Evolution of High Andean Puna Ecosystems: Environment ...ecaths1.s3.amazonaws.com/argentina1/MRD.1993.13.145.156 (Baied...Mountain Research and Development, Vol. 13, No. 2, Mountain

Evolution of High Andean Puna Ecosystems: Environment, Climate, and CultureChange over the Last 12,000 Years in the Central Andes

Carlos A. Baied; Jane C. Wheeler

Mountain Research and Development, Vol. 13, No. 2, Mountain Geoecology of the Andes:Resource Management and Sustainable Development. (May, 1993), pp. 145-156.

Stable URL:

http://links.jstor.org/sici?sici=0276-4741%28199305%2913%3A2%3C145%3AEOHAPE%3E2.0.CO%3B2-C

Mountain Research and Development is currently published by International Mountain Society.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtainedprior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content inthe JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/journals/intms.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. Formore information regarding JSTOR, please contact [email protected].

http://www.jstor.orgMon Apr 23 18:07:51 2007

EVOLUTION OF HIGH ANDEAN PUNA ECOSYSTEMS: ENVIRONMENT, CLIMATE, AND CULTURE CHANGE OVER THE LAST 12,000 YEARS

IN THE CENTRAL ANDES

ABSTRACT In the High Andes, much of the traditional knowledge related to natural resource utilization and agricultural practices was lost with the arrival of the Spaniards. That which has survived has been transformed, together with the landscape. Based upon ongoing research on the paleoecology and archaeology of the central High Andes, this paper presents the first results of pollen analysis from late-Pleistocene/early Holocene Laguna Seca (latitude 18'11' South, longitude 69"14'30" West) in the Chungara- Cotacotani lake district of Lauca National Park. The Laguna Seca record covers two crucial events in the evolution of high Andean puna ecosystems: (1) the period preceding the earliest human occupation, and (2) the period of change from a food-gathering to a food-producing economy. The results, though preliminary, indicate the contribution that these studies can make to the management and preservation of fragile mountain ecosystems.

R%SUME Evolution des kosystimes de la haute puna des Andes: changements de l'enwironnement, du climat et de la culture dans les Andes centrales au coun de 12 000 derniiws annies. Dans les hautes Andes, une grande partie des connaissances traditionnelles relatives i I'utilisation des ressources naturelles et aux pratiques agricoles a dispaGpar suite de la colonisation espagnole. Ce qu'il en restait s'est transformi de la m&me manisre que le paysage. Cet article s'appuie sur les rksultats de travaux actuels portant sur la palkokcologie et l'archiologie des hautes Andes centrales et prksente les premiers rksultats de I'analyse du pollen datant de la fin du PliistocZ.ne/dibut de l'Holoc6ne en provenance de la Laguna Seca (latitude 18"ll ' Sud, longitude 69"14'30" Ouest) situie dans le district lacustre de Chungar-Cotacotani 5. I'intkrieur du parc national de Lauca. Les donnkes enregistrkes pour la Laguna Seca couvrent deux kvinements d'une grande importance pour l'ivolution des kcosysti5mes de la puna des hautes Andes: (1) la pkriode prickdant la premisre occupation humaine et (2) la pkriode d'kolution d'une iconomie de cueillette i une kconomie de production de denrkes alimentaires. Bien que priliminaires, les risultats donnent une idke de la contribution possible de ces ktudes pour la gestion et la priservation des fragiles kcosystsmes de montagne.

ZUSAMMENFASSUNC Evolution der ~una-Okosysteme in den Hochanden: Urnwelt-, Klima- und Kulturueriinderungen in den Zatralanden w a h d der vetgangenen 12.000 Jahre. Mit der Ankunft der Spanier gingen viele der fiberlieferten Erfahrungen bezuglich Nutzung der natfirlichen Ressourcen und landwirtschaftlicher Methoden in den Hochanden verloren. Was davon erhalten blieb, hat sich, ebenso wie die Landschaft selbst, verindert. Diese Ver6ffentlichung bringt erste Resultate einer Pollenanalyse aus dem Spatpleistozin/ F ~ h h o l o z i nLaguna Seca (geogr. Breite 18" 11' Sud, Linge 69" 14' 30" West) im Ch~gar-Cotacotani Seengebiet des Lauca National Parks. Die Ergebnisse berucksichtigen neuere Forschung auf dem Gebiet der Palio-Okologie und der Archiologie in den zentralen Hochanden. Die Laguna Seca Aufzeichnung schlieBt zwei entscheidende Vorginge bei der Entwicklung des Puna-Okosystems der Hochanden ein: (1) die Zeit vor der ersten Besiedelung und (2) die Periode, in der sich die Nahrungserzeugung vom Sammeln zur Produktion verinderte.

Die ersten Ergebnisse sind geeignet zu zeigen, welchen Beitrag solche Untersuchungen beim Management und bei der Erhaltung eines fragilen Gebirgs-Okosystems leisten k6nnen.

RESUMEN Evolucih de ecosistemas de la puna andino: Cambios de medio ambiente, clima, y cultura s o h 12,000 alios en las Andes Centrales. El presente trabajo resume la informaci6n obtenida en recientes investigaciones arqueol6gicas y paleoecol6gicas en la regi6n de 10s Andes Centrales. Se presentan 10s registros polinicos de Laguna Seca (18"111S lat., 69'14'30"W long.) que se extienden desde la primera ocupaci6n de cazadores del sector andino de la puna, desde aproximadamente 11,000 afios A.P., abarcando el period0 de transici6n hacia una economia pastoril basada en la cria de llamas y alpacas, y el impact0 de esta actividad productiva sobre la cubierta vegetal. Los resultados que aqui se presentan, si bien preliminares, integran y aportan informaci6n para la preservaci6n y manejo de recursos naturales enecosistemas de altura.

INTRODUCTION

The European settlement and colonization of the New present. The introduction of Old World cultigens, ani- World launched an environmental crisis of unprece- mals, farming practices, and legislation modified the dented magnitude with lasting consequences up to the landscape and impoverished genetic resources through

'Department of Geography, University of Montana, Missoula, MT 59812-1018,USA. 2Macaulay Land Use Research Institute, Hartwood Research Station, Shotts, Lanarkshire ML7 4JY, United Kingdom. Present address: Facultad de Medicina Veterinaria, Universidad Nacional Mayor de San Marcos, Apartado 5137, Lima 3, Peru.

O International Mountain Society and United Nations University

146 / MOUNTAIN AND DEVELOPMENTRESEARCH

the elimination of native species, local traditions, and knowledge. In the Andes, for example, previously un- known diseases reduced the human population by an estimated 8096, leading to economic and social disin- tegration of native societies (Wachtel, 1977). Within little more than a century following the conquest of Cuzco in 1532, many llama and alpaca herds were also decimated (Flores Ochoa, 1977, 1982). Newly introduced European livestock (sheep, goats, cattle, and pigs) rapidly replaced llamas and alpacas in the coastal and highland valleys, and eventually even displaced them in much of their original puna habitats. Early tax records, laboriously kept by the Spaniards, illustrate the extent of this devastation. In 1567, Garci Diez de San Miguel, inspector for the Spanish Crown, reported a single, privately-owned herd of 50,000 llamas and alpacas in the province of Chucuito on the western shore of Lake Titicaca. Five years later in 1572, Pedro Gutierrez Flores counted only 160,000 heads in the entire province. By the end of the sixteenth century llamas and alpacas had practically disappeared throughout the Andes (Flores Ochoa, 1977).

The Spanish intervention in the Andes vividly exem- plifies the consequences of human impact upon fragile mountain ecosystems. But what were the conditions before 1532? It would be unrealistic to consider puna ecosystems to be pristine. Twelve thousand years of continuous human occupation preceded European con- tact, and during this period natural resources were used, managed, and modified, initially by low impact hunter- gatherer groups, then later, some 7,000-6,000 years ago, by increasingly sedentary pastoralist-agriculturalist communities.

Given the long history of human activity in the High Andes and the radical changes that occurred starting in the sixteenth century, it is not possible to reconstructthe evolution of puna ecosystems, or even to evaluate the impact of traditional llama and alpaca pastoralism, based on the study of present-day conditions alone. Much of the traditional knowledge related to natural resource utilization and agricultural practices was lost with the arrival of the Spaniards, while that which has survived has been transformed, together with the landscape. Successful management and preservation of these fragile, intensively used ecosystems requires an understanding of both prehispanic and more recent historic, ethno-historic, and current events, as well as reconstruction of the environmental and climatic history of the puna during late-glacial and post-glacial times.

SOtJTH AMERICA

PACIFIC OCEAN

FIGURE1. Puna life zones (modified after Winterhalder and Thomas, 1978).

Past environmental conditions can be documented in several different ways. On a regional scale, lacustrine and peat-bog sediments provide pollen records of past vegeta- tional changes, while on the scale of the archaeological site, botanical and archaeo-zoological materials docu- ment specific use of the environment. Taken together, these lines of inquiry can provide a comprehensive picture of the nature of past human impact on the environment.

Based upon ongoing research into the paleoecology and archaeology of the High Andes of northern Chile, this manuscript presents the first results of pollen analysis from Laguna Seca in the Chungara-Cotacotani lake district of Lauca National Park. This record covers two crucial events in the evolution of the puna ecosystem: (1) the earliest human occupation, and (2) the period of change from a food-gathering to a food-producing economy. The results, though preliminary, indicate the contribution that these studies can make to the manage- ment and preservation of fragile, endangered habitats.

PUNA ECOSYSTEMS

Lying between 3,500 and 5,500 m above sea level, puna ecosystems extend along the high ridge of the Andes between approximately 7" and 27" South. Within this belt, conditions of reduced atmospheric pressure, widely fluctuating diurnal temperature, and reduced moisture availability produce intense environmental stress. From north to south, precipitation varies considerably and, while seasonal, is unpredictable in onset and volume. Extended droughts are common. Under these conditions,

productivity is low and fluctuates from area to area, affecting energy flow from year to year. Carl Troll (1968), for example, distinguishes three distinct puna life zones differentiated through vegetative characteristics and dis- tinct patterns of human behavior (Figure 1).

Running from northwest to southeast, the Moist Puna extends from approximately 7" South in Peru to the east- central Bolivian plateau. Annual precipitation varies fi-om 500 to 1,000 mm decreasing from east to west and north

to south, and is sufficient to support agriculture at altitudes up to 4,200 m. Human occupation of this zone may have begun some 15,000 years ago, and it was here that llama and alpaca domestication took place and the first urban centers in the puna subsequently evolved (Wheeler, 1984, 199 1). The agricultural and pastoral wealth of the circum-Titicaca area sustained the develop- ment of what has long been recognized as one of the most important foci of human activities in the Americas, from viliage farming to urbanism and civilization. At the time of Spanish contact this area was the meeting place of two powerful Andean language groups, the Quechua and the Aymara.

Southwest of the Moist Puna extends the Dry Puna belt where mean annual precipitation is between 300 and 500 mm. Agriculture becomes impractical above 4,000 m, but

relatively rich pasture lands provide abundant grazing for livestock of native camelids all year round. The cumu- lative effects of several climatic factors-scarce precipita-tion, perennial frost, intense solar radiation, and reduced oxygen and carbon dioxide levels-make the Dry Puna a moderately risky habitat for human populations.

Extending to the south and west of the Dry Puna, from the Aymara village of Lirima at 20" South and well into the uplands of the Atacama Desert, lies the Salt Puna. Here, extreme aridity imposes significant limitations to year-round occupation. Precipitation drops progressively from 300 mm in the north and east to almost zero in the Atacama Desert to the west. The upper limit of perma- nent agricultural settlement is substantially lower than in the ~ o i s t and Dry punas, plunging to 2,000 m south of 26"s.

EARLY HUMAN UTILIZATION OF THE PUNA: THE IMPACT OF HUNTING

Relatively little information concerning human utiliza- tion of what are today the puna ecosystems is available for the period prior to 10,000 yr B.P. Late-glacial faunal assemblages from highland archaeological sites and pale- ontological localities in Peru and Bolivia are dominated by extinct fauna including Parahipparion, Scelidotherium, Agalmaceros, and Megatherium (Hoffstetter, 1986). Paleo- climatological sequences from Lake Junin in central Peru (Hansen et al., 1984), various localities in Bolivia (Graf, 1981 a,b; Ybert, 1984, 1987; Ybert and Miranda, 1984) and Argentina (Markgraf, 1985, 1987; Fernandez, 1986 a,b; Fernandez et al., 1991) generally indicate the ex- istence of a cold and humid climate prior to 13,000 yr B.P., followed by a cold and dry episode lasting approx- imately 3,000 years. Around 10,000 yr B.P., a return to more humid conditions is registered, together with the appearance of a fully modern faunal assemblage and the first unquestioned evidence of human occupation of the puna.

The most extensive documentation of early human adaptation to the high-altitude ecosystem comes from archaeological sites in what is today the Moist Puna. In the puna ofJunin, 170 km northeast of Lima, Peru, eight cave and rock-shelter sites have been studied (Matos, 1975; Wheeler Pires-Ferreira et al., 1976; Rick, 1980; Lavallee et al., 1985; Wing, 1986; Moore, 1989), and a pollen sequence has been obtained from both sediment and archaeological deposits (Hansen et al., 1984; van der Hammen and Noldus, 1985). Taken together, the archaeo-zoological and paleo-vegetational data from these sites indicate a progressive development from

generalized hunting of all available ungulates-guanaco (Lama guanicoe), vicuiia (Vicugrza vicugna), and huemul deer (Hippocamelus antisiensis) -towards specialized hunt- ing of guanaco and vicuiia. This occurred prior to the beginning of domestication some 6,000 years ago (Wheeler, 1984). The decrease in cervid remains in the puna of Junin record appears to be a cultural rather than an environmental phenomenon (Wheeler, 1984, 1985) ; this is also observed in faunal assemblages from other sites in the area (Wheeler Pires-Ferreira et al., 1976; Wing, 1986; Moore, 1988, 1989).

As in the Moist Puna, earliest occupation of the Dry and Salt punas began some 10,000 years ago (Barfield, 1965; Yacobaccio, 1985 a,b; Nuiiez and Santoro, 1989). Excavations at various archaeological sites in northern Chile (Santoro and Nufiez, 1988; Nuiiez and Santoro, 1989) and southern Peru (Ravines, 196'7, 1972; Aldender- fer, 1988) have produced materials of this period. Inter- pretation of the adaptive strategies practiced by the early inhabitants of these sites has been based primarily upon site location, density of occupation, and artifact assem- blages. Although an apparent abandonment of the zone between 8,000 and 5,000 yr B.P. has been attributed to climatic change and volcanic activity, no paleoecological research has been carried out in association with these excavations. Detailed studies of both the botanical and faunal remains are either lacking or, as at Patapatane Cave, the samples are too small to permit reliable inter- pretation (Wheeler and Dennis, unpubl.), so it is as yet impossible to evaluate archaeologically the impact of hunting on the environment at this time.

CAMELID DOMESTICATION AND THE IMPACT OF NATIVE ANDEAN PASTORALISM

The puna ecosystem was a primary center of ungulate domestication, comparable to the Near East (sheep and goats), Mediterranean Europe (cattle), and Tibet (yaks). Evidence from sites located in the Moist Puna of central Peru indicates that llamas and alpacas were first brought

under human control some 6,000 years ago (Wing, 1974, 1986; Wheeler, 1984, 1985, 1988 a). A possible second center of domestication has been reported in the Salt Puna Salar de Atacama, between 4,800 and 4,000 yr B.P. (Hesse, 1982 a, 1986; Santoro and Nuiiez, 1987; Nuiiez

and Santoro, 1989),but data are lacking for the Dry Puna of northern Chile.

As is the case with the preceding period of hunting adaptation, the most complete evidence on the origin and evolution of native Andean pastoralism available to date comes from the puna of Junin in the central Peruvian Andes (Figure 2). Excavations at the sites of Pachamachay and Panaulauca (Wheeler Pires-Ferreira et al., 1976; Rick, 1980; Wing, 1986; Moore, 1988, 1989), as well as Telarmachay rock-shelter (Wheeler, 1984, 1988 a; Lavallee el al., 1985) have all contributed evi-dence documenting the shift from hunting to herding. At Telarmachay rock-shelter, located at 4,420 m above sea level, more than a ton of animal bones were recovered, providing detailed evidence of the domestication process. At this site the percentage of fetal camelid remains prior to 7,000 yr B.P. corresponds to the observed frequency of pregnant females in contemporary vicuiia and guanaco populations, about 35%, suggesting a hunting economy. By 7,000 yr B.P. fetal remains are largely replaced by the bones of newborn animals. The latter comprise 57% of all camelid remains at this period (Wheeler, 1985, 1988 a), and are thought to reflect disease-induced mortality associated with domestication, as opposed to hunted animals taken in utero. Early attempts at domestication, perhaps involving corralling, would certainly have re-sulted in crowding and unsanitary conditions, ideal for the propagation of epidemic disease(s), and newborn animals would be the most susceptible segment of the population. A similar pattern of neonatal mortality still exists among llamas and alpacas. Herders may loose up to 70% of each year's young due primarily to Clostridium perfringens Type A enterotoxemia (Ramirez, 1987) and complications arising from the failure of passive im-munoglobulin transfer from mother to offspring (Gar-mendia et al., 1987). The probable emergence of epizo-otic diseases at this time is, in all likelihood, a by-product of the domestication process itself. Changes in dental morphology suggest that the animal being brought under control at Telarmachay was the vicuiia (Wheeler, 1984, 1985). Evidence of llama domestication is less clear, but it is likely to have occurred at the same time somewhere to the south (Wing, 1974, 1986), although both alpaca and llama domestication must have occurred many times in different parts of the Moist Puna. The pollen data from Telarmachay may indicate that overgrazing within the vicinity of the site became a problem shortly after domestication occurred (van der Hammen and Noldus, 1985).

A second set of data concerning early camelid herding comes from the Salar de Atacama in the Chilean Salt Puna (Figure 2). This is based on the analysis of animal bones from the site of Puripica-1 where Hesse (1982 a) documents the existence of domestic camelids at 4,500-4,000 yr B.P., as well as research on the fiber remains which confirms their presence at 3,200 to 2,600 B.P. (Dransart, 1991). It has been suggested that these animals represent an independent center of domestica-tion, separate from the Moist Puna (Hesse, 1982 a,b),but the lack of time depth makes it impossible to determine

i PERU ,_-,

, , BRASIL

---. \ ,?:----'!

1 \ PANAULAUCA PUNA OF JUNlN A 0 XLARMACHAY L.+-- ,-.-.--,L,MAbC0 PACHAMACHAY

I

I BOLIVIA i PACIFIC OCEAN .LA PA2

I I

>

1 70e

P E R U

K a f a n l l e a c o t n p a m p a

A m a r e l s

Puno VUNGAS

LAKE TITICACA

Chuqulngulllo

Asnna

Toquepale 0

la0-

20' B O L I V I A 20"-

,l '.-_r-- . _ ,'Bsrro Negro "

- - - - - - - - - -TIODIC 01 Capricorn

0 100 200

FIGURE2. General location maps showing puna archaeological sites (o),modern pollen rain records (A), and fossil pollen sections (A).

if the herds were introduced into the area already domesticated or if they were brought under human control locally. Whatever the case, it has been suggested that the shift to herding occurred at this time in response to oscillating climates (Druss, 1978; Hesse, 1982 a,b), but direct paleoecological evidence is lacking, as are data from the adjacent Dry Puna zone which might document the spread of domestic animals from the north.

The gap between initial domestication, some 6,000 years ago, and the development of specialized breeding to produce llamas and alpacas similar to those we know today may have been as long as 4,000 years. By the end of the pre-ceramic period some 3,800 years ago, domestic

llamas and alpacas were common in the highland valleys of Peru and northern Chile (Hesse, 1982 a; Wing, 1986); by 1,400 years ago llamas were being bred on the coast of Peru (Pozorski, 1979; Shimada and Shimada, 1985) and in Ecuador (Wing, 1986; Stahl, 1988; Miller and Gill, 1990); and 400 years later there is evidence of alpaca production on the south coast of Peru (Wheeler, Russel, and Stanley, in press). But it was under the Inca empire, between A.D. 1430 and 1532, that llama and alpaca distribution reached its maximum Andean extension as llama pack trains accompanied the conquering armies from southern Colombia to central Chile.

CONTEMPORARY HERDING PRACTICES

There has been considerable difference of opinion about the extent to which the puna ecosystems have been modified since the arrival of the Spanish (Bernhardson, 1986). At present, a pattern of mixed herding in which llamas, alpacas, and sheep graze jointly is common throughout the puna. Although we have considerable information concerning the foraging habits of all three species and their impact on the environment today (San Martin and Bryant, 1987), the extent to which this information can be used in reconstructing the evolution of puna ecosystems is limited. Without data on the composition of the vegetation prior to the introduction of European livestock, it is not possible to evaluate precisely what their impact, as well as that of native pastoralism, has been.

In comparison to sheep, the Andean domesticates, llamas and alpacas, are environmentally low-impact graz- ers. They differ from ruminants of the suborder Pecora in stomach morphology, absence of horns and antlers, and the presence of soft, nail-covered digital pads instead of hooves. This structure minimizes impact on the fragile

puna groundcover, in contrast to the hooves of sheep which cut into the ground. Additionally, their prehensile split upper lip permits them to be highly selective in their diet, grasping the desired plant or plant part and neatly clipping it off. Llamas and alpacas do not dig up vegeta- tion with their feet and do not eat plants down to the root, again in contrast to sheep. Llamas and alpacas are also much more efficient in utilizing the low-protein- content vegetation of the puna, due inpart to the longer time in which food is retained in the rumen (San Martin and Bryant, 1987).

Today, under conditions of pure native pastoralism, llamas and alpacas forage separately; the former prefer the dryer bunch grasses and browse them where available, the latter graze the moister plants of the highland bogs. When sheep are introduced they compete directly with alpacas for pasture resources, a fact which has un-doubtedly contributed to overgrazing of the puna wet- lands. Whether, and/or to what degree, such overgrazing occurred prior to the introduction of sheep remains to be determined.

EVOLUTION OF THE DRY PUNA ECOSYSTEM: RESEARCH IN LAUCA NATIONAL PARK, CHILE

Lauca National Park presents ideal conditions for examining the evolution of the Dry Puna ecosystem under conditions of early hunting-gathering, native Andean pastoralism, and European origin herding (Fig- ure 2). Previous research on the natural environment and cultural resources under the Man and Biosphere Pro- gramme (Bustos and Veloso, 1982) provides an important starting point, together with the archaeological investiga- tions conducted by the Universidad de Tarapaca ( Rivera, 1975; Santoro and Chacama, 1982, 1984; Dauelsberg, 1983 a,b).

Located within the Dry Puna life zone (see Troll, 1968), Lauca National Park encompasses a section of the Altiplano plateau perched at 4,000-4,500 m above sea level. The park was created in 1970 and modeled after the United States conservation archetype (Torres et al., 1978). In 1981, Lauca was incorporated into UNESCO's

World Biosphere Reserve System, and its present-day area of 137,883 hectares was set by CONAF-the national parks administration-in 1983. The park encompasses a variety of natural ecosystems, including lacustrine envi- ronments of lakes Chungara and Cotacotani, high-altitude grasslands (Pajonales), shrublands ( Tolares), cushion-peat bog habitats (Bofedales), and Quenoa dwarf forests (Figure 3). The average altitude is 4,000 m, and the highest peaks are the volcanoes Pomerape and Parinacota, located on the northern edge of Lake Chun- gara. Precipitation is sparse, averaging from 300 to 500 mm annually, and is restricted to the late spring and summer months from November to March. Mean annual temperature recorded at the 4,400-m Parinacota meteor- ological station in 1972 was loC, with daily fluctuations of as much as 30" C, and freezing temperatures occurring during every month of the year (Torres et aL, 1978).

FIGURE 3. The ChungaraCotacotani lake district, Chile. In the foreground, a typical marsh (Bofcdal) by the village of Parinacota during the dry season (3,900-4,000 m). The hummocks in the background are avalanche debris (17,000-15,000 yr B.P.) from a partial cone collapse of Parinacota volcano, shown in the foreground (right) t e gether with Mount Pomerape (left).

The area is known to contain important archaeological Santoro and Chacama (1982, 1984), and Hakenasa Cave sites (Santoro and Nufiez, 1987). Although no major by Santoro and Dauelsberg (Santoro and Nufiez, 1987). archaeological excavations have been carried out, several Excavation units range in size from one to six square sites have been tested by archaeologists from the Uni- meters, and radiocarbon age determinations range from versidad de Tarapaca. The most important of these 9,500 to 2,500 yr B.P. These dates span the crucial period include Patapatane Cave, excavated by Santoro (1987), for the origin and early development of native Andean Tojo Tojone by Dauelsberg (1983 a), Las Cuevas by pastoralism.

PAST VEGETATION AND ENVIRONMENTAL CHANGE: THE EVIDENCE FROM LACUNA SECA

At 4,000 m altitude, Laguna Seca (latit~del8~11" South, longitude 69O14'30 West) is a small, shallow lagoon located in one of several hundred depressions and hummocks that originated after the partial collapse of Parinacota volcano sometime between 17,000 and 15,000 years ago (Francis and Self, 1987). The lagoon is critically located in relation to glacial advances but was never overrun by ice. A two-meter deep sample of peat and lacustrine deposits containing a pollen-stratigraphic se- quence was recovered for study with the goal of doc- umenting the vegetational and environmental history of the area (Figure 4).

The local vegetation at Laguna Seca is dominated by wiry tussock-like grasses, which include h a , Festuca, and

Dqreuxia often in association with chamaephytes of the genera ~mophyllum, Tdmglochin, Adesmia, W m ' a , and Astmplus (Figure 3). Shrubs, including several Composi- tae (Pamtrephia, Baccharis, Senecio, C h u q u i ~ ) , Solana- ceae (Fabiana), and a dwarf tree of the Rosaceae family (Polylepis) are also found. Extensive marshy environments (Bofedales) supporting rich hydrophilous communities surround the site at elevations between 4,200 and 4,600 m. Cyperaceae and Juncaceae are the dominant families, together with water plants such as IsoetRF, MMyriophyUum, Elodea, A w h , and Lilaeopsis. Bofedales are considered the best grazing land for camelids in the Chilean Altiplano, followed in palatability by the tussock-like grassland (Troncoso, 1982).

MATERIALS AND METHODS

SEDIMENT DESCRIPTION Two major depositional units were visually identified

in the two-meter deep section (Figure 4): a lower lacus- trine unit for which a basal 14C date of 11,155 i 100 yr B.P. (GX-14,023) was obtained; and an upper peat unit sealed on top by 6 cm of loose sand. The distinctive transition in sediment stratigraphy, from strongly lami- nated lacustrine deposits to organic peat sediments, occurs at a depth of 90-92 cm and has been 14C dated to 7,030 * 245 yr B.P. (GX-14,359).

Examined under a stereo-microscope, the lacustrine

unit shows large concentrations of fossil freshwater snails and osrracods. The upper unit (0-92 cm) is composed of light brown fibrous peat with sedge and grass frag- ments, roots, and varying amounts of sand and silt. There are virtually no visible changes in humification or plant composition within the peat unit.

POLLEN-ANALYTICAL METHOD Sediment samples were taken from the previously-

cleaned surface of the exposed profile at Laguna Seca. Volumetric (1.8 ml) samples were treated following stan-

FIGURE 4. Exposed sediments at Laguna Seca. The lacustrine unit (below) underlies an upper peat unit.

dard procedures with hydrofluoric acid to remove sili- cates and with acetolysis solution to reduce organic matter (Faegri and Iversen, 1975). Samples were spiked before chemical treatment by adding three tablets of Eucalyptus sp. to allow estimation of pollen concentration per volume of sample. A smear of the pollen residue was

stained with safranin, mounted in glycerin, and sealed with wax. Pollen counts were carried out at a magnifica- tion of x 400 to x 1000 on Olympus Vanox and Wild Heerbrugg M20 light microscopes. As many as three slides were used in the tabulation of each pollen sDec- trum to reach a minimum of 200 grains when possible. Identification of pollen grains was checked against our pollen reference collection, photographs, and descrip tions published elsewhere (Heusser, 1971; Markgraf and D'Antoni, 1978; Wingenroth and Heusser, 1985). In general, pollen preservation is excellent but, while pollen is very abundant in the lacusvine unit, it is relatively scarce in many of the peaty sediment samples.

The pollen record from Laguna Seca is presented as a percentage diagram (~igures 5 and 6). ~ & a excluded from the pollen sum are Cyperaceae-Juncaceae, aquatics (My?iophyllum, Iridaceae), algae (Botryococcus, Miastmm), and monolete and trilete fern spores. Pollen listed under the "unidentifiedn category include grains that were not recognizable owing to poor preservation and grains whose affinities could not be ascertained.

The Compositae family is an important component of the puna shrubland, thriving in marshy environments and on sandy soils, ranging from herbs to relatively woody shrubs. Compositae pollen were differentiated pollen- morphologically into Tubuliflorae (long spine), Hapb pappus type, short spine among which Ambrosia and Arkmisia were isolated into types, Liguliflorae, and Muti- sieae of which Leuch.eria and Nassauuia were also segre- - gated into separate types.

Distinction between Blylqbis and A c m pollen was impossible because of the extensive overlap in size, sculpture, and pore and furrow pattern. onet the less, Pdylepts rather than Acuena is the more important compo- nent in the Dry Puna (Troncoso, 1982; Kalin Arroyo d al., 1988). In the pollen diagrams, both genera are grouped together as Polylqbts-Acaena type although it is assumed that the majority of this pollen represents Polylepis.

POLLEN STRATIGRAPHY

The dominant pollen taxa are Gramineae and Compo- sitae. Pollen productivity and dispersal of grasses and Compositae Tubuliflorae is high relative to the total production and dispersal of all other families, which in consequence are significantly under-represented in the pollen spectra. Thus, the Laguna Seca pollen-percentage diagram is dominated by grasses, accounting for more than 60% of the total sum at many levels. Compositae Tubuliflorae pollen percentage amounts to between 10 and 20% of the total pollen sum. Occurrence of other taxa, such as Umbelliferae, AzmUa, Chenopodiaceae, Caryophyllaceae, Solanaceae, Cruciferae, and Pdylqbis- Acaena is low, generally remaining below 5%. Pollen of Cyperaceae and Plantago remains below 2% at all levels -generally at and below 1%-except in the upper spectra when algae percentages drop to completely dis- appear from the record in the upper 30 cm. The limnic

environment is represented by freshwater algae; Bo- tryococcus and Miastrum are prevalent throughout the record except in the upper 27 cm of the section. Below 30 cm depth, Botryococcus percentages remain between 80 and 90 and those of Pediastrum between 10 and 20%. Cyperaceae at 25 cm depth, Plantago at 17 cm depth, and trilete spores replace algae in the upper segment of the Laguna Seca record. A few aquatics such as MyriophyUum and Iridaceae are also present but in minor percentages. Both experience two peaks, in the lower 25 cm of the section and between 90 and 30 cm depth. They are indicators of open, shallow waters, common in ponds and tarns throughout the High Andes. Long-distance pollen in the Laguna Seca record is represented by Alnus, Celtis, I%docarpus, Jugluns and Acalypha common to the Yungas phytogeographic region.

Based on changes in the relative frequency of pollen types, three major environmental zones are identified in the Laguna ~ e c a section:

Zone 1 (205-145 cm) represents the time interval between 12,000 and about 9,000 yr B.P. Gramineae pollen dominates with more than 55%, Compositae Tubuliflorae attains between 10 and 30%, and minor percentages of Compositae Liguliflorae (-1 %), Mutisieae type (including Lacheria and Nassauvia types), and among the other herbaceous taxa especially Umbelliferae (mostly Azorella) are consistently with over 5%.Malvaceae, Cheno- podiaceae, Caryophyllaceae, Cruciferae, and pollen of Polylepis-Acaena are present almost continuously, but with vaiuis of less than 5% of the ~ o l l e n sum. LO&-distance " arboreal pollen reaching a maximum of 5% is repre- sented by Acalypha, Alnus, Celtis, Podocarpus, and Juglans indicating distant pollen transport from the subtropical mountain and lowland forest of the eastern slopes of the Andes. The basal sample (205 cm) shows a relatively different spectrum, low percentage of Compositae Tubuli- florae (12%), high percentage of Gramineae (75%), and Umbelliferae (+3%) pollen. Gramineae pollen is present in a slightly higher percentage than in overlying zone 2, reaching here one of the record's peak. Comparable percentages of aquatics Botryococcus (51%) and Pediastrum (48%) suggest a relatively freshwater environment. This basal spectrum suggests the possibility of a late-Pleistocene environment just before 11,000 yr B.P.

In zone 2 (145-120 cm), beginning at about 9,000 yr B.P., long-distance arboreal pollen temporarily disappears from the record for an estimated 1,000 years. Gramineae pollen initially reaches almost 90% of the pollen sum, Compositae Tubuliflorae are represented by less than 9% Azorella pollen values and of other Umbelliferae, but increase-to maximum levels of more than 15% in the upper part. On the other hand, Compositae short spine

INTERPR

Although a detailed reconstruction of late-Quaternary environmental conditions of puna ecosystems is not yet possible, the gross floristic relationships of the past vegetation can be inferred from the pollen stratigraphy of Laguna Seca and other regional pollen records in the Moist and Dry Punas. Overall, the Laguna Seca record indicates the existence of Holocene-type climates and relatively stable ecosystems since at least 11,000 yr B.P. Late-Pleistocene/early Holocene lacustrine conditions at Laguna Seca, together with the occurrence of Andean forest taxa in the pollen spectra, suggest a climate cooler and moister than today, but somewhat drier than during previous late-Pleistocene times. The decline of arboreal exotic pollen towards about 8,000 yr B.P. and the rise of grass pollen indicate a change in the annual distribution of precipitation and a trend towards drier and warmer conditions in the high puna. The change from clayey-silty to peat sediments at Laguna Seca (7,000 yr B.P.), is indication of the final establishment of dry and probably

(mostly Ambrosia type) and Chenopodiaceae are persis- tently higher than before. Pollen of Polylepis-Acaena tends to decrease to between 2 and 3% percent of the pollen sum.

Zone 3 (120-0 cm), probably representing the interval between 8,000 yr B.P. and the present-day, is characterized by the decline of Ambrosia type pollen, Chenopodiaceae, Umbelliferae, including Azmella, Cruciferae (type I), and Poblepis-Acaena. Gramineae, between 50 and 60%, and Compositae Tubuliflorae, between 20-25%, continue to dominate the section. The Cruciferae (Menonvillea type) curve reaches relatively high values. Ephedra is also repre- sented. The change from lacustrine to peat sediments at about 7,000 yr B.P., together with changes in percentages of selected taxa (Cruciferae, Umbelliferae, including Azorella and Polylepis-Acaena),suggest an environmental change at the regional level.

Lack of radiocarbon dates above 90 cm, together with scarce pollen in most of the samples, makes the environ- mental interpretation difficult for the upper part of the section, the last 7,000 years. The uppermost part of the peat unit shows a peak in long spine Compositae to more than 50%, together with increases in Umbelliferae and Gentianacea, followed by peaks in Caryophyllaceae, of about lo%, Ranunculaceae, and Plantago pollen.

The estimated date for this assemblage is about 3,000-2,000 yr B.P. and probably indicates traces of human activity and manipulation of forage resources. Iridaceae and Myriophyllum are also well represented above 80 cm depth. Algae (Bot~ococcus-Pediastrum)com-pletely disappear from the record in the upper 30 cm. Gramineae pollen rebounds to its average representation, and Compositae Tubuliflorae increase up to 18% of the pollen sum. These upper spectra suggest water deficit in the local marshes, a trend that may be further ex-trapolated at the regional level.

warm conditions. Between 5,000 and 4,000 yr B.P., a short period of increasing moisture is suggested by the sig- nificant increase in pollen of aquatics and Andean arboreal pollen reaching the Dry Puna. Pollen spectra remain relatively stable up to about 3,000 yr B.P., when a sharp increase in Compositae Tubuliflorae pollen and a decrease in Gramineae is followed by a increase in Cyperaceae and Pluntago pollen. This is interpreted as a trend towards much drier conditions, but it may also suggest human manipulation of the environment through grazing of domestic camelids.

The Laguna Seca pollen record compares well with the nearby record of Sajama (Ybert and Miranda, 1984) and El Aguilar (Markgraf, 1985) in terms of timing and paleoenvironmental trends through the Holocene. The Pleistocene-Holocene transition is blurred at Laguna Seca, but fluctuating percentages of aquatics and algae in the basal spectra suggest changes in the water budget for the zone.

154 / MOUNTAIN AND DEVELOPMENTRESEARCH

Overall, the Laguna Seca palynological record suggests a moist and cold climate between 11,000 and 8,000 yr B.P., followed by progressively drier conditions. Late- Pleistocene/early Holocene Laguna Seca turns into a BofedaLtype ecosystem at about 7,000 yr B.P. suggesting a regional trend towards dry and probably warm condi- tions that lasted until about 5,000 yr B.P.. After 5,000 yr B.P., a moister but probably warm episode occurs. At about 3,500 yr B.P. a drastic dry episode precedes what we identified as the first human impact in the Dry Puna,

estimated to have occurred between 3,000 and 2,000 yr B.P. This coincides with the environmental changes due to human impact seen in El Aguilar (Markgraf, 1985) at 2,000 yr B.P. and in the Bolivian Moist Puna records of Cotapampa and Katantica (Graf, 1981 a) at 3,000 yr B.P. The Laguna Seca record does not allow an estimation of the establishment of present-day conditions in the Dry Puna, suggested to have occurred at about 4,000 yr B.P. in El Aguilar (Markgraf, 1985), and at about 2,000 yr B.P. in Sajama (Ybert and Miranda, 1984).

DISCUSSION AND CONCLUSION

Pollen correlations between puna localities after 11,000 yr B.P. become complex as it is apparent that lacustrine basins of the Tauca episode became isolated by lower water levels. The algae (Botryococcus-Pediastmm)spectra at Laguna Seca, for-example, suggest the influence of individual drainage basins, local topography, and spring- water systems dominating their limnological character by the Pleistocene-Holocene transition. By 11,000 yr B.P., Holocene-type climates become established in the puna, at 18" South. Pollen records throughout the puna in- dicate relatively stable environmental conditions capable of supporting a population of game animals and Paleoindian/Early Archaic-type adaptation. In fact, both Laguna Seca (Baied, 1991) and Sajama (Ybert and Miranda, 1984) pollen records indicate high puna envi- ronments dominated by open grasslands patched with still relatively large water reservoirs during the early- Holocene. Nonetheless, the early Holocene sedimentolog- ical, lake-level, and pollen records show a reduction in the water budget whkn compared to the late-Pleistocene lake level records of the puna. Pollen records (Graf, 1981 a; Baied, 1991) suggest for this interval a cool and moist climate with precipitation well above the present. It is possible that plant communities may have resembled those found today in the Moist Puna. The fact that the first human occupations have been dated as starting sometime between 10,000 and 9,000 yr B.P. appears not to have any direct correlation to climate change, as the plant communities may have been established two thou- sand years prior to this early occupation. Furthermore, late-Pleistocene/early Holocene Dry Puna ecosystems would have been more receptive to human occupation than they were during the mid-Holocene and than they are today.

The pollen and lake-level records indicate a gradual transition towards drier and warmer climates, a process that starts at 11,000 yr B.' and becomes established between 8,000 and 7,000 yr B.P. Thus, the disruptive Pleistocene/Holocene transition, seen, for exampie, at Barro Negro (23" S; 3,900 m) (Fernandez et al., 1991), is not obvious in the Laguna Seca record some 5" to the north. The replacement of the American horse (Hi@i- dion) at Barro Negro (Fernandez, 1986 a, b; Fernandez et al., 1991) may also be seen as a product of latitudinal differences in late-Pleistocene atmospheric circulation, producing more disruptive conditions in the southern latitudes of the puna belt.

At Laguna Seca, the first major environmental change

is recorded between 8,000 and 6,500 yr B.P. Evidence of increasing desiccation is seen in the transition from lacustrine to peat sediments, 14C dated to 7,000 yr B.P., setting a major change in the water budget of the high plateau (Altiplano). A reduction in the depth of Lake Titicaca shortly after 8,000 yr B.P. has been recorded which lasted for approximately 4,500 years. The Sajama section studied by Ybert and Miranda (1984), while not showing a change in sedimentation, suggests the estab- lishment of a dry phase by 7,000 yr B.P., while data from El Aguilar in northwestern Argentina indicate a similar event at 8,000 yr B.P. (Markgraf, 1985). Although still not confirmed by the archaeological record, mid-Holocene dry conditions in the Dry Puna had the potential of sustaining a hunter adaptation in the High Andes. Environmental conditions may have been similar to the present ones, with sufficient forage resources to support a diverse game animal population. Ash flows from re- gional active volcanoes and the variable accumulation of tephra in different sectors of the puna may have played a limiting role to human and animal populations during this period. Although not substantiated in the Laguna Seca sediments, tephra accumulation in cave sites, radio- carbon dated to the middle Holocene, has been observed (Santoro and Nufiez, 1987).

Dry environmental conditions prevailed between 6,500 and 5,000 yr B.P. with a short wet episode following. The first indication of possible human manipulation of the environment occurs at about 4,000-3,000 yr B.P., sug- gested by an increase in the percentage of herbaceous taxa which may indicate the onset of native camelid pastoralism. A similar pollen sequence is recorded at El Aguilar (Markgraf, 1985) where shifting percentages of herbaceous taxa at the expense of Gramineae and Com- positae pollen suggest manipulation of forage resources. Botanical studies conducted on present-day vegetation in the Bolivian plateau (Ruthsatz, 1983; Ruthsatz and Fisel; 1984) have shown a process of species replacement due to selective grazing by domestic camelids and sheep.

The Laguna Seca paleoenvironmental data are but a preliminary step towards providing a framework within which we can study how the impact of human activity has affected the evolution of puna ecosystems. Continuing research will be directed, not only to further refine pollen and paleoenvironmental records, but also to refine the details of past human-environment interaction through the study of botanical and faunal remains from archae- ological sites.

C. A. BAIEDAND J. C. WHEELER/ 155

ACKNOWLEDGMENTS

Fieldwork in Chile was made possible by the coopera- tion of Mario Rivera Diaz and Calogero Santoro Vargas, professors of Archaeology a t the Universidad d e Tarapaca (UT) in Arica; Rudolf Thomann of the Institute of Agriculture, Universidad d e Tarapaca; and Eduardo Nufiez Araya of the Corporacion Nacional Forestal of Chile (CONAF). A joint agreement between U T and CONAF covered all necessary permits for this research. A fellowship ( to C. A. Baied) of the Perez Companc Foundation of Argentina, funds of the Explorers Club,

REFERENCES Aldenderfer, M., 1988: Middle Archaic Period domestic archi-

tecture from southern Peru. Science, 241: 1828-1830. Baied, C., 1991: Late-Quaternary Environment, Climate, and

Human Occupation of the South-Central Andes. Ph.D. dis- sertation, University of Colorado. Boulder.

Barfield, L., 1965: Recent discoveries in the Atacama Desert and the Bolivian Altiplano. American Antiquity, 27: 93-100.

Bernhardson, W., 1986: Campesinos and conservation in the Central Andes: indigenous herding and conservation of the Vicuna. Environmental Conservation, 13 (4) : 31 1-318.

Bustos E. and Veloso, A. (eds.), 1982: El Hombre y los Ecosistemas de Montana. Man and the Biosphere Programme B 6: El Ambiente Natural y las Poblaciones Humanas de 10s Andes del Norte Grande de Chile (Arica, latitude 18"28'S). Mon- tevideo, UNESCO.

Crosby, A. W., 1972: The Columbian Exchange Biological and Cultural Consequences of 1492. Westport, Greenwood Press.

, 1986: Ecological Imperialism: The Biological Expansim of Europe, 900-1900. Cambridge. Cambridge University Press.

Dauelsberg, P., 1983 a: Tojo-Tojone: un paradero de cazadores Arcaicos. Chungara, 11: 11-30.

, 1983 b: Investigaciones arqueologicas en la Sierra de Arica, Sector Belen. Chungara, 11: 63-83.

Dransart, P., 1991: Llamas, herders and the exploitation of raw materials in the Atacama Desert. World Archaeolo~, 22(3): 304-319.

Druss, M., 1978: Environment, Subsistence Economy, and Settlement Patterns of the Chiuchiu Complex (ca. 2700 to 1600 B.C.) of the Atacama Desert, Northern Chile. University Microfilms Inter- national, Ann Arbor.

Faegri, K. and Iverson, J., 1975: Tmtbook ofPolla Analysis. Haffner, New York.

Fernandez, J., 1986 a: Hippidion: New data on the now extinct American horse in northwest Argentina: palaeoenvironment and paleoclimatic implications. Current Research in the Pla'sto- cene, 3: 64-74.

, 1986 b: Reemplazo del caballo americano por camelidos en el limite Pleistocenico-Holocenico de Barro Negro, Puna de Jujuy. Relaciones de la Sociedad Argentina de Antropologia, 16(1984-1985): 137-152.

Fernandez, J., Markgraf, V., Panarello, H., Albero, M., Angiolini, E., Valencio, S., and Arriaga, M., 1991: Late Pleistocene/

Early Holocene environments and climates, fauna, and human occupation in the Argentina Altiplano. Geoarchaeology, 6: 251-272.

Flores Ochoa, J. A., 1977: Pastures de Puna. Lima, Instituto de Estudios Peruanos.

, 1982: Causas que originaron la actual distribuci6n espacial de las alpacas y llamas. In Millones, L. and Tomoeda, H. (eds.), El Hombre y su Ambiente en los Andes Centrales. Osaka, Senri Ethnological Series 10, National Museum of Ethnology, pp. 63-92.

Sigma Xi-The Scientific Research Society, the Ibero-Latin American Studies Center, and the Department of Anthro- pology and Graduate School of the University of Colo- rado, all supported this investigation. Additional funding was provided by the FONDECIT/CONICYT, National Research Council of Chile, Grant Number 89-12 to C. Santoro Vargas, Coordinator. J. C . Wheeler gives special thanks to Angus Russel and the staff of the Macaulay Land Use Research Institute.

Francis, P. and Self, S., 1987: Collapsing volcanoes. Scientific Amkcan, 255(6): 90-97.

Garmendia, A. E., Palmer, G. H., DeMarrtini, J. C., and McGuire, T. C., 1987: Failure of passive immunoglobulin transfer: a major determinant of mortality in newborn alpacas (Lama pacos). American Journal of Veterinary Research, 48(10): 1472-1476.

Graf, K., 1981 a: Palynological investigations of two Post-glacial peat bogs near the boundary of Bolivia and Peru. Journal of Biogeography, 8: 353-368.

-, 1981 b: A Quaternary pollen profile from the Upper Valley of Cochabamba (Bolivia).In Evenson, E., Schlochter, C., and Rebassa, J. (eds.), Tills and Related Deposits. Rotterdam, Balkema, pp. 365-372.

Hansen, B., Wright, H. Jr., and Bradbury, J. F'., 1984: Pollen studies in the Junin area, Central Peruvian Andes. Geological Society of America Bulletin, 95: 1454-1465.

Hesse, B., 1982 a: Archaeological evidence for Camelid exploita- tion in the Chilean Andes. Saugethierfiundliche Mitteilungen, 30: 201-211.

, 1982 b: Animal domestication and oscillating climates. Journal of Ethnobiology, 2: 1-1 5.

, 1986: Buffer resources and animal domestication in prehistoric northern Chile. Archaeozoolo@ca (Melanges): 73-85.

Heusser, C., 1971: Pollen and Spores of Chile. Tucson, University of Arizona Press.

Hoffstetter, R., 1986: High Andean mammalian faunas during the Plio-Pleistocene. In Vuilleumier, E and Monasterio, M. (eds.), High Altitude Tropical Biogeography. Oxford University Press, pp. 218-245.

Kalin-Arroyo, M., Villagran, C., Marticorena, C., and Armesto, J., 1982: Flora y relaciones biogeograficas en 10s Andes del norte de Chile (18"-19" S). In Bustos, E. and Veloso, A. (eds.), El Hombre y los Ecosistemas de Montana. Two Volumes, Man and the Biosphere Programme, 6: El Ambiente Natural y las Poblaciones Humanas de 10s Andes del Norte Grande de Chile (Arica, Lat. 18" 28's). Montevideo, UNESCO, pp. 71-92.

Kalin-Arroyo, M., Squeo, E, Armesto, J. J., and Villagran, C., 1988: Effects of aridity on plant diversity in the northern Chilean Andes: results of a natural experiment. Annals o f t h Missouri Botanical Garden, 75: 55-78.

Lavallee, D., Julien, M., Wheeler, J. C., and Karlin, C., 1985: Telunnachay Chasseurs et Pasteurs Pre'histmiques des Andes I. Paris, Editions Recherches sur les Civilisations, ADPE

Markgraf, V., 1985: Paleoenvironmental history of the last 10,000 years in northwestern Argentina. Zentralblatt fur Geologie un Palaontologie 11/ 12: 1739-1749.

- , 1987: Paleoclimates of the southern Argentine Andes. Current Research in the Pla'stocene, 4: 150-157.

Markgraf, V. and D'Antoni, H., 1978: Pollen Flora of Argentina. Tucson, University of Arizona Press.

Matos, R., 1975: Prehistoria y ecologia humana en las punas de Junin. Revista del Museo Nacional, 41: 37-74.

Miller, G. R. and Gill, L. A., 1990: Zooarchaeology at Pirincay, a Formative Period site in highland Ecuador. Journal of Eeld Archaeology, 17: 49-68.

Moore, K. M., 1988: Hunting and herding economies on the Junin Puna. In Wing, E. S, and Wheeler, J. C. (eds.), Economic Prehistory of the Central Andes. Oxford, B.A.R. International Series 427, pp. 154-166.

, 1989: Hunting and the Origins of Herding in Peru. Ph.D. dissertation. Department of Anthropology, University of Michigan, Ann Arbor, Michigan.

Murra, J. V., 1965: Herds and herders in the Inca State. In Leeds, A. and Vayda, A. I? (eds.), Man, Culture and Animals. American Association for the Advancement of Science Publication, 78: 185-215.

Nuliez, L. and Santoro, C., 1988: Cazadores de la Puna Seca y Salada del area centro-sur andina (norte de Chile). Estudios Atacamenos, 9: 11-60.

Pozorski, S., 1979: Late prehistoric llama remains from the Moche Valley, Peru. Annals of the Camgie Museum, 48(9): 139--170.

Ramirez, A,, 1987: Alpaca Clostridium perfn'ngens Type A Enter- otoxemia: Purification and Assays of the Enterntoxin. Ann Arbor, University Microfilms International, Ann Arbor, Michigan.

Ravinnes, R., 1967: El abrigo de Caru y sus relaciones con otros sitios tempranos del sur del Peru. Nawpa Pacha, 5: 39-57.

-, 1972: Secuencia y cambios en 10s artefactos liticos del sur del Peru. Revista del Museo Nacional, 38: 133--184.

Rick,J. W., 1980: Prehistoric Hunters of the High Andes. New York, Academic Press.

Rivera, M. A., 1975: Una hipotesis sobre movimientos po- blacionales altiplanicos y transaltiplanicos a las costas del norte de Chile. Chungara, 5: 7-31.

Ruthsatz, B., 1983: Der einfluss des menschen auf die vegetation und standorte arider tropischer hochgebirge am beispiel der hochanden. Bet: dt. Botan. Ges., 96: 535-576.

Ruthsatz, B. and Fisel, U., 1984: The utilization of natural resources by a small community on the highlands of Bolivia and its effects on vegetation cover and site conditions. Erdwissenschaftliche Forschung, XVIII: 21 1-234

San Martin, E and Bryant, E C., 1987: Nutrition de bs Camelidos Sudamericanos: Estado de nuestro conocimiento. Lubbock, Texas Tech University Technical Article T-9-505.

Santoro, C. and Chacama, J., 1982: Secuencia cultural de las tierras altas del area centro-sur Andina. Chungara, 9: 22-45.

, 1984: Secuencia cultural de las tierras altas del extremo norte de Chile. Estudios Atacamenos, 7: 85-103.

Santoro, C. and Nuiiez, L., 1987: Hunters of the Dry Puna and the Salt Puna in northern Chile. Andean Past, 1: 57-109.

Shimada, M. and Shimada, I., 1985: Prehistoric breeding and herding on the north coast of Peru. American AntiquiQ, 50: 3--26.

Torres, H., Hernandez, L., Weber, C., Palma, R., and von Buch, K., 1978: Plan para el Manejo del Parque Nacional Lauca. Santiago, Corporacion Nacional Forestal Publicacion Tecnica No. 6.

Troll, C., 1968: The Cordilleras of the Tropical Americas. In Troll, C. (ed.), GeeEcology of the Mountainous Regions of the Tropical Amkcas. Colloquium Geographicum Band 9, Geo- graphischen Institut, Universitat Bonn, pp. 15-56.

Troncoso, R., 1982: Evaluacion de la Capacidad de Carga Animal del Parque Nacional Lauca. Informe de Consultoria. Corpora- cion Nacional Forestal, Arica.

van der Hammen, T. and Noldus, G. W., 1985: Pollen analysis of the Telarmachay Rockshelter. In Lavallee, D., Julien, M., Wheeler, J. C., and Karlin. C. (eds.), Telarmachay Chasseurs et Pasteurs Prehiston'ques des Andes I. Paris, Editions Recherches sur les Civilisations, ADPF, pp. 379-387.

Wachtel, N., 1977: The Vision of the Vanquished. Barnes and Noble, New York.

Wheeler, J. C., 1984: On the origin and early development of camelid pastoralism in the Andes. In Clutton-Brock, J. and Grigson, C. (eds.) Animals and Archaeology, Vol. 3,Early Herders and TheirFlocks. Oxford, British Archaeological Reports, BAR International Series 202, pp. 395-410.

, 1985: De la chasse a I'elevage. In Lavallee, D., Julien, M., Wheeler, J. C., and Karlin, C., Telarmachay Chasseurs et Pasteurs Pre'historiques des Andes I. Paris, Editions Recherches ,

sur les Civilisations, ADPE pp. 63-79. , 1988 a: Origin and evolution of the South American

Camelidae. Selected Papers Westem Veterinary Confmce. Western Veterinary Conference, Las Vegas, pp. 290-300.

, 1988 b: Llamas and Alpacas of South America. Selected Papers Western Veterinary Confmce. Western Veterinary Con- ference, Las Vegas, pp. 301-310.

, 1991: Origen, Evolucion y Status Actual. In Fernandez-Baca, S. ( ed.), Avances y Perspectiwas del Conocimiento de los Camelidos Sud Ama'canos. Santiago, Oficina Regional de la FA0 para America Latina y el Caribe, pp. 11-48.

Wheeler, J. C. and Dennis, B. J., 1988: Analysis of animal bones from Patapatane, Chile. Unpublished manuscript.

Wheeler, J. C., Russel, A. J. F., and Stanley, H. F., in press: A Measure of Loss: Prehispanic Llama and Alpaca Breeds. Proceedings of the World Meeting on Domestic Animal Breeds Related to the Discovery of America, Cordoba, Archiuos de Zootemia.

Wheeler Pires-Ferreira, J. C., Pires-Ferreira, E., and Kaulicke, I?, 1976: Preceramic animal utilization in the Central Peruvian Andes. Science, 194: 483-490.

Wing, E. S., 1974: Animal domestication in the Andes. In Reed, C ~ A .(ed.), Origins of Agrinclture. The Hague, Mouton, pp. 837-859.

, 1986: Domestication of Andean mammals. In Vuilleu-mier, E and Monasterio, M. (eds.), High Altitude Tropical Biogeography. New York, Oxford uni;ersity Press, pp. 246-264.

Wingenroth, M. and Heusser, C.,1985: Flora Polinica de la Queb rada de Matienzo, Mendoza, A~ent ina. Mendoza, Consejo Nacio- nal de Investigaciones ~ienGficas y Tecnicas.

Winterhalder, B. and Thomas, R. B., 1978: Geoecology ofSouthern Peru. Occasional Paper, 26. Inst. of Arctic and Alpine Re- search, Boulder, Colorado.

Yacobaccio, H. D., 1985 a: Explotacion complementaria de recursos en sociedades cazadoras-recolectoras Sur-Andinas. Cuadernos del Institute Nacional de Antrojbologia, 10 (1983-1985) : 493-514.

, 1985 b: Una adaptacion regional de cazadores-recolectores en 10s Andes centro-sur. Relaciones de la Sociedad Argentina de Antrojbologia, lG(1984-1985): 165-173.

Ybert, J. I?, 1984: Diagramme sporopollinique de la Coupe Holocene du Rio Chuquiaguillo sur I'Altiplano Bolivien. Cahiers ORSTOM (Geologic), 14: 29-34.

, 1987: Spectres palynologiques de tourbieres et d'sedi- ments lacustres de la fin du Pleistocene et de 1'Holocene des Andes de Bolivie. Geodynamique, 2: 108--109.

Ybert, J. P. and Miranda, Y., 1984: Analisis palinologico de un corte del Sajama. Actas, Segundo Conpso Geologic0 de Bolivia, pp. 691-695.