http://online.wsj.com/news/articles/SB1036110025940498271
But we do -- because we tested it. We packed sample vials and took to the skies, hopping on 14 different flights everywhere from Atlanta to Sydney, Australia. On each, we collected water from the galley and lavatory taps, sealed them up and sent them to a lab for analysis. The results of our water-quality snapshot: a long list of microscopic life you don't want to drink, from Salmonella and Staphylococcus to tiny insect eggs. Worse, contamination was the rule, not the exception: Almost all of the bacteria levels were tens, sometimes hundreds, of times above U.S. government limits. "This water is not potable by any means," says Donald Hendrickson, the director of Hoosier Microbiology Laboratories in Muncie, Ind., which tested our samples.
The good news, of course, is that this water isn't the main drinking supply for passengers, who usually get bottled H2O from the beverage carts. But plenty of people depend on the plane's taps to wash their hands and brush their teeth. And while the airlines say they rarely serve tap water, many flight attendants say it isn't that uncommon: When the bottled water runs out, they turn to the tanks -- which, under federal regulations, are supposed to provide drinkable water. "It's the way our service works," says Sara Dela Cruz, a spokeswoman for the union of United Airlines attendants.
For their part, the airlines say they closely follow federal guidelines for drinking water, and say no passengers have ever complained about getting sick from it. "It's absolutely drinkable," says a United spokesman. They called our water tests unscientific, and said our own samplers could have contaminated the results. "Someone with dirty hands must have used that sink," said a spokesman from National Airlines, where the lavatory sample came back positive for coliform.
But our experts said human contamination wouldn't explain all our results. Some of the water we collected on a short flight to St. Louis, for example, contained Pasteurella pneumotropica, a bacterium primarily carried by rodents. Similarly, our Chicago-to-Los Angeles trip turned up Pseudomonas, a highly resistant bacterium associated with a range of infections. And while the U.S. government sets a maximum bacterial level of 500 "colony-forming units" per milliliter for municipal drinking water, our lab counted more than four million per milliliter in a single sample alone. That's roughly the same bacterial concentration you find in a tainted raw hamburger, Dr. Hendrickson says.