Before anything else: look at it. Not sniff it — look.
Pink, orange, or red streaks, or any fuzzy growth in any colour, means throw it out. That's mould or bacterial contamination, it can't be fed back to health, and scraping off the top doesn't remove what's grown through the rest. This is the only failure mode on this page that's terminal, and it's rare.
Everything else — grey or black liquid, a hard crust, a smell that makes you recoil, complete absence of bubbles, separation into layers — is a hungry starter, not a dead one. Smell is a terrible diagnostic here. A badly neglected starter smells appalling and is usually fine.
-
Pour, scrape, and take a littleDay 1 · 5 minutes
Pour off the liquid on top and discard any dried crust around the rim. Then dig down and take a tablespoon from the middle — the centre of the mass is where the culture is in the best shape, away from the surface and the walls.
Put that spoonful in a clean jar and throw the rest away. You're not trying to rescue the volume; you're rescuing the organisms, and there are plenty in one spoonful.
Feed it generously — 1:5:5 by weight, so 10 g of old starter, 50 g of flour, 50 g of water at about 85°F. The big dilution matters here: it drops the acidity fast, and it's accumulated acid that's suppressing the yeast.
Use rye if you have it For the first two or three feeds, make it half rye. Rye carries more microbial life and more available minerals, and it wakes a struggling culture faster than white flour will. Switch back once it's rising. -
Feed again, expect nothingDay 2 · 5 minutes
Probably no visible activity, or a few sad bubbles. Feed again at 1:5:5, keep it warm — 78–82°F if you can manage it. Warmth does more here than any other single factor.
The smell should already be shifting from solvent toward something more like sour dairy. That change is the first real sign, and it usually arrives before any rise does.
-
Something happensDay 3 · 10 minutes
Bubbles, a modest rise, a cleaner smell. Move to twice-daily feeding at 1:2:2 now that there's a population worth feeding. Two feeds a day at a smaller ratio builds numbers faster than one big one.
If there's still absolutely nothing on day 3, don't give up — go back to § 01, confirm it isn't contaminated, and check the two things that stall a revival: temperature, and chlorinated water. Chloramine in particular doesn't off-gas overnight the way people assume.
-
Back to normalDay 4–5
You want the same standard as a new starter: doubles in 4–8 hours, domes, then flattens — three feeds running. Predictability, not vigour.
Don't bake with it before that. A revived starter that rises once impressively and then stalls will produce a dense loaf and make you think you'd failed at the bread, when the problem was upstream.
| Stuck on | Usually | Do |
|---|---|---|
| No activity by day 4 | Too cold | Find 78–82°F — oven with the light on works |
| No activity, kitchen is warm | Chloramine in the water | Switch to filtered or bottled. It doesn't off-gas |
| Rises then collapses fast | Still over-acidified | Bigger dilution, 1:10:10, twice daily |
| Smells like acetone throughout | Underfed for the temperature | Feed more often, not more |
| Any pink, orange, or fuzz | Contamination | Stop. Start a new one |
Notes from the counter
The thing worth internalising is how lopsided the odds are. A starter is a dense, acidic, competitive environment that has spent weeks or months actively excluding anything that isn't supposed to be there. That's exactly why it survives neglect that would ruin almost any other living thing in your kitchen.
So the honest advice on a jar you found at the back of the fridge with black liquid on it is: it's probably fine, it'll take four days, and the only thing that would make me throw it out is a colour that shouldn't be there.
Worth doing once you're back up and running: dry some of it as insurance. Dried starter keeps for years and takes ten minutes to make. It's the difference between a bad week and starting over.