"It's a Snark!" was the sound that first came to their ears,
    And seemed almost too good to be true."
            -Lewis Carroll, "The Hunting of the Snark"

Sometimes the biggest obstacle to making a discovery can be knowing what to look for.

Researchers at the Brookhaven National Laboratory recently released the results of experiments they said might have produced an exotic form of matter called a quark-gluon plasma, which scientists believe existed moments after the birth of the universe.

But Brookhaven has scrupulously avoided claiming that the quark-gluon plasma had actually been created, leaving the nuclear physics community has been buzzing with speculation over what exactly Brookhaven made, and how their early results could be confirmed.

Now, the experimenters at Brookhaven are making more and more precise data on the new material, trying to establish whether what they have is a quark-gluon plasma -- even as the theoretical community debates just what that is, and what evidence would be conclusive enough prove that one had it.

When the findings were discussed at the Quark Matter 2004 conference in Oakland, Calif., days of debate failed to bring about a consensus on what Brookhaven physicists had created.

"We haven't succeeded in converging yet," said Miklos Gyulassy, a Columbia University theorist, after the conference.

Observing anything at the sub-nuclear level can be difficult: at that scale, there's no such thing as a thermometer, and even particles of light become powerful actors, streaking through the microworld like Star Wars lasers. Most measurements of very, very small things can only be made by smashing them together, or bombarding them with other particles, in the vicinity of detectors that pick up and record information on the debris that the collision throws off.

It's a little like firing a rifle at an object hidden in a cardboard box, and then trying to figure out what it was based on what you find splattered on the walls of the container.

The difficulty is exacerbated when researchers are looking for something, like quark-gluon plasma, that has never been seen before. Like the Snark hunters in Lewis Carroll's poem, they are pursuing a quarry whose nature is not fully known, armed with only a collection of distinguishing traits.

"We had a discussion on Friday, [at the conference] has the quark-gluon plasma been discovered?" said Krishna Rajagopal, a theorist from MIT. "Nobody has any other hypothesis that makes sense," he said, but added that some critical elements of what would constitute proof were missing.

Others, however, said there were still alternative explanations, including the possibility that a kind of plasma could be created out of quarks and gluons that were still bound together to form the subatomic particles called hadrons.

"What I would like to see is for someone to say: 'The following measurements are diagnostic of a quark-gluon plasma,' rather than some soup of hadrons" said Andreas Kronfeld, an elementary particle physicist at the Fermi National Accelerator Laboratory in Batavia.

Most people are familiar with how, as temperatures increase, the structure of matter breaks down: solids melt into liquids, which evaporate into gases. At extremes of temperature and pressure, even the structure of the atoms inside gases fails. Electrons are stripped away from atomic nuclei, creating a hot "soup" of charged particles called a plasma.

Under different conditions, physicists said, plasmas can behave like liquids, or like gases; they can be tenuous and flowing or extremely dense. We see plasmas every day, from the shining substance of stars and the flash of bolts of lightning to the glowing material inside a neon sign.

But nuclear physicists say there is an even more extreme physical state. The protons and neutrons that make up atomic nuclei are, in turn, made up of smaller particles called quarks and gluons. The strong force that binds them together is so powerful that physicists are unable to separate quarks and gluons long enough to study them in isolation.

But the theory that describes the strong force predicts that at high enough temperatures and pressures, even quarks and gluons will break free of their "bound states" and form a substance that behaves more like a continuous fluid than a collection of individually interacting particles.

Colossal energies would be required to create such a material. According to Rajagopal, it would require creating "conditions last seen a few microseconds after the Big Bang."

Physicists at Brookhaven's Relativistic Heavy Ion Collider, nicknamed RHIC, try to approximate those conditions. Powerful magnetic fields accelerate the nuclei of gold atoms to 99.995% of the speed of light, and then smash them into each other.

The temperature of the collisions has not been precisely measured - in fact, it's hard to say whether individual particles in such collisions can be said to have a "temperature" at all - but theory predicts the plasma would exist at an energy above 200 million electron-volts, said William Zajc, a professor at Columbia and spokesman for Brookhaven's PHENIX experiment. That's comparable to about 2 trillion degrees Celsius.

Even with a powerful accelerator like RHIC, creating a quark-gluon plasma would be no mean feat. Physicists have devoted years to finding ways to study quarks and other elementary particles under more and more extreme conditions.

Gyulassy compared studying quark-gluon states to the exploration of Mars. "We've known it's there for years, but we keep exploring it with more and more precise tools," he said.

It's clear both why some physicists would be very excited by the announcement that a QGP might have been found, and why others would be skeptical. On the surface, these conflicts could be seen as a conflict between the enthusiasm of discovery and the caution of a careful researcher, but there are deeper issues involved.

What constitutes "proof" of the discovery of something no one has ever seen? As Rajagopal said, it cannot simply be merely the absence of competing theories.

One test, physicists said, would be seeing how well experimental results match what theory predicts about the behavior of quark-gluon plasmas. But, as Kronfeld and Rajagopal both pointed out, quantum chromodynamics, the theory that governs the strong force, is often complicated and difficult to calculate.

They also noted that the conditions under which theory is created are often quite different from reality. Theorists can imagine creating quark-gluon plasmas under steady, sustainable applications of heat and pressure that are well beyond the technology available to today's researchers.

"We theorists have a very good understanding of the quark-gluon plasma at very high temperatures," Rajagopal said, but the conditions in RHIC are considerably cooler, and result in a substance with different properties than would be expected from a very hot quark-gluon plasma.

So what proofs have experimenters found?

Physicists widely agreed that two pieces of the evidence Brookhaven presented were strong indicators of a plasma state.

The first is something called jet quenching. When particles collide at high energies, they throw off debris, in the form of quarks that immediately break apart into a spray, or jet, of lower-energy particles.

Typically, researchers will see two jets of material ejected from a collision in opposite directions. In some of RHIC's gold-on-gold collisions, however, only one jet was seen, evidence that some kind of cohesive material was being created that was dense enough to stop the energetic particles thrown off by the collision.

A number of scientists think the material responsible for this jet quenching is a quark-gluon plasma.

Another oft-cited result was something called elliptic flow, observed in cases when two gold nuclei collided at a slight angle rather than head on. Researchers described an almond-shaped region in such collisions where the particles momentarily overlapped.

If individual protons and neutrons were colliding, physicists would expect the energy ejected from the region of overlap to be essentially even in all directions, like sparks from an exploding firework. However, the results of some collisions showed more energy being ejected along one axis than another - an indicator that material in the overlap region was "flowing" in a collective manner, like a fluid or plasma, rather than a collection of independent particles.

The liquid-like behavior of the plasma is, according to Rajagopal, one of the ways in which the relatively cool material being created in RHIC differs from what is expected of a quark-gluon plasma at higher temperatures.

"Theory predicts that at very high temperatures, the quark-gluon plasma is tenuous and gas-like. Such a plasma would slow jets, but not completely quench them. And, it would produce some elliptic flow, but not as much as observed," he said.

Other, more esoteric measures of particles ejected from the collisions also indicated that a plasma was being formed, possibly of quarks and gluons, physicists said.

For example, Rajagopal said, researchers have seen a much higher number of protons ejected from some of the RHIC collisions than is usually seen from collisions at that energy level. One explanation is that the protons could be coalescing from a quark-gluon plasma as it cools, which might account for some of the excess.

"What is the ultimate proof?" said Timothy Hallman, spokesman for Brookhaven's STAR detector, one of the instruments that measures the results of the collisions at RHIC. "That's a question that as a community we've been discussing."

The discussion has had its benefits. "On the theory side, everyone comes home sort of charged up with new ideas," Rajagopal said.

But for experimenters like Hallman, Zajc, and their colleagues at RHIC, the work of producing results that can clarify what is really going on in those microseconds in the accelerator chamber is just beginning.

This year, Hallman said, RHIC entered a mode called "high statistics running," in which RHIC will create many more collisions. Hallman said the experimenters on the project were hoping to increase their volume of data by a factor of ten. With more data, more precise measurements can be made, and more statistical information gathered.

One thing most experts agree on is that accurately measuring the temperature of the central region of the collision will be a major step toward confirming or rejecting a quark-gluon plasma as the explanation of the observed phenomena.

At the moment, Zajc said, researchers only know the temperatures of the end products - the stuff splattered on the walls of the box. At about 1 trillion degrees Celsius, these particles are much cooler than would be expected from a quark-gluon plasma.

But physicists expect that, like a conventional plasma, a quark-gluon plasma would "glow," or radiate photons. Judging the temperature of a hot object from the radiation is gives off is one of physicists' oldest tricks.

The measurement is tricky, Rajagopal said, because the collision also throws off a large number of other particles, creating "an almost blinding blizzard of photons, making it hard to discern the glow of the plasma."

Hallman and Zajc said the increased number of collisions in the current run should provide them with enough data to make a precise measurement of the number of photons coming from particle decays. That number can be subtracted from the total number of photons, isolating the radiated light from the collision.

There are Zajc added, other, rarer events that can help physicists define "the nature of the beast."

Studying the emission of rare objects called J/Psi particles can provide a second measure of the temperature of the collisions, Zajc said.

Experiments underway at Brookhaven to examine the movements of a rare type of heavy quark, called a charmed quark, could give physicists a better idea of the temperature, density and flow properties of the substance.

With jet quenching as a measure of the density of the substance, photon radiation telling them the temperature, and elliptic flow and heavy quarks giving further clues, physicists would have a fairly complete picture of what is being created at RHIC, and how it behaves.

That, Hallman believes, will help them eliminate the possibility that what they're seeing are the "normal" reactions of particles in which quarks and gluons remain in their bound states

So in the next year, physicists should know if what they've created "really is" a quark-gluon plasma.

But, Rajagopal said, that isn't necessarily the point. "The goal of the whole field," he said "is to study the properties of quark-gluon plasma at higher and higher energies, and to examine the phase transitions back from these excited states."

According to the Big Bang theory of cosmology, all the matter in the universe -- and in fact spacetime itself -- exploded from a singularity, a point of infinite density and temperature. Physicists believe that in the microseconds after that event, all that existed was a uniform soup of the most fundamental particles: a quark-gluon plasma.

As the universe expandedand cooled, matter as we know it, and the forces that bind it together, crystallized out of that plasma, Zajc said, much as ice crystals form when water freezes. Ninety-eight percent of the mass of protons and neutrons, he said, is "frozen energy" from the big bang.

The point is to develop that complete picture of particle interactions, not just to make goo in a particle accelerator. Understanding the properties of high-energy plasmas means understanding the story of the creation of all matter.

"You couldn't ask for more as an experimentalist," Zajc said. "It's the most exciting time you could envision."