Behavioral Ecology Advance Access originally published online on April 28, 2006
Behavioral Ecology 2006 17(4):622-627; doi:10.1093/beheco/ark006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Developmental divergence: neglected variable in understanding the evolution of reproductive skew in social animals
a Behaviour and Genetics of Social Insects Laboratory, School of Biological Sciences, A12, University of Sydney, Sydney, NSW 2006, Australia, b Laboratoire d'Ecologie, UMR CNRS 7625, Université Pierre et Marie Curie, 7 quai Saint Bernard, 75005 Paris, France, and c Department of Zoology, University of Cape Town, Rondebosch 7701, South Africa
Address correspondence to M. Beekman. E-mail: mbeekman{at}bio.usyd.edu.au.
Received 25 May 2005; revised 12 February 2006; accepted 30 March 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
Unequal reproduction is widespread in animal societies. Over the last decade, there have been several attempts to derive a conceptual framework that allows diverse taxa of invertebrates and vertebrates to be compared with the aim of identifying common causes underlying reproductive division of labor. Sherman et al. (1995) used reproductive skew values, derived from variation in lifetime reproductive success, to classify taxa. Crespi and Yanega (1995), on the other hand, focused on the loss of "totipotency"––the potential throughout a lifetime to exhibit the full behavioral repertoire of the species––and argued that comparisons can only be made between taxa with and without totipotency among group members. Here we argue that the effect of irreversible morphological differences on the variation in lifetime reproductive success of breeders and helpers has been neglected. Such phenotypic specialization originates during development and is not equivalent to behavioral differentiation among adults. We classify social animals into 3 developmental pathways to examine the ontogenetic stage at which irreversible morphological differences arise. This allows for heuristic comparisons of reproductive skew across taxa; it is more useful to compare skew in termites and bees without morphologically specialized breeders and helpers to cooperative breeding birds and mammals than it is to compare them with other insects with morphological castes. Once physical castes have evolved, natural selection will lead to further morphological specialization for increased efficiency at both helping and breeding (e.g., physogastry in insect queens and complete sterility of some social insect workers). Importantly, extreme reproductive skew can be achieved with and without morphologically specialized helpers and breeders, indicating that grouping societies based on their reproductive skew value alone is not useful.
Key words: caste, development, morphology, physogastry, reproductive skew, social insects, social vertebrates.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
In many social animals, reproduction is not shared equally among group members but is monopolized by a minority. Societies that exhibit reproductive skew are found across a wide range of taxa including vertebrates, such as wild dogs, meerkats, white-fronted bee eaters, and invertebrates, such as snapping shrimps, termites, all ants, and some bees and wasps. Given that many highly diverse taxa have evolved reproductive skew, several authors have hypothesized that such skew has been reached via similar mechanisms (see, e.g., Alexander 1974
; Andersson 1984; Gadagkar 1990; Alexander et al. 1991; Crespi 1994; Crespi and Choe 1997). However, a comparative study of the causes of reproductive skew has long been hindered by the lack of a "one size fits all" conceptual framework. A decade ago, Crespi and Yanega (1995) and Sherman et al. (1995) attempted to fill this hiatus, urging that common features underlying reproductive skew should be considered across all social taxa. Both papers contributed to our understanding of similarities and differences among animal taxa, although their approaches differed fundamentally (reviewed in Costa and Fitzgerald 1996). Sherman et al. (1995) introduced the "eusociality continuum" that, in theory, enables all social taxa to be compared using one measure of reproductive skew, that is, the variation in lifetime reproductive success among group members. In contrast, Crespi and Yanega (1995) focused on the loss of totipotency––the potential throughout a lifetime to exhibit the full behavioral repertoire of the species––in a proportion of group members.
We argue that the above papers, as well as others that aimed to provide a framework for identifying the causes of reproductive division of labor in social animals (Gadagkar 1994; Wcislo 1997; Helms Cahan et al. 2002) have neglected a key variable, namely developmental divergence. Ontogeny needs to be appreciated in particular when distinct developmental pathways lead to irreversible morphological divergence between breeders and helpers. Indeed, such irreversibility affects the scope of natural selection on individuals (e.g., Bourke 1999). When helpers have permanently lost the option to breed, natural selection will favor those individuals best adapted to increase their inclusive fitness via nonreproductive activities. Conversely, when helpers are behaviorally specialized but not morphologically specialized, that is, they still have the possibility to reproduce later in life, selection will not favor adaptations aimed solely at increasing inclusive fitness via helping. Similarly, morphologically specialized breeders that cannot perform the helper role will be selected for increased fecundity.
Development has a crucial effect on the evolution of irreversible morphological specialization. Growth is a continuous process in vertebrates, whereas in all invertebrates, it is punctuated by molting. Importantly, arthropods are enclosed in a solid cuticle that prevents any growth until the next molt. In holometabolous insects (such as ants, bees, and wasps), metamorphosis is complete––immatures (larvae) are legless grubs that undergo a final molt (pupation) to produce allometrically distinct adults. Once the adult has emerged, further morphological change is impossible. Thus, any differences in cuticular morphology among holometabolous adults must result from variations in the developmental process during the immature stages (O'Donnell 1998). In contrast, hemimetabolous insects (which include the termites) exhibit incomplete metamorphosis––older immatures already look like "adults" except for being wingless, and they grow larger with each molt. Thus, despite development being punctuated by molting, the developmental process in termites is more similar to that of vertebrates than to holometabolous insects' because the ontogenetic changes are gradual. In both holometabolous and hemimetabolous insects, further morphological change is impossible once the adult has emerged, whereas adult morphology changes continuously in vertebrates.
In this paper, we first consider the details of the different approaches adopted by Sherman et al. (1995) and Crespi and Yanega (1995) and discuss their relative merits. We then illustrate how the existence of irreversible morphological divergence is not equivalent to behavioral differentiation. Even though the occurrence of morphological castes is mainly restricted to the social insects, the points that we make are equally important to researchers working on social vertebrates. Finally, we divide social animals into 3 groups based on their developmental pathway and show that high reproductive skew can occur without irreversibility.
|
THE EUSOCIALITY CONTINUUM VERSUS LOSS OF TOTIPOTENCY |
|---|
|
TOPABSTRACTINTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
Sherman et al. (1995)
ignored morphological divergence among group members and instead argued that differences in the distribution of lifetime reproductive success serve as a reliable and biologically relevant measure for comparing species that exhibit unequal reproduction. Using this eusociality continuum, animals with a similar level of reproductive skew are grouped together, no matter how they differ in other aspects of their general biology. In their view, the grouping of disparate animals based on their variation in lifetime reproductive success reveals cross-taxonomic similarities of the causes and consequences of reproductive partitioning.
Although reproductive skew is the most evolutionarily intriguing aspect of animal sociality, we question the use of skew values as the sole defining characteristic of a social species. Apart from the fact that measuring an individual's variation in lifetime reproductive success remains an elusive objective in most species, there are likely to be many conditions unrelated to social interactions that result in variation in reproductive success among group members (Crespi and Yanega 1995; Costa and Fitzgerald 1996). But in our view, the most important shortcoming of the eusociality continuum is that a society with a single reproductive individual can result from 2 dramatically different processes: behavioral regulation or morphological specialization. Compare the females of the queenless ponerine ant Dinoponera quadriceps and the fungus-growing ant Atta columbica. The colonies of both species have only one breeder, and using the continuum concept of Sherman et al. (1995), they both approximate a skew value of 1. However, the 2 species show crucial differences in their social organization. Dinoponera quadriceps has colonies of less than 100 workers, and aggressive interactions determine which worker mates and reproduces (Monnin and Peeters 1999). Every adult worker can, in theory, mate and replace the dominant breeder. In sharp contrast, the queen in A. columbica is a highly fecund egg layer who starts a new colony on her own and remains the sole breeder for up to 20 years, producing millions of adult workers who can never mate (Hölldobler and Wilson 1990). The mechanisms leading to the evolution of similar reproductive skew values in D. quadriceps and A. columbica are remarkably different, and yet according to the continuum of Sherman et al. (1995), they are lumped together.
Contrary to Sherman et al. (1995), Crespi and Yanega (1995), following Wilson (1971), divided all social systems into either "cooperative breeders" or "eusocial," depending on whether a proportion of group members lose their totipotency, that is, become fixed as helpers or breeders, at some point prior to "reproductive maturity" (our italics), such that transitions between helper and breeder phenotypes cannot occur. This approach makes the fundamental point that morphological specialization into helpers and breeders affects the scope of natural selection on individuals. We agree with the insightful classification of Crespi and Yanega (1995) but argue that it is insufficient for 2 reasons. Firstly, the notion of reproductive maturity is inadequate for definitions that aim to encompass both invertebrates and vertebrates—in arthropods, only the final molts produce individuals able to breed, whereas in birds and mammals, there is no fixed point at which individuals reach sexual maturity. Secondly, Crespi and Yanega (1995) failed to distinguish between behavioral and morphological divergence (we restrict the latter to describe allometric changes, i.e., differences in body shape, not just size). Indeed, they defined castes as "groups of individuals that become irreversibly behaviorally distinct ...." Insects that have morphologically divergent adults to perform the helper (workers) and breeding (queens) roles satisfy the criterion of irreversibility, unlike species (including many insects) in which reproductive skew results from behavioral differences due to age, mating, dominance rank, or ecological constraints. In the latter case, behavioral divergence is not truly irreversible as helpers have not necessarily lost the ability to reproduce later in life if the opportunity arises. In our view, it is essential to distinguish between irreversible and reversible divergence because loss of totipotency only results from morphological divergence.
|
MORPHOLOGICAL DIVERGENCE AND THE EVOLUTION OF QUEENS |
|---|
|
TOPABSTRACTINTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
In termites, most ants, and some bees and wasps, breeders and helpers correspond to 2 castes that can be distinguished by discontinuous, allometric differences in morphology, that is, body shape and internal organs. Ant queens and workers are the most spectacular example of morphological castes because workers never develop wings (termite workers also lack wings, but in some species they retain the ability to develop them if they undergo their final molt). However, the terms "queen" and "caste" are used inconsistently in the social insect literature, with many authors oscillating between functional and morphological meanings (see Peeters and Crozier 1988
).
In Ropalidia wasps, all adult females are morphologically similar but are called queens if they reproduce (Gadagkar 2001). O'Donnell (1998) used insemination as the criterion for distinguishing queens and workers. Because the production of offspring and insemination are common to all sexually reproducing animals, such criteria lack heuristic value, and in our opinion, these inconsistencies in terminology obfuscate cross-taxonomic comparisons and trivialize the significance of the evolution of queen and worker castes. Moreover, researchers working on social vertebrates are likely to use different terminology to those studying the same phenomenon in social insects. Hence, we advocate that the term "queen" be restricted to breeders that are allometrically distinct from helpers (Wheeler 1986; Peeters and Crozier 1988) because the evolution of castes represents a key point in the elaboration of sociality as it leads to differential selection pressures on queens and workers (Bourke 1999). Even though Crespi and Yanega (1995) and Crespi and Choe (1997) already mentioned differential selection pressures on specialized helpers and breeders, they did not distinguish between irreversible morphological and reversible behavioral differentiation.
|
THREE DEVELOPMENTAL PATHWAYS TO BECOMING A BREEDER |
|---|
|
TOPABSTRACTINTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
Indeterminate growth
In social birds and mammals, growth is indeterminate and thus characterized by a clear inflection point and subsequent asymptote. Individuals become reproductively mature prior to the cessation of growth, but typically they reproduce only years later. Older individuals in mammals are often larger and more dominant than younger individuals, and aggressive interactions lead to a dominance hierarchy in which high-ranking (older, larger) individuals monopolize reproductive opportunities (e.g., meerkats [Clutton-Brock et al. 2001
]; dwarf mongoose [Creel et al. 1992]; wild dog [Creel et al. 1997]; naked mole-rats [Margulis et al. 1995]; and savannah baboons [Altmann et al. 1988]). In birds, it is not so much size that matters but experience, lack of mates, or access to a territory (Emlen 1991; Koenig and Dickinson 2004). Helpers learn by assisting a dominant pair for one or more breeding seasons that might increase their chances of breeding independently in later years. Moreover, helpers have a chance to inherit the territory when the same-sex dominant dies (Cockburn 1998).
Importantly, the morphologies of vertebrate helpers and breeders vary continuously and are not associated with allometric differences. When all individuals remain totipotent and growth is indeterminate, selection does not favor the evolution of a sterile helper caste as the probability of a given group member breeding is sufficiently high to have favored traits that maximize direct fitness (a similar point was made by Alexander et al. 1991). Thus, most adult vertebrates are totipotent (although menopause is arguably a mammalian exception), lack castes, and rarely attain a high skew value.
Incomplete metamorphosis
In termites, the immatures are able to work for the colony from an early age. Here, helpers and breeders are separate developmental stages, and in a proportion of species, helpers retain the ability to molt and metamorphose into winged sexuals (Noirot and Pasteels 1987; Roisin 2000). These sexuals are the final molts and are morphologically specialized for aerial dispersal and the foundation of new colonies. Interestingly, in other species, death of a founding queen triggers molting by some immatures who become reproductives, although they do not grow wings or disperse (neotenics). The termites (±2600 species) exhibit great diversity in the degree of irreversibility of the workers' developmental path, according to whether this differentiation occurs at an early or late instar. In Kalotermitidae and Termopsidae, colonies cease to exist once they have eaten their wooden nest, whereon all helpers will molt into sexuals and disperse (Shellman-Reeve 1997). In other termites that have more perennial colonies, late instar immatures cannot metamorphose into sexuals and therefore form an irreversible worker caste. The social structure of the latter species is more comparable to social Hymenoptera with castes as the termite helpers no longer have the option to breed. In contrast, termites with "reversible" workers are similar to social vertebrates having helpers at the nest as each individual retains the option to become a breeder later in life. Hence, helpers in these taxa will not be selected to lose totipotency as they should not forgo the option to breed themselves.
Complete metamorphosis
In all social Hymenoptera, the immatures are legless larvae that are completely dependent on the adults and cannot contribute to colonial labor. Metamorphosis produces adults that will either breed or work, or do both, because the existence of queen and worker castes is far from universal among these holometabolous insects.
In many social bees and wasps, female adults vary in size but exhibit no allometric differences, and reproductive division of labor is not determined by morphology but by extrinsic (e.g., mating opportunities) or intrinsic (e.g., relative fighting ability) factors. Here, disparities in reproductive ability (e.g., the amount of metabolic reserves) are typically established during larval development, but all adults can become breeders (O'Donnell 1998). In halictid bees, females differ in size but not in allometry and are all capable of mating, although females that emerge outside the period of male activity will remain virgin and become helpers (Yanega 1992). Females born when males are active leave the maternal nest, mate, and start breeding on their own. Social Hymenoptera that lack queen and worker castes are therefore more similar to social vertebrates, relative to social Hymenoptera that have queen and worker castes. The latter insects exhibit rigid organization because workers and queens are completely interdependent, except during the short period when queens found a new colony independently. Because the presence of irreversible morphological castes results in divergent selection pressures (helpers will be selected to increase their inclusive fitness and breeders to increase their direct fitness), cross-taxonomic comparisons that ignore developmental divergence are unlikely to yield insights into common causes of reproductive division of labor.
Spectacular fecundity: postbreeding morphological specialization
The evolution of morphological castes seems a pivotal factor in the attainment of large, complex societies among insects (Bourke 1999; Jeon and Choe 2003). Not surprisingly, theories attempting to explain the origin and maintenance of morphological castes were based on aspects of insect biology such as metamorphosis, short generation time, small biomass, and extraordinary fecundity. However, the discovery of queens in one vertebrate, the naked mole-rat (O'Riain et al. 2000), sheds new light on the evolution of castes.
Just how similar are the queens of insects with those of the naked mole-rat? In the latter, a female that was previously reproductively inactive undergoes a remarkable transformation after the onset of reproduction. Her lumbar vertebrae begin to lengthen with the result that her abdomen increases significantly in volume. The adaptive significance of this is that the breeder is capable of carrying more fetuses in utero and litter size increases from a mean of 7 to a maximum of 28 pups. These physically transformed breeders can justifiably be referred to as queens because the changes are irreversible, but their developmental pathway is distinct from that of queens in social Hymenoptera. The latter become queens as a result of being reared differentially, whereas the former fight to the bitter end with siblings, and if successful, the vertebral column will start to lengthen with her first pregnancy. Breeders that have not yet had offspring are morphologically indistinguishable from nonbreeding females (O'Riain et al. 2000).
The postbreeding morphological specialization in naked mole-rats has parallels in the physogastric queens of higher termites, army ants, and some stingless bees. In these spectacularly fecund females, the soft intersegmental membranes between the cuticular plates of the abdomen stretch to accommodate the active ovaries. In Macrotermes termites, an established queen has an abdomen about 500 times larger than that of a young, newly fertile queen (Bordereau 1982). Similarly, in army ants of the genus Eciton, physogastry in the queens allows the simultaneous activity of thousands of ovarioles (Hagan 1954; Whelden 1963). In both cases, physogastry is an adaptation for the production of very populous colonies (up to a million nest mates). Physogastry is clearly comparable to the elongation of the vertebral column documented in the naked mole-rat because both are consequences of successful breeding, not developmental divergence. However, it is important to recognize the different origins of physogatric queens in insects and naked mole-rats as the proximate mechanisms are fundamentally different despite strong similarities.
|
INTEGRATING DEVELOPMENTAL DIVERGENCE WITH REPRODUCTIVE SKEW |
|---|
|
TOPABSTRACTINTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
We have focused on the effect of developmental divergence on reproductive skew as this has not always been given due emphasis (although Alexander et al. [1991]
did mention the importance of "gradual metamorphosis" for the evolution of sociality). Because the evolution of irreversible morphological specialization dictates the reproductive options of adult individuals, comparisons of reproductive skew across social taxa should consider developmental divergence. This approach, illustrated in Figure 1 in which we used most of the example taxa from Figure 1 of Sherman et al. (1995), shows that we find high skew in all developmental pathways, including taxa without morphological castes. Hence, loss of totipotency is not always selected for, even in taxa that share the same developmental pathway. The dominant female in halictid bees and the honeybee queen may both monopolize reproduction and undergo complete metamorphosis, but only the honeybees evolved castes whereas halictid bees did not. It is therefore more informative to compare halictid bees with cooperative breeding birds than it is to compare halictid bees with honeybees when attempting to understand the causes underlying high skew in both species. Similarly, all the termites exhibit high skew, but irreversible workers have evolved in some species only. This leads to the conclusion that reproductive skew and the evolution of irreversible morphological differentiation are 2 different evolutionary processes.
|
The fact that skew can evolve with and without irreversible divergence indicates that other factors are also likely to determine the extent of reproductive division of labor in social animals, and in our conceptual framework, idiosyncrasies become obvious. For example, dampwood termites and vertebrates such as African wild dogs and dwarf mongooses have similar levels of skew, and all individuals are totipotent (Figure 1). However, there is a large disparity in longevity between breeders and helpers in dampwood termites compared with the vertebrates. In termites, reproductive individuals outlive helpers by orders of magnitude, whereas such differences are negligible in most social vertebrates (naked mole-rats being the exception [Sherman and Jarvis 2002
]). Differences in longevity between helpers and breeders, together with much larger colonies in termites than in vertebrates, have dramatic consequences for the probability that a given helper can fill a breeding vacancy (Bourke 1999) and hence greatly affect lifetime variation in reproductive skew.
The evolution of queen and worker castes represents an evolutionary threshold, as previously recognized by Crespi and Yanega (1995). This is because natural selection becomes circumscribed to a specific, limited range of behaviors. This, then, opens the possibility of even greater caste divergence with the size of workers often being reduced and ovaries or sperm reservoirs lost. Queens, in contrast, will be selected for higher reproductive potential (e.g., physogastry) and are often so well adapted to their role that they can no longer exist without workers (e.g., wingless "ergatoid" queens in ants, Peeters and Ito 2001).
Searching for a unifying framework that encompasses all taxa exhibiting reproductive skew should not be at the expense of essential biological details. Beekman and Ratnieks (2003) and Beekman et al. (2003) argued that a better understanding of the diversity of reproductive characteristics of animal societies can only come from combining inclusive fitness theory with a thorough understanding of the biology of the taxa under study. A similar appreciation of biological details is, in our view, essential for an understanding of the causes of reproductive skew in social animals. It is only through this approach that we can understand why in some social taxa that share the same developmental pathway and reproductive skew totipotency has not always been lost.
|
ACKNOWLEDGEMENTS |
|---|
We would like to thank Laurent Keller, Bernie Crespi, Mark Elgar, and members of the laboratories in Sydney and Paris for useful comments and discussions. The ideas for this paper were, in part, conceived during C.P.'s stay at Sydney, financed by Direction des Relations Internationales of CNRS. M.B. is supported by the Australian Research Council. M.J.O. is funded by the National Research Foundation and University Research Corporation.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONTHE EUSOCIALITY CONTINUUM VERSUS...MORPHOLOGICAL DIVERGENCE AND THE...THREE DEVELOPMENTAL PATHWAYS TO...INTEGRATING DEVELOPMENTAL...REFERENCES |
|---|
Alexander RD. 1974. The evolution of social behavior. Annu Rev Ecol Syst 5:325–83.[CrossRef]
Alexander RD, Noonan KM, Crespi BJ. 1991. The evolution of eusociality. In: Sherman PW, Jarvis JUM, Alexander RD, editors. The biology of the naked mole rat. Princeton, NJ: Princeton University Press. p 1–44.
Altmann J, Hausfater G, Altman S. 1988. Determinants of reproductive success in savannah baboons, Papio cynocephalus. In: Clutton-Brock TH, editor. Reproductive success: studies of individual variation in contrasting breeding systems. Chicago, IL: The University of Chicago Press. p 403–18.
Andersson M. 1984. The evolution of eusociality. Annu Rev Ecol Syst 15:165–89.[Web of Science]
Beekman M, Komdeur J, Ratnieks FLW. 2003. Reproductive conflicts in social animals: who has power? Trends Ecol Evol 18:277–82.[CrossRef]
Beekman M, Ratnieks FLW. 2003. Power over reproduction in social Hymenoptera. Philos Trans R Soc Lond B Biol Sci 358:1741–54.
Bordereau C. 1982. Ultrastructure and formation of the physogastric termite queen cuticle. Tissue Cell 14:371–96.[CrossRef][Web of Science][Medline]
Bourke AFG. 1999. Colony size, social complexity and reproductive conflict in social insects. J Evol Biol 12:245–57.[CrossRef][Web of Science]
Clutton-Brock TH, Brotherton PNM, Russell AF, O'Riain MJ, Gaynor D, Kansky R, Griffin A, Manser M, Sharpe L, McIlrath GM, Small T, Moss A, Monfort S. 2001. Cooperation, control, and concession in meerkat groups. Science 291:478–81.
Cockburn A. 1998. Evolution of helping behavior in cooperatively breeding birds. Annu Rev Ecol Syst 29:141–77.[CrossRef][Web of Science]
Costa JT, Fitzgerald TD. 1996. Developments in social terminology: semantic battles in a conceptual war. Trends Ecol Evol 11:285–9.
Creel S, Creel N, Mills M, Monfort S. 1997. Rank and reproduction in cooperatively breeding African wild dogs: behavioral and endocrine correlates. Behav Ecol 8:198–306.
Creel S, Creel N, Wildt DE, Monfort SL. 1992. Behavioural and endocrine mechanisms of reproductive suppression in Serengeti dwarf mongooses. Anim Behav 43:231–45.
Crespi BJ. 1994. Three conditions for the evolution of eusociality: are they sufficient? Insectes Soc 41:395–400.[CrossRef]
Crespi BJ, Choe JC. 1997. Explanation and evolution of social systems. In: Choe JC, Crespi BJ, editors. The evolution of social behavior in insects and arachnids. Cambridge, UK: Cambridge University Press. p 499–524.
Crespi BJ, Yanega D. 1995. The definition of eusociality. Behav Ecol 6:109–15.
Emlen ST. 1991. Evolution of cooperative breeding in birds and mammals. In: Krebs JR, Davies NB, editors. Behavioural ecology. An evolutionary approach. 3rd ed. Oxford: Blackwell. p 301–37.
Gadagkar R. 1990. Evolution of eusociality: the advantage of assured fitness returns. Philos Trans R Soc Lond B Biol Sci 329:17–25.[CrossRef]
Gadagkar R. 1994. Why the definition of eusociality is not helpful to understand its evolution and what should we do about it. Oikos 70:485–8.[CrossRef][Web of Science]
Gadagkar R. 2001. The social biology of Ropalidia marginata: toward understanding the evolution of eusociality. Cambridge, MA: Harvard University Press.
Hagan HR. 1954. The reproductive system of the army-ant queen, Eciton (Eciton) part I. General anatomy. Am Mus Novit 1663:1–8.
Helms Cahan S, Blumstein DT, Sundström L, Liebig J, Griffin A. 2002. Social trajectories and the evolution of social behavior. Oikos 96:206–16.[CrossRef][Web of Science]
Hölldobler B, Wilson EO. 1990. The ants. Cambridge, MA: Belknap Press.
Jeon J, Choe JC. 2003. Reproductive skew and the origin of sterile castes. Am Nat 161:206–24.[CrossRef][Medline]
Koenig WD, Dickinson JL. 2004. Ecology and evolution of cooperative breeding in birds. Cambridge, UK: Cambridge University Press. 293 p.
Margulis S, Saltzman W, Abbott DH. 1995. Behavioural and hormonal changes in female naked mole-rats (Heterocephalus glaber) following removal of the breeding female from a colony. Horm Behav 29:227–47.[CrossRef][Medline]
Monnin T, Peeters C. 1999. Dominance hierarchy and reproductive conflicts among subordinates in a monogynous queenless ant. Behav Ecol 10:323–32.
Noirot C, Pasteels JM. 1987. Ontogenetic development and evolution of the worker caste in termites. Experientia 43:851–60.[CrossRef][Web of Science]
O'Donnell S. 1998. Reproductive caste determination in eusocial wasps (Hymenoptera: Vespidae). Annu Rev Entomol 43:323–46.[CrossRef][Web of Science][Medline]
O'Riain JM, Jarvis JUM, Alexander R, Buffenstein R, Peeters C. 2000. Morphological castes in a vertebrate. Proc Natl Acad Sci U S A 97:13194–7.
Peeters C, Crozier RH. 1988. Caste and reproduction in ants: not all mated egg-layers are ‘queens’. Psyche 95:283–8.[CrossRef]
Peeters C, Ito F. 2001. Colony dispersal and the evolution of queen morphology in social Hymenoptera. Annu Rev Entomol 46:601–30.[CrossRef][Web of Science][Medline]
Roisin Y. 2000. Diversity and evolution of caste patterns. In: Abe T, Bignell DE, Higashi M, editors. Termites: evolution, sociality, symbioses, ecology. Dordrecht, The Netherlands: Kluwer Academic. p 95–119.
Shellman-Reeve JS. 1997. The spectrum of eusociality in termites. In: Choe JC, Crespi BJ, editors. The evolution of social behavior in insects and arachnids. Cambridge, UK: Cambridge University Press. p 52–93.
Sherman PW, Jarvis JUM. 2002. Extraordinary life spans of naked mole-rats (Heterocephalus glaber). J Zool (Lond) 258:307–11.[CrossRef]
Sherman PW, Lacey EA, Reeve HK, Keller L. 1995. The eusociality continuum. Behav Ecol 6:102–8.
Wcislo WT. 1997. Are behavioral classifications blinders to studying natural variation? In: Crespi BJ, Choe JC, editors. The evolution of social behavior in insects and arachnids. Cambridge, UK: Cambridge University Press. p 8–13.
Wheeler DE. 1986. Developmental and physiological determinants of caste in social hymenoptera: evolutionary implications. Am Nat 128:13–34.
Whelden RM. 1963. The anatomy of the adult queen and workers of the army ants Eciton burchelli Westwood and Eciton hamatum Fabricius. J N Y Entomol Soc 71:158–78.
Wilson EO. 1971. The insect societies. Cambridge, MA: Belknap Press of Harvard University Press.
Yanega D. 1992. Does mating determine caste in sweat bees? (Hymenoptera: Halictidae). J Kans Entomol Soc 65:231–7.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  W. O. H. Hughes, B. P. Oldroyd, M. Beekman, and F. L. W. Ratnieks Ancestral Monogamy Shows Kin Selection Is Key to the Evolution of Eusociality Science, May 30, 2008; 320(5880): 1213 - 1216. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||