An Infinity of Viruses

When I talk about viruses, I have to struggle with big numbers.

If you get sick with the flu, for example, every infected cell in your airway produces about 10,000 new viruses. The total number of flu viruses in your body can rise to 100 trillion within a few days. That’s over 10,000 times more viruses than people on Earth.

If there can be so many viruses in a single person, how many viruses are there in total on our planet? I’ve hunted around for a number, and the one I’ve seen most often is 1031. As in, 10000000000000000000000000000000. As in over 10 million times more viruses than there are stars in the universe. As in, if you were to stack one virus on top of another, you’d create a tower that would stretch beyond the moon, beyond the sun, beyond Alpha Centauri, out past the edge of the Milky Way, past neighboring galaxies, to reach a height of 200 million light years.

Now it seems that I may have been lowballing that number.

Scientists don’t come up with a number like 1031 by counting every single virus on Earth. They take surveys in different environments–the soil, the water, the ocean floor–and then extrapolate. As they take more samples, they can adjust their estimate up or down.

It used to be that scientists could only conduct these surveys by squinting down a microscope and actually spotting the individual viruses. That turns out to be a bad way to count viruses, in part because their hosts–such as bacteria and protozoans–often don’t grow in the artificial confines of a laboratory. Without viable hosts, the viruses can’t reproduce, and so they go unnoticed.

Things got a lot better once scientists found a new way to count. Instead of looking for full-fledged viruses, they just looked for pieces of their DNA. The ocean, once considered a desert for viruses, turned out to be a broth of viral DNA. Given the sheer volume of the oceans, most of the world’s viruses reside there. And most of the viruses that scientists discovered in the ocean turned out to be parasites of bacteria, known as bacteriophages.

But scientists have long known that there’s another kind of virus out there, one that uses a somewhat different molecule for its genes.

DNA is a double helix, which encodes genes along both strands. When our cells make a protein from a gene, they make a copy on a single-stranded molecule called RNA. DNA-encoded genes can also make RNA molecules that do other jobs, such as sensing the concentration of various elements inside a cell.

In some viruses, RNA also takes on DNA’s job, and encodes genes. Influenza is just one example of the many RNA viruses out there (or in you). Unlike bacteriophages, RNA viruses almost never infect bacteria. Instead, they infect us, plants, fungi, protozoans–all members of the same great lineage of life known as eukaryotes.

The methods that scientists use to tally up DNA viruses often miss RNA viruses, so it’s been hard for scientists to estimate how many RNA viruses are on Earth. In the current issue of the ISME Journal, scientists at the University of Hawaii describe a new method they developed to fill that hole.

The scientists first scooped up water from off a pier on Oahu (about 115 liters all told). They then passed the water through fine filters to exclude bacteria and larger organisms. Through a series of additional steps, they concentrated the viruses down further and then extracted all the DNA and RNA from their samples. The scientists then measured how much of each kind of molecule was present in the sea water. Based on the mass of RNA in an average virus, they could then estimate how many RNA viruses were in the sample.

Their conclusion was remarkable: RNA viruses made up between 38 and 63% of the viruses in the sea water. In other words, about half of the viruses in the ocean are RNA viruses.

This can be surprising when you consider the number of hosts that RNA and DNA viruses can infect. In the ocean, bacteria (hosts to DNA-based bacteriophages) far outnumber eukaryotes (hosts to RNA viruses).

But it’s important to bear in mind that eukaryotes are better as virus incubators. A single eukaryote cell can spew out far more viruses than a single bacterium. Bacteria can’t churn out viruses the way you can, my dear eukaryotic reader.

This new study is just a first cut at estimating the number of RNA viruses on Earth. It’s possible that for some reason the waters off the coast of Hawaii are weirdly good places to find them. The scientists themselves acknowledge that they may be overestimating the number of RNA viruses through some flaw in their methods. But when I contacted an expert on global virus surveys, Curtis Suttle of the University of British Columbia, for his opinion, he gave a thumbs-up. “I think it is an interesting and provocative paper,” he told me. “We will have to see if it holds up.”

If it does, I’ll have to find some new metaphors to describe 10000000000000000000000000000000 + 10000000000000000000000000000000.

[For more on viruses, here’s a recent talk I gave, and here’s my book, A Planet of Viruses.]


*Some people say viruses are truly alive, and some say they aren’t. But if they aren’t, then we need some name that can encompass the things that are “really” alive, along with the things that carry genes, can evolve, and carry out all the other life-ish tasks that viruses do.

There are 22 Comments. Add Yours.

  1. Claudiu Bandea
    February 21, 2013

    The finding that half of viruses in the environment might be “RNA viruses” is a significant finding, and this is a great post about it. However, even if this finding is correct, it should have little effect, if any, on our view about the abundance of viruses on Earth, which is estimated to be at 10 to the power of 31. Obviously, this astronomical number could have been easily underestimated or overestimated by a factor of 10, so doubling or halving this number is well within the margin of error of any estimation of this kind. Is that true?

  2. zmil
    February 21, 2013

    There are a number of RNA bacteriophages, although they seem to be less common than DNA bacteriophages- only two of the 19 known bacteriophage families (Cystoviridae and Leviviridae) are RNA viruses, and I only know of one report of a marine RNA phage. That said, very little is known about marine phages, so an alternate interpretation of this paper is that RNA phages may be more prevalent (in the ocean at least) than previously thought.

  3. Nikhil B. Bhattacharya
    February 21, 2013

    Oh my God!

  4. David B. Benson
    February 22, 2013

    Claudiu Bandea — Yes, these estimates are quite crude.

  5. Gary Grady
    February 25, 2013

    A quibble, estimates of the “number of stars in the universe” usually are estimates of the number in the observable universe. There are good reasons to suppose that the entire universe is vastly bigger and very possibly infinite. See any recent text on cosmology.

  6. Torbjörn Larsson, OM
    February 25, 2013

    Thanks! Very interesting and very useful.

    But if they aren’t, then we need some name that can encompass the things that are “really” alive, along with the things that carry genes, can evolve, and carry out all the other life-ish tasks that viruses do.

    Couldn’t agree more!

  7. Ed Rybicki
    February 26, 2013

    There is not that much difference between 1x10exp31 and 2x10exp31 – not when the margin of error for either DNA or RNA viruses is around ten-fold! So I wouldn’t worry too much about getting it wrong…B-) I also got 150 million light years, but then that’s an even smaller margin! It’s still hugely impressive when you can illustrate that distance with a map of the Universe, with a line stretching from us all the way to the next galactic supercluster.

    In fact, the virus total is probably even higher than 2x the previous estimate: current sampling methods seem to drastically underestimate the population of small circular ssDNA viruses – which have representatives infecting bacteria / archaea as well as marine and terrestrial plants and animals.

    Genuinely a Virus Planet!

  8. Claudiu Bandea
    March 2, 2013

    What is a virus Carl?

    With the recent realization that viruses are the most abundant organisms on Earth and that the repertoire of viral genes exceeds that of cellular species, not to mention the significant role played by viruses in the evolution of their hosts (e.g. more than half of the human genome is composed of endogenous viruses and transposable elements), we can confidently say that Earth is “A Planet of Viruses”. Carl Zimmer, our host, has even written a book about this.

    When it comes to planet Earth, the realization that it is ‘round’, not ‘flat’, has been one of the major breakthroughs in human knowledge. Although for most people, even for scientists studying various aspect of our planet, this knowledge is not essential, nevertheless, they recognize and appreciate this knowledge as highly significant.

    When it comes to viruses, I would think is it important to know what a virus is. Several months ago, I asked virologist Dr. Robin Weiss the question: “What is a virus Dr. Weiss?” The question was in the context of a comment I posted to Weiss’ review of Carl’s book “A Planet of Viruses” (
    For whatever reason, but not unexpectedly, Dr. Weiss did not answer the question. However, I expected that Carl, who is a scholar and a science writer about viruses, would be curious about this question. Probably, Carl did not read my comment and previous papers on the subject (1, 2), or maybe he did not agree with my thesis and that of other virologists (e.g. 3, 4) that the nature of viruses has been misinterpreted ever since they were discovered more than a century ago (metaphorically, that’s more than a hundred years of ‘flat-earth’).

    In a recent post, Carl wrote: “Viruses…they’re just protein shells that package a few genes, which they insert into a host cell” ( I’m sure Carl knows that many viruses have more than “a few genes”; he knows that some viruses have more than a thousand genes, which is many times more than some cellular species. But, I don’t think Carl knows what a virus is.


    1 Bandea CI. A new theory on the origin and the nature of viruses. J Theor Biol.; 105:591-602, 1983.
    2 Bandea CI. The origin and evolution of viruses as molecular organisms. Nature Precedings. 2009.(
    3 Claverie JM. Viruses take center stage in cellular evolution. Genome Biol. 2006; 7:110.
    4 Forterre P. Giant viruses: conflicts in revisiting the virus concept. Intervirology. 53:362-78, 2010.

  9. Claudiu Bandea
    March 2, 2013

    Are viruses alive?

    As pointed out by Torbjörn, Carl’s remark on the question whether viruses are alive or not is certainly valuable: you can write an article in Scientific American, if not a book, expanding on it.

    Most of the problems and confusion associate with this question are associated with the misleading view about the nature of viruses as virus particles and the evolutionary origin of viral lineages.

    Since their discovery more than a century ago, viruses have been identified with the virus particles and defined based on the physical, biochemical and biological properties of these particles (see my comment above and references therein). Because of this misleading concept, the question about the living or non-living nature of viruses has remained in disarray.

    For example, after Wendell Stanley crystalized the tobacco mosaic virus (i.e. the virus particles produced by the virus), for which he was generously rewarded, the obvious view was: how can crystalized proteins (that’s how Miller erroneously described the biochemical composition of the crystalized virus) be considered living entities? Or to use a more recent perspective about the nature of viruses, the one posted by Carl a few days ago, if viruses are “just protein shells that package a few genes”, how can they be regarded as organisms comparable to the cellular microbes, such as Bacteria or Achaea?

    Similarly, the hypotheses about the evolutionary origin of the ancestral viral lineages are critical in addressing the question whether the extant viruses are living or non-living. For example, if viral lineages originated from ‘genetic elements’ that existed before the origin of the cellular organisms, currently the prevalent hypothesis, or if the viral lineages originated later from ‘endogenous cellular genetic elements’, then defining them as living or non-living is more of a judgment call. That’s because it is not clear at what stage during evolution from simple to complex these ‘genetic elements’ can be defined as viruses. On the other hand, if viruses originated from bona fide cellular species (which everyone agrees that are living ) by reductive evolution as predicted by the fusion model on the origin of viruses (1, 2), then it easier and more intuitive to regard viruses as organisms and, therefore, living.


    1 Bandea CI. A new theory on the origin and the nature of viruses. J Theor Biol.; 105:591-602, 1983.
    2 Bandea CI. The origin and evolution of viruses as molecular organisms. Nature Precedings. 2009.(

  10. Ralph Dratman
    March 3, 2013

    “What is a virus” and “Are viruses alive” seem like empty questions to me. I could make up lots of puzzling questions. What species is a liger? Are mushrooms animal or vegetable? What is the IQ of a cunning squirrel?

    Just because the answer to a question is unclear, does not necessarily mean it is an interesting or useful question.

    • Ed Rybicki
      March 4, 2013

      @Ralph Dratman: there is an element of “how many angels can dance on the head of a pin?” in ANY discussion of “what is life?”, because most answers tend to be anthropomorphic and/or reductionist – so it verges on theology, and has about as much point.

      As for “what is a virus?”, however – that CAN be answered, and has been – and it is not very complicated. I have a number of definitions here (; understandably, I prefer my own…B-)

      As for answers to YOUR questions:
      1) a liger is a hybrid, so a member of no species
      2) mushrooms are neither animal nor vegetable; they are fungi
      3) still too low to be smart enough not to cross a road.

  11. Claudiu Bandea
    March 4, 2013

    The issues I addressed in my previous comments might be relevant for some people and not for others. I would predict, for example, that they are highly significant for Carl, who is scholar interested in viruses and has written some wonderful articles and books about them. I would also predict that Carl will continue to write about viruses, but because of the issues addressed here, he will never write again that viruses are “just protein shells that package a few genes”, because that would be scientifically (and metaphorically) inaccurate and misleading. However, it would be great to have Carl let us know his thoughts on these issues.

  12. Ralph Dratman
    March 4, 2013

    @Ed Rybecki: Your definition (“A virus is an infectious acellular entity composed of compatible genomic components derived from a pool of genetic elements”) sounds very good to me — but it might not satisfy Claudiu Bandea. Probably he is looking at viruses in some other way. My point was just that I think it is silly to argue about definitions as if they were independent facts rather than human conventions. And my silly questions were just examples of potentially confusing riddles, like “what is the sound of one hand clapping” — insisting there is a sound and the answer must describe it. Your answers to my #1 and #2 are, of course, correct, but I’m not sure you have completely nailed the squirrel query!

    • Ed Rybicki
      March 5, 2013

      @Ralph Dratman: B-) Yes, I got the one-hand-clapping idea. And there is undoubtedly some angels-on-a-pin stuff going on; I usually tell my students not to worry about it; viruses will prove themselves to be independent organisms with them, frequently.

      And a cunning squirrel is still a stupid mammal.

  13. Claudiu Bandea
    March 5, 2013

    @Ralph Dratman: certainly, any question or issue can be trivialized or obscured; there is an art of doing that. But science doesn’t work that way, at least not in long term.

    To be specific to our conversation, the issue of understanding the nature of viruses is not trivial. For example, because of the misleading concept of viruses as viral particles, even our host Carl Zimmer, a scholar of viruses, described viruses as “just protein shells that package a few genes.” Obviously, Carl knows that his description of viruses is incorrect and misleading; however, the century-old dogma about the nature of viruses is so strong that it overcomes even the most basic knowledge. And, as I previously discussed (see references in previous comments) entire generations of virologists and other scientists have been victims of this dogma.

    Indeed, due the dogma of viruses as virus particles, thousands of scientific articles and books written during the last century contain embarrassing errors that border the pseudo-science realm. Take for example, the following quote from James Watson’s book “Molecular Biology of the Gene” which is highly representative of the modern, scientific description of viruses: “all viruses differ fundamentally from cells, which have both DNA and RNA, in that viruses contain only one type of nucleic acid, which may be either DNA or RNA”. Despite the common knowledge, that within their host cells, the so called ‘DNA viruses’ have both nucleic acids, even James Watson, the eminent scientist, who arguably knew nucleic acids better than anyone, has been a victim of this dogma.

    @Ed Rybicki: in regard to squirrel query, I think the problem with the squirrels is not that they cross the road, but that it crosses the road in front of a car; to be fair, the cars have only become part of their environment a few generation ago, so we should give them a break on the IQ issue; I predict that in a few millions years from now, very few squirrels will cross the road in front of a car (for multiple reasons).

  14. sabas moshy
    March 6, 2013

    Thanks very much Mr carl zimmer for your discovering a very excited scientific news, i real congratulate.

  15. Cees Timmerman
    December 18, 2013

    200 million lightyears? Not 100 million?
    Or 100 thousand?

    You’re spreading the struggle meme, Carl.

    [CZ: Thanks for pointing out that error on my web site–I’ll fix that. I can’t load your link, but all I can say is that the figure of 200 million light years is what you’ll find in the scientific literature.]

    • Ed Rybicki
      December 19, 2013

      I think a back-of-the-envelope (or preferably, these days, front-of-calculator) calculation based on average virion size = 100 nm, and number of viruses = 10exp31, will give you 170 million light years.
      As I have mentioned above, however, the number of viruses is estimated from concentrations in seawater, and probably seriously underestimates both the number of RNA and ssDNA viruses – so there are probably a LOT more viruses than we thought there were.
      Meaning they would stretch even further. It’s a virus universe, folks…as we will discover, of we ever get to Mars, or preferably Europa, Ganymede and Enceladus.

  16. H. (Bart) Vincelette
    March 11, 2014

    From a more grassroots level: My understanding of a virus is that it is an incomplete organism due to its lack of sufficient DNA to replicate itself. So, viruses have to hijack the DNA of a host cell in order to reproduce themselves.They were so difficult to study until we had electron microscopy & later, the polymerase chain reaction of duplicating the viral DNA to such numbers that it could be studied. There are likely viruses out there with the potential to wipe out our species. Further destruction of rain forests may expose us to some of them. So, there are viruses that give you a headache & sniffles (coryza, or a cold), some that invade the lower respiratory system , causing viral pneumonias, & some that invade the cells of the circulatory system & can kill in less than 72 hours, such as Ebola.The intense scrutiny of HIV has given us tremendous insights into the mysteries of viruses.With the innumerable number of diseases & disabilities for which the cause is unknown; it is possible that many are caused by yet unidentified viruses. I am an expert in nothing; but I have buries fifty-seven good friends to the Human Immunodeficiency Virus which cripples parts of the human immune system that are routinely involved in fighting off various cancers & infections.

  17. mary m. fraraccio
    October 23, 2014

    So why worry so much about one when there are countless other’s everywhere? It’s like if we knew how many gazillions, surely it would kill us from the shocking truth! Our exquisite human immune systems, our defensive antibodies, are something else and there’s much to be said about there ability to function so remarkably. I’m surprised I haven’t killed over already! Dear Heavenly Father I’m having a holy cow, we all going to die there’s no question about it, all in our Season….but not EARTHA !

  18. Lary Robinson, M.D.
    March 29, 2015

    Viruses likely cause most all cancer–20% of cancers are currently proven to be viral induced. Many of the other currently-unexplained, non-malignant fatal illnesses that plague mankind likely will be eventually be shown to be viral in origin. In fact the aging process itself with cellular senescence and death likely is related to endogenous retroviruses. Doesn’t this all sound like the major “purpose” of virus is to keep humans from living too long and overpopulating the Earth. And as medical science slowly learns to protect us from the many viral illnesses so that we do live longer, simultaneously our society is maturing. Perhaps it is no coincidence that humans “social” evolution is occurring to decrease the rate of reproduction and population growth. So if there truly is a “designer” (as over 90% of Americans believe), it seems that giving the world viruses to cull the herd of humans (in addition to perhaps generating genetic diversity as outlined in this article) in a rather ingenious idea that is effective!

    • D Gary Grady
      March 31, 2015

      @Dr Larry Robinson: Interesting speculation, but as I’m sure you know, most cancer deaths occur in people past reproductive age. Infectious disease (viral, bacterial, fungal, parasitic) have historically killed lots of children and thereby acted as a much more efficient constraint on population growth, but the human population was growing anyway even before the relatively recent development of vaccines, antibiotics, and now antivirals. I would hope than an intelligent designer would realize that a disease affecting fertility would be a far less murderous way of controlling population. Also, the latest General Social Survey from the University of Chicago and multiple other sources (Pew, for example) suggest that far less than 90% of the U.S. public believes in a designer.

Add Your Comments

All fields required.

Related Posts