Skip navigation Harvest of hunger


1. Putting food on the table

2. A need to breed

3. Competing, or meeting?


Zambian boy stands amid the remains of the usually-drought resistant sorghum. 2002 Photo by Brenda Barton, World Food Program.







New plant varieties and more efficient production are a key reason for the 146 percent rise in world grain production since the 1950s.



These bits of genetically engineered corn plants will be grown into whole plants, then tested for fitness in the field. © David Tenenbaum


Overemphasizing the untested
After 10,000 years of success, plant breeding has suddenly taken a back seat to genetic engineering, says University of Wisconsin-Madison corn breeder James Coors, even though the new technology has yet to prove its utility. One way the shift pinches is in a drying-up of interest in the field.

Thin boy, hand on hip, shirt torn, hanging off of him, gazes off to his left. Dry, brown field behind him. Twenty years ago, says Coors, it was easy to attract graduate students. Today, "Students have this impression ... about the wonders of biotech, what it represents in terms of how fast we can do things."

Before you yawn and grab another potato chip, remember that new varieties of plants, coupled with more efficient production systems, are a key reason for the 146 percent rise in world grain production between the 1950s and 1990s.

While we're doing statistics, 60 percent of the 11 million childhood deaths in developing countries each year are associated with malnutrition, and 160 million children under five are stunted by protein malnutrition. That's from Gro Harlem Brundtland, director of the World Health Organization.

Add it up: With a soaring world population, hundreds of millions already malnourished, a shift toward grain-gobbling animal agriculture, and little prospect for plowing more acreage, the only way to keep dinner on the table is to increase productivity on each hectare.

In other words, to keep breeding better plants.

Uses sex, but not sexy
Seeds, some beginning to sprout, in a round laboratory dish.Plant breeding takes money, effort and dedication. Moving traits with conventional breeding requires a lot of plant sex, to mix up the genes from two parents, and then many years worth of trial and error. In today's fast-paced scientific realm, the notion that a plant breeder might spend a lifetime gradually improving a crop seems like black-and-white television -- useful in its day, but "entirely last week."

"In the old days," says Coors, an agronomy professor, "people would write in grant applications, 'We are doing such and such so by the time we're done, we will have seed that can be used.' Now a grant, says, 'I'm going to deal with such and such a [biochemical] pathway, so in the future, plant breeders can use the possible genetic mechanism to produce a seed that might be of use."

The shifts are obvious -- and momentous, Coors says. "You are no longer concerned with producing the seed, but with a technology that somebody else might use." But, he adds, the role of the public university has changed from a source of seedstock to a supplier of highly trained talent for the private sector, so "Nobody in the academic sector is interested in producing the seed. That's regarded as straightforward and mundane work, not as a legitimate scientific endeavor."

A cautionary tale
A fundamental difference separates biotech from conventional breeding: the notion that moving one gene will produce predictable benefits. "Many of these things that are being done in biotech are really neat gene manipulations, " Coors says, "but in many cases, it's very difficult to get these things working in plants. They have to be grown, harvested by farmers, go through conventional grain channels and actually make it to market."

Neat rows of round containers holding soil, worms and seeds.
Worms in each container are exposed to genetically engineered corn containing an insecticide gene. If the worms die, the variety may be able to kill insects in the field. © David Tenenbaum

One promising single-gene discovery grew from work by the late Oliver Nelson, a University of Wisconsin-Madison geneticist. Nelson identified a gene in corn that increased the production of the amino acid lysine, yielding a grain with more complete protein.

CIMMYT, the international center for corn and wheat research, adopted Nelson's high-protein corn in its program to breed crops for agricultural programs in the developing world. The organization's specialists used conventional breeding to move the gene into a useful variety, Coors says, but the task was difficult. "The gene softens the kernel, and that brings on insect and fungal problem, so the grain can't be stored well."

After nearly giving up, CIMMYT is finally releasing high-lysine varieties. Although the transfer involved the use of conventional techniques, not genetic engineering, Coors says the story shows the complexity of single-gene manipulations.. "It took 30 years of breeding to cope with all the problems that came with that one gene."

Editing evolution
In the olden days, plant breeding involved crossing varieties, looking at the progeny, selecting the best ones, crossing, selecting, and crossing again in a cycle that might be repeated dozens of times. "We were dealing with many genes at once, endlessly," says Coors, "but that's what evolution is about."

Red and yellow pots of wheat stalks behind a fence. Wheat is a huge crop worldwide that is constantly being bred for improved yield and disease resistance. © David Tenenbaum

The single-gene focus of biotech could not be a stronger contrast. "It's so fundamentally different from how evolution has proceeded over the last three billion years," Coors adds. Natural selection - like conventional plant breeding, has not typically acted on single genes, but rather "on the whole organism's performance. We have altered the rules of the game."

Coors acknowledges that the single-gene approach will probably succeed in some cases. "We are going to be making progress, but it will be a different type of progress, it will involve different talents."

Anybody think about crossing genetic engineering with conventional breeding?



  The Why Files  

There are 1 2 3 pages in this feature.
Bibliography | Credits | Feedback | Search

©2002, University of Wisconsin, Board of Regents.