At an open house at DuPont Pioneer’s Dallas Center Corn Research Center near Des Moines, Iowa, retired corn breeder Bill Ambrose marveled at the tools available today to do the job he did for nearly 40 years.
“We could do a few hundred things and they do mega thousands of things,” Ambrose said.
In his day, he said, much more was done by hand—a team of five might harvest 250 plots in a day, while now “these guys that work in this place here have got huge combines that they can harvest 250 plots an hour,” he said.
For decades now, farmers and seed scientists have seen yields improve. But they’re not satisfied. Fundamentally, what drives corn breeders is the one thing farmers want. And that desire persists.
“Grain yield, grain yield, grain yield,” said Dave Bubeck, research director for the central business unit at DuPont Pioneer.
After all, Bubeck said, that’s what farmers get paid for—the amount of sell-able crop grown on their land. “And we want to maximize that, so that’s always foremost in our selection criteria every step of the way.”
Seed companies today use the advance computational power of computers and databases to analyze the genetics of individual seeds and eliminate anything that’s inhibiting yield. But exactly what to breed for varies depending on where the seeds will be planted and other factors. Bubeck said there’s no one answer for all farms.
“It really comes down to an individual farmer’s product management approaches, the soil type, the whole regimen of farming practices that an individual farmer has,” he said.
The vital seed has evolved from humble origins into big business. At DuPont Pioneer, grain yield has increased about 1-2 percent per year for decades and the company projects consistent, continued yield growth. That means in another generation we could be seeing yields as unimaginable today as today’s yields were 25 or 50 years ago. Weed, insect and drought resistant seeds are readily available now.
In addition to traditional breeding techniques, of course, the seed companies now rely on genetic modification. David Fischhoff, Monsanto’s technology, strategy and development lead, said in addition to those two tools the company also works closely with farmers to achieve agronomic improvements.
“An example of that would be the work we’re doing in what we call integrated farming systems, which is really the application of information technology to enable growers to better place different hybrids and varieties on their fields,” Fischhoff said, “to plant them, if you will, at the right density—that is, how closely spaced the plants are to one another, which has a big impact on yield.”
The future could include seeds that process nitrogen differently, for example, or produce greater nutrition. But these developments don’t appear overnight. The seed companies typically expect a new product to travel a 10 year path from idea to commercialization, with in-field testing during several of those years.
Iowa State University agronomy professor Kan Wang says even with genetic engineering, scientists still have a lot to learn. It’s no magic bullet.
“Sometimes we introduce a piece of a gene there and then oh, well that plant indeed performed better, a little bit better for one aspect,” Wang said, “but on a cost of the other aspect.”
Seed companies don’t blindly breed for higher yield because the result could be a plant with lots of grain that can’t be harvested because the stalks don’t stand.
At Iowa State’s seed lab, lab manager Mike Stahr sees myriad potential directions for seed development.
“They’re going to keep on looking for ways to make the seeds grow better, be easier to harvest, be easier to store,” Stahr said.
Perhaps eventually farmers will plant various seeds within a field, or even row.
“So someday there may be such a thing as they apply different seed types in a different part of the field,” Stahr said. “Maybe one part is low and needs a variety that can handle wet feet and another part of the field might be ideal conditions.”
Scientists and farmers both know this: the quest for the perfect seed won’t end with just one.