In a briefing for reporters, Majidi said the sample kept at the FBI lab was destroyed because the bureau believed it might not have been allowed as evidence at trial.
"Looking at hindsight, obviously we would do things differently today," Majidi said.
The science that let investigators look for tiny genetic mutations in the kind of anthrax used in the attacks was only becoming available around 2004, Majidi said. Not until then did investigators trace strains of genetically-unique anthrax back to Ivins' biodefense lab at Fort Detrick, Md.
"There were a lot of lessons learned," said Majidi, who heads the FBI office in charge of investigating weapons of mass destruction. "Were we perfect? Absolutely not. We've had missteps, and those are the lessons learned. ... It was over the last few years that we were able to incorporate all of the lessons learned that we have throughout this investigation."
As part of a February 2002 subpoena, Ivins gave investigators two samples of the unique anthrax strain known as RMR-1029 that he created in his lab. One went to the FBI lab, where it was destroyed. The other went to the lab of Dr. Paul Keim, a geneticist at Northern Arizona University in Flagstaff, Ariz.
Keim still had his RMR-1029 sample in 2006 when the FBI realized it could match Ivins to two batches of anthrax-laced letters that were mailed in the weeks after the Sept. 11, 2001, terror attacks. The anthrax letters killed five and sickened 17 after turning up on Capitol Hill, in newsrooms and postal facilities.
Ivins took a fatal dose of acetaminophen last month as prosecutors prepared to indict him for murder.