Darwinian Agriculture and a potential problem to be avoided

I’m partway through the interesting book Darwinian Agriculture by R. Ford Denison. In it, he makes the point that individual species have been more or less optimized by millions of years of selection, and so it is unlikely that biotechnology can significantly improve them. On the other hand, he points out that assemblages of plants have not been similarly optimized, at least not optimized for yielding food for us.

This leads to an interesting point that may be a problem for some landrace plant breeding. Denison says that in a crowded field, a genotype that allocates more resources to growing taller, and less resources to producing seeds, will eventually dominate over shorter but more productive genotypes in the same crop. The taller plants will each shade several of the shorter plants, and over time the population will shift. Which is to say that if we bulk select landrace grain crops, we may end up selecting for more stems and less grain. This would be even more likely if we are impressed by the taller plants and select for them, intentionally or unintentionally. Denison points out that this inter-crop “arms race” is counterproductive from our point of view.

Now, there might be many reasons why we would want a taller, less productive grain; maybe we want to use the straw for compost or thatching, or we want it to smother out the weeds, or we want greater ease of hand harvest, or we want taller corn stalks to keep the ears away from raccoons. In all those cases, we might be willing to put up with lower grain production. But if we don’t want this to happen, we should work to counteract it. And as a general point, we might want to think of ways to select against interspecies competition in our crops. What are some strategies that would work for this purpose? I’m thinking that head-to-row planting would be one way of counteracting it; we would be able to select families of plants that yielded the most, and competition between families would be reduced.

The reverse of this is also true. When selecting squash or greens, we should try to avoid saving large amounts of seeds from plants that prioritized producing seeds over producing fruit or leaves. It is only in the crops for which the seed is the desired harvest that the mechanism described by Denison would be problematic. A taller, less-seedy kale might be perfectly fine.

4 Likes

I can see how that might be a problem on a farm or in a very large garden, but I garden on a backyard scale. It’s easy for me to observe pretty much every plant, every day, and restrict or completely cull anything I don’t like the looks of.

Maybe, maybe not. The evolutionary imperative of all annual plants is to make seeds and the more they make the better the chances the next generation will succeed. It’s not likely any a robust plant will make a significantly lower amount, still even if it does. I would replace the “is” in that statement with “may be”.

How they allocate resources may be the operative phrase there.

Imagine a discussion between two plants. Stretch says, I’m going to grow big and tall with big deep roots and fat juicy stems and make me a hundred fat seeds. Shorty laughs and says, I’m going to bloom my rear end off and make five hundred seeds. My family has been bred for generations to just make lots of seeds, it’s in my genes and I don’t care if it stops raining for a month and they all dry up, or if it rains every day for a month and they all rot, I gave it a shot.

1 Like

The seediness issue came up in another thread recently too. The reason it came up as a record keeping question was that I thought my solution was going to require tracking harvests plant by plant, but folks suggested a much better way.

Seediness could be a problem depending on how you select seeds, pushing plants to play the numbers game to an excessive extent (though we still want them to be at least somewhat prolific) at the expense of germination rates/fruit quality/etc. A thoughtful breeder can counteract that tendency, though.

1 Like

I’m halfway through this book too! You and I are on the same reading schedule. The tall plant/short plant tradeoff was also discussed in Robinson’s Return to Resistance as one of the few successes of the Green Revolution. Definitely true that natural selection does not always favor the interest of the species. Or our interest, of course.

@MarkReed I think this more height/less seediness problem would only be an issue if everything else was equal. In other words, between two plants that are equally good at harvesting resources, resisting pests, etc., we should select for the one that produces more edible seeds and less inedible structure—but unchecked natural selection will do the opposite.

Unless, of course, we have a particular use for the extra height. If the tall plant produces 10% less seeds, but its extra height means that all of its seeds end up as human food rather than feeding the local raccoons, that would be a good trade-off.

As far as the evolutionary imperative to produce seeds, that’s true—but Denison’s point is that the taller, less fruitful plant will have an advantage in consistently producing seeds—until all the plants are the same height, at which point the lower-growing and more inherently productive plants have been selected out. So in this case, the imperative actually works against our (and the plants’) best interests. (This is assuming that the main competitors are plants of the same species.)

@ChlorophyllDragon , Thanks for linking to that other thread!

2 Likes

@MashaZ That’s funny that you are also reading Darwinian Agriculture!

It has been a very informative read so far. I think he’s a little too harsh on the argoecologists sometimes, but much of his criticism seems justified from my experience. Too often, garden and farming writers will wax lyrical about how much biomass is produced by a redwood forest, while ignoring the fact that all that biomass is inedible! Or they will argue that “nature doesn’t till”, which is sort of true—except that our crop species are mostly cultivated weeds that come from “tilled” riverbanks and burns. Or worse, they will say “nature doesn’t plant in straight rows”—but “nature” isn’t trying to keep track of things and has unlimited time!

Where I disagree with Denison, I disagree with him for the same reason I occasionally disagreed with Robinson; the assumption that we are just stuck with mechanized commodity farming to feed distant cities. That leads Denison to assume that closing nutrient loops is not possible, or that the small advantage of intercropping is not worth trying to retool all our machinery, etc.

His criticism of biotechnology is fascinating, and similar (though more detailed) to that proposed by Carole Deppe. We just can’t beat the power of natural selection operating on huge scales of space and time.

I’m reading Denison’s book because it was recommended by Chris Smaje, author of A Small Farm Future. Smaje integrates Denison’s positions into a more holistic whole; his position is that our future is one dominated by small, labor-intensive, low-input farms, whether we like it or not. It is well worth a read.

Come to think of it, somebody should introduce him to the landrace concept, if he hasn’t already found out about it. It would fit very well into his worldview.

I’m finding the criticism of agroecology very disturbing, just because I’d read so much of it and been so influenced by it … hate to think I wasted all that time. I’m still not totally convinced - but his emphasis on the limited size of the Earth and the growing demands of the human population is sobering. I had less trouble with his critique of biotech because I never liked that anyway.
I’m also a big fan of Chris Smaje - I used to follow him on Twitter and now I miss him. Should just sign up for his blog. Yes, I agree, he’s a potential recruit to landracing!

@MashaZ I’m not 100% sold on it either, but I think there is a lot to this criticism. Smaje makes many of the same criticisms without going down the “agribusiness” route, so I was primed to accept it.

And my experience as a “permaculture zealot” also primed me to accept it. I had a lot of practical problems with some permaculture techniques, and then found that there were conceptual problems underlying the practical problems.

I still think the basic permaculture concept is valid. The problem, as far as I can tell, is that while they draw inspiration from traditional farming practices in many parts of the world, they tend not to draw as much from traditional farming practices in Europe and other temperate climates. Using more tropical-adapted practices in temperate climates is a recipe for failure. The farther north one is, the more important annuals will become, I think.

And, permaculturists seem to have bought into the idea that disturbance is a bad thing. So when they are mimicking nature, they are actually mimicking a very idealized concept of nature. We’re animals (biologically speaking) and animals tend to create disturbance. That’s our natural role, just like it is for beavers and bison. Annual plants like disturbances, and so do mid-succession species like most of our fruit and nut trees, bramble fruits, etc. Late succession habitats are not actually very productive from the standpoint of many animals; closed conifer forests are a particularly good example of this.

I don’t totally agree with Denison about natural ecosystems being unselected. I think that an ecosystem with unfilled niches or inefficiency will tend to spur the evolution or in-migration of new species to fill that need; that’s why the great evolutionary radiations happened right after the mass extinctions, when there were lots of niches to be filled. Still, he’s right about them not being as selected as individual species.

And most of all, what we’re building will never be an unmodified ecosystem, any more then an ecosystem with beavers will be an unmodified ecosystem. We are building an “ecosystem plus humans” of which by definition the only examples will be ones created by humans. We can and should look at how other animals disturb their habitats to figure out how to properly disturb ours.

That’s where I’m in favor of much smaller farms, a patchwork landscape; natural disturbances are patchy. And even within farms, this is important. My garden now has annually dug beds with vegetables and some tolerated annual weeds, surrounded and interspersed by borders and patches of various perennials, stone walls, brick paths, and other varied habitats. And my yard is bursting with helpful and fascinating invertebrate life. But I’m now better able to keep aggressive perennial weeds from taking over my vegetable beds.

3 Likes

All really interesting thoughts - thanks. There’s also the fact that perennials will grow in places unsuitable for annual cropping. (Did you ever read J. Russell Smith’s “Tree Crops”?) But I’ve just started Denison’s Chapter 8, on what will work, so I will suspend judgment now until I’ve finished the book!

1 Like

Yes, I have read Tree Crops; it is a great read. And that’s a good point; the fact that perennials will always yield less than annuals doesn’t matter if the annuals can’t grow in a given location! Steep slopes and rocky ground should definitely be in perennial cover, as should fragile soils in the tropics and semi-arid areas.

Also, perennials have other benefits. We can’t turn urban parks into cornfields, but we could turn them into nut orchards! Similarly, if we need a woodlot anyway, it might as well be a nut-bearing woodlot.

I think that sort of thing would be the very essence of the kind of “bet-hedging” that Denison advises.

To expand a bit more on my comments above about annuals; here in Denver, I’ve found that woody perennials suffer greatly from late frosts and unseasonable mid-winter warm spells, not to mention the occasional early snow that breaks down trees still in leaf. There are hardly any nuts that will bear here at all, and fruit trees generally fail to bear in at least two or three years out of ten. Meanwhile, my squash and bean seeds are happily spending the winter and spring inside.

As the climate becomes more chaotic, plants that invest heavily in perennial above-ground “infrastructure” may suffer more than the adaptable annuals. I think that the perennial crops best suited to a climate like mine (other than grass pastures that Dension favors) are perennial tuber-bearing plants. In practice, such plants are generally treated almost like annuals, but they have the advantage of a large store of energy to quickly get growing in the spring. Perennial tubers such as the potato or sweet potato provide more calories to the acre than any grain crop. They do, of course, take more work, but in a “small farm future” there will probably be plenty of labor available.

2 Likes

I’ve finished Darwinian Agriculture; the last few chapters were particularly interesting. And they contained a lot of stuff that could be relevant for landrace plant breeding. Among other things:

Breeding plants that will guide the evolution of their own beneficial symbionts, rather than trying to culture and apply improved symbionts directly. (This makes more sense because the symbiotic organisms evolve faster than the plants.) The book included explanations of how plants actively shape their microbiome.

The importance of “bet-hedging” by working with a wider range of crops, including little-known and wild crops.

The description of some really elegant strategies that plants use to resist pests. For instance, some wild potatoes release a chemical that mimics an aphid alarm signal. Aphids release this signal when they are under attack, warning other aphids to vacate the area, so aphids that land on these potatoes quickly fly away. Many of our attempts to breed lasting immunity to pests and disease into our crops have quickly broken down due to ongoing evolution of the pests; but this strategy won’t break down, because any aphid that stops paying attention to the alarm signal will more likely be eaten by predators.

Speaking of which, the book also includes strategies for keeping immunities from breaking down and for slowing the evolution of crop pests.

The importance of selecting groups of plants, rather than just individuals. (This would work against the kinds of “arms races” discussed in the first post of this thread.)

A discussion of phenotypic plasticity in plants, by which they can adapt to cyclically changing conditions.

Some thoughts on how agricultural incentives can be changed to support better outcomes—specifically by tying incentives to outcomes rather than to practices; this would allow for evolution in best practices to meet the desired outcomes.

And last but not least, a reference to Carol Deppe and the potential importance of amateur plant breeding and on-farm innovation.

One ironic point was a brief discussion on how crops aren’t actively evolving in the farmers’ fields—they are in our landrace populations!

I certainly didn’t agree with Denison on everything. I thought he got a bit reductionistic at points, in particular his emphasis on selfish genes. While reductionistic ways of thought can be useful, it is important to integrate any insights into a coherent whole. Not that Denison doesn’t realize this; he stressed the point, for instance, that plant genetics can only be fully understood in their ecosystem contexts. A genetic sequence abstracted from that greater whole might not be intelligible.

Overall, I think this is a very valuable book for anyone thinking about crop breeding and agricultural improvement.

3 Likes

Thank you for this very detailed summary and analysis! I’ve been enjoying the conversation. I went back to my notes because I remembered my biggest takeaway from the book (I don’t think you mention it here?) the formula for adaptation:
“The rate of evolution depends on
how much genetic variation there is within a population and
The intensity of selection. More intense selection leads to faster evolution".

From this I simplified: The speed of adaptation =
genetic diversity of starting population x the intensity of the selection x size of population.

However the intensity of the selection pressure is not a direct line, more like a U curve because at extremely intense selection pressure the genetic diversity declines.

And bam! This is why sharing seeds (with selection criteria like flavor and earliness) and having big pools like our seed project could have a very fast adaptation rate: Each gardener can be extremely selective without narrowing the genetic base. Every participant sends in seeds from their top 5%, and no loss to genetic diversity. The get back everyone else’s top 5%. I wonder what Denison would say about projects like this. Can you ask him?? :slight_smile:

Anyway, I digress. I’m copy and pasting some of my notes below, but I do sort of think I was overthinking things too much a year ago, and that not everything here is actually true, so take it with a grain of salt for anyone who reads this.

“Less beneficial indigenous strains often end up outcompeting the applied strains”. Refering to nitrogen fixing bacteria:
Kiers found that older soybean varieties did just as well when faced with mixed good and bad rhizobia, the difference was enough to eliminate all the yield advantage of the modern varieties.
They are unsure whether this was due to sanctions.
Darwinian agriculture says that plants impose sanctions on rhizobia strains that don’t produce as much nitrogen, thereby limiting the proliferation of “cheaters”. BUT could it be that the plant is being selective in its forming of relationships, only allowing the ones in that it deems beneficial. And how would nitrogen fertilizer and herbicides, selecting for yield etc affect how that plant decides? And in that case, he says the older varieties outperformed modern varieties when there was a mix of rhizobia, he is saying that the modern varieties had the relationships, but weren’t able to be benefitted. So… modern plants not able to form relationships/benefit, and our takeaway is that the bacteria aren’t as good. In fact, our plants aren’t as good.

Bet hedging
A plant that is conservative around water use will be less able to grow fast when conditions are good. Plant that uses a lot of fertilizer will grow extra good with fertilizer, extra poor without it. A conservative plant will outperform in lean times.

Consider the trade-offs that may be associated with

  • Selecting for flavor and vigor (reduced yield)

  • Selecting for yield and uniformity (Does poorly under low resource conditions, Reduction in flavor and nutrition)

  • Bet hedging strategies sacrifice some expected performance under average conditions in order to reduce possible outcomes under unusual conditions to an acceptable range. We sacrifice the best imaginable outcome to avoid the risk of the worst imaginable one.

  • Breeding for individual fitness often results in group decrease tradeoff: Example, chickens selected for higher egg laying reduce group laying because the whole group become more aggressive over time and peck each other.

  • Wheat- competitiveness against weeds causes reduction in yield-- higher stalk plants less able to carry heavier weight. Breeding for increased yield in wheat includes tradeoff of lower protein concentration.

  • Toxin production is costly to plants. So increased pest resistance potential likely to include lower yield. (only compared to herbicide sprayed comparison).

  • Most plants only make toxins once they are under attack. So having them present all the time has been evolutionarily rejected

We need to identify trade-offs and decide if they are worth the benefits

  • Tradoffs link to performance under past co2 levels may be restraining photosynthesis today.

  • For example the human gene that protects people from malaria, but also reduces blood ability to carry oxygen. A detriment when malaria is not present.

4 Likes

Your bit on chickens selected for higher laying rate isn’t true.
“Breeding for individual fitness often results in group decrease tradeoff: Example, chickens selected for higher egg laying reduce group laying because the whole group become more aggressive over time and peck each other.”

Aggressive chickens isn’t related to their lay rate. If they aren’t being fed enough they will stop laying. If you have aggressive birds it’s more likely not related to feed.

I don’t know about chickens, just an interesting quote I found, but it does remind of what Temple Grandin says in her animal bahaviour book-- each gene corresponds to another gene so you might get unintended consequences for selecting for a particular trait. It’s been a long time, but I remember something about how the high laying genes are related to the aggressive genes, and that’s why roosters have gone from doing mating dances to attract hens to something more like chicken rape. And how the heritage breeds still do the mating dance, and modern ones don’t. I have zero personal opinions or experience on that matter though! That’s in this book if you haven’t read/listened. Soo good.

https://www.audible.com/pd/Animals-in-Translation-Audiobook/B00C7YEGW0?eac_link=enp27qbTGTDw&ref=web_search_eac_asin_1&eac_selected_type=asin&eac_selected=B00C7YEGW0&qid=QEUP0R7dR4&eac_id=143-2835664-9181010_QEUP0R7dR4&sr=1-1

@julia.dakin Yes, I think that formula for adaption is very relevant to our project! I could see what Denison would think about the project; at the very least, he would probably see it as a useful kind of “bet-hedging”. I could ask him!

The point about trade-offs is particularly important. Just about everything in life involves trade-offs, and so if we are not careful we may find ourselves accepting trade-offs that are detrimental.

I thought the part about microbial symbionts was perhaps the most interesting part of the book. Who knew that plants could shape the microbial community to that extent? Did you mention Denison in the Going to Seed course on microbial symbiosis?

@KadenceLunemann If I properly understood the part of the book about chickens, he wasn’t claiming that as a general rule. Rather, the experimenter in question found that selecting groups of chickens that laid the highest average number of eggs worked better than selecting the individual chickens that laid the highest absolute number of eggs. By selecting groups, they brought up the performance of the whole group, largely due to minimizing competition. By contrast, selecting individuals didn’t have as strong of an effect.

I’m wondering how we landrace plant breeders could effectively practice group selection. How could we ensure that different groups were growing in largely similar conditions? Maybe it would be easy for people with acres of land to work with, but many of us are growing in crowded gardens where conditions are highly variable.

Maybe if we planted a few seeds in each hole and then treated each clump of plants as a group, saving from the highest-yielding clumps? (To make this work, the clumps would have to be spaced apart a bit so that they didn’t influence one another as much.)

1 Like

Finished the book today - yes, lots of fascinating insights. Not as many direct takeaways as I had hoped for, but the mindset and general principles are valuable, I think. It’s encouraging that he sees so many opportunities for improvements. I do wish he had talked more about local adaptations, though. I get the feeling that plant breeders are looking for varieties that will work everywhere, which will usually require too many tradeoffs and therefore forecloses a lot of possible (local) improvements.
So what’s next on the reading list, @MalcolmS ?

Yes, that’s a good analysis of the book. He’s very much coming at this from the point of view of standard cash-crop agriculture and how it can be made more sustainable, rather than from a more “peasant” outlook of self-provisioning.

I’m going to be reading Rivers of Change: Essays on Early Agriculture in Eastern North America once I get it from the library. I’m not sure if it will be interesting or not, but I hoping that it might throw light on the origins, strengths, and weaknesses of annual agriculture by visiting an often overlooked center of early agricultural development. I get the feeling that it will be quite scholarly.

Rivers of Change sounds like it could be useful. It’s expensive, though, and not in my local libraries - I think I’ll wait for you to report on it before deciding whether to buy it! Along the same lines, Cronon’s “Changes in the Land” is very good.

@MalcolmS Just found “Rivers of Change” on the Internet Archive lending library!