How We Can Tame Overlooked Wild Plants to Feed the World

A HAND-PAINTED WOODEN sign marks the entrance to Steven Cannon's community garden, tucked between a sidewalk and some train tracks in Ames, Iowa. It depicts the iconic image of a seedling poking from a mound of dirt. At the far end of the garden, Cannon, a tall and reedy geneticist for the US Department of Agriculture, digs into the soil with a shovel and then his bare hands, pulling up fistfuls of lumpy roots. Strip the scene to its essence—ignore the cars driving past and the power lines strung overhead—and you could be watching a Neolithic farmer. They collected seeds from wild plants, buried them near their homes, and harvested the crop, hoping it would be bigger and better than the last one. That simple act—agriculture—came to define us as a species.

Cannon isn't trying to re-create the past, though. He's inventing the future. On this fall afternoon, his team is harvesting tubers that resemble dark-skinned fingerling potatoes. They're called Apios americana, the potato bean—a legume endemic to North America. Native Americans gathered them and may even have served them at the first Thanksgiving. European settlers found them thriving in their cranberry bogs—places with low light, few nutrients, and bad soil. But they didn't bother domesticating them into an agricultural staple.

After a couple hours of labor, Cannon's harvest is complete. A dozen rubber bowls overflow with dirt-crusted tubers. Still, he is disappointed. “We were hoping for a little better yield,” he says. “This is about average.” Average is fine if you're just messing around in a kitchen garden. But Cannon is up to something far more essential. The potato bean is part of his plan for remaking our food supply from the ground up. He doesn't want to just grow Apios. He wants to turn it into a new crop that could help feed the world.

MASHED POTATO BEANS

  • Yield: 6 servings
  • Ingredients:
  • 2 LBS potato bean tubers
  • 1 C half-and-half or milk
  • 8 TBS unsalted butter (Potato beans, which have three times the protein of their namesake starches, might be a little dry, so this recipe compensates with extra fat.)
  • 6 TBS soft goat cheese
  • ¼ TSP freshly ground black pepper
  • ¼ TSP freshly ground nutmeg
  • salt to taste
  • Prep:
  • Peel tubers, then boil until soft, about 10 to 15 minutes. Drain and mash. Add half-and-half or milk. Mix in butter and then pepper, nutmeg, and salt. Serve topped with goat cheese.

We need new crops. Thousands of years of breeding and decades of genetic modification have made the crops we sow predictable, easy to harvest, and capable of feeding more than 9 billion people. But they are also vulnerable to disease, pests, and the whims of weather. That's troubling, because global warming is bringing more disease, more pests, and more whimsical weather. On current trend lines, global wheat and soybean harvest yields could fall by nearly 30 percent by midcentury. Corn yields could drop by 7.5 percent. In the baking-hot European summer of 2003, plant growth fell by 30 percent. By 2050, that kind of summer will be the new normal. “Suppose the US breadbasket ends up with a climate like Texas,” Cannon said at a genetics meeting last year. “We need to look to species already adapted to extremes.”

The potato bean is one of those species. Versatile like a potato, protein-rich like a bean, with a flavor vaguely like a starchy peanut, Apios does well in both dry and soggy soils. And there are plenty of others like it. Roughly 18,000 species of legumes grow around the world. They're packed with protein and help fertilize the soil. Yet people have domesticated fewer than 50, and commonly eat only half that many. Cannon has assembled a short list of additional candidates: marama beans, yehub nuts, lupine, and a bunch of other so-called orphan crops, wild edible plants that could change the face of agriculture if someone could just turn them into reliable crops.

Domestication made humans the first species on earth to have a secure, reliable food supply, enabling the development of culture and technology and medicine. Every facet of modern society is built on its back. Yet somewhere along the way, we stopped innovating. Cannon is one of a small but committed band of researchers who are quietly trying to create new crops. They're messing with new versions of wild sunflowers that have bigger, oilier seeds and don't need as much water. They're working on a genetic rewrite of the chickpea, selecting traits that will help it thrive in a warming world.

Climate change is making the mission imperative; the genetic revolution is making it feasible. That harvest from Cannon's neighborhood garden might have been disappointing, but it was also the first shot in the next green revolution.

Domestication is evolution—with human beings at the controls. When nature is in charge, evolution selects living things based on traits that favor their survival; we humans select instead for traits that promote yield, flavor, predictable growth, and resistance to attack. We started doing it mainly because we could. In the Paleolithic era, fluctuations in climate made it hard for groups of humans to rely too heavily on particular plants. But when the climate stabilized after the last ice age, around 12,000 years ago, we could pick and choose. Around the world, farming impulses stirred in different societies. People from divergent cultures began actively managing many of the same wild plants during the same period. From the several thousand plant species that prehistoric people regularly relied on for food (out of roughly 50,000 species that are edible), our ancestors chose just a handful, all grasses, to form the bedrock of their diets.

Those domesticated organisms often bear little resemblance to their wild ancestors. Ten thousand years ago in what's now Mexico, for instance, farmers took a weed called teosinte and created corn. Teosinte's tiny ears contain only about a dozen kernels; an ear of modern-day maize has about 800. Selective breeding turned a scruffy grass into a starch-packed staple of the global meal.

It sounds simple, but humans haven't domesticated a new staple crop for thousands of years.

What domesticated food crops do look like, however, is each other. Many of the traits we humans selected are the same regardless of the species. We want plants that hold on to their seeds rather than drop them to the ground, which plant scientists call “shattering.” We want those seeds to be large and to sprout when sown, and we want all the seeds to mature at roughly the same time. Together, these features make up what scientists call “domestication syndrome,” the combined qualities that distinguish, say, corn from teosinte.

Much of what we eat today was domesticated when people were just learning to weave clothing and still thousands of years from developing an alphabet. Today we tinker—but only at the margins. Maybe we get higher yield here or resistance to corporate herbicide over there. As far as transforming plants from the wild into new well-bred row crops, though, progress pretty well stopped a millennium before Jesus ate a matzo. Macadamia nuts, kiwifruit, vanilla bean: All arose in the Common Era. But as far as major crops go? Zilch.

Today, humans rely on fewer than 150 plants for nourishment, and just three cereal crops—wheat, rice, and corn—make up more than two-thirds of the world's calories; along with barley, they own three-quarters of the global grain market. Those crops, by and large, aren't suited to a changing world. Modern humans have a chance—an imperative, even—to do it better. That's the appeal of domesticating a whole new set of plants. It's a back-to-our-roots undertaking that goes way beyond seed-saving hippies or even postapocalyptic seed banks like the vault in Svalbard. “There's something quite romantic that does appeal both to the foodies and the biotech crowd,” says Susan McCouch, a plant geneticist at Cornell University. “It's ancestral, and also critical for the future.”

Taking plants from the wild and bending them to our will is a painstaking process. But a geneticist named Lee DeHaan already has results to show—and taste. That's because today's domesticators have tools our ancestors never could have conceived of: DNA sequencing technology allows researchers to choose exactly what traits they want—guiding evolution in a fraction of the time it took our Neolithic predecessors.

On a warm day in January, DeHaan rummages around in a freezer at the Land Institute, an agricultural research lab in Salina, Kansas, and digs out a loaf of bread for me. It's sheathed in brown paper and entombed in a ziplock. I tuck it gently into my bag like something precious, because this isn't any old frozen provision. It was made with a grain that DeHaan invented. It's derived from a distant uncle of wheat called intermediate wheatgrass. Modern wheat is an annual, a plant that farmers have to sow anew every year, but DeHaan's grain is a perennial. It lives through multiple growing seasons, which means it requires less fertilizer (which in turn means a reduction in toxic runoff). Our prehistoric ancestors turned to annuals because they typically produce more seed and do well in disturbed soils, and their need for replanting from seed each year made them easier to domesticate in the first place. But perennials don't waste energy putting down new roots each year, which also means farmers don't have to expend energy breaking up topsoil.

Perennial grasses dominated the famed prairie ecosystem that once spread across Kansas. The plants, deep-rooted and tall, resisted disease and grew in thick mats that left no room for weeds. They lock carbon in the soil and are better at coping with erratic rainfall (again: climate change). Yet today, acre upon acre of former grassland in North America is planted instead with namby-pamby, needy annuals: wheat, corn, soybeans. The same goes for China, Brazil, and Russia.

Climate change is making the mission imperative; the genetic revolution is making it feasible.

Back in 1976, the Land Institute set out to reengineer existing grain crops into perennials. They started by crossbreeding conventional annual wheat with intermediate wheatgrass. It worked, but only sporadically. Then, in 2001, DeHaan showed up. He grew up on a farm in Minnesota, and he had wanted to tinker with perennial grain ever since he was a kid. (Some boys dream of driving tractors, others of creating new plants by meticulously transferring motes of pollen.) He began by working on the wheat crosses, but on the side he started from scratch, selectively breeding the wheatgrass.

Initially, DeHaan thought a perennial wheatlike grain would take 50 to 100 years of artificial selection. But as genomic technology has become faster and cheaper, the work has accelerated. Breeders sequence the DNA of seeds and then use genetic markers to select plants with the traits they want. Genes that confer low shattering in corn will likely do the same thing in wheat, or potato bean. The method is still traditional breeding—not genetic modification—but the DNA code opens a shortcut.

By 2010, what DeHaan's team had was brand-new and very wheatlike—except in the roots. Wheat roots are thin and extend down just a couple of feet; the new crop sat atop a huge root system extending 10 feet down, tapping water deep underground and effectively stealing it from weeds. The deep roots are also better at absorbing soil nutrients that might otherwise be completely lost.

The institute named the new plant Kernza, a nod both to kernels and to Kanza, the native tribes of the region and the etymological root of Kansas. “We don't want a niche market that succeeds economically,” DeHaan says. “We want something that changes agriculture in a significant way.” Kernza looked to be it.

Kernza has since attracted a growing band of collaborators, from universities (University of Kansas, University of Georgia, Kansas State, University of Minnesota) to the federal government (the USDA) to big corporations (General Mills). Lately, the World Bank and the Gates Foundation have both come calling. And this summer, 90 acres of Kernza planted in Minnesota will be harvested for Patagonia Provisions, the outdoor-clothing company's fledgling line of sustainable foods. A distillery in Ventura, California, and a brewery in Lawrence, Kansas, are also experimenting with the stuff.

Still, Kernza isn't ready to upend agriculture yet. The seeds are too small and the yield too low; it shatters, and the hulls stick to the seed, impeding milling. “Those are major barriers,” DeHaan says. He estimates it'll take 20 years to get it perfected. Not bad, considering that it took between 2,000 and 4,000 years to domesticate wheat, rice, and barley.

The bread DeHaan gave me, made from Kernza, defrosts in my car on the drive home to Colorado, and even before I unpack I take the loaf into my kitchen, slice it into thick slabs, and eat it with butter. Despite having spent several months in the freezer, it is some of the tastiest bread I've ever eaten, with a rich, earthy flavor reminiscent of rye or pumpernickel but with the lighter texture of a peasant loaf.

DeHaan also gave me a bag of Kernza flour from a shelf in the institute's underground, concrete-walled seed vault. Inspired by the bread, I bake some chocolate chip cookies with it, using half Kernza and half standard all-purpose flour, roughly following a recipe developed by a Kansan named Elizabeth Peuchen. The verdict from friends and family: thumbs up. The cookies have a slightly nutty flavor, a complexity similar to whole wheat but without its unpleasant chewy quality. Eating them feels satisfyingly subversive, like I baked the future of the world's food supply into a sugary treat.

 

When European settlers came to North America, they looked out on a vast, edible landscape … and mostly ignored it, passing over the potato bean, mesquite, and yucca fruits for the seeds they brought from home. Those settlers knew that their seeds would deliver reliable nutrition, and domestication is, frankly, hard work. Almost none of the crops we eat in this country originated here. Most of the New World foods—corn, tomatoes, potatoes, beans—came from what's now Central and South America. Even the few native crops that were arguably domesticated—a handful of fruits and nuts like blueberries, cranberries, and pecans—were probably spread naturally by birds and squirrels, not actively cultivated by humans.

These days, people are much more likely to go “food prospecting,” grabbing what their neighbors have domesticated—garlic, cashews, quinoa, mango—and planting it themselves. It's what Timothy Crews, director of research at the Land Institute, calls a food slump. “We've been satisfied with increasing our diet with everybody else's foods,” he says. That's a bummer, because local plants are the ones most likely to be adapted to local conditions.

The potato bean is a prime example. A scientist in Louisiana named Bill Blackmon spent the 1980s collecting and breeding it, assessing the features of more than 2,000 varieties. (This one grew quickly; this one had tiny tubers; this one tasted too much like dirt.) Cannon took the most promising 50 of those strains, plus a few that he and his team collected around the Northeast, and began training the weed to act like a crop.

Breeding legumes generally involves collecting pollen from male flowers with tweezers and a magnifying glass and moving it, by hand, to the females. Potato beans don't tolerate the process particularly well. “It's a complicated, almost orchidlike little flower,” Cannon says. The various pollination-related parts are hidden and uncooperative, and the pollen is only viable for a few hours a day. So Cannon is relying on the kindness of passing insects to pollinate pairs of selected varieties that are grown apart from all the other plants.

The best strains of Cannon's potato bean currently yield about half as many tubers as a potato plant. But Apios americana has three times as much protein as a potato of the same weight, making it much more nutritionally efficient. Still, as Cannon puts it, at this point Apios “isn't as nicely behaved as a potato or a sweet potato.”

Harvesting is also a challenge. Potato beans grow on long underground stems, called stolons, that require a lot of digging. “It is a vigorous vine, so it's not just going to stand up nice for a harvester,” Cannon says. Domestication of the regular old potato involved selecting for dwarf varieties, with shorter stolons. But more recently scientists have identified the genes that control dwarfing in other crops, like beans and wine grapes. With that information, Cannon is trying to turn off the right genes in Apios in the hope of dwarfing it. He thinks it will take him a few more years.

So the work continues. With each harvest, Cannon's team measures things like the ratio of aboveground material (the vines and leaves) to the underground tubers, the number of tubers, and how far apart they grow from one another (tighter-spaced tubers are easier to harvest). They sequence the DNA of plants from each line, looking for genetic markers to make the selection easier. “In major crops like corn or soybeans, tens of thousands of varieties—with known and described characteristics—are stored and available to breeders,” says Cannon. “For a ‘new’ crop like Apios, we need to start from zero.” And then they eat their research.

One afternoon, on the long wooden table in Cannon's kitchen, atop quilted white place mats adorned with pictures of apples, pears, and other fruit, Cannon lays out anApios feast—bowls of leek and potato-bean soup, plates piled with boiled and smashed potato beans topped with olive oil and sheep cheese, and a South Indian-type dish of potato beans simmered with mustard and cumin seeds, cashews, turmeric, coconut, and chilis.

The dishes are far more flavor-packed than their equivalent standard potato versions. The soup is vaguely nutty, and the mashed beans are more satisfying, like something nutritious rather than a buttery pile of starch. The Indian dish is rich and substantial without being overly filling. In all its presentations, the potato bean has a distinct legume quality, almost as if you'd crossed a lentil with a Yukon gold.

Those poor European settlers had no idea what they were missing. Sitting in Cannon's airy kitchen, I feel like I've been given a peek at an alternate reality. It's corny, I know, but those potato beans give off a faint aroma of nostalgia, a lingering hint of a lost parallel course for American agriculture. It's so close you can almost taste it.

A few miles away, in a glass-door fridge near Cannon's office at Iowa State, several dozen plastic bags crammed with potato beans wait on shelves. They are last year's harvest, measured and sequenced and ready to be roasted, fried, and sautéed. Someday they might be as unremarkable as sacks of potatoes.

CHOCOLATE CHIP KERNZA COOKIES

  • Yield: about 60 cookies
  • Ingredients:
  • 1 ½ C Kernza flour
  • 1 C all-purpose flour (Scientists are still tweaking Kernza's gluten levels. Since gluten provides elasticity, mix Kernza with wheat flour to keep things chewy.)
  • 2 eggs
  • 1 C butter, softened
  • 1 C brown sugar, packed
  • ½ C granulated sugar
  • ½ TSP baking soda
  • 1 TSP vanilla extract
  • 12 OZ semisweet chocolate chips
  • 1 C chopped walnuts, pecans, or hazelnuts (optional)
  • Prep:
  • Preheat oven to 375° F.
  • In a large bowl, beat butter with electric mixer on medium to high for 30 seconds. Add sugars and baking soda. Beat until combined, scraping sides of bowl. Beat in eggs and vanilla.
  • Beat in flour. Fold in the chocolate chips and nuts.
  • Drop dough in rounded teaspoons onto ungreased cookie sheets, about 2 inches apart. Bake 8 minutes or until the edges are lightly browned. Transfer to wire rack to cool.