ACTA NEUROBIOL. EXP. 1978, 38: 353-359

Short communication

SOME QUESTIONS RELATED TO MECHANISMS OF SLAVE-RAIDS IN AMAZON- RUFESCENS LATR.

Jan DOBRZAmSICI and Janina DOBRZAflSKA

Department of Neurophysiology, Nencki Institute of Experimental Biology Warsaw, Poland

Abstract. A new hypothesis concerning the mechanism of the P. rufescens' slave-raids, is proposed. (i) The so-called pseudo-scouting does not contribute to the slave-raids. (ii) The direction of the raid js determined by the random movement of the most easily aroused individuals - the so-called activists. (iii) Once the direction has been set the army marches straight ahead and returns by the same road orienting itself by visual stimuli; single individuals who lag behind can however follow the scent-trial which had been left by the army. (iv) The slave-capturing instinct of the amazons is not directed at the abduction of pupae alone but also and even perhaps primarily at the abduction of mature individuals belonging to the slave-; that the amazons bring almost only pupae is caused by the fact that the mature individuals affer resistance and are consequently a more troublesome prey.

The data taken from observations and studies carried out over a period of many years on the raiding-behavior of the slave-making ant P. rufescens, indicate a need for a revision of the opinions expressed in our earlier studies (6) and in the studies of other authors. It has become necessary since the publications contain opinions and generalizations which are almost always different and often contradict each other. This study is an attempt to confront and discuss certain aspects of those opinions on the basis of the published material and some new data which we obtained an accurate picture of the behavior of those social-parasitic could have important implications for many general considerations in biology.

5 - Acta Neurobiologiae Experimentalis Direction of movement of an army-column in amazons. To this day, many authors accept with outdoubt the opinion, expressed by Fore1 (8, vol. 4, p. 127) that there exist scout-ants and that P. rufescens' army is commanded by "officer" ants. Wilson in his latest monograph about social (14, p. 368) adopts a similar position by an a priori extrapolation onto the whole Polyergus , the results obtained by Talbot (12) who has discovered the existance of scouts in American Mayr - and does so without taking into consideration the data obtained by Dobrzanska. and Dobrzanski (6) and Koehler (9) (qithuzgh he quotes both articles). However, even if there exists a very close aifinity between two species in the same geographical zone, the possibility of great behavioral differences can not be excluded. It is possible to enumerate a great many examples, but it should suffice to mention only the very closely related species of F. fusca L. and F. cinerea Mayr. which have been the subjects of the previous work (4). On the other hand, already in Brun's paper (1, p. 137) it is possibls to find doubts concerning the validity of Forel's thesis, as for example when: "Oft (the authors' remark) ist die Armee von vornherein mit bemerkenswerter Genauigkeit auf ein bestimmtes Nest der Sklavenart orientiert". He assumes, therefore, that is not a general rule that P. rufescens march out without being previously given information about a specific goal for attack. Later on, Brun states clearly: "Die Vorhut der Armee ist nicht konstant, sondern fortwahrenden Wechsel unterworfen; jedenfalls wird sie nicht etwa durch ein einzelnes Individuum 'ange- fiihrt' ". Thus, our thesis that the so-called scouts (namdd by us pseudo-scouts) do not play a role in the finding of slave-species' nests and in the directing of amazons' armies towards them (6), was preceded by Brun's observations (1). We have introduced, however, a new opinion, that is, the mass of slave-making ants is stimulated for marching out by a group of the most excitable individuals (activists), and that the final directior of attack is random, with the activists moving in a circular, chaotic way. The fact that the direction of expeditions is a matter of chance has been confirmed by Koehler's studies (9). Koehler laid down pupae belonging to the Serviformica species in a straight line from the nest and in the direction different from the one which had already been marked out by the activists. In this way he cou-ld influence the direction of ihe expedition. He had obtained, with the use of very weak stimuli (he laid out only 25 pupae), an effect similar to the one which we have obtained by passing over the activists from one place to another (6). The influence, that even such weak stimuli can have on the direction of the attack, constitutes an indirect proof that a choice of direction is not determined by the previously collected information about a specific goal but is induced by quite accidental stimuli. Very clear evidence for this was provided by Koehler's comparison between the number of pupae captured in naturally occurring attacks with the gains of attacks artificially steared in a direction chosen at random. The numbers turned out to be similar which definitely indicates that prey-finding in nature is also a matter of chance. The probability of finding prey, by an army directed towards nests previously discovered by scouts, would be greatly increased. It. should therefore be accepted that our conclusions (6) which were hased on a rather indirect evidence, have been confirmed by Koehler's studies (9). A few other observations of ours could be added in this connection. At a distance of 4 m away from the studied P, rufescens' nest there was a big road belonging to F. rufa. From time to time, the slave-making ants went on an raid across that road which resulted every time in an attack by F. rufa ants who defend their roads with an almost equal ferocity to that with which they defend their nests. Although those ants cannot match the amazons in a duel, and thus many more of them were killed, their practically unlimited quantitative superiority resulted in the slave-making ants being decimated each time. The continual occurrence of such fight led by the end of the summer, to such a decrease in the amazons' numerical force that the colony ceased to make raids and, in the following year, ceased to exist at all. It is abolutely clear, that the direction across the F. rufa path could not have been indicated to the amazons by a scout, because it could not have had a chance to go and return across the enemy's busy road after having made a reconnaissance trip. More important, they would have lacked the incentive to proceed in the very direction where they would encounter such a powerful and dangerous obstacle. It is beyond doubt that the direction of those expeditions was not predetermined but was chosen by an army unaquainted with the potential danger. This evidences the completely spontaneous character of the proces by which an army of the amazons moves in a randomly chosen direction. Also there do exist normally functioning P. rufescens' nests which are completely or almost completely deprived of pseudo-scouts, i.e., of individuals who move around a territory giving the impression of making a reconnaissance which has given a rise to the theory about scouts. The existance of nests without pseudo-scouts is an additional argument in support of our hypothesis according to which they are an atavistic group of individuals who no longer play a part in the life of a community. Their movements around a territory superficially resemble foraging behavior. Although foraging has completely fallen out of the P. rufescens' model of behavior, the pseudo-scouts not only do not bring anything but also do not even show sings of trying to find prey. Raignier and Boven (11) have found a similar mechanism of the direction choosing in the hunting raids of African Dorylinae ants. However, the African Dorylus have nothing taxonomically in common with Polyergus, moreover, its extremely predatory nomadic way of life is quite opposite to the slave-making behavior of Polyergus. Thus, similarity in the direction findings stems solely from the similar structure of raids. On the other hand, slave-making ants, Polyergus lucidus, which are close relative to differ considerably as to the army behavior, since, according to Talbot (12) they have scouts responsible for direction of the expeditions. Directional orientation. It is worthwhile to add a few words concerning the problem of orientation in the slave-making ants. Our thesis (6) that the amazons' army does not follow a trail, but having once chosen a direction moves straight ahead and returns by the same route by visually finding the direction, appears to be contradicted by some observations. For instance, it has been observed that a singular ant runs out of a nest at a time when a column has already moved a long distance away, beyond its range of vision. Usually, such stragglers do not show any hesitation and run in the direction taken by the expedition but only begin to lose their way when the scent-trail left by the column has erased. The erasure of their own trail does not, however, make any impression on the returning army: burdened with prey, the slave-making ants make their way back to'the nest along their own road, even if in their absence the road had been dug over (6). The thesis about the visual orientation of an army should be thus supple- mented by saying that single amazons who have not been directed during the general incitement can use the scent-trails left by the F, rufescens' column. A new look at the slave-capturing instinct. It is generally thought that the slave-capturing instinct is ~reciselydirected toward the robbing of pupae only, and that it is for the lack of pupae that the amazons take larvae and even as a last resort, freshly hatched ants. In contradistinction to this view, might be our observations performed on P. rufescens with slave-ants F. cinerea. F. cinerea Mayr., one of the more frequent slaves, is a difinitely polycalic species. When one of the nests is entered by a P, rufescens female who sets up there a family on the basis of socially-parasitic relations, the nest remains a part of the colony and, through its slaves, maintains indirectly social links with the other nests inhabited by normal F. cinerea families. The existence of such social links is confirmed also by Czechowski's report (2). It follows, from his and our observations, that the slave-making ants - in their raids - do not attack nests belonging to the same colony. This is perfectly understandable because, as the result, of constant interactions, all ants inhabiting a colony treat each other as equal nestmates, together with th slave-making ants living in that colony. Pisarski's (10) and Dobrzaliska's (3) studies on F. exsecta Nyl. indicate that there exist very close ties between the nests of a polycalic colony but that those ties are not uniform, with not all the nests necessarily communicating with each other and exchanging workers. However, fcr as long as there exists an indirect social link between them, it prevents alienation from the colony. We have every reason to assume that similar relations occur in polycalic colonies of F. cinerea (4). The P. rufescens are, however, in contrast to F. cinerea - monocalic ants. Treating as their nestmates those inhabitants of the colony who constantly come into contact with them, they can lose the persona! scent-link (weaker in them than in polycalic ants) with a nest that they do not have a direct contact with. This kind of relative isolation can undoubtedly explain the mistaken raid of amazons on a nest belonging to their own colony. A slave-making ants' army, by their normal procedure, forced their way into a nest positioned 1.5 m away from their own, and carried out only several pupae and many mature workers who were marked with dyes when they were being cerried by the amazons. The carried F. cinerea did not of course offer any resistance because they were caught by their nestmates from the same colony. They folded up in a specific position for carrying and passively let themselves be transported to the nest inhabited by the slave-making ants. Most of the F. cinerea who had been brought as trophy very soon .went calmly back to their own nest although not all the individuals marked during the transportation were noticed on the return trip. It would be quite normal if some of them were too passive to go self-dependently out from the nest and some of them stayed behind having found actual stimuli for taking up some activity in the new nest, similarly to what takes place in F,exsecta colonies (3). This event throws some new light onto the issue of the slave- capturing instinct in the slave-making ants. It should be remembered that the amazons during their attacks on strange nests do not display active aggression or a hostile attitude to the inhabitants, fighting and killing only when attacked themselves. It is also characteristic that the easy to carry larvae are brought relatively infrequently and only when pupae cannot be found; but it does happen that the amazons bring young white ants who struggle and run away, are troublesome in every respect and at the same time do not evoke in the carrying soldiers them, any expressions of aggression (5). A compfetely different behavior is displayed by F. sanguinea Latr. the another slave-making species. Being, unlike the P. rufescens, a facultative parasite, they treat their prey also, or even primarily, as a source of nourishment (7, 13). This is reflected in the F. sanguinea's behavior during raids: they actively attack the inhabitants of raided nests, killing them and bringing them back as quarry, together with the pupae into their own nest. The comparison between the two very different behaviors suggests that the bringing back of pupae into the nest without aggressive or food-consummatory reflexes must be connected in the P. rufescens with the loss of the negative reaction to strangers. Every worker belonging to the slave-species could constitute a potential nestmate to the amazons with their slave-capturing instinct being reduced to a tendency for bringing every one of them into their own nest. It could even be infered, from analyzing the known facts, that the amazons are predisposed towards catching the mature individuals but the mature individuals run away or fight and the very young ~truggleand escape thus significantly slowing down the speed of action which is the necessary condition for the success of a raid. The P. rufescens' slave-capturing instinct which demands thousands of captures is most easily satisfied by pupae. It is to be expected that described attitude to the individuals in the form of imago is operating only in relation to a given species of slaves which actually inhabits the slave-making ants' nest.

This investigation was supported by Project 10.4.1.01 of the Polish Academy of Sciences.

1. BRUN, R. 1924. Das Leben der Ameisen. B. G. .Teubner, Leipzig, Teubners Naturwis. Bibliothek 31: 211 p. 2, CZECHOWSKI, W. 1975. Mixed polycalic colony of (Serviformica) cinerea Mayr. and Polyergus rufescens Latr. Ann. Zool. 33: 67-76. 3..DOBRZA~~SKA, J. 1973. Ethological studies on polycalic colonies of the ants Formica exsecta Nyl. Acta Neurobiol. Exp. 33: 597-622. 4. DOBRZANSKA, J. 1978. Problem of behavioral plasticity in slave-making amazon-ant P. rufescens Latr. and its slaves-ants F. jusca L. and F. ci- nerea Mayr. Acta Neurobiol. Exp. 38: 113-132. 5. DOBRZAfiSKA, J. 1978: Evaluation of functional degeneration of the amazon- ant Polyergus rufescens Latr. under an influence of socially parasitic way of life. Acta Neurobiol. Exp. 38: 133-138. 6. DOBRZA~SICA,J. and DOBRZANSKI, J. 1960. Quelques nouvelles remarques sur 1'Cthologie de Polyeryus rufescens Latr. Ins. Soc. 7: 1-8. 7. DOBRZA~SKI,J. 1961. Sur l'hthologie guerriere de Latr. Acta Biol. Exp. 21: 53-73. 8. FOREL, A. 1921-1923. Le monde social des fourmis du globe. 5 vols. Libraire Kundig, Geneve, 948 p. 9. KOEHLER, F. 1966. Untersuchungen zur Orientierung der raubzeuge der Amazonenameise Polyergus rufescens Latr. Inst. Soc. 13: 305-309. 10. PISARSKI, B. 1972. La structure des colonies polycaliques de Formica (Copto- formica) exsecta Nyl. Ekol. Pol. 20: 111-116. 11. RAIGNIER, A. and BOVEN, J. K. A. 1955. Etude taxonomique et biometrique des Dorylus du Sous-Genre Anomma (, Formicidae). Annls. Mus. R. Congo Belge, Sci. Zool. 2: 359 p. 12. TALBOT, M. 1967. Slave-raids of the ant Polyergus lucidus Mayr. Psyche (Camb Mass.) 74: 199-313. i3. WHEELER, W. M. 1905. An interpretation of 'the slave-making instincts in ants. Bull. Am. Mus. Nat. Hist. 21: 1-16. 14. WILSON, E. 0. 1972. T-he societies. Belknap Press of Harvard Univ. Press, Cambridge, 548 p.

Accepted 26 June 1978

J. DOBRZANSKIand J. DOBRZARSKA, Nencki Institute of Experimental Biology, Pasteura 3, 02-093 Warsaw. Poland.