For more than a century, Mark Sundberg's family has had bee hives on the family farm.
In a good year, Sundberg Apiaries produces 800 barrels of honey, or about half a million pounds. But a bad year can cut its production in half — and recent years have largely been bad.
Honeybees are in trouble and their population is steadily declining — in part because of the increasing use of pesticides and a drop in the number of flowering plants. A growing problem is the prevalence of a pest called varroa mites, tiny parasites that attack bee larvae and adult bees in the hives.
Some beekeepers are trying to fight disease by improving bee genetics. Among them is Sundberg, a third-generation beekeeper who for several years has been selectively breeding hygienic bees that recognize sick bees and remove them from the hive. That cleaning helps break the mites' reproductive cycle and limits the spread of disease in the hive.
After University of Minnesota experts help Sundberg identify bee larvae with the hygienic trait, he transfers those larvae to other hives to create new queens with the trait he wants.
To check on whether the process is working on his farm, Sundberg fires up a smoker to calm the bees. He then pries the wooden tops off the hives.
"It seems like some of the hives are figuring it out. Some you'll get to and they look great. Then there's something like that!" he said, peering into a hive where he found no bees. "It's horrible."
Sundberg suspects that mites killed the bees, even though the hives were treated with an insecticide to kill mites.
Although Sundberg prefers to use oil of thyme or formic acid, which he describes as soft treatments, he does occasionally use an insecticide treatment when mites threaten to overwhelm a hive.
His next step is to see how widespread this infestation is. From a nearby, healthy-looking colony, Sundberg shook about 300 bees into a quart jar and sprayed ether into the jar to kill the bees and mites.
He hates doing this test, but when he dumped out the dead bees, tiny brownish dots peppered the sides of the glass jar. Sundberg counted more than 30 dead mites, twice as many as had hoped to see.
"I would conclude the treatments aren't working very well," he said. "So this hive looks good now. If they die this winter I'll know it's probably because of the mites."
Sundberg said it's frustrating to see a dead hive next to an apparently healthy one. But he said that's part of the natural selection process he hopes will make bees healthier.
"Quite frankly, some of the colonies need to die, because they are not able to survive," he said. "And the ones that are able to survive, we need to reproduce from them."
Sundberg said by selecting for traits like the hygienic behavior, he's helping bees help themselves. But it's a painfully slow process
Selectively breeding bees is much more difficult than breeding better cows or hogs. Queen bees can fly miles to mate with 15 to 20 drone bees.
It takes at least four generations to develop the desirable trait, Sundberg said. Artificial insemination isn't practical on a commercial scale because it is a complex, difficult process.
As hives replace queen bees every couple of years, the selection process must be ongoing.
University of Minnesota bee expert Marla Spivak, who started the Minnesota Hygienic bee line, said natural selection offers the best solution because chemical treatments no longer are effective against bee diseases and parasites.
That's why she expanded the program to help individual beekeepers develop their own hygienic bees, an approach that creates more genetic diversity among bees.
Beekeepers like Sundberg need to keep using chemical treatments to keep bees alive while slowly improving bee genetics.
If nature were left to its own devices and beekeepers stopped using chemical treatments, Spivak said 80 to 90 percent of bees would die.
"Then after about ten years the remaining bees would develop resistance to these problems so that we would never have to treat again," she said. "But in the meantime we would not be able to pollinate fruits and vegetables. So the tradeoff is not realistic."
More beekeepers are now trying selective breeding, which Spivak called a hopeful development. She said that should have happened 20 years ago, but didn't.
"I think the reason is because it's too easy to buy a treatment," she said. "So they can purchase an antibiotic for a disease problem or they could put a pesticide strip in the colony to control the mites. So it's taken a while for beekeepers to say, 'OK, all of these treatments are not really sustainable so we're going to have to do something to help the bees help themselves.'"
U of M bee teams travel the country helping beekeepers select for the hygienic trait. The university also is starting a study to measure the effectiveness of the bee breeding effort. Two other lines of hygienic bees have been developed in Louisiana. But genetic development of bees is very limited, Spivak said.
Sundberg thinks improved bee genetics has reduced or eliminated some diseases in his hives — including American foulbrood and chalkbrood. But he worries that while he's developing healthier bees, they won't survive a looming pesticide threat.
Research is starting to show the popular neonicotinoid insecticide can kill bees or cause neurological damage.
"With the neonicotinoids we don't really fully know what the long term effects are yet," Sundberg said. "If everything is going to turn to corn and beans and the ground is poisoned, what am I doing here? Why am I bringing these bees here?"