There's a diary on the rec list at the moment about a topic that I've been following for quite a number of years: Colony Collapse Disorder, the catastrophic die-off of honeybee colonies worldwide. I started to write a comment there, but it rapidly became clear that I sometimes have trouble staying in "reasonable comment length".
Based on the press release for an upcoming paper, the diarist announces the mystery of CCD solved. Every one in the world, much less Daily Kos, who has been following this story wants the mystery solved. Plus, it's hard to like the agricultural chemical industry, so diaries that point fingers at them always do well here. But science isn't like a courtroom; one opinion doesn't mean it's all over. And the science here is not over.
To that diary's author: please don't take this response the wrong way. You're reporting on cutting-edge research on a critically important topic, and I always consider such things one of Daily Kos's strengths. I just want to give some wider context here.
First off, here's a link to a pre-press proof of the actual paper. Reading science isn't like reading the news. Every research team -- no matter what is being researched -- wants the theory they are working on to be the right one, and so very nearly every press release ever issued about a forthcoming paper describes it as the convincing solution to whatever problem is at hand. Go to the source; most research papers outside of very niche fields are written in fairly approachable language these days.
This study is certainly worrisome. The data is strongly suggestive that imidacloprid, in doses formerly considered below the threshold for hard, may cause delayed death in honeybees. But there are some concerns with this study, and the authors were very professional and straightforward noting them. The most serious, in my opinion, is that none of the studied hives were reproductively healthy, including their controls. I'm sure most DKos readers aren't really interested in the breeding habits of bees. But basically, when their hives should have been having lots of little baby bees, they ... weren't. And that got worse as the study ran, for both the hives they treated with imidacloprid and their controls.
Before going any further, that is, in my mind, a pretty serious problem with this study. Something was wrong with their bees. Maybe it was some other environmental factor? Maybe something genetic? We don't know, and neither do the scientists involved. Bees weather accumulative stressors poorly, and there's lots of science on that topic. Would we see the same reactions to imidacloprid in hives without ... whatever was troubling these hives? We don't know. At least, not yet.
The authors also note that they didn't exactly duplicate the conditions seen in CCD hives. They had more dead bees, and less capped brood that has been seen in the "wild". They sealed dead hives, so there was no way to determine if their hives had the same resistance to looting as in observed CCD. They theorized that imidacloprid may interfere with larval development, somehow, and cause delayed adult mortality, but they didn't give us anything to support that hypothesis, and the leap they make to explain how CCD is still prevalent in France (where imidacloprid is and has been banned) struck me as particularly weak, as is the handwaving away of the 10 year delay between widespread US adoption of imidacloprid and the first significant CCD reports. And the way that imidacloprid may have killed bees in this study is very unlike how the similar chemical thiamethoxam appeared to kill bees in a study earlier this year.
This is evidence, certainly. But despite what they say, I don't think it's "convincing evidence". This is one small study that found a way to kill bees and hopes that's the smoking gun for CCD. We've been there before. In 2008, Israeli acute paralysis virus was "strongly correlated" with CCD. In 2009, it was problems with RNA transcription (maybe caused by IAPV, or maybe not, but probably not by pesticides) that was a "diagnostic marker" for CCD. But also in 2009, maybe depopulation was due to infection with the microsporidian Nosema ceranae instead.
There are probably a dozen more studies pointing the finger at other causes, or groups of causes. These studies are not generally fatally flawed, although many, like the one currently being discussed, have some significant shortcomings. Several of them fairly conclusively rule out things that are pretty conclusively ruled back in by others. All of them demonstrate that, unfortunately, it isn't really very hard to kill a lot of bees. CCD almost certainly does not arise from a single factor. Is imidacloprid exposure involved? Well, maybe. It absolutely doesn't help, and its ridiculously profligate use (thiamethoxam, too) should be reined in (it's bad for a lot of things).
But the mystery isn't over. Not yet.