NEW YORK (FORTUNE) - It's been seven years since Henry Blodget became a household name. On Dec. 16, 1998, when he was a relatively unknown stock analyst at Oppenheimer & Co., he put a price target of $400 on Amazon.com.

We all know what happened next: Blodget became the embodiment of the Internet bubble -- his fortunes rising into the stratosphere for 15 months before crashing with all the stocks he recommended. The government accused him of fraud -- for touting stocks he privately trashed to colleagues. He paid a fine, and, without admitting or denying guilt, agreed to a lifetime ban from the securities industry.

It's all worth remembering for two reasons. First, that kind of mania has taken hold of another stock -- Google. Last week, Piper Jaffray analyst Safa Rashtchy, one of the best known experts on the search industry, put a $600 price target on the stock, only to be topped two days later by Mark Stahlman at Caris & Co., who said that over the longer term Google would likely hit $2000. The stock, which had been trading at about $425, instantly shot up to near $475. It closed Thursday at $463.

Second, Henry Blodget himself has become one of Google's most vocal bears. Indeed, on Tuesday, readers of his blog, InternetOutsider, found the following next to a picture of a grizzly: "No one else is writing this piece, so it will have to be me. I should say upfront that I'm not predicting that this will happen (yet), and I'm certainly not making a recommendation. I'm just laying out a scenario that could kneecap Google and take its stock back to, say, $100 a share."

It's a pretty simple, but intuitively powerful argument. It basically says that at Google's size -- $4 billion in annual net revenues and counting -- it gets increasingly hard to grow at rates that would justify paying 50 times earnings for the stock. Competition eventually drives up costs and drives down the prices one can charge. Hot new markets -- like online advertising -- inevitably hit bumps that make operating in them seem more challenging. And -- a point that Blodget doesn't make but should -- the bigger and more powerful a company like Google gets, the more it becomes everyone's bogeyman.

"Let's say click fraud continues to increase as a percent of total clicks (which seems perfectly plausible to me)," Blodget writes. "Eventually, all else being equal, ROIs will start to decrease, as the $1.00 keyword that delivers a profitable sale today will deliver an unprofitable one tomorrow. Then, two things will happen: First, marketing dollars will begin to flow back offline or at least flow online at a slower rate. Second, keyword prices will start trending down."

Or let's say Google loses its suit with the Association of American Publishers. The group has sued Google because Google is scanning books to create a searchable, and presumably monetizeable online card catalog. Does that mean that every Web site Google crawls to put in its index of billions of pages can demand Google get permission first? Even if they don't succeed, would it lead to a flurry of distracting and expensive lawsuits?

Should either of these events transpire, the investor revaluation could slam the entire Internet economy as hard as the dot-com meltdown, and Google would be hit the hardest. Despite all the neat products it releases seemingly weekly, Google effectively has only one revenue stream -- pay-per-click online advertising. (Yahoo has a big display advertising and fee business too)

Meanwhile, Google, with all the people it has been hiring and all the computers it has been buying, is not in a position to cut its costs quickly.

Even a small stumble could cut Google's stock in half. For example, to believe Rashtchy's price target of $600 a share you have to believe that the company will keep its PE of 50 and more than double its earnings per share over the next two years. But what if EPS grows only 30 percent in both those years -- still huge growth for a company Google's size -- and, as a result, its multiple contracts to 30? You get EPS of $8.79 at the end of 2007 and a $263 stock.

It all seems hard to fathom today. Indeed, when Google reports fourth quarter 2005 earnings at the end of January, Wall Street expects profits twice those of Q4 2004. But it doesn't take much to change the market's perception of a company, Blodget says. Few people on the planet understand that fact as well as him.