When the fuel runs out
Gut bacteria do not go dormant when the fibre runs out. They switch to the next available food, and in the distal colon the next available food is protein. That switch changes what gets made.
The switch
Bacteria in the colon prefer carbohydrate. Given fermentable fibre, that is what they use. When it is gone, they turn to what is left: protein that escaped digestion in the small intestine, plus protein from mucus and from shed cells of your own lining.
Fermenting protein is sometimes called putrefaction, an old word that has survived because it is accurate about the products. Carbohydrate fermentation yields short-chain fatty acids. Protein fermentation yields ammonia, hydrogen sulfide, phenols and p-cresol, indoles, and branched-chain fatty acids (Smith and Macfarlane, 1996; Windey et al., 2012).
Several of those are studied for effects on the colonic lining — hydrogen sulfide and p-cresol in particular — at the concentrations reached in the gut. This is not a claim that they are toxins in any simple sense. Some have normal physiological roles. It is a claim about direction: as carbohydrate falls, the metabolite profile shifts, and it shifts towards compounds that are harder on the epithelium rather than easier.
Where the switch happens
Predictably, in the distal colon. Model systems of the human colon show carbohydrate becoming limiting in the distal compartments, with proteolytic activity rising there as it does (Macfarlane et al., 1992). The pH gradient in the previous chapter is a readout of the same thing. Carbohydrate fermentation acidifies; as it falls off, pH drifts up, and a less acidic environment suits the protein-fermenting organisms better.
So the two processes are not independent. Carbohydrate fermentation actively suppresses the protein route by outcompeting it and by holding pH down. The question is not whether protein fermentation happens — some always does — but how far down the colon carbohydrate keeps the upper hand.
What acidity does besides suppress
A mildly acidic colon has other consequences worth knowing.
It shapes who grows. In controlled fermentation studies, pH near 5.5 favoured butyrate-producing bacteria, while a shift to 6.5 favoured Bacteroides instead (Walker et al., 2005). Small pH differences change the community, not just its output.
It restrains secondary bile acid formation. A small group of gut bacteria converts primary bile acids into secondary ones, including deoxycholic acid, and this conversion is sensitive to pH — it proceeds poorly in acidic conditions. Secondary bile acids are among the more studied outputs of the distal microbiome, and at high concentrations they are an irritant to the lining.
It assists mineral uptake. Some absorption of calcium and magnesium occurs in the colon, and it is improved by fermentation-driven acidification. The effect is real but modest, and most of your mineral absorption happens further upstream.
Go deeper: what the evidence does and does not establish
The mechanisms above are well described. The step that is not settled is the one people most want: whether deliberately extending carbohydrate fermentation further along the colon makes a measurable difference to how a person feels or fares.
What exists is mechanism, plus a large body of observational evidence that people eating more fibre from whole foods tend to do better on a range of long-term measures. Those are associations drawn from populations followed over years.
Note carefully what that is not. It is not a trial of a fibre supplement, and it does not isolate the mechanism described here. The mechanism is a plausible and well-supported explanation for the association. It is not the proof of it.
Anyone telling you a fibre prevents something is going further than the evidence goes. The honest position is that the mechanism is sound, the epidemiology points the same way, and the causal trial has not been run.
Bacteria that run out of fibre switch to protein, and the products change from short-chain fatty acids to ammonia, sulfides and phenols. Carbohydrate fermentation suppresses that switch — for as far along the colon as it lasts.
What to hold on to
This is the argument for fermentation reaching the whole colon, and it does not depend on butyrate being higher distally. It depends on something simpler: for as long as carbohydrate fermentation continues, the protein route stays suppressed, and where it stops, the other one takes over.
Which sets up Part IV. No single fibre ferments across that whole distance at a useful rate. So what would a fibre chosen for the whole path actually look like?