Kinship and conflicts over male production in Formica ants

(1)

Department of biological and environmental sciences University of Helsinki

Finland

Academic dissertation

To be presented, with permission of the Faculty of Biosciences of the University of Helsinki, for public criticism in the Auditorium 1041 of Biocenter 2, Viikinkaari 5,

on November 12th, 2004 at 12 o’clock noon.

Helsinki 2004

(2)

©

Heikki Helanterä (chapters 0, II, IV, V)

©

Blackwell Publishing (chapters I, III, VI)

Cover illustration by Voitto Vuorio 2004.

Technical editing by Potra Girl Design.

Author’s address:

Department of biological and environmental sciences P.O. Box 65 (Viikinkaari 3)

FI-00014 University of Helsinki Finland

e-mail heikki.helantera@helsinki.fi

ISBN 952-91-7898-0 (paperback) ISBN 952-10-2162-4 (PDF) http://ethesis.helsinki.fi Edita Prima Oy Helsinki 2004

(3)

Kinship and conﬂ icts over male production in Formica ants

HEIKKI HELANTERÄ

This thesis is based on the following articles, which are referred to in the text by their Roman numerals:

I Hannonen, M., Helanterä, H. & Sundström, L. 2004: Habitat age, breeding system and kinship in the ant Formica fusca — Molecular Ecology 13: 1579-1588.

II Helanterä, H., Jaakkola, M., Trontti, K. & Sundström, L.: Hierarchical analysis of population structures in the ant Formica fusca — Manuscript.

III Helanterä, H. & Sundström, L.: Worker reproduction in the ant Formica fusca — Journal of Evolutionary Biology, in press.

IV Helanterä, H. & Sundström, L.: Worker policing and nestmate recognition in the ant Formica fusca – Manuscript.

V Helanterä, H. & Sundström, L.: Male production in Formica ants, testing the hypotheses

— Manuscript.

VI Wenseleers, T., Helanterä, H., Hart, A. & Ratnieks, F. 2004: Worker reproduction and policing in social insect colonies. An ESS analysis. — Journal of Evolutionary Biology 17:

1035-48.

(4)

Contributions

Supervised by:

Prof. Liselotte Sundström, University of Helsinki, Finland

Reviewed by:

Dr. Andrew Bourke, Insitute of Zoology, Zoological Society of London, UK Dr. Janne Kotiaho, University of Jyväskylä, Finland

Examined by:

Dr. Stuart West, Edinburgh University, UK

I II III IV V VI

Original idea MH, HH, LS HH, MJ, LS LS HH HH, LS TW

Study design MH, HH, LS HH, MJ, LS LS, HH HH HH TW

Data collection MH, HH HH, MJ, HH HH HH -

KT, LS

Analyses HH, MH HH HH HH HH TW

Manuscript preparation HH, MH, LS HH, LS HH, LS HH, LS HH, LS TW, HH,

AH, FR

(5)

For example genes have evolved co-operation in chromosomes, cells co-operate in multicellular individuals and individuals co-operate in human and insect societies. Such collectives then represent new levels of organisation and consequently selection has to be considered also on this higher level (Lewontin 1970, Reeve & Keller 1999, Leigh 1999, Michod 2004)

Seemingly harmonious collectives are, however, also ground for confl icts. Individuals may be attracted to use the group resources to their selfi sh benefi t without contributing to the group output. While this increases the success of the selfi sh individual, it disrupts the group success. This is the so called “Tragedy of the commons” (Hardin 1968, Leigh 1999): if each individual maximises its own success within the group, the whole group carries a cost that may destroy the co-operative whole.

The same principle applies in some form to all

kinds of co-operation from human co-operation to endosymbionts of eukaryotic cells, and from insect societies to mutualisms between species (recent reviews in e.g. Dugatkin 2002, Hammerstein 2004, Sachs et al 2004). For example, within a genome, there is a selective advantage for any gene that increases its transmission in meiosis (Hamilton 1967, for a review see Pomiankowski 1999) at the cost of other genes, and in an unrepeated prisoner’s dilemma, there are always higher payoffs for the defecting strategy (Axelrod & Hamilton 1981).

However, it is obvious that co-operation is not always disrupted by selfi shness and that the coalitions we see today are the ones where the confl ict has been successfully restrained.

There are numerous possible mechanisms for confl ict suppression, for example direct (Trivers 1971, Axelrod & Hamilton 1981) or indirect reciprocity (Nowak & Sigmund 1998), by-product benefi ts (Dugatkin 1997), partner choice (Nöe &

Hammerstein 1994, 1995), pleiotropy (Foster et al in press), kinship (Hamilton 1964, 1972) and mutual policing (Ratnieks 1988, Frank 1995, 1996).

From these the last two, kin selection theory and its subsection, policing theory, are the most important in the context of social confl icts in insect societies (reviews in e.g. Bourke & Franks 1995, Crozier

& Pamilo 1996, Queller & Strassmann 1998, Sundström & Boomsma 2001). In my thesis, I have

Summary

HEIKKI HELANTERÄ

University of Helsinki, Department of Biological and Environmental Sciences, PO BOX 65, FI – 00014 University of Helsinki, Finland

(7)

8

investigated the effects of kinship and worker policing on the confl ict over male production in the ant genus Formica.

Kin selection

Even though Darwin mentioned some kin selection ideas in Origin of Species (“…selection may be applied to the family, as well as to the individual…”

Darwin 1859), evolution of co-operation was usually explained by arguments based on the good of the species or the population (e.g. Wynne- Edwards 1962, see Williams 1966) until the mid 1960’s. However, at the time this view met strong opposition because it was recognized that cooperative groups are vulnerable to free-riders, as explained above (Hamilton 1964, Williams 1966).

The fi rst explanation for stable co-operation that was compatible with the modern, individual-based view of evolution was kin selection theory. W. D.

Hamilton (1964) showed that co-operation can be stable if the benefi ts are shared among related individuals. This is because relatives carry the same genes disproportionately often compared to random individuals in the population. Thus, if rb –c

> 0, i.e. the benefi t b to the recipient(s) weighted by r, the coeffi cient of relatedness is larger than c, the costs carried by the co-operative actor, the altruistic trait will spread in the population. The higher r within the group is, the more genes individuals share in comparison to the random individuals in the population, and the weaker the Tragedy of the commons will be for the group.

The further developments in kin selection methodology (Hamilton 1970, Grafen 1985, Queller 1994, Taylor & Franks 1996, Franks 1998) are based on the covariance approach of Price (1970). The great advantage of these methods is that they partition the selection clearly into selection among individuals in the group on the one hand and among groups on the other. Thus, in terms of the Tragedy of the commons, a selfi sh individual gains in selection within the group, but the as a consequence the group thrives less well in competition among groups because resources are

spent on within group competition. Thus, in its present form kin selection theory is fully compatible with the modern version of group selection or multi-level selection (e.g. Wilson 1997). This is most simply put forward in Frank (1995): fi tness of each individual is represented by the product of two terms wij = (zij / zi) * (1 – zi), where zij is the competitiveness of the focal individual and zi is the mean competitiveness of individuals in the ith group. The fi rst term thus determines the proportion of group resources an individual gains by competition, and the second term determines the group output, which decreases in proportion to average competitiveness in the group.

Even though relatedness has gained a central role in empirical studies of kin selection, kin selection does not concern only relatedness (see e.g. Griffi n & West 2002 for a critical review).

The central role of relatedness is understandable from an empirical point of view, as coeffi cients of relatedness are easy to estimate, either from pedigrees or from genetic marker data (Queller &

Goodnight 1989, Queller et al 1993). However, as is evident from the original formulation, r is merely the conversion factor between costs to self and benefi ts to kin and is useless if benefi ts and costs are not carefully considered. Moreover, measuring relatedness is not always straightforward. That is because relatedness is a relative measure and it must be measured against the correct background population allele frequencies (Queller & Goodnight 1989, Queller 1994). The correct relatedness coeffi cient also depends on the scale of cooperation and competition such that if the relatives who benefi t from co-operation are also competing against each other to some extent, the relatedness of potential competitors has to be evaluated (Taylor 1992, Queller 1994, West et al 2002).

Moreover, a separation must be made with potential and actual confl ict (Ratnieks & Reeve 1992). Potential confl ict is the genetic basis of confl icts and is directly evident from asymmetries in relatedness coeffi cients. For example workers in a social insect colony with one singly mated queen are related to their sisters by 0.75 and to

(8)

their brothers by 0.25, while the queen is related to both by 0.5. Thus, there is a potential confl ict over the sex ratio so that the workers prefer a sex ratio of 3:1 (females to males) but the queen prefers an equal 1:1 ratio (Trivers & Hare 1976). However, the actual confl ict, the expression and outcome of the potential confl icts, depends on the relative power of the parties, and the constraints on their actions that arise from the biological details of each particular system (Ratnieks & Reeve 1992, Sundström & Boomsma 2001, Beekman et al 2003).

For example, if workers are unable to assess the sex of offspring and thus bias the sex ratio, their power to capitalize on the potential confl ict is limited (Nonacs & Carlin 1990, Nonacs 1993). The concept of power is essential to understanding the confl ict outcome and why potential confl icts do not necessarily translate into actual confl icts (Ratnieks

& Reeve 1992, Beekman et al 2003, Beekman &

Ratnieks 2003).

Furthermore, costs are not only important for selection within groups and the confl ict outcome.

Costs of confl ict are an essential part of Tragedy of the commons -arguments, as they determine how greatly the group suffers from selfi shness, compared to a completely cooperative group (Frank 1995, Foster 2004). The shape of the cost functions has extensive effects on the evolution of cooperation. For example, diminishing returns from either cooperative or selfi sh behaviour may restrain the selfi shness of individuals, and thus signifi cantly decrease the group level costs of confl ict (VI, Foster 2004).

Mutual policing

Policing theory is one of the recently most successful areas of kin selection logic and has gained a lot of support from empirical studies (e.g. Ratnieks 1988, Ratnieks & Visscher 1989, Frank 1995, Foster &

Ratnieks 2000, 2001, Martin et al 2002, Endler et al 2004). The concept of policing was originally developed for social insects (Ratnieks 1988) but the most general models are the ones by Frank (1995, 1996, 2003). All these models predict that the lower

the relatedness among individuals (i.e. workers in the social insect case) is, the more they are predicted to spend resources on policing selfi sh individuals.

Furthermore, the higher the costs of selfi shness are, the more easily policing is predicted to evolve even when relatedness is high (Frank 1995).

Such policing can occur on any level of a hierarchy. For example, in the context of within genome confl ict, an allele would benefi t from increasing its own transmission rate in meiosis, but if this happens at a cost to the rest of the genome, other genes are selected to prevent such selfi shness by policing (Leigh 1999, Pomiankowski 1999). As another example, chimpanzees Pan troglodytes have various mechanisms of enforcing co-operation and effi cient group functioning in their troops (de Waal 1996).

Policing behaviour can be divided into two subcategories, namely punishment and sabotage (Clutton-Brock & Parker 1995, Reeve & Keller 1997). The difference is that sabotage only prevents benefi ting from selfi shness, but does not prevent selfi shness in the fi rst place. For example, honeybee workers destroy the eggs laid by other workers, but this does not prevent workers from laying another egg. Punishment, however, also decreases the future chance of selfi shness. For example, in naked mole rats, reproductive females “shove”

lazy workers to make them work harder (Reeve 1992) and workers in honeybees and queenless ponerine ants aggress or even kill workers that try to challenge the dominant breeder, which results in decreased fi tness for the selfi sh individual (Visscher

& Dukas 1995, ponerine ants reviewed in Monnin

& Ratnieks 2001).

From the group point of view policing has dual effects. On the immediate time scale policing can be costly to the colony, but it prevents individuals from benefi ting from selfi shness, and may stop them from trying again (as in punishment). On evolutionary time scales policing may select for self policing, i.e. individuals refraining from selfi sh exploits, provided that selfi sh behaviour carries a cost to the selfi sh individual (Frank 1996, 2003).

In this case also colony level costs of confl ict

(9)

10

are alleviated, and the Tragedy of the commons decreases.

Social insects

Since Hamilton’s seminal work, social insects have been a main object of kin confl ict studies (reviews in e.g. Hamilton 1972, Fletcher & Ross 1985, Bourke

& Franks 1995, Crozier & Pamilo 1996, Queller &

Strassmann 1998, Sundström & Boomsma 2001).

Obvious practical reasons for this are their huge diversity and economical importance (Hölldobler

& Wilson 1990), and the wide knowledge of their natural history and behaviour (e.g. Hamilton 1964, Wilson 1971, Hölldobler & Wilson 1990).

Furthermore, as objects of confl ict studies eusocial insects have some special features that distinguish them from most other cooperative systems.

The most striking feature, which is used also in defi ning advanced sociality of Hymenopterans, is the remarkable reproductive division of labour (Hölldobler & Wilson 1990, Crespi & Yanega 1995).

In extreme cases, such as leaf cutting attine ants or Echiton army ants, a single queen is the only reproductive individual in the colony of hundreds of thousands of individuals. Thus, social insects have in their polymorphism evolved far from being analogous with the basic Tragedy of the commons -situation, where similarly endowed individuals share a common resource. Even if workers are able to reproduce, their reproductive value is likely to be orders of magnitude lower compared to that of the queen. There are two reasons for this, the higher fecundity of the queen (Hölldobler & Wilson 1990), and her potentially much longer life-span (Keller

& Genoud 1997). For example, workers of the ant Formica exsecta usually live for one year, while the queen may live and reproduce for 30 years (Pamilo 1991a). Furthermore, because the only role for the queen in the colony is reproduction, her reproduction is cost-free (in terms of kin selection theory, not life history theory), while worker reproduction most likely decreases colony effi ciency (Oster & Wilson 1978, Franks et al 1990, Cole 1986).

As was already noted in the earliest discussions of reproductive confl ict in hymenopteran societies (Trivers & Hare 1976), asymmetry in reproductive value has far-reaching effects on expression of potential confl icts. First, this asymmetry increases the power of the queen, because she is often invaluable for the colony. This is not only because of her high reproductive capacity, but also because she is in monogynous societies the only source of new females and thus necessary for long term survival of the colony. Thus workers should not risk the life of the queen e.g. by aggressively confronting her in fi ghts over dominance. Conversely, the asymmetry offers also the workers an advantage.

This is because they are far more numerous than the queens even in the most polygynous species and because they have the control over rearing the brood. This gives them the power to bias both the sex ratio of the brood and the caste ratio of developing females, in addition to the possibility of laying their own eggs.

This far, social insects have offered the most successful examples of policing. Worker policing of eggs has been shown to be remarkably effi cient in honeybee Apis mellifera and also its wild relatives A. cerana and A. fl orea. In these policing is predicted by relatedness alone, as queens are highly multiple mated (Palmer & Oldroyd 2000) and relatedness among workers consequently low. Vespinae wasps, the other model group of policing, however also include species where costs arguments are needed to explain policing, for example the hornet Vespa crabro (Foster et al 2002) and the common wasp Vespula vulgaris (Foster &

Ratnieks 2001a).

Successful worker reproduction seems to be an exception rather than the rule in social Hymenoptera. Based on male parentage data in 50 species of social insects, it seems that worker policing and self policing are both common, and that costs arguments are a better explanation than relatedness for the queen dominance in male production (Hammond & Keller in press). However, this study is based only on data of the confl ict outcome, and more detailed studies of proximate

(10)

mechanisms are needed to show the importance of worker policing and self policing.

Costs of worker reproduction are a widely used explanation for the lack of worker reproduction (e.g. reviews by Tóth et al 2004, Hammond

& Keller in press), but these costs have rarely been thoroughly quantifi ed. It has been shown numerous times that reproductive workers work less than non-reproductives (Ross 1985, Monnin

& Peeters 1999, Martin et al 2002, Hartmann et al 2003), but the actual colony level costs of this have rarely been shown (but see Cole 1986, Reeve 1991, Lopez-Vaamonde et al 2003). Thus, the costs of worker reproduction are an obvious target for future studies. However, to know where costs are likely to differ crucially, careful studies of confl ict expression and relatedness are needed.

Formica

Even if worker reproduction has been studied in a number of species, tests among large numbers of closely related species are few, with the exception of Apis bees (reviewed in Barron et al 2001) and Vespinae wasps (reviewed in Foster &

Ratnieks 2001b). Furthermore, the studied species are not a very representative sample of social hymenopterans. Polygyny (i.e. the presence of multiple reproductive queens in a mature colony) is widespread in ants and occurs in other social insects as well (Hölldobler & Wilson 1990, Keller 1993). Especially in ants it has strong correlations with ecology and dispersal (Bourke & Franks 1995, Keller 1993, V), but worker reproduction has been studied in only two polygynous ant species (Evans 1998, Hammond et al 2003), the present work (III, IV, V) excluded. Also in the large comparative study of worker reproduction (Hammond & Keller in press), there were only 2 species where worker relatedness is lower than 0.5, even though relatednesses in ants are often very low, even zero (Bourke & Franks 1995, Crozier & Pamilo 1996). Similarly, colony sizes of the species studied this far are, with a few exceptions, small (few hundred workers or less, Hammond & Keller in press) compared to the wide

range of colony sizes found in ants (Hölldobler &

Wilson 1990, Bourke 1999).

The wood ant genus Formica is in many ways a good choice for a study of confl ict determinants.

First, Formica, especially the Palaearctic species studied here, are one of the best known ant groups with respect to population and colony kin structures (see e.g. Pamilo 1982, 1983, 1984, Rosengren &

Pamilo 1983, Pamilo et al 1978, 1992, Rosengren et al 1993, Sundström 1993, Gyllenstrand 2002).

Second, as shown by this great body of work there is extreme variation in several key determinants of confl icts, namely queen number, relatedness and colony size. This allows a comparative approach to the problems, which is essential in separating effects of phylogeny from the other confl ict determinants. Furthermore, there is even great variation within species (Sundström 1993, Goropashnaya et al 2001, Gyllenstrand 2002, Zhu et al 2003, I & II) in relatedness, which allows tests within species. Also the relations among species (i.e. competition, Savolainen & Vepsäläinen 1990, social parasitism Hölldobler & Wilson 1990, Czechowski et al 2003, community succession Punttila et al 1991, 1994, Vepsäläinen et al 2000), which may have implications on kin structures (I, II) or confl ict resolution (IV) are well known.

Before this study, worker reproduction in Formica has only been genetically studied in a few species. Based on parentage of males reared to adulthood, worker reproduction seems to be absent in monogynous F. exsecta and F. truncorum, and possible in slightly polygynous F. rufa (Walin et al 1998). Thus, further studies are clearly needed that look at a larger number of species and look at the confl ict expression beyond the fi nal brood.

Aims of the thesis

This work had three main aims. The fi rst aim was to study the variation in kin structure within and among populations of Formica fusca and the genetic differentiation and gene fl ow among populations (I, II). This gives us information on the possible genetic basis of behavioural differences and tells

(11)

12

us if gene fl ow among populations may prevent optimal confl ict behaviour. The second aim was to study the extent of confl ict expression on several different levels, from variation among species and its correlates (kin structure, colony size, phylogeny, V) to proximate mechanisms of confl ict resolution within colonies (egg policing, worker worker aggression, self policing, III, IV, V). The third aim was to study theoretically how worker policing, relatedness and colony size together determine the optimal level of worker fertility (VI).

Material and methods

In order to accomplish a diverse picture of both ultimate and proximate determinants of the confl ict, a combination of several methods is needed. Thus this thesis combines theoretical work with genetic studies, experimental work and a comparative analysis that incorporates phylogeny into the study of confl icts.

1. Genetic analyses

We uses microsatellite markers developed for Formica ants (Chapuisat 1996, Gyllenstrand et al 2002) for the analysis of genetic differentiation (I, II), kin structure (I, II, III, V) and parentage (III, V).

Microsatellites are a powerful and effective tool for these questions (Queller et al 1993, Balloux &

Lugon-Moulin 2002, Shlötterer 2004).

2. Experimental work and bioassays

Experiments were used for assessing the effects of social conditions (i.e. the presence of a queen) for all aspects of confl ict expression and resolution, i.e. worker ovary development (IV, V), worker egg laying (III, V), policing behaviour (IV) and parentage of the males reared (III). The use of laboratory nests was necessary for several reasons.

First, in laboratory nests the social conditions, i.e.

the presence of the queen can be controlled, and the queen is available for genotyping and parentage analysis (III, IV). The presence of multiple queens

in fi eld colonies, and the large size of Formica colonies make estimation of worker reproduction in natural colonies virtually impossible (V). Second, colonies of F. fusca move easily if disturbed, which makes continued studies in the fi eld diffi cult. Third, detailed observations of behaviour and transfers of eggs (IV) are impossible in the fi eld due to the underground nesting habits of F. fusca.

3. Comparative analysis

Comparison of traits across species is a powerful means of detecting possible adaptation (Futuyma 1998). Because closely related species do not represent independent data points, additional measures must be taken to make the right conclusions concerning correlations of traits across species (Felsenstein 1985). Thus we analysed the patterns of worker fertility, relatedness and colony size in the light of a recent molecular phylogeny of Formica ants (Goropashnaya 2003, V). All traits were tested for phylogenetic independence using a randomization based method (Abouheif 1999, Reeve & Abouheif 2003). For traits that were phylogenetically dependent, independent contrasts (Felsenstein 1985) were then analysed instead of the actual trait values. The correlations between worker fertility, relatedness and colony size are then used to indirectly infer the possible effects of worker policing on worker fertility (VI, V)

4. ESS-analysis

The optimal worker behaviour was analysed using an ESS (Evolutionary stable strategy, Maynard Smith 1982) method with kin selection extensions (sensu Taylor and Frank 1996, Frank 1998). This type of analysis allows the costs and benefi ts of selfi sh behaviour to depend on the actions of other individuals in the colony and the population (Frank 1998, VI). Furthermore, this type of analysis allows any relatedness values to be analysed, whereas in population genetics models a separate ESS model for each kin structure has to be constructed (Wenseleers et al 2003, VI).

(12)

ESS models of this kind do not aim at exact numerical predictions, because optimal behaviour of organisms is subject to various constraints (Oster

& Wilson 1978, Frank 1998). Rather the model points out into which direction the key parameters are likely to affect the outcome (Frank 1998). Thus, best support for the models should come from comparative studies among species where key parameters (such as relatedness and colony size) differ (V).

Main results and conclusions

Genetic variation and gene fl ow among populations: implications for adaptive confl ict behaviour

The studies of genetic structuring in F. fusca showed that kin structure varies widely both within and among populations (I, II), but that genetic differentiation among populations is not high (II). This has two implications for the evolution of confl ict behaviour. First, gene fl ow among populations with different average kin structures may prevent individuals from behaving optimally in kin selection contexts, if the optimal behaviour depends on population averages (II). This may be one reason why worker fertility is not correlated with relatedness across or within species (V).

Gene fl ow among populations with differences in average kin structure has been shown to occur also in F. exsecta and F. cinerea (Goropashnaya et al 2001, Zhu et al 2003, Gyllenstrand 2002). Such gene fl ow suggests that studies concerning optimal worker behaviour based on a single population have to be interpreted very cautiously, especially in species where variation among populations has not been assessed. Second, even though dispersal in F. fusca is to some extent limited on local scales, differentiation among populations is not great (II).

This shows that the differences in kin structure among populations and the underlying behaviour are unlikely to be based on genetic differences.

Furthermore, because kin structure varies widely within populations and populations are short-

lived (I, II), it is unlikely that F. fusca would evolve into genetically differentiated social forms, such as those found in F. truncorum (Sundström 1993, Gyllenstrand 2002), and Solenopsis invicta (Ross

& Shoemaker 1997). Considered together, our results suggest that plastic, facultative responses are the best explanation for variation in confl ict behaviour in Formica ants, This agrees with the fi nding of responses in behaviour to variation in relatedness in several studies of Formica ants (Sundström 1994, Sundström et al 1996, Hannonen

& Sundström 2003).

We did not fi nd unambiguous support for the predicted connection of kin structure and habitat age, even though in the two populations studied in (I) habitat age seemed to be the crucial parameter.

Relatedness was highest in old populations (II), but inbreeding was only found in the two populations already studied in (I). Inbreeding does not seem to be common in ants (Pamilo et al 1997, but see Cole

& Wiernasz 1997 and Sundström et al 2003), and high relatedness has usually been considered to be associated with large scale mating fl ights and outbreeding populations (Pamilo et al 1997, Ross 2001). Our study shows, however, that this is not always the case (see also Hasegawa &Yamaguchi 1995, Sundström et al 2003, Oberstadt & Heinze 2003), and that the connection of kin structure and differentiation among populations may be is not determined only by variation in queen number, but also reproductive partitioning and relatedness among reproductives have to be considered (I, II, Ross 1993, 2001, Pamilo et al 1997).

Our results (I, II) agree with the earlier knowledge of dispersal and colonizing abilities of F. fusca. On the one hand, dispersal of F. fusca has been shown to be limited, especially over open water (Vepsäläinen & Pisarski 1982), which is supported by the clusters of related colonies found within several populations and the larger average differentiation of island populations (II). On the other hand, F. fusca is known to be an effi cient colonizer of newly opened habitats (Punttila et al 1991, 1994), which is supported by the lack of signifi cant bottlenecks in recently expanded

(13)

14

populations (II). However, more detailed analyses of spatial genetic patterns within populations, and preferably also with maternally inherited genetic markers (see e.g. Ross et al 1997, 1999) are needed to assess the extent of local colony founding and the possibility of colony founding by budding.

This will have implications for social evolution of confl icts in F. fusca, e.g. for example the sex ratio evolution in the species, because the extent of local resource competition among related colonies that result from budding is likely to affect optimal sex ratios (Pamilo 1991b). Furthermore, predictions of reproductive skew models may be affected by the possibility of budding. If a queen does not gain a large enough share of reproduction, she may have the possibility of leaving the colony with some workers from the original colony. This may limit the power of a dominant queen to force her interests on submissive queens (Reeve & Keller 2001, Kokko 2003).

Ultimate confl ict determinants:

theory and results from Formica ants

The comparative study (V) shows that worker fertility is widespread in the genus Formica.

Worker laid male eggs were found in all but two of the species sampled, and the proportions of fertile workers were high compared to other species with comparable colony sizes (such as Apis species and Vespula vulgaris). According to theory (Ratnieks 1988, Frank 1995, 2003, VI), there are two possible ultimate explanations for this. First, it may be that costs of worker fertility are for some reason low in Formica, and thus selection pressures for self policing are weak (Foster 2004, VI). Second, it may be that worker policing is weak in Formica, and thus workers may actually succeed in reproduction.

Possible reasons for the lack of policing include a low need for policing, e.g. when costs of worker fertility are low (Ratnieks 1988), or high costs of policing, due to e.g. recognition errors (Ratnieks 1988, Ratnieks & Reeve 1992). These explanations are not mutually exclusive and further studies of colony level costs of worker reproduction, the

effi ciency of policing and also the parentage of males reared are needed for the full picture of the confl ict. This far the data shows effi cient worker policing in Formica (III, Walin et al 1998) which suggests that low or non linear colony level costs are the ultimate reason for high worker fertility (V, IV). Thus self policing would not be strongly selected for, even if worker policing was effi cient.

We found that worker fertility increased in the absence of a queen (V), which suggests self policing by workers in the presence of a queen. The model (VI) offers two mutually non exclusive ultimate explanations for self policing. First, worker policing may select for acquiescence, i.e. workers giving up reproduction in response to social policing.

This route into decreased confl ict expression was already verbally considered by Ratnieks (1988) and formally confi rmed by Frank (1996, 2003) in the general model of policing. Our detailed model shows that self policing should indeed occur also in social insect colonies whenever worker policing is effi cient (VI). Second, self policing may be a response to the high fertility of a queen. If the queen honestly signals her high fertility to workers, they may be selected to refrain from reproduction without coercion (VI). This idea has been verbally framed by Seeley (1985) and Keller

& Nonacs (1993) and it has been widely accepted as an explanation for worker acquiescence (Bourke

& Franks 1995, Keller & Reeve 1999, Sundström &

Boomsma 2001) The kind of information required for such signalling has been shown to exist in Formica ants and numerous other species (Liebig et al 1999, 2000, Ortius & Heinze 1999, Hannonen et al 2002, Heinze et al 2002), but the idea has not been previously proved formally (VI). The model (VI) and the indirect empirical evidence (III, V) show that effective communication, which is crucial for functioning of social insect colonies in the contexts of e.g. foraging and nest defence (Hölldobler & Wilson 1990), also plays a large role in the outcome of confl icts.

The model also entails an interesting feedback between queen fertility and self policing. The more fertile the queen is, the less fertile are the workers

(14)

predicted to be (VI). Thus the relative queen fertility increases, which again selects for less fertile workers. This feedback is further enhanced by the likely correlation of high queen fertility and large colony size (Bourke 1999), both factors which should select for less fertile workers. This feedback may be crucial for alleviation of direct confl icts over reproduction and the evolution of large, effi cient colonies with effi cient division of labour among workers (Oster & Wilson 1978, Bourke 1999).

Worker ovary development and egg laying were not correlated to phylogeny, colony size or relatedness (V). This agrees with the key result of the ESS analysis (VI) which shows that not only relatedness, but more importantly the effi ciency of worker policing and costs of worker reproduction determine the optimal level of worker fertility.

The model prediction seems to hold also in bees and wasps where worker policing has been closely studied and where worker policing occurs independently of relatedness (Foster et al 2002, Foster & Ratnieks 2001a). In species where worker policing is the most effi cient, such as Apis mellifera (> 99 %, Visscher 1996), A. fl orea (~98%, Halling et al 2001) and Vespula vulgaris (>99%, Foster &

Ratnieks 2001a), the proportion of fertile workers is correspondingly low (0.1, 0.5 and 1% respectively).

Such low levels of selfi shness can not be explained by relatedness alone, unless costs of worker reproduction were extreme. In species with less effi cient policing, such as Dolichovespula species, fertile workers are respectively more common but still lower than expected in the absence of policing (e.g. 6% fertile workers and policing effi ciency of 86% in D. media, and 9% fertile workers and policing effi ciency of 87% in D. sylvestris, Foster et al 2001).

Ultimately worker fertility may also be governed by the risk of colony orphaning (V).

Workers of F. aquilonia and F. polyctena, where worker fertility was clearly decreased, are very unlikely to end up in queenless situation where worker reproduction would be unchecked because of the permanent, obligate polygyny in these

species (Rosengren et al 1993). Thus there is no need for the workers to maintain functional ovaries as an insurance against queenlessness (Franks et al 1990). This contrasts the pattern in other highly polygynous species (F. cinerea, F. exsecta and F. truncorum), where occasional queenlessness is presumably more common due to more variable colony structures and colony founding strategies (Pamilo 1991a, Sundström et al 1996, Goropashnaya et al 2001, Liautard & Keller 2001, Gyllenstrand 2002, Zhu et al 2003, Elias, M., Rosengren, R. & Sundström, L., unpublished data). The fi nding that the highest amounts of fertile workers (F. cinerea) and worker egg laying (F. truncorum) were found in unicolonial species, where relatedness is zero, suggests that selfi shness may be selected for in unicolonial populations, where co-operative behaviour no longer pays off (Queller & Strassmann 1998).

Proximate mechanisms:

studies of Formica fusca

Proximately the confl ict seems to be resolved by a combination of worker policing and self policing, probably facilitated by queen signalling (III, IV, V).

The microsatellite data from F. fusca shows that although offspring of workers have only remote chances of being reared in the presence of a queen, workers commonly lay eggs (III). However, the observations show that these eggs are removed by other workers (IV). Worker policing of eggs (sabotage) is not complemented by aggressive policing (punishment), probably because the potential costs of such aggression are high when such a large proportion of workers is fertile (IV).

It seems that because worker laid eggs have slim chances of being reared, workers have also evolved a level of self restraint, though incomplete, in queenright conditions (III, IV). However, without estimating costs of worker reproduction we are unable to tell if this actually results in colony level costs of the confl ict, or are the costs of worker fertility alleviated e.g. due to non-linear cost functions (Foster 2004, VI, V).

(15)

16

Interestingly, the proximate basis of worker policing in F. fusca seems to be completely different from the species studied this far (IV). In Apis bees (Ratnieks & Visscher 1989, Halling et al 2001, Oldroyd et al 2001), Vespinae wasps (Foster

& Ratnieks 2000, 2001), and the ants Camponotus fl oridanus (Endler et al 2004), and Pachycondyla inversa (D’Ettorre et al 2004) workers accept eggs laid by non-nestmate queens, which suggests that all eggs bearing a universal queen chemical are accepted (Ratnieks 1995, Martin et al 2002, Endler et al 2004). In F. fusca, only eggs laid by a nestmate queen were accepted (IV). There are two mutually non-exclusive ultimate explanations for this behaviour. First, species of subgenus Serviformica, such as F. fusca, are commonly parasitized by queens of other Formica species (Hölldobler &

Wilson 1990, Czechowski et al 2003). Thus there is a strong selection pressure for workers to discriminate against eggs laid by alien queens. This interpretation is supported by the fi nding that F.

fusca workers also discriminate against eggs laid by F. rufa queens (Helanterä, unpublished data).

Second, the relatedness of queens in polygynous colonies of F. fusca varies widely, from zero to full sisters (I). This means that workers may greatly benefi t from discriminating among offspring of different queens. This kind of discrimination has been shown to occur in F. fusca (Hannonen

& Sundström 2003) and it might as a side effect produce a nestmate recognition effect. This shows that in addition to its ultimate effects (i.e. changes in kin structure and the probability of colony orphaning (V)), polygyny may also have effect on the proximate mechanisms behind confl ict resolution. Also ecological pressures like social parasitism may strongly infl uence the mechanisms.

The fi nding of nestmate recognition of eggs is novel and demands further studies in several aspects, e.g..

the occurrence of such behaviour in other species within and outside Formica, and the correlations of the behaviour to polygyny, genetic variability within colonies and parasite pressure.

Conclusions

The combination of approaches in this study reveals a complex picture of the confl ict. First, it is evident that it is not enough to study the outcome of the confl ict, i.e. the parentage of males reared.

Although this tells about the individual fi tness benefi ts of worker reproduction, it does not tell us anything about the confl ict expression and the associated inclusive fi tness costs incurred via decreased colony success. Thus, in order to get a full picture of the Tragedy of the commons and the effect of confl ict expression on the group level, quantifying confl ict expression at lower levels of worker policing and self policing is an essential fi rst step. This must be given emphasis in future studies of reproductive confl icts. Second, the studies of genetic differentiation revealed that variation in kin structure and queen number within the range of substantial gene fl ow may keep the workers from achieving optimality in confl ict behaviour.

Thus studies of confl ict behaviour that are based on a single population should be generalized with caution.

Our studies show that Formica ants, especially the polygynous ones, are in many respects different from the monogynous model systems of worker reproduction and policing, namely honeybees and wasps. Not only does polygyny affect the relatedness structure and size of the colonies, but it may have implications for the proximate mechanisms as well, as illustrated by the nestmate recognition effect in Formica fusca (IV). Furthermore, queen number is closely linked to the extent and direction of gene fl ow among populations and the probability of colony orphaning. Thus future studies of worker reproduction to in polygynous and socially variable species are called for, since they may give a different picture of the confl icts than monogynous ones.

The resolution of the confl ict over male parentage in Formica ants is a combination of worker policing, self policing and honest communication. Proportions of fertile workers are high compared to species such as Apis mellifera, the

(16)

epitome of colony integration and superorganism views (Seeley 1997, Queller 2000). Further studies in Formica will have to assess if colony level costs of high worker fertility are high enough to make Formica ants an example of strong Tragedy of the commons in social insects or if they will be an example of low costs of worker reproduction and not-so-tragic commons (Foster 2004).

ACKNOWLEDGEMENTS

First I of course want to thank Lotta Sundström, my supervisor. Lotta introduced the world of confl icts on her lectures and there has been no return for me.

She has always had time for discussions and her input to my work has been tremendous, throughout the years and most importantly during the last chaotic weeks of writing up. Furthermore, she let me work without extra pressure from her, which is the best way to get me to do any work. Importantly, Lotta’s authority does not come from a hierarchy but her knowledge. Acknowledging this authority has often meant extra work with manuscripts but in retrospect it has proven to be for the best.

Also the other members of Team::Antzz have been a tremendously enjoyable group of personalities to work and have a good time with.

Especially I want to thank Kalle, Katja and Minttu who have shared with me the offi ce, work and play in Tvärminne, and diverse discussions, sometimes very helpful scientifi cally, sometimes just plain entertaining. Especially Katja’s friendship and enthusiasm have cheered up many days. Luppa was invaluable help in the fi eld and hilarious company during long evenings in Tvärminne. Everybody in the group over the years, Vienna, Perttu (thanks for population genetics advice and swedish coffee), Chris, Emma, Cathy, Marianne, Kriko, Elina, Paul, Riitta, Tuomo and Hanni have in their own way made life at and outside work friendly and nice.

Rainer Rosengren provided priceless help, pleasant company and discussions of Caribbean religion in the fi eld and deep insight into the Formica

world. Jouni Sorvari discussed ants and interesting research plans. Soile Kupiainen has been a tremendous help in the lab, genotyping hundreds and hundreds of individuals reliably and carefully, and making everything run smoothly. In addition, Mikko Putkonen, Salla Heiman, Hanna Paulomäki, Kata Valosaari, Salla Suomensaari, Heidi Viljanen and Timo Honkasalo helped in sorting ants and the experiments.

During my PhD work I’ve had the opportunity to travel and meet remarkable people. I want to thank the people at Sheffi eld University for fruitful collaboration and good times. Tom showed me a glimpse of theoretical work, let me stay at his and Kristien’s home during my visit and deserves a mention as a great initiator of parties. I thank Francis for fruitful and interesting discussions and new ideas (which did not always concern honeybees), beekeeping outings to the Peaks and healthy breakfasts. Adam H., Adam T., Elva, Nic, Nigel, Margaret and others made me feel at home in Sheffi eld. Furthermore, I want to acknowledge Kevin for entertaining moments and discussions (sometimes about policing, occasionally loud and non-scientifi c) all over the world from a bar in Hanko to under a waterfall in Hokkaido. Everybody in the network meetings proved that social insect research can be a lot of fun.

The Department of Ecology and Systematics/

Biological and Environmental Sciences has been an excellent place to work at. Special thanks to all the other PhD students for Wednesday morning coffee, and the social footballers, both on the pitch/beach and into UCL. Those involved in the PhD-meeting in Lohja taught me something novel, i.e. that organising can actually be fun. From all the teachers, I especially want to thank Hannu Pietiäinen for being an excellent motivator and showing a genuine interest in the progress of students, and Kari Vepsäläinen for a lot of food for thought over numerous discussion groups. Ilkka Teräs, Hannu Pietiäinen, Jari Niemelä and Veijo Kaitala have taken care of a lot of bureaucracy related to my studies. Jouni Vainio has patiently taken care of computers and software. In the

(17)

18

MES-lab Anne Aronta and Tupu Kuismin took care of a lot of things I have no idea of, and everybody working there made the atmosphere pleasant.

Tvärminne Zoological station provided facilities and beautiful surroundings for fi eld and experimental work. Especially Raija, Lallu and Eva were always helpful when practical matters needed taking care of. Outside working hours, I want to thank the other inmates, Christoph, Peter, Kim, Topi, Marja, Riggert, Markus, Kai, Mika, Olli, Kongo, you are too many to mention… for making evenings enjoyable and the following mornings more bearable.

For making me forget work, I want to thank all my friends, especially Mikko for sharing a fl at and a lot of rioutously funny but also more serious moments and Arttu, Jani and Jussi for friendship over the years. Furthermore, all the friends I have been making music with have been important for my mental well-being. The mighty FC Tolsa, FC OssiKumpu and FC Tolsan Torso have provided physical exercise and recreation in pleasant company.

My family has supported me in all respects, and well beyond. Kiitos vanhemmilleni loppumattomasta tuesta ja huolehtimisesta, veljilleni Antille ja Ilkalle (sekä Ludmilalle ja Pauliinalle) lukemattomista korvaamattoman arvokkaista hetkistä, ja lisäksi veljenpojalleni Otolle horjumattomasta uskosta biologiseen tietämykseeni.

Finally I want to thank Krista for love and everything else.

This work has been funded by the Evolutionary Ecology graduate school, Academy of Finland, Oskar Öfl unds stiftelse, the INSECTS network, the Chancellor of Helsinki University and Hämäläis- Osakunta. Dr. Andrew Bourke and Dr. Janne Kotiaho are gratefully acknowledged for effi cient reviewing of this thesis. Voitto Vuorio drew the ants on the cover.

References

Abouheif, E. 1999: A method for testing the assumption of phylogenetic independence in comparative data.

— Evol. Ecol. Res. 1: 895-909.

Axelrod, R & Hamilton, W.D. 1981: The evolution of cooperation. — Science 211: 1390–1396.

Balloux F. and Lugon-Moulin, N. 2002: The estimation of population differentiation with microsatellite markers.

— Molecular Ecology 11: 155-165.

Barron, A.B., Oldroyd, B.P. & Ratnieks, F.L.W. 2001:

Worker reproduction in honey-bees (Apis) and the anarchic syndrome: a review. — Behav. Ecol. Sociobiol.

50: 199-208.

Beekman, M. & Ratnieks, F.L.W. 2003: Power over reproduction in social Hymenoptera. — Phil. Trans. R.

Soc. Lond. B. 358: 1741-1753.

Beekman, M., Komdeur, J. & Ratnieks, F.L.W. 2003:

Reproductive confl icts in social animals: who has power? — Trends Ecol. Evol. 18: 277-282.

Bourke, A.F.G: 1999: Colony size, social complexity and reproductive confl ict in social insects. — J. Evol. Biol.

12: 245–257

Bourke, A.F.G. & Franks, N.R. 1995: Social evolution in ants. — Princeton university press. Princeton, New Jersey. 529 p.

Chapuisat, M. 1996: Characterization of microsatellite loci in Formica paralugubris B and their variability in other ant species. — Molec. Ecol. 5: 599–601.

Clutton-Brock, T.H. & Parker, G.A. 1995: Punishment in animal societies. — Nature 373:209-216.

Cole, B.J. 1986: The social behavior of Leptothorax allardycei (Hymenoptera: Formicidae): time budgets and the evolution of worker reproduction. — Behav.

Ecol. Sociobiol. 12: 191–201.

Cole, B.J. & Wiernasz, D.C. 1997: Inbreeding in a lek- mating ant species, Pogonomyrmex occidentalis.

— Behav. Ecol. Sociobiol. 40: 79-86.

Crespi, B.J & Yanega, D. 1995: The defi nition of eusociality. — Behav. Eco.. 6: 109-115.

Crozier, R & Pamilo, P: Evolution of social insect colonies.

— Oxford University Press. Oxford. 306 pp.

Czehowski, W., Radchenko, A. & Czehowska, W. 2002:

The Ants of Poland. — Warszawa, Warsaw.

Darwin,C. 1859: Origin of species.

D’Ettorre. P., Heinze, J. & Ratnieks, F.L.W. 2004: Worker policing by egg eating in the ponerine ant Pachycondyla inversa. — Proc. R. Soc. Lond. B, in press.

De Waal, F.B.M. 1996: Good natured: the origins of right and wrong in humans and other animals. — Harvard University Press, Cambridge, Massachusetts.

Dugatkin, L.A. 1997: Cooperation among animals: an evolutionary perspective. — Oxford University press, NY.

—. 2002: Animal cooperation among unrelated individuals. — Naturwissenschaften 89: 533-541.

Endler, A., Liebig, J., Schmitt, T., Parker, J.E., Jones, G.R.,

(18)

Schreier, P. & Hölldobler, B. 2004: Surface hydrocarbons of queen eggs regulate worker reproduction in a social insect. — Proc. Nat. Acad. Sci. 101 (9): 2945-2950 Evans, J. D. 1998: Parentage and sex allocation in the

facultatively polygynous ant Myrmica tahoensis.

— Behav. Ecol. Sociobiol. 44: 35–42.

Felsenstein, J. 1985: Phylogenies and the comparative method. — Am. Nat. 125: 1-15.

Fletcher, D.J.C. & Ross, K.G. 1985: Regulation of reproduction in eusocial Hymenoptera. — Ann. Rev.

Entomol. 30: 319–43.

Foster, K.R. 2004: Diminishing returns in social evolution:

the not-so-tragic commons. — J. Evol. Biol. 17:

Foster, K.R. & Ratnieks, F.L.W. 2000: Facultative worker policing in a wasp. — Nature 407: 692–693.

—. 2001a: Convergent evolution of worker policing by egg eating in the honeybee and common wasp. — Proc.

R. Soc. Lond. 268:169-174

—. 2001b: Paternity, reproduction and confl ict in vespine wasps: a model system for testing kin selection predictions. — Behav. Ecol. Sociobiol. 50: 1-8.

Foster, K.R., Ratnieks F.L.W., Gyllenstrand, F. & Thorén, P.A. 2001: Colony kin structure and male production in Dolichovespula wasps. — Mol. Ecol 10: 1003-1010.

Foster, K.R., Gulliver, J. & Ratnieks, F.L.W. 2002: Worker policing in the European hornet Vespa crabro. — Ins.

Soc. 49: 41-44.

Foster, K. R., Shaulsky, G., Strassmann, J.E., Queller, D.C. & Thompson, R.L.: Pleiotropy as a mechanism to stabilize co-operation. — Nature, in press.

Frank, S.A. 1995: Mutual policing and repression of competition in the evolution of cooperation groups.

— Nature 377: 520–522.

—. 1996: Policing and group cohesion when resources vary. — Anim. Behav. 52: 1163–1169.

—. 1998: Foundations of social evolution. — Princeton university press. Princeton, New Jersey.

—. 2003: Repression of competition and the evolution of cooperation. — Evolution 57:693-705.

Franks, N.R., Ireland, B. & Bourke, A.F.G. 1990: Confl icts, social economics and life history strategies in ants.

Futuyma, D.J. 1998: Evolutionary biology.— Sinauer Associates, Sunderland, Massachusetts.

Goropashnaya, A.V. 2003: Phylogeographic structure and genetic variation in Formica ants. — PhD Thesis, Uppsala Universitet.

Goropashnaya, A.V., Seppä, P. & Pamilo, P. 2001: Social and genetic characteristics of geographically isolated populations in the ant Formica cinerea. — Mol. Ecol.

10: 2807-2818.

Grafen, A. 1985: A geometric view of relatedness- — In Dawkins, R. & Ridley, M. (ed.): Oxford Surveys in evolutionary biology, vol. 2: 1-18.

Griffi n, A.S. & West, S.A. 2002: Kin selection: fact and fi ction. — Trends Ecol. Evol. 17: 15-21.

Gyllenstrand, N. 2002: Effects of social organisation on spatial genetic structures in Formica ants. — PhD

Thesis, Uppsala Universitet.

Gyllenstrand N., Gertsch, P.J. & Pamilo. P. 2002:

Polymorphic microsatellite DNA markers in the ant Formica exsecta. — Mol. Ecol. Notes 2: 67-69.

Halling, L.A., Oldroyd. B.P., Wattanachaiyingcharoen, W., Barron, A.B., Nanork, P., et al. 2001: Worker policing in the bee Apis fl orea. — Behav. Ecol. Sociobiol. 49:

509-513.

Hamilton, W.D. 1964: The genetical evolution of social behaviour I & II. J. Theor. Biol. 7: 1–16, 17–52.

—. 1967: Extrarordinary sex ratios. — Science 156: 477- 88.

—. 1970: Selfi sh and spiteful behaviour in an evolutionary model. — Nature 228: 1218-20.

—. 1972: Altruism and related phenomena, mainly in social insects. — Ann. Rev. Ecol. System. 3: 193–232.

Hammerstein, P. (ed.) 2004: Genetic and cultural evolution of cooperation. — The MIT press, Cambridge, Massachusetts. 485 p.

Hammond, R.L. & Keller, L. 2004: Confl ict over male parentage in social insects. — PloS Biology, in press.

Hammond, R.L., Bruford, M.W. & Bourke, A.F.G. 2003:

Male parentage does not vary with colony kin structure in a multiple-queen ant. — J. Evol. Biol. 16: 446-455.

Hannonen, M. & Sundström, L. 2003: Worker nepotism among polygynous ants. —Nature 42:921.

Hannonen, M., Sledge, M.F., Turillazzi, .S & Sundström, L. 2002: Queen reproduction, chemical signalling and worker behaviour in polygyne colonies of the ant Formica fusca. — Anim. Behav. 64: 477-485.

Hardin, G. 1968: The tragedy of the commons. — Science 162: 1243-1244.

Hasegawa, E. & Yamaguchi, T. 1995: Population structure, local mate competition, and sex allocation pattern in the ant Messor aciculatus. — Evolution. 49: 260-265.

Hartmann, A., Wantia, J., Torres, J. A. & Heinze, J. 2003.

Worker policing without genetic confl icts in a clonal ant. — Proc. Nat. Acad. Sci. 100: 12836-12840.

Heinze, J., Stengl, B. & Sledge, M.F. 2002: Worker rank, reproductive status and cuticular hydrocarbon signature in the ant, Pachycondyla cf. inversa. — Behav.

Ecol. Sociobiol. 52: 59-65

Hölldobler, B. & Wilson E.O. 1990: The ants. — The Belknap press of Harvard university press. Cambridge, Massachusetts. 732pp.

Keller, L. (ed.) 1993: Queen number and sociality in insects. — Oxford University Press. Oxford. 439 pp.

Keller L. (ed.) 1999: Levels of selection in evolution.

Princeton university press. Princeton, New Jersey.

Keller, L. & Nonacs, P. 1993: The role of queen pheromones in social insects: queen control or queen signal?. — Anim. Behav. 45: 787–794.

Keller, L. & Genoud M. 1997: Extraordinary life spans in ants: a test of evolutionary theories of aging. — Nature 389: 958-960.

Kokko H. 2003: Are reproductive skew models evolutionarily stable? —

Proc. R. Soc. B 270: 265-270.

(19)

20

Leigh, E.G., Jr. 1999: Levels of selection, potential confl icts and their resolution: the role of the ”common good”. — In: Keller L. (ed.): Levels of selection in evolution. Princeton university press. Princeton, New Jersey. 529 p

Lewontin, R.C. 1970: The units of selection. — Ann. Rev.

Ecol. Syst. 1: 1-18.

Liautard, C. & Keller, L. 2001: Restricted effective queen dispersal at a microgeographic scale in polygynous populations of the ant Formica exsecta. — Evolution 55: 2484-2492.

Liebig, J., Peeters, C. & Hölldobler, B. 1999: Worker policing limits the number of dominance in a ponerine ant. — Proc. R. Soc. Lond. B 266: 1865–1870.

Liebig, J., Peeters, C., Oldham, N.J., Markstädter. C. &

Hölldobler, B. 2000: Are variations in cuticular hydrocarbons of queens and workers a reliable signal of fertility in the ant Harpegnathos saltator?. — Proc.

Natl. Acad. Sci., U.S.A. 97: 4124.

Lopez-Vaamonde, C., Koning, J.W., Jordan, W.C. &

Bourke, A.F.G 2003: No evidence that reproductive workers reduce the produciton of new queens. — Anim.

Behav. 66: 577-584.

Martin, S.J., Beekman, M., Wossler, T.C. & Ratnieks, F.L.W.

2002: Parasitic Cape honeybee workers, Apis mellifera capensis, evade policing. — Nature 415:163-5.

Maynard–Smith, J. 1982: Evolution and the theory of games. — Cambridge University Press, New York.

Maynard–Smith, J. & Szathmary, E. 1995: The major transitions in evolution. — Oxford University Press.

Oxford.

Michod, R.E. 2004: Cooperation and confl ict mediation during the origin of multicellularity. — In: Hammerstein, P. (ed.) :Genetic and cultural evolution of cooperation.

The MIT press, Cambridge, Massachusetts. 485 p.

Monnin, T. & Peeters, C. 1999. Dominance hierarchy and reproductive confl icts among subordinates in a monogynous queenless ant. — Behav. Ecol. 10: 323- 332.

Monnin, T. & Ratnieks, F.L.W. 2001: — Policing in queenless ponerine ants. — Behav. Ecol. Sociobiol. 50:

97-108.

Nöe, R. & Hammerstein, P. 1994: Biological markets:

supply and demand determine the ffects of partner choic on cooperation, mutualism and mating. — Anim.

Behav. 39: 78-90.

—. 1995: Biological markets. — Trends Ecol. Evol. 10:

336-339.

Nonacs, P. 1993: Male parentage and sexual deception in the social Hymenoptera. — In: Wrensch, D.L. & Ebbert, M.A. (ed): Evolution and diversity of sex ratio in insects and mites: 384–401. Chapman & Hall. New York.

Nonacs, P. & Carlin, N.F. 1990: When can ants discriminate the sex of brood? A new aspect of queen- worker confl ict. — Proc. Natl. Acad. Sci., U.S.A. 87:

9670–9673

Nowak, M.A. & Sigmund, K. 1998: Evolution of indirect reciprocity by image scoring. — Nature 393: 826-829.

Oberstadt, B. & Heinze, J. 2003: Mating biology and population structure of the ant, Leptothorax gredleri.

— Insect. Soc. 50: 340-345.

Oldroyd, B.P., Halling, L.A., Good, G., Wattanachaiyingcharoen, W., Barron, A.B., Nanork, P.

& Ratnieks, F.L.W. 2001: Worker policing and worker reproduction in Apis cerana. — Behav. Ecol. Sociobiol.

50: 371-377.

Ortius, D. & Heinze, J. 1999: Fertility signaling in queens of a North American ant. — Behav. Ecol. Sociobiol. 45:

151–159.

Oster, G.F. & Wilson, O.E. 1978: Caste and ecology in the social insects. — Princeton university press. Princeton, New Jersey. 352 pp.

Palmer, K.A. & Oldroyd, B.P. 2000: Evolution of multiple mating in the genus Apis. — Apidologie 31: 235-248.

Pamilo, P. 1982: Genetic population structure in polygynous Formica ants. — Heredity 48: 95-106.

—. 1983: Genetic differentiation within subdivided populations of Formica ants. — Evolution 37: 1010- 1022

—. 1991a: Life span of queens in the ant Formica exsecta.

— Ins. Soc. 38: 111-119.

Pamilo, P. 1991b: Evolution of colony characteristics in social insects. I. Sex allocation. — Am. Nat. 137: 83- 107.

Pamilo, P. & Rosengren, R. 1984: Evolution of nesting strategies of ants: genetic evidence from differnet population types of Formica ants. — Biol. J. Linn. Soc.

21: 331-348.

Pamilo, P., Rosengren R., Vepsäläinen, K., Varvio-Aho, S.- L. & Pisarski, B. 1978: Population genetics of Formica ants. I. Patterns of enzyme gene variation. — Hereditas 89: 233-248.

Pamilo, P., Chautems, D. & Cherix, D. 1992: Genetic differentiation of dicjunct populations of the ants Formica aquilonia and Formica lugubris in Europe.

— Insect. Soc. 39: 15-29.

Pamilo, P., Gertsch, P., Thoren, P. & Seppä, P. 1997:

Molecular population genetics of social insects. — Ann.

Rev. Ecol. Syst. 28: 1-25.

Pomiankowski, A. 1999: Intragenomic confl ict. — In:

Keller L. (ed.): Levels of selection in evolution. Princeton university press. Princeton, New Jersey. 529 p.

Price, G.R. 1970: Selection and covariance. — Nature 227:

529-531.

Punttila, P., Haila, Y., Pajunen, T. & Tukia, H. 1991:

Colonisation of clearcut forests by ants in the southern Finnish taiga: a quantitative survey. — Oikos 61:, 250- 262.

Punttila, P., Haila, Y., Niemelä, J. & Pajunen, T. 1994:

Ant communities in fragments of old-growth taiga and managed surroundings. — Annales Zoologici Fennici 31: 131-144.

Queller, D.C. 1994: Genetic relatedness in viscous populations. — Trend Ecol. Evol 8: 70-73.

Queller, D.C. 2000: Relatedness and the fraternal major transitions. — Phil. Trans. R. Soc. Lond. B. 355: 1647-

(20)

1655.

Queller, D.C. & Goodnight, K.F. 1989: Estimating relatedness using genetic markers. — Evolution 43:

258–275.

Queller, D.C., Strassmann, J.E. & Hughes, C.R. 1993:

Microsatellites and kinship. — Trends Ecol. Evol. 8:

285–288.

Queller, D.C. & Strassmann, J.E. 1998: Kin selection and social insects. – BioScience 48: 165–175.

Ratnieks, F.L.W. 1988: Reproductive harmony via mutual policing by workers in eusocial Hymenoptera. — Am.

Nat. 132: 217–236.

Ratnieks, F.L.W. & Visscher, P.K. 1989: Worker policing in the honeybee. — Nature 342: 796–797.

Ratnieks, F. L. W. 1995. Evidence for a queen-produced egg-marking pheromone and its use in worker policing in the honey bee. — Journal of Apicultural Research 34: 31-37.

Ratnieks, F.L.W. & Reeve, H.K. 1992: Confl ict in single- queen Hymenopteran societies: the structure of confl ict and processes that reduce confl ict in advanced eusocial species. — J. Theor. Biol. 158: 33–65

Reeve, H. K. 1991. Polistes. — In: Ross, K.G. & Matthews, R.W. (eds): The Social Biology of Wasps Cornell University Press, Ithaca, New York. pp: 99-148.

Reeve, H,K. 1992: Queen activation of lazy workers in colonies of the eusocial naked mole-rat. — Nature 358:

147-149.

Reeve, H.K & Keller, L. 1997: Reproductive bribing and policing as evolutionary mechanisms for the suppression of within-group selfi shness. — Am. Nat.

150: S42–S55.

Reeve, H.K & Keller, L. 1999: Levels of selection: Burying the units-of-selection debate and unearthing the crucial new issues. — In: Keller L. (ed.): Levels of selection in evolution. Princeton university press. Princeton, New Jersey. 529 p.

Reeve, H.K. & Keller L. 2001: Tests of reproductive-skew models in social insects — Ann. Rev. Entomol. 46: 347- 385

Reeve, J. & Abouheif, E. 2003: Phylogenetic independence 2.0. — Department of Biology, McGill University.

Distributed freely by the authors on request.

Rosengren, R. & Pamilo, P. 1983: The evolution of polygyny and polydomy in mound-building Formica ants. — Acta Entomol. Fenn. 42: 65-77.

Rosengren, R., Sundström, L. & Fortelius, W.: Monogyny and polygyny in Formica ants: the result of alternative dispersal tactics. — In: Keller, L. (ed.), Queen number and sociality in insects. Oxford University Press.

Oxford. 439pp.

Ross, K. G. 1985. Aspects of worker reproduction in 4 social wasp species (Insecta, Hymenoptera, Vespidae).

— J. Zool. 205: 411-424.

Ross, K.G. 1993: The breeding system of the fi re ant Solenopsis invicta: effects of colony genetic structure.

— Am. Nat. 141: 554–576.

—. Molecular ecology of social behaviour: analyses of

breeding systems and genetic structure. —Mol. Ecol.

10: 265-284.

Ross, K.G, & Shoemaker, D.D: 1997: Nuclear and mitochondrial genetic structure in two social forms of the fi re ant Solenopsis invicta: Insights into transitions to an alternative social organization. — Heredity 78:

590-602.

Ross, K.G, Shoemaker, D.D., Krieger, M.J.B., DeHeer, C.J.

& Keller, L. 1999: Assessing genetics structure with multiple classes of molecular markers: a case study involving the introduced fi r ant Solenopsis invicta.

— Mol. Biol. Evol. 16:525-543.

Sachs, J.L., Mueller, U.G., Wilcox, T.P. & Bull, J.J. 2004:

The evolution of cooperation. — Q. Rev. Biol. 79: 135- 160.

Savolainen, R. & Vepsäläinen, K. 1988: A competition hierarchy among boreal ants: impact on resource partitioning and community structure. — Oikos 51:

135-155.

Schlötterer, C. 2004: The Evolution of molecular markers – just a matter of fashion? — Nat. Rev. Gen. 5: 63-69.

Seeley, T.D. 1985: Honeybee ecology. A study of adaptation in social life. — Princeton University Press.

Princeton, New Jersey. 201pp.

—. 1997: Honey bee colonies are group-level adaptive units. — Am. Nat. 150: S22-S41.

Sundström, L. 1993: Genetic population structure and sociogenetic organisation in Formica truncorum (Hymenoptera: Formicidae). — Behav. Ecol. Sociobiol.

33: 345-354.

Sundström, L. 1994: Sex ratio bias, relatedness asymmetry and queen mating frequency in ants. — Nature 367:

266–268.

Sundström, L., Chapuisat, M. & Keller, L. 1996: Conditional manipulation of sex ratios by ant workers: a test of kin selection theory. — Science 274: 993–995.

Sundström, L. & Boomsma, J.J. 2001: Confl icts and alliances in insect families. — Heredity 86: 515-521.

Sundström L., Keller L. & Chapuisat, M. 2003: Differential dispersal patterns in males and females of the ant Formica exsecta. — Evolution 57:1552-1561

Taylor, P. D. Altruism in viscous populations – an inclusive fi tness model. — Evol. Ecol. 6: 352-356.

Taylor, P.D. & Frank, S.A. 1996: How to make a kin selection model. — J. Theor. Biol. 180: 27-37.

Toth, E., Queller. D.C., Dollin, A. & Strassmann, J.E. 2004:

Confl ict over male parentage in stingless bees. —Ins.

Soc. 51: 1-11.

Trivers, R.L. 1971: the evolution of reciprocal altruism.

— Q. Rev. Biol. 46: 35-57.

Trivers, R.L. & Hare, H. 1976: Haploidiploidy and the evolution of social insects. — Science 191: 249–263.

Vepsäläinen, K. & Pisarski, B. 1982: Assembly of island ant communities. — Annales Zoologici Fennici 19:

327-335.

Vepsäläinen, K., Savolainen, R., Tiainen, J. & Vilén, J.

2000: Successional changes of ant assemplages: from virgin and ditched bogs to forests. — Annales Zoologici

(21)

22

Fennici 37: 135-149.

Visscher, P.K. 1996: Reproductive confl ict in honey bees: a stalemate of worker egg-laying and policing.

Visscher, P.K. & Dukas, R. 1995: Honey bees recognize development of nestmates’ ovaries. — Anim. Behav.

49: 542–544.

Walin, L., Sunsdström, L., Seppä, P. & Rosengren, R. 1998:

Worker reproduction in ants — a genetic analysis. — Heredity 81: 604–612.

Wenseleers, T., Ratnieks, F.L.W. & Billen, J. 2003: Caste fate confl ict in swarm-founding social Hymenoptera:

an inclusive fi tness analysis. — Journal of Evolutionary Biology 16:647-658.

West, S.A., Pen, I. & Griffi n, A.S. 2002: Cooperation and competition between relatives. — Science 296: 72-75.

Williams, G.C. 1966: Adaptation and natural selection:

a critique on some current evolutionary thought. — Princeton University press, Princeton, NJ. 308 pp.

Wilson, E.O. 1971: The insect societies. — Harvard University press, Cambridge, Massachusetts.

Wynne-Edwards V.C. 1962: Animal dispersion in relation to social behaviour. — Edinburgh & London, Oliver &

Boyd.

Zhu, D., Chapuisat, M. & Pamilo, P. 2003: Highly variable social organization of colonies in the ant Formica cinerea. — Hereditas 139: 7-12.