The pursuit of science is distinguished by a commitment to the systematic exploration of nature, since that is the way to discover the relations among the myriad phenomena of the natural world. Science is an enormously powerful aid to thought because it uncovers these relations, the so-called "laws" of, say, physics and chemistry.

The scientist is a species of map-maker; maps locate the features of some new terrain relative to one another. If we pressure the scientist into being guided less by the quest for understanding, for map making, and more by criteria of utility, such as the shortest distance from A to B, he will produce inferior maps. A line on a piece of paper, as the early explorers found, is not a map. Stray from it, and you are lost.

The things we discover as scientists, and shape into devices as engineers, should be things in which we can take pride, things, one might say, of beauty. I am speaking partly of a beauty the layman can recognize, as in the lines of a supersonic aircraft.

Still more, I am referring to a beauty that only the professional sees; economy and generality in a scientific theory; simplicity and unity in a device. These qualities hit the professional viewer in the solar plexus, eliciting a gasp of delight. "It fits, it must be right," says the scientist. "It fits, it has to work," says the engineer. No one but they can make that judgment.

Nor is it a judgment that can be made by any scientist, in the one case, or any engineer in the other. It must be made by a jury of those who have experience in the area in question, who know first hand what the difficulties are, where the opportunities and obstacles lie, and what the competition is doing. They are tough judges, as they should be, being the best equipped to recognize the genuine breakthrough as distinct from a meretricious one.

I once remarked that it is easier to make a discovery than to recognize that you have made one. Discoveries abound for those who have the eyes to see them. Most of us, however, are destined to walk around blindly, crushing diamonds under foot.

Having lived through that experience once or twice, scientists are in danger of making the opposite error—mistaking broken glass for gems. We could perhaps have committed such an error in the past weeks; it is too early to say. I am referring to the saga of room-temperature nuclear fusion.

Fortunately for the world, scientists are human. We continually make fools of ourselves. We rely on our colleagues to tell us when we have been blinded by the light emanating from some glittering scientific prize. Our colleagues, being also human, can be depended on to tell us how wrong we are.

Briefly stated, this is the process of checks that we term peer review.

It is a more complex process than it sounds, since the scientist who pays too much heed to criticism falls victim to the opposite error; he is unlikely to discover anything.

What we have seen recently are some courageous and competent scientists daring to think along heterodox lines. This is the essential prerequisite for discovery.

However, the "due process" of science was circumvented by the University of Utah press conference of March 23. As a result the normal procedures for establishing scientific truth, which constitute a process of examination resembling what goes on in a court of law, have been temporarily replaced by the notorious procedure of "trial by headline."

The facts that have been presented, particularly by Stanley Pons and Martin Fleischmann at the University of Utah, are hard to believe. Heavy water electrolysed in a cell, with a palladium negative electrode, was found to fuse deuterium atoms in a fusion reaction that generated substantially more energy than it consumed.

One would, a priori, expect metals to pull atoms of deuterium apart, not put them together. Moreover, if some deuterium atoms were nonetheless trapped as pairs, one would not expect the metal to push them so energetically at one another as would be required for significant nuclear fusion. This was the skeptical chemist's contribution to the debate.

The physicists then weighed in with the mystery of the missing neutrons. Given the energy release measured by Pons and Fleischmann there should be a billion times more neutrons emitted per second than are observed. Where did they go?

These questions were clearly enough raised, but I do not recall a headline based on them.

They are not casual questions, but represent a collision between what we think we know on the basis of many experiments performed in several fields of research over a period of decades, as against what the new findings would require us to believe. The verdict at present must therefore be, "unproven."

The thing that makes scientific discovery difficult is that there are more ways of doing an experiment wrong than of doing it right. This is the reason that, in science, seeing is not believing.

To insist on such troublesome questions now that we are triumphantly into the age of room-temperature fusion is to appear to be a scold. Yet, when the dust settles, it could be that we shall complain, with some justice, that we were insufficiently warned.

By this time an impressive number of groups in about a dozen countries have been reported in the press to have confirmed aspects of Pons and Fleischmann's findings.

Very few scientists have, of course, seen their results. Nonetheless, at the annual meeting of the American Chemical Society on April 12, 7,000 chemists gave Pons what was reported to be a thunderous ovation.

Newspapers tell us that engineers are debating the best means of exploiting the new technology in electric power plants, retrofitted home heaters, fusion-fueled cars and human implants.

The Utah state government is committing $5 million to the University of Utah to permit rapid expansion of room-temperature fusion research there. Canadian mini-science is not far behind. On April 14, a senior vice-president of British Columbia Hydro expressed keen interest in the new development ("it's a breakthrough") and announced a $10,000 grant to the University of British Columbia.

In the four days April 11-14, Ontario Hydro sold about 600 kilograms of heavy water to about 200 university and industrial laboratories in North America and Europe for fusion research projects based on the new principle. Palladium prices jumped on the stock market from US$40 to about $178 per ounce.

The best and most welcome joke of all would, of course, be if all of these seemingly preposterous dreams proved correct. Not only would we have witnessed a great discovery, but also (at least until the military demanded an entirely new panoply of weapons) a tremendous boon to mankind.

However attractive this prospect, we should not go on suspending the laws of nature in order (as the Greeks phrased it) to "save the phenomenon." The courts of science exist to try the defendant—a purported discovery—according to laws that actually exist.

Sometimes this will result in laws being refashioned, but first we must weigh the evidence by suitable refereeing, and communicate it in a systematic and informative fashion according to procedures which have evolved over centuries.

Since I have, in the course of this moral tale, been speaking about the conduct of science, I have also been speaking about its government. We have to insist in both contexts that the best professional judgment select the winners from among the multitude of projects that vie for attention and support. If the facts are unfavorable, our wishes must wait.

This entails stopping to ascertain the facts—an academic pursuit that we all too often equate with irrelevance, but that we should begin to recognize instead as the essential guarantee of utility.

John Polanyi, who won the Nobel prize in 1986, is a professor of chemistry at the University of Toronto. This article is based on his address to the Association of Professional Engineers of Ontario.