigations at Seed Cave (Windust CaveH—45FR46), Franklin County, Washington, by Robert S. Thompson. Eastern Washington Univ. Rep. in Arch. and Hist. 100-141. Cheney. MEHRINGER, PETER J., JR. 1985. Late-quaternary pollen records from the interior pacific northwest and northern great basin of the United States. P. 167-189. In Pollen Records of Late-Quaternary North American Sediments, V.A. Bryant and R.G. Holloway, eds. Amer. Assoc. Strat. Palynol., Dallas. NEWCOMB, R.C. and C.A. REPEN- NING. 1970. Occurrence of mam- moth fossils in the touchet beds, south-central Washington. North- west Sci. 44:16-18. NICHOL, RK. and CJ. Wild. 1984. “Numbers of Individuals” in faunal analysis: the decay of fish bone 17 archaeological sites. J. Archaeol. Sci. 11:35-51. OLSON, DEBORAH L. 1983. A descrip- tive analysis of the faunal remains from the Miller Site, Franklin County, Washington. Unpubl. thesis, (Anth.), Washington State Univ., Pullman. PIELOU, E.C. 1979. Biogeography. Wiley-Interscience, New York. RAUP, DAVID M. and REX E. CRICK. 1979. Measurement of faunal simt larity in paleontology. J. Paleon- tology 53:1213-1227. RENSBERGER, JOHN M., ANTHONY D. BARNOSKY and PATRICK SPENCER. 1984. Geology and Paleontology of a Pleistoceneto Holocene loess succession, Benton County, Washington. Eastern Wash- ington Univ. Rep. in Arch. an Summer 1986 JOURNAL OF ETHNOBIOLOGY 81 LITERATURE CITED (continued) Hist. 100-139. Cheney. SMITH, BRUCE D. 1976. ‘’Twitching’”’: A minor ailment affecting human paleoecological research. Pp. 275-292. In Cultural Change and Continuity, C.E. Cleland, ed. Academic Press, New York. THOMPSON, R.S. and J.I. MEAD. 1982. Late Quaternary environments and biogeography in the Great Basin. ternary Res. 17:39-55. TURNER, LARRY W. 1972. Habitat dif- ferences between Spermophilus beldingi and Spermophilus colum- bianus in Oregon. J. Mammalogy 53:914-917. UNDERWOOD, LARRY and JAMES A. MOSHER. 1982. Arctic Fox. Pp. 491- 503. In Wild Mammals of North America, J.A. Chapman and G.A. Feldhamer, eds. Johns Hopkins Univ. Press, Baltimore. VISHER, STEPHEN S. 1954. Climatic Atlas of the United States. Harvard Univ. Press, Cambridge. TYSER, ROBIN W. and TIMOTHY C. MOERMOND. 1983. Foraging behavior in two species of different- sized sciurids. Amer. Mid]. Natur. 109:240-245. J. Ethnobiol. 6(1):83-98 Summer 1986 ETHNOBIOLOGY, COGNITION AND THE STRUCTURE OF PREHENSION: SOME GENERAL THEORETICAL NOTES R. F. ELLEN Eliot College University of Kent at Canterbury Canterbury, Kent CT2 7NS, United Kingdom ABSTRACT.—Over the last decade taxonomy has been shown increasingly to be an agi enema ipaviepies inadequate representation of the results and devi ty in folk biology. re = rapt | Its concepts of rank, level and contrast, and pti ght p do not exhaustively account for observed behavior, while flexibility and contextual con- siderations have been under-stressed. Taxonomic approaches, moreover, have tended to produce patterns which reflect method as much as anything intrinsic to the data. This paper summarizes some of these difficulties, and suggests that a methodology focused on the notion of prehension may provide the basis for a broader approach encompassing both cognitive and social processes involved in the generation of classifications. Man and life and nature are none of them domains that present themselves to the curiosity of knowledge spontaneously and passively. [Foucault 1970:72} In its strict and technical sense, taxonomy is an hierarchical metaphor involving linked notions of rank, level and contrast. In European thought, it finds its first known historical expression in Aristotle, but has since passed via Linnaeus into modem biological usage. As a model of classification it had gained such wide currency by the middle of the present century that a great many ethnographers assumed that it must also necessarily order the folk schemes they were beginning to describe and analyze. This assumption has given rise to the formal theories associated most closely with Brent Berlin and the American school of ethnosemantics (see Ellen 1979b:12-13). In the last decade these theories have come under sustained attack. The objections have been various and interconnected (eg. Ellen 1979a, 1979b; Fox 1975:118-119; Friedberg 1968, 1970; Healey 1978-79; Hunn 1976, 1977b, 1982—despite Hunn 1975). However, it is helpful to present them under a limited number of headings, even at the risk of appearing to conflate certain matters and separate others that are clearly linked. The headings are: definition, rank and level, contrast, flexibility, context, taxonomizing as a thought process, taxonomic artifacts and taxonomy as theory. In this Paper I wish, first, to provide a resume of the various criticisms and then, in the light of my own analysis of Nuaulu ethnozoology (Ellen 1972, 1975; 1976a; Ellen, Stimson and Menzies 1976a, 1976b; Ellen 1978, 1979a, 1985 forthcoming, in press), to outline a broader alternative approach focusing on a concept of prehension. AGAINST A GENERAL TAXONOMIC THEORY OF CATEGORIZATION 1. Definition. It has not helped matters greatly that there has been some confusion over what, Precisely, is meant by “taxonomy.” For some it is no more than a synonym for Classification, and in this respect folk taxonomies are seen as equivalent to folk classi- fications. Thus, Berlin, Breedlove, and Raven (1973:214) use taxa to refer to all “linguistically recognized groupings of organisms of varying degrees of inclusiveness.” The confusion is compounded by the different meanings attached to the term and its cognates in different European languages. In French, taxonomie is broadly equivalent to the narrow definition of “taxonomy” in English; while taxinomie (from the Greek 84 ELLEN Vol. 6, No. 1 taxis) has a much broader reference. As Claudine Friedberg (1968:315) points out, while the distinguished nineteenth century botanist, A. P. de Candolle, used taxonomie, Lévi- Strauss uses taxinomie. It is, of course, possible that this distinction i is responsible for the different usages of English language writers. Here I have employed ‘‘taxonomy” only in its strict Aristotelian sense. That is, in the sense understood by Kay (1971), as a model which owes its form to the Linnaean analogy expressed in set theoretic terms. 2. Rank and level. The notion of hierarchy is integral to the taxonomic model found in ethnobiology. Berlin, Breedlove, and Raven (1973:214] pursue the logic further and suggest that taxa of differing degrees of inclusiveness can be placed in a limited number of ranked categorical types (unique beginner, life-form, generic, specific, varietal] and that these characteristically occur at the same taxonomic level. Upon these distinctions and definitions they erect a complex theoretical edifice which is then used to describe particular ethnographic cases. The notion of hierarchy, at least as it has been applied to entire classifications, and the insistence on assigning categories to different levels, has sometimes given rise to problems which are altogether spurious. The concept of “level,” except in a limited and rather crude way, is very difficult to demonstrate beyond particular local regions of classificatory space in particular domains of particular peoples. Similarly, although, used in a loose metaphorical way, all folk classifications are “hierarchic,’”’ none are of great depth in any absolute sense (Hunn 1976:509). If we wish to retain some semblance of taxonomic organization we might preferably do so by opting for the tree-diagram model rather than that of a true taxonomy. That is, not by making a priori assumptions as to level. Thus in Fig. 1, although it may well be possible to demonstrate “local” contrasts ao sini AANAA FIG. 1.—Standard abstract representation of a taxonomy. Upper case letter, lower-case letters and figures respectively indicate items at the same Jevel. Items at the same level are regarded as being in a relationship of contrast. Items at each subordinate level are contrasting segregates of a more-inclusive class at the next highest level. Summer 1986 JOURNAL OF ETHNOBIOLOGY 85 (1:2), we cannot always show with the same ease that 1 contrasts laterally with 8 even though 1 and 8 may share the same number of categorical links from 0, as in: OBd8, OAal. The recognition of taxonomic rank requires “inelegant complications of formal taxonomic models” (Hunn 1976:510}, provides no basis for either distinguishing induction from deduction, for explaining how taxa or other categories are actually generated, for handling non-transitivity, or for the fact that paired and contrasted items do not necessarily imply more inclusive categories (Brown 1979:794-795; Lancy and Strathern 1981:78). I summarize some of the reasons why this should be so in the remainder of this section. Hierarchy is not the only metaphor that can express inclusivity in the relations between categories, and there are many ethnographic demonstrations of this. Notions of “broadness” and “narrowness” may, for example, be more appropriate. Moreover, there appears to be a curious notion that classificatory space is consistent with the dimen- sions of a type of two-dimensional graphic representation which, historically and ethnographically, is of limited extent. Indigenous conceptual arrangements and their linguistic expression may be least violated by employing other means of graphic represen- tation: networks of focal points around which categories cluster, Venn diagrams in two or three dimensions, “sphere of influence” models, or “type-token representations (Bright and Bright 1965; Ellen 1979a:354-357, 1979b:12-14; Friedberg 1970; Tyler 1978:278-279). Hunn (1976:515) has employed the notion of a system of differences in “classification space.’ Distances within this space are assessed in terms of overall similarity and difference between organisms with respect to perceptible attributes of morphology and behavior. Categories are defined by reference to patterns perceived within this system of differences. 3. Contrast. The model of taxonomy employed by Berlin, Breedlove and Raven (1973:214) tequires that categories at the same level be mutually exclusive and contrasting. This ethnosystematicist notion developed in conjunction with componential analysis, which has proved inadequate for defining such concepts (Hunn 1976:509; cf. Turner 1974:16-17). Folk biological categories in general are not defined by reference to verbalizable feature contrasts, let alone single characters. They are semantic primitives, at their lower levels Senerated by induction (Hunn 1975:313, 1976:515). In some classifications not even the notion of contrast need be present, at least not in the sense that we normally understand it. Thus, rather than something being x or y it is common for it to be more x than y, Or more y than x. The idea of gradation eliminates distinct boundaries and has been termed by Lakoff (1972) “‘fuzziness.”’ Thus we are invited to speak of ‘‘shrubness” or “treeness, “birdness” or “snakeness” (Randall 1976:549-551). The vagueness can sometimes exceed even such fuzzy expressions as these, especially in the form of such hedges as “well, it might be,” “it’s a sort of bird” (where the stress is on ‘‘sort’”’ rather than “bird’’}, and “‘may be.” All of this fits well with the idea that the prevailing relations between Categories are through polythesis. 4. Flexibility. The problem of applying notions of hierarchy and contrast consistently lead us to suggest that taxonomies {and particularly those which are imputed to rest Predominantly upon morphological distinctions) have often been assumed incorrectly to be the only (or if not the only then certainly the primary or dominant) means of Classification. Berlin’s approach has been strongly and widely criticized for attempting to impose a form of taxonomic rigidity on a cultural apparatus the general characteristics of which are quite antithetical: namely fludity, flexibility and elasticity (Bulmer 1974:24; Dwyer 1976:442; Ellen 1979b; Healey 1978-1979}. For Friedberg (1974:327), there are a multiplicity of systems of reference, and there is always the possibility of disjunction between the separate spheres of nomenclature, identification and classification; or we 86 ELLEN Vol. 6, No. 1 might say between “models for’ and ‘‘models of,” or between keys and classifications. It would appear then that taxonomies have to be extracted and created from a muc more complex classificatory web of which they are part, and in the process even constructed in the interests of neat presentation. The evidence for the inadequacy of taxonomy has come to light partly through the so-called “special problems,’ which many have seen as simply the artifacts of method. Examples of these include multiple and inter-locking hierarchies, the employment of radically different principles at different levels, and extra-hierarchic relations (eg. Conklin 1969:50; Perchonock and Wemer 1969:232, 234). Thus, to say that a taxonomy has cross-cutting classes is to beg the primacy of the taxonomy itself, especially when such categories are highly significant in cultural terms, however we may agree to measure this. Synonymy, homonymy, polysemy and anomaly, all lead us to question the legitimacy of an approach in which there are distinct ranked levels and clearly-bounded contrasting categories. Covert categories, in so far as they can be demonstrated to exist at all, seem to contradict the very idea of taxonomy meena soe Paiavit. 1976; Healey 1978-1979:364). 5. Context. Trenchant also has been the criticism of taxonomic and allied ethnobiological studies for becoming divorced from the situational considerations of ethnography, of the context in which folk classifying takes place (Ellen 1979b; Martin 1975). Thus, for Friedberg (1974:320), “une classification d’objets naturels n’est indépendante de ces derniers; ils existent en dehors de la perception que ]’on peut avoir d’eux dans une culture particuliére.”” The problem with decontextualization is not simply that it isolates clas- sifications from the rest of culture and thus presents us with something which is abstract, but that by shedding extraneous cultural information it presents us with the illusion that knowledge consists merely in understanding resemblances (cf. Foucault 1970:1 11). In other words, it is a complete reversal of the ethnoscience position in which adequate description should provide enough information to know how to perform ; culturally acceptable manner. If the decision is made to isolate classifications as form tions of that reality. The elimination of context enables the assumption that what ' being examined is, in fact, a formal system. Identifying a taxonomic, or any eae classificatory system, is just like identifying a religious “system” in a tribal society ; is always possible to isolate it if you want to, but to do so may lead to a complete misunderstanding of its structure, function and position in the social and cultural fabric- Social and cultural anthropologists, of all people, cannot make this assumpto? (Harrison 1970, 1977). i. i ll, ec, gl cm i ———— 1. ————z, Fe a ee ll, ey, Ay a ~_ Summer 1986 JOURNAL OF ETHNOBIOLOGY 87 6. Taxonomizing as a thought process. There has been an assumption in much formalistic research that the reconstruction of taxonomic hierarchies provides us with a basis for understanding thought processes involved in classifying behavior, that categories are classified in a particular way as a reflection of how people think. It is now clear that taxonomic classifications play a lesser role in human thinking than has been hitherto assumed, especially outside the area of biological kinds (Wierzbicka 1984:325). Strictly taxonomic categories have often not been distinguished from other types of category, the conceptual relation “kind of’ has not been clearly dissected from the referential relation of set inclusion (Wierzbicka 1984:313, 315}. But even in ethnobiology there has been considerable scepticism. Randall (1976) has criticized the taxoomic model on the grounds that its assumptions about transitivity suggest incorrectly that elicited taxonomies represent structures involved in memory storage. Rather, he suggests, taxonomic trees are the result of classifying behavior, and not the means by which information is stored. It is also now clear that certain hierarchical relations are not transitive, and that complex hierarchies are often a product of the procedures used. Connections between different categories are regularly made using short-cuts, in ways which seem to violate taxonomic reconstructions. Thus, in Fig. 1, the relationship between 1 and, say 6, is commonly not arrived at via the route aABd (as would be necessary using taxonomic logic), but often directly or through another level category. We may therefore conclude with Hunn (1977a:12) that, “it is more valid psychologically to describe the classificatory structure as based on non-hierarchical relations of perceived similarity.” This is why network and spatial models are somehow more attractive than the undirectional, duodimensional taxonomy. Moreover, as Hunn has pointed out, “taxonomic theory provides no basis for distinguishing induction from deduction in logical thought.’ He argues that more inclusive categories, such as “birdoid,” are deter- mined deductively, while less inclusive categories, such as ‘‘sparrow,” are determined inductivly (Hunn 1976:510, 519). The evidence for this now seems overwhelming. 7. Producing taxonomic arti facts. If folk taxonomies do not reflect the actual psychological Processes involved in many of those activities we describe as classifying, it is necessary to ask what it is they do reflect. They were at first thought to reflect a true emic model. This, after all, was the aim of the “new ethnography” of which folk biology has been a noteworthy part. It does, however, begin to look more like an etic model of the emic, and then of a very particular kind, where the analyst has already decided what the model should contain and selected results accordingly, even rejecting some statements on the grounds that they appear to represent the idiosyncratic interests of particular informants. What they do reflect, of course, to a very considerable extent, is an observer's model of taxonomy, which in its most “tight” stereotypical Linnaean form has been outlined by Kay (1971). The attraction of such a model lies partly in its intrinsic aesthetic appeal, Partly in a desire for parsimony, lucidity and rational order, partly in its demonstrable resemblance to some processes in folk classifying, but partly also in its implicit, and occasionally explicit, mimicry of scientific natural historical taxonomy and linguistics. This latter influence is evident both in terms of how the enterprise is phrased, the ways in which data are conceived, in the technical procedures for elicitation, in the formal Precision of analysis and in the formulation of the problematic. In some cases, I would Suggest that individuals with a grounding in natural history have a hidden bias towards finding “natural” categories, and towards an under-emphasis of variation, with a Corresponding stress on the taxonomic approach. One gets the impression = ethno- biological taxonomists are by inclination collectors, and as Bulmer (1974:82) has pertinently remarked: .. . almost all collectors like nice, perfect specimens, and derive considerable aesthetic pleasure from the ordering they impose on them. A danger which the 88 ELLEN Vol. 6, No. 1 ethnobiologist must guard against very consciously is that of letting his personal aesthetic judgements override his data... But more generally, while the specific scientific model is certainly the immediate and most obvious source for representations of folk models, it is well to remember that the taxonomic model and its close classificatory congeners is grounded in a special cultural tradition, in which the graphic and written representation of the relations between categories markedly alters, rigidifies and directs their conceptualization (Ellen 1979b). In addition to the specific conventions of literacy, graphic representation and the scientific tradition, there is the general fact that in all inter-cultural communication (of which ethnobiological ethnography is just one, very specialized, example) all local explanations have to be rendered in a form which is in some sense meaningful to the ethnographer. Although we pride ourselves on our grasp of alternative worldviews and organizing principles, we cannot be sure that we ever known everything that is relevant, since knowledge may be caste in an idiom with which we are quite unfamiliar, and therefore ill-placed to understand. The history of anthropological theory itself provides ample confirmation of this, as successive explanatory frameworks are able to indicate previously unexpected ionships and g in old data. Moreover, within the confines of our own range of conceptualization we are always predisposed (by virtue of our cultural socializati nd professional training] to favor one view rather than another. The attraction of the taxonomic model is that once you have data devoid of contextual considerations (in their widest sense) it is virtually impossible not to put a taxonomic construction upon them. It is easy enough to represent a classification as taxonomically ordered and based entirely on morphological criteria if you a priori assert that this is what you are looking for. If you are attempting to reconstruct a native conceptual universe as it applies to plants and animals then you cannot begin by excluding categories and arrangements based on non-morphological criteria since informants do not, in the coursee of their ordinary lives, necessarily make suc discriminations. In my experience people do not regularly make judgements which suggest that they operate with an all-purpose classification which is recognized as being in any way separate from, or different to, classifications organized in some other way. As soon as you begin to exclude certain categories an g ts from considera- tion then, of course, you begin to yield regularities which look much more like contrasting, hierarchically-ordered and ranked taxa. Thus, the unambiguous all-purpose, morphologically-based taxonomy is something which the ethnographer or linguist extracts, but even then seldom perfectly. What is extracted may serve aS a basis for establishing the existence of universal processes, schemes of categories, but equally it may do no more than reflect an artifact created by common techniques of extraction and representation. What is more, the very character of taxonomizing as a process generates anomalies. To s a opinion poll style survey questions. But in addition to this, techniques such as card os slip-sorting, the drawing of tree diagrams and other techniques which mechanically encode an implicit or explicit assumption to dichotamize successively, will unfailingly Summer 1986 JOURNAL OF ETHNOBIOLOGY 89 produce formal hierarchies, while eliciting definitions ruthlessly erases any fuzziness between categories as ordinarily used. It is only possible to approach the essential cultural reality of categories if techniques are for the most part basically non-directive, that is much looser and more varied (see Perchonock and Werner 1969:236-238). Hence, in my own work I have increasingly relied on simply listening to people talking about animals, using prompts which are less inclined to force the data into formal taxonomies. I am particularly troubled that some scholars should believe so readily that taxonomies are incontravertibly ‘in the data,’ that they emerge “from the data as a consequence of their natural properties’ (see Lakatos and Musgrave 1970:98). The taxonomic view of classification is in this sense like Lévi-Straussian structuralism, in that if you try hard enough it is possible to discern the kind of order you are seeking, wherever you wish. My field data (by which I understand what was written down in my notebooks, on cards, on specimen tags, in photographs and drawings, and recorded on magnetic tape) were obtained through a mixture of directive and non-directive methods. In this form they consist of fragmented, sometimes contradictory statements, and have to be “processed,” transformed into clear generalizations, testable hypotheses and descriptions. There is nothing in the data in this form which would suggest that a distinction between morphological and non-morphological characteristics, or between taxonomic and non-taxonomic processes, is justified, or that there exists something called “the Nuaulu classification of animals,” which is conceived of as some kind of structured totality. This is a construction which I have placed upon the data. Certainly, one of the properties of the Nuaulu data (as I have presented them) is that many permit a taxonomic construction; but they may permit others as well. The ‘natural’ properties referred to are ambiguous and the term contentious. I would wish strongly to resist the kind of empiricism which uncritically sees taxonomies as simply facts out there waiting to be collected, like so many herbarium specimens. Such an approach begs crucial questions in the understanding of classification, and (ironically) its general application has only been possible through the systematic neglect of the full range of factors at work in “classifying.” In the highly particular social world of professional biology the principle of “taxonomic rigidity” is an important working assumption; in the context of ethnobiology it has become simply dogma (Healey 1978-1979:379). But you cannot work from taxonomy as if it provided a set of axioms (Hunn 1976:510). 8. Taxonomy as theoretical icon. A final problem, and perhaps a factor explaining the tenacity with which its practitioners defend the taxonomic approach, is linked to the fact that it is not simply that taxonomy and formal elicitation has become indelibly linked to certain kinds of substantive investigations, but that ‘‘taxonomy” has become central to, even to emblazon, particular theories of culture (Conklin 1969), conceptions of ethnography (Spradley 1979), and models of thought processes (Bruner, Goodnow and Austin 1956}. In other words, it has become reified; to some extent at least because it Provides a universal model to counter cultural relativism (Brown, Kolar, Torrey, Truong- Quang and Volkman 1976}. Curiously (and paradoxically), the approach is based on the rigorous definition of cultural boundaries in order to provide the basic units for the construction of pan-human hypotheses. However, this notion of the boundedness of cultures is linked to an organic model and crude functionalist assumptions, without being genuinely sociological, systematic or contextual. This becomes readily apparent if it is compared with the treatment of classification in, say, historical linguistics and Philology. In the latter, cultural boundaries disintegrate, and diffusion and historical explanation rule, and the patterns themselves are contingent upon particular social and tural processes. 90 ELLEN Vol. 6, No. 1 BESIDES TAXONOMIES While we may agree that attempts to define, rigorously, the principles of classi- fication and nomenclature in folk biology are certainly useful for particular ethnographic populations, the making of inductive generalizations about certain types cross-culturally assumes that variability is according to a limited number of well-understood criteria along parallel axes. The restricted check-list approach exemplified by the work of Berlin and his associates cannot, then, cope with the wider dimensions of variation between systems. It not only tends to reify a particular kind of classification (that which we call taxonomic], but seems to claim that a large number of semantic fields are at all times similarly organized. ‘Taxonomy has been elevated to an artificially high status as the mode that humans employ to organize and act upon discrete elements in the environment” (Rosch, Mervis, Gray, Johnson and Boyes-Braem 1976). It is compelling because it is a stylish (p. 790) representation of relationships among natural elements and because taxonomizing appears to us as an efficient strategy for organizing, storing, and retrieving elements (especially words) in memory (see eg. Ericsson, Chase and Faloon 1980). Even if we agree that the taxonomic mode typifies Western culture (and I am not convincd of this either), we cannot assume that all cultures have exactly the same formulations of resemblance, relationship, class or contrast (see Hobart 1982:56). Nevertheless, many field researchers, sceptical of the claims for an all-embracing taxonomy in human categorization of nature, have reasonably argued in favor of a “limited natural taxonomy,” or have found taxonomies a convenient descriptive framework (Ellen, Stimson and Menzies 1976b; Taylor 1980:285). e may agree that, as one available common process, taxonomy is universally available in the classifying repertoires of all people. It is, however, more important in some societies than in others (see also Super, Harkness and Baldwin 1977), and although the taxonomic mode is useful in describing the structure of some systems it 1s not entirely adequate (Hunn 1977a:13), and in some cases limited to particular domains (Lancy and Strathern 1981:780). Thus, taxonomy works quite well for Nuaulu vertebrates (and among vertebrates for birds and reptiles), but it works less well with mammals, invertebrates and fungi. Elsewhere, although the Tobelorese use non-taxonomic features they appear to prefer taxonomic structure, especially “below” basic terms (Taylor 1980:276-277). Using procedures for testing taxonomic thought (see Bruner, Olver and Greenfield 1966), and using tests to reflect presence or absence of taxonomic thought, it has been suggested that there is a close correlation between the tendency of children to employ taxonomic-like strategies and the degree of depth and complexity of folk taxonomy 1? a language (Lancy and Strathern 1981:774). Work undertaken by Lancy and Strathem in two New Guinea populations suggested that Ponam children improved their taxonomic thought with age whereas Melpa only improved with education (p. 777). “Melpa aad are in English and advances in learning names of things (as opposed to categorizing) 10 mother tongue should be associated with general improvement in fluency” (p. 778). Melpa appear to mute a taxonomizing tendency as socialization advances (p. 778) and other modes of representation are employed, in tests pairing was interpreted as blocking taxonomizing. In the case of the European tradition, the taxonomic approach is firmly linked to the development of literacy and scientific culture. So, where literacy has eve? just a toehold (as among the Tzeltal) it cannot but help encourage taxonomic expressi0?- Thus, two things become abundantly clear: firstly, that the cognitive and linguistic constructs employed in classifying are varied and combined in different ways in different cultures, and, secondly, that the ways in which they are employed are exceedingly flexible. In addition to hierarchic class inclusion, folk biological classifying activity involves indices, keys, paradigms, typologies (Conklin 1964:39-40), non-hierarchic binary Summer 1986 JOURNAL OF ETHNOBIOLOGY 91 opposition and pairing (Brown 1979:794-795, Lancy and Strathern 1981:782), simple dichotamous division, and possibly other forms as well (Tyler 1978:290). We must also be prepared for categories to be expressed in different ways on different occasions in different places. Thus, pairing may involve a wide variety of principles and these will be realized differently in different cultures: difference, duality, complementarity, alliance, hostility, equality (Lancy and Strathern 1981:788}. Similarly, pairs may be elaborated to form more complex constructs, analogies, series of paired opposites, series of vertical similarity, alternation, and more complex symmetries. These basic relations in ordering social classifications are well-known (Needham 1979). Moreover, items may be assigned to different categories, arranged in different ways, according to different principles, depending on context. In other words, culture enables various forms of alternative orientations, organizations and actions; culture in this sense is a tool-kit (Salzman 1981). Friedberg (1971) has shown this for Bunaq plant categories. For the Nuaulu, it occurs not only in terms of the allocation of certain salient terminal Categories to more inclusive ones (Ellen 1975), but also in the identification of the content of terminal categories themselves (Ellen, forthcoming). How these different modes are employed will vary. Taxonomy will, in certain cultures, be dominant. In some cultures styles of classification may vary between domains, which may have their own special Organizational structure; elsewhere they may vary according to situation. UNDERSTANDING THE GROUNDS FOR PREHENSION Rather than documenting taxonomies or other kinds of classifications and categories as so many butterflies (Leach 1961:2), it is necessary to focus upon the processes which generate them; not detached cognitive processes, but those rooted in particular situations. To distinguish it from the arid abstraction of the notion of classification, we might call this prehension. Prehension refers to those processes which through various cultural and other constraints give rise to particular classifications, designations and repre- sentations. What results depends on the input at all stages in the process (elicitory techniques, etc.) and the interaction of various factors. Prehension stresses the situa- tional bias of classification, whereas cognition and perception suggest purely cerebral processes. Indeed, classification itself may be deemed too narrow a definition of what is involved (Reason 1979), and too easily ends in psychological reductionism, in a discussion of states of mind. Prehension recognizes, without the necessity of qualifi- Cation, the difficulty of distinguishing mind from matter, thinking from doing or speaking, individual from group, cerebral from social, natural from cultural. Thus, prehen- sion entails individual acts of perception, but is not (and cannot be) confined to them. We can only begin to approach a realistic understanding of classifying behavior if we begin by observing people assigning items to categories and using names in natural ethnographic settings, as well as experimental ones. Moreover, if we use experimental techniques which we might reasonably expect to produce particular results, we should try also to devise other techniques which might produce other, different, results. . The structure of prehension is as follows. People bring to situations in which classifying activity takes place, and from which verbal statements about classifying behavior result, information of diverse kinds acquired through both informal and formal socialization experiences of the world in general and of earlier classifying situa- tions. How they then classify depends upon the interplay of this past knowledge (including prescriptions and preferences with regard to particular cognitive and linguistic idioms) with the material constraints of the classifying situation, between conscious and subconscious, the purposes of the classifying act, and the inputs of others. Thus, thinking, saying and doing are not separate activities but inter-penetrating ones, while cognitive bricolage provides us with both models “‘of’’ and models ‘‘for’’ (in terms of 92 ELLEN Vol. 6, No. 1 Geertz’s distinction (Geertz 1966).2 Practical problems do not exist on their own, and in a very real sense all classifications are, therefore, “practical” (see also Hunn 1982). There is a further important aspect of prehension. This arises from the fact that the processing and storage of information in the mind is imperfect, and communication of that information less perfect still. Paradoxically, there is a connection between this short- coming and the considerable capacity of the human mind to re-order information in different ways, replacing irrelevant information with that of greater and more immediate utility. That classifications are messy, cross-cutting and changing is a reflection of this. Consider also the paradox that while the human mind always strives for order, the reality it deals with is so complex that it can never fully attain it. Concepts are often used, operationalized, without defining them. On the other hand, however, in order for communication to take place, classification must have at least some intersubjective structure, some agreed cultural mules, some ‘‘doxa’”’ (Bourdieu 1977). Prehension is an inherently social process. Classifying activity may be solitary or inter-personal; that is we may wish to communicate not simply with others but with ourselves as well. But even solitary behavior is modeled on that hypothetically occuring between individuals. In a solitary classifying act there is no communication with others, but, nevertheless, thought usually takes the form of linguistic expression, though not necessarily verbalized..The problem is that at the point of transformation into the lexical output of language there is a fundamental simplification of cognitive and semantic relations. Linguistic expression necessarily entails both decomposition and facilitation; decomposition because language faculty is unable to encode at one instance the totality of informational relations in the mind, and facilitation because that same complexity must be translated into a form which makes communication and expression possible. It is, if you like, the same as translating from the machine code of a computer, designed for the internal organization and manipulation of information, to a program: ming language. Thus, it is impossible for a sender to communicate everything to a receiver. We must, therefore, distinguish the intended message from the outward signs of the message and it is these latter which will vary depending on who the recipient is. Receivers will encode the message, not in terms of the intentions of the sender, but in the light of the recipient’s own expectations and knowledge. Sender A may say ‘X 1S a kind of Y,” based on the unverbalized information that X and Z are types of Y, although recipient B (with no knowledge of the Y-Z relation) may interpret this statement, through his or her own experience as a fourth item W, to imply that X (like W) is a kind of Y. There is, then, a degre of ambiguity, and people must interpret and operate with respect to the codes and outward signs of others without knowing what inner processes are taking place and the information which generates them (Wallace 1970). We may, therefore, agree (with Reason, n.d.) that ‘linguistic utterance is not, generically precise at all; it is generically sufficiently precise.” Interpretation depends upon whether the person of the same culture interacting in a particular activity will be different from that of a member of the same culture but one who is not prepared by previous mutual experience. The interpretation of a non-member of the culture will be different yet again, and that of the specialist ethnographer especially so. All this will affect the degree of possible ambiguity. Interpretation will also depend upon the questions or commands of the interlocutor (recipient). For example, an individual may get from A to B according to a variety of ad hoc conscious and unconscious procedures, and certainly without the use of any mental map. On the other hand, if asked to draw one post facto there may be no problem, even though the map had no bearing on the original decision. As Reasom (Reason, n.d.:7-8)(see also Crick 1976:159) has pointed out: “ambiguity, ambivalence, metaphorization, are not peripheral and arcane aspects of language use, but central and essential . . . It is hypostatized, reified classificatory usages which require special social conditions to obtain.” What this implies, and what I have tried to focus on here, 18 Summer 1986 JOURNAL OF ETHNOBIOLOGY 93 classifying (as an historically situated activity) rather than an emphasis on classifications. “Classifications,” says Reason in continuing the passage just quoted, “as such are, i at all, only derivatively meaningful.” Indeed, formal representations may be “positively misleading, for they purport to, but cannot incorporate the grounds of such interpreta- tions” (cf. Tyler 1978:290). THE COGNITIVE ACQUISITION OF THE INSTRUMENTS OF PREHENSION The dynamics of prehension cannot simply be understood in terms of the interplay of factors at the instance of classification or verbal expression. The outcome depends on the life experiences of the classifier: learned cultural behavior, personal experience, adaptation and individual socialization. In this respect it is important to acknowledge the significance of materiality and the child’s acquisition of that materiality. Thus, in the early development of every child, bodily discovery, experience and perceptual salience will determine a cultural dominance of front over rear, above over below, hands over feet, and so on (Clark 1973}. Linguistically, the second item in each of these pairs will be accordingly marked. But not only does materiality affect the handling of knowledge, but the experience of time also. Thus, the past appears to markedly dominate the future. This must be is so in two senses. First, past experience has cognitive priority and is only displaced through repeated contrary cases. Secondly, temporal ordering itself serves as a basis for serial signification (as in the contrast set older:younger, or in birth order names}.3 Thus, overall, the dominant cognitive relations brought to prehension are material, historical and biographical, rather than non-material and contemporary. In addition to such processes ingrained in early socialization, we must add the linguistic and classificatory idioms resulting from cultural convention, but what we must not then do is simply to accord to the mind a mechanistic model. The mind itself organizes information extensively in terms of paradigms, and is an active rather than 4 passive system in which images are connected and constantly transformed. What is certain is that unless there is clear and explicit cultural evidence for a total unitary classification of animals, it is as unlikely to be generated subconsciously in the mind as it is to be a logical consequence of the structure of language. For Ardener (1980) “it will be no wonder if we cannot sometimes tease out in real life whether we are dealing with a “social” or a “linguistic” phenomenon. Language is to the social as a measuring rod is to the measured, where, however, the inches or centimetres stretch or contract at the same time as the object itself deforms in related or independent directions.” SEMANTIC UNIVERSALS AND CLASSIFICATIONS IN SOCIETY Once we have understood the process of prehension, and the degree of predictability as to its outcome among particular populations, we can return to the level of societal generalization and cross-cultural comparison, and at once the debate between univer- salists and relativists is seen as the caricature it inevitably must be; an entirely false Opposition sustained through ideological mystification and polemic (Ardener 1982:3; see also Ellen 1979p, Hollis and Lukes 1982 [particularly Gellner}). Classifying behavior does teflect social organization, but the degree to which we can discover close correlations ill depend upon the constancy in application of a particular mode. One possibility is the employment of grid-group forms of analysis pioneered by Douglas, 1982. I have elsewhere (Ellen 1979b) attempted to list the mix of variables for any given society that is likely to affect the structure, content and function of classification. It is possible to demonstrate correlations bet irs of cl istics: between division of labor and classificatory complexity, literacy and arbitrariness, semantic field integration and social 94 ELLEN Vol. 6, No. 1 integration, and so on; linking the formal properties of particular classifications with the substantive ones of the societies in which they are found.* If we could eliminate all other variable we might reasonably expect such horizontal pairs to show a regular correlation. The problem for the anthropologist, and one reason why I am suspicious of attempts to seek constant macro-relations between classifications and types of society defined in terms of vague and general criteria, is that it is not always possible to find predictable regularities in the vertical relations between variables. For example, literacy does not always accompany hierarchy, while rigorous expression of inclusiveness may eliminate anomaly as well as generate it. The pattern observed and the extent to which particular pairs of correlation are evident depends upon the entire nexus 0 variables. The attempt to generate a neat concordance demands care, since for one thing it seeks cultural generalizations on the basis of very limited information about the behavior of individuals. Since, as I have indicated, the process of prehension operates through individuals in the context of collective social experiences, it can hardly be expected to coincide with statistical generalizations or necessarily reflect what is culturally dominant. I detect a confusion of the individual with the collective level in work in the taxonomic tradition, in which culture is assumed to be some mythical omnipresent speaker-hearer, both the sum of its component individuals and a constant from which we might infer the classifying behavior of individuals. On the other hand, we cannot deny that universals can be extracted, though their character must be subject to considerable qualification. Brown's universalist-evolutionary arguments (eg. Brown 1984) have, for example, been severely criticized, along with his polythetic, structurally diverse, and much more likely to involve special purpose significata; as categories become more general so they become more cultural, less biological (Ellen 1977; Hunn 1982:12-16; Randall and Hunn 1984). Similarly, the detailed wiring for some aspects of color classification sheds little light on how basic naming principles shape language, and it seems unlikely that detailed neural specifica tions will find much of a place in explanations of language universals. Howeve!, it 1S possible to detect apparently universal ordering principles underlying the character of lexica: conjunctivity (including binary opposition), criteria clustering, marking, 2m dimension salience (Witkowski and Brown 1978:443-444]. It appears that a “rich cognl tion” model, one permitting both the intrusion of general underlying principles and possibly domain-specific ones, is warranted by the little evidence available. But while such a model is attractive, our attempts to tease out convincing domain-specific semantic universals, other than for color, have not yet met with much success. So, rather than stressing the patent substantive invariance of semantic universals through formalism, it is equally important to stress isomorphic patterns, that is the latent relational aspect (Bateson 1973:615; Lévi-Strauss 1966). It is this, rather than the former, which accounts for the feasibility of cross-cultural communication, and the substantive semantic continuities which give rise to the very real problem of misunderstanding 2 the level of close interactions between individuals. This is so much the case, suggests Ardener (1982:4), that it is itself a human universal. Thus, before we can, with confiden®® make claims for the existence of semantic and lexical continuities, it is first necessary to consider the limits of cultural discontinuity. In other words, the formalists have got it the wrong way round. Rather than making a priori essentialist assumptions {wrappe : ci subject to analysis, we should instead follow the practice of Bayle (Flug 1971:5), se in his Dictionnaire historique et critique sought universals which encompassed conceivable appearances, including the most obscure and atypical. Summer 1986 JOURNAL OF ETHNOBIOLOGY 95 NOTES 'One of the most barren attempts to employ notions of hierarchy and contrast with respect to a particular domain must surely be Stark’s (Stark 1969) analysis of body parts. In a domain (the body) where classifying procedures are necessarily analytic rather than synthetic owing to material continuity of the parts (see Ellen 1977b), the notion of level becomes absurd and entirely arbitrary. Fale . what sense does ‘‘face” contrast with “knee” at all, and why should it not contrast with ead’’? y) . , As representational models are not neutral in their relation to action, it might even be argued that the notion of representation should be avoided altogether. Moreover, Geertz’s distinction is a product of a literate tradition in which representational models and plans for action are more obviously separate. 3} am grateful to Kevin Durkin for pointing out to me that the dominance of past over future is controversial among developmental psychologists and philosophers of time. Thus, we must contrast a “moving time’ view, in which ego is static and time passing by, with a “moving ego” view. There is also some disagreement in the literature as to which of, say, “before” and “after” should be said to be the marked term. Nevertheless, in mundane classificatory events, as in much individual interaction and subsistence decision-making, practical experience (of the past) on (future) action is crucial. 4 Re ; For a specific ethnobiological example see Dwyer, 1979:19, 25. ACKNOWLEDGEMENTS An earlier version of this paper was presented in the Department of Social Anthro- pology of the University of Stockholm, May 1983. I should like to thank Tomas Gerholm and other members of the seminar for their hospitality and comments on that occasion. I am also indebted to Kevin Durkin, Michael Fischer, David Reason and Bob Veltman for their help and suggestions, although they would dissent, for various reasons, from the totality of what I have to say. LITERATURE CITED ARDENER, E. 1980. Comprehending others: some problems of approach. Unpubl., Edinburgh. . 1982. Social anthropology, language and reality. In Semantic Anthroplogy (Association of Social Anthropologists Monograph 22} (D. Parkin, ed.|. Academic Press, London. BATESON, G. 1973. Steps to an ecology of mind: collected essays in anthro- pology, psychiatry, evolution and epistemology. Paladin, St. Albans. BERLIN, B., D. BREEDLOVE and P.H. RAVEN. 1973. General principles of classification and nomenclature in ri biology. Amer. Anthr. 75:214- ae BOURDIEU, P. 1977. Outline of a theory of practice (Cambridge Studies in Social Anthropology 16) (Richard Nice, trans.}. Cambridge Univ. Press, Cambridge. BRIGHT, J.O. and W. BRIGHT. 1965. Semantic structures in Northwestern California and the Sapir-Whorf hypothesis. Amer. Anthr. 67:249-258. BROWN, C.H., J. KOLAR, B.J. TORREY, T. TRUONG-QUANG and P. VOLK- MAN. 1976. Some general principles of biological and non-biological folk classification. Amer. Ethnol. 3:73-85. . 1979. Folk zoological life forms: their universality and growth. Amer. Anthr. 81:791-817. 96 ELLEN Vol. 6, No. 1 LITERATURE CITED (continued) 1984. Language and living DWYER, P.D. 1976. An analysis of things: uniformities in folk classi- fication and naming. Rutgers Univ. Press, New Brunswick, New Jersey. BRUNER, J.S., J. GOODNOW and G. AUSTIN. 1956. A study of thinking. Wiley, New York. OLVER and P.M. a GREENFIELD. 1966. Studies in cognitive growth. Wiley, New york. BULMER, R.N.H. 1967. Why is the casso- wary not a bird? A problem of zoological taxonomy among the Karam of the New Guinea highlands. Man 2:5-25. and J.1. MENZIES. 1972-3. Karam classification of marsupials and rodents. J. Polynesian Soc. 81-82: 1, 4, 472-499, 86-107. 1974. Memoirs of a small Guinea. Journal d’ Agriculture Tropi- cale et de Botanique Appliquee 21: 79-99. , J... MENZIES and F. PARKER. 1975. Kalam classification of reptiles and fishes. J. Polynesian Soc. 3:267-308. CLARK, H. 1973. Space, time, semantics and the child. In Cognitive develop- ment and the qcquisition of | , (T.E. Moore, ed.). Academic Press, New York. CONKLIN, H.C. 1964. Ethnogenealogical method. Jn Explorations in cultural anthropology. (W.H. Goodenough, ed.). McGraw-Hill, New York. 1969. Lexicographical treatment of folk taxonomies. In Cognitive anthropology. (S.A. Taylor, ed.). Holt, Rinehart and Winston, New York. CRICK, M. 1976. Explorations in langu- age and meaning: toward a semantic anthropology. Malaby Press, London. DOUGLAS, M. 1982. Essays in the sociology of perception. Routledge and Kegan Paul, London, Boston and Henley. Rofaifo mammal taxonomy. Amer. Ethnol. 3:425-445. . 1979. Animal metaphors: an evolutionary model. Mankind 12: ELLEN, RF. 1972. The marsupial in Nuaulu ritual behaviour. Man 7: 22.3-2.38. 1975. Variable constructs in Nuaulu zoological classification. Soc. Science Info. 14:201-228. A.F. STIMSON and J. MENZIES. 1976. The content of categories and experience. The case for some Nuaulu reptiles. Journal d’ Agriculture Tropicale et de Botan! que Appliquee 23:125-138. 1976. Structure and incon- sistency in Nuaulu categories for amphibians. Journal d’Agriculture Tropicale et de Botanique Appliquee 23°-125-138. ss _ 1977. Anatomical classifi- cation and the semiotics of the body. In The anthropology of the body (Association of Social Anthro- pologist’s Monograph 15) (J. Blacking, ed.|. Academic Press, London. 1978. Restricted faunas and ethnozoological inventories in Wal- lacea. In Man and Nature in South- east Asia. (P.M. Scott, ed.). London School of Oriental and African Studies. Repr. from Working Papers in Southeast Asian Studies. 10:1-16. 1979. Omniscience and ignorance. Variation in Nuaulu knowledge, identification and class fication of animals. Lang. in Soc. 8: 337-364. 1979. Introductory Essay: In Classifications in their Social Context. (R.F. Ellen and D.A. Reason, eds.). Academic Press, London. 1985. Species transforma tion and the expression of resem blance in Nuaulu ethnobiology: Ethnos 50:5-14. Summer 1986 JOURNAL OF ETHNOBIOLOGY 97 LITERATURE CITED (continued) . (in prep.). The social rela- tions of prehension: a substantivist analysis of animals, categories and society in central Seram —___.._._. {im press), Review ai “Language and living things,” by Cecil H. Brown. Lang, in Soc. ERICSSON, R.A., W.G. CHASE and S. FALOON. 1980. Acquisiton of a memory skill. Science 208:1181- 1182 FLUG, G.P. 1971. The development of a method. History and Theory FOUCAULT, M. 1970. The order of things: an archaeology of the human sciences. Tavistock, London. FOX, J.J. 1975. On binary categories and primary symbols: some Rotinese perspectives. Jn The interpretation of symbolism. (Association of Social Anthropologists, Studies 2) (R. Willis, ed.). Malaby, London. FRAKE, C.O. 1980 (1977). Plying frames can be dangerous: some reflections on methodology in cognitive anthro- pology. In Language and cultural description. (C.O. Frake, ed.). Stan- ford Univ. Press, Stanford. FRIEDBURG, C. 1968. Les méthodes denquéte en ethnobotanique. Journal d’Agriculture Tropicale et de Botani- que Appliquée 15:297-324. ——_______.. 1970. Analyse de quelques sification botanique Bunag. In Echanges et communications: mé- langes offerts 1 Claude Lévi-Strauss (J. Pouillon and P. Maranda, eds.). Mouton, The Hague. , 1971, mig sur la classi- fication botanique Bunaq (Timor Bere oe es ae Francais 118:255 ce Les processus classi- ficatoires appliqués aux objets Naturels et leur mise en évidence: quelques principes méthodologiques. Journal d’Agriculture Tropicale et de Botanique Appliquée 21:313-334. GARDNER, P. 1976. Birds, words and a requiem for the omniscient infor- mant. Amer. Ethnol. 8:446-468. GEERTZ, C. 1966. Religion as a cul- tural system. Jn Anthropoligical approaches to the study of religion (Association of Social Anthropolo- gist’s Monograph 3} (M. Banton, ed.). Tavistock, London. HARRISON, B. 1970. Translation as ee - Philos. Ling. 1:29-30. 7. On understanding a general 1 name. Jn Communication and understanding. (G. Vesey, ed.}. Harvester Press. HEALEY, C. 1978-9. Taxonomic rigidity in folk biological classification: some examples from the Maring of New Guinea. Ethnomedizin 5:3-4, 361-384. HOBART, M. 1982. Meaning or moaning? An ethnographic note on a little understood tribe. Jn Semantic Anthropology (Association of Social Anthropologist’s Monograph 22) (D. Parkin, ed.). Academic Press, London. HOLLIS, M. and S. LUKES, (ed.). 1982. Rationality and relativism. Black- well, Oxford. HUNN, E. 1975. The Tenejapa Tzeltal version of the animal kingdom. Anthropol. Quart. 48:14-30. ______ «1976. Toward a perceptual model of folk biological classifica- tion. Amer. Ethnol. 3:508-524. . 1977. An outline of Sahap- tin bird classification. Omak, Wash- ington: A working paper for the XII jaan — on Salishan Languages, August. "1977. eka folk zoology: the classification of discontinuities in nature. Academic Press, London. 1982. The utilitarian factor in folk ‘biological classification. Amer. Anthropol. 84:830-847. KAY, P. 1971. On taxonomy and semantic contrast. Language 47:886-887. 98 ELLEN Vol. 6, No. 1 LITERATURE CITED (continued) LEVI-STRAUSS, C. 1966. The savage mind. Weidenfeld and Nicolson, London. LAKATOS, I. and A. MUSGRAVE, (eds.}. 1970. Criticism and the growth of knowledge. Cambridge Univ. Press, London. LAKOFF, G. 1972. Hedges: a study of meaning criteria and the logic of fuzzy concepts. Papers from the 8th Regional Meeting, Chicago Linguistic Society, Chicago LANCY, D.F. ae mv STRATHERN. 1981. Making twos: pairing as an alternative to the taxonomic mode of representation. Amer. Anthropol. 83:773-795. LEACH, E.R. 1961. Rethinking anthro- pology. In Rethinking Anthropology (London School of Economics Mono- graphs on Social Anthropology 22) (E.R. Leach, ed.). Athlone, London. MAJNEP, [.S. and R. BULMER. 1977. Birds of my Kalam country. Auck- land Univ. Press, Auckland, Oxford Univ. Press, Oxford. MARTIN, M.A. 1975. L’ethnobotanique science per se? A propos d’un livre de B. Berlin, D.E. Breedlove, P.H. Raven: ‘‘The principles of Tzeltal plant classification.” Journal d’ Agriculture Tropicale et de Botanique Appliquée 7-9:237-276,. NEEDHAM, R. 1979. Symbolic classi- fication. Pepe, Santa Monica, Californi PERCHONOCK, N. and O. WERNER. 1969. Navaho systems of classifi- cation: some implications for ethno- science. Ethnology 8:229-243 RANDALL, R.A. 1976. How tall is a taxonomic tree? Some evidence for dwarfism. Amer. Ethnol. 8:229-242. E.S. HUNN. 1984. Do life- forms evolve or do uses for life? Some doubts about Brown’s univer- sals hypothesis. Amer. Ethnol. 11: 329-349. REASON, D. 1979. Classification, time and the organization of production. In Classifications in their social context. (R.F. Ellen and D. Reason, eds.). Academic Press, London. _ n.d. On the essential futility of formal ‘representations of folk classifications. Unpubl. manu. ROSCH, E., C.B. MERVIS, W.D. GRAY, D.M. JOHNSON and P. BOYES- BRAEM. 1976. Basic objects in natural categories. Cognitive Psych. 8:382-439. SALZMAN, P. 1981. Culture as enhabil- mentis. In The structure of folk models (Association of Social Anthro- pologist’s Monograph 20) (L. Holy and M. Stuchlik, eds.). Academic Press, London. SPRADLEY, J.P. 1979. The ethnographic interview. Holt, Rinehart and Wins ton, New Yor STARK, L.R. 1969. The lexical structure of Quechua body-parts. Anthropol. Ling. 11:11-15. SUPER, C.M., 5. HARKNESS and L.M. BALDWIN. 1977. Category behaviors in natural ecologies and in cognitive texts. Quart. News. hang for Compar. H TAYLOR, P.M. 1980. ee ethno- biology: the folk classification of “biotic forms.’ Unpu h (Anth.) Dissertation. Yale Univ., New Haven, Connecticut. TURNER, NJ. 1974. Taxonomic systems and ethnobotany of three conten porary Indian groups of the ie Peers (Haida, Bella Coola, an Lilloet). Syesis 7, Supplement L TYLER, S. 1978. The said and the unsaid. Academic Press, London F WALLACE, A.F.C. 1970. Culture an personality. 2nd ed. Random House, New York. WIERZBICKA, A. 1984. Apples are not a “kind of fruit: the semantics sf human categorization. Amer. Ethnol. 11:313-328. ¥ WITKOWSKI, S.R. and C.H. BROW 1978. Lexical universals. Annu. Rev- Anthr. 8:353-371. J. Ethnobiol. 6{1):99-120 Summer 1986 TOPICS AND ISSUES IN ETHNOENTOMOLOGY WITH SOME SUGGESTIONS FOR THE DEVELOPMENT OF HYPOTHESIS-GENERATION AND TESTING IN ETHNOBIOLOGY DARRELL ADDISON POSEY Laboratério de Etnobiologia Departamento de Biologia Universidade Federal do Matanhdo 65,000 Sao Luiz, Maranhdo (Brazil) ABSTRACT.—This paper defines ethnoentomology, briefly traces the history of the field, and afferc ] f, 1 4 h surveys the literature in major subject areas gg Hypothesis-g ion/testing is suggested as an important “intellectual bridge” to a world science that builds upon knowledge syst fallh ieties. Exampl P ted INTRODUCTION Definitions, even for ethnoentomology, are often difficult to formulate, and, once formulated, are usually unsatisfactory. Insight and understanding is sometimes increased through a comparison with a related term or concept, hence the juxtaposition of “cultural entomology” and “ethnoentomology” in the discussion that follows. Cultural entomology treats the influence of insects upon the ‘essence of humanity as expressed in the arts and humanities” (Hogue 1980). Cultural anthropologists usually restrict their studies to “advanced,” industrialized, and literate societies, maintaining that entomological concerns of “primitive” or ‘‘noncivilized” societies are in the domain of ethnoentomology. They are principally interested in written forms of cultural expres- sion and limit their studies to physically recorded sources of literate societies. It is well to note that this, like many divisions, is an artificial one, and it implies an ethnocentric “welthey’”’ bias built upon assumptions of fundamental differences between ‘‘primitive”’ and “civilized” classification and thought. Thus far, anthropological research has not substantiated such assumptions. Although the prefix “ethno” generally indicates knowledge of “folk” societies and the word cell “ento” refers to insects (thus ethnoentomology is concerned with the knowledge and use of insects in different human societies), defining the term is not as easy as might be expected. A fundamental problem is that of delimiting entomology itself. Even though the concept “insect” is clearly defined by Western science, entomologists also frequently study “related arthropods.” Since these two concepts gradually developed in Western science, it cannot be assumed that they are universal and, in folk societies, must be elicited using emic procedures that “discover” conceptual paradigms rather than methods that impose preconceived concepts upon the society under study. There are a number of areas within ethnoentomology which can be successfully researched through analyses based upon observations and data collection using the Categories of Western science, ie., using the etic approach, without diminishing their ethnoscientific contribution. Examples include studies of insects as food, the role of in disease transmission, hallucinogenic insects, the use of insects for omamen- tation, problems in contamination of food with insects, etc. Few studies have passed 100 POSEY Vol. 6, No. 1 from the etic to the cognitive emic level. Yet the native (folk) view of insects—their naming, classification, and use—is surely the ultimate goal of ethnoentomology. This paper gives a general survey of both emic and etic topics in ethnoentomology, utilizing the general Western concept “insects and related arthropods” as a unifying category for comparative study. Cultural entomology is treated as a subdivision of ethnoentomology that deals with recorded sources in literate societies. Cultural entomological interests will, therefore, be incorporated throughout the paper, although no attempt is made to review the vast literature. The purpose of this review is to outline areas of interest for ft g investigation, with an attempt made to establish ethnoentomology, and ethnobiolo in general, as a hypothesis-generating and testing mechanism. That is, to show how folk knowledge and beliefs can serve to generate new ideas and hypotheses which can then be investigated and tested by our own science. This approach provides an intellectual bridge between Western and folk sciences as well as the basis for a non-culturally biased world science. The paper argues that folk specialists must be treated as scientists, with their respective systems regarded as invaluable codifications of human observations of natural phenomena. ~~ 1 1 ae | A BRIEF HISTORY OF ETHNOENTOMOLOGY Development of entomology as a folk science has been traced for Egypt (Efflaton 1929}, the Middle East (Harpaz 1973), Greece and Rome (Scarborough 1979), and other parts of the world (Essig 1931; Montgomery 1959; Wilson and Boner 1937). Modern entomology acquired a distinctively humanistic flavor (and perhaps its “ethno” ten- dencies) from entomologist-philosophers such as William Morton Wheeler, Maurice Maeterlinck, and Jean Henri Fabre, who “not only described insect phenomena with imagination and brilliance, but wrote and spoke of their meaning on a human intellectual plane” (Hogue 1980). Contemporary ethnoentomology began in the Nineteenth Century with the works of Wallace (1852), Daoust (1858), Bates (1862), Hagen (1863), Katter (1883), Librecht (1886), Glock (1891), Marshall (1894), and Wagner (1895). Writings by Armbruster (1926), Arndt (1923), Barrett (1925), Caudell (1916), Dammerman (1929), Ealand (1929), Gudger (1925), Knortz (1910), Laufer (1927), and Nordenskiold (1929) brought the subject into the Twentieth Century. Essig’s (1934) survey of the importance of insects to the Indians of California established the traditional categories of ethnoentomological interest. Zinsser’s (1935) Rats, Lice and History remains a classic because of its perspective of insects as forces in human social and biological history. Insects as Human Foo (Bodenheimer 1951) likewise brought insects to world attention in a more positive light as a potential and important source of protein. Wyman and Bailey (1952) were the first to use the term ethnoentomology in print in their seminal work on the Navajo Indians. The writings of Schimitschek (e.g. 1968, 1977) certainly establish him as a major force in cultural and ethnoentomology. Other general works include those by Clausen (1954), Cloudsly-Thompson (1976), Hitchcock (1962), Hogue (1980), Kevan (1974, 1979, 1980), Posey (1976, 1977, 1978, 1979, 1980, 1986), and Ritchie (1979). Conklin’s (1973) general bibliography of folk classification offers an important section of entries on ethnozoology (including ethnoentomology) and provides 4 bibliographic framework to link ethnoentomology with its theoretical roots 17 ethnoscience. INSECTS AND HUMAN HISTORY a Zinsser’s (1935) work popularized the knowledge of the association of insects with Aiesk ef emg Pe, Kmart. meses he iy He Pi oe . ee {himman ae tt, a, a, i ma Summer 1986 JOURNAL OF ETHNOBIOLOGY 101 history. Subsequent works (Cloudsly-Thompson 1976; Hare 1954; McNeill 1976; Ritchie 1979; Sigerest 1951; Smith 1973) trace the plagues and pestilence caused by insect-borne diseases such as bubonic plague, typhus, yellow fever, and trypanosomiasis. Bushvine (1976) details the effects of ectoparasites on human hygiene and medical history. Crosby (1972) analyzes the complexities of trans-Atlantic exchanges of insect- transmitted diseases and emphasizes the destructive impact of such on aboriginal popula- tions of the New World. Often such devastation extended well into regions with no direct contact with Europeans. This was due to extensive aboriginal trade routes that brought goods infested with insect vectors deep into the hinterlands (Posey 1976). The complete impact of insect-related diseases is still little known for the Americas (Dobyns 1966). Certainly the role of insects in human evolutionary history is indisputable. Students interested in this broad area should begin their studies by consulting the bibliographies of the above works. INSECTS AND HUMAN FOOD € most extensive literature in any subject of ethnoentomology concems the relationship between insects and human food. The study of Entomophagy, the direct use of insects as human food, has a long and varied history. Why Not Eat Insects? (Holt 1885) stimulated a series of studies concerning the nutritional potential and importance of insects to the human diet. Subsequent general surveys (eg., Bergier 1941; Bodenheimer 1951; Conconi et al 1981; Curran 1939; Dufour 1981; Gorham 1976a,b; Harlan 1976, Hoffman 1947; Meyer-Rochow 1973, 1975, 1976, 1985; Ruddel 1973; Taylor 1975) have investigated the variations in cultural practice of entomophagy. Other studies have documented the biological efficiency of insect reproduction and the consequent Production of protein (DeFoliart 1975; Dufour 1981; Meyer-Rochow 1975, 1976). Recent works discuss the practical problems of insect foods for Western societies, including socio- €conomic factors, manpower, preparation, handling, and marketing (Conconi 1982; Dufour 1981; Gorman 1979; Ramirez et al 1973; Kok 1983). Insects are also consumed indirectly through the ingestion of contaminated foods. This is because of the impossibility of complete removal of insect parts from food pro- ducts (Caron 1978). Contamination necessitates the establishment of a complex set of rules and standards utilized by government food- and drug-regulating agencies (Taylor 1975). Detailed works outline the hazards of insect ingestion which include allergic reac- tions, poisoning, tumorigenic stimulation and related health problems (Chooviva- thanavanich et al 1970; Dufour 1981; Gorham 1975; Pimental et al 1977; Taylor 1975). The major factors affecting insect consumption are not health hazards, however, but cultural biases. Bodenheimer’s (1951) book on insects as human food stimulated a Series of works regarding cultural traditions and taboos of insect eating (eg., Aeschlimann 1982; Catley 1963, Meyer-Rochow 1973, 1978; Ruddel 1973; Taylor 1975; Tihon 1946). Although Western societies have a particularly strong bias against insects as food, with honeybees being the only Arthropod systematically exploited for human food (Dufour 1981), many other societies have a long and extensive inventory of useful and edible species (eg., Aldrich 1921; Catley 1963; Daoust 1858; Tindale 1953; Wallace 1852; Silow 1976, 1983). Techniques for the evaluation of insect nutritional qualities have been developed (Coconi 1977; Coconi et al 1981, 1984; DeFoliart 1975; Tetotia and Miller 1974) and allow for the generation of numerous lists of species and their dietary poten- tials (eg., Dufour 1981 ; Redford and Dorea 1984; Taylor 1975). Such techniques are not without problems and refinements in protein and nutrient evaluation are still needed (Redford 1986). | Other indirect effects of insects include the enormous cost of agricultural chemicals used to control them. A dramatic worldwide increase in mechanized monocultural 102 POSEY Vol. 6, No. 1 planting has led to sharp rises in epidemic outbreaks of insect pests and the resultant rise in crop loss (Altieri 1983; Cooper and Tinsley 1978). This trend, combined with soaring energy costs, has created serious global problems and threatened the stability of food prices in both developed and under-developed countries (Altieri 1985). Other farming trends, such as “no-till” planting, have led to increased vulnerability to some crop pests and greater dependency on herbicides; these herbicides, in turn, often increase the susceptibility of some crops to other insect and microbial pests (Oka and Pimentel 1974). This situation stimulates even greater dependency upon insecticides. All of these have contributed to re-establish the viability of traditional agriculture and the necessity of studying folk agriculture in detail. Apart from the high costs of chemical agents, health hazards are alarming and ominous. Fatal and non-fatal poisoning from pesticides is common, and the long-term effects of ingestion through contaminated food and water are of wide concern (Gorman 1975, 1979; Pimental et al 1977; Taylor 1975). Another problem is that for cosmetic and psychological reasons many types of insects are idered repulsive (Hosen 1980). That certain species are relished as edible delicacies in one society and viewed with dread or horror in another is a question of cultural ‘‘tastes.’’ Yet as Dufour (1981) points out, attitudes toward insects can change. As world food supplies dwindle and long-term space travel becomes a way of life, consumption of insects may have to become acceptable (Pimental 1976). Insects are innovative marketing to introduce insect-based products (Dufour 1981). INSECTS AND MEDICINE It is in China that we find the most ancient and complete record of the use of insects in medicinal preparations. Read (1935) gives a detailed inventory of useful medicinal species. Chinese veterinary medicine had evolved to a point that curative diets and remedies were used to treat ailing crickets and silkworms (Laufer 1924; Read 1935}. Scarborough (1974b, 1981) gives evidence of the importance of certain insects 10 ancient Greek and Roman medicine. Numerous other surveys record diverse examples of insects in various cultures (Kevan 1979, Greenlee 1944, Meyer-Rochow 1985; Posey 1978; Swanton 1928). For example, in Brazil termites are used to treat bronchitis, catarrh and influenza, constipation, dog bite, goiter, incontinence, measles, protruding umbilicus, rheumatism, whooping cough, sores, boils, ulcers, etc. The treatments range from teas made from crushed insects or their nests to inhalation of smoke from burning termite cartons (Mill 1982). Principal insect groups listed in Brazil by Lenko and Papavaro (1979) for the variety of their medicinal uses are cockroaches (to treat alcoholism, asthma and bronchitis, colitis, constipation, tooth ache, etc.) and wasps (for stomach ache, wounds, spider bites, constipation, burns, etc.). Bees are important in Kayapé Indian medicine. Different honeys are though to have different medicinal properties and are used for a variety of diseases. Pollen (collected by bees), larvae and pupae likewise have medicinal qualities. Smokes from different waxes are the most important and powerful curative substances: patients are either “bathed in the smoke or inhale it. Houses are also “cleansed’”’ by smokes from burnt beeswa%, batumen, and resin (Posey 1983b, e, f; Posey and Camargo 1985). Mixtures of wasps are though to be aphrodisiacs. Parts of the horns of the rhinoceros beetle (Megasoma acaeon, Dynastidae) are thought to give great sexual strength (Lenko and Papavaro 1979}. Ant and wasp infusions are widely used to cure goiter, paralyses, Summer 1986 JOURNAL OF ETHNOBIOLOGY 103 and rheumatism (Ealand 1915}. Perhaps most amazing is the use of stinging ants and wasps as cures for crippling arthritis (cf. News and Comments, Journal of Ethnobiology 3(1):97, 1983). Stings from these Hymenoptera are apparently effective in curing arthritis. Cures for certain types of blindness are also attributed to wasp stings (Aratjo 1961). The Uapixana and Tirié Indians also use ant stings to cure various maladies (Lenko and Papavaro 1979). In Brazil the enormous mandibles of Atta are used to suture wounds. The ants are allowed to bite the sides of the wound; when they close their jaws, their heads are broken off and the closed mandibles hold the would together (Gudger 1925). Other diverse uses include Tindale’s (1953) observation of grubs being used in Australia by the aboringines as a “substitute teat” to wean their children. The aborigines also commonly treat stomach ache and colds with a liquid prepared from crushed green- tree ants and larvae (McKeown 1944); crushed cockroaches are used to treat cuts (Rudell 1973). Clousdly-Thompson (1976) provides a long list of medicinal uses of insects in the ancient and modern world. BEEKEEPING AND INSECT REARING From ancient cave paintings of honey raids to beekeeping in Babylon, and the use of beeswax to embalm the dead in Assyria, Ransome (1937) traces the importance of “the Sacred Bee” in ancient times. Crane (1984) describes the “archaeology” of bee- keeping recorded in historical texts and art. Crane (1979) also provides a comprehensive survey of production, collection and use of beeswax in many parts of the world. The literature is so extensive on the keeping of Apis in ancient and contemporary times that it cannot be reviewed herein. Keeping of stingless bees (Meliponinae} is a much less known area of ethno- entomology. Yet the keeping of meliponine species was a highly developed science amongst native peoples of Africa and the Americas (Parent et al 1978; Schwarz 1945, 1948). Schwarz (1948) provides one of the most complete studies on the domestication of meliponines by the Maya of Central America. These Indians were expert in the genetic manipulation of different bees to increase honey and wax productivity and perfected many methods for the division of colonies and the rearing of numerous species. Highly omamented man-made hives were employed in special shelters constructed for the sacred bees and bee gods. The Maya had several methods to attract and “tame” wild swarms, which included attraction with plantations of flowering plants preferred by the bees. Such practices continued into modern times and are still observed in Mexico, Panama, and other parts of Central America (Bennett 1964, 1965; Hendricks 1941; Weaver and 1981; Weaver 1981). ; Stingless beekeeping was also highly developed in pre-Colombian South America. Nordenskiéld (1929) provides an interesting survey of South American folk apiculture observed during the first half of this century. For stingless bees in Brazil, Lenko and Papavaro (1979) record 171 folk names, the Majority of which were of indigenous origin. Many scientific names are actually taken directly from their Tupi Indian language origin (Nogeira-Neto 1970). ee though some species of Meliponinae were undoubtedly fully ted in South America, many species were only semi-domesticated. Chagnon (1968) and Metraux (1948b) describe bee management by the Yanomamo and Guarani Indians, but they fall Short of describing the bees as fully domesticated. Contemporary Kayapé Indians of Brazil name and classify at least 56 folk species of stingless bee (Posey 1983e, f}; nine species are semi-d ticated. Hives of these species are raided, and a portion of the brood comb (with some honey and pollen) is returned to the next before resealing, so that the bees re-establish the colony and the Indians : +4: rey Continue t years. In addition to the nin = 104 POSEY Vol. 6, No. 1 species, the Kayap6é mark the nests of several others and carefully observe the progress of the colonies. When the Indians think that the quantity of honey and/or wax suffi- cient, the nest is raided. Nests of some species, when found in the forest, are actually —— to the village to be observed on a daily basis (Posey and Camargo 1985). Other to be “managed” include the sauva (Atta spp.) (Lenko and Papavero en various wasp species (Baldus 1937; Chagnon 1973; Metraux 1984a), and honey- producing — (Brachygaster) (Lenko and Papavaro 1979). \ are raised by several South A i tribes (Chagnon 1968; Stewart and Metraux 1948). Palm trees are deliberately cut to provide a fodder for egg- laying adults. The Indians know exactly when to return to the decaying pith to extract the numerous, large grubs. Coimbra (1984) offers detailed information on the rearing of four species of Bruchidae and Curculionidae larvae by the Suri Indians of Rondonia (Brazil). A sizeable bibliography exists regarding insect-rearing for laboratory experiments as well as livestock food (Calvart et al 1969; Chambers 1977; MacHargue 1917; Vander zant 1974). One interesting study deals with the commercial management of Hermetia illucens larvae, which are used as fish bait in Brazil (Santos and Coimbra 1984). Insects have been — for a number of purposes. Laufer (1927) describes in detail how el houses were prepared for singing and fighting crickets with China (Fig. 1). Special shelters were prepared for them, during the summer, with clay beds built for each individual. Elaborate diets were recognized for different species in different lunar cycles. Special diets and medicines were available for ailing crickets. Intricate porcelain dishes were created for feeding the prized individuals. Even delicately carved “ticklers” were created to urge reluctant cricket warriors to battle. — (1979) notes that American Indians likewise raised crickets “simply to enjoy their songs.” Similar reports are found in Bates (1862), Caudel (1916), and Floericke (1922). Posey and Camargo (1985) report on the keeping of stingless bees purely because of the Kayap6 Indians’ fascination with social insects. Of course, cultivation of silk worms (Bombyx FIG. 1.—Special basket kets in Asia. Courtesy of Dr. Nelson Papavero, Museu de set Key ‘S30 Paulo, Brazil. Summer 1986 JOURNAL OF ETHNOBIOLOGY 105 mori) is another ancient Chinese tradition in insect rearing. The details of this folk science are described by Read (1935) and Cloudsly-Thompson (1976). Scale insects (Coccus cacti) that feed on prickly pear (Opuntia spp.) are still raised in Mexico, Honduras, the Canary Islands, Algeria, Spain, and Peru because of their use in the production of the carmine-red pigment cochineal (Cloudsly-Thompson 1976; Ealand 1915). Similarly, “lac insects” (Laccifer lacca) are reared in Thailand, Burma, and India for their production of shellac, polishes, and sealing wax (Cloudsly-Thompson 1976). PESTS AND PEST CONTROL Was it a tarnished plant bug that caused the potato rot which led to the great Irish potato famine? Quite possibly, according to Wheeler (1981). If so, it probably was not the first ecological and social disaster wrought by insect pests. Although the roles of insects in agricultural history remains little known, we do know that today worldwide deforestation and the dominance of crop monocultures have provoked a sharp rise in insect pests (Cooper and Tinsley 1978; Thresh 1982). Likewise there have been dramatic increases in insect-borne diseases caused by blood parasites and arboviruses. These situa- tions, along with the high costs of pesticides and energy for their applications, have Stimulated a refreshing new emphasis upon studies of pest management in traditional agriculture (Altieri 1985). Western agriculturalists have generally assumed traditional agricultural systems to be of low productivity and have used “bigger yield” as their justification for expensive technologies and chemical dependency (Alverson 1984). Yet in many cases, native agriculture has been shown to be both productive and efficient in its use of local skills, available energy, and materials (Egger 1981; Kerr and Posey 1984; Parker et al 1983; Posey 1983c, d; Wilken 1977}. One of the major reasons for this effectiveness is efficient pest Management. Traditional cropping systems have “built-in suppression mechanisms” (Altieri 1983a, b). These include: (a) arrangement of crops, (b) composition and abundance of non- Crop vegetation in and around fields, (c) genetic diversity of domesticates and semi- domesticates, (d) matching of soil varieties with crop varieties, (e) ‘natural corridors” between fields, and (f) variation of field sites and long-term management of old fields (Altieri 1985; Denevan 1971; Denevan et al 1984, Parker et al 1984; Posey 1984). Brown and Marten (1984) point out that crop losses in native fields may be as high as 40%, and such losses are still within the range of losses in moder agriculture using pesticides. One major difference exists: elimination of pesticides from modern systems can produce losses approaching 100% (Schwarz and Klassen 1981), whereas pest ¢ in traditional agricultural systems almost never exceeds reasonable bounds (Altieri 1985). The relationship between agricultural polycultures and lower pest incidence is currently under investigation (Altieri and Letourneau 1982; Perrin 1980; Risch et al 1983). Maintenance of a broad genetic base certainly diminishes attacks from host-specific pests (Brush 1982; Gliessman et al 1981; Pimentel and Goodman 1978). A variety of manage- ment techniques have been described for different societies. Use of resistant native cultivars, crop rotation, variation in planting times, and use of shade to shelter useful insects are only a few keys to successful traditional agriculture. In Nigeria, for example, okra is planted to divert flea beetles (Podagria spp.) from cotton (Perrin 1980). Variations in relative corn and bean planting dates are also used to reduce leafhopper and armyworm ge (Altieri and Letourneau 1982). Many studies detail other management techniques (Altieri 1983a, b, 1985, Bunting 1972; Glass and Thurston 1978; Golob et al 1982; Huis et al 1982; Khan et at 1978, Litsinger et al 1978a, b; Matteson et al 1984; Wilken 1977). The effects of spatial arrangement, eg. row spacing, are still little known but appear to have significant impact on pest management. Matteson et al (1984) document a signi- 106 POSEY Vol. 6, No. 1 ficant difference in crop loss between cowpeas (Maruca testulalis) planted in intra-rows rather than inter-rows with maize. Kerr and Posey (1984) report that interdispersal of aria (Calathea aloua) with tuber crops reduces nematodes and Collembola-bome virus attacks in Kayapo fields. Management of “weeds” is also an important factor in the overall practices of tradi- tional agriculturalists. “Relevant weeds,” according to Altieri (1983a, b; 1985), support rich natural enemy fauna that provide alternative prey/hosts, pollen or nectar, or favorable microhabitats unavailable in weed-free fields. Most of what Western agriculturalists would consider “weeds” in a Kayapé field are, in fact, useful semi-domesticates for the Indians (Posey 1986). Altieri and Letourneau (1982) provide examples of cropping systems in which the presence of weeds has enhanced biological control of insect pests. Western science has only begun to seriously study traditional agricultural science. Yet existing evidence already points to the richness of ideas and data available to interested researchers (Brokenshaw et al 1980; Posey et al 1984; Parker et al 1983). Some laboratories have begun the serious study of toxicological potentials of native pesticides. Results have been promising (Ganjian et al 1983; Kubo and Matsumoto 1984; Kubo et al 1984). Other entomologists, agriculturalists, and ethnoentomologists should devote more attention to the investigation of integrated pest management by native peoples. MYTHOLOGY, RITUAL, AND ‘““NATURAL MODELS” Reports of insects in mythology and ritual are widespread. Bushnell (1910) and Mooney (1972) discovered many insects as key figures in the belief systems of Indians of southeastern North America. For the Louisiana Chocktaw, for example, grasshoppers and men were created at the same time and were once brothers; ants were likewise con sidered as having human ancestors (Bushnell 1910). Ant clans existed in many tribes (Gilbert 1943; Grinnel 1899) and ant people were thought to have been the first to have inhabited the underworld (Bushnell 1910). Water beetles (Hydrophilidae) were respon sible for the formation of the earth because they had brought up the mud from beneath the waters to form the first dry land (Mooney 1972). The Cherokee attributed the origin of “Sacred Fire” to the heroic efforts of the watet spider, which brought fire on its back while corssing the ocean (Mooney 1972). Diseases and crop pests were cast upon people, according to legend, by Grubworm, who organize his fellow insects to punish humans for their abuse of nature (Posey 1977; Swanton 1928). Insects also play an important role in Australian aboriginal lore (Meyer-Rochow 1985}. Spencer and Gillen (1899) reported 30 insect totems; Berndt and Berndt (1984) provide further evidence of insect-named clan and totemic groupings. One of the majo cosmogenic myths of the aborigines refers to the famous witchey grub that served - humankind’s first food. Numerous examples of insects in mythology can be found 1n various compendia (eg., Armstrong 1970; Clausen 1954;, Cowan 1865; Denton 1968; Ealand 1915; Griaule 1961; Kevan 1974, 1979, 1980; Posey 1978, 1980, 1981; Reim 1962 Ritschky 1981; Schimitschek 1968, 1977; Wyman 1973). Insects are also important components in many ceremonies. The Cherokee shamans employed many insect names in their sacred chants (Kilpatrick and Kilpatrick 1970) and had an elaborate “extraction” ritual to remove disease-causing insects from their patients bodies (Greenlee 1944; Lawson 1937; Morphi 1932). Fortune-telling rituals used insects as indicators of the future (Mooney 1972). In Brazil, one of the most dramatic rituals is that of the Maue marriage ceremony (Biard 1862). Young boys are submitted to an ordeal of pain in which tocandeiro ants (Paraponera clavata), known for their extremely powerful stings, are placed into a woven mitt (Fig. 2). The boys receive dozens of pa stings when they don the ceremonial glove. When the severe swelling subsides from his arm, a body is considered free to marry Summer 1986 JOURNAL OF ETHNOBIOLOGY 107 HIG. 2.—Ceremonial mitt used by the Maué Indians of amazonia. Stinging ants are placed in the mitt, which is then worn by boys in the marriage ceremony. Courtesy of the Museu Paulista, S40 Paulo, Brazil. Accounts such as those previously related are generally recorded out of cultural context and, consequently, are of limited significance to the folk entomologist. Recent studies (eg., Brown and Chase 1981 ; Gregor 1983; Luhrmann 1981; Malkin 1956; Posey 1985; Waddy 1982; Wilbert 1981) have attempted t ide a broader cultural framework for the interpretation of insects in myth and ceremony. Natural ‘‘models” based upon insect examples and recognized by native peoples themselves have been shown to be useful for the organization of folk scientific data and are significant alternatives to the imposed models of traditional and Western science (Posey 1981}. ORAL LITERATURE AND ECOLOGICAL KNOWLEDGE Oral literature is a major vehicle for ecological information. Santos and Posey (1986) witnessed an old man in the Ilha de Lencois (Brazil) describing his pursuit of a mythological animal. The “plot’’ of the story could take only three minutes to relate, but the master story-teller kept his audience of local youth spellbound for nearly 45 minutes. Analysis of the folk tale reveals that the minute details used to embellish and add credence to the story is also instruction in local ecology and survival. . Myths are concentrated symbolic codes that transmit cultural information, including social rules and standards of behavior. Ecological information, such as knowledge of animal behavior and “coevolutionary complexes,” can also be communicated in myth form (Posey 1983c). Baldus (1937, 1970) recorded Taulipang myths describing the com- mensal relationships between birds and wasps. Kayapo lore describes commensalsim between stingless bees and acrids (Posey and Camargo 1985). Mill (1982) points out that widespread stories of “weeping termites” in Brazil reflect folk knowledge of the biological fact that ground nesting termites (Nasutitermes, Velocitermes, and Cortaritermes) exude droplets of exocrine secretions for chemical defense when disturbed. A Kamaiura myth describes termite nests that glow in the night (Villas-Boas and Villas-Boas 1972). These glowing nests are not superstitious nonsense, but rather recognitions of a natural phenomenon caused by periodic invasions of termite mounds by phosphorescent Lampyridae larvae (Redford 1982). aie Oral literature has not been sufficiently studied as a transmitter of biological infor- mation. This is because of the highly symbolic language of myth and folklore, which is frequently considered as nonsense by those who do not understand the linguistic and Cultural codes. Researchers who take the time to learn the language of the societies they 108 POSEY Vol. 6, No. 1 study and prepare themselves with training in folkloristics can indeed make a signifi- cant contribution to myth interpretation and ethnoentomology. MISCELLANEOUS TOPICS Several miscellaneous topics deserve mention. Brief examples will be given to illustrate each topic. Use of insects as ornaments and decorations—Berlin and Prance (1978), Covarrubias (1971), Kennedy (1943), Lothrap (1964), and Outram (1973) review the importance of insects in art and ornamentation in the New World. Meyer-Rochow (1975) reports the use of green tenebrionid beetles, as well as scarabaeids and buprestids, among the Wahgi Valley people of New Guinea. The Kayapé inherit the right to use irridescent elytra of Euch goliatha and make elaborate ceremonial hats from meliponine batumen (Posey 1983e). Butterfly wings are commonly used in the Americas for adornment and decora tion (Posey 1986). Klots and Klots (1959) report the use of luminous beetles (Pyrophorus spp.) to decorate the hair of Indian girls. Insects as objects of entertainment—Dragonfly catching is a favorite and developed sport in the Banda Islands (Simmons 1976). Butterfly wings are important play objects for Trobriand Island youth (Meyer-Rochow 1985). In Papua New Guinea, large weevils (Rhyn- chophotus ferruginius) are used as musical instruments by letting the human mouth serve as a variable resonance chamber for the wing vibrations of the beetle (Meyer-Rochow 1973). Staged fights between lucanid beetles are reported in Thailand (Meyer-Rochow 1975}. Posey and Camargo (1985) report the keeping of stingless bees by the Kayapo Indians purely because of their fascination with social insects. Lenko and Papavaro (1976) give several examples of the keeping of Pyrophorus spp. beetles for entertainment, 4S well as for their light. Dances inspired by insect movements are reported in several North American Indian groups (Bushnell 1910; Gilbert 1943; Schoolcraft 1851; Swanton 1928, 1946). Insects as indicators—Due to the sensitivity of the head louse to minute changes in body temperatures, some native groups diagnose illness of patients by the presence Or absence of lice. Slight fevers can cause an exodus of body lice that indicates oncoming illness (Malinowski 1929; Raths and Biewald 1974). Absence of certain insects can be taken as a sign of envi 1 pollution (Englehardt 1959), while the presence of other species (such as green flies that are attracted to decaying matter) can indicate unhealthy condi tions (Meyer-rochow 1985). Water striders, for example, are indicators of polluted water to the Trobrianders (Meyer-Rochow 1985). Meyer-Rochow (1985) reports how Australian aborigines use the contents of spider webs to indicate the proximity of honey bees. The presence of “mutucas” (Tabanus) near river banks indicates the presence of game © indigenous hunters of Brazil (Lenko and Papavaro 1976). Insects as UFOs—Many tribes in Papua New Guinea report the presence of “flying light spots” in areas where lumi t are not reported by entomologists (Callahan and Mankin 1978). These sightings may be explained by flying insects that enter into electric fields caused by thunderheads; the result is that ordinary insects appear to ea sparks” (Meyers-Rochow 1985). Insects and utilitarian concepts—Insects are frequently used as fish bait in preliterate cultures (Kevan 1979). Nasutitermes mounds are used for construction material ' Brazilian Indians, who prize the natural insulating qualities of the nests extensive galleries (Posey 1979). Nests of Azteca ants are buried with newly planted crops to stimulate plant growth (Kerr and Posey 1984). Beeswax and batumen are eitenaively used for artifact production and paint bases by indigenous tribes of South America (Crane 1979, 1984; Summer 1986 JOURNAL OF ETHNOBIOLOGY 109 Posey 1978, 1980; Schwarz 1945, 1948). In Papua New Guinea and Northern Australia, ants and maggots are used to clean skeletons and bones. Cantharid beetles are the source of poisons for arrow points for some South American Indians (Meyer-Rochow 1985). Hallucinogens—Insects have been found as sources of hallucinogens used by some indigenous groups (Meyer-Rochow 1985}. It is unclear if the hallucinogenic properties are due to the insects themselves or the plant sources upon which they feed (Blackburn Insects and Archaeology—Insects are frequently found in archaeological sites. The presence of seasonal species has been shown to be useful to the archaeologist in deter- mining seasonality of site use and the historic ecological setting (Gilbert and Bass 1967; Hevly 1982; Hevly and Johnson 1974). Urban Ethnoentomology—A current topic in entomology is the ecology of insects in urban environments (Frankie and Ehler 1978). Studies in this area focus upon the adap- tations of insects to the special climatic and edaphic conditions created by intense human manipulation of the natural environment. ‘‘Synanthropy” describes the nature of this Coexistence with humankind over an extended time (Povolny 1971); a formula for determining the degree of synanthropy has even been developed (Nuorteva 1963). This specialized area of human-insect relationships might also be called “urban ethnoentomology.” HYPOTHESIS GENERATION: THE ETHNOBIOLOGICAL BRIDGE The interdisciplinary Kayapé project has developed methodological procedures to scientifically test hypotheses generated through its ethnobiological investigations of indigenous ecological knowledge (Posey 1986). Native concepts and beliefs are used by Westem scientists as emic guides for their research designs (Posey 1983a, 1985}. Data collection utilizes indigenous categories for floral and faunal inventories, while ethnoecological concepts (often couched in myth and natural symbols) establish the basis for interdisciplinary dialogue and research. In this manner, indigenous knowledge of biological communities and ecological relationships can be studied; when non-Western Notions arise, these are formulated as hypotheses and tested by respective specialists. Posey (1983b), for example, reports the discovery of nine new species of stingless bees (Meliponinae} through the comparison of Kayapé and Western taxonomic systems. Posey and Camargo (1985) record the utility of Indian knowledge about bee behavior in the development of studies in areas little known to ethnoentomologists, such as: differences in odor characteristics, swarming behavior, flight patterns, and habitat choices between or within meliponine species. They also propose scientific investigations based on indigenous knowledge of bee species distribution in relation to ecological zones and habitat sharing by certain species clusters. Indian ideas of acrid commensalism and use of odor trails by species for which such activity is unreported have also spurred further studies by entomologists of stingless bee behavior. Overal and Posey (1986) have effected a large inventory of arthropod agricultural pests based on Indian information and confirmed by field collections. They also report the development of research into the highly effective control of agricultural pests in indigenous gardens through inter-cropping, use of trap crops and natural predators. The Indians attribute much of this natural control to predatory ants, wasps, and termites, all of which are glorified in Kayapd myth and song. Roles of these insects in crop pest control are being investigated following indigenous guidelines. Kerr and Posey (1984) report how the Kayapo utilize Azteca spp. ants to repel leaf- cutting sauva (Atta spp.). Likewise, Kerr and Posey (1986) report the indigenous use of 110 POSEY Vol. 6, No. 1 several natural pesticides and call for their testing by Western science. At least in the case of Azteca spp., Overal and Posey (1986) report very positive results from scientific tests to determine their effectiveness in the protection of Amazonian citrus. Anderson and Posey (1986) and Posey (1984) report the intentional planting of certain plant species by Indians to attract bees. Such knowledge can be helpful in the investigation of tropical pollination and aid in the improvement of apiculture. Many bee species are thought by the Kayapé to have important medicinal properties (Posey 1983e). Although almost practically unknown by pharmacologists they need to be investigated for their effecti and potential for a natural pharmacopeia (Elisabetsky and Posey 1986). These are but a few examples from a single ethnobiological project of how indigenous knowledge can stimulate new ideas for Western science. No researcher is expected to accept prima facie all native beliefs. Much indigenous knowledge, as we have already seen, is highly symbolic and difficult for even the most experienced anthropologist to interpret; however, nothing can be dismissed by the ethnobiologist no matter how ridiculous it may intially sound. The most seemingly ludicrous ideas today may offer the greatest insights tomorrow when their symbols are finally decoded. Refusal by Western scientists to study native beliefs is, after all, not a very scien- tific attitude. It is much more scientific to test the validity of native observations through the testing of hypotheses generated by ethnobiological study. CONCLUDING REMARKS Knowledge, classification, and use of insects in human societies is diverse but relatively unstudied in a systematic manner. Lack of anthropological and linguistic train- ing by entomologists—and entomological training by anthropologists and linguists— hampers ethnoentomological research. A true science of ethnoentomology will not develop until researchers have sufficient expertise in all three fields to investigate the native emic view of ‘natural worlds.” This situation does not prevent the elaboration of studies in cultural entomology that attempt to investigate insect importance in literate societies. Nor does it inhibit important research into the potential uses of insects as foods and medicines. Indeed, insects have played a significant role in human history and may be even more impor tant in the future. Whether as protein sources in space flight or as key elements 10 integrated biological pest control, insects will continue to be studied and manipulated for human welfare From the theoretical side, folk biological studies can discover ‘natural models” used by other peoples to define their own world in their own terms. Instead of imposing paradigms of anthropological structuralism and Western science upon non-Westem peoples, we must lear to elicit and organize our data within the cognitive bounds of the societies we study. Folk systems of knowledge have in most cases developed for many millenia and are frequently more ancient than Western science. They reflect the diversity of ways in which the natural world can be ordered and provide detailed information of ethology, ecological communities, useful species, and biological diversity. Folk knowledge can also serve to generate new ideas and hypoth that can be investigated and tested w ith the rigorous controls of occidental science. Studies of folk knowledge as outlined in this paper offer a powerful “intellectual bridge” between different peoples. Understanding the sciences of other cultures enriches Western science and provides the philosophical bases for the understanding and apprec!# tion of other peoples on and in their own terms. Summer 1986 JOURNAL OF ETHNOBIOLOGY 111 ACKNOWLEDGEMENTS I wish to thank Dr. Murray Blum for his many years of encouragement of my ethnoentomological research. Special thanks to Dr. Charles Hogue, Dr. H. Weidner, and Dr. Nelson Papavaro for their assistance in acquiring useful materials for the prepara- tion of this paper. I am grateful to Daniel Linger, Dr. Warwick E. Kerr, and an anonymous reviewer for their useful criticisms and comments on this paper, and to Carol Jones for her typing and proofing of the final manuscript. LITERATURE CITED AESCHLIMANN, J. P. 1982. Du role des insectes dans 1’alimentation humaine. Mitt. entomol. ges. 32(4): 99-103. ALDRICH, J. M. 1921. Coloradia pandora Blake, a moth of which the caterpillar is used as food by the Mono Lake Indians. Ann. Entomol. Soc. Am. 14:36-38. ALTIERI, M. A. 1983a. Agroecology: The Scientific Basis of alternative agri- culture. Univ. California Div. Biol. Control, Berkeley. 85 pp. 1983b. Pest manage- ment technologies for peasants: a ae systems approach. Crop Prot. 1985. Developing pest Management strategies for s farmers based on traditional know- ledge. Dev. Anthro. Network 3(1): 13-18. eee 4 fie 1. E PETOURIN- EAU. 1982. Vegetation management and biological control in agroeco- systems. Crop Prot. 1:405-30. ALVERSON, H. 1984. The wisdom of tradition in the development of dry- land farming: Botswana. Hum Organ. 43:1-8. ANDERSON, A. B. and D. A. POSEY. 1986. Manejo de campo-cerrado pelos indios Kayapo. Bol. Museu Goeldi. In press. ARAUJO, A. M. 1961. Medicina rustica. Serie Brasiliana, Vol. 300. Cia. Ed. Nac., S40 Paulo. 205 pp. ARMSTRONG, E. A. 1970. Insects. Man Myth Magic 52:1445-1451. ARNDT, W. 1923. Bemerkungen uber die rolle der insekten in Arzneischatz der alten kulturvolker. Dtsch. Entomol. Z. 1923:553-570. BALDUS, M. 1937. Ensaios de etnologia brasileira. Series Brasiliana, Vol. 101. Cia. Ed. Nac., S40 Paulo. 235 pp. . 1970. Tapirape, tribo tupi no Brasil Central. Series Brasi- liana, Vol. 17. Cia. Ed. Nac., Sao Paulo. 210 pp. BARRETT, S. A. 1925. The Cayapa Indians of Ecuador. Indian Notes and Monographs, No. 40. Mus. Am Indian. New York. 180 pp. BATES, H. W. 1862. Description of a remarkable species of singing cricket. J. Entomol. 1:474-477. BENNETT, C. F., Jr. 1964. Stingless-bee keeping in western Mexico. Geogr. Rev. 54(1):85-92. . 1965. Beekeeping with stingless bees in western Panama Bee World 46(1):23-24. BERGIER, E. 1941. Insectes Comestibles e Peuples Entomophages. Impr. Rul- liére Fréres, Avignon. 229 pp. BERLIN, B. and G. T. PRANCE. 1978. Insects galls and human ornamenta- tion: Ethnobotanical significance of a new species Licania in Amazonas, Peru. Biotropica 10:81-86. BERNDT, R. M. and C. H. BERNDT. 1964. The World of the First Austra- lians. Ure Smith, Sydney. 238 pp. BIARD, A. F. 1862. Deux Annees au Brésil. Hachette, Paris. 197 pp. BLACKBURN, T. 1976. A query regarding the possible hallucinogenic effects of ant ingestion southcentral Califomia. J. California Anthr. 3:78-81. BODENHEIMER, F. S. 1951. Insects as Human Food: A Chapter in the 112 POSEY Vol. 6, No. 1 LITERATURE CITED (continued) Ecology of Man. Junk, The Hague. S2 pp. BROK ENSHA, D. W., D. M. WARREN, RNER, O. WERNER, eds. 1980. Indigenous Knowledge Systems and Development. tanurresh Univ. Press America, Lanham. 4. BROWN, C. H. and P. K. aa 1981. Animal classification in Juchitan, Zapotec. J. Anthropological Res. BROWN, B. |. and G. G. MARTEN. 1984. The Ecology of Traditional Pest Management in Southeast Asia. seh Pap., East-West Center, Hawaii, 23 pp. BRUSH, e B. 1982. The natural and human environment of the central Andes. Mt. Res. Dev. 2:14-38. BULMER, R. N. H. 1968. Worms that croak and other mysteries of Karam natural history. Mankind 6:621-639. BUNTING, A. H. 1972. Pests, population and poverty. Trop. Sci. 14:37-50. BUSHNELL, D. I. 1909. The Choctaw of Bayou Lacomb, St. Tammany Parish, Louisiana. In Bureau of American Ethnology, Bul. 48. Smithson Inst., Washington, D.C. 1910. Myths of the Louisiana Chocktaw. Am. Anthro- pol. 12:526-535. BUSHVINE, J. R. 1976. Insects, Hygiene and History. Athlone, London. 262 pp. CALLAHAN, P. S. and R. W. MANKIN. 1978. Insects as unidentified flying objects. i 3 alge 17:3355-3360. CALVERT, C. and N. O. eee 1969. House fly pupae as food for poultry. J. Econ. Entomol. 62:938-939. CARON, D. M. 1978. Insects and human nutrition. Ame. Bee J. 118(6):388-389. CARVALHO, J. C. M. 1951. RelacGdes entre os indios do alto Xingi e a fauna regional. Publ. Avulsa. Mus. Nac., Rio de Janeiro, 7:1-32. CATLEY, A. 1963. Notes on insects as food for native peoples in Papua New Guinea. Trans. Papua New Guinea Sci. Soc. 4:10-12. CAUDEL, A. N. 1916. An economic con- sideration of Orthoptera directly affecting man. Proc. Entomol. Soc. Wash. 18:84-92. CHAGNON, N. 1968. Yanomamo: The fierce people. Holt, Rinehart, and Winston, New York. 135 pp. 1973. Yanomamo. Pp. 234- 247 i in Peoples of the Earth, (EE. oo Pritchard, eds.). Danbury, Vero CHAMBERS, D. L. 1977. Quality control in mass rearing. Ann. Rev. Entomol. 22:289-308. CHOOVIVATHANAVANICH, P., P. SUWANPRATEEP, N. KATHA- VICHITRA. 1970. Cockroach senst- tivity in allergic Thais. Lancet 2:1362-1363. CLAUSEN, L. W. 1954. Insect Fact and . Macmillan, New York. 194 pp. CLOUDSLY-THOMPSON, J. L. 1976. Insects and history. St. Martins, New York. 242 stu COIMBRA Jr. 1984. Estudos de ecologia Pe entre os Surui do parque indigena Aripuna, Rom donia. 1. O uso de larvas de Cole6p- teros (Bruchidae e Curculionidae} na a Rev. Bras. Zool. 2(2):35 CONCONI, ri R. E. 1977. Valor nutritovo de ciertos insectos comestibles de México y lista de algunos insectos comestibles del mundo. An. Inst. Biol. Univ. Nacional. Auton. Méx. 48(1):165-185. . 1982. Los Insectos Como Fuente de Proteinas en el Futuro. Limusa, México, D. F. 144 PP- M. P. dO. MM. GONZALES. 1981. Digestibilidad en vitro de algunos insectos comestibles en México. Folia Entomol. Mex. 49:141-154. MORENO, CM, MAYAUDON, F. VALDEZ, Summer 1986 JOURNAL OF ETHNOBIOLOGY 113 LITERATURE CITED (continued) and M. A. PEREZ, et al. 1984. Protein content of some edible insects in Mexico. J. Ethnobiol. 4(1):61-72. CONKLIN, H. C. 1973. Folk Classifi- cation: A topically arranged bibli- ography of contemporary and back- ground references through 1971. Yales Univ. Press, New Haven. COOPER, J. I. and T. W. TINSLEY. 1978. Some epidemiological conse- quences of drastic ecosystem changes accompanying exploitation of trop- ical rain forest. Terre Vie 32(2):221-240. COVARRUBIAS, M. 1971. Indian Art of Mexico and Latin America. A. A. Knopf, New York. 386 pp. COWAN, F. 1865. Curious Facts in the History of Insects. J. B. Lippincott, Philadelphia. 405 pp. CRANE, E., ed. 1979. Honey: A Compre- hensive Survey. Intl. Bee Res. Assn. Gerrads Cross, Buckshire, England. 624 pp. CRANE, E. 1984. The Archaeology of Beekeeping Duckwort, London. 611 CROSBY, A. W., Jr. 1972. The Columbian Exchange: Biological and Cultural Consequence of 1492. Greenwood, Westport. 286 pp. , C. H. 1939. On eating insects. Nat. Hist. 613:84-89. CURRAN, C. H. 1937. Insect lore of the Aztecs revealing early acquaintance- ship with many of our agricultural pests and therapeutic measures against so currently prominent a Creature as the black widow spider. Nat. Hist. 39:196-203. DAMMERMAN, K. W. 1929. The Agri- cultural Zoology of the Malay a. J. H. Buxy. Amsterdam. DAOUST. M. V. 1858. On some eggs of insects employed as human food, and giving rise to the formation of Oolites in Laucustrine limestone in Mexico. Ann. Mag. Nat. Hist. 2:78-80. DEFOLIART, G. R. 1975. Insects as a source of protein. Bull. Entomol. Soc. Am. 21(3):161-163. DENEVAN, W. M. 1971. Campa Subsis- tence in the Gran Pajonal, Eastern Peru. Geogr. Rev. 61(4}:496-518. DENEVAN, W., J. TREACH, J. ALCORN, C. PADOCH, J. DENSLOW, and J. PAITAN. 1984. Indigenous agrofor- estry in the Amazon: Bora Indian management of swidden fallows. Interciencia 9(6):346-357. DENTAN, R. K. 1968. Notes on Semai ethnoentomology. Malay. Nat. J. 21(1):17-28. DOBYNS, H. 1966. Estimating aboriginal American population. Curr. Anthro- pol. 7:395-416. DUFOUR, R. A. 1981. Insects: A Nutri- tional Alternative. Dep. Med. Pub. Affairs, George Washington Univ. Med. Cent., Washington, D.C. 64 pp. EALAND, C. A. 1915. Insects and Man. Richards. London. 343 pp. EFFLATON, B. 1929. The development of entomological science in Egypt. Trans. Ist. Int. Congr. Entomol, 1928. 2:737-742. EGGER, K. 1981. Ecofarming in the tropics—Characteristics and poten- tialities. Plant Res. Dev. 13:96-106. ENGELHARDT, W. 1959. Was lebt in Tiimpel, Bach und Weiher? Verlag- shdlg. Stuttgart. 387 pp. ELIZABETSKY, E. and D. A. POSEY. 1986. Conceito de animais com os are Kayapo. Rev. Bras. Zool. in ESSIG, E. O. 1931. A History of Ento- mology. Macmillan. New York. 1029 1934. The value of insects to the California Indians. Sci. Mon. 38:181-186. FLOERICKE, K. 1922. Huechrecken und libellen. Kosmos. Stuttgart. 235 pp. FRANKIE, G. W. and L. E. EHLER. 1978. Ecology of insects in urban environ- ments. Ann. Rev. Entomol. 23:367- 388. 114 POSEY Vol. 6, No. 1 LITERATURE CITED (continued) GANJIAN, L, I. KUBO and P. FLUD- ZINSKI. 1983. Insect antifeedant elemanolide lactones from Vernonia amygdalina. Phytochemistry 22(11): 2525-2526. GIACONE, A. 1949. Os Tucanos e outras tribos de Rio Uaupés affluente do Negro-Amazonas. Imprensa Off. Estado. Sao Paulo. 135 pp. GILBERT, W. H. 1943. The Eastern Cherokee. In Bureau of American Ethnology, Bull. 133, 169-413. Smithson. Inst. Washington, D.C. GILBERT, B. M. and W. H. BASS. 1967. Seasonal dating of burials from the presence of fly pupae. Am. Antigq. 32(4):534-535. GLASS, E. H. and H. D. THURSTON. 1978. Traditional and modern crop protection in perspective. Bioscience 28(2):109-115. GLIESSMAN, S. R., E. R. GARCIA and A.M _ AMADOR. 1981. The ecologi- cal basis for the application of tradi- tional agricultural technology in the management of tropical agroeco- systems. Agroecosystems 7:173-185. GLOCK, J. 1891. Die symbolik der bienen und ihrer produkte in sage, dichtung, kultus, kunst und Brau- chen der Volker. Heidelberg. 411 pp. GOLOB, P., J. MWAMBULA, V. MHANG and F. ‘NGULUBE. 1982. The use of locally available materials as protec- tants of maize grain against insect infestations during storage in Malawi. J. Stored Prod. Res. 18:67-74. GORHAM, J. R. 1975. Filth in foods: Implication for health. J. Milk Food Technol. 38:409-418. 76. Insects as food. Bull. Soc. Vector Ecol. 3:11-16. 1976. A rational look at insects as food. FDA Guidelines 5:231-239, 1979. The significance for human health of insects in food. Annu. Rev. Entomol. 24:209-224. GREENLEE, R. 1944. Medicine and cur- ing practices of the modern Florida Seminoles. Am. Anthropol. 46:897- GREGOR, T. 1983. Dark dreams about the white man. Nat. Hist. 8 pp. GRIAULE, M. 1961. Classification des insectes chez les Dogon. J. Soc. Afr $1:7-71. GRINNEL, G. B. 1899. The butterfly and the spider among the Blackfoot. Am. Anthropol. 1:194-196. GUDGER, E. W. 1925. Stitching wounds with the mandibles of ants and beetles. J. Am. Med. Assoc. 84{24): 1861-1864. HAGAN, H. 1863. Die insektennamen der Tupi sprach. Stettiner Entomol. Ztg. 24:252-259. HARE, R. 1954. Pomp and Pestilence. Gollantx. London. 224 pp. J. R. 1976. The plants and animals that nourish man. Sci. Am. 254(3}:89-97. HARPAZ, I. 1973. Early entomology in the Middle East. In History of Entomology, (R. F. Smith, ed.), Palo Alto, California. Annual Reviews. 456 pp. HENDRICKS, P. K. 1941. Cultivo = abejas indigenas en el Estado Guerro. Méx. Antiguo 5:365-373. HEVLY, R. H. 1982. Analysis of flotation samples from the Coronado tr. transmis- sion line corridor. In The Specialist Volume: Biocultural Analysis, Cor onado Ser. 4, Paper rae sng na, Flagst Northern Arizo: ae g oe 1974. Insect ta from pre historic pueblo in Arizona. Pan-Am- Entomol. 50(3):307-308 4 HITCHCOCK, S. W. 1962. ‘Insects an Indians of the Americas. B Entomol. Soc. Am. 8:181-187. HOFFMAN, W.E. 1947. Insects as human he Proc. Entomol. Soc. Wasb- 7. 9:223-23 shat C. L. 1980. Commentaries e cultural entomology—definition ° cultural entomology. Entomol. News 91(2}:33-36. Summer 1986 JOURNAL OF ETHNOBIOLOGY 115 LITERATURE CITED (continued) HOLT, V. M. 1885. Why Not Eat Insects? E. W. Classey. Middlesex, England. 325 pp. HOSEN, H. W. 1980. Factors associated with the attribution of human traits to nonhumans. J. Soc. Psychol. 112(1):161-162. HUIS, A. VAN, R. §. NAUTA and M. E. VULTO. 1982. Traditional pest Management in maize in Nicaragua: A Survey. Meded. Landbouwhogesch. Wageningen 82-86, pp. 43. KATTER, F. 1883. Die canthariden spec. Meloe als Meilmitel der Tollwurth. Entomol. Nachr. 9:156-183. KENNEDY, C. H. 1943. A dragonfly nymph design in Indian pottery. Ann. Entomol. Soc. Ame. 36:190-191. KERR, W. E. and D. A. POSEY. 1984. Nova informacdo sobre a agricultura dos Kayap6. Interciencia 9(6}:392-400. . 1986. “Um Cipd” que mata abelha. Rev. Bras. Zool. in press. KEVAN, D. K. 1974. The Land of Grass- hoppers. Lyman Entomol. Mus. Ste-Anne-de-Bellevue, Quebec. 328 pp. KEVAN, K. M. 1979. The place of grass- hoppers and crickets in Amerindian cultures. Proc. of the 2nd Triennial Meet. Pan-Am. Acridol. Soc. 74 pp. KEVAN, D. K. 1980. Grigs, graces, graphics and graffiti. Metaleptea 2(2): -/2. KHAN, M. M., D. RAJAGOPAL, and P. HANUMAPPA. 1978. Plant protec- tion practices and problems of chilly growers of Kolar District. Mysore J. Agric. Sci. 12:159-163. KILPATRICK, J. and A. KILPATRICK. 1970. Notebook of a Cherokee 120 pp. KLOTS, A. B. and E. B. KLOTS. 1959. Insekten. Miinchen-Ziirich: Droe- mersch Verlagsanst. 367 pp. KNORTZ, K. 1910. Die insekten in sage, Sitte, und literatur. Annaberg/Sach- sen. Graeser. 151 pp. KOK, R. 1983. The production of insects for human food. J. Can. Inst. Food Sci. Technol. 16(1):5-18. KUBO, I. and T. MATSUMOTO. 1984. Abyssinin, a potent insect antifee- dant from an African medicinal plant, Bersame abyssinica. Tetrahedron Lett. 25(41):4601-4604. KUBO, I., T. MATSUMOTO, J. A. KLOCKE and T. KAMIKAWA. 1984. Molluscicidial and insecticial activi- ties of isobutylamides isolated from Fagara macrophylla. Experientia 40:340-341. LAUFER, B. 1927. Insect-musicians and cricket champions of China. Field Mus. Anthropol. Leafl. x Field Mus. Nat. Hist. Chicago. 2 LAWSON, J. 1937. History of Carolina. Richmond, Virginia. 367 pp. LENKO, K. and N. PAPAVARO. 1979. Insectos no Folclore. Cons. Estad. Artes Cienc. Hum. Sdo Paulo. 518 pp. LIEBRECHT, F. 1886. Tocandyrafestes. Z. Ethnobiol. 18:350-352. LITSINGER, J. A., E. C. PRICE and | ae iF HERRERA. 1978a. Filipino farmer use of plant parts to control rice insect pests. Int. Rice Res. News 3(5):15-16. 1978b. How the farmers in three provinces control crop pests. we 8(8):6-16. 0. Small farmer pest control atlas for rainfed rice, corn, and grain legumes in three Phil- ippine provinces. Philipp. Entomol. 765-86. LOTHRAP, S. K. 1964. Treasures of Ancient America. Geneva:SKIRA 329 pp. , T. M. 1981. Bee as meta- phor: Psychodynamic tensions in Maya culture. Unpubl. B.A. thesis, Harvard Coll., Cambridge, Massa- chusetts. 75 pp. MACHARGUE, J. S. 1917. A study of proteins of certain insects with reference to their value as food for poultry. J. Agric. Res. 19:633-637. 116 POSEY Vol. 6, No. 1 LITERATURE CITED (continued) MCKEOWN, K. C. 1944. Insect Wonders of Australia. Angus & Robertson, Sydney. 256 pp. MCNEILL, W. H. 1976. Plagues and People. Anchor, Garden City. 369 pp. MALINOWSKL B. 1929. The Sexual Life of Savages in Northwestern Mela- nesia. H. Ellis, New York. 437 pp. , B. 1956. Seri ethnozoology: A preliminary report. Davidson J. Anthropol. 2:73-83. MARSHALL, W. 1894. Neueroffnetes, wundersames Arzenei-Kastlein. Leipsig. 127 pp. MATTESON, P. C., M. A. ALTIERI, and W. C. GAGNE. 1984. Modifi- cation of small farmer practices for better pest management. Ann. Rev. Entomol. 29:383-402. MEGGITT, M. J. 1962. Desert People: A study of the Walbri Aborigines of Central Australia. Angus & Robert- son, Sydney. 298 pp. METRAUX, A. 1948a. The hunting and gathering tribes of the Rio Negro Basin. In Handbook of South American Indians, 3:816-67. Smithson. Inst., Washington, D.C. 1948b. Tribes of the Middle and Upper Amazon River. In Handbook of South American Indians, 3:668-712. Smithson. Inst., Washington, D.C. MEYER-ROCHOW, V. B. 1973. Edible insects in three different ethnic groups of Papua New Guinea. Am. J. Clin. Nutr. 26:673-677. 1975a. Can insects help to ease the problem of world food shortage? Search 6(7):261-262. 1975b. Local taxon- omy and terminology for some ter- restrial arthropods in five different ethnic groups of Papua New Guinea and Central Australia. J. Rev. Soc. West Australia 58:15-30. 1976. The use of insects as human food. Food & Nutr. 33:151-152. . 1985. The diverse uses of insects in traditional societies. J. Ethnommed. In press. MILL, A. E. 1982. Amazon termite myths: Legends and folklore of the Indians and Caboclos. Trans. R. Entmol. Soc. London 6(2):214-217. MILLER, J. A. 1981. Space gardening. Sci. News Lett. 119:330-334. MODELL, M. 1977. Sustaining life in a space colony. Technol. Rev. 79(8): MONTGOMERY, B. F. 1959. Arthropods and ancient man. Bull. Entomol. Soc. Am. 5:68-70. MOONEY, J. 1972. Myths of the Chero- kee and Sacred Formulas of the Cherokee. Reprinted from the seventh and ninteenth Annual Reports of the Bureau of American Ethnology. Elder Bookseller, Nash- ville. 426 pp. MORPHI, F. J. A. 1932. History of the Province of Texas. 156 pp. __ NOGEIRA-NETO, P. 1970. A Criac4o de Abelhas Indigenas sem Ferrdo (Meliponinae). Tecnapis, S40 Paulo. NORDESKIOLD, E. 1929. L’apiculture indienne. J. Soc. Americanistes 21(1):169-182. P NUORTEVA, P. 1963. Synanthropy © blowflies in Finland. Annu. Entomol. Fenn. 29:1-49. OKA, I. N. and D. PIMENTEL. 1974. Cor susceptibility to com leaf aphids and common com smut after herbicide treatment. Environ. Ento- mol. 2(6):911-915. : 1973. Insects in the art and myth of the northwest coast Indians. Bull. Entomol. Soc. Can- 5(1):20-26. OVERAL, W. L. and D. A. POSEY. ai c VERSTRAETEN. 1978. Les miels dans la foret claire du shaba méri- Summer 1986 JOURNAL OF ETHNOBIOLOGY 117 LITERATURE CITED (continued) dional. Bull. Rech. Agron. Gembloux 13(2):161-176. PARKER, E., D. A. POSEY, J. FRECH- IONE and L. F. DA SILVA. 1983. Resource exploitation in Amazonia: Ethnoecological examples from four populations. Ann. Carnegie Mus. 52 (Art 8}:163-203. PERRIN, R. M. 1980. The role of environ- mental diversity in crop protection. Protopathic Ecol. 2:77-114. PIMENTEL, D. 1971. World food crisis: Energy and pests. Bull. Entomol. Soc. Am. 22:20-26. —__________ and N. GOODMAN. 1978. Ecological basis for the man- agement of insect populations. Oikos 30:422-437,. (itu EME. Me, DRITSCHILO, D. GALLAHAN and N. KINNER. 1977. Pesticides, insects in foods, and cosmetic stan- dards. BioScience 27(3):178-185. POSEY, D. A, 1976. Entomological considerations in southeastern abori- ginal pear a oe J. Ethnohist. 23(2):147-160. 1977. An ethnoento- mological perspective of the south- eastern Indian belief system. Hum. Mosaic 11(1):1-9. 1978. Ethnoento- mological survey of Amerind groups of lowland South America. Florida Entomolo. 61:225-229 —_——___.___.. 1979. Ethnoento- mology of the Kayapé Indians of Cen- tral Brazil. Unpubl. Ph.D. dissert., Univ. Georgia, Athens ———________.. 1979. Kayapé controla iNseto com uso adequado do am biente. Rev. Atual. Indig. 3(14):47-56. . 1980. Consideraciones €tnoentomologicas sobre los grupos amerindios. Am. Indig. 40(1):105-120. . 1981. Wasps, warriors, and fearless men: Ethnoentomology of the Kayap6é Indians of central Brazil. J. Ethnobiol. 1:165-174. 1983a. Ethnomethod- ology as an emic guide to cultural systems: The case of the insects and the Kayapd Indians of Amazonia. Rev. Bras. Zool. 1(3):135-144. 1983b. Folk apiculture the Kayapo Indians of Brazil. Biotropica 15(2}:154-158. 983c. Indigenous knowledge and development: An ideological bridge to the future. Ciénc. Cul. 35(7):877-894. . 1983d. Indigenous knowledge and development of the Amazon. Pp. 135-144 in The Dilem- ma of Amazonian Development, (E. Moran, ed.). Westview. Boulder, Colorado. 1983e. Keeping of stingless bees by the Kayapé Indians of Brazil. J. Ethnobiol. 3(1):63-73. . 1983f. The importance of bees to an Indian tribe of Ama- zonia. Fla. Entomol. 65(4):452-458. . 1984. A preliminary report on diversified management of tropical forest by the Kayapé Indians of the Brazilian amazon. Ethnobotany in the Neotropics: Advances in Ecomonic Botany 1:112-126. 1985. Hierarchy and utility ina folk biological taxonomic system: Patterns in classification of Arthropods by the Kayapé Indians of Brazil. J. Ethnobiol. in press. 1986. Etnoentomol- ogia dos indios Amazonicos. In Suma Brasileria de Etnologia. Rio de Janeiro: FINESP. in press. and J. M. F. Camargo. Additional notes on beekeeping of Meliponinae by the Kayapé Indians of Brazil. Ann. Carnegie Mus. in press. | J. FRECHIONE, J. EDDINS and L. FRANCELINO DA SILVA. 1984. Ethnoecology as Ap- plied Anthropology in Amazonian Development. Human Organization 43(2):95-107. 118 POSEY Vol. 6, No. 1 LITERATURE CITED (continued) POVOLNY, D. 1971. Synanthropy. In Flies and Disease, (B. Greenberg, ed.) Princeton Univ., Princeton. 856 pp. RAMIREZ, J. P., P. ARROYO and A. CHAVEZ. 1973. Aspectos socio- econdmicos de los alimentos y la alimentaci6n en México. Rev. Comer. Ext. del Bco. de Comer. Ext. 1:675-690. RANSOME, H. M. 1937. The Sacred Bee in Ancient Times and Folklore. George Allen & Unwin London. 285 pp. RATHS, A. and D. BIEWALD. 1974. Tiere im Experiment. Aulis Verlag Deubner & Co., Kéln. 316 pp. READ, B. E. 1935. Insect a Peking Nat. Hist. Bull. 94:8-8 REDFORD, K. H. 1982. Be attraction as a possible function of biolumi nescence in the larvae of mans es termitilluminans s (Coleoptera: Elateridae]. Rev. Bras. Zool. 1(1):31-34. 1986. Insects as food for humans: Some cautionary com- ments. J. Ethnobiol. in press. and J. DOREA. 1984. The nutritional value of inver- tebrates with emphasis on ants and termites as food for mammals. J. Zool. London 203:395-395. REED, A. W. 1965. The flies and the bees. Pp. 40-41 in Aboriginal Fables and Legendary Tales. Reed, Sydney. 144 REIM, H. 1 62. Die insektennahrung der Australischen ureinwohner. Academie Verlag, Berlin. 224 pp. RISCH, S. J., ANDOW and M. A. ALTIERI. 1983. system diver- sity and pest control: Data, tentative conclusions, and new research direc- tions. Environmental Entomol. 5-629. RITCHIE, C. 1979. Insects, the creeping conquerors and human history. Soc. Sci. 54(2):122-132. ROHEIM, G. 1974. Children of the Desert: The Western Tribe of Central Australia. Basic Books, New York. 309 pp. RUDDELL, K. 1973. The human use of insects: Examples from the Yukpa. Biotropica 5(2):94-101. RUTSCHKY, C. W. 1981. Arthropods in the lives and legends of the Pennsyl- vania Indians. Mels. Entomol. Ser. 30:39-42. SANTOS, P. B. and D. A. POSEY. 1986. Conceitos de saiide, adoecer, vura e morte em relacSo as plantas medi- cinais e o aparecimento de D. Sebas- tiZo, Rei Messianico, na ilha de Len- cois, Maranh4o. Ciencia e Cultura in press. and C. COIMBRA, JR. 1984. CriacZeo e comercializacéo de larvas de Hermetia illucens (Dip- tera: Stratiomydae) em uma comurr idade do Distrito Federal. Cienc. Cult. 36(12):2211-2215. SCARBOROUGH, J. 1979a. On the history of early entomology chiefly Greek and Roman, with a prelimi nary bibliography. Melsheimer En- tomol. Serv. 26:17-27. _1979b. Nicander’s ers, - ions, in- toxicology: spid rp eae sects, Hist. 21(1):3-34. ; 1979c. Nicander’s toxicology: spiders, scorions, insects d myriapods. P. 2. Pharm. Hist. 21(2):73-92. . 1981. Ancient medi- ine: Some recent books. Clio Med. vores eps -149, SCHIMITSCHEK, E. 1968. Insekten abs n , in Brauchtum, Kukt, und Kultur, 2nd ed. Pp. 162 Kukenthal’s Handbook der oe J. C. Helmcke, D. Starck, R. Wek- muth, eds.). Walter de Gruytet Berli 7. Insekten in der bildenden asd im Wandel det Summer 1986 JOURNAL OF ETHNOBIOLOGY 119 LITERATURE CITED (continued) Zeiten in psychogenetischer Sicht. Veroff. Naturhist. Mus. Wein. 14:119 ff. SCHOOLCRAFT, H. R. 1851. Historical and Statistical Information Respect- ing the History, Condition, and Pros- pects of the Indian Tribes of the United States. Bur. Indian Affairs, Philadelphia. 485 pp. SCHWARZ, H. F. 1945. The wax of stingless bees (Meliponidae) and the uses to which it has been put. J. New York Entomol. Soc. 53:137-139. 1948. Domestication of stingless bees and rites connected with bee culture. Bull. Am. Mus. Nat. Hist. 90:142-160. SCHWARZ, P. M. and W. KLASSEN. 1981. Estimate of losses caused by in- sects and mites to agricultural crops. In CRC Handbook of Pest Manage- ment in Agriculture. (D. Pimentel, ed.). CRC Press, Boca Raton. SINGEREST, H. E. 1951. Civilization and Disease. Cornell Univ. Press, Ithaca, New York. 255 pp. SILOW, C. A. 1976. Edible and Other Insects of Mid-western Zambia. Occ. Pap. 5. Inst. Allm. Jamforand Etnogr., Uppsala. —__—____________. 1983. Notes on Ngan- gela and Kkoya ethnozoology, ants, and termites. Etnol. Stud. 36:1-177. SIMMONS, P. 1976. A specific visual tesponse in dragonflies. Odonato- logica 5:285. SKINNER, A. 1910. The use of insects and other invertebrates as food by North american Indians. J. New York Entomol. Soc. 18:264-267. SMITH, K. G. V., ed. 1973. Insects and Other Arthropods of Medical Impor- tance. Br. Mus. Nat. Hist. London. 561 pp. SPENCER, B. and F. J. GILLEN. 1899. The Native Tribes of Central Austra- lia. Macmillan, London. 387 pp. STEWARD, J. and A. METRADX. 1948. The Peban tribes. In Handbook of South American Indians 3:816-867. Smithson. Inst. Washington, D.C. SWANTON, J. R. 1928. Religious beliefs and medical practices of the Creek Indians. Ann. Rep. Bur. Am. Ethnol. No. 42:437-672. 1946, The Indians of the southeastern United States. Bur. Am. Ethnol. Bull. 137. Washington, D.C. 445 pp. TAYLOR, R. L. 1975. Butterflies in a4 Stomach: Or Insects in Human Nut tion. Woodbridge, Santa cae. 224 pp. TEOTIA, J. S. and B. F. MILLER 1974. Nutritive content of house fly pupae and manure residue. Br. Poult. Sci. 15:177-182. THRESH, J. M. 1982. Cropping practices and virus spread. Ann. Rev - pathol. 20:193-218. TIHON, L. 1946. A propos des termites au point de vue alimentaire. Bull. Agric. Congo Belge 37:865-890. TINDALE, N. B. 1953. On some Austra- lian Cossidae, including the moth of the Witjuti (Witchey) grub. Trans. R. Soc. South Aust. 76:56-65. VANDERZANT, E. S. 1974. Develop- ment, significance and application of artificial diets for insects. Annu. Rev. Entomol. 19:139-160. VILLAS-BOAS, O. and C. VILLA-BOAS. 1972. Xing, os indios e seus mitos. Zahar, Rio de Janeiro. 246 pp. WADDY, J. 1982. Biological classification from a Groote Eylandt aboriginal point of view. J. Ethnobiol. 2(1):63-77. WAGNER, M. 1895. Das zeidelwessen und sein ordnung im Mittelalter und in der neuren Zeit. Munchen. 235 pp. WALLACE, A. R. 1852. On the insects used for food by the Indians of the Amazon. Trans. Entomol. Soc. Lon- don 2:241-244. WEAVER, N. and E. WEAVER. 1981. Beekeeping with the stingless bee Melipona beecheii by the Yucatecan Maya. Bee World 62(1):7-19. 120 POSEY Vol. 6, No. 1 LITERATURE CITED (continued) WHEELER, A. G., Jr. 1981. The tarnished lant bug: Cause of potato rot? J. Hist. Biol. 14(2):317-338. WILBERT, J. 1979. Geography and tel- luric lore of the Orinoco Delta. J. Lat. Am. Lore 5(1):129-150. 1981. Warao cos- mology and Yekuana roundhouse symbolism. J. Lat. Am. Lore 7(1): 37-72. WILKEN, G. C. 1977. Interpreting forest and small-cale farm systems in Middle America. Agroecosystems 3: 291-302. WYMAN, L. C. 1973. The Red Antway of the Navajo. Mus. Navajo Art. Santa Fe, New Mexico. 274 pp. and F. L. BAILEY. 1952. Native Navaho methods for the con- trol of insect pests. Plateau J. Mus. North. Arizona 24(3}:97-103. . 1964. Navajo Indian Ethnoentomology. Univ. New Mex- ico Pub. Anthropol. No. 12., Albu- querque. 143 pp. ZINSSER, H. 1935. Rats, Lice, and History. Blue Ribbons Books, New York. 259 pp. J. Ethnobiol. 6(1):121-128 Summer 1986 NEW DIRECTIONS IN ETHNOPHARMACOLOGY DR. ELAINE ELISABETSKY Laborat6rio de Etnofarmacologia Depto. de Fisiologia Universidade Federal do Para 66.000 Belém Para Brazil ABSTRACT.—New directions in ethnopl logy must be concerned with more than the recording and preservation of folk knowledge of plant and animals species with biodynamic compounds. Native concepts of disease and illness, collection and prepara- tion procedures, religious and ceremonial contexts, etc. must also be studied and understood. Such additional information will aid in identifying prototype drugs as well as their development and return to the indigenous cultures from which they came. The growth of Science would depend on man’s willingness to believe the improbable, to cross the dictates of common sense. Daniel J. Boorstin INTRODUCTION Several workers (Farnsworth and Morris, 1976; Farmmsworth and Pezzuto, 1983; Holmestedt and Bruhn, 1982) in discussing the historical importance and potential of ethnopharmacological research have emphasized that such studies may lead to (1) the discovery of prototype drugs! ; (2) the recognition of new therapeutic actions of com- pounds already commonly in use for other purposes (Peigen 1983); (3) the discovery of new sources of crude material for semi-synthetic drugs; and, (4) the utilization of in natura plants or their simple pharmaceutical formulations as a means of attaining the develop- ment of low cost medicines. Some 7,000 compounds currently used in modern medicine are derived from natural products; most had been used for centuries by European, Asiatic and Amerindian healers. This can be better appreciated when one considers that even in highly industrialized countries, 40-45% of the commercial pharmaceutical products come from natural sources. For the USA in 1980, this amounted to US$ 8 billion (Jaroszewski 1984, Famsworth 1985}. The contribution of natural products to world health care is even greater in Asia, Africa and Latin America, where a large portion of the world’s Population lives. There, either by choice or necessity, plants are used as medicine directly from nature (Svendsen and Scheffer 1982). a Vari are oft itori f a great deal of knowledge of natural medicinal sources within isolated indigenous and peasant cultures. There is now a considerable danger of loss in the oral transmission of medical plant lore to the youth of the societies because of pressure from economic development and accelerated interaction with domi- nant cultural systems. Unless this trend changes or unless the information is recorded, this valuable knowledge will be lost. Reversing this trend will require detailed ethnographic research which is crucial for the understanding of drug use and medical Practices of a given society, because these are inextricably integrated with religion, ritual, social relations, etc. Native knowledge and understanding of medical concepts is 122, ELIZABETSKY Vol. 6, No. 1 fundamental for ethnopharmacological research, for only then can we understand why and how such a treatment is used. Interaction between anthropology and ethnopharmacology is the basis upon which holistic research on medicinal plants in particular, and medical care in general, must emanate (Elisabetsky & Posey 1985; Elisabetsky & Setzer 1986). The purpose of this paper is to elaborate the linkages between anthropology and ethnopharmacology, stressing their relation with further laboratory procedures. ETHNOLOGY AND DRUG EFFECTIVENESS The concepts of health and disease, as well as the medical system as a whole, are articulated within a given society, as cultural systems (Kleinman 1978). Although anthropologists frequently stress that indigenous societies deal with “sickness” as interrelation between nature, supernature, society and the individual (Fabrega_ a Tsunoo 1978; Worsley 1982; Young 1982 and others), etl often select traditional medicine to be screened by laboratory procedures i in hope “f ihe discovery of new universally applicable drugs without considering their cultural contexts. This leads toa shortcoming because not every cea remedy 3 is necessarily = Seer the anroani Since a truly holistic view of the individual fein is atte FS basis for ee medical reasoning, treatments are often directed to possible sources of sickness other ina physical ones. — religious/ceremonial aspects integrated into indigenous medi way to cope with psychological needs, social control, and eae of eee steps (Fabrega 1975) instead of just “folk” explanations or “superstitions’ Although the necessity of f interdisciplinary work i is frequently stressed i . Bruhn and Holmstedt 1981), most consider merely as a catalogue of remedies, methods of preparation or simple lists of oe plants Even this very fi is often not considered in the overall research designs. It is because medicinal plants (or animals) are used in different therapeutical contexts that ene requires an interdisciplinary approach that includes anthropologis Ethnological studies deal with disease concepts, belief systems and attitudes as Con: ceptual frames of references, stressing their internal logic and coherence (Quin uintanilla 1978; Young 1982; Worsley 1982). The position of many anthropologists may thus be limiting to ethnopharmacology for they preclude the possibility of attaining universal medical advances based on different cultural experiences. We cannot disregard the possibilities of attaining new medical ideas by integrating traditional and modern knowledge. For instance, this comprehensive view of the individual, so common and parr apse in indigenous societies in its approach to health or sickness, is NOW only ving much attention in our own medical science, thanks to recent development a such 1 new disciplines as psychosomatic medicine, medical ecology and the recogni tion of the t 1 conditions in relation to individual and public health. It is well to note that, Rama i Ws medical practices would likely chant for “more frequently than not, it is the bioch physiological and arate pharmacological knowledge to accrue. This in turn often allows for the theoretical rationalization of human disorders” (Maxwell 1984:377). The success of this interdisciplinary approach lies, therefore, in the understam of the physical and cultural treatments of a holistic medical system. Only then can we isolate and scrutinize particular aspects of medicine (Elisabetsky and Setzer 1986). Summer 1986 JOURNAL OF ETHNOBIOLOGY 123 GOALS OF ETHNOPHARMACOLOGY Although to many “the goal of ethnopharmacologic research is to find new prototype drugs” (Malone 1983:134), this is not and should not be considered the only one. If, for example, a country dependent on foreign sources of a certain pharmacological compound (see Gereffi 1983) detects, through ethnopharmacologic research, another source of this compound in a native species, a critical first step will have been taken in making that drug available for lower costs. For developing countries, where importation of pharma- ceuticals is is one of the fastest growing drains on hard foreign currency (Antezana 1981:175), this extremely valuable contribution from ethnopharmacological research, i.€., a national source of crude material, is necessary before the drug can be developed and promoted (see WHO 1979). To take specific examples, in China there are 200,000 people engaged with the planting, processing and distribution of medicinal herbs, including 800 pharmaceutical industries producing 2,000 types of remedies (Wang 1983). In India, there is a network of more than 600 hospitals and 15,000 dispensaries providing health care facilities based on traditional medical systems such as Ayuveda, Unari and Sidha (Akerele 1985). From those plants, today used as crude materials, many will eventually be trans- formed in the so-called prototype drugs (Peigen 1983). Nevertheless, a great part of China and India’s populations are already profiting from their herbal lore, long before Western science achieves its goal. Different goals may require procedures other than those now used by pharmaceutical industries and/or some academic laboratories where the com- plete pharmacological evaluation of each plant coupled with the complete phytochemical analysis of the whole plant (or the part used by natives] is undertaken (see Malone 1977, 1980). This is an unrealistic goal for most laboratories in developing countries because of high cost and unavailable technology. The ultimate goal of complete evaluation and analysis is the patent of a new molecule and this may not be the only—or even the most important— goal to Third World countries. Although complete laboratories may be rare in developing countries, some specialized in specific pharmacological/therapeutical actions can be found. Indigenous knowledge, and its evaluation through ethnological research, coupled with biomedical data, becomes crucial since therapeutic action indicated by users can give the clues to those physiological processes most likely related to the action of the active compound(s) of the indigenous medicine (Peigen 1983). The possibility of correlating the native therapeutical use with our own biomedical concepts would therefore fit important roles for ethnopharmacological research in developing countries: first, it permits the selection of the correct disease experimental model, that is, the one that would most likely reveal the pharmacological activity necessary to attain the described therapeutical action; secondly, it would allow researchers to direct plants to the appropriate laboratory for analysis and testing. Once this laboratory is provided with the means to make the complete analysis regarding one specific phar- macological action, it would be able to analyse many other plants used for the same Purpose with medium costs. Reduction of costs at the p tage would be reflected in the cost of the final product (Elisabetsky 1985). INDIGENOUS PHARMACOTECHNICS Bernardi (1980) reminds us “ . . . we (scientists) cannot but consider medicines as natural objects, but we have to visualize them in their cultural background as the Products of man, aimed at being used for the health of man” (Bernardi 1980:95}. As Evans- Pritchard (1937:448) put it: “medicines are not natural objects but human artifacts. Native users, though, are well acquainted with the importance of different steps to be 124 ELIZABETSKY Vol. 6, No. 1 followed in order to transform a plant into a medicine. As different parts of the same plant may vary according to their chemical constituents (qualitative or quantitatively), the part selected by natives is of critical importance for phytochemists as well as phar- macologists. It is known that environmental factors may alter chemical constituents (Brown 1984). Users, for example, may state that a certain plant is more effective when collected along the river than this very same species collected in any other place. Or, they may note that a plant is suitable for making the remedy only if collected in the summer, because it is toxic when gathered in winter. Since we are increasingly becom- ing aware of the importance of such environmental factors as sources of modification of chemical profiles of plants, such information should be recorded and carefully analysed. There are several traditional ways to prepare remedies, e.g., concoctions, infusions, cold or hot teas, baths, alcoholic infusions, fermented beverages, heated leaves for plastering, roasted seeds for coffee-like beverages, cigars made out of dried flowers or leaves, enemas, extracted oils and saps, syrups and still many others made by natives (Elisabetsky and Setzer 1986; Posey and Santos 1986). The quality and/or quantity of the substances that will be extracted varies significantly with each of these procedures. Methods of preparation may indicate the means by which an indigenous population deter- mines the best therapeutical use of a particular species (i.e., by extracting the active substance, eliminating the toxic ones and, determining the appropriate mode of administration). Indigenous methods may, and most likely will, be improved through the use of modern technology, but should be taken as important clues in beginning stages of research. It is a common practice among natives to mix more than one species of plant in making a medicine. This can have two basic consequences to further laboratory analytical procedures: (1) it may be possible that only one of the several plants used is actually the one responsible for the desired pharmacological action, although this plant may not be reported as the most important in the remedy; (2) since drugs interact with each other, the presence of one may alter the pharmacological properties of another. There is also the possibility that all or several of the plants used in the mixture have the same phar- macological action that adds to overall effectiveness. Posology (the administration mode and dosage regime} determines the effectiveness of any kind of medicine. Natives can give specific information regarding the posology of their medicines. The same plant, prepared and/or ingested in different ways, is often used for different purposes. Even baths and massages, widespread practices in several indigenous medical systems, as well as among caboclos, are now being reevaluated since new evidence regarding skin permeability may offer a scientific basis for their claimed efficacy (Lewis and Elvin-Lewis 1977). I suggest that, if all these items are carefully taken into account, the results of ethnopharmacological studies can be improved. INTERACTION BETWEEN MEDICAL SYSTEMS Contacts between Westem medical care and indigenous concepts of disease, diagnosis, and cure are continuosly growing throughout the world. Contradictions in conceptual frameworks are apparent and it is clear from ethnographic studies “how ineffective modem therapeutic programs can be when medical personnel fail to gain 4 ing of the principles and concepts of traditional medicine that gove™ the behavior of these patients” (Logan 1978; Rubel 1967). Chen (1978:273] properly con- cludes: “If the dominant system recognizes that it is unable to provide adequately for the care of the population and that other medical systems are filling in this gap for 4 sizeable part of the population, it has little choice but to accommodate this fact.” Summer 1986 JOURNAL OF ETHNOBIOLOGY 125 There is not necessarily conflict between indigenous curers and their biomedical counterparts. This can be seen in the integration of traditional and Wester trained prac- titioners in China, India, Brazil or Malasya where folk and Western care are both institutionally provided (Elisabetsky and Setzer 1986). It is both possible and necessary to improve the interaction between folk and modern medical systems. It should be possible to return ethnopharmacological knowledge, improved through scientific analysis, to the people that most contributed to it and most desperately need it. We can use this knowledge for their benefit through the distribution of scientifically tested active medicinal plants, advising against the use of toxic ones, providing dried plants (cultivated under controlled conditions in order to optimize and maintain a constant chemical con- stitution) making them available through the whole year and/or to places where they do not originally exist; and providing simple pharmaceutical (Gallenic) preparations. Contacts between native and Western medical systems are inevitable. If correctly done, anthropologists can propose health programs that teach the importance of an understanding of and appreciation for local medical theories to western biomedical prac- titioners (Alvarado 1978; Logan and Hunt 1978; Press 1978; Rubel 1967; Spector 1979; Scrimshaw and Burleigh 1978; Worsley 1982). In addition, programs can be devised for indigenous populations with the purpose of informing them about the deleterious effects of allopathic drugs that are more likely to occur due to their misuse (Illich 1975). Such effects are more widespread in the Third and Fourth Worlds as the result of, among other causes, failures in the control of sales, delivery, distribution, and consumption of drugs. This problem is further complicated by the misinformation that accompany remedies sold in these areas (Silverman et al. 1982). Needless to say, ideological influences and economic pressures that result from multinational pressures, colonialism, and imperialism make the resolution of these problems even more difficult (see Silverman et al. 1982; Van der Geest 1981). CONCLUSIONS The discovery of prototype drugs (Malone 1977) is just one of the ways ethnophar- macology can contribute to the improvement of health systems throughout the world. Although a desirable goal, another important contribution is the development of low cost medicines. Figure 1 shows a proposed research model for the various possibilities deriving from ethnoph logical studies and how these goals can be achieved through interdisciplinary collaboration. It stresses the cruicial role of ethnographic research into the practical aspects of indigenous health care as the first step of such a model. It specifies the sequence of steps needed for collection and classification of specimens involved in plant selection, together with the different levels of treatment that can be worked out om the data. aoe At this point it should be emphasized that medicinal plants can be used in different therapeutical contexts. This is the area in which ethnobotanists, botanists, chemists, pharmacologists and anthropologists should work together (Bruhn and Holmstedt 198 1). The outcome of such interdisciplinary research perspective will be (1) the preservation of knowledge that would be otherwise lost because of the abrupt urbanization and develop- ment process occuring throughout the world (Prance 1972; Schultes 1984); (2) the rationalization of ethnop! logical and phytochemical laboratory procedures with a consequent reduction of investments and a greater possibility to develop low cost medicines (Elisabetsky 1985); (3) the promotion of understanding of the impact and Possible damage that a careless and negligent use of allopathic medical care may have ioe native populations that comprise the bulk of the planet’s population (Van der Geest 981, 1983). / 126 ELIZABETSKY Vol. 6, No. 1 Step 1 Step 2 Step 3 Step 4 ANTHROPOLOGY/ BOTANY ETHNO- CLINICAL MEDICAL PHARMACOLOGY! PHARMACOLOGY ANTHROPOLOGY ETHNOBOTANY/ Scientific Pre-clinical Controlled ETHNO- taxonomy harmacological clinical PHARMACOLOGY evaluation of the evaluation of : pharmacological , remedies as used Collection of data action indicated by by indigenous concerning medical users and toxicology groups. concepts, uses and of indigenous practices of remedies. medicinal plants. Level A Level B Level C Level D Return of results Commercialization Commercialization As proposed by to original users of in natura active of phytotherapeutic Malone (phyto- (medicinal gardens, studied plants (dried products (simple chemical studies leaflets, etc.) roots or leaves along pharmaceutical directed to the with information preparations, isolation, purification about therapeutic standardized and and structure action, preparation stabilized) elucidation of active and posology) principles, synthesis FIG. 1.—Propose line of ethnopharmacological research. Based on Carlini (1983). NOTE ly one that has a wholly different chemical structure from existing agents and a wholly different medical application. Every time such a drug is discovered, there will be major changes in the practices of medicine.” (Malone 1983:128). LITERATURE CITED AKERELE, OLAYWOLA. 1985. The thirty-eight world health assembly. Int'l. Tradition Med. News. 1:1. ALVARADO, ANITA. 1978. Utilizacion de Agentes y conceptos Etnomedicos el o de referencia de la medicina occidental. In La Medicina Moderna y la Antropologia Medica en la Poblacion Fronteriza Mexico-Esta- dunidense. (B. Velimirovic, ed.} Organizati Ticana de la tion Paname Salud. Publicaciones Cientificas numero 359. ANTEZANA, FERNANDO S. 1981. Essential drugs—whose responst bility? J. Royal Soc. Med. 74: 175-177. BERNARDI, B. 1980. An anthropological approach: the problem of plants 1» traditional medicine. J. Ethnopharm. 95-98. BROWN, JR., KEITH S. 1984. Adult: obtained Pyrolizidine Alkaloids Summer 1986 JOURNAL OF ETHNOBIOLOGY 127 LITERATURE CITED (continued} defend Ithomiine butterflies against a spider predator. Nature 309:707- 709. BRUHN, JAN C. and BO HOLMSTEDT. 1981. Ethnopharmacology: objec- tives, principles and perspectives. Pp. 405-430. In Natural products as medicinal agents., (J.L. Beal, E. Rein- hard eds.) Hippokrates-Verlag, 405- 430. CARLINI, ELISALDO ARAUJO. 1983. Pesquisas com plantas brasileiras usadas em medicina popular. Revista da Associacao Médica do Brasil, 29: 109-110. CHEN, PAUL C. Y. 1979. Traditional and modern medicine in Malaysia. Amer. J. Chinese Med. 7:259-275. ELISABETSKY, ELAINE. 1985. Etnofar- macologia. In Suma Brasileira de Etnologia, volume de Etnobiologia Tess, DARRELL A. POSEY. 1984. Pesquisa etnofarmacologica e recursos naturais no trépico timido: © caso dos indios Kayapéd e suas implicacoes para a ciencia médica. Anais deo I Simposio eas do oe Umido. In Pr Se RACHEL “SETZER. 1986. Caboclo concepts of disease, diagnosis and therapy: implications for ethnopharmacology and health Systems in amazonia. Jn the Amazon caboclo: historical and contemporary perspectives (Eugene Parker, ed.). tudies in Third World Societies Series, Vol. 32. Williamsburg: In ess. EVANS-PRICHARD, E. E. 1937. Witch- craft, oracles and magic among the de. Clarendon Press, Oxford. FABREGA, HORACIO. 1975. The need for an ethnomedical science. Science 189:969-975. FARNSWORTH, NORMAN R. 1985. Cited in Catherine Caufield, A reporter at large: the rain forests. The New Yorker, January 14, pg. 61. and R. W. MORRIS. 1976. Higher plants—the sleeping ii of drug eon Am. J. P " March-April:46-5 and J. ws PEZZUTO. 1983. Rational approaches to the develop- ment of plant-derived drugs. Paper presented at the 2° Simpdsio de Produtos Naturais, Joao Pessoa, rasil. GERETTI, G. 1983. The pharmaceutical industry and dependence in the third world. Princeton Univ. Press, XIV, Princeton, New Jersey. HOLMSTEDT, BO and JAN G. BRUHN. 1982. Is there a place for ethno- pharmacology in our time? Trends in of geal Sciences, 3(5}: 181-1 ILLICH, TaN, 1975. Medical nemesis: the expropriation of health. Calder & Boyars, Lon JAROSZEWSKI, J. W. 1984. Natural pro- ducts and drug development. Phar- macy International 5:27-28. KLEINMAN, ARTHUR. 1978. Concepts an a model for comparisos of medical systems as cultural systems. Soc. Sci. and Med. 12:85-93. LEWIS, WALTER H. and MEMORY P.F. ELVIN-LEWIS. 1977. Medical botany —plants affecting man’s health. John Wiley & Sons, New York. LOGAN, MICHAEL H. 1978. Humoral medicine in Guatemala and peasant acceptance of modern medicine. Pp. 363-375 In Health and human condition: perspectives on medical ies {Editors}. 1978. Health and human condition: perspectives on medical anthropology. Duxbury Press, Massachusetts MALONE, M. H. 1977. Pharmacological ing gand { evaluation. = 23-53 In Ne natura. ugs with 128 ELIZABETSKY Vol. 6, No. 1 LITERATURE CITED (continued) pharmacological, biological or therapeutic activity. (H. Wagner and P. Wolff, eds.). Springer-Verlag, Berlin. . 1980. Common problems encountered in ethnopharmacolog- ical investigations and how to solve them. Ciencia e Cultura 33:19-31. 1983. The pharmacological evaluation of natural products— general and specific approaches to screening ethnopharmaceuticals. J. Ethnopharm. 8:127-147. MAXWELL, R. A. 1984. The state of the art of the science of drug discovery —an opinion. Drug Dev. Res. 4: 375-389. PEIGEN, XAIO. 1983. Recent develop- ments on medicinal plants in China. J. Ethnopharm. 7:95-109. PRANCE, GILLEAN T. 1972. Ethno- botanical notes from onian Brazil. Econ. Botany 26(3):221-237. PRESS, IRWIN. 1978. Bureaucracy versus folk medicine: implications from Seville, Spain. Pp. 376-387 In Health and human condition: perspectives on medical anthropology., (M. Logan and E. Hunt, eds.) Duxbury Press, Massachusetts. 1980. Problems in the definition and classification of medical systems. Soc. Sci. Med. Bull. 14:45-57. QUINTANILLA, ALVARADO. 1978. Effect of rural-urban migration on beliefs and attitudes toward disease and medicine in Southern Peru. In Medical anthropology. (F. Grollig and H. Hunt, eds.} Mounton, The Hague. RUBEL, ARTHUR J. 1967. The role of social science research in recent h Latin American Research Review 66:37-56. SCHULTES, RICHARD E. 1984. pie cy to progress in medicine. Penk: presented at the VIII Simposio de Plantas Medicinais do Brasil, SCRIMSHAW, SUSAN and ELIZABETH BURLEIGH. 1978. Possibilidades de integracion de la medicina indigena y occidental en America Latina y en la poblaciones hispanicas de los Estados Unidos de America. In La medicina moderna y la antropologia medica en la poblacion fronteriza Mexico-etadunidense. (B. Velimi- rovic, ed.) Organization Panamenri- cana de la Salud, Publicaciones Cientificas n° 359:35-48. SILVERMAN, MILTON, PHILLIP R. LEE and MYA LYDECKER. 1982. Pre- scriptions for death—the drugging of the Third World. Univ. California Press, Berkley. SPECTOR, RACHEL E. 1979. Cultural Diversity in Health and Illness. Appleton Century-Crofts, New York. SVENDSEN, A. B. and SCHEFFER, J. J. C. 1982. Natural Sie in therapy: prospects, goals and means in modem research. te Weekblad 4: 93-103. VAN DER GEEST, S. 1981. Under-the- counter-medicines in developing countries. Pharmacy International 2(12):280-283. 1983. Non-information for patients: selling drugs in developing countries. > hag International, February YOUNG, reseene 1982. The anthro- pology of illness and sickness. Annu. Rev. Anthr. 11:257-285. WANG, P. 1983. Traditional chinese medicine. Pp. 68-75 In Traditional eds.}. World Health Organization, Geneva. WHO. 1979. Drug policies including traditional medicines in the context of primary health care. New Dehli. WORSLEY, PETER. 1982. Non-Westem medical systems. Annu. Rev. Anthr. 11:315-348. eS J. Ethnobiol. 129-149 Summer 1986 AN INTRODUCTION TO ETHNOVETERINARY RESEARCH AND DEVELOPMENT CONSTANCE M. McCORKLE Department of Rural Sociology University of Missouri-Columbia Columbia, MO 65211 ABSTRACT.—One of the newest directions in ethnobiology, ethnoveterinary research and development (ERD) is no more than a decade old. As this label suggests, ERD constitutes the systematic investigation and application of folk veterinary knowledge, theory, and practice. C topics in the field include: veterinary et! i d ethnotaxonomy; ethnoveterinary pharmacology, manipulative techniques, and magico- both present and future ERD Largely stimulated by international livestock development concems, anthropologists and veterinarians have joined forces to tackle the real-world complexities of eth inary systems from a holistic but comparative and production- systems-specfic perspective which gives equal attention to emic and etic analyses of animal health-care problems and their solutions. With the integrated knowledge this inter- disciplinary endeavor yeilds, developers can more readily design and implement socioculturally acceptable and ecologically and economically sound interventions to improve animal health and productivity—and with it, the well-being of human groups whose livelihood depends in whole or in part upon animal husbandry. INTRODUCTION Ethnoveterinary research and development (hereinafter ERD) constitutes such a “new direction” in ethnobiology that as yet there is not even consensus on a label for the field. “Ethnozootechnics” has been suggested as one possibility (Schillhorn van Veen, pers. com.). Sollod and Knight (1983) and Sollod et al. (1984) have coined the epithet “veterinary anthropology.” And here I opt for the more generic rubric that forms the title of this review. If labeling this domain of study is somewhat problematic! defining it is even more so. Its boundaries are diffuse, shading off at the edges into a variety of different disciplines and subdisciplines in both the hard and the “soft” sciences, and in both “pure” and applied research. If ERD cannot be easily bounded disciplinarily, neither can it be expediently defined—as sometimes done for other “fuzzy” fields—as ‘‘whatever an ‘ethnoveterinarian’ does.” No such creature exists! However, as Sollod et al.’s (1984) label indicates, the principal actors in ERD are veterinarians and anthropologists, working both singly and jointly. The latter are almost exclusively sociocultural antl pologists, although occasionally a folklorist, linguist or €ven an archaeologist may investigate a topic directly or tangentially related to animal health, Among veterinarians, a number of fields are represented: epidemiology, immunology, parasitology, pathology, pharmacology (or pharmacognasy) and physiology. There is also room in ERD for contributions from: many of the biological sciences, €.g. botany, ecology, ethology, entomology, zoology; certainly from specialists in animal 130 McCORKLE Vol. 6, No. 1 herders about veterinary techniques ment practices, while also observing forage an! ge hs water conditions and e examining general herd health. Below, the author (unseen) cee The ethnoveterinary researcher on the job. Above, riding the range: interviewing alpaca and animal es for the f folk versus “‘modern’ ’ remedi many ills afflicting their sheep. = mmm ll Summer 1986 JOURNAL OF ETHNOBIOLOGY 131 husbandry, range science and water management, and at the level of veterinary policy, planning, and extension, from rural sociologists, economists, agricultural economists, communications experts and others. Given this range—both actual and potential—of researchers and their research orien- tations, a strict definition of ERD is difficult and perhaps not even desirable. However, a very broad definition can be offered: ERD constitutes systematic research and develop- ment which takes as its principal subject or its major departure point folk knowledge and beliefs (theories, taxonomies, definitions, diagnoses, etc.}, practices, technology and resources, social organization and so forth pertaining to any aspect(s| of animal health among species raised or managed by human beings. In this definition I have opted for the term “folk” (or in Francophone writings, “populaire”) rather than, e.g., “traditional” or “indigenous” merely in the interest of historical precision. The latter two terms frequently appear in ERD titles, but a people’s veterinary beliefs and practices are not always entirely or demonstrably traditional or indigenous. Instead, they may represent a melange which incorporates elements from other ethnic groups and/or from modern veterinary science. In the latter regard, folk systems may have absorbed these elements (albeit often imperfectly) through word-of- mouth diffusion, by contact with commercial livestock operations, or from veterinary extension services. In fact, as extension efforts intensify, folk veterinary medicine around the world tends to become ever more syncretic. Leaving aside this minor terminological point, as for “‘aspect(s) of animal health,” these naturally incorporate all features of livestock production systems which can impact—whether positively or negatively, directly or indirectly—upon the physical con- dition of the animals being managed. At the broadest level, this includes all husbandry techniques involving: feeding, watering, range and pasture management; manipulation of breeding, reproduction and herd composition and dynamics; housing and supervision; prevention, control, curing of disease and, relatedly, sanitation in all management opera- tions; and harvesting of animal products. From an emic perspective, supernatural husbandry techniques—like reproductive, protective, or propitiatory rites and magical cures for animals—must also be included in this list. Ultimately, too, the larger ecological, €conomic, political, sociostructural and ideological contexts of the animal production System itself are implied in ERD in its fullest formulation—at which point it in truth €comes “veterinary anthropology.” Having dealt at least provisionally with labeling and defining ERD, the next step is to identify the corpus of work falling within its purview. Here, the definitional qualifiers “principal,” “major,” and especially “systematic” come into play. Desultory references to folk veterinary beliefs and practices or related husbandry ee can be found Scattered throughout many works. These include: ethnographies* of peoples whose livelihood depends upon animals; accounts by travelers, missionaries, former sseseiug authorities or other officers (e.g., de St. Croix 1972); writings in medical anthropology, archaeological treatises,4 field-based studies in veterinary medicine and range management;” and still others.© Naturally, all such sources of information should be consulted by the ERD researcher in preparation for work among a given ethnic group or on a specific animal health issue. However, they do not fit any definition of ERD per se. Either their treat- ment of matters ethnoveterinary is asystematic, anecdotal, and very much subordinate to a different principal topic (the most common case); or their data base falls wide of the “folk” mark. Just the opposite is true of the works reviewed here. As a first effort at drawing together ERD worldwide, the following introductory review is perforce non- comprehensive. ” Nevertheless, the studies referenced and discussed below do constitute the bulk of the literature to date, and they accurately represent the variety of thrusts in the field. 132 McCORKLE Vol. 6, No. 1 DISCUSSION ERD background, development, and goals.—With one qualification, studies which take folk veterinary beliefs and practices as a primary topic of scientific investigation first began to appear in the mid-1970s. In veterinary medicine this statement is qualified by 1 id 1 1 the longstanding study and use in veterir } of herbal lary Pp gy and p remedies for animals (e.g. Bairacli-Levi 1984; Schillhorn van Veen, pers. com.). In anthropology, however, it seems to be unqualified—despite an established interest in the study of domesticated animals from a number of perspectives (Shanklin 1985b). Between the mid-70’s and now, ERD can indeed be said to have burgeoned. Predicatably, it is difficult to arrive at many generalizations about the field overall. Researchers come from a variety of countries and disciplines; their research issues, emic/etic emphases, and theoretical approaches (where these exist) vary accordingly; their geographic areas of investigation girdle the globe; the species involved can include any animal domesticate or semi-domesticate; and, of course, the field itself is still in a phase of rapid growth and change. Where this decade of diversity acquires coherent focus, definition and purpose, however, is in the arena of international livestock development and extension. Here, ethnoveterinary research has as its explicit, overarching goal the enhancement of livestock productivity through improved management of animal health, as informed by an understanding of folk veterinary medicine and related husbandry techniques. Largely with the impetus from development projects like the Small Ruminant Collaborative Research Support Program (SR-CRSP) and the Niger Range and Livestock Project, as of the 1980’s a handful of ‘“‘core’’ works and workers in ERD have emerged. This core of ethnoveterinary endeavor is characterized by its holistic, systems analysis, and therefore interdisciplinary orientation. That is, it recognizes the impor- tance and interconnectedness of the physical, cultural, social, economic, political and historical matrices in which animals and their owners are embedded. It therefore seeks to integrate findings from correspondingly appropriate but disparate disciplines in the biological and social sciences (after Sollod et al. 1984:285-286). Additionally but not distinctively, core ERD emphasizes the need for firsthand field research among stockowners themselves, under real-world husbandry conditions, in order to arrive at any meaningful comprehension of this systemic complexity on the ground. To this end, it draws heavily upon anthropological method and theory, combining these with the technical skills and knowledge of animal scientists. It is, in fact, ‘veterinary anthropology.” This core thrust in ERD has come to the fore only in the last five years, and it clearly charts the course of the field’s future growth. As noted earlier, to date it has almost exclusively involved veterinarians and sociocultural anthropologists. And mainly due to present policy priorities in international development, it has so far concentrated upon herd animals (cattle, sheep, goats, alpaca, llama) in Africa and, to a lesser extent, Latin America. ; In contrast, the first half of the field’s formation displays a greater diversity 10 researchers, species and geographic locales, although many of the research topics are the . The iverse’’ studies continue to increase® in quantity and quality, and much of the data they produce are immediately relevant to core ERD concerns. But again, they are differentiated by their more delimited and disciplinary-specific research goals and approaches. In this respect, the holistic, systemic and ultimately practical thrust of core ERD has lent fresh meaning to the congeries of studies in the field as a whole, placiné them into a more unified heuristic framework. The following discussion is organized by general topical areas which have been addressed in any part of ERD to date. Throughout, the relevance of each area to develop- _s EE ee a —— ee ee OO A IE tt a a a Summer 1986 JOURNAL OF ETHNOBIOLOGY 133 ment and extension is highlighted. The topical categories themselves are not discrete; they merely serve as an organizational device. Many studies in fact span a variety of categories. Due to their holistic orientation, this is particularly true of core works. In such cases, studies are often cited and/or discussed in several sections. Veterinary ethnosemantics” and ethnotaxonomy. I begin with this area because it forms the backbone of almost any endeavor in ERD. The importance of even the most basic semantic and taxonomic researches for d ing and analyzing indigenous veterinary and husbandry concepts and how these guide behavior, for identifying different types of native veterinary practitioners, and for communicating with stockowners and extending new health-care information and techniques to them is recognized by virtually every core work. The major theme in such research has been the relationship between folk and scien- tific taxonomies—especially in the domain of livestock diseases, where an in-depth, empirical appreciation of the shape, scope and accuracy of a people’s etiological, anatomical, physiological, diagnostic, curative and epidemiological knowledge is essen- tial before developers can even begin to evaluate what, how, and if native veterinary practices should be altered. A considerable number of ERD studies therefore devote attention to trying to sort out and ‘match up” folk disease identifications and/or taxa with their scientific equivalents (Ba 1982, Grandin 1985, Ibrahim 1984, Ohta 1984, Maliki 1981, McCorkle 1982a, 1983b, Sollod 1983, Sollod et al. 1984, Wolfgang 1983, and Wolfgang and Sollod 1986; possibly also *Cabrol 1984 and *Noirtin 1975). Predictably, this is not an easy task. Medical science classes diseases according to the etiological information afforded by sophisticated laboratory analysis. In contrast, at least pending practical necropsy, folk disease distinctions typically rely on the recogni- tion of morbid signs, more rarely on epidemiology, sometimes on sorcery, or on any com- bination of these. Moreover, as Ohta (1984) points out, when pathogenic explanations for disease are lacking, it is often difficult to distinguish ‘disease names” from ‘ terms of symptom” since both may reference morbid signs. Further complicating this picture is the fact that, as among the Twareg of Niger (Wolfgang and Sollod 1986), the same morbid condition may have several appellations depending upon the species afflicted. The result is that a single folk disease category—like g’icha ‘diarrhea’ among the sheep and camelids of the Quechua of Peru (McCorkle 1982a), wilsere ‘bush disease’ among the cattle of the FulBe of Upper Volta/Burkina Faso (Wolfgang 1983), or azania ‘too much blood’ among Twareg camels (Wolfgang and Sollod 1986)—often glosses a wide array of etiologically distinct ailments. Conversely, folk classifications may also assign the scientifically ‘same’ disease to different categories on different occasions, based on vary- ing configurations of the clinical, epidemiological and sur tural information available to the native diagnostician and on the species involved. Nevertheless, it is clear from these and other studies that pastoral peoples possess a rich store of knowledge about many livestock diseases. To take but one example, Schwabe and Kuojok (1981) describe the extensive appreciation of cattle diseases (and of bovine anatomy and physiology) held by traditional Dinka healers and stockowners. This lore derives from practical experience—e.g., personal observations of clinical signs, Sacrificial dissections and specific instances and modes of contagion—coupled with a “rational empirical process’”’ (Schwabe and Kuojok 1981:237) which integrates these and other sources of information. Still, as nearly all researchers of ethnoveterinary €pistemology have remarked, some of the resulting folk surmises, explanations and Curative or preventive actions are “incorrect in major or minor parts” (Schwabe and Kuojok 1981:237). _—_ Simple semantic and taxonomic investigations can help to pinpoint where stockowners could most benefit from increased etiological and epidemiological infor- 134 McCORKLE Vol. 6, No. 1 mation, more astute diagnoses, and new treatment, prevention and control options. For the same reason, research into other semantic domains of the animal production system (Anderson 1978, Ba 1982, Flores-Ochoa 1978, Maliki 1981, McCorkle 1983b, Meneses T. in progress) is valuable insofar as many husbandry practices impact upon the occur- rence and spread of livestock diseases. Finally, all such research is critical for effective communication between stockowners and development/extension workers. As so many authors have pointed out, the labors of both groups would be eased if they can learn to comprehend and utilize each other’s veterinary concepts, techniques and vocabulary. Ethnoveterinary pharmacology.—This is the investigation of a people’s use of plants and other materials in preventing and treating animal diseases, wounds, fractures, in encouraging fertility, appetite, productivity, and so forth. Most core studies make at least mention of this very basic aspect of veterinary care, and some go into considerable detail (Ba 1982:55, 87 ff.; Maliki 1981:47 ff.). Works whose specific focus is the ethnoveterinary pharmacopoeia can range from the folkloristic to the “high tech.” Many have an essen- tially descriptive aim—ie., identification of the materials, their appellation, categoriza- tion, acquisition, preparation, indication, administration (including both natural and supernatural operations) and reported efficacy. Such works may take a purely ethnographic approach. An example is Brisebarre’s (1984a) study of the therapeutic use of boquets hung in the sheepfolds of Cevennes, along with her examination of more empirical curative applications of plant and other materials to Cevenol ovines (Brisebarre 1978). Alternatively, descriptive studies may have a more strictly pharmacological end in view, as in Nwude and Ibrahim’s (1980) detailing of 92 plant species employed in traditional veterinary medicine in Nigeria for every type 0 domestic livestock (possibly also *Gourlet 1979). Likewise for Chavunduka’s (1976) identification of 53 plant species of ethnoveterinary medicinal importance in southern and eastern Africa, along with their uses, preparation and administration. For veterinary pharmacologists, identification and description are but the first steps toward controlled scientific screening of local plants in order to establish their real utility if any, optimal rene ie and effective frequency of application (e.g. Ibrahim et al. 1984, *Mourier-Ballon 1983). While research of this sort can add useful new drugs to the modern veterinary pharmacopoeia, its ERD importance lies in improved folk pharmacotherapy which is culturally appropriate, economically feasible and consistently available. At this level, its relevance to development and extension is evident. An example is provided by the SR-CRSP/Peru. Building upon existing ethnoveterinary pharmacological knowledge, the project has worked with one peasant community in the central highlands to test the efficacy of a wild tobacco as a botanical for ovine ectoparasites (Bazalar and Arevalo, in progress). As per the longstanding and widespread use of nicotine-based parasiticides in both folk and modern veterinary medicine (Schillhorn van Veen, pers. com.}, initial trials have proved successful; and work is now being done to establish the minimum effective compound and to secure supply of the plant (Fernandez 1985). The project also plans to test these tobacco compounds in combination with tarwi (Lupinus mutabilis) water. Tarwi is a bitter, alkaloid-laden legume which is edible only after prolonged steeping. The trials Bustinza Ch. (1985) performed on this indigenous cultigen’s use in southern Peru as a folk cure for ectoparasites of alpaca have already demonstrated its efficacy. Working in conjunction with SR-CRSP social scientists, project veterinarians are conducting similar trials on other plant materials in the ethnopharmacopoeia which are employed to combat ovine endoparasitism (Arevalo and Bazalar, a, b, in progress). ghout, emphasis is placed on compounds and applications which can be readily prepared and comprehended within the peasant community itself. Summer 1986 JOURNAL OF ETHNOBIOLOGY 135 Ethnoveterinary manipulative techniques.—This topic is distinguished from pharma- cotherapy above and magico-religious procedures below by its primarily mechanical nature—although no such distinctions may be drawn emically. Of course, all these approaches may be used conjointly—as when a stockowner surgically cleans and then sutures a wound, poultices it, and offers up a prayer for the animal’s speedy recovery. For convenience, here I lump vaccination and other prophylactic measures with the healing arts—bonesetting, surgery, wound treatment, chiropractic-like manipulations and, at least in China (Metalie 1984), acupuncture. As before, ERD’s concern is to identify and describe, discover the emic rationale for, and evaluate the appropriateness and effectiveness of such manipulations. Ethnoveterinary prophlaxes may be of an essentially empirical, managerial sort, e.g.: smudge fires to drive away disease-bearing pests; manual removal of ticks; avoidance of infested pastures and unclean water, quarantine of contagious individuals; mineral feedings, protection from extremes of weather; and general sanitation measures like cleaning, disinfecting or rotating animal quarters. They may also include various magico- religious performances, taboo observances and so forth (see below). But a more classic example of ethnoprophylaxis is traditional vaccination. For instance, some FulBe vac- cinate their cattle against rinderpest by inserting a bit of lung from an infected animal into an incision in the nose, leaving the material in place until the wound festers; others inject a solution in which the lung tissue has been soaked (Wolfgang 1983:58). Fulani (Ba 1982:75) and WoDaaBe (Maliki 1981:60) follow similar procedures for bovine pneumonias. Upon completion of the vaccination process, WoDaaBe also excise the rotting flesh and cauterize the wound. As a healing art, cauterization appears to be a routine and multi-purpose technique among all Sahelian pastoralists. For example, FulBe treat livestock sprains with a series of tiny burns in the sprained area—much like the “pinfiring” performed on Western racehorses with leg problems, to increase blood flow to the injured part (Wolfgang 1983:57). FulBe, Fulani, Twareg, and WoDaaBe, whether rightly or wrongly, all use branding in treating a galaxy of ills. Across the three ethnic groups, these ills include, e.g.: anthrax, trypanosomiasis, rickettsiosis, epilepsy, edemata, botulism, scabies, bloat, diarrheas, toothaches, fevers, blows to the body, digestive and hoof ailments, muscle pains, sprains and lizard bites. Venesection or bleeding is another popular healing art in African veterinary practice. All of the foregoing authors plus Evans-Pritchard (1969), Ohta (1984), Schwabe and Kuojok (1981), Wolfgang and Sollod (1986) and others note its use. Bonesetting and wound-treatment skills are found in folk veterinary toolkits worldwide—as are, too, effective surgical and obstetric techniques. These latter run the gamut from relatively simple operations (such as marking, castration, excision of tumors, ttain amputations) through a variety of obstetric procedures (e.g., episiotomy, Caesarean section, embryotomy] to complex cosmetic surgery like horn training (Schwabe 1984). Magic, religion and ethnoveterinary medicine.—This topic has received considerable attention in ERD for a variety of reasons. Admittedly, it is precisely the sort of exotica ich anthropologists dote on, and it readily captures the veterinarian’s curiosity as well. More importantly, however, magico-religious beliefs and practices appear to form a part of folk veterinary systems everywhere; and in many, emic distinctions between natural and supernatural matters in animal health are blurry.!0 If for no other reason than its Pervasiveness, the supernatural must be acknowledged in any ERD study aspiring to a holistic, systems-analysis approach. As an overarching ideological SEEK, the super- tural can impinge upon every facet of livestock production. However, from an €xamination of the literature, magic and religion seem to figure most prominently in two areas pertaining to animal health: in the supernatural promotion of livestock 136 McCORKLE Vol. 6, No. 1 fertility and productivity;!2 and more significantly, in ethnoetiology—which in tum informs folk diagnosis, treatment and prevention of animal disease and accident. Maliki (1981:65 ff.) presents one of the most thorough-going descriptions of a people’s supernatural pastoral repertoire. Writing on the WoDaaBe of Niger, he discusses: fertility, protective and curative rites for animals; hexes, curses and broken taboos which can bring on livestock disease and accident; divination procedures for predicting herd misfortunes; “good and bad luck” days for performing veterinary and other management operations; and more. McCorkle’s (1983b) treatment of these same phenomena for the Peruvian Quechua is equally detailed. However, in addition to describing these Amerind’s panoply of super- natural explanations for animal ills, she seeks to analyze them etically. The Quechua etiological category of “evil winds” is illustrative. Indeed, ‘“winds’’ are common etiologies in a number of folk veterinary systems, including FulBe, Fulani, Twareg and WoDaaBe. This comes as little surprise since certain livestock diseases in fact can be transmitted aerially (e.g., anthrax, foot and mouth disease, rinderpest) and/or promoted by environ- mental stresses (e.g., a variety of respiratory ailments). Among the Quechua, however, ethnodiagnosis of attack by an evil wind may or may not correspond with any plausible scientific equivalent. Sometimes this diagnosis appears to gloss plant poisoning; sometimes it references a tumorous growth; at still other times, it cannot be linked to any specific clinical signs. Nonetheless, it can often lead to appropriate prophylactic or treatment measures—e.g., keeping animals away from the haunts of evil winds cum toxic plants, or surgically removing tumors. Whether etically translatable or not, as already noted, magico-religious belief and practice figure in folk veterinary systems worldwide, in both developing and developed milieux. To illustrate, Wolfgang (1983) mentions FulBe magical techniques for con- trolling, avoiding, or curing certain cattle diseases and ethnoetiological agents such as genies. Ibrahim (1984) comments on ‘‘spirits” and ‘‘the unseen” as explanations among Nigerian Fulani for livestock diseases with unknown (microscopic) causes = neurological signs. Chavunduka (1976:8) notes Manyika tribal beliefs in ancestor spirits and “evil dreams” as origins of disease. “Evil beings” plague Turkana livestock (Ohta 1984). Recurrent themes in Kimball’s (forthcoming) observations on Brunei Malay hantu spirits, and Islamic prayers such as the ‘neutralizing harm verse” to forestall various kinds of livestock problems. For Irish stockowners, Shanklin (1985) describes evil-eye theories of animal ills, and their associated ritual and behavioral precautions. Brisebarre (1978, 1984b) and others (cited in Brisebarre 1985b) document a pantheon of French “veterinary saints’’ to whom provincials still turn to bless, protect, cure and m tiply their livestock. And Brisebarre (1985c) analyzes the principles of sympathetic magic behind French stockowners therapeutic use of boquets. Finally, many of the foregoing and other studies (e.g., Schwabe and Kuojok 1981) further indicate what social types 0 individuals (priests and shamans, sorcerers, herbalists, smiths, heads of household or lineage, wives, etc.) are traditionally responsible for the various supernatural—as we as naturalistic—operations related to animal well-being. For development and extension, the importance of understanding supernatural aspects of folk veterinary systems is threefold. The most obvious consideration 1s 4 diplomatic one. If ERD personnel ignore, belittle, or worse still, unwittingly outrage indigenous ideology, their work is not likely to meet with much success. A second ideration is that ti nagico-religious practice and idiom in fact em! : tical veterinary and management acumen. Treatments like feedings of saint-blessed salt (Brisebarre 1984b) are potentially effective for some maladies; and seemingly outze ethnoetiologies like ‘evil winds” (McCorkle 1982a) can nevertheless dictate appropriate curative or preventive action. Developers must therefore be careful about dismissing Summer 1986 JOURNAL OF ETHNOBIOLOGY 137 “superstitions” out-of-hand. Third, extension efforts can directly build upon an under- standing of the supernatural in folk veterinary systems. Useful management techniques can be reinforced with added information as to how genies, spirits, evil winds/dreams/eyes or what-have-you accomplish their nefarious aims, and new skills can be introduced in a cultural idiom which makes sense to stockowners or at least does not threaten ideological, and related sociostructural, integration. Ethnoveterinary extension.—With regard to social structure—and as Halpin (1981), McCorkle (1982a), Schwabe and Kuojok (1981), and others have pointed out—one of the most logical choices for recruiting and training effective veterinary extension personnel is local healers who have traditionally dealt with animal (or human) health problems. These specialists or semi-specialists typically share the same language and culture as their clientele; already enjoy their confidence and esteem (albeit to varying degrees); occupy a recognized role in the ethnomedical system; and often control a wide range of empirical medical skills and knowledge. Identifying these individuals, their established domains of practice, their real expertise, and their potential as veterinary extension workers is yet another important task in ERD. Along the lines proposed for use of traditional healers in human health care in many developing countries (e.g., Dunlop 1975), Schwabe and Kuojok (1981) emphasize that, with some training and organization, such individuals could provide effective and relatively cheap grassroots delivery of basic health services to livestock, and possibly even to humans. Halpin (1981) advises that these “barefoot vets’ can be drawn from among stockowners as well as healers. He further notes that a trained coterie of camp- level veterinary extensionists could be particularly effective in nomadic areas, where other types of delivery are so problematic for so many reasons (cf. Imperato 1974). In developing nations, these “paravets” could additionally function as a unique component in a ‘badly needed disease intelligence system” (Schwabe and Kuojok 1981:237) and as accurate interpreters of stockowners’ primary veterinary ‘troubles, constraints, fears and aspirations” (Halpin 1981:5). As these authors point out, such information would in tum permit more rational design, performance and evaluation of livestock disease control programs. Grandin (1985), Halpin (1981), Loutan (1984), D. Sandford (1981), S. Sandford (1983), Schwabe (1980), and Schwabe and Kuojok (1981) all offer suggestions and observations on how selection, training, supervision, motivation an tion, logistics, supplying, Teporting and accounting procedures, and etc. of paravets can or has been organized vis-a-vis: multilinguistic realities; complex national government and local social structures—including household, camp, village-chief, interethnic and common-interest group organization, the veterinary worker's specific role within and responsibility to these Structures; epidemiological profiles; and animal management and ee Se Summarizing the lessons learned from the Niger Range and Livestock Project’s pilot Paravet program, Loutan (1984) provides a particularly thorough and insightful case study which addresses a majority of these issues. Animal health and livestock production systems research. —All of the foregoing considera- tions and topics are implied in this final category, which embodies the core of current ERD. Works in this vein may naturally differ in their topical emphasis and scope, often depending upon the author's disciplinary training and subdisciplinary interests. They may highlight veterinary, management, or socioeconomic and sociocultural aspects of the animal-health and production-system issues examined. They can also vary in their Primary, immediate goals of research: thorough-going description, disciplinary theory- building and validation, policy planning, advocacy of a given research design, or investiga- tion of a specific animal-health question. However, all studies in this group share two 138 McCORKLE Vol. 6, No. 1 defining features: an explicit recognition of the holistic, systemic complexity of the phenomena under study; and an ultimate commitment to making research results useful for livestock development and extension. Among the first works in this group to reach print is Maliki’s (1981) report on WoDaaBe cattle herders in central Niger. The range of topics he treats is indicative of these studies’ holistic outlook. To illustrate as briefly as possible, he details: herd composition, dynamics, and ownership and use rights, along with all Fulfulde semantic distinctions in race, sex, age, reproductive and productive state, and personal names for cattle, plus additional categorizations for sheep and camels; every aspect of basic animal management such as p ing/mineral-feeding/watering patterns and_ selection/ breeding/fertility/g ion/abortion/calving/milking, WoDaaBe description and classi- fication of plants according to their palatability and nutritive value for the different animal species and at different stages of plant growth, plus their veterinary medicinal and other uses; similarly for identification of livestock diseases and other health problems—their ethnoetiology, the clinical signs herders recognize, the specific cures and controls they seek to apply; herd movements during the eight emic seasons of the pastoral year and their impact upon the social groupings and activities of families, camps and clan; relatedly, the role of animals in rites of passage, friend-and kinship, social status and recreation; harvesting, consumption and distribution of all pastoral products; magical beliefs, songs, proverbs, origin myths and etc. pertaining to herds; and still more—all with precise transcriptions of the hundreds of lexemes in the WoDaaBe herding vocabulary. Ba’s (1982) treatise on the ‘‘veterinary arts’’ among Sahelian Fulani (Peul) follows a similar format, but with a tighter focus on veterinary and related management practices, and a more limited discussion of social, economic and cultural correlates of Fulani animal hus- bandry. Both studies are essentially descriptive. McCorkle (1983b} covers largely the same topics as Maliki—plus others such as the social organization of labor for herding (1982b), and management issues in sheltering, shearing, docking, castration and predator control (1983a}]—for Quechua Indians of Peru. However, she has a theoretical as well as a descriptive aim: to correct neofunctionalist analyses of agropastoral subsistence systems. Using a New World data base to refine and validate the cross-cultural applicability of a ‘dialectic’ model of preindustrial agropastoralism in Europe, she demonstrates how Andean herding and cropping stand in a simultaneously complementary and competitive relationship to each other. In the process, she outlines how veterinary care, in particular, is constrained by the low pro- ductivity and multiple competing demands of paleotechnic agriculture. Under their present ‘meat and potatoes” production system, this leaves Andean peasants short of land, labor, capital, technology and technical information for significantly increased attention to herd health problems—certainly insofar as intensive, costly, “‘tech-fix’’ solu- tions derived from Western commercial practice are concerned. For livestock develop- ment and extension, McCorkle further discusses some of the systemic potentials and problems posed by ecological, sociostructural, and sociopolitical factors relating to, €8- communal land tenure and pasture/field usufruct rules, traditional reciprocal labor patterns and centuries-old ethnic dominical mechanisms. The ultimate implication for livestock development is that only a global, systems analysis which acknowledges the dialectical tensions between preindustrial cropping and herding can forestall the error of “robbing Peter to pay Paul’’”—i.e., of upping pastoral production at the expense of agriculture, or vice versa. € paramount concern of Wolfgang’s (1983) work among the FulBe of west-central Upper Volta is to arrive at specific recommendations for veterinary extension and policy planning. To this end, she focuses her research on three major areas: (1) FulBe classi- fications, etiologies and treatments (both folk and Western) for cattle diseases, plus herders’ own assessment of the socioeconomic impact of different diseases; (2) the Summer 1986 JOURNAL OF ETHNOBIOLOGY 139 current structure and functioning of animal health-care delivery services in the region; and (3) a survey of the country’s major veterinary diagnostic laboratory facilities. Additional topics of investigation include certain non-disease-related health problems of cattle and (especially in Sollod et al. 1984) women’s role in maintaining herd well- being. Findings from all these areas inform Wolfgang's final recommendations for veterinary extension and policy in Upper Volta (now Burkina Faso). A sampling of these recommendations is of interest because they reflect needs common to many developing countries. One is an immediate improvement in epidemiological information, so that planners can concentrate scarce resources on the most prevalent, economically damaging livestock diseases. Another is educational outreach to correct folk misunderstandings about and consequent misuses of expensive Western drugs. A closely related concern is to remove communication, and even simple translation, barriers between stockowners and extension agents—a problem which, theoretically at least, could be resolved by incorporating some herdsmen into the livestock service, as has been done in other parts of Africa. Finally, Wolfgang notes a need for modest improvements in regional laboratory diagnostic facilities, and in other technological and infrastructural aspects of health care delivery. Throughout, however, she emphasizes that including stockraisers themselves—both women and men—as substantive parti- cipants in the extension process should greatly enhance diagnostic, delivery, and treat- ment effectiveness and cost-efficiency. Sollod et al.’s (1984) aim is somewhat more didactic and programmatic than that of the foregoing studies. These authors seek, first, to define and codify the exciting new trend in ERD which tackles animal health and production system research through “veterinary anthropology.”” Then, drawing upon the fieldwork of Sollod and Knight (1983) (a veterinarian and an anthropologist) among herding groups of central Niger, plus Wolfgang’s investigations (which were in part supervised by Sollod), they demonstrate how this fusion of perspectives and methodologies can greatly enrich analyses of Patterns, problems, and control options in livestock health. Sollod and Knight’s Niger research is particularly instructive. There, the inter- disciplinary team was able to relate epidemiological profiles of livestock diseases—their incidence, prevalence, seasonality and geographic distribution—directly to differing Systems of animal production (Twareg versus WoDaaBe) and to specific husbandry prac- tices within these systems which promote or discourage expression of a given ailment. These practices, in turn, were linked to concrete ecological, cultural, commercial and subsistence parameters of Twareg and WoDaaBe life. For example, it was found that stress- telated pneumonia and protein-caloric inanition were severe problems among WoDaaBe, but not Twareg, sheep. This finding was related to the seasonal timing of ovine births. The Twareg control breeding through penile sheath ligation of rams, thus ensuring that ewes do not give birth towards the end of the dry season, when forage is scarce and nutri- tion poor. In contrast, the WoDaaBe—who consider themselves to be cattle herders— expend little effort of any sort on their sheep. The only “control” they exercise on breeding is sales of rams in response to market demands for mutton, especially at the time of the annual Id festival. Depending on whether this moveable feast falls before or after the first breeding season, WoDaaBe ewes and their lambs suffer or thrive accordingly (after Sollod et al. 1984:291). Bee The veterinary anthropology which these authors espouse highlights the dynamic interplay of endogenous and exogenous determinants of disease—the latter defined as factors external to etiological agents or their hosts. The contextualized, culture-specific information which this comparative stance yields is critical for the design of successful development and extension programs because ‘It makes possible the use of nonmedical approaches to animal health which include marketing and management interventions, and allows the use of a simplified package of veterinary commodities for each pro 140 McCORKLE Vol. 6, No. 1 duction system” (Sollod et al. 1984:292). It has long been recognized, and repeatedly demonstrated, that changes in management alone are sufficient to control many livestock diseases. Yet as Schillhorn van Veen (1984:306-308) has observed, despite the fact that such interventions can be highly beneficial at relatively low cost and risk, management is rarely defined for indigenous stock operations. The interdisciplinary, holistic and production-systems approach of veterinary anthropology works to fill this definitional and empirical lacuna. Finally, by virtue of its holistic, production-systems orientation, Shanklin’s (1985a) work among farmer-stockowners of northem Ireland also falls in this last group of studies. The ethnoveterinary portion of her monograph is designed to test a single hypothesis: “that if different types of animals are kept in a given environment, susceptibility to disease will be a factor in the decision to keep a specific type of animal and selective breeding will be largely determined by this consideration” (Shanklin 1985a:215). To this end, she marshals comparative data on bovine as versus ovine production with regard to: stabling, pasturing, seasonal supplemental feeding, both folk and scientific veterinary knowledge and care, breeding practices, land and labor requirements, economic value and market outlets, government regulation and historical shifts in these and other pro- duction parameters. Her larger aim is to review theoretical debates in ecological anthropology relating to the adaptive value of traditional and non-traditional elements in the animal production system, and to identify ecological constraints to indefinitely upping livestock production. CONCLUSIONS ERD is still in its infancy—or perhaps with the appearance of conceptually and disciplinarily more integrative papers like Schillhorn van Veen (1981, 1984), Sollod et al. (1984), and the present review—its early adolescence. As is to be expected of a young area of research, many ERD works are still focused on the descriptive level; and across the field as a whole, there is a healthy diversity of topics and approaches. Again, where this diversity finds a unifying form and function, however, is in international livestock development. Here, ERD is of critical importance because without improvements 10 animal health (and nutrition), rarely can any improvements in livestock productivity be achieved. In response to this need, a contemporary core of development-oriente ethnoveterinary research has emerged. Within this core, a number of shared themes, methodologies, and perceived needs for future research can be distinguished. First and most salient, of course, is an emphasis on the “ethno” in ERD. As recognized for other development sectors (cf. Brokensha et al. 1980), a thorough-going understanding of and respect for folk veterinary knowledge, concepts, practice and practitioners is a must. While clearly not all elements of oveteri {and their associated management, sociostructural, and etc.) systems are accurate or effective, their ensemble represents a rich resource for developers seeking to enhance animal health and productivity in ways which are readily compre- hensible and culturally acceptable to the client audience and which are ecologically an socioeconomically sound. In other words, existing folk practice and belief should always be the starting point for veterinary research, development and extension—as, inde they were in the evolution of Western veterinary medicine. Second, as we have just seen, there is an invigorating move in ERD towards analyzing veterinary development issues within a holistic but comparative and Pro" duction-systems-specific framework. Production systems or subsystems may be defined ture area, ethnic group, agroeconomic sectors (e.g. cropping versus herding, sale versus subsistence}, intraethnic household characteristics, species or other parameters like ecozone. This new dynamic in ERD has brought with it an explicit recognition that Summer 1986 JOURNAL OF ETHNOBIOLOGY 141 the constellation of endogenous and exogenous variables impinging upon animal well- being ultimately lies beyond the ken of any one technical or social science. Ideally, research into the complex, real-world coordinates of livestock health should therefore be carried out by concerted interdisciplinary action. Veterinarians and anthropologists have together risen to this challenge; and there is both room and need for collaboration with other disciplines, as well. Such research naturally calls for in-depth field-based studies rather than just laboratory analyses ‘divorced from the realities of pastoralism’’ (Sollod et al. 1984:285). In this regard, the usefulness of time-tested methods of anthropological fieldwork is undisputed in ERD. Likewise for the ethnographic expertise and the emic, bottom-up perspective of anthropology. In the findings and hypotheses of animal health and production systems research to date, some consensus on development and extension strategies is also emerging. To wit, that educational, managerial, marketing, and other such interventions may often prove more appropriate, economical, and effective than modern drug therapy, eg., as applied in mass vaccination and treatment schemes. In the rush to implement costly top-down, ‘‘tech-fix” programs which offer immediate short- term benefits, developers, policy planners and stockowners alike may lose sight of longer-term drawbacks to such solutions in third-world countries. These drawbacks can include: ecological degradation and depletion, as from overgrazing; relatedly, escalating social and political tensions over competition for scarce feed and water; spasmodic breakdowns in veterinary supply and delivery lines due to an unstable economy and/or government, or to infrastructural inadequacies; political and financial machinations within the livestock service; loss of genetic tolerance to disease in stock, and increasing drug resistance in vectors and etiological agents; and more. (For an interesting case study of some of these problems, see Lawrence et al. 1980.) There is also agreement in core ERD on the wisdom of employ- ing local healers and stockowners themselves as extension agents or assistants, although equally it is recognized that their use is not problem-free and requires careful selection and organization. As for future research needs, there is a clear consensus on the vital necessity of everywhere acquiring more, and more accurate, epidemiological data—data which must be collected, compared and analyzed in both emic and etic terms. This very basic sort of information is obviously imperative if valid correlations are to be drawn between pat- terns of livestock disease and the physical and human ecologies which animals and their keepers inhabit. It is also imperative for meaningful communication between stockowners and ERD personnel. Beyond the need for improved epidemiological information, I would like to add several Other areas which I perceive as requiring more attention. One is the formal, ethno- scientific study of folk classifications for livestock diseases/etiologies/cures, types of pastures and rangelands, species and races of animals, and so forth. To the best of my knowledge, ERD investigations of ethnobiological categories have so far been carried out largely by individuals inexpert in the rigorous procedures of formal linguistic analysis. Yet such analyses, we know, can reveal not only the underlying logic of folk conceptual systems, but sometimes also crucial biological and sociological facts and interrelation- ships overlooked by Western science. This untapped source of systematic information could be of great potential value to ERD because, as one author has convincingly argued, in some cases folk practice or conceptualization of a problem may prove comparable Or superior to that of established science. In others, the two perspectives may diverge but may both embody important insights which can be synthesized. In either case, it is desirable to transcend the conventional science/indigenous, active/passive dichotomy to allow greater indigenous participation in determining development goals and means (after Howes 1980:342). Formal ethnosemantic analysis has an obvious role to play in this discovery process. 142 McCORKLE Vol. 6, No. 1 Another area which has received surprisingly little attention is the many parallels between human and animal ethnomedical systems. The vast literature on human tradi- tional medicine (see, eg., Harrison and Cosminsky 1976) rarely mentions any link between the two. Yet the ethnoveterinary literature contains repeated hints that they are not always highly differentiated. Indeed, as Schwabe and Kuojok (1981) and Schwabe (1978) observe, knowledge derived from folk veterinary experience may be analogy inform human ethnomedical concepts and practice. Moreover, Homo sapiens and their ill-serviced regions. inally, I suggest that it is time at least to begin substantively integrating and theoretically synthesizing ERD findings to date. An overview of the literature reveals many commonalities—and even some startling identities—in folk veterinary beliefs and practices cross-culturally. Unfortunately, there is not space in this review to launch a discussion of these congruencies and their probable causes. Clearly, though, both the similarities and differences in ethnoveterinary systems worldwide need to be catalogued, systematically compared with their correlates in human ethnomedicine and Westem veterinary science, and explained. arriving at larger explanatory models of etl inary phenomena, the relatively more advanced field of medical anthropology holds forth some pertinent analytic frameworks. As noted above, folk medical theory and practice for animals is both emically and etically related to that for humans. Consequently, general research topics and approaches in medical anthropology and ERD frequently overlap. (For a survey of medical anthropology concerns see, eg., Colson and Selby 1974, Foster and Gallatin Anderson 1978, McElroy and Townsend 1979.) Again, there is neither the space, nor perhaps the need, to detail these touchpoints here. Suffice it to reiterate that ‘veterinary anthropology” can profit from much of the analytic groundwork already laid in its sister field of medical anthropology. Likewise for programs of veterinary extension vis-a-vis social science models of cultural change and development, theories of innovation and modernization, the FSR&E literature (farming sy hand extension), and communications theory. As ERD begins to compile and integrate its holistic knowledge of folk veterinary medicine in a production-systems context and to apply itself to hands-on extension, perspectives derived from these cognate areas of research can do much to insure that its insights into the real-world complexity of ethnoveterinary systems are appropriately and effectively utilized. These analytic and synthetic tasks now facing ERD offer an even newer direc- tion for this ‘new direction in ethnobiology.” ACKNOWLEDGEMENTS Preparation of this review was supported by the Title XII Small Ruminant Collaborative Support Research Program under Grant No. AID/DSAN/XII-G-0049 through the SR-CRSP’s Rural Sociology Project; additional support was provided by the University of Missouri, Columbia. The SR-CRSP also funded the author's ethnoveterinary researches in Peru in 1980 and 1985-86. Iam grateful to Drs. Jere Gilles, Tjaart Schillhorn van Veen, and Albert Sollod for their comments 0? early drafts of the paper. In particular, the latter two—both veterinary scientists who have worked extensively in the field—provided balance between technical and social science perspectives 07 ERD and its evolution, plus added references for review. I would also like to acknowledge the many researchers worldwide who so graciously responded with their letters and reports to my call for contributions in the Anthropology Newsletter, the ODI's Pastoral Development Network, the Summer 1986 JOURNAL OF ETHNOBIOLOGY 143 Lettre of the Societe d’Ethnozootechnie, and other periodicals. Finally, thanks are due the Journal of Ethnobiology for providing the stimulus to this review, which has brought many ERD workers into contact for the first time. NOTES 1 although i it avoids the academically | f Latin and Greek , “ethnozootechnics”’ is perahps a too-narrow term. It could be taken to imply the study of folk: veterinary knowledge and technique to the exclusion of larger considerations (ideological, socio-organizations, economic, etc.) which also influence the management of animal health. “Veterinary anthropology,” while linguistically inelegant, obviates this problem. As “the study of ‘man’ from a veterinary viewpoint,” it focuses attention upon the importance of animal health and productivity for human well-being rather than as decontextualized ends in ner of themselves. Moreover, it precisely captures the core of inquir RD (see text). And, it forms a nice analogy to ‘‘medical anthropology,” since ERD in many ways ea for animals this domain of study among human populations. Also, the term is innately appealing to ee MANE like myself — dian in heros: area. Still, a slightly less disciplinary- om label migh definition and antecedent forms—and hence, too, more cect ic. So despite its prone. mix, “a T employ the overarching taoaeetatinesy? (McCorkle 1982a) to reference the field as a whole. 2This is most likely where the agroeconomic base or, better still, the animal production system itself forms one of the foci of research. To give but a few yet representative ethnographic examples, in an extensive study of Saami ethnoecology, tend (1978) details both past and pre- sent systems of reindeer management, their sociostructural correlates and physiographic setting, and touches upon Rangifer nutrition and health. Evans-Pritchards’ (1937, 1969) classic investiga- tions among the Nuer document many health-related aspects of their cattle husbandry, although unfortunately he mentions little of Nuer vete erinary medicine per se. And works like those by Flores- Ochoa (1979) and West (1981) on alpaca herders in Peru, Bernus (1981) and Nicolaisen (1963) on the Twareg, Okaiyeto (1980) and Stenning (1959) on the Fulani, Dow ‘Hudson and seecnten (1985) on the Turkana, and many others offer occasional ob diseases, health-relat Pp , contact poi ag he ang so forth. 1 - ett = | j veterinary i information is also tucked away in va fieldnotes and “heads” of ethnological and h loot Spear nt s. Thi L ind, for a variety of animal aos in Ecuador, Frank Hole, on reproductive, ae ethnodiagnostic, and other aspects of sheep and goat husbandry among the Lur or the Zagros mountains of Iran; Joel Knipers, for folk theories of equine health in eastern Indonesia; and David Lonergan, on the veterinary beliefs and practices of shepherds in central Sardinia. 3snippets of ethnoveterinary lore may appear in works that touch upon witchcraft, ritual, and teligion as these relate to health and healing (e.g., Buxton 1973 and Richards 1927, cited in Schwabe and Kuojok 1981, *Jalby 1974) or in addenda to discussions of human ethnomedical systems (e.g., the appendix to Heke: 1980 4For example, birt in a variety of disciplines mars sought to reconstruct elements of eterinary know g., Bodson 1984, Roquet 1984, Schwabe et al. 1982, rer 1978, 1984). Also, recent pa bine work provides some insights into ees pets len. Wheeler 1984) and stresses (e.g., Pollard and Drew 1975] among early animal est such studies may allude to folk beliefs or, more typically, disease-related husbandry practices, €.g.: in veterinary medicine, Fazil and Hofmann 1981; Higgins 1983; Reed et al. 1974; Schillhorn van Veen 1981, 1984, Schneider 1977; Sollod 1981; and in range science, Glazier 1982. 144 McCORKLE Vol. 6, No. 1 The extensive historiography of Western veterinary medicine and its a pacaainansl documents much of the discipline’ s folk Lp cppbeocal (e.g., Smithcors 1957, and many others). And specialized n a given animal domesticate (e.g., Law 1980) sometimes mention ike eats techniques and theories applied to that species. 7An exhaustive review was the initial ideal, but this was thwarted by a number of factors. For one, rg limited waned of researchers whom I have been able to identify as working specifically in ERD is flung ‘round the world—most notably in the U.S., Europe, and Africa. Moreover, we appear to heel me only partially aware of one anothers’ work, especially when we step outside our primary geographic epinge of research. There as yet exists no formal network, or even an infor- mal community, of E Nor is there any recognized group of journals in which inary information regularly a Indeed, a good deal of the ERD literature exists gcd in fugitive” form: in unpublished mimeo or xerox, in recondite newsletters and journals, in USAID and other project technical reports, in theses and dissertations from third- as well as first- and second-world countries—and of course, all in a variety of languages. ing through the literature that is pear presents yet another basic problem. Because, as noted ens there is no one label for the field, from titles alone it is often impossible to distinguish between works with and without an ethnoveterinary orientation. Of course, titles which contrapose “veterinary” with “anthropology,” “ethno,” “traditional,” “indigenous,” “popular,” or ‘“folk’’ pose no problem. But more amorphous appellations like “Epidemiology of Animal Disease X in nae X” or ‘Herding Among the X People” may or may not have an “ethno” and a veterin po- nent, respectively. Each such work must be carefully examined for its perspective and pcos 8among th li d 1 of such effort those of French researchers investigating LLU the folk v veterinary medicine, both past and present, of France (e 8 Societe d’Ethnozootechnie 1984). part or in whole with this in the space available here. In any case, many o offer fe saan seaaeae ee ae and only a few of the remainder were available for firsthand examination. Nevertheless, based on their titles and annota- ‘ tions, a number of these publications are clearly relevant to sections of the discussion. In such instances, these studies are cite test with an asterisk. Finally, Brisebarre (1985a) has ae Laem * companion, but unannotated, bibliography ig 63 veterinary medi between 1970 and 1984 which deal wie pastoral sesearch in Africa. 9 Ethnosemantics” is employed here in a simple, non-technical sense. I do not mean to a the formal linguistic discipline known variously as ethnosemantics, ethnoscience, or componen tial analysis. See concluding remarks. 10For contemporary Nilotic cattle-culture —— Schwabe (1984:140) remarks that “the practices of animal husbandry, religion and healing are thoroughly ant ” Maliki (1981:54) notes that ‘there is a thin line between” pharmaceutical se poring terinary treatments among WoDa McCorkle (1982a:7) writes ies Quechua villagers make ‘little or no distinction between natural and supernatural ills and c llp ] fideo] L ‘ erwities like culling, a. marketing, restocking, p pasturing, eerste docking, and predator control, see McCorkle "Detailed seen ~ symbolic analyers of livestock “fertility rites” are fairly — - ture. Howev: ny larger issues in animal ear meer husbandry, I do not reference them here. Summer 1986 JOURNAL OF ETHNOBIOLOGY 145 LITERATURE CITED ANDERSON, MYRDENE. 1978. Saami Ethnoecology: Resource Manage- ment in Norwegian Lapland. Unpubl. Ph.D. Dissert. (Anthrop.}, Yale Univ. AREVALO T., FRANCISCO and HER- NANDO BAZALAR R. In progress a. Ensayo de la eficacia contra Fasciola hepatica de la shepita y alcachofa en Ovinos alto andinos naturalmente infectados. Paper to be presented at V Congreso sobre Agricultura Andina, March 1986, Puno, Peru. . In progress b. Ensayo sobre la eficacia de la pepa de zapallo contra nematodos gastrointestinales y pulmonares en ovinos alto andinos naturalmente infectados. IVITA, Huancayo, Peru. BA, ABOU SIDI. 1982. L’Art Veterinaire des Pasteurs Saheliens. Environne- ment Africain, Serie Etudes et Recherches 73-82:1-90. . 1984. L’Art Veterinarie en Milieu Traditionnel Africain. Agence de Cooperation Culturelle et Technique, Paris. BAIRACLI- LEVY, JULLIETTE DE. 1984 (1952). The Complete Herbal Hand- book for Farm and Stable. Faber and Faber, London. BAZALAR R., HERNANCO and FRAN- CISCO AREVALO T. In progress. Ensayo de la eficacia del Utashayli contra Melophagus ovinus en ovinos alto andinos naturalmente infestados. Paper to be presented to LAPA, November 1985, Huancayo, Peru. BERNUSE, EDMOND. 198]. Touaregs Nigeriens: Unite Culturelle et Diver- Site Regionale d’un Peuple Pastuer. Memoire ORSTEM No. 94. Editions de l’Office de Recherches Scienti- fiques et Techniques Outre Mer, aris. BODSON, L. 1984. La medecine veteri- naire dans l’antiquite greco-romaine: Problemes, composantes, orienta- tions. Ethuidzootechnie 34:3-12. BRISEBARRE, ANNE-MARIE. 1978. La medecine veterinaire traditionelle du berger de transhumance en Ceven- nes. Le Courrier de la Nature 5:4-12. . 1984a. A propos de l’usage therapeutique des boquets suspendus dans les bergeries Cevenoles. Bulletin d’Ethnomedecine 32(4):129-163. . 1984b. Le recours a Saint Fleuret, guerisseur des bestiaux, a Estaing (Aveyron). Ethnozootechnie 34:59-85. 1985a. Elevage et medecine veterinaire dans les pays d'Afrique et Madagascar. Production Pastorale et Societe 16:109-112. . 1985b. Le medecine veterinaire populaire en France: apercu bibliographique. Production Pastorale et Societe 16:101-108. . 1985c. Les boquets therapeutiques en medecine veteri- naire et humaine: Essai de synthese. Bulletin d’Ethnomedecine 35(3):3-38. BROKENSHA, DAVID, D. M. WARREN, and OSWALD WERNER. 1980. Indigenous Knowledge Systems and Development. Univ. Press of erica, Lanham, Maryland, New York, London. BUSTINZA CH., VICTOR, et al. 1985. Piel de Alpaca. IDSA, Puno, Peru. sane hae J. 1973. Religion and Healing andari. Oxford Univ. Press, fh rd. CABROL, A. 1984. Lexique pastoral du oo oriental. Folklore 38:30- Gaia D. M. 1976. Plants regarded by Africans as being of medicinal value to animals. Rhode- sian Veterinary Journal 7(1):6-12. DE ST. CROIX, F. W. 1972. The Fulani of Northern Nigeria. Gregg Inter- national Publ., Westmead, England. DUNLOP, DAVID W. 1975. Alternatives to “modern” health delivery systems in Africa: Public policy issues of traditional health systems. Soc. Sci. and Med. 9:581-586. 146 McCORKLE Vol. 6, No. 1 LITERATURE CITED (continued| DYSON-HUDSON, RADA and J. TER- RENCE MCCABE. 1985. South Turkana Nomadism: Coping with an Unpredictably Varying Environ- ment. HRAFlex Books FL17-001, Ethnography Series. Human Rela- tions Area Files, Inc., New Haven, Connecticut. EVANS-PRICHARD, E. E. 1937. Eco- nomic life of Nuer: Cattle. Sudan Notes and Records 20:209-245. 1969 (1940). The Nuer: A Description of the Modes of Liveli- hood and Political Institutions of a Nilotic People. Oxford Univ. Press, New York and London. FAZIL, M. A. and R. R. HOFMANN. 1981. Haltung und Krankheiten des Kamels. Tierarztl. Prax. 9:389-402. FERNANDEZ, MARIA E. 1985. Partici- patory Action Research and the Farming Systems Approach with Highland Peasants. Unpubl. M. A. thesis, (Rur. Soc. Dev.) Reading Univ. FLORES-OCHOA, JORGE A. 1978. Classification et denomination des camelides sud-americans. Annales Economies Societes Civilisations 5-6:1006-1016. 1979 (1968). Pas- toralists of the Andes. ISHI, Phila- delphia. FOSTER, GEORGE and BARBARA GALLATIN ANDERSON. 1978. Medical Anthropology. John Wiley & Sons, New York. GLAZIER, DANA. 1982. Herding Dy- namics Study: Profile of a WoDaaBe Herding Unit of the North Dakoro Region, Niger. Niger Ministry of Rural Development/USAID, Tahoua. GOURLET, S. 1979. Les Plantes en Medecine Veterinaire Populaire. These de Doctorat (Vet. Sci.), Toulouse. IN, BARBARA. 1985. Towards a Maasai ethno-veterinary. Unpubl. Doc., ILCA, Nairobi. HALPIN, BRENDAN. 1981. Vets—bare- foot and otherwise. Overseas Devel- opment Institute Agricultural Administrative Unit Pastoral Net- work Paper 11c, London. HARRISON, IRA E. and SHEILA COS- MINSKY. 1976. Traditional Medi- cine: Implications for Ethnomedi- cine, Ethnopharmacology, Maternal Health, and Public Health—An Annotated Bibliography of Africa, Latin America, and the Carribean. Garland Publ., Inc., New York and London. HIGGINS, A. J. 1983. Observations on the diseases of the Arabian came (Camelus dromedarius) and their control: A review. Veterinary bul- letin 53(12):1089-1100. HOCKINGS, PAUL. 1980. Sex and Disease in a Mountain Community. Vikas Publ. House, New Delhi. HOWES, MICHAEL. 1980. The uses of indigenous technical knowledge in development. Pp. 335-351 in In- digenous Knowledge Systems and Development. (David Brokensha, D. M. Warren, and Oswald Werner, eds.). Univ, Press of America, Lanham, MD, New York, London. IBRAHIM, M. A. 1984. Veterinary tradi- tional practices in Nigeria. Paper presented at the ILCA/NAPRI Sym: posium on Livestock Production in the Subhumid Zone of Nigera, Kaduna. _ N. NWUDE, Y. O. ALIU, and R. A. OGUNSUSI. 1983. Traditional concepts of animal disease and treatment among Fulan! herdsmen in Kaduna state of Niger. Overseas Development Institute Agricultural Administration Unit Pastoral Development Network n. Paper 16c, cua — DE, K. OGUNSUSI, and Y. O. ALIU. 1984. Screening of West African plants for anthelmintic activity. ILCA Bulletin 17:19-23. IMPERATO, PASCAL J. 1974. Nomads of the West African Sahel and the Summer 1986 JOURNAL OF ETHNOBIOLOGY 147 LITERATURE CITED (continued) delivery of health services to them. Soc. Sci. and Med. 8:443-457. JALBY, R. 1974. Sorcellerie et Medecine Populaire en Languedoc. Ed. de l’Aygues, Nyons. KIMBALL, LINDA AMY. Forthcoming. Brunei Malay ethno-veterinary prac- tice. Borneo Research Bulletin. LAW, ROBIN. 1980. The Horse in West African History: The Role of the Horse in the Societies of Pre-colonial West Africa. Oxford Univ. Press, Oxford. LAWRENCE, J. A., C. M. FOGGIN, and R. A. I. NORVAL. 1980. The effects of war on the control of diseases of livestock in Rhodesia (Zimbabwe). Veterinary Record 107: 82-85 LOUTAN, L. 1984. Veterinary auxiliaries Pp. 763-781 in Pastoral Development in Central Niger: Report of the Niger Range and Livestock Project (Jeremy Swift, ed.), Niger Ministry of Rural Development/USAID, Niamey. MAILIKI, ANGELO B. 1981. Ngaynaaka: Herding According to the WoDaaBe. Niger Range and Livestock Project, Rapport Preliminaire/Discussion Paper No. 2. USAID/Niger Ministry of Rural Development, Tahoua. McCORKLE, CONSTANCE M. 1982a. Management of Animal Health and . 1982b. Organizational Dialectics of Animal Management. SR-CRSP Publ. No. 5, Dept. Rur. Soc., Univ. Missouri, Columbia. . 1983a. The Techno- environmental Dialectics of Herding in Andean Agropastoralism. SR- CRSP Publ. No. 7, Dept. Rur. Soc., Univ. Missouri, Columbia. . 1983b. Meat and Potatoes: Animal Management and the Agropastoral Dialectic in an Indigenous Andean Community, with Implications for Development. Unpubl. Ph.D. Dissert. (Anthrop.} Stanford Univ. McELROY, ANN and PATRICIA K. TOWNSEND. 1979. Medical Anthro- pology: An Ecological Perspective. Duxbury Press, N. Sciuate, MA. MENESES T., NORMA. In progress. Vocabulario Pecuario Quechua de la Comunidad de Quishuara. B.A. thesis (Ling.), Univ. San Marcos, ima. MOURIER-BALLON, M. 1983. Essai d’Ethnobotanique en Bas-Dauphine. These de Doctorat de 3eme Cycle, Univ. Lyon II. METAILIE, G. 1984. Apercu des prin- cipes de la medecine veterinaire tradi- tionnelle en Chine. Ethnozootechnie 743-50. NICOLAISEN, JOHANNES. 1963. Ecol- gy and Culture of the Pastoral Tuareg, with Special Reference to the Tuareg of Ahaggar and Air. National Museum of Copenhagen Ethnographical Series No. 9, Copen- agen. NOIRTIN, C. 1975. Vocabulaire patois en pathologie bovine. These de Doctorat (Vet.) Ecole Nationale Veterinaire, Alfort. NWUDE, N. and M. A. IBRAHIM. 1980. Plants used in traditional veterinary medical practice in Nigeria. J. Vet. Pharmacol. Therap. 3:261-273. OHTA, I. 1984. Symptoms are classified into diagnostic categories: Turkana’s view of livestock diseases. African Study Monographs, Supplementary Issue 3:71-93. OKATYETO, P. O. 1980. A Descriptive Study of the Major Economic Activ- ities of the Settled Pastoral Fulani in Three Zaria Villages of Kaduna State. Unpubl. M.S. thesis, Ahmadu Bello Univ. POLLARD, GORDON C. and ISABELLA M. DREW. 1975. Llama herding and settlement in prehistoric northern Chile: Application of an analysis for 148 McCORKLE Vol. 6, No. 1 LITERATURE CITED (continued} determining domestication. Amer. Antiquity 40(3):296-305. REED, J.B.H., D.L. DOXEY, A.B. FORBES, R.S. FINLAY, I. W. GEERING, S.D. SMITH, and J. D. WRIGHT. 1974. Productive performance of cattle in Botswana. Trop. Anim. Hlth. and Prod. 6:1-21. RICHARDS, M.C. 1927. Medical treat- ment by Bor witchdoctors. Sudan Notes and Records 10:241-242. ROQUET, GERARD. 1984. De l’icono- graphie du velage a Videntification foetus. Societe d’Ethnozoologie et dEthnobotanique Bulletin de Liaison 15:3-20. SANDFORD, DICK. 1981. Pastoralists as animal health workers: The range development project in Ethiopia. Overseas Development Institute Agricultural Administration Unit Pastoral Network Paper 12c, London. SANDFORD, STEPHEN. 1983. Manage- ment of Pastoral Development in the Third World. John Wiley and Sons, Chichester, New York, Brisbane, Toronto, Singapore. SCHILLHORN VAN VEEN, T. W. 1981. Livestock systems and animal health Farming Systems Research Group — Paper No. 3, Michigan State 84. Observations on animal em especially on a proaches to identify and overcome constraints in the subhumid zone of West Africa. Pp. 303-316 in Livestock Development in Subsaharan Africa: Constraints, Projects, Policy (James R. Simpson and Phylo Evangelou, eds.), Westview Press, Boulder. SCHNEIDER, H. 1977. Analyse der Tiergesundheitssituation in Sudwest Africa (Namibia). Unpubl. thesis (Vet. Med.), Univ. Giessen, West Germany. SCHWABE, CALVIN W. 1969. Veterinary Medicine and Human Health. The Williams & Wilkins Co., Baltimore. 1978. Cattle, Priests, and Progress i in Medicine. Univ. Min- nesota, Minneapolis. . 1980. Animal disease control, part TL Newer methods, with possibility for their application in the Sudan. Sudan J. Veter. Sci. and Animal Husbandry 21(2):55-56. . 1984. A unique surgl- cal operation on the horns of African bulls in ancient and modern times. Agricul. Hist. me 138-156. AAC MAKUET © KUOJOK. 581. eee and beliefs of the traditional Dinka healer in rela- tion to provision of modern medical and veterinary services for the southern Sudan. Human Organiza- tion 40(3):231-238. ADAMS and C. T. HODGE. 1982. Egyptian beliefs about magical and physiological pro- perties of the bull’s spine suggest a rational anatomical origin and mean- ing for the ankh. Anthropol. Linguis- tics, Winter:445-479. SHANKLIN, EUGENIA. 1985a. Done gal’s Changing Traditions: An Ethno- graphic Study. Gordon & Breach Science Publishers, New York. 1985b. Sustenance and symbol: Anthropological studies of domesticated animals. Annu. Rev: Anthr. 14:375-403. SMITHCORS, J. F. 1957. Evolution of the Veterinary Art: Narrative Account to 1850. Veterinary M Medi- cine Publishing Co., Kansas City, ssouri. SOCIETE D’/ETHNOZOOTECHNE. 1984, La Medecine Veterinaire Popu: laire. ape Issue, Ethnozootechnie, . Pari SOLLOD, ALBERT E. 1981. Patterns f Zone. Summer 1986 JOURNAL OF ETHNOBIOLOGY 149 LITERATURE CITED (continued) Niger Ministry of Rural Develop- ment/USAID, Niamey. 1983. The influence of trypanosomiasis on the animal disease taxonomies of the Fulbe. Paper presented to the 26th Annual Meeting of the African Studies Assoc., Boston. and JAMES KNIGHT. 1983. Veterinary anthropology: A herd health study in central Niger. Pp. 482-486 in Proceedings of the Third International Symposium on Veterinary Epidemiology and Economics, Veterinary Medicine Publ. Co., Edwardsville, Kansas. , KATHERINE WOLF- GANG, and JAMES KNIGHT. 1984. Veterinary anthropology: Inter- disciplinary methods in pastoral systems research. Pp. 285-302 in Livestock Development in Sub- saharan Africa: Constraints, Pros- pects, Policy (James R. Simpson and Phylo ihe al eds.), Westview STENNING ¥ G. 1959. Savannah Nomads: A Study of the Wodaabe Pastoral Fulani of Western Bornu Pro- vince, Northern Region, Nigeria. Oxford Univ. Press, London. WEST, TERRY L. 1981. Alpaca Produc- tion in Puno, Peru. SR-CRSP Publ. No. 3, Dept. Rur. Soc., Univ. Missouri, Columbia. WHEELER, JANE C. 1984. On the origin and early development of camelid pastoralism in the Andes. Pp. 395-410 in Animals and Archaeology: Early Herders and Their Flocks (Juliet Clutton-Brock and Caroline Grigson, eds.), BAR International Series 202. WOLFGANG, KATHERINE. 1983. An Ethnoveterinary Study of Cattle Health Care by the Fulbe Herders of South Central Upper Volta. Unpubl. thesis, Hampshire College/Tufts Univ., Amherst, Massachusetts. and ALBERT SOLLOD. 1986. Traditional Veterinary Medical Practice by Twareg Herders in Cen- tral Niger. Integrated Livestock Pro- duction Project, Tahoua, Niger and Tufts Univ. School of Veterinary Medicine, North Grafton, Massa- chusetts 7 «: eee 3 : ; = a ~ ¢ = " 7 : : : es ay Sal = 5 OC yee ; Xe; te wt - ; E 7 7 Car =~ =: y i . p : ; : ew ¥ * 5 7 7 i : : j . ge oo ie at ard oe pat 7 ‘- =n oer se. oi ae i na Pei Bie! 7 FO ed q Sigs re - ee ple ici 7 eer & oe ee - 7 ; ‘ar, J. Ethnobiol. 6(1):151-167 Summer 1986 GENETIC DIVERSITY AND CONSERVATION IN TRADITIONAL FARMING SYSTEMS! STEPHEN B. BRUSH Department of Applied Behavioral Sciences University of California Davis, CA 95616 ABSTRACT.—A few farming systems have carried forth the legacy of Neolithic in the thousands of ancestral crop varieties known as “landraces.” Our debt to this heritage is immeasurable, yet the very success of modern agriculture is contributing to the disap- pearance of landraces. A handful of scientifically bred varieties has replaced thousands of ancestral types in twenty years as part of the worldwide effort to develop agriculture. Although landraces and the farming systems that maintain them are part of long evolu- tionary processes, rapid change threatens their survival. Given the value of landraces for genetic resources, their replacement has been termed “genetic erosion.” This “1 1 e. ee 1 : * : Ben of genetic resources of crops in two areas of genetic diversity, the Andes and Southeast Asia. Similarities and differences in the pattem of genetic change relating to the adop- tion of modern crop varieties are observed. The pattern shows a dichotomy between commercial and subsistence farming systems and between micro-regions. By describing conservation and loss of genetic resources we can obtain a more accurate understanding of agricultural change in underdeveloped and marginal areas. GENETIC EROSION Genetic erosion refers to the loss of genetic resources because of human activity. It occurs in “natural” ecosystems, for instance with the clearing of tropical forests, as well as anthropogenic systems, for instance when ancient landraces of crops are replaced by modern varieties. Concern for genetic erosion arose with the recognition that biological diversity is being lost at an unprecedented rate because of the rapid demographic, economic and technological changes of the late Twentieth Century (Myers 1981). This paper treats the loss of germplasm from traditional farming systems. In thinking about genetic erosion, we confront the familiar issue of environmental Costs resulting from economic growth. The value of genetic resources, particularly crop germplasm, is undisputed. Crop germplasm is a renewable resource, but agricultural Cc € in areas of genetic diversity may involve replacing one type of germplasm (“‘landraces’’} with another (modern varieties”). This replacement destroys the older type, unless it is collected for storage in germplasm collections of seed banks. Two issues Surround this replacement. One is the efficacy of conservation outside of living agroecosystems (Frankel and Soule 1981). The other is the issue of environmental degradation in ancient agricultural systems that existed for millennia without external subsidies. Although increases in productivity can be gained through germplasm substi- tution, they are usually accompanied by increased management costs and reliance on Purchased inputs, especially fertilizer and seed. Genetic diversity in agriculture is also theoretically associated with stability (VanderPlank 1968), and a decrease in diversity thus represents environmental degradation to a less stable system. However, the diver- Sity/stability argument has not been adequately investigated in relation to agriculture, contrary views are prevalent (e.g. Brown 1983, Altieri, et al. 1983). 152 BRUSH Vol. 6, No. 1 In the early 1970s, two phenomena coincided to change fundamentally the way we think about the genetic resources for agriculture. First was the recognition that the “Green Revolution” was taking hold in many parts of the world, especially Asia. Predictions of famine that were common in the 1960s were replaced by concerns over the impact of the abundant harvests from the high yielding varieties (HYVs) of rice and wheat (e.g. Wharton 1969). The Green Revolution was succeeding in areas of genetic diversity which are the primary providers of the genetic resources for agriculture everywhere. The suc- cess of HYVs showed us that “traditional” cultures were rapidly changing. Second was the spread of the southern Corn Leaf Blight through maize fields of the southeastern United States. This lead us to recognize how vulnerable modern agriculture is to exotic pathogens, owing to its reliance on a narrow genetic base (National Academy of Science 1972). The context for this coincidence included the ideas of a population bomb about to explode, the finite nature of resources, the OPEC oil shock of 1973, and the environ- mental movement’s notion of the “Spaceship Earth” and its celebration of traditional cultures. This coincidence linked two popular notions: the fragility of modern technology and lack of control over a base of critical resources. The benefits for poor people in Jerdeveloped countries reaped from the Green Revolution represented a “hidden peril” for agriculture: the loss of genetic resources that serve to protect crops from pests and pathogens (Chedd 1970). The term “genetic erosion” represents the wedding of these notions, implying a limited resource base being destroyed by peasant farmers as they modernize. The response to this perceived peril was to increase the collection and conservation of genetic resources. The first half of the 1970s produced landmark studies of the status of genetic resources in agriculture (e.g. Frankel and Bennett 1970; Frankel and Hawkes 1975). National germplasm systems were strengthened, and an international system for monitoring, collecting, conserving, and distributing genetic resources was organized in various ways, including establishing the International Board for Plant Genetic Resources (IBPGR) (Plucknett and Smith 1982). A systematic program was planned for the world’s major food crops, to set priorities for collection and to evaluate the status of germplasm collections. The growth/conservation debate continues, and it is apparent that information on the nature, status and trend of genetic erosion is much needed (U.S. Strategy Conference 1981). Research is required for policy and planning, whether to accelerate collection oF to develop in situ conservation. The use of ethnobiological research tools has great poter tial for studying the dynamics of genetic resources from ancient farming systems. These resources are cultural as well as biological objects, and selection is a function of social, economic, and cultural factors as well as a function of the germplasm’s biological characteristics. THE ORIGINS OF CROP GENETIC DIVERSITY Centers of crop origin were first recognized by their high levels of genetic diversity, both at the species and variety level (Vavilov 1951). Crop germplasm from these centers of diversity forms the foundation of the crop breeding industry that is essential to modem agriculture, and thousands of cultivars have been observed and collected from these centers. Research on the geographic origins of agriculture (Ucko and Dimbleby 1969) indicates a climatic prerequisite of seasonal variability and a well-marked dry season (Hawkes 1983), and it showed that mountainous or hill land areas were especially significant. 1] factore can he cited i £ g to this diversity: (1) the physical a, sity of centers of origin, (2) the long history of cultivation, (3) the large number of coevolve —— —-_ lel —_— Ee ee Summer 1986 JOURNAL OF ETHNOBIOLOGY 153 pests, pathogens, and competitors, and (4) cultural practices of selection, maintenance and distribution. These factors have created numerous niches for distinct cultivars, isolating mechanisms, and selection pressure for diversity. Ethnobotanical research shows that all farming cultures classify and select plants according to many criteria: agronomic, culinary, medicinal, ritual. The rich lexical domains associated with traditional agriculture in areas of domestication are indicative of the positive role that farmers have played in selecting and maintaining genetic resources (Brush et al. 1981). Diversity appears to be one of several strategies used to create stable subsistence by farmers practicing low-energy, non-industrial farming. Other strategies include terracing and other slope modification, mulching and other soil amendments, and frequent field rotation with long fallow periods. Diversity is manifested in low-energy farming systems in numerous ways: in the mix of cultivation, gathering, and grazing as productive activities, in nutrition strategies (DeWalt 1983) in polyculture whereby different crops are produced in association with one another (Kass 1978), in the posses- sion and cultivation of numerous small plots (Carter and Mamani 1983), in the agronomic technologies of field preparation and irrigation (Wilken 1972), and in large numbers of cultivars of single species (Brush et al. 1981). Genetic diversity in crops continues in many environments, but its continuity is particularly marked in mountain areas. Genetic diversity continues here for some of the same reasons that made hill lands centers of crop domestication: well-marked environ- mental change over small distances, marked seasonality, and isolation between small production zones. Equally significant, however, is the relative marginality of mountain regions in world agriculture. This marginality is the result of numerous factors: lower population densities, greater environmental risk, less accessibility to markets, limited arable land, and lower investment in agricultural development. These environmental and economic factors have retarded the pace of change, including the substitution of improved crop varieties common to lowland areas. Mountain areas are, however, under considerable pressure to change. Population growth, economic development, greater communication, and political reform stimulate change, leading to some alarm over potential environmental degradation in the highlands (Messerli 1983). Well-tuned traditional agricultural practices, such as crop diversity, terracing, small irrigation systems, and long fallow periods, are being replaced by more intensive, and perhaps less stable practices. - While genetic diversity is inherited from ancient agricultural traditions and may be explained as a response to a complex and competitive environment, it is also associated with low productivity. Jennings and Cock (1978) show that crop productivity in centers of origin is appreciably lower than outside these centers. This is true even after economic and cultural differences are taken into consideration. Agricultural development for greater productivity requires increased management and control over the various components of the farming system. Genetic control and improvement has been accomplished by national and international breeding and seed multiplication programs, exemplified by the release of HYVs of rice and wheat. Breeding and multiplication programs have been developed in most underdeveloped regions since World War II, and by the early 1960s the results of these programs were evident. Farmers were beginning to adopt HYVs in large numbers, leading to inquiries about the rate pattern of adoption (Perrin and Winkelman 1976). Concern about whether new technology could or would be adopted on small as well as large farms and concern about negative impacts of adoption lead to an extensive research effort in the social sciences (Feder et al. 1982). This effort revealed that adoption cannot be predicted by the scale of the farming enterprise and that negative influences such as decreased labor demand in agriculture do not result. It also revealed that a number of factors in the general agricultural economy influence adoption. These include risk aversion (Roumasset 1976), human capital (Evenson 1973), labor availability 154 BRUSH Vol. 6, No. 1 (Norman 1969), credit (Lipton 1976), tenure (Parthasarathy and Prasad 1978}, and the availability of complementary components such as fertilizer and water (Clay 1975). This social science research on the adoption of HYVs also revealed that the pattern and rate of adoption was very uneven across geographic regions. Following the classic analysis of hybrid corn diffusion in the U.S. by Griliches (1957), a sigmoid pattern of adoption of HYVs in the Third World is evident (e.g. Herdt and Capule 1983; Dalrymple 1978). The rate and ceiling of adoption is variable. Adoption in some places proceeds rapidly toward 100% of farms, while in others it proceeds slowly to a much smaller percentage. The implications of different rates and ceilings of adoption are very significant for our understanding of genetic erosion. They suggest that simple extra- polation from early adoption experience may not be valid to describe the eventual pattern of genetic replacement of traditional varieties of HYVs. In order to elucidate this conclusion, it will be helpful to review two cases of the diffusion of improved varieties. these are the cases of potatoes in the Andes and of rice in Southeast Asia. SELECTION FOR DIVERSITY: TWO CASE STUDIES Andean Potato Agriculture.—The potato (Solanum spp.) is the staple in the Andes, where it was domesticated. Seven species are cultivated from four ploidy levels (Hawkes 1978). Diversity at the species level distinguishes bitter species (S. x juzepczukii, S. x curti- lobum and S. ajanhuiri) from non-bitter species (S. tuberosum, S. x chaucha, S. stenotomum and S. goniocalyx). The former are frost resistant and must be processed by freeze drying (into chujio) before consumption. Besides species level diversity, the Andes is also a region of tremendous varietal diversity in the potato. The International Potato Center has a collection of some 12,000 named accessions, representing rougnly 5,000 distinct clones (A. Huam4n, personal communication). Andean farmers employ a systematic folk taxonomy to identify and select varieties (Brush et al. 1981). It 1s common for farmers to be able to identify 40 different varieties, and single fields may be planted in as many varieties. The size of household inventories of potato varieties and amount of field diversity differ greatly between households and regions. dean potato agriculture is diverse in many ways besides the number of species and varieties under cultivation. The management of potato farmland is frequently under simultaneous control by households and communities. Households are responsible for selecting varieties, for most of the labor employed in potato cultivation, and for post harvest processing, storage and marketing. The community controls sectoral fallowing which determines where the household will grow its crop; it regulates the agricultural calendar; and, in some cases, community-based irrigation is used for potatoes. As in all potato agriculture, the maintenance of viable seed is crucial. This depends on the regular rotation of seed tubers between ecological (altitudinal) zones. This sytem may be managed at the household level by the ownership and use of different fields at various altitudes or by communities and market systems that move large quantities of seed across wide regions. Traditional potato agriculture is also associated with a highly fragmented land- holding system. Carter and Mamani (1983), for instance, describe a farming community in Bolivia in which the average household owns 21 different plots in the potato zones. ite 3 techniques (Tapia 1983) and post harvest technology (Werge 1979) are similarly vari ¢ ain ade eribustn Orr aeeierslerirel ch } here are felt in the Andes- Population has grown at over 2.5%/year for three decades, popular pressure for develop- ment and incre. iving standards are evident, reform has been undertaken to stimulate . s 5 1 , eee FS 1 re ee a hac en created to promote change. Urban demand has grown in the highlands stimulating the growth of market systems (Appleby 1976). Summer 1986 JOURNAL OF ETHNOBIOLOGY 155 Highland potato agriculture in Peru has been the object of development efforts for over three decades. Improved varieties from native germplasm were first released in the early 1950s, and since then over 30 varieties have been released. These are now ubiquitous throughout Peru. Besides the release of improved varieties, the Peruvian National Potato Program has promoted the diffusion of selected native clones and encouraged the adoption of modern techniques such as the application of chemical fertilizers and pesticides. The availability of credit for these agrochemical inputs has been tied to the adoption of improved varieties. Three decades of these development efforts have had a clear impact in the Andes; the use of agrochemicals is common, improved varieties have been adopted to some extent almost everywhere, and the infrastructure to promote these changes (markets, roads, extension services, seed production systems) has been expanded. The extent of these changes has led to predictions of serious genetic erosion in the Andes (Ochoa 1975), and some of the limited data available from the region indicates that these predictions are valid. Research in the Mantaro Valley, for instance, indicates that roughly 65% of the potato land is planted in improved varieties (Horton 1984). This valley, however, is not typical of most of the highlands, being one of the most commer- cial and intensively cultivated anywhere in highland Peru. The poor performance of Andean agriculture (Caballero 1981; Gonzales 1984) suggests that the overall adoption of new technology, including improved varieties, is low. My impression supports this conclusion, but systematic data on the use of improved and traditional varieties are insufficient to calculate regional or national rates of the use of different types of potato. Horton (1984) reports high adoption rates for well developed and important commercial production areas, but these rates should not be extrapolated over the general Andean region. Bidegaray and Schmidt (1985) report a low 25% adoption rates in the Cuzco area. Changes, including genetic erosion, are evident everywhere in highland potato growing regions of Peru, but their distribution is very uneven. In some areas diverse native collections have been entirely replaced by improved varieties, while in others native types continue to dominate. The distribution of this change follows socioeconomic and environmental contours. The greatest loss has occurred in lower potato growing zones and those within major valley systems with urban centers and markets. The least genetic erosion occurs in higher zones more distant from urban centers and markets. Research conducted by the International Potato Center (CIP) and by Brush, Mayer and Fonseca, indicates that the adoption of improved potato varieties is only one of several possible Options open to small farmers, and it may not be the first option taken to increase Production. CIP studied three agroclimatic zones, based on altitude, in the Mantaro Valley of central Peru. The lowest zone (3000-3450 m) is adjacent to the major urban and market centers of the valley. Although small farms are most typical, this zone has more medium and large scale farms than the other zones, and these types produce potatoes for improved and native varieties and frequently reserve one or more fields for mixed native varieties alone (Brush et al. 1981). 156 BRUSH Vol. 6, No. 1 The distribution of innovations in potato agriculture that is found in the Mantaro Valley appears to be typical of most of the Peruvian highlands. Current research being conducted by Brush, Mayer, and Fonseca along the eastern slopes of the Andes cor- roborates this. In the Tulumayo Valley, east of CIP’s Mantaro study area, improved varieties and selected native varieties are grown for commercial purposes in both high and mid-altitude zones. Mixed fields of native varieties are retained under traditional tillage practices, although they are grown in only the highest fields. The total area devoted to native varieties is smaller, and these are grown under greater environmental stress than previously. In the Paucartambo Valley, east of Cuzco, improved potato varieties have made less impact, but much land once grown in native potato varieties has been converted to barley as a cash crop in the regional beer industry. As adoption occurs, farmers tend to subdivide their farming system into commercial and subsistence sectors. The former are located on the best land, often with irrigation, and receive higher capital inputs. Although the subsistence sector is also important, It is relegated to poorer land and receives fewer inputs. It is possible that families experien- cing land and/or labor shortage might eventually abandon the subsistence sector altogether. This relates to the frequency of off-farm employment and to greater reliance on purchased food. Andean farmers have many reasons for preferring native over improved varieties, but culinary and agronomic ones stand out. Improved varieties are considered “watery , a negative attribute compared to the preferred “floury” quality of native varieties. The almost universal consensus among farmers interviewed about variety choice was that the new varieties were inferior eating potatoes, suited for soups, frying, oF undis- i ting urban consumers. For subsistence farmers, whose diets remain predominately reliant on potatoes, keeping traditional varieties is a culinary necessity. No nutritional differences, other than water content, are noted between traditional and improved varieties. Improved varieties also do not remain viable as seed potatoes for more than a few seasons. Farmers observe that they degenerate and ultimately fail as seed. This can be by purchasing seed, but this is a costly requirement for subsistence farmers. Native seed potatoes are kept viable indefinitely by rotating seed between altitudinal zones, but the seed rotation system used for native seed are generally not applied to improved varieties. Althonah farmerc h £ . : & ing traditional varieties, they have other reasons for adopting modern ones: their greater productivity and their resistance to the two greatest risks, frost and late blight (Phytophthora infestans). Traditional varieties are generally recognized to be more susceptable to these risks than improved varieties. This challenges the notion that older types are more “adapted” than new ones. Genetic erosion is undoubtedly occurring because farmers are changing their system. What is not known is the rate of genetic erosion relative to this change. Potato agriculture in commercially developed valleys such as the Mantaro and Cajamarca valleys has rates of replacement above 60%, but away from these areas the rate appears much lower, perhaps below 25%. There seem to be no areas in the higher altitudes, away from market centers or where subsistence production continues, where farmers completely abandon native varieties for improved ones or for other crops. Genetic change is as inevitable and essential to subsistence agriculture as it 1s 4 to commercial agriculture. Subsistence farmers consciously select among clones, eliminating unwanted ones and emphasizing others. This selection is evidenced by the distribution of varieties within fields, by the politan distribution of certain varieties over very large regions, by the use of certain types for gifts and exchange, by their cultivation for seed, and by the commercial status of a few native varieties. The ontem patte: Ss ine _ ae. improved varieties ed = el 2 —_—— Rg ! | Summer 1986 JOURNAL OF ETHNOBIOLOGY 157 and development programs, but with these new elements, the rate of change has accelerated, justifying the concept of erosion. Besides the introduction of new varieties and other innovations for potato agriculture over the last thirty years, another major change is the rise of commercial potato production on the coast. Coastal production involves larger farms than are typical in the highlands, relies on purchased inputs, is commercially oriented, and employs improved seed produced under certification. The average yield on the coast is 17.85 metric tons per hectare while in the highlands it is only 5.86 metric tons (Peru 1975:2). Coastal potatoes are grown largely for the Lima market, only a short distance from the major center of coastal production in the Cafete Valley. e successful development of potato agriculture on the coast is a potentially limiting factor in the rate of agricultural change in the highlands. Farmers in the central highlands compete in some of the same markets as coastal farmers, and the economies of scale, more optimal agroclimatic conditions, and proximity to markets of coastal producers give them great advantages over the highlands. This increases the risk to highland farmers of investing in new agricultural technology. Besides competition with coastal producers, small farmers in the highlands face two Opportunity costs in adopting improved potato varieties. First is the cost of diverting resources to change potato agriculture. Capital to purchase new seed can be invested in other crops, in livestock, or in other household production such as artisan activities. The Andean household economy is extremely diversified and variable, and given the uncompetitiveness of potato agriculture at higher altitudes, it is likely that available capital would not be diverted there. Second is the cost in time needed to make the adoption decision, learn about new varieties, locate and purchase seed. A significant portion of household income in all sectors of the Andes comes from off-farm employ- ment, and these activities make more difficult an already complex agricultural calendar for farmers involved in a vertical economy. The time involved in the adoption decision may be an additional cost too great to bear. The result of these two opportunity costs is the decision to keep growing traditional varieties. Although the data is insufficent to predict future trends of potato germplasm conservation and loss, we can identify factors that both promote and constrain genetic change. It is promoted by continued high population growth, increasing demand from urban markets, improvement in transportation and communication, changing con- sumption patterns, and active and well organized national and international crop improvement programs. It is constrained by population loss from some areas, consumer preferences, and the marginality of certain areas because of high risk, land fragmentation, and isolation from markets. The balance of these factors has resulted in the uneven pattern of change described above. It is reasonable to assume that the balance will not be altered rapidly, for instance by technological breakthroughs that make remote highland regions more active in commercial potato markets or by a return to native varieties. It is possible that develop- ment programs could overcome these constraints to adoption and accelerate genetic erosion of potatoes in the highlands. The likelihood of this is diminished, however, by several features of the agricultural development policy of Peru. These include a serious inancial crisis in the government, an emphasis on development in the tropical lowlands of eastern Peru rather than in the highlands, and a steady migration away from the highlands, especially from the small farm sector. These features are compounded by political pressures for cheap urban food from the coast as well as from abroad (Ferroni 1980), by ethnic differences between regions, and by failure of reforms to improve productivity in the highlands (Caballero 1981). These features indicate that the distribution of agricultural change and the adoption of improved potato varieties will not change in the near future. 158 BRUSH Vol. 6, No. 1 Southeast Asian Rice Agriculture.—Rice (Oryza sativa) cultivation in Southeast Asia is a useful comparison to potato cultivation in the Andes. Unlike the Andes, where virtually all of the research on genetic diversity has been done in one country (Peru), Southeast Asia has similar research in several countries. Case study material from Thailand, Indonesia, and the Philippines demonstrates their similarity to Peru in terms of agricultural, economic and cultural diversity. Although the true center of domestication for rice is in northern India and southern China, the area of rice diversity extends into Southeast Asia (Chang 1976). Rice has been the object of intensive breeding and develop- ment programs for several decades, and modern varieties have diffused widely throughout the region. These include HYVs from the International Rice Research Institute (IRRI) and earlier releases of national breeding programs. Rice agriculture and the diffusion of new varieties have been examined in detail (Grist 1965, Hanks 1975; Barker 1978; Herdt and Capule 1983). Southeast Asia is characterized by different agroclimatic zones associated with hill and mountain environments. As in the Andes, the diffusion of improved varieties of rice corresponds to this zonation. Reliable aggregate data on improved variety diffusion is more readily available for Asia than for the Andes. Economic and agronomic research on rice provides a good overall picture of the presence or absence of modern rice varieties (Herdt and Capule 1983), but little research has been done on the retention of diverse stocks that might accompany some adoption. Most research on adoption has concentrated in areas with relatively high acceptance of new varieties. These areas tend to be optimal for rice production and not “traditional” by virtue of their cultural diversity, ecological diversity or agroecological inality (Huke and Duncan 1969). An exception to this pattern of research is work done in the Chang Mai Valley of northern Thailand (Rerkasem and Rerkasem 1984). While lacking the varietal distinctiveness of a sexually propagated potato clone, rice is self-pollinating and thus typified by distinctive varietal populations (Grist 1965). Asian rice is subdivided into three major subspecies (indica, japonica or sinica, and javanica) which are geographically distinct and differ according to their grain. Each of these species is subdivided into numerous varieties whose grand total may be as high as 120,000 (Swaminathan 1984). In Indonesia, an estimated 8,000 varieties, primarily javanica, are found (Bernsten et al. 1982), while in the Philippines and Thailand, 1,500 and 3, varieties respectively are found with indica varieties predominating (IRRI 1978). As in Andean agriculture, the genetic diversity of rice is recognized culturally in Asia. This recognition occurs at both the general level in the distinction between the major subspecies over large regions and at the specific local level. This specific cultural recognition is evident in naming and folk taxonomic systems, in planting patterns, 10 culinary aspects and in rituals. Among the Ifugao, H. Conklin (personal communt cation) has elicited a folk classification of five levels for rice. In 1973, this system was applied to the classification of 78 varieties grown by the Ifugao. Among the Iban me Sarawak, Sutlive (1978), found that each family maintained a variety of strains which were named according to such criteria as place or origin, bouquet, and culinary property. Sutlive describes the Iban practices of planting varieties in separate portions of the field and locating a special ritual segment in the middle of a field where a special variety 1s planted. At harvest time, the Iban gather each variety separately. Long and short grain rices are stored separately, and varieties that have not produced well that year are gathered and stored separately (V. Sutlive, personal communication). In Thailand, Rerkasem and Rerkasem (1984) note that farmers have a good appreciation for the concepts of “variety” and “selection”. Farmers in Chang Mai identified 42 varieties from 55 grain samples according to color, shape and size of the grain and panicle branching and size (Rerkasem and Rerkasem 1984:304-305}. All of this indicates the high level of cultural awareness and care for the genetic diversity of rice that is generalized across a wide area. —_— ———— | Summer 1986 JOURNAL OF ETHNOBIOLOGY 159 Rice cultivation in insular Southeast Asia is customarily divided into two major subsystems: upland (dry) rice and lowland {irrigated} rice. Variations exist (Padoch 1983), and irrigated areas may add a dry season rice crop which is not irrigated (Grist 1965). Upland rice is directly seeded and intentionally grown under aerobic conditions (McIntosh et al. 1984). Upland rice covered the majority of rice acreage in Asia, accounting for 61%, while irrigated rice covers 33% and deep water rice 6% (Huke 1982). In Southeast Asia, however, upland rice accounts for only a small perdentage of the total area—in Indonesia 11% and in the Philippines 14% in 1977 (IRRI 1984). Upland rice is associated with lower inputs (labor, capital) and also with lower productivity, averaging less than 1ton/ha compared to 4ton/ha for irrigated rice (IRRI 1984; Barker 1978). Seden- tary agriculture accounts for most upland rice, although it is also found in swidden agriculture. In each it is primarily associated with subsistence production in which rice is supplemented by other crops (IRRI 1984). Upland and lowland rice varieties belong to the same species but are distinguished along a continuum of morphoecological groups. Lowland rice exceeds upland in the number of varieties and genetic diversity. This would appear to contradict the image of a homogeneous agroecosystem for irrigated rice. Upland rice is distinguished by diver- sity relating to drought tolerance (IRRI 1984). It is associated with cropping systems that are generally more diverse and “traditional” than lowland ones. Chang (1976) suggests that the smaller diversity among upland types indicates that they are not ancestral to lowland types. Lowland rice in Asia is famous for intensity and productivity. Geertz (1963) describes this as a product of “involution,” an evolutionary process whereby increased labor is absorbed by fine-tuning the complex paddy system to achieve increased productivity. Although involution may be disputed (Collier 1981), the evolution of irrigated rice in the Philippines and Indonesia has resulted in high population densities, land fragmen- tation, and much higher yields than found in the upland systems. Densities associated with upland swidden systems generally fall below 50 persons per square kilometer while irrigated systems support 500 or more (Geertz 1963; Gourou 1966). On Java, which is dominated by wet rice, 63% of the land under cultivation is in farms of less than 0.5 hectares, and the average farm size is 0.64 ha (Birowo and Hansen 198 1:5). On the Outer Islands of Indonesia, where irrigated rice is secondary to upland systems, the figures are the opposite, with 70% of the farms larger than 0.5 ha. On Kalimantan, where swidden is common, the average farm size is 2.71 ha. For dry rice production in Indonesia, Gourou (1966) reports production of 812 kg. per hectare, while the Philippines wet rice employing traditional methods yields an average 2427 kg/ha (Conklin 1980). The disparity is even greater when moder varieties and other subsidies are applied to wet rice, and yields of over 4000 kg/ha are achieved (Barker 1978). Another distinctive characteristic of irrigated rice is the dominance of rice as a cultigen and subsistence item. Although rice is culturally significant among the Hanunoo, it accounts for less than 20% of their diet. Irrigated rice, on the other hand, is associated with monocropping and with heavy reliance on rice as the staple (Gourou 1966). Although the relative hegemony of rice decreased in Java during the Colonial period, it has continued in other parts of Southeast Asia (Geertz 1963). A final distinction between the two systems is that irrigated rice is associated with an elaborate and stratified land tenure system in contrast to the simple and egalitarian system of tenure of swidden. The tenure system of paddy reflects the density of the popula- tion and the complexities of the water control system. Differential access is determined by ownership and a variety of rental and share tenancy arrangements (Utami and [halauw 1978; Conklin 1980). These create disparities between different sized farms and between landowners and the landless. Although the distribution of paddy is relatively egalitarian 160 BRUSH Vol. 6, No. 1 among those who own land in Java, over 40 percent of the population is landless or near landless (Birowo and Hansen 1981). The rate of change is greatest in lowland systems, that is in those that originally had the greatest diversity. As expected from this difference, the rate of genetic erosion in lowland rice greatly exceeds that of upland rice. Special agronomic characteristics, such as deep water or pathogens limit adoption, and cultural characteristics, such as preference for glutinous rice, may also be limiting. Considerable research has been devoted to the question of farm size as a limiting condition, concluding that it is not a signi- ficant factor (Herdt and Capule 1983). Lowland rice systems have proved to be among the most dynamic small farm economies in the world, measured by the adoption rate of HYVs and other inputs. Herdt and Capule (1983:5) report that between 1966 and 1981, the overall percentage of Asian rice area planted in modem varieties rose from 1.4% to 39.5%. In countries where lowland rice predominates this rate is much higher. In the Philippines, for instance, by 1981 77.4% of the total rice area was planted in modern varieties. Although the Philippines represents an unusually high national average, numerous areas within other countries have achieved similar rates (Herdt and Capule 1983; Bernsten et al. 1982; Chang 1984). The extent of genetic erosion in lowland rice in the Philippines is unusual in Asia, although the potential for replacement exists elsewhere. Indonesia, for instance, had 60% of all its rice land in modem varieties by 1981 (Herdt and Capule 1983). There is, however, some reason to speculate that HYV replacement may be limited in some lowland or irrigated rice systems. Prabowo and Sajogyo (1981) compare two villages on Java with widely differing adoption rates. In the East Java village, virtually complete adoption was reported, while in the West Java village, adoption rose and then sharply declined to 17 % following an outbreak of gall midge. Poor water control and lack of fertilizer have limited adoption elsewhere (Herdt and Capule 1983). Thailand is a striking contrast to the rapid diffusion of HYVs in the Philippines and Indonesia. By 1980, less than 10% of the Thai rice area was planted in HYVs, even though lowland rice predominates (Herdt and Capule 1983). Several reasons have been suggested for this low rate. Thailand is famous for high quality and glutinous rices, especially for export, and HYVs are not regarded as competitive in quality. Glutinous rice is preferred as a staple, and although glutinous modern varieties are available, they are interior to native varieties (Rerkasem and Rerkasem 1984). Fukui (1975) notes that native varieties performed as well as HYVs, given limited fertilizer use and underdeveloped water control. He observed that Thai farmers adopted HYVs not as a replacement to indigenous varieties but as a second crop because of their short growing period an aE: photosensitivity. Rerkasam and Rerkasem (1984) describe the ethnobiological basis of non-adoption in the Chang Mai Valley. Their case study reveals that HYVs are grow? on 5% of the land in the valley. Traditional varieties are grown on 20%, and two selected local varieties cover 55% of the land. These authors note that numerous criteria are used to evaluate rice varieties and that the farming system has many specific agr oclimatic niches for which specific varieties are selected. It is unlikely that any single variety ca? meet all of these criteria or perform well across all niches. eal Data from other lowland rice growing areas in the Asian center of diversity 18 insufficient to judge whether we should regard Thailand as typical of resistance to t : diffusion of HYVs on the Asian mainland. Farmer (1979) notes that the new varieties have limited use during the wet season in India, Sri Lanka and Bangladesh, although they are widely used during the dry season, because of their non-sensitivity to photoperiod and their short duration. Regional wet season problems, such as deep water are important in limiting diffusion, as are problem soils in some areas (Farmer 1979). On Indonesia’s outer islands, the percentage under improved varieties declines in higher elevations and areas under tidal influence and with toxic soils (Bernsten et al. 1982). PRET ee, erence Summer 1986 JOURNAL OF ETHNOBIOLOGY 161 In South Kalimantan, for instance, modern varieties account for only 30% of the area in wet seasons and 20% in dry seasons (Bernsten et al 1982:21). Besides these agronomic issues that limit diffusion of HYV rice, economic issues can also be identified. The first economic issue to be studied in relation to adoption was scale of farming, and this proved to be a non-issue. Small farmers were repeatedly found to be adopters, and in some areas, they actually exceeded larger farmers in adopting HYVs (Herdt and Capule 1983). In lowland rice, more marginal zones (as measured by degree of water control, access to major markets, etc.) are less likely to adopt than more optimal areas. This difference would appear to explain the relatively low adoption of modern varieties in Orissa (25%) compared to other Indian States, Andhra Pradesh (63%), West Bengal (41%) (Herdt and Capule 1983). € most significant economic issue dividing high and low adoption is the difference between lowland and upland rice. Unlike lowland rice, the upland version is characterized by continued use of traditional varieties. Nepal with 75% of its rice in upland systems only achieved a 25% adoption by 1980 (IRRI 1984; Herdt and Capule 1983). In India, states with the highest proportion of upland rice show generally lower adoption of modem varieties. In Madhya Pradesh, with 40% of the rice area in upland, less than 2% is planted to modern varieties (IRRI 1984; Herdt and Capule 1983). Two limiting factors for adop- tion of improved rice varieties in upland farming systems have been observed: the dependence of improved varieties on water and the requirement for technical knowledge of and heavy investments in fertilizers (Kunstadter et al. 1978). These two characteristics make the new varieties inherently unacceptable to poor upland farmers. Besides the inappropriate nature of HYVs for upland systems, the particular characteristics of those systems must be understood in relation to technological adop- tion. As noted above, these systems have always been less intensive than lowland ones. They are associated with subsistence rather th ket product, margi agroecological zones, and culturally distinct groups. Upland agriculture is often located in mountainous regions where slopes, soils and, socioeconomic conditions impede the construction of irrigation and terrace systems for paddy rice. Moreover, upland rice is usually only one component, often secondary, in diverse farming systems. The investment to learn about and acquire new varieties thus represents an opportunity cost to the other components. DISCUSSION Similarities and differences in the pattern of adoption of improved crop varieties are evident in comparing these two centers of genetic diversity. In both the Andes and Southeast Asia, the pattern of adoption is associated with a dichotomy between farming Systems and with an upland/lowland dichotomy. In both, the genetic diversity of the Major crop is retained in more marginal agroecological zones where farming systems mixed and oriented toward subsistence. In the Andes, these zones are associated with higher altitudes where the environmental limits to the crop are more proximate. In Asia, these zones are also associated with hill agriculture and higher altitudes. In each area, advantages of the improved varieties are more apparent in the more intensive lowland system, where they are adopted. These systems are characterized by commer- cial production, greater control over the various inputs of agriculture, and better access to markets. The size of a particular farm does not seem to deter adoption in these areas. In some Asian rice systems, as in Andean potato systems, the adoption of new varieties is accomplished because they can fill special niches, such as in dry season rice farming. In upland farming systems in both regions the advantages of improved varieties are outweighed or made irrelevant by other features of the system. In the Andes, high altutude potato farmers adopt new technology, but they do not replace older varieties with 162 BRUSH Vol. 6, No. 1 improved types. The costs of seed, poor market access and culinary preference are the major limiting f In Asia, improved varieties of rice are not suited for specific lowland environments, such as deep water farming, and they are generally not acceptable to upland farmers who lack water control and other intensive management techniques. In each area, the availability of other critical inputs, such as fertilizers, influence the adoption of crop varieties. The similarities between potatoes and rice in respect to genetic erosion may be summarized by reference to the importance of agroclimatic zone, cultural preference, and farming system differences regarding control over inputs, markets access, farm diver- sity, and subsistence orientation. Genetic change in each is more rapid in areas where intensive and commercial agriculture existed prior to improved varieties. In marginal areas, adoption is slower and is often accomplished by fitting improved varieties into a diverse farming system, rather than by simple replacement of the older seed stock. While potato and rice farming systems share several attributes that indicate a general pattern of genetic erosion, it is also important to note that they seem to differ in the extent of change. Although it is impossible to generalize across all of Asia or the Andean region, rice in some Asian countries has experienced far greater change than potatoes anywhere in the Andes. This may be because new potato varieties do not out- perform older ones to the same extent as new rice varieties. No Andean country in the original hearth of potato domestication has seen a majority potato fields planted in improved varieties, while several Asian countires have passed this mark in rice. These differences show clearly in a comparison between Peru and the Philippines. Unfortunately, regional and national data on potato agriculture in Peru is not comparable to informa- tion about Philippine rice, but the limited data available suggests that genetic erosion there is perhaps less than half the Philippine adoption rate of 77%. Although a majority of Peruvian farms in regions close to urban areas have converted to modern varieties, this rate drops sharply with distance from cities. With the exception of the upland/lowland rice dichotomy, the Philippine case shows a high overall adoption rate, regardless of location in relation to urban markets. Several differences between the two crops and farming systems may be cited in rela- tion to the more dynamic performance of rice in the Philippines. A primary one has to do with the national markets of each crop. Entering the 1960s, the Philippines experienced a serious deficit in rice production for its national consumption, and demand for rice has remained high. Peru, on the other hand, has only faced potato shortages in time of severe crop failure, and it has imported potatoes only in emergencies. The urban popu- lation of Peru consumes fewer potatoes than the rural one, and this keeps the potato market relatively weak. The strong national market in the Philippines seems to have reduced the disadvantages experi i by more isolated farms that are important in Peru. Transportation costs are relatively less for rice than for potatoes, and there 1s no comparable need to market the crop as a fresh vegetable. Besides these market differences, two other factors may be cited to account for the lower rate of genetic erosion in Peru: a) different emphasis given to crop development in the two countries and b) fundamental agronomic differences between the crops. The high overall adoption of improved varieties in the Philippines followed large scale promotion campaigns which made credit, fertilizers, information, and improved vaneties available. Faced with a serious deficit in rice production in the late 1960s, the Phlip- pines undertook a successful agricultural extensi prog! , the Masagana 99 program. This has been heralded as responsible for the dramatic changes in Philippine "ce agriculture (Merrick 1981), typified by the adoption of HYVs. The success in impro Philippine rice production to self sufficiency through technological innovation is closely correlated with institution building in national research and extension programs. Summer 1986 JOURNAL OF ETHNOBIOLOGY 163 Peru, on the other hand, has a generally poor record in improving agricultural productivity and a mixed record in institution building and extension. For the last 40 years, Peru has conducted national agricultural research programs aimed at improving potato production, and these programs have turned out over 30 improved varieties since 1950. Yet from 1954 to 1972, potato yields actually fell by 25% (Eastman and Grieshop 1986). No single campaign comparable to the rice programs in the Philippines was ever mounted in Peru. Peru has always been self sufficient in potato production (Eastman and Grieshop 1985) and urban food habits emphasize rice and bread rather than potatoes (Ferroni 1981). Agricultural research and extension suffered major setbacks in the 1968 military revolution, that attacked rural underdevelopment through agrarian reform rather than technological change. The National Potato Program ceased to function in the early 1970s, and the national agricultural extension service was abolished shortly after the military revolution. The National Potato Program was reorganized with CIP’s help in the early 1980s, and the national agricultural extension service has been refashioned. The adoption rates of new crop varieties in Peru and the Philippines are also affected by differences between the two crops. Perhaps the most important difference relative to genetic erosion is the ability of new rice varieties to outperform older varieties. In contrast, the newer potato varieties have some advantages, such as precosity or disease resistance, but their overall performance is not as clearly advantageous as rice HYVs. As an asexually reproduced and vegetatively propagated crop, the potato is disadvantaged by very slow seed multiplication. The time that it takes to produce a sufficient amount of new seed material in tuber crops is roughly ten times what sexually reproduced cereal crops require. The promotion of improved varieties in rice is enhanced by the ease of seed duplication, and by less cumbersome seed production and distribution systems. An important result of these differences is the availability and acceptability of greater amounts of improved rice seed to all types of farmers. Another difference between the two crops is the complexity of seed maintenance in potatoes. As discussed above, native Andean agriculture employs a system of seed rotation between ecological zones. New varieties can be incorporated into this system, but farmers who adopt usually depend on regular purchase of new seed because of the degeneration of new varieties. This is a major limiting factor on the rate of adoption. In rice, on the other hand, no such cumbersome seed rotation system is required, and improved varieties have been easily incorporated into the pre-existing seed system. A third difference between potatoes and rice concerns storage. Potatoes, of course, are a fresh vegetable with a very high water content. In storage, they lose weight at water ‘vaporates. Weight loss may be controlled by refrigeration, but this is not an option available to small farmers. As noted above, small potato farmers in Peru are in direct competition with larger farmers in the highlands or on the coast. Larger farmers who can afford refrigerated storage can play the market to their advantage by waiting until the price is advantageous. Small farmers without this storage capacity cannot afford to delay selling their crop. The result is that traditional varieties are kept for consumption and household income is sought from other sources. ; The patterns of adoption and diffusion of new varieties of rice and potatoes described above are derived from the agroecology of the different farming systems producing each ctop, by the socio-economic characteristics of these systems, and by the nature of the Particular crop. Although retention of diverse, native germplasm in each crop is still evident, the ongoing evolution of these farming systems may later this picture. Systemic Pressures for the adoption of improved varieties are present in both regions. These include rapid population increase, changing expectations a d puon patterns, and € generation of new technology more appropriate to marginal farm areas. This paper Suggests that it is now advantageous for the farmers in these mixed agroecosystems to retain their ancestral landraces. Genetic change in these marginal areas will take renewed 164 BRUSH Vol. 6, No. 1 and increased research and promotion and will require change in the overall agricultural system to make other technological inputs available. Competition with the more optimal farm regions already changed by the green revolution will retard this. The conclusion is that genetic erosion does not occur evenly and will probably not proceed as rapidly as once expected. Agricultural development in marginal areas is but one of many alternatives for developing countries, and there are few institutional or political pressures to invest heavily in this. On the other hand, the people in these areas are as eager as any to improve living standards. As the world looks to ways to protect its genetic resource base, the continued cultivation of traditional varieties in m al areas may be seen as a positive rather than a negative feature. The challenge for develop- ment planners is to devise ways to improve living standards of poor farmers and to encourage continued cultivation of traditional varieties. Conservation cannot and should not be pursued at the expense of the well-being of farmers who today manage the Neolithic legacy. NOTE lResearch for this paper was supported by grants from the National Science Foundation, the International Potato Center, and the College of William and Mary. The author wishes to acknowledge the collaboration of Enrique Mayer and Cesar Fonseca in the fieldwork. He is also grateful to Benjamin Orlove, Douglas Horton, Deirdre Bradley and Margaret Brush for their comments on earlier drafts of this paper. LITERATURE CITED ALTIERI, M. A., D. K. LETTOURNEAU and J. R. DAVIS. 1983. Developing sustainable agroecosystems. Bio- Science 33:45-49. APPLEBY, G. 1976. The role of urban food needs in the regional develop- ment of Puno, Peru, Pp. 147-178 in G. E. Hansen, ed.). Westview Pr., Regional Analysis Vol. 1 (C. Smith, ed.). Academic Press, New York. BARKER, R. 1978. Yield and fertilizer input in Interpretive Analysis of Selected Papers from Changes in Rice Farming in Selected Areas of Asia. IRRI (ed.). Los Bafios, Philippines. BERNSTEN, R. H., B. H. SIWI and H. M. BEACHELL. 1982. The development and diffusion of rice varieties in Indonesia. IRRI Research Paper Series No. 71 o. 71. BIDEGARY, P. and E. SCHMIDT. 1985. Seleccion y adopcion de variedades de papa en Cusco. Unpubl. ms. Inter- national Potato Center, Lima. BIROWO, A. T. and G. E. HANSEN. 1981. Agricultural and rural development: an overview, Pp. 1-27 in Agricultural and Rural Development in Indonesia Andean Potato Agriculture. Econ. Botany 35(1):70-85. CABALLERO, J. M. 1981. Economia Agraria de la Sierra Peruana Antes ee la Reforma Agraria de 1969, Lima, IEP. CARTER, W. and M. MAMANI. 1982. Irpa Chico: Individuo y comunidad or la cultura aymara. La Paz, Juventud. iJ ie CHANG, T. T. 1976. The origin, evo tion, cultivation, dissemination am diversification of Asian and African rices. Euphytica 25:425-441. _ _ 1984. Conservation 0 rice genetic resources: luxury OF necessity? Science 224:251-256. CHEDD, G. 1970. Hidden peril 0 pe green revolution. New Scientist *®- 171-173. Summer 1986 JOURNAL OF ETHNOBIOLOGY 165 LITERATURE CITED (continued) CLAY, E. J. 1975. Equity and productivity effects of a package of technical innovations and changes in social institutions: tube wells, tractors, and high yielding varieties. Indian J. Agric. Econ. 4:74-87. COLLIER, W. L. 1981. Agricultural evolu- tion in Java, Pp. 147-176 in Agricul- tural and Rural Development in Indonesia. (G. E. Hansen, ed.). Westview Pr., Boulder, CO. CONKLIN, H. C. 1980. Ethnographic Atlas of Ifugao. Yale Univ. Press, New Haven. DALRYMPLE, D. G. 1978. Development and spread of high-yielding varieties of wheat and rice in the less developed nations. Foreign Agric. Econ. Report. 6th Edition, Washing- ton, D.C. DEWALT, K. M. 1983. Nutritional Strategies and Agricultural Change in a Mexican Community. Univ. Michigan Press, Ann Arbor. EASTMAN, C. and J. GRIESHOP. 1986. Technology development and diffu- sion: potatoes in Peru. In The Transformation of International Agri- cultural Research and Development: Some U.s. Perspectives. (L. Compton, ed.). Westview Press: Boulder, CO. EVEASON, R. 1973. Research, extension _. Rural Development. (P. Foster and J. R. Sheffield, eds.). Evans Bros., London. FARMER, B. H. 1979. The “green revo- lution” in South Asian ricefields: €nvironment and production. J. Development Studies 15:304-319. FEDER, G., R. E. JUST and D. Z. Zilber- MAN . 1982. Adoption of agricultural innovations in developing countries: a survey. Washington, D.C., World Bank Staff Working Papers, No. 542. FERRONI, M. 1980. The urban basis of Peruvian food policy. Ithaca, NY. Cornell University, Ph.D. thesis. , O. H. and E. BENNETT (eds.. 1970. Genetic Resources in Plants— Their Exploration and Conservation. IBP Handbook No. 11. Blackwell Scientific Pub., Oxford. FRANKEL, O. H. and J. G. HAWKES (eds.). 1973. Crop Genetic Resources for Today and Tomorrow. IBP 2. Cambridge University Press, Cambridge. FRANKEL, O. H. and M. E. SOULE. 1981. Conservation and Evolution. Cam- bridge University Press, Cambridge. FRANCO, E., D. HORTON and F. TAR- DIEU. 1979. Produccion y utilizacion de la papa en el valle del Mantaro- Peru. Social Science Unit Work Document No. 1979-1. International Potato Center, Lima, Peru FUKUI, H. 1975. Paddy production technology, present and futur Pp. 246-271 in Thailand: A Rice Growing Society. (Y. Ishii, ed)). University Press of Hawaii, Honolulu. GEERTZ, C. 1963. Agricultural Involu- tion: The Process of Ecological Change in Indonesia. Univ. Califor- nia Press, Berkeley. GOUROU, P. 1966. The Tropical World. Fourth Ed. Wiley, New York. GONZALES DE OLARTE, E. 1984. Economia de la Comunidad Campesina. Instituto de Estudios Peruanos, Lima. GRILICHES, E. 1957. Hybrid corn: an explanation in the economics of technological change. Econometrica 25:501-522. GRIST, D. H. 1965. Rice. Fourth Ed. Longmans, London. HANKS, L. 1972. Rice and Man. Aldine, Chicago. HAWKES, J. G. 1978. Biosystematics of the Potato, Pp. 15-69 in The Potato Crop. (P. M. Harris, ed.). London, Chapman & Hall. . 1983. The Diversity of Crop Plants. Harvard University Press, Cambridge, MA. HERDT, R. W. and C. CAPULE. 1983. 166 BRUSH Vol. 6, No. 1 LITERATURE CITED (continued) Adoption, Spread, and Production Impact of Modern i cafera in Asia. IRRI, Los Banos, Philippine HORTON, D. E. 1984. Social as in Agricultural Research: Lessons from the Mantaro Valley Project, Peru. IDRC, Ottawa, Ont HUKE, R. E. 1982. Rice Area by Type of Culture: South, Southeast, and East Asia. IRRI, Los Banos, Philip- pines. and J. DUNCAN. 1969. Spatial aspects of HYV diffusion. Studies in the Diffusion of Innova- tion, Discussion Paper No. 6. Dept. Geography, Ohio State Univ. IRRI. 1984. An Overview of Upland Rice Research. IRRI, Los Banos, Philip- pines. JENNINGS, P. R. and J. H. COCK. 1978. Centres of origin of crops and their productivity. Econ. Botany 31:51-54. KASS, D. C. L. 1978. Polyculture crop- ping systems: review and analysis. Cornell International Agriculture Bulletin 32. KUNSTADTER, P., S. SABHASRI and a. SMITINAND, eds. 1978. Farmers in the Forest. Univ. Hawaii Press, Honolulu. re M. 1976. Agricultural finance rural credit in poor countries. World oo 4:543-554. MCINTOSH, J. L., Z. HARAHAP and B. H. SIWI. oe Asiety upland rice cropping system, Pp. 503-510 in An Overview of Upland Rice Research. (IRRI, ed.). Los Banos, Philippines. MERRICK, J. 1981. Masagna 99: pro- moting a miracle; an agricultural campaign in the Philippines. Academy for Educational Develop- ment, Washington, D.C. MESSERLI, B. 1983. Stability and insta- bility of mountain ecosystems: intro- duction to a workshop sponsored by the U.N. University. Mountain Research and Development 3:81-94. MYERS, N. 1981. The Sinking Ark. Pergamon Press, Oxford. NATIONAL ACADEMY OF SCIENCES. 1972. Genetic Vulnerability of Major Crops. Washington, D.C. NORMAN, D. W. 1969. Labor inputs of farmers: A case study of the Zaria Province of the North-Central State of Nigeria. Nigerian Journal o Economic and Social Studies 11:1-13. OCHOA, C. 1975. Potato collecting expeditions in Chile, Bolivia and Peru, and the genetic erosion of in- digenious cultivars, Pp. 167-173 in Crop Genetic Resources for Today and Tomorrow. (O. H. Frankel and J. G. Hawkes, eds.). IBP Vol. 2. Cam- bridge University Press, Cambridge. PADOCH, C. 1983. Agricultural prac- tices of the Kerayan Lun Dayeh. Borneo Research Bulletin 15:33-38. PARTHASARATHY, G. and D. S. PRA- SAD. 1978. Response to the impact of new rice technology by farm size and tenure: Andrea Pradesh, India. Los Banos, Philippines, IRRI. PERRIN, R. and D. WINKELMAN. 1976. Impediment to technical pro- gress on small vs. large farms. Amer. J. of Agri. Economics 58:888-894. PERU, MINISTERIO DE ALIMENTA- CION. 1975. Cultivo de la Papa en la Sierra. Informe Especial No. 39. Lima. PLUCKNETT, D. L. and N. SMITH. 1983. Crop germplasm conservation and developing countries. Science 220: 163-169. PRABOWO, D. and SAJOGYO. 1981. Sidoarjo, East Java, and Subang, West Java, Pp. 68-78 in Agricultural and Rural Development in Indonesia. (G. E. Hansen, ed.). Westview Press, Boulder, CO. REKASEM, B. and K. RESHAM. 1984. The agroecological niche and farmer selection of rice varieties in the Chiang Mai Valley, Thailand, Pp. 303-311 in An Introduction t0 Human Ecology Research on Agricul- sees Systems in Southeast Asia. (A. Rambo and P. E. Sajise, eds.). Univ- vsieiessea Laguna, Philippines. Summer 1986 JOURNAL OF ETHNOBIOLOGY 167 LITERATURE CITED (continued) ROUMASSET, J. A. 1976. Rice and risk: decision making among low income farmers. North Holland Publ. Co., Amsterdam. SUTLIVE, V. H. 1978. The Iban of Sara- wak. AHM Publ. Corp. Arlington Heights, Illinois. TAPIA, M. 1983. Diagnostico de 8 com- unidades alto-andinas del Peru: Cuzco, Puno, Ayachucho. Cuzco, Peru, Proyecto PISCA, CIID, IICA. mimeo. UCKO, P. J. and G. W. DINBLEBY (eds.). 1969. The Domestication and Exploi- tation of Plants and Animals. Duck- worth, London. U.S. STRATEGY CONFERENCE. 1981. Proceedings of the U.S. Strategy Con- ference on Biological Diversity. U.S. Department of State Publication 9262, Washington, D.C. UTAMI, W. and J. IHALAUW. 1978. The relation of farm size to production, land tenure, marketing, and social structure—Central Java, Indonesia, Pp. 129-140 in Interpretive Analysis of Selected Papers from Changes in Rice Farming in Selected Areas of Asia. (IRRI, ed.). IRRI, Los Banos, Philippines. VANDERPLANK, J. E. 1968. Disease Resistance in Plants. Academic Press, New York. VAVILOV, N. I. 1951. The origin, vari- ation, immunity and breeding of cultivated plants. Chronica Botanica 13:1-366. WERGE, R. W. 1979. Potato processing in the central highlands of Peru. Ecol. Food and Nutrition. 7:229-234. WHARTON, C. R. JR. 1969. The green revolution: cornucopia or pandora’s box? Foreign Affairs 47:464-476. WILKEN, G. C. 1972. Microclimate management by traditional farmers. Geog. Rev. 62:544-566. he aa Poe 7 aoe aoe ae 7 ‘aah ai 7 + i i ao ae fo, joss ne _s . nies aa J. Ethnobiol. 6(1):169-204 Summer 1986 SECTORAL FALLOWING SYSTEMS IN THE CENTRAL ANDES BENJAMIN S. ORLOVE Division of Environmental Studies University of California Davis, CA 95616 RICARDO GODOY Harvard Institute of International Development Harvard University Cambridge, MA 02138 ABSTRACT.—Sectoral fallowing systems from 51 communities (Fig. 1) in highland Peru and Bolivia have been examined, focusing on altitudinal and latitudinal variation in these systems. In th inst f ] g f Pr : ra irom || : 1g, a significant proportion of lands is left fallow. This practice has been studied for implications related to maintenance of soil quality and reduction of pathogen impacts. Recent changes in several systems are noted. Comparison with other Andean patterns of culture and social organization reveal similarities with the present study area. INTRODUCTION Many peasants in the Andean highlands practice a complex pattern of crop and Pasture management called sectoral fallowing systems. These systems, found in a large number of communities! distributed over a wide territory, are of interest to ethnobiologists for several reasons. They show that human knowledge and use of plants and animals can lead to the coordination of behavior of many individuals. They also demonstrate the complex interaction of human populations and the environments in which they live. Since there are a large number of these systems, they are particularly well-suited to statistical analysis. Anthropologists who specialize in the Andean region have been drawn to study these Systems, not only because they are an important aspect of economic organization in the region, but also because they appear to exemplify important forces which shape peasant life there. Some writers select an adaptationist perspective, viewing human activity as constrained by the difficulties of extracting resources from fragile and unproductive mountain ecosystems. Others emphasize the cultural continuities between earlier periods and the present, and see in agriculture and pastoralism, as in other areas of activity, underlying Andean patterns of thought and belief. Still others stress the importance of political economy, of the ways in which peasants and pastoralists, as one class in a dependent, stratified society, cope with the pressures from elites, governments and markets. This article examines these sectoral fallowing systems in detail. It describes the basic features of these systems and analyzes the variation among the systems. The evidence Provides general support for the adaptationist view of these systems, but also indicates the significance of Andean cultural patterns. It does not directly address the political economy perspective, neither confirming or challenging it. The evidence hints at recent shifts in these systems associated with population increases and commercialization of agriculture, but the lack of quantitative information does not permit a detailed exploration of these important links. 170 ORLOVE & GODOY Vol. 6, No. 1 SECTORAL FALLOWING SYSTEMS: DEFINITIONS Early ethnographic accounts of rural Andean life comment on community-wide management of cycles of planting and fallowing (Bandelier 1910:83, 85), but it has only been recently that these systems of land use have been described in detail. The account of this system by Matos Mar (1964) is one of the earliest full ones. The community of Paracaos in the province of Canta, department of Lima in the central Peruvian highlands may be taken as typical (Table 1). The 163 households which make up this community have access to some irrigated plots, which are cultivated each year, and to natural pastures. In addition, they have 929 hectares of land which they cultivate in some years and let lie fallow in others. These lands are divided into ten sectors. Most households own plots in all of the sectors. These households plant potatoes in their plots in a given sector in one year. In the next year, those who wish to do so plant other Andean tubers (0ca, ABLE 1.—Cultivation cycle of sectoral fallowing systems in Paracaos (Degregoti and 5 T Golte 1973:44-45). YEAR Name of Sector 1958 1959 1960 1961 1962 Huayatama pasture pasture pasture pasture pasture Cushurumachay pasture pasture pasture pasture pasture Shujuncha min. tubers! pasture pasture pasture pasture Chuyochacra potatoes min. tubers pasture pasture pasture Acopuquio pasture potatoes min.tubers pasture pasture Ayal and Ushtuna pasture pasture potatoes min.tubers pasture Tamburhuasi pasture pasture pasture potatoes min. tubers Liuli and Mitaysinsan pasture pasture pasture pasture potatoes Sinsanchacra pasture pasture pasture pasture pasture Milacancha pasture pasture pasture pasture pasture Name of Sector 1963 1964 1965 1966 1967 Huayatama pasture pasture potatoes min. tubers pasture Cushurumachay pasture _ pasture pasture potatoes min. tubers Shujuncha pasture pasture pasture pasture potatoes Chuyochacra pasture pasture pasture pasture pasture Acopuquio pasture pasture pasture pasture pasture Ayal and Ushtuna pasture pasture pasture pasture pasture Tamburhuasi pasture pasture pasture pasture pasture Liuli and Mitaysinsan min. tubers _ pasture pasture pasture pasture Sinsanchacra potatoes min.tubers pasture pasture pasture Milacancha pasture potatoes min.tubers pasture pasture 1 minor tubers are olluco, oca, mashua Summer 1986 JOURNAL OF ETHNOBIOLOGY 171 Oxalis crenata; ollucu, Ullucus tuberosa; and mashua, Tropaeolom tuberosum) in the same plots, although others choose not to plant in this second year. All of them let these Plots lie fallow for the next eight years. Although individuals in the community recognize the boundaries and ownership of the plots, households have exclusive rights to use their plots only during the years when they are planted. During the fallow years, the plots are used as pasture by all the members of the community. The plots in all of the sectors pass through the same sequence of uses. However, the timing is staggered, so that no two sectors begin the sequence in the same year. When a particular sector is about to be culti 1 after eight years of fallow, the community authorities indicate the day when agricultural activities will begin; they also ratify the ownership of particular households, and can redistribute plots which are abandoned or left vacant by owners who died without leaving heirs. Other sectoral fallowing systems vary in the details of the crops which are grown, the length of the cycle, and the nature of the communal control, but they share with Paracaos a number of imporant features. (1) Each of the sectoral fallowing systems is a land-use system which consists of a set of lands associated with a set of households; (2) The set of lands are divided into a number n of sectors. The lands which make up each sector are contiguous; (3) All households own plots in most or all sectors. Most or all households own plots in all sectors; (4) There is a sequence of n year-long uses for the lands. Some of these uses consist of the planting of a specific annual crop or small number of annual crops. The set of crops may differ among the successive cropping uses out of the total set of uses. The other uses are fallowing, combined with grazing. All the fallowing uses occur after all the planting uses; (5) All m sectors pass through the same Sequence of n year-long uses. In any given year, one and only one sector will have each Cropping use, and the number of sectors which are fallowed is equal to the number of fallowing uses in the n year cycle; (6) When a sector is used for planting, each household has access and usufruct rights to its plot or plots in that sector; (7) When a sector is used for fallowing and grazing, all households have access and grazing rights to the entire sec- tor; and (8) This land use pattern is maintained and enforced by institutionalized means. These systems have awakened the interest of a number of anthropologists and S8eographers. They seem large in their spatial, temporal and social scale. Many specialists, accustomed to thinking of peasants as competitive and uncooperative, are surprised by this coordination of activity. How widespread are sectoral fallowing systems? How may their presence and distribution be explained? Along what dimensions do the sectoral fallowing systems differ from one another? Are there any regularities in their patterning? As in other areas of human activity in the Andes, several lines of explanation may be Presented. The sectoral fallowing systems could be seen as adaptations to mountain environments, as a historical product of the political economy, or as an expression of a cultural patterns. These explanations could be integrated in a variety of ashions METHODS Literature reviews conducted independently by the two authors (Custred and Orlove 1974, Campbell and Godoy in press) showed several types of sources: some detailed ethnographic studies of single sectoral fallowing systems, brief references to sectoral fallowing systems throughout the Andes, and a few studies of changes within sectoral fallowing systems in particular regions. We also sent a questionnaire to 119 anthro- Pologists and geographers who have conducted research in the Andes. We received 27 teplies. There were some instances of incomplete or missing information in both Published materials and completed questionnaires. Our sample of 51 communities (Appendix 1) is thus not random. We hope that it represents a fairly complete presen- tation of available research on communities with sectoral fallowing systems. L772 ORLOVE & GODOY Vol. 6, No. 1 There are several basic variables which we consider in detail and which are sum marized in Table 2; though simple, they have not been fully distinguished in earlier writings. The cropping and fallowing cycle in any given system is as many years long as there are sectors in the system, since each sector must pass through every year of the cycle and since no two sectors are at the same point in the cycle in any given year. Thus the number of sectors which are in fallow at any given time is the same as the number of years which each sector lies fallow. We will discuss in particular the follow- ing variables: the number of years which each sector lies fallow, which we call “number of fallow years” and the number of years in which each sector is cropped, which we call the “number of crop years’. The sum of these two variables is the number of years 1n the cycle, identical to the number of sectors in the system; we refer to it as “number of sectors”. The ratio of the number of fallow years to the number of sectors is called the “fallow ratio”. Because of the great importance of tuber crops, particularly the potato, in Andean agriculture, we constructed several measures. The “number of tuber years” refers to the number of years in the cycle in which tuber crops are grown exclusively or predominantly. This definition raises certain measurement problems. If we counted only years in which tuber crops were grow? TABLE 2.—Summary of variables utilized in this study. Description Value Name Basic features: Number of tuber years iF TUBERYR Number of other (non-tuber} rop years re) OTHCRPYR Number of fallow years F FALLYR Number of crop years T40 CROPYR Number of non-tuber years O+F NOTUBYR Number of sectors T+O+F NUMSECTS Percentage of time in fallow F(T + O + F FLRATIO Percentage of time not in T tuber cultivation (O + F(T + O + F) NOTBRA Tuber years as percentage of crop years TI(T + O} TUBINDX Other features: Mean elevation of plots in sectoral fallowing system ELVSFS Community elevation ELVCOM Number of sectoral fallowing systems associated with a specific community NUMSFS Summer 1986 JOURNAL OF ETHNOBIOLOGY 173 exclusively, we would have to reject cases in which a few other crops were occasionally interspersed in a sector otherwise planted with tubers. However, we did not wish to include cases where tubers accounted only for a small proportion of the crops which were planted in a specific year. In practice, this problem tumed out not to be difficult; most sources described the cropping patterns in sufficient detail to permit the coding of years in which both tubers and other crops were grown as tuber years or non-tuber crop years. The sum of the number of fallow years and the number of years in which crops other than tubers were grown is called the “number of non-tuber years”. The ratio of the number of non-tuber years to the number of sectors is called, somewhat awkwardly, the “non- tuber ratio’; the ratio of the number of tuber years to the number of crop years is called the “tuber index”. The tuber index ranges from 0.25, in the cases in which tubers account for only one of the four cropping years, to 1.0, when only tubers are grown. In addition we sought to establish the mean elevation of plots within the sectoral fallowing systems; for this purpose, we used the much more readily available average of the highest and the lowest elevations within the sectoral fallowing system. To establish the elevation of the community, we took the elevation of the central settlement, if there was one, or the average elevation of the permanent dwellings, if there was not. These variables are respectively called “system elevation” and “community elevation”. The examination of the data revealed three difficulties which might be termed the multiple-system problem, the zero-fallowing problem, and the multiple-community problem. They are all illustrated by the community of Irpa Chico, in the province of Ingavi, department of La Paz, Bolivia. A recent monograph about this community offers one of the best descriptions of land use systems anywhere in the Andean highlands (Carter and Mamani 1982). The inhabitants on this community distinguish between the sayaiia, a plot held by an individual or a household which can be used for grazing, cultivation or house construction; each household may decide independently on the use to which it will put each sayafa plot. An aynuqa2, by contrast, is a set of communally controlled lands managed as a sectoral fallowing system. Local community authorities retain con- siderable discretionary powers in reallocating plots within these aynuga to poor families. With a total of 513 households and 3109 hectares of land in the aynuga, the land- Population ratio for these lands is close to that in Paracaos. The total number of plots within the aynuga exceeds 11,000. In Irpa Chico, unlike Paracaos and many other com- munities, there are several sectoral fallowing systems or aynuqa rather than just one. Carter and Mamani suggest that there are eight aynuga within the community. The multiple-system problem consists in the necessity of deciding whether to consider each of the aynuqa as a distinct system. If all the aynuqa had the same sequence of planting anf fallowing, they might easily be considered to be a single sectoral fallowing system in which the sectors were non-contiguous. For example, if eight different aynuqa all went through a sequence of one year of potatoes, one of barley and six of fallow, then they could be considered to be a single sectoral fallowing system in which the plots in each Sector were dispersed among sub-sectors rather than being placed contiguously. However, there are at least six different sequences (op. cit. 87-88). Some of these resemble other sectoral fallowing systems; in one, for instance, three years of cultivation, first of potatoes, then of quinoa (Chenopodium quinoa) and finally of barley, are followed by three fallow years. However, in other systems there are no fallow years at all. In one aynuga, two years of potatoes are followed by four of barley, after which potatoes are planted once again. This zero-fallowing problem presents a dilemma: if the cases of zero-fallowing are excluded, then an artificial distinction is imposed between aynuga which the local Population all calls by a single term and manages by the same rules; their inclusion, ough, seems to contradict the basic character of the vast majority of sectoral fal- lowing systems—the alternation of planting and fallowing. 174 ORLOVE & GODOY Vol. 6, No. 1 Finally, the multi-community problem is presented by cases in which members of more than one community own plots within a single sectoral fallowing system. In Irpa Chico, each of the aynuqa are associated with a specific sub-community, which Carter and Mamani term zona or zones. Although Irpa Chico as a community shows a high degree of endogamy, there is also a fair degre of endogamy within the zones. Most households own the majority of the plots in the aynuga associated with their zone, although many of them own plots in other aynuqa as well. If Irpa Chico did not possess other strong communal organizations, each specific zone might be considered to be a distinct community. There are several other cases, all within Bolivia, where there is some ambiguity over the question of whether the term “community” should be assigned to a larger, more inclusive social unit, which might manage several sectoral fallowing systems, or to a smaller, more exclusive one with a single system (Alb6 1972, Buechler 1980, Campbell and Godoy in press). Their absence in Peru might be attributed to the policy of the Peruvian government of granting official recognition to communities; to receive such a title, a community must indicate precisely its boundaries and its compo- nent households. Their presence in Bolivia may also reflect a stronger continuity with the multi-level Andean ayllu form of organization (Platt 1982a). Because we both had field experience in the specific regions for which this multi-community problem appeared (Orlove in the Lake Titicaca area, Godoy in the northern portion of the department of Potosi], we believed that we could reasonably assign the limits to communities. It 1s quite possible that this multi-community problem appears in other areas, but has not been adequately documented. The more troubling zero-fallowing problem appears to be less frequent than the multiple-system problem. After some reflection, we decided to exclude the zero-fallowing cases, in part because they proved to be very difficult to detect; Inpa Chico is the single instance in the literature in which we could be certain that plots in zero-fallowing systems were managed sectorally, rather than at the level of the individual household, although other sources suggest that it may occur elsewhere (Painter 1981). There appear to be many such non-sectoral zero-fallowing cases of potato cultivation on unirrigated land and of intensive maize cultivation on irrigated land. The second justication for excluding zero-fallowing cases was the sense of Cartet and Mamani’s informants that such systems were recent and atypical, to be explained as the unfortunate consequence of land fragmentation. It thus seemed appropriate t0 consider such cases as former sectoral fallowing systems which would no longer be classified as such. Even this hedge did not completely eliminate the difficulty, because there were some cases in which fallowing was neither clearly present nor clearly absent. The island of Amantani in the Peruvian portion of Lake Titicaca, with a four-sector system, is one such ambiguous case. When a particular sector is first cultivated, it 18 called papasuyu or ‘potato-sector’, since potatoes are the exclusive crop. The next three WN. a8 ogasuyu Or ‘oca sector’, siwarasuyu or ‘barley sector’ and uywasuyu OT munities, for two reasons: the research upon which this study is based was —— : at the level of communities rather than of sectoral fallowing systems, and there are so Summer 1986 JOURNAL OF ETHNOBIOLOGY 175 communities in which sectoral fallowing systems have been identified but in which the precise number of such systems is unknown. Most communities had only one system, but some had several. Of the total of 51 communities for which some information was available, 13 lacked sufficient information to establish precisely the number of sectoral fallowing systems. For the 38 communities with sufficient data, there were 55 sectoral fallowing systems. If analysis were conducted on one system per community, the systems from multi-system communities might be underrepresented in the sample; however, if the analysis were based on the total sample of systems, then the systems from multi- system communities might be overrepresented. This problem was far less intractible, since the analysis could be conducted both ways, and in fact gave similar results. The difference between single-system and multiple-system communities are slight, and the systems in both types of communities are similar. These topics are discussed in greater detail in the next section. RESULTS Distribution.—As earlier reports have indicated, the sectoral fallowing systems are restricted to a band in the central Andes (Table 3). They are found in central and southern eru and in western Bolivia, ranging from 10° 20’.S to 18° 50’ S. Of the 51 cases, the elevations of the plots in the sectoral fallowing systems are available for 31; these range from 2400 to 4200 meters above sea level. The range of the average elevations of plots in the different sectoral fallowing systems is smaller, from 3000 to 4100 m. The rainfall, reported for 15 of the 51 cases, varies from 300 to 1270 mm/yr; excluding the highest and lowest figures, which may be suspect, reduces the range to 500 to 1100 mm/yr. At higher elevations, crop production is limited to small plots of potatoes, predominantly the bitter varieties which are processed by freeze-drying into a product called chuiio; the natural and cultivated grasslands in these areas support livestock. Fallowing is more extensive and is not managed communally. The fields at lower eleva- tions are often irrigated. They are usually cropped yearly, in a more intensive agricultural regime. Other crops such as maize are more common, though some potato cultivation is also found. Fallowing in these zones is also not managed on a sectoral basis. Aridity seems to limit sectoral fallowing systems in the south, where fields are more often irrigated or fallowed on a non-communal basis. The factors which restrict sectoral fallowing systems to the north are less clear. Heavier precipitation and warmer temperatures affect soils and vegetation, favoring the formation of sod, quite different from the bunch grasses which characterize the main zone of sectoral fallowing systems. The management of this sod is relatively little known; shorter fallowing periods might be needed, for instance, or the greater labor requirements to prepare fallow fields might not be as well suited to sectoral fallowing systems. The apparent absence of sectoral fallowing systems north of 10° S. may also be the result of underreporting. Fewer anthropologists and geographers have worked there. Much of the published research, Particularly for the northern Peruvian highlands, provides little detailed inf the spatial and temporal patterning of cropping, fallowing and grazing, although some sources clearly document the existence of non-sectoral fallowing systems (Brush 1977). The absence of these systems in Ecuador appears to be less in doubt. ae is distribution places the sectoral fallowing systems well within the limits of cultivation of the potato, the most important crop in these systems. However, it is not quite appropriate to equate the sectoral fallowing system with potato cultivation, since Andean peasants at latitutdes outside the range of sectoral fallowing systems in northern Peru (Brush 1977), Ecuador (Knapp 1984} and southern Bolivia (Rasnake 1982) grow Potatoes on plots that are fallowed. Other crops that are grown in sectoral fallowing systems, such as barley, quinoa, and broad beans, are also more widely distributed than — ee Vol. 6, No. 1 ORLOVE & GODOY 176 ee ee Oe 2 Dee dilteeeceneeiaamen me “PAWUTLUIT]D UIAq DALY SAN[LA IS2MO] FU YIM % S IYI PUB SaNyTeA saysty ayI YIM S9Sed Jo %S ay YDIYM Woy UOLLNdod dy) jo UBS JU) JO SYSISUOD Pau ULz ‘soaidap JO SYparpuNny pue ssarZap Ul prieys aiv sapnisuoy pur sapnaye’y; 000'T ste 0 00sZ0 00S 000'T 1ZS°0 009 00'e 000°C EBSE O0TE T9S 0869 +36 i a ayruend) Il 000°C e220 1ZS80 006 000°T 9920 00°01 002 000"? O16e OO8E OS8 ceed e9'sl ayruend Pay 000'T OST 0 00090 OST 000'I cee 0 00° 00'T 000'T 0008 OOLT 008 O99 Ov Ol anyeA UINUTUTYy 000° 000'T 49160 O0¢I 000°C 298°0 O0'ST 00'e1 000'P 860P OS6e OZT1 OO'LZ OZL'8I anyeA WINUWITXeY vTZ0 9¢7'0 Zes00 ssc 6. 0 6e10 9L'T LSC 9b6'0 187 188 (646 067 £07 uonPIAa prepuras ese’ 88S5'0 9808°0 t9°9 6S £v9'0 SB Ils T1I9T eSze Sse 614 CLIL ee vl queoul peur 000'T 00S'0 T8I8'0 009 00S'T 1e9'0 00° 00°S 000°¢ OO8E S198 STL OST ST Pl uUeIPaW Ler 16S'0 ce08'0 38029 eon’ Ov9'0 162 v7S TO9'T STLE Lese 8TZ OL TZ perl uBoW el Ol (ai al Ol 6 6 8 Z OT 8 9€ t £ Byep 3no YIM saseo jo saquinyy 8e Iv 6e 6£ Iv (67 (44 ev ad Ie ev SI 6v 8Y Biep YIM Saseo jo saquinyy (uray) (ua) (us) 1reyurey a pn S4SWON XCNIGM.L LVYALON YASMLLON YANTANL OLLVYTd SLOISWNN UYATTVd WAdOWD = SHSATZ =0WODATA ico 4 he nts ‘suzaish¢ SuLmol[v4 [01019ag fo suolsuauig awmos—'¢e FIAVL Summer 1986 JOURNAL OF ETHNOBIOLOGY 177 these systems. We wish to stress that fallowing is found in many parts of the Andes where sectoral fallowing systems are absent. Other Features. —A number of specialists on the Andes have emphasized the importance of the potato in agriculture in that region (Brush et al. 1981). In particular, potatoes are associated with agriculture at high elevations; in lower areas, maize is a major crop. Special frost-tolerant bitter varieties of potatoes are adapted to elevations too high for other crops. We were therefore surprised to find a slight negative relation? between the tuber index and the elevation of the community; we had expected a strong positive relation. In other words, potatoes account for a slightly higher proportion of the crops in communities at lower elevations. TUBINDX = 2.04 — 0.000413 ELVCOM (n =37, t35 =-3.81, corrected R2 =27.3%). No significant correlation was found between the tuber index and the elevation of the sectoral fallowing system. An examination of the cropping sequences in Appendix | sug- gests that the importance of grains, particularly barley and quinoa, at high elevations may account for this unexpected result; it should be remembered that communities manage sectoral fallowing systems jointly with other crop lands at higher and lower eleva- tions, so that cropping patterns within sectoral fallowing systems may be influenced by the availability of lands outside them. We also note that tuber indices decrease with distance from the equator: TUBINDX = 1.24 - 0.0461 LATD (n=40, t3g =-2.68, corrected R2 = 13.7%]. The Mann-Whitney U Test finds the difference between the median tuber indices for Peru and Bolivia, 0.6 and 0.417, respectively, to be significantly different from Zero at the 0.05 level. Another aspect of the sectoral fallowing systems remains to be discussed: the case of multiple systems. Of the 51 cases, sufficient data were available for 38 of them to determine the number of sectoral fallowing systems. There was one system in 25 cases, two systems in 10 cases, three systems in 2 cases, and there was also one case of a four- system community. In most communities with multiple systems, each system is located at a specific elevation. Camino et al. offer one of the best descriptions of such a com- munity, Cuyo-cuyo in the province of Sandia in the Peruvian department of Puno (1981). In the highest system, the Juki manda from 3800 to 4100 meters above sea level, a sector is first planted with special frost-resistant varieties of potatoes, called Juki in Quechua and papa amarga (‘bitter potato’) in Spanish, which are used for the prepara- tion of freeze-dried potatoes or chuyio; the sector then remains in fallow for five years. In the uray manda from 3200 to 3800 meters, in the first year potatoes are planted, in the second oca, olluco and mashua, in the third oca, ulluco and mashua, and in some Cases a few rows of broad beans, and in the fourth broad beans and some barley; the four crop years are followed by two years of fallow. The manda del anexo from 3200 to 32400 meters is located at some distance from the other two systems; it also follows a sequence of one year of potatoes, one of oca and four of fallow, although a number of other crops, including olluco, mashua and broad beans, may be included in both years. There is an association between the elevation of the community and the number of sectoral fallowing systems. The equation is: NUMSES = 4.36 - 0.000815 ELVCOM {n =35, t33 =-2.30, corrected R2 = 11.2%). A Mann-Whitney U Test does not find significant at an 0.05 level the difference of the median elevations for single-system and multiple-system communities, which are 3720 meters and 3500 meters, respectively. This relation reflects the fact, discussed later in more detail, that the sectoral fallowing systems tend to be located higher than the community of peasants who cultivate plots in them. Since most multiple systems Consist of systems more or less stacked vertically, the lower communities have more room in which to place these systems. The principal exceptions to these vertical arrangements are several multiple-system communities in the Lake Titicaca basin which, 178 ORLOVE & GODOY Vol. 6, No. 1 like Irpa Chico, utilize distinct systems in locations whose differences, such as soil fer- tility, do not depend on large variations in elevation, in addition, the relative flatness of the Lake Titicaca basin impedes the downward drainage of cold air and create signi- ficant differences in the risk of frost between flat areas and adjacent slopes which are less than 100 meters higher in elevation. We examined the different systems within multiple-system communities, and did not find significant correlations between elevation and the other features of such systems. In addition, Wilcoxon Signed-ranks Tests showed no significant differences between dimensions (number of sectors, number of crop years, tuber index, etc.}, systems in single- system communities and multiple-system communities. On the basis of these results, we conducted our statistical analysis by taking the mean values of the systems in the multiple-system communities. Since 65.8% of the communities for which data were available had only one system, we did not believe that we were reducing variability by using means. Adaptationist Hypotheses: Claims.—The adaptationist claims for the presence and distribution of sectoral fallowing systems have been stated on several occasions. Such claims tend to rest on rather loose notions of adaptation rather than on more rigorous extensions of biological concepts. Adaptations are seen as means by which populations draw resources from their environments while maintaining long-term productivity and by which they avoid the hazards present in their environments. One adaptationist discussion of sectoral fallowing systems states that fallowing, in combination with deposi- tion of dung by animals grazed on fallow sectors, “is essential to the regeneration of the land” (Guillet 1981:144). Another article claims that these systems permit “maintenance of soil fertility and removal of crop pathogens” (Brush and Guillet 1985:26). It is argued that Andean soils tend to be low in nutrients, that cultivation removes nutrients from the soils, and that fallowing must be maintained to prevent degradation of the soils. Fallowing and Soil Management.—These adaptationist claims address issues that are more complicated than they might appear to be. Agriculture can lead to decreasing nutrient levels in the soil, because crop plants directly consume nutrients and because tillage can increase microbial activity which can lower nutrient levels and impair the physical structure of soils. However, fallowing does not necessarily restore nutrients to the soil, although it may do so, if nitrogen-fixing plants are cultivated or occur natural- ly. Fallowing may also increase the organic content of the soil, if atmospheric carbon fixed in photosynthesis during fallow returns to the soil; however, crop plants derive the bulk of their carbon from carbon dioxide in the atmosphere rather than from the soil. Fallowing nonetheless appears to present several advantages in the case of Andean agriculture. First, by reducing the proportion of the time when the soil is stripped of its plant cover, fallowing protects soil against some of the dangers of nutrient leaching and erosion that accompany agriculture. Second, fallowing may allow some soil nutrients to be redistributed in a fashion that makes them more avilable to crop plants. The deep- rooted perennial grasses which grow during fallow years reach to lower levels of the soil than annual crop plants do; they may draw nutrients from these lower levels and, when reincorporated into the soil during the preparation of fallow fields for planting, increase the concentration of these nutrients in the upper layers of the soil. Third, there m4y be agronomic benefits from the increase in organic matter which occurs as plants grow during fallow and then are partially or fully reincorporated into the soil; in particular, the physical structure of the soil may improve, enhancing its water-holding potential and favoring root growth. Fourth, the use of fallow land for grazing may also improve the soil. There may be a net transfer of nutrients to the fallow fields as animals grazed A ee ee =O ene a rr ee rr ee, ee eee Summer 1986 JOURNAL OF ETHNOBIOLOGY 179 on both fallow fields and on permanent pasture (often located adjacent to and above the sectoral fallowing systems) deposit dung on the fields; it should be noted, however, that the dung is often not deposited evenly throughout the fields, so that grazing in some areas might still lead to a loss of nutrients. (It is also possible, though unlikely, that the presence of grazing animals damages soils because of compaction.) Finally, fallowing may enhance the moisture content of the soil; this is its principal advantage in other regions of the world. It is noteworthy that a textbook, Soils and Soil Fertility, places its discussion of fallowing in the chapter on water management, rather than in any of the chapters on nutrients (Thompson and Troeh 1978). It is quite possible, though, that the addition of fertilizer to soils may enhance soil fertility more than fallowing does. The relation between elevation and soil fertility is complex and mediated by several variables. Temperature is the most important of these, since it affects the rate of plant growth, the rate at which microorganisms nutrients such as nitrogen, phosphorus and sulfur found in organic matter in the soil into forms available for plants, and the tate of oxidation, which breaks down organic matter, and may therefore reduce the quality of the physical structure. At low temperatures, below about 5°C, biological activity in the soil is reduced to very low levels or stops entirely, so that the organic content of the soil does not increase and its physical structure does not improve (Price 1981:235). igher temperatures cause organic compounds to oxidize and break down, at temperatures above about 25°C, this loss overtakes the formation of organic matter, and soil fertility declines. The meximum rate of accumulation of organic matter occurs between these extremes, These rates, however, are influenced by soil moisture. Biological activity proceeds more slowly in very dry soils, reducing the decompositon of organic matter and the conversion of nutrients into forms available to plants. However, excessive moisture also reduces the oxidative decompositon of organic matter and the release of nutrients into the soil; it may also lead to the leaching of nutrients which are already present. As with temperature, there is an ideal range of moisture, somewhere on the wet rather than dry There is thus no universal direct relation between elevation and soil fertility. Price (1983:248} states “soils in the humid tropics . . . generally improve with elevation (up to a certain point),” and offers as an example the contrast between the heavily leached soils commonly found in tropical lowlands and the more fertile soils in adjacent and zones at intermediate altitudes. This difference is due primarily to the lower temperatures at higher elevations. In the case of the relatively high altitude regions where sectoral allowing systems are found, however, it seems that soil fertility is lower at higher elevations because of low temperatures which slow the growth of bunch grasses and which impede decomposition of organic matter. It thus seems that sectoral fallowing systems are generally beneficial to Andean soils, because they assure a certain length of fallow period; it might also be important for fallow periods to be longer at higher elevations. Fallowing and Pest Management.—The relation between fallowing and the reduction of losses due to nematode attacks is also somewhat unclear. The round cyst nematodes (Globodera rostochiensis and Globodera pallida) attack potato roots, reducing the abili- ty of the plant to absorb water and nutrients (Evans et al. 1975). Losses increase with greater nematode density, in some cases, virtually no potatoes can be harvested. At one Stage in the nematode’s life cycle, the female nematode produces eggs which she retains within her body. When the female dies, her cuticle becomes a dark hard layer, forming @ cyst inside which the eggs can survive for as long as 20 years (Hooker 1981:96). Nonetheless, nematode populations increase most rapidly when their populations are arge and when food supplies are abundant; limitation of their food supply may lead their 180 ORLOVE & GODOY Vol. 6, No. 1 populations to decline (op cit.:95). They are favored by moist cool soil conditions. Hooker states that ‘‘(w)hen nematode densities are high, rotation with potatoes grown once in five years is necessary to assure profitable potato yields. Resistant potato cultivars in rotation with susceptible cultivars and nonhosts considerably reduce the required length of rotation.” (op cit.:96). This source does not provide a definitive fallow period necessary to prevent damage, since Andean peasants might well be able to subsist on potato yields which Western experts would not consider “profitable,” and because many native cultivars may be resistant. The number of non-tuber years is a better measure of the effect of particular sectoral fallowing system on the nematodes than the number of fallow years, since the effect on nematodes of a year in which grain and legume crops are grown is more likely to resemble that of a fallow year than that of a tuber year. It seems likely that a non- sectoral form of fallowing, in which the fallowing cycles of individual plots are not coordinated, would be less effective than sectoral forms, since plots adjacent to an area where nematode-infested potatoes remain at risk for several years. It would be more common in non-sectoral systems to find the planting of potatoes in plots adjacent to areas where potatoes had been planted in the previous year; in such plot, there is a risk that nematodes remaining from the previous year would disperse from the adjacent area to attack the new potato crop. Other Claims.—In addition, sectoral fallowing systems have been considered to be adaptations in other ways. By grouping together a large number of plots that will be cultivated in a given year, they reduce the total perimeter of these plots which border on fallow land, thus reducing the frequency with which domesticated animals enter the plots and damage crops. This feature reduces the amount of labor which a household must allocate to the supervision of its animals (Mayer 1979:71). Skar’s study (1982) of the Peruvian community of Matapuquio, in the province of Andahuaylas, department of Apurimac, shows the importance of limiting livestock damage. When an ani eaks into a plot in a sector currently under cultivation, the owner of the animal pays compensation to the owner of the plot; if the same animal were to damage crops in a plot that had been planted in a sector which was currently in the fallow portion of the cycle, no such payment is required. As a member of the community stated, ‘If you Sow the other fields [in fallow sectors] then the animals will just come and eat it and no one is responsible for it.” (1982:82) In addition, herders take particular care to keep animals away from cultivated sectors. This juxtaposition of plots also permits the more experienced and knowledgeable members of the community to have a large role in making decisions about the timing of agricultural activities (Custred and Orlove 1974). If the plots were scattered, it would be more difficult for all members of community to obtain their advice. Adaptationist Hypotheses: Results.—These adaptationist claims are generally supported by the fact that sectoral fallowing systems are located at higher elevations than areas of continuous cropping, where the maintenance of soil fertility is more problematic, and that they are associated with the cultivation of tubers, the crops most sensitive to round cyst nematodes. A more detailed examination of these systems does not support these claims unequivocally, however. The adaptationist claims refer to related but distinct aspects of the system. The fertility argument generally supposted that a significant proportion of the land will lie fallow at any moment, and usually does not address the question of how many years any one sector must remain without being cultivated. The pathogen argument, by contrast, states that each sector must remain out of tuber cultivation for a certain number of years, and usually does not address the question of what proportion of the time the sectors will remain without tuber cultivation. Summer 1986 JOURNAL OF ETHNOBIOLOGY 181 The typical features of sectoral fallowing systems offer some support of the general claims. The mean fallow ratio in the 42 cases for which information is available is 0.640, a figure which seems high enough to permit the maintenance of the nutrient levels and the physical structure of the soil. One recent source makes the adaptationist claim that in sectoral fallowing systems “more land is left fallow than is being cultivated at any one time.” (Guillet 1981:143-144). A use of the Wilcoxon Signed-ranks Test permits a non-parametric test of the hypothesis that mean fallow ratios are greater than or equal to 0.5; parametric tests, which require fallow ratios to be distributed normally, are less appropriate. This hypothesis is supported at the p<.005 level. However, 9 out of 42 cases have fallow ratios of 0.5 or less. Although these cases are part of a population whose mean is significantly greater than 0.5, they do not conform to Guillet’s claim. In the 39 cases for which information is available, the average number of non-tuber years is 6.70 and 35 or 89.7% have at least 4 non-tuber years, probably enough to reduce pathogen populations (Hooker 1981:96). The hypothesis that the mean number of non-tuber years is greater than 4 is supported by the Wilcoxon Signed-ranks Test at the p <.001 level; nonetheless, 4 out of 39 cases or 10.3% had fewer than 4 non-tuber years. These claims would be difficult to test definitively, since levels of soil nutrients and pathogen populations vary continuously, without sharp breaks at particular fallow or number of non-tuber years. The data do seem, though, both to support the claims and to indicate the existence of cases which challenge them. A second difficulty in definitively testing these claims lies in the fact that the same level of nutrient increase might be reached by different fallow ratios in different communities; a similar problem exists for pathogen protection. This issue could be addressed by examining variation by elevation. Since the benefits of fallowing accrue more slowly with an increase of eleva- tion, because the colder temperatures at higher elevations reduce plant growth and microorganism activity, the fertility claim would suggest that the fallow ratio would increase the elevation. These relations do not hold; there is no significant relation between this feature of sectoral fallowing systems and elevation, whether measured by elevation of community (n = 37) or, more accurately but with a smaller sample, by average elevation of plots in the sectoral fallowing systems (n = 26). One might expect that more Southerly communities, where average temperatures are lower and frost. are more common, to require higher fallow ratios in order to achieve levels of soil fertility similar to those at equivalent elevations further north. In fact, the fallow ratio decreases with distance from the equator: FL O = 0.979 - 0.0236 LATD (n=40, t3g =-2.12, corrected Ry = 8.2%}. Since Bolivia lies generally further south than Peru, this difference could also be associated with the nation in which communities are located. The Mann-Whitney U Test comes Close to finding a significant difference between the median fallow ratios, 0.571 in Bolivia and 0.652 in Peru, it assigns a p-value of 0.052. The lower fallow ratios in Bolivia might in part reflect the dominance of the Lake Titicaca region, where some fallow ratios are low, in the sample of communities from that country; the overall relation of fallow ratio with latitude is more puzzling. ra , Similarly, it should be noted that the number of crop years is positively associated with latitude: 2 ‘ ROPYR = 0.006 + 0.0186 LATD (n =42, tap =2.80, corrected R= 14.3 70), and that the differences for the median values for Bolivia and Peru, 3.5 and 2.5 respectively, is Significant at the level of p< 0.005. There is no significant relation between number of sectors and latitude or nation. Noting the following relations: FLRATIO = (NUMSECTS - CROPYR}/NUMSECTS CROPYR = NUMSECTS - (NUMSECT S*FLRATIO), . it is reasonable that one has a negative and the other a positive relation with latitude. 182 ORLOVE & GODOY Vol. 6, No. 1 It seems even more difficult to construct reasonable hypotheses for the influence of elevation or latitude on the nematode-reducing effects of fallowing. We did note an association between the number of non-tuber years and the elevation of the sectoral fallowing systems: NOTUBYR = -8.09 + 0.00394 ELVSES (n =25, ta3 =2.05, corrected Rg = 11.7%). The relationof the number of non-tuber years and the elevation of the community was weaker. Contrarily, we found a strong relation between non-tuber ratio and community elevation, though not with system elevation: NOTBRAT = 0.422 + 0.000109 ELVCOM (n = 35, t33 =2.63, corrected R2 = 14.8%). These relations generally indicate an increase in non-tuber years and ratios with elevation. Agricultural Intensification and Sectoral Fallowing Systems. —This examination of the adaptationist hypotheses led to another set of considerations. The adaptationist position might have been more strongly supported in the past than it is at present. Because of the extensive changes in highland Andean agriculture, it seems possible that some sectoral fallowing systems have been altered in certain ways. In this view, these weaknesses in the adaptationist position could be explained by the presence of forces which have shifted sectoral fallowing systems from a greater to a lesser degree of adap- tation. In other words, it might be possible to accept the adaptationist position even though there are some cases which do not support it; to do so, one would need to find some means to distinguish between the systems for which it would hold and those for which it would not. One promising argument could be called the intensificationist position, since the reduction of fallow ratios resembles what Boserup (1965) calls agricultural intensification, and parallels her classification of agricultural systems. Population pressure and commercialization of agriculture have been identified as causes which might lead to more intensive land use within agricultural systems in general and sectoral fallowing systems in particular, reducing fallow ratios and the number of fallow years. Three sources which diachronically examine sectoral fallowing systems support these arguments. Godoy (1984) describes sectoral fallowing systems in the northern portion of the department of Potosi, Bolivia, in which the total number of sectors has been maintained, but each sector is cultivated for an increasing number of years an left in fallow for a correspondingly shorter period. He attributes this decline in fallow ratios, as well as the intensification in lands not cultivated by sectoral fallowing systems, to population increase; he also suggests that taxation placed additional demands on the yields of a finite land base. Mayer’s (1979) studies of the Mantaro Valley in the depart- ment of Junin show not only that fallow ratios declined, but also that strong pressures for commercialization led to a breakdown of the coordination of fallowing and planting by different households, so that sectoral fallowing systems no longer exist. The previously- ioned study by Carter and Mamani offers a detailed description of changes in the sectoral fallowing systems or aynuga in one Bolivian community. In some cases, the number of years in the cycle is retained, but households cultivate crops in years that were formerly allocated to fallowing, much as some residents of Amantani do. To follow the earlier terminology, the number of sectors in such cases remains unchanged, but a fallow year is replaced by a crop year. In other instances, an entire sector is taken out of the cycle and converted into private household plots, known locally as sayana, SO that by one. In either case, the fallow ratio declines; in some instances, when the number of fallow years falls to zero, the lands are no longer considered to be sectoral fallowing systems. Anecdotal information also provides evidence for this decrease in fallow ratios: an entire sector may be removed from the system by appropriation by a landlord (Orlove 1976) or by other communities (Godoy 1984}. Since the number of sectors is reduced, the number of years in the cycle must decline as well. It seems that the number of croP Summer 1986 JOURNAL OF ETHNOBIOLOGY 183 years remains the same and the number of fallow years is reduced, lowering the fallow ratio. These pressures are documented by the sources already mentioned but not by the set of 51 cases. Information is scarce on the population variables, for example, none of the 51 cases have information on population density at the community level, and population densities at the next higher administrative level, of the district in Peru and canton in Bolivia, would be a poor substitute. Few cases had information on such possible measures of commercialization as distance to markets or presence of metal roofs. Fallow ratios and the number of fallow years also are not significantly related to other, less reliable measures of commercialization, such as the distance to the nearest town or official recognition of community. The 51 cases do offer some indirect support. There is a significant positive relation between the number of sectors and fallow ratio in sectoral fallowing systems. FLRATIO = 0.426 + 0.0271 NUMSECTS (n = 42, t4g = 4.04, corrected R2 = 27.2%). The fact that there are few systems with a high number of sectors and a low fallow ratio Suggest the generality of the following sequence: a system might start with a large number of sectors and a high fallow ratio. Demographic and commercial pressure might lead to a decrease in the number of fallow years and a corresponding increase in the number of crop years, leading to a decline in fallow ratio. After a certain point, the entire number of sectors might be halved and the area of each sector doubled (Mayer 1979:67-70). This possibility, though, is quite speculative. It should be stressed that the previously- mentioned areas where fallowing has declined are subject to high rates of population increase, increased sale of crops, or both of these influences. Fallowing may well be main- tained in communities where populations have remained relatively steady, due to lower fertility, to higher mortality, or to higher out-migration, and where commercialization of agriculture has been less, due to acceptance by peasants of low cash incomes or to the earning of income through migratory wage labor. This uneven support of the adaptationist position and the difficulty of directly testing the intensificationist position still leaves open the possibility that demographic or €conomic pressures have altered some features of the sectoral fallowing systems. It would therefore be useful to examine features of these systems which are less prone to these Pressures, to search for regularities in them, and then to try to explain them. There are two such features: the relative elevations of the communities and the systems, and the sequencing of crops. It would seem fairly likely that the former would change less than other features such as fallow ratios. A community might expand the sectors somewhat, add crop years and delete fallow years, or might lose sectors to haciendas, but it is less likely that the community or the sectoral fallowing system would be relocated entirely. The latter feature also displays certain regularities, for reasons that are less immediately evident. In the case of the former, two variables, system elevation and community eleva- tion, were compared. These variables are closely related: ELVSFS = 2117 + 0.464 ELVCOM (n =27, t25 =3.27, corrected R? = 27.2%). The System elevation is generally greater than community elevation. The mean difference between the system elevation and the community elevation is 229 meters; the Mann- Whitney U Test supports the hypothesis that this mean is greater than zero at the P<.01 level. The mean difference between system elevation and community elevation is unequal to the difference between the mean system elevation and the mean community elevation because of the different sample sizes for the two variables. There are several teasons for locations of communities at lower elevations: the importance of proximity to irrigated lands lying below the sectoral fallowing systems, the importance of deep tiver valleys in shaping transportation networks, the need for access to water. In some Cases, this location also reflects the sixteenth-century Spanish policy of forcibly 184 ORLOVE & GODOY Vol. 6, No. 1 resettling Andean populations in nucleated settlements called reducciones, which were often at lower elevations than the fields. The crop sequences also show strong regularities. Commercial crops such as onions, cabbage and carrots are absent; such crops tend to be grown in lower zones, where warmer temperatures and more abundant irrigation water encourage investment in such crops. Conversely, subsistence crops dominate. More specifically, root crops have a position of unique importance. In 7 of 41 cases, only tubers are cultivated. In the remaining 34 cases, potatoes are always the first crop grown. In 11 cases, information is insufficient to determine precise crop sequenes. In the cases where there are several years of exclusive or mixed tuber cultivation, these years occur before the others in the cycle. To test the hypothesis that this pattern did not occur by chance, one could take the number of tuber years and the number of non-tuber crop years for each community and assume that the communities selected the order of tuber and non-tuber years at random. There is 4 probability of less than .001 that all 34 communities would place all tuber years before all non-tuber crop years. This sequence of crops reflects the great importance of potatoes in Andean diets. They grow well over a wide range of environmental conditions (Gade 1975). Potatoes offer high yields, whether measured by unit area or by labor (Painter 1981; Brush * al. 1981). they account for a large proportion of the calories and protein which local peasants consume (Orlove 1986). In addition, they store well. It could be argued along adaptationist lines that Andean cultivators plant potatoes in the beginning of the cycle, because the levels of nutrients in the soil are highest at that time. If the concern Kae solely to provide potato crops with high levels of nutrients, though, they might be plante after, rather than before, nitrogen-fixing crops, such as the native tarwi (Lupinus mutabilis) and the introduced broad bean. ; d A second adaptationist perspective suggests that tubers are planted before grains a> legumes because the harvesting of tubers destroys the pattern of furrows in the field, while the harvest of the other crops does not (Orlove 1977a). The purported advantage of this arrangement is that the furrows facilitate drainage of water in the rainy season, thus reducing erosion and permitting the earlier return of wild grasses. The ae for this argument is anecdotal, and in any case this effect would be unlikely to hold wit significance in the wide variety of degrees of slope and of soil structure in the regions where sectoral fallowing systems are found. Culturalist Hypotheses. —Where adaptationists would view the needs and preferences of peasants as unproblematic consequences of environmental, technological - f demographic causes, culturalists would seek to explore the origins and ramifications o these needs and preferences. In the instance of the priority given to the potato, a culturalist argument holds some plausibility. Andean peasants, it could be said, plant potatoes ae because it seems right to them to do so. In this view, potatoes have a value whic transcends their agronomic or nutritional significance. This culturalist argument © supported by the importance of potato-planting ceremonies in some regions ee sectoral fallowing systems (Barstow 1979). Although the ritual emphasis on maize o the Andes is widely acknowledged to be greater than on potatoes (Murra 1960), potatoe e hold a greater importance than the other crops with which they alternate in the _— fallowing system. No other crop would be as appropriate for the rituals that 1m man cases form an integral part of the preparation of a sector after it has completed its fallow years. There are other cases where Andean cultural patterns structure fallowing cycles. As previously mentioned, in Paracaos, potatoes are planted in the first crop year, sao as turno (Degregori and Golte 1973); the second year, known as Jana or yana, the ave tubers are grown, and a smaller proportion of the plots are planted. The term yan@ Summer 1986 JOURNAL OF ETHNOBIOLOGY 185 great significance in the Andes; it refers to the second, and lesser, or two elements which form a complementary duality (Platt 1978). This correspondence might suggest that a certain number of sectors represents the cultural ideal; one might expect an even number of sectors, granted the importance of dualism in Andean culture. Although the link between cultural pattern and observable behavior need not be direct, this view nonetheless is not supported by the empirical evidence. The proportion of even values of the number of sectors, the number of crop years and the number of fallow years is not significantly different from the proportion of odd values, and in some areas where Andean traditions are held to be particularly strong, there are odd numbers of sectors; the Jukumani in northern Potosi; with a seven- sector system, are a case in point. The culturalist perspective may account for particular details, but its greatest Significance is its accounting for relative smoothness with which sectoral fallowing systems operate. This view resembles Bourdiew’s discussion of culture as habitus, which he defines as “systems of durable, transposable dispositions” (1977:72; Ortner 1984). Three features of sectoral fallowing system which are common in the Andes are boundedness, universal participation and rotational turns. The existence of well-bounded communities, all of whose members participate in many activities, is reflected in a number of spheres: other communal pattems of resource management, such as pasture and irrigation water (Brush and Guillet 1985}; the importance of assemblies in which all resident households and no outsiders have a vote (Alberti and Mayer 1974); collective work projects such as road and school construction, in which all resident households must contribute equally in labor and in materials (Brush and Guillet 1985); and the well-known ritual complex of the annual cycle of fiestas (Isbell 1978; Rasnake 1982). Many aspects of Andean organization show similar patterns of the division of a whole into equal parts, and the assignment of these parts on a rotational basis (Albé 1974). Common examples include the rotation of political and ritual offices on a yearly basis among the sections of a community (Rasnake 1982) and the rotation of political offices among villages in multi-community systems (Alb6 1972). The management of irrigation water, in which water rights are assigned sequentially to a number of component households, is another example of this pattern of rotational turns (Mitchell 1976). This instance also reflects the boundedness and universal participation in the limitations on use of water by outsiders and in the ritualized annual cleaning of irrigation canals, in which all households Participate. An unusual example of rotational turns may be found in bands of musicians who play panpipes at fiestas around Lake Titicaca; Buechler (1980:55) reports that these Political Economic Hypotheses.—Writers who emphasize political economy view these forms of organization somewhat differently. They argue that the tight internal gover- nance and great emphasis on ritual in Andean communities are to be understood p as responses to pressures from external elites which seek to control peasant land and labor (Wolf 1982:145-149). In this sense, the sectoral fallowing system serves to mark arge extensions of land as being equally under the protection of a set of households. In plots managed by households, it would be more difficult to protect them against encroachment from others. Although only 7 sources report the extent of sectoral fallowing systems, the average number of households, about 350, may well represent a balance between the advantages of small size (easy management) and large size 186 ORLOVE & GODOY Vol. 6, No. 1 (protection against outsiders}. Similarly, it could be argued that sectoral fallowing systems are well-suited to regions like the Andes, where seasonal migration encourages peasants to manage their plots in a highly predictable manner. In fact, the areas where sectoral fallowing systems are now found correspond loosely to the provinces from which in the colonial period mita workers were sent to Huancavelica and Potosi, though this spatial overlap is far from precise. Nonetheless, this emphasis on political economy is not directly supported by the case material. Many such systems are found in areas where there is relatively little pressure from outsiders on peasant lands and where there is relatively little outmigration. Though political economy in general has influenced Andean agriculture strongly (Orlove 1977b), sectoral fallowing systems do not seem to provide evidence for this view. SUMMARY There are certain regularities among sectoral fallowing systems. They are found within specific altitudinal and latitudinal limits. They may serve to maintain soil fertility, to reduce nematode damage to potato crops, and to facilitate a mixed agropastoral economy. Specific diachronic information from selected localities indicates that fallowing decreases with population increase and with commercialization, since such demographic and economic data are available for only a few communities, it was not possible to examine these relations by a statistical analysis of a large number of sectoral fallowing systems. The sectoral fallowing systems also correspond to other social and cultural patterns in the Andes, this link may help explain their relatively smooth functioning. ACKNOWLEDGMENTS The authors would like to thank the many colleagues who gave generously of their time in filling out questionnaires. We also wish to acknowledge the useful comments which we received from several colleagues, particularly Stephen Brush, Michael Singer and Karl Zimmerer. We greatly appreciate Neil Willits’ valuable advice on the use 0 several statistical tests, Thomas Turrentine’s diligent efforts in tracking down a number of sources which were difficult to obtain and in coding these sources and the question naires, Aileen Tsai’s prompt entry of the data, Rod Thompson’s helpful assistance in guiding a notive through the complexities of packaged statistical programs, and Sharon Lynch’s patient and careful transformation of numerous drafts into a final manuscript. Maria Dudzak prepared the map. The authors are responsible for any errors of fact OF interpretation which may remain in the article. NOTES L households whose lands are continguous and whose members participate in institutionalized economic, political and religious activities. The term “community” is a common and convenient label for such groups. It is important to mention, though not to dwell upon, the major debates amon Andean speciali h ing and appropriate use of this term: to what extent do communitics reflect Andean cultural principles and patterns of social organization; to what extent are acne wa nity a L I 4 rg : seh 3 p ial and national states; to what extent are patterns of community organization a spontaneous outgrowth of adaptation to cisco , a SE a vat } ee ya eee os ] iq] ization (Gt 1978; Guillet and Brush 1985; Murra 1984, Salomon 1982; Yambert 1980). Summer 1986 JOURNAL OF ETHNOBIOLOGY 187 The transcription of Quechua and Aymara terms is not fully standardized. We have adopted the spelling of the source which we cite most frequently. The same term also appears in published sources as aynoca and, less frequently, aynogqa. 3We do not offer exact p values for the acceptance or rejection of the relations, since a precise t-test is inappropriate. The variations are not normally distributed, and the necessary condition for equality of variance is not met. Nonetheless, we have included regressions for which t-ratios are greater than 2.0, since such levels seem to indicate some importance. igin f + 1 fall 4We have deliberately avoided a discussion of the ori i they are so difficult to detect in both the archaeological record and historical sources. Their conformity with Andean patterns of organization suggests an Andean origin, although there are similar patterns of organization in Spain (Fernandez 1981, Espinoza 1981). f wer a 7 8 { Baer i] : \ ri \ BRAZIL Ne pie J } / YOY Yg> Pasco 4 re 7% we.) es PERU i 3. \~ (eds f \ Junin “di / j a \, ) fi * ' ips _ Nes ~*~ Lima Sy f ed he Z a 6: 7 / hd ‘ \\ Cuzco \ J ‘\ Huancavelica yo N yeh all ack on af / % i i er a8 pe ‘io 4, v La Paz f > ‘ A rimac } 42s? a Te ate i, 137 ah Puno j ( ) ra st. St, L IVIA L~>r-k “41¢ 30+ BO \ Ayacucho fs bas : = ‘ is. ae wt \ SD Tie age » r 6 et ~ om 2 Titicaca £ ka “~ “Dt es SO ‘i Ry, al fi. g. 8 D O 7 § x ' oP i ae ) aa A ae ee ad hy m= International Boundaries Oh eo —-,* ™ r . ~——- Departmental Boundaries 3 ' Nec iat a “48 VS 0 100 200 4 : a Miles 2 CHILE a ; “sue Potosi 4 76 72 68 ee eeenaneeteeetemenesn n n FIG. 1. Locations of communities studied. See text and appendix 1 188 ORLOVE & GODOY APPENDIX 1.—Communities with sectoral fallowing systems. aS OW Elevation Total No. Average Name of Depart- Country Lati- Longi- of Com- of SFS Elevation Community ment tude tude munity perCom-_ of SFS (meters munity (meters) Paucartambo Pasco Pem 10° 769 3044 1 N/A (Andrews 1963} 40’ 20 Paucarcolla Puno Perm 15° 70° 3830 1 N/A (Bourricaud 40’ 10’ 1967) Santa Lucia Lima Perm 11° 76° 3200 1 3425 de Pacaraos 15‘ 40’ (Degregori & Golte 1973) Marcapata Cuzco Perm 13° 70° 3100 3 3300 (Yamamoto 30’ sf 1982) Same 3850 Same 4100 Miraflores) Lima Peru 12° 750 3000 1 3500 Laraos (Mayer 20’ 50’ and Fonseca 1979) Kaata La Paz Bolivia 15° 69° 3500 l 4100 (Bastian 1978) 0’ Q’ Jukumani Potosi Bolivia 18° 66° N/A 1 3750 (Godoy 1982) 35’ 25’ Jesisde LaPaz Bolivia 16° 68° 3800 l 3800 Machaca 35’ 5! (Vellard 1963) Irpa Chico La Paz Bolivia 16° 68° 3800 2 N/A Carter & 40’ 20’ ( r Mamani 1982) Vol. 6, No. 1 a Summer 1986 JOURNAL OF ETHNOBIOLOGY 189 —— : 5 of Total %of Crops Crops Crops Crops Crops ears No.of Years in in in in in Fallow! Culti- Sectors Fallow Ist 2nd 3rd 4th Sth vated Year Year Year Year Year 4 1 5 80% Potatoes Fallow Fallow Fallow Fallow N/A NIA N/A N/A N/A N/A N/A N/A N/A 8 2 10 80% Potatoes Oca Fallow Fallow Fallow Mashua Olluco 3 2 5 60% Potatoes Oca Fallow Fallow Fallow Olluco Mashua 3 e 5 60% Potatoes Oca Fallow Fallow Fallow Olluco Mashua l 1 2 50% Bitter Fallow _— Bitter ‘Fallow _ Bitter Potatoes Potatoes Potatoes 7 3 10 70% Potatoes Oca Barley Fallow Fallow Mashua Olluco 5 3 8 63% Potatoes Oca Barley Fallow Fallow 3 4 7 43% Potatoes Oca Wheat Barley Fallow 3 4 7 43% Potatoes Quinoa Cafiihua Barley Fallow Canihua 3 3 6 50% Potatoes Quinoa Barley Fallow Fallow 1 : In some instances, some plats will be cultivated in the first fallow years. Note discussion in text. 190 ORLOVE & GODOY Vol. 6, No. 1 APPENDIX 1. Communities with sectoral fallowing systems. (continued) 9S ow Elevation Total No. Average Name of | Depart- Country Lati- Longi- ofCom- of SFS Elevation Community ment tude tude munit er Com- of SFS (meters) munity (meters) NIA Santa Barbara Cuzco Peru 14° 71° NIA 1 N/A (Orlove & 1” 10 Custred 1974) Quehue Cuzco Peru 14° 71° 3675 2 N/A (Orlove & Custred 1974) N/A Pata Anza Cuzco Peru 14° 71° NIA 1 N/A (Orlove & Custred) Pampa Marca Cuzco Perm 14° 71° 3829 1 N/A (Orlove & 5 30’ Custred) Mosocllacta Cuzco Peru 14° 71° 3815 1 N/A (Orlove & 10’ 30’ Custred) Laymis Potosi Bolivia 18° 66° NIA 1 3900 (Harris 1982) 30’ 15’ San Pedrode Lima Peru 11° 76° 3168 1 3000 Casta (Gelles 45’ 35’ 85 Question.) Rinconada Puno Perm 15° 69° 3820 2 4050 (Chaquilla 85 45’ 55’ Question. } Same 3820 Summer 1986 No. of No. of Years Years Fallow! Culti- d vate 3 3 10 3 8 1 7 q 6 3 3 3 : 3 NIA NIA 2 y) 3-5 2 5 y) JOURNAL OF ETHNOBIOLOGY 19] Total % of Crops Crops Crops Crops Crops No. of Years in in in in in Sectors Fallow Ist 2nd 3rd 4th 5th Year Year Year Year Year 6 50% Potatoes Habas Barley Fallow fallow Ocas Arvejas Mashua Quinoa 13 76% Potatoes Olluco Barley Fallow Fallow Oca Habas Habas Quinoa Cafiihua 9 88% Potatoes Fallow Fallow Fallow Fallow 9 77% Potatoes __ Barley Fallow Fallow Fallow 2 67% Potatoes Oca Barley Fallow Fallow Olluco Habas Habas Quinoa Cafiihua 6 50% Potatoes Barley Barley Fallow Fallow Olluco Olluco Quinoa Quinoa F 57% Potatoes Barley Habas_ Fallow Fallow N/A N/A N/A N/A N/A N/A N/A 4 50% Potatoes Arvejas Fallow Fallow Potatoes Habas 8 63% Potatoes Potatoes Oca Fallow Fallow Olluco 7 71% Potatoes Barley Fallow Fallow Fallow ty some instances, some plats will be cultivated in the first fallow years. Note discussio M in text. APPENDIX 1. Communities with sectoral fallowing systems. (continued) ORLOVE & GODOY ss °W = Elevation Total No. Name of Depart- Country Lati- Longi- of Com- of SFS Community ment tude tude munity per Com- (meters) munity Matapuquio Apu- Peru 13° Li 3000 1 (Skar 1982) — rimac 35’ 10’ Ch’egec Cuzco Peru 13° 71° 3300 3 (Webster 85 20’ 10’ Question. } (Pseudonym) Same Same Usi (McCorkle Cuzco Peru 13° 75° 3700 2 85 Question.} 45’ 30’ (Pseudonym) Same Chinchero Cuzco Peru 13° 72° N/A 1 (Nijfiez del 25! 0’ Prado 1949) Caraybamba Apu- Peru 14° 73° 3140 1 (Fujii/Tomoeda rimac 10’ x 1981) Chanajari Puno Peru 20° 30° 3740 1 (Collins 85 Question.} Huallay Grande Huan- Perm 12° 45° 3600 1 (Bradby 85 Cave- — Question.} lica Chiripa La Paz Bolivia 16° 69° 3835 1 (Browman 85 30’ 0’ Question.} Alccavitoria Cuzco Perm 149 72° N/A 1 (Custred 1974) 30’ 20’ Vol. 6, No. 1 Average Elevation of SFS (meters) 3750 3950 3910 3800 Summer 1986 JOURNAL OF ETHNOBIOLOGY 193 No. of No.of Total % of Crops Crops Crops Crops Crops Years Years No.of Years in in in in in Fallow! Culti- Sectors Fallow Ist Ind 3rd 4th 5th vated Year Year Year Year Year 5-8 4 12 66% Potatoes Oca Habas Wheat Fallow Olluco a 1 10 90% Potatoes Fallow Fallow Fallow Fallow 5 | 6 83% Potatoes Fallow Fallow Fallow Fallow 2 1 3 66% Maize Fallow Fallow Maize Fallow 3 2 5 60% Potatoes Barley Fallow Fallow Fallow Habas 3 2 5 60% Potatoes Barley Fallow Fallow Fallow Habas 4 3 vi 57% Potatoes Olluco Barley Fallow Fallow Ocu Habas 13 2 Ihe 87% Potatoes Volunteer Fallow Fallow Fallow Olluco _— Potatoes Ocas - + 11 64% Potatoes Ollucos Habas_ Barley Fallow Ocas 6 I 7 86% Potatoes Fallow Fallow Fallow Fallow 4 3 N/A N/A _ Potatoes Barley Habas’ Fallow Fallow Wheat Tarni Quinoa Canihua NIA N/A N/A N/A N/A N/A N/A N/A N/A lin some instances, some plats will be cultivated in the first fallow years. Note discussion in text. ORLOVE & GODOY Vol. 6, No. 1 APPENDIX 1. Communities with sectoral fallowing systems. (continued) 5 OW ~=Elevation Total No. Average Name of Depart- Country Lati- Longi- of Com- of SFS_ Elevation Community ment tude tude munity perCom- of SFS (meters) munity (meters) Ambana La Paz Bolivia 15° 69° 3800 2 3700 (Aguirre a7 0’ 1980) Same 4500 Rumipata) Cuzco Peu N/A N/A — 3100+ 1 3500 (Pseudonym) (Guillet 1979) San Jose’ Puno Peru 149 70° 3800 1 N/A (Jacobsen 50’ 5! 1982) laulli Puno Peru 149 70° 3800 1 N/A (Jacobsen 50’ 5! 1982} Soras Aya- Peru 140 730 3433 l 3400 (Turpaud & ~~ cucho 5! 35! Murrugarra 1966) Tangor Pasco Peru 10° = 760 2700 l N/A (Mayer 1979) 25’ 25 Yacan Pasco Peru 10° 769 =. 3667 1 3780 (Fonseca 30’ 30’ 1972) Yancao Lima Peru 11° 770 NIA 2 N/A (Galdo & Q’ 0’ Martinez 1966) N/A (Matos Mar 40’ 35' 1957) Summer 1986 JOURNAL OF ETHNOBIOLOGY 195 No. of No. of Total % of Crops Crops Crops Crops Crops Years Years No.of Years in in in in in Fallow! Culti- Sectors Fallow Ist 2nd 3rd 4th 5th vated Year Year Year Year Year 5 o 10 50% Potatoes Ocas Habas’ Barley Oats Peas 10 a 13 77% Potatoes Ocas Oats Fallow Fallow 4 3 7 57% N/A N/A N/A Fallow Fallow 2 4 6 33% Potatoes __ Barley Repeat Opt. Fallow Oats Repeat Quinoa Ocas Habas 5 NIA 6 83% N/A Fallow Fallow‘ Fallow’ Fallow 4 3 Z 57% Potatoes cas Barley Opt. Fallow Ollucos Repeat Mashua Habas 7 5 12 58% Potatoes Ocas Trgo NIA NIA 9 3 8 63% Potatoes Oca Barley Fallow Fallow Olluco Habas Mashua Quinoa 4 3 7 57% Potatoes Ollucos Barley Fallow Fallow Ocas 2 1 3 66% Fallow Fallow. Bitter Fallow Potatoes otatoes N/A N/A 6 N/A N/A N/A N/A N/A N/A l , In some instances, some plats will be cultivated in the first fallow years. Note discussion in text. 196 ORLOVE & GODOY Vol. 6, No. 1 APPENDIX 1. Communities with sectoral fallowing systems. (continued) 9S WW Elevation Total No. Average Name of Depart- Country Lati- Longi- of Com- of SFS_ Elevation Community ment tude tude munity perCom-_ of SFS (meters) munity (meters) Tukiwasi Cuzco Peru N/A N/A 3100 1 3304 (Pseudonym) (Guillet 1974) Moccoraise Cuzco Perm 13° 71° 3500 1 N/A (Gade 1975) ao 30’ Macha Potosi Bolivia 18° 66° N/A l 3850 (Platt 1982) 30’ 20’ Yucay. Cuzco Perm 13° 72° 2930 1 N/A (Fiorante 0’ 0’ Molinie 1982! Mantaro Junin Peru 11° 73° N/A 2 4100 Valley 45’ 20’ eastside (Mayer 1979) 2nd 3750 Mayer Surimana Cuzco Peru 149 710 3586 2 3700 (Orlove 30’ 95’ 1976) PPE N/A Pomacanchi Cuzco Peru 14° 7 3700 1 3825 (Gade 85 0’ 30’ Question.} Ichu Puno Peru 16° 69° 3830 1 NIA (Bourricaud Q’ 50’ 1967) Huayapampa Lima Perm 11° 760 3047 2 3275 1968) Summer 1986 JOURNAL OF ETHNOBIOLOGY 197 No. of No. of Total % of Crops Crops Crops Crops Crops Years Years No.of Years in in in in in Fallow! Culti- Sectors Fallow Ist 2nd 3rd 4th 5th vated Year Year Year Year Year ] 1 2 50% Potatoes Fallow Repeat Fallow Repeat Ocas Quinoa Habas , 4 11 64% Potatoes N/A N/A N/A Fallow 7 4 11 64% Potatoes Oca Quinoa Barley Fallow N/A N/A 9 N/A Potatoes N/A N/A N/A N/A 10 Z 12 83% Potatoes Barley Fallow’ Fallow Fallow ats 4 3 7 57% Potatoes Oca Wheat Opt Fallow Ollucos Barley Repeat S 3 8 63% Potatoes Barley Habas_ Fallow Fallow r 2 9 77% Potatoes Barley Fallow Fallow Fallow 2 3 5 40% Potatoes Barley Habas_ Fallow Fallow 2 3 5 40% Potatoes Barley Quinoa Fallow Fallow Oats 3 a 6 50% Potatoes Olluco Wheat Fallow Fallow Oca Barley 5 2 7 71% Tuber Tubers Fallow Fallow Fallow ‘in some instances, some plats will be cultivated in the first fallow years. Note discussion in text. 198 ORLOVE & GODOY Vol. 6, No. 1 APPENDIX 1. Communities with sectoral fallowing systems. (continued) ss ow Elevation Total No. Average Name of Depart- Country Lati- Longi- of Com- of SFS Elevation Community ment tude tude munity perCom-_ of SFS (meters) munity (meters) Huayhuahuasi Cuzco Peru 14° 71° 3950 1 N/A (Orlove 1977) 40’ 30’ Huancaire Lima Peru 12° 76° 3200 1 N/A (Soler 1964) 25’ 10’ Pampahuarca Cuzco Perm 149 719 3900 1 4100 (Orlove 1977] 50’. Escara Oruro Bolivia 18° 68° N/A 1 N/A (Prescott 1974) 40’ 10’ Acora Puno Peru 16° 69° 3861 es N/A (Tshopik 1946} 0’ 50! NIA Unipa Apu- Perm 13° 73° 2903 3675 (Fonseca & = rimac 35’ 25’ Murrugarra 1966} Cuyo Cuyo Puno Peru 14° 69° 3450 3 3950 (Camino 1978} 30’ ~—«-30 3500 3300 — San ho a ee le ae rh * a ri 7 _ - Summer 1986 JOURNAL OF ETHNOBIOLOGY 225 NEWS and COMMENTS SOCIETY OF ETHNOBIOLOGY NEWS Tenth Annual Conference The 10th Annual Conference of the Society of Ethnobiology will be held March 5-8 1987 at the Florida State Museum, University of Florida. Papers are invited on the following and related topics: cultural ecology, plant and animal domestication, ethno- zoology, zooarchaeology, ethnobotany, archaeobotany, palynology, ethnopharmacology, human diet and nutrition, folk taxonomy. For further information please write to Elizabeth S. Wing, Florida State Museum, Gainesville, FL 32611 / (904) 392-1721. Barbara Lawrence Prize Announced The Society will award a prize in honor of Barbara Lawrence for the best paper sub- mitted by a student for presentation at the 10th Annual Meeting. The competition is open to any member who considers themself a student and has not held the PhD degree at the end of the preceding summer session. The paper can be presented in an oral or a poster session and will be considered for publication in the Journal of Ethnobiology. Manuscripts submitted for this competition should be single authored only; joint efforts will not be considered. Manuscripts are judged solely on quality, originality, and presentation of research. They should follow the Journal of Ethnobiology format and should be sufficiently precise and documented to enable the reviewing committee to judge their merits. Manuscripts are limited to eight doubled-spaced, typed pages, including a required abstract but excluding copies of figures, tables, and references. Please include a cover letter indicating the you are a Society member and meet the criteria listed above and send it and your paper to the 10th Annual Conference com- mittee, Florida State Museum, Gainesville, FL 32611. ANNOUNCEMENTS and REQUESTS The Society for Economic Botany announces the appointment of a new editor of the Society’s journal Economic Botany. John Thieret replaces Oswald Tippo as of June 1986. The Society’s 27th Annual Meeting was held 13-16 June this year at the New York Botanical Garden in the Bronx, New York. Highlights included a symposium on palms and a keynote address by this year’s Distinguished Economic Botanist, Efraim Hernandez- Xolocotzi. The 28th Meeting is scheduled for Chicago 22-25 June 1987, with a symposium on traditional medicine as focal point. Contact Susan Verhoek, Society president, Depart- ment of Biology, Lebanon Valley College, Annville, PA 17003, for more information. The School of American Research sponsored a seminar on “Bone Chemistry and Past Behavior’ chaired by T. Douglas Price, 3-7 March 1986. The chemical analysis of bones teveals information about diet, nutritional status, disease, even social status. Papers are to be published in the School’s Advanced Seminar Series. The School of American Research also wishes to announce a 1 December 1986 deadline for applications to their anthropology resident scholars’ fellowship program. Contact Susan Bodenstein at the school, P.O. Box 2188, Santa Fe, NM 87504, for details. The Laboratorio de Etnobiologia, Universidade Federal de Maranhao announces plans for the First International Congress of Ethnobiology, 15-21 June 1988 at the University in Sao Luiz, Maranhao, Brazil. The six-day congress will be divided into three parts: (1) 226 NEWS & COMMENTS Vol. 6, No. 1 three days devoted to specialized areas of ethnobiological research, (2) two days for inter- disciplinary discussions arranged around geographic interests and area studies, and (3) a final day reserved for presentations of applied ethnobiological projects and discussions of the potential of ethnobiological research for the improvement of world conditions. Organized symposia are invited. for additional information please write: Prof. Dr. Darre ] Addison Posey, Coordinator, Laboratorio de Etnobiologia, Largo dos Amores, 21, Univer- sidade Federal do Maranhao, 65,000 Sao Luiz, Maranhao (Brazil) or phone 098-221-1796. Celestial Seasonings Laboratory requests assistance in building a collection of herbarium voucher specimens and seeds. They are most interested in plants used in teas but are interested also in expanding their herbarium collections of toxic plants that may be adulterants in teas and in other economically valuable plants. For information contact Trish Flaster, Curator, at the Laboratory, 1780 55th St., Boulder, CO 80301, or call (303) 9-3779. ETHNOBIOLOGY in the NEWS Gentians outlawed in Tennessee: The Tennessee House voted 95 to 0 this past February to make it a misdemeanor to deliver, sell, or possess on school grounds the seeds of Gentiana lutea, equated in the legislation with “jimson weed.” Not so, says Nashville botanist Milo Guthrie. The Latin designation for jimson weed is Datura stramonium. What the bill bans is a harmless plant bearing yellow flowers that grows in the Alps and Pyrenees. The legislature wanted to outlaw jimson weed seeds because chewing them produces a hallucinogenic effect and has made several Nashville students ill recently. The bill’s sponsor, State Rep. Shelby Rhinehart, said it was drafted by the legislature’s legal staff and can be fixed by the state Senate. “I just told them to outlaw jimson weed. I didn’t tell them to put that Latin in there,’ he said. (Adapted from an AP report in the Washington Post on 17 February 1986.) Robert Bye, who sent the clipping, suggests that the confusion may have arisen from a phonological conflation of “jimson” with “gentian” in the local dialect. Bye recom mends that the lawmakers consult G. A. Mead’s 1970 note ‘On the Improper Usage of Common Names when Giving Botanical Data,” American Antiquity 35:108-109. Are Fungi the Stradivari Violin’s Secret Ingredient? A Seattle P.I. Science Brief this spring reported that the secret of the pure tone A Stradavari violins may be due to the effect of a microscopic fungus that grows within the cells of the wood used to make these classic instruments. Microscopic examinations of the samples showed that traces of fungi—which could have grown only through water immersion—had altered the shape of the wood cells, according to Joseph Nagyvary, Texas A&M professor of biophysics and biochemistry. Water fungi eat gummy material in the wood and make it lighter and drier, he said. They also force the cells to separate am loosen up. Nagyvary traveled to Europe and examined old shipping records stored in villages and monasteries. The records showed that logs had been sent downstream — the rivers leading from the Tyrolean Alps, where most of the wood was cut, to the Italian towns where the instruments were made. Haitian Pig Roast: A. Oppenheimer reports from Port-au-Prince (Seattle Post-Intelligencer of 15 — 1986) that peasants are protesting the consequences of a $22 million joint 2 nena? program put into effect in 1984 to improve Haitian pig herds. Under the direction © the International Institute of Agricultural Cooperation Haiti’s 1.2 million pig popula- Summer 1986 JOURNAL OF ETHNOBIOLOGY 227 tion was exterminated in 1984 to be replaced by U.S.-bred pigs. Western experts had characterized the Haitian pigs as ‘‘a degenerate species (sic), a pig that hadn’t been properly fed in 100 generations,” a long-legged breed that ‘‘roamed freely” and subsisted on “orange peels, mango seeds, and garbage.” These pigs were infected with African swine fever. The new breed of pigs, however, requires imported feed, concrete-floored stalls, expensive vaccines, and is too short of leg to walk to market. Critics of the program argue that the peasants, who constitute 70% of Haiti’s population, traditionally relied on pigs as an investment, a way of saving for their children’s education. The delicate new breed cannot be properly cared for and often dies prematurely, forcing Haiti's impoverished farmers to sell their land. Zapotec and High Tech: Mexico City News of 27 January 1986 carried an account of Gary Martin's efforts in collaboration with local herbalists in Oaxaca, Mexico, to establish demonstra- tion medicinal herb gardens and medicinal herb “libraries” in rural village centers, using a micro-computer to collate information on medicinal applications of Oaxaca’s some 4000 plant species by indigenous curers. For further information contact Martin at G.A.D.E., A.C., Apartado Postal 379, 68000 Oaxaca, Oaxaca, Mexico. GRUPO ETNOBOTANICO LATINO AMERICANO (GELA) CHARTERED During the Fourth Latin American Botanical Congress, which was celebrated in Medellin, Colombia, from June 29 to July 4, 1986, the “Grupo Etnobotanico Latino Americano” (GELA) was chartered as a section of the Latin American Botanical Association. This council consists of one representative from each country in Latin America and one from the United States, including its coordinating committee. Mexico City, D.F., is the seat of the first coordinating committee, which is chaired by Javier Caballero, with Victor Manuel Toledo, Monserrat Gispert, Arturo G6mez-Pompa and Armando Contreras as members. The representatives from each of the other countries, at this point, include Anthony Anderson (Brasil), Eduino Carbono (Colombia), Sonia Lagos (Honduras), David Diaz-Miranda and Maximina Monasterio (Venezuela), Gary Martin and Arturo G6mez-Pompa (USA). The principal function of GELA is to promote scientific exchange between different researchers who are actively working in the field of ethnobotany. This council will use the Bulletin of the Latin American Botanical Association as its medium of information dissemination. € primary tasks of the Coordinating Committee of GELA are the following: 1. Form a directory of ethnobotanists from Latin America, the U.S. and other parts of the world, who are active in researching this topic. To move toward this goal, it will soon distribute a registration form to be entered into a computerized data base. 2. Form a bibliographic clearinghouse of publications by those researchers included in the directory, to be placed in the care of the Secretary of the Library of the Mexican Botanical Society. This library is actually located within the facilities of the Botanical Garden of the National Autonomous University of Mexico. This collection will be available to researchers of any country, via a photocopying service. 3. Promote the organization of a meeting of Latin American ethnobotanists just prior to the celebration of the next Latin American Botanical Congress. This meeting will be celebrated in Mexico in 1988. For more information of GELA, write to: Javier Caballero, Jardin Botanico, UNAM, Ciudad Universitaria, Coyoacan 04510 México D.F. [Translated by Gary Nabhan]. 228 Vol. 6, No. 1 Each spring semester the San Diego Natural History Museum and the Museum of Man jointly sponsor a one-day ethnobiology seminar honoring the late Dr. Raymond M. Gilmore (see J. Ethnobiol. 4:97-99, 1984). The first, in 1985, featured Dr. Johanes Wilbert on the cosmology of the Warao Indians of Venezuela. The following year Dr. Thomas R. Van Devender spoke on changes in climate and Sonoran Desert begetation during the last 20 thousand years based on his studies of packrat middens. Our next seminar, scheduled for Saturday, 7 February 1987, at the Natural History Museum, will be presented by a panel actively researching Cucurbita spp. (pumpkins and squash). Dr. Thomas W. Whitaker, Ms. Laura Merrick, and Dr. Gary Nabhan will present archaeologic, ethnographic, and genetic aspects of this segment of the traditional New World corn-beans-squash trinity. society of Ethnobiology 1987 Annual Meeting Notice The 40th Annual Meeting of the society of Ethnobiology will be held on March 5-8, 1987 the Florida State Museum University of Florida Please see News & Comments on page 225 for details NOTICE TO AUTHORS The Journal of Ethnobiology accepts papers on original research in ethnotaxonomy and folk classification, ethnobotany, ethnozoology, cultural ecology, plant domestication, zooarchaeology, archaeobotany, palynology, Saini Gs and ethnomedicine. Authors should follow the format for article from articles in this issue. All papers should be typed double- spaced a pica or elite type on 8% x 11 inch paper with at least one inch margins on all sides. The ratio of tables and figures to text pages should not exceed 1:2-3. Tables should not duplicate material in either the text or graphs. All illustrations are considered figures and should be submitted reduced to a size which can be published within a journal page without further reduction. Photo- graphs should be glossy prints of good contrast and sharpness with metric scales included when appropriate. All illustrations should have the author(s) name(s) written on the back with the figure number and a designation for the top of the figure. Legends for figures should be typed on a separate page at the end of the manuscript. Do not place footnotes at the bottom of the text pages; list these in order on a separate sheet at the end of the manuscript. Metric units should be used in all measurements. Type author(s) name(s} at the top left corner of each manuscript page; designate by handwritten notes in the left margin of manuscript pages where tables and graphs should appear. If native language terminology is used as data, a consistent phonemic orthography should be employed, unless a practical alphabet or a phonetic is justified. A brief characterization of this orthography and of the phonemix inventory of the language(s) described should be given in an initial note. To increase readability native terms should be indicated as bold-face italics to contrast with the normal use of italic type for foreign terms, such as latin binomials. If necessary, the distinction between lexical glosses, i.e., English] ing, and precise English equivalents or definitions should be indicated by enclosing the gloss in single quotation marks. Authors must submit two copies of their manuscript plus the original copy and original figures. Papers not sul the correct format will be returned to the author. Submit your manuscripts to: DR. WILLARD VAN ASDALL, Editor Journal of Ethnobiology Arizona State Museum, Building 26 University of Arizona Tucson, Arizona 85721 NEWS AND COMMENTS Individuals with information for the ‘‘News and Comments” section of the Journal should submit all appropriate material to Eugene Hunn, Department of Anthropology, — DH-05, University of Washington, Seattle, Washington 98195. 3 bas aEvIEWs | bine Please send sug ;, comments, or reviews to Charles H. Office of Arid Lands Studies, U University of Arizona, en Arizona 85721. SUBSCRIPTIONS | ce Subscriptions to the Journal of Ethnobiol ogy sh nena caddies t piicdeseag vac id Univesity of Foie Gainesville, Florida - CONTENTS SKETCHES IN THE SAND sae DEVELOPMENT OF A SOCIETY: introduction to the special issue ola A. Weber VOUCHER SPECIMENS IN ETHNOBIOLOGICAL STUDIES AND PUBLICATIONS Robert A. Bye, Jr. VERIFICATION AND REVERIFICATION: PROBLEMS IN ARCHAEOFAUNAL STUDIES Amadeo M. Rea FOOD SAMPLE COLLECTION FOR NUTRIENT ANALYSES IN ETHNOBIOLOGICAL STUDIES Harriet V. Kuhnlein GUIDEPOSTS IN ETHNOBOTANY Vorsila L. Bohrer NEW DIRECTIONS OF PALYNOLOGY IN ETHNOBIOLOGY Richard G. Holloway and Vaughn M. Bryant, Jr. ON THE ANALYSIS AND ATION OF igre LIST DATA IN ZOOARCHAEOLOGY R. Lee L ETHNORIOLOGY, COGNITION AND THE STRUCTURE — : SOME GENERAL THEORETICAL NOTES ae ing Herc ISSUES IN ETHNOENTOMOLOGY Journal of Ethnobiology VOLUME 6, NUMBER 2 | es s e Journal and Society Organization EDITOR: Willard Van Asdall, Arizona State Museum, Building 26, University of Arizona, Tucson, Arizona 85721. ASSOCIATE EDITOR: Karen R. Adams, Department of Ecology & Evolutionary Biology, University of Arizona, Tucson, Arizona 85721. NEWS AND COMMENTS EDITOR: Eugene Hunn, Department of Anthropology, DH-05, University of Washington, Seattle, Washington 98195. K REVIEW EDITOR: Terence E. Hays, Department of Anthropology and Geo- graphy, Rhode Island es Providence, Rhode Island 02908. : Steven A. Weber, Department of Anthropology, University of Pennsylvania, Philadelphia, iaobamien 19104. SECRETARY/TREASURER: Cecil Brown, Department of Anthropology, Northem Illinois University, Dekalb, Illinois 60115. CONFERENCE COORDINATOR: Jan Timbrook, Department of Anthropology, Santa Barbara Museum of Natural History, 2559 Puesta Del Sol Road, Santa Barbara, California 93105. EDITORIAL BOARD BERLIN, Department of Anthropology, uphicoa! of California, Berkeley, California 94720, ethnotaxonomies, linguistic ROBERT A. BYE, JR., Department of Environmen ee ee and Organismic Biology, University of Colorado, Boulder, Colorado 80309, ethnobotany, ethnoecology. RICHARD I. FORD, Director, Museum of Anthropology, University of Michigan, Ann Arbor, Michigan 48109; archaeobotany, cultural ecology B. MILES GILBERT, Box 6030, Department of Geology, Northem Arizona University, Flagstaff, Arizona 86011; Sowa ac HAYS, Depart f Antt and Geography, Rhode Island College, Providence, Rhode ‘Island 02908, sihistonnne < ethnotaxononiies. EUGENE HUNN, Department of Anthropology, University of Washington, Seattle, Washinshod 98195, ethnotaxonomies, zooarchaeology, cultural ecology. HARRIET V. KUHNLEIN, Director, School of Dietetics and Human Nutrition, Mac "donald College of McGill University, 21,111 Lakeshore Road, Ste. Anne de Bellevue, Quebec H9X 1C0, Canada; ethnonutrition. DARRELL A. POSEY, Camegie Museum of Man, Pittsburgh, Pennsylvania 15206; ethnoentomology, tropical cultural ecology AMADEO M. REA, Curator of Birds and Mammals, San Diego Museum onan History, P.O. Box 1390, San Diego, Calif zooarchae- , De r Ont. le Florida State Museum, University ville , Florida 32611, zooarchaeology. logy/Ar haeology Program, asacieueesis ssachusetts ahi eee tropical SSOUR] BOTAN JUN 3 0 19 BARDEN LBRAR) Journal of Ethnobiology VOLUME 6, NUMBER 2 WINTER 1986 Recently, in a telephone conversation with Linda Desmond, typesetter for the Journal, I mentioned having difficulty in deciding upon a topic for this issue. She volunteered that she liked the column on apple butter and added “Give us more apple butter.” (See Volume 4, Number 2). That article had been inspired by the holiday season, and having just distributed petite fruit cakes as holiday gifts, I’ll tell you about the recent evolution of the recipe I used. Let’s suppose your situation is this. The idea of fruit cake appeals to you; it is, afterall, traditional holiday fare and they are festive in appearance. You and your family have never enjoyed the heavy, moist texture of the fruit cakes you’ve sampled, not even the blue ribbon winners at the county fair. You cannot justify the expense and effort of baking fruit cakes because of their festive appearance alone, so you have a dilemma. The easy solution is to firmly declare—as many do—that fruit cakes (or anything else you've decided you don’t like) are overrated, aren’t worth the effort, are too expensive, etc. and you might even look with disdain upon those who enjoy them. The more creative way, and happily for Society many also take this course, is to use your abilities as a domestic scientist: decide what you don’t like in the fruit cakes you’ve sampled, identify those qualities you do like and be sure to emphasize those, invent innovations of your own and otherwise experiment. Since fruit cake is usually made only at holiday time, it may be several years before you have perfected a family tradition. Let’s further suppose you live on a farm in the Mid-west of the U.S.A. You have available, free for the taking, black walnuts, hickory nuts, hazel nuts, cherries, currants, etc. You also have a number of children who are just the right age to harvest these com- modities and prepare them for baking (and, incidentally, in the case of black walnuts this involves a rather considerable amount of time and effort). And, enjoying the flavor of nuts, you decide to feature them, that is, your fruit cake will have more nuts than fruit. But, what about the texture? Was there a reason, originally, for the moistness of the cake? Did it serve a function that isn’t important to you? In what ways can you eliminate the unwanted heaviness? If increasing the amount of leavening isn’t the course you want to pursue, do you have another option? You realize, perhaps subconsciously, that doing something about the texture requires a re-thinking of the concept of the batter. You identify that it’s essentially a combina- tion of wet and dry ingredients. If one reduces the amount of batter and incorporates its two components separately, will this achieve the desired effect. So, you try it and discover that it does! You now have a fruit cake which is mostly nuts and fruit with just enough batter to hold the really tasty ingredients together. In effect, the nuts and fruits are glazed with the batter in the same manner, more or less, that puffed popcom grains are stuck together with the sorghum molasses glaze in the popcorn balls for which you are so well known. You even wonder if this, subconsciously, is from whence the idea came. As the years pass and the family matures, you lose that source of labor. You find edient hased nuts and discover that almonds and Brazil nuts are wonderful too. Some of the children leave the area, even as far away as Arizona. it exn MANLIL LU i But hickory trees do not in sunny Arizona grow. Nor are hickory nuts or black walnuts readily available for purchase, so the traditional family recipe becomes modified i with further substitutions: pecans for hickory nuts, English walnuts for black walnuts, raisins for currants, purchased candied fruit, etc. Perhaps this anecdote of the evolution of a recipe for fruit cake will remind us that family traditions are modified with changes in circumstances. So, too, traditions within a larger cultural setting are far from static, changing with time and circumstances, as Jan Timbrook pointed out so well (see Volume 4, Number 2:143-144). —-, W.V. -_ ~~ —_é, —e — wy ll atte — - ae. _ te a atin, we? — ay NY a =- en, te fn, — ail ' ‘ | J. Ethnobiol. 6(2):229-238 Winter 1986 RECOGNITION OF VARIABILITY IN WILD PLANTS BY INDIANS OF THE NORTHWEST AMAZON: AN ENIGMA RICHARD EVANS SCHULTES Jeffrey Professor of Biology and Director Botanical Museum of Harvard University, Emeritus Cambridge, MA 02138 “There is evidence that herbal doctors disti gui h between diff varieties, anatomical parts, stage of maturity (for all of which they have a vocabulary) and times of collection . .. This information may be very significant in terms of medical effectiveness of the remedy.” Abbot and Shimazu (1985) | The Indians of the northwest Amazon—that region lying in Colombia and Ecuador and adjacent areas of Brazil and Peru—have an uncanny familiarity with the plants of the ambient vegetation and with their biodynamic activity. This knowledge and skill is obviously the result of millennia of intimacy with and utter dependence upon the flora, combined with the intense curiosity about natural phenomena that characterizes these people. One of the enigmas that botanists have not yet had much success in understanding concerns the native’s recognition of ‘‘kinds” or “varieties” of many of the wild species of their useful plants. The variants are so well established in the Indians’ classifications that they usually have distinguishing native names. This skill is manifest not only to those few of the 80,000 species native to the region that are economically important but is found in the aboriginal classification of a number of plants which apparently have little or no importance as utilitarian, ceremonial, magical or mythological species. In most cases, it is botanicaliy impossible to discern morphological differences on which subspecific taxonomic categories might be recognized. Yet often—nay, usually—an Indian can tell at once and frequently on sight and at a significant distance, without feeling, tasting, smelling, crushing, tearing or other physical manipulation, to which category a plant belongs. The “identification” of these “kinds” is indeed a complex interdisciplinary problem, but, while it is obviously of deep significance to the anthro- pologist and psychologist, it is of extraordinary importance to the botanist and phytochemist. Little research on this fascinating aspect of ethnobotany seems to have been attempted. Most of the “explanations” offered are pure conjecture. It has been suggested that these recognized named “varieties” may be simply different parts of a large plant, sundry age forms or portions growing in shade, sun or under other environmental conditions. It is very probable—especially with food, medicinal, narcotic or toxic species—that some of these “varieties” may represent chemovars. But, if so, how can a native visually identify which chemovar it is and give it the name that his language has for that variant? I have tested the perspicacity of the Indians in this respect on many occasions and have rarely found them hesitant, doubtful or in error. And Indians of different tribes and living at appreciable distances from one another will identify these variants with amazing consistency. 230 SCHULTES Vol. 6, No. 2 There are no better examples of this aboriginal perspicacity than the stimulant of the westernmost Amazon of Colombia and Ecuador—yoco, the sapindaceous vine Paullinia Yoco (Schultes 1942)—and the source of the western Amazon hallucinogenic drink variously known as ayahuasca, caapi, natema, pindé, yajé—the malpighiaceous liana, Banisteriopsis Caapi (Schultes 1957). Evaluation of field studies and studies of voucher specimens of these two major economic plants illustrate the enigma surrounding the aboriginal ability to identify “variants” of the species in their rich ambient vegetation. peci ited are p din the Economic Herbarium of Oakes Ames and/or the Gray Herbarium, both at Harvard University, and/or in the Herbario Nacional de Colombia in Bogota, Colombia. Il. Paullinia Yoco R. E. Schultes et Killip in Bot. Mus. Leafl., Harvard Univ. 10 (1942) 302, t. XXvil. Paullinia (Fig. 1) is a genus of the Sapindaceae comprising more than 180 species of lianas of tropical or subtropical America. One species, P. Cupana Kunth is cultivated in the Brazilian Amazon as the source of a caffeine-rich drink prepared from the seeds. Anthropological writings and reports of travellers frequently made mention of yoco in the Colombian Putumayo. Material collected by the Belgian scientist Florent Claes in the late 1920s was chemically examined and found to contain caffeine. It was misiden- tified as Paullinia scarlatina Radlk., a species that does not occur in South America. In 1942, the liana was definitively identified and described as a new species of P ~osumecnal P. Yoco (Schultes 1942). ty years ago, the Colombian writer Zerda Bayon reported that the Indians of that “there is a yoco blanco [‘‘white yoco”] and a yoco colorado [{‘red yoco"’]. His specimens have long since been lost, and he did not attempt to explain the differences between these two named kinds of yoco. In 1942, when the binomial Paullinia Yoco was published, I commented as follows on the several named and recognized variants. “During my ethnobotanical studies 10 the Putumayo, I repeatedly questioned nati ning the differences between yoce blanco and yoco colorado with conflicting replies. While it is true that the sap expressed from some stems makes a light chocolate-brown mixture when added to cold water, that from other stems makes a whitish milky mixture. Both taste the same, and both are equally effective as a stimulant. The Indians do not prefer one to the other. I find it impossible to distinguish botanically the liana which gives yoco blanco from that — yields yoco colorado, but the natives can identify them immediately by slashing the bark with a machete. I have noticed that yoco colorado nearly always is a much stouter and apparently older plant than yoco blanco. It is not possible that the differences a due to seasonal or soil conditions, for yoco blanco and yoco colorado grow side by side and can be collected at the same time” (Schultes 1942). ” One of the recent collections (Pinkley 380) reports that totoa-yoko (“white yoco “has more leche (‘milk’) than other types and is therefore the best type.” Klug 1955 indicates that huarmi yoco is the “strongest” kind. Schultes et Cabrera sine num. notes that yoco de brujo has unusually large leaves. Further studies in the field and herbaria have done little to advance our understanding of the reasons for the Indians’ recognition of these named “variants.” On the contrary, : 1 dis > a pe y more the problem has been 1: WAP VeLy Winter 1986 JOURNAL OF ETHNOBIOLOGY 231 XY x ws we ee iy = “23 +} ines a) at nated (vi PAULL Lae FIG. 1.—Paullinia eo Drawn by the late Gordon W. Dillon, artist, Botanical Museum, Harvard University 232 SCHULTES Vol. 6, No. 2 named kinds of yoco. The botanists who have gathered the names and have associated them with collections of yoco cannot offer distinguishing characters. The named “variants” now known number fourteen: ape blanco yoco; canaguche yoco (“‘yoco of the palm Mauritia flexuosa’); huarmi-yoco; po-yoko; tigre yoco (‘‘yoco of the jaguar’), taruco yoco; totoa-yoco (“white yoco”’); verde yoco (‘green yoco”’), yagé-yoco or yoco- yaje ( yoco of the hallucinogen Banisteriopsis Caapi’’); yoco colorado; yoco-cu (‘‘red- dish yoco”); yoco de brujo (medicine-man’s yoco”); yoco negro (‘black yoco”). Two of these names might suggest that that kind of yoco was used with products of other plants (with chicha de cananguche, a fermented drink prepared from the fruits of Mauritia and with yajé, a narcotic drink prepared from Banisteriopsis), but that does not explain the uncanny skill of the natives in distinguishing the kind of yoco from afar. COLOMBIA: Comisarid del Putumayo, Umbrid. “Liana. Verde yoco.” January-February 1931. G. Klug 1930—Same locality and date. ‘Liana. Blanco yoco.” Klug 1933. Same locality and date. Klug 1935—Same locality and date. Klug 1937—Same locality and date. “Liana. Yagé yoco.” Klug 1946—Same locality and date. “Vine. Cananguche yoco.” Klug 1947—Same locality and date. ‘Vine. Petals white. Strongest yoco. Huarmi yoco.” Klug 1955—Same locality and date. “Vine. Taruco yoco.” Klug 1957—R{o Putumayo, Piftuna Negra.“ Yoco. Arbusto de 1.5 m.” November 20, 1940. Cuatrecasas 10708—Mocoa. “Bark used as source of stimulant drink. Large liana.” December 3-7, 1942. Schultes et Smith 3045—Same locality and date. “Large woody vine. Stimulant and antifebrugal plant.” Schultes et Smith 3045A4—Rio Sucumbiés, between Rfo Putumayo and Quebrada Teteye. “Yoco colorado. Extensive liana 1 in dense forest. Sap of cambium scraped and used also as a purge before taking yajé and to expel stomach parasites.’ March 19, 1942. Schultes 3426—Departamento del Cauca, Rio Caquet4, Puerto Limoh. “Large liana. Bark used to prepare stimulant. Whitish latex-like sap.” February 28-29, 1942. Schultes 3341—“ Yoco colorado. Extensive vine, in dense forest. Used as a stimulant and febrifuge.’”” May 29, 1942. Schultes 3476—Rio Sucumbidés, Conejo. April 2-5, 1942. Schultes 3543— Rio Putumayo, Puerto Ospina. “Extensive liana; basal diameter 3 inches; sap expressed as stimulant. Bark contains a white, latex-like sap. July 6, 1942. Schultes 4028—Mocoa. “Large forest liana.” March 1953. Schultes et Cabrera s.n.—Between Mocoa and Pepino. March 1953. Schultes et Cabrera s.n.—Mocoa. “Yoco negro.” March 1953. Schultes et Cabrera s.n. Near Mocoa. “Tigre yoco.” March 1953. Schultes et Cabrera s.n. Along road 12 km. below Mocoa. “Yoco colorado .” March 1953. Schultes et Cabrera s.n.— Pepino. ‘‘Extensive vine in forest. Yoco de tigre.’ March 1953. Schultes et Cabrera s.n. Rid Uchupayacu. “ Yoco blanco. Flowers white. Large liana.” Schultes et Cabrera s.n.— Mocoa and vicinity. ‘Yoco colorado.” March 1953. Schultes et Cabrera s.n.—Pepino. “Unusually large leaves. Yoco de brujo .”’ March 1953. Schultes et Cabrera s.n. —Mocoa and vicinity. “Large forest liana. Yoco yajé.’” March 1953. Schultes et Cabrera s.n. Rio Guamués, San Antonio. ‘Tree 4 m. Secondary growth. Bark for stimulant and to allay hunger.” September 6, 1963. —Mocoa, old road to Rumiyaco. Alt. 700-850 m. “Los indigenas lo toman en maceracion en frio. Yoco.”” October 10, 1965. Garcid-Barriga, Hashimoto et Ishikawa 18695—R{o Putumayo Remanso. “Primary forest. Bark of stem is scraped and soaked to make a beverage which is drunk by the men early mornings while making plans for the day. Totoa-yoko (‘white yoco’) has more leche than other types of yoco, therefore is the best ee ” August 22, 1966. Pinkley 380—Buena Vista. “Po-yoko. This vine, growing wild in the jungle, can grow very tall and have a diameter as thick as 4-5 inches when old. Its flower was described as ‘medio-blanco’ (whitish) by Francisco. It has a seed which splits open and is red inside with a black seed in the middle. This is the most common form of yoko taken by the Siona. They scrape the inner wood and drink the juice that comes out. They drink one or two mouthfuls in Eee ETT ee acre cccc acca a aS r Winter 1986 JOURNAL OF ETHNOBIOLOGY 233 the early dawn to twist the chambira fibre or perform other tedious tasks.” September 13, 1972. Piaguaje 5. ECUADOR: Provincia Napo, Rio Aguarico, Dureno. “Primary forest. Bark of stem scraped and soaked in cold water to make a beverage which is drunk by the men very early each morning. Liana. The men drink usually together in the morning while planning the day. Kofan: Yoko-cu (‘reddish yoko’).”’ January 3, 1966. Pinkley 72—Same locality. “Cortex of stem reddish (older than vine of no. 311). Kofan: yoko-ca (‘reddish yoko ’).” June 22, 1966. Pinkley 312 [Note: no. 311 not available for study].—Same locality. “Totoa yoco (‘white yoco’).”” September 6, 1966. Pinkley 428—Rio Napo, amongst Secoya Indians. “Yoco.”’ November 29, 1971. Louthian s.n. Il. Banisteriopsis Caapi (Spr. ex Griseb.) Morton in Journ. Wash. Acad. Sci. 21 (1931) 485. Banisteriopsis (Fig. 2) is a genus of approximately 100 species of forest lianas of tropical America. Several species have been reported as the source of an hallucinogenic drink prepared by Indians in South America—B. Caapi and B. inebrians Morton. Recent taxonomic research has indicated that the second binomial is a synonym of the first. Vague references to this drug were made in missionary writings of the late 17th Century in Peru and Ecuador, but little was known about it until the mid 19th Century. In 1858, Villavicencio wrote about the hallucinogenic drink ayahuasca in his Geografia del Ecuador, but he mentioned no botanical identification beyond the fact that it was a liana (Villavicencio 1858 and 1984). The first scientific identification of the drug was done by the British plant explorer, Richard Spruce, who in 1852 had discovered that the Tukanoan tribes of the Rio Uaupés of Brazil prepared an intoxicating drink of caapi from the bark of a liana. He collected botanical material and identified it as a new species of Banisteria—B. Caapi—now nomenclaturally known correctly as Banisteriopsis Caapi. The description of the species was not published until 1858, and Spruce’s account of the preparation of caapi did not appear until 1873 (Spruce 1873, 1908 and 1970). 1853, Spruce met with the use of caapi amongst the Guahibo Indians of the upper Orinoco of Colombia and Venezuela—Indians who ‘‘not only drink an infusion .- . but also chew the dried stem.” In 1857, when he travelled and collected in the Ecuadorian Andes, he found the Z4paro and other Indians taking a narcotic drink called ayahuasca, and he reported that he considered it to come from “the identical species of the Uaupés, but under a different name” (Langdon 1985). . Since this early scientific work, many specialists and amateurs have written about the malpighiaceous narcotic of the western Amazon. We now are rather certain that it is prepared basically from the bark of one species, Banisteriopsis Caapi (Gates 1982). Occasionally additives may be put into the brew to alter, strengthen, change or lengthen the intoxication (Spruce 1873, 1908 and 1970}. Many have been reported, but two are of importance and are widely used: the rubiaceous Psychotria viridis R. et P. and the malpighiaceous Diplopterys Cabrerana (Cuatr.) Gates (formerly known as Banisteriopsis Rusbyana (Ndz) Mart.). Both of these plants contain tryptamines, the first known in both ilies, which actually do alter and intensify the effects of the hallucinogenic 6 -carbolines in the basic bark of B. Caapi (Schultes and Hofmann 1973 and 1980). There is no doubt that Indians in the northwest Amazon can “identify” different “kinds” of caapi or ayahu t a distance without feeling, tasting or smelling the liana. Sundry field studies have noted this peculiarity, and there is a long list of native names 134 SCHULTES Vol. 6, No. 2 BANISTERIOPSIS Caapi iM Spruce ex Griseb,) Morton FIG. 2.—Banisteriopsis Caapi (ayahuasca). Drawn by the late Elmer W. Smith, artist, Botanical Museum, Harvard University. , = a Winter 1986 JOURNAL OF ETHNOBIOLOGY 235 that are presumed to designate these numerous variants. The natives maintain that they are able to use these kinds of caapi or yajé or ayahuasca to prepare drinks of different strengths, for different purposes or in connection with different ceremonies or dances or magico-religious needs, or what the partaker wishes to kill in the hunt. At least 30 “kinds” are recognized and have native epithets in the western Amazon. This aspect of ethnobotanical studies certainly requires much more intensive and interdisciplinary field research. Are these kinds different age forms, are they due to hardly perceptible soil or other ecological factors; are they the result of growing in semi-open or secondary situations, as opposed to the dense forest; are the specimens taken from various parts of the liana; are the cultivated specimens specially selected clones with varying chemical composition and, consequently, varying physiological effects; or are they chemovars? Langdon (1985) has written: ‘Apparently the native populations. . . recognize many different kinds of caapi with different hallucinogenic properties; I consider these to be chemical variants. The ease with which caapi can be vegetatively propagated . . . makes it possible for clones of such variants to be maintained.” This statement might explain a limited few cases of cultivated plants, but it cannot satisfac- torily be considered an explanation of the problem for wild lianas, nor can it in any way clear up the Indian’s ability to identify these variants ocularly from a distance. Langdon (1985) further states that the western Siona of the Putumayo of Colombia, Tukanoan- speaking peoples, with whom she worked, use “finer distinctions than the botanist in classifying plants. One feature employed in their classification is botanical: length, breadth, size and leaf pattern and whether or not the plant flowers; another feature concerns the part of the plant used; another refers to phases of growth; still others are distinguished . . . on the kind and colours of the visions induced; the strength of the intoxication also enters as a factor. Other differences are taken into account: each plant has a spiritual guardian and a shaman owner, and shamans often trade kinds. Further- more, if a shaman finds a wild liana in the forest, he will prepare a drink to ascertain its worth for inclusion in his own repertoire, especially in regard to what visions it can induce; if he takes a cutting, he will then and there name and classify it.” It is difficult for the scientist to understand or accept many of these “criteria”, as teal as they may be to the Indian, but native perspicacity in the finer classification of many plants—both wild and cultivated—should be critically examined for the possible Practical values of some of the points of evaluation to taxonomists and phytochemists. On the basis of extensive field work in Peru, Rivier and Lindgren report that the Sharanahua Indians distinguish three types—red, black and white “kinds”, and that the distinction is based more in the differentiation in colour of the drink than in the appearance of the plant (Rivier and Lindgren 1972) - According to another field i igator, Deltgen, there are in the Colombian Vaupés six kinds of caapi, based primarily on their effects (Deltgen 1978-79). : Still another specialist, Reichel-Dolmatoff, states that there is in the Vaupes “a large series of kinds” distinguished mainly on the basis of psychoactive effects (Reichel- Dolmatoff 1975). The most comprehensive field investigation of the Indian recognition of “kinds” of the narcotic liana is that of anthropologist Langdon amongst the western group of Tukanoans—the Siona—who live in the Colombian Comisaria del Putumayo, far to the west of the Tukanoan tribes of the Vaupés. Langdon was able to collect 18 different vernacularly named “kinds”. Botanical material of almost all was collected and identified by Dr, Timothy Plowman of the Field Museum of Natural History in Chicago (Langdon 1985). Almost all are botanically referrable to a single species: Banisteriopsis Caapi. Siona classification “is seen’, according to Langdon, ‘‘as more complex than that of botany and depends on the conjunction of botanical features, chemical effects of the 236 SCHULTES Vol. 6, No. 2 mode of preparation and cultural suggestions in the visions experienced” (Langdon 1985). It is undoubtedly true that all of these criteria are employed by the Indians for eventual “classification” of the kinds of Banisteriopsis Caapi, but it is still not possible for a native in the forest ocularly to identify with certainly a kind of caapi by vernacular name from such features as its chemical tituents and culturally—what he believes that the drug may induce. The meticulous field research of Langdon has indeed produced a good start, but, in her own words, “ er exploration bet thi junction of botany-chemistry-culture warrents further investigation”. It is still an enigma. COLOMBIA: Comisarfa del Putumayo, Rfo Uchupayaco, southeast of Puerto Limon. “Yajé, Narcotic. Liana.” February 27-28, 1942. Schultes 3346—R{o Sucumbiés, Conejo, March 29, 1942. Schultes 3475—Mocoa, August 28, 1963. Juajibioy C. 256a—Mocoa. “Bichemia (‘bejuco’), amarr6n huasca (‘vine of the boa’).” August 28, 1963, Juajibioy C. 279—Mocoa. “Inde huasca.” August 28, 1963. Juajibioy C. 280—San Antonio del Guamués. “Yajé del monte’. August-September 1963. Naranjo et Wiederhold 4—San Antonio del Guamués. “Yajé sembrado.”’ August, September. Naranjo et Wiederhold 5—Buena Vista. “‘Celima’s wai yagé. Celimo grows it in his field. It is a tall and thick vine. It is cooked before drinking, with yagé-oko added. Francisco Piaquaje said that it is another class, but he can’t identify it. Luciano Piaquaje recognized it as wai’ yagé.” September 1972. Langdon 21—Buena Vista, “Beji yage. This yagé is cultivated in the fields. It is tall and very thick at full growth. It is cooked with yagé-oko and is supposed to be one of the strongest yagés”’ September 26, 1972. Langdon 23—Buena Vista “Wai yage or wahi-yagé collected and grown by Francisco Piaquaje. The yagé is a mata (bush) instead of a vine. It is about 1 meter tall. It is grown in their fields. The plant grows as a thick bush. This yagé is usually prepared simply by grating the stems and without cooking. It is used to see the pinta de caseria—visions of animals. The leaves may possibly be mashed and their juice used also.” September 26, 1972. Piaquaje 25—Buena Vista. “So’om-wa-wa’i yagé. Collected and grown by Ricardo Yaiguaje. This yage, which Richard grows in his field, is given its name because of the way it grows. The vine, which is about 3-4” thick at maturity, grows very, very long: so-om-wa. To make it grow thick, Ricardo prunes the branches. This is a class of wahi-yagé and they drink it raw by grating the vine. However, they also mix it with various classes of kwa’ku-yagé. It shows the visions of hunting. Leonides, Ricardo’s father, told him to plant only this in a field, and the yagé would be like the seed of the wild pig sesé so that even though the Siona no longer drink yagé, they would always have a lot of game. I asked him if this was sese yagé, and he said: ‘Lo mismo’.”” September 27, 1972. Langdon 30—Buena Vista “ Usebo yagé—collected by Ricardo Yaiguaje. The vine of this class of yage is very thick, about 6” in diameter. The vine grows straight up instead of twisting, until its thickness begins to diminish. This is one of the kwa’ku class and 1s cooked with yage-oko.” September 27, 1972. Langdon 31—Buena Vista. “Wai yage. It is a bush about 1-1¥ meters high.” September 28, 1972. Langdon 32—Buena Vista. “ Weki- yage. Grown by Celimo Amo. It is a large vine. Is a class of kwa’-ku yagé.” September 29, 1972. Langdon s.n.--Between Mocoa and Pepino. “Stout vine on trees, 45 ft. tall. Leaves membranaceous, dark green, shining above, pale green beneath. Stems boiled with chagropanga (Diplopterys Cabrerana) to make narcotic drink. Indehuasca. Yajé.” July 28, 1960. Schultes 22553. INTENDENCIA DEL GUAINIA: Rio Inirida, El Remanso. August 12, 1975. Garcia- Barriga 20805. ——ti emma ee EG | | | Winter 1986 JOURNAL OF ETHNOBIOLOGY 237 ECUADOR: Provincia Pastaza, Rio Chico, affluent of Rio Pastaza, Village of Rio Chico. Alt. c. 1000 m. “Shredded stem boiled with fine leaves of Diplopteris Cabrerana (Shemluck et Ness 218) until conc. and ca. 3 oz. taken on empty stomach for hallucinations. The i oa pias bananas and salt can be eaten. Ayahuasca.’’ August 1979. Shemluck ess 219. PERU: Departamento de Loreto, Iquitos. “Ayahuasca. Woody vine. A tea brewed from the leaves and stem produces fanciful dreams. Also used as a cure for many ills and as an intoxicating beverage. A strong narcotic.” August 2-8, 1929, Killip et Smith 27385—Iquitos region, Rio Nanay, Picuruyacu “Ayahuasca. Vine growing near garden. Chopped vine boiled with yajé and samiruca [Psychotria viridis]." July 5, 1966, Martin et Lau-Cam 1089. Rio Amazonas. “Trepadora sin flores ni frutos. Haya huasca.” October 25, 1966. Torres 223. Zapote, Alto Rio Puris, “Lowland forest. The stems are mashed and boiled with water during one hour with one or more additives to prepare ayahuasca.” August 22, 1968. Rivier 1—Zapote, Alto Rio Puris. “Lowland forest. Rami- ee (‘yellow ayahuasca’). Culiba name” August 22, 1968. Rivier 2. Marcos, Alto Rio Purus. “The stems are mashed and boiled in water for about one hour with one or more additives to prepare ayahuasca. Lowland forest.” August 8, 1968, M68. Rivier 3. Marcos, Alto Rio Puris. Shurioshinipa (‘red ayahuasca’) = name in Sharanahua. August : 1968. Rivier 4. Rio Amazonas, Caballacocha. “Ayahuasca. E] remedio. Extensive woody lana 6 ft. high with trunks to 10 cm. diameter. Cultivated in chacra near lake. Stems chopped up and boiled for 8 hours to prepare hallucinogenic beverage, mixed with chacruna leaves (Psychotria vividis).” March 22, 1977. Plowman, Schultes et Tovar 6430. Departamento de San Martin, Huahuiva. “Ayahuasca negra.” July 6, 1985. Woytkowski 5045; 5074.—Monte real. Shillinto o ayahausca amarilla. Medicinal.” July 10, 1958. Woytkowski 5076—San Alejandro, Rio de Loreto. “Monte real. 300 m. Ayahuasca amarilla.” July 24, 1958. Woytkowski 5119—San José de Sisa, c. 550 m. “Trepador voluble (soga) cultivado. Flores rosadas. Shimba-ayahuasca.” July 26, 1958. Velarde N. 6577—Tarapoto. Rio Schilcayo. Alt. 350 M. “Extensive cultivated liana, 6 in. tall, growing in full sun. Sterile. Ayahuasca.” May 4, 1976. Plowman 6041. LITERATURE CITED LANGDON, J. 1985. Siona classification ABBOTT, LA., and C. SHIMAZU. 1985. of yagé, Ethnobotany, ethnochem- The geographic origin of the plants most commonly used for medicines y Hawaiians. J. Ethnopharma. 14:213-222. DELTGEN, F. 1978-79. Culture, drug and personality—a preliminary report about the results of a field research among the Yebamasa Indians of Rio Piraparana in the Colombian Com- isaria del Vaupés. Ethnomedizin 5(1,2):57-81. GATES, B. 1982. Banisteriopsis, Diplo- Pterys (Malpighiaceae) in Flora Neotropica, New York Botanical oe New York. Monograph No. istry, visions and history. Unpubl. lecture. Congreso Internacional Americanistas, Bogota. REICHEL-DOLMATOFF, G. 1975. The shaman and the jaguar—a study of narcotic drugs among the Indians of Colombia. Temple Univ. Press. Philadelphia. RIVIER, L., and J.E. LINDGREN. 1972. ‘Ayahuasca’, the South American hallucinogenic drink: an ethno- botanical and chemical investigation. Econ. Botany. 26:101-129. SCHULTES, R.E. 1942. Plantae Colom- bianae Il. Yoco: a stimulant of 238 SCHULTES southern Colombia in Bot. Mus. Leafl., Harvard Univ. 10:301-324. . 1957. The identity of the malpighiaceous narcotics of South America in Bot. Mus. Leafl., Harvard Univ. 18:1-56. and A. HOFMANN. 1973 Vol. 6, No. 2 Orinoco in Ocean Highways: the Geographical Review, v. 9., no. 55 (1873) 184-193. [Ed. A.R. Wallace] Notes of a botanist on the Amazon and Andes. Macmillan and Co., Ltd. London. 2 vol. Reprinted ed. Johnson Reprint Corp., New York. 2 vol. and 1980. The Botany and Chemistry of Hallucinogens. Ed. 1 and Ed. 2 Charles C. Thomas, Publisher, (1970). VILLAVICENCIO, M. 1858. Geografia de la Rep&blica del Ecuador. R. Craigs- head, New York 371. Reprinted edi- tion: Corporacion Editora Nacional, Quito (1984). Springfield, Ill. SPRUCE, R. 1908. On some remarkable narcotics of the Amazon Valley and The Cactus Primer. A.C. Gibson and P.S. Nobel. Pp. vi + 286; illustr.; Harvard University Press, Cambridge, Mass. 1986. $39.95. As the authors state in their preface: “People around the world and from all walks of life are hopelessly susceptible to a condition called cactophily, the love of cacti.” This interest has led to a plethora of books on the cactus family— mostly dedicated to classifica- tion or horticulturally curious specimens and their care as house plants. Here we have a different book—one that, to my knowledge, is the first to present in a single volume such a wealth of data on the biology and structure of this most misunderstood family of xerophytes. The topics discussed span a broad spectrum: general features; early evolutionary trends; special features; chemistry; how structure and chemistry can help unravel phylogeny; and, finally, relationships of the family. The extensive glossary is a very helpful addition; the detailed index unlocks with ease much of the information in the volume. Each chapter has its own bibliography. As a contribution of true biological value, this volume will be welcomed by botanists, horticulturists and amateur cactophiles alike, and especially by those interested in the drier parts of the world, for it is not possible to find such a mass of scattered information in one book—and amassed and expertly evaluated by two recognized specialists in the group. Richard Evans Schultes Professor Emeritus Botanical Museum of Harvard University Oxford Street Cambridge, MA 02138 J. Ethnobiol. 6(2}:239-255 Winter 1986 ABORIGINAL EXPLOITATION OF PRONGHORN IN THE GREAT BASIN BROOKE S. ARKUSH Department of Anthropology University of California Riverside, California 92521 ABSTRACT.—Ethnographic, historic, and archaeological data concerning pronghorn exploitation in the Great Basin are presented in a framework that allows for a revision of prevailing models concerning this activity. It is proposed that various ethnographic and historic accounts do not accurately reflect the prehistoric contexts of pronghorn exploitation. Exception is taken with the Pendleton and Thomas (1983) model, which Proposes that the use of permanent, labor-intensive drive structures diminishes through time in the Great Basin. INTRODUCTION The American pronghorn (Antilocapra americana), distinctive in both its evolution and behavior, was an important source of food and clothing for Indian peoples of central and western North America. Many ethnographic and historic sources mention both the ceremonial preparation for, and the actual hunting of pronghorn; and pronghorn remains are present in the faunal assemblages of numerous Great Basin sites. In spite of this, archaeological data concerning the exploitation of pronghorn by aboriginal groups in the Great Basin has not been synthesized with ethnographic data on the pronghorn. A revision of several prevailing concepts concerning pronghorn hunting that appear in the literature, and which have apparently been accepted by many North American anthropologists in lieu of a critical examination of the available data, is suggested. Various models concerning this important economic and social activity appear to be based largely on both post-contact- and negative data, and probably are not applicable to the Pre-contact practice of pronghorn exploitation. EVOLUTION AND ETHOLOGY Based on the geological contexts of fossil specimens, it has been estimated that several 8enera of antil ids }