I poured a hefeweizen for my brother-in-law last summer, handed it over with a two-finger head, and by the time he had finished telling me about his fence project the glass was flat on top. Ninety seconds, maybe less. The beer tasted right. It was carbonated — a hiss on opening, a proper rush of bubbles up the side — and it still went bald in the time it took to say hello. That gap between carbonation and foam is the thing I had misunderstood for years. Carbonation is how much gas is dissolved. Foam is whether anything in the beer can hold a bubble wall together once the gas comes out.
What a head is actually holding up
A bubble in beer is a thin film of liquid stretched around carbon dioxide. Left alone, that film drains downward, thins, and tears, and the bubble merges with its neighbor. Big bubbles then eat smaller ones, the head coarsens into a few soapy craters, and it collapses. Everything that improves head retention does one of two jobs: it makes the film more viscous and elastic so it drains slowly, or it cross-links the molecules in the film so the wall resists tearing.
That is why the fix is almost never more gas. Overcarbonating a beer with a weak film gives you a violent, short-lived foam and a glass that is two-thirds head on the pour and bare a minute later. I had been treating a structural problem as a pressure problem.
The proteins doing the lifting, by name
The molecules that build the wall come almost entirely from malt. John Palmer's rundown in Brew Your Own on the science behind beer foam names the three that matter and is worth reading slowly, because knowing which protein you are protecting changes what you do on brew day.
- Protein Z, an albumin around 40,000 in molecular weight, which Palmer describes as having the highest surface viscosity and elasticity of the group. This is the one that makes a film stubborn about draining.
- Lipid transfer protein 1, or LTP1, roughly 9,700 in molecular weight and produced while the barley germinates. It foams in its native state, stabilizes foam once the boil denatures it, and — usefully — binds up lipids that would otherwise wreck the film.
- Hordeins, the barley storage proteins, spanning a wide range of sizes. A variant near 23,000 concentrates in the head. Palmer notes some hordeins foam more readily than protein Z but give a less stable result, which matches the beer that throws a gorgeous head and loses it anyway.
Two other ingredients act as the glue. Isomerized alpha acids — the same bitter compounds I spent an entire batch learning to calculate — hydrogen-bond to those proteins and cross-link them, with isohumulone better at it than isocohumulone. And divalent metal ions help: Palmer cites zinc additions as small as 2 ppm producing significant improvement. A very pale, very lightly hopped beer is working with less glue than a bitter one, which is part of why a well-made pilsner is harder to keep a head on than an IPA.
The foam killers, ranked by how often they are your fault
Lipids are first and the margin is not close. Fats and oils are strongly hydrophobic; they punch into the bubble film, rupture it, and the bubbles coalesce almost instantly. The dose required is tiny. A fingerprint of butter on the rim of a glass, a splash of milk left in a rinsed pitcher, a bag of crisps eaten over the beer — any of those will strip a head in seconds, and none of them are a brewing mistake at all.
Second is surfactant residue. Detergents and some sanitizers are designed to reduce surface tension, which is precisely the property foam needs. A glass rinsed but not fully cleared of dish soap will pour a good head and then drop it. This is a different failure from the one my sanitizing routine is meant to prevent: the sanitizer is doing its job, and the residue is still eating the foam, so serving glassware gets a hot rinse and nothing else.
Third is the beer itself running short on protein. Corn, rice, and plain sugar contribute fermentables and no foam-positive protein, so every point of gravity from them dilutes what the malt brought. Overlong boils and concentrated extract boils cause similar trouble, which is why Palmer recommends the late-extract method over boiling a thick wort and topping up.
A beer-clean glass, and the ten-second test
Both the AHA and Palmer land on glassware as the piece homebrewers underrate, and it is the only variable you can fix after the beer is already in the bottle. Oil and grease occupy the surface where bubbles want to form, so a film you cannot see reads as a beer problem. The test is quick: fill the glass with water, dump it, and watch the inside wall. Water sheeting off in an even film means the glass is clean. Water beading into droplets, or bubbles clinging in a ring at the waterline, means there is something on the glass.
My routine now is a hot rinse with a soft brush and no soap, air-dried upside down on a rack rather than towel-dried, since a kitchen towel carries oil from everything else it has touched. Tall, narrow glasses also help, both by limiting the surface where gas escapes and by giving the head walls to lean on. And serve the beer cold: film viscosity rises as temperature falls, so the same beer at 40°F holds a head longer than at 55°F.
Yeast that eats your foam after the fact
The detail that finally explained my hefeweizen is one I would never have guessed. Stressed yeast excrete proteinase A, an enzyme that degrades proteins in exactly the 10,000-dalton neighborhood where LTP1 lives — and Palmer points out it keeps being secreted after fermentation finishes and stays active in the packaged beer. A batch left a long time on the trub at a warm temperature is a batch whose foam proteins are being quietly digested while it sits.
That reframes several habits I already had for other reasons. Pitching enough healthy yeast keeps the culture from getting stressed in the first place, which is the same argument for making a starter on a big batch. Getting beer off the yeast cake within a reasonable window and then keeping it cold slows the enzyme down. And a bottle-conditioned beer that spent four months warm in a cupboard pours thinner-headed than the same beer stored cold, whatever its gas level.
The order I check things in now
When a beer of mine goes bald, I work down this list rather than reaching for the priming sugar:
- Pour the same beer into a freshly hot-rinsed glass. If the head holds, the beer was never the problem and the dishwasher was.
- Check serving temperature. Warm beer sheds foam, and a glass straight from a hot rinse is a warm glass.
- Look at the grain bill for foam-positive malt. Ten percent wheat malt, or a half pound of a dextrin malt such as Carapils, is the cheapest structural fix available and does not change the flavor much.
- Count the sugar. Every pound of corn sugar or honey is gravity with no foam protein attached, so a beer built on adjuncts needs the malt side carrying more.
- Check how long the batch sat warm on the yeast, and whether the bottles have been stored cool. This is the slow, invisible one.
- Only then look at carbonation, and look at it as a separate question from foam. Under-gassed beer cannot make a head, but over-gassed beer with a weak film makes a head that will not last.
The hefeweizen got a rebrew: twenty percent more wheat malt, a mash at 156°F instead of 150°F, a rack out of the fermenter on day twelve instead of day twenty-five, and a tall glass that had never met dish soap. The head lasted the whole fence conversation, which is the only benchmark I care about.