Darwinian agriculture II
Last week, I was at a meeting in the Netherlands on "Darwinian agriculture: the evolutionary ecology of agricultural symbiosis." Topics included: the effects of cows on human evolution, the independent invention of "agriculture" by ants and termites, and some disadvantages of diversity. As promised, here are a few highlights.
Dan Bradley (Ireland) discussed the evolution of cows since domestication by humans. How closely related are cows to the extinct wild aurochs? In other words, how recently did they share a common ancestor? In other words, how long have humans limited interbreeding between wild aurochs and the animals they raised for milk and meat? (Limiting interbreeding allows two groups to evolve differences, as discussed previously for fish.) Different methods, both based on DNA from old auroch bones and cows, give different answers. When they compared mitochondrial DNA (inherited only from mothers) they estimated that cows and aurochs last interbred about 10,000 years ago. So humans domesticated aurochs around the same time as crops like wheat. But comparing Y chromosomes (inherited only from fathers) suggests much more recent crossing. Which is right? Probably both. If an aurochs bull sneaked into a field and mated with a cow, the calf would be raised as part of a farmer's herd. But calves from reverse matings stayed with their wild aurochs mother. Similar patterns are seen in some human populations. For example, in the British Isles, Viking Y chromosomes are more common than Viking mitochondria.
Apart from the occasional auroch raid, humans have controlled the evolution of cows for thousands of years. But cows have influenced our evolution, too. Albano Beja-Pereira (Portugal) mapped the distribution of lactose tolerance in humans. Other mammals (except house cats) can't digest this milk sugar as adults. In Europe, the ability to digest lactose is concentrated in northern Europe, centered on Denmark. People in this region have been using cows to convert grass into meat and milk for thousands of years. If you have only one cow, the choice between meat for a few days or milk for months seems pretty clear. Among humans who raised cows, the few who didn't lose the ability to digest milk as they got older were more likely to survive and reproduce, so the frequency of lactose tolerance increased over generations. A similar genetic pattern has been seen in some Africans whose ancestors raised cows. Beja-Pereira also looked at genetic variations in cow genes for milk proteins. These genes were more diverse in the same part of Europe where lactose tolerance was most common in humans. This diversity may reflect some combination of deliberate selection (choosing which cows and bulls to breed) and the evolutionary effects of larger cow populations in this region.
Mike Jeger (UK) explained how computer models of the spread and evolution of pathogens (microbes that cause disease) can help us defend crops. A new fungicide may work for a few years, until mutant fungi resistant to it become common. If we switch to a new fungicide (or some other control method) soon enough, the resistant mutants usually disappear after a few years. There is often some cost of resistance - a fungicide-resistant enzyme may not work as well -- so they can't compete with nonresistant members of their species. But once resistant mutants are common, some may acquire new mutations that reduce the cost of fungicide resistance. After that, we will have to avoid using the fungicide for much longer, before resistant fungi disappear. He also discussed the "evolutionary epidemiology�? of viruses that attack crops. We can breed crops for virus resistance, but some kinds of resistance will be beaten by viral evolution faster than others. With high-value crops, hand removal of plants that show virus symptoms can be practical. Then the only virus strains that survive to the next generation are those that cause disease too mild to be noticed. This approach reminds me of the use of window screens to select for less severe forms of malaria.
The remaining four talks covered various topics, but the effect of diversity was a common theme. Benefits from crop diversity are well known. Disease may spread more slowly in a mixture of two crops than in one-crop "monoculture.�? Mixtures may use soil resources more completely than either crop alone. But all four talks discussed cases where less diversity may actually be better.
Koos Boomsma (Denmark) and Duur Aanen (The Netherlands) discussed ants and termites that depend on fungus "gardens�? for food. In both cases, different insect colonies may grow different strains, or genotypes, of fungi. But each colony grows only one strain. In ant gardens, contact between two different fungal strains triggers a negative reaction that reduces growth. Even manure from ants that ate one strain will trigger this reaction in a second strain. In termite gardens, different fungal strains don't fight. But they don't bond, either, and this also limits growth. Over tens of millions of years, ants and termites have evolved behaviors that maintain their gardens as fungal monocultures. Ants remove alien fungi, even strains that might be grown by another ant colony. Termites prevent their fungi from reproducing sexually, by eating fruiting bodies that could produce sexual spores. Without sex, one strain gradually takes over.
My former student, Toby Kiers (US and Netherlands), talked about how human farming methods affect the evolution of microbes that help crops. Rhizobia are bacteria that provide some crops with nitrogen. Mycorrhizal fungi provide some crops with phosphorus. Both are symbiotic, in long-term physical contact with their plant hosts. If there were only one strain of microbe per plant, plants with better microbes would grow more. Microbes associated with bigger plants would reproduce more. So better microbes would be more common in the next generation. In other words, the microbe populations would evolve greater mutualism. That's what would happen if the microbe population associated with each plant were a monoculture, but it's not. With many strains per plant, a strain that helps the plant indirectly helps its worst competitors. Microbes from the same plant are, after all, most likely to compete for the next host plant. Toby has shown that soybeans, at least, have evolved a defense against bad rhizobia. She kept some rhizobia from giving the plant any nitrogen, as if they were a bad strain. She did this by surrounding the nodules (bumps on the roots that house the rhizobia), with gas lacking nitrogen. The plant shut off oxygen to those nodules, and the rhizobia inside grew less. My last post has an MP3 file describing this work. Toby also compared different soybean varieties, in the field. Some did better with a mixture of good and bad rhizobia than others, perhaps due to differences in the "sanctions�? they impose on bad rhizobia. Sanctions against nodules that don't supply nitrogen may solve the problem of too much rhizobium diversity within a plant. Diversity within a nodule may still be a problem, though. We don't think the plant can impose sanctions on some rhizobia within a nodule without hurting them all.
Finally, I talked about breeding crops that yield more per acre (or hectare) because individual plants compete less with each other. The best-known example is plant height. Short plants make more grain because they waste less on stems. This works well if you have a whole field of short plants. But, in a mixture, the taller, low-yield plants shade out the shorter high-yield plants. Plants that branch less can yield more, in monoculture, but can't compete against plants that branch more.
I discussed similar tradeoffs, especially for roots in a previous post. At the meeting, Hans de Kroon questioned the paper I cited on wasteful aggression among roots. Two plants sharing a large pot may grow more roots, relative to each plant would in a pot half the size, just because they make more roots in a bigger pot, not because of root interactions. Breeding for less root is probably still a good idea, but root interactions may not be the key.
Whether we look at ant or termite fungus gardens, microbes that help crops, or crops themselves, diversity can lead to interactions that reduce growth. Should we work to reduce diversity in agriculture, then? Not exactly. Diversity may be useful at some scales, but harmful at others. If the world grew more different crops, a disease that killed any one crop would have less effect. But that may not mean that every field should contain more than one crop. We may also benefit by growing different crops in a field in different years, the well-known practice of crop rotation.
Shortly after returning from the meeting, I got a notice for another meeting on "Plant breeding for organic and sustainable, low-input agriculture." I agree that organic farmers need cultivars that are competitive with weeds. I hope speakers at this planned meeting will recognize the tradeoffs between competitiveness and yield, discussed in previous entries and in our paper on Darwinian Agriculture. For a farmer, higher prices for organic products may balance lower yields. But lower yields require using more land to grow the same amount of food, perhaps draining wetlands or clearing forests. Population growth and using crops to make ethanol put still more strain on food supply, raising prices and increasing demand for land to farm.
Related papers by the speakers:
Beja-Pereira A., G. Luikart, P. R. England, D. G. Bradley, O. C. Jann, G. Bertorelle, A. T. Chamberlain, T. P. Nunes, S. Metodiev, N. Ferrand, and G. Erhardt. 2003. Gene-culture coevolution between cattle milk protein genes and human lactase genes. Nature Genetics 35:311-313.
Denison R. F., E. T. Kiers, and S. A. West. 2003. Darwinian agriculture: when can humans find solutions beyond the reach of natural selection? Quarterly Review of Biology 78:145-168.
Gotherstrom A., C. Anderung, L. Hellborg, R. Elburg, C. Smith, D. G. Bradley, and H. Ellegren. 2005. Cattle domestication in the Near East was followed by hybridization with aurochs bulls in Europe. Proceedings of the Royal Society B 272:2345-2350.
Hess K. and H. de Kroon. 2007. Effects of rooting volume and nutrient availability as an alternative explanation for root self/non-self discrimination. Journal of Ecology 95:241-251.
Jeger M. J., S. E. Seal, and F. Van den Bosch. 2006. Evolutionary epidemiology of plant virus disease. Advances in Virus Research 67:163-203.
Kiers E. T., S. A. West, and R. F. Denison. 2002. Mediating mutualisms: farm management practices and evolutionary changes in symbiont co-operation. Journal of Applied Ecology 39:745-754.
Kiers E. T., R. A. Rousseau, S. A. West, and R. F. Denison. 2003. Host sanctions and the legume-rhizobium mutualism. Nature 425:78-81.
Poulsen M., J. J. Boomsma. 2005. Mutualistic Fungi Control Crop Diversity in Fungus-Growing Ants. Science 307:741-744.