Not all types of mass transit are at equal risk from bioterrorism, however. “Buses are pretty leaky,” says Dr. Tara O’Toole, deputy director of the Johns Hopkins Center for Civilian Biodefense Studies. They travel through the open air and open their doors every couple of blocks. That would dilute most surreptitious releases enough to limit casualties. Commuter rail lines also travel aboveground and stop often. Subways, however, are another matter. Because they are underground, they do not receive the ultraviolet radiation that can kill many pathogens (but not chemical agents). An invisible cloud of anthrax spores could therefore infect some people immediately and, after spores eventually fell to the ground, infect again when crowds or trains relaunched them. “They’re also at a pretty optimal temperature”–far from the extreme heat or cold that can damage chemical or biological agents, says Jason Pate, a terrorism expert at the Center for Nonproliferation Studies in Monterey, Calif. And as the 1966 Army study found, drafts created by speeding subways can spread an agent from station to station in minutes.

If mounting an attack is relatively easy, the consequences of one depend on many factors. The virulence of the agent matters, as does whether it is contagious (smallpox is, botulism and anthrax aren’t). So does the delivery system and, of course, emergency response. “When you put all those factors together,” says Michael Moodie, president of the Chemical and Biological Arms Control Institute in Washington, D.C., “in some cases you come up with a very high [toll] and in some cases with a low one. It’s important not to assume everything is going to go right for the terrorist.” The nerve agent VX is not very volatile, for instance, so releasing the liquid would not produce a particularly deadly fog. And Aum Shinrikyo didn’t do very well with sarin, which evaporates after two hours at 50 degrees. For the highest toll, a terrorist would have to spread a poison or pathogen with an aerosol generator. Even busy New Yorkers would likely notice that.

Worst case? A group of subway riders is unknowingly infected with a biological agent. They go their separate ways. Days later, people in far-flung places become sick. “Releasing something on public transportation makes it more difficult to figure out what the site of release was and hence who else is at risk,” says O’Toole. Not knowing the source of exposure–where did these 87 people cross paths?–makes it impossible to tell the public who else is at risk. In a smallpox attack, that hurdle would cost precious days and allow the contagion to spread.

Transit officials got serious about bugs and gas after Aum Shinrikyo’s sarin strike. “It was a dramatic event,” says Mike Healy, spokesman for the Bay Area Rapid Transit System (BART) in San Francisco. “It’s something our people have been cognizant of ever since.” The next year BART sent a team of police officers to the Army’s chemical-weapons facility in Fort McClellan, Ala., for training in bioterror prevention. That spurred BART to develop a response plan, with regular drills to practice evacuations and identify agents based on passengers’ symptoms, which is crucial for knowing what first aid is needed. Other cities have similar plans in place.

Transit systems will spend hundreds of millions of dollars in the next few years to upgrade security. Washington, D.C.’s plan is a model: its transit system is unique in having a sensor system for chemical agents in several stations. The city has now asked the federal government for $81 million to add biosensors. Reliable biosystems suitable for subways, however, are still years away.

FIRST STEPS: Refine biosensor technology. Pump up video and undercover-police surveillance. And make sure that transit and emergency workers are trained in terrorism response.