WASHINGTON (AP) - It takes special testing and time to prove if anthrax or its harmless cousins are really in suspicious letters cropping up from Nevada to Capitol Hill - and frustrated anthrax experts say it's no wonder those quick checks making headlines are confusing people.

Anthrax is a member of the large Bacillus family of bacteria. So are germs that cause mild food poisoning and others that are harmless to people.

Emergency workers checking suspicious letters like the one Senate Majority Leader Tom Daschle got Monday can do quick tests to determine if signs of the large Bacillus family exist to give reason to worry about anthrax. But this rapid first test is very often wrong because it cannot tell if B. anthracis, the specific anthrax germ, is the culprit or if less harmful Bacillus bugs are present, say biologists.

Checking if people have been exposed to anthrax is even muddier. All those TV images of Floridians getting their noses swabbed suggest that's a medically important test - it's not - and reports over the weekend that blood tests found five more were exposed were wrong, says the Centers for Disease Control and Prevention, which hasn't even finished the blood testing.

"It is confusing, isn't it?" said CDC Deputy Director David Fleming, who is struggling to figure out how to stem some of the public panic by better explaining the myriad tests.

"We've got to do a better job of communicating," adds bioterrorism expert Dr. Michael Osterholm of the University of Minnesota, who said the situation is spurring research into better rapid tests.

Anthrax spores measure between 1 and 5 microns in size - far too small to see with the naked eye, or, unless there are a lot of them, even with a quick microscope check.

So how do scientists prove exposure to the real anthrax bug - B. anthracis, not its harmless cousins?

As a "field test," scientists dissolve a little sample of, say, powder from a suspicious letter into a special fluid and run it through a gadget that checks for Bacillus family genetic markers.

"A positive would be suspicious" but not necessarily proof of anthrax, Fleming stressed. "A negative would be meaningless" because there simply might not be enough bacteria in the sample to register.

For proof, scientists culture the powder - put a sample in a lab dish with a special mixture of fluid and proteins that spur any bacteria present to grow. Then they use a technique called PCR to identify the genetic material of any bacteria there.

That can take as little as six hours if significant anthrax is present. The CDC won't declare a substance anthrax-free for 24 to 48 hours, allowing extra time for tiny amounts to grow.

How is a person tested?
People can be exposed to anthrax without getting sick, but there is no definitive test for exposure. People who handled a contaminated letter or were in a contaminated area are prescribed anthrax-preventing antibiotics on that basis alone, Fleming said.

Nasal swabs are simply for investigative purposes, a clue to how many people were exposed and where. Someone prescribed preventive antibiotics should not quit the medication if their nasal tests are clean, Fleming said, because the spores might have disappeared by the time they were swabbed.

If someone has possible anthrax symptoms, doctors take a sample of skin, blood or spinal fluid, culture it and do PCR testing, Fleming said.

Just as with environmental testing, the culture can be quick or slow, depending on the amount of bacteria in the sample. If they find anthrax, the CDC and a handful of other labs can take the next, more laborious step of growing enough bacteria to identify the anthrax's genetic strain. That's a key not to treatment but to determining the bacteria's origin.

What about testing blood for antibodies, which if they're present indicate the immune system has mustered an attack against the bacteria? Reports over the weekend falsely suggested antibody tests had found five more Floridians were exposed.

Antibodies don't develop for several weeks after exposure, no help if quick answers are necessary, Fleming said. The tests require two blood samples, taken a week or two apart. A rise in antibody levels suggests exposure has occurred, but CDC hasn't yet done those second tests on the Floridians.

The false reports were "a bit of a mystery to us. At some point in the chain, it's not clear where, there was some miscommunication," Fleming said.

Testing isn't perfect in any event. Also last weekend, Nevada health officials declared a suspicious letter mailed to Microsoft contained anthrax despite conflicting PCR results. The CDC's more sophisticated lab is redoing Nevada's tests to settle whether the Microsoft letter really was tainted.

Copyright 2001 The Associated Press. All rights reserved.