Research into potent bioagents increases the risk - Hundreds of universities and labs have joined the study of toxic microbes. Since 2003, there have been 111 accidents.

Her work was in a different type of bacteria, but when asked to help clean a chamber that had been used to create an aerosol version of Brucella, she leaned inside and wiped it down.

The bacteria entered her body through her eyes, investigators later surmised. She was infected for more than a month before doctors diagnosed her with brucellosis and put her on a regimen of strong antibiotics.

The incident last year was part of a small but unsettling number of laboratory accidents that has followed a boom in research funding after the Sept. 11, 2001, terrorist attacks and the still-unsolved anthrax mailings that came a week later.

The burst of money has spread biodefense work to hundreds of university and research laboratories. In some cases, the labs have been ill-prepared to work on the exotic microbes.

"Universities aren't set up to handle these programs," said Edward Hammond, U.S. director of the Sunshine Project, a nonprofit group in Austin, Texas, that tracks information on biological weapons research. "I think we made a serious mistake putting 400 labs, thousands of people in the U.S., in the driver's seat behind biological weapons."

All told, there have been 111 cases involving potential loss of bioagents or human exposure reported since 2003 to the national Centers for Disease Control and Prevention or the U.S. Department of Agriculture.

The incidents include the potential exposure of 12 laboratory workers to live anthrax bacteria after an incorrect sample was sent to Children's Hospital Oakland Research Institute in 2004, the infection of three researchers at Boston University in 2004 after they mistakenly handled a sample of live tularemia bacteria, and the disappearance of a mouse infected with Q fever at Texas A&M in 2006.

Federal officials say that the overall number of incidents is small, and they emphasize that no one has died -- and that no one beyond laboratory workers has been infected.

"If you're looking at the total amount of work in these labs, it strikes me that 100 incidents is very low," said Dr. Richard E. Besser, director of the CDC's Coordinating Office for Terrorism Preparedness and Emergency Response. "Full investigations were done, and none of the events were thought to put the public at risk."