Thesis Title: "The evolution of brain size and structure in birds." Thesis Title: "The evolution of skilled forelimb movements in carnivorans." Thesis Title: "An analysis of skilled forelimb movements during feeding in members of Macropodoidea." • Neuroecology of birds and mammals • Evolutionary neurobiology • Neurobehavioural effects of exposure to environmental contaminants

The mechanisms utilised by the newborn quoll to move from the uterus to the teat within the pouch are unknown. The ability to sense gravity and odour have been suggested and it is possible that temperature may also assist the young in this migration.2. An increasing temperature gradient was observed from the sinus at 28.98 degrees C increasing to 29.38 degrees C on the skin between the sinus and the pouch and further increasing to 30.96 degrees C within the pouch. This temperature gradient was not as apparent during lactation.3. Hairs may also play an important role in allowing the newborn to leave the gelatinous material emanating from the uterus and travel to the pouch. The hairs form a tunnel between the sinus and the pouch and may assist the young in the moving from uterus to the pouch. (c) 2005 Elsevier Ltd. All rights reserved.

Diversity in vertebrate brain size and composition is thought to arise from either developmental constraints that cause coordinated changes between brain regions or a mosaic model, whereby changes in individual brain regions are independent of changes in other brain regions. These two mechanisms were tested in birds using multiple regression analyses. Across 13 orders, significant correlations were present between some brain regions, but not all. Most of the correlated changes reflect the connectivity between different brain components, such that regions with the most interconnections are correlated with one another but not other brain regions. Whether mosaic changes are characteristic of brain regions or systems in birds, however, to our knowledge, remains to be investigated.

Secondary flightlessness has evolved independently many times in birds. Morphological changes in the pectoral girdle and flight feathers and changes in body size have been associated with the evolution of flightlessness, and it has also been suggested that flightless birds have relatively small brains. We therefore tested whether flightlessness is related to changes in relative brain size. Relative brain size was compared between volant and flightless species using both conventional statistics and modern comparative methods within nine taxonomic groups. No significant difference was found between flightless and volant species in six of these groups, regardless of whether body mass or tibiotarsal measurements were used as estimates of body size. Species with relatively smaller brains compared with their volant relatives were the great auk Pinguinus impennis , the kakapo Strigops habroptilus and some species of penguin. Thus, we found no evidence of a general correlation between the evolution of secondary flightlessness and the evolution of relatively small brains in birds. This suggests that neural requirements are not significantly different between flightless and volant species, although our methods may have overlooked subtle neurological changes that do not result in markedly different endocranial volumes.

Despite significant progress in understanding the evolution of the mammalian brain, relatively little is known of the patterns of evolutionary change in the avian brain. In particular, statements regarding which avian taxa have relatively larger brains and brain regions are based on small sample sizes and statistical analyses are generally lacking. We tested whether psittaciforms (parrots, cockatoos and lorikeets) have larger brains and forebrains than other birds using both conventional and phylogenetically based methods. In addition, we compared the psittaciforms to primates to determine if cognitive similarities between the two groups were reflected by similarities in brain and telencephalic volumes. Overall, psittaciforms have relatively larger brains and telencephala than most other non-passerine orders. No significant difference in relative brain or telencephalic volume was detected between psittaciforms and passerines. Comparisons of other brain region sizes between psittaciforms and other birds, however, exhibited conflicting results depending upon whether body mass or a brain volume remainder (total brain volume – brain region volume) was used as a scaling variable. When compared to primates, psittaciforms possessed similar relative brain and telencephalic volumes. The only exception to this was that in some analyses psittaciforms had significantly larger telencephala than primates of similar brain volume. The results therefore provide empirical evidence for previous claims that psittaciforms possess relatively large brains and telencephala. Despite the variability in the results, it is clear that psittaciforms tend to possess large brains and telencephala relative to non-passerines and are similar to primates in this regard. Although it could be suggested that this reflects the advanced cognitive abilities of psittaciforms, similar studies performed in corvids and other avian taxa will be required before this claim can be made with any certainty.

The grey-headed flying fox Pteropus poliocephalus, a nationally vulnerable species, is endemic to coastal eastern Australia, from Maryborough, Queensland, through New South Wales to Melbourne, Victoria. Pteropus poliocephalus forages at night, primarily on eucalypt blossom within 50 kin of traditional camps (day roosts), usually in dense, riparian vegetation. Several attempts have been made to track long-distance movements of P. poliocephalus across its extensive and climatically highly variable range, but the technology has been inadequate for tracking at the required speed and scale. Satellite tracking was used to monitor movements of a 2-year-old male P poliocephalus, trapped at the Currie Park camp in the north of the species' range (28degrees48'S) from January 2000, and another from its most southerly camp, Melbourne (37degrees50'S), from April 2000. The first flying fox moved camp 50 km noth-east to Dallas Park, where it stayed until March, then transited at least 15 other camps between 28degrees 12'-32degrees44'S, before returning to Dallas Park in September. The second flying fox remained in Melbourne until July, then transited at least six other camps to 33degrees44S, before returning to Melbourne in January 2001. Both animals made round trips of at least 2000 km, traversing > 4degrees latitude, before returning to their camps of origin, while other animals remained. The study suggests that P poliocephalus is a partial migrant that uses winds to facilitate long-distance movements, and underlines the importance of management at a national scale.

The possible relationships between relative brain size and developmental mode and between relative brain size and five measures of the length of the development period were tested across over 1400 species of birds. Using both conventional statistics and phylogenetically based comparative methods, significant differences in relative brain size were detected among modes of development. Across all species, there were significant relationships between relative brain size and each of the following developmental traits: incubation period, age of fledging, duration of postfledging parental care, and total period of parental care. In contrast, the age of first flight was not significantly correlated with relative brain size. The relationships between these five developmental traits and relative brain size varied among developmental modes and orders such that significant relationships were present within some modes and orders but not in others. Thus, developmental differences play a significant role in the evolution of brain-size differences, but the role depends upon the taxonomic level being investigated. This is likely due to the differential lengths of periods of neural and behavioural development in young birds. Our conclusions support the contention of previous studies that developmental differences have played a key role in avian brain evolution.

Endocranial volumes of vertebrate skulls and brain masses are often used interchangeably in comparative analyses of brain size. We test whether endocranial volume can be used as a reliable estimate of brain size in birds by comparing endocranial volumes with brain masses across 82 species using absolute values and with respect to body size. The results of paired tests across all 82 species and within two orders, Passeriformes and Psittaciformes, did not yield a significant difference between the two measures. These results were supported by correlational analyses that showed a significant positive relationship between endocranial volume and brain mass. Unpaired tests within short-tailed shearwaters (Puffinus tenuirostris) and paired tests within budgerigars (Melopsittacus undulatus) also yielded no significant differences between endocranial volume and brain mass. Thus, a combination of interspecific and intraspecific comparisons indicates that endocranial volume does provide a reliable estimate of brain size. Although this may enable more rapid collection of avian brain size data, endocranial volume should be used with caution because it cannot account for seasonal and age-related variation and cannot be used to measure differences in brain structure.

Variation in relative brain size was examined in 55 species of waterfowl (Anseriformes). Using both conventional statistics and phylogenetically based comparative methods, the extent of variation in relative brain size and possible relationships with mode of foraging and diet were examined. The results indicate that although brain size does vary considerably between closely related species of waterfowl, it is not reliably related to either foraging mode or diet. There are a number of possible reasons for the lack of relationships between brain size and foraging mode and diet. Firstly, subtle changes in foraging mode and diet may favor relatively large changes in brain size. Secondly, foraging mode and diet could be correlated with the expansion of an individual brain region without affecting overall brain size. Thirdly, other behavioral/ecological traits may be more important with respect to brain size evolution in waterfowl. For example, the relatively large brain of the musk duck (Biziura lobata) and altriciality of their young in comparison to other stiff-tailed ducks (Oxyura spp.) indicates that developmental rate plays a significant role in the evolution of brain size. Given the difference between our results and that reported in inter-order comparisons of brain size in birds, further research is required into other avian orders to assess how brain size and behavior might be related within orders as well as between them.

It has been hypothesized that play is more likely to be present in larger brained species. We tested this hypothesis in mammals using independent contrasts, a method that controls for phylogenetic relatedness. Comparisons across 15 orders revealed that the prevalence and complexity of play was significantly correlated with brain size, with larger brained orders having more playful species. Three orders, Rodentia, Marsupialia, and Primates, were used for within-order comparisons among species and, where possible, among families. The comparisons were not significant for rodents or primates, and those for marsupials yielded inconsistent results. Therefore, although a strong relationship is present at the highest taxonomic level of comparison, it diminishes or evaporates at lower level comparisons.

The ghost bat, Macroderma gigas, and the orange leaf-nosed bat, Rhinonycteris aurantius, occupy similar ranges across northern Australia and are often found in the same roost caves. Both species are considered rare and vulnerable to further population decline. A third small species, the large bent-wing bat, Miniopterus schreibersii, has a similar body mass to R. aurantius, but has one of the largest ranges of any Australian mammal. In the present study we examine the effect and sensitivity of the animals' roosting microclimates on their energy and water balance. M. schreibersii exhibits a basal metabolic rate about 40% greater than other bats of similar body mass, whereas the other two species are close to predicted levels. R. aurantius shows a decrease in body temperatures below thermoneutrality. R. aurantius has levels of pulmocutaneous water loss among the highest seen for a mammal, and calculations based on nasal tip temperatures suggest that most of this loss is across the skin. Calculated ambient temperatures at which metabolic water production is equal to pulmocutaneous water loss in dry air are -14.7 degrees C for R. aurantius, 9.8 degrees C for M. schreibersii and -0.3 degrees C for M. gigas. Exposing the animals to relative humidities of between 80% and 90% shifted these calculated temperatures to 5.6 degrees C, 25.2 degrees C, and 2.9 degrees C, respectively. For each species the ratio of metabolic water production to evaporative water loss has been treated as a joint function of humidity and ambient temperature. The resulting surface plot shows that under known roosting conditions in caves R. aurantius and M. schreibersii remain in positive water balance, whereas M. gigas does not.

A behavioural index of forelimb dexterity and comparative statistics were used to analyse the relationships between proximal (shoulder, upper and lower forelimb) and distal (wrist, forepaw, digits) forelimb dexterity and four aspects of brain morphology (overall brain, cortex, cerebellum and telencephalon sizes) in 18 species of marsupials. On the basis of the principle of proper mass, it was expected that an increase in forelimb dexterity (either proximal or distal) would be positively correlated with the size of the brain and the three brain components. Using independent contrast analysis to remove the effects of phylogeny revealed three significant correlations between: cortex size and distal dexterity, cerebellum size and proximal dexterity, and telencephalon size and distal dexterity. The relationship between cortex size and distal dexterity was subsequently corroborated by Spearman rank correlations. These results suggest that the execution of finely coordinated forelimb movements may not be dependent upon overall brain size, but may be dependent upon the size of brain components, thus supporting the principle of proper mass.

Two visual areas, V1 and V2 (first and second visual areas), appear to be present in the posterior neocortex of all eutherian mammals investigated so far. However, previous studies have not established whether an area homologous to V2 also exists in metatherian mammals (marsupials). Using electrophysiological techniques, we mapped the visual receptive fields of neurons in the striate and peristriate cortices of the northern quoll, an Australian marsupial. We found that neurons in a 2-mm-wide strip of cortex rostrolateral to V1 form a single, relatively simple representation of the complete contralateral hemifield. This area resembles V2 of eutherians in several respects: (i) neurons in the medial half of the peristriate area represent the lower visual quadrant, whereas those in the lateral half represent the upper visual quadrant; (ii) the vertical meridian of the visual field is represented adjacent to V1, while the visual field periphery is represented along the lateral and rostrolateral borders of the peristriate area; (iii) there is a marked anisotropy in the representation, with a larger magnification factor parallel to the V1 border than perpendicular to this border; and (iv) receptive fields of multiunit clusters in the peristriate cortex are much larger than those of cells in V1 at comparable eccentricities. The cortex immediately rostral and lateral to V2 did not respond to visual stimulation under our recording conditions. These results suggest that V1 and V2 together form a 'core' of homologous visual areas, likely to exist in all therian mammals.

This study examined manual dexterity and skilled limb movements in two species of tree kangaroos (Dendrolagus lumholtzi and D. matschiei) in order to evaluate the claim that claws are detrimental to object handling and the view that all marsupials conform to a ‘typical’ set of movements. The tree kangaroos demonstrated two main differences from previously studied species: (1) a high degree of freedom of movement of the upper forelimb; and (2) in one species (D. matschiei), some independent digital movement. The two species differed from one another in the mode of picking up food items and the type of grasp used. These differences appear to be linked to differences in feeding and foraging strategy between the two species, as well as to anatomical differences. This study indicates that marsupials do not have a common set of skilled forelimb movements and that claws do not impede prehension.

The development of the mesonephros and metanephros was examined in the marsupial Northern native cat, Dasyurus hallucatus, from birth through to the end of lactation. The mesonephros was present at birth, reached a maximum volume 11 days after birth and had regressed completely by day 30. The metanephros was present on day 2; glomeruli were first seen on day 8, and nephrogenesis continued until day 89 post partum. The newborn native cat is at a very primitive stage of development compared to other marsupials. However, at weaning, like other marsupial species, the native cat has developed a fully morphological and functional urinary system.

The development of righting was studied in the young of Dasyurus hallucatus, a small marsupial from northern Australia. Young were tested from birth to weaning. Righting began at 40 days, when tactile input on the snout triggered rotation to prone. Over the next 15-20 days, asymmetrical tactile input on the body triggered righting movements by the hindlegs (and later by the forelegs). Vestibular righting reflexes developed after these tactile righting reflexes. Furthermore, asymmetrical vestibular righting (i.e., when the young are held laterally in the air) developed before symmetrical vestibular righting (i.e., when held downward by the pelvis or placed supine in water). Vestibular righting triggered by falling supine in the air did not develop until about 80 days. This study further demonstrates that righting behavior does not consist of a single, integrated motor pattern, but a suite of motor patterns having independent control mechanisms and patterns of development.

The influence of afferent input on the survival of target neurons in mammals has been examined by the removal of one eye of pouch young of the marsupial native cat (Dasyurus hallucatus). The ages at eye removal spanned the period of neurogenesis of the ascending visual pathway, and were earlier than the time of maximal axon number in the optic nerve. Autoradiography following the injection of tritiated proline into the intact eye of adult animals shows that the lateral geniculate nucleus contralateral to the injected eye of the earliest enucleates retains its laminated structure, despite the total absence of binocular competition throughout development. However, we find a dramatic, age-related reduction in the volume of those parts of the lateral geniculate nucleus and superior colliculus which would normally receive a contralateral-only projection from the enucleated eye. The effects of the enucleation are not restricted to the primary termination sites of the optic axons but ramify throughout a large part of the neo- and archicortex.

The pattern of retinogeniculate connections has been examined in a range of diprotodont marsupials, including wallabies, possums, forest wombat and koala. The lateral geniculate nucleus (LGd) in most species has alternating bands of ipsilateral and contralateral retinal terminal fields, with considerable interspecific variability. The number of terminal bands of retinogeniculate input varies from eight to eleven in most species, with little binocular overlap. By contrast in one species, the feather-tailed glider, the most significant feature of LGd organisation is binocular overlap. No relationship is apparent between LGd organisation and the life-style of the animals.

Behavioural estimates of the visual acuity of the Northern Native Cat (or Northern Quoll) - Dasyurus hallucatus - were made using the Mitchell jumping stand technique. A value of 2.3–2.8 cycles per degree was obtained. This functional acuity compared well with predictions based on the peak ganglion cell density (2600 cells/mm2) determined from the retinal ganglion cell density map.