The University of Arizona

Cactus-Specialist Bees Differ When Choosing Pollen-Doctored Flowers

By Lori Stiles, August 8, 2000

Very little is known about how most bees select their host plants. But it is important and relevant to baffling questions in evolutionary biology.

Why, for example, do the vast majority of bees and other herbivorous insects specialize in particular host plants? Of the roughly 20,000 bee species, most bees are specialists. Relatively few species, including the well known honeybee and bumblebee, are generalists that collect pollen and nectar from many different types of plants to maintain communal hives. Just like smart investors diversify portfolios to reduce financial risk for economic advantage, generalist, social bees would seem with a diversified diet to have a big evolutionary advantage.

So why do solitary specialist bees abound in nature? Maybe there are no evolutionary advantages to specialization, some biologists speculate. Possibly specialization results because insects have adapted over time to a single available food source and have even lost their ability to recognize other plants as food. In which case, should their plant disappear, so do the insects.

Margrit McIntosh, a doctoral candidate in ecology and evolutionary biology at the University of Arizona, is studying how several distantly related or unrelated species of cactus-bee specialists select their host plants. Her subject species include Diadasia rinconis, Diadasia opuntiae, Idiomelissodes duplocincta and Lithurge apicalis. Her novel approach has been in studying the bees as herbivores, rather than as pollinators, to discover how they specialize.

Through a series of experiments designed to "trick" the cactus bees into collecting non-cactus pollen, McIntosh has discovered that host-plant selection is a two-part process. Flowers are showy visual cues that attract a passing cactus bee in search of a host plant, but pollen determines whether the bee ultimately rejects or forages on the flower.

McIntosh is talking about the work this week at the 85th annual meeting of the Ecological Society of America in Snowbird, Utah. Her talk focuses on experiments with ground-nesting Diadasia bees, where she discovered striking individual differences.

The bees will dine on a wide range of plants when foraging for nectar, their flight fuel. They stick to cacti when collecting pollen to provision their nests. However, McIntosh discovered remarkable differences among individuals within both species of Diadasia bees when it came to pollens they accept. That flexibility is noteworthy about these bees, given that pollen choice for specialist bees determines whether their offspring survive or perish.

Each bee tunnels underground to create her nest, in which she lays her larvae, each larva in its own separate adobe-like cell. The bee lays each egg on a pollen "loaf," or pollen mixed with nectar, then seals the cell. When the larvae emerge, they have no choice on what they feed. If the female cactus bee has unwisely chosen a non-nutritious pollen, her offspring will starve.

POLLEN - DOCTORED FLOWERS

Realizing that the chances of getting the bees to land on non-cactus flowers are slim, McIntosh ingeniously switched pollens within the flowers for her experiments.

Once she established the location of nesting aggregations, she began the meticulous task of removing cactus pollen from the cactus flowers and replacing it with other pollens.

She offered the bees unchanged cactus flowers (the control), cactus flowers with novel pollen (pollen from non-cacti or non-host cacti plants), and cactus flowers in which the stamens were cut or "ripped" and left inside the flower.

Because it was necessary to rip out all the cactus stamens before placing non-cactus pollen into cactus flowers, McIntosh used the "ripped control" to check whether damage to the plant could itself affect bee foraging preferences.

She used a variety of pollens and stamens, both freshly collected and frozen. She incorporated pollen from three different plant families, Cactaceae (cacti) , Malvacaeae (mallows) and Convolvulaceae (morning glories). She chose flowers from prickly pear, barrel cactus, saguaro, globe-mallow, prickly poppy, mesquite, acacia, Arizona poppy, and sunflower.

McIntosh studied uncaged D. rinconis bees at the lot near her house and D. opuntiae bees in 6' x 6' x 6' flight cages at the Santa Rita Experiment Station 35 miles south of Tucson in spring and early summer of 1999.

D. rinconis accepted some non-cactus pollen, especially globe-mallow pollen.

But some of her most interesting 1999 results came from D. opuntiae bees. They accepted frozen cactus pollen, mallow and even saguaro.

"I only had five D. opuntiae individuals and I had a really wide range of host acceptance just within that group of individuals," McIntosh noted.

"One particular bee in the group collected only saguaro pollen while the other four did not. She would just visit the flower over and over again, so you can see that she actually preferred saguaro over the other control cactus flowers." (No saguaros grow in this area north of the Santa Rita Mountains.)

"Then there was an individual that did not accept any non-cactus pollens, but she did accept frozen barrel cactus pollen. Then I had another individual that would only accept fresh prickly pear pollen, no other cacti, no other frozen pollen."

SPRING 2000 RESULTS

McIntosh also discovered significant foraging differences in individual D. rinconis bees when cacti bloomed in Tucson during April and May 2000. In April, she quickly caged 7 newly-emerged D. rinconis bees in the vacant lot near her home for a 4-stage experiment.

First she offered them only globe-mallow flowers. All 7 bees took nectar from the flowers, but none initiated nesting. Next she offered them only cactus flowers. All 7 bees dug nests within 2 days of access to these flowers. Then McIntosh again offered them only globe-mallow flowers. Six of the 7 bees switched to globe-mallow and continued their nesting.

"This surprised me, because they had refused to initiate nesting with globe-mallow." One of the bees refused to switch to globe-mallow at this time.

Finally, McIntosh offered the bees one cactus flower, and a plentiful supply of globe-mallow. "Within a few hours, all 7 bees collected pollen only from the cactus flower, ignoring the globe-mallow."

"These bees were so different (in accepting pollen), yet they came from the same species, even the same nesting aggregation. They were in the same cage, under the same conditions.

"When there is a range of behaviors like that within a population, it means that, depending on the circumstances, bees that stick with and become even more specialized to cacti will do well if other plants disappear. Or, if the cacti failed for some reason, then the strict cacti specialists would go extinct, but the more flexible bees will survive," McIntosh said, "The range of behaviors within the population suggests that although these bees specialize on cacti, they are not on some evolutionary dead end."

That some bees did accept non-cactus pollen offered in doctored flowers is further evidence that, contrary to popular thinking, insects are not little "automatons" ruled only by instinct, lacking the ability to learn, she added.

Her research approach to cactus bees as herbivores rather than as pollinators puts a different spin on the insect-plant host relationship. Insect-plant host relationships are usually considered "mutualistic" (mutually beneficial) or "antagonistic" (one organism benefits at the expense of the other). There is a potential conflict of interest between the bee and the plant. The plant depends on the bee to spread pollen among flowers, but the bee is collecting pollen to provision her nest. "Is this a mutualistic or an antagonistic relationship, or a combination of both?" McIntosh asks.

A further mystery: Why is it apparently so easy for the bees to switch from cacti to mallow pollen, plants only distantly related?. Malvacae are thought to be "ancestral" plant hosts for the genus Diadasia, McIntosh said. Is there just some similarity between cactus and mallow pollen? Or do the cactus bees "remember" mallow flowers from their evolutionary past?