Early in the 20th century, Mr Durant founded General Motors Corp, one of a revolutionary group of auto companies that in some ways were the dotcoms of their era. Investment capital poured in and GM stock soared more than 5,500 per cent from 1914 to 1920.

But when the overcrowded automobile industry failed to deliver on inflated expectations in the early 1920s, auto stocks plunged. GM lost two-thirds of its stock value in six months. A panicked Mr Durant borrowed money and feverishly bought the shares in a futile attempt to prop them up.

GM, of course, eventually recovered and soared again - too late for its founder, who lost his entire fortune and wound up running a bowling alley.

Many other once-great auto companies, such as Packard, Studebaker and Hudson, fared less well than GM, disappearing one by one in the ensuing decades.

And therein lies a cautionary tale for investors who have sat in shock as Cisco Systems Inc stock has lost more than 70 per cent of its value and Yahoo! Inc shares have fallen close to 90 per cent. The current technology boom and bust, it turns out, is nothing new. It is simply the latest in a long history of investment bubbles that have plagued shareholders since investing began.

From the Dutch tulip-bulb craze of the 17th century, to the locomotive revolution of the 19th, to the rise, fall and resurrection of personal-computer stocks and biotechnology stocks in the 1980s and 1990s, investors have fallen madly in love with - and then madly out of love with - the hot technology of the moment.

These past debacles offer lessons for investors trying to figure out what to do after the Nasdaq Composite Index's 57 per cent fall from its peak last March 10 of 5048.62. Bubble watchers say the worst may not be over. Just as the stocks tend to overshoot on the way up, bubble stocks tend to overshoot on the way down as well.

Through Thursday, the Nasdaq index has lost 3.5 per cent this week alone, and is below where it ended the year in 1998. History does suggest that, eventually, many of the pieces of a burst stock bubble get patched back together. Some of the stocks will soar again. But the recovery typically takes longer than investors hope - usually years, not months. Many of the stocks caught up in the bubble fail to survive, and of those that do survive, all but the most robust wind up behaving less exuberantly than they did before.

Technology companies "are going to take a good whack, a good hard landing, and you can't expect that they are just going to shrug it off and go up to new highs and forget it," warns George Roche, chairman of Baltimore mutual-fund group T Rowe Price. "You are going to require a period to repair it."

Some optimists hope the recovery this time may come faster than it has after past bubbles. They point to the intervention of the Federal Reserve to stimulate the economy, and to today's pace of technological change, information flow and inventory adjustment, which is far faster than anything known in the past.But even the most bullish observers aren't predicting an immediate snapback.

In fact, a ghoulish fascination is developing among professional investors over the parallels between the current tech wreck and past stock disasters.

Many of the pros have been sending one another charts overlaying an index of Internet stocks or the Nasdaq Composite against charts showing the biotechnology bubble that collapsed in 1992, the Dow Jones Industrial Average around 1929, or the 1990 collapse of Japan's Nikkei 225 Average.

The basic truth market historians draw from past bubbles is that it almost always takes longer to repair the damage than people expect. After the economic crash of 1837, for example, it took a handful of railroad stocks a solid decade to regain precrash levels - and most didn't. GM's recovery took six years, about the same period it took most personal-computer and biotech stocks to rebound in recent times. The Japanese stock market topped out at the end of 1989, and remains 67% below its peak. Most recently, investors point to the 1992 biotechnology collapse. Biotech stocks, as measured by the American Stock Exchange Biotechnology Index, didn't stop falling until March 1995. They didn't regain old highs in a lasting way until December 1998.

There is also the fate of the electric utilities in the 1920s. Bearers of a breathtaking new technology that transformed the world just as the Internet is doing today, it was 1964 before they returned to their 1929 peak, and they never regained the vigor of their youth.

Radio Corp of America was another high flier of the 1920s. America's fascination with the radio propelled RCA's shares to a high of $500 just before the '29 crash - an increase of 2,000 per cent over six years. The crash wiped out all those gains, and it took more than three decades for the stock to revisit its highs. Even then, RCA, which had become a television company, ultimately was undone by competition, and wound up being split up and sold.

Today's technology pessimists, such as Jeremy Grantham, co-founder and chief strategist at Boston money-management group Grantham, Mayo, Van Otterloo & Co, argue it could be a decade before technology stocks turn convincingly upward. Mr Grantham has tracked more than a dozen past bubbles involving gold, oil, nickel, the dollar, the British pound, Japanese stocks in the 1980s and US stocks in the 1960s.

In every case, he says, the stock, commodity or currency gave back all the extra, inflation-adjusted gains beyond what it would have attained if it had simply maintained its slower, prebubble performance. He believes tech stocks have still farther to fall, and that they will pull the broader market down into a lasting slump as well.

The Wall Street Journal