Predators with Pouches:

The Biology of Carnivorous Marsupials

Predators with Pouches

Predators with Pouches provides a unique synthesis of current knowledge of the world’s carnivorous marsupials—from Patagonia to New Guinea and North America to Tasmania. Written by 63 experts in each field, the book covers a comprehensive range of disciplines including evolution and systematics, reproductive biology, physiology, ecology, behaviour and conservation.

Predators with Pouches reveals the relationships between the American didelphids and the Australian dasyurids, and explores the role of the marsupial fauna in the mammal community. It introduces the geologically oldest marsupials, from the Americas, and examines the fall from former diversity of the larger marsupial carnivores and their convergent evolution with placental forms.

The book covers all aspects of carnivorous marsupials, including interesting features of life history, their unique reproduction, the physiological basis for early senescence in semelparous dasyurids, sex ratio variation and juvenile dispersal. It looks at gradients in nutrition—from omnivory to insectivory to carnivory—as well as distributional ecology, social structure and conservation dilemmas.

  1. Page xii
    1. Page 3

      Molecular phylogenetic studies (albumin microcomplement fixation, DNA hybridisation, and DNA sequence analyses) since 1990 have refined the hypothesis of dasyuromorphian relationships inferred from morphological and allozyme data in 1982. DNA sequences weakly support Thylacinidae and Dasyuridae as sister-families apart from Myrmecobiidae, but this problem requires further study. All molecular studies agree on the placement of two endemic New Guinean groups within larger clades of dasyurids. Specifically, Phascolosorex and Neophascogale (‘phascolosoricines’) form a monophyletic group with dasyurines (quolls, false antechinuses, and allies), and Murexia (including all New Guinean ‘antechinuses’) forms a monophyletic group with phascogalines (phascogales and antechinuses). Otherwise, the basic phylogenetic structure of Dasyuridae inferred from molecular data is consistent with earlier morphological and allozyme results. DNA sequences place ‘phascolosoricines’ as sister to a Dasyurus-Sarcophilus clade within Dasyurini, and Murexia as sister to Antechinus within Phascogalini; both results have major implications for morphological evolution and require further evaluation. Within Dasyurini, DNA sequences show several early lineages (including the Dasyurus-Sarcophilus group) that may have resulted from an episode of rapid cladogenesis. Within Antechinus, DNA sequences recover monophyletic species groups that are identical to those proposed on the basis of allozymes. Within Sminthopsini, Sminthopsis and Ningaui form a clade apart from Antechinomys. Although DNA sequences fail to resolve a monophyletic Sminthopsis, they do recover dunnart species groups that are partially congruent with those suggested by morphology. Within Planigalini, Planigale maculata and perhaps Pl. novaeguineae are sister to other planigale species. Relationships among early dasyurin lineages, Antechinus species groups, Murexia species, Sminthopsis species groups, and most Planigale species are currently unresolved. We endorse a revised suprageneric classification of dasyurids that reflects the most robust molecular results. Cladogenic dates estimated from a molecular clock indicate that the four major radiations of modern dasyurids took place in the late mid-Miocene, perhaps in response to climatic drying across Australia. DNA sequence studies concur with recent morphological and allozyme results in suggesting the existence of several (perhaps many) undescribed or cryptic dasyurid species.

    2. Page 21

      Until now, phylogenetic studies regarding the evolution of marsupials have included morphological, cytogenetic, biochemical and molecular approaches. Within American metatherians, the large-sized didelphids seem to constitute a well supported clade in contrast with mouse opossums, which are not grouped in a natural assemblage. The phylogenetic position of the medium-sized Metachirus has long been a point of confusion, but recent molecular studies confirm earlier hypotheses that this genus is a taxon belonging to the large-sized didelphid opossum clade. South American shrew opossums (the caenolestids) appeared as the sister taxon to didelphids, although some molecular reconstructions using different molecular markers show them in close association with the Australian peramelids. Finally, one of the long-standing issues regarding the evolution of Australian and South American marsupials has been the affinities of Dromiciops gliroides, a microbiotheriid mouse opossum better known as the ‘monito del monte’. According to different data set analyses, this species should be part of a clade that includes the Australasian dasyurids, diprotodontians and notoryctemorph marsupials. Current paleontological calibration, as well as geological and geophysical evidence, would sustain a passage of microbiotheriids from South America to Antarctica with further differentiation in the latter continent. Alternatively, microbiotheriids could constitute the ancestors of Australasian marsupials, having dispersed from Antarctica to Australasia and subsequently differentiating and undergoing further extinction in the latter region.

    3. Page 30

      The three classic groupings of South American marsupials–sparassodonts, pseudodiprotodonts, and didelphimorphians – may not constitute natural groups. The affinities of borhyaenoids and allies are currently under intense scrutiny and debate. The molar morphology of an early form, Mayulestes, suggests peradectian affinities. The monophyly of Paucituberculata + Polydolopimorphia is not supported on the basis of available information. Instead, it is suggested the possible belonging of microbiotherians, glasbiids and polydolopimorphians to a natural group. Didelphimorphia, as currently understood, is probably another waste basket taxon that includes several lineages of still uncertain affinities.

    4. Page 43

      The comparison of the postcranial skeletons of the only three subcomplete marsupial skeletons (Pucadelphys, Andinodelphys, and Mayulestes) from the Palaeocene (Tiupampa, Bolivia) allows an appraisal of their locomotory habits and biology. The anatomy of the cervical, lumbar and caudal vertebrae, shoulder, elbow, hip, knee, and ankle joints is successively analysed. Several features are indicative of arboreal abilities (e.g. prehensile tail, posterodorsal extension of the posterodorsal angle of the scapula, anterior position of the acromion, relatively spherical head of the humerus, prominent medial epicondyle, ulna bent anteriorly, widely opened trochlear incisure, open acetabulum with a concave dorsal edge, large lesser trochanter of the femur, medial orientation of the ectal facet of the calcaneum). However, an increasing gradient of arboreal ability was observed from Pucadelphys to Mayulestes. Several other features indicate that the three genera were very agile (e.g. anterior position of the anticlinal vertebra, large neural and transverse processes of the lumbar vertebrae, strongly everted ilium, elevated crests of the femoral trochlea). The Tiupampa marsupials were clearly more agile than most living didelphids (except Metachirus) and resemble the condition observed in living dasyurids. It is hypothesised that the Tiupampa didelphimorphs could have approached what is regarded as the probable generalised plesiomorphic pattern for marsupial locomotion, i.e. agile terrestrial animals but with good climbing ability. The less specialised taxa (Pucadelphys and Andinodelphys) could be structurally ancestral to both didelphids and dasyurids.

    5. Page 63

      We examine the phylogeographic structure of nine genera of rainforest didelphid marsupials (Didelphis, Philander, Metachirus, Gracilinanus, Marmosa, Marmosops, Micoureus, Monodelphis, and Caluromys) based on sequences from the mitochondrial cytochrome b gene. Multiple geographic representatives of many of the currently recognised species in each genus provide a backdrop to questions concerning the nature of species boundaries and the geographic ranges of these species, as well as the proper application of the available names. We use phylogeographic data to understand the historical connections between the major wet forest biomes of South America, specifically connections between southern Amazonia and the Atlantic Forest of coastal Brazil. Species diversity in several genera is greater than current taxonomy would suggest. While some lineages appear relatively recent (lowland species of Didelphis), the majority of extant species are surprisingly divergent, suggesting species formation well before the Pleistocene. Thus, there appears to be little support for the putative role of Pleistocene refuges in generating species diversity in these genera. Geographic samples are limited, but there are areas of strong phylogeographic concordance among lineages within Amazonia and historical connections occur between southern Amazonia and the Atlantic Forest through both the Paraná Basin and around the ‘horn’ of eastern Brazil.

    6. Page 82

      Marsupials of the genus Thylamys Gray 1843 (Didelphimorphia: Didelphidae) include small mouse opossums with distinctive morphological traits, mainly distributed on dry open and semi-arid biomes of southern South America. Similar to other Neotropical mammals, its species diversity and distribution are poorly documented. Thylamys, with Didelphis elegans Waterhouse 1839 as type species, was early on used as a subgenus of Marmosa, with a broader definition and including species now in the genera Marmosops and Gracilinanus. By the 1980s, it became clear that Marmosa was not a natural group, and Thylamys was raised to full genus and associated to the elegans group of Tate (1933). However, the situation for most taxa included in the genus remains obscure because good series are not available.

      Peruvian specimens show a wide variation; however, they may represent to two different taxa instead of a single species as it is currently recognised: Thylamys elegans. Through comparison and evaluation of discrete morphological traits and morphometric analyses of the variation within and among Peruvian and Chilean populations, I assign Peruvian specimens to tatei and pallidior. By comparing tatei and pallidior to elegans, I show this taxon to be more restricted than previously thought.

      Because elegans is the only valid known species from the west of the Andes, it is supposed that tatei and pallidior should have closer affinities to it. Although elegans and tatei are alike externally, pallidior shares more cranial characters with tatei. In addition, the latter two species do not show sexual dimorphism, which is evident in elegans. I recommend the use of tatei as a valid species, restricted to the western slopes of central Peru. Morphological comparisons included other taxa from the whole geographic range of the genus to update the diversity and distribution of the genus on western South America.

      Morphological characters used to distinguish species proved so effective that I have used the same set of characters to group the species. Used in combination with their distributions, I group the seven recognised species in three geographic units, as a first approach to natural groups. I propose a biogeographic scenario to explain the colonisation of the western side of the Andes by the genus, typically found at temperate areas to the east of the Andes. Two different types of dispersal are supposed to have occurred in this migration, favoured by the climatic fluctuations and final uplift of the Andes during the Plio-Pleistocene.

    7. Page 102

      Over the last two decades, the study of new fossil material has yielded major insights into the evolution of Australia’s marsupial carnivores. This material includes the first complete or near-complete crania for Tertiary representatives of Dasyuridae, Thylacinidae and Propleopinae (Hypsiprymnodontidae), as well as the first skulls known for the thylacoleonid genera, Priscileo and Wakaleo. Also of significance has been the discovery of a new species of marsupial carnivore, Djarthia murgonensis, from the Early Eocene Tingmarra Local Fauna. Regarding dasyuromorphian evolution, the study of well-preserved material from Oligocene–Miocene deposits of Riversleigh has been particularly illuminating. Many findings have been unanticipated. For example, it is now clear that during the early and middle Miocene, the now ubiquitous dasyurids were rare, while the recently extinct Thylacinidae were unexpectedly diverse. Furthermore, at present there is no hard evidence for the existence of any extant dasyurid genera or subfamily greater than Pliocene in age. The oldest confirmed dasyurid is from early to middle Miocene deposits and forms a sister clade to the three living subfamilies. Moreover, it is argued that dasyurids are highly specialised among dasyuromorphians, particularly with respect to their basicranial morphology and not ‘primitive’ Australian marsupials as has often been supposed. On the other hand, from results of analysis of late Oligocene–Miocene cranial material, it is now evident that Thylacinidae constitutes a very conservative lineage with the recently extinct Thylacinus cynocephalus little more derived than some late Oligocene–Miocene taxa regarding either cranial or dental features. Propleopinae (giant rat-kangaroos), previously known only from dental remains, are now represented by two skulls. Interpretation of this evidence supports the hypothesis that at least some species included significant amounts of meat in their diets, while phylogenetic analysis hints at the possibility of a special relationship with balbarines. With respect to Thylacoleonidae, the study of cranial material leaves the issue of the family’s ordinal level affinities uncertain, while the previously accepted tenet that Wakaleo could not be ancestral to Thylacoleo is also questioned. On the basis of dental evidence to hand, the Early Eocene Djarthia murgonensis can not be placed in any marsupial clade with confidence. Thus, biogeographic scenarios excluding the possibility that ameridelphians ever colonised Australia are considered premature. Growing evidence for a diversity of marsupial carnivores in pre-Pleistocene Australia is considered to diminish, if not contradict the argument that the continent’s large terrestrial carnivore niches have long been dominated by reptiles. The flip side of this debate is that arguments for reptilian supremacy are commonly based on assumptions regarding the biology and behaviour of fossil varanid, snake and crocodilian taxa that in many cases are highly speculative. A tendency to consider only estimated maximum dimensions for extinct reptilian species may also have generated false impressions with respect to the their significance in the ecology of their respective communities.

    8. Page 124

      The Dasyuridae of Australia and New Guinea exhibits one of the most spectacular species-level diversity of any marsupial group. This chapter examines the biogeographical areas of Australia and postulates their role in the subspecific variation and speciation of dasyurid taxa. Physical barriers (such as rivers), as well as fluctuations of the arid-zone during the Pleistocene, explain much of the diversity within dasyurid species and genera. However recent molecular work suggests that most speciation within the Dasyuridae was a result of earlier environmental factors during the Miocene and Pliocene. Evidence is also presented that competition between dasyurid taxa lead to variation within species, and in the case of Antechinus, sympatric speciation. Finally it is concluded that further studies on dasyurid ecology, systematics and biogeography are needed to quantify their true specific diversity, as well as explain the factors that shaped their evolution.

    1. Page 133

      The carnivorous and insectivorous marsupials exhibit a number of ancestral morphological features that are not evident in most other Australasian families. In the present review the structural design of the spermatozoon and egg, and their maturation and interaction in the female reproductive tract at the time of fertilisation, are briefly considered for these species.

      The evidence suggests that the gametes of the dasyurids and didelphids appear to be highly derived in their structural organisation as well as in being very different from each other. In didelphids sperm pairing occurs, whereas in dasyurids post-testicular sperm maturation is elaborate but no sperm pairing takes place. In the latter group, but not the former, prolonged sperm storage occurs in the higher reaches of the female reproductive tract after mating. Oocytes of dasyurids, but not didelphids, have a large central yolk mass and in the zona pellucida the distribution and abundance of oligosaccharides appear to differ between the species. Thus in both these families of marsupials some unique features of gamete design, organisation and behaviour appear to have evolved, whereas other features are shared by at least a few other marsupial groups.

    2. Page 147

      Reproduction in many carnivorous marsupials is seasonal; however the proximal cues for reproduction are known for a minority of species. Species from the genera Antechinus and Sminthopsis appear to rely on photoperiodic change for timing of reproduction, with other environmental factors such as pheromones (Antechinus) and rainfall (Sminthopsis) of secondary importance. For most other carnivorous marsupials the proximal cues for reproduction have not been determined, although photoperiod is likely to be the major cue for reproductive timing. Unfortunately, almost nothing is known about reproductive activity for many carnivorous marsupials, especially those from South America, and only further studies can clarify how their reproductive life history is controlled, requiring significant further effort on the part of researchers.

    3. Page 169

      Reproductive data has been obtained for seven species of dasyurid marsupials endemic to New Guinea, with the aim of establishing the pattern of reproduction of each. Observations have been made on animals in the field as well as in captivity, and it appears that all seven are capable of year-round breeding. This contrasts with what is known for the Australian species, which, with two exceptions, are strictly seasonal breeders. Aspects of the reproductive anatomy and behaviour of males and females of each species have been documented.

    4. Page 183

      This chapter presents an overview onthe gross anatomy and histology of the male genital system of several South American didelphid species of the genera Didelphis, Philander, Metachirus, Lutreolina, Caluromys, Monodelphis, Marmosops, Gracilinanus, Marmosa, Micoureus and Glironia, with emphasis on the white-belly opossum Didelphis albiventris. A comparative description of the testes, testicular-epididymal pedicle, epididymides, ductus deferens and spermatic cord, urethra, penis, prostate, two or three pairs of bulbourethral glands, as well as the cloaca and paracloacal glands, are herein presented. Particularly in Didelphisalbiventris, a few other aspects such as innervation of the male genital system, puberty, biometry and testicular activity during the annual reproductive cycle are also dealt with. Further studies involving other South American marsupials of the genera Thylamys, Chironectes, Caluromysiops, Caenolestes and Dromiciops are fundamental for a better understanding of the male genital system and phylogeny of American marsupials.

    5. Page 205

      The sensory and motor development of the early pouch young of the Northern Quoll (Dasyurus hallucatus) are described in some detail to give a basis for the relationship between the motor behaviour of the newborn marsupial and its neural and anatomical development. Comparisons are then made with other newborn marsupials and with eutherians at stages of development similar to those of the newborn marsupials. The differences between marsupials and eutherians at these stages are not as great as might be assumed. All have developed parts of the vestibular and somatosensory systems but the trigeminal nerve provides the main sensory input (including a chemical sense) at these stages and it, and the more posterior cranial nerves, are the controllers of the motor behaviour at these stages. Different marsupials are born at different stages and the mothers adopt different birth positions. Thus the journey from the urinogenital sinus to the pouch poses differing challenges for the newborn. Different species have different sensory and motor development and so have different mechanisms to achieve the different challenges involved in reaching the teat.

    1. Page 221

      Carnivory includes both faunivory (eating of vertebrates) and insectivory (eating of invertebrates). Faunivory is associated with larger body size, insectivory with smaller body size, but not exclusively. When scaled for metabolic body size, the energy requirements of marsupial carnivores are higher than those of most other marsupials but the numbat, an ant eater, is an exception; like eutherian ant eaters it has very low energy requirements. Little is known of the specific nutrient requirements of carnivorous marsupials. Carnivore diets are generally high in water, protein, vitamins and minerals, low in carbohydrate, and highly variable in fat content. Muscle and viscera are generally highly digestible but mammalian hair and teeth, reptilian dermal scales and avian feathers are poorly digested and consequently help in identification of prey remains in carnivore faecal scats. The digestive tract of carnivores is conservative, and the small intestine (the main site of digestion and absorption) dominates. Digesta passage is rapid, but decreases with increasing body size of the carnivore. Strict carnivory is a narrow nutritional niche, with disadvantages as well as advantages. The metabolism of carnivores conforms to the nutrient profile of animal tissues, which eliminates the energetic costs of synthesising redundant enzymes but limits the ability to exploit diets of other than animal material. These general principles of carnivory, derived mainly from eutherians, are applied to marsupials in this chapter. The plea is made for more specific nutritional information on carnivorous marsupials in order to better understand the basis for niche separation among grossly similar species, and thus manage habitats for maximum biodiversity of marsupial carnivores.

    2. Page 229

      Nutritional requirements of 12 species of Neotropical opossums (Didelphidae) were inferred through a laboratory food preference experiment. Nutritional contents of experimental diets follow predictions based on field diet data, with more frugivorous species showing high non-structural carbohydrate proportions and more carnivorous species high protein contents in their selected diets. Mouse opossums selected nutritional and fibre proportions that suggested that they may be more frugivorous and less strictly insectivorous than previously thought. The species form a gradient of differential food specialisation from frugivory to carnivory, and significant differences are found only between extremes of this gradient. Nutritional contents established here are consistent with natural diet patterns and can be used as important complementary data to field diet studies and to prepare diets for captive maintenance of the species.

    3. Page 238

      Extant carnivorous marsupials are small (4–13,000 g) and almost all are nocturnal or crepuscular and are active during the coldest part of the day. Heat loss and gain via their relative large surface area is therefore likely to be substantial and to have implications on thermal biology and energy expenditure. In this review, data on thermoenergetics in carnivorous marsupials, with an emphasis on the families Dasyuridae and Didelphidae, are summarised and compared with data on other marsupials and mammals in general.

      All carnivorous marsupials have low basal metabolic rates (BMR) when compared to most placental mammals, but similar BMR to those of omnivorous/herbivorous marsupials. Thermal conductances of carnivorous marsupials are similar to those of other similar-sized mammals. Carnivorous marsupials have a high metabolic scope and endogenous heat production is achieved to a large extent by shivering thermogenesis and some poorly understood non-shivering component. During exposure to heat carnivorous marsupials use predominantly panting and licking of fur and appendages for evaporative cooling. Carnivorous marsupials have relatively high field metabolic rates (FMR) and, especially in the small species, high FMR/BMR ratios. To minimise daily energy expenditure many carnivorous marsupials use communal nesting and huddling and torpor extensively and thus can lower energy expenditure substantially. While thermal biology and energetics of carnivorous marsupials generally is well known, most of the information is based on laboratory work. Thus, more fieldwork is needed to put physiological data of carnivorous marsupials into an ecological context.

    4. Page 254

      This chapter covers some of the basic ecological, behavioural and physiological changes that have been reported in studies of dasyurid marsupials exhibiting the unusual post-mating male mortality (life-history Strategy 1). Rather than attempting to cover all reports, the chapter concentrates upon changes that the author considers to be most relevant in attempting to explain why post-reproductive males in the wild die, yet females survive. Evidence relating to a defect in feedback of corticosteroid hormones in discussed in relation to recent findings for eutherian stress models. The relevance of these studies for explaining the truncated lifespan of Strategy I dasyurid males is discussed and an integrated flow diagram is attempted to synthesise the physiological changes that are likely to occur in the weeks preceding the death of males. Considered together, the adaptive physiological changes that occur during the last few weeks of life of the males are remarkable in enabling them to maximise their reproductive potential prior to a rapid physiological decline that involves stress related dysfunction and pathologies involving renal, gastrointestinal (GI), neuroendocrine and central nervous systems.

    1. Page 271

      The extent to which the term ‘carnivory’ can be applied to extant Neotropical marsupials is somewhat difficult to determine clearly. Although no Neotropical marsupial feeds exclusively on plant matter, the relative importance of animals in the diet of each species may vary. At present there is no living counterpart to the extinct carnivorous marsupials once found in many regions of South America, specially members of the Borhyaenidae, the most diverse family. The old explanation for the extinction of all exclusively predatory forms of South American marsupials through competitive exclusion by North American newcoming eutherian carnivores, arrived during the Great American Interchange, is no longer accepted nowadays. In fact, the decline of the marsupial predatory forms began much earlier than the arrival of eutherians, and many were already completely extinct by the time of the rise of the Panama isthmus. The living marsupials possess similar body form and do not show great niche diversity. These animals can be placed in five different classes in relation to their degree of carnivory/insectivory. The most carnivorous genera are Lestodelphys, Lutreolina, and Chironectes that feed mainly on vertebrates and invertebrates. The genera Philander, Metachirus, Thylamys, Monodelphis, Caenolestes, Rhyncholestes, Lestoros, and Dromiciops feed mainly on invertebrate animals, but also on vertebrates and on plant matter. For Neotropical marsupials there is a negative and significant correlation between arboreal activity and degree of carnivory. Carnivorous/insectivorous marsupials should also have relatively low numerical densities and low biomasses than more omnivorous species. Nevertheless feeding requirements may also provide some advantages for carnivorous/insectivorous marsupials, as high-energy food enables them to survive in less favourable environments at higher latitudes in South America. Basic studies focusing on diet of carnivorous/insectivorous marsupials are still necessary. Moreover, only long-term field studies (i.e. =3 years) will be adequately address several relevant ecological aspects of these marsupials.

    2. Page 285

      The extent of ecomorphological convergence in function and form between marsupial and placental carnivores is analysed. Convergence is apparent in trophic functional morphology, with the genera of marsupial carnivores grouping with the families of placental carnivores with which they seem convergent from external appearances. Convergence in trophic morphology or form is superficial, however. A major difference in canine tooth shape relates to different killing behaviour in the marsupial carnivores. Other phylogenetically based differences in skull morphology may represent different solutions to the same biomechanical problems. In terms of morphologies associated with locomotor and hunting behaviour, the marsupial carnivores tend to group together, as does each family of placental carnivore. The skeletal indicators of running speed, activity substrate and prey capture mode appear to be more useful in separating and assigning major groups to their predominant locomotory type than in assigning different species within those groups to their known hunting type and activity substrate classifications. This suggests a strong influence of phylogenetic history on skeletal morphology. Similar morphological size patterns, in the trophic structures that are proximal to prey capture methods, have been found in guilds of both marsupial and placental carnivores. These patterns have an underlying basis in food size partitioning and competitive pressure (evidence from character release). These similar patterns attest to convergence in the structuring mechanisms that influence morphological size relationships within ecological guilds.

    3. Page 297

      We studied latitudinal patterns of species diversity, trophic guild structure, habitat use, body size and use of tail in South American marsupials. We regressed ratios of biological characters on temperature and precipitation parameters as surrogates for latitudinal changes in the physical environment. Our results indicate that mean minimum temperature is positively related with species diversity, the proportion of frugivores, the percentage of marsupials with a long prehensile tail, and the proportion of arboreal and scansorial marsupials forms, and is negatively correlated with the proportion of carnivores, the percentage of marsupials with incrassated tails, the proportion of terrestrial forms, and the proportion of small-sized marsupials. Variability in temperature has higher explanatory power than variability of precipitation. Major life forms in South American marsupials show latitudinal patterns.

    4. Page 318

      Assemblages of small dasyurid marsupials (<500 g) are usually more diverse in arid regions of Australia than in non-arid coastal and sub-coastal areas, whereas the converse is true for larger species. To quantify species overlaps at the regional scale, species density maps were constructed using distributions of all species predicted from bioclimatic modelling. Highest species densities of small dasyurids were predicted to occur in hummock grassland and desert complex habitats (mean species richness 7.0–8.2; maximum 14), and the lowest to occur in forest and heathland (mean species richness 3.6–5.4; maximum 7). Actual field surveys showed that local species richness in most habitats was less than half that predicted at the regional scale, but approached two-thirds the regional level in arid woodland and hummock grassland. Local richness in hummock grassland was the highest of any habitat, averaging 5.3 species with a maximum of eight. Species densities of large (>500 g) dasyurids were greater regionally than locally, but nonetheless ranged from 0–3 at both scales. No large species now occur in arid habitats following the decline of the western quoll, Dasyurus geoffroii, and assemblages of three species are restricted to Tasmania.

      For small dasyurids, species richness at the local scale depends on the structural complexity of vegetation and other, abiotic components of the environment. Structurally complex habitats allow species to achieve separation of foraging niches and hence reduce dietary overlap. In non-arid habitats interspecific competition appears to maintain foraging niche separation and hence limits the number of species that coexist, but in arid habitats population densities of most species are so low that competition provides little constraint on coexistence. In hummock grassland other potential constraints such as predation appear unimportant; emerging evidence suggests instead that one species, the mulgara Dasycercus cristicauda, has a facilitatory effect on the local richness of smaller species. Local richness is further enhanced in arid habitats by climatic events such as fire and rainfall that can stimulate movements of species across the regional landscape. Large dasyurids have been more affected than their smaller relatives by the shocks of European settlement. On mainland Australia there is evidence that remaining species segregate by habitat, while in Tasmania competition appears to drive complementary niche separation along both habitat and prey size axes.

      Future work should focus on the enigmatic dasyurids of New Guinea, and attempt to resolve the distributional limits of species with patchy or restricted ranges. The outstandingly rich assemblages of small dasyurids in hummock grasslands also provide a beacon for future research, as do several poorly-known taxa that are restricted to coastal forest and heathland habitats. More long-term and experimental studies are essential to disentangle the factors that influence the distributions and diversity of dasyurids generally.

    5. Page 332

      The behaviour of carnivorous marsupials is reviewed with a particular focus on new knowledge gathered from the 1980s and the interpretation of behaviour in the framework of behavioural ecology. The general life-sustaining behaviour of individuals is first described by examining activity patterns, maintenance activities, nest building and use of shelter, predatory behaviour, predator avoidance and exploration and object play. Carnivorous marsupials are typically small, nocturnal, secretive and exploit a wide range of prey through agile and flexible behaviour. Social behaviour is reviewed with a focus on communication; spacing; agonistic; and sexual behaviour; parental care, and socialisation and play. Although most carnivorous marsupials are solitary, they express a large repertoire of social acts with some unusual behaviour. Knowledge about the behaviour of carnivorous marsupials has progressed well beyond simple description of form and function to evolutionary insights through comparative study and experimentation. There is scope for much more research in this direction, especially with American and New Guinean species.

    6. Page 347

      This chapter gives a brief overview of studies that have been carried out to describe the way in which dasyurid marsupials communicate by chemical means. This involves the production of chemical substances from both cutaneous scent glands and glands associated with the reproductive tract, the dispersal of these substances by various morphological adaptations, and the detection of these airborne chemicals (pheromones) by the special sensory system, the vomeronasal organ (VNO). Gas chromatography coupled with mass spectroscopy GC-MS is increasingly being applied to identify substances used by animals in chemical communication. While many of the structures used by animals to detect pheromones have been known for many years, it is only now with the availability of sensitive techniques such as functional magnetic resonance imaging (fMRI) that we are able to visualise regional changes in brain activity in temporal sequence in response to these pheromones.

    7. Page 358

      The limited relevant data that are available which may pertain to the occurrence of sperm competition and mating systems in carnivorous marsupials are presented. A positive allometric relationship was found between body mass and testis mass (n = 46 species) with some variation around the mean. There was also a positive allometric relationship between body mass and numbers of stored sperm in the male reproductive tract (n = 8). The limited data available suggest that length of copulation varies markedly between species and the sperm, once deposited in the female tract, undergo efficient transport to the higher reaches of the female tract, with storage taking place in the isthmus of the oviduct until ovulation. Females ovulate many oocytes and, where investigated in Antechinus, multiple paternity was found to occur in the animals in both laboratory experiments and in the wild. These data are discussed in relation to the apparent breeding system, and the possibility of inter-male sperm competition.

    8. Page 376

      Several studies have now established that biased sex ratios occur in litters of some species of dasyurid and didelphid marsupials. It is presumed that these biased ratios have an adaptive significance and they have been interpreted within various theoretical frameworks, predominantly the Local Resource Competition Hypothesis (LRC), the Trivers-Willard Hypothesis (TWH), and the First Cohort Advantage Hypothesis (FCAH). No single framework has been successful in explaining all cases observed, even within a single species, and none is likely to. One barrier to providing adaptive explanations for sex-biased litters is that we know of no low-cost mechanism of sex ratio manipulation. Recent studies with Antechinus agilis have shown a bias of 2F:1M that is generated prior to birth, and the full level of bias cannot be explained by sex-selective embryo loss alone. Pre-fertilisation mechanisms must contribute to the generation of these sex-biased litters. As we continue to narrow down the stage(s) at which sex bias is generated, we come closer to determining the mechanism(s) that generate sex bias in litters, but at present we can only speculate. One possibility, in A. agilis at least, is that sexes of sperm respond differently to the period of sperm storage between copulation and fertilisation.

    9. Page 383

      The parasites of dasyurid marsupials are far better known than those of other carnivorous Australian or of South American marsupials. The known parasite fauna of dasyurids is diverse and includes representatives of eight genera of trematodes, nine genera of cestodes, 33 genera of nematodes, a single species of acanthocephalan, 17 genera of fleas, one genus of lice, two genera of ticks, 27 genera of mites and five species of protozoans. The families of parasites represented are often cosmopolitan but there is a high level of endemicity at the genus and particularly at the species level. Several genera of helminths and arthropods provide links between the South American and Australian parasite fauna, but there are many instances of acquisitions of parasites from other host groups. There is a remarkable similarity between types of parasites present in dasyurid marsupials and those found in eutherian carnivores or insectivores, illustrating the importance of diet and parasite life cycles as well as host phylogeny in determining the structure of parasite assemblages. Many species of parasites cause observable lesions in their hosts, but their overall effect on health and population parameters is poorly understood. The study of parasites of carnivorous marsupials is embryonic, but offers considerable scope for future investigations.

    1. Page 399

      Marsupials are well represented in the New World with three orders. Here we discuss the conservation status of New World marsupials by considering their geographic ranges, association with major biomes and vegetation types, and the latest IUCN’s Species Survival Commission Red List assessment. We also discuss marsupial geographical distributions with respect to the global conservation strategy that focuses on global biodiversity hotspots. The IUCN Red List considers 23 New World marsupials, or about a third of all extant species, as threatened with extinction. We highlight that an additional 11 species with restricted ranges should also require timely attention. New World biodiversity hotspots contain 78% of threatened species recognised by IUCN’s Red List, as well as 80% of the remaining species whose conservation status may be in need of further revision. These hotspots comprise ecosystems that have already lost 70% or more of their native habitat, and are in most need of conservation attention.

    2. Page 407

      Dasyurid marsupials are distributed throughout the major terrestrial environments of Australia, but since European settlement have suffered local and regional extinctions, range reductions and population declines. In this paper we examine the conservation status of small dasyurids (<500 g) and the threats they face. We also evaluate recovery procedures for threatened taxa and assess their success. Twenty-four percent of smaller dasyurids are classified as vulnerable, endangered or data deficient. Large body size and occupancy of one or two habitat types are correlated strongly with endangerment; species currently considered as ‘low risk, near threatened’ group closely with vulnerable and endangered species, indicating a risk of further declines. The processes contributing most to declines include habitat loss and fragmentation, altered fire regimes and predation. As of April 2001, no Recovery Plans had been adopted by the Commonwealth Government for any small dasyurid species. There is much information on the reproduction and development of smaller dasyurids, making them suitable for captive breeding. However, captive breeding programs have been limited, the dibbler Parantechinus apicalis being the only species bred systematically for reintroductions. There is a need for integration between captive breeding programs and recovery planning, as well as for more information on the population viability and metapopulation structures of small dasyurids, genetic diversity of populations and inbreeding depression. We suggest a program of survey, research, management and education to improve conservation outcomes for all small dasyurids.

    3. Page 422

      Anthropogenic declines have been reported in all of Australasia’s eight larger marsupial carnivores (genera Dasyurus, Sarcophilus, and Thylacinus). One species is now extinct (T. cynocephalis), one subspecies is Endangered (D. maculatus gracilis), one species and one subspecies are classified as Vulnerable to extinction (D. geoffroii and D. m. maculatus, respectively), two species are Lower Risk – Near Threatened (D. hallucatus, D. viverrinus), two are of unknown conservation status (the New Guinea quolls, D. albopunctatus and D. spartacus), and only one species is classified as Lower Risk – Least Concern (S. laniarius). A successful recovery program has been executed for one species (D. geoffroii – formerly Endangered), which will shortly be removed from threatened fauna lists. While the causes of decline are multiple and complex, the overriding factor is probably habitat loss, degradation and fragmentation, and the associated factors: loss of protective cover from predators, reduced food availability, and increased contact with humans. Introduced predators, cats (Felis catus), foxes (Vulpes vulpes) and dingoes (Canis lupus dingo), play a not well understood role. Other important mortality factors include persecution, non-target poisoning and road mortality. In this chapter, we conduct a critical review and summary of 1) the causes and correlates of decline, 2) the issues related to trophic level and life history that predispose marsupial carnivores to anthropogenic decline, 3) a range of solutions, some of which are being implemented, and 4) future directions for the large amount of work still to be done.

    4. Page 435

      This chapter provides an account of a successful recovery program that has substantially improved the conservation status of a threatened dasyurid species. The chuditch Dasyurus geoffroii was once widespread throughout much of the Australian continent, and like many of Australia’s medium-sized mammals, has suffered a dramatic reduction in abundance and range. Predation by the introduced European red fox has been shown to be a significant threatening process in the conservation of the chuditch and other medium-sized mammals in Western Australia (WA). Competition with the fox for food may also be involved. By the 1980s, it was estimated that fewer than 6000 chuditch remained, all confined to the south-west of WA, most being in the jarrah forest. In 1983, the chuditch was listed as a threatened species in WA and in 1991 it was listed as an Endangered species under Commonwealth legislation. Consequently a recovery plan was prepared and a recovery team appointed to co-ordinate the implementation of six recovery actions considered necessary to improve the conservation status within 10 years. This review describes progress of these recovery actions. The abundance of chuditch in the south-west of WA has increased significantly and current logging and prescribed burning practices in the jarrah forest do not appear to detrimentally impact on this. Fox control using poisoned (1080) dried meat baits over large areas of the south-west of WA has been one of the major factors in the recovery of the chuditch. The ability to breed chuditch in captivity and to translocate the progeny to areas where they once occurred has also contributed to the recovery of this species. Based on this recovery, a review in 2003 may recommend a change in its conservation status and removal from State and Commonwealth threatened species lists.

    5. Page 452

      At the time of European settlement in Australia, the distribution of the numbat extended across much of southern Australia. By 1985 only two small populations survived, at sites 160 km apart in the south-west of Western Australia. The decline of the numbat coincided largely, although not wholly, with the spread of the fox (Vulpes vulpes) from its point of introduction in Victoria.

      During the early 1980s, an experimental fox control and numbat monitoring program compared changes in numbat abundance in baited and unbaited habitat at Dryandra Woodland. This experiment showed that growth in the numbat population followed the introduction of monthly fox baiting. In 1985 reintroduction of numbats into Boyagin Nature Reserve commenced in conjunction with regular fox control and a population persists there today.

      A program of fox control and numbat translocation has resulted in the establishment of six further reintroduced populations in Western Australia. Action by Earth Sanctuaries Limited has resulted in the establishment of two additional populations in fenced sanctuaries in South Australia and New South Wales. In 1994, an assessment of the status of the numbat under the existing IUCN Red List criteria recommended a change from Endangered to Vulnerable.

    6. Page 464

      Marsupial moles (Notoryctes) are amongst the most unusual, enigmatic and elusive vertebrates that inhabit Australia. They occupy the sandy deserts of central and north western Australia and are the most fossorial of marsupials, rarely venturing to the surface. While they have been known to Aboriginal peoples for thousands of years, and to science for over a century, very little has been documented about Notoryctes ecology, reproduction, or behaviour. The genus exhibits extreme morphological specialisation and shows a high degree of convergence with the unrelated eutherian golden moles (Chrysochloridae) which, like Notoryctes, are insectivorous and live underground in sandy deserts. While the phylogenetic affinities of marsupial moles are not yet clearly resolved, molecular and palaeontological studies confirm that they have been evolving separately from other marsupials for a very long time, perhaps from the beginning of the Tertiary, and that the genus warrants placement in its own order. Two species are currently recognised, N. caurinus in northwestern Australia and N. typhlops in central Australia, although the taxonomy of the group is currently under review. Both species are regarded as rare and endangered, although so little is known about this elusive animal that this even this is uncertain. In any case, there is grave reason for concern for the conservation of marsupial moles due to high predation by introduced foxes and cats, and to changes in their habitats resulting from introduced herbivores and changed fire regimes. We conclude that studies are urgently needed on the ecology of marsupial moles, and that these would benefit from collaboration with Aboriginal people who have knowledge of these animals’ habitats and habits.

    7. Page 475

      Conservation genetics is a relatively new field within both realms of conservation biology and molecular genetics. The application of molecular genetic principles to the conservation of any species of carnivorous marsupial is still in its infancy and there is extremely limited information available for many species. Indeed, there are probably a number of as yet unrecognised cryptic species which may only be identifiable through molecular means. In this chapter, I present an overview of some of the concepts of conservation genetics including the applicability of phylogenetics and phylogeography, population genetics, and molecular ecology to conservation. The limited number of case studies in which these principles have been applied to the conservation of carnivorous marsupials is highlighted and directions for future research are suggested.