More Recent Comments

Monday, February 12, 2018

Dirty bacteria

Did you know that the dirt in your local park is full of bacteria? Each scoop of soil contains millions of bacteria. And it's not just in your local park, soil bacteria are everywhere. This is part of the reason why the total mass of bacteria on the planet outweighs all of the eukayotes combined, including elephants and whales.

There are hundreds of different species of bacteria in your local dirt. They are as different from each other as moose and mushrooms.

Did you ever wonder whether the bacteria in Australian soil are similar to the bacteria in Austrian soil? Delgado-Baquerizo and his colleagues did, so they tested soils from all over the world. The results are published in a recent issue of Science (Delgado-Baquerizo et al., 2018).

The answer is yes ... and no. They looked at 237 locations on all continents except Antarctica. Most samples had about 1000 different species—the authors call them "phylotypes" because it's hard to define what a species is in bacteria. Only a small number of species (phylotypes) were found in all locations (511 out of 25,224 = 2%) but they accounted for almost half of the total mass. Here's how the authors describe their result ...
Together, our results suggest that soil bacterial communities, like plant communities, are typically dominated by a relatively small subset of phylotypes.
Most of those 511 dominant phylotypes fall into two large and diverse clades (phyla?): Proteobacteria and Actinobacteria. The distribution is shown in Figure 1 of the paper (left). It illustrates a little-known fact about bacteria; namely, that they are a very diverse group. Scientists are only beginning to explore this diversity. Only 18% of the 511 dominant phylotypes were previously known to science!

Image Credit: Bacillus Sp. soil bacteria from The ecology of soil-borne human diseases

Delgado-Baquerizo, M., Oliverio, A.M., Brewer, T.E., Benavent-González, A., Eldridge, D.J., Bardgett, R.D., Maestre, F.T., Singh, B.K., and Fierer, N. (2018) A global atlas of the dominant bacteria found in soil. Science, 359(6373), 320-325. doi: doi: 10.1126/science.aap9516


  1. Possibly off thread. The need to say phylotypes instead of species actually can be seen as a creationist point.
    A YEC creationist would see biology as just phylotypes originally. the great diversity, within kinds, being shades of everything. The species segregation only being a later act of poverty . So in bacteria they simply are the REAL way biology segregates itself. they are not the special case but the rest of biology is!
    Another example, possibly, of this was stephen Goulds work on, help, carribeab land snails. the diversity was so great and yet they could reproduce together and so species names were denied.
    I suspect in these tiny creatures one sees biology as it actually works in reacting to environment and bigger creatures, today, as poor remains of a previous greater diversity.

  2. Most samples had about 1000 different species—the authors call them "phylotypes" because it's hard to define what a species is in bacteria.

    Nope, it's just hard to identify species. A type is defined by characteristics of individuals, which means that you can be certain about whether individual X is a member of type Y. What you aren't certain of is whether type Y corresponds to a species. On the other hand, species are defined by evolutionary dynamics and some members of a species may have different properties from others (i.e. some form of polymorphism) or members of several species may share a lot of properties (see cryptic species). That makes it pretty much impossible to identify which species an individual belongs to, because that would take observing the dynamics over time.

    I had this graphic to show the difference:
    (left side has species bottom as a dated phylogeny, right side has two different type assignments schemes - note that there is one correct way to divide this into species, but there are multiple ways to divide it into types - species are a systematics, types are a classification scheme).

    1. You might want to contact the authors. Here's what they said in the paper ....

      What is currently missing is a detailed ecological understanding of common soil bacterial species, which we refer to as phylotypes (as bacterial species definitions can be problematic).

    2. But a phylotype is defined by a sequence and a minimum level of similarity (chosen arbitrarily). In this case 16S rRNA and a cutoff at 97% similarity was used. The part you quote ends with a citation (Konstantinos et al. 2006), which notes that the phylospecies concept is in principle applicable to bacteria, but it is hard to diagnose(!) these. Bacterial species definitions are problematic because you can't easily identify them. You can however easily identify types. And types can be defined in such a way that they are useful proxies for species. In contrast you claim that it is hard to define what a species is in bacteria. There's a difference between being hard to define and being hard to apply in an ecological study.

    3. Your saying you can identify TYPES easily but not species!
      THEN you say types can be defined in such a way as proxies for species.
      I don't understand you!
      the thing is that in biology "kinds' where there are so many and such variety that it sUGGESTS, as a op[tion, this is the true classification for biology.
      just easy divergence from some original population.
      No hard and fast species due to selection in competition.
      its just that later, under stress there is so much extinction a ILLUSION of species is seen but the poveroty of survivors.
      i see the fossil record as showing this.
      diversity is the original state. Diversity from easy variation in rich environments. like in the modern amazon. the great speciation is really just leisure clasess. its not from violent competition.
      Just a idea that suits creationism better.

    4. I don't understand you!

      That's not very surprising...

      Types are easy to identify, because we make them for the explicit purpose of being able to easily identify them given the data we have available. But they are also arbitrary, which means that any claim to being the "true classification of biology" is begging the question. When a type is defined through 97% or more similarity in 16S rRNA, why not another seuqnece? Why not 98% or 96%?

    5. I still misunderstand your previous post. yUet i agree with you that types are created.
      i question all classifications in biology, I mean i reject them as poorly thought through, suspect that there is just a spectrum in a created kind.
      I agree there are populations maintaining a body plan but the origin of body plans I suspect is from easy living. not competition.
      So in creatures that live easy, shown by numbers(bacteria) and diversity(amazon) the populations being types or species loses all meaning.
      In creatures/biology that live harder libves the extinction in the spectrum is great. So a illusion of species/types takes place.
      just like in people.
      all manjkind if a spectrum of breeding from a original pair/repeated on the ark, but we have clearly species of mankind.
      clearly populations maintaining body plans. We can see the merging on the boundaries yet results are real in the spectrum.
      just thinking.

  3. Are there plausible or proven methods to estimate the mass of bacteria present in different areas? If so, how does it compare to the mass of plants and animals? Presumably such estimates are behind Stephen J. Gould's claim that Bacteria dominate life on earth.

  4. Oh, sorry that I missed your stating the premise: "This is part of the reason why the total mass of bacteria on the planet outweighs all of the eukayotes combined, including elephants and whales."

    So my question, restated, is: How do you measure (locally) those masses? Heck, I'd be happy to know how you measure the mass of the bacteria within a given quart of dirt.

    1. Take some dirt. Count the bacteria. Multiply by the average mass of a typical bacterium. Extrapolate to the entire planet.

  5. The concept of species is so fraught when it comes to bacteria - phylotype is not necessarily better though it works in this context. It's a pain in the neck because it's very hard to talk about individual kinds of bacteria without a firm species name. Nevertheless, I suspect species will never entirely make sense for bacteria.

    Given that the world is basically made out of bacteria, I'm actually more interested in what a geographical compendium of bacterial genes might look like. Some years ago we published a paper showing that Streptomyces coelicolor, which was isolated from soil on the Rutgers campus in 1944, encodes a specific receptor (kijR) and receptor-regulated resistance gene (kijX) for the glycosylated angucyclinone antibiotic kijanimicin. Kijanimicin, as far as I know, is only produced by one known bacterium, namely Actinomadura kijaniatis, isolated in North Africa in the 1970s. The geographical disconnect suggests either that one set of genes is highly mobile or that there are other producers, or both. Or that I'm thinking about this all wrong. Nevertheless, ever since then I've been wondering about what lives where and how it talks to its neighbours, which neighbours etc... a global map of signals and receptors. Fascinating and highly relevant to the propagation of clinical resistance to antibiotics by horizontal gene transfer....

    1. If species never makes sense for bacteria then why is that not important? Is it possible bacteria reveal a truth. A true equation about biology changing bodyplans.
      If bacteria "types" have different bodyplans then why are they not species with names?
      If not ready for prime time species , EVER, then this implys bodyplans easily change in bacteria without being restrained by population boundaries including any natural selection.
      I'm not sure but its possible bacteria, other biology, show how the origin of species actually comes. It doesn't. there is just a easy spectrum and later on extinctions, poor envirorments eliminate some of the spectrum and we call the survivors species.
      As in newtons spectrum however its just white light to the eye. Then a closer look shows colours and closer more colours.

    2. It is important to understand why the concept of "species" doesn't work well for bacteria. It does reveal a truth, though you and I may not agree what truth.

      In bacteria, rampant horizontal gene transfer produces a web of relationships. Certain core genes that produce proteins for DNA replication and proteins synthesis seem to "stay home" more often, and have a sort of tree-like evolutionary relationship, but most genes are different are different. Individual gene sets (operons) and plasmids get transferred around and evolve while treating bacterial cells more like habitats than selves. A bacterial "species" is like one combination in a "mix-and-match" set.

      Bacteria are perhaps even harder to organize into species than plants like blackberries that mix asexual (clonal) seed set with just enough sexual reproduction to drive plant taxonomists to drink.

    3. How is this gene transfer happening? By reproduction or besides it?
      Your still saying the population is not constraining a bodyplan.
      So bodyplans change so quick as to reject the segregational idea of species.
      I'm not sure but its a option this shows how all biology works.
      Its all a spectrum in kinds/types. this from a fluid breeding results. species only appear when poverty appears. only then identified but its a error.
      Possibly this is the reason in many parts of the world they have plants etc that have great numbers of "species" living in one area.
      It simply is a healthier area and more survived from the original spectrum.
      Possibly the cichlid fishes in those african lakes show this equation.