The first thirty seconds: before you swallow

Digestion starts before food reaches your stomach and, remarkably, before it reaches your mouth.

The cephalic phase is the body's anticipatory response to the sight, smell, taste and chewing of food. It is not a theory; it has been measured directly. In one study, fifteen minutes of sham feeding (chewing food and spitting it out, never swallowing) produced a simultaneous stimulation of stomach and pancreatic secretion, antroduodenal movement, and pancreatic polypeptide release, lasting thirty minutes. Atropine, which blocks the relevant nerve signalling, abolished all of it.

A review estimates that anticipatory signalling accounts for more than half of the overall post-meal response. The companion article in this set on the first thirty minutes covers it in more detail.

Saliva meanwhile contains an enzyme that begins breaking down starches, and the tongue secretes a lipase that starts on fats and keeps working in the acidic stomach.

Hours 0 to 4: the stomach

The stomach is not a passive bag. It grinds, mixes, acidifies and, most importantly, meters food into the small intestine slowly, because the small intestine can only handle so much at once.

Stomach acid normally holds an empty stomach at a pH of about 1.5 to 2.0. That acid denatures protein, activates the enzyme pepsin from its inactive precursor, breaks down plant cell walls, and kills most microorganisms that arrive with food. A companion article covers the detail.

How long does a meal take to leave? It depends enormously on what the meal is.

In a study of around 73 healthy volunteers using a standardised test meal and gamma scintigraphy, the time for half the meal to leave the stomach was:

• Liquid portion: median 44 minutes, with most people between 28 and 78 minutes
• Solid portion: median 162 minutes

A separate study of 83 healthy adults using radiopaque markers found gastric emptying at a median of about 2.4 hours in men and 2.9 hours in women, with a 10th-to-90th-percentile range running from 0.7 to 4.9 hours. That is a nearly sevenfold spread among people who are all entirely healthy.

Notice also that this study found differences between men and women at every stage of the gut (stomach, small intestine and colon), and all three were statistically significant. It is a single-centre study using one technique, so treat it as one good measurement rather than a settled fact.

Hours 2 to 8: the small intestine

This is where nearly all absorption happens.

The pancreas delivers enzymes for carbohydrate, fat and protein, along with bicarbonate to neutralise the incoming acid. Bile from the liver and gallbladder emulsifies fat so the fat-splitting enzymes can reach it. The intestinal wall contributes its own enzymes for the final steps, the ones that break disaccharides such as lactose, maltose and sucrose into single sugars.

Small intestinal transit in that 83-person study ran to a median of about 3.2 hours in men and 4.4 hours in women, with the healthy range spanning roughly 1.5 to 11 hours.

A different technique (MRI in 26 healthy adults) measured the time for a meal to travel from mouth to the start of the colon at a median of 135 minutes.

That study contains a useful warning about a common test. A breath-hydrogen method, widely used for the same measurement, gave a systematically earlier result, and in 8 of 26 people (31%) the marker was still in the terminal ileum, not yet in the colon, when the breath reading suggested it had arrived. Different methods give different numbers, and they are not interchangeable.

One honest gap: you will often read that fat specifically slows gastric emptying, or that the stomach releases calories at a fixed rate. We could not retrieve a human source establishing either as a figure, so we are not stating one. The liquid-versus-solid contrast above is what the retrieved evidence supports.

Hours 8 to 60: the colon

By the time material reaches the colon, most of what was useful has been taken. What remains is water, fibre, bacteria, bile components and cellular debris.

The colon's jobs are to reclaim water and electrolytes, to host the fermentation of fibre, and to store the residue.

Colonic transit in that 83-person study ran to a median of about 1.3 days in men and 1.5 days in women, with the healthy range extending to 3.7 days in women and 1.9 days in men.

A larger study using an electromagnetic capsule in 111 healthy volunteers mapped where that time is spent. In women, roughly 32% in the ascending colon, 34% in the transverse, 17% descending and 17% rectosigmoid. In men, 33% / 25% / 14% / 28%. So the first half of the colon takes up the majority of the journey. That study also found colonic transit lengthens with age and that whole-gut transit was shorter with higher BMI. One of its authors is affiliated with the company that makes the measuring device, which is worth knowing.

Put end to end, a typical journey is roughly one to three days, and being at either end of that band is ordinary.

The variability nobody warns you about

Earlier work on this project found something worth repeating here. When the same healthy people were measured twice, a week apart, whole-gut transit correlated only moderately with itself, and the researchers concluded that the variability reflects genuine biological variability rather than measurement error.

Your own transit time is not a fixed personal constant. It moves, week to week, for reasons that are mostly internal.

That single finding explains a great deal of ordinary bodily confusion. If your digestion feels different this week than last, the most likely explanation is that it is different this week than last, and that this is what a normally functioning gut does.

What changes the pace

A few things have been measured, and they are mostly not the things people expect.

Sex. The clearest difference in the study that produced most of the figures above, and one that is rarely mentioned. Women were slower than men at every stage (stomach, small intestine and colon), and all three differences were statistically significant. The colonic difference was roughly 1.5 days against 1.3.

Age. Colonic and whole-gut transit lengthen with increasing age, in the 111-person capsule study.

Body mass. The same study found whole-gut transit was shorter with higher BMI, which is the opposite of what most people would guess.

What the meal is. Liquid and solid components of the same meal emptied at medians of 44 and 162 minutes respectively. That is the largest single effect in this article.

Not much else, reliably. The study that established the reference ranges for gastric emptying noted that no consistent effect of anthropometric measures on emptying was present. Height, weight and build did not predict it.

There is one more structural observation from the capsule study that is quietly interesting: colonic and whole-gut transit times cluster at intervals separated by roughly 24 hours, which the researchers took as evidence that the process is not continuous. The colon does not trickle steadily; it moves things along in bursts, on something close to a daily rhythm. That fits with what a companion article covers: colonic activity is minimal overnight and picks up sharply on waking.

What the numbers are for, and what they are not for

We have given specific figures because they are interesting and because vagueness invites people to fill the gap with worse information.

But they are not a self-assessment tool. Every figure here is a median with a wide healthy range, drawn from particular test meals and particular measurement techniques in particular groups of volunteers. There is no number in this article that makes anyone's digestion abnormal, and we are not going to suggest otherwise.

If something about your digestion has changed and stayed changed, that is a conversation with a doctor, who can ask the questions an article cannot.

A few things that follow, practically

Timing meals around the previous one. If half a solid meal is still in the stomach at nearly three hours, the feeling of not being hungry three hours after lunch is not a failure of appetite. It is accurate.

The overnight gap is when most of the colon work happens. Colonic transit is measured in days, so what you eat today is being processed well into tomorrow. There is no same-day relationship between a meal and what happens in the bathroom.

Chewing. The research is weaker than you would guess: small average effects on appetite, substantial inconsistency between studies, documented publication bias, and chewing-gum industry affiliations on the main meta-analyses. Nothing we found addresses whether it affects nutrient absorption. The companion article on the first thirty minutes gives the figures.

The system has enormous spare capacity. One figure captures this better than any other: clinically significant pancreatic enzyme insufficiency requires the loss of almost 90% of enzyme output. A healthy gut absorbs about 93% of the fat in the diet. That is not a system operating at its limit.

One thing the numbers cannot tell you

A closing caution about measurement itself, because it explains why figures from different sources disagree so much.

Every number in this article came from a particular technique. Gamma scintigraphy tracks a radiolabelled meal. Radiopaque markers are counted on X-rays. An electromagnetic capsule reports its own position. MRI images the gut directly. A breath test infers arrival in the colon from hydrogen production by bacteria.

These do not agree with each other. The MRI study measured mouth-to-colon transit at a median of 135 minutes; the breath test in the same people gave 127.5 minutes, and in nearly a third of participants the marker was still in the terminal ileum when the breath reading said it had arrived in the colon.

None of those techniques is wrong. They measure slightly different things, in slightly different ways, and they are not interchangeable. So if you encounter a transit figure somewhere that disagrees with one here, the explanation is usually the method rather than a factual dispute.

The same applies to test meals. The 44-minute and 162-minute half-emptying figures are for one standardised research meal. A different meal produces different numbers, and the researchers who published those figures say so explicitly.