May 2004, Issue 2

Social Wasps and Bees

Philip T. B. Starks is an assistant professor of behavioral ecology in the Biology Department. After receiving his PhD in neurobiology and behavior from Cornell in 1999, Starks held a three-year postdoctoral fellowship at the Miller Institute for Basic Research in Science at UC Berkeley, where he was hosted by Wayne Getz. Starks now works in the Dana Laboratories on the Medford Campus, and his group includes graduate students as well as postdoctoral research associates. Their primary interests are the evolution of eusociality, invasion biology, recognition systems, and host-parasite interactions.

Eusocial animals such as bees, wasps, and ants live in multigenerational groups, provide cooperative care to their young, and divide the work of reproduction. Starks tests models that explain reproductive skew, or the degree of reproductive monopolization by the dominant female, in paper wasps. "Although the alpha female dominates reproduction, she allows some reproduction by subordinate females," he says. To understand how this behavior can be evolutionarily stable for the subordinates, Starks' group estimates genetic relatedness between cooperating wasps by microsatellite analysis. Models incorporating ecological constraints, genetic relationships, and enhanced group productivity have yielded some counter-intuitive predictions, one being that sisters of the dominant female would be allowed to reproduce less than would subordinates whose relation to the alpha female is more distant.

Invasion biology is the study of non-native species and how they spread into a new environment. The paper wasp Polistes dominulus is a "European invader." First seen in Cambridge, Massachusetts, in 1978, the paper wasp has spread across the US. "People have generated different hypotheses for invasion success in social insects. Some suggest that an extreme genetic bottleneck is necessary and others suggest that this bottleneck is not necessary but rather that you need to have some release from ecological restraints," says Starks. When a group of individuals is separated from the main population, genetic diversity is reduced greatly: this makes for the genetic bottleneck. "Preliminary results suggest that there was no genetic bottleneck for the P. dominulus wasp, which indicates that there were probably multiple, independent introductions of this animal." Often an invasive population will also take on a new social behavior, one not displayed in the species' native area. One of Starks' postdoctoral research associates is hoping to spend six months in Italy teasing out behavioral differences between the native P. dominulus and its North American cousin.

Wasps use cuticular hydrocarbons to distinguish nestmates from non-nestmates. The lab has approached this problem in recognition systems by analyzing hydrocarbons with gas chromatography. "Depending on the retention times, we can identify different compounds that indicate which nest individuals come from," Starks says. Although wasps may indeed be able to recognize degrees of relatedness within a colony, "we can't see it if my wasps recognize but don't discriminate [show that they recognize]. So it's been very hard for us to find intracolonial kin discrimination."

Enter the hermit crab. Because hermit crabs are solitary animals, they don't offer much in the way of intracolonial kin discrimination. But they do answer other challenges in recognition systems research thanks to their affinity for the empty shells of gastropod mollusks. Starks has been looking at hermit crab shell selection in collaboration with Peter Wong, PhD, and Brian Gravel, MS, both in the School of Engineering, and Jan Pechenik, PhD, in the Department of Biology. Wong and Gravel develop computer models of shells that are then fabricated from resin polymer. Says Starks, "we're poised to determine what characteristics shells have that hermit crabs really select on. You've got the production, perception, and action components [of recognition systems], and it can be extremely difficult to tease them apart if you want to study them individually. With hermit crabs, they select inanimate shells. There is no response of the shell. So we can actually work out the preferred characteristics of shells, and with that determine the 'ideal' template used in shell selection behavior."

Starks is excited about possibilities for cross-disciplinary research. In addition to working with geneticists, he would welcome collaborations with chemists to work out the hydrocarbon signaling model, with engineers to devise better equipment for studying bees and wasps, and with sociologists and computer scientists to model multi-level behavior. He cites collaborations with other social biologists in France and the US as sources of great satisfaction and insight.

For more information, go to



Tufts University, Office of the Vice Provost
Health Sciences Campus: (617) 636-6550
Medford Campus: (617) 627-3417
Copyright 2005 Tufts University. All Rights Reserved.

Please send questions/comments about this site to Webmaster.