Restoring Our Seed: Extension Program to Train Farmers in Ecological Seed Crop Production

Final Report for LNE02-160

Project Type: Research and Education
Funds awarded in 2002: $135,000.00
Projected End Date: 12/31/2004
Matching Federal Funds: $15,000.00
Matching Non-Federal Funds: $54,000.00
Region: Northeast
State: Maine
Project Leader:
Expand All

Project Information


As markets for organic produce expand, the limited availability of organic seed bred for organic systems has become a major bottleneck. Just as the management of soil fertility, disease and pest control is fundamentally different in organic and in conventional farming systems; so too, crop genetic management is not the same. The majority of commercially available seed is bred by conventional seed companies that use agrochemicals for disease and pest protection.

Restoring Our Seed (ROS) trains New England organic farmers in the knowledge and skills for organic seed production and on-farm crop improvement. Farmers learned how to integrate seed production into an ecological whole farm system, incorporate habitats for pollinators, select seed-crops for disease resistance and local adaptability, harvest and clean seed, and market or use their improved seed for elite vegetable lines.

What is Ecological Plant Breeding?
What Darwin observed, and ecological plant breeders harness, is the power of selection from variable gene-pool populations. When the ecological breeder selects under the natural pest and disease pressure of the organic field, plants with favorable traits survive or are chosen.  Weaker plants die or are removed. Heirloom or ‘farmer’ varieties carry wide, variable gene-pools that have evolved survival mechanisms by adapting to their environment year by year. Farmer varieties tend to have richer diversity, flavor and natural disease resistances but lower yields than modern pedigree cultivars. Modern cultivars are not screened by natural pressures, but are bred in conventional fields for uniformity, appearance and shelf-life, with ample water and agrochemical protection, and become dependent on irrigation and agrochemicals to survive.

Durable Horizontal Resistance
“Most of the plant breeding programs of the twentieth century have totally failed to achieve their objective of increasing resistance to disease and pests. We are actually increasing the susceptibility of many of our crops to disease and parasites.” Dr. Raoul Robinson

Restoring Our Seed is teaching ecological plant breeding using evolutionary selection principles. In contrast to the conventional breeding approach that selects for single gene traits, we are selecting for complex linked traits that enable crops to thrive under the combined pressures in organic fields. Single gene resistance is often short-lived because pathogens evolve. It lacks a complex of genes that can adapt to the new strains of pathogens or pests that constantly evolve. Ecological breeding draws on the genetic biodiversity of heirloom and indigenous varieties, with complex genetic diversity. Multiple-linked genes have wider gene pools to survive evolving diseases and to adapt to the variable weather extremes of climate change. Participatory crop improvement to select for taste, nutrition and adaptability to local climate conditions is conducted with traditional farmers worldwide; however little work has been done in the United States.

Quality, particularly flavor, unlike yield and disease resistance, is not directly influenced by natural selection. For example, the market for locally marketed tomatoes demands sweet, rich flavor with some firmness. Organic farmers need high quality cultivars to compete in value-added niche markets. Therefore, introgressing and selecting high quality germplasm in a bulk population is important. The successful introduction and long-term sustainability of participatory breeding depends on direct benefits in improved crops that increase income for farmers and related businesses.

Gene-pool-based Plant Breeding
Since many heirlooms today lack the genetic diversity of early landrace progenitors, our strategy to restore landrace characteristics for ecological plant breeding is to generate  ‘composite cross’ gene-pools by inter-crossing varieties successful in local organic field conditions combined with breeding in lines with specific favorable features, then exposing the populations to natural selection. In this way the populations have an enhanced genetic capacity to respond to fluctuating local conditions and unanticipated stresses, such as global climate change or the evolution of pests and pathogens.


Restoring Our Seed (ROS) is a participatory plant-breeding program to generate locally-adapted organic seed produced and screened in organic systems in New England. ROS coordinates a team of organic farmers, cooperative and ‘lay’ extension and seed companies to conduct a training program in organic seed production and on-farm participatory crop improvement. The program is developing the scientific knowledge, technical skill and inspiration for ecological seed crop management – to reduce use of crop protection chemicals, increase yield, enhance on-farm sustainability and increase profits. ROS conducted conferences, seminars and field days, produced an on-line manual posted on, established a regional seed-saving network, and generated participatory breeding projects. A ‘Seed Stewards’ educational curriculum was piloted in the Belfast Middle School, and subsequently recognized as Maine’s exemplary model agriculture education program. The project represents a community-based approach to strengthen our regional seed supply.

Performance Target:

1. To conduct a farmer-training program in organic seed crop production to increase on-farm sustainability and farmer profit, with supporting community education.

We achieved and surpassed our targets to conduct a training program, to increase on-farm sustainability and to provide community education. We conducted a year-round training program for four cycles involving farmers primarily in Maine, Massachusetts, Vermont, Connecticut and New York, as well as in New Hampshire, Canada and the Southern states. See Section 8 for outreach details.  The greatest profit will be realized by farmers adapting varieties to local conditions, not from sale of unimproved seed to companies.

2. To increase the number of farmers and gardeners trained in organic seed production, on-farm selection and crop improvement.

ROS achieved and surpassed our target of training farmers and gardeners.

3. To increase the quality and quantity of organically grown open-pollinated vegetable seed by growing-out and improving open-pollinated seed.

We achieved Target 3 by conducting a program to improve crops, whose greatest results require at least five years to bear significant fruit. Breeding projects and distribution of seed are addressed in Section 5.


Click linked name(s) to expand
  • Jeremy Barker-Plotkin
  • Will Bonsall
  • Bryan Connolly
  • Brett Grohsgal
  • Dr.Mark Hutton
  • Jack Kertesz
  • Jay Leshinsky
  • Frank Morton
  • Dr. John Navazio
  • Bryan O'Hara
  • Dr. Raoul Robinson


Materials and methods:

Materials: open pollinated genetically variable vegetable seed planted in organic fields with local disease and pest pressures, pedigree cultivars with desirable traits.

Original Methodology:
Each demonstration site had a plot of a self-pollinated crop (tomatoes) and one plot with cross-pollinated species (cucurbits and brassicas). At our first Winter Conference, farmers discussed which crops to improve. Farmer field days demonstrated each step in how to observe, evaluate and select for horizontal resistance (HR) to the disease complexes found in each demonstration site. We demonstrated how to test and increase the resistance of an heirloom vegetable to a local disease by deliberately creating ‘disease nurseries’ in which to challenge and screen the vegetables we sought to improve. The disease was encouraged by growing the same crop for several years and by growing varieties with degrees of susceptibility to the disease. Varieties of plants with degrees of disease resistance were bred together for an increased complexity of resistances. Pedigree cultivars with total resistance to a strain of a pathogen were not grown. This approach can be used by farmers to select and breed for desired traits.

Demonstration Sites
Each plot included:

1) a variety of the crop to improve,
2) a variety resistant to the disease, and
3) a variety extremely susceptible to the disease.

Each plot had three replications (reps). The purpose of the reps was to affirm the validity of observations by minimizing variations caused by changes in soil type, fertility levels, moisture and solar exposure. Each field trial was in a block design as close to a square as possible. Border rows in each plot were planted with a cultivar of the same crop, in this case the susceptible variety, to minimize the edge effect. Outside rows of crops perform better than interior rows because of less competition for sun, nutrients and water.

Trial Evaluations: Measuring Resistance
Disease pressure, insect attack, heat, drought or cold stress provide an opportunity to measure which varieties exhibit higher resistance/tolerance to environmental pressure. Plants were measured on a 1-9 scale for the presence of disease by farmers and extension staff. A score of ‘1’ represented the lowest level of a trait, and ‘9’ represented the strongest. We scored individual plants and varieties, measuring the severity of disease symptoms and the rate that disease spread through each variety. At the end of the season, we had a total score for each plant

Selections: Survivors of Adversity
We selected seeds from plants that exhibit higher levels of resistance, nature’s survivors. In self-pollinated tomatoes, selections were made several times each season, since there is little risk of crossing with susceptible plants. In cross-pollinated crops, the selection of individual plants from the improving variety was made early in the season to prevent crossing with susceptible plants. After the selections were made, all other individuals from the improving variety, the susceptible variety and the resistant variety were rogued out. Seed saved was used the following year as the new improving variety. In subsequent years we included the original unselected variety to compare with the selected version, to assess the progress we made. After three years we measured improvement resulting from three successive generations of selection that strengthened the multiple genes conferring resistance. Our trial site methodology was adapted to fit the constraints of typical market farmers.

What worked and what didn’t:
Timing: Market farmers do not have much time in the summer to attend field days. Attendance was far higher in fall and winter events. We adapted by photo- documenting the summer field selection methods for use in winter trainings. Methods: Our original demonstration site methodology was too complex. We adapted our methods to be more feasible for busy market farmers, by eliminating the susceptible and resistant checks. Below is our revised methodology:

                                        On-Farm Crop Improvement Guidelines
a. Decide what crop or cultivar has potential for improvement. Select traits to improve based on the variations of the plants in your field and your market need, such as cold tolerance for year-round harvest, resistance to disease or attractive appearance.

b. Trial and compare the same crop from different companies. Work on the best lines with characteristics you seek from the widest gene pool you collect. Use your intuition.

c. Plant the crop in well-spaced rows to evaluate each plant. Grow as large a population as possible to maintain a diverse pool of traits. Isolate to prevent accidental cross-pollination, unless you are deliberately crossing to create new crop combinations. Grow islands of native habitats to attract beneficial pollinators and predators of insect pests.  

d. Screen out the weaker plants using your typical management for all plants. Don’t baby the crop. Remove or market less desirable plants before flowering to prevent cross- pollination with superior plants. Keep the whole plant in mind as you select so as not to unwittingly select out valuable but less visible traits. Save the best plants for seed.

Tip: For cross-pollinating crops of pre-flower green leaves (ie: brassicas), evaluate, taste and rogue out the less desirable plants to sell or eat. Let only the best plants cross-pollinate. For crops of post-flower fruits (ie: cucurbits) evaluate and taste the first fruits of all plants, tag the best plants, then rogue out the poorer plants (alas nothing available to sell at this stage) to prevent lower quality plants from pollinating the better plants.

e. Harvest the improved line. Clean, remove any smaller, lower quality seed. Air-dry and store. Repeat your selection process year-by-year.

Research results and discussion:

a. Farmers implement whole farm cropping system with appropriate spacing and beneficial insect habitats
b. Farmers conduct on-farm selective seed-saving and crop improvement
c. Farmers learn small-scale seed harvesting and cleaning

By the completion of the funding period:

a. About 200 farmers and gardeners had participated in the whole farm cropping system training sessions conducted by Frank Morton, a national leader in ecological seed cropping systems.  In our final survey of 28 of our most active participants, 20 adapted practices to improve pollinator ecology of their farms, that include: establishing and maintaining hedgerows, keeping honey bees, building, ‘orchard mason’ bee houses, adding small ponds for wildlife diversity, inter-cropping more annual and perennial insectary habitats for beneficials, allowing brassicas to flower through the seasons, and sowing cover crops preferred by pollinators. We can conservatively extrapolate that at least 50 more farmers that attended ROS conferences adapted these practices, based on verbal interviews.

b. At least 60 farmers have been involved in an average of four out of a dozen possible on-farm selection and breeding projects.

c. Seed cleaning training, conducted by Will Bonsall, Tom Stearns and Nathan Corymb, was taught at all four conferences, reaching about 400 farmers and gardeners. In addition, the Public Seed Initiative (PSI) brought their mobile seed cleaning unit to summer NOFA and Fall MOFGA events, and Hampshire College partners established a seed cleaning center.  These hands-on teachings were enthusiastically embraced by attendees at all skill levels. In our final survey 19 out of 28 reported learning seed cleaning techniques that increased their efficiency. We extrapolate that at least 100 more farmers benefited.

Each year many of the same farmers and gardeners returned to the conference to exchange experiences, new knowledge and share seed. At each conference we attracted people with wide differences in skills and experience levels. We provided instruction at these events suitable for all levels to demystify the art of selection and breeding. While that was daunting to manage well, the experienced seed-savers and larger-scale seed producers in attendance inspired the beginners.

Course Corrections – Noteworthy Unexpected Learning and Change of Methodology
From the top-down training approach of our first conference in Maine in 2003, the training process evolved into a farmer-led participatory program. At the inception of the project, we were taught the breeding methods of university and industry-trained breeders, including crossing desirable traits to produce hybrids and recurrent mass selection for horizontal resistance. We discovered through our own experience how to create composite cross gene pools, and then learned that this is called ‘evolutionary plant breeding’ in the literature. Our gene pools are being selected to be delicious and attractive, and consequently are highly marketable, non-uniform varieties. We also discovered through our own experience that over-wintered brassicas develop a systemic resistance to flea beetles. Flea beetles are a critical pressure in spring organic mesclun production. The most significant result of our project is the farmers’ discovery of the gene-pool-based breeding approach that recreates the dynamic genetic resilience of landrace and heirloom varieties for today’s organic market farmer.

Market farmer Bryan O’Hara reports, “There is some resistance to insects which is nothing short of astounding where the flea beetles do not assault over-wintered brassicas in our field. I could plant mizuna in the spring, right next to this over-wintered mizuna, and it would be absolutely devastated by flea beetle. The over-wintered mizuna has no flea beetle damage.”

O’Hara continues, “Not only are these plants resistant to the flea beetle in terms of some physiological change or some difference in the plant itself, but the flowering of the plants attracts a phenomenal number of beneficial insects that are the predators of the caterpillars that assault my brassicas later in the season. It became clear that the wasps that are your allies are going to feed on the flowers as a nectar source for food for themselves where they intend to lay their eggs, and their larva are going to develop inside the caterpillar that feeds on that crop. It makes all too much sense and I started thinking that there might be something vastly missing in not producing flower and seed crops. By cutting the plant growth short and always harvesting all the vegetables and never allowing for flower and seed production, how would these wasps proliferate on my farm?”

There are very good reasons to think that over-wintering annuals for seed crops can bring great benefits to many regional farms. Thus, we have increased our emphasis on cold-hardy greens among our breeding projects. As a result of the new discoveries and benefits, there is a high level of ownership and enthusiasm generated from our project by the participating farmers.

Participation Summary


Educational approach:

Instructional materials have been produced at increasing levels of sophistication:
1. An instructional brochure was distributed to about 1000 growers.
2. A seed-saving curriculum was distributed to 230 teachers. It is available free on-line on <>. It was downloaded 5,868 times in 2006 alone. An educational seed-saving course was conducted in ME, CT, MA and Toronto.
3. A handbook on Organic Seed Production and Saving was written by Bryan Connolly with CR Lawn, published by NOFA-MA.
4. An illustrated Restoring Our Seed Manual of 200 pages was distributed to all 400 attendees at ROS conferences. 35 copies have been sold. The entire manual is posted on our website for free download. 420 participants attended the three-day ROS winter conferences. 280 people attended the 14 ROS field days at MOFGA’s Scionwood Exchange/Seed Swap, Small Farm Field Days, Common Ground Country Fair, at Maine Cooperative Extension’s Highmoor Farm, at the NOFA Mass Summer Conference and in Connecticut, as well as students at on-site Middlebury College and Hampshire College workshops. Presentations were given at the New England Vegetable and Berry Conference, Carolina Farm Stewardship Conference, North American Community Garden Conference in Toronto, Community Food Security Conference, Midwest CSA Conference, Maine School Garden Conferences, the two Seeds and Breeds conferences and the European Organic Plant Breeding conference <cost8>, 60.

ROS EXHIBITS at NOFA Summer Conference Amherst, MA 2003-5, at MOFGA Common Ground Country Fair Unity, ME 2003-6, at Seeds & Breeds Conference Des Moines, IA 2005.

Articles and Ads: Articles in Fedco Seeds catalog, MOF&G (MOFGA newspaper), Saving Seeds (GE-Free Maine newspaper), Acres USA, advertisement in the Natural Farmer (NOFA newspaper), coverage on Seeds of Change website, MOFGA and all NOFA websites, link to Organic Seed Alliance website. Mailings to core mailing list of over 400 farmers and gardeners who have attended at least one event during our project.

CR Lawn and Eli Rogosa Kaufman attended the CFSA Conference as consultants to the SARE-funded Saving Our Seed (SOS) project at its inception. Lawn, Kaufman and Bryan Connolly attended SOS Field Day at Even’ Star Farm in Maryland to coordinate further with the project. ROS collaborated with Organic Seed Partnership in New York. Elizabeth Dyck from OSP presented at our fall conference. Matt Falise and Michael Glos provided seed for Brandy Rose tomato, and made the initial crosses for the Wonder Pickle cucumber breeding project.

ROS Demonstration Sites
Mark Hutton, Highmoor Cooperative Extension Farm, Monmouth, ME
MOFGA Education Center, Unity, ME
Bryan Connolly, Mansfield Center, CT
Bryan O’Hara, Lebanon, CT
Jeremy Barker-Plotkin, Simple Gifts Farm, Belchertown, MA
UMASS Farm, Ruth Hazzard, UMASS Extension, Deerfield, MA
Middlebury College Garden, Middlebury, VT
Shooting Star Farm, Canaan, ME

Massachusetts Demonstration Site
The MA site established cooperation with University of Massachusetts Cooperative Extension, Ruth Hazzard and Plant and Soil Science Dept. faculty Robert Bernatsky, to produce cold-hardy greens seed crops and to investigate flea beetle preferences. We also provided funding and technical support for UMASS ‘GardenShare’ student garden to purchase open-pollinated seed, and conducted a seed-saving training program.

Additional Project Outcomes

Project outcomes:

Impacts of Results/Outcomes

The farmers who participated in ROS worked with the following breeding methodologies:

a. Selecting to stabilize a hybrid,

b. Introgressing desirable traits by crossing modern cultivars and heirlooms to create a stabilized open pollinated variety,

c. Selecting a variable open-pollinated heirloom to improve flavor and horizontal disease resistance,

d. Recurrent mass selection under typical field pressures in cold-climate New England, for multiple-linked traits of cold-hardiness, disease resistance, flavor and appearance

e. Composite crosses or gene-pool-based ecological breeding to increase genetic variability, then screen through natural selection for over-wintering, then selecting the survivors for the quality characteristics of flavor and appearance

ROS Participatory Breeding Projects:

1. Pruden’s Purple
A thick-skinned, firm, early tomato producing good market yields. Selected for early blight resistance and attractive shape.

2. ‘Brandy Rose’ (Brandywine x Rose de Berne)
Brandywine is larger, later, less uniform, tender, less productive with rich flavor. Rose de Berne is medium size, beautiful, almost blemish-free, pink translucent skin covering luscious, rose-pink flesh that is complexly sweet, spicy, and juicy. We selected for best qualities of each.

3. Wonder Pickle
Selected for: plant health, reduced foliar diseases, dark color, good yield over the season, and flavor. (Conquest x Clinton x Wautoma)

4. Potato Dance – A combination of Blossom (the maternal plant) with Caribe, Island Sunshine, Prince Hairy, Green Mountain, Purple Peruvian. We selected for resistance to Colorado potato beetle combined with good flavor.

5. Winter Luxury Punpkin -Saved seed from fruits storing longest with strongest stems.
6. Dancing Greens
Let cross-pollinating varieties grow together. Select for unique salad greens.
a. Mizpoona (Mizuna x Tatsoi),
b. Purple Mizuna <>
c. Scarlet Mizpoona (Mizuna x Tatsoi x Scarlet Ohno Turnip)
d. Dancing Kales: John Sokoloski over-wintered 30 kales in Connecticut with no protection, and let the surviving plants go to seed and cross. We are selecting for a diverse gene pool of cold hardy kales with good flavor.

Even’ Star Cold-Hardy Winter Greens: We have established a close collaboration with Maryland market grower Brett Grohsgal, who has developed his own lines of winter-hardy salad greens over the past fifteen years. Eleven farmers received breeding packets of one or more of his Arugula, Thick-stem mustard, Tatsoi, Smooth Kale, Collards, Tenderleaf and Segregating mustards. Grohsgal provided guidelines for New England farmers to select for cold-hardiness in our more challenging climate. Four NE farmers succeeded in over-wintering the Arugula in 2005-6. Even’ Star Arugula proved so superior in summer trials that Fedco Seeds added it to their catalog.

Bryan O’Hara’s Winter-Hardy Greens Gene-Pools: CT market grower O’Hara successfully over-winters greens in low tunnels. After multiple cuttings, he allows diverse brassica varieties to cross, creating gene pools with hybrid vigor yielding high production of a variety of leaf shapes, colors, textures and flavors highly valued by his customers. At our fall conference O’Hara provided seed for a mizuna/tatsoi cross, a maruba/mizuna/tatsoi cross, and a gene pool of 7 mustards which have been selected yearly since 1999 for winter hardiness, flavor and texture.

Heirloom populations were observed to differentiate according to environmental field pressures and genetic diversity. A wide range of responses to pathogen pressures was observed in both self and cross-pollinating populations. An accumulation of resistance responses in the heirloom tomato, brassica and cucurbit genotypes was measured over the four year period, however the dramatic weather variability over the four years did not allow for even scoring comparisons. This may also be related to both the diversity within the gene pools, and rate of evolution of the pathogens. In the first season, the composite cross gene pool populations yielded more than the parents. After four years of selection, an average of 10 to 15 percent less disease was observed compared to the original parent population. In the heirloom tomato cross of Brandywine x Rose de Berne, about half of the farmers selected primarily for flavor, and the other half for attractive appearance. Those who selected for flavor observed a measurable improvement in about 35 percent of the lines. The less desirable tasting lines were discarded. In the fourth year of the project, we bulked the superior tasting lines with the most attractive lines, and continued the selection process. In the cucumber cross of three modern cultivars with superior flavor and superior disease resistances, a great segregation of traits was measured in the first three years. In the fourth year, our consistent selection for flavor and yield measured an improvement of flavorful cucumbers with about 15 percent higher yield than the original parent lines. 50% of the farmers in our final survey reported learning practices and techniques that increased disease resistance in their crops.  40% reported improved flavor and appearance in their crops.

Distribution of Seed
Seed was distributed at all ROS conferences, field days and through the mail-order branch of ROS called the ‘Dancing Seed Company’. Each year at least 7 growers have produced sufficient seed to share with ROS and with the growers at our conference. Many more saved seed for themselves. We sent out a survey to assess the success rate, and to determine what are the key obstacles. See Section 9 – Farmer Adoption. For the 2004 growing season we distributed 67 packets of breeding project seeds to 34 growers. For 2005 we distributed 156 packets to 45 participating farmers. For 2006, we distributed 247 packets to at least 60 farmers at events and through the mail.

Restoring Our Seed generated five levels of participation:

The majority of participants are saving seed for their own use. They benefit by having greater control over their seed supply, saving money, and maintaining varieties that might otherwise be unavailable.

A second group produces seed as a value-added crop for sale. One of our performance targets was to increase the number of farmers engaged in that enterprise and to increase the quantity and quality of the seed being offered. Due to economics of scale and more favorable weather conditions for seed production, western seed producers have a competitive edge over New England’s growers. Seed production possibilities are limited in the northeast. Although only 5% of ROS participants are currently marketing seed to companies, we anticipate that proportion to increase gradually as they gain further skills, and 5 ROS participants have recently entered the market.  However, one company in our region reports that only 3 of its 40 growers are from the Northeast SARE region and another is shifting more of its most important production to western growers as a result of a 17% failure rate in our region.  Three regional companies keep at least half of their own production in house.

The third group is farmers selecting and breeding on their own farms to create varieties uniquely adapted to their conditions, climates and markets. When we wrote our proposal, we did not even consider this option, but we have documented that it offers the greatest potential benefit for farmers. Brett Grohsgal in MD, Jeremy Barker-Plotkin in MA, and Bryan O’Hara in CT, are lead ROS farmer-breeder trainers who teach other farmers from their own practical on-farm crop improvement models. ‘You cannot afford not to select and save your own seed.’ states Grohsgal.

The fourth group consists of extension agents and researchers involved in scientific collaboration on a regional, national and international scale. In New England, Dr. Mark Hutton, Maine Cooperative Extension, a ROS trainer, presented workshops at each ROS winter conference and conducted on-farm trainings in Maine each summer. Dr. Kim Stoner and Ruth Hazzard have actively cooperated and developed their own related projects as a result of working with ROS. ROS co-coordinator Eli Rogosa has become a member of the EU-funded SUSVAR program for organic plant breeding, presented a lecture at their summer 2006 conferences and strengthened international scientific cooperation.

The fifth group consists of seed companies. Representatives from seven seed companies attended or presented at our conferences and field days. They discussed the market for organic seed, networked with and recruited potential seed growers, and came away with new information about how to strengthen their own in-house training programs for seed growers. At least two companies substantially improved their grower training and farm inspection protocols as a result of attending ROS events. That alone made the grant worthwhile as all farmers in the region depend on high seed quality.

Economic Analysis

ROS has identified six potential economic benefits to growing organic seed crops:

1) To save money on the farm seed budget.
2) To acquire desired varieties regardless of their commercial availability.
3) To meet or exceed standards of the new organic rule.
4) To adapt and improve varieties to unique on-farm conditions and climate.
5) To diversify farm income by marketing seed crops to seed companies or directly through farmers markets or starting own mail-order business.
6) To reap two harvests from one crop, the vegetable plus the seed harvest.

Most of the growers we documented were motivated primarily by the first two reasons. Our research indicates that 4) and 5) are not economically profitable unless the farmer makes a significant commitment to on-farm seed production. Only 18% of the farmers surveyed reported increased profits or hourly wages from selling seeds as a result of our program. However, 36% reported improved production efficiency and 36% improvements in the quality of their crops. These data suggest that growers who are willing to acquire the skills, to place a high priority on seed growing in their operations and to devote substantial time and space to seed crops, can reap economic benefits. See Section 9 for barriers to farmer adoption.

Farmer Adoption

ROS lead-farmers have demonstrated that on-farm seed production and crop improvement enhances the quality of their vegetables, and gives a competitive edge. There is a growing interest in developing cold-hardy varieties that extend the selling season. Allowing plants to mature into flowering and seed stage increases the diversity of pollinators and insect predators, directly enhancing on-farm sustainability. As more NE farmers grow seed, the biodiversity of our organic farms will be enhanced. The participatory model that ROS introduced strengthens a decentralized seed production that increases local seed and food security. Building an ongoing regional seed-saving and seed-growing community through the exchange of stories, knowledge, seed and breeding lines will have a lasting effect in empowering farmers towards regaining control over one of their key inputs. In the long term, continued cooperation will result in the implementation of one of Raoul Robinson’s most cherished goals: local breeding networks to develop and disseminate improved germplasm.

Transfer of the knowledge and skill of on-farm crop improvement is a long-term process. Farmers learn best from other farmers. ROS surveys indicate that the biggest barrier to farmer adoption is lack of time (40-50% of respondents), and the next biggest the availability of land and adequate space for isolations to preserve crop purity (30-40% of respondents).

The poor weather the past two growing seasons in our region exacerbated the time pressure, forcing farmers to focus their efforts on survival of their most important economic crops and lower the priority of experimental work. Production space is valuable so it is hard for farmers to commit land for speculative ventures that may prove unable to compete in economic viability with their primary production. For these reasons, it is of utmost importance that would-be seed growers get the best possible training from seed companies, extension agents, university faculty and programs such as ROS, so that they can increase in the confidence, skills and willingness to try new ventures such as seeds.

Assessment of Project Approach and Areas of Further Study:

Areas needing additional study

a. Optimizing seed-crops into the whole farm plan requires more concrete research data. Farmers need to have well researched cropping systems (with isolation by time, physical barriers, and spacing) to profitably integrate seed cropping into the farm operation in a way that does not interfere with production tasks.

b. Pollinator/predator insect/vegetable flower dynamics: We need more hard research data on what are the pollinators of each specific vegetable and the predators of that plant’s insect pests. We need to know what pollinator-insects come onto the farm as a result of growing what seed crops. We need entomologists studying the insects and advising us on how to optimize the specific benefits of integrated pest management.

c. The dynamics of over-wintering brassica greens and resulting flea beetle resistance in spring needs extensive research to optimize profit potential for NE growers. Spring flea beetle damage is currently a critical constraint for growers.

d. Breeding bread wheat varieties for yield and quality in New England is an unmet research need. Organic farmers in New England lack locally adapted cold-hardy wheat cultivars.  Nutrition and flavor are not key traits in modern wheat cultivars but are a primary concern for local artisan bakers and discerning consumers. Future publicly funded research is needed to use the ROS ‘gene-pool breeding’ approach to generate genetically diverse wheat populations that are anticipated to respond rapidly to on-farm screening for tolerance to New England climate pressures combined with selection by farmers for nutrition and flavor.

e. Long-term funding is needed to address the critical unmet need for training in participatory plant breeding. PPB puts the farmer at the center of the breeding process. PPB is a highly effective proven strategy to generate locally-adapted crops that is urgently needed to feed a growing population in a period of climate change with fluctuating weather extremes and altered seasonal patterns.

f. Research on the impact of climate change on crop genetic management in the northeast is needed. There is need for disease resistance breeding for solanaceous crops as the weather gets more humid in the NE. Public funding is needed to trial varieties from the Mid-Atlantic states and further South.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.