(FORTUNE Magazine) – In 1955, the year FORTUNE published its first list of the 500 largest industrial companies in America, three brothers moved from the Florida Panhandle to Columbus, Ga., to start an insurance company. Their names were Bill, John, and Paul Amos. They knew next to nothing about insurance and had no ties to their new hometown, which they picked because it was the largest city in the Southeast without its own insurance company. Paul Amos, who at 79 is the only one of the three still alive, says they always knew their new venture was "a big gamble." But in truth, the brothers started their new life in Columbus--and their new business, which they called American Family Life Assurance Co.--not because they were gamblers but because they were dreamers.

Today American Family Life Assurance Co. is known as AFLAC--yes, the one with the duck commercials. It has a market cap in excess of $20 billion and assets of well over $50 billion. Though its headquarters remain in Columbus, its largest market, by far, is Japan. In the past decade alone its stock has returned over 23% annually. And then there's this: AFLAC is one of only three companies founded in that year of the first FORTUNE 500 to actually make it onto the FORTUNE 500, where it comes in this year at No. 172.

That fact, surely, is worth dwelling on for a moment. Out of what must have been thousands of startups that began in 1955, only three made it to the pinnacle. (The others are American Financial Group and Dover Corp., an industrial conglomerate.) What combination of grit and talent, foresight and dumb luck, I wondered, caused those three dreamers from Florida to succeed so emphatically when so many others either failed or came up short? And it also made me wonder something else: Fifty years after the inauguration of the FORTUNE 500, what will it take for today's startups to make their way to that same wonderful place?

Dan Amos is sitting in his office talking about his father, Paul, and his two uncles. "My dad was the salesman," he says with an easy smile. "Uncle Bill was the accountant--he was tight, like me. And Uncle John was the visionary. He never liked operations. He liked thinking up what the company was going to do next."

Amos's office is in the executive suite of the AFLAC building--the tallest building in Columbus. He's been CEO since 1990, when he was 39, and his uncle John, who had been the company's driving force, died of lung cancer. Dan Amos wants you to know the job wasn't just handed to him; he'd been a highly successful salesman and manager for the company before becoming president in 1982. At AFLAC, salesmen have always been paid on commission, and Dan Amos had to take a $100,000 pay cut when he joined his uncle in the executive suite. (His son Paul, who's 28, is the state sales coordinator for north Georgia.)

Would John Amos have been surprised to find AFLAC on the FORTUNE 500? Unlikely. "He liked to make predictions about how big we would be," says Dan Amos--though in the early days such predictions did not seem very credible. Like most startups--then, now, and forever--the company struggled. There was no venture capital industry in 1955; the brothers borrowed around $50,000 to meet Georgia's capital requirements, and raised additional capital by knocking on doors and selling shares of stock. That was also how they tried to sell insurance--door to door--but without much success. "We were broke many times," says Paul Amos. "We just never declared bankruptcy."

"Mr. Paul," as he's known at AFLAC, believes the company made it because the brothers weren't "steeped in insurance tradition." They were free to think about the business in new and different ways. So, just as Ray Kroc discovered a new approach to selling burgers and Sam Walton found a new approach to the retail business, John Amos came up with a genuine insurance innovation.

Here's what happened: A few years after the founding of American Family Life, the Amos's father died of cancer. Realizing how much of the cost of his father's care was not covered by insurance, John Amos conceived the idea of supplemental cancer insurance. Customers would pay a small premium--at that time it was $24 a year for a family and $12 for an individual--and would receive a lump sum if they came down with cancer; the money could be used for everything from extra nursing care to hotel rooms for family members.

Then the next breakthrough. The main distribution mechanism for insurance back then was the door-to-door salesman. But building an insurance company one blessed customer at a time was agonizingly slow--far too slow for men as ambitious as the Amos brothers. There had to be another way.

American corporations were adding health and other benefits as part of the way they paid employees. What if they could offer supplemental cancer insurance too? American Family was too small to offer it to big companies, so it began selling the insurance to companies with 100 to 500 employees. The business took off.

By the 1970s, AFLAC was growing at a steady clip--steady enough to be listed on the New York Stock Exchange--but it was still primarily based in the South, and it was still a one-product company. It had built its headquarters by then, and John Amos had become a figure of considerable stature in Columbus, where he was known for the white three-piece suits he always wore. He and his brothers had done what they'd come to Columbus to do: They'd built a company and made themselves rich.

And it wasn't enough, at least not for John and Paul. (Bill retired in the late 1970s.) It's not that they were greedy; the Amoses were generous benefactors. It's that there was something inside them that pushed them to find new territory to conquer, to build a company that wasn't large just by Columbus's standards, or even Georgia's, but by America's.

To do that, AFLAC had to go all the way to Japan. John Amos had visited Osaka in 1970 to attend the World's Fair. All over the city he saw people wearing white surgical masks. When he asked why they did it, he was told that it was because the Japanese were trying to avoid catching colds. He immediately thought: Any population this health-conscious will jump at the chance to buy supplemental cancer insurance. And he was right.

Japan in those days didn't allow foreign insurance companies in the country. After a four-year struggle, AFLAC became the first post-war insurer granted a license by the Japanese government; Paul Amos believes it was because the company was so small, and few believed it would succeed.

Once it got the license, AFLAC had to set up a completely new operation, one with its own executive team and way of doing business. In Japan, AFLAC did almost no advertising because cancer was a subject people simply didn't talk about. It marketed to giant corporations instead of small businesses. It had a corporate culture the Columbus AFLAC employees would scarcely have recognized. "I didn't even realize that AFLAC had a U.S. headquarters," recalls Akitoshi Kan, a Japanese executive who began working there in 1980. "AFLAC was viewed in Japan almost like a Japanese company. You just didn't see Americans over there." Within seven years, AFLAC was generating as much revenue in Japan as in the U.S.; today Japan accounts for nearly 80% of revenues. As for Akitoshi Kan, he is the second Japan-based executive to move to Columbus and take a job in the executive suite of the AFLAC Tower, which he did in 2001. He is now the company's executive VP for operations.

You lose things when you grow to be a FORTUNE 500 company. You have to put in systems that eliminate errors--but can also diminish the human touch. You have to do things by the book, which means adding bureaucracy. It's hard, if not impossible, for the CEO to know everybody in the place. The company outgrows treasured traditions. Inevitably, all this has happened at AFLAC.

Old-timers still remember when John Amos's wife would bring her homemade ice cream to the office, and when John and Paul Amos would go deep-sea fishing and fry the fish for everybody. John Amos used to live in an amazing house he built on the top of the company parking lot--he always wanted to be close to the office. It's empty now. They remember how sad they all were when he died. Mostly, they remember when AFLAC was smaller and homier, and they miss it.

On the other hand, AFLAC now has a diversity day, and an employee appreciation week. The company started a day-care center in 1991. It may not give out cookies every Friday anymore, but every employee gets a bonus the week before Thanksgiving. Yes, it has systems and bureaucracy and the other trappings of modern corporate life, but most people at AFLAC understand why.

The business has modernized as well. AFLAC still offers supplemental cancer insurance, but it is only one of a wide array of offerings, and no longer the largest. A short-term disability product is now AFLAC's most successful in the U.S.

As for Dan Amos, he travels to Japan every few months. He talks to investors, analysts, reporters, and corporate customers. Unlike his uncle John, Dan Amos professes to "love operations." Mostly, though, he prefers to give his executives autonomy and focus on "strategic things." And he knows there's no guarantee that his son will succeed him as CEO. The modern corporation just doesn't work that way. Which is fine by him.

Even though AFLAC is now sleek and modern, Dan Amos hasn't stopped dreaming. He has ideas for new products and plans for growth. "I think we'll double in the next five years," he says bluntly.

If AFLAC represents what it once took to make the FORTUNE 500, what will it take for companies starting up today? A few weeks after my visit to Columbus, I flew to Seattle to meet the founders of NanoString. Only five months old, it doesn't even have an office; Amber Ratcliffe, the 28-year-old chief operating officer, mainly works out of her apartment. The other founders, Krassen Dimitrov, 36, and Dwayne Dunaway, 37, are disengaging from the Institute for Systems Biology, a privately funded operation run by a biotech pioneer named Lee Hood, where they've worked the past few years. NanoString will be the fourth company to be spun out of the institute.

I met the four NanoString principals--the fourth, Seattle Genetics chairman Perry Fell, has agreed to become CEO of the new company--in a conference room at the institute. The technology they hope to commercialize is difficult for a layman to understand; the phrase the founders kept using was "molecular bar code." As best I understand it, ever since the human genome was decoded, scientists have been relying on something called microarrays to analyze genes. But microarrays require large amounts of biological samples, take days to complete, and still offer, at best, only approximate results. By contrast, NanoString's technology will be able to simultaneously identify and digitally count millions of single gene transcripts, automate the processing, complete the task in 20 to 30 minutes, and yield results that are absolutely accurate. If the technology is as good as the NanoString folks believe, it will be a huge step forward.

What's easier to understand is the genesis of the company. Ratcliffe had worked with Dimitrov, both at the University of Washington and then at the institute. She left in early 2001 to get her MBA, but before she did, Dimitrov, who was running the microarray lab, came up with his "bar code" idea. As Dimitrov and Dunaway tested the "proof of principle" in the lab, Ratcliffe worked up a business plan based on Dimitrov's technology as part of her MBA. She then entered some national business plan competitions. At the Northwestern University contest, hers was named Best High-Growth Business Plan. In all, she won $46,000 in three contests. By then it was pretty clear that she, Dimitrov, and Dunaway were going to start a company. In addition to luring Fell as CEO, the NanoString founders have received a funding commitment from OVP Venture Partners, one of Seattle's better-known VCs. They'll soon get $8 million in venture money and be off and running.

For most of the interview the founders of NanoString could not have seemed more different from the Amos brothers. Whereas the brothers from Florida were strivers who could have started just about any kind of company, these young founders were the ultimate specialists: Biotech is what they know. Their new company, incubated in an entrepreneurial laboratory, envisioned in a prize-wining business plan, funded by bigtime venture capital, was the classic modern startup.

But then I asked what, in their wildest dreams, they thought NanoString could become. Without hesitation, Fell shouted, "Microsoft!" He explained: NanoString technology could be a kind of operating system for gene analysis. It wouldn't matter what field you were in; if you needed gene analysis, you would use the NanoString process. "Just think of all the applications," he said, reeling off a few: early cancer detection, bioterror prevention, agriculture. "You could use it to come up with a SARS detection kit," he said excitedly. "You could create a handheld device that will allow doctors to get a blood sample and determine almost immediately whether a patient is well or sick, and what he or she has."

"The sky's the limit," Ratcliffe chimed in. Dimitrov and Dunaway both smiled in agreement. And in this one crucial respect, they're just like the Amos brothers in 1955: They're dreamers too.

FEEDBACK jnocera@fortunemail.com