Scientists have discovered traces of the Marburg virus in specimens drawn from a species of fruit bats, the strongest evidence to date that some bats are the source of the deadly infection.

The discovery, based on bats collected in the rain forests of Gabon, was reported Tuesday in the journal PLoS One by researchers from the U.S. Centers for Disease Control in Atlanta and the Centre International de Recherches Medicales de Franceville and the Institut de Recherche pour le Developpement, Franceville, Gabon.

Coincidentally, several of the scientists involved in the finding are collecting bats outside a cave in Uganda where the most recent human cases of Marburg hemorrhagic fever were believed to have been infected.

"It's great.... We're really excited to see what comes out of these studies here and see if our theories hold up,'' Dr. Jon Towner, a microbiologist with the U.S. Centers for Disease Control said from Ibanda, in western Uganda, where the work is underway.

Scientists have been trying to discover the natural niche or reservoir of the virus that causes Marburg since the first outbreak of the disease in 1967. But with outbreaks few and far between, finding the source of the killer virus has been elusive.

In this work, Towner and others were conducting a fruit bat survey in Gabon and the Democratic Republic of Congo with Dr. Eric Leroy of the Centre International de Recherches Medicales de Franceville.

It is in this area where Leroy and colleagues reported finding traces of viral RNA and antibodies to the Ebola virus in three species of fruit bats in a study published in 2005.

Those fruit bat species were not the same as the one found to carry evidence of Marburg infection.

Towner explained that this survey wasn't undertaken to look for the source of the Marburg virus, but the researchers decided to put to use a diagnostic test they had.

"Sure enough, we got some hits,'' he said.

They found genetic traces of virus in four bats and antibodies to the virus in 29 others. Three of the four that tested positive for viral RNA also carried antibodies. All the bats appeared healthy and were foraging for food when trapped.

And all the bats were from one species of fruit bats, Rousettus aegyptiacus. Unlike other fruit bats in the area, these bats are cave dwellers, Towner said. And at least five of the seven known outbreaks of Marburg have had links to caves or mines.

"It's not proof positive that these bats are the reservoir. But it's kind of: where there's smoke, there's fire,'' Towner said.

Proving that a species is the reservoir of a pathogen is a long and laborious process. And experts say that finding live virus -- which this team did not do -- would be more conclusive that the bats are the source, not just a species that can survive infection.

Still, experts who aren't involved in the work believe this focus on bats is a promising line of exploration.

"If you find evidence in bats and you don't see a massive death associated with it, I think that's a good start,'' Dr. Heinz Feldmann, director of the special pathogens program at Canada's National Microbiology Laboratory in Winnipeg said in an interview last week.

Towner and colleagues from South Africa and the World Health Organization are collecting bats from a mine in Uganda where two men were infected with Marburg earlier this summer.

One of the men survived. He was with the researchers in the mine on Tuesday, pointing out where he'd been working before he fell ill, Towner said.

"We're really anxious to see what we get here, in this cave,'' Towner said, noting opportunities to search for the source of the virus so soon after the initial infection are rare.

So far the team has collected about 600, he said, with about half of those being fruit bats.