How Many Cells in the Human Body? About Half Are Bacteria

Bench card

Question

How many bacteria live in and on a human body, compared with human cells?

Answer range

about 3.8×10¹³ bacteria and 3.0×10¹³ human cells in a 70 kg reference man, a ratio near 1.3 (estimate uncertainty about 25 %); a second route on this page gives 4×10¹² to 3×10¹³ bacteria

Method

Sender 2016 chain recomputed with units, plus an independent route from stool solids; human count checked against Hatton 2023

Inputs

10 (4 for the colon route, 4 for the stool route, 2 for the human red-cell count)Source: inputs table and CSV download on this page

Weakest input

colon content volume (0.4 L, standard error 17 %) and the assumption that bacteria per gram of stool stands for the whole colon

Key source

Sender, Fuchs & Milo 2016, PLoS Biology

how-many-bacteria-in-the-human-body-bench.csv (CSV) · where each value comes from, and its license (JSON)

Data credits for this card

Data: Sender R, Fuchs S, Milo R 2016, PLoS Biology 14(8):e1002533, CC BY 4.0, https://doi.org/10.1371/journal.pbio.1002533 (retrieved 2026-10-24); Rose C, Parker A, Jefferson B, Cartmell E 2015, Critical Reviews in Environmental Science and Technology 45(17):1827-1879, CC BY 3.0, https://doi.org/10.1080/10643389.2014.1000761 (retrieved 2026-10-24); Hatton IA et al. 2023, Proceedings of the National Academy of Sciences 120(39):e2303077120, CC BY 4.0, https://doi.org/10.1073/pnas.2303077120 (retrieved 2026-10-24); answer range and input count computed 2026-10-24 for this page

The ratio has a long history and a short arithmetic. In 2016 Ron Sender, Shai Fuchs and Ron Milo published a recount in PLoS Biology that lays out every input, its source and its uncertainty. This page is built on their figures. Where it recomputes something, it says computed; where it has only a bibliographic record of a paper, it says so.

How many bacteria, and how many human cells?

The two headline numbers belong to one specific body. That body is the "reference man" of radiation-protection tables: 20–30 years old, 70 kg, 170 cm tall. For him, Sender and colleagues estimate about 3.8×10¹³ bacteria and about 3.0×10¹³ human cells, a bacteria-to-human ratio of about 1.3.

They put the uncertainty of that ratio at about 25% and the spread among 70 kg men at about 53%. The two percentages measure different things. One is the estimate's error; the other, the variation between people. Computed for this page, 3.8×10¹³ out of 6.8×10¹³ is 56%, which is the "about half" in the title.

Almost all of the bacteria are in one place. Sender and colleagues multiply measured bacterial concentrations by organ volumes to get order-of-magnitude upper bounds. The colon comes to 10¹⁴ at most, dental plaque to 10¹², and saliva, skin and the lower small intestine to 10¹¹ each. The stomach and upper small intestine reach only about 10⁷.

Illustration

In the stomach and upper gut, bacteria number 10³–10⁴ per millilitre. Between them, the two hold fewer bacteria than a single millilitre of the colon's contents, which run near 10¹¹ per millilitre. Every organ outside the colon together adds at most about 10¹², under 4% of the colon's share.

Slide the diagram sideways to read its labels, or open it full size.

The same lopsided picture appears in a drop of seawater, where plankton counts are dominated by the smallest cells. It is why a log axis is the only honest way to draw the chart above. On an ordinary ruler scaled to 10¹⁴, the stomach's 10⁷ would sit a ten-millionth of the way along, pressed against zero.

Where did ten to one come from?

Sender and colleagues traced every paper they could find that gave a source for the number of gut bacteria back to a single sentence. In 1972 Luckey assumed about 10¹¹ bacteria per gram of gut content and multiplied by 1 litre, about 1 kg, of alimentary tract. That gave 10¹⁴.

This page has the 1972 paper only as a bibliographic record. The derivation above is as Sender and colleagues describe it. Savage's 1977 review repeated the 10¹⁴ figure (bibliographic record); set against a human count of roughly 10¹³, it gave the ten-to-one ratio that Sender and colleagues date to the 1970s.

A common headline version of the claim says the body is 10 times more bacterial than human, or carries 1–3 kg of bacteria. For the 1–3% body-mass version, Sender and colleagues cite a 2012 news release from the US National Institutes of Health (bibliographic record only). By their account, a 2014 letter in Microbe by Rosner questioned the ratio, and in 2016 the recount followed, first as a letter in Cell and then as the full PLoS Biology paper; the Cell letter is a bibliographic record here.

Slide the diagram sideways to read its labels, or open it full size.

What changed was not the density but the volume. The 10¹¹ bacteria per gram of the 1972 sentence has held up well. The recount removed the assumption that a whole litre of gut content carries that density, when only the colon does.

Route one: counting the colon

The recount multiplies three numbers, and each carries a unit. The volume of colon content comes from imaging studies of the undisturbed colon, normalised to a 1.70 m man. It is about 0.4 L, with a standard error of 17% and a spread of 25% between men. The reference-man tables put its density at 1.04 g/mL, so the colon holds about 416 g of content.

For bacterial density, the recount turns to stool. The figure is the geometric mean of 14 stool studies from 1966 to 2008, converted to a common wet-weight basis. It is 9.2×10¹⁰ cells per gram, with a standard error of 19% and a spread of 46% between people.

Multiplied through with units carried, 416 g × 9.2×10¹⁰ cells/g gives 3.8×10¹³ cells, computed for this page and matching Sender and colleagues' figure. Combining the two standard errors in quadrature gives about 25%, so one standard error spans 2.9×10¹³ to 4.8×10¹³. The same unit discipline that keeps a water strider estimate honest applies here: a litre has to become grams before it can meet cells per gram.

Slide the diagram sideways to read its labels, or open it full size.

The weakest input is colon volume. The imaging studies behind 0.4 L were few; one reported a volume that included an unmeasured amount of gas, and the most detailed had to estimate the stool fraction, about 70%, to remove it. Sender and colleagues also note that using stool density for the whole colon is probably an upper limit. Bacterial concentration climbs from about 10⁸ per gram where the small intestine empties into the colon to about 10¹¹ per gram at the exit.

Route two: the same count another way

A second route should not reuse the colon volume or the bacteria-per-gram figure. So this one starts from what leaves the body. A review of 60 studies of healthy adults puts the median dry mass of stool at 29 g per day; study means ranged from 12 to 81 g. The same review puts bacterial biomass, living and dead, at 25–54% of those dry solids (Rose and colleagues 2015).

Among the sources they cite for that fraction is Stephen and Cummings 1980, a bibliographic record here. An average gut bacterium has a dry mass of about 1–2 pg (Sender and colleagues 2016). Sender and colleagues quote colonic transit, the time content spends in the colon, at a normal 25–40 hours.

Computed for this page: 29 g/day × 0.40 ÷ 1.5 pg per cell gives about 7.7×10¹² bacteria leaving per day. A residence time of 32.5 hours turns that flow into a standing population of about 1.0×10¹³. Holding stool output at the 29 g median and pushing the other three inputs to their favourable ends gives 2.6×10¹³; pushing them the other way gives 3.8×10¹².

At 3.8×10¹³, route one is about 3.7 times the central value of route two, within the factor of ten this site requires before it publishes an estimate.

Slide the diagram sideways to read its labels, or open it full size.

Route two comes in lower mostly because stool output times transit time under-fills the colon. At 29 g of dry solids a day for about 1.4 days, the colon would hold about 40 g of solids. Yet 416 g of content at 27% dry matter, the figure Sender and colleagues use, is about 110 g of dry solids. They met the same gap when they checked the colon volume against stool flow and got 150–250 mL instead of 400 mL.

Both routes support one conclusion: a body holds on the order of 10¹³ bacteria, somewhere between 1 and 4×10¹³. The "about half" of the title rests on the better-documented colon route.

Why most of your cells are red blood cells

The human count has its own two routes. By Sender and colleagues' account, Bianconi and colleagues summed 56 categories of cell in 2013 and reached 3.7×10¹³ with a standard deviation of 0.8×10¹³; the 2013 paper is a bibliographic record here. Sender and colleagues re-examined the six largest categories, revised four of them, and reached 3.0×10¹³. In 2023 Ian Hatton and colleagues rebuilt the count from 1,264 cell groups across 60 tissues.

They reported about 36 trillion cells in a 70 kg male, about 28 trillion in a 60 kg female and about 17 trillion in a 32 kg ten-year-old. The three compilations agree to within about a quarter.

In all three, red blood cells dominate. Blood volume of 4.9 L times a red cell count of 5.0×10¹² per litre gives 2.5×10¹³ red blood cells, computed for this page; Sender and colleagues put red blood cells at 84% of their total. That red cell count is the one on a blood test, where the normal range for men is 4.6–6.1×10¹² per litre.

Red blood cells and platelets carry no nucleus, so nearly 90% of your cells, about 26×10¹², have none. That matters for the length of DNA per cell: most cells have no DNA to measure. Only about 3×10¹² human cells are nucleated.

Slide the diagram sideways to read its labels, or open it full size.

Two revisions in the Sender count show how a convention can inflate a number. Glial cells were long put at ten times the number of neurons, about 3×10¹². A 2009 count found about 86 billion neurons and about 85 billion non-neuronal cells in the adult male brain (Azevedo and colleagues 2009, abstract). That ratio is near 1 to 1, so the glial figure fell to 8.5×10¹⁰.

Dermal fibroblasts fell from 1.85×10¹² to 2.6×10¹⁰, a factor of about 70, once the thickness of the dermis entered the calculation. Compare the bacteria to nucleated cells only and the ratio returns to about 10 to 1. As Sender and colleagues point out, that is a matter of definition rather than measurement.

How much do all those bacteria weigh?

About 0.2 kg, wet. Half of the 0.4 kg of colon content is bacterial mass, by the fraction Sender and colleagues take from Stephen and Cummings 1980 and a 1984 study of bacterial dry matter. The same 0.2 kg comes back from the other side: 3.8×10¹³ cells at about 5 pg each. That wet 0.2 kg is about 0.3% of a 70 kg body, against the 1–3 kg of the headline claim. Dry, the body's bacteria amount to about 50–100 g.

Set the two scales side by side and the body looks different on each. By count, bacteria slightly outnumber human cells. By mass, human cells total about 46 kg once the 24 kg or so of extracellular water and solids are removed. Three quarters of that mass sits in fat and muscle cells. Those cells are usually over 10,000 µm³ apiece and make up about 0.2% of the count.

A red blood cell, at about 100 µm³, is among the smallest cells in the body. A full table of cell sizes is the job of How Big Is a Cell. Here the point is that a count and a mass of the same body answer different questions.

Slide the diagram sideways to read its labels, or open it full size.

How big is one gut bacterium?

The average gut bacterium behind these figures weighs about 5 pg wet. It is several times heavier than the 1 pg, 1 µm³ bacterium of textbook estimates. In culture, Escherichia coli cells measured 1.0–1.5 µm wide and 1.6–3.9 µm long, shortest when starved and longest when growing fast on rich medium (Volkmer and Heinemann 2011, 22 growth conditions).

A cell of that size swimming in water at 20 °C at about 30 µm/s sits far below the syrup line, where viscosity wins over inertia. It is also the size of the 1 µm bead that never sits still in Brownian motion. A baker's yeast cell is a few times larger; diatoms, glass-walled cells mostly 20–200 µm, sit one to two rungs further up the ladder.

Slide the diagram sideways to read its labels, or open it full size.

Every bacterial figure on this page leans on one unmeasured thing: the density of bacteria along the colon itself, rather than in the stool that leaves it. Concentration rises a thousandfold from the start of the colon to the exit, through water absorption, growth during the 25–40 hour transit and shedding from the gut wall. As of the 2016 recount, that gradient had not been measured in a living person well enough to replace the stool proxy. Until someone does, the count sits at about 3.8×10¹³, give or take a quarter, with a second, cruder route pointing lower.

Sources

1. Sender R, Fuchs S, Milo R. "Revised Estimates for the Number of Human and Bacteria Cells in the Body." PLoS Biology 14(8):e1002533 (2016). https://doi.org/10.1371/journal.pbio.1002533 Read: full text. Text license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).
2. Hatton IA, Galbraith ED, Merleau NSC, Miettinen TP, Smith BM, Shander JA. "The human cell count and size distribution." Proceedings of the National Academy of Sciences 120(39):e2303077120 (2023). https://doi.org/10.1073/pnas.2303077120 Read: full text. Text license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).
3. Rose C, Parker A, Jefferson B, Cartmell E. "The Characterization of Feces and Urine: A Review of the Literature to Inform Advanced Treatment Technology." Critical Reviews in Environmental Science and Technology 45(17):1827–1879 (2015). https://doi.org/10.1080/10643389.2014.1000761 Read: full text. Text license: CC BY 3.0 (https://creativecommons.org/licenses/by/3.0/).
4. Volkmer B, Heinemann M. "Condition-Dependent Cell Volume and Concentration of Escherichia coli to Facilitate Data Conversion for Systems Biology Modeling." PLoS ONE 6(7):e23126 (2011). https://doi.org/10.1371/journal.pone.0023126 Read: full text. Text license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).
5. Bianconi E, Piovesan A, Facchin F, Beraudi A, Casadei R, Frabetti F, et al. "An estimation of the number of cells in the human body." Annals of Human Biology 40(6):463–471 (2013). https://doi.org/10.3109/03014460.2013.807878 Read: bibliographic record only.
6. Luckey TD. "Introduction to intestinal microecology." The American Journal of Clinical Nutrition 25(12):1292–1294 (1972). https://doi.org/10.1093/ajcn/25.12.1292 Read: bibliographic record only.
7. Savage DC. "Microbial ecology of the gastrointestinal tract." Annual Review of Microbiology 31:107–133 (1977). https://doi.org/10.1146/annurev.mi.31.100177.000543 Read: bibliographic record only.
8. Stephen AM, Cummings JH. "The microbial contribution to human faecal mass." Journal of Medical Microbiology 13(1):45–56 (1980). https://doi.org/10.1099/00222615-13-1-45 Read: bibliographic record only.
9. Sender R, Fuchs S, Milo R. "Are We Really Vastly Outnumbered? Revisiting the Ratio of Bacterial to Host Cells in Humans." Cell 164(3):337–340 (2016). https://doi.org/10.1016/j.cell.2016.01.013 Read: bibliographic record only.
10. Cummings JH, Bingham SA, Heaton KW, Eastwood MA. "Fecal weight, colon cancer risk, and dietary intake of nonstarch polysaccharides (dietary fiber)." Gastroenterology 103(6):1783–1789 (1992). https://doi.org/10.1016/0016-5085(92)91435-7 Read: bibliographic record only.
11. Azevedo FAC, Carvalho LRB, Grinberg LT, Farfel JM, Ferretti REL, Leite REP, et al. "Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain." Journal of Comparative Neurology 513(5):532–541 (2009). https://doi.org/10.1002/cne.21974 Read: abstract.
12. National Institutes of Health. "NIH Human Microbiome Project defines normal bacterial makeup of the body." News release, 2012. https://www.nih.gov/news-events/news-releases/nih-human-microbiome-project-defines-normal-bacterial-makeup-body Read: bibliographic record only.

Cite this page

Ants, Cells & Atoms. “How Many Cells in the Human Body? About Half Are Bacteria.” Published 2026-10-25. https://antscellsandatoms.com/scale/how-many-bacteria-in-the-human-body/. Accessed (add the date you read it).

No DOI is issued for entries. Figures made here are CC BY 4.0; see each caption for the license of photographs.

How this entry is kept

Published

25 October 2026

Last substantive revision

none since publication

• How entries are checked
• Corrected entries
• Contact and feedback
• Report an error in this entry

Related entries

Scale & Estimates · Mechanism explainer · [10⁻⁴–10⁻³ m]

Reynolds Number: Why a Bacterium Can't Coast and a Fish Can

Image: “Oczlik (Cyclops) - samica” by Mikron86 (2025) · CC BY 4.0

"Oczlik (Cyclops) - samica" by Mikron86 — https://commons.wikimedia.org/wiki/File:Oczlik_(Cyclops)_-_samica.jpg — CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)

Scale & Estimates · Scale estimate · [10⁻⁷–10⁻⁴ m]

How Big Is a Cell? Most Are a Hair's Width or Far Narrower

Image: “SEM blood cells” by Bruce Wetzel and Harry Schaefer, National Cancer Institute (1982) · Public domain

"SEM blood cells" by Bruce Wetzel and Harry Schaefer, National Cancer Institute (1982) — https://commons.wikimedia.org/wiki/File:SEM_blood_cells.jpg — Public domain (work of United States federal employees at the National Cancer Institute (NIH), public domain in the United States under 17 U.S.C. §105; Commons tag PD-USGov; the NCI Visuals Online source states that all images are in the public domain and may be reproduced without permission, with credit to the listed source or author; dated February 1982; used in the EU on the source's explicit public-domain statement)

Illustration

Scale & Estimates · Scale estimate · [10⁻¹⁰ m]

How Small Is an Atom? For Most Elements, About 0.1–0.5 nm

Image: Illustration: Ants, Cells & Atoms, all rights reserved

Illustration: Ants, Cells & Atoms. Generated schematic illustration — not a photograph, micrograph or specimen record. © Ants, Cells & Atoms, all rights reserved.