Landraces and poorly-interbreeding plants

I'm wondering how to identify progress in developing a landrace among species which are self-compatible and mostly self-polinate.  Let's take beans, which are my favorite example (and favorite plant to grow).  Depending on who you ask, they have a 1-5% rate of natural interbreeding (you can encourage the higher rates by encouraging pollinators and dense inter-planting of varieties).  In this regime, if you start with a genetically diverse set of "parents", it would seem like a long time before you could say you had something like a landrace.  If you were good at identifying the natural hybrids that occurred and making sure to preferentially plant the resulting F2 seeds, maybe you could speed the process along, but it would seem to me to take a long time to create something that could credibly be called a "landrace".  When would you identify a population as being a landrace rather than a bunch of different varieties being grown an improved together?  Does it matter?

Beans are an interesting case study....

When I plant beans, about 80% of them die without producing offspring. That is an immediate, survival-of-the-fittest phenomenon that happens without any effort on my part. In succeeding years, the surviving varieties will mostly continue to survive. A few varieties will really thrive, while the rest eek out a meager existence. The varieties that thrive tend to predominate in the population.

I grow beans landrace style, which is all jumbled together. That encourages cross-pollination between genetically distinct varieties. The naturally occurring hybrids tend to grow much more vigorously than either parent, so they tend to produce many more seeds than their parents. This effect continues for 3-4 years. The new varieties that are created by natural crossing between the locally-suitable parents, are even better at producing seeds in the local ecosystem. So while the natural crossing rate is low, it has a higher impact than the 1% to 5% crossing rate might indicate.  

One thing that happened due to the heirloom preservation meme in recent decades, is that people inadvertently selected against a plants ability to cross-pollinate. Growing landrace style, or intentionally saving seeds from observed natural hybrids, selects for a population of plants with higher cross pollination rates.

In the Beautifully Promiscuous and Tasty Tomato project, we are selecting for flowers that are open, instead of the closed up flowers of domestic tomatoes. This one change might move the cross pollination rate of tomatoes from 3% to 30%.

My definition of landrace is a crop that is locally-adapted, genetically-diverse, promiscuously-pollinated, and beloved by the local community.

I use the term "grex" to describe a bunch of varieties growing together that don't fit into the above definition for one reason or another.

Chad Meyer wrote:I'm wondering how to identify progress in developing a landrace among species which are self-compatible and mostly self-polinate........... When would you identify a population as being a landrace rather than a bunch of different varieties being grown an improved together?  Does it matter?

Good questions for sure.  Just dropping in this description from https://ucanr.edu/blogs/blogcore/postdetail.cfm?postnum=43295  which I think is in-step with Joseph's concepts:

"What is a Landrace Cultivar?

To put a landrace in context, let's quickly look at the alternatives. Heirloom seeds have been selected over time to reliably produce fruit with specific characteristics like red, green, or black skin on a tomato, and these characteristics have been stabilized for a period of time. They are open-pollinated, so the seeds will produce plants that look more or less just like the two-parent lines.

Hybrid seeds result from crosses between different varieties for specific purposes. Tomatoes hybrids are bred for disease resistance. Corn is hybridized in particular for sweetness and size. These crosses produce uniform plants and seeds in the first generation, but the second generation of seeds breaks down unpredictably into those two-parent lines again.

A landrace comes from a process that moves in the opposite direction in terms of selection. Researchers have defined a landrace in a number of specific ways, but at the center of each of them is the idea of high genetic diversity and a connection to the specific soils and conditions where it was developed over time.

A landrace population hasn't been selected for stable, uniform characteristics like an heirloom. Instead, a landrace has promiscuously incorporated genetic material from plants with different characteristics, evolving over time to thrive in specific growing conditions. In the garden, that means you don't know if your landrace tomatoes will be red but you do know that they will probably grow well.

Genetic diversity provides some real benefits when the alternatives are hyper-selected varieties like hybrids that might have lost their flavor or hyper-stable heirlooms that might be vulnerable to every bug and disease that shows up. A landrace is an opportunity to adapt a cultivar to your own garden and conditions, choosing the best tasting and best-growing plants to literally make them your own.

Geneticist Jaime Prohens writes, “Given that adaptability to particular environments, general resilience, cultural value, tolerance to local stresses, sensorial value, etc. are traits often controlled by multiple genes, the most reasonable material for the initial development of new generations of local cultivars consists of historical landraces, found in seed banks or under local cultivation.” ..." (end quoted material)

Just to add some additional commentary on an issue this is quite fascinating to me.  Recent research indicates that even clonal populations, when followed for several generations, increase the genetic diversity between the constituent individuals.  Although it's early days in the investigations to be sure, the data point to the fact that plants will accumulate mutations to varying degrees as they are growing.  If any of the mutations produces greater growth or fitness in the population, that individual might catch of the eye of the gardener and be saved for seed produciton.....thus, selection has been made on the population due to a genetic change that results in traits more favorable to the gardener than those in the founding seeds that had been started with years earlier.  It takes longer to get to the same genetic diversity this way as compared to starting out with high diversity in the population like Joseph has been doing, but it's not nothing and can contribute to crop improvement even in species that are self-pollinating.

This, I think, clears up a misconception I had about landraces. In my mind, a landrace might have been what you would call a “pre-cultivar” where rather than all plants being essentially uniform they exhibit a narrow range of similar characteristics. The plants that I’ve seen called “landraces” (except for the “Lofthouse Landrace” varieties) seemed to fit that category.  Either of the above definitions helps me to wrap my head around how plants producing beans and bean pods of all different colors can be (and become) one unified “thing”.

My beans are unified in being able to grow in my ecosystem.

They all thrive in a high altitude desert garden, in an open field, with full sunlight all day long, all summer long. They all tolerate the silty alkaline soil. They reliably survive all of the local bugs, diseases, and animals. They thrive without fertilizer or weeding. They all produce seeds in a short growing season. They all survive the local climate, including an occasional light frost. They survive on once a week irrigation.

Additionally, the farmer has put heavy selection pressure on them to grow as bush beans, or semi-bush beans.

They all have pods that are non-shattering in the field, yet are easily threshed.

The multi-colored seed coats, and assorted shapes are two traits that are easily observed where differences are allowed. If I wanted, I could create a landrace of beans in which only white seed coats are allowed.

Having thought about this over the past week, and in light of other comments made this week, I think it's safe to say that I'm overthinking things (that's what I always do)!

If I am growing out 10 different (named or otherwise) varieties of common bean that would "breed true" as they say, and if I'm growing them side-by-side or all interplanted, and I save seeds and try to regrow whatever produced beans while maintaining some notion of diversity (e.g. at minimum planting all the colors/patterns of beans I've saved), then I've already got more diversity in my garden than in most seed packets -- and the corresponding resilience that brings -- independent of how much the bees manage to produce hybrids and independent of whether I end up preferentially selecting those hybrids or not.

So, whether what I'm growing ends up being a "landrace" or "grex" or "hybrid swarm" or just a bunch of varieties grown together, I'm still reaping the benefits of diversity in my garden and likely to have that pay dividends as time goes on.