"We discovered something important today", John Bardeen mumbled to his wife on the evening od Dec 16, 1947, as she peeled carrots over the kitchen sink.
"That's great", she remarked as he strolled into the living room.
Such was the start of a scientific revolution. A half-century ago, the transistor era dawned, spawning Silicon Valley, the space age, high-tech weaponry, personal computers, an pocket-size radios that blared rock music to boogeying beachgoers.
"The transistor is what made Silicon Valley into Silicon Valley", says science writer and transistor historian Michael Riordan.
"The transistor became the nerve cell of the information age. It's the gizmo that makes the personal computer possible and the Internet possible, and allows millions of computers to talk to each other around the globe."
Gazing proudly at the first transistor - a thumb-size mélange of a gnarled paper clip, gold foil, a coiled wire, and a translucent slice of polystyrene - inventors John Bardeen and Walter Brattain knew it was important.
Yet they and their associate, William Schockley, failed to fully grasp what they had wrought at AT&T's Bell Laboratories in Murray Hill, N.J.
The transistor allowed them to control and amplify electrical signals inside materials called "semiconductors".
The achievement was to the 20th century what the steam engine was to the 18th century and the telegraph to the 19th century. The steam engine inaugurated the age of mass production and rapid train travel; the telegraph, the age of instant connunication. And the transistor brought about the information age.
From stock quotes to hit songs, from Mars-bound rockets to Hollywood blockbusters encoded on laser discs, from supercomputer simulations of exploding stars to cellular phone chats with TV psychics, oceans of data are stored and distributed, thanks to the transistor.
It also accounds for much of the world economy. Today, the United States economy dominates the planet, to a considerable extent because of its excellence in microelectronics.
"The U.S. chip industry is one of the biggest industries in the world today. It's an economic Goliath, globally as well as in the United States", said Jeff Weir of the Semiconductor Industry Association. "In 1990, the global chip industry was a $50-billion industry. Today, it's a $140-billion industry."
In 1965, Silicon Valley guru and Intel co-founder Gordon Moore showed that the number of transistors that could be jammed on a microchip - a fingernail-size silicon chip with thousands or millions of transistors - was doubling every year. The growth rate was so steady that people named it "Moore's Law".
The doubling rate has slowed slightly, to once every 18 months. Yet, the microelectronics boom continues, as engineers squeez more and more transistors onto tinier and tinier chips.
"It's really an astonishing rate of progress over a long period of time. To keep on that rate, we have to move faster and faster", said Mark Pinto, chief technical officer of Lucent Technologies Microelectronic Group in Berkeley Heights, N.J.
According to present projections, Pinto says, around 2012 experts will be able to shrink a quarter of a trillion transistors onto a single "memory chip", which stores data that can be written over.
The microchip is an outgrowth of technologies now taken for granted - telephones and radar - Michael Riordan and Lillian Hoddeson say in their book, Crystal Fire, a history of the transistor. Riordan is a physicist and science writer at the Stanford Linear Accelerator Centre; Hoddeson is a historian of physics at the University of Illinois.
In the 1910s, they say, AT&T wanted to dominate the long-distance telephone business in the United States. But calls over distances more than 1,000 miles were difficult: Voices tended to disappear in a sea of electrical noise.
So the company used high-power vacuum tubes to amplify the voices. The first cross-continent telephone connection was officially opened between San Francisco and Washington and New York City in 1915.
Vacuum tubes, however, consumed a lot energy, produced a lot of heat, tended to be slow, and tended to break down. AT&T sought a way to amplify signals without tubes. It also hoped to find a way to automatically transfer calls without using human operators or trouble-prone electro-mechanical switches.
The answer arrived in the form of a substance that had attracted attention during the Second World War - silicon.
Silicon, one of the universe's 100-plus elements, combines with other elements to form the silicates in the Earth's crust - for example, sand. It is a semiconductor, meaning it can conduct only a trickle of electricity. That trickle can be manipulated to amplify or suppress the electrical flow. Germanium is also a semiconductor.
During the war, Riordan and Hoddeson say, scientists at AT&T Bell Labs and in England discovered that highly purified silicon could be used in crystal detectors to detect high-frequency radar signals.
By a mixing of theoretical insights and messy hands-on laboratory tinkering with germanium and other materials, the Bell Labs scientists Bardeen and Brattain invented the transistor. Schockley later helped develop a more modern version of the device.
But Schockley's aggressive personality repelled associates, including Bardeen and Brattain. He might have become a rich man if his paranoia hadn't driven away eight business partners - "the transistor eight", he reportedly called them - who fled to found the fabulously profitable company Fairchild Semiconductor. From their success sprouted much of Silicon Valley.
Another breakthrough came in the late 1950s: Robert Noyce of Fairchild and Jack Kirby at Texas Instruments independently discovered how to form numerous transistors and other electrical components on a slice of silicon. thus was born the microchip, the workhorse of modern microelectronics, including personal computers.
Schockley has been unfairly denied credit as founder of Silicon Valley because of his controversial claim that blacks tended to be less intelligent than whites, which he espoused later in life, says Schockley's biographer, the noted science writer Joel Shurkin.
The microelectronics revolution has come far since Schockley's heyday/ How much farther can it go? Experts warn that as chips shrink, they require much costlier manufacturing methods.
Others think silicon has plenty of untapped computing power.
"When I joined the company over 20 years ago, the field was awash with warnings that `silicon is running out of gas' and `we've got to go with other technologies'", says Randy Isaac, a vice-president at IBM's Thomas J. Watson Research Centre in Yorktown Heights, N.Y.
"In fact", Isaac says, "what's happened over the last 20 years is we've had a significant reduction of talk about silicon `running out of gas'.