September 2, 2005
F.B.I. Abandons Disputed Test for Bullets From Crime Scenes
By ERIC LICHTBLAU
WASHINGTON, Sept. 1 - F.B.I. scientists said Thursday that they would abandon a controversial bullet-matching technique that had been used in thousands of investigations.
The Federal Bureau of Investigation said it still had confidence in the scientific reliability of the technique, which is known as bullet lead analysis and analyzes the chemical composition of a bullet. But in light of criticism of how the results were interpreted in court, the bureau said it would stop conducting the tests.
The bureau's laboratory in Quantico, Va., is the only one in the country that performs the analysis, an expensive process that seeks to determine how a particular bullet found at a crime scene compares with other bullets in the possession of a suspect.
A major study last year by the National Research Council found that the bureau's examiners had sometimes overstated the test results in concluding that a bullet had come from a certain box or batch of ammunition.
"It was a difficult decision to abandon a technique," said Dwight Adams, the director of the F.B.I. laboratory, in an interview. "This is a valid science, but the issue here goes beyond the science into the courtroom and the way evidence is presented, and there was the potential for the results to be overstated or misunderstood when an examiner would say that two bullets were analytically indistinguishable."
The F.B.I. has conducted about 2,500 bullet-lead tests since the 1980's as part of local, state, federal and foreign investigations. It imposed a moratorium on the technique last year after the research council's study before announcing Thursday that it would drop the practice.
The bureau is notifying some 300 law enforcement agencies that have received positive test results from the technique since 1996 to help them determine whether further review of their cases is needed.
A number of defendants who were convicted with the help of the bullet-testing technique have challenged the evidence because of questions about its reliability, and experts said the F.B.I.'s decision would probably lead to more legal challenges.
"This whole episode is a huge black eye for the F.B.I.," said William Thompson, a professor of criminology at the University of California, Irvine, who has studied the issue. "There are many cases where examiners were giving testimony that was wholly unreliable in claiming they could determine the same bullets came from the same box."
Ken MacFadden, a chemist who headed the National Research Council panel that produced last year's report, said juries were often led to believe from F.B.I. testimony that a "match" of two bullets could be likened to a conclusive DNA match, when it might be no more definitive than two people sharing a blood type.
The F.B.I.'s decision to stop using the technique went beyond the panel's own recommendations, Mr. MacFadden said. But, he added, "This decision make sense given the serious questions that are out there."