In an extraordinary briefing yesterday, the FBI described in detail how a small army of scientists managed to trace samples of anthrax from the 2001 letter attacks to the bureau's chief suspect, microbiologist Bruce E. Ivins, at the Army's biodefense laboratory at Fort Detrick in Frederick.
A panel of six microbiological experts joined top officials from the FBI's laboratory in the briefing - apparently an attempt to address concerns expressed by some scientists and others about the strength of the evidence linking Ivins to the high-profile case.
Ivins committed suicide last month as investigators were preparing to indict him, authorities have said.
"This is very extraordinary for the FBI and the Department of Justice to discuss a case that has not gone through due process," said Vahid Majidi, assistant director of the FBI's Weapons of Mass Destruction Directorate.
The officials said the investigation ultimately involved scientists from a long list of federal, private and academic institutions, including the Defense Department, the FBI, the National Institutes of Health, the Centers for Disease Control and Prevention, and the Institute for Genomic Research in Rockville.
Five people were killed and 17 became ill from anthrax that was sent in letters to Congress and news organizations in September and October 2001. The lengthy federal probe into the attacks hit several dead ends and led to a $5.8 million payment to a man wrongly identified as a person of interest in the case.
Officials said yesterday that the scientific work began with the collection of more than 1,000 anthrax samples from around the world and led to several years of effort to characterize in detail the DNA of the anthrax used in the attacks. Then the scientists had to look for matches among the samples they had gathered.
That demanded the development of highly sophisticated tests. By early 2007, bureau officials said, scientists were able to identify eight subtle and unique genetic mutations in the anthrax letters, and by testing the 1,079 anthrax samples in their repository, they eventually narrowed the possible source to eight samples in their collection and finally to a single flask in Ivins' lab.
"Obviously, for us, the science is strong enough that we're disclosing it to you today," Majidi said.
Again yesterday, Ivins' lawyer disputed the government's account. "Because of their admitted mishandling of evidence, they are accusing him," said Paul F. Kemp, who has been representing Ivins' interests in the case.
The scientists said that other evidence, including Ivins' failure to provide investigators with anthrax samples they requested, helped to focus the bureau's attention on him as its chief suspect.
Careful study of the anthrax used in the 2001 attacks revealed that it was derived from a strain called Ames RMR 1029, used in government labs and a handful of other institutions.
Majidi said that investigators, before issuing subpoenas to everyone known to possess the material, asked Ivins in February 2002 to submit a pair of samples of any RMR 1029 that he had worked with in his lab. They asked that the samples be prepared according to a protocol Ivins himself had helped the FBI design.
One sample from the pair went to the FBI's repository. The second went to Paul Keim, a biologist at Northern Arizona University and director of the Translational Genetic Research Institute.
But the samples as Ivins submitted them, Majidi said, did not comply with the protocol. They were prepared in the wrong medium and in the wrong container. They would never stand up to scientific or legal scrutiny, the investigators agreed, so the FBI's sample was destroyed.
Asked in an April 2002 subpoena to submit samples again, Ivins did so.
Years later, after the development of more sophisticated genetic testing, investigators discovered that Ivins' first sample - the one sent to Keim's lab, where, luckily, it had been preserved - was a match to the eight mutations found in the attack letters.
The anthrax from Ivins' second sample, it was discovered, was not from the RMR 1029 strain that investigators had asked for. Asked whether they believed Ivins was deliberately trying to throw off investigators or cover his tracks, FBI officials declined to answer. They said they had called the briefing "to discuss science," not a suspect's motivation.
In the weeks since Ivins' suicide brought the investigation back into the public eye, the bureau has come under intense scrutiny.
Earlier leads had led investigators to another Fort Detrick microbiologist, Steven Hatfill, who was never charged with anything. He sued the FBI for harassment and won the $5.8 million settlement from the government.
One of the targets of the 2001 attacks, former Senate Majority Leader Tom Daschle, also received the scientific briefing. He said he found the evidence "complete and persuasive."
Still, he said, some open questions remain and the FBI's work needed to be scientifically reviewed.
FBI officials said their work was thoroughly vetted by independent scientists as it progressed. And several peer-reviewed papers on the technology used to reveal and match the eight genetic mutations in the anthrax letters have already been published. More are in preparation.
Majidi described the technology developed during the investigation as "revolutionary," with strong potential for further use in forensic pathology, biodefense and public health.