I'm not a scientist and really can't understand why no-one wouldn't have done a controlled test or comparison in laboratory conditions.
In my mind, it seems like declaring that a plant is growing gold, without testing the ground (or doing a lab test), and later finding out that its roots are picking up the gold from a seam near the surface.
It's impossible to cover everything, so you have to get an idea of what's interesting enough to publish.
Then when you do submit it for publication, the reviewers will say "have you tried X?", where X is something which takes a lot more time and funding than you have, and you really just want to be done with the paper.
Bob's - http://vimeo.com/album/2192316/video/51178467
The whole series - http://vimeo.com/album/2192316/page:2/sort:preset/format:thu...
I've just watched it again to double check, it wasn't a chore.
This year's - http://edibleschoolyard.org/library/edible-education-101-ris...