FORGET mini and never mind micro. Those once titillating terms are virtually obsolete in the rapidly expanding field of ultra-small-scale science. Every month, more investigators are penetrating the formerly impregnable sub-microscopic world of individual molecules and atoms. At that level, prefixes like "micro," meaning a mere thousandth of a standard measurement unit, are as inapt as serving peas with a snow shovel. Research in the '90s demands a new generation of hyper-diminutives such as nano, pico and femto -- respectively a billionth, trillionth and quadrillionth of a second or meter.

But if the scope of inquiry has shrunk, its significance has exploded. Scientists are now starting to get answers to questions that couldn't even be asked a decade ago: What really happens at each incremental stage of a chemical reaction? How tightly does a single atom grab onto a surface? What does a molecule feel like? And exactly how do the infinitesimal subcomponents of individual cells work to move the human body around?

In the not-so-long term, the nanotechnology boom will mean fuels that burn with ferocious power but leave only the faintest puff of pollution, computers smaller than shirt buttons and such complex molecular re-engineering of body tissues that mutant human forklifts and 90-mph hamsters are by no means unthinkable. Watching the Ties That Bind Since the first bewildered hominid began wondering why sticks turned into charcoal, mankind has developed a formidable understanding of how chemical reactions transmute substances into new combinations and how to harness that process to make new materials. Unhappily, scientists generally have been restricted to observing only the beginning and end phases of the event and inferring what happens in between.

But that was before the "femtosecond molecular camera" and related devices that reveal step-by-step the mating dance of atoms. "We can now see the birth of a molecule in real time," says Ahmed Zewail, a physical chemist at Caltech. "On the most fundamental level, we are able to see electrons and nucleii interact with one another."

Like people, atoms and molecules don't generally combine or break up unless they get excited. The stimulus can be as simple as stirring a solution or heating a pot of soup, and most often the added energy is expended in rearranging electrons. Just as piping electric power to an elevator allows it to lift workers two or three floors, adding energy to atoms boosts electrons from their normal orbits around the atomic nucleus to higher energy levels. Once there, they raise the atom's potential for reacting with its neighbors to either break or form a bond.

Fortunately, these changes are easy to see. Just as the average desk jockey will change color dramatically if suddenly forced to run a mile, atoms and molecules give off different sorts of light when excited. For example, blasting a phosphor glob with the right kind of energy will make it fluoresce -- which is how your TV screen works. The frequencies that molecules absorb or emit depend on the kind of energy they receive, and different elements respond to different kinds of stimulus. So in order to observe very specific reactions, researchers can't excite compounds with something as gross and messy as heat.

Instead, they fire narrow beams of fine-tuned radiation at specimens; and scientists have known for decades which rays correspond to which altered chemical states. So in order to see exactly what happens, all that is theoretically necessary is to bombard Molecule A and Molecule B with a beam of energy (such as microwaves, X-rays or ultraviolet radiation) and watch how their emitted frequencies change at each stage as they are distorted to form Molecule C.

The problem is that chemical reactions often last less than a trillionth of a second. And, as any photographer knows, in order to "stop" the action of an event, the camera shutter has to be faster than the event's motion or the image will be blurred. For chemists, that means seeing reactions in fractions of picoseconds, far quicker than any mechanical device can operate.

In order to get intervals that short, Caltech scientists recently harnessed the fastest thing in the universe -- light. They devised a way to shoot two simultaneous laser beams at a molecule, but delay the second beam a few femtoseconds by forcing it to travel a couple of extra millionths of a meter farther then the first beam.

The first pulse "pumps up" the molecule to an energized state; the second or "probe" pulse arrives just as the molecule is in the process of reconfiguring itself part-way into the reaction. A detector monitors and records the frequency changes as they happen, giving researchers snapshots of how the molecules combine in "frames" as brief as six quadrillionths of a second.

Another way to get dependably measured micro-bursts of controllable radiation is by using a synchrotron -- a device that accelerates electrons in a ring by speeding them up and then bending their path with powerful magnets which are spaced at intervals along the ring. Thanks to a convenient law of physics, when charged particles are accelerated near the speed of light in a vacuum, they produce a band of radiation extending from the long wavelengths of infrared to short, high-energy X-rays. The electrons spin around in bunches, giving off a burst of light every time they pass a magnet. These closely spaced pulses serve a "shutter" function similar to the delayed laser probe.

Using the National Synchrotron Light Source at Brookhaven National Laboratory on Long Island, scientists are examining the way gas molecules interact -- with an eye to producing cleaner, more efficient burning of fuels. Says Jack Preses, a physical chemist at Brookhaven, "Essentially what we're doing is trying to look at the most elementary reactions that, when put together, make up the complex reactions that form the real world." Learning precisely how hydrocarbons combine and break apart may go a long way toward reducing the wheezing miasma given off by even the cleanest modern auto engines.

By laser standards, synchrotron pulses are excruciatingly slow -- on the order of a few hundred picoseconds, with the newest machines capable of pulses every half dozen picoseconds. But synchrotron beams are much easier to fine-tune at short wavelengths -- the only ones that certain organic molecules such as hydrocarbons will absorb. Combining results from both technologies will transform the kinds of materials we use. The Heart of the Matter Nanotechnology also has spectacular potential for medical research. To cite only one example, reported last month, scientists succeeded in using exquisitely fine beams of laser light to investigate in molecular detail how motion is created in the body.

All forces great and small -- from the large muscle contractions needed to lift a suitcase to the infinitesimal nudges that move components within a single cell -- ultimately arise from the action of individual motor molecules in a three-stage process:

A unit of fuel called ATP attaches itself to the motor molecule; the motor molecule then latches onto the object it is going to shove by building a hasty chemical "cross-bridge"; and finally the ATP breaks down into byproducts, releasing enough energy to push the motor molecule (and its attached load) a small distance. In a fraction of a second, hundreds of thousands of such tiny biological motors rev up in concert, enabling our hearts to pump blood and our arms to swat tennis balls or houseflies.

The general outline of the ATP process is well understood. But until recently, scientists didn't know exactly how hard each motor molecule pushed. Two different teams, reporting in Nature, used "optical tweezers" to measure the force exerted by motor molecules called dynein and kinesin in moving tiny objects along the microscopic filaments within a cell.

By employing specially tuned beams of laser light that were strong enough to shine into one individual cell and apply pressure -- but not powerful enough to destroy the delicate membranes -- they aimed the rays at a blob of matter that was being transported by as few as one or two motor molecules and increased the power until they had stopped the motion. Then they gradually decreased the power until the object "escaped" and continued on its way. Thus the experimenters were able to determine the range of force generated by single molecules.

This knowledge is particularly valuable because once a baseline for motor-molecule power is established, experimenters can begin to introduce mutations in the motor chemistry and watch how they affect molecular force.

Already "we are beginning to be able to manipulate the biological machine responsible for strength," says James A. Spudich, a cell biologist at Stanford Medical School.

Far down the experimental road, this might mean super-athletes. In the more plausible near term, however, nanotechnology may aid in treating dangerous weakness in critical organs by introducing specific strengthening mutations into common motor molecules such as myosin, found in virtually all human muscle tissue.

"It is now a trivial matter," Spudich says, "to alter the cell chemistry involved in cardiomyopathy," a weakening of heart muscles. And "given the rate at which things are going in molecular genetics, we may soon be able to replace that kind of myosin even in an individual." Moving Atoms One at a Time Elsewhere, a different kind of micro-manipulation is underway. Following recent innovations in a device called the scanning tunneling microscope (STM), scientists are now able to pick up and move individual atoms. Earlier this year, Donald Eigler and Erhard Schweitzer of IBM's Almaden Research Center in San Jose, Calif., reduced the temperature of a tiny nickel plate to a brisk 4 degrees above absolute zero (around -456

F) to minimize atomic vibration, sprayed it with a few atoms of a gas called xenon, and then dragged the atoms around until they spelled out the company name.

The STM, invented in the early '80s and used to sense the properties of objects much too small to be seen with the comparatively big, clumsy rays of visible light, takes advantage of the eerie properties that matter exhibits on the subatomic scale. For example, it is ordinarily handy to think of electrons as particles; but under the right conditions, they can also act like waves and "tunnel" their way through barriers that would be impassable to solid particles.

The STM exploits this effect. A superfine needle is given a tiny electric voltage and "scanned" across the top of an object, traveling only a few atomic diameters above the surface. As the needle moves, electrons respond to the voltage differential by tunneling across the gap, creating an extremely small current which is sensed by the STM. This current varies according to how far the tip is from the object; by recording the current changes, the STM "sees" the shape of the surface.

If the voltage on the tip is increased, the needle will attract whole atoms strongly enough to pull them around, as the IBM team did with xenon. Last month Almaden scientists dragged mounds of gold atoms into a map of the Western Hemisphere about one-fiftieth the width of a human hair, and did it at room temperature. Eigler has since managed to move individual metal atoms -- essential for experiments involving atomic-size electrical circuits. Another team is working with an STM-like machine called a magnetic force microscope that measures the magnetic properties of a surface.

The latter is of particular interest to computer companies such as IBM, says James Kaufman, manager of the condensed-matter experiments group at the Almaden center, "because we store data by magnetizing tiny domains." (That is, of course, how PC disk drives write information onto plastic floppy disks or metal hard-drive platters.) If data could be stored at the level of individual atoms, an encyclopedia would fit very comfortably on the head of a pin, with room left over for a few thousand grocery lists.

Curt Suplee covers science for The Washington Post.