a magazine on ecological agriculture
a one stop treasure of practical field experiences

Local seed systems: For enhancing food security and farm resilience

There is a declining trend in the species diversity among small millets. While varietal improvement research conducted by the government hardly reaches those who need them, efforts made by NGOs are sporadic and on a very small scale. There is not only a need to bring in these institutions to work together but work in such a way that the local seed systems are protected and nurtured. RESMISA model is an attempt towards strengthening local seed systems integrating the strengths of various varietal improvement efforts.

Baby trial plot at Bero, Jharkhand. Photo: Dhan Foundation

Baby trial plot at Bero, Jharkhand. Photo: Dhan Foundation

Small millets are hardy, superior in nutritional qualities, meet food and fodder requirements, require few external inputs and also sequestrate carbon. More importantly they have significant cultural value due to their long history in the South Asian Region. Despite these advantages, there is a drastic decline in the area under small millets along with loss of varietal diversity.

In the last two decades decline in inter and intra-species diversity among small millets is being observed in various degrees in India. Further, the released varieties, output of varietal improvement research from national agricultural research systems, have not penetrated most of the cultivation areas.

Presently in India, the national agricultural research system (NARS), is the primary actor focusing on varietal improvement and comes out with broad general recommendations. The research is conducted in research stations located far away from the cultivating locations, with limited participation by farmers. Their penetration to major cultivating regions has been very limited. On the other hand there is limited number of non-governmental organisations (NGO) which focus on on-farm conservation of local varieties, working with farmers in the remote areas. These are very location specific and project specific programmes and have limited spread. Besides these two, informal seed systems prevail in the villages. These systems have an essential role in promoting food security, and are central to the conservation of biodiversity, in the face of rapidly dwindling global genetic diversity. While all the three systems having their own strengths are playing significant roles, which are complementary in nature, they have been doing so independently within their frameworks. There is a need to bring them together, and build on their strengths so that a synergy is created in achieving the needed varietal improvement and enhancing varietal diversity of small millets in the cultivating regions.

Understanding the need for integration of these complementary roles, an integrated model was designed and attempted by DHAN Foundation, a development NGO, under ‘Revalorising Small millets in Rainfed Regions of South Asia’ (RESMISA) project funded under CIFSRF by IDRC and DFATD, Canada. The project was initiated in 2011 and focuses on four small millet crops, namely, finger millet, little millet, barnyard millet and kodo millet in five sites (3 in the state of Tamil Nadu, one each in Odisha and Jharkhand states) located in different agro-climatic regions of India. The project has created a platform where in farmers and their organisations like federation/ association of self help groups (SHGs) from the sites, scientists and field staff of DHAN Foundation, and scientists from Tamil Nadu Agricultural University and All India Coordinated Small Millets Improvement Project of ICAR are involved to continually interact through the course of study. The guiding methodology of the project is farmer-led research that builds on indigenous knowledge systems and is complemented by gender sensitive scientific and participatory methods.


RESMISA model (see flow chart) integrates on-farm conservation, participatory varietal selection (PVS) and community based seed systems. The research begins with understanding the present status of varietal diversity and seed dissemination system in the sites. Different tools such as field surveys, transect walks, biodiversity contests and interaction with the local farmers were used for scouting the varieties of focused crops that are under cultivation in each of the study sites. This is followed by biodiversity blocks and morphological characterisation of local varieties.

The varieties identified are classified into popular and vanishing varieties. While special attention is given for on-farm conservation of the vanishing varieties through nodal farmers, the popular varieties enter the PVS trials as local checks. In PVS, experimentation on acceptability of suitable materials in farmers’ fields included mother trials, baby trials and informal research and development. Mass multiplication and promotion of preferred varieties emerging from PVS and onfarm conservation of vanishing local varieties are carried out in community seed systems.

Integrated model for On-farm conservation, Varietal improvement and locally embedded seed systems
Integrated model for On-farm conservation, Varietal improvement and locally embedded seed systems

Varietal diversity and seed systems

Though there were many small millet varieties in the sites, not more than two varieties covered majority of the area in each of the four crops studied. Further varietal diversity at hamlet level was very limited in all the sites. This situation clearly indicated the need for increasing varietal diversity of small millets in the sites.

Further, more than 90% of the farmers use farm saved seeds of small millets. They do not follow seed selection procedures leading to mixing of varieties. Under these conditions the best strategy for varietal improvement and enhancing varietal diversity is creating more options at the individual farmer level regarding the preferred varieties and promoting simple quality seed selection methods among the farmers. For generating more options of preferred varieties, PVS was attempted in the sites.

Participatory varietal selection

Ragi varieties evaluated at Anchetty, Tamil Nadu. Photo: Dhan Foundation
Ragi varieties evaluated at Anchetty, Tamil Nadu. Photo: Dhan Foundation

A set of promising varieties (8-10), comprising both traditional as well as released varieties of each small millet crop, were short listed based on farmers preferences for the desirable traits. In 2011 and 2012, these varieties were evaluated in each site on farmers field following farmers’ management practices (Table 2).

These varietal trials, known as mother trials, were monitored by the trained field staff of DHAN Foundation under the guidance of technical personnel. The participating farmers, numbering on an average of 15 to 20, took the responsibility of raising the crop and harvesting, apart from participating in the evaluation process. Varieties suitable for the site were identified based on the results of quantitative analysis and farmers’ preference analysis. In quantitative analysis, the standard procedures in collecting and interpreting the data of growth and yield parameters were followed. The preference analysis used the preference scores given by the groups of farmers, male and female groups separately, for each of the varieties in the trials.

From two cycles of varietal evaluation in mother trials it was possible to identify 1 to 4 most preferred varieties (both released and traditional varieties; see Table 1) in each crop for respective sites. The performance of these identified varieties was further assessed individually against the farmers’ varieties in more number of farmers’ fields (34-64) in baby trials under farmers’ management practices. The plot size for each variety was larger (minimum of 200 m2) than in mother trials. Such of the varieties whose performances were accepted by majority of the local farmers were finally selected, mass multiplied and promoted by reaching more number of farmers under informal research and development activities. Hence, it was possible to enhance the varietal diversity in each project site within a short period of 3 years. Also, a large number of farmers had seeds of their preferred varieties in their possession. This is in contrast to conventional breeding programme, wherein it would take a minimum of 8 to 10 years for improved seeds to reach the farming community.

Project sites No. of varieties present Popular varieties
Traditional Released Number Name
Finger millet
Anchetty 2 3 2 GPU 28 (R), INDAF 5 (R)
Bero 4 2 Demba (T), Lohardagiya (T)
Jawadhu Hills 2 1 Muttan kelvaragu (T)
Semiliguda 19 2 4 Bati (T), Mati (T), Kalakarenga (T), Sunamani (T)
Little millet
Jawadhu Hills 9 3 Sittan (T), Karusittan (T), Vella samai (T)
Semiliguda 8 2 Bada saon (T)
Barnyard millet
Peraiyur 3 1 Sadai (T)
Kodo millet
Peraiyur 4 1 Siru varagu (T)
Table 1: Status of varietal diversity in small millets at the study sites
R- Released variety; T- Traditional variety

On-farm conservation

To conserve the local genetic resource of a crop, two bio-diversity blocks were established in the fields of interested farmers. These blocks also served as a means to create awareness among the farmers to witness the varietal diversity available within their own area.

Seed purity is maintained by collecting selected panicles during harvest. Also, the seeds of these varieties are sent to the research centres for further purification and for conducting morphological characterization studies. Further, to support interested farmers in on-farm conservation, an initiative to generate biodiversity funds in each of the project sites, has been undertaken.

Community based seed systems

Specific efforts were made for involving local community organisations like SHGs and their federation /associations, in the project sites. Exposure trips were organised for the trial farmers to widen their awareness on the loss of varietal diversity. They were also involved in managing the biodiversity fund. Federation/ association of SHGs are in a position to take the responsibility for on-farm conservation and promotion of preferred varieties emerging from PVS due to their social and financial capital accrued over the years. Interested trial farmers from these SHGs from various parts of the research site formed into a farmers’ group named RESMISA research coordination committee (RRCC). RRCC will handle the responsibilities of quality seed production, seed purification and seed dissemination on a sustainable basis.


A woman farmer doing preference analysis in Anchetty, Tamil Nadu. Photo: Dhan Foundation
A woman farmer doing preference analysis in Anchetty, Tamil Nadu. Photo: Dhan Foundation

The integrated model is unique in several aspects. It is a model where in large number of farmers (578 men and 333 women) were involved at various stages of the whole process, which helped them to understand the basic and simple principles behind each of their actions.

Opportunity for the scientists and farmers to work together helped to produce meaningful results through better understanding and appreciating each others experience. Though yield appeared to be the main criteria for assessing the superiority of a variety, the trial farmers also took into consideration several other traits/ dimensions before preferring a variety like crop duration, nonshattering of grains at maturity, non-lodging, uniform maturity, and good fodder yield. Women farmers, in particular, were more concerned with grain quality traits, such as colour, taste, grain hardiness and keeping quality.

The experience proved that the traditional varieties have potential to address the needs of farmers which is generally ignored by the formal system. Collaboration of partners made it possible to share elite germplasm across the states, thereby widening the varietal options available for the farmers living in remote sites. The model also made it possible to integrate varietal improvement and varietal development with the existing development activities of SHGs, like microfinance.

Varietal improvement has to be a continuous process to meet emerging requirements of the farmers (though happens on an episodic manner in the project). Housing it within the community seed systems gives scope for continuing the same beyond the project period. Further this model creates space for inclusion of potential local germplasm in the formal research stream of NARS and for productive collaboration between formal and informal seed systems.

Integrating on-farm conservation, PVS and community seed systems can effectively result in varietal improvement and increasing varietal diversity of small millets. Collaboration of multiple stake holders namely farmers’ organisations, development organisations and research organisations are prerequisite for implementing the integrated model. Further strong adherence to gender sensitive farmer-led participatory research methodology and facilitation to set an effective working relationship by overcoming the existing power equations are essential for making the collaboration successful. This approach is replicable across the geographies and across the crops.

Recognising the need for integrated approach and collaboration across the partners, there is need for institutionalising these in the ongoing efforts for enhancing the livelihoods of small farming households. In the process, necessary efforts need to be taken for including potential traditional varieties in the formal public seed system and in other State funded crop support systems, and securing farmers’ rights related to them.

M Karthikeyan and C S P Patil

M Karthikeyan is the Principal Investigator, RESMISA project and Program Leader for Rainfed Farming Development Program, DHAN Foundation, India.

C S P Patil is the former Principal Scientist, University of Agricultural Sciences, Bangalore, India.


Gill, T. B., Bates, R., Bicksler, A., Burnette, R., Ricciardi, V., & Yoder, L., Strengthening informal seed systems to enhance food security in Southeast Asia, Journal of Agriculture, Food Systems and Community Development, 2013, 3(3), 139–153.