Nanotechnology, the practice of manipulating matter on the atomic scale, may demand an exact science. But so far, nanotech investing has not.

A few short years after Wall Street's first flirtation with the science routinely touted as the next big wave of innovation, there is more misunderstanding about nanotechnology among investors -- and more confusion than information. And it's not just the little guys who get befuddled; it's also big investment banks like Merrill Lynch.

On April 1, 2004, Merrill and its highly respected tech analyst, Steve Milunovich, launched the Merrill Lynch Nanotech Index, injecting a jolt of volatility into many of the 25 small-cap components. A week later, the firm quietly swapped out Cabot Microelectronics ( CCMP) for Cabot Corp. ( CBT). A week after that, Merrill pulled six companies out of the index after some of them -- including Applied Film ( AFCO), Emcore ( EMKR) and Three-Five Systems ( TFS) -- complained that they didn't see themselves as nanotech plays. It also added three others.

In one of the most eloquent admissions of error in the history of analyst reports, Merrill wrote, "Given the difficulty in defining nanotechnology, along with the dynamic nature of this emerging technology, we believe these changes will enhance our index, making it more useful to the marketplace."

By Merrill's own definition, the index has only grown more "useful." Nearly a year on, nine of the companies in the original Merrill Nanotech Index have been jettisoned, without the mergers or delistings that normally occasion such changes. Ten of the 27 companies in the current version were not included at the start.

Given all the volatility and uncertainty about nanotech stocks, why does anyone bother? Because as nanoscience reaches its adulthood and more laboratory breakthroughs transfer into commercial applications, the ability to engineer at the nanoscale has the potential to transform nearly every industry, just as assembly-line manufacturing did in the 19th century and networked computers are doing now. Merrill's index may have been embarrassing, but the fundamental point was right: Something big is shaping up in nanotechnology -- and Merrill was smart to pursue it.

Last year, companies and governments invested $8.6 billion in nanotechnology research and development, up 10% from 2003, according to Lux Research, one of the most respected research firms to rise up around nanotech. Companies such as IBM ( IBM), Hewlett-Packard ( HPQ), General Electric ( GE), NEC ( NIPNY) and BASF ( BF) are among the cutting-edge researchers in the field, although nanotech involves only a sliver of their operations today.

Some 20,000 people around the world are working in nanotechnology, a figure that the National Science Foundation expects to rise tenfold in 15 years. Forecasts for the nanotech market range all over the spectrum, but the most commonly cited is the NSF's figure of $1 trillion within the next decade.

So the interest in and commitment to nanotechnology are widespread, but beyond that, the uncertainty begins. Experts talk about the potential breakthroughs -- the ability to deliver life-saving drug proteins that the human body has so far resisted, batteries to power laptops and devices with memory chips that are many times tinier and more powerful than anything available today, elevators stretching into outer space. But no one can tell you when, or who will make them.

"There will be new companies formed that will create a phenomenal amount of value because of the revolutions they're creating," says Warren Packard, a venture capitalist at Draper Fisher Jurvetson, which has been actively investing in nanotechnology startups. "At the same time, there will be many, many companies that won't make it. The fundamentals of investing and of analyzing a company's prospects don't change. Just because someone calls itself nanotechnology, that could mean nothing for its prospects."

In the same way that desperate companies such as Zapata, a Texas fish-oil company, or K-Tel, a third-rate record label, tried to reinvent themselves as Internet stars in the mid-'90s, drawing the inevitable flurry of speculative trading, many companies will don a "nanotechnology" sash without necessarily having more to do with nanotechnology beyond the fact that they're housed in buildings composed of atoms.

But separating the fakes from the real thing isn't easy. "Right now, picking nanotech stocks is pulling needles from the haystack with very few needles and a pretty big haystack," says Matthew Nordan, vice president of research at Lux Research, which has its own index of large-cap and small-cap nanotech stocks (LUXNI).

Even defining nanotechnology is a good way to spur a debate among people familiar with it. One of the more commonly accepted definitions is from the National Nanotechnology Initiative, which considers nanotech the ability to control or manipulate objects on the atomic scale, with novel properties and functions because of their size.

But that raises confusion over the standard explanations of the nanoscale. Nanotechnology involves things larger than a nanometer but smaller than 100 namometers. To put that in context, people always reach for the statistic that a human hair is 80,000 times wider than 100 nanometers (or 50,000, or more than a million, depending on who's talking). But some human hairs are ten times wider than others.

So, how can the layperson understand the scale that this science involves? One way is to consider that ten hydrogen molecules fit into a nanometer. A better way is to consider that it would take 100 objects 100-nanometers wide to be seen by people with good eyesight. Another is to take a ride down an order of magnitude scale. Looking at the differences in scales, you can imagine that if you were a nanometer tall, the average person would be roughly as large as the orbit of the moon is to us.

At the lower end of the nanoscale, quantum properties take over, following laws that differ from bulk properties in some startling and unpredictable ways. Gold doesn't appear gold anymore, it looks red. The ratio between an object's surface area and volume rises. Very weak energy forces that are insignificant on the macroscale we live in start to play a significant role. The new laws are good news -- by opening the door to new possibilities -- and bad. Chipmakers are creating 90-nanometer chips, but once they get below 50 nanometers, they may need to rethink the way they make chips, investing in costly new processes that are not yet proven to work.

Turning these new properties into new products will take some time. Some observers point to the failure of Nanosys to go public last summer as a watershed moment in nanotech investing. And not a bad moment, but a very encouraging one that signaled that Wall Street wasn't about to get too manic about emerging nanotech companies. (Nor are venture capital firms, which invested $200 million in nanotech startups last year, down 39% from 2003.) One big success could shatter that sense of sobriety. But for now, investors have plenty of time to think rationally about nanotechnology.

"I think the market is smarter," says Tim Harper, president of U.K.-based nanotech-research firm Cientifica. "The Nanosys IPO was rejected for a very simple reason: Where's the product? You've got a bunch of IP intellectual property . So what?"

This is the first story in a multi-part introduction of nanotechnology to investors. Next: Common myths about nanotechnology.