Local varieties of cassava: conservation, cultivation and use in Uganda

The study explored the theory that on-farm conservation of cassava germplasm is influenced by farmers’ traditional and cultural preferences of particular varieties. Traditional knowledge practices that are used for on-farm conservation of cassava germplasm as well as cassava attributes for selection were assessed. The findings obtained from the study indicated that farmers use traditional knowledge to select and preserve cassava germplasm for future use. It was also clear that farmers have their preferences such as culinary attributes, storability in the ground, early maturity and cooking quality to mention but a few that influence the decisions taken to retain or abandon cultivation of varieties. Therefore, by planting varieties in multiples plots, replanting immediately after harvesting, sharing with others in the community and planting disease-free materials, farmers ensure that they preserve varieties of interest for decades. The information generated during this study could inform development policies tailored toward ensuring sustainable on-farm conservation of cassava genetic resources.


Introduction
Crop genetic resources are important in ensuring food security since they provide the raw materials needed for crop improvement (FAO 2010). Farmers over the years have cultivated a variety of crops on their farms that are adapted to particular needs and conditions (Tripp 1996). It is from the human selection practices that crop genetic resources have been preserved for many decades (Gwali et al. 2011). However, the introduction of improved varieties originating from crop improvement programs to combat emergence of diseases and pests as well as other environmental stresses are subjecting several crops genetic resources to genetic erosion (Peroni and Hanazaki 2002;Legg and Fauquet 2004;Alicai et al. 2007). It is important to support conservation of crop genetic resources by fully understanding the traditional knowledge and practices that influence their selection maintenance and conservation (Negi 2010;Parajuli and Das 2013;Wilder et al. 2016). It is important to be mindful of the notion that farmers may not preserve varieties for the sake of conservation but rather because they are adapted to particular needs and conditions that may change at any given time (Tripp 1996). Traditional farmer knowledge is not only important for its socio-cultural value but also for a number of reasons most of which are relevant to the conservation of crop biodiversity. While it has been the responsibility of genetic resource and research centers to recover germplasm before it is lost and to ensure its introduction into germplasm banks (FAO 2010), it is important that the role of local farmers in genetic resource conservation is recognized since they aid in ensuring food and agricultural diversity, valuable landscape, livelihoods and food security. Nevertheless, traditional livelihoods and indigenous plant varieties are increasingly endangered by largescale commercialization of agriculture, population dynamics, land-use/cover changes and the impacts of climate change (FAO 2009).
Cassava (Manihot esculenta Crantz) is one of the crops that have been conserved by farmers for decades, and it is thought to have been cultivated from 3000 to 7000 years ago in South America (Ng and Ng 2002). Cassava is an annual root crop that is widely thought to have originated from the Amazon basin (Nassar 2000). The crop was later introduced to West Africa from which it spread to other African nations (Hillocks 2002;Okogbenin et al. 2007). In Uganda Cassava was introduced between 1862 and 1875 by the Asian traders and spread to almost all parts of the country (Langlands 1972). Its spread and cultivation was due to its adaptability to a variety of agro-ecological conditions and its tolerance to drought (Nassar and Ortiz 2006;OECD 2014). Cassava is therefore regarded as a food security crop in Uganda and Africa as a whole (Balyejusa Kizito 2006). Currently, Cassava is grown in many parts of Uganda as one of the major food crops and is globally gaining economic importance for its starch utilization in food, feed and industry (Legg 1999;Jansson et al. 2009;Nuwamanya et al. 2009;Turyagyenda et al. 2012).
The duration of cassava cultivation in Uganda implies that it has evolved over time and has undergone both environmental and human selection on farmers' fields. Cassava is consequently thought to be harboring important genes that will be of use in the future improvement of the crop (Turyagyenda et al. 2012). However, its genetic diversity is threatened by diseases most notably cassava mosaic and cassava brown streak diseases that have resulted into reduced productivity and loss of germplasm (Legg and Fauquet 2004;Alicai et al. 2007;Kawuki et al. 2016). The current global focus of breeding cassava varieties for industrial purposes will further threaten the cassava genetic resources as farmers will focus on growing cassava for commercial purposes (Jansson et al. 2009). For instance, in Teso region, the farmers are mainly growing NaSe 3 , the improved variety from cassava breeding program. It is difficult to find landraces like Ebwanatereka and Jaribu which were once common (Mr. Okaasai Opolot. personal communication February 10 2014). The loss of cassava genetic resources might compromise future breeding of the crop thus the need for support toward its ex situ and in situ conservation.
Previous studies on management and differentiation of local varieties by farmers in Uganda revealed that there could be variation in on-farm selection and cultivation of varieties influenced by cultural views . For instance, some cultures do not cultivate bitter varieties because they do not consider them as food, whereas other cultures mostly in mid-northern Uganda and northwestern cultivate mostly bitter varieties that are considered to be tastier after processing than sweet varieties . Bitterness is also a security measure as thieves cannot readily use/sell them. Therefore, farmers have perceptions about local varieties that need to be fully understood and integrated into future on-farm conservation policy that is acceptable and legitimate to the local communities. Understanding the socio-cultural factors that influence farmer decision making during the selection and retention of cassava varietal diversity is crucial for future improvement of the crop. The current study was aimed at assessing the farmer knowledge and practices that influence decision making in selection and retention of cassava local varieties in 6 agro-ecological zones of Uganda. The objectives were to document farmer varieties based on farmers' knowledge, to determine the farmers' preferences for cultivation, utilization and selection, and to assess conservation practices undertaken by farmers on the cassava varieties. Undertaking a study that will generate knowledge on conservation of food security crop like cassava is useful in the quest for attaining sustainable development goal 2 which focuses on ending hunger and all sorts of malnutrition by 2030. Furthermore, the study responds to agenda 21 of the earth summit that calls for programs and policies for strengthening biodiversity conservation for sustainable economic development and environmental protection.

Study area
The study was conducted in Uganda which is located in East Africa and lies astride the Equator, between latitudes 4°12 0 N and 1°29 0 S and longitudes 29°34 0 W and 35°0 0 E. Temperatures are in the range of 15°-30°C. The country can be suitably divided into seven broad agro-ecological zones which have similar economic and social backgrounds, and in which ecological conditions (soil types, topography, rainfall), farming systems and practices are fairly homogeneous (UBOS 2010). Six agro-ecological zones southwest, midwest, mid-north, eastern, central and northwestern Uganda were surveyed. In eastern agroecological zone, districts of Kaberamaido, Soroti, Amuria, Serere, Mbale, Manafwa, Busia and Iganga were surveyed. Southwestern districts of Rukungiri, Bushenyi, Buhinga and Ntungamo were surveyed. Mid-western districts surveyed include Kasese, Hoima, Kibaale, Kabalore, Buliisa, Kiryandongo, Kyenjojo, Kyegegwa and Kiryandongo. In northwestern Uganda, districts of Koboko, Nebbi, Arua were surveyed, whereas in central agro-ecological zone, districts of Kalangala, Butambala. Masaka, Kayunga, Luwero, Mukono, Rakai, Nakaseke, Bukomansimbi and Kalungu were surveyed. Mid-northern districts of Gulu, Lira and Apac were surveyed. The study focused on collecting cassava local varieties and the associated traditional knowledge for in vitro conservation. In this paper, a ''local variety'' is a cassava landrace identified by farmers under a single ethnic name which had been grown in the area for at least 20 years. The data collection was conducted between 2013 and 2016.
Local varieties of cassava: conservation, cultivation and… 2429

Sampling strategy
Districts from each of the 6 agro-ecological zones ( Fig. 1) were selected purposively based on their track record in cassava growing. Sub-counties and households were purposively sampled using snowball method where prior information was obtained from extension workers at the districts that they possess cassava local varieties in their gardens. From each district, 4 sub-counties were further sampled. Households within sub-counties were sampled based on information from key informants or from the previous household about the presence of cassava local varieties. A total of 384 respondents were interviewed using a questionnaire across all the selected regions. Variables investigated included farmer preferences for cassava cultivation, utilization and cassava conservation practices as well as cassava local varieties cultivated by farmers. Varieties were recorded as named by farmers in their native languages of the different ethnic groups in the different sites selected.

Data analysis
Data were coded in Microsoft Excel and analyzed using the Statistical Package for Social Scientists (SPSS) version 24.0 (2016 release, _IBM Corp., Chicago, IL, USA) using descriptive statistics. Multiple response data for conservation practices and production of cassava local varieties were grouped together using the multiple response command of SPSS. The descriptive words provided for each use were categorized and counted through the multiple responses. Data were presented as frequencies in the households sampled, percentages and cross-tabs. Cassava local varieties were also listed using the ethnic names provided by farmers (Fig. 2).

Duration of cassava cultivation
The information generated during this study was obtained from farmers who had been cultivating cassava for a period of not less than 5-50 years and more. Mid-western Uganda and central Uganda recorded the longest period of cassava cultivation in the study area with central Uganda having the highest number of farmers who reported having been growing cassava for the last 50 years and more. The mean age for respondents was 47 male and 46 female.

Acreage under cassava cultivation
In the current study, 60% of all households interviewed grow cassava on land between 1 and 4.9 acres followed by 32.7% that grow cassava on \1 acre, 4.9% on 5-9.9 acres and 1.8% on 10 or more acres. In mid-western, central, eastern, northwest and mid-north Uganda, majority of farmers grow cassava on 1-4.9 acres, at 19.7, 16.1, 7.5, 11.4 and 3.4%, respectively (Table 1).

Number of varieties cultivated by farmers
The current study indicated that farmers grow up to a maximum of 14 varieties on their farms. However, most farmers grow between 1 and 4 local varieties ( Table 2). The highest number of varieties grown on farm was reported from northwestern and mid-western Uganda.

Utilization of cassava local varieties
Primary purpose of cassava local varieties in study areas was home consumption and sale (76.2%). 23.3% responded that cassava local varieties are cultivated mainly for home consumption and 1.3% grow local varieties for sale only (Table 4).

Reasons for decreased cultivation of cassava local varieties
Some farmers also believed that there was decreased cultivation of local cassava varieties due to reduced yields (211.3%), pests and disease susceptibility (67%), bitter taste (51.5%), lack of market (19.4%), introduction of improved varieties (220.50%), limited land for cultivation (19.70%) and long maturity period (10.6%). In southwestern and mid-western Uganda, farmers thought that the major reason for decreased production was reduced yields at 42.9 and 42.10%, respectively, while in central, mid-northern and northwestern Uganda, most farmers 35.3, 66.7 and 42.9%, respectively, believed that introduction of improved varieties contributed most to reduced cultivation of local varieties. In eastern Uganda, reduced yields and introduction of improved varieties at 35% were reported as the most contributing factors to decreased cultivation of cassava in the areas surveyed (Table 6).

Conservation practices undertaken by farmers and sources of planting material
Farmers reported that once a variety is introduced in the area for the first time they ensure that it is not lost if it possesses the traits of interest. This is achieved by replanting immediately after harvesting (331.10%), planting in multiple plots (171.70%), sharing with others in their networks to grow as backups (55.6%) and use of clean planting material (41.6%) (Table 7). Therefore, 385.4% reported that they use their own planting materials. But in case there is need to plant a specific variety that is not available on their farms, they can obtain it for free from other households within their social networks (161.2%).
Occasionally stakes are obtained from outsiders for free (24.7%), are supplied by authorities (13.4%) and also purchased from other communities (15.6%) ( Table 8). Local varieties of cassava: conservation, cultivation and… 2439

Cassava local varieties cultivated by farmers
Farmers are known to possess systems of categorizing and recognizing plant and animal species in their communities (Mtunguja et al. 2014). Similarly the naming of varieties as reported by farmers in the current study was based on the different attributes possessed by a variety, the origin of the variety (area where it was obtained) and the person who introduced the variety to the community and the morphological variation of the variety (Table 3). For instance, Welobediyo from mid-western Uganda means ''relax,'' farmers reported that the variety was named because it cooks very fast thus a saying that goes like ''with Welobediyo, a visitor should relax since food will be available in the shortest time possible''. Sibampali from mid-western Uganda meaning ''tighten your trouser'' because variety is high yielding one should have enough energy to uproot it. Mercury variety named because farmers believe that it is as highly marketable as mercury. Gilgil variety from northwestern Uganda named after a village where it was mostly grown. Ocol meaning ''dark'' in Alur because leaves and stems of that variety are dark in color. Omoo meaning ''coming together'' in Lugbara, Clans that used to have conflicts were united after sharing Omoo variety. Katebe meaning ''small chair'' in Bantu language, because the variety grows short stems and branches thus the idiom ''Mumpi nga Katebe translated, as short as a chair''. The naming of cassava varieties as reported by farmers in this study is similar to what has been reported in other regions and crop varieties. For instance, Mehta et al. (2009) reported that the naming criteria of rice varieties in the Himalayas is based on morphological traits, environmental adaptability, agronomic traits, place of origin and local recipes. In Tanzania, however, Mtunguja et al. (2014) observed that famers use mainly well adapted morphological descriptors to distinguish between cassava varieties. They could successfully distinguish between bitter and sweet varieties. In Africa, some communities breed local crop varieties on the basis of indigenous knowledge and use local taxonomy in selecting and naming varieties (Haugerud and Collinson 1990;Almekinders et al. 1994). In mid-west agro-ecological zone specifically in the district of Buliisa, some communities reported a practice of exploring cassava attributes by germinating cassava seeds and evaluating the seedling specifically for culinary attributes, disease susceptibility and yield. In this region, some varieties are named according to the individual who sprouted, evaluated and later distributed them to the entire community. Varieties such as Nyarale ''introduced by Rale'', Nyaru-ucha ''introduced by Ucha'', Nyaruzele ''introduced by Zele'' and Nyaeva ''introduced by Eva'' were generated from such practices. This raises the importance of local breeders within communities and the importance of establishing community-based breeding programs for cassava. The study has shown that, farmers involved in local breeding are highly appreciated to an extent that the varieties generated after such practices are named after them. Therefore, farmers' involvement in planning and execution of breeding strategies could enhance adoption rates of improved varieties.
Occasionally, cassava seedlings randomly germinate and grow among vegetatively propagated plants. However, due to the highly heterozygous nature of cassava it is most likely that the seeds are genetically dissimilar from the parents. And it's also possible that the seedlings are disease free since vegetatively propagated material serve as the major source of diseases (Elias et al. 2000). Farmers normally get interested in these seedlings, evaluate and protect them for the next growing season if they possess agronomic and culinary values. In this way, farmers incorporate genetic variability from sexual reproduction into the local cassava variety gene pool.

Utilization of cassava varieties
The study indicated that most farmers 76.2% cultivate cassava local varieties for both home consumption and sale. With this category being dominated by farmers from midwest (32.5%), northwest (17.8%) and central Uganda (11.5%), respectively (Table 4), in mid-western Uganda farmers with small acreage of land grow cassava mostly for home consumption. Farmers from central Uganda where Banana is a major food crop also commonly grow cassava for home consumption, consumed fresh as snacks and making flour for pancakes for sale. During the survey it was noted that commercial farmers mostly grow improved varieties that are known to be high yielding with early maturity period and do not care much about selection and on-farm retention or conservation since improved varieties are distributed from government and non-government organizations (NGO) periodically. The growing of local varieties mostly for home consumption has ensured onfarm conservation, since farmers always keep a few but a variety of plants for food security and also to complement the positive and negative traits that could be in the local varieties. Similar to previous findings from other regions, in the current study most farmers believed that cassava local varieties are better for food security than improved varieties because they have a longer storability in the soil. Bitter varieties make good flour for food and local beer and therefore helpful during drought and also thought to be less susceptible to disease (Mtunguja et al. 2014;Akintunde and Obayelu 2016). Bitterness is also thought to be a food security measure since it is avoided by thieves. These findings emphasize the need to incorporate farmer preferences in conservation and breeding strategies in order to improve productivity and sustainable use of cassava genetic resources.

Production dynamics of cassava local varieties
The farmers that participated in the study had been cultivating cassava for not less than 5 years up to 50 years and more. They were therefore believed to be knowledgeable about cassava production dynamics over the years. Most farmers believed that cultivation of cassava local varieties has increased in the past few years (Table 5). The major reasons for increased cassava production in general included: tastiness (144.20%), high yields (165.1%), early maturity (76.10%), disease tolerance (48.10%), good cooking quality (103.9%), marketability (27.5%) and storability (35.10%). Farmers emphasized culinary attributes (taste), high yield and cooking quality as the major reason for variety cultivation and retention. Elias et al. (2000) observed that while farmers generally prefer high yielding varieties, they may preserve lower yielding varieties in parallel with more productive varieties, due to cultural preferences such as taste or cooking quality. In the process, farmers manage the risk of a calamitous crop loss by keeping several different varieties in production at the same time and often in the same field. In northwestern, mid-northern and some parts of eastern Uganda, cassava is mostly processed prior to consumption due to the high cyanogenic compounds in the cultivated varieties . In these regions, bitter varieties are considered tastier after processing than the sweet varieties and thus high starch content is a key trait that farmers look for in the varieties cultivated.

Conservation practices undertaken by farmers on the cassava varieties
It is not unusual for farmers to exchange stem cuttings with their neighbors and neighboring communities, resulting in fields with a mixture of local cassava varieties (Andersson and de Vicente 2010). And it is not common that cassava planting materials for local varieties are sold or bought. Although farmers may not deliberately conserve cassava genetic resources, they do so through traditional practices that ensure on-farm retention of varieties over years.
Cassava farmers in Uganda use various practices to ensure retention of varieties on farm. They plant varieties in multiple plots (40.3%), replant immediately after harvesting (84.2%), share with others (21.1%) and use clean planting material (17.4. %). In case planting for some reason does not happen immediately after harvesting, farmers reported that stakes can be kept viable for some time under the shade preferably in an upright position until the next planting. These methods ensure that in case of any catastrophic events, varieties are not lost. With the prevalence of viral diseases, some farmers have learnt to recognize symptoms associated with cassava viral diseases and select planting material from plants that are asymptomatic (Table 7). This strategy not only ensures availability of food and seed for the next planting but also reduces the spread of diseases on farm and within the community. The farmers were commended and encouraged to continue planting disease-free cuttings in order to reduce the spread of diseases.
The information generated during this study indicates that farmers in most areas where cassava is grown have more or less similar practices of maintenance and preservation of cassava genetic resources Elias et al. 2001;Mtunguja et al. 2014;Akintunde and Obayelu 2016). However, it was observed during this study that farmers from the study areas are not aware of the scientific importance of conservation of cassava genetic resources rather preserve varieties for the sake of supporting their households' needs. Therefore, government and development partners should invest in sensitizing farmers about the importance of on-farm conservation of crop genetic resources for future use.

Conclusion
The information obtained from this study indicates that farmers are still growing cassava local varieties despite the distribution of disease-tolerant and high yielding improved varieties. It is also clear that farmers have preferences that influence the decisions taken to retain or abandon particular varieties. Based on their local knowledge, farmers understand the need to grow a diversity of cassava as this can help in times of scarcity. Although improved varieties have enabled continued cassava production in Uganda and the region, their cultivation and adoption may reduce the number of varieties cultivated by farmers leading to erosion of genetic diversity (Tripp 1996). As the government of Uganda promotes research toward cassava improvement, measures should be put in place to ensure protection and sustainable utilization of its genetic resources.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.