THE FBI has suspended use of a controversial technique for matching bullets from crime scenes to unused ammunition in a suspect's possession, it emerged last week.
The technique in question relies on variations in the levels of trace elements in refined lead to identify the batch to which a bullet belongs. But as New Scientist revealed three years ago (20 April 2002, p 4) there are serious doubts about the technique's validity. The ratios may not be unique, and the batches of lead are so large that it may be impossible to distinguish bullets made months apart.
"The number of bullets from a single batch might be anything up to 35 million
A year ago, a US National Research Council review endorsed the concept, but said the FBI needed to improve its quality control and the statistics it used to assess the significance of matches. The panel warned that the number of bullets from a single batch might be anything from 12,000 to 35 million. Around 9 billion bullets are made each year in the US.
Last month, a New Jersey appeals court overturned the conviction of Michael Behn for murder and armed robbery, ruling that the lead analysis that helped secure the conviction "was based on erroneous scientific foundations".
The FBI is not abandoning the lead analysis technique. But after receiving the NRC report it has stopped using it in new cases "until it could address the statistical issues raised in the report", FBI special agent Ann Todd told New Scientist.
All comments should respect the New Scientist House Rules. If you think a particular comment breaks these rules then please use the "Report" link in that comment to report it to us.
If you are having a technical problem posting a comment, please contact technical support.
