Williams, et al (2015) Neonicotinoid pesticides severely affect honey bee queens.
Updated Nov 2: The science of the effects of neonics upon pollinators is so contentious, that I am often asked to comment on new published studies. This study, which claimed an effect of neonics on queen mating, immediately garnered such requests. Unfortunately, the lead author was at conference at the time, and was unable to fully discuss the paper and answer all my questions until recently. I now wish to revise my comments accordingly.
I’ve now had the chance to discuss the paper at length with the author, of whom I think highly (and with whom I previously coauthored a paper). We’ve discussed the shortcomings in the original design of the experiment, which come back to haunt us when we try to interpret the results.
In this study, queens were reared in either Test or Control cell builder colonies–the Test group being fed neonic-spiked pollen; the Control group received unspiked pollen.
The main finding of the study was that after four weeks of being in mini mating nucs stocked with roughly 1000 workers, fewer queens were successfully laying in the Test group that had been fed neonics during the cell building period. The question then, is why?
Since nurse bees typically produce jelly free of pesticide residues, and since the researchers did not test the jelly in the queen cells, we have no way of knowing whether the developing queens were directly exposed to neonics. Without such confirmation of exposure, one cannot say that the queens were actually exposed to the treatment. The bees that were clearly exposed were the nurses that reared the queen larvae and later used to stock the nucs.
For the experiment, the newly-emerged queens were placed into mini nucs stocked with roughly 1000 workers.
The bees used to stock the Test nucs came from colonies that had been fed neonics for 36 days, ensuring that those workers had been reared under continuous neonic exposure, and then when they emerged had likely consumed neonic-spiked pollen for their entire nursing period up ’til when they were transferred to the mating nucs.
The bees used to stock the Control nucs had received no exposure to the insecticides.
Interpretation of the results is thus confounded by our not knowing whether the observed effects were due to changes in the queens or changes in the workers in the nucs.
The result was that there was greater mortality and/or failure to lay eggs of the queens in the neonic-treated nucs.
But it was difficult to tease out exactly why. The Treated queens flew and mated exactly the same as the Control queens and had significantly more ovarioles than the Control queens, but may have had somewhat fewer stored spermatozoa(it was difficult to tease out statistical significance due to the small numbers and high variability).
The question then is, were the apparent differences between queen performance due to direct effects of exposure to the queens during their larval development, or were they due to the performance of the workers in the nucs?
Could the apparent differences in the number and viability of the spermatozoa be due to poor queen care by those 1000 treated workers, or due to poor thermoregulation by the neonic-exposed workers, or simply due to their premature death leading to rapid dwindling of those 1000 bees in the nucs? The experiment was not designed to answer those questions.
In answer to my questions, Dr. Williams tells me that none of the queens were observed to be drone layers. Those which had not laid worker eggs had simply not laid any eggs at all (although there is the possibility that the workers cannibalized them; this often occurs in hungry nucs).
I also asked whether the worker population had dwindled in any of the nucs. Answer: “If the queen was there, regardless of egg-laying, there were always ample numbers of workers.” Unfortunately, this does not exclude the possibility that the workers had abandoned some nucs.
Although the findings of this study are strongly suggestive, the small number of subjects, plus the innate variability of mating success (I rear thousands of queens each spring), makes me hesitant to draw firm conclusions until we see the experiment replicated, which Dr. Williams intends to do. I have also volunteered to collaboratively replicate it in California next spring.
This paper is open access, and can be downloaded at Neonicotinoid pesticides severely affect honey bee queens