The Ebola virus that struck Kikwit, Zaire, last spring invaded the skin cells, sweat glands and connective tissue of its victims, researchers now say, suggesting it was probably capable of infecting people through mere skin contact.

Additionally, evidence found in an ongoing probe into the outbreak suggests the virus might be transmissible through the air by coughing. Scientists said they found loads of viruses inside and on the outside of alveolar cells, which exchange carbon dioxide for oxygen in human lungs. Previously, scientists thought the only modes of transmission of Ebola involved direct contact with contaminated blood or bodily fluids.

"We see it [Ebola] in the alveoli on both sides, so that means that it's growing in the macrophages [white blood cells] around the alveoli and inside the alveolar cells," said Dr. C. J. Peters of the Special Pathogens branch of the U.S. Centers for Disease Control and Prevention in Atlanta. "That means that, theoretically, virus is available to be spread from the lungs."

But "availability" doesn't always equal actuality: the mere presence of virus in human lungs doesn't mean that it is actually spread on the air exhaled from the lungs.

"When you cough you don't cough out of your alveoli. You cough from your bronchus," Peters explained. "So it doesn't prove the virus is spread, it says it's there. It's the kind of finding that you don't like to see, but you don't know what to make of it."

The CDC is only at the beginning of a lengthy process of scrutinizing nearly 50,000 animal samples and hundreds of human blood and tissue samples gathered in Zaire during the May outbreak.

For two decades, scientists have been fine-tuning methods for finding viruses in tissue samples kept in formalin preservative, rather than in much more dangerous-to-handle blood samples. Using refined staining methods, Dr. Sherif Zaki -- a colleague of Peters at the CDC -- was able to spot large clusters of Ebola viruses in all nine sets of human samples, from individuals whose Ebola illness was previously confirmed by blood tests.

The same method confirmed the presence of viruses in their lungs.

Finding viruses in the sweat glands and skin "makes us a lot more concerned about touching the bodies [of Ebola victims]," Peters said. World Health Organization virologist Jim LeDuc agreed in a telephone interview from Geneva, adding that it seems the precautions taken by scientists and citizens in Kikwit adequately allowed for the possibility of skin transmission.

The key to stopping Kikwit's outbreak, which claimed just under 250 lives, was isolation of Ebola patients and proper disposal of their bodies.

Still unanswered is the question of where Ebola comes from -- what animal or insect it hides in when not causing human epidemics. The international team of researchers that combed the rain forest region outside Kikwit this summer sent more than 18,000 animal samples and 30,000 insects for analysis, Peters said.

So far, Peters said, the CDC has been overwhelmed by the task of cataloguing and identifying all the species, most of which are totally unfamiliar to North American scientists. Researchers from Zaire and other countries are assisting the CDC and U.S. Army scientists at Ft. Detrick, Md., in the identification effort.

"It's going to be months before we know anything," Peters said. "And let's face it, it's a massive undertaking."

Next week, experts from all over the world who worked on the ground in Kikwit or in laboratories in three countries will gather in Geneva to review their findings. No surprises are expected, but WHO officials plan to look at lessons learned by this outbreak and announce creation of a new rapid response force to react to future epidemics all over the world.