Back in April, right after I decided that it was time to join the tribe of cell phone owners, I heard something on the radio about cell phones being implicated in Colony Collapse Disorder in bees.  Honeybees would leave the hive and disappear, with no sign of dead bees.  The radio report indicated that researchers had found that cell phones were somehow interfering with the bees’ navigation system and preventing them from finding their way home.

Great.  Now my new cell phone was hurting pollinators – my graduate thesis was on pollination biology so conserving plant and pollinator interactions is something that’s been very important to me for a while.

I mentioned it to a few friends, but kept saying that I needed to read the paper – was it the cell phones themselves, or the cell towers?  If the latter, I’d at least only be indirectly at fault.

This week I finally tried to find the paper online.  Wikipedia has a nice site that provides an overview of Colony Collapse Disorder and has links to primary documents as well as media reports.

The good news is that the research in question had nothing to do with cell phones – a good reinforcement of the idea that it always pays to examine the research directly.  In two sets of experiments, scientists at the University of Koblenz-Landau in Germany took cordless phone bases (not cell phones) and put them right in honeybee hives.  These emit electromagnetic radiation all the time. 

In the 2005 study, four of 16 hives experienced Colony Collapse Disorder during the experiment, but one of these was a control hive with no cordless phone present.  The researchers recorded lower mean honeycomb weights and areas, and longer return times, for hives that had the phone bases than for the control hives, but these differences were not statistically significant (and, as my grad school advisor would say, therefore they really weren’t differences).  One of the main suggestions for improvement that the authors had was that next time around they should not place the exposed and control hives in blocks; instead they should distribute them randomly.

In 2006 they did so, and they added a couple of other new features to the study.  They made some physical barriers to create hives that had only 50% of the electromagnetic exposure that other exposed hives had, and now that the hives with phone bases were interspersed with the controls, they made some metal barriers to keep the electromagnetic waves from affecting the controls.  The effectiveness of these barriers sounds a bit sketchy and is not explained well in their report (both of these studies are only available in English as rough translations from the German provided by the authors); in fact, one of their main suggestions for future research is to actually measure the intensity of the electromagnetic radiation within each hive, which seems pretty critical.

Their sample size was tiny again – five fully exposed hives, three partially exposed hives, and eight controls.  They found similar patterns again – the return rates that they recorded for the controls were better than for the exposed hives (not much difference between the different intensity of exposure), but the differences were not statistically significant.  They then developed an index that combined not just how many bees returned but how long it took them to get there, and finally got a significant result – the fully exposed group had worse performance than the controls.  But it seems like they may have done quite a few statistical tests without providing a correction for multiple comparisons (the more tests you do, the greater the chance that one will turn up significant even if it’s not biologically meaningful, so you should have a higher standard for significance when you’re running a bunch of tests using the same data).  On the other hand, a larger sample size would have increased their power to detect significant differences, so the trends that they were seeing might actually stand up with a larger study.

What does this all mean?  To me it suggests that a statistically robust study on the impacts of electromagnetic radiation on bee behavior is warranted.  At the same time,  having a radiation source right in the hive seems extreme, but it also seems plausible that a cordless phone base in the hive might be comparable to the radiation emitted by some high tension powerline nearby – I don’t know enough about what levels of electromagnetic radiation bees would be likely to run into naturally.  It also means that while I don’t need to carry around direct guilt about my cell phone for the moment, electromagnetic radiation may still be taking a toll on wildlife.

The researchers themselves have been open with the press about the fact that they are not claiming that their study provides any kind of explanation for Colony Collapse Disorder – the original reporter that got the whole firestorm started never even contacted this lab.  Instead, the researchers suggest that Americans worry more about the contributions that herbicides and genetically modified crops may make toward honeybee declines than about electromagnetic radiation's effects.

The Colony Collapse Disorder Working Group’s Frequently Asked Questions page explains that they don’t think that bee feed, bee antibiotics, bee use (pollination vs. honey production), or source of queens are causal factors.  Instead, they are focusing on chemicals, pathogens, parasites, poor nutrition, stress, and lack of genetic diversity as possible culprits.  They also indicate that it’s more likely than not that there isn’t a single cause but rather that a combination of factors is weakening honeybee stocks.

So, I’m off the hook on my phones (cell and cordless) for the moment, but unfortunately there is no remedy for the disappearing bees yet, or for the crops and wildflowers that depend on them.

As we get closer to winning this year's legislative fight on the Endangered Species Act, we expect to see the Bush Administration turn its attention to attacking the Act by weakening the regulations that govern its implementation.  Late last week, the Administration suffered a serious setback to this "if we can't gut the law itself then we'll just gut the regulations" strategy when a federal court threw out its new rules regarding pesticides and endangered species.

“Pesticides are driving America’s wildlife toward extinction, and this administration wants to remove the checks and balances that hold them accountable,” Earthjustice attorney Patti Goldman told the L.A. Times in an August 25 story.  “It’s time for them to stop trying to sidestep the law, and start addressing this problem in a serious and systematic way.”

The court found that the Bush Administration "plainly violated" the Endangered Species Act by eliminating reviews by the U.S. Fish and Wildlife Service for new pesticides.