(Fortune) -- "It's fascinatin' because we're in uncharted terri-tree," says the genial, Scottish, entertainingly named Hugh Grant, 52. He is the CEO of Monsanto, possibly America's most feared corporation. Monsanto dominates the agricultural biotechnology industry, whose audacious mission is to transform the genetic composition of the world's food supply. More than 80% of the soybeans and cotton harvested in this country now have at least one patented Monsanto gene in them, as does more than 70% of the field corn.
Monsanto (MON, Fortune 500) is the company best known, depending on which documentary film one has seen, for feeding Frankenfoods to our children, courting ecological catastrophe, or bringing ruinous patent suits against struggling family farmers.
On the other hand, those who don't vilify Monsanto tend to rhapsodize about it, or at least about its mission. In a world on pace to spawn 9.1 billion mouths to feed by 2050, the black magic of ag-biotech offers the only apparent prospect of salvation: crops that will be, it is promised, ever more resistant to insects, disease, and climatic stress; that will require ever less water, fertilizer, and pesticide; and that will bring forth ever more abundant, hardy, and nutritious harvests.
Whichever school of thought one subscribes to, there is now a fresh reason to fear Monsanto. It relates to the "uncharted territory" Monsanto's Grant is alluding to through his rolled r's and glottal stops. The most important genetic trait ever engineered -- the Monsanto herbicide-tolerance gene known as Roundup Ready -- is about to come off patent and, in a testament to just how young this entire business sector is, that's never happened before.
Monsanto competitor DuPont (DD, Fortune 500) claims, and federal and state antitrust regulators are probing whether, Monsanto is using abusive patent license provisions and other tricks of the trade to hobble the development of both the generic versions of Roundup Ready and a patented rival DuPont product, known as Optimum GAT. DuPont sells genetically modified seeds through its subsidiary Pioneer Hi-Bred International of Johnston, Iowa. DuPont/Pioneer asserts that Monsanto, if successful in postponing the availability of those competing products, will force farmers to switch to Monsanto's own second-generation offering, Roundup Ready 2 Yield, whose patents won't run out till 2020. Roundup Ready 2 currently costs about 40% more than its predecessor.
"The cost to farmers and consumers if Monsanto succeeds will be in the billions of dollars," asserts David Boies, DuPont's lead outside lawyer and perhaps the most sought-after American trial lawyer of his generation. "But the worst cost," Boies says, "will be the restrictions on innovation and the restraints on yields that prevent American growers from being more competitive around the world and keep us from feeding more people with less resources."
Monsanto protests that DuPont's contentions are nonsense and that its rival is cynically manipulating both public opinion and the antitrust regulators in an effort to gain leverage in a mundane business dispute. DuPont's real problem, Monsanto maintains, is that it suffered a setback in its product pipeline: A key DuPont genetic trait simply didn't work as well as hoped. Now DuPont needs Monsanto's Roundup Ready 1 to fix its problem, and it's refusing to pay Monsanto a fair licensing fee.
"I expect payment inside my patent life," says Grant, who claims that the dispute is really that simple. "The story takes longer to tell than it really should."
At the core of this controversy is something lawyers refer to as the "patent bargain." In the Constitution, the Founding Fathers gave Congress the power "to promote the progress of science and useful arts" by rewarding inventors with a lawful monopoly on the sales of their innovations for a limited period -- currently set at 20 years from the day the patent application is filed. In exchange, the inventor must teach the world how to duplicate his invention. At the end of the patent term, this know-how enters the public domain.
As Roundup Ready nears the end of its patent life -- for soybeans, that will happen in 2014 -- the concern is that Monsanto could, by forbidding rivals from even starting the multiyear process of developing, testing, and internationally registering generic versions of Roundup Ready until its patent term has expired, effectively extend its monopoly an extra five to seven years, if not longer.
The controversy springs from a legislative void. While the pharmaceutical industry has the Hatch-Waxman Act of 1984 to tell its players exactly how to transition seamlessly from patent monopolies to generic competition, agricultural biotech has no equivalent.
Far more is at stake here than a simple attempt to finagle a few extra years of patent protection for a successful product -- a common enough drill in the pharmaceutical industry. The stakes are higher because this industry differs from the drug business in one crucial respect. With pharmaceuticals, a patient with two medical needs can purchase one company's patented pill for his first condition and another firm's patented gelcap for the other.
In the world of ag-biotech, in contrast, a consumer lacks that freedom. He is constrained by the fact that a plant can grow from only one seed. If a farmer wants his crop to contain more than one patented genetic trait -- say, soybeans with the Roundup Ready trait and an insect-resistance gene engineered by a competitor like DuPont, Bayer (BAYRY), Syngenta (SYT), Dow AgroSciences (DOW, Fortune 500), or BASF -- the traits must be bred into a single seed, that is, "stacked" on top of each other.
Many farmers regard the Roundup Ready trait as so crucial that they will not buy other genetically engineered traits unless those are stacked in seeds that already contain Roundup Ready, a gene that makes crops immune to the most widely used weed killer. Trait developers and seed breeders, in turn, can't stack a trait on top of Roundup Ready unless Monsanto gives them permission to do so in a patent license.
In the view of DuPont and other alarmed observers, this situation makes Monsanto an industry gatekeeper, capable of deciding which new genetically modified traits can be introduced and which cannot. Put another way, Roundup Ready has become a monopoly platform product, much like what the Microsoft Windows operating system became in the market for personal computer software in the late 1990s.
The parallels are not lost on DuPont's longtime outside counsel Boies, chairman of Boies Schiller & Flexner, who was the Justice Department's lead trial counsel in its antitrust case against Microsoft. He argues that Monsanto's stacking restrictions are actually more objectionable than the conduct that got Microsoft (MSFT, Fortune 500) into trouble 15 years ago. "Microsoft's actions were designed to inhibit the stacking of the Netscape browser on the Microsoft Windows operating system," says Boies. "Here, you have outright prohibition of stacking genetic traits on top of Monsanto's Roundup Ready trait."
Monsanto officials charge that Boies's arguments are part of a DuPont-instigated campaign of "noise" and "fabrication." Monsanto routinely licenses stacking rights to trait developers, they say, pointing to at least nine commercial seed products on the market today in which non-Monsanto traits (including one of DuPont's) are stacked atop a Monsanto trait. The problem, they say, is that while DuPont licensed the right to stack most genes on top of Roundup Ready, it's now stacking a gene that falls outside its license. That's a stacking right "they never paid for," says Grant. "That's the gist of this thing."
Monsanto's dispute with DuPont surfaced in litigation in May 2009, when Monsanto sued DuPont for patent infringement in federal court in St. Louis, near Monsanto's 504-acre corporate campus in Creve Coeur, Mo. (pronounced as if France had never existed). DuPont promptly counterclaimed, alleging antitrust violations. Shortly thereafter, the Justice Department commenced an antitrust inquiry of Monsanto focusing on allegations that closely parallel DuPont's, and soon several state attorneys general, led by Iowa's Tom Miller, were sniffing around too.
The St. Louis litigation is proceeding slowly, though, and Monsanto has won two preliminary victories there that bode ill for DuPont. If there is any merit to DuPont's claims that Monsanto is imminently, irrevocably, and illegally extending its monopoly in this crucial industry, only swift intervention by the Justice Department is likely to make any difference. That department's antitrust division is now led by assistant attorney general Christine Varney, who has criticized her "overly cautious" Bush-era predecessors for insufficient vigilance in policing dominant firms. On top of that, she and her boss, Attorney General Eric Holder Jr., have vowed to make antitrust enforcement in the agribusiness sector a priority of their tenure. In this context, it seems all but foreordained that another shoe will drop.
Though Hugh Grant leads today's Monsanto, it is Robert Fraley, 57, who personifies it. Now chief technology officer, Fraley joined Monsanto almost 30 years ago when it was still an industrial chemical company that made plastics like safety glass and sun tints embedded in auto windshields. Fraley was hired to strengthen the company's nascent ag-biotech unit and pioneer an industry that didn't yet exist. He succeeded.
At an interview, Fraley peers levelly into the eyes of a reporter as if bracing for an unpleasant but accustomed task. He is tall, trim, fit, and poker-faced, and his head is shaved and smooth. His utilitarian office is adorned mainly with hunting and fishing trophies. Neither his temperament nor his appearance fits the stereotype of a Ph.D. microbiologist, but that's what he is. At a ceremony at the White House in 1998, President Bill Clinton presented him and three other Monsanto scientists with the National Medal for Technology. (Other recipients: Apple (APPL) CEO Steve Jobs, Polaroid (PRDCQ) founder Edwin Land, and Internet pioneer Vint Cerf.)
Fraley grew up on a farm near Hoopestown, Ill., about 100 miles south of Chicago, where his dad grew soybeans and corn. As a kid, he remembers doing a chore for the local farmers. "You'd walk up and down all the rows of soybean fields," Fraley recalls, "and cut the weeds out by hand. You'd do that from morning till night. That was a tough job."
Because of Monsanto -- and to no small extent Fraley -- American farm boys don't need to do that anymore. In 1974, Monsanto introduced the herbicide Roundup, which was made up mainly of the salt known as glyphosate. Glyphosate has very low toxicity for animals, but for a short duration -- before rainwater dissolves it into harmless residues -- it will kill almost any green plant onto which it is sprayed.
Though a farmer had to be careful about exactly when he applied Roundup to his fields -- lest he kill his crop along with the weeds -- the product achieved cheaper, safer, and more effective weed suppression than anything before it. Roundup revolutionized farming and became Monsanto's most successful product. (In 1981, Monsanto's Leicester, England, outpost hired a young Glaswegian with a graduate degree in rural science, Hugh Grant, to show Scottish farmers how to apply Roundup to fields of malting barley, used to make Scotch whisky.)
Seeking to build on the Roundup juggernaut, Monsanto's nascent biotech unit -- which hired Fraley out of the University of Illinois in 1981 -- decided to look for a gene that could make crops immune to its own blockbuster product. If one could be found and introduced successfully into crop DNA, it would make the farmer's weed-killing task even simpler, spurring still greater sales of Roundup.
Looking in Monsanto's waste pools, its scientists found a bacterium that thrived in Roundup-laden sludge, and then isolated the gene that enabled it to do so. The team then had to introduce that bacterial gene into embryonic soybean cells. Fraley was part of the Monsanto team that pioneered a way to accomplish the feat. They used a parasitic microorganism known as Agrobacterium tumefaciens, whose natural life cycle involves injecting some of its own DNA into the nucleus of a host plant's cells. Though the bacterium in nature injects harmful genes into cells, Monsanto plucked those out beforehand and replaced them with the Roundup-resistance genes they sought to introduce. In a petri dish they then let the Agrobacterium do its thing with embryonic soybean cells. The embryonic cells were then allowed to divide naturally until they developed into a seedling. The seedling grew into a mature plant, and when it bore its own seeds, they incorporated the "transgenic" DNA -- the gene transplanted from the waste-pool bacterium into the soybean genome. Monsanto had created Roundup Ready, and saw that it was good.
"I can still remember those little green blobs of soybean tissue culture cells that became eventually Roundup Ready soybeans," Fraley says. "That was the first glimmer that this would be possible."
After more than a decade of additional research and development, Monsanto introduced its first two commercial transgenic crop traits in 1996: Roundup Ready for soybeans and Bollgard cotton, which provided resistance to three loathed insect pests.
Farmer acceptance was astonishingly rapid. Last year 91% of the soybeans planted in the U.S. were genetically modified, and at least 92% of them contained Monsanto's Roundup Ready trait. Similarly, about 88% of the country's cotton is now genetically modified, and 95% of those plants contain at least one patented Monsanto trait.
Yes, genetically engineered seed costs more -- easily 30% to 80% more than conventional seed -- but not nearly what farmers judge they can earn in increased yield or save in reduced pesticide, herbicide, tilling, and airplane-spraying costs.
"Farmers pencil out every purchase decision," stresses Fraley. "My dad would spend more time trying to figure out which seed he was going to grow than anything else. The fact that farmers have adopted this technology so rapidly and so extensively is due to one simple thing: It really works."
A moment later, though, Fraley adds a second factor. "What facilitated that broad adoption," he continues, "was the company's absolutely pivotal decision to broadly license" its traits rather than, for example, making them available only through the company's own proprietary seed companies. In other words, Monsanto followed the business model of Microsoft, which licensed Windows to all comers, rather than, say, the closed model of the early Apple, which kept its operating system to itself. "Somewhere between 200 and 300 seed companies have access to our Roundup Ready or to our bug-control traits," says Fraley, including major competitors like Pioneer, Syngenta, and Dow. "It's benefited the industry," he says. "Then I hear these shrill voices saying, "This technology is so broadly adopted that can't be good.' But to me it's a sign of the benefits it's provided."
Nevertheless, having had the foresight to license your product broadly is no defense to an antitrust suit, as Microsoft itself found out in the late 1990s. Similarly, having come by a monopoly honestly (and ingeniously and daringly) is no defense to a Sherman Act charge, in which the government alleges that a company is artificially trying to maintain or extend a monopoly.
While Monsanto officials still see their company as the scrappy underdog it was a decade ago, the rest of the world seems to see it as an arrogant, big-footed bully.
Today, Monsanto's Roundup Ready is the only glyphosate-tolerance trait commercially available in a world where glyphosate is the overwhelmingly prevalent herbicide. (Monsanto's patent on Roundup weed killer expired in 2000. Monsanto's gross profits from Roundup Ready and other traits eclipsed those from Roundup in 2003.)
Farmers and seed companies, having paid dearly for Roundup Ready's benefits throughout its patent life, are now eager to begin enjoying their half of the patent bargain -- the point when Monsanto's legal monopoly expires and Roundup Ready enters the public domain.
In 2008, however, some farmers and seed companies started to wonder whether they would ever see such a day. That year, according to DuPont, Monsanto told some independent seed companies that when their patent licenses on Roundup Ready 1 soybeans ran out in 2012 -- two years before the patent expired -- Monsanto would refuse to renew them. Instead, seed companies that wanted to have any glyphosate-tolerant soybeans in stock would have to switch to Monsanto's patented successor product, Roundup Ready 2 Yield.
Roundup Ready 2 -- whose first commercial release is taking place this spring -- consists of the very same gene that was used to make Roundup Ready 1, though this time Monsanto introduced it into a different spot in the soybean genome and also tweaked its "genetic cassette" -- that is, the material surrounding a gene that can turbocharge its expression. Monsanto claims that the changes will pack a 7% boost in yield. DuPont sharply disputes that claim, however, alleging in papers submitted to the Justice Department that Roundup Ready 2 "offers little, if any, documented additional value to customers." In an interview with Fortune, Bill Niebur, DuPont's vice president for crop genetics research, denigrates "RR2" as "a new acronym" posing as a new invention.
If independent seed companies couldn't keep breeding Roundup Ready 1 past 2012, then they weren't going to have any generic Roundup Ready 1 available to sell farmers for the spring planting in 2015, after the patent expired. In addition, under the terms of Monsanto's standard seed licenses, farmers are supposed to destroy any seed left over after planting, and they are barred from saving seed from harvests. So how exactly would farmers obtain generic Roundup Ready 1?
Last, who was going to take over the burdensome task of keeping current all the foreign import licenses on Roundup Ready 1 once Monsanto no longer had an incentive to do so, and when no other company had access to Monsanto's vast informational archives about its product?
Monsanto insists that all those fears stemmed from misinformation or misunderstandings. To begin with, it flatly denies it ever tried to force seed companies to switch from Roundup Ready 1. And to allay any lingering concerns, in December 2009, Monsanto publicly pledged that all existing Roundup Ready 1 licenses would automatically be extended through the end of the product's patent life in 2014, so any seed company that wanted to continue breeding Roundup Ready 1 into its seed varieties could do so. Similarly, the company declared, no farmer would be barred from saving Roundup Ready 1 seed in the last year of the patent term.
Finally, Monsanto promised to keep up all Roundup Ready 1 international regulatory licenses through 2017 -- three years after the patent expires. After that, it has urged that the industry come together and figure out a mechanism for maintaining those licenses long term.
The takeaway, CEO Grant says, is simple: "By spring of 2015 there will be free Roundup Ready 1, or the grower will have a choice of writing a check for Roundup Ready 2. I'm hoping they'll write a check for Roundup Ready 2, because it yields more."
What's more, Monsanto protests, there's no need for any Hatch-Waxman-type law in the ag-biotech world. That's because the moment the patent expires on a transgenic seed trait, the generic seeds already exist! As of 2015, says Monsanto in-house attorney Scott Partridge, "you can take a handful of soybeans with Roundup Ready 1 in them, take that and a bucket of dirt, and you're now manufacturing Roundup Ready seeds."
Well, true. But today's seed market is highly sophisticated, and very likely most farmers aren't going to start saving their own seed for replanting even if Monsanto lets them do so. That's because the yield of any crop is determined not just by the transgenic traits artificially introduced into it but also by the quality of its "germplasm" -- all the other genes that make up a given seed variety's genome. Germplasm developers are continually improving each seed variety through conventional breeding techniques -- that is, crossing high-performing plants with one another. By doing so, they can improve most corn and soybean variety yields by about 1% to 1.5% a year. If a farmer saves seed from his own harvest, he'll miss out on that annual boost.
It is essential, therefore, that germplasm developers continue to actively breed Roundup Ready 1 in preparation for generic release. And DuPont insists that, notwithstanding public reassurances, Monsanto is still pressuring and incentivizing germplasm developers into dropping their Roundup Ready 1 seed lines, either through restrictive provisions in patent licenses or through alluring rebates and discounts. In January, DuPont subpoenaed 26 seed companies seeking information pertinent to this question. Monsanto has moved to quash the subpoenas, calling them premature. (Seed firms contacted by Fortune declined to comment.)
While the future of Roundup Ready 1 as a generic product is what farmers worry about most, it is not, in the end, what is driving DuPont. DuPont is not, after all, in the business of selling generics. Rather, DuPont's key gripe is that Monsanto is allegedly using provisions in its Roundup Ready 1 patent licenses to hamper DuPont's development of its own alternative platform product -- one that, it claims, would offer farmers and consumers something far superior to either Roundup Ready 1 or 2.
Since at least 2006, DuPont's Pioneer seed subsidiary has been developing and testing a product it calls GAT, an acronym for a trait that achieves glyphosate tolerance through a different mode of action from that employed by Roundup Ready. Its innovative approach results in both better yields and better protection under most circumstances, DuPont claims.
To add to GAT's appeal, DuPont hard-wired it together with a second DuPont gene that confers tolerance to a different class of herbicides -- herbicides that, yes, DuPont happens to sell. This extra feature could become quite important to farmers now that glyphosate-resistant weeds have begun to spring up in some fields.
To test its GAT product, Pioneer bred it into its best germplasms, explains John Soper, Pioneer's senior research director for soybeans. Those already happened to include the Roundup Ready 1 trait, which Pioneer has been licensed to breed into its seeds since 1992. Pioneer's original plan was to eventually breed out the Roundup Ready and sell its best seed varieties with just GAT in them, unburdened by any Monsanto patent-license obligations. But in the course of testing GAT, Soper recounts, "we started to see some interesting things."
The good news was that when GAT and Roundup Ready 1 were stacked, they had a synergistic impact on yield, boosting it 6% over what the same germplasm would yield with just Roundup Ready 1.
The bad news was this: "While in most environments GAT alone worked just fine," recounts Soper, "there were a few environments where we ended up with severe stress on the plants, where it didn't protect to the degree we really wanted the customers to have." In a January 2009 video presentation often cited by Monsanto, Pioneer acknowledged that GAT, if used alone, posed an "unacceptable risk" to farmers.
So DuPont/Pioneer decided to make its new platform product the combination of GAT and Roundup Ready 1. Pioneer foresaw no problem with this course correction, it says, since it had long been licensed to use Roundup Ready 1 and to stack any other traits on top of it.
Monsanto didn't see it that way. As Monsanto read its patent license with Pioneer, Pioneer could stack any trait on top of Roundup Ready except one: another gene designed to confer glyphosate tolerance. Since GAT was exactly that, DuPont couldn't stack it on Roundup Ready unless it agreed to pay still more licensing fees on top of those it was already paying. Monsanto claims that DuPont had failed to license this particular form of stacking earlier because no one had anticipated any reason to stack a glyphosate-tolerance gene on top of another glyphosate-tolerance gene until DuPont discovered the "unacceptable risks" of relying on GAT alone.
In May 2009, Monsanto sued Pioneer to stop it from developing the stacked product and warned independent seed developers not to breed it into their seed varieties. Monsanto alleged that the stack infringed on Monsanto's patents because it made use of Roundup Ready 1 in a way that was not authorized under Pioneer's existing patent licenses.
In January, U.S. District Judge E. Richard Webber sided with Monsanto on that issue, finding that Pioneer was, indeed, breaching its patent license. While DuPont/Pioneer takes sharp exception to that ruling and is seeking reconsideration of it, in the meantime the decision tees up very cleanly an important and novel antitrust question.
"Under the antitrust laws," asks DuPont/Pioneer counsel Boies rhetorically, "what conceivable justification is there for Monsanto to say, "You can't combine my monopoly product with your product in order to give consumers a better overall product'"
During a 90-minute interview, Boies alludes 11 times to the Microsoft litigation. "The thing that's remarkable to me," he observes, "is how directly the court of appeals decision in the Microsoft case bites on exactly what Monsanto's doing and the arguments Monsanto's making. They're saying, "This is our intellectual property, and we can do what we want with it.'"
But isn't Monsanto correct? Its patent extends till 2014; DuPont is using the patented product; a judge says DuPont isn't licensed to use it that way; case over. Right?
"This is, again, exactly what Microsoft said," protests Boies. Microsoft argued that its intellectual property rights -- copyrights, in that instance -- in the Windows operating system gave it power to dictate what computer makers could and couldn't do with it, including the right to forbid them from pre-installing the Netscape browser in preference to Microsoft's own Internet Explorer browser. But, Boies continues, the June 2001 ruling of the federal appeals court in Washington, D.C., rejected Microsoft's argument, finding that it bordered on the "frivolous."
"The company claims an absolute and unfettered right to use its intellectual property as it wishes," that court wrote. "That is no more correct than the proposition that use of one's personal property, such as a baseball bat, cannot give rise to tort liability."
Monsanto counsel Partridge replies that the Microsoft ruling does not apply here because it involved copyrights rather than patents, and the U.S. Supreme Court reiterated as recently as 2004 that antitrust laws impose no duty on patent holders to deal with their rivals.
"After patent expiration in 2014," Partridge notes, "DuPont will have the unfettered right to stack GAT and Roundup Ready." Even before then, he adds, Monsanto "has offered to allow DuPont to stack GAT with Roundup Ready at a reasonable royalty -- like any other licensee."
"Yes," says Boies mockingly, "they said they'd license it for some astronomical price. But you can't do indirectly what you can't do directly. The court in the Microsoft case said, 'No, you can't prohibit people from stacking onto a monopoly product. When your patent monopolizes an area of trade like corn seed or soybean seed, or like personal computing, then you've got special restrictions. You can use a patent to exploit the potential of your product. But you can't use it to exclude development of new competition.'"
Rarely before has the conflict between the patent laws and the antitrust laws been so squarely joined, and never, certainly, with so much at stake.