During a lecture at the National Institutes of Health in 2005 in the leadup to the landmark Human Microbiome Project, David Relman, a microbiologist at Stanford, explained that humans were made up of “communities.” By peering deep into what he called “inner space,” scientists discovered that we were never alone: Our bodies have 10 trillion cells, but we are host to 100 trillion microbes. “In other words,” Relman said, “we are ten parts microbe, and one part human. We are clearly outnumbered.” The figure has been repeated by researchers in scientific journals, books, and TED talks. Scientists joke that it is practically illegal not to cite some version of the figure in a presentation on the human microbiome.
Ten parts microbe, one part man is a humbling demonstration of biological magnitude. The research has redefined what it means to be human and reflects one of the great decentralizations of 20th-century biology: We wouldn’t be us without them. The biological is the new astronomical; the deeper we probe, the more we see. And the ratio of them to us is overwhelming.
There’s only one problem: The oft-cited 10-to-1 figure is almost certainly inaccurate. It’s a crude estimate from 1972 that established itself as a fact through repetition and generations of citations. As such, it reveals the staying power of convenient falsehoods—and how an obscure scientist who left a peculiar legacy has been cited by top researchers for decades.
Earlier this year, Judah L. Rosner, a researcher at the National Institutes of Health, delved into the apparent origins of the figure. He had long harbored doubts and, in a page-long letter published in Microbe Magazine, he persuasively dismantled the ratio. Recent estimates, he pointed out, suggest the human gut contains between 30 trillion and 400 trillion microorganisms, whereas the human body has an estimated 37 trillion cells—with a considerable range that goes from 5 to 724 trillion. Based on these approximations, the human body could have nearly the same number of cells as microbes or, at the more extreme end, nonhuman cells might outnumber our own almost 100 to 1.
To Rosner, the many repetitions of the assertion stem from laziness. Either that, he said, or scientists were so wowed by the dramatic ratio that they had suspended disbelief. “Science tries to give best estimates, so simply quoting an erroneous estimate is like saying the moon is made of green cheese,” he told me in an e-mail. Facts get updated all the time. This one had not.
Where did the figure originate? The long trail of citations dead-ends in a curiosity: a paper by Thomas D. Luckey, a microbiologist who made his calculations in 1972 based on 1 gram of feces. The paper was presented as part of a visionary symposium he convened that helped set the field of intestinal microecology in motion.
Luckey, a biochemist who spent much of his career at the University of Missouri, did far more in his unusual life than set this fact in motion. In a autobiography from 2008, he’s shown wearing a white samurai suit. Sir Samurai T. D. Luckey, as he came to call himself, explained that he’d been a farm boy, an expert milker of rattlesnakes, a doll collector, and the first scientist ever to demonstrate that farm animals grew faster when given low doses of antibiotics—a practice that has since become widespread in agriculture. Later in life, he studied radiation and drew intense criticism for a radical belief that low doses were good for health. (For years, he slept next to a jagged brown rock flecked with uranium-92.) When he died at the age of 94 earlier this year, he had faded into relative obscurity; the sole obituary, in Missouri’s Columbia Tribune, makes little mention of his extraordinary career. But Luckey’s guess—ten parts microbe, one part man—nonetheless persevered.
Within the last year or so, some microbiologists have gingerly backed away from the early estimate. In a 2014 report issued by the American Society of Microbiology, the ratio of bacteria to human cells is said to be closer to 3 to 1. Peter Turnbaugh, a microbiologist at the University of California San Francisco who was on the ASM report’s steering committee, said that determining absolute numbers is a challenge. Simply put, there’s a tremendous diversity and abundance. “The most important thing is that much of what makes us human—many of important aspects of health and the predisposition to disease and recovery—depends on metabolic activity of these microbes,” he told me. “I don’t think it’s all about numbers.”
Martin J. Blaser, a doctor and researcher who is author of the recent book “Missing Microbes” and a contributor to the NIH’s Human Microbiome Project, told me in an e-mail, “We don’t yet know the exact number and it will be a range. There are other estimates that it is 3:1. Everyone has other research priorities, so as far as I know, no one has (yet) gone back to actually do the study using modern methods to provide a more up-to-date estimate.”
The “‘fake’ fact,” as Rosner calls it, still circulates in part because it is easy to find—a lot easier than it is to turn up the original citation. That problem plagues citation practices, according to Ole Bjorn Rekdal, a social anthropologist in Bergen, Norway. Rekdal believes misleading or outdated references perpetuate these types of academic urban legends. “At times,” he wrote in one recent paper, “I get the feeling that references have been placed in quantities and with a degree of precision reminiscent of last minute oregano flakes being sprinkled over a pizza on the way to the oven.”
Ten parts microbe and one part man vividly captures our imagination. Even if the estimate is off, it is innocuous—certainly it has no obvious negative consequences, nor is it a result of outright deception in the sciences, as in the case of debunked research connecting vaccines and autism. Perhaps the crude estimate endures because it serves the practical purpose of astonishing those who hear it, in the same way that bogus Martian canals inspired a greater curiosity about the solar system, or the myth that all humans only access 10 percent of their brains might foster a greater appreciation for neuroscience.
Rather than trumpet how great we are, in this case, the insistence is on being outnumbered, as if we we’re not as human as, maybe, we are. After all these years, we’ve been won over by the microbial perspective. As Luckey wrote in 1972, “If we abandon anthropomorphism for the microbic view, we must admire the efficiency of these microbes in using man as a vehicle to further their own cause.”
Peter Andrey Smith is a reporter based in Brooklyn.
Correction: Due to a reporting error, an earlier version of this story neglected to state the full range of estimates of the number of cells in the human body; estimates span from 5 trillion to 724 trillion.