Chapter 5 — Cheese & the Curd

"A cheese may disappoint. It may be dull, it may be naive, it may be oversophisticated. Yet it remains cheese, milk's leap toward immortality." — Clifton Fadiman, Any Number Can Play, 1957

Milk is not built to last. It leaves the udder warm, sterile in the strictest sense for perhaps an hour, and thereafter embarks, with or without a cheesemaker's help, on a rapid and entirely predictable decline: its sugars fermenting, its proteins souring, its fats turning, within a day or two at most, into something no household would willingly drink. A ewe or a goat gives milk in a narrow, seasonal window tied to lambing and kidding, and for the greater part of the pastoral year gives none at all; a cow's lactation is longer but no less finite. For any people whose wealth walked about on four legs rather than sitting still in a granary, this presented an old and pressing problem — a superabundance of a rich, nourishing liquid for a few months of the year, spoiling within days of milking, and nothing at all to show for it the rest of the time. Cheese is the answer that pastoral cultures the world over arrived at, independently and repeatedly, long before anyone understood the chemistry behind it: a means of taking milk's fleeting abundance and locking it, through curdling, salting and time, into a solid that could travel, that could wait out a winter, and that grew, if anything, more interesting the longer it was left alone. Fadiman's epigraph calls it milk's leap toward immortality, and for all its epigrammatic polish the phrase is not mere cleverness; it is, essentially, a fair description of the biochemistry.

The archaeological record for this leap runs back further than most kitchens realise. Perforated pottery vessels excavated across the Kuyavia region of Poland, coated in residues of dairy fat and dated to around 5500 BC, closely resemble in shape and function the cheese-strainers still used in traditional dairies today, and their chemical analysis, published in Nature in 2012, offers the earliest secure evidence anywhere in the world of cheesemaking as a distinct craft, separate from the simple drinking of milk — evidence, in other words, that Neolithic farmers in northern Europe were straining curds from whey some seven thousand years before this book was written. Further south and later in date, but rather more vivid, a sealed jar recovered from the tomb of Ptahmes, a high-ranking Egyptian official of the thirteenth century BC whose burial site near Saqqara was rediscovered in 2010 after being lost to drifting sand for over a century, was found in 2018 to contain a solidified white mass that laboratory analysis identified, by its surviving milk proteins, as a genuine cheese made from a blend of cow's and either goat's or sheep's milk — the oldest solid cheese ever recovered intact, some three thousand three hundred years old, and by more than one archaeologist's account bearing at least a family resemblance, in its method if hardly in its present state, to a soft cheese not unlike a modern Camembert. Between the Polish strainers and the Egyptian tomb lies a gap of some four thousand years and half a continent, and the honest conclusion the evidence supports is not that cheese was invented once and carried outward from a single hearth, but that it was discovered, independently and repeatedly, by pastoral peoples everywhere who kept milking animals and needed somewhere to put the surplus. The often-repeated legend that cheese was discovered by accident, when a nomadic herdsman carried milk in a pouch fashioned from a calf's stomach and found it had curdled into curds and whey by the day's end, is worth mentioning for its charm and its plausibility — rennet is, after all, exactly what a calf's stomach lining contains — but it survives in the historical record as folklore rather than as a documented event, repeated in cheese writing for centuries without a securely dated original source, and this book prefers to say so plainly rather than dress up a pleasing story as settled fact.

Milk transformed: the chemistry of a colloid

To understand why milk curdles at all, and why it curdles in two quite different ways depending on how it is provoked, it helps to know what milk actually is, chemically, before anyone interferes with it. Milk is, at root, an oil-in-water emulsion carrying a second, separate colloidal system suspended within it: droplets of butterfat, stabilised by a thin membrane of phospholipid and protein, floating in a watery serum that itself carries dissolved lactose, whey proteins, minerals, and, most importantly for this chapter's purposes, casein — a family of proteins that make up roughly eighty per cent of milk's total protein and that do not dissolve in the ordinary sense at all, but instead cluster into small spherical structures called micelles, tens of thousands of individual casein molecules bound loosely together around a core of calcium phosphate, each micelle a few hundred nanometres across and small enough to scatter light rather than transmit it, which is the entire reason milk looks opaque and white rather than clear.

What keeps these micelles from clumping together into a single mass, despite the crowded, jostling suspension they exist in, is a particular casein subtype called kappa-casein, which sits predominantly at the outer surface of each micelle and projects a short, flexible, negatively charged chain — sometimes called the micelle's "hairy layer" — out into the surrounding fluid. This hairy fringe does two things at once: its negative charge repels the equally negative fringe of every neighbouring micelle, and its physical bulk simply gets in the way, providing what colloid chemists call steric stabilisation, holding the whole suspension apart by force of crowding as much as by charge. Remove that hairy layer, or neutralise its charge, and the underlying attraction between micelles — largely mediated by the calcium phosphate at their cores — is free to assert itself, and the micelles begin, of their own accord, to clump. Curdling milk, whatever method a cheesemaker chooses to achieve it, is nothing more or less than defeating kappa-casein's stabilising fringe by one of two entirely different routes, and the route chosen determines, more than almost any other single decision in the dairy, what kind of cheese results.

The first route is acid. Lower milk's pH, whether by the direct addition of an acid such as lemon juice or vinegar, or — the older and, for cheesemaking proper, far more common method — by allowing lactic acid bacteria to ferment milk's own lactose into lactic acid over a period of hours, and the accumulating hydrogen ions gradually neutralise kappa-casein's negative charge, weakening the electrostatic repulsion that keeps the micelles apart. As the pH falls toward roughly 4.6 — the isoelectric point of casein, the precise pH at which the protein carries no net charge at all — the micelles lose their mutual repulsion entirely and aggregate into a soft, fine-curded, faintly tangy mass: the method behind fresh cheeses such as paneer, queso fresco, and the simplest farmhouse curd cheeses, none of which require any enzyme at all, only bacteria, warmth, and time — or, for the impatient cook, a splash of vinegar and considerably less of either.

The second route is enzymatic, and it is the one behind the great majority of the world's aged and structured cheeses, because it achieves something acid coagulation cannot: a firm, elastic, cohesive curd strong enough to be cut, stacked, pressed and matured for months or years without simply collapsing into paste.

Rennet and coagulation

Rennet is, traditionally, an extract of the fourth stomach chamber, the abomasum, of an unweaned calf, kid or lamb — the chamber whose job in the living animal is to curdle its mother's milk and hold it in the gut long enough for slow, thorough digestion to occur, which is precisely why a milk-fed calf's stomach carries such a potent supply of the enzyme in the first place. The active ingredient is chymosin, and its mechanism, worked out through the second half of the twentieth century by dairy biochemists, is a model of enzymatic economy: chymosin does not attack the casein micelle wholesale, but performs a single, highly specific cut, severing kappa-casein's hairy fringe at one precise point along its chain — the bond between the 105th and 106th amino acids, phenylalanine and methionine — clipping away the water-soluble outer portion of the fringe, called the caseinomacropeptide, which drifts off into the surrounding whey, and leaving behind a shortened, hydrophobic stub, para-kappa-casein, no longer capable of holding the micelles apart. Once enough kappa-casein across the milk has been cut in this way — cheesemakers generally reckon the gelation point arrives once somewhere under eighty per cent of it has been cleaved — the naked, destabilised micelles begin to aggregate, drawn together by the calcium bridges the hairy layer had been suppressing, and the whole liquid sets, within twenty to forty minutes at a gentle warmth, into a coherent, custard-like gel: the curd proper, holding the great bulk of the milk's fat and protein suspended within a continuous protein matrix, with the liquid whey — water, lactose, whey proteins, and the cast-off caseinomacropeptide — still trapped throughout it, waiting to be drained away.

Not every culture that made cheese had ready access to a calf's stomach, and the historical record shows considerable ingenuity in finding substitutes with a comparable protein-cleaving action. The Iberian peninsula and parts of the Balkans developed a long tradition of vegetable rennet pressed from the dried flowers of the cardoon thistle, Cynara cardunculus, whose enzymes — cardosins, chemically distinct from chymosin but functionally similar enough — still produce cheeses such as Spain's Torta del Casar and Portugal's Serra da Estrela with a distinctively soft, almost spoonable curd quite unlike calf-rennet cheeses of comparable age. Fig-tree latex, nettle, and a handful of other plant extracts saw similar, more localised use wherever animal rennet was scarce, expensive, or, for some communities, religiously unacceptable alongside meat. The twentieth century added a further, now dominant alternative: microbial and, since the 1990s, genetically engineered "fermentation-produced chymosin," in which the calf gene for chymosin is inserted into a laboratory culture of yeast or fungus, which then manufactures the identical enzyme in a fermentation vat without any animal involvement at all. The United States Food and Drug Administration approved the first such product for commercial use in 1990, and equivalent regulatory approval followed shortly after in Europe; today the substantial majority of cheese made in the industrialised world — including, somewhat to the surprise of cooks who assume otherwise, most commercial Cheddar, mozzarella and a great many other everyday cheeses — is set with this fermentation-produced enzyme rather than with rennet drawn from an actual calf, a genuinely underappreciated piece of modern food biotechnology that has, among its other virtues, made a reliably vegetarian cheese possible at industrial scale for the first time in the craft's long history.

Curds and whey

However the coagulation is achieved, the cheesemaker's next task is to persuade the newly set curd to release the whey trapped within it, a process called syneresis, and the tools available for hastening it — cutting, heating, stirring, salting, pressing — account, between them, for a great deal of the difference between one cheese family and the next. Cutting the set curd into smaller cubes exposes more surface area through which whey can escape, and the size of the cut is itself a deliberate lever: a fresh chèvre or a Brie is cut, if at all, into large, gentle pieces that retain a good deal of moisture, while a hard grating cheese such as Parmigiano-Reggiano is cut into curds no larger than a grain of rice or wheat, releasing far more whey and yielding, eventually, a far denser, drier paste. Gentle heating of the cut curd — "cooking," in the cheesemaker's own vocabulary, though the temperatures involved rarely exceed the upper thirties or low forties Celsius — accelerates syneresis further still, contracting the curd's protein matrix and squeezing out yet more liquid, which is why the hardest, longest-aged cheeses are almost invariably cooked curds and the softest are almost never cooked at all. Cheddaring, the technique that gives Cheddar cheese its name, adds a further mechanical step: the drained curd is cut into slabs, stacked and repeatedly turned under its own weight, a slow, patient pressing that expels yet more whey while knitting the curd's texture into the dense, sliceable, faintly stringy consistency cheddar is prized for.

The whey that drains away through all of this is not waste, whatever its watery, faintly greenish appearance might suggest to the uninitiated cook; it carries the milk's whey proteins — chiefly beta-lactoglobulin and alpha-lactalbumin, entirely distinct from casein and untouched by rennet's cutting action — along with most of the original milk's lactose and a modest residue of butterfat too fine to have been caught in the curd. Heat that whey further, past roughly 80°C, and the dissolved whey proteins, which the initial coagulation left entirely alone, finally denature and coagulate in their own right, forming the soft, delicate, faintly sweet curds of ricotta — the name means, quite literally, "recooked," a fair description of a cheese made by heating what a first cheese has already left behind, and a tidy demonstration that even milk's fugitive protein has a second chance at solidity before it is finally lost to the drain.

The great families

Once a curd has been formed and drained, its onward fate divides, broadly, into four great families, and the difference between them is governed as much by microbiology as by mechanics.

Fresh cheeses — chèvre, ricotta, paneer, mozzarella in its unaged state, cottage cheese — are eaten within days of the curd's formation, with no further ripening asked of them at all; their character comes almost entirely from the milk itself, the coagulation method, and whatever salt and seasoning the maker adds, and it is this family, requiring the least equipment, the least time, and the least specialised knowledge, that most home cheesemaking begins and often ends with. Mozzarella earns its own small mention within this family for a technique unlike any of its fresh cousins: after acid and a little rennet together set the curd, that curd is not simply drained and eaten but heated in hot water or whey to somewhere near 60°C and physically stretched and folded — the pasta filata, "spun paste," method the cheese takes its Italian name from — a mechanical working that aligns the curd's casein strands into long, parallel fibres and gives the finished cheese its distinctive elastic, sliceable pull, a texture no other fresh cheese on this list shares, achieved by muscle and heat rather than by any further microbe or enzyme at all.

None of these four families is hermetically sealed from the others, and it is worth saying so before moving on, since the tidy taxonomy a book must impose for clarity is considerably tidier than the cheeses themselves. A cheese may be washed and also mould-ripened at its core, cooked and also long-aged, fresh in method but salted and dried until it behaves, on the table, like something considerably harder; regional traditions the world over have spent centuries inventing hybrids that sit happily astride two or more of the categories above, and the four families offered here are best read as poles a real cheese may lean toward, rather than boxes any real wheel is obliged to fit inside.

Bloomy-rind cheeses — Camembert and Brie chief among them — are instead deliberately inoculated, after draining, with spores of Penicillium camemberti, a domesticated mould related to, but distinct from, the wild moulds that would otherwise colonise an ageing cheese at random. Sprayed or dusted onto the curd's surface, the mould germinates over the following days into the soft, white, velvety rind these cheeses are named for, and as it grows it secretes enzymes — proteases and lipases — that work inward from the surface, breaking down the firm, chalky young curd beneath into the soft, oozing, ammoniated paste a properly ripe Camembert is prized for; the ripening proceeds, quite visibly, from rind to centre, which is why a young wheel cuts firm and pale at its heart while an older one has softened almost to the rind.

Washed-rind cheeses — Munster, Taleggio, Époisses, Livarot, and the notoriously pungent Limburger among them — take a different microbial route entirely, one this book's naturalist sympathies find genuinely delightful to explain. Rather than a mould, these cheeses are ripened chiefly by bacteria, above all Brevibacterium linens, deliberately encouraged by the maker's habit of regularly washing the developing rind with brine, sometimes cut with beer, wine or spirit, keeping the surface salty, moist and warm — precisely the conditions this bacterium favours. Brevibacterium linens is not a specialist of the cheese cave alone; it lives, in nature, on human skin, and with particular enthusiasm between the toes, where it feeds on sweat and shed skin cells much as it feeds on a cheese's surface proteins, in both cases producing, via an enzyme called methionine gamma-lyase, a family of volatile sulphur compounds — chief among them methanethiol — responsible for the sharp, unmistakably foot-like aroma these cheeses share with an unwashed sock, a coincidence of shared microbiology rather than of shared ingredients, and one every cheesemonger has had to explain, gently, to at least one alarmed customer.

Blue cheeses — Roquefort, Gorgonzola, Stilton — take yet a third route, inoculating the curd itself, rather than merely its surface, with spores of Penicillium roqueforti, a mould that, unlike its bloomy-rind cousin, requires oxygen to grow and therefore develops only where the cheese is pierced or naturally fissured, threading itself through the paste along these air channels to produce the characteristic blue-green veining, sharp aroma and peppery bite these cheeses are named for. Roquefort's own legend — that a shepherd, called away from his lunch of bread and ewe's-milk curd by the sight of a passing shepherdess, returned months later to find his abandoned meal transformed by the wild mould native to the limestone caves of Combalou into something extraordinary — is, again, a charming story rather than a documented one, though the underlying biology it dramatises is sound enough: Penicillium roqueforti does occur naturally in cool, damp caves, and Roquefort's own production remains legally tied to ripening within those specific caves to this day. What is more securely documented is the craft's own transition, over the nineteenth and twentieth centuries, from relying on wild airborne mould to deliberate inoculation: cheesemakers first propagated spores by growing the mould on stale bread and working it, ground to powder, into the curd, a practice recorded well into the twentieth century before more controlled laboratory culturing methods took over.

Hard, long-aged cheeses — Parmigiano-Reggiano, aged Gouda, Comté, mature Cheddar, Gruyère — depend on none of these visible surface organisms at all, or only lightly, and instead on the slow, patient work of enzymes, both the residual rennet trapped within the curd and enzymes released by the lactic bacteria used in the initial fermentation, breaking casein down over months or years into shorter peptides and, eventually, free amino acids. It is this slow proteolysis that gives a well-aged hard cheese its savoury depth and, in the most extreme cases, its audible crunch: the small white crystals scattered through a well-aged Parmigiano-Reggiano or Gouda are not, as diners sometimes assume, crystallised salt, but tyrosine, one of the amino acids released by this long protein breakdown, precipitating out of solution once its local concentration exceeds what the cheese's remaining moisture can dissolve — a literal, edible signature of age that no young cheese, however good, has yet had time to earn.

The microbiology of the rind

Stand back from any one of these families and a common thread emerges that the naturalist in this book's persona finds genuinely compelling: a ripening cheese's rind is not a fixed thing but a living, changing microbial community, succeeding through a predictable sequence of dominant organisms as the cheese ages, not unlike the ecological succession of a cleared field returning to forest. A newly drained curd's surface is at first colonised by yeasts and the mould Geotrichum candidum, tolerant of the curd's initial high acidity and low oxygen, which soften and deacidify the surface just enough to permit less acid-tolerant organisms — the Penicillium moulds, the Brevibacterium and its bacterial associates — to establish themselves in turn over the following days and weeks, each wave of colonisers altering the surface conditions in ways that favour the wave that follows. A skilled affineur, the specialist who manages a cheese's ripening in a cave or cellar rather than a dairy, is managing this succession as much as managing the cheese itself, adjusting humidity, temperature and washing schedule to favour one organism's advance over another's, and the particular blend of organisms present in any one cave's air and walls — its resident microbial population, built up and stabilised over years or decades of continuous use — genuinely does impart a distinct, reproducible character no other cave, however carefully its recipe is copied, can quite replicate: a microbial terroir every bit as real, and every bit as jealously guarded by the makers who depend on it, as the terroir this book's third chapter attributed to wine and to the wild ferments generally.

Field Note — A Wheel in Section

Cut a whole cheese in half, cave-aged and true to its type rather than the pre-sliced wedge a supermarket offers, and the wheel reads, if the reader looks properly, like a cross-section through geological strata rather than a single uniform food. At the outermost edge sits the rind itself — bloomed white, washed orange-red, or simply hard and waxed, depending on the family — a distinct microbial zone, often only a few millimetres deep, where the whole visible drama of mould or bacterium has done its work. Beneath it, in a ripe bloomy or washed-rind cheese, lies a soft, almost liquid layer where the rind's enzymes have already worked their way inward, its texture and often its very colour distinct from the firmer, paler paste still holding its original shape nearer the centre — a visible record, cut open for inspection, of ripening still travelling inward from the surface it began at, precisely as the crumb of a loaf records how far the oven's heat ever reached beneath its crust. In a long-aged hard cheese the geology is subtler but no less real: fine white crystals of tyrosine, scattered unevenly through the paste like a hard frost, denser wherever the cheese has had the longest or the warmest spell to ripen in, and a texture that crumbles rather than bends nearer the rind, where years of slow moisture loss have concentrated the protein most. Nothing about a wheel of cheese, cut fresh, is quite as uniform as its smooth exterior suggests; it is, every time, a small and edible record of a microbial argument conducted, layer by layer, from the outside in.

<figure> <svg xmlns="http://www.w3.org/2000/svg" width="100%" viewBox="0 0 640 400" role="img"> <title>From Milk to Cheese — the five stages of the curd</title> <style> .ink{stroke:#43331f;fill:none;stroke-width:1;} .mid{stroke:#6f5a40;fill:none;stroke-width:0.8;} .light{stroke:#a9906c;fill:none;stroke-width:0.7;} .hair{stroke:#c8b79a;fill:none;stroke-width:0.6;} .fillmid{fill:#6f5a40;stroke:none;} text{font-family:Georgia,'Times New Roman',serif;fill:#43331f;} .title{font-variant:small-caps;letter-spacing:2.5px;font-size:16px;text-anchor:middle;} .roman{font-style:italic;font-size:10px;text-anchor:middle;fill:#9c5a2c;} .label{font-style:italic;font-size:11px;text-anchor:middle;} .sub{font-style:italic;font-size:9.5px;text-anchor:middle;fill:#6f5a40;} </style> <defs> <path id="fl" class="mid" d="M0,0 C3,-7 13,-7 15,1 C16,7 9,11 4,9 C1,8 0,3 2,2"/> <path id="ar" class="fillmid" d="M0,-3.2 L8,0 L0,3.2 Z"/> </defs> <rect x="12" y="12" width="616" height="376" class="ink"/> <rect x="17" y="17" width="606" height="366" class="hair"/> <use href="#fl" x="24" y="24"/> <use href="#fl" x="616" y="24" transform="rotate(90 616 24)"/> <use href="#fl" x="616" y="376" transform="rotate(180 616 376)"/> <use href="#fl" x="24" y="376" transform="rotate(270 24 376)"/> <line x1="60" y1="36" x2="580" y2="36" class="hair"/> <text x="320" y="52" class="title">From Milk to Cheese</text> <line x1="60" y1="60" x2="580" y2="60" class="hair"/> <text x="320" y="76" class="sub">Chapter V · The Art of the Curd</text>  <line x1="140" y1="235" x2="540" y2="235" class="hair"/>  <g transform="translate(100,235)"> <circle r="34" class="ink"/> <text y="-46" class="roman">I</text> <path class="ink" d="M-14,-4 L-14,20 Q-14,26 -8,26 L8,26 Q14,26 14,20 L14,-4 Z"/> <rect x="-6" y="-18" width="12" height="14" class="ink"/> <path class="mid" d="M10,-18 L17,-22 L11,-13 Z"/> <path class="mid" d="M16,-1 C27,-1 27,15 16,15"/> <line x1="-9" y1="5" x2="9" y2="5" class="light"/> <line x1="-9" y1="12" x2="9" y2="12" class="light"/> <text y="52" class="label">Milk</text> </g> <use href="#ar" x="130" y="235"/>  <g transform="translate(220,235)"> <circle r="34" class="ink"/> <text y="-46" class="roman">II</text> <path class="ink" d="M0,-16 C9,-4 9,10 0,17 C-9,10 -9,-4 0,-16 Z"/> <line x1="-5" y1="0" x2="5" y2="0" class="mid"/> <line x1="0" y1="-5" x2="0" y2="5" class="mid"/> <text y="52" class="label">Rennet or Acid</text> </g> <use href="#ar" x="250" y="235"/>  <g transform="translate(340,235)"> <circle r="34" class="ink"/> <text y="-46" class="roman">III</text> <path class="ink" d="M-18,2 Q0,20 18,2 L14,-6 Q0,8 -14,-6 Z"/> <circle cx="-8" cy="-3" r="4" class="mid"/> <circle cx="2" cy="-5" r="4.5" class="mid"/> <circle cx="10" cy="-1" r="3.5" class="mid"/> <circle cx="-9" cy="9" r="0.8" class="fillmid"/> <circle cx="1" cy="12" r="0.8" class="fillmid"/> <circle cx="9" cy="9" r="0.8" class="fillmid"/> <circle cx="-3" cy="14" r="0.8" class="fillmid"/> <text y="52" class="label">Curds & Whey</text> </g> <use href="#ar" x="370" y="235"/>  <g transform="translate(460,235)"> <circle r="34" class="ink"/> <text y="-46" class="roman">IV</text> <circle cx="0" cy="-24" r="5" class="ink"/> <line x1="0" y1="-19" x2="0" y2="-14" class="mid"/> <rect x="-18" y="-14" width="36" height="5" class="ink"/> <line x1="-14" y1="-9" x2="-14" y2="14" class="light"/> <line x1="14" y1="-9" x2="14" y2="14" class="light"/> <rect x="-18" y="14" width="36" height="5" class="ink"/> <text y="52" class="label">Pressing & Ageing</text> </g> <use href="#ar" x="490" y="235"/>  <g transform="translate(580,235)"> <circle r="34" class="ink"/> <text y="-46" class="roman">V</text> <path class="ink" d="M-16,14 L16,14 L4,-14 Z"/> <circle cx="-2" cy="6" r="2" class="mid"/> <circle cx="6" cy="9" r="1.6" class="mid"/> <circle cx="-7" cy="11" r="1.3" class="mid"/> <text y="52" class="label">Cheese</text> </g> <line x1="120" y1="350" x2="520" y2="350" class="hair"/> <text x="320" y="366" class="sub">Curd follows culture, as form follows curd</text> </svg> <figcaption><em>Plate 9. Milk to curd to cheese: casein micelles, their stabilising kappa-casein fringe intact, destabilised by rennet's single precise cut, aggregating into a coherent gel, cut and drained of whey, and finally salted and set to ripen</em></figcaption> </figure>

<figure> <svg xmlns="http://www.w3.org/2000/svg" width="100%" viewBox="0 0 640 640" role="img"> <title>The Cheese Families — fresh, bloomy-rind, washed-rind, blue, and hard-aged wheels</title> <style> .ink{stroke:#43331f;fill:none;stroke-width:1;} .mid{stroke:#6f5a40;fill:none;stroke-width:0.8;} .light{stroke:#a9906c;fill:none;stroke-width:0.7;} .hair{stroke:#c8b79a;fill:none;stroke-width:0.6;} .fillmid{fill:#6f5a40;stroke:none;} text{font-family:Georgia,'Times New Roman',serif;fill:#43331f;} .title{font-variant:small-caps;letter-spacing:2.5px;font-size:16px;text-anchor:middle;} .label{font-style:italic;font-size:12px;text-anchor:middle;} .sub{font-style:italic;font-size:9px;text-anchor:middle;fill:#6f5a40;} </style> <defs> <path id="fl" class="mid" d="M0,0 C3,-7 13,-7 15,1 C16,7 9,11 4,9 C1,8 0,3 2,2"/> </defs> <rect x="12" y="12" width="616" height="616" class="ink"/> <rect x="17" y="17" width="606" height="606" class="hair"/> <use href="#fl" x="24" y="24"/> <use href="#fl" x="616" y="24" transform="rotate(90 616 24)"/> <use href="#fl" x="616" y="616" transform="rotate(180 616 616)"/> <use href="#fl" x="24" y="616" transform="rotate(270 24 616)"/> <line x1="70" y1="36" x2="570" y2="36" class="hair"/> <text x="320" y="52" class="title">The Cheese Families</text> <line x1="70" y1="60" x2="570" y2="60" class="hair"/> <text x="320" y="76" class="sub">Chapter V · Five Wheels, Five Paths of Rind</text>  <g transform="translate(160,210)"> <circle r="58" class="ink"/> <circle cx="-14" cy="-10" r="1" class="fillmid"/><circle cx="10" cy="-16" r="1" class="fillmid"/> <circle cx="18" cy="8" r="1" class="fillmid"/><circle cx="-6" cy="18" r="1" class="fillmid"/> <circle cx="4" cy="-2" r="1" class="fillmid"/><circle cx="-20" cy="10" r="1" class="fillmid"/> <circle cx="22" cy="-6" r="1" class="fillmid"/><circle cx="-10" cy="-24" r="1" class="fillmid"/> <text y="88" class="label">Fresh</text> <text y="104" class="sub">soft curd, no rind</text> </g>  <g transform="translate(320,210)"> <circle r="58" class="ink"/> <circle r="58" class="light" stroke-dasharray="1.5 3"/> <line x1="0" y1="-58" x2="0" y2="-64" class="mid"/><line x1="41" y1="-41" x2="45" y2="-45" class="mid"/> <line x1="58" y1="0" x2="64" y2="0" class="mid"/><line x1="41" y1="41" x2="45" y2="45" class="mid"/> <line x1="0" y1="58" x2="0" y2="64" class="mid"/><line x1="-41" y1="41" x2="-45" y2="45" class="mid"/> <line x1="-58" y1="0" x2="-64" y2="0" class="mid"/><line x1="-41" y1="-41" x2="-45" y2="-45" class="mid"/> <text y="88" class="label">Bloomy-Rind</text> <text y="104" class="sub">soft white mould</text> </g>  <g transform="translate(480,210)"> <circle r="58" class="ink"/> <path class="mid" d="M-50,-30 Q-30,-38 -10,-30 Q10,-22 30,-30 Q50,-38 50,-30" /> <path class="mid" d="M-50,-8 Q-30,-16 -10,-8 Q10,0 30,-8 Q50,-16 50,-8" /> <path class="mid" d="M-50,14 Q-30,6 -10,14 Q10,22 30,14 Q50,6 50,14" /> <path class="mid" d="M-50,36 Q-30,28 -10,36 Q10,44 30,36 Q50,28 50,36" /> <text y="88" class="label">Washed-Rind</text> <text y="104" class="sub">brined & sticky</text> </g>  <g transform="translate(220,420)"> <circle r="58" class="ink"/> <path class="mid" d="M-30,-38 Q-20,-10 -34,4 Q-14,10 -22,32"/> <path class="mid" d="M0,-42 Q10,-18 -6,-2 Q14,6 4,30"/> <path class="mid" d="M30,-36 Q18,-12 32,0 Q16,14 26,36"/> <circle cx="-30" cy="-38" r="1" class="fillmid"/><circle cx="-22" cy="32" r="1" class="fillmid"/> <circle cx="0" cy="-42" r="1" class="fillmid"/><circle cx="4" cy="30" r="1" class="fillmid"/> <circle cx="30" cy="-36" r="1" class="fillmid"/><circle cx="26" cy="36" r="1" class="fillmid"/> <text y="88" class="label">Blue</text> <text y="104" class="sub">veined with culture</text> </g>  <g transform="translate(420,420)"> <circle r="58" class="ink"/> <circle r="44" class="light"/> <circle r="30" class="light"/> <circle r="16" class="light"/> <circle cx="-8" cy="-6" r="0.9" class="fillmid"/><circle cx="10" cy="4" r="0.9" class="fillmid"/> <circle cx="-2" cy="16" r="0.9" class="fillmid"/><circle cx="18" cy="-14" r="0.9" class="fillmid"/> <circle cx="-20" cy="12" r="0.9" class="fillmid"/><circle cx="4" cy="-24" r="0.9" class="fillmid"/> <text y="88" class="label">Hard-Aged</text> <text y="104" class="sub">grainy, long-ripened</text> </g> <line x1="90" y1="560" x2="550" y2="560" class="hair"/> <text x="320" y="576" class="sub">Every family begins the same; only time and moisture divide them</text> </svg> <figcaption><em>Plate 10. The cheese families and the organisms that make them: bloomy rind ripened by Penicillium camemberti from without, blue cheese veined by Penicillium roqueforti from within, washed rind coloured and scented by Brevibacterium linens, and the hard-aged wheel transformed slowly by its own trapped enzymes alone</em></figcaption> </figure>

A note on safety: the raw-milk question

No chapter in this book has a better claim to this volume's hard rule on honesty than the question of raw-milk cheese, because it is a genuinely contested question, argued in good faith by knowledgeable people on both sides, and this book owes the reader the argument rather than a false consensus.

Pasteurisation itself is not, at root, a cheesemaking invention at all; it takes its name from Louis Pasteur, who developed the technique in 1864 not for milk but for wine, commissioned by Napoleon III's government to solve the spoilage plaguing the French wine trade, and who demonstrated that heating wine to a temperature well short of boiling — commonly cited at 60 to 100°C for a controlled period — killed the spoilage organisms responsible without ruining the wine's flavour, a discovery soon extended, by Pasteur and others, to beer and then to milk. Applied to milk, pasteurisation reliably kills the pathogenic bacteria of greatest historical concern in dairy — among them Brucella, Salmonella, and pathogenic strains of E. coli and Listeria monocytogenes — and it is this reliability that led American regulators, in the wake of documented outbreaks of typhoid and brucellosis traced to raw dairy products in the early twentieth century, toward one of two paths for any cheese made with unpasteurised milk: pasteurise the milk before making the cheese, or, if raw milk is used, mature the finished cheese for a minimum period during which the cheese's own falling pH and slowly changing internal chemistry can be relied upon to suppress most pathogens on its own. The United States Food and Drug Administration formalised the second option in 1949, in the regulation still known informally as the sixty-day rule: any cheese made from milk that has not been pasteurised must be aged at a temperature above 35°F (roughly 1.7°C) for a minimum of sixty days before sale. The scientific basis for that particular figure appears to trace to a handful of mid-century studies — most influentially a 1946 investigation into Brucella abortus surviving in unpasteurised Cheddar for up to six months, alongside earlier data suggesting typhoid outbreaks were not attributed to cheeses cured beyond about sixty-three days — rather than to any single, comprehensive risk assessment, and the regulation has stood substantially unrevised for three-quarters of a century since.

Whether sixty days remains an adequate safety margin is precisely where the honest disagreement lives. Proponents of the current rule, and of raw-milk cheesemaking more broadly, point to the genuinely enormous body of raw-milk cheese produced and consumed without incident across continental Europe — France and Italy in particular permit, and take considerable pride in, raw-milk versions of cheeses such as Comté, Roquefort and Parmigiano-Reggiano that would fall outside America's rule if made and aged for less than the mandated period, and note that the cheesemaking process itself, through acidification, salting and moisture loss, creates conditions genuinely hostile to many pathogens well before sixty days have passed. Critics, including food-safety researchers who have published in the peer-reviewed literature since the late 1990s, point to more recent and more troubling evidence: a study published in the Journal of Food Protection in December 2010 found a strain of E. coli O157:H7 remaining viable in Gouda and Cheddar cheese for roughly one hundred days, well beyond the sixty-day threshold, and documented outbreaks — among them a multistate E. coli illness cluster traced to a raw-milk Gouda that had, in fact, been aged the full sixty days required by law — have demonstrated in practice that the rule's protection is not absolute. The United States Food and Drug Administration itself has acknowledged, through its own spokespeople, that researchers first raised questions about the sixty-day threshold's adequacy in the late 1990s, and has since undertaken its own microbiological surveillance of raw-milk cheese aged the mandated period, without, as of this writing, revising the underlying rule.

Listeria monocytogenes deserves particular mention in its own right, because it is the pathogen of greatest continuing concern in soft cheese specifically, raw-milk or otherwise, and because its risk falls with genuine unevenness across the population. Listeria tolerates refrigeration, salt and moderate acidity far better than most foodborne pathogens, giving it an unusual foothold in exactly the soft, moist, low-acid environment a bloomy-rind or washed-rind cheese provides; roughly one per cent of soft cheeses carry detectable Listeria contamination by some estimates, a figure that rises considerably, by a factor of three or four, when the cheese in question is made from raw rather than pasteurised milk. For most healthy adults, Listeria infection is mild or entirely asymptomatic; for pregnant women, who contract listeriosis at roughly ten times the rate of the general population, and for whom the infection can cross the placenta and cause miscarriage, stillbirth or severe illness in a newborn, health authorities on both sides of the Atlantic are essentially unanimous in advising against soft, mould-ripened and raw-milk cheeses during pregnancy, regardless of that cheese's country of origin or reputation. This is, in this book's judgement, the one piece of the raw-milk argument that carries no honest counter-case at all: whatever position a reader takes on the broader merits of raw-milk cheese for the general adult population — and reasonable, informed people continue to disagree — the specific advice for pregnancy and for the immunocompromised is not a matter of taste or tradition, but of a documented and occasionally fatal risk, and this book declines, in keeping with its hard rule on safety, to romanticise any hazard past the point of honesty.

The reader is entitled, in short, to a plain summary rather than a verdict this book has no standing to deliver: raw-milk cheese, properly made by a careful producer under a functioning regulatory regime, carries a real but generally small risk for the healthy general population, weighed by its defenders against a flavour and microbial complexity pasteurisation genuinely does diminish; that same cheese carries a materially higher and better-documented risk for pregnant women, infants, the elderly and the immunocompromised, for whom the pasteurised alternative is the only defensible choice; and the sixty-day rule that governs the American middle ground between an outright ban and the freer European tradition remains, on the evidence this chapter has reviewed, a reasonable but imperfect compromise rather than a guarantee — worth knowing, by any diner reaching for a wedge of unfamiliar cheese, exactly as it is, and not as either its enthusiasts or its critics would prefer it to be.

Butter and cream: an emulsion inverted

It is worth closing this chapter with the one great dairy transformation that involves no microbe and no enzyme at all, only mechanical force applied to a colloid this section has already introduced: cream, agitated past its own stability, into butter. Cream itself, skimmed from milk by simple settling or by the more efficient modern centrifuge, is an oil-in-water emulsion, its butterfat globules — each one wrapped in a thin, protective membrane of phospholipid and protein, the milk fat globule membrane, that keeps them from coalescing into a single mass — dispersed through a watery, protein-rich serum. Churning, whether by the mechanical paddle of an industrial dairy or the vigorous shaking of a jar in a domestic kitchen, does one simple thing over and over: it batters the fat globules against one another and against the churn's own walls with enough force to rupture that protective membrane, and once ruptured, the exposed fat begins to coalesce — small globules merging into larger ones, larger ones trapping air and forming a rising foam, until the fat clumps grow too large for that foam to sustain itself, the foam collapses, and a mass of visible butter grains separates out, swimming in the thin, faintly milky liquid every home baker recognises as buttermilk.

What has actually occurred, in the strict language of colloid chemistry, is a phase inversion: cream begins as an oil-in-water emulsion, fat suspended as the minority phase within a continuous watery phase, and churning drives it, by main force, into its structural opposite — butter is a water-in-oil emulsion, tiny droplets of water and buttermilk suspended as the minority phase within a continuous matrix of fat, the very reverse arrangement of the cream that produced it. That a violent, sustained battering of one stable colloid should reliably flip it into the opposite of itself is, on reflection, no less strange a transformation than curdling milk into cheese, and a good deal faster: where a curd may take hours to set and months to ripen, a jar of cream, shaken by an impatient child at a kitchen table, will invert into recognisable butter within about ten minutes, no bacterium, mould or enzyme required at any stage, only cold cream and a sufficiently tireless arm. Cultured butter, prized across much of continental Europe for a flavour American butter generally lacks, restores a microbial element to the process after all: the cream is deliberately soured with lactic acid bacteria before churning, in essentially the same acid-fermentation logic behind a fresh cheese, lending the finished butter a faintly tangy, complex depth that sweet-cream butter, churned straight from fresh cream without that fermentation, never quite develops — a small, final reminder, as this chapter closes, that even the dairy's simplest mechanical transformation has, more often than not, a microbe waiting patiently just offstage to make it better still.

Milk, defended by nothing more than a fringe of stabilising protein a few molecules deep, proves no more resistant to transformation than the wheat this book examined a chapter ago — curdled by acid or by a single enzymatic cut, colonised by mould or bacterium into a dozen distinct families, or simply battered until its very structure turns itself inside out. What curdling, ripening and churning all share, in the end, is the same principle this whole volume keeps rediscovering in new materials: that transformation is rarely destruction, and very often the opposite — milk's fleeting abundance, condensed by craft into something that keeps, that travels, that grows only more itself the longer it is patiently left alone. Not every transformation in this book's larder is so forgiving of patience, however, nor so quick to reward a careless hand with nothing worse than a disappointing wheel of cheese. There remain crafts in which the margin between a safely preserved food and a dangerous one is measured in a single degree of temperature or a single point of acidity, where "wait and see" is no longer sound advice at all — and it is to that sterner, older craft of preservation, salt and smoke and the sealed jar, that this compendium turns next.

บทถัดไป · 9 / 15Chapter 6 — The Art of Preservation