The threat of environmental degradation and the prospects for conservation and sustainable development in Latin America are interlaced evermore within the worldwide web of the global economy. Signals, shifts, swings, and sometimes seizures in local and global economic systems daily set in motion the human activities that lead to environmental change in the region and that reset many conditions controlling the prospects of conservation and development, i.e., conservation-with-development. Surely the orb of a worldwide economy is not completely new for most Latin Americans and for that matter their environments. But, with equal certainty, the current period of neo-liberal economic reform spins a setting for environments and environmental management unlike its historical precedents.

Driven by aggressive trade policies and government deregulation and privatization, markets are clearly triumphing in the new economies of Latin America. The Andean countries--Venezuela, Colombia, Peru, Ecuador, and Bolivia--differ only in detail with respect to the trend toward further economic globalization found throughout Latin America. In each of these countries, rapidly growing markets for goods, labor, and capital presently fuel impressive economic growth in aggregate terms and even a more prodigious promise of future rewards to be reaped. But what of the environmental consequences and conservation implications of local ties to these expanded global economies?

My research has led me to examine the changing fortunes of two environmental resources that play crucial roles in rural ecosystems and livelihoods in the Andean countries, soils and the biological diversity--or biodiversity--of agricultural plants. Soils are crucial to many of the 50 million-plus people living in the Andes. Yet severe soil erosion in many cases hinders rural development and exacerbates poverty. It fuels unplanned urban growth and social crises such as the coca-cocaine industry and terrorist insurgency and threatens to degrade the major watersheds of South America through siltation and increased flooding.

The biodiversity of Andean crops provides nutritional, cultural, and agroecological resources to the region's mountain-based farmers. And this biodiversity surpasses that of any other world region. Peasant farmers residing in the Andean highlands sow at least forty crop species that evolved there prior to the 16th century onset Spanish rule (note 1.). Even more staggering are the number of diverse varieties belonging to the major Andean crops: cultivated potatoes hold as many as 5000 landraces while maize in Peru supports as many as 6000 cultivars, a bewildering biological spectrum that exceeds any other country and that contains roughly one-quarter of the world's main maize groups. Plentiful minor crops that are uncommon or rare elsewhere suffuse the diversity of this mountain farming yet further: for example, the tuber-bearing triad of ulluco, mashua, and oca, the grain-yielding quinoa and amaranth, and the leguminous tarwi or Andean lupine.

In studying each of these resource systems I have directed my attention to their ecological and biophysical properties, their management by land users, and changes incurred over time, including recent alterations due to strengthening ties to national and international markets, in short, the global economy. Each case study involves a non-deterministic style of analysis insofar as peoples' environment-related behavior is recognized not to relate in a deterministic fashion to human population numbers, technology, culture, or political economic organization. Various studies show soil erosion, for example, resulting from either population increase or decrease though neither is necessarily the case. Concrete conditions pressuring environment-related changes in farmer behavior must therefore be identified rather than being inferred on the basis of assumed relationships. Also, each study is based on analysis undertaken at the regional scale. Many studies indicate that the human-induced causes most responsible for changes in soils and biodiversity are likely to vary markedly across even short distances and thus that studies of these environmental features integrate a local scale of analysis.

II. A CASE STUDY OF SOIL EROSION AND CONSERVATION IN THE BOLIVIAN ANDES: THE CALICANTO WATERSHED

The 64 km2 watershed of the Calicanto River is one of the largest in Esteban Arce Province (Cochabamba Department in central Bolivia) and currently sites construction of an intermediate-size dam (charge capacity 100,000 m3) (note 2.). Draining a rugged upland west of Tarata village, the Calicanto River debouches alongside the village into the montane basin known as the "High Valley" (Valle Alto). The rugged terrain of the upland watershed together with the Tarata alluvial fan at its base have been peopled for several thousand years, although land use patterns in each area differed. Rainfed potato production and grazing have predominated in the mountainous upland between 2700 and 3500 masl while maize-growing based on floodwater irrigation techniques has characterized the lower alluvial fan. During the colonial period, both areas were occupied by agricultural estates (haciendas), although subsequently the maize-growing lands in the "High Valley" were acquired by independent peasant smallholders beginning in the late 19th century.

Considerable soil erosion had occurred in the Calicanto watershed prior to 1953 (note 3.). Water erosion had produced sizeable gullies (as large as 8-10 m. deep, 20 m. wide, and 2.2 km. long) and rill and sheet erosion were common. The formerly wide channel of the lower Calicanto river on the Tarata alluvial fan was filled with alluvial terraces built of river-transported deposits carried from the watershed. Wind erosion too was considerable. Much soil loss that occurred in the Calicanto watershed prior to 1953 was set in motion by erosive conditions in the physical environment. Steep slopes, semi-arid climate, and high winds that funnelled between the "High Valley" and the adjacent Santivañez basin disposed the watershed toward severe soil erosion (note 4.).

Various types of land use activities also had accentuated soil erosion in the watershed for centuries prior to 1953, as the estate owners there produced potatoes, wheat, and livestock for regional and national markets. In addition, estate owners carried out a substantial trade in firewood, an activity remembered by present inhabitants and documented in testimonials. Yet the extent of soil erosion in the Calicanto watershed prior to 1953 was less than at present (note 5.). As discussed in the following section, soil loss there has continued at a high rate and has probably worsened during ensuing decades.

Soil erosion in the Calicanto watershed during recent decades has been severe. Rough approximations using the Universal Soil Loss Equation (USLE) yielded estimates of 114 metric tons/hectare/yr. in the valley floor and 173 metric tons/hectare/yr. on the valley sideslopes. Although the USLE method offers limited accuracy in such settings as the Calicanto watershed, high estimates are confirmed by a variety of other evidence. Three proofs of severe recent erosion in the watershed are the following: 1) abundant soil surfaces with elevated pedestals, capped by rocks and vegetation, that have formed recently. 2) abundant gullies with freshly cut walls; and 3) truncated soil profiles with exposed subsurface horizons, including widespread surface areas of rocks formerly in the subsoil.

Numerous changes in land use during the past four decades led to worsening soil erosion in the Calicanto watershed. Reduced implementation of tasks that either directly or indirectly had served to conserve soil were especially noticeable. The construction and maintenance of diversion ditches, protective rock walls, and agricultural terraces in particular diminished during the period 1953-1991. Forty-six percent of peasant households residing in the watershed quit practicing at least one of these soil conservation techniques. These changes formed part of the noticeable lessening of labor inputs per unit of agricultural land. Disintensification of labor inputs was also apparent in other changes that occurred during recent decades such as elimination of the second mounding of furrows in potato and maize agriculture. Although this latter technique was viewed primarily as a cultivation practice that enhanced crop yields, it also had served to limit soil loss.

Reduced practice of soil conservation techniques occurred as peasant inhabitants of the Calicanto watershed encountered a growing shortage of labor available for their local farm production. This labor shortage was especially acute among the households least able to recruit labor, due to small family size and to limited money resources for paying day-wages (Table 3). Using the local designations of "poor peasants" and "middle peasants" (a distinction based on the land, labor, and livestock possessed by a household) in order to estimate labor availability, it is apparent that a disproportionate share of persons abandoning soil conservation techniques were labor-scarce households of "poor peasants." Unable to recruit sufficient labor for labor-intensive conservation tasks, these households decided to by-pass them. These decisions permitted the households to maintain or even increase field areas and extent of cropping while less labor was allocated per unit area, with the impact of the latter concentrated on abandonment of labor-intensive conservation techniques.

Demographic conditions and wage levels underpinned labor shortages in peasant production in the Calicanto watershed, although they were not the most direct causes of these shortfalls in labor supply. Population in the watershed grew slightly during the 1953-1991 period.(note 6.) Together with adjacent areas in the province of Esteban Arce, an average growth of .21% occurred each year between 1950 and 1975 (INE 1982), and a similar rate was likely to have continued subsequently. Low-moderate population growth was the result primarily of permanent out-migration, whereby inhabitants of the Calicanto watershed and nearby areas relocated outside the province, especially to the capital city Cochabamba (1950 population 82,397, 1976 population 222,067l; INE 1982). Demographically, low-moderate growth in the Calicanto area did not provide the additional hands that might otherwise have countered the local labor shortage.

Wage rates in farm work also underpinned the labor shortage in peasant production. Day wages offered in the Calicanto watershed, like other nearby upland areas, were the lowest in the region (in 1991, $1.52/day). This wage rate was significantly less than in the nearby "High Valley" and in Tarata village where day laborers earned $2.42 in 1991 (160% of the upland rate). Wage rate differences reflected the lower profitability of agriculture in the upland, where field owners chose to forego labor inputs rather than increase wages. At the same time, other measures for recruiting extra-household labor, such as reciprocal labor exchange, had become less effective. Although reciprocal labor exchange was widely practiced throughout the 1953-1991 period, the pool of workers participating in these local networks shrunk over time. This trend toward fewer workers laboring in upland production was echoed in the conditions within households, where the origins of the local labor shortage were perhaps most discernable.

Within the peasant households of the Calicanto watershed, labor was increasingly diverted to economic activities other than local farming and livestock-raising during the 1953-1991 period. Following the national Agrarian Reform administered in the watershed during the mid- and late 1950s, former tenant households gained land ownership and direct control of their personal labor. Like peasant households in other Cochabamba regions, Calicanto residents responded to economic opportunities in the post-reform period by increasing labor-time in commercial activities, including commercial farming and livestock-raising. In addition, they expanded labor allocation in other work activities such as piecework knitting and, most importantly, working for day wages. Although the latter activity included local employment, either in Tarata village or the nearby "High Valley," the majority of day wage opportunities involved extra-regional labor markets.

Work in extra-regional labor markets entailed temporary migration, an activity that increasingly characterized the economic strategies of Calicanto peasants during the 1953-1991 period. By 1991, approximately 82% of Calicanto households included at least one person who migrated for longer than two months each year in search of work. Among young adults (especially unmarried individuals), the incidence of temporary migration neared 100%. The character of temporary labor migration from the watershed underwent significant changes between 1953 and 1991, led by regional economic changes discussed in Section IV. Rapid growth of the Cochabamba urban area drew increasingly large numbers of migrants throughout the period. In addition, much temporary migration during the 1960s and the '70s was directed toward lowland Santa Cruz. In the 1980s, this periodic labor movement became oriented principally toward the Chapare region. In 1991, the Chapare predominated as the major destination for Calicanto migrants, with many households sending one or more persons to work there during several intervals each year.

The frequency and overall duration of temporary migration from the Calicanto watershed rose during the 1980s. Labor markets in the Chapare region were wider, deeper, and considerably more accessible than previous ones. As temporary labor migration increased, Calicanto households shifted responsibility for local production tasks. In approximately 74% of households, women undertook greater amounts of farm work. They both carried out previously male tasks (plowing, threshing, mounding, wheat harvest) and amplified their labor-time in others (maize and potato harvest, household chores).7 The greater labor-time of women in farm work reflected seasonal as well as year-round aspects of labor shortages in local production. Work routines were most transformed seasonally at peak periods of labor demand, such as plowing, planting, and harvest. But year-round work routines also changed, and the withdrawl of labor from conservation techniques represented the most important of these with respect to resource management.

Worsening soil erosion in the Calicanto watershed has mutually reinforced labor shortages in local production. Decreased agricultural productivity due to worsened soil loss has led farmers to extend field areas and to expend less labor per area cultivated. This mutually reinforcing relation hinged on the condition of labor shortage in the peasant economy, for otherwise intensification and perhaps soil conservation, rather than disintensification, might have occurred. Decreased productive capacity due to erosion pressured households to diversify more widely their economic activities, expend greater labor in non-farm activities, and make erosion-inducing modifications in local production.

At the same time, peasant inhabitants of the Calicanto watershed have remained highly aware of both soil erosion processes and conservation measures. Their well-developed knowledge of erosion processes has been founded on the observation of erosion-induced changes in soil properties, the ecological invasion of erosion-tolerant plants, and the formation of eroded landforms. Primary rationale for this local knowledge has been to improve the productive capacity of fields and grazing lands. Local conservation measures have been based on combined erosion-protection and sediment-trapping. Detailed knowledge of soil erosion and conservation measures has continued to be widely practiced in the local language (Quechua) among both elderly persons and young adults. With respect to the causes of Calicanto erosion, this evidence amply negates suggestion that broad-based diminution of soil management knowledge has occurred.

Labor shortages in peasant land use in the Calicanto watershed arose due to the growing diversion of labor-time away from local production. This economic diversification resulted from cumulative decisions made by peasant farmers in response to unfavorable economic conditions. Since 1953 government policies promoted capital accumulation in economic activities, regions, and social sectors other than the agropastoralism carried out by peasant producers in the Calicanto watershed and in other highland regions. Disadvantageous policies were manifested particularly in declining terms of trade. Peasants reacted by expanding non-traditional labor activities in efforts to maintain or improve their accustomed standards of living. Termed the "scissors effect" or the "simple reproduction squeeze," these chronically unfavorable economic conditions and associated pressures on peasant producers in the Calicanto watershed and other highland Cochabamba areas took on distinct forms during recent periods (1953-1980, 1980-1985, 1985-1991).

Following downfall of the ruling oligarchy and rise to state power of the National Revolutionary Movement party in 1952, various governments in Bolivia implemented economic policies to convert the nation's economy from a traditional export enclave to a modern industrial one through Import Substitution Industrialization. Modernization of peasant production in the Andean highlands was planned to result from the agrarian reform that the national government legislated in 1953. Between that year and 1970 approximately 11.5 million of the 32 million hectares designated for reform were deeded to 433,000 peasant families. Although highland peasants gained rights to allocate their labor free of landlord claims and to own a portion of former estate land, the 1953 agrarian reform did not lead to economic modernization. During decades following the reform, peasants inhabitants of the Calicanto watershed and other highland Cochabamba regions faced diminishing economic prospects for agricultural production and livestock-raising, leading them to divert growing labor-time to non-farm economic activities.

Unfavorable terms of trade, limited availability of farm credit, and a limited land base plagued peasant production in highland Cochabamba and fueled the diversification of household economies during the 1953-1980 period. For peasants, the changing prices gained by their land-based products vis-à-vis the costs of consumption items declined by more than 50% between 1963 and 1976. Moreover, peasant agriculture was placed at a considerable disadvantage with respect to non-peasant farm production (commodities originating in the former showed only 40% of price increases registered in the latter). Disadvantageous conditions were found too in the limited credit available to peasant land users, who received less than half of the agricultural credit dispensed in the country yet produced a majority of farm goods. Unfavorable terms of trade and deficient agricultural credit were created through government policies that ensured inexpensive foodstuffs for expanding numbers of poor urban dwellers and low-cost labor for city-based industries and rural agribusiness. Peasant producers were unable to counter the unfavorable markets for goods and capital though increases in farm area; by the early 1980s more than half the region's peasants counted farm and grazing lands less than 3 hectares.

Temporary labor migration increasingly distinguished the livelihood strategies of peasant producers in highland Cochabamba during the 1960s and the '70s. Much temporary migration was directed toward the eastern lowlands of Santa Cruz Department, where as many as 40,000 temporary migrants labored annually (note 8.). Economic growth in the Santa Cruz lowlands during the 1960s and the 1970s was the planned outcome of government economic policy, which had conferred large land grants through the agrarian reform (note 9.). Government import-substitution policies stimulated the production in Santa Cruz of rice, sugar, and cotton, with the addition in the 1980s of soybeans when export-oriented crops received stimulus. Extreme unevenness in regional development thus contrasted the Santa Cruz lowlands and highland Cochabamba, with many peasant households in the latter periodically sending migrants to work in the lowland realm of wage-earning opportunities.

Urban populations and industry too expanded considerably during the 1960s and the '70s through the stimulus of import substitution policies. Between 1950 and 1976 the urban populace increased from 27 to 42% of the national total. Although manufacturing employment (textiles, shoes, urban-based agroindustry) grew moderately, their expansion was exceeded by the growth of "service activities," mostly informal sector jobs (construction, street vending, domestic service). These latter opportunities for employment drew migrant streams from the struggling peasant sector, some permanently and others for temporary intervals. As many peasant households integrated work in the city by one or more members-in many cases girls and young women who labored as domestic servants-into diversified economic strategies, they continued to carry out farm production in such rural regions as the Calicanto watershed. By the late 1970s, the extent of non-farm economic activities provided roughly one-half the income of typical peasant households. It also is important to note that economic diversification included greater commercialization of specialized agricultural and livestock products (e.g., potato production during the off-season).

A national economic crisis in Bolivia was precipitated in the late 1970s and the early '80s by collapse of the international tin market, weakening demand for other major exports (cotton, natural gas), and the concurrent rise of service payments on the country's mounting debt. Between 1980 and 1984 the GNP fell 17% and by 1985 the annual inflation rate reached a record 22,000%. Yet although the capitalist sector of the national economy fell during the early 1980s, its condition remained more favorable than that of peasant agriculture and livestock-raising. Between 1974 and 1984, for instance, annual growth of 4.2% in the capitalist sector far outstripped estimated growth of .8%/yr. in peasant production. Faced with worsening constraints, peasant producers in highland Cochabamba further diversified their economies. New income-earning opportunities appeared in the booming coca-cocaine industry, which flourished during the same period that the official economy collapsed. Motivated by growing international demand and infused with commercial skills and capital ventured by wealthy businessmen (many from Santa Cruz and Beni Departments), the coca-cocaine industry expanded production severalfold each year during the early 1980s (note 10.).

The lowland Chapare region of Cochabamba became the geographic center for expanding coca leaf production during the late 1970s and the early 1980s. Production of coca leaf (raw material for cocaine) was carried out by peasants who labored intensively in cultivation and risked repression, extortion, and field eradication. Many peasant laborers in the Chapare migrated there for short periods from the nearby highlands of Cochabamba. Widespread, close links between the two regions were represented in the annual movement of an estimated 400,000 persons between them (note 11.). For peasant producers in highland Cochabamba, work in either one's own plot in the Chapare or that of another landholder there was the principal means of shoring up the household economy in the face of acute economic uncertainty and thus driving further its diversification (note 12.).

Advised by the International Monetary Fund and the World Bank, Bolivia decreed a "new political economy" in 1985 that implemented structural adjustment policies according to a neoliberal model. Reform measures deregulated the country's economy, reduced state spending, and abandoned the earlier metal-export emphasis. New policies continued to disadvantage peasant agriculture and livestock-raising in highland Cochabamba (note 13.). Minimal government expenditures, for instance, were reflected in the provisioning of one agricultural extension agent for every 25,000 rural inhabitants and by increased international funding for agricultural development (62% by 1990). Pricing policies too undermined the economic interests of peasant farmers and herders, who saw their terms of trade continue to decline. In addition, certain food imports (especially wheat) claimed a significant share of urban markets that earlier were provisioned by peasant producers.

Further diversification of peasant economies in highland Cochabamba following the 1985 "new political economy" have mostly followed the earlier routes for income-earning. Temporary migration to the Chapare and the Cochabamba urban area have remained commonplace, and limited migration to Santa Cruz persisted. Although employment opportunities in the regional coca-cocaine industry (Chapare), the complex urban economy (Cochabamba), and lowland agroindustry (Santa Cruz) dropped during the late 1980s, they nonetheless maintained ample extra-regional attraction to impoverished highland peasants. In addition, economic diversification by peasant producers since 1985 entailed expanded work in piecework arrangements for textile-making (knitting and weaving).

III. A CASE STUDY OF BIODIVERSITY LOSS ("GENETIC EROSION") AND CONSERVATION IN THE PERUVIAN ANDES: PAUCARTAMBO PROVINCE

The Peruvian province of Paucartambo covers a Corsica-size area that separates the Amazon lowland from mountainous Cusco. Dissected by the Paucartambo River and its tributaries, the region contains a range of elevation-related climatic environments between 2700 and 4100 meters above sea level (masl). Moreover, the northern ("lower-valley") portion is considerably more humid than the western and southern ("upper-valley") areas due to the contraction of the easterly Cordillera Paucartambo. Notable gradients of vegetation and soil types track the elevational as well as areal axes. The peasant agricultural systems of highland Paucartambo support concentrated diversity of both maize and potatoes. Roughly 20,000 Quechua-speaking peasant farmers carry out nearly all agricultural production in the study region.

Households, which belong to peasant communities of 50-500 people, comprise the chief unit responsible for directly organizing agricultural production in highland Paucartambo. In 1987, the majority of households in the region cultivated eight fields, although this figure varies from ones posssessing zero (the rural landless) to others that produce more than 15 parcels each year. Fields are small in size (most between .1 and .5 hectares) and those of a single household tend to be scattered among a striking array of elevation-related environments--the "generalist" or "compact" mode of montane agricultural organization. Production traits of single fields-such as crop type, rotation sequence, calendar, physical environmental parameters, level and type of inputsÑtend to be shared with most nearby fields, leading to a spatial coalescence of land use into "production zones."

Similar agricultural production types coalesce into five areas or production zones that occur throughout most of the study region. Production zones partially correspond to elevation. The actual degree of areal cohesion and the dynamics involved in modifying spatial organization, however, are brought about by both informal and formal social mechanisms governing land use. The former stem principally from pressures to coordinate agriculture and livestock-raising whereas the latter entail community institutions. As a result, when households attempt to change their agricultural production, their capacity to act autonomously is often constrained. Areal changes in the two production zones containing native maize and potato cultivars (the kheshwar and yuñglla, respectively) therefore must be seen as mediated by the socio-spatial organization of land use.

Diverse calendric strategies complement the spatial layout of agriculture in highland Paucartambo. A staggered demand for labor matches the array of production-zone types managed by most single households. Temporal dispersion of work permits households to carry out more cultivation than would otherwise be possible given the labor-intensive character of peasant agriculture in the Central Andes. The capacity to provision labor for agricultural tasks depends not only on the demographic composition of the household but also on its ability to wield social and economic power in recruiting extra-household workers via reciprocal labor exchange (ayni; the most common form of payment in the region), wage labor, and payment-in-kind (minka; the least common local form of payment). Requiring distinct endowments of labor as well as land, different crops of course are maintained in cultivation only as long as households allocate sufficient resources. In highland Paucartambo, the resource-allocation strategies of households are shaped strongly by recent changes in the regional economy and society.

Increased commercialization of produce and labor has impelled most households in the study region to reallocate their land and labor resources. Commercial activities have been amplified as a result of several socio-economic changes occurring during the past two decades. Major changes include the following: the 1970 Agrarian Reform that converted most of Paucartambo from manorial estates (haciendas) to official "peasant communities" (Comunidades Campesinas, Fonseca and Mayer 1988); a declining land base; the increased availability of agricultural inputs and credit through regional markets and the state (especially the national "Agrarian Bank"); and the proliferation of "contract farming" arrangements. Altered demand conditions in the regional economy have also driven the growth of commercialization. Expansion of both the urban food market in the departmental capital of Cusco and the labor market in the neighboring lowlands of Pilcopata provided important stimuli for local change.

Diversifying the type of farm commodities and the scheduling of their production comprises the major form of agricultural change that has recently occurred in highland Paucartambo. Diversification offered one of the few opportunities whereby peasant farmers might raise their earnings. Disadvantaged by the bias of national agricultural policy in favor of commercial agriculture in the lowlands as early World War II, highland production has been weakened further by a stagnation of market demand for sierran staples during the last twenty years. Over this period, two types of commercialization have become common in highland Paucartambo agriculture. First, faced by a chronic depression in the price of staple products at the height of harvest, many peasant farmers in the region initiated cultivation of an off-season crop of potatoes (known as mahuay [Q. maway]). Secondly, the farming of barley based on contract arrangements between agriculturalists and a national-international beer company (Compañía Cervecería del Sur) became widespread. The production of both off-season potatoes and barley is now common among households in highland Paucartambo.

Temporary migration from highland Paucartambo in order to earn wages has enlarged since the early 1970s. Forged in the crucible of regionally uneven development, the expansion of temporary migration is fueled by both economic growth in the nearby selva of Pilcopata and Madre de Dios as well as stagnation of the highland economy. Unlike lowland commodities oriented towards export, highland staples such as potatoes have undergone a regular decline in the terms of trade except for a brief interim following the 1969 coup of the Revolutionary Military Government. (The drop-off reached 2% annually in the mid- and late 1970s, a rate far exceeding those of non-staple agricultural products.) Pronounced intra-regional variation in the frequency of temporary migration is combined with sharp local differences in its destination and scheduling. Most migrants from the "lower valley" area of highland Paucartambo are employed in the plantations of Pilcopata during February-March and July-August while those of certain "upper valley" communities work from January through March in the gold mines of Cinco Mil (Madre de Díos department).

Diversified commercial activities consisting principally of off-season potato production, "contract" barley farming, and temporary labor migration have forced highland Paucartambo agriculturalists to adjust their farming strategies. Adjustments in the allocation of land and labor resources have entrained disparate consequences for different native crops. The case of maize illustrates these adjustments. Present in South America by at least 4000 years ago, maize subsequently underwent evolutionary divergence there that was as multi-faceted as in Mexico and Central America. The differentiation of intra-specific maize variants was especially pronounced in the deeply dissected valleys of the Central Andes. Maturation period, which conferred ecological specialization to a wide range of bio-physical environments, diverged markedly within the Andean maize crop. Its evolutionary diversification is represented by a greater than twofold variation among the 32 well-defined races that comprise the basis of Peruvian maize.

A total of 27 maize cultivars were identified in highland Paucartambo; the regional total represents 11 races, more than one-third of the foundation underlying Peru's maize crop. Differing significantly in the combination of row number and ear length, the 27 cultivars comprise "ideal types" within an extremely complex continuum of visible ear and kernel morphologies. (The morphological complexity of maize is reflected by the local naming system which is characterized by a high level of synonymy; more than 150 different linguistic labels were found to be associated with the 27-cultivar collection.) The widespread mixing of maize types is ensured by the nearly ubiquitous intercropping of multiple cultivars within fields designated for subsistence (mean=2.9 cultivars/field; n=67). Distinct sets of intercropped cultivars are grouped according to differences in maturation period.

Three calendar-based plantings each contain a separate group of maize cultivars. Agriculturalists in highland Paucartambo refer to seed size, a relatively good indicator of maturation period, in naming the plantings: "big seed" (hatun muhu), or slow-maturing maize cultivars; "medium seed" (chaupi muhu), the intermediate group; and "small seed" (uch'uy muhu), or relatively fast-maturing cultivars. By confining each calendar-based planting to a separate field, agriculturalists guarantee that maturation period continues to differentiate intra-specific groups. The elevational range of plantings vary in conjunction with maturation period. In the lower valley, for instance, slow-maturing "big-seed" cultivars are restricted to elevations below 2850 masl. "Little seed" cultivars, on the other hand, are planted between 2700 masl and 3100 masl. (In the semi- arid upper valley of the region, maize production reaches an upper limit of 3550 masl. "Medium seed" cultivars, which possess an intermediate maturation period, occupy an elevational zone that overlaps with the other plantings.

The areal distributions of maize cultivars differ due to exchange practices. Widespread distributions, defined as occurrence in "upper-valley" areas as well as the "middle valley" and the "lower valley," are demonstrated by roughly two-thirds (17) of the cultivar total. All cultivars belonging to this group mature at relatively rapid or intermediate rates (the "small seed" and "medium seed" plantings). Their comprehensive distributions are the result of seed exchange practices that permit agriculturalists in upper-valley areas, marginal for maize production, to obtain seed frequently from their downstream counterparts. In marked contrast, slow-maturing ("big seed") cultivars are restricted to the lower valley where a warmer and more humid climate extends the growing season and permits their cultivation. As a consequence, slow-maturing cultivars evince endemic distributions within a portion of the study region while the other types are widely distributed.

Native maize is one of the most labor-intensive crops grown in highland Paucartambo. Annually, it receives 121 person-days/hectare (based on a sample of 30 fields where oxen were used for plowing and planting), a figure that exceeds previous estimates by a significant margin (e.g., 92.8 person-days/hectare and 20-50 person-days/hectare). The present study's focus on fields intercropped with multiple cultivars ("polycultivar") and its inclusion of tasks involved in post-harvest processing probably accounts for the significant differences. At least 12 distinct agricultural steps are used by maize producers in the study region (plowing, raking, planting, first mounding, second mounding, weeding, harvest, sheathing, husking, drying, shelling, storage). Finally, labor requirements associated with each of the three main maize plantings are concentrated at the beginning and end of the growing season. While the latter concentration of work occurs during May and June in the case of all three plantings, the former is more temporally dispersed due to differences of four-six weeks in planting (note 14.).

"Big seed" maize production has recently been quit by more than twenty-seven households inhabiting the lower-valley portion of highland Paucartambo (Peasant Communities of Acobamba, Totora, and Huaqanqa). The households attribute their altered agricultural strategy to a lack of labor available during the crucial period of planting and cultivating (August and September). Seasonal scarcity of labor in the lower valley corresponds closely to a pair of recent changes in the local economy. First, responding to the opportunity of obtaining wage work in the planting and harvest of rice in nearby Pilcopata, as well as the eroded capacity to earn income locally, increased numbers of young married males with families and fields now migrate between July and September (see also Baca 1985). Secondly, the demand on households for agricultural labor during the same period has grown due to the increased production of off-season potatoes, which are planted in July, August, and September.

Socio-spatial linkages in land use have begun to undermine further the production of slow-maturing cultivars. Even households not strapped by a seasonal shortage of labor have ceased producing "big-seed" fields. Inter-household concordance arises principally from the close coordination of livestock and agriculture in the local economy. Although maize production and the kheshwar production zone do not presuppose a high degree of communal or formal control (Table 3), informal agreements governing the use of land in nearby parcels abound. The practice of pasturing cattle in recently harvested and fallowed fields while avoiding damage in adjacent crop parcels encourages land users to coordinate informally the production of plots in close proximity. With many former "big-seed" parcels now cropped using different calendric regimes, the extremely long growing season of the slow-maturing maize cultivars (8-10 months) has become more incongruous with its surroundings.

Biologically, the loss of "big-seed" cultivars in the lower valley is significant due to the restricted distributions of these taxa. Representing unique racial combinations within the study region, "big seed" cultivars display restricted ranges. As a result, reducing "big-seed" plantings produces a biological loss more pronounced than would occur in the case of the other maize plantings ("medium seed" and "small seed"). It is worth noting that the racial category has been considered a useful taxonomic unit for estimating cultivar loss in the maize crop. The existing classification of maize races, however, is considered inadequate for types exhibiting traits purportedly derived from hybrid combinations. Rather than representing hybrids, such types might constitute unique forms. If subsequent taxonomic revisions demonstrate a greater evolutionary distinctness of "big-seed" cultivars, loss of native maize types in the lower valley will assume even greater biological import.

IV. IMPLICATIONS FOR SUSTAINABLE ENVIRONMENTAL MANAGEMENT

Labor shortages in environmental resource management in the Bolivian and Peruvian Andes have led to worsening soil erosion and the loss of biodiversity during recent decades. Though this role of labor scarcity cannot be assumed a priori, it likely characterizes many regions of peasant production. Similarly, though such scarcity does not derive necessarily from strengthening ties to the global economy, it often in fact does. Nonetheless, development economics has tended to overlook this condition, as it fixes attention on the labor surplus that appears from the viewpoint of the industrial and direct-wage sector and high rates there (compared to lower wage-equivalents in peasant farm production). Implications of the environmental role of labor shortages in peasant production for conservation-with-development are evaluated below with respect to three policy-related issues: (1) non-farm employment and (2) technology innovation.

Expansion of non-farm work is frequently considered necessary for policy directed toward alleviating rural poverty and promoting environmental conservation among farmers in developing countries. Findings of soil erosion and biodiversity loss due to modification of conservation practices and local labor scarcity--together with similar, combined occurrence of labor-intensive conservation practices (e.g., agricultural terraces) and "full-employment peasant economies" in many world regions--warn that this recommendation needs to be critically re-examined for conservation-with-development. Its re-examination needs to be especially thorough with regard to peasant production that takes place in environments that are subject to degradation and lacking ecological resilience (or the capacity to recover from disturbance, e.g., semi-arid mountains).

Attention to the environmental consequences of expanded non-farm labor activities is necessary also because these latter activities have expanded notably during the past decade. The increasing variety of contractual arrangements and expanding spatial dispersion found in informal sector work have shaped the character of non-farm employment. Like Calicanto inhabitants in the Bolivian Andes, peasant producers in many world regions allocate increasing labor-time to non-farm employment by migrating frequently for short periods. The serious environmental deterioration associated with this development trend suggests altering price, credit, and investment policies that typically exacerbate labor shortages in peasant production, perhaps especially with respect to land use in fragile environments.

In any case, the likely shift toward non-farm work, even if slowed, suggests that policy promoting conservation-with-development focus on the innovation of farm technologies that are both labor-saving and environmentally sound. Two aspects of technology innovation merit evaluation: possible limitations on the appropriateness of local-style technologies and the need for research and development on "blended" technologies. Criticisms of the social and environmental consequences of high-capital, low-labor Western-style technologies have led to growing interest in the advantages of local-style or so-called indigenous technologies. Innovation and adoption of local technologies are often associated with little environmental degradation and constrained socio-economic differentiation. In Andean regions, for example, the favorable environmental and socio-economic attributes of stone-wall agricultural terraces and excavated raised fields have led to policy recommendations for their expanded use. But, for peasant producers, the feasibility of adopting these technologies and continuing other labor-intensive local technologies is likely to be limited by labor shortages.

Policy making related to technology innovation for conservation-with-development would be advised, therefore, to consider the combined criteria of environmental suitability and labor requirements. Rather than assessing only low-labor Western-style technologies and highly labor-intensive local-style technologies, a third alternative emerges. "Blended technologies" that combine local and non-local attributes, such as improved oxen-drawn farm implements, in many cases offer peasant producers the advantages of both labor-savings and environmental suitability. Innovation of "blended technologies" frequently occurs among peasant farmers, although research and development coordinated closely with these producers would further enhance the innovation and diffusion of "blended technologies." Policies promoting conservation-with-development would invest soundly by supporting research on and development of this technological orientation in peasant production, which previously has been overlooked.

1. If we add to this list the numerous plants domesticated in the easterly foothills the total of Andean crops rises to at least seventy species and perhaps as many as 200. Notable variation in these estimates hinges partly on the delimitation of "domesticated plants," which strictly speaking are dependent entirely on farmers for survival, versus "cultivated plants," which are planted and harvested but may survive and reproduce in the absence of these interventions (Harlan 1975). The figure of 40 agricultural species in the highlands refers mostly to domesticates whereas the estimate of 200 encompasses many cultivated but undomesticated plants. Because of dispersals between Andean regions during the process of plant evolution I desist from using the descriptor "native," which generally indicates "a pedigree that back forever in association with a particular geographical region" (Crosby 1991: 84).

2. The dam is intended to provide irrigation for fields on the Tarata alluvial fan.

3. Various techniques were used to assess soil erosion in the Calicanto watershed. Methodologies dealing with physical environmental evidence included the examination of historical aerial photographs (1961) at a scale of 1:50,000, oral histories provided by elderly inhabitants, and the geomorphic analysis of soil profiles, channel deposits, and alluvial terraces, including dating techniques based on tree rings and carbon-14 samples. Study of land use practices, their historical change, and the articulated perceptions (discourses) of peasant inhabitants about these changes was based on extended participant observation over the course of February-December 1991, open-ended interviews, and an economic survey involving the voluntary cooperation of 100 peasant households.

4. The watershed area (64 km2) contains roughly 48% with slight and moderate slopes (0-10%) and 52% with steep slopes (greater than 10%). The average annual rainfall in the watershed exceeds slightly that of Tarata (441 mm/yr.; Pinto 1987).

5. The Calicanto watershed does not evince dramatic worsening of soil erosion concurrent with the Agrarian Reform that was legislated in 1953 and administered there during the late 1950s.

6. In 1991, roughly 450 households owned land in the Calicanto watershed (1991 population circa 2,250 inhabitants). The person-land ratio in the Calicanto watershed was 2.67 hectares per person or roughly 1.0 hectare of arable land per person based on estimates of overall population and area. But considerable disparities in landholding size existed among households.

7. Increased allocation of women's labor-time in local production did not infer their absence from temporary migration, although Calicanto women migrated less frequently than their male counterparts.

8. The resident population in Santa Cruz also grew rapidly, averaging a yearly increase of 8-10% in the 1960s and the '70s (Casanovas 1981, Maletta 1988). Because many permanent migrants in the Santa Cruz lowlands relocated from the countryside of highland Cochabamba, residents that remained in the latter regionsÑprimarily peasant producersÑwere able to establish social networks with family members and former neighbors who subsequently facilitated temporary migration.

9. Roughly 18.5 million hectares were distributed in mostly large grants (over 500 hectares) during 1971-80 (Maletta 1988).

10. The rule of the drug-trafficking narcomilitary regime of General Garc'a Mesa (1980-1982) also fueled the boom in the coca-cocaine industry. The production of coca leaf in Bolivia is estimated to have grown from about 9,000 metric tons in 1974 to 150,000 metric tons in 1985 (Maletta 1988).

11. Another plain evidence of abundant economic ties between the Chapare and the nearby highlands is the estimate that 95% of peasant colonizers in the Chapare continue to own lands in the highlands (Rivera 1990: 12).

12. Resembling earlier migration to Santa Cruz, temporary work in the Chapare by highland Cochabamba peasants was facilitated by the permanent migration of family members and former neighbors, who subsequently helped establish migration networks. As early as 1981 the number of permanent inhabitants in the Chapare numbered 83,525, 40% of whom had resided there for less than five years (Flores and Blanes 1984), and by 1985 the resident population of the Chapare reached 125,000 (Healy 1986). Most Chapare residents had emigrated from such economically stagnant rural regions of highland Cochabamba as the Calicanto watershed, which alone furnished scores of colonists.

13. Gierhake (1992) shows that although Cochabamba received public expenditures between 1987 and 1990 that were proportionate to its population, the large part of these funds were directed to the lowland Chapare region where programs attempted to reduce coca cultivation by substituting other crops.

14. Timed according to religious festivals, "big-seed" fields are planted following August 15th [Virgen de la Asunta, locally known as "Mama Asunta"], "medium-seed" fields after September 18th [the "Mama Chanca," a local patron saint], and "small-seed" fields during the month subsequent to All Saints Day ["Todos Santos," a moveable date].

ARA (Archivo de la Reforma Agraria, La Paz). "Documentos referentes al fundo Durasnuni." (Packet 2, Document 3552, March 18,1958).

Ashby, Jacquiline A. "The social ecology of soil erosion in a Colombian farming system." Rural Sociology, Vol. 50, No. 3 (1985), pp. 377-396.

Bebbington, Anthony. "Indigenous agricultural knowledge systems, human interests, and critical analysis: Reflections on farmer organization in Ecuador." Agriculture and Human Values, (xxx) (1991), pp.14-24.

Bernstein, Henry. "Notes on capital and peasantry," in John Harriss (Ed.), Rural development: Theories of peasant economy and agrarian change (London: Hutchinson, 1982), pp. 160-177.

Biggs, Stephen D., and Edward J. Clay. "Sources of innovation in agricultural technology." World Development, Vol. 9, No. 4 (1981), pp. 321-336.

Binswanger, Hans. "Agricultural mechanization: A comparative historical perspective." Research Observer, Vol. 1, No. 1 (1986), pp. 27-56.

Bishop, B. C. Karnali under stress: Livelihood strategies and seasonal rhythms in a changing Nepal Himalaya. (Chicago: University of Chicago Geography Research Papers Nos. 228-229, 1990).

Blaikie, Piers. The political economy of soil erosion in developing countries. (London: Longman Scientific, 1985).

__________. "The explanation of land degradation," in J. Ives and D. C. Pitt (Eds.), Deforestation: Social dynamics in watersheds and mountain ecosystems (London: Routledge, 1988), pp. 132-158.

Blaikie, Piers, and Harold Brookfield. Land degradation and society. (New York: Methuen, 1987).

Blasco, Mario. "La tierra en el desarrollo rural de la zona andina." Desarrollo rural en las Américas,Vol. 11, No. 3 (1979), pp. 155-164.

Blaut, James M., Blaut, Ruth P., Harman, Nan, and Michael Moerman. "A study of cultural determinants of soil erosion and conservation in the Blue Mountains of Jamaica." Social and Economic Studies, Vol. 8, No. 4 (1959), pp. 403-420.

Boserup, Ester. The conditions of agricultural growth. (New York: Aldine Publishing, 1965).

Brown, Lester R. "Eroding the base of civilization." Journal of Soil and Water Conservation, Vol. 36, No. 5 (1981), pp. 255-260.

Brush, Stephen B. Mountain, field, and family: The economy and human ecology of an Andean valley. (Philadelphia: University of Pennsylvania, 1977a).

Brush, Stephen B. "Myth of the idle peasant: Employment in a subsistence economy," in Rhoda Halperin and James Dow (Eds.), Peasant Livelihood (New York: St. Martin's, 1977b), pp. 60-78.

Carney, Judith. Indigenous soil and water management in Senegambian rice farming systems. Agriculture and Human Values (xxx) (1991), pp. 37-48.

Casanovas S., Roberto. Migración interna en Bolivia: Orígen, magnitud, y principales caracteristicas. (La Paz: Ministerio de Trabajo y Desarrollo Laboral, 1981).

CIDRE. Monografía de la provincia Esteban Arze. (Cochabamba: CIDRE, 1985).

Collins, Jane L. Unseasonal migrations: The effects of rural labor scarcity in Peru. (Princeton, NJ: Princeton University, 1988).

Collins, Jane L. "Labor scarcity and ecological change," in Peter D. Little and Michael Horowitz (Eds.), Lands at risk in the Third World: Local-level perspectives (Boulder, CO: Westview, 1987), pp. 19-37. Conservation Foundation and FAO (Food and Agriculture Organization of the United Nations), "Soil erosion survey of Latin America." Journal of Soil and Water Conservation, Vol. 9, No. 5 (1954), pp. 214-229.

Cotlear, Daniel. "Cambio institucional, derecho de propiedad y productividad en las comunidades campesinas." Revista Andina, Vol. 6, No. 1 (1988), pp. 7-67.

Crosby, Alfred W. The Columbian Exchange: Biological and cultural consequences of 1492. (Westport, CO: Greenwood, 1972).

Dandler, Jorge. "Diversificación, procesos de trabajo y movilidad espacial en los Valles y Serranías de Cochabamba," in Olivia Harris, Brooke Larson, and Enrique Tandeter (Eds.), La participación indígena en los mercados surandinos (Cochabamba: CERES, 1987), pp. 639-682.

Deere, Carmen Diana, and Robert Wasserstrom. "Ingreso familiar y trabajo no agrícola entre los pequeños productores de América Latina y El Caribe," in Andrés R. Novoa B. and Joshua L. Posner (Eds.), Agricultura de ladera en America Tropical (Turrialba, Costa Rica: CATIE, 1981), pp. 151-167.

Donkin, R. A. Agricultural terracing in the aboriginal New World. (Tucson, AZ: University of Arizona, 1979).

Eckholm, Erik P. Losing ground: Environmental stress and world food prospects. (New York: Norton, 1976).

Ellenberg, Heinz. "Man's influence on tropical mountain ecosystems in South America." Ecology, Vol. 67 (1979), pp. 401-416.

Ellenberg, Heinz. Desarrollar sin destruir. (La Paz: Instituto de Ecolog'a, 1981).

Erickson, Clark L., and Kay L. Chandler. "Raised fields and sustainable agriculture in the Lake Titicaca Basin of Peru," in John O. Brower (Ed.), Fragile lands of Latin America (Boulder, CO: Westview, 1989), pp. 230-250.

Flores, Gonzalo, and Jose Blanes. Dónde va el Chapare? (Cochabamba: CERES, 1984).

Foster. G. R. "Modelling soil erosion and sediment yield," in R. Lal (Ed.), Soil erosion research methods (Ankeney, IA: Soil and Water Conservation Society, 1988), pp. 97-118.

Friedmann, John. Planning in the public domain: From knowledge to action. (Princeton, NJ: Princeton University, 1987)

Friedmann, John. Empowerment: The politics of alternative development. (Oxford: Basil Blackwell, 1992).

Fuenzalida, Fernando. "La estructura de la Comunidad de Indigenas tradicionales," in R. G. Keith, F. Fuenzalida, J. Matos Mar, J. Cotler, and G. Alberti (Eds.), El campesino en el Perú (Lima: IEP, 1970), pp. 61-104.

Gallopín, Gilberto. "Science, technology and the ecological future of Latin America." World Development, Vol. 20, No. 10 (1992), pp. 1391-1400.

Gierhake, Klaus. "Die rolle öffentlichen investitionen im prozess der regionalentwicklung in Bolivien (1987-1990). Geographische Zeitschrift, Vol. 79, (1991), pp. 239-245.

Gill, Lesley. Peasants, entrepreneurs, and social change: Frontier development in lowland Bolivia. (Boulder, CO: Westview, 1987).

Godoy, Ricardo A. "Ecological degradation and agricultural intensification in the Andean highlands." Human Ecology, Vol. 12, No. 4 (1984), pp. 358-383.

Golte, Jürgen. La racionalidad de la organización andina. (Lima: Instituto de Estudios Peruanos, 1980).

Haque, C. Emdad, and M. Q. Zaman. "Human responses to riverine hazards in Bangladesh: A proposal for sustainable floodplain development." World Development, Vol. 21, No.1 (1993), pp. 93-107.

Harden, Carol. "Mesoscale estimation of soil erosion in the Río Ambato drainage, Ecuadorian sierra." Mountain Research and Development, Vol. 8, No. 4 (1988), pp. 331-341.

___________. "Andean erosion conditions." National Geographic Research and Exploration, Vol. 7, No. 2 (1991), pp. 216-231.

Healy, Kevin. "The boom within the crisis: Some recent effects of foreign cocaine markets on Bolivian rural society and economy," in Deborah Pacini and Christine Franquemont (Eds.), Coca and cocaine: Effects of people and policy in Latin America (Ithaca, NY: Cornell, 1986), pp. 101-144.

Hecht, Susanna B. "Indigenous soil management in the Latin American tropics: Neglected knowledge of native peoples," in Miguel A. Altieri (Eds.), Agroecology and small farm development. (Boca Ration, FL: CRC Press, 1990), pp. 151-158.

Holling, C. S. "Resilience and stability of ecological systems." Annual Review of Ecology and Systematics, Vol. 4, (1978), pp. 1-23.

I.I.D.E. (International Institute for Development and the Environment) and U.S. A.I.D. Perfil ambiental de Bolivia. (La Paz: U.S. A.I.D., 1986).

I.N.E (Instituto Nacional de Estadística). Atlas Censal de Bolivia. (La Paz: I.N.E., 1982).

Ives, Jack D., and Bruno Messerli. The Himalayan dilemma: Reconciling development and conservation. (London: Routledge, 1989).

Knapp, Gregory. Andean ecology: Adaptive dynamics in Ecuador. (Boulder, CO: Westview, 1991).

Larson, Brooke. Colonialism and agrarian transformation in Bolivia: Cochabamba, 1550-1900. (Princeton, NJ: Princeton University, 1988).

Lassen, Cheryl A. Landless and rural poverty in Latin America. (Ithaca, NY: Rural Development Committee of Cornell University, 1980).

LeBaron, G., L. K. Bond, P. Aitken, and L. Michaelson. "Explanation of the Bolivian highlands grazing-erosion syndrome." Journal of Range Management, Vol. 32, No. 3 (1979), pp. 201-208. Lewis, W. A. "Economic development with unlimited supplies of labour." Manchester School of Economic and Social Studies Vol. 22 (1954), pp. 139-191.

Low, Frank K. "Estimating potential erosion in developing countries." Journal of Soil and Water Conservation, Vol. 22, No. 4 (1967), pp. 147-148.

Maletta, Héctor. El subempleo en el Perú: Una visión crítica. Apuntes, Vol. 4, No. 8 (1978), pp. 3-48.

Maletta, Héctor. "Agricultura y política económica en Bolivia 1985-1987." Debate Agrario, Vol. 2 (1988), pp. 87-130.

Malloy, James M. Bolivia: The uncompleted revolution. (Pittsburgh, PA: University of Pittsburgh, 1970).

Malloy, James M., and Eduardo Gamarra. Revolution and reaction: Bolivia 1964-1985. (Oxford: Transaction, 1988).

McDonald, Scott. B. Mountain high, White avalanche. (Westport, CO: Praeger, 1989).

Metz, John J. "A reassessment of the causes and severity of Nepal's environmental crisis." World Development, Vol. 19, No. 7 (1991), pp. 805-820.

Millones, Jose O. "Patterns of land use and associated environmental problems of the central Andes." Mountain Research and Development, Vol. 2, No. 1 (1982), pp. 49-61.

Mitchell, William, and David Guillet, eds. Irrigation in high-elevation mountain communities. (New York: Association of Latin American Anthropology, forthcoming).

Monheim, F. "The population and economy of tropical mountain regions: Illustrated by the examples of the Bolivian and Peruvian Andes," in D. Klaus-Müller (Ed.), International Workshop on the Development of Mountain Environment (Munich: German Foundation of International Development, 1974), pp. 133-146.

Morales, Juan Antonio. "Ajustes estructurales en la agricultura campesina boliviana." Debate Agrario, Vol. 9 (1990), pp. 121-162.

Parsons, James J. "The northern Andean environment." Mountain Research and Development, Vol. 2, No. 3 (1982), pp. 253-262.

Pearse, Andrew. Seeds of plenty, seeds of want. (Oxford: Clarendon, 1980).

Pérez-Crespo, Carlos. "Why do people migrate? Internal migration and the pattern of capital accumulation in Bolivia." (Binghamton, NY: IDA Working Paper No. 74).

Pinto Amaya, Eddy. "Estudio Pluviométrico del Departamento de Cochabamba." (Cochabamba: Universidad Nacional Mayor de San Simón (Thesis), 1982).

Posner, Joshua L. Cropping systems and soil conservation in the hill areas of tropical America. Turrialba, Vol. 32, No. 3 (1982), pp. 287-299.

Posner, Joshua L., and Malcolm F. McPherson. "Agriculture on the steep slopes of tropical America: Current situation and prospects for the year 2000." World Development, Vol. 10, No. 5 (1982), pp. 341-353.

Preston, David A. "The revolutionary landscape of highland Bolivia." The Geographical Journal, Vol. 135, No. 1 (1969), pp. 1-16.

Preston, David A. "Migración y desarrollo rural en América Tropical," in Andrés R. Novoa B. and Joshua L. Posner (Eds.), Agricultura de ladera en America Tropical (Turrialba, Costa Rica: CATIE, 1981), pp. 169-182.

Pulgar Vidal, Javier. Geografía del Peru: Las ocho regiones naturales del Perú. (Lima: Universidad Nacional Mayor de San Marcos, 1968).

Redclift, M. Sustainable Development: Exploring the contradictions. (London: Methuen, 1987)

Reyes, M. Medio ambiente y desarrollo en el Ecuador. (Quito: Salvat, 1983).

Richards, Paul. Indigenous agricultural revolution: Ecology and food production in West Africa (London: Unwin Hyman, 1985)

Rist, Stephan, and Juan San Martín. Agroecología y saber campesino en la conservacion de suelos. (Cochabamba: AGRUCO, 1991).

Rivera P., Alberto. "El Chapare actual," in CERES (Ed.), El chapare actual; sindicatos y ONGs en la región. (Cochabamba: ILDIS, 1990).

Rivière d'Arc, Helene. "Public and private agricultural policies in Santa Cruz (Bolivia)," in F. Barbira-Scazzocchio (Ed.), Land, people, and planning in contemporary Amazonia (Cambridge: Cambridge Centre of Latin American Studies, 1979), pp. 154-161.

Sage, Colin. "Intensification of commodity relations: Agricultural specialization and differentiation in the Cochabamba serranía, Bolivia." Bulletin of Latin American Research, Vol. 3, No. 1 (1984), pp. 81-97.

Sauer, Carl O. Agricultural origins and dispersals: The domestication of animals and foodstuffs. (New York: American Geographical Society, 1952).

Southgate, Douglas, and Morris Whitaker. "Promoting resource degradation in Latin America: Tropical deforestation, soil erosion, and coastal ecosystem disturbance in Ecuador." Economic Development and Culture Change xxx, xxx (1992), pp. 787-807.

Stadel, Christoph. "The perception of stress by campesinos: A profile from the Ecuadorian sierra." Mountain Research and Development, Vol. 9, No. 1 (1989), pp. 35-49.

Staver, C. P., A. C. Byers III, A. C. Ravelo, and J. C. Dickinson III. "Refining soil conservation strategies in the mountain environment-the climatic factor: An Ecuadorian case study." Mountain Research and Development, Vol. 11, No. 2 (1991), pp. 127-144.

Stocking, Michael, and Scott Perkin. "Conservation-wtih-development: An application of the concept in the Usambara Mountains, Tanzania." Transactions of the Institute of British Geographers, Vol. 17 (1992), pp. 337-349.

Thrupp, Lori Ann. "Legitimizing local knowledge: "Scientized packages" or empowerment for Third World people," in D. Michael Warren, L. Jan Slikkerveer, and S. Oguntunji Titilola (Eds.), Indigenous knowledge systems: Implications for agriculture and international development. Studies in Technology and Social Change, No. 11. (Ames, IO: Iowa State University, 1989).

Todaro, Michael P. Economic development in the Third World. (London: Longman, 1989).

Treacy, John M. "Agricultural terraces in Peru's Colca Valley: Promises and problems of an ancient technology," in John O. Brower (Ed.), Fragile lands of Latin America (Boulder, CO: Westview, 1989), pp. 209-229.

Urioste Fernández de Córdova, Miguel. El estado anticampesino. (La Paz: ILDIS, 1984).

Vayda, Andrew P. "Progressive contextualization: Methods for research in human ecology." Human Ecology, Vol. 11, No. 3 (1983), pp. 265-281.

Vogt, William. The population of Venezuela and its natural resources. (Washington, D. C.: Pan American Union, 1946).

Warren, D. Michael, L. Jan Slikkerveer, and S. Oguntunji Titilola (Eds.). Indigenous knowledge systems: Implications for agriculture and international development. Studies in Technology and Social Change, No. 11. (Ames, Iowa: 1989).

Weil, Connie. "Migration among landholdings by Bolivian campesinos." Geographical Review, Vol. 73, No. 2 (1983), pp. 182-197.

White, Stuart, and Fausto Maldonado. "The use and conservation of natural resources in the Andes of Southern Ecuador." Mountain Research and Development, Vol. 11, No. 1 (1991), pp. 37-55.

Xu, Cheng, Han Chunru, and Donald C. Taylor. "Sustainable agricultural development in China." World Development, Vol. 20, No. 8 (xxx) pp. 1127-1144.

Zimmerer, Karl S. "Labor shortages and crop diversity in the southern Peruvian sierra." Geographical Review, Vol. 81, No. 4 (1991a), pp. 414-42.

_____________. "Informe Diagnóstico: Uso de la tierra y erosión de suelos en la cuenca del Río Calicanto," Mimeo (Cochabamba: CIDRE, 1991b).

_____________. "Local soil knowledge and conservation-with-development: Advances from highland Bolivia." Mimeo (Madison, Wisconsin: Department of Geography, December 1992).

This study was completed with the collaboration, assistance, and support of numerous persons and institutions. Special gratitude is expressed to the Cochabamba peasants who offered considerable time, interest, and patience in collaborating with the field study. The peasant unions in Cochabamba and many development institutions there were equally helpful. Field research was made possible by my affiliations with five Non-Government Organizations in Cochabamba, including CIDRE, and their cooperation is sincerely appreciated. Financial support for planning, field work, and library research involved in the present study was provided by a Post-Doctoral Fellowship from the Social Science Research Council and research grants from the National Science Foundation and the Graduate School of the University of Wisconsin-Madison. My deep gratitude and intellectual debt to those mentioned above and others who cannot be mentioned due to confines of this acknowledgement does not, however, absolve me of complete responsibility for the present study.

Published April 27, 1996.
File converted to HTML by GCH.