For a long time, one number did a great deal of work in popular accounts of human biology. Ten to one: for every cell that was genetically yours, ten bacteria were along for the ride, making you, in some loosely held sense, more microbial than human. The figure appeared in university textbooks, in science journalism, in conference presentations, and in the kind of enthusiastic public talks that explain the body as an ecosystem. It had the quality that the best scientific factoids tend to have: it was startling, it was tidy, and it made you reconsider something you had taken for granted.
It was also wrong, or at least considerably overestimated, and the correction, when it came, turned out to be as instructive as the original claim.
A 1972 Estimate That Became a Textbook Fact
The ten-to-one ratio is traceable, in the main, to a 1972 calculation by the American biochemist Thomas Luckey — though it reached the broader literature primarily through a 1977 paper by Dwayne Savage, which cited Luckey’s estimate and became the reference most subsequent work cited in turn. Working from available data on bacterial density in gut samples, Luckey extrapolated an estimate for the total bacterial load across the human body. The figure that emerged, roughly 1014 bacteria against around 1013 human cells, was an order-of-magnitude comparison, the kind of rough approximation that makes sense as a starting point for further inquiry.
The assumptions embedded in that calculation were, by later standards, quite coarse. Luckey’s estimate took bacterial density measurements from colonic content samples and applied them broadly across the gastrointestinal tract, when in fact bacterial density varies enormously along the digestive system. It also treated both “a standard bacterium” and “a standard human cell” as fixed-size units, which neither is. Most consequentially, the estimate did not adequately account for red blood cells, which are considerably smaller than most other human cell types and make up approximately 84 per cent of human cells by count. That omission alone substantially distorted the final ratio.
None of this made the calculation dishonest. Luckey was working at the limits of what was then measurable. The problem came later, when the figure was absorbed into the literature not as a rough estimate with known limitations but as an established ratio, passed from paper to paper, textbook to textbook, without being traced back to its foundations.
Going Back to Check the Arithmetic
In 2016, researchers Ron Sender, Shai Fuchs, and Ron Milo at the Weizmann Institute of Science decided to do exactly that. Their first paper, “Are We Really Vastly Outnumbered? Revisiting the Ratio of Bacterial to Host Cells in Humans,” published in Cell in January 2016, questioned whether the ten-to-one figure could be sustained on current evidence. The short answer was that it could not.
Sender, Fuchs and Milo then rebuilt the estimate from the ground up, drawing on more recent measurements of bacterial populations across different regions of the gut, better data on human cell counts by type, and a clearer accounting of body composition. Their companion paper, “Revised Estimates for the Number of Human and Bacteria Cells in the Body,” published in PLOS Biology later that year, provided the corrected numbers.
For a reference adult male of 70 kilograms, the revised estimate put the bacterial count at approximately 3.8 × 1013 cells and the human cell count at approximately 3.0 × 1013. That is a ratio of roughly 1.3 to 1 in bacteria’s favour, not 10 to 1. The microbiome remained numerically dominant, but only just, and the margin was close enough to prompt Milo to describe the original estimate as a rough order-of-magnitude figure that, over decades, had been dressed up as established fact.
A Ratio That Moves
One of the more counterintuitive details in Sender et al.’s revised accounting is that the ratio is not a fixed property of the body at all. It fluctuates over the course of a day, and one of the main reasons is mundane: bowel movements. A significant fraction of the body’s bacteria are concentrated in the colon, and a single bowel movement can remove roughly one-third of the colon’s bacterial population, temporarily shifting the ratio back toward parity, or even in favour of human cells, before bacterial numbers recover.
The total bacterial load also varies considerably depending on assumptions about body size, diet, and health status, and the 38 trillion figure should be understood as a central estimate within a range that runs from roughly 30 to 50 trillion depending on the model’s inputs. Despite their numerical weight, those bacteria are not particularly heavy: the entire microbiome amounts to roughly 200 grams, less than 0.3 per cent of body weight. A figure that once seemed to suggest a kind of microbial takeover turns out to describe something more like a close and unequal cohabitation.
Why the Error Survived
The intellectual history here is worth pausing on, because the ten-to-one figure’s longevity was not an accident. It had several properties that made it sticky.
It was memorable. A ten-to-one ratio is the kind of number that lodges in the mind; a ratio of 1.3 to 1 does not carry the same rhetorical weight. It was also intuitively satisfying in a way that aligned with a broader cultural moment: the microbiome was, in the early 2000s, emerging as a subject of genuine scientific interest, and a figure that made you more microbe than human gave that research a compelling frame. The factoid served a narrative purpose, and that purpose kept it in circulation even as the tools available to check it improved.
There is also a structural point about how scientific claims propagate. A figure that enters the textbook literature is rarely re-examined unless someone has a specific reason to question it. Most researchers who cited the ten-to-one ratio were not making claims about bacterial cell counts; they were using the ratio as contextual colour, a way of conveying the scale of the microbiome before moving on to their actual subject. When a figure functions as backdrop rather than argument, the incentive to audit it is low.
Sender, Fuchs and Milo had that incentive, and the result was a correction that made headlines not because the underlying science had failed but because a rough estimate had been promoted, through accumulated repetition, into something harder than it was.
What the Correction Does and Does Not Change
The revised ratio changes nothing about the microbiome’s biological significance. Approximately 38 trillion bacteria, distributed across the gut, skin, mouth, and other surfaces, represent an enormous community of metabolic activity, and the evidence for its relevance to digestion, immune function, and other physiological processes does not depend on whether bacteria outnumber human cells by a factor of ten or by a factor of 1.3. The importance of those organisms is a question of function, not of enumeration.
Nor does this episode support any general conclusion about the unreliability of science. It is the story of one specific propagated error, traced to a single underspecified estimate, corrected in the ordinary course of scientific scrutiny. The system, eventually, worked.
What it does invite is a more careful habit of mind around inherited figures: the ones that appear in introductory paragraphs, that do the work of orienting readers without themselves being the subject of investigation. Those figures are rarely wrong in ways that matter for the science that follows them. But they can be wrong, and the willingness to check, to follow a number back to its source and ask whether the assumptions that generated it still hold, is not scepticism for its own sake. It is ordinary due diligence, applied to the kind of claim that usually escapes it.
The ten-to-one ratio had an exceptionally good run for an extrapolation from a 1972 gut sample calculation. Its correction is a small episode in the history of biology. But it is a useful one.
