Feeding small amount of pesticide killed slowly but thoroughly wrote: Past research has shown that neonicotinoid pesticides, which target insects' central nervous system, do not instantly kill bees. However, to test the effect of even small amounts of these pesticides on western honeybees (Apis mellifera), Harvard researchers treated 16 hives with different levels of imidacloprid, leaving four hives untreated. After 12 weeks, the bees in all twenty hives (treated and untreated) were alive, though those treated with the highest does of imidacloprid appeared weaker. But by 23 weeks everything had changed: 15 out of the 16 hives (94 percent) treated with imidacloprid underwent classic Colony Collapse Disorder: hives were largely empty with only a few young bees surviving. The adults had simply vanished. The hives that received the highest doses of imidacloprid collapsed first. Meanwhile the four untreated hives were healthy.

"There is no question that neonicotinoids put a huge stress on the survival of honey bees in the environment," lead author Chensheng (Alex) Lu, an associate professor at the HSPH, told mongabay.com. "The evidence is clear that imidacloprid is likely the culprit for Colony Collapse Disorder via a very unique mechanism that has not been reported until our study,"

That mechanism? High-fructose corn syrup. Many bee-keepers have turned to high-fructose corn syrup to feed their bees, which the researchers say did not imperil bees until U.S. corn began to be sprayed with imidacloprid in 2004-2005. A year later was the first outbreak of Colony Collapse Disorder.

Bumblebees exposed to pesticide foraged less and made less queens wrote: A U.K. team exposed buff-tailed bumblebees (Bombus terrestris) to small doses of imidacloprid, and placed the bees in an enclosed natural setting where they could forage free. After six weeks (a far shorter time than Lu's study), the team weighed the nests and compared them with control colonies that had not been exposed. Treated colonies were on average 8-12 percent smaller than the colonies that had not been exposed, which implies that exposed bees were not gathering as much food. However, even more alarming was the case of the missing queens. Pesticide-exposed colonies produced 85 percent fewer queens: a total of only around 1-2 queens per hive. Queens are the most important bees in a colony, since they found new colonies after winter when all the other bees perish.

Pesticide-exposed foraging bees don't go home wrote: A second study, this one conducted in France, shows directly how neonicotinoid pesticides may be impacting bees, eventually leading to the collapse of the colony. Researchers glued tiny microchips to free-ranging western honeybees in order to track their movements. They then administered small does of a different neonicotinoid pesticide, called thiamethoxam, to a portion of the bees. Exposed bees were two to three times more likely to not return from foraging excursions than unexposed bees.

It has taken a long time to understand the link between Colony Collapse Disorder and neonicotinoid pesticides, because scientists were looking for an instant-killer, and not something that caused slow deaths over several months, says Lu. In addition he adds that scientists ignored "the fact that the timeline of increasing use of neonicotinoids coincides with the decline of bee populations."

Lu says policy makers "need to examine the effect of sub-lethal doses of pesticides throughout the life cycle of the test model (in this case honey bees)." He further notes the depending on LD50 findings (i.e. a lethal dose that results in the death of half of the specimens tested) "is not relevant to the modern day chemical toxicity testing." In other words, regulators need to start testing the long-term impacts of chemicals in the environment, and not simply focused on whether or not they instantly kill test subjects.