DNA on trial.

Imagine a scientific test so sensitive it could identify a crime suspect from a single cell. A microscopic drop of blood, a fallen hair, even the saliva on an envelope flap would provide a positive ID.

Sound like science fiction? It's not. The test, called DNA fingerprinting--which looks at the genetic material that makes each of us unique--may provide the evidence that sets O.J. Simpson free . . . or puts him behind bars for life.

Simpson is on trial for allegedly killing his ex-wife and her friend last June. Prosecutors claim DNA samples taken from blood at the crime scene and at O.J. Simpson's house will tie the suspect to the murders.

But some scientists and lawyers argue that DNA evidence shouldn't be allowed in court. DNA fingerprinting is a powerful technique that can help send dangerous felons to jail, they agree, but if used improperly, the test might help convict an innocent person.

Do you think this genetic evidence should be used? Join the debate as the lawyers and scientists battle it out in court.

Despite its name, DNA fingerprinting has nothing to do with the marks on our fingertips. Investigators still consider those unique patterns of ridges and whorls the most accurate crime-scene evidence. But the use of DNA fingerprints is on the rise.

First used in court eight years ago, DNA fingerprints are a record of a person's genetic makeup. The DNA molecules we inherit from our parents determine our individual characteristics, from foot size to eye color to whether or not you have a freckle behind your left ear.

CELLULAR ID

Except for identical twins, "everybody's DNA is different," explains Mark Stolorow of Cellmark Diagnostics, the company that is testing Simpson's DNA. That uniqueness makes DNA a valuable tool for identification, he says. It's as if your ID photo were visible on every cell in your body.

In the most basic DNA fingerprinting method, investigators isolate DNA from blood or tissue samples taken from the suspect. Using chemical "scissors" that cut DNA only at certain places, they chop the genetic material into tiny fragments (see diagram, above). They use the same chemical scissors to chop up the DNA found at the scene of the crime.

Since no two individuals have the same genetic makeup, the scissors cut each person's DNA differently, says Stolorow. If the cut-up fragments from two sources (the crime scene and a suspect) match, they almost certainly come from the same person, he says. A DNA match doesn't prove a suspect committed a crime, experts say, but it shows he or she was at the crime scene.

Defense lawyers can use DNA evidence too: If DNA from a crime scene doesn't match a suspect's sample, it raises serious doubt about the person's guilt. And in a criminal trial, a "reasonable doubt" is all you need to set a suspect free.

"That's what the defense is hoping for in the O.J. Simpson case," says Case Western University law professor Paul Gianelli. Last year, Gianelli says, DNA evidence helped free a Virginia man who had spent seven years in jail for a crime he did not commit.

DNA fingerprinting also requires very little physical evidence, says Stolorow of Cellmark Diagnostics. In one type of DNA test, he says, "we can analyze blood stains as small as the head of a pin."

QUESTIONABLE QUALITY

DNA fingerprinting may sound like the ideal evidence for an open-and-shut case--if the tests are done accurately. But that, say opponents of using DNA evidence, is a big "if."

"There have been instances where the tests are not done as carefully as they should be," says William Thompson, a law professor at the University of California at Irvine. If a distracted lab worker accidentally contaminates the suspect's DNA with DNA from the crime scene, the fingerprints could match, even though the suspect is innocent.

Professor Gianelli says passing laws requiring lab workers to be specially trained and tested could reduce the chance of such errors. "It's a disgrace," he says,"that crime labs are not regulated" to insure accuracy.

Another problem with DNA fingerprinting, says biologist Daniel Hartl of Harvard University, is that labs do not test all of the 100,000 genes in human DNA. Instead, they check for matches in only four or five DNA fragments. Sometimes, just by chance, two people will have similar DNA in those particular segments. Thus, the DNA fingerprint from an innocent suspect could match that of the real criminal.

How often do such false matches happen? Experts disagree. In one case, FBI experts testified that the odds of a false match were 1 in 6 million. A defense expert in the same case said the chance of a random match was more like 1 in 94. A jury hearing the second number is far more likely to have doubts about a suspect's guilt, says law professor Thompson.

Some scientists say juries shouldn't even be given error estimates. "You shouldn't try to convict anybody based on statistics," says biologist Hartl. Instead, he proposes expanding DNA tests to look for matches in more fragments of the DNA. This would take and cost more money, he says, but it would reduce the chance of an innocent person going to jail.

Experts on both sides emphasize that the scientific theory behind DNA fingerprinting is sound. Everybody agrees that every person's DNA is different, and that a perfectly performed test can detect those differences. "What we've fought about," says Gianelli, "are [the accuracy of] the statistics and the [scientific] procedures."

The public focus on the O.J. Simpson case might force scientists and the government to find ways to resolve some of these disputes. Until then, do you think DNA evidence should be used in court? Debate and decide.