One of the central, unresolved controversies in biology concerns the distribution of primitive versus advanced characters at different stages of vertebrate development. This controversy has major implications for evolutionary developmental biology and phylogenetics. Ernst Haeckel addressed the issue with his Biogenetic Law, and his embryo drawings functioned as supporting data. We re-examine Haeckel's work and its significance for modern efforts to develop a rigorous comparative framework for developmental studies. Haeckel's comparative embryology was evolutionary but non-quantitative. It was based on developmental sequences, and treated heterochrony as a sequence change. It is not always clear whether he believed in recapitulation of single characters or entire stages. The Biogenetic Law is supported by several recent studies – if applied to single characters only. Haeckel's important but overlooked alphabetical analogy of evolution and development is an advance on von Baer. Haeckel recognized the evolutionary diversity in early embryonic stages, in line with modern thinking. He did not necessarily advocate the strict form of recapitulation and terminal addition commonly attributed to him. Haeckel's much-criticized embryo drawings are important as phylogenetic hypotheses, teaching aids, and evidence for evolution. While some criticisms of the drawings are legitimate, others are more tendentious. In opposition to Haeckel and his embryo drawings, Wilhelm His made major advances towards developing a quantitative comparative embryology based on morphometrics. Unfortunately His's work in this area is largely forgotten. Despite his obvious flaws, Haeckel can be seen as the father of a sequence-based phylogenetic embryology.

In the 1790s, Robert Townson established the main features of the water economy of terrestrial amphibians: rapid evaporative water loss in dry surroundings, ‘drinking’ by absorption of water through the abdominal skin pressed against moist substrates, and use of the urinary bladder as a reservoir from which water is reabsorbed on land. This knowledge was of little interest to the establishment in the first half of the nineteenth century of experimental physiology as a basic medical discipline, when frogs became models in the elucidation of general physiological processes. Townson's pioneer contributions to amphibian physiology were forgotten for 200 years (Jørgensen 1994b). Durig (1901) and particularly Overton (1904) restored knowledge about amphibian water economy to the level reached by Townson, but the papers had little impact on the young science of animal physiology because they primarily aimed at elucidating the transport of fluids across membranes. Frog skin remained a model preparation in such studies throughout the century. With the establishment of terrestrial ecology early in the century, the relations of animals, including amphibians, to water became a central theme. Concurrently with comparative studies of amphibian water economy in an ecological setting, the subject proceeded as an aspect of animal osmoregulation. Adolph (1920–1930) and Rey (1937a) established the highly dynamic nature of water balance in amphibians in water and on land. Their observations indicated functional links between environment, skin and kidneys, the nature of which remained to be explored. Thorson & Svihla (1943) reopened the ecological approach in a comparative study of the relations between amphibian habitat and tolerance of dehydration. By mid-century, the central themes of amphibian adaptations to terrestrial modes of life were re-established, except for the function of the bladder as a water-depot. During the following decades, a rich literature appeared, particularly focusing on adaptations of amphibians to arid environments. Thus, in the 1970s, it was found that ‘waterproofing’ of the highly permeable skins by means of skin secretions had evolved independently in several families of tropical arboreal frogs, and that a number of amphibians that aestivate whilst burrowed in dry soil could reduce evaporation by forming cocoons from shed strata cornea. In 1950–1970 the role of bladder urine as a water depot in terrestrial amphibians was recognized: this did not change the established view of water balance in terrestrial amphibians as alternating between dehydration on land and rehydration in response to the deficit in body water. Amphibians may, however, maintain normal water balance whether the ambient medium is water or air by means of little understood integrated mechanisms in control of cutaneous drinking behaviour, water permeability of the skin and bladder wall, and urine production.

Theory concerning the evolution of life history (the schedule of reproduction and survival) focuses on describing the life history which maximises fitness. Although there is an intuitive link between life history and fitness, there are in fact several measures of the ‘black box’ concept of fitness. There has been a debate in the bio-mathematical literature on the predictive difference between the two most commonly used measures; intrinsic rate of increase r and net reproductive ratio R0. Although both measures aim to describe fitness, models using one of the measures may predict the opposite of similar models using the other measure, which is clearly undesirable. Here, I review the evolution of these fitness measures over the last four decades, the predictive differences between these measures and the resulting shift of the fitness concept. I focus in particular on some recent developments, which have solved the dilemma of predictive differences between these fitness measures by explicitly acknowledging the game-theoretical nature of life-history evolution.

The ecosystem approach to fisheries recognises the interdependence between harvested species and other ecosystem components. It aims to account for the propagation of the effects of harvesting through the food-web. The formulation and evaluation of ecosystem-based management strategies requires reliable models of ecosystem dynamics to predict these effects. The krill-based system in the Southern Ocean was the focus of some of the earliest models exploring such effects. It is also a suitable example for the development of models to support the ecosystem approach to fisheries because it has a relatively simple food-web structure and progress has been made in developing models of the key species and interactions, some of which has been motivated by the need to develop ecosystem-based management. Antarctic krill, Euphausia superba, is the main target species for the fishery and the main prey of many top predators. It is therefore critical to capture the processes affecting the dynamics and distribution of krill in ecosystem dynamics models. These processes include environmental influences on recruitment and the spatially variable influence of advection. Models must also capture the interactions between krill and its consumers, which are mediated by the spatial structure of the environment. Various models have explored predator-prey population dynamics with simplistic representations of these interactions, while others have focused on specific details of the interactions. There is now a pressing need to develop plausible and practical models of ecosystem dynamics that link processes occurring at these different scales. Many studies have highlighted uncertainties in our understanding of the system, which indicates future priorities in terms of both data collection and developing methods to evaluate the effects of these uncertainties on model predictions. We propose a modelling approach that focuses on harvested species and their monitored consumers and that evaluates model uncertainty by using alternative structures and functional forms in a Monte Carlo framework.

Owing to the close phylogenetic relationship of Platyrrhini (New World monkeys) and Catarrhini (Old World monkeys) to man, nonhuman primates are often used as models for the study of male reproductive physiology and endocrinology. This review aims at providing new data and insights into comparative primate spermatogenesis, dealing specifically with quantitative aspects of germinal epithelial organisation and germ cell production, and with the roles of gonadotrophic hormones in this process.

Typically, the seminiferous epithelium is composed of specific germ cell associations (spermatogenic stages). In rodents, prosimians and most Catarrhini, tubular cross sections contain a single spermatogenic stage whereas in Platyrrhini, great apes and man multi-stage tubules are present. Since Platyrrhini represent a more basal type of primate, this spermatogenic feature must have developed convergently. The primate multi-stage tubular arrangement was previously believed to be associated with low spermatogenic efficiency. However, recent studies using new methodological approaches and comparing primate species from all taxa have revealed that multi-stage organisation is compatible with highly efficient spermatogenesis. In fact, meta-analysis demonstrated that the efficiency of spermatogenesis in several nonhuman primate species is comparable to that of rodents which are considered as species with highly efficient germ cell production. The duration of the spermatogenic process was not related to organisation or efficiency of spermatogenesis. Sertoli cell work load was species-specific but had no impact on germ cell numbers and on the efficiency of spermatogenesis.

The gonadotrophic hormones, luteinizing hormone (LH) and follicle stimulating hormone (FSH) are the primary regulators of primate testicular function. Recent studies revealed that in New World monkeys chorionic gonadotrophin (CG) – the primate pregnancy hormone – regulates testosterone production instead of LH. Receptor studies demonstrated a dual action of the closely related hormones LH and CG in primates. It is hypothesised that following the divergence of the Platyrrhini lineage from Catarrhini, the LH/CG system evolved independently with ancestral functions of the LH/CG system retained in the neotropical taxa.

In summary, key spermatogenic features are preserved across all primate taxa whereas male reproductive endocrinology features appear substantially different in the neotropical primates compared to other primate lineages.

The gametes of man and some other Eutheria have been manipulated successfully for practical reasons, but many gaps remain in our basic understanding of the way that they function. This situation stems not least from a failure to recognize the extent to which eutherian spermatozoa and eggs, and elements related to their operation, have come to differ from those of other groups. Novel features in the male that reflect this include a radical design of the sperm head with the acrosome seeming to function primarily in egg-coat binding rather than its lysis, a multifaceted post-testicular sperm maturation and an androgen/low-temperature-regulated system of sperm storage – both tied to the epididymis, a variable male accessory sex gland complex, and descent of the testis and epididymis to a scrotum. In the female, such novelties are represented in a need for sperm capacitation, in an unusual regulation of sperm transport within the oviduct, in the cumulus oophorus and character of the zona pellucida around the small egg, and in a unique configuration of gamete fusion.

The collective evidence now suggests that many of these features reflect a new fertilisation strategy or its consequences, with most being causally linked. One initial ‘domino’ in this regard appears to be the small yolkless state of the egg and its intolerance for polyspermy, as determinants of the unusual mode of oviductal sperm transport and possibly the existence and form of the cumulus oophorus. However, a particularly influential first ‘domino’ appears to be the physical character of the eutherian zona pellucida. This differs from the egg coats of other animal groups by virtue of a resilient elasticity and thickness. These qualities allow this primary and often only coat to stretch and so persist during later expansion of the blastocyst, usually until close to implantation. At the same time, the dimensions, physical character, and particularly the relative protease-insensitivity of the zona appear to have had profound effects on sperm form and function and, more indirectly, on sperm-related events in the male and the female tract. Marsupials display some similarities and also some strikingly different features, against which the enigmas of the eutherian situation can be evaluated.

Parasitic exploitation occurs within and between a wide variety of taxa in a plethora of diverse contexts. Theoretical and empirical analyses indicate that parasitic exploitation can generate substantial genetic and phenotypic polymorphism within species. Under some circumstances, parasitic exploitation may also be an important factor causing reproductive isolation and promoting speciation. Here we review research relevant to the relationship between parasitic exploitation, within species-polymorphism, and speciation in some of the major arenas in which such exploitation has been studied. This includes research on the vertebrate major histocompatibility loci, plant–pathogen interactions, the evolution of sexual reproduction, intragenomic conflict, sexual conflict, kin mimicry and social parasitism, tropical forest diversity and the evolution of language. We conclude by discussing some of the issues raised by comparing the effect of parasitic exploitation on polymorphism and speciation in different contexts.

Impact craters contain ecosystems that are often very different from the ecosystems that surround them. On Earth over 150 impact craters have been identified in a wide diversity of biomes. All natural events that can cause localized disruption of ecosystems have quite distinct patterns of recovery. Impact events are unique in that they are the only extraterrestrial mechanism capable of disrupting an ecosystem locally in space and time. Thus, elucidating the chronological sequence of change at the sites of impacts is of ecological interest. In this synthetic review we use the existing literature, coupled with our own observations at the Haughton impact structure, Devon Island, Nunavut, Canada to consider the patterns of biological recovery at the site of impact craters and the ecological characteristics of impact craters. Three phases of recovery are suggested. The Phase of Thermal Biology, a phase associated with the localized, ephemeral thermal anomaly generated by an impact event. The Phase of Impact Succession and Climax, a phase marked by multiple primary and secondary succession events both in the aquatic realm (impact crater-lakes) and terrestrial realm (colonization of paleolacustrine deposits and impact-generated substrata) that are followed by periods of climax ecology. In the case of large-scale impact events (> 104 Mt), this latter phase may also be influenced by successional changes in the global environment. Finally, during the Phase of Ecological Assimilation, the disappearance of the surface geological expression of an impact structure results in a concomitant loss of ecological distinctiveness. In extreme cases, the impact structure is buried. Impact succession displays similarities and differences to succession following other agents of ecological disturbance, particularly volcanism.

More realistic approaches are needed to understand the complexity of ecological systems. Emergent properties of real systems can be used as a basis for a new, neither reductionist nor holistic, approach. Three systems, termed here BUBBLEs, WAVEs and CRYSTALs, have been identified as exhibiting emergent properties. They are non-hierarchical assemblages of individual components, with amplification and connectedness being two main principles that govern their build-up, maintenance and mutual relationships. Examples from various fields of biological and ecological science are referred to, ranging from individual organisms to landscapes.

The typical sperm is comprised of a head, midpiece and flagellum. Around this theme there is an enormous diversity of form – giant sperm, multi-flagellate sperm and also sperm that lack flagella entirely. Explaining this diversity in sperm morphology is a challenging question that evolutionary biologists have only recently engaged in. Nonetheless, one of the selective forces identified as being an important factor in the evolution of sperm form is sperm competition, which occurs when the sperm of two or more males compete to fertilize a female's ova. In species with a truly monandrous mating system, the absence of sperm competition means that the selection pressure on males to produce motile sperm may be relaxed. Potentially aflagellate sperm are less costly to produce, both in terms of energy and time. Thus, selection may therefore favour the loss of the sperm flagellum and any other motile mechanisms in monandrous taxa. A review of the literature revealed that 36 taxonomic groups, from red algae to fish, were found independently to have evolved aflagellate sperm. I review what is known about the mating systems of each of these taxa and their nearest sister taxa. A sister-group analysis using this information provided weak evidence suggesting that the evolution of aflagellate sperm could be linked to the removal of selective pressures generated by sperm competition.

The evolution and origin of primate social organisation has attracted the attention of many researchers, and a solitary pattern, believed to be present in most nocturnal prosimians, has been generally considered as the most primitive system. Nocturnal prosimians are in fact mostly seen alone during their nightly activities and therefore termed ‘solitary foragers’, but that does not mean that they are not social. Moreover, designating their social organisation as ‘solitary’, implies that their way of life is uniform in all species. It has, however, emerged over the last decades that all of them exhibit not only some kind of social network but also that those networks differ among species. There is a need to classify these social networks in the same manner as with group-living (gregarious) animals if we wish to link up the different forms of primate social organisation with ecological, morphological or phylogenetic variables. In this review, we establish a basic classification based on spatial relations and sociality in order to describe and cope properly with the social organisation patterns of the different species of nocturnal prosimians and other mammals that do not forage in cohesive groups. In attempting to trace the ancestral pattern of primate social organisation, the Malagasy mouse and dwarf lemurs and the Afro-Asian bushbabies and lorises are of special interest because they are thought to approach the ancestral conditions most closely. These species have generally been believed to exhibit a dispersed harem system as their pattern of social organisation (‘dispersed’ means that individuals forage solitarily but exhibit a social network). Therefore, the ancestral pattern of primate social organisation was inferred to be a dispersed harem. In fact, new field data on cheirogaleids combined with a review of patterns of social organisation in strepsirhines (lemurs, bushbabies and lorises) revealed that they exhibit either dispersed multi-male systems or dispersed monogamy rather than a dispersed harem system. Therefore, the concept of a dispersed harem system as the ancestral condition of primate social organisation can no longer be supported. In combination with data on social organisation patterns in ‘primitive’ placentals and marsupials, and in monotremes, it is in fact most probable that promiscuity is the ancestral pattern for mammalian social organisation. Subsequently, a dispersed multi-male system derived from promiscuity should be regarded as the ancestral condition for primates. We further suggest that the gregarious patterns of social organisation in Aotus and Avahi, and the dispersed form in Tarsius evolved from the gregarious patterns of diurnal primates rather than from the dispersed nocturnal type. It is consequently proposed that, in addition to Aotus and Tarsius, Avahi is also secondarily nocturnal.

Most processes or forms of learning have been treated almost as special creations, each as an independent process unrelated to others. This review offers an evolutionary cladogram linking nearly one hundred forms of learning and showing the paths through which they evolved.

Many processes have multiple forms. There are at least five imprinting processes, eleven varieties of Pavlovian conditioning, ten of instrumental conditioning, and eight forms of mimicry and imitation.

Song learning evolved independently in at least six groups of animals, and movement imitation in three (great apes, cetaceans and psittacine birds). The cladogram also involves at least eight new processes: abstract concept formation, percussive mimicry, cross-modal imitation, apo-conditioning, hybrid conditioning, proto-pantomime, prosodic mimicry, and image-mediated learning.

At least eight of the processes evolved from more than one source. Multiple sources are of course consistent with modern evolutionary theory, as seen in some obligate symbionts, and gene-swapping organisms. Song learning is believed to have evolved from two processes: auditory imprinting and skill learning. Many single words evolved from three sources: vocal mimicry, discrimination learning, and abstract concept formation.

A sigmoidal transcriptional response (STR) is thought to act as a molecular switch to control gene expression. This nonlinear behaviour arises as a result of the cooperative recognition of a promoter/enhancer by transcription factors (TFs) and/or their synergy to attract the basal transcriptional machinery (BTM). Although this cooperation between TFs is additive in terms of energy, it leads to an exponential increase in affinity between the BTM and the pre-initiation complexes. This exponential increase in the strength of interactions is the principle that governs synergistic systems. Here, I propose a minimalist quasi-equilibrium model to explore qualitatively the STR taking into account cooperative recognition of the promoter/enhancer and synergy. Although the focus is on the effect of activators, a similar treatment can be applied to inhibitors. One of the main insights obtained from the model is that generation of a sigmoidal threshold is possible even in the absence of cooperative DNA binding provided the TFs synergistically interact with the BTM. On the contrary, when there is cooperative binding, the impact of synergy diminishes. It will also be shown that a sigmoidal response to a morphogenetic gradient can be used to generate a nested gradient of another morphogen. Previously, I had proposed that halving the amounts of TFs involved in sigmoidal transcriptional switches could account for the abnormal dominant phenotypes associated with some of these genes. This phenomenon, called haploinsufficiency (HI), has been recognised as the basis of many human diseases. Although a formal proof linking HI and a sigmoidal response is lacking, it is tempting to explore the model from the perspective of dosage effects.

A number of plasma membrane receptor types originally thought to be specific to neurons have been found in other somatic cells. More surprisingly, the mammalian sperm and neuron appear to share many of these ‘neuronal’ receptors. The morphology, chromosome number, genomic activity, and functions of those two cell types are as unlike as any two cells in the body, but they both achieve their highly disparate goals with the aid of a number of the same receptors. Exocytosis in neurons and sperm is essential to the functions of these cells and is strongly influenced by similar receptors. ‘Neuronal’ receptor types in sperm may also play a role in the control of sperm motility (a function of course not shared by neurons).

This review will consider the evidence for the presence of sperm plasma membrane ‘neuronal’ receptors and for their significance to mammalian sperm function. The persuasiveness of the evidence varies depending on the receptor being considered, but there is strong experimental support for the presence and importance of a number of ‘neuronal’ receptors in sperm.

Recent studies on humans show that too much choice can make subjects less likely to choose any item. I consider general adaptive and non-adaptive explanations of why such choice aversion, or its converse, might occur in animals. There are three questions: is more choice always preferred, does it ever lead to less consumption (or a lower probability of consumption), and may it result in worse items being selected? A preference for choice is one of the main explanations for lek formation and I draw attention to previously unrecognised parallels with models of human shopping behaviour. There is indeed evidence of female preference for larger leks, although much of the observational data are open to other interpretations. Unfortunately nobody has looked for choice aversion where it is most to be expected, in leks larger than normally occur. Evidence that too much choice of males confuses females is strongest in acoustically advertising frogs, but the widespread decrease of mating skew in larger leks might also have this explanation. A model reanalyses data on skew in black grouse Tetrao tetrix and suggests that considering only a random subset of a large lek may increase the chances of selecting the better males: larger leks are more likely to include better males, but these are less likely to be selected. These opposing effects may lead to an optimum lek size, but only with a sufficient decline in choice accuracy with size. With food choice, very few studies have avoided confounding choice with food quality, by manipulating only flavour. The widespread phenomena of stimulus-specific satiety and novelty seeking imply that monotonous diets are aversive, but no studies test whether animals choose sites where they know food diversity to be greater. Operant experiments that demonstrate mild preferences for free choice concern choice about the means to get food rather than the food itself. In some insect species even moderate choice of diet can be deleterious, and studies on search images and the confusion effect may be evidence of this in vertebrates. Environmental enrichment of captive animals often relies on increasing the options available, but it need not be the choice itself that is beneficial. I consider briefly further areas in biology where choice preference or aversion are potentially important.

We hypothesize that forms of antagonistic coevolution have forged strong links between positive selection at the molecular level and increased cancer risk. By this hypothesis, evolutionary conflict between males and females, mothers and foetuses, hosts and parasites, and other parties with divergent fitness interests has led to rapid evolution of genetic systems involved in control over fertilization and cellular resources. The genes involved in such systems promote cancer risk as a secondary effect of their roles in antagonistic coevolution, which generates evolutionary disequilibrium and maladaptation. Evidence from two sources: (1) studies on specific genes, including SPANX cancer/testis antigen genes, several Y-linked genes, the pem homebox gene, centromeric histone genes, the breast cancer gene BRCA1, the angiogenesis gene ANG, cadherin genes, cytochrome P450 genes, and viral oncogenes; and (2) large-scale database studies of selection on different functional categories of genes, supports our hypothesis. These results have important implications for understanding the evolutionary underpinnings of cancer and the dynamics of antagonistically-coevolving molecular systems.

The use of available morphological characters in the interpretation of the flight of insects known only as fossils is reviewed, and the principles are then applied to elucidating the flight performance and techniques of Palaeozoic palaeopterous insects. Wing-loadings and pterothorax mass/total mass ratios are estimated and aspect ratios and shape-descriptors are derived for a selection of species, and the functional significance of wing characters discussed. Carboniferous and Permian ephemeropteroids (‘mayflies’) show major differences from modern forms in morphology and presumed flight ability, whereas Palaeozoic odonatoids (‘dragonflies’) show early adaptation to aerial predation on a wide size-range of prey, closely paralleling modern dragonflies and damselflies in shape and wing design but lacking some performance-related structural refinements. The extensive adaptive radiation in form and flight technique in the haustellate orders Palaeodictyoptera, Megasecoptera, Diaphanopterodea and Permothemistida is examined and discussed in the context of Palaeozoic ecology.

Among the air-breathing vertebrates, the avian respiratory apparatus, the lung-air sac system, is the most structurally complex and functionally efficient. After intricate morphogenesis, elaborate pulmonary vascular and airway (bronchial) architectures are formed. The crosscurrent, countercurrent, and multicapillary serial arterialization systems represent outstanding operational designs. The arrangement between the conduits of air and blood allows the respiratory media to be transported optimally in adequate measures and rates and to be exposed to each other over an extensive respiratory surface while separated by an extremely thin blood-gas barrier. As a consequence, the diffusing capacity (conductance) of the avian lung for oxygen is remarkably efficient. The foremost adaptive refinements are: (1) rigidity of the lung which allows intense subdivision of the exchange tissue (parenchyma) leading to formation of very small terminal respiratory units and consequently a vast respiratory surface; (2) a thin blood-gas barrier enabled by confinement of the pneumocytes (especially the type II cells) and the connective tissue elements to the atria and infundibulae, i.e. away from the respiratory surface of the air capillaries; (3) physical separation (uncoupling) of the lung (the gas exchanger) from the air sacs (the mechanical ventilators), permitting continuous and unidirectional ventilation of the lung. Among others, these features have created an incredibly efficient gas exchanger that supports the highly aerobic lifestyles and great metabolic capacities characteristic of birds. Interestingly, despite remarkable morphological heterogeneity in the gas exchangers of extant vertebrates at maturity, the processes involved in their formation and development are very similar. Transformation of one lung type to another is clearly conceivable, especially at lower levels of specialization. The crocodilian (reptilian) multicameral lung type represents a Bauplan from which the respiratory organs of nonavian theropod dinosaurs and the lung-air sac system of birds appear to have evolved. However, many fundamental aspects of the evolution, development, and even the structure and function of the avian respiratory system still remain uncertain.

Once the inoculum of B. cinerea comes into contact with the host and starts to be active in the phyllosphere of a susceptible host tissue, a series of events take place. These events may develop into a process that leads to necrosis of the host, or may end in an arrested infection with minimal damage to the host tissue. Increased susceptibility to the pathogen is associated with factors that enhance ageing of the host tissues, such as the plant hormones ethylene and abscisic acid and elevation of free radical levels in the host tissue. Decreased susceptibility is obtained by inhibiting the production or activity of such factors in the presence of increased levels of plant hormones such as gibberellic acid, and by increasing the calcium content of the cell walls and by scavenging of free radicals in the host tissue. There is evidence for the induction of resistance in hosts affected by B. cinerea. Host tissues challenged by B. cinerea react at the DNA, RNA and protein level and accumulate pathogenicity related proteins, phytoalexins or other phenolic compounds. Deposition of polymers in cell walls and lignification have also been recorded in various hosts. The role of each of these factors in relation to protection is not clear. Moreover, some of the phenomena may occur too late to protect the host tissue against infection. Although the inhibition of specific proteins such as polygalacturonases has been suggested as a mechanism by which to inhibit disease, it is unlikely that the inhibition of one enzyme, would lead to significant restriction of infection. However, simultaneous inhibition of several hydrolytic enzymes produced by the pathogen should result in disease suppression. Possibilities of reducing the susceptibility of hosts or arresting further development of localized infections are discussed.

The aim of this review is to summarize newly available information on lemur social systems, to contrast it with the social organization of other primates and to relate it to existing models of primate social evolution. Because of their evolutionary history, the primates of Madagascar constitute a natural experiment in social evolution. During millions of years of isolation, they converged with other primates only in the most fundamental way in the evolution of solitary, pair-living and group-living species, but deviate in several respects within these basic categories of social organization. Solitary lemurs remain poorly studied, but their social organization appears to be broadly similar to that of other solitary primates, even though the unexpected lack of sexual dimorphism may indicate that similar types of social organization can give rise to different mating systems. The determinants of a solitary lifestyle remain elusive. Pair-living lemurs show striking convergences with other monogamous primates in several behavioural traits, but also deviate in that the majority of species are at least partly nocturnal and do not exhibit direct paternal care of dependent young. Group-living lemurs have not evolved single-male groups, male-bonded and multi-level societies, and polyandrous groups may also be lacking. Female philopatry is common, but female bonds are generally weakly developed and eviction of females from natal groups is not unusual. Group-living lemurs also differ from anthropoids in that their groups have even adult sex ratios, smaller average size and may split up on a seasonal basis. Feeding competition, predation risk and reproductive competition can not fully explain these unusual aspects of lemur social organization. It has therefore been suggested that the social consequences of the risk of infanticide and of recent changes in activity may be ultimately responsible for these idiosyncracies of group-living lemurs, an explanation largely supported by the available evidence. Thus, social factors and fundamental life-history traits, in addition to ecological factors, contribute importantly to variation in social systems among lemurs, and possibly other primates. However, neither the diversity of lemur social systems, nor the evolutionary forces and mechanisms operating in these and other primates are yet fully understood.