Why white shirts go yellow under the arms

You sweated more this summer than you meant to, and your shirts are only telling you about it now.

It was that kind of summer nearly everywhere. June was the second-hottest June ever recorded, July tied for the second-hottest July, and in both months the sea was warmer than it has ever been for the time of year.

None of that stained anything, though. Heat does not mark a shirt. What marks a shirt is something you put on yourself every morning on purpose, and the reason nobody blames it is that everybody calls the result by the wrong name.

This is not me being fussy about a word. Getting it right changes what you should do about the shirt, and it rules out the first thing most people reach for.

It is not your sweat

Nearly all the sweat you make is a clear fluid with no smell and no colour in it. There is nothing in it to leave a mark.

Coloured sweat does exist, and it is rare enough to be a diagnosis rather than a laundry problem. It is called chromhidrosis, which you say as krome-hih-DROH-sis, it was first described in 1709, and it turns sweat black, blue, green, brown or yellow, either because the body is making a pigment or because of a dye or a drug that has been swallowed. There is a second version where the sweat leaves you clear and gets its colour afterwards, from bacteria living on the skin.

If you have neither, and almost nobody does, then whatever is on the shirt did not leave you that colour. It turned yellow afterwards, on the cloth, and something else did it.

What it actually is

Sweat is not just water. About a quarter of it is sebum, the oil your skin makes, and that is the half that matters here.

Antiperspirant is built to meet it. The aluminium in the stick reacts with that oil and the two of them turn into a gel, and the thing to know about that gel is that water barely touches it.

On your skin, that gel is the product doing its job. It settles into the opening of a sweat gland and plugs it, which is the whole difference between an antiperspirant and a deodorant: one stops the sweat, the other just handles the smell. On your shirt, it is exactly the same gel. Not a side effect of the product working. It is the product working, on cloth instead of on skin.

Which is also why washing does so little. Something that will not dissolve on your skin is not going to dissolve in a machine either.

A team of textile researchers proved that the hard way. They made the stain on purpose in a lab, with real antiperspirant and a stand-in for sweat, and then put it through eight different wash routines, changing the temperature, the amount of detergent, and what they treated it with first. Eight. So if you have tried a hot wash on this and been disappointed, that is the expected result and not something you did wrong.

It takes a season, not a day

The other half of the mechanism is that nothing happens quickly. Cotton Incorporated's technical bulletin on why fabrics yellow gives the underarm as its worked example: the build-up of aluminium chloride from certain deodorants embeds into the fabric, is "very difficult to remove in laundering", and "over a period of time builds to a level that severely affects the garment".

So the shirt was not ruined on the hot day in July. The hot day was one morning out of about ninety, and every one of them left a little behind.

Nothing happened on the day you noticed it. That was just the day it finally got dark enough to see.

The dryer is what makes it permanent

Here the advice becomes simple and almost nobody follows it. Texas A&M's extension service puts it in four words: heat sets stains. Do not put a stained garment in the dryer, and do not iron it, until you are sure the mark has gone. Cotton Incorporated adds the other half from the opposite direction: over-drying a laundered garment can yellow it on its own, with no help from anybody's armpit.

Which means the machine most likely to have set this permanently is the one people think of as the neutral step at the end. It is the argument this guide keeps arriving at from different directions, most fully in the piece on what happens between washes: the wash is the smallest part of looking after clothes, and what surrounds it decides more.

And bleach is on the list of causes

The advice you will find first, on page after page written by people selling undershirts or deodorant or stain remover, is chlorine bleach. The one technical source in this field puts chlorine among the causes of yellowing rather than the cures for it. Cotton Incorporated calls it the "unseen assailant" and describes it being held in the cloth by the finishes put there at the mill, building up slowly, and yellowing it in the end.

Those same pages offer a specific reason why bleach makes underarm marks worse, which is that it reacts with what is already in the cloth and locks it in harder. It may well be right. It has no source outside pages selling something, so this piece will not assert it. What can be said without inventing anything is narrower and quite enough: the substance being recommended to you as the fix is independently documented as a cause of the problem.

What that leaves you

Not a trick, which is the honest answer and the reason most pages on this subject are longer than this one. Four things follow from the evidence rather than from anybody's product.

Treat it early, because the same extension service notes that old stains may be impossible to remove, and this one is defined by accumulation. Keep heat away from it, the dryer and the iron both, until the mark has actually gone rather than faded. Do not reach for chlorine. And do not expect an ordinary wash to do much, because the people who study this had to design a procedure to beat one.

Worth knowing that even the reference tables agree it is a strange one. In the extension service's stain chart, perspiration and deodorant are not filed under protein or oil or dye with everything else. They sit under Special, next to rust and scorch, which is the tidy version of everything above.

The prevention that follows from the mechanism is not something I found tested, so take it as reasoning rather than as a finding: less product, and letting it dry before the shirt goes on, gives the gel less to form from and less opportunity to transfer. The rest of it is the same answer as always. The mark is made by a product doing its job, fixed in place by heat, and made permanent by the thing you reached for to remove it.