Cite as: [Author(s), 2000. Title.] In: K. McElwee, D. Burke, M. Niles, X. Cummings, and H. Egna (Editors), Seventeenth Annual Technical Report. Pond Dynamics/Aquaculture CRSP, Oregon State University, Corvallis, Oregon, [pp. ___.]

Fish Culture in the Peruvian Amazon: Producer Perceptions and Practices in Three River Systems

Eighth Work Plan, Adoption/Diffusion Research 1-1 (8ADR1-1)
Final Report

Joseph J. Molnar
Department of Agricultural Economics and Rural Sociology
International Center for Aquaculture and Aquatic Environments
Auburn University, Alabama, USA

Fernando Alcántara Bocanegra and Salvador Tello
Instituto de Investigaciones de la Amazonia Peruana (IIAP)
Iquitos, Peru

Abstract

The Instituto de Investigaciones de la Amazonia Peruana (IIAP) is the leading governmental organization working in aquaculture and fisheries research in the Peruvian Amazon region. In addition, IIAP produces fingerlings, offers training courses, and works with nongovernmental organizations (NGOs) endeavoring to promote fish culture. This report summarizes fieldwork and survey results from rural communities in the Iquitos area of the Amazon served by NGOs assisted by IIAP. The researchers visited rural communities and interviewed fish farmers, community residents, and public and private agency officials to discover the strategies and approaches to small-scale, community-based aquaculture employed in the Peruvian Amazon. Subsequently, data were collected from a sample of 146 practicing fish farmers in the Napo, Tamishiyacu, and Tahuayo river systems areas north and south of Iquitos, as well as in the Iquitos-Nauta Road area directly south of the city. Fish farmers were identified in selected communities provided technical assistance in aquaculture by CARE/Peru and several other nongovernmental organizations. Results portray the species cultured, marketing strategies employed, and the perceived impact of fish culture on families and farming systems. The data show that fish farmers are in an advantageous situation for fish culture. They encounter few barriers to building ponds, obtaining fingerlings, feeding their fish, or marketing the product. Fruits and other forest-based fish foods are widely available to support extensive production systems. A number of NGOs are providing regular farm visits and advice on fish culture. The natural cycle of the Amazonian river systems ensures a market period of relatively high prices for farm-reared fish. Additional attention is needed on identifying and communicating production practices that will reduce risk and enhance the benefits of aquaculture.

Introduction

Loreto, Peru's largest department, is entirely forested except for water bodies and urban areas. Iquitos is a city of more than 300,000 people accessible only by boat or airplane. No roads connect the city in northeast Peru to Lima on the western coast. Founded during an earlier period of rubber boom (Barnham and Coomes, 1996), Iquitos is now a regional trade center that serves the many small communities that line the Amazon and its tributaries (Chibnik, 1994). Many humanitarian and environmental nongovernmental organizations (NGOs) operate in the Loreto area and are based in Iquitos (Rainforest Conservation Fund, 1999). The Peruvian Amazon has been subjected to large-scale commercial exploitation for the last two centuries. As Nauta, Tamishiyacu, and then Iquitos grew during the late 1800s, they became centers of urban consumption and international export. While fine rubber—jebe fino (Hevea brasiliensis)—was recognized worldwide, Peru produced large amounts of weak rubber—jebe debil—from upland varieties of Hevea sp. Coomes (1992b) provides an analysis of the rubber trade and the difficulties involved in the business. Over 3,000 metric tons was exported annually from the Peruvian Amazon between 1902 and 1917 (Villarejo, 1988). This export economy crashed after the Second World War, but the tire industry in Lima rejuvenated the rubber trade during the 1960s and early 1970s. Petroleum-based tires have effectively ended the rubber trade in the northeastern Peruvian Amazon.

The Amazon River fishery plays a fundamental role in the livelihoods and survival of rural populations in this region (McDaniel, 1997). Fishing is by far the most important source of animal protein in the Amazon Basin and the main generator of cash for people living along the river. Araujo-Lima and Goulding (1997) argue that fishing is the most promising means for increasing animal protein within the Amazon Basin with a minimum of environmental degradation. Aquaculture plays a unique and dynamic role in the forest-and-river-based farming system of the Peruvian Amazon or Selva (Tomich et al., 1995; Pinedo-Vasquez et al., 1992).

The number of fish species in the Amazon hydrographic basin has been estimated at 2,000. Only 1,400 of these have been described scientifically. These represent approximately 10% of the planet's ichthyofauna (Rainforest Conservation Fund, 1999). The percentile distribution of the species in the main fish families is as follows: Sirulydeous 44%, Characoids 42%, Cichlidae 6%, and other species comprising 8%. Araujo-Lima and Goulding (1997) maintain that while Colossoma macropomum (called gamitana in Peru) is not the only Amazon fish species deserving of special attention, it is the first species about which enough is known to both manage wild stocks and develop aquaculture.

Fishing Strategies

Successful fishing depends on the fisherman's knowledge of fish biology and physiology, of the peculiarities of the river or lake, and of water level variations. This knowledge and the necessary tools, which vary according to the species to be caught, have been largely developed by the Indians and inherited by the ribereños—the multi-ethnic river community dwellers. These native fishing techniques, which are carried out with bow and arrow, harpoon, spear, hooks, nets, traps, or even by poisoning parts of the floodplains with a plant poison (of the Lonchocarpus, Phyllanthus, and Tephrosia genera, which paralyzes the fish gill, forcing them to come up to breathe) do not damage the fishing stock significantly (Rainforest Conservation Fund, 1999; Hiraoka, 1986, 1989).

In Amazonia, the introduction of dip nets and their use in the flooded parts of the forest, the use of ice to store the fish, and the improvements in refrigeration and transport systems all led to increased exploitation of river fisheries. In turn, these technical innovations were a response to the increasing demand from foreign markets and facilitated exports. Boats with large-scale gear operated by urban interests are often a threat to stocks in local rivers and oxbow lakes. Moreover, fish with skin—traditionally not consumed by Indians and ribereños—began to be included in the number of edible fish. Consequently, predatory practices tended to replace traditional fishing techniques (Rainforest Conservation Fund, 1999).

The Amazon's fishing potential, previously regarded as inexhaustible, was overestimated. Though most waters in the Iquitos area are brown, black and clear water rivers are poor in fish. The fishing output of the Amazonian basin in 1980 was estimated at 150,000 t yr^-1, but the fishing potential could be approximately twice this figure, that is, 300,000 t. This estimate encompasses all species and sizes, not only those with commercial value (Rainforest Conservation Fund, 1999).

Under these circumstances, it is foreseeable that fish supplies for the rapidly growing Amazonian population cannot be guaranteed for long. Management of fish for subsistence use or local consumption is done by the inhabitants of the areas around the lakes, whose interests in preservation conflict with those of the professional fishermen. The river people supplement fishing with subsistence agricultural activities, jute plantations, and extraction of wood and other products. Now aquaculture is widely perceived as a farm-based activity that complements traditional sources of food and livelihood.

Aquaculture in the Selva

There is no fish-breeding tradition in Amazonia. The aboriginal populations kept fish, manatees, and turtles in large corrals for periodic consumption, but no techniques for reproduction in captivity were developed. Previous efforts by government agencies, NGOs, missionaries, and others, however, have led to a certain level of indigenous knowledge and interest in aquaculture.

There is a unique relationship between aquaculture and fisheries in many parts of the Selva (Hall, 1997). The abundance of large, rapidly growing fish species supports an extensive capture fishery in the Amazon, its tributaries, and a large number of oxbow lakes. The fishery, however, is cyclic, as fishing is more difficult during the high water period of December through March. At this time, fish prices for some species are as much as twice the low water period price. This cyclical deficit in the supply of fish coupled with a widespread perception that river and lake fish stocks have declined and will continue to do so are the primary motivations for fish culture in the Selva. Commercial-scale fishers using large-scale fishing gear have depleted fish stocks in many oxbow lakes, further encouraging pond-based fish production.

Abundant supplies of warm water, generally available pond inputs, and easily obtainable grow-out stock are some of the favorable conditions for fish culture in the Amazon river system. Fingerlings can be obtained through individual effort in rivers or oxbow lakes or through purchase from fishermen. Longstanding efforts by the Department of Fisheries and other agencies that provided technical assistance in aquaculture has stimulated interest in fishponds. Current NGO efforts build on previous government efforts in some communities. In others, NGOs are introducing pond aquaculture as a new activity.

For example, the upper Tahuayo is mainly várzea forest with abundant and productive aquatic habitat. Nonetheless, many farmers on this other river systems construct the regionally popular piscigranjas (small dams on creeks or wetlands). While fish are scarce during some parts of the year and often difficult to catch in others, the flooding of the forest limits where farmers can build ponds.

An environmental NGO (Rainforest Conservation Fund) reported building a small pond project with some of its staff and members of the Chino Lakes ranger (vigilante) group. The environmental group's conclusions emphasized the risks involved in fish farming (theft, dike breakage, predators, etc.) and the high labor costs. It concluded that the construction of piscigranjas (fishponds) was unproductive on the Tahuayo. Nonetheless, CARE/Peru and other agencies are undertaking a program of pond construction in this and similar locales. Our fieldwork suggests that ponds are popular among farmers and do contribute to income and food security in these communities.

Fishing is not an important activity on the Tamishiyacu due to the lack of lakes and natural habitat for fish. While fish are abundant there, little commercial fishing occurs due to the lack of an accessible market. On the upper Tahuayo River, however, fishing is a very important economic and subsistence activity (Rainforest Conservation Fund, 1998).

On the upper Tahuayo, scores of ponds and swampy areas as well as large lakes and the river itself make fishing a primary economic activity. In this part of the buffer zone, seven lakes are presently, and have historically been, of utmost importance to the people here. These lakes, as well as the Tahuayo River, are blackish in color and support species considered to be "blackwater fish," but the pH and nutrient content of the water is similar to "white water" (see Coomes, 1992a, 1992b).

Rainforest Conservation Fund (1998) identifies a number of uses for the oxbow lakes. Charo Cocha (cocha is the word for "lake" in the local Quechua dialect) is over 500 ha in size, and supplies residents of Esperanza and the village of Charo with fish and income. Cunshicu Cocha, lying within the jurisdiction of Buena Vista, has an area of 65 ha, and provides subsistence and income to Buena Vista and the village of Cunshicu. The NGO reports that some 100 families use these two lakes, with often-considerable pressure from non-locals who come from as far as the Amazon River and Iquitos.

The local people have long contested the use of freezer boats, illegal nets, traps, poisons, and explosives. Just upriver, five small lakes, ranging from 1 to 10 ha in size, are of vital importance to the village of Chino (population 310). Besides being subjected to the same problems as Charo and Cunshicu Cocha, their small size makes them extremely sensitive to environmental conditions (water levels, temperature, predators, etc.). The strong drop in the Tahuayo River's water levels during July to November nearly eliminates fish from these lakes. Consequently, the community of Chino has implemented an effective vigilance and management system since 1986 (Penn and Alvarez, 1990). These conditions underscore the importance of fishponds as a means for countering the cycles of the river.

The reciprocal relationship between fisheries and aquaculture in the Peruvian Amazon is further enhanced by the well-established patterns of fish marketing present in the region. Fish are a central part of the ribereños' diet, many species are accepted for consumption, and fish sales seem to be readily accomplished locally or at market centers. Tello's (1998) study of fish landings in Iquitos illustrates the diversity of fish in the markets and the centrality of Prochilodus nigricans (boquichico) as the most heavily harvested fish. Colossoma macropomum (gamitana) has had a steady, albeit slightly declining level of reported fish landings over the extended period of data that were available (see also Alcántara, 1994).

Methods and Materials

Sample and Data Collection

Fish farmers were identified in selected communities that were provided technical assistance in aquaculture by CARE/Peru and several other NGOs in the Napo, Tamishiyacu, and Tahuayo river systems that combine to form the Amazon, as well as in the Iquitos–Nauta Road area south of Iquitos (Figure 1). Structured interviews were conducted with a sample of 146 fish farmers having accomplished at least one harvest in the past two years (Casley, 1988; Townsley, 1996). The sample was drawn from NGO program participants in selected communities provided technical assistance in aquaculture by CARE-Peru and other selected NGOs.

The survey instrument was adapted from previous research conducted by Molnar et al. (1996) in five PD/A CRSP countries—Honduras, Thailand, the Philippines, Rwanda, and Kenya. The Peru survey, however, reflects the unique conditions and context of Amazonian fish culture, the diversity of species, and the singular relationship of aquaculture to the river fishery in the region. Ponds were identified in communities on three river systems north and south of Iquitos as well as the Nauta Road area south of Iquitos. Data collection took place in early 1999 and was conducted by graduate students from the Department of Fisheries at the Universidad Nacional de la Amazonia Peruana.

Analysis

The analysis tabulates the survey responses across the three locationsnorth, south, and central to Iquitoswhere data were collected for the study. From this information, central patterns of comparison and difference in practice and approach to fish production and technology utilization can be discerned.

Results

Respondent Characteristics

Table 1 describes the individual and household characteristics of study respondents. Women comprised about 45% of the respondents in Nauta Road, but only about one-tenth of the Napo River and Tamishiyacu fish farmers who we contacted. The Napo River farmers were somewhat older and the Tamishiyacu farmers slightly younger than those in the other locations.

Women and children are central beneficiaries of enhanced protein availability, income, and food security associated with fish ponds in the Selva. While pond construction labor seems primarily to be the province of men, it was sometimes reported that women provided pond inputs and participated in fish harvest.

Over 75% of the study households had children under age ten. Tamishiyacu families had the most children under age 10 and the smallest proportion with children over age 18. Tamishiyacu farmers also had the largest households, as 72% reported six or more members. About two-thirds of the respondent households in the other locations were that large.

Most respondents in the study were associated with individual family ponds. In some communities, NGO technicians worked with groups of families that shared pond construction labor and pond management responsibilities. About 14% of the study respondents were associated with such group ponds.

Few group or collective ponds were reported or encountered during our field visits, but the survey data show that group ponds are an important mechanism for introducing aquaculture into poor rural communities. CARE/Peru officials indicated that they used community groups to construct an initial demonstration pond in some locales, but that family-based ponds were the preferred strategy. In some communities, groups of families cooperate in a minga—a labor exchange arrangement—to jointly construct a fishpond on one family's land. Minga is a Quechua word for the collaboration of communities on specific tasks (e.g., harvesting, sowing, and house building). The completion of the pond is typically followed by a small celebration hosted by the beneficiary family. Subsequent group efforts may build additional fish ponds on other farms.

Landholding

Table 2 profiles the landholding of study respondents. Nearly all owned one or two parcels of land. Napo River holdings tended to be divided between the two larger categories, but 70% of the Tamishiyacu farmers reported holding more than 10 ha. Holders of multiple parcels were primarily pond groups organized for constructing and maintaining a fishpond.

Nauta Road farmers were most diverse in size as one-fourth reported holdings less than 1 ha in size. About two-thirds of the farmers in each location indicated that they owned less land than other farmers did.

Farm Enterprises

The farm enterprises maintained by fish farmers are portrayed in Table 3 . Chickens were the most commonly reported animal enterprise, followed by pigs, ducks, and cattle. In terms of cash income, 60% of the Nauta Road farmers cited fish as their primary source. Furthermore, Nauta Road farmers were more likely to integrate animal production with fish culture. Fish was identified as the primary source of cash income by about 20% of the farmers in the other locales. Chickens were the primary source of cash income for most farmers in the study.

More Nauta Road farmers raised animals near or with their fishponds than farmers from other areas. Such integrated operations recycle the undigested feed and other nutrients from penned animals.

Pond Location and Water Source

More than 85% of the fish farmers had but a single pond, as shown in Table 4 . In contrast, about 45% of the Nauta Road farmers had more than one pond. None of the Napo ponds were located near the farm residence, but one-third of the Nauta Road ponds were. The location of the fishpond relative to the household is significant; ponds near households are easier to monitor. Family members can attend to the pond as well as give regular surveillance to deter theft. About 19% of the fishponds were located near the house. Farmers primarily filled their ponds from springs using gravity flow methods. No farmers experienced problems obtaining water for their fishponds.

Species Cultured

Table 5 shows that Prochilodus nigricans (boquichico) was the most frequently cultured fish in the Peruvian Amazon, grown by about 75% of the farmers. Colossoma macropomum (gamitana) was the next most frequently culture fish. Piaractus brachypomus (pacu) was grown by more group operations (71%) than on individual farms (58%), the largest difference in the comparison. Similar proportions grew Brycon sp. (sabalo) and Leporinus sp. (Lisa), but 9% more individual farmers favored Semaprochilodus insignis (yaraqui) and 11% more individual farmers reported growing Astronotus ocellatus (Oscar or acarahuazú). Alcántara (1994) presents a detailed analysis of fish landings at Iquitos ports and the species that are brought to market.

Fish Feeding

Farmers in the four locations fed their fish a variety of different items reflecting differences in the species cultured and items available (Table 6) .

About 73% of the Napo farmers fed their fish several times a day. The greater incidence of integration with poultry and duck production that requires multiple daily feedings literally spills over to the fish crop. Poultry houses are typically located directly over the fishpond, so feed and litter are nearly continuously deposited into the pond.

Nauta Road resident farmers reported a high level of attentiveness to their ponds, but more Tamishiyacu farmers spent long periods at their ponds when they visited them. All the Nauta Road farmers fed their fish several times a week or more often. Three-quarters of the Nauta Road farmers spent less than an hour on each visit as they were making more visits to attend to their fish. Feeding refers primarily to the provision of fruits, household scraps, and other items directly consumed by the fish. Patterns of fish feeding parallel patterns of pond visitation.

Pond Management

Table 7 profiles some of the pond management practices used in the Peruvian Amazon. Nauta Road farmers tended to not fertilize their ponds, whereas half the farmers at the other sites used cattle manure. Curiously, 15% of the Nauta Road farmers limed their ponds, a fundamental fish culture management practice not followed by many farmers in the other areas. These farms had truck delivery access to Iquitos port facilities where lime and other bulk agricultural chemicals are more readily available. Lime increases the alkalinity (pH) of the pond and fosters primary productivity. Tamishiyacu farmers spent the most time with their ponds when they visited them, Napo farmers the least.

Stocking Practices

Table 8 suggests that fingerlings are typically acquired from local fisherman, and few farmers are using farm-produced fingerlings from private or government sources. About two-thirds of the farmers experienced some problems finding fingerlings when they wanted to restock their ponds. Most were stocking fingerlings between 3 and 10 cm in size.

Farm Equipment

Table 9 shows the farm equipment reported by the surveyed farmers. Wheelbarrows were owned by about 87% of the sample. About one-quarter had a scale and a similar proportion a fish net.

Harvest Practices

Farm labor for harvesting fish was usually supplied by family members, particularly in the Tamishiyacu-Tahuayo communities, as shown in Table 10. Few operators employed laborers or used the services of a buyer to harvest their ponds. Few reported difficulties obtaining labor. Farmers harvested large fish, most favoring sizes 20 to 45 cm in length and greater than 750 g. Thus, most operators achieved only one cycle of production each year.

Marketing

Only a small proportion of each sample reported fish harvested solely for home consumption or barter, i.e., not sold for cash (Table 11) . Most sold some fish for cash and only a few said that they sold less than half the harvest for cash. About 80% of the Napo River respondents sold all their harvest for cash, but only 21% of the Tamishiyacu did. In Nauta River, 80% said they sold more than half the harvest for cash. Nearly 100% did so in Tamishiyacu-Tahuayo Rivers.

Middlemen purchased fish from about 41% of the farmers overall. In the Napo River 56% used a middleman. Only 18% of Tamishiyacu farmers sold to middlemen. Few farmers sold fish to restaurants.

Tamishiyacu farmers were less likely to sell fish in a community marketplace than farmers from the Napo River and Nauta Road regions. Direct marketing by way of pond bank sales was most common for farmers of Tamishiyacu-Tahuayo Rivers. The most common marketing method for farmers from all regions was pond bank sales to neighbors and others coming to the ponds at harvest. Word-of-mouth knowledge about prospective harvests or the willingness to partial-harvest for immediate sale remain primary means for marketing fish for most small- and medium-size farmers.

Marketing Problems

Table 12 shows that very few farmers experienced trouble marketing their fish. Few reported price problems, though one-fourth of the Nauta Road farmers did so. Most felt they could move their product at a lower price. Most respondents felt that larger fish are easier to sell.

Fingerling sales to other farmers were most common in Napo River, where more than half the respondents reported such transactions. Private fingerling sales among farmers are an important indicator of sustainability, especially where government services are unreliable or unavailable in much of the locale. The respondents did not frequently report fingerling sales. Most were growing species that do not naturally reproduce in ponds, so fingerlings would not be an expected by-product of their ponds.

Impacts on Households

Table 13 shows a series of questions profiling the impacts of fish culture on households. Almost none of the farmers thought that there were points in the annual farm cycle when the pond was too much work. Few respondents noted problems associated with the fish enterprise in taking care of other crops, taking care of the family, or completing other household work. Most thought the pond fit well with other activities of the household. Previous work suggests that women are much more likely to report these difficulties (Molnar et al., 1996). About 84% of the respondents noted the benefits of additional cash for their families as something associated with the fish crop.

Pond Conflicts

Table 14 shows respondent experiences with a series of problems sometimes encountered by fish farmers. No farmers noted problems concerning conflict with others about water resource available for fish crops. Tamishiyacu operators had the most problems with predators eating their fish, but this was also an issue for many farmers in each of the other locations.

Theft was a concern for about two-thirds of the Napo River and Nauta Road farmers, but much less so on the Tamishiyacu-Tahuayo Rivers. Napo farmers were most likely to agree that fish were easier to steal than other crops, though one-third of the other respondents thought so as well.

In our field visits, farmers reported few specific production problems. Nutria was most frequently mentioned as a source of production losses, with some occasional mentions of alligators and birds. In general, farmers seemed happy with the fish enterprise, several were in the process of planning or constructing additional ponds, and we encountered only a few unused or abandoned ponds. These were typically small ponds with water supply problems. Ponds that fail to keep water during the dry season thwart the counter-cyclical possibilities of fish culture.

Prospects for the Pond

Nearly all respondents thought their fish pond produced enough to be worth the work they put into it, though a few Nauta Road respondents were skeptical (Table 15) . Nearly all respondents thought fish was the best use of the land it occupied. Nauta Road and Tamishiyacu residents were most likely to report themselves as planning to build new ponds (over 90%). In Napo River, only 72% thought so. Nearly all were happy with fish as a crop to raise, but less so on the Tamishiyacu River (84%). Overall, 86% of the farmers surveyed felt that fish was more profitable than other farm activities, but 92% of the Napo River farmers felt this way.

Technical Assistance

Aquacultural extension services were making frequent contacts with farmers on the Napo and Tamishiyacu Rivers, but less so on the Nauta Road. In each location, NGO technicians were working in aquaculture in one way or another. Most farmers reported some kind of extension contact in the past year. Most farmers received regular visits if they wanted them. Nearly all respondents indicated that they wanted extension help in the future.

Link to Table 16 (Sources of technical assistance reported by fish farmers in the Peruvian Amazon, 1999).

Conclusions

There are few obstacles to at least limited success with fish culture in Amazonia. We observed a high degree of variability in the quality of pond construction and culture conditions, as well as in the type of species employed and level of management applied. Farmers typically stock a polyculture of three to five different species of wild–caught fry or juveniles for grow-out. Farmers typically reported counts of the number of fish of each species that was stocked.

Peru is in an advantageous situation for fish culture. The data show that farmers encounter few barriers to building ponds, obtaining fingerlings, feeding their fish, or marketing the product. Fish farmers in the Peruvian Amazon are interested in and receptive to technical assistance. Fruits and other forest-based fish foods are widely available to support extensive production systems. A number of NGOs are providing regular farm visits and advice on fish culture. The natural cycle of the Amazonian river systems ensures a market period of relatively high prices for farm-reared fish. The research agenda is appropriately focused on enhancing the availability of hatchery-reared fry. Nonetheless, additional attention is needed on identifying and communicating production practices that will reduce risk and enhance the benefits of aquaculture for the many small- and medium-scale farms in the Selva.

Anticipated Benefits

The communication process linking experimental pond practice to farm practice involves several layers of translation and transmission (Cernea, 1991a, 1991b). Many factors interact to affect the nature and extent of impact that PD/A CRSP scientists and research programs have on national aquacultural institutions and farm practice (Huisman, 1990). Experimental findings are at base, experimental; that is, they reflect controlled conditions and careful measurement of a focused set of factors. Farm conditions reflect variable physical and management situations that often mitigate the impact of effects identified by repeated experimental trial. The data presented here provide empirical specification of the needs and preferences of the actual intended beneficiaries of PD/A CRSP activities in Peru.

As such, they provide a baseline or template for interpreting the cumulative impact of PD/A CRSP and NGO partner activities, as well as a starting point for identifying new directions and emphases that will help realize the promise of aquaculture for farmers and their families in developing countries.

Acknowledgments

We thank the several PD/A CRSP Principal Investigators, Host Country counterparts, and NGO colleagues who facilitated our work and advised us on the many problems and issues. In particular we are grateful to Juan Guerrero of CARE/Peru for his help in obtaining the interview data. Without this assistance, our work would have been much narrower in scope.

Literature Cited

Alcántara, F., 1994. Diagnóstico de los recursos hidrobiológicos de la Amazonía. Report. Tratado de Cooperación. Amazonica, Lima, Peru, 120 pp. Available at <www.fao.org/ag/agl/agls/aaweb/public/22/22000020.htm#I3>.

Araujo-Lima, C. and M. Goulding, 1998. So Fruitful a Fish: Ecology, Conservation, and Aquaculture of the Amazon's Tambaquí. Columbia University Press, London, 213 pp.

Barnham, B.L. and O.T. Coomes, 1996. Prosperity's Promise: The Amazonian Rubber Boom and Distorted Economic Development. Westview Press, Boulder, 189 pp.

Casley, D.J., 1988. The Collection, Analysis, and Use of Monitoring and Evaluation Data. Johns Hopkins University Press, Baltimore, 156 pp.

Cernea, M., 1991a. Putting People First: Sociological Variables in Rural Development. The World Bank, Washington, D.C., 358 pp.

Cernea, M.M., 1991b. Using Knowledge from Social Science in Development Projects. Discussion Paper Number 11. The World Bank, Washington, D.C., 42 pp.

Chibnik, M., 1994. Risky Rivers: The Economics and Politics of Floodplain Farming in the Amazon. University of Arizona Press, Tucson, 232 pp.

Coomes, T., 1992a. Making a living in the Amazon rain forest: Peasants, land, and economy in the Tahuayo river basin of northeastern Peru. Ph.D. thesis. University of Wisconsin-Madison, 450 pp.

Coomes, O.T., 1992b. Blackwater rivers, adaptation, and environmental heterogeneity in Amazonia. Amer. Anthropol., 94:698–701.

Hall, A., 1997. Sustaining Amazonia: Grassroots Action for Productive Conservation. Manchester University Press, Manchester, UK, 234 pp.

Hiraoka, M., 1986. Zonation of Mestizo riverine farming systems in northeastern Peru. Natl. Geog. Res., 2:354–371.

Hiraoka, M., 1989. Ribereños' changing economic patterns in the Peruvian Amazon. J. Cult. Geog., 9:103–119.

Huisman, E.A., 1990. Aquacultural research as a tool in international assistance. Ambio, 19:400–403.

Kohler, C.S. Kohler, M.J. DeJesus, F. Alcántara Bocanegra, E. Rios Isern, and G. Llosa Talavera, 1999. Development of sustainable pond aquaculture practices for Piaractus brachypomus in the Peruvian Amazon. In: K. McElwee, D. Burke, M. Niles, and H. Egna (Editors), Sixteenth Annual Technical Report. Pond Dynamics/Aquaculture CRSP, Oregon State University, Corvallis, Oregon, pp. 99–113.

Maino, R.A., B. Cam, and D. Davis-Case, 1996. Participatory Analysis, Monitoring an Evaluation for Fishing Communities, a Manual. FAO Technical Paper 364. Food and Agriculture Organization, Rome, 142 pp.

Martinez-Espinosa, M., 1997. Report of the Expert Consultation on Small-Scale Rural Aquaculture. FAO Fisheries Report No. 548. FIRA/R548. Food and Agriculture Organization, Rome, 138 pp.

McDaniel, J., 1997. Communal fisheries management in the Peruvian Amazon. Human Organization, 56:147–152.

Molnar, J.J., T.R. Hanson, L.L. Lovshin, 1996. Social, Institutional, and Economic Impacts of Aquacultural Research: The Pond Dynamics/Aquaculture CRSP in Rwanda, Honduras, the Philippines, and Thailand. , International Center for Aquaculture and Aquatic Environments Research and Development Series No. 40. Auburn University, Alabama, 63 pp.

Penn, J.W. and J.A. Alvarez, 1990. Comunidad campesina protege sus cochas. Kantari, 258:3–10.

Pinedo-Vasquez, M., D. Zarin, and P. Jipp, 1992. Economic returns from forest conversion in the Peruvian Amazon. Ecol. Econ., 6:163–173.

Rainforest Conservation Fund, 1998. A brief view of market-oriented extraction in the Peruvian Amazon. Available at <homepage.interaccess.com/~rcf/brief view.html>.

Tello, S., 1998. Analysis of a multispecies fishery: The commercial fishery fleet of Iquitos, Amazon Basin, Peru. M.S. thesis. Oregon State University, Corvallis, 83 pp.

Tomich, T.P., P. Kilby, and B.F. Johnston, 1995. Transforming Agrarian Economies: Opportunities Seized, Opportunities Missed. Cornell University Press, Ithaca, New York, 354 pp.

Townsley, P., 1996. Rapid Rural Appraisal, Participatory Rural Appraisal, and Aquaculture. FAO Fisheries Technical Paper 358. Food and Agriculture Organization, Rome, 142 pp.

Villarejo, A., 1988. Asi es la Selva. CETA Press, Iquitos, Peru, 330 pp.