Barbara Bowen

Pollen dispersal is the step in higher plant mating systems over which the parent plant has the least control as it is dependent on the vagaries of weather conditions (anemophily) or animal behaviour (animal pollinator activity). While many families have passive release from the anther the Proteaceae has a diversity of pollen dispersal methods. Flowers from a range of species in the Proteaceae, covering the majority of genera from each of the four main subfamilies, were examined to determine how pollen is dispersed and to gain an overall view of how male function varies within the family. This provides a basis for predicting the degree of the six likely fates of pollen released in this family. Only one group (subfamily Persoonioideae) and four genera in one other subfamily (Proteoideae) dispense pollen directly from the anthers onto a flower visitor. Five genera in the Proteoideae have explosive pollen release, while Symphionema may require vibration to release the pollen. All the remaining Proteoideae genera, the single species of Bellendenoideae, and all genera of the Grevilleoideae (except Sphalmium) have a pollen presenter where pollen is dispersed from the style of the flower. Ancestral Proteaceae were likely to have been insect pollinated and had relatively small flowers. Taxa with explosive pollen release may have evolved early in the family and may have been more abundant early in the fossil record. However, the taxa with pollen presenters became much more abundant throughout the Tertiary when many developed robust gynoecia that can accommodate larger vertebrate pollinators.

The black-flanked rock-wallaby (Petrogale lateralis lateralis) is a threatened species, once widespread throughout Western Australia but now restricted to disjunct populations including those of Cape Range National Park. It is a herbivore with a foraging range restricted to rocky outcrops and, as such, may be impacted by competition for resources from other native or introduced herbivores. This study compared the diet of the black-flanked rock-wallaby with those of co-occurring species, the euro (Macropus robustus erubescens) and the feral goat (Capra hircus), from two gorges at Cape Range National Park, to determine whether there is overlap in dietary niches. Diet composition was determined using microhistological analysis of faecal pellets in comparison with reference plant material. The black-flanked rock-wallaby diet consisted predominately of browse/forbs with some variation across seasons (63% in summer months 74% in winter months) this overlapped significantly with the diet of goats (Schoener Index: 0.79-0.88), but not euros (SI: 0.41-0.57), whose diet consisted predominantly of grasses (72-78%). There was, however, a significant overlap in the consumption of plants with stellate hairs for all three herbivores (SI: 0.89-0.98). Dietary overlap between rock-wallabies and goats may indicate a mechanism for competition, supporting continuing measures for reduction of goat numbers in Cape Range National Park.

The ancient Gondwanan family Proteaceae has its greatest speciation in fire-prone environments of Australia. Fire response is either by seedling recruitment from parent plants that succumb to fire (obligate seeders), or survival and resprouting from protected buds (resprouters). Starch is the main source of energy for resprouting and in roots is restricted to parenchyma tissue. This study compared the size and distribution of storage parenchyma and the magnitude of starch reserves in roots of several proteaceous species from different genera in relation to their fire response and taxonomy. Cross-sections (2 μm) of roots of 51 resprouter and 42 seeder species from 12 genera were stained for starch. Areas of cortex and ray parenchyma along with starch grain density were measured using image analysis software (Assess 2.0) and comparable samples of root tissue were assayed chemically for starch. Starch, where present, predominated in ray and cortex tissue with a greater percentage in resprouters (13.4 ± 1.03) than seeders (1.8 ± 0.26); these results correlated significantly with the chemical assay for starch (r = 0.93, P < 0.0001). Resprouters also had more storage parenchyma (56.9 ± 1.72%) than seeders (41.9 ± 1.91%) mostly due to broader rays (17.5 ± 1.22%) compared with seeders (8.2 ± 0.16%). Percentage of cortex tissue was similar for seeders and resprouters (39.4 ± 2.24 and 33.7 ± 2.04 respectively). Anatomical preferences for storage site were consistent within genera and broad suprageneric groupings. This study shows that histological analysis of root starch is a reliable predictor of resprouting capacity in Proteaceae and that patterns of storage tissue within genera, together with the persistence of parenchyma devoid of starch in seeders, are consistent with response to fire and suggests homoplastic evolution of this response within the family.

Fires are features of ecological communities in much of Australia; however, very little is still known about the potential impact of fire on plant diseases in the natural environment. Phytophthora cinnamomi is an introduced soil-borne plant pathogen with a wide host range, affecting a large proportion of native plant species in Australia and other regions of the world, but its interaction with fire is poorly understood. An investigation of the effects of fire on P. cinnamomi activity was undertaken in the Stirling Range National Park of south-western Australia, where fire is used as a management tool to reduce the negative impact of wildfires and more than 60% of the park is infested with, and 48% of woody plant species are known to be susceptible to, P. cinnamomi. At eight sites confirmed to be infested with P. cinnamomi, the proportion of dead and dying susceptible species was used as a proxy for P. cinnamomi activity. Subset modelling was used to determine the interactive effects of latest fire interval, average fire interval, soil water-holding capacity and pH on P. cinnamomi activity. It was found that the latest and average fire interval were the variables that best explained the variation in the percentage of dead and dying susceptible species among sites, indicating that fire in P. cinnamomi-infested communities has the potential to increase both the severity and extent of disease in native plant communities.

This study investigates whether tree decline in Eucalyptus gomphocephala (tuart) is associated with the functional diversity of soil bacterial communities. We selected 12 sites with different stages of decline and assessed crown health [Crown density (CD), Foliage transparency (FT), Uncompacted live crown ratio (ULCR), Crown dieback ratio (CDR) and Epicormic index (EI)] and soil bacterial functional diversity based on Biolog EcoPlates™ incubation [Average well colour development (AWCD), Shannon diversity (H'), richness (S) and Shannon evenness (E)]. Crown health indices differed between sites with EI being the most robust indicator of decline in crown health followed by CDR and CD (P<0.05). Soil bacterial indices collected at 0-10 and 20-30. cm soil depth between December (summer, dry season) and May (autumn, start of wet season) differed between sites (P<0.05), and significant relationships between crown health indices, except ULCR, and all soil bacterial indices were observed. Principle component analysis (PCA) showed that a decrease in the utilization of carbohydrates, carboxylic acids, amino acids and amines by the soil bacterial communities correlated to sites with poor crown health, indicating some changes in physiological responses of bacterial groups with declining tree health. Using stepwise regression analyses, in the 0-10. cm soil layer in December, itaconic acid had a 46% contribution to the EI. Carboxylic acids, including itaconic acid, have a strong ability to solubilize soil minerals in calcareous soil, and these possibly increased the availability of soil mineral nutrients in the healthier sites compared to the declining sites, particularly in the dry season. In addition, lack of soil water in the declining sites limited soil bacterial diversity and was positively correlated with EI in the 0-10. cm soil layer in December. In conclusion, soil bacterial functional diversity has a strong relationship with tuart decline and the importance of soil microbes in tuart ecosystem health must be considered in the future.

Management interventions are needed to reverse the decline of Tuart (Eucalyptus gomphocephala) woodland in the Yalgorup area of south-west Western Australia where the largest intact remaining example of this ecosystem is located. Although the cause of the decline is uncertain and several factors may be involved, management action should not be withheld because the decline process is not fully understood. We contend that the reduction in fire frequency over the last 50 years has led to an increase in understorey density, particularly of Western Australian Peppermint (Agonis flexuosa), resulting in greater competition for resources, which may in turn have increased the susceptibility of healthy woodland to decline. In contrast to Tuart regeneration, which is usually tied to fire, Western Australian Peppermint can establish readily in unburnt woodland. Further, once Western Australian Peppermint seedlings develop to the lignotuberous stage, they can resprout vigorously after fire. Therefore, a combination of fire and the physical removal of understorey in sites where this species has formed extensive thickets is required to: (i) provide an opportunity for regeneration of Tuart in both healthy and declining stands; (ii) improve the chances of sustained recovery of Tuart trees in declining stands; and (iii) ensure heterogeneity in the vegetation at multiple scales, a recognized strategy for conserving biodiversity and increasing ecosystem resilience. We propose that this approach may also be relevant to other tree decline syndromes in southern Australia. However, fostering community support for active intervention using thinning and fire in conservation reserves and staging the operations within an experimental framework will be important for such action to gain both the social and scientific acceptance necessary for it to be applied widely.

During the late 19th and early 20th centuries, frustrated pastoralists in South Australia, New South Wales and Queensland fenced their properties to stop the Australian dingo (Canis lupus) killing their livestock. Eventually links formed between individual fences, creating a continuous barrier now covering over 5000 km. Known variously as the Dingo Fence ‘Dingo Barrier Fence or ‘Wild Dog Fence it is of wire mesh standing 1.8 in high, with a further 30cm buried. There is a 5 m wide cleared buffer on each side and the entire structure is well maintained. On the New South Wales side of the fence where sheep are the main livestock, dingoes are controlled and numbers are low, whereas on the South Australian side dingoes are tolerated alongside cattle husbandry. Unintentionally, the fence initiated a large-scale experiment, allowing biologists to assess the biological consequences of removing a large predator. Alan Newsome and his colleagues from the Commonwealth Scientific and Industrial Research Organisation (CSIRO) studied vertebrate abundances on either side of the fence by counting animal tracks at stock watering points. One striking finding was that the introduced predator the red fox (Vulpes vulpes) was present on both sides, but in higher numbers in the absence of dingoes. Dingoes eat foxes and drive them away, explaining the differences in fox numbers across the fence. Foxes threaten several native Australian mammals (see Chapters 2 and 16), including many borrowers and diggers important in soil turnover, nutrient cycling and dispersing plants. It may be that, by regulating fox numbers, dingoes are protecting native fauna that in turn modify the environment to the benefit of many soil organisms. The case of the Dingo Barrier Fence shows how relationships between organisms determine the range and relative abundance of species.

It has been an exciting 12 months since the last update on the CPSM. Since that time there have been 3 PhD, 1 Masters and 3 Honours student completions. There have been new species of Phytophthora described from our native woodlands with many more waiting to be described, we have fished successfully for Phytophthora throughout Western Australian waterways, we've learnt how to eradicate P. cinnamomi from some of our native ecosystems across Australia, how some of our native plants respond to treatment to phosphite and how this treatment can be optimised, how satellite and airborne remote sensing technologies can be used to monitor and detect decline in tree health, the effect of Phytophthora on our remarkable fauna, and how P. cinnamomi appears to survive in black gravel soils of bauxite mines. Newly initiated projects will lead to exciting discoveries, including the sequencing of the entire P. cinnamomi genome, the interaction between P. cinnamomi and woylies and bandicoots, the potential spread of P. cinnamomi by feral pigs, the susceptibility of Australian native plants to P. ramorum, the biology, ecology and management of Phytophthora spp. in nurseries, and the mechanisms of survival of P. cinnamomi in a range of different soils in our natural ecosystems. Finally, the CPSM was instrumental in a successful bid for the State Centre of Excellence for Climate Change and Woodland and Forest Health with 27 national and international collaborating partners and this will be hosted by Murdoch University and have a close alliance with the CPSM. This talk will give a brief overview of some of these exciting research developments.