By Gail Doesken

6 minute read

NOTE: The plant breeding processes addressed in this post are known as “traditional plant breeding” and are techniques generally recognized by the United States Department of Agriculture (USDA) to be non-GMO. While emerging technologies continue to push the lines of what is and is not genetically modified, we will leave that topic of discussion for another day. 

Humans began practicing agriculture and saving seeds over ten thousand years ago. In the process of eating and seed saving, humans chose which ones tasted best, looked nicest, kept well, or produced best by surviving pests, diseases, or other stresses. In making these observations and choosing accordingly, they began selecting some traits over others. They were plant breeders, and they were farmers.

While plant breeding has become increasingly more precise and efficient over the years, it continues to be a science driven creative process based upon observation. 

Goals of plant breeding

Plant breeding persists today with the goals of providing food, fiber, and fuel to feed, clothe, and power the world. The National Association of Plant Breeders (NAPB) aptly uses the tagline “improving plants to improve lives”.  Or, as some would say, the goal of plant breeding is “to feed the world.”

Breeding efforts have contributed to large increases in production over the years, resulting in up to 50% yield increases in some crops1. Improving food crops has typically meant working to improve overall quality, yield, storage capacity, and tolerance to abiotic and biotic stresses. Over the last few decades, specialty crop markets – including vegetable crops – have seen plant breeding trend toward improving nutritional quality and flavor characteristics, and selecting for more regionally adapted varieties.

But breeding plants to be high yielding, disease resistant, or highly nutritious is not the end of the cycle. The effort, money, and time spent developing a new variety is wasted when it is not grown and used. If the goal of plant breeding is to improve lives and feed the world, new varieties must be adopted and grown by farmers in sufficient numbers to make a difference. A variety will improve lives and feed the world only when utilized in real world food production.  

Studies show that it typically takes between five to six years from official new variety release to significant grower adoption 2. In addition to the time it takes for adoption, the process of developing a new variety often takes between eight to ten years. In the decade of time between when a plant related “problem” is identified, and a breeding “solution” is found, market demands, agronomic conditions, policies, and farmers priorities are likely to change leaving some breeding projects with years of work and no futher value. It is simply very difficult to predict if new varieties will be adopted with a 13-16 year breeding process (including years to adoption). The potential for inefficiency and loss is enormous.

Technology, use of greenhouses, and breeding programs spanning hemispheres to optimize year around growing seasons are just a few of the ways plant breeders increase efficiency and reduce the time it takes to complete a full breeding cycle. While these efforts help, they are not enough when we look at the mounting pressures of the present and future: climate change, increasing populations, and rapidly changing consumer preferences. Constant change is the new normal. As plant breeders, farmers, gardeners, and eaters, we need to develop resilience – an ability to quickly adapt to change.

Greenhouse production used in the early stages of the plant breeding cycle.
Greenhouse production used in the early stages of the plant breeding cycle.

So, how do we make plant breeding more efficient in response to the changing conditions and needs of the future? Well, one way is to ensure that the work we do is actually being used. This means a greater understanding of what drives growers to adopt new varieties, and a commitment to appreciable adoption as the true end of the breeding cycle.  

Factors driving new variety adoption

A farmers perception of a variety is one of the leading factors associated with its adoption3. Involving farmers in the process of selection results in released varieties with the greatest likelihood of appreciable adoption. Breeding programs looking to have the greatest efficiency in terms of successful adoption of released varieties need to look at how to effectively involve farmers (and even backyard gardeners!) in the breeding process. Internal surveys from the Seed to Kitchen Collaborative breeding network have found that farmers who tested varieties through their collaborative trials were 68% more likely to adopt the varieties as a regular part of their production.

Numerous studies (see further reading below) address the benefits and challenges of participatory, or collective, plant breeding. The great benefits of high adoption rates and rapid feedback from farmers in collaborative trialing has traditionally been stymied by the difficulty of managing these decentralized breeding and trialing initiatives. Until recently, the difficulty of distributing seed, collecting and sharing results, and managing the communications necessary to have meaningful farmer participation rates has prohibited widespread use of participatory breeding among many plant breeding organizations, particularly in specialty crops such as vegetables.

Collaborative breeding is the future of plant breeding

So, what is the future of plant breeding? A positive future looks like more efficient breeding programs that address the increasing variability facing our world and food production systems. A positive future looks like breeding programs that respond quickly to change. A positive future looks like resilient seed and agricultural systems.  

So, how do we start building this future? We’re working on it, and we hope you’ll join us!

SeedLinked was born with the goal of creating more efficient, transparent, and resilient seed systems. Leveraging widespread smart phone use and availability, and coupling that with big-data analytics, SeedLinked began building a tool in 2017 to remove the management and logistical barriers common to participatory breeding programs. Three years later, it is now possible to efficiently set up and run collaborative trials with thousands of participants across widely distributed geographical areas for minimum cost and effort.

SeedLinked’s current trialing network with over 1,400 participants.

Using SeedLinked, farmers and gardeners can easily participate in the selection process and keep up with emerging varieties of interest, all with minimal extra work. And plant breeders can leverage an important resource to build more efficient breeding programs able to respond quickly to change. Unlike centralized breeding programs where data comes from just a few locations each year, collaborative trialing allows hundreds of data points to be generated at 10-12 times less than traditional station trialing costs. This means that with the same resources plant breeders can do more, see more regional variation (and select in response to regionally-specific problems), and ensure that the varieties they are developing will be used.   

SeedLinked began beta testing to ensure our tricot statistical methods and analytical models used by the platform would be able to generate quality data on par with traditional replicated variety trialing. In the past two years of beta testing with over a thousand participants, we’ve demonstrated that the data quality coming from these trials is as predictive of on-farm performance as data generated through traditional trialing methods (replicated trials across a few individual locations). We’re excited to share that we’re not alone in these findings.

As with all crowdsourced data, the more data points, or participants, the better the predictive ability. We hope you’ll consider joining the collaborative trailing network of the future!

If you’d like to dive further into this topic, we’ve included links to some further reading along with references below. 

Let us know what you think. Email us at 

References & further reading

  1. Fehr, W.R. (ed.) 1984. Genetic contributions to yield gains of five major crop plants. CSSA Spec. Publ. 7. ASA and CSSA, Madison, WI. 
  1. Witcombe, J.R., A.J. Packwood, A.G.B. Raj, and D.S. Virk. 1998. The extent and rate of adoption of modern cultivars in India. In: J.R. Witcombe, D.S. Virk, and J. Farrington, editors, Seeds of choice: Making the most of new varieties for small farmers. Oxford and IBH Publishing Co., New Delhi, and Intermediate Technology Group, London. p. 53–68. 
  1. Ceccarelli, S. (2015), Efficiency of Plant Breeding. Crop Science, 55: 87-97. doi:10.2135/cropsci2014.02.0158 
  1. Etten, J. V., Sousa, K. D., Aguilar, A., Barrios, M., Coto, A., Dell’Acqua, M., . . . Steinke, J. (2019). Crop variety management for climate adaptation supported by citizen science. Proceedings of the National Academy of Sciences, 116(10), 4194-4199. doi:10.1073/pnas.1813720116