Nosfer, thanks for the links.

Unfortunately, I'm leaning towards that first link.  It does a nice job in explaining some of the problems.  However, there was also this:

Microbiologist Rosie Redfield was quoted saying: "If this data was presented by a PhD student at their committee meeting, I’d send them back to the bench to do more cleanup and controls."

I know peer reviews can be brutal, but I think that was a low blow.  No reason to go there.

The other link…this quote stood out to me:

"There's dozens and dozens of things that could be done, and should be done," he (co-author of the paper Ron Oremland) said. "We can't do everything."

That is "science-speak" for "I know we didn't follow every single procedure, but our results are still valid."  That…doesn't sit well with anyone reviewing the paper.  We don't like missed steps.  There are procedures for experiments for a reason.  In my labs, I had a saying "There are two ways to do any experiment.  The right way.  And the way that leads to the unemployment office.  Choose wisely."  One of my guys actually made that into a poster that hung in one of the labs.  Anyway…people skip certain procedures to either save time or really it's just laziness.  Either way…it is unacceptable…especially with something as big as this paper….

Well with any new "discovery", the experiment really needs to be replicated by other scientists.

I once asked on the JREF forum what would it take for "you" to believe in ghosts or whatever, can't remember the exact wording now. Many answers came back that the experiment has to be replicated by respected scientists. This replication is necessary to rule out mistakes made by the original researchers. As careful and as professional as they are, everyone makes mistakes once in a while eh?