Friday, October 30, 2009

The Myth of Monocultures

Brendan sent me a link to the new PBS video special based on Pollan's The Botany of Desire. Ironically, I had been eyeballing the same book on my shelf the past few weeks wondering if it was due for a re-read.

The PBS special was very well done and I definitely recommend it.

The last 15 minutes though (an uninformed tirade against modern agriculture), were unwelcome and irritating. One telling moment was a local ag scientist lamenting the "tragic" public rejection of the environmentally-friendly Bt potato, followed by more fear-mongering by a partisan lobbyist.

Pollan should limit his lectures to topics he understands - which wears thin as he waxes on the horrors of monocultures. Plants (including crops) are in a continual arms race with pests and pathogens. The plants keep finding new ways to protect themselves while the parasites "keep trying to pick the lock." Picture a landscape filled with many different kinds of potatoes. According to the classic version of this story, a parasite that stumbles upon a new way to break into a certain potato variety may wipe out all individuals of this variety but it won't kill EVERY potato on the landscape.

The problem with this story is that plants varieties that look very different don't necessarily have different disease and pest resistance genes. Just like humans, which are known for tremendous phenotypic diversity, can be extremely susceptible to the same flu and HIV germs across all ethnicities, all the multicolored potatoes in the above picture may be similarity susceptible to diseases like late blight.

It's not so surprising that crops are notoriously vulnerable to parasites when you meet their wild relatives. The fruits and grain of the undomesticated relatives are usually barely edible - tiny, fibrous and full of poisons. It should be no surprise that the same qualities that humans love in their food (lots of easily-digested calories) are appreciated by many other creatures as well. Plant breeders have countered this inherent susceptibility by introgressing (moving) key resistance genes from wild relatives of crops into the crops themselves. This process has actually made modern industrial tomatoes more genetically diverse than either traditional Latin American or U.S./European heirloom varieties (although the traditional and heirloom varieties look more diverse).

Theoretically, it would be useful to maintain crop populations with diverse resistance genes. The industrial application of this (multilines) invovles breeding many different versions of a favorite crop variety that are identical except for their resistance genes. As appealing as this idea is, it hasn't really worked out in the real world. The alternate approach (pyramiding) seems to be more effective. Here, many different resistance genes are combined into a single crop variety. Pests and pathogens may be able to overcome a single gene at a time, but it's usually almost impossible to simultaneously overcome several.

PBS' The Botany of Desire also makes a lot of Bt resistance developing in insects exposed to genetically-engineered Bt crops, but this is nothing new. Insects and pathogens develop resistance to all kinds of pesticides - whether they're natural plant chemicals, artificial sprays, or genetically-engineered genes.

Evolution is a fact of life. The pests and pathogens will continue to adapt, and they will eat up our crops if we don't keep ahead of them. It's a problem you can never completely solve - which is good for me and my scientist colleagues!


  1. Will have to make the time to watch that sometime this weekend.

    Had no idea modern tomato breeds were more diverse than heirlooms, that's AWESOME.

    I'd guess what would happen with a diverse population of say potatoes is that some super-disease would come along and wipe out 95% of the potatoes, leaving a few that happened to contain a resistance gene. Next year people could plant those potatoes, and they'd be resistant, but most of the genetic diversity would be lost. To save it, they'd have to introgress the resistance allele into the rest of the potatoes, which is what breeders already do without having to wait for a disease to wipe out everything that can't resist it.

  2. Yeah, and this probably happened automatically when 5000 varieties of potatoes were kept and actively improved in the Andes, surrounded by diverse wild, weedy potatoes - but if you move a dozen or so varieties to North American farms (where the wild relatives don't grow), the farm will look diverse but it won't have very diverse resistance.

    Part of the story I didn't elaborate on is that moving resistance genes from wild relatives often lowers the quality of the crop by making it more weedy. It's like your post on Gros Michel vs. Cavendish bananas. Gros Michel was the favorite because it tasted better and was incredibly easy to ship and store - but when a new disease hit, the choice was second rate bananas or no bananas.

  3. Interesting. Do you know if the issue with resistance genes bringing in other weedy traits is a result of multiple genes in close linkage or that those traits directly affecting disease resistance? Either would make sense.

  4. I think you may be playing fast and loose with the definition of "diverse" here. Even if a modern farming operation is planting potatoes with a thousand resistance genes, it's still a monoculture wherein every plant is identical to its neighbors. The genes may be diverse, but the farm itself is not, unless multiple cultivated varieties with different ancestral plants are being planted on the same farm, which doesn't appear to be how it's usually done. It may be less likely that a pathogen can pick all the locks, but the result is the same if one of them does.

    The other thing I find weird about this is that if the modern cultivated crop is so tough, then whence all the pesticides and stuff? Surely a plant with the combined pest-resistance of twenty wild species should be able to handle itself without so much help, if things work the way you're implying.

    I'm not so much trying to attack as trying to understand: Pollan's argument against monocultures makes a lot more intuitive sense to me, being basically the argument that you don't put all your eggs in one basket, and if I'm reading you correctly (which I am perhaps not), you seem to be saying, oh, but this is a fucking awesome basket. Which somehow does not reassure me.

  5. Paragraph 1:
    I absolutely agree. Crop diversity is important for lots of reasons, including cultural ones. My point is that some little boutique farm that grows 15 different types of potatoes that all look, taste and grow extremely different will likely share susceptibility to bugs and disease. People like Pollan suggest that apparent and functional diversity are the same thing - but they're not.

    Paragraph 2:
    I'm arguing that all domesticated crops are inherently weak and there's almost nothing you can do about it. Wild plants defend themselves by filling their vulnerable parts (e.g. fruits) with fibers and poison. When you don't allow a plant to do this (because you want easily-digestible fruit) and then you force the plant to make the fruit even bigger and filled with carbs, the plant becomes an easy target for bugs and microbes. All plant traits are incredibly complicated and difficult to control precisely. It is incredibly difficult to use breeding to improve the resistance of plants - which is why our crops need our constant tending and weeds don't. And, despite the hype, current genetic engineering is laughably primitive and can't manipulate these complex traits well yet.

    Paragraph 3:
    I want farmers to plant lots of different varieties for lots of reasons (including that I want to be able to buy more than 1 kind of potato). The fallacy of Pollan's argument is that diversity doesn't exist on a single scale. A set of plants can be very diverse in say, the color or size of their tubers, and all be incredibly homogenous in their susceptibility to late blight.

    Our attempts to hold back pests and diseases with resistance genes is like using duct tape to fix a leaky pipe. It's not a perfect solution, but if duct tape is all you have, it's what you use - and using duct tape of colors isn't going to help.

    In a way, Pollan advocates for us to plant lots of different varieties with the hope that they won't all get wiped out. The better solution (that breeders use) is to constantly survey for pests and pathogens that have developed new ways to kill plants, find an appropriate resistance gene that counters this new ability, and get it into popular varieties before any of them get wiped out.

  6. James:

    Those are both HUGE problems whenever resistance genes are moved.

  7. I think I get what you're saying now. You're saying that species with common ancestry also probably have similar weaknesses in common, as far as disease (and particularly as regards diseases which weren't present in the ancestral environment of the genus). So planting a large number of varieties of species potatoes, for example, isn't likely to work out well because the species likely all have similar vulnerabilities.

    Cultivated varieties, on the other hand, are likely to be more resistant to a broader array of diseases, because any variant in a domesticated variety of plant that showed resistance to any given disease was probably used to breed resistance into multiple other varieties. So a cultivated variety of food plant (at least one with an ordinary genetic situation, i.e. leaving out plants like bananas and apples which are mostly cloned) is likely to have the accumulated resistance of several hundred years' of breeding behind it.

    Is that more or less it?

  8. Yep!

    Thanks for helping me to clarify my post!

  9. Good.

    Now I'm wondering something else. At a Garden Rant thread earlier today, I doubted the claim that eating GMO food has completely sterilized lab animals in experiments and was directed to read the books Genetic Roulette and Seeds of Deception by Jeffrey M. Smith as proof of same. I found Smith's website and was wondering if you knew of any reviews (friendly or not) of the book(s) by people who were competent to talk about the science involved in his claims.

  10. It is kind of interesting to discuss the genetic diversity of food crops. From the best information, it appears that many food crops evolved from one selected line of a native plant that was more edible, sweeter, etc. If this is true, which evidence does suggest, than the genetic diversity of crops has always been low. Jared Diamond explains this well in his discussion of the edible line of almond.

    Therefore if current ag science attempts to cross in new genes from a native cultivar, than in a way we are increasing our genetic diversity. We can let the outcrossing occur naturally, but if we don't know the parentage of the parents, than how do we know if we are increasing genetic diversity?

    I think we like to forget about the parentage of our crops. I dont understand how we can claim that adequate genetic diversity that can prevent new diseases can occur "naturally" in just 3500 years. Especially since it has taken many native crops several 10s of thousands of years to "naturally" form resistance to pests and diseases present in the environment.

    Great post Matt!

  11. Mr. S.

    I've given up on tracking down the truth for every crazy GM claim because they inevitably are total b.s.

    The number one thing that people don't understand about genetic engineering is that it's simply a collection of tools and techniques - just like mechanical engineering or chemical engineering. Vehicles are often dangerous either by accident (e.g. poor design leading to recalls) or intentionally (e.g. tanks). The same is true of plant breeding and genetic engineering. Solanine is a poisonous alkaloid naturally present in potatoes. If you are careless with breeding or genetic engineering, you may accidental produce a potato that contains too much solanine. In fact, traditionally bred potato varieties have had to be recalled in the past for this reason - so now we check for it. The only difference between genetic engineering and mechanical engineering is that no one is calling to ban mechanical engineering because they don't like tanks.

    Fundamentally, there's no reason to think that moving genes from one organism into another is INHERENTLY dangerous - it happens in nature all the time. In fact the most common tool for genetic engineering is a bacterium that's been doing it for millions of years on its own.

    You've seen galls on plants caused by this bacterium, Agrobacterium tumefaciens. It injects a piece of DNA into plant cells that "genetically engineers" the plant to do two things - form giant tumors and produce huge amounts of a nutrient source that Agro needs. A huge proportion of walnut trees, roses and cane berries are infected with this pathogen, so in a very real sense, humans have been consuming genetically engineered plants for millions of years.

    All genetic engineering does is write-over the genes that cause tumors and nutrient production with ones that cause pest and herbicide resistance, and let agro do its thing. It's the SPECIFIC GENES that are inserted, not the technique that is potentially dangerous. Genetic engineering is currently heavily regulated by the USDA, FDA AND EPA (unlike breeding or mechanical engineering btw) and should remain so.

  12. We were just talking about this post here:

    I thought I would share a few more thoughts.

    I think one of the things you're missing when you talk about the importance of modern breeding for things like late blight resistance in potatoes, is just who the competition is and what they are up against.

    First of all, if you want to study traditional, Burbank, plant breeding techniques in the US on a formal basis, you will almost certainly do this at the U of Wisconsin at Madison. That's because there are almost no other programs in existence in the US anymore, or any other countries for that matter. No one wants to fund them, and no one wants to hire graduates.

    Because there are no paid positions, the only alternative is public domain plant breeding. Many people who do this have degrees by the way, including a number of PhDs, but these are almost all in other fields.

    Because there is no money in public domain plant breeding, if you do this full time you have to have another source of income. This leaves nearly all of these people asking for donations from others, or running their own seed company effectively selling hours and hours of work for next to nothing.

    When they do manage to create something interesting, it gets picked up by larger seed companies and sold without any royalties going back to the original breeder.

    I know there are many others, but I'm personally aware of about 10 plant breeders who are skilled enough to have created widely accepted varieties and have name recognition. Most of these have spouses who work, and live from paycheck to paycheck. They have to struggle to find money to fund their projects, and sometimes don't even own their own land. Often personal emergencies or accidents mean they lose years or decades of work, because they don't have access to good storage facilities for their breeding materials and don't have ways to cope with illness.

    While there's been a bit of a resurgence of this kind of breeding work in recent years, most of the skills rest with people who are getting on in their years, and no real efforts are being made to replace them.

    If a relatively small pot of grant money could be found, say a few million dollars, it could really make a big difference to these people, in their lives and work. I think you would be astonished at the kind of biodiversity they could produce, and quality of varieties they create.

    It's not as hard as you say to produce and maintain things like blight and other disease resistance in potatoes, but it is an ongoing and full time job someone has to do. Like you pointed out, modern seed companies are good at producing different colored potatoes, but aren't very good at true biodiversity.

  13. Thanks for the link, Patrick,

    1) Demand for plant breeders in industry currently far outstrips supply. The big seed companies have been funding grad fellowships and (due to shortage) have been hiring barely qualified students for outrageous salaries - e.g. $120k! This, of course does nothing for 99% of crops/varieties we care about since these companies only work on industrial varieties of corn, cotton and soybeans.

    2) I'd like to hear more about the breeders not getting paid royalties for their creations. Wasn't it Burbank himself that lobbied congress to extend some kind of patent protection to bred varieties? Why aren't freelance breeders getting royalties?

    3) There's no doubt that public sector support for plant breeding is completely evaporating (and in fact it's in the process of getting much worse as some of the few standing departments will be lucky to survive the recession).

    I am hugely opposed to this trend, which I addressed previously and would absolutely support any proposal to increase funding for academic or freelance plant breeders. The value we would get from this would be many times more valuable than the trivial cost of a small number of salaries.

    3) I'm glad to hear that potato germplasm contains enough diversity to keep up with pest and pathogen resistance - and I expect this would be true for most crops, though I definitely agree that it's a full time job. Last time I was in an industrial organic spinach field, it was explained to me that they needed new downy mildew resistance pretty much every year!

  14. I see elsewhere on the Internet Burbank was a motivation of the Plant Patent Act of 1930. I can't pretend to know a lot about it. It's for plants propagated asexually, and excludes potatoes. I guess it's mostly for things like fruit trees.

    I can tell you a lot more in general about the state of varietal protection in Europe than I can about the US. I don't really understand how it all works for you. I'm also at a disadvantage, my BS is in Computer Science and Mathematics and you are a recent graduate in an agriculture related field. I'm sure you are more on top of this kind of thing than me.

    In general terms, variety protection is not available for Burbank style plant breeders. The costs are too high, and after they've registered their variety the barriers are too high to bring them to market. This can only be done effectively by the larger seed companies, who are looking for more marketable traits.

    In my opinion, there's an important difference between freelance breeders and Burbank-style breeders, the former would likely produce commercial varieties for mono-culture farmers, and the later work on generally improving varieties and increasing biodiversity.

    I'm not really sure there's enough potato germplasm to keep up with pest and pathogen resistance like you say. I think in general this depends on taking a more wholistic approach to the problem, including better enviornmental protection and increased biodiversity.

    The German word for plant breeder is quaker, known in the US for the religious community and brand of oats. It's also the same word for grower. That's because in the era when the word was more commonly used, there was no difference between a breeder or grower, they were just one and the same. I think most germanic languages, and perhaps other European languages, have a single word for plant breeder and grower. In Dutch it's kweeker.

    It's not that we have to ban seed companies, but we really need to return to a time when it was normal to do a little plant breeding and selection, along with growing crops for market. That's really how we are going to deal with blight in potatoes or downy mildew in spinach, is if farmers take some responsibility themselves for breeding and selecting this resistance for their own farms and local climates.

    We really need to return to the time where, in part, biodiversity means every farmer is growing slightly different plants.

  15. Thanks, Patrick.

    That's all very interesting. There's been a lot of buzz recently about "participatory breeding," usually focused on developing world farmers, but I don't know to what extent it's actually occurring...

  16. Interesting thread. I would add that certainly in Food Inc and perhaps to some extent in Botany of Desire, and thirdly throughout this thread, a less than ideal proposition is being attached to diversity and that is: different cultivars of the same plant in the same field.

    This is NOT diversity. I don't care if you plant one each of every type of potato on this earth in the same field, all you have is a potato monoculture.

    As we are doing here, Food Inc. speaks about a grower with multiple types of potatoes in one field, which might prove useful but it defies a true polyculture.

    There are growers on less than an acre with over a thousand different plant types - corn, beans, tomatoes, squash, jackfruits etc. A pest will not find a key that fits every door in that neighbourhood.

    Of course industrial scale farming as it is currently practised en masse in North American and elsewhere, is ill equipped to manage polycultures without either employing more people or developing more sophisticated harvesting technology. There isn't a machine in production that can pick beans corn and squash from the same field at the same time.

    It is noteworthy that no where in nature do we see fields and fields of one plant type. The benefits of following the rules learned from observing the many millenia of success found in nature extend far beyond mimimizing pestilence and would include higher yields, less run off and fertilizer use, minimizing the energy required in annual tilling, greater variety of beneficial insects and predators among other things.

    To look at monocultures and expect them to outperfrom true polycultures, all else being equal, is very unrealistic.

  17. Yeah, I agree with all of that.

    I'm all for perennial/polyculture, but (as you point out) in the Industrialized world machines are affordable and labor is very expensive (and often completely unavailable).

    Precision ag (with computerized robots hooked up to GPS) is really starting to make inroads. If a machine can pick an orange today, maybe it can tend interplanted rows in a few years.

    ...And the more we get the ecology to work on our side, the less we need farmers/machines to micromanage their fields.



Related Posts with Thumbnails