« July 2011 | Main | September 2011 »

August 24, 2011

Education as a tragedy of the commons

The New York Times is discussing whether we spend too much on education, or not enough. It might help to phrase the question more explicitly, either as:

1) Would individual families benefit from investing more in their own education?

2) Would society as a whole benefit from investing more in education?

Two commentators point out that individuals with more education have substantially higher salaries, on average. Maybe those individuals would have had higher salaries even if they hadn't gone to college, of course, or even if they'd gone to a less-expensive college. That question has been debated elsewhere, and I don't have anything to add.

But even if college is a good investment for individuals, it isn't necessarily true that we would be better off, collectively, if we spent more, collectively, on education. Assume, for the sake of argument, that the number of high-paid jobs is fixed at 10% of the working-age population. If those with more education are more likely to get those jobs, then investing in education makes sense for individuals and their families.

Education could be a sound individual investment even if the education itself was worthless and incomes were determined only by credentials relative to others. But, if producing more people who are qualified to be medical doctors doesn't increase the number of jobs for doctors, then the higher pay of college-educated doctors relative to less-educated nondoctors doesn't make educating more people a good investment for society as a whole. Individual benefits from investing in education don't necessarily translate into societal benefits from investing in education.

How might society benefit if more people were more educated? My guess is that, if more people were better educated, we would have less crime, a stronger economy, and better political decision-making. My guess is that these societal benefits would outweigh the societal costs of investing more in education. But you can't estimate the total return on these societal investments from the economic return to individuals from investing in their own education. Societal benefit:cost ratios could be lower or higher than those for individuals.

Here's another important point. The return on investments in one kind of education for one group doesn't necessarily predict the return on investments in other kinds of education for other groups. Years ago I saw a comparison of countries that started with similar economies and invested similar amounts of public money either in universities or in primary education. Investing in primary education led to lower birth rates, so the education budget was spread among fewer kids. They developed educated workforces and their economies grew, so that eventually they could afford universities as well. (Until then, citizens wanting more education got it in other countries, thereby making contacts that improved trade.) In contrast, the countries that focused on universities only educated -- "credentialed" might be a better word -- the children of the ruling kleptocracy. The overall population stayed ignorant, birth rates stayed high, and economies stagnated.

Fortunately, I live in a rich country that can afford both great universities and great primary and secondary schools. An honest cost:benefit analysis would be a good place to start.

August 19, 2011

Reciprocity maintains cooperation between plants and mycorrhizal fungi

Mycorrhizal fungi attach to plant roots and trade phosphorus (and sometimes other benefits) for photosynthates. But is it really "trade"? In other words, if one partner defects, will the other continue its contribution? A recent paper in Science, by Toby Kiers and colleagues, shows that the term, "trade" is justified.(Kiers et al. 2011)

Earlier, as my first PhD student, Toby showed that symbiotic rhizobia, living inside soybean root nodules, can't stop supplying the plant with nitrogen without suffering a decrease in their own fitness.(Kiers et al. 2003) We called the plant response "sanctions", although this could be misinterpreted as an attempt by the plant to improve the behavior of the rhizobia. We assume that the behavior of a given rhizobial genotype is programmed by its DNA, but sanctions against less-beneficial strains will decrease their frequency in subsequent generations. More recently, Ryoko Oono demonstrated a subtler form of sanctions against potentially reproductive rhizobia when their nonreproductive clonemates stop fixing nitrogen.(Oono et al. 2011)

Different strains of rhizobia on the same plant are (mostly) segregated in different root nodules, so "shutting down" one nodule preferentially hurts a single strain of rhizobia. But multiple strains of mycorrhizal fungus typically infect the same root. Can plants preferentially allocate resources to the most-beneficial fungi? Jim Bever and colleagues showed that they can, when different strains are attached to different parts of the root, but apparently not when the strains are more mixed.(Bever et al. 2009) Results from Kiers' group, however, suggest that mixing is not a problem.

They grew Medicago truncatula (a model species related to alfalfa) with mixtures of mycorrhizal fungi that they classified as more- or less-beneficial. To track photosynthate allocation, they let plant photosynthesis take up carbon dioxide containing the heavier 13C isotope of carbon. To see which fungi got more of this carbon, they used a clever technique.

They extracted RNA from the fungi and separated it into lighter and heavier (more-recent-plant-carbon) fractions, using a centrifuge. Then they used difference in the RNA base sequences of their different fungi to measure their relative representation in the light and heavy fractions. Fungi with greater representation in the heavy fraction were getting more recently supplied carbon from the plant, so they were benefiting more from symbiosis. A more-beneficial fungal species got more carbon than either of two less-beneficial species. This result is consistent with host sanctions (or, since each fungus interacts with multiple plants, with a "biological market" where individuals trade more with those that offer the best deal).

It's hard to measure all of the benefits a mycorrhizal fungus provides, however, so they also manipulated the exchange of specific resources. The species they classified as more beneficial got more carbon when it had access to more phosphorus, and presumably supplied it to the plant. A less-beneficial species didn't get more carbon when it had access to more phosphorus, perhaps because it didn't give much of the phosphorus to the plant. Similarly, the cooperative strain apparently responded to differences in carbon supply from the plant, allocating more phosphorus to where it was getting more carbon. So sanctions appear to go both ways.


Bever J. D., S. C. Richardson, B. M. Lawrence, J. Holmes, and M. Watson. 2009. Preferential allocation to beneficial symbiont with spatial structure maintains mycorrhizal mutualism. Ecology Letters 12:13-21.

Kiers E. T., M. Duhamel, Y. Yugandgar, J. A. Mensah, O. Franken, E. Verbruggen, C. R. Felbaum, G. A. Kowalchuk, M. M. Hart, A. Bago, T. M. Palmer, S. A. West, P. Vandenkoornhuyse, J. Jansa, and H. Bücking. 2011. Reciprocal rewards stabilize cooperation in the mycorrhizal symbiosis. Science 333:880-882.

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.

Oono R., C. G. Anderson, and R. F. Denison. 2011. Failure to fix nitrogen by non-reproductive symbiotic rhizobia triggers host sanctions that reduce fitness of their reproductive clonemates. Proceedings of the Royal Society B 278:2698-2703.

August 12, 2011

This week's picks

Reciprocal Rewards Stabilize Cooperation in the Mycorrhizal Symbiosis Toby Kiers, who previously demonstrated host sanctions against cheating rhizobia, now shows that plants give less carbon to less-beneficial mycorrhizal fungi. I hope I can find time to discuss this paper in more detail soon.

Natural variation in Pristionchus pacificus dauer formation reveals cross-preference rather than self-preference of nematode dauer pheromones "strains may have evolved to induce dauer formation precociously in other strains in order to reduce the fitness of these strains"

Nest Inheritance Is the Missing Source of Direct Fitness in a Primitively Eusocial Insect

Polyandrous females benefit by producing sons that achieve high reproductive success in a competitive environment

Kin selection in den sharing develops under limited availability of tree hollows for a forest marsupial

Aging of the cerebral cortex differs between humans and chimpanzees "significant aging effects in humans were... individuals that were older than the maximum longevity of chimpanzees. Thus... brain structure shrinkage in human aging is evolutionarily novel and the result of an extended lifespan"

Bacterial persistence by RNA endonucleases

Host-parasite local adaptation after experimental coevolution of Caenorhabditis elegans and its microparasite Bacillus thuringiensis

Sperm chemotaxis, fluid shear, and the evolution of sexual reproduction

August 7, 2011

Nitrogen-fixing cereals?

This week's paper was published in Science (Beatty and Good 2011). It discusses the prospects for developing cereals, like wheat or rice, that can use ("fix") atmospheric nitrogen. This paper was brought to my attention both by my wife, Cindy, by Andy McGuire. As my first graduate student, at UC Davis, Andy tested the use of legume "green manures", which form symbioses with nitrogen-fixing rhizobia bacteria, to supply nitrogen to a subsequent wheat crop. (McGuire et al. 1998)

This week's paper suggests three approaches. Maybe we could engineer cereals to host rhizobia in root nodules, similar to those found on legumes. Maybe we could encourage looser associations between cereals and bacteria to fix nitrogen in or on the plant, even without nodules. Or, maybe we could engineer the plants themselves to fix nitrogen.

The authors note that the proposed cereal-root nodules would need to provide a low-oxygen environment, to protect the key enzyme, nitrogenase. Just keeping oxygen low isn't that difficult. Legume nodules have a physical barrier to gas diffusion that limits oxygen influx to the nodule interior, and it might not be that hard to do something similar in cereals. But rhizobial nitrogen fixation is powered by respiration, which requires lots of oxygen. So rhizobia need a high oxygen flux, in addition to a low oxygen concentration. Diffusion is slow at the low concentrations beyond the diffusion barrier, so diffusion of free oxygen within rhizobia-infected cells can't meet demand. Much of the oxygen is carried by diffusion of a plant hemoglobin, whose high concentration makes the inside of legume root nodules red.

And that's not the end of the challenges we would need to overcome to put functional root nodules on cereals. Suppose the diffusion barrier were just thick enough that respiratory uptake drops oxygen concentration from atmospheric (21 kPa) to the targeted near-zero concentration (1 kPa, say) across the diffusion barrier. What if the respiration rate drops to 90% of it's current value? (A slight decrease in soil temperature could have this effect, as could a shortage of photosynthate.) The concentration drop across the barrier is proportional to the flux (Fick's Law of Diffusion, similar to Ohms law for electricity), so the concentration drop would now be 18 kPa instead of 20 kPa. So the nodule-interior oxygen would now be 3 kPa instead of 1 kPa. 3 kPa is probably high enough to destroy nitrogenase.

Evolution has solved this problem, too, at least in legumes. As conditions change, legumes adjust the gas permeability of their nodules to keep oxygen low enough to protect nitrogenase, but high enough for respiration to meet their nitrogen-fixation needs. When sheep graze clover plants, the resulting photosynthate shortage decreases nodule respiration, so you might expect nodule-interior oxygen levels to rise, perhaps endangering nitrogenase. But clover plants decrease their nodule gas permeability decreases so much that nodule-interior oxygen actually decreases, rather than increasing (Hartwig et al. 1987, Denison and Okano 2003). Cereal root nodules would need something similar.

If we could engineer cereal crops to allow infection by rhizobia and support their reproduction inside root nodules that actively regulate oxygen supply, maybe the rhizobia would fix nitrogen there... at first.

But, like all living things, rhizobia evolve. Any mutation that reduced rhizobial investment in nitrogen fixation would free resources for additional rhizobial reproduction. If there were only one rhizobial genotype per plant, this "cheating" would be self-defeating, because a nitrogen-starved plant would have less photosynthate to support rhizobia. But, with multiple genotypes per plant, we have a "tragedy of the commons", favoring cheaters.(Denison 2000, West et al. 2002)

Legume evolution has found at least a partial solution to this problem, too, although there's probably room for improvement. Although moderate levels of cheating may be tolerated (Kiers et al. 2006), major diversion of resources from nitrogen fixation to rhizobial reproduction triggers host "sanctions", which reduce the fitness of rhizobial cheaters (Kiers et al. 2003, Oono et al. 2011). If we don't want nitrogen-fixing cereals to waste photosynthate supporting nonfixing rhizobia, they would need to impose sanctions, too. Or maybe growing them in rotation with sanction-imposing legumes hosting the same rhizobia would be enough to keep rhizobial cheaters rare.

What about the second option, cereals without nodules, but associating with nitrogen fixers? To meet a significant fraction of a cereal's nitrogen needs, we would need to regulate oxygen supply, as in nodules. And we would soon face the same problem of cheaters (Kiers and Denison 2008), so we'd also need some form of sanctions. If we're have to duplicate the functions of nodules anyway, why not copy nodules, rather than starting from scratch?

Engineering the cereals themselves to fix nitrogen, without rhizobia, would solve the problem of cheaters. The article suggests that it might be possible to add nitrogen fixation to plant mitochondria, the current site of respiration, or to chloroplasts, responsible for photosynthesis. Photosynthesis generates oxygen, which would tend to destroy nitrogenase. But the authors say that "some cyanobacteria perform photosynthesis and nitrogen fixation in the same space but at separate times" - maybe chloroplasts could somehow shield nitrogenase from oxygen during the day, when they're photosynthesizing, and activate nitrogenase only at night. This seems possible, at least in theory. But could it be done soon enough to help avert looming food shortages?

The article closes with the hope that "if nitrogen supply and carbon metabolism can be closely coupled, excess nitrogen would not be lost to the environment." That is already true of legumes, which shut down nodules when they have as much nitrogen as they need (Denison and Harter 1995). But this feedback control evolved - legumes that wasted photosynthate on "extra" nitrogen fixation were out-competed by more-frugal mutants - it would not be an "automatic" outcome of any of the approaches proposed.

So, again, any attempt to develop nitrogen-fixing cereals would benefit from copying what legume nodules already do. It's too bad that we don't yet know the mechanisms that regulate oxygen supply in legume nodules, impose sanctions on rhizobial cheaters, or adjust nitrogen-fixation rate to match nitrogen needs. Regulation of nodule gas permeability appears to be involved in all three, but we don't know how gas permeability is regulated. It's too bad that nobody wants to fund that kind of research any more.


Beatty P. H., A. G. Good. 2011. Future Prospects for Cereals That Fix Nitrogen. Science 333:416-417.

Denison R. F. 2000. Legume sanctions and the evolution of symbiotic cooperation by rhizobia. American Naturalist 156:567-576.

Denison R. F., Y. Okano. 2003. Leghaemoglobin oxygenation gradients in alfalfa and yellow sweetclover nodules. Journal of Experimental Botany 54:1085-1091.

Denison R. F., B. L. Harter. 1995. Nitrate effects on nodule oxygen permeability and leghemoglobin. Nodule oximetry and computer modeling. Plant Physiology 107:1355-1364.

Hartwig U., B. Boller, and J. Nösberger. 1987. Oxygen supply limits nitrogenase activity of clover nodules after defoliation. Annals of Botany 59:285-291.

Kiers E. T., R. F. Denison. 2008. Sanctions, cooperation, and the stability of plant-rhizosphere mutualisms. Annual Review of Ecology, Evolution, and Systematics 39:215-236.

Kiers E. T., R. A. Rousseau, and R. F. Denison. 2006. Measured sanctions: legume hosts detect quantitative variation in rhizobium cooperation and punish accordingly. Evolutionary Ecology Research 8:1077-1086.

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.

McGuire A. M., D. C. Bryant, and R. F. Denison. 1998. Wheat yields, nitrogen uptake, and soil water content following green manure vs. fallow. Agronomy Journal 90:404-410.

Oono R., C. G. Anderson, and R. F. Denison. 2011. Failure to fix nitrogen by non-reproductive symbiotic rhizobia triggers host sanctions that reduce fitness of their reproductive clonemates. Proceedings of the Royal Society B 278:2698-2703.

West S. A., E. T. Kiers, E. L. Simms, and R. F. Denison. 2002. Sanctions and mutualism stability: why do rhizobia fix nitrogen? Proceedings of the Royal Society B 269:685-694.

August 1, 2011

Are we doomed?

New Statesman asked a bunch of scientists this question. Bjørn Østman blogged that:

"it is likely that the human population will be decimated some time in the future, perhaps even as soon as within the next hundred years. A cataclysmic event caused by global climate change, perhaps? If anyone survives at all, I predict it will be in a number of smaller populations separated from each other geographically.... Assuming that the environmental changes caused by the global cataclysmic event are severe enough that the separated populations won't be able to interact (i.e., have sex) for long enough, the different populations will continue to diverge from each other, and speciation will eventually occur. 6-7, say, million years hence..."
I can't tell if this is supposed to be a joke, but I can't imagine humans surviving, yet staying isolated that long. And I can't imagine climate-change-related catastrophe reducing world population below a billion or so. To eliminate all but a few isolated populations would take a major extraterrestrial impact, global nuclear war, or maybe some disease.

I don't think humans are likely to go extinct within the next few thousand years. But is civilization doomed?

Can we define civilization as a continuum? At the high end:
* there is some form of collective decision-making that respects factual information and serves everyone's long-term interests, to the extent possible,
* literacy is nearly universal, as is access to accurate scientific and historical information, and to differing opinions on politics, religion, etc.
* original research to increase scientific understanding and improve technology is encouraged
* criticizing the government or the dominant culture is not a crime
* those accused of crime are entitled to some form of due process based on public examination of the evidence

At the low end:
* good and services are exchanged mostly through voluntary trade, at least within groups, but groups war among themselves and individuals can be executed for "witchcraft", "blasphemy", or "sedition."

The question is, if some natural or man-made disaster (a major epidemic, say, or failure of moderates to vote) reduces civilization to a given level, somewhere between these extremes, which direction will it tend to change, over the next thousand years or so? Will within-group processes tend to increase or decrease civilization? Some current trends in the US are not encouraging.

What about between-group processes? Will countries that respect science and other secular values be strong enough (economically? militarily?) to impose their values on countries that rely on "God, guns and guts", or will it be the other way around?

Carnival of Evolution

Larry Moran's Sandwalk has an extensive roundup of recent blog posts on evolution.