DeCode genetics makes headlines for its pioneering discoveries linking genes to heart disease, cancer, diabetes and other deadly diseases. It hopes its genetic tests will one day save lives by spotting people at risk years before they get sick. That is, if the company itself can survive.

DeCode shares are worth 23 cents each, down from $18 at their initial public offering in 2000. The company's most advanced heart drug is stalled, its genetesting business faces all sorts of competition, and it is running out of cash fast. DeCode needs more money by early 2009 to continue operations and is exploring asset sales. A white knight acquirer may still emerge. But don't bet on it, says Morningstar analyst William Buhr, who predicts bankruptcy within six months or so. DeCode founder Kari Stefansson declines comment; in a conference call in November he blamed DeCode's woes on auction-rate securities in which money manager Lehman Brothers had invested much of its cash balance.

DeCode has company. Almost half of 370 small-cap biotech companies have less than one year of cash left, according to the Biotechnology Industry Organization, and prospects for raising more funds are slight. Big, profitable biotech companies like Amgen are holding up fine. But many small companies rely on repeated equity financings to keep their research going for the 15 years it can take to design a new drug. In the past investors have been willing to gamble on the drug lottery. Now they're fleeing.

Shares prices of 70 U.S. biotech companies have slid below a dollar in the last year. Total financing for biotechs fell 54% for the first three quarters of 2008 to $8.2 billion, versus $17.9 billion in the same period last year, according to Burrill & Co. The industry has gone a year without an initial public offering of at least $6 million, the second-longest period since 1990, says Jefferies & Co.

Two dozen biotechs have announced layoffs in the last two months, while others are closing down nonessential programs in a desperate bid to survive. Former highflier AtheroGenics filed for bankruptcy in October; its heart drug fizzled in a big trial in 2007. CombinatoRx of Cambridge, Mass. just laid off 65% of its 180 employees after a promising arthritis drug fell short of its main goal in a trial, despite some signs of efficacy. The company doesn't have money to repeat the trial. Says Chief Executive Alexis Borisy, "The business model is broken at the moment." Ernst & Young partner Glen Giovannetti calls the current period "the deepest downturn in the history of the industry."

Zack Lynch, whose Neurotechnology Industry Organization represents 50 smaller companies pursuing treatments for brain disorders, worries that if the downturn continues, "we may lose an entire generation of potential brain drugs" as companies vanish.

The industry has joined the list of supplicants hoping for a federal bailout of sorts. Profitable companies now get a tax credit for R&D expenses; biotech firms are lobbying for moneylosing firms to get, essentially, an advance on this tax credit. "There is a lot on the line—we don't just make widgets, we make drugs that save lives," says James Greenwood, who heads the Biotechnology Industry Organization.

Big drug companies are benefiting from small-company woes, as early-stage products suddenly go on sale. "It is an opportunity," says GlaxoSmithkline Senior Vice President Adrian Rawcliffe. In October his company agreed to buy Genelabs Technologies, which is working on hepatitis drugs, for $57 million, or $1.30 per share.

In November Roche inked a deal to buy Memory Pharmaceuticals, which invented two Alzheimer's drugs Roche is testing, for $50 million, a mere 61 cents a share. That's a nice gain over the 15-cent quote on the shares the day before but quite a comedown from the $7 new-issue price in 2004. Biotech companies' fundraising woes means "we will see more biotech companies looking for an acquisition," says Daniel L. Zabrowski, Roche's global head of drug deals.

Biotech Exelixis in South San Francisco took a different approach in mid-December. It agreed to join with Bristol-Myers Squibb to help develop two of Exelixis' cancer drugs. Bristol-Myers is paying Exelixis $240 million under the deal. Exelixis' shares had plunged 58% in 2008 until the announcement and, according to Chief Executive George Scangos, the company had largely given up on selling equity to raise cash. Its shares soared 33% the day the deal was announced.

Biotech is a business mixing a few fabulously profitable winners like Genentech and Amgen with a large population of dream-stage outfits that lose money for years and years. Since 1990 publicly traded biotech companies have lost a cumulative $50 billion, according to reports from Ernst & Young. (The number would have been smaller if it had included the profits of successful ventures after their acquisition by big drug companies.) Biotechs have done better recently; in 2007 U.S. biotech firms nearly broke even on $65 billion in revenue, up 11%.

Bayer HealthCare Chairman Arthur Higgins says many investors have been naive about the long odds small companies face. They "convinced themselves we have made massive progress in understanding biology, when in reality we have made limited progress," he says.

Indeed, many people assume that biotechs are more efficient at inventing drugs than drug companies are. But that "is a complete myth," says Harvard Business School professor Gary Pisano. He pored over 20 years of data for his 2006 book Science Business and found that biotech companies spend just as much as drug companies for each compound that reaches the market, roughly $1.3 billion. While biotechs are more nimble, they often lack the broad skills needed to develop drugs, he says. "Every idea getting its own company may not be the best way to organize the sector," he says.

Hedge fund manager Larry N. Feinberg, whose Oracle Partners specializes in health care, used to invest in young biotech companies with no products. But he stopped in 2002 after concluding there was little reward for all the risk. What drugs are likely to work? "You can have Nobel laureates advising you [about that], and they are wrong half the time," says Feinberg.

Still, investors who buy when the market is depressed can make money; witness the quadrupling of Memory's share price. The table on this page lists companies that Leerink Swann or JMP Securities consider good gambles: These are smaller biotechs with drugs that have already shown signs of effectiveness. Says biotech banker Steven Burrill, "If you buy a good sampling of the industry now, five years from now you are going to look like a genius."