The Case of Honduras after Hurricane Mitch 
This paper is based on interviews with farmers that received emergency bean-seed provisions in Yorito, Honduras following Hurricane Mitch. It analyses the need for and appropriateness of these provisions. Although crop losses were extensive in Yorito, most farmers were able to secure a harvest of at least one major crop, and many of those who lost all their beans had access to local seed sources. No bean varieties were lostin Yorito as a result of Mitch. Conversely, as provisions introduced improved varieties that were previously unavailable, they augmented thelocal gene pool. However, the seeds provisioned were of a narrow genetic base (due to the low variation in the varieties distributed). Taking account of the differences in agroecological and socioeconomic variables between communities and households, the seeds provisioned were not appropriate to all farmers.
Hurricane Mitch struck Central America in October 1998, leaving Honduras with thousands of dead and an economy in ruins. Landslides, floods and winds devastated houses, infrastructure and cropland. Rural communities of subsistence producers experienced severe harvest failures (UN 1998).
As most subsistence farmers depend on their own agricultural production, or other local sources, for their planting material for the next cropping season, it was expected that the harvest failure would severely restrain people’s ability to save or secure such planting material (CIAT et al 1998). In response, various institutions distributed seeds with the intention of restoring agricultural production and reducing the dependency on external food provision (de Barbentane and Fowler, 2003).
However strong, the linkage between agricultural production and food/seed security of subsistence households is only indirect. Coping strategies often make it possible to secure food and seeds despite harvest failures. These capacities may make programs for emergency seed provision not only unnecessary, but even counterproductive (Pottier 1997, Sperling 1997, Longley and Richards 1999).
The present study, based on research in villages that received emergency seed provisions in Honduras following Hurricane Mitch, assesses the need for such provisions. In addition, it discusses the appropriateness of the approaches that were employed in these programs.
The fieldwork this text is based upon, was undertaken in the municipality of Yorito in central Honduras in February, March and April of 2000 (see Haugen 2001). A survey was undertaken with 85 respondents in six different communities, using a questionnaire with open-ended questions. The resulting quantitative data were supplemented by information obtained in interviews with individuals and groups. These small-holder communities cultivate beans in two seasons a year, and maize once a year as their staple crops.
In the following text, the communities are placed in three groups, depending on their altitude. In the ‘low’ communities of Luquigue and Jalapa, households cultivate lands at 600-900 meters above the sea level. In the ‘intermediate’ communities of La Ladera and Vallecillos, and in the ‘high’ communities of Santa Cruz and Mina Honda, households cultivate at 900-1200 meters and 1200-1600 meters, respectively. Access to markets was generally poorer at higher altitudes.
The households that were interviewed were classified with respect to their well-being, using 12 different indicators  . Households in Luquigue were found to be relatively well-off, while poor households were most frequent in Vallecillos and Mina Honda. In Jalapa, La Ladera and Santa Cruz, households were generally intermediately well off.
Cropping system Food security and crop traits
Farmers have various “interests” that their cropping system is meant to fulfil. While the ultimate goal is to achieve a high level of well-being, the more immediate interests may be to achieve a good harvest in terms of quantity and quality. Farmers undertake a continuous evaluation of their seed lots  to secure that the seeds/crops serve these interests satisfatorily (see Bellon 1996). A number of different criteria can be considered in this evaluation, reflecting the different interests involved.
Table 1 shows the criteria that were most frequently reported among respondents to the survey in Yorito, and how farmers evaluate the most common bean-varieties in the area relative to these criteria. The table shows that varieties are most frequently evaluated with respect to criteria like ‘production’, ‘market acceptance’ and ‘taste’, indicating that, overall, these criteria are most important to farmers.
|Adaptation to zone||3||2||10||2||14||7|
|Adaptation to production system||8||5||4||2||3||1||1||6||2||5|
|Resistance to drought||3||3||1||3||4||1||1|
|Resistance to diseases||14||2||1||9||6||5||9||7||4||7|
|Resistance to pests||6||2||2||4||5||2||1||3|
|Time for maturity||5||6||1||2||6||9||1|
|Maintenance of tradition||2|
*The figures in this table cannot be directly compared across zones or across varieties as the sample size may differ from case to case. For instance, almost all the farmers in the low zone have tried Tio Canela and could evaluate this variety, while the number of farmers who could evaluate Chingo and Retinto was much lower. As fieldwork did not register sample sizes, it is impossible to give the data in proportions and compare columns directly. However, the reader will be able to use the table to infer various information, as presented in the text.
Table 1 reveals that farmers’ experiences with different varieties of beans differ strongly. Furthermore, experiences with Tio Canela, the variety most commonly supplied after Mitch, differ markedly between zones. The criteria farmers use to judge their seed lots vary as well, both between communities and between households, and may even differ within households (Synnevåg et al 1999). While a significant number of respondents at higher altitudes in Yorito were interested in varieties with ‘adaptation to zone’, ‘storability’, ‘simultaneous maturation’ and ‘nutrition’, respondents at lower altitudes put much less emphasis on such criteria.
The differences in the criteria farmers use to judge their seed lots reflects differences in interests between farmers. Lack of homogeneity in such ’sets of interests’ may be analysed in the light of differences in socioeconomic conditions between zones. While households in zone 1 are relatively well-off and have easy access to markets for inputs and produce, households in zone 3 are poorer and have less access to markets. Easier access by farmers in zone 1 to fertilisers and pesticides may reduce the demand for varietal adaptation. The widespread use of silos for storage similarly reduces the relevance of storability. An interest for nutrition may be reduced when more foodstuffs can be purchased in the market.
Strategies for securing seeds
Most subsistence farmers depend on their own production, or other local sources, for their planting material for the next cropping season (Wright et al 1994). Among respondents in Yorito, only 11 % (9 out of 82) said they have ever purchased bean seeds in the market. On the other hand, only 21 % said they exclusively use seeds saved domestically, so the majority of the respondents (68 %) base their seed security on a combination of domestic saving and local, informal procurement.
Seeds can be procured locally in a number of different ways, through seed exchange, or exchange for money or labour, but also as loans. Small amounts may be procured in the form of gifts, but this form of social support has not been widespread in recent years.
Seed procurement strategies vary strongly between zones in Yorito. While saving seeds domestically is most frequent in Santa Cruz and Mina Honda, purchases in the market are employed most frequently in Luquigue and Jalapa. This observation is relevant for research on the intensity of seed shortages: While households in Santa Cruz and Mina Honda were underrepresented among the households that acquired seeds externally after Mitch, this alone does not indicate that the seed shortage was less intense in these communities, as these households have a tendency towards saving seeds domestically.
Dynamics in PGR use
Table 2 shows the number of farmers growing different bean varieties across different cropping seasons. While some varieties have fallen out of use during the last ten years, others have been introduced and are now widely used.
|Variety Season||1990, prim||1990, post||1998, prim||1998, post||1999, prim||1999, post|
|Concha Blanca rojo||1|
|Negro Concha M.||1||1||1|
|Pansa de Mono||1|
|Number of varieties||14||10||11||13||14||13|
|Total number of seed lots5||104||81||83||99||137||127|
2 Improved varieties introduced 10-20 years ago
3 Improved varieties introduced 1-3 years ago. Tio Canela and Dorado were widely distributed in post-Mitch emergency seed provisions
4 Unclear origin
5 As some farmers manage more than one seed lot, the total number of seed lots exceeds the number of farmers.
The effects of Mitch
Effects on varietal diversity
Mitch struck in October 1998, affecting the postrera of beans. FAO (1996) calls attention to the danger that disasters, such as Mitch, may cause the loss of crop genetic diversity. As shown in Table 2, two varieties that were cultivated in the postrera of 1998 were not cultivated in 1999, and may therefore be candidates for such loss. However, these two varieties have always been cultivated very sparingly, and it is impossible to conclude from the survey that these have become extinct. Conversely, in-depth interviews brought information that these varieties were still in use on the time of the fieldwork.
Effects on food security
The proportion of respondents in Yorito that reported crop losses of maize and beans as a result of Mitch is provided in Tables 3 and 4, respectively. The tables report the proportion of respondents that claimed any loss whatsoever. In the case of beans, the proportion of farmers that reported a total loss is also given.
Losses varied across zones and between crops. Farmers contended that the differences observed are related to differences in the agroecological conditions between zones. As the second crop of beans at lower altitudes was sown well before Mitch, bean plants were at intermediate stages of development when Mitch struck. Most farmers in this zone cultivate level terrain, which became waterlogged as a result of Mitch. This caused heavy bean losses. Maize, on the other hand, was already mature and the maize that was not already harvested tolerated waterlogging. At higher altitudes, the maize was not yet mature, and the stronger winds at these altitudes caused lodging. The second crop of beans, however, was in many cases not yet sown, while in other cases the germinating plants could survive, as steep fields, common in these zones, did not become waterlogged.
|Zone||Proportion of farmers experiencing any loss whatsoever|
|Zone 1 (Luquigue/Jalapa)||31,0% (n=29)|
|Zone 2 (La Ladera/Vallecillos)||66,7% (n=24)|
|Zone 3 (Santa Cruz/Mina Honda)||79,3% (n=29)|
Table 4: Proportion of respondents in different zones in Yorito that reported crop losses of beans.
|Zone||Proportion of farmers experiencing any loss whatsoever||Proportion of farmers experiencing total loss|
|Luquigue||92,3% (n=13)||61,5% (n=13)|
|Jalapa||69,2% (n=13)||30,8% (n=13)|
|Zone 2||59,1% (n=22)||36,4% (n=22)|
|Zone 3||55,6% (n=27)||18,5% (n=27)|
Evidence suggests that most farmers in Yorito were able to secure a harvest of at least one major crop, reducing problems with food shortages. This was the case even in the southern region of Choluteca, which was severely affected by Mitch (Haugen 2001). This indicates that cultivating different crops, thus diversifying cropping systems, was important in preventing food shortages after Mitch. However, preventive strategies are not the only way to reduce vulnerability towards harvest failures. Most households have livestock, an important reserve when the crops fail. Off-farm employment exists. In Yorito, day-labouring is widespread in the coffee-sector, for example. Informal mechanisms for the provision of seeds to seed-short households also exist, and seed loans are widespread.
In some cases, however, the use of certain coping mechanisms started to erode the very basis of long-term household food security: some sold land in exchange for food. Thus, there is no doubt that food provisions that were given in the aftermath of Hurricane Mitch were crucial for some households.
Seed security after Mitch
Despite the crop losses caused by Mitch, only 9,3 % (7 out of 75) of the respondents who wanted to cultivate beans in the Primera 1999 reported that they had insufficient amounts of seeds for this season. Of these seed-deficient households, three reported that they are chronically deficient, meaning that they probably would be deficient even in the absence of Mitch. The absence of significant seed-shortage after Mitch is confirmed by studies of price-changes for seeds after Mitch. There was no major price-increase for seeds in Yorito after Mitch, indicating that seeds were accessible to farmers. However, seeds were extensively distributed in Yorito. How would the picture have looked if no such distributions had taken place?
Need for seed provisions
About one-third of the respondents to the survey reported that they had lost their entire harvest of beans (Table 4). However, the experience of harvest failures is not uncommon in this area. The overall proportion of farmers who occasionally experience harvest failures of such intensity that they need to look for seeds elsewhere, is slightly above half. People are accustomed to securing seeds elsewhere when the household supply fails. As two-thirds of the households in the region were able to secure some harvest and the bean harvest of more than one-third was unaffected, seed shortages would not have been acute even in the absence of relief supplies.
Choice of beneficiaries
Emergency seed provisions were undertaken at a high scale in Yorito after Mitch. Of 75 households that planned to cultivate beans in the first cropping season after Mitch, 49 received provisions. How efficient were agencies in targeting those households in need?
Households whose domestic supply was deficient were over-represented among households that received emergency seed provisions (75 % of non self-sufficient households received seeds, compared to 56,4 % of self-sufficient households). However, this tendency is not statistically relevant (p-value for Chi-square-test = 0,091 (n=75)), so the relief efforts lacked a strong bias towards seed-deficient households.
More prosperous households were more likely to receive provisions than poor households (p-value = 0,77 (n=75)). This might be accounted for by the fact that Luquigue, which is comprised of relatively well-off households, was most severely affected by Mitch. However, it has already been shown that farmers in this community supply themselves with seeds from formal markets even in regular years, so these farmers may not have needed seed relief. As noteworthy is the fact that households that were already actively taking part in farmer networks organised by the organisations that undertook seed provisions were far more likely to receive such provisions. This latter tendency is statistically relevant (p-value <0,01 (n=75)).
Choice of varieties
As shown, even self-sufficient households took advantage of the opportunity of receiving seeds from the relief supplies. This enthusiasm for trying out new material may simply be based in curiosity, but may also suggest that there was dissatisfaction with traditional varieties. Tio Canela, the most commonly supplied variety, increased in use from the first to the second season in 1999. Farmers generally liked it. Clearly, making this new variety accessible to farmers was an important outcome of assistance efforts.
At higher altitudes, however, Tio Canela enjoys few advantages compared to traditional varieties. Tio Canela is less productive, and has a higher degree of variation in yield than traditional varieties (Haugen 2001). As farmers generally appreciate productive and stable varieties, it may be argued that focusing solely on Tio Canela and similar improved varieties, could not be optimal for restoring the short-term food security of seed-short households after Mitch.
Though traditional varieties were not lost as a direct result of Mitch, the introduced varieties may gradually replace traditional ones. Table 2 shows that Retinto, Chingo and Pedreño are all in danger of being replaced.Chingo may be most vulnerable. According to Table 1, Chingo has no advantages relative to Tio Canela, suggesting that farmers have good arguments for discarding the variety. Thus, unless new advantages emerge, farmers cannot be expected to maintain this variety.
The threat of genetic erosion, resulting from the abandonment of traditional varieties when new varieties are introduced, should not be an argument against the introduction of new varieties. Rather, the introduction of new varieties may, through making more material accessible to farmers, strengthen farmers’ traditions for experimenting with material and thus adapting to changing conditions. However, there is a different question of whether post-disaster turmoils are an appropriate occasion for such introductions.
Mitch did not restrict farmers access to material in the way some severe or repeated disasters may. However, it is important to recognise that disasters, and post-disaster relief, may undermine farmers’ capacities to experiment and adapt even in other manners. The ever-increasing presence of external organisations, inter alia, in supplying seeds post-disaster, may undermine farmers’ confidence in their traditional systems for securing seeds. This might also cause erosion of traditions for experimenting with and selecting germplasm. Thus, there is a danger that the traditional system of varieties being adopted and abandoned on the basis of farmers’ own experiences is becoming substituted by a system where institutions external to the agricultural system is making the decisions. Thus, relief agencies and other external actors may trigger genetic erosion even when disasters themselves do not have an impact.
The overall picture of communities that remained largely seed-secure even after Hurricane Mitch leaves no evidence of hurricane-caused genetic erosion in bean and maize systems in Yorito. Moreover, the simultaneous cultivation of maize and beans made farmers less vulnerable to the disaster, and generally, food security was protected in the short run. This suggests that the massive seed relief efforts in Yorito may not have been entirely necessary.
Taking account of the complexity of and differences in ‘concern profiles’ between households and communities, the strategy of the post-Mitch seed relief, built on a narrow genetic base, may have been simplistic. Lack of targeting towards seed-deficient and poor households suggests either that relief agencies could have discovered there was no major seed emergency, or that they were simply unable to distinguish those in need from others.
In the long run, the distribution of new varieties augmented the existing gene pool and may ultimately promote more effective farmer plant breeding and selection. However, programs of emergency seed provision may also undermine informal institutions that people employ to cope with food and seed shortages, and others that people employ to adapt to changing environmental conditions. Time will tell.
Barbentane, S. De and C. Fowler (2003) Seed Relief after Hurricane Mitch in Honduras: A Critical Analysis of Institutional Responses. (Manuscript)
Bellon, M.R. (1996) The Dynamics of Crop Infraspecific Diversity: A Conceptual Framework at the Farmer Level. Economic Botany 50 (1) 26-39
CIAT, CIMMYT, CIP and IPGRI (1998) Semillas de Esperanza para Honduras y Nicaragua. CIAT, Tegucigalpa, Honduras
FAO (1996) Global Plan of Action on the Conservation and Sustainable Utilization of Plant Genetic Resources for Food and Agriculture. FAO, Rome
Haugen, J.M. (2001) Whatever the Will of the Weather: A Study of Seed Systems in Honduras, and their Importance for Food Security and Agrobiodiversity in the Aftermaths of Hurricane Mitch. Thesis submitted for the Cand. Agric. Degree, Agricultural University of Norway
Longley, K. and Paul Richards (1999) Farmer Seed Systems and Disaster, in Restoring Farmers’ Seed Systems in Disaster Situations: Proceedings of the International Workshop on Developing Institutional Agreements and Capacity to Assist Farmers in Disaster Situations to Restore Agricultural Systems and Seed Security Activities. : FAO, Rome. 123-137.
Louette, D. and M. Smale (1996) Genetic Diversity and Maize Seed Management in a Traditional Mexican Community: Implications for In Situ Conservation of Maize. NRG Paper 96-03. CIMMYT, Mexico D.F.
Pottier, J (1997) Agricultural Rehabilitation and Food Insecurity in Post-War Rwanda. IDS Bulletin 27 (3) 56-75
Sperling, L. (1997) The effects of the Rwandan war on crop production and varietal diversity: A comparison of two crops. AgREN Network Paper No. 75 19-30. Agricultural Research & Extension Network, ODI, London
Synnevåg, G., T. Huvio, Y. Sidibé and A. Kanouté (1999) Farmers’ Indicators for Decline and Loss of Local Varieties from Traditional Farming Systems: A Case Study from Northern Mali. in Serwinski, J., I. Faberová (eds.): Proceedings of the Technical Meeting on the Methodology of the FAO World Information and Early Warning System on Plant Genetic Resources. FAO, Rome
UN (1998) UN Inter-agency Transitional Appeal for Hurricane Mitch. United Nations, Tegucigalpa, Honduras
Wright, M., T. Donaldson, E. Cromwell and J. New (1994) The Retention and Care of Seeds by Small-Scale Farmers. Natural Resources Institute, Chatham UK
 This article is based on a thesis submitted for the cand. agric. degree in Nature Conservation – Tropical Ecology and Management at the Agricultural University of Norway.
 Campaign Coordinator, Friends of the Earth Norway. Contact information: firstname.lastname@example.org
 Professor, Center for International Environment and Development Studies, Agricultural University of Norway, Aas, Norway, and, Senior Advisor to the Director General, International Plant Genetic Resources Institute, Rome, Italy.
 The following indicators were employed: ‘Ownership and standard of housing’, ‘Ownership of land’, ‘Amount of land cultivated’, ‘Engagement in day-labouring’, ‘Destination for the on-farm agricultural production (domestic vs. market) and need to purchase food’, ‘Health conditions and access to health services’, ‘Access to non-agricultural sources of income’, ‘Ownership of livestock’, ‘Ownership of cattle’, ‘Experience of food shortages’, ‘Use of day-labourers on own farm’ and ‘Capacity to lend money to others’
 A seed lot is a particular population of seeds (or crops) that is managed separately because it is seen as having qualities that are distinct from those of other populations of seeds/crops (Louette and Smale 1996). In general, this separate management is based on the seed lots being recognised as belonging to different varieties. A variety is composed of all the seed lots used and recognised as distinct units by farmers, and sharing the same name.