These ants become slaves when workers from the slavemaking ant colony attack the nests of the host species Temnothorax longispinosus, kill the adult ants and steal the brood. Back in the masters' nest, often located in hollow acorns, nutshells, or twigs, the brood care behaviour of the emerging slave workers is exploited to the advantage of the slavemaker species. As Susanne Foitzik and her work group have shown, the enslaved worker ants feed and clean the larvae, raising the offspring of their social parasite – but only to a certain point.
Enslaved host ants | kill a slavemaker pupa. The host workers belong to the species Temnothorax longispinosus
"Probably at first, the slaves cannot tell that the larvae belong to another species", explains Foitzik. As a result, 95 percent of the brood survive the larval stage. But the situation changes as soon as the larvae pupate. "The pupae, already looking like ants, bear typical cues on their cuticles concerning their species. We have been able to show that a high fraction of the slavemaker pupae are killed by slave workers." The pupae are either neglected or actively killed by being attacked and torn apart. Several slaves at once may assault a pupa, which is unable to move or defend itself during the pupal stage, and is not protected by a cocoon.
In parasite nests in West Virginia, only 27 percent of the pupae survived, and in the New York colonies, only 49 percent. In Ohio, the survival chances of the American slavemaking ant was a bit higher at 58 percent, but this figure is still well below the survival rate of 85 percent for host pupae in their own nests. "The enslaved workers do not directly benefit from the killings because they do not reproduce," explains Susanne Foitzik. But, through the killing of slavemaker offspring, their neighbouring relatives – which might very well be the sisters of the worker slaves – indirectly benefit, as their chances of survival increase. Slavemaker colonies damaged by slave rebellions grow slower, and smaller slavemaking colonies conduct fewer and less destructive slave raids.
The large differences in the death rates in colonies from different regions fits the predictions of the "geographic mosaic theory of co-evolution": This theory claims that populations differ because they are subjected to different local selection pressures, and because they possess different attack or defense traits due to mutations, which in turn means that evolution can go in different directions in different geographic areas. While the host ants in New York are very aggressive and often successfully thwart slave raid attempts, the hosts in West Virginia profit more from the slave rebellion behavior because, as genetic analyses have shown, the neighbouring colonies are more often close relatives to the rebelling slaves.
This study on the evolution of slave rebellion has been financed since October 2011 by the project "The evolution of resistance and virulence in structured populations", funded by the German Research Foundation. (© Johannes Gutenberg University Mainz)