The Global Cookie Jar

Nobody manufactures a chocolate-chip or cream-filled-sandwich cookie better than the U.S. cookie industry. And nobody appreciates that excellence more than the American cookie consumer, who, in 2003, helped Nabisco's Oreo and Chips Ahoy!, as well as Keebler's Chips Deluxe, rake in a full 25% of the cookie category's annual take, making these the nation's top-selling brands another year running, according to Information Resources, Inc., Chicago. So why should cookie manufacturers stray from the chocolate chippers and cream sandwiches they do best?

Because of biscotti. It seems like only yesterday that we wrote off the dried-out wedges as Italy's scheme for selling burnt toast to naïve Yanks. But nowadays, many wouldn't dream of leaving a coffee shop without one to dunk in a double-decaf soymilk latte. And biscotti aren't alone: The bakery case at my local coffeehouse now carries as many madeleines, Russian teacakes, and oat-filled Australian Anzac biscuits as it does peanut butter cookies and snickerdoodles. In supermarkets, too, imports from the likes of France, Germany and Britain make strong showings alongside our homegrown brands, suggesting that American consumers might welcome increasing cookie immigration quotas.

But will they welcome the Marie? The American cookie has evolved into an icon of richness that reflects the land of plenty. That evolution reaches its apex in the typical food-court bakeshop cookie: an indulgence the size of a dinner plate that's so sweet and chewy -- and so packed with nuts, chips and candies -- that we can't wait until the oven timer stops before we sink our teeth into one. When the average American dreams about cookies, this is what they see.

Meanwhile, the rest of the world doesn't dream about "cookies," it dreams about "biscuits." And when it does, it sees a thin, round disc called a Marie (or a Maria, if you speak Spanish). Along with its siblings the Osborne, the caraway-scented Abernethy, France's golden-edged petit beurré and the deceptively named Rich Tea (which isn't very rich and contains no tea), the Marie counts itself among a class of "hard, sweet biscuits" that, to much of the world, personify the cookie -- or, rather, the biscuit -- in its purest form.

But not to us. While immigration may have widened its profile a little, "a few years ago, if you'd asked an American what a Marie is, I'd think maybe one in 100 would know," says Jeffrey A. Zeak, pilot plant manager, American Institute of Baking (AIB), Manhattan, KS. But if you posed the question to Hispanics, Europeans or South Asians, "they would know straightaway." It's a telling lesson, he says, in the disconnect between "what we as Americans know as cookies and what other peoples of the world know as cookies."

The differences are striking when comparing formulations. A typical Marie contains about 21% sugar and 16% fat on a flour-weight basis. No wonder that it tastes more like an animal cracker than a "proper" cookie. By contrast, one of our own chocolate-chip cookies, and not necessarily a premium one, might pack 52% sugar and 48% fat. This doesn't surprise Zeak. "Traditionally, cookies from other parts of the world are leaner in formula -- much lower in fat and sugar than what we're used to," he says. "A high use of sugar is definitely an American thing." In fact, the chief complaint he hears from international colleagues who visit the AIB is that "American baked goods are way too sweet."

They're also much softer -- again unlike a Marie, which doesn't just taste like an animal cracker, but is also flat and crisp like one, too. "It's processed very much like you would process a cracker," Zeak continues, with its dough temperature rising upward of 100?F during mixing, and the cookies rotary-cut on a sheeting line. And because a hard, sweet-biscuit dough develops more gluten than a fat- and sugar-rich dough for a wire-cut cookie, manufacturers often supplement the formula with sodium metabisulphite to reduce the disulphide groups in the gluten network and reign in the dough's springiness.

A nation weaned on jumbo chocolate chips and co-extruded soft-center cookies may have a hard time swallowing a hard, sweet biscuit like the Marie. But then how do you explain our enthusiastic embrace of biscotti -- hardly a soft sweet biscuit?

A biscotti, like its cousins zwieback, mandelbrot and the British rusk, undergoes a dual-baking procedure that originated in ancient times to intentionally rob the biscuit of its moisture. To recreate that process, Zeak says, manufacturers first use a depositer to extrude chubs of dough onto sheet pans that go into the oven for the first bake. "There might be larger operations that use band ovens," he explains, but "for the most part, biscotti are done on an intermediate scale in rack ovens." Once the loaves have finished their initial bake, manufacturers remove them from the oven for slicing, perhaps in a standard bread slicer that they've modified to produce a thicker, biscotti-appropriate wedge. Then, they pop the slices back into the oven for a longer time at a much lower temperature, driving off that last bit of moisture and giving the cookies their crunch.

They're a far cry from the gooey bakeshop ideal, but so long as we have something sweet and liquidy in which to soften our biscotti, we give them our endorsement. So, too, do manufacturers, says Zeak, who love the product from a shelf-life standpoint. "They can just sit in the jar in the coffee house for a while, provided they don't start to pick up moisture and go soggy on you" -- a fate that proper storage and packaging prevent on the consumer's end, but that sound product design can preempt upfront.

For example, in response to the clamor for low-carbohydrate baked goods, Bonnie Gorder-Hinchey, director of culinary services at the Hazelnut Council, Seattle, has experimented with swapping some of the wheat flour in a biscotti formula for ground hazelnut meal, available as a free-flowing flour ground to a particle size of 2 mm. The results, she says, have been impressive. In addition to lending the cookies an earthy, hazelnut flavor, the meal doesn't absorb moisture from the air. "You can't completely eliminate the flour," she cautions, as "you need something to hold the dough together, and flour provides the gluten structure to bind the biscotti." But replacing anywhere from 25% to 75% of the wheat flour with the nut meal has led to success.

A light hand with humectant sweeteners, like honey, molasses and high-fructose corn syrup (HFCS), also keeps out atmospheric moisture that would otherwise turn biscotti limp. Besides, holding down sweetener levels in general is consistent with the biscotti's stated mission: to be a carrier for sweet liquids, rather than to carry the sweetness itself. Admittedly, biscotti have taken on considerably more sugar since hitting the Starbucks circuit. "They probably didn't have the variety that they have now, as far as the chocolate coatings and cranberries and inclusions go," observes Zeak. "But it's just another excellent example of how an ethnic cookie filtered its way to the United States and became adapted to U.S. tastes."

When it comes to matters of taste, Klaus Tenbergen, a chef-instructor in culinary arts at Kendall College, Evanston, IL, can't emphasize enough that everything is relative. Schooled in the European baking tradition, he believes that a gulf the size of the Atlantic separates the European palate from the American, especially on the subject of cookies. And he's not just talking about our penchant for sweetness and softness, either.

"For example," Tenbergen says, "what they call speculaas in Holland -- here we call them windmill cookies -- are thin, crispy little spice cookies. And they are not as flavorful here as they are in Europe. They're very different, in my opinion. They just don't taste the same." Namely, they're not as spicy. Americans may know no limits for heat when it comes to habanero salsas and extreme cinnamon red-hots, but often hit a wall of cognitive dissonance when they find too much spice in traditionally sweet baked goods.

That wall, however, shows signs of cracking. White-tablecloth restaurants have long promoted blending sweet with savory, and they've reversed the flow, bringing the spice into the sweet, and giving us dried chiles in chocolate cakes or cracked-black-pepper crème brûlée. And if diners ooh and aah at the restaurant, might they not give such combinations a go at the cookie counter?

With the proper implementation, they just might. Tenbergen, in particular, appreciates black pepper's ability to bounce flavors off one another in a cookie, setting up a contrast between its piquancy and the basecake's mellow, buttery richness, or its deep, caramelized notes. "It's not there to give a taste on its own," he emphasizes, "but to bring out the flavor of the other ingredients." Translation: You're not formulating a black-pepper butter cookie, you're formulating a butter cookie that happens to benefit from black pepper.

The presence of spices and herbs in cookies may seem novel to most Americans, but it actually has a history as long and storied as the trade routes that carried those seasonings East to West. Since the Middle Ages, fennel, sage, lavender, and rosemary, as well as cardamom, caraway, coriander, and ginger, have been fair game in cookies, as have citrus oils and flavored waters. "We have a lot of orange and lemon peel, and a lot of candied peel," in European cookies, Tenbergen points out. "And in Greek baking, they use rose water and orange-flower water, which they sprinkle on their cookies once they come out of the oven." In a very real sense, the notes of blossoming orange in a croquant du Périgord or Poland's poppy-seed-filled ciastka makowe come from the same rootstock as the seemingly unrelated kâk bi halîb, a yeast-leavened Lebanese cookie shot through with the flavors of dried marjoram, anise seed and mahlab, a seasoning made from the ground pits of black cherries.

Cookies in this classic European tradition "were made from ingredients available in the home at the time," points out Mark Cocco, technical manager, sweet goods business unit, Givaudan Flavors, Cincinnati. "Even today, many of the commercially prepared European cookies that we see on store shelves still contain these same ingredients." But factors such as the source plant's maturity, variety and growing conditions can send the quantity, quality and cost of its output swinging. What's more, unsophisticated expression and distillation techniques used to extract spices and essential oils can sacrifice the freshness and vibrancy of flavor.

Therefore, flavor systems designed to deliver the right notes at the right time grant the advantage of stability and standardization. This, Cocco says, "reduces the processing variation typically experienced from natural materials." Newer techniques, such as supercritical carbon-dioxide extraction, which keeps volatile flavor oils below 50?F throughout the extraction process, improve upon methods notorious for obliterating the most-vulnerable, yet desirable, flavors. By encapsulating those flavors in gum or maltodextrin carriers, flavor suppliers add another measure of security. Some encapsulates even partially dissolve upon hydration and then, like double-acting baking powder, delay releasing the remaining flavor payload until the application of heat.

Technology has made cookie manufacturing easier vis-à-vis fats, too. In a perfect world, we might make all our cookies with sweet creamery butter the way Grandma did. Even in a perfect world, butter isn't perfect. It has incredible flavor going for it, and it makes shortbread incomparably rich and crumbly. But from a commercial processing standpoint, Grandma didn't know best.

That's because cookies need fat for reasons other than just flavor and texture. When we cream fat with the other ingredients in a cookie formula, the liquid-oil fraction surrounds the sugar granules and flour proteins in a hydrophobic coat, delaying the former's solubility and preventing the latter from developing too much gluten. And like oil on a bicycle chain, it makes for a smoother, gentler mix and keeps the dough from sticking to surfaces during subsequent processing. Meanwhile, the solid fat traps air in tiny dough pockets that, as the cookie bakes, fill with carbon dioxide released from leavening reactions and steam from water vapor. The upshot is a cookie with good lift, a fine crumb and the consistent size that the folks in packaging covet.  

A fat's ability to perform these creaming functions, as well as others, depends in large part on its crystalline structure as it solidifies. In general, the smaller the crystal is, the better the fat's suitability to cookies. Of the alpha, beta and beta-prime crystalline forms into which fats tend to solidify, beta-prime shows the best potential for producing an ideal cookie fat. Butter's crystalline structure doesn't follow this model and, consequently, butter doesn't display the best creaming properties. "You want a plastic fat that will actually entrap air and hold that air," says Lynn Lawrence, technology manager, HUMKO Oil Products, Cordova, TN. "Obviously, if you've got a product that is too firm, it's not going to entrap air. It's going to be brittle and create lumps in the cookie dough. On the other hand, if you've got a product that's too soft -- say, butter at a warm temperature -- it's just going to liquefy and that won't entrap air, either."

Another problem with butter, Lawrence continues, is that "it has a very narrow plastic range, whereas a lot of the hydrogenated vegetable oils have a very wide plastic range." In other words, while butter goes from a solid to a liquid with only a few degrees rise in temperature -- melting out at about 90?F -- other solid fats, such as hydrogenated vegetable oils and many tropical oils, will remain solid over a wider temperature range.

The temperature range in which a plastic fat goes from a solid to a liquid (or vice versa) is known as the solid fat index (SFI) and, for the purposes of a cookie, we want it to span a roomy gap. "An all-purpose shortening would have more tolerance over a wide temperature range than something with a steeper melting curve, like some of the specialty fats," Lawrence says. "There are some fats that are designed specifically to have melting points around body temperature -- say 95? up to maybe 100? or slightly over 100? -- and that's preferred for eating qualities so you get a better mouthfeel." However, they aren't as compatible with high-temperature mixing operations or in production facilities without precision temperature control as something with an SFI in the neighborhood of 95? to 115?F.

A fat's plastic range takes on particular significance in the production of delicate puff-pastry cookies, such as the heart-shaped palmiers and jam-filled puff-pastry shells common in European bakeries. Traditionally, pastry chefs would repeatedly roll out a butter-rich sheet of dough and fold it back in on itself to create alternating leaves of butter and pastry that, as they baked, puffed and separated with the release of steam. Scaling up to commercial proportions, manufacturers now laminate the cookies, usually co-extruding a sheath of fat atop the dough and then proceeding with fold-and-sheet operations that yield hundreds of flaky layers.

The process calls for the use of a roll-in fat whose consistency parallels that of the dough. "In other words," Lawrence says, "you don't want the fat to be much firmer than the dough, or once you run it through the sheeting or folding process, you can actually puncture the dough layers and not get any volume. On the other hand, if the fat is much softer than the dough, when you go through the sheeting and folding, it's likely to work into the dough so much that you don't create the distinctive layers of dough and fat." Also, the fat has to melt cleanly and quickly in the mouth, or the cookies risk hitting the palate as waxy.

Manufacturers like working with roll-in margarines because they not only clinch these criteria, but at roughly 20% water, they offer a burst of steam during baking. Options based on blends of palm and vegetable oils are particularly effective thanks to the palm oil's tendency toward beta-prime crystallization and its plasticity, which suit the needs of puff pastry without requiring hydrogenation. As Salleh Kassim, executive director, American Palm Oil Council, Torrance, CA, says, that means a product with fewer trans fats to apologize for. And, he adds, "The good thing about palm is that it has the highest content of vitamin E and of tocotrienols. Among other oils and fats, including soybean oil, we see very little content of tocotrienols. Palm has about 600 ppm, which is very high." In combination with palm's high degree of saturation, these antioxidants make it one of the most naturally stable oils available.

That kind of stability comes in handy in cream fillings, a mainstay in European sugar wafer and sandwich cookies, that are susceptible to oxidation. Chalk it up to the amount of fat the creams contain -- the fat-to-sugar ratio runs as high as 2:1 -- coupled with their exposure to oxygen during aeration and as they sit between the basecakes in the finished cookie. Not surprisingly, stable hydrogenates and tropical fats, like coconut oil, see a lot of play in cream fillings, as do fats protected with the antioxidants butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT).

The fat in a cream filling not only needs stability, but has to help it "glue" the cookies together. Given those requirements, Lawrence says that an appropriate fat will have a "high solids content at room temperature to allow the cream to set." Go with too high a solids content, however, and the filling could crumble away from the cookies. Then again, one that's too low in solids, he notes, may be "so soft that it oozes out from between the two basecakes." Adding it all up, he says, the fat has to produce a filling that "can be deposited on that basecake and stay put."

It comes back to the fat's SFI, which is also instrumental in determining the cream's eating qualities. While an ideal basecake filling should have a broad SFI for processing flexibility, in a filling, Lawrence explains, "You want the fat to melt out rather quickly. If you have a fat with a very flat slope, you're going to get that waxy mouthfeel. So you're looking for something that has a melting point around body temperature. Something around a 95? to 100?F melting point would be ideal."

Back in the olden days, European bakers didn't fuss over hydrogenates versus tropical oils or steep versus shallow SFI curves. They just sandwiched their cookies with marzipan, a paste of sweetened ground almonds and egg white, or with similarly sweet and creamy hazelnut praline. Manufacturers today can source industrial grades and quantities of marzipan and hazelnut praline, as well as unsweetened almond butter and hazelnut paste -- which, at a minimum of 90% roasted hazelnuts ground to about 25 microns, looks, tastes, feels and behaves like hazelnut butter.

"If you're adding hazelnut butter to make a filling, it has some really positive aspects," Gorder-Hinchey says. "For example, if you're making gianduja (a hazelnut-flavored confectionery chocolate developed in Switzerland and popular as a cookie filling or icing), the fats in the hazelnuts have a lower melting point than the fats in the chocolate. So, by adding hazelnut butter to the chocolate, you're lowering the melting point and giving yourself a smoother, creamier mouthfeel." That's a eutectic mixture: a combination of two dissimilar fats that, when melted, blended and recrystallized as one, melt at a lower temperature together than they would melt at on their own.

Manufacturers have to take care when blending dissimilar fats, however, as changes in the fats' crystalline structure can lead to problems ranging from unwanted softening to unsightly fat bloom. Gorder-Hinchey also warns that without proper stabilization, oils can separate from the hazelnut paste and wander into the basecake. "You definitely want to have something to hold everything together so you don't get the separation," she says. Once again, fats with a beta-prime crystal formation do the trick: Hydrogenated cottonseed oil, hydrogenated palm oil and hard palm stearin emulsified with vegetable oil-derived mono- and diglycerides are all worthy candidates. Usage levels in the 2% to 3% range should suffice, as higher concentrations dry out the butter's texture and compromise its flavor.

The companions of European nut-paste and cream-filled sandwiches are Linzer cups, kolachy, and buttery cookie-press rosettes topped with a thumbprint of fruit. Simple jams served bakers of the past in good stead, but manufacturers concerned about shelf life, smudging and moisture migration need to formulate their fruit fillings with a little more science in mind.

The key, says Scott Summers, director of technical services, Tree Top Inc., Selah, WA, comes down to water activity (aw). "You want your product shelf stable at room temperature, and that means that the water activity in the fruit material has to be pretty low. Typically, that 0.60 water activity is the magic number: If you're below that, then you're shelf stable."

Hovering below this value not only avoids microbial issues but also produces a filling whose viscosity and solids content adhere it to the cookie and hold remaining water out of reach of the basecake. Starch, gum and hydrocolloid stabilizers confer moisture control beyond what the sugar solids provide, but as with the use of stabilizing fats in nut butters, too much of a good thing makes a filling so tough that it might have come from the fruit of the rubber plant.

In fact, many low-water, high-viscosity fruit fillings come from a more prosaic source: apples. As Summers explains, hygroscopic, low-moisture apple powders, either drum- or air-dried and often treated with sulfur dioxide, supply the filling base. "Then they add a little water and a lot of sugar to the powders, which bind that water and become shelf stable."

Apple grind particulates or evaporated fruit pieces go into the mix to give the fillings more fruit identity, says Summers, and "because the water activity in the filling is very low from the low-moisture powders and all the sugar, it takes the 22% to 24% moisture of the grinds and equalizes it lower into a shelf-stable situation." While the air-dried apple powders stay the same, it's the fruit concentrates, dried fruit, colors, flavors and dried-fig or dried-plum pastes that change perceptions of which fruit the filling actually is. "Basically, you've taken the exact same filling and you've characterized it. And it works pretty well. Apple was pretty cheap and blueberries and strawberries were not," he adds.

As for which sweeteners work best in fruit fillings, Summers says that perceived flavor and cost have more bearing on the decision than anything else. Dextrose, with its heat of solution at -25.2 calories per gram at room temperature, effectively "steals" heat from the surrounding system in order to dissolve. This gives it a "cooling" taste that some claim complements a fruit filling. HFCS also frequently shows up in U.S. formulations, but invert syrup is the sweetener of choice overseas -- perhaps, Summers suggests, because of European concerns about genetically modified corn. "We do some business with the United Kingdom where we are using fructose-infused apples in certain applications, but we are getting these high-amylase corn derivatives that suppliers can guarantee are not GMO. And we have to certify that these corn syrups are coming from those raw materials."

Sweetener choice is more functionally relevant to the production of the tubular, cream-filled cookies called pirouettes or gauffres. These thin wafers resemble crêpes more than cookies in that they're made not from a stiff, oven-baked dough, but from a fluid batter low in flour and fat that's deposited onto a heated flattop surface. There, the wafer "bakes" very quickly, after which the finished product is scraped from the surface, rolled immediately while still soft, allowed to cool and harden, and then is injected with cream. Timing is of the essence because if the wafer rests too long before shaping and the sweetener starts to crystallize, it loses its flexibility, turns brittle and cracks.

So which sweetener works best? Because a pirouette succeeds or fails on its pliability, it needs one that shows sufficient fluidity. That suggests a syrup, which enters the mix already in solution form and only becomes more fluid as it heats. However, the quicker the pirouette firms up after rolling -- and the longer it stays that way -- the better. So, a humectant corn or invert syrup, although it promotes batter fluidity, also scavenges atmospheric moisture that threatens the cookie's texture. Traditional recipes often split the difference and go with castor or superfine sugar: Because of their surface-area-to-volume ratio, these sweeteners -- equivalent to about 10X granulated sucrose -- dissolve quickly for a flexible batter. And while they are more hygroscopic than coarser granulations, they're still not going to absorb as much water as sweetener syrups.

In general, most cookie formulas use either standard granulated sucrose or a granulation known as "bakers' special," because, Zeak says, "it's economical and it's functional. There are certainly other types of sweeteners out there, but can you use them at the same levels you would normally use sucrose and get the same functional properties -- the same type of spread, the same type of processing characteristics?" Corn and invert syrups are valuable for moisture retention in some cookies, but "we're never going to have a cookie that's made out of 100% high-fructose corn syrup or 100% standardized invert," he says. "The cookie dough starts taking on properties like stickiness that make it more difficult to process."

Specialty sweeteners, such as honey, molasses and treacle (a sugar-refining byproduct similar to molasses and common in traditional British cookies) retain moisture, too. They also contribute color and unique flavor. But can a manufacturer formulate a cookie, such as an old-fashioned German Lebkuchen, sweetened wholly with honey? After all, people did it for centuries before sugar came into wide currency.

Zeak says that it can be done. But, he asks, "What are the ramifications going to be?" For one, given honey's cost, it'll be a very pricey cookie. Number two, because honey is high in reducing sugars, he says "the product is going to brown an awful lot no matter how low a temperature you use." That triggers another set of economic considerations because, he continues, "we're making cookies as rapidly as we can. If an ingredient causes something to brown more, yes, we could adapt our process and bake at a lower temperature for a longer time. But that's going to slow down production. So instead, I'm going to want to use something that won't give me as much browning. Now, I might want a honey characteristic in my product -- like in a honey graham cracker, where there's a little bit of honey. But that little bit of honey is just added for its flavor. It's not the main sweetener. The main sweetener is going to be sucrose."

The main sweetener may be sucrose, but the main ingredient in a cookie -- the basis that sets the structure -- is wheat flour. Unless the cookie is a meringue. That we've even come to consider the meringue a cookie is probably an accident of history, as the puffy clouds of sweetened, whipped egg whites resemble confectionery more than what we normally think of as a baked good. In fact, because the egg-white foam in a meringue insulates the product so efficiently, we don't so much bake meringues as slowly dry them out in a low oven. Two to three hours at 200?F isn't an uncommon protocol, as the time and temperature combinations seen in most other cookie baking would parch the meringue's surface while leaving the center raw.

Meringues are finicky creatures in other respects, too. The formulation is straightforward enough: egg whites, sugar, perhaps cream of tartar. But they demonstrate in real time the peculiarities -- and the wonders -- of protein science. The simple act of beating the whites begins denaturing the albumen proteins, and as they unravel and coagulate with each other, they form a structure that encloses finely dispersed air bubbles that have been whipped into the emulsion. Ovomucin proteins form an insoluble film around the bubbles, stabilizing the foam; globulins give it viscosity; and, as the meringue bakes and the air pockets expand, the coagulation of ovalbumin helps the network.

Overbeating throws off the whole system because it prematurely sets the foam and prevents its expansion under heat; extreme overbeating collapses the protein network entirely. Salt weakens protein bonds, encouraging moisture loss from the foam. And because high temperatures damage the albumen proteins and reduce their foaming properties, egg processors carefully control pasteurization temperatures (federal law requires pasteurization for all egg products) and, when manufacturing dried egg whites, spray-drying conditions.

But the real bugbear to achieving a stable, buoyant meringue is fat. By interfering with the proteins' ability to coagulate, even the smallest amount of lipid -- whether from other ingredients or even as residue on equipment -- can destroy the emulsion. That makes yolk contamination of the white a serious consideration. Glenn W. Froning, Ph.D., professor emeritus, Department of Food Science & Technology, University of Nebraska, Lincoln, says: "It's very important in a meringue that you specify a foaming-type egg white. You don't want an egg white that has any substantial contamination of yolk, for instance. Egg-product manufacturers normally push to make sure that they don't have more than 0.05% yolk contamination. If they get over that, they start to get a very big influence on the foaming properties of the egg white." Surfactant foaming aids, such as sodium lauryl sulfate, help restore lost foaming properties, but they can't work miracles.

On the bright side, certain ingredients, including acids like cream of tartar, actively stabilize egg-white foams. They protect against overcoagulation by keeping albumen proteins from bonding too firmly, lending an elasticity to the foam that, like the give in a structure built in earthquake country, actually makes it stronger. Acids also improve the white color of the foam.

Sugar acts as a sort of spacer between egg white proteins, which delays foam formation, Froning says, but also prevents overcoagulation and stabilizes the meringue during baking. The sweetener's moisture-retention properties account for that latter function, as it holds water in the foam until it has had time to set. Any granulated sugars will do -- even brown -- although finer granulations produce less-voluminous meringues. In an Italian meringue, a sugar syrup heated to the hard-ball stage, or 250?F, both sweetens the cookie-confection and gives it a finer eating texture.

That's pretty impressive for a mixture of egg white and sugar. But meringues are impressive specimens -- indeed, as are all cookies, whether made of puff pastry, rolled-up wafers, a hard sweet biscuit dough, or lots of sugar, fat, and big, gooey chocolate chips. As Zeak says, "We could take a cookie in its purest form, where we're talking about flour, sugar, fat, and that's basically it, and from there, we can go in a million different directions."

Kimberly J. Decker, a California-based technical writer, has a B.S. in Consumer Food Science with a minor in English from the University of California, Davis. She lives in the San Francisco Bay area, where she enjoys eating and writing about food. You can reach her at [email protected] .

Back to top