Microbiologist Adam Roberts went digging through men's beards in search of poop.
Instead, he found a tantalising suggestion of a breakthrough in the global fight against drug-resistant infections.
The research began, as very few good things do, with a fake viral story from the internet.
Last year a slew of articles were published with headlines like, "Some beards contain more poo than a toilet, shocking study reveals".
The "study" these stories cited was actually a project carried out by a TV journalist in Albuquerque, in which she swabbed a handful of men's beards and then had a microbiologist examine the cultures, explained the Nick Evershed of the Guardian. Among other things, the scientist found "enterics", bacteria that are often found in the intestine. As bacteria are not the same thing as faecal matter, and it's not all that uncommon for specific types of bacteria to be present in both the gut and on the skin, this wasn't really news - but try telling that to someone who's finally persuaded her editor to let her use the word "poo" in a headline.
Regardless of its scientific merits, the story did pique the interest of Roberts, a microbiologist at University College London who is part of a team of researchers searching for new antibiotics that could help stave off the growing drug-resistance crisis.
For almost a year now, he has been running a programme called "Swab and Send", which invites ordinary people to submit swab samples from the world around them to be tested for antimicrobial properties. He's gotten samples from computer keyboards, Egyptian bank notes, the floor of the London Undergound's Circle Line train, the inside of a toilet in a public bathroom and has so far isolated 20 promising strains capable of killing bacteria and yeast.
This is important, Roberts said in a phone interview, because the overuse and misuse of antibiotics has spurred disease-causing bacteria, such as Staphylococcus and E. coli, to evolve resistance to most of the drugs we prescribe to treat them.
"Humans have used antibiotics pretty indiscriminately since we first discovered them, and we haven't really found that many," he said.
Roberts was asked by the BBC to conduct a small study on the "poop in facial hair" theory for its Trust Me, I'm a Doctor series. Taking swab samples from 20 men's beards, he set about sorting through the 100 or so types of bacteria he found.
Roberts didn't find any evidence of fecal bacteria in beards. But he and his colleagues ran tests on all of the isolates as a matter of course. Amazingly, about 25 percent of beard isolates started killing off the indicator strains.
"They were producing their own antibiotics," Roberts said.
He and his colleagues will need to conduct more tests to figure out how exactly the antimicrobial isolates work and whether they're effective against more-serious multi-drug-resistant strains.
The hope is that the beard bacteria and other isolates from Roberts' "Swab and Send" project might aid in the development of new antibiotics, an area of science that has stalled since the "golden age" of antibiotic discovery in the 1950s and 1960s.