Case Study 1 — Maya Bakes Her First Sourdough

The Setup

Maya Okonkwo has been keeping a sourdough starter for fourteen weeks. The starter lives in a one-quart glass jar in the bottom drawer of her refrigerator. It rises and falls predictably between feedings. It passes the float test. By every objective measure, it is ready to bake.

She, by her own admission, is not.

What Maya has been doing for fourteen weeks is reading. She has read forum posts. She has watched videos. She has bought a book by an Oakland baker that is full of technical terms — autolyse, bulk fermentation, bench rest, cold retard, coil fold, poke test, Dutch oven preheat at 500°F for 45 minutes minimum. The book is intimidating in the way that engineering documentation is intimidating — every step seems both essential and impossible to verify in real time. Maya, who is a software engineer, has read internal documentation for distributed systems with less anxiety than she has read this book.

The problem, she has finally diagnosed, is not the starter. The problem is fear of waste. She is afraid that she will spend three days on a loaf and end up with a brick.

Tonight is the night she is going to do it anyway. Aisha, her partner, has agreed to eat whatever comes out of the oven on Sunday morning, even if it is, in fact, a brick.

The Plan

Maya has chosen a recipe she trusts: a country loaf, 80% bread flour, 20% whole-wheat flour, 75% hydration, 2% salt, 20% starter (all baker's percentages — she has had to look up the math). The recipe calls for:

• Friday evening: feed the starter at 1:5:5 (10 g starter, 50 g flour, 50 g water). Wait 8–10 hours.
• Saturday morning: combine 100 g of the active starter with 400 g flour and 300 g water; rest 30 minutes (the autolyse).
• Saturday morning, after autolyse: add 10 g salt and 50 g additional water; mix; bench-rest 30 minutes; do a coil fold; rest 30 minutes; coil fold again; repeat for 4 hours total.
• Saturday afternoon: shape into a tight ball; place seam-up in a flour-dusted basket (a banneton); cover; refrigerate overnight.
• Sunday morning: preheat oven and Dutch oven at 250°C / 480°F for 45 minutes. Score the loaf. Bake covered 25 minutes. Bake uncovered 20 minutes more, dropping to 230°C / 450°F. Cool one hour minimum before cutting.

She has read the recipe seventeen times. She has set timers. She has put the salt and the flour and the digital scale in a row on the counter. She is, by all the standards of an engineering review, ready.

What Happens

Friday, 8:00 PM. Maya feeds the starter. She removes 10 g of the existing starter from the jar and discards the rest. She adds 50 g whole-wheat flour and 50 g room-temperature filtered water. She stirs vigorously, marks the side of the jar with a rubber band at the level of the new mixture, and puts it on the counter (warm — about 24°C / 75°F).

She watches it for ten minutes. Nothing happens visibly. She tries not to be disappointed.

She goes to bed.

Saturday, 6:00 AM. Maya wakes and goes to the kitchen. The starter has tripled in volume. The rubber band is now well below the surface of the bubbling, dome-topped culture. The smell is fruity-tangy, faintly alcoholic, the way ripe pears smell after a few days on a counter. The texture, when she pokes a clean spoon into it, is full of small bubbles.

She drops a spoonful into a glass of water. It floats.

She is, suddenly and unexpectedly, almost in tears. She has been keeping this thing alive for fourteen weeks, and this is the first time she has seen it perform the way she has been told it would perform.

She takes a deep breath. She gets to work.

Saturday, 6:30 AM. Autolyse. Maya combines 100 g of the active starter with 400 g bread flour, 100 g whole-wheat flour, and 300 g water. The dough is shaggy, sticky, raggedy. She covers it and sets a timer for 30 minutes. She makes coffee.

7:00 AM. Maya adds 10 g of fine salt and 50 g of additional water (the rest of the recipe's hydration). She mixes by squeezing the dough through her fingers, the way the book describes. The salt dissolves quickly. The dough comes together. The hydration goes from manageable-shaggy to very slack. Maya has a small panic — did I add too much water? — and pushes through.

She sets the dough in a clean bowl, covers it with a damp towel, and starts the bulk fermentation timer for 4 hours.

7:30 AM. Coil fold #1. Maya wets her hands, lifts one side of the dough, stretches it up, and folds it under itself; rotates the bowl 90 degrees; repeats. The dough is fragile and sticky but it holds together, and she sees small bubbles starting to develop.

8:00 AM. Coil fold #2. The dough has tightened slightly. The folds go more cleanly. There are visible bubbles on the surface.

8:30 AM, 9:00 AM, 9:30 AM, 10:00 AM. Coil folds #3 through #7. Each fold is a little easier. The dough has gone from a slack puddle to something that holds a shape. Bubbles are everywhere. Maya, who has been checking the dough every 30 minutes, has gone from anxious to fascinated. She is watching the gluten develop in real time.

11:30 AM. End of bulk. The dough has roughly doubled in volume. The surface is glossy, dotted with bubbles. The sides of the bowl show stretchy strands when she peels the dough away from the wall. By every indicator in her book, it is ready to shape.

11:45 AM. Shaping. Maya turns the dough out onto a lightly floured counter. She uses a bench scraper (also bought specifically for this) to gently pre-shape the dough into a round. She lets it bench-rest for 20 minutes. Then she does the final shape — a tighter round, with the seam pinched on the bottom — and places it seam-up into a banneton dusted heavily with rice flour (which she has been told doesn't stick the way wheat flour does).

12:15 PM. She covers the banneton with plastic and slides it into the refrigerator.

She does not cook anything else this weekend that requires intellectual effort. She has used up her cooking energy.

Sunday, 8:00 AM. Oven preheats with Dutch oven inside, at 250°C / 480°F. Maya does the math: 45 minutes preheat is going to coincide perfectly with her coffee.

8:45 AM. Bake. Maya pulls the cold dough out of the refrigerator. It is firm from the cold; the gas has held; the surface is taut. She turns it out onto a piece of parchment paper, scores a single deep slash across the top with a razor blade. She lifts the lid of the screaming-hot Dutch oven (with thick gloves), drops the parchment-and-dough inside, slams the lid back on, and shuts the oven door.

⚠️ Safety reminder. Cast iron at 250°C will cause immediate burns. Use thick oven gloves on both hands. The lid is the heaviest object you will lift; lift with both hands. Do not allow any contact with skin.

25 minutes later. Maya removes the lid. The loaf has visibly risen. The slash has opened into a deep tear with a crisp ridge of flour-dusted dough peaked along its edge. She drops the temperature to 230°C / 450°F and bakes another 22 minutes.

9:32 AM. Maya pulls the loaf out of the Dutch oven and onto a wire rack. The loaf is dark brown, almost mahogany. The bottom thumps when she taps it. The smell — and this is the part she has not been prepared for — is overwhelming. A complex fermented aroma, with notes of nuts and roasted grain and something almost wine-like.

10:30 AM. Maya cuts the loaf. The crumb is open, with irregular medium-sized holes. The texture is springy and moist. She eats a piece without butter, and then another piece with butter. She brings the third piece to Aisha, who is still in bed.

"It's not a brick," Aisha says.

"It's not a brick," Maya agrees.

She eats the rest of that loaf over the next two days. On Tuesday evening, after work, she pulls the starter out of the refrigerator and feeds it again.

The Science Audit

What happened in Maya's loaf, in chemistry terms?

Starter: a wild fermentation community. The 10 g of starter she fed on Friday night contained roughly 10⁸ yeast cells and 10⁹ lactobacilli. Over 10 hours of feeding at 24°C, those organisms grew on the available flour and water, multiplied severalfold, and reached the active state where their CO₂ output is high and their metabolism is in steady-state.

Autolyse (30 minutes). During autolyse — a French baking term meaning "self-cleaving" — water hydrates the flour. The starches absorb water; the gluten-forming proteins (gliadin and glutenin) begin to associate; endogenous wheat enzymes (especially proteases and amylases) start their slow activity. The autolyse pre-builds gluten without any mechanical work, making the later mixing easier.

Salt addition. Salt does several things: it tightens gluten (NaCl ions interact with charged side chains on the gluten proteins, neutralizing some intramolecular repulsion); it controls yeast (slowing fermentation slightly); it adds flavor.

Bulk fermentation (4 hours, 22°C). The yeast eats the available sugars (glucose, fructose, maltose — generated by the wheat's amylase enzymes acting on damaged starch granules), running glycolysis and producing CO₂ + ethanol + side products (esters, higher alcohols). The lactobacilli simultaneously ferment sugars to lactic and acetic acid, dropping the dough's pH from about 5.8 (fresh-mixed) to around 4.5 (end of bulk). The acid environment selects against spoilage organisms and contributes flavor.

The coil folds redistribute the developing gluten without tearing it. Each fold builds dough strength and incorporates trapped air into the network.

Cold retard (20 hours, 4°C). Refrigeration slows the yeast's metabolism dramatically (Q₁₀ effects — yeast activity roughly halves with every 10°C drop). The lactobacilli also slow, but less dramatically — different microbes respond differently to cold. The relative ratio of "interesting flavor compounds" to "just CO₂" rises during the retard. The dough also firms up structurally, making it shapeable.

Bake (250°C / 480°F, then 230°C / 450°F). - First seconds: Dough surface heats; yeast activity surges briefly (oven spring) before the yeast dies at ~50°C internal. - First 5–10 minutes: Steam from the dough's water saturates the closed Dutch oven; surface stays moist and pliable; gas expansion gives the loaf its volume; gluten begins to set as it heats past 75°C. - 10–25 minutes: Starches gelatinize fully (Ch 9); the crumb structure locks in. - 25–47 minutes (uncovered): Surface dries; Maillard reaction and caramelization develop the dark crust; volatile flavor compounds form and infuse the loaf; ethanol mostly evaporates.

Cooling. Starches retrograde; the crumb's structure stabilizes; the loaf reaches eatable texture.

The chemistry was not, in any individual step, doing anything we have not described in earlier chapters. What was new for Maya was watching it all happen in sequence in her own kitchen, with her own hands, in her own pots. The chemistry was always going to work. The question was whether she would let it.

Analyze This

1. The sourdough recipe specified an autolyse step before adding salt and starter. Why? What chemistry happens in those 30 minutes that improves the final loaf?

2. Maya was anxious about adding "the rest of the water" because the dough became very slack. Was she right to be anxious, or is high hydration normal for this style of bread? What does extra water do to the final crumb?

3. The cold retard ran 20 hours at 4°C. Sketch a graph of "yeast activity vs. temperature" and "lactobacillus activity vs. temperature." Where on each curve are 4°C and 22°C? What does this predict about the flavor profile the cold retard would produce, relative to a warm, fast ferment?

4. Maya did not measure the pH at any step, but the recipe was working in the background: pH around 5.8 to start, around 4.5 by the end of bulk, perhaps 3.8–4.0 after the cold retard. (a) What does this pH change select for biologically? (b) If a starter mistakenly grew an unwanted organism, where in the timeline would it most likely show up — and why is the pH protective?

5. Suppose Maya had used 100 g of commercial yeast (way too much, but suppose) instead of 100 g of starter. The bread would still rise — probably more, faster. But describe three ways the finished loaf would taste different from the sourdough, and link each to a specific chemical or microbial mechanism.