(FORTUNE Magazine) – What killed the cat makes us rich. To be more specific: If it weren't for the intellectual curiosity of engineers and scientists thinking things through, trying out ideas, and taking wild guesses in R&D labs across the country, we'd have half the economy we have today. In a recent report on R&D, the American Association for the Advancement of Science estimates that as much as 50% of U.S. economic growth during the half-century since the FORTUNE 500 came into existence has been due to advances in technology.

For corporations, R&D has been a fabulous investment. The Industrial Research Institute in Arlington, Va., an association that brings together research managers to share best practices, calculates that each dollar spent on R&D brings an average return of about three dollars. Of course, the focus of R&D in America has changed a lot over the years, as the defense-driven Cold War era gave way to the electronics and communications explosion of the Internet age, and as biotech has taken off. Lately, homeland security spending has unleashed another wave of investment.

As the focus of the research has changed, so has the style of the organizations doing the work. Early giants of invention such as Bell Labs and the research wings of companies like DuPont and GE found themselves in a world where technical knowledge was becoming widely dispersed. At the same time, growing global competition meant that R&D labs had to turn from pursuits that might possibly win Nobel Prizes to ones that more directly supported their parent companies' need to get better products to market quicker. No longer could any organization hope to employ all the smart mystery solvers in a given field.

Today some of America's greatest companies view R&D less as a strictly in-house process than as a kind of international bazaar of ideas. Intel, for example, has an "acquire and develop" strategy; its Intel Capital unit invests in numerous startups working on potentially hot technologies in order to have access to those that succeed. Eli Lilly established and spun out a venture called InnoCentive, which posts on the Internet technical challenges it wants solved and the money it is willing to pay. Nobel laureate physicist Arno Penzias, a former chief scientist at Bell Labs who is now a venture capitalist, sums up the trend in R&D management as being governed by the same "make or buy" template that manufacturers use to decide where to get ingredients and components for products. "Make or buy is what it's about now in R&D," he says. "You've got to decide what to acquire vs. what to grow in your own greenhouse."

On the following pages, we give a big tip of the hat to the crown jewels of the FORTUNE 500, R&D labs and the marvels they've spawned.

Bell Labs, Laser

When scientists Arthur Schawlow and Charles Townes at Bell Labs published a technical paper in 1958 describing the principles of what would come to be known as the laser, they couldn't have imagined how widespread their invention would become. Telecommunications, surgery, welding, computer printers, bar-code scanners, and CD players are some of the uses for the beams of coherent light zillions of lasers produce. In the 1964 photo at right, researcher George Douglas adjusts a prism at the end of a 33-foot laser.

Lockheed, SR-71

Blackbird America's deadly cold war rivalry with the Soviet Union unleashed a torrent of funding for developing military technologies, many of them airborne. No machine better exemplifies the ingenuity engineers brought to bear than the SR-71 Blackbird, a superfast spy plane born at Lockheed's Skunk Works lab under the leadership of aeronautical genius Clarence "Kelly" Johnson. He and his engineers conceived the Blackbird as a faster, higher-flying successor to Lockheed's 475-mph U-2, one of which was shot down over the Soviet Union in 1960. With a cruising speed of 2,100 mph, the Blackbird stymied Russian surface-to-air missile designers when it began flying in 1962.

Nothing about building the Blackbird was easy. Pratt & Whitney had to develop exotic turbo-ramjet engines to provide propulsion. The aircraft's skin and most of its internal structure had to be fabricated from hard-to-work titanium to withstand aluminum-melting temperatures of 800° F created by air friction at ultrafast cruising speeds. The landing gear needed special nonmelting tires.

The most alarming thing about the Blackbird was the way it wept fuel when sitting on the ground--no temperature-resistant material existed to seal the gaps between its skin panels. To reduce the fire hazard, a special high-flashpoint fuel called JP-7 was developed. As the SR-71 climbed and accelerated after takeoff, its skin warmed and expanded, causing the panel joints to close. Then the pilot would rendezvous with an aerial refueling plane, top off the tanks, and rocket to 80,000 feet and beyond, where Blackbird plied its trade.

Above, Air Force pilots board an SR-71 in 1969.

Corning, Fiber Optics

Artisans have long known that molten glass can be drawn into an exquisitely thin thread. But it was not until 1970 that Corning scientists Donald Keck (far right) and Robert Maurer and colleagues learned to make optical fibers that can transmit information, in the form of pulses of light, over long distances.

DuPont, Kevlar

The eventual uses of an invention can be impossible to anticipate. When scientists at DuPont's Experimental Station struggled in the early 1960s to come up with a fiber similar to nylon but tougher and stiffer, the hoped-for market was belts for automobile tires. In 1965 chemist Stephanie Kwolek (above left, wearing heat-resistant gloves) devised a process for cold-spinning the aramid fiber that came to be known as Kevlar. But tiremakers proved slow in adapting their manufacturing processes to work with the stuff; instead, bulletproof vests became the first major application for Kevlar, which is five times stronger than the same weight of steel. In subsequent years it found uses in aircraft, military armor, helmets, boats, sporting goods, ropes and cables--and, yes, tires.

Procter & Gamble, Disposable Diapers

Every patent is accompanied by a detailed technical drawing, and the one above shows the design for the disposable diaper, which P&G brought to market in 1961 under the Pampers brand. Pampers got their start in 1956 when P&G researcher Vic Mills, who'd been spending evenings and weekends babysitting a newborn grandson, assigned a group to study the possibility of making a leakproof diaper that kept babies dry and could be thrown away. Washing diapers wasn't his thing. The team came up with a design that used absorbent paper to soak up baby effluvia and a plastic backing to contain it. In 1971 adhesive tape was added. Pampers Feel 'n Learn Advanced Trainers with Spider-man and Hulk designs? Those came later.

GM and Engelhard Corp., Catalytic Converter

Engelhard Corp. figured out the basic science--using metal "catalysts" to transform pollutants into water vapor and CO2. GM figured out how to integrate it into a car and mass-produce it. And because Washington mandated that starting in 1975 all cars use catalytic converters, the skies have been spared hundreds of billions of tons of pollutants.

Identix, Automated Facial Recognition

Homeland security is a goldmine, especially for startups hatching R&D that big companies may want to acquire. The U.S. wants visitors from 27 nations to soon have scannable biometric information in their passports. Identix, near Minneapolis, uses facial-recognition software to capture the geometric features of a person's face, which it stores as a spiderweb-like template. These templates can be matched against others in a master database or watch list.

Eli Lilly, Bioengineered Insulin

The microscopic shapes shown floating above are crystals of the first batch of human insulin produced by drugmaker Eli Lilly using recombinant DNA. Introduced in 1982 under the name Humulin, Lilly's diabetes treatment was the first medicine to be produced by genetically modified bacteria. Scientists inserted the human gene that directs insulin production into the DNA of humble E. coli, inducing the microscopic critters to churn out the protein, which diabetics can't produce enough of in their own bodies.

Biotech protein medicines like Humulin are produced in a high-tech version of beer-brewing. It's a fermentation process wherein the special bacteria are nurtured in stainless-steel "bioreactors" filled with a nutrient broth. Before Lilly's breakthrough, diabetics had to rely on insulin harvested from the pancreases of pigs and cows.

SUBSCRIBERS ONLY Get the full FORTUNE 1,000, including company profiles, industry data, up-to-date stock information, and FORTUNE stories.