Aug. 7 (Bloomberg) -- Scientists and legal experts questioned the reliability of novel genetic tests that the FBI says link deadly anthrax letters to an Army bioweapons scientist who authorities allege carried out the 2001 killings by himself.

Because the FBI has never offered such tests in criminal cases, it's uncertain the results would have been admitted in court as evidence against Bruce E. Ivins, who committed suicide last week, experts said.

``Microbial forensics is still a nascent field, and, as far as I know, no one has ever been convicted in a U.S. court on the basis of microbial forensic evidence,'' said Peter Hotez, a microbiologist at George Washington University. ``It's untested. In all likelihood he's the guy, but you can't say it's beyond a reasonable doubt.''

Gene sequencing that linked the highly refined anthrax spores used in the anthrax letters to a flask in Ivins's laboratory was a key piece of evidence cited by the Justice Department yesterday as proof Ivins acted alone.

The government said it will close the case soon. Before he died, Ivins, a researcher at the Army's bioweapons lab at Fort Detrick, Maryland, told people he expected to be charged with the five anthrax killings, court papers said.

The uncertainty about whether a trial judge would admit the test results into evidence may explain why the FBI didn't move faster to arrest Ivins after learning he had been hospitalized for exhibiting homicidal and suicidal tendencies, said one former federal prosecutor. Instead, Ivins was placed under 24- hour surveillance after his release from a mental hospital.

`Wanted Him to Confess'

``They never arrested him because they wanted him to confess,'' said Joseph diGenova, a former U.S. attorney for the District of Columbia. He said the FBI used a tactic designed to pressure a suspect into confessing because prosecutors knew ``there would have been all sorts of problems on the reliability of the scientific analysis.''

Ivins died July 29 after taking an overdose of Tylenol. He had been scheduled to meet last week with prosecutors, who planned to ``lay our cards on the table'' and tell him ``here's what we have,'' Jeffrey Taylor, the U.S. attorney in Washington, told reporters yesterday.

Taylor would not explain why the government chose not to arrest Ivins after learning he had been released from the mental hospital. ``Our job in law enforcement is to pursue our criminal investigation,'' he said.

Bulletproof Vest

In a court document released today, investigators said they seized a bulletproof vest, homemade body armor and ammunition during a July 12 search of his house, office and cars.

A federal judge today signed a warrant authorizing FBI agents to search computer terminals at a Frederick public library that Ivins was seen using while under surveillance. The FBI said it was seeking ``any indication'' of ``writings identifying a plan to kill witnesses or names of intended victims'' or ``suicide letters.''

The reliability of the evidence against Ivins would have been ``a major, contested battle'' if the case had gone to court, said Paul Giannelli, who teaches scientific evidence at Case Western Reserve University's law school in Cleveland.

``The government would have to establish the reliability of its methodology,'' Giannelli said. The FBI may have used ``traditional scientific methods, but their application would probably be new.''

Close the Book

Giannelli said the hurdle is not insurmountable for prosecutors, adding, ``I wouldn't say it couldn't come in'' as evidence.

Others said it's premature to close the book on the case and that investigators should share more evidence with the scientific community.

``I heard first a solid claim from the FBI that they are absolutely sure the genetic analysis pinpointed it, and then later on I heard some weasel wording that made it a little less crystal clear,'' said Philip K. Russell, a biodefense expert who directed emergency preparedness for the U.S. Department of Health and Human Services after the attacks.

``The proof is going to be in an independent scientific analysis of that data,'' said Russell, chairman of the Sabin Vaccine Institute in Washington. ``It's pretty complex stuff and will take some experts in bacterial genetics and bacterial molecular epidemiology to opine with confidence on it.''

Lab Reports

Ivins's lawyer, Paul F. Kemp, said the government never showed him the laboratory reports that officials said link the letters to the anthrax handled by his client.

``If they're so proud of these reports, why didn't they release those yesterday?'' Kemp said.

At yesterday's briefing, Taylor said the ``key breakthrough'' in the investigation came in 2005 when the FBI developed a test that linked spores found in the envelopes to the flask.

Investigators ``then focused their attention laser-like onto that flask, and the person who had control of that flask,'' Taylor said. He said it took another two years to exclude as suspects others at Fort Detrick who had access to the flask.

DiGenova said it was ``telling'' that Attorney General Michael Mukasey and FBI Director Robert Mueller allowed lower- ranking officials to announce the government had solved ``as big a case as everybody says it is.'' Mueller conducted private briefings yesterday for victims and members of Congress on the investigation's findings.

Their absence at the public announcement ``underscores the department's arms-length view of the facts,'' diGenova said. ``I don't think the attorney general or the director wanted to be facially associated with the conclusions reached in the case.''

In a statement, Justice Department spokesman Dean Boyd said Mukasey and Mueller ``were well aware of the conclusions reached in this investigation and fully support them.'' Boyd said it was decided that officials who were most familiar with the ``detailed and vast'' facts of the investigation should respond to media questions.

To contact the reporters on this story: James Rowley at jarowley@bloomberg.net; Avram Goldstein at agoldstein1@bloomberg.net

Last Updated: August 7, 2008 18:07 EDT