Morphological Adaptations of Polyergus rufescens that Aid in Slave-Making
David James
ABSTRACT
The European slave-making ant Polyergus rufescens performs numerous cloak and daggeresque behaviors during its life cycle. In no other caste of P. rufescens is this more apparent than in the queen. Unable to care for her own brood, the queen must infiltrate a Formicine nest, kill the resident queen, and fool the remaining colony of workers to do her bidding. The queen uses several strategies to carry out this dangerous mission. She follows fellow workers on a larval raid of a Formicine nest, in order to use this distraction to gain access to the host queen. Formicine workers who attack the parasitic queen are repelled by emissions from her Dufour’s gland. These emissions have been characterized as esters of butanoic acid and acetic acid. Decyl buanoate comprises 80% of the esters in the Dufour’s gland, and has been shown to have a repellent/appeasement effect on Formicine workers. The invading queen’s ability to defeat the host queen is augmented by an important morphological adaptation, sharp mandibles. With the old queen dead, the behavior of the host nest workers towards the usurping queen changes from aggression to grooming. This change in worker attitude toward the new queen is facilitated by a change in hydrocarbon profile of the new queen. P. rufescens workers are also skilled in the clandestine arts. Colonies of P. rufescens penetrate neighboring ant nests of the genus Formica and kidnap brood. They find their way to and from target nests by using a combination of chemical cues and polarized skylight. These stolen larvae are taken back to P. rufescens nest to be reared into a life of colony maintenance slavery. Morphological and behavioral adaptations of P. rufescens workers make them ideal for nest raiding but take away from their ability to perform domestic tasks. As is the case with their queen, kidnapping, slave-making and deception are all integral behaviors needed for survival.
Introduction
Slavery in ants was first discovered by Pierre Huber in 1810 (Huber, 1810), and this behavior has been a favorite topic of discussion among myrmecologists ever since. This form of social parasitism is subdivided into two main categories, facultative and obligatory. These categories reflect the dependence of the parasitic ant on its slave host. Facultative slave-makers, such as Formica sanguinea, will enslave other ants if possible, but they can utilize alternative strategies for survival (Holldobler and Wilson, 1990). Obligatory slave-makers must enslave other ants to survive (Holldobler and Wilson, 1990).
Charles Darwin was fascinated enough by ant slavery to include a possible evolutionary hypothesis for it in his On the Origin of Species (Holldobler and Wilson, 1990). Darwin suggested predation as the cause for slave-making (Holldobler and Wilson, 1990). In this scenario, pre-parasitic ants attacked other nests to obtain food (Holldobler and Wilson, 1990). Some larvae brought back as food were not eaten and allowed to develop into slaves (Holldobler and Wilson, 1990). A second hypothesis suggests territoriality was the driving force in ant slavery propagation (Holldobler and Wilson, 1990). Pre-parasitic ants stole larvae and pupae from nearby colonies as a means of defending their turf (Holldobler and Wilson, 1990). Some of these stolen larvae were not eaten and developed into slaves (Holldobler and Wilson, 1990). A final theory proposes ant slavery evolved from larval transporting behavior (Holldobler and Wilson, 1990). Larvae are often transported between the nests of single polydomous colonies (Holldobler and Wilson, 1990). If this behavior was extended to include other unfamiliar colonies, an early form of slavery would be created (Holldobler and Wilson, 1990).
Among the Formicidae, six ant genera have been reported to practice slave-making: Leptothorax, Strongylognathus, Harpagoxenus, Formica, Polyergus, and Rossomyrmex (Regnier and Wilson, 1971). Of these genera, Polyergus is recognized as, "the pinnacle (or nadir if you prefer) of the slave-holding way of life (Holldobler and Wilson, 1990)." This formicine ant genus contains five obligatory parasitic species which direct their slave-making raids against ants in the related genus Formica (Mori et al., 2000a). Workers and queens in the genus Polyergus are completely dependent on slaves as they are unable to feed themselves, care for their brood, or perform colony maintenance (Holldobler and Wilson, 1990). This parasitic relationship in which workers of a parasitic ant species capture brood from other colonies and rear them as enslaved nestmates is referred to as dulosis.
Polyergus rufescens is commonly known as the European Amazon ant (Mori et al., 2000a) These ants enslave Serviformica workers that belong to the Formica fusca group, such as Formica cunicularia and Formica rufibarbis (Mori et al., 2000a) P. rufescens ants infiltrate, deceive, and kidnap as a way of survival. Numerous morphological traits can be viewed as aiding or propagating P. rufescens ants usurping, clandestine lifestyle.
Discussion
Mandibles of P. rufescens Workers. The main mission of the P. rufescens worker is to raid neighboring ant colonies (usually Formica cunicularia), steal larvae, and return them to their nest (Mori et al., 2000b). This function of the P. rufescens worker is reflected in the form of its mandibles. An infiltrator storming a stronghold would be wise to carry along a formidable weapon. This weapon is articulated by Wheeler (Wheeler, 1910):
The worker is extremely pugnacious, and, like the female, may be readily distinguished from the other Camponotine ants by its sickle-shaped, toothless, but very minutely denticulate mandibles. Such mandibles are not adapted for digging in the earth or for handling thin-skinned larvae or pupae and moving them about in the narrow chambers of the nest, but are admirably fitted for piercing the armor of adult ants.
These sword-like mandibles are quite common in ant genera that practice slave-making (Holldobler and Wilson, 1990). Facultative slavemakers, such as Formica sanguinea, still possess dentition on the mandibles (Figure 1), and these ants are quite capable of carrying out nest maintenance chores (Holldobler and Wilson, 1990). Without this dentition, P. rufescens workers are rendered useless in accomplishing any such domestic tasks. The contrasting behaviors of P. rufescens workers, augmented by their highly specialized mandibles, while at home and when raiding is described here again by Wheeler (1910):
While in the home nest they sit about in stolid idleness or pass the long hours begging the slaves for food or cleaning themselves and burnishing their ruddy armor, but when outside the nest on one of their predatory expeditions they display a dazzling courage and capacity for concerted action compared with which the raids of sanguinea resemble the clumsy efforts of a lot of untrained militia. The amazons may, therefore, be said to represent a more specialized and perfected stage of dulosis (parasitism) than that of the sanguinary ants. In attaining to this stage, however, they have become irrevocably dependent and parasitic.
Thus, the degree of reliance upon slave-making (facultative: F. sanguinea, obligatory: P. rufescens) is reflected in the grade of dentition in the mandibles.
The Role of Dufour’s Gland in P. rufescens Workers. A second adaptation to aid slave-making ants was discovered by Regnier and Wilson (1971) in the ant Formica subintegra. These researchers discovered that the Dufour’s gland of F. subintegra was extremely large in comparison to that of its host, Formica subsericea (Regnier and Wilson, 1971) (Figure 2). The Dufour’s gland is common to all social Hymenoptera (Billen, 1987). This sac-like structure is associated with the sting of workers and queens (Figure 2). In F. subintegra, Regnier and Wilson (1971) described the three major substances in this gland to be decyl acetate, dodecyl acetate and tetradecyl acetate. This hypertrophied gland was found to contain 700ug of these substances, which is equal to 10% of F. subintegra body weight (Regnier and Wilson, 1971). These three acetates were found in large quantities on host workers (F. subsericea) immediately following a raid by F. subintegra (Regnier and Wilson, 1971). It was put forth by Regnier and Wilson (1971) that these compounds account for the disorientation and panicked retreats of F. subsericea observed frequently in slave-making raids by F. subintegra. Recently, similar substances have been reported in the Dufour’s gland of P. rufescens workers (D’Ettorre et al., 2000). The Dufour’s gland secretions contained 15 identifiable compounds with octadecyl butanoate as the major compound (D’Ettorre et al., 2000). It is believed that these acetates are sprayed in such a high quantity onto the host workers that it causes an alarm reaction (Regnier and Wilson, 1971). This reaction was first described by the discoverer of slave-making ants, Pierre Huber (1810):
One of the principal features of the wars levied on the Ash-colored ants (host) seems to consist of exciting fear, and this effect is so strong that they never return to their besieged nest, even when the oppressors (F. sanguinea) have retired to their own nest; perhaps they realize that they could never remain in safety, being continually liable to new attacks by their unwelcome visitors.
Not wanting to face a unified response from the host defenders, these "unwelcome visitors" utilize their Dufour’s gland secretions to create a chemical diversion.
Plasticity of the Cuticular Hydrocarbons of P. rufescens Workers. Another feature of P. rufescens workers that aids in slave-making is their cuticular hydrocarbon disguise. The type and amount of hydrocarbons present in an insect’s cuticle create a species-specific signature (Cougoudan et al., 1997). This chemical fingerprint communicates to other ants (and researchers) information concerning the ants species, colony, and social caste (Cougoudan et al., 1997). One problem P. rufescens might have with their slave captives is rebellion. In order to lessen the likelihood of such an event, P. rufescens modify their cuticular hydrocarbons to more closely resemble their slave species (Cougoudan et al., 1997). To test this hypothesis, two slave species of ants, Formica rufibarbis and F. cunicularia, were swapped in a P. rufescens nest (Cougoudan et al., 1997). Hydrocarbons were extracted from P. rufescens workers before and after the swap (Cougoudan et al., 1997). Analysis of these hydrocarbons revealed an adjustment in proportions of certain hydrocarbons by P. rufescens towards the newly introduced slave species (Cougoudan et al., 1997).
Orientation of P. rufescens During Raids. One last aspect of P. rufescens worker behavior that should not be overlooked is their orienteering during slave raids. An early account of a P. rufescens slave raid is given by Wheeler (1910):
The ants leave the nest very suddenly and assemble about the entrance if they are not, as sometimes happens, pulled back and restrained by their slaves. Then they move out in a compact column with feverish haste, sometimes, according to Forel, at the rate of a meter in 33 seconds or 3 cm. per second. On reaching the nest to be pillaged, they do not hesitate like sanguinea but pour into it at once in a body, seize the brood, rush out again and make for home.
From numerous observations of P. rufescens nests, Talbot (1967) discovered that this quickly assembled attack force was formed in response to the return of an ant scout. Grasso et al. (1997) decided to test whether or not the invading column of ants use a chemical trail left by the ant scout to orient to the target nest. To test this hypothesis, large sections of earth were removed in front of the advancing P. rufescens troops (Grasso et al., 1997). This did not have any effect on the raiding ants ability to find the target nest (Grasso et al., 1997). Grasso et al. (1997) also artificially altered the position of the sun and filtered the effects of UV rays on advancing columns of P. rufescens raiders. From these various experiments Grasso et al. (1997) concluded that ant scouts lead the invading army to the host nest using a combination of cues. During the outward journey to the nest, the sun’s position and polarized-light patterns lead the invaders to the host colony (Grasso et al., 1997). Upon returning home, P. rufescens workers rely more on celestial parameters and chemical trails (Grasso et al., 1997).
P. rufescens Queens. As is the case with P. rufescens workers, the unique anatomy of a queen’s mandibles makes slave-making a necessary evil. Her inability to care for even her first brood allows her only one plan to achieve reproductive success (Mori et al., 2000b). This plan involves locating a host nest, infiltrating it, and usurping the queen found within (Mori et al., 2000b). A P. rufescens reproductive hoping to find a host colony to usurp will often follow her fellow workers on a slave raid (Mori et al., 2000a). This not only provides an easy means to locate a host colony, but it also provides a enormous distraction to aid in her gaining access to the colony (Mori et al., 2000a). Once the invasion has begun, the queen must rely on numerous morphological traits to aid in her mission, a single insect overthrow of an ant colony.
Cuticular Hydrocarbons in P. rufescens Queens. P. rufescens queens can modify their cuticular hydrocarbons in order to render themselves "invisible" or mimic their hosts. Hydrocarbon analysis of newly mated P. rufescens queens before usurpation revealed little to no hydrocarbons (D’Ettorre et a., 2000). It is believed that by possessing no hydrocarbons the queen is less likely to draw attention from the host workers (D’Ettorre et al., 2000). While wandering the corridors of the host’s nest, P. rufescens queens attempt to hide themselves in a camouflage of chemical insignificance (Lenoir, 1997). Upon finding the resident queen, the lack of hydrocarbons on the P. rufescens queen makes it easier for her to mimic her host (D’Ettorre et al., 2000). The P. rufescens queen acquires the hydrocarbon profile of the resident queen through physical contact when they duel to the death (D’Ettorre et al., 2000). In only five days following the usurpation of the resident queen, the hydrocarbon profile of the P. rufescens queen already resembles that of her vanquished foe (D’Ettorre and Errard, 1998). The lack of a hydrocarbon signature is not always enough to allow the parasitic queen to accomplish her mission. In fact, the P. rufescens queen is sometimes attacked and killed before replacing the resident queen (D’Ettorre et al., 2000). Luckily, she is outfitted with another morphological adaptation to defend herself from those who are not fooled by her trickery.
Dufour’s Gland in P. rufescens Queens. While the Dufour’s gland of P. rufescens workers is considered large compared to other ant species, it is even larger in the P. rufescens queen (D’Ettorre et al., 2000). As is the case with P. rufescens workers, the Dufour’s gland secretions of the P. rufescens queen have also been chemically analyzed. The queen secretion contains esters of butanoic acid and acetic acid, 80% of which is decyl butanoate (D’Ettorre et al., 2000). There are two different theories as to the induced response of this substance on host workers. One such theory reported by Topoff et al. (1998) and Mori et al. (2000b) is that decyl butyrate serves as an appeasement allomone. After killing the resident queen, the attacks on the parasitic P. rufescens queen by host workers become both less frequent and less violent (Mori et al., 2000b). During this phase of the usurpation, the P. rufescens queen vibrates her gaster to release more secretions from her Dufour’s gland (Mori et al., 2000b). This secretion changes the host worker’s behavior from that of aggression to grooming (Mori et al., 2000b). A similar strategy has been reported in Polyergus breviceps queens (Topoff et al., 1988,). Bioassays performed by Mori et al. (2000b) consisted of introducing an alien ant species Camponutus ligniperda coated with decyl butyrate into a colony of F. cunicularia. Results from these experiments revealed that aggressiveness by the F. cunicularia workers was drastically decreased towards the coated C. ligniperda when compared to controls (Mori et al., 2000b). Coated C. ligniperda workers were observed freely moving about the host colony, and they were even allowed to remain near the brood (Mori et al., 2000b). This appeasement allomone affords the parasitic queen enough time to disguise herself by acquiring the residents queens hydrocarbon profile (Mori et al., 2000b).
Another possible effect of the P. rufescens queen secretions is that they act as a repellent (D’ Ettorre et al., 2000). The fundamental problem with the appeasement allomone hypothesis is presented here by D’Ettorre et al. (2000):
The idea of an appeasement pheromone raises an evolutionary problem. Why would the host respond in a way that reduces its fitness, unless it too uses the same compounds but for other purposes, e.g. its own communication system? In preliminary analyses of the host species we could not detect the major product present in Dufour’s gland of P. rufescens queens as predicted from the above hypothesis, and therefore tested an alternative hypothesis that the secretion acts as a repellent for host (or any other ant species) workers.
This repellency of decyl butanoate was confirmed through a series of feeding bioassays. Droplets of honey were treated with decyl butanoate, octadecyl butanoate (major worker compound), limonene (an ant repellent), and a solvent control (D’Ettorre et al., 2000). Only the honey treated with decyl butanoate exhibited repellency by F. cunicularia workers (D’Ettorre et al., 2000). D’Ettorre et al. (2000) suggests that decyl butanoate buys time for the P. rufescens queen to prepare her disguise by repelling aggressive host defenders. Whether the P. rufescens queen is repelling or appeasing her defenders, the outcome of a successful invasion is consistent. By utilizing morphological and chemical adaptations, the P. rufescens queen enslaves an entire nest of ants to care for her and her young.
The unique behavior and morphological adaptations of P. rufescens ants have made this uncommon species one of the most frequently studied ants of all time (Holldobler and Wilson, 1990). Their cloak and dagger behavior aided by morphological and chemical weapons should evoke envy in spy agencies worldwide. As is the case with most research, the more an organism is studied the more questions arise. In regards to P. rufescens, if decyl butanoate alone can repel host workers, why does the P. rufescens queen’s Dufour’s gland contain a mixture of compounds (D’Ettorre et al., 2000)? What evolutionary path was followed to facilitate slave-making in P. rufescens? And finally, does decyl butanoate serve as an appeasement allomone or a repellent to host defenders?
REFERENCES
Billen J., Naturwissenschaften, 74, 340 (1987).
Cougourdan A. B., Bagneres A. G., Provost E., Dusticier G., and J. Clement, Comp. Biochem. Physiol., 116b, 287 (1997).
D’Ettorre P., and C. Errard, Ins. Soc. Life, 2, 71 (1998).
D’Ettorre P., Errard C., Ibarra F., Francke W., and A. Hefetz, Chemoecology, 10, 135 (2000).
Grasso D. A., Ugolini A., Visicchio R., and F. Le Moli, Animal Behavior. 54, 1425 (1997).
Holldobler B., and E. O. Wilson, The Ants (Harvard University Press, Cambridge, MA, 1990) pp. 450-465.
Huber P., Recherches sur les moeurs des fourmis indigenes (Paschoud, Paris, 1810).
Kutter H., Vierteljahrsschrift der Naturforschenden, 113, 1 (1969).
Lenoir A., Fresneau D., Errard C., and A. Hefetz, Information Processing in Social Insects (Birkhauser Verlag, Basel, 1999) pp. 219-237
Mori A., Grasso D. A., Visicchio R., and F. Le Moli, Insectes Sociaux, 47, 7 (2000a).
Mori A. et al., Ethology, Ecology and Evolution, 12, 315 (2000b).
Regnier F. E. and E. O. Wilson, Science 172, 267 (1971).
Talbot M., Psyche, 74, 299 (1967).
Topoff H., Cover S., Greenberg L., Goodloe L., and P. Sherman, Ethology, 79, 209 (1988).
Wheeler W. M., Ants: their structure, development and behavior (Columbia University Press, New York, NY, 1910) pp. 472-473.