Biodiversity publications archive

Biodiversity and Fire: The effects and effectiveness of fire management

Proceedings of the conference held 8-9 October 1994, Footscray, Melbourne
Biodiversity Series, Paper No. 8

Biodiversity Unit
Department of the Environment, Sport and Territories, 1996

10. Effects of fuel reduction burning on fauna in a dry sclerophyll forest

Kevin Tolhurst,
Fire Research, Dept. Conservation & Natural Resources, Victoria

10.1 Abstract

This study looked at the effects of low intensity spring and autumn burning on species composition and population structure of birds, bats, small terrestrial mammals, reptiles and epigeal invertebrates in a dry sclerophyll foothill forest in north-central Victoria.

Birds were little affected by burning in either season because no habitat was completely removed and no new habitats were formed by the fires. Birds were easily able to avoid the immediate effects of the fire and returned to the burnt areas soon after burning. Bats were the most diverse group of mammals in the area but no conclusions could be drawn on the effects of fire on them because of problems of sampling them effectively. Small terrestrial mammal populations were reduced by burning, but had recovered to pre-burn levels after two breeding seasons. Small unburnt patches within the fire area were important habitat for skinks and other ground foraging animals. Unburnt patches were significantly associated with logs and branches on the ground. Spring burning had a greater impact on litter and soil invertebrates. The abundance of some groups was reduced after spring burning for a period of up to two years after burning.

Key words: fuel reduction, sclerophyll forest, vertebrates, low-intensity burning, invertebrates, bats, reptiles, mammals, birds, Victoria

10.2 Introduction

Fire ecology literature is largely based on observations of different ecosystems following burning. Very few studies have included preburning observations, so there has been a heavy reliance on the study of fire effects by comparing areas with different burning histories (mainly comparing recently burnt areas with nearby unburnt areas). The inherent problem with this approach is accounting for the natural variation in population densities between areas.

Effects of fire on forest vertebrates have been studied mainly in relation to single events (wildfires or fuel reduction burns), and mammals have been the main animals studied (e.g. Catling & Newsome 1981; Christensen et al. 1981; Heislers 1974; Newsome et al. 1975; Recher et al. 1974; Suckling & Macfarlane 1984).

Small ground mammals are the best studied faunal group in relation to fire responses. Very little is known about arboreal mammals, macropods, large ground mammals and invertebrates. Virtually no data exist on the effects of burning on reptiles, amphibians or bats, even though these species may be quite common. More is known about the effects of fire on birds, and this has shown that bird populations change with changes in habitat.

The results presented here relate to the fauna of Wombat State Forest, which is less diverse than some other forest types such as the box/ironbark forests in dryer country north of the divide. The forest has been previously disturbed by extensive wood harvesting activities associated with gold mining in the late 19th century and a long history of previous fires.

This study aimed to improve the knowledge of the response of all faunal groups to low-intensity burning. To achieve this, techniques were developed or refined to determine the population size, community composition, and seasonal and site variation of each faunal group. Further development is needed, particularly for reptiles, bats and invertebrates.

10.3 Vertebrates

10.3.1 Birds

Review and methods

Birds are a conspicuous and important element of Victoria's forest fauna and a major predator of forest insects. Their response to different burning regimes needs to be understood and managed.

The effects of fuel reduction burning on forest birds have been studied in south-west Western Australia by Kimber (1974) and Christensen and Kimber (1975), and in Victoria (Wombat Forest) by Cowley (1974). The approach of looking at areas burnt under various regimes in the past (retrospective studies) has been used by Woinarski (1990) in tropical woodlands of northern Australia.

The present study offered an opportunity to collect information on the responses of bird populations to low-intensity prescribed burning in a particular forest type, using a properly replicated design using unburnt treatments as controls. An important advantage over previous studies was that detailed information would be collected simultaneously on vegetation, fire behaviour and other animals including invertebrates. Another advantage was that burns were planned to be repeated over time, at two different frequencies.

This summary reports selected results after a single prescribed fire. Further analysis is needed to relate these results to changes in other parameters, although preliminary observations are offered. The results deal with just three of the study areas (Blakeville, Kangaroo Creek and Burnt Bridge) as only baseline data have been collected from the other two (Barkstead and Musk Creek). A complete account of methods and results presented here is given in Loyn et al. (1992). Bird populations were assessed at various times before and after treatment by an area search technique (Loyn 1986, Hewish & Loyn 1989). Unmarked areas were searched for 20 minutes, and numbers of all birds seen or heard were recorded. Counts were made to assess the bird fauna in winter and summer. No counts were made in heavy rain nor strong wind. Data collection will continue and future results will provide information on the effects of repeated prescribed fires in spring and autumn.

Results and discussion

Bird abundance and species composition remained remarkably stable on burnt and unburnt areas. The main changes observed after burning were influxes of some species which fed from recently exposed soil, small decreases in some species inhabiting understorey, and influxes of some species (e.g. scarlet robin, flame robin) which fed on eucalypt nectar. The latter is of interest, as such influxes have not been reported in previous studies of prescribed burning. Quantitative observations show that the flowering of messmate stingy bark in 1987 was quite exceptional, and lasted from late March to early May throughout the study areas. Yet the main concentrations of nectar feeding birds were much more localised. The areas where concentrations were found coincided with areas that had been burnt the previous spring (Burnt Bridge) or more recently (a fortnight before the counts at Kangaroo Creek). This suggests that the burns may have affected the quality of flowering (e.g. nectar flow or nutritional value) and birds responded to that rather than simply to the amount of flowering. More data are needed to establish that this apparent link with burning is not coincidental and to determine whether any effects persist in the long term.

The other changes recorded on the three study areas are similar to those observed by Cowley (1974), in a swampier, shrubbier part of this same forest, and by Kimber (1974) and Christensen and Kimber (1975) in Western Australia. Cowley (1974) studied survival of banded birds and his observations relate mainly to nine species inhabiting the understorey. Numbers of understorey birds declined initially in all studies, although Kimber (1974) found that they exceeded previous levels after two years and were then higher than in adjacent forest that had been unburnt for 40 years. These results all relate to the structure of the forest understorey.

Some important caveats must be made if results are to be extrapolated elsewhere. Firstly, the study areas were small (approximately 10 ha/site) and no part of any burnt site was more than 300 m from unburnt forest. The burns were well conducted and some shrubs remained intact (albeit often scorched or killed), as did the eucalypt canopy. Hence birds were easily able to survive the burning and move back into habitat as it suited them. It is not known whether adjacent unburnt forest was a necessary part of the habitat for any species. This should not be seen as a design defect of the study, because it may reflect a common situation where prescribed burns are carried out on a small-scale mosaic pattern. But it does not reflect the equally real situation where prescribed burns are carried out on a largescale, with greater variation in fire intensity due to site and weather conditions over the period of the burn. Cowley (1974) considered that most prescribed fires in the Wombat State Forest were so patchy that they would have little effect on birds, and he designed his experimental burn to be more complete than usual. Nevertheless, at least two thirds of his banded birds survived and remained on the burnt study areas.

Another qualification arises from the nature of Wombat State Forest. Before treatment, the areas contained scattered shrubs (which were less numerous after burning) and areas of bare ground (which were larger and more numerous after burning). Hence there were no dramatic changes in habitat as a result of burning; no obvious habitats were lost or created. In shrubbier forest, birds such as scarlet robins and buff-rumped thornbills would usually be absent and enter only after burning (Loyn 1985; Christensen 1974), whereas here they remained present throughout. Flame robins have been observed elsewhere to increase after wildfire (Stokes 1975; Reilly 1991). The birds that did appear to enter the study areas after burning (e.g. white-winged chough) were also present initially in nearby forest, and were just responding to burning on a different spatial scale to the smaller birds. The results of this study must therefore be viewed in the context of the initial vegetation structure, the relatively short-term period of study and the size of the
burnt areas.

10.3.2 Bats

Review and methods

In Australia, less is known about bats than any other order of mammals, though there have been great advances in the last decade. They were overlooked in mammal surveys, despite the fact they often occur in large numbers and may be more diverse than other mammal groups present. The major reason for this lack of knowledge is that bats are generally difficult to catch or observe. The techniques for catching bats in flight have particular limitations, and researchers can put in much effort for few results. These limitations are due to the variety of foraging behaviour exhibited by different bat species, with some species foraging above the canopy, some below and some even taking insects from the ground (McKenzie 1983). Because of these limitations, early research on Australian bats was mainly restricted to the study of colonial or cave-dwelling species with easily accessible roosts (Hall 1981). Bats are included in most current studies of forest wildlife in Victoria (e.g. Squire 1990; Earl & Lunt 1989), but there is still a dearth of basic knowledge about their ecology.

The aim of studies over the past three years has been to provide information on the bats of the Wombat State Forest to understand more readily
the possible impact of particular fire regimes if widely applied. Comparisons are difficult to make for bats as their home range may cover several treatment areas. This work is covered more fully in Kemp et al. (1992).

Results and discussion

Seven bat species were identified (Table 1), making bats the most diverse group of mammals in the study area. These species represented seven out of the ten species of the Microchiroptera that were recorded within a sixty kilometre radius of Ballarat (Gilmore et al. 1979; Parnaby 1977).

All species identified were aerial insectivores, with the lesser and Gould's long-eared bats being foliage gleaners as well. The abundance of these bats was therefore likely to depend on the insect population in the forest understorey and tree canopy as well as availability of sites for roosting and breeding.

Table 1: Bat species recorded from the Blakeville, Barkstead and Musk Creek FESAs
Scientific Namea Common Name
Tadarida australis White-striped mastiff-bat
Nyctophilus gouldi Gould's long-eared bat
Nyctophilus geoffroyii Lesser long-eared bat
Chalinolobus gouldii Gould's wattled bat
Chalinolobus morio Chocolate wattled bat
Eptesicus darlingtonii Large forest eptesicus
Falsitrellus tasmaniensis Great pipistrelle

Note: a Nomenclature follows Menkhorst (1989)

Data on the comparative abundance of bat species were not conclusive as the trapping methods were selective and air temperature at the time of assessment was critical to the results (Fig. 10.1).

Figure 10.1: Relationship between bat activity and air temperature showing the importance of monitoring temperature during bat surveys.

Figure 10.1: Relationship between bat activity and air temperature showing the importance of monitoring temperature during bat surveys.

Insufficient data were collected to comment on the effects of fuel reduction burning on bats. Research into the effects of burning on bats is difficult and will require improvements in assessment methodologies and careful experimental design before any conclusions about population changes can be drawn.

10.3.3 Small Mammals

Review and methods

The effects of fire on fauna have been reviewed by Suckling and Macfarlane (1984) who found that little attention has been paid to the effects of repeated burning or the season of burning on fauna. These factors may have a significant effect due to the seasonal fluctuations in small mammal populations and their usually well-defined breeding periods. Many studies of prescribed burning have looked at differences between sites with different time since fire rather than changes on the same sites subjected to different fire regimes. The conclusions drawn from these studies need to be considered with caution; differences in fauna may be due to climatic differences following burning, site differences, pre-burning differences or the effect of the burning being studied. Any of these may significantly influence plant regeneration and therefore faunal habitat.

This summary describes the effects of single spring and single autumn prescribed fire on brown antechinus and bush rats at three of the FESAs. A detailed account of the study is given in Humphries and Tolhurst (1992). Data collection and analysis is continuing and future results will provide information on the effects of repeated prescribed fires in spring and autumn.

Results and discussion

A single prescribed fire in either spring or autumn had little effect on brown antechinus populations. Populations were initially reduced as a result of direct mortality in the fire, probable predation and some emigration from the burnt areas in the first few months after treatment. The high breeding potential of brown antechinus, and an adequate availability of nesting sites, combined to allow populations to recover to pre-burning levels within two breeding periods after spring burning, but they had not recovered after a single breeding period following autumn burning.

Populations of bush rats took three breeding seasons to recover when between 50 per cent and 80 per cent of their habitat in gullies was burnt in spring or autumn, but there was no recovery in the same period when all the habitat was burnt. When less than 50 per cent of bush rat habitat was burnt, populations recovered after a single breeding season. The variability in fires, as indicated in this study, highlights the importance of both the large and small-scale burning mosaic as limiting factors for the short-term survival and subsequent recovery of some species.

Recovery of both species occurred within a two year period of spring or autumn burning when about 40 per cent of gully vegetation and about 10 per cent of the mid-slope ground cover was left unburnt. This rapid recovery was attributed to the successful survival and breeding of individuals during this period in unburnt patches. Because the animals remain within defined ranges except when dispersing, recolonisation of areas totally burnt by fire would occur only when their habitat had recovered sufficiently to provide adequate food, shelter and breeding sites. The results suggest that fire management prescriptions should aim to minimise the burning of gully vegetation and burn no more than 7080 per cent of an area overall if rapid recovery (within three years) of bush rats and brown antechinus is a management objective.

10.3.4 Reptiles

Review and methods

Less is known about the response of Australian reptiles to wildfire or prescribed burning than is known about the response of birds or small mammals. The few studies reported in the literature compare species assemblages in areas of known burning history, but without knowledge of species complements before the fire event. These studies have concentrated on the reptilian fauna of central Australia (Cogger 1969; Fyfe 1980; Caughley 1985) and northern Australia (Braithwaite 1987), areas noted for their reptilian diversity. Few studies have been conducted in the temperate regions of south-eastern Australia where prescribed burning is a common and frequently used management tool in forest ecosystems. These environments have low reptilian species richness (Rawlinson 1971). Both fine and heavy fuels temporarily removed by prescribed burning, are parts of reptile habitat required for thermoregulation sites, shelter, food; and oviposition sites (Rawlinson 1971).

This summary documents the short-term responses of eight common scincid lizards to single spring and autumn low-intensity prescribed fires. Full details of the methods and results are given in Humphries (1992). Data collection will continue and future results will provide information on the effects of repeated prescribed fires in spring and autumn.

Results and discussion

The abundance of southern water skinks remained relatively stable following both spring and autumn burning because their primary habitat, fallen logs and branches, had not been significantly affected in terms of abundance or condition. Coventry's skinks and McCoy's skinks forage on and in the ground/litter layer, the microhabitat most affected by burning. The relatively high recorded abundances of these species after burning were probably due to the increased ease of detecting them as a result of reduced cover. The abundances of skinks at different times since fire are shown in Figure 10.2.

Figure 10.2: Skink abundance at the Musk Creek FESA at different times since fire in the spring, autumn and control treatments.

Figure 10.2: Skink abundance at the Musk Creek FESA at different times since fire in the spring, autumn and control treatments.

The few studies of the effects of burning on reptiles in southern Australia have dealt with the survival of particular species after high-intensity wildfire (e.g. Lunney et al. 1991; Newsome et al. 1975). In the present study observations were made of species surviving or being killed by low-intensity fire. During the autumn burn at the Barkstead FESA in April 1987, the comparatively strong-limbed species, southern water skink, was observed moving ahead of the fire front, which had a forward rate of spread calculated at a maximum of 1.2 m.min-1, and taking refuge in hollow logs and stumps. The security of these refuges depended on a number of factors including: size, degree of decomposition, seasonal dryness and fire intensity. Individual southern water skinks were also occasionally observed crossing a two metre wide bulldozed mineral earth break separating the treatment areas from unburnt sites. At the Musk Creek FESA, the day after both spring and autumn burns, numerous individuals of McCoy's skink were observed under large logs. These specimens of McCoy's skink, a short limbed and slow moving species, often showed signs of being burnt in the fire with scales and limbs melted and fused together. Many individuals probably died as a direct result of the burning. At the Barkstead FESA before the burning, a group of eight grass skinks were sheltering in a small pile of decaying timber. The day after the autumn burning, the wood was blackened all over but not totally burnt, but the skinks were no longer present. It was assumed the skinks would have survived, but this was not verified.

The effects of single spring and autumn prescribed fires in this study were similar. Four of the five species studied either depend on or largely rely on the litter layer for food and shelter. These species showed the greatest fluctuations in abundance. Litter loads reached pre-burn levels two to four years after burning. The importance of this rapid recovery of a major habitat component is shown by the increase in the relative abundances of Coventry's skink 28 months after spring burning. McCoy's skink, a species dependent on deep litter, is likely to be the most vulnerable to high-frequency and moderate- to high-intensity burning regimes. The only species that has a preferred semi-arboreal (fallen logs) microhabitat preference, southern water skink, showed little variation in abundance. It has been suggested that this species may actually benefit, in the short term, from infrequent fires and some forestry practices through a reduction in the shrub and ground cover and an increase in available basking sites (Rawlinson 1971; Mather 1978).

In a fuel reduction burn, the result, and indeed the aim, is not the complete consumption of all litter. Fires generally achieve 60 to 80 per cent of the area being burnt to some degree (Hodgson & Heislers 1972). The small-scale mosaic of burnt and unburnt areas that follow a fuel reduction burn are crucial to the local survival of many litter invertebrates and can act as the source of future recolonisation by invertebrates and small mammals (Leonard 1972; Heislers 1980; Humphries & Tolhurst 1992). These unburnt patches are also likely to act as refuges for small skinks. Most lizard species studied have definite home ranges (Heatwole & Taylor 1987). These permit lizards to familiarise themselves with their immediate environment and thus the nearest point of shelter in case of attacks from predators. The size of home ranges for some lizards is very small. Mather (1978) determined the mean home range of southern water skink to be 7.6 m² in open-forest at Toolangi, central Victoria. It is conceivable that in a low-intensity prescribed fire, entire or large proportions of existing home ranges of southern water skink and other small skinks would be left relatively intact.

Clearly no one fire regime is favourable to all species. There are no data on the optimum frequency of burning for conserving any reptile species. However, a fire regime that allows the build-up of litter and other sheltering and basking sites (e.g. logs), and thereby an adequate food supply (litter invertebrates), should enable most reptiles to thrive (Suckling & Macfarlane 1984). Future observations during this study will test this hypothesis.

10.4 Invertebrates

10.4.1 Methods and Review

Arthropods of the Phyla Chelicerata, Crustacea and Uniramia (Manton 1977) constitute by far the largest and most diverse components of the mesofauna (>100 µm to <1cm length) within litter and upper soil horizons of forests. These small animals, together with earthworms (Phylum Annelida) and microbes (<100µm), are essential in (1) regulating the decomposition of organic matter, (2) aerating soils, (3) recycling nutrients, and (4) serving as prey or acting as predators, parasites or parasitoids in food chains (Crossley 1970, 1977; Cromack et al. 1977; Greenslade& Greenslade 1983).

Litter-surface frequenting (epigeal) arthropods are especially vulnerable to surface fires, which destroy their habitats and desiccate upper soil layers. Earthworms are also highly fire-sensitive, because of their intolerance to ambient temperatures above 25°C (Reynolds 1973) and low moisture levels in the litter/upper soil horizons (Fig. 10.3). These characteristics of earthworms are due to their essential requirement of a continuous film of moisture within the permeable outer skin (cuticle) for respiratory exchange (Lee 1983).

Figure 10.3: Correlation between the annual mean values of earthworm density and soil dryness index, based on 40 composite litter/upper soil samples examined along 100 m transects in the Blakeville control treatment.

Figure 10.3: Correlation between the annual mean values of earthworm density and soil dryness index, based on 40 composite litter/upper soil samples examined along 100 m transects in the Blakeville control treatment.

The results reported here refer to progress made during the first three years of the study, and specifically relate to impacts of single spring and autumn burns. A complete description of methods and results is given in Neumann (1992).

10.4.2 Results and discussion

The study identified 29 arthropod taxa (families) at the litter surface in the Blakeville FESA. Seven of these were trapped frequently at all study sites and 15 taxa were trapped moderately often to occasionally. Only seven taxa all rare and representing 24.1 per cent of the total number of taxa trapped were restricted to one or two of the study sites. At the control treatment, the mean number of specimens in composite samples varied little between the sampling years for both noninsect arthropods and insects. This reflected a high degree of stability among the epigeal arthropod fauna in the undisturbed dry sclerophyll forest sampled. It also indicated that pitfall trapping produces consistent results with respect to the activity of epigeal noninsect arthropods and insects, and therefore seems appropriate for this study, which uses sampling over time to assess the effects of low-intensity prescribed burns on litter arthropods.

The spring burn caused short-term reductions in activity among the common major taxa Collembola (springtails) and Diptera (flies) for up to one year. The rarely trapped minor taxa [Opilionida (harvestmen), Lepidoptera (moths), Apocrita (parasitic wasps)] were also affected for 12 months. These reductions were associated with low litter loads in the first year after the burn. Populations of earthworms also declined substantially, but recovered within 2.9 years of the burn. The autumn burn suppressed the Collembola and the minor taxa Blattodea, Polydesmida, Thysanura and Tettigoniidae for up to ten months. Earthworms were not affected. Very dry soil conditions were associated with depressed collembolan activity at all study sites irrespective of burning.

Given the importance of Collembola, larval Diptera and earthworms among decomposers in forest litter it appears that the spring burn, and to a lesser extent the autumn burn, may have temporarily reduced the decomposer cycle. Research at species level is required to substantiate this hypothesis. These results suggest that fuel reduction burning on rotations of three years or less in forest similar to that studied here should be scheduled for autumn rather than spring to minimise adverse impacts on the overall invertebrate fauna inhabiting litter/upper soil.

10.5 Conclusions

Birds were little affected by a single spring or autumn low-intensity burn. Bird abundance and number of species were remarkably stable in burnt and unburnt areas. No new habitats were created and no existing habitats were destroyed, although there was a change in the frequency of each habitat type. These results however, cannot be applied to vegetation types where burning may create or remove particular habitats.

Seven bat species were found in the study areas, making bats the most diverse group of mammals in the area. The data on the comparative abundance of bats associated with fire treatments were inconclusive due to difficulties in the methodology of studying these mammals. Bat activity was significantly related to air temperature, so any bat assessment must take account of the air temperature at the time.

Recovery of the small mammals, brown antechinus and bush rats, occurred within a two year period of the burning when at least 50 per cent of gully vegetation and ten per cent of the midslope ground cover was left unburnt. This rapid recovery was attributed to the successful survival and breeding of individuals during this period in unburnt patches. Because the animals remain within defined ranges except when dispersing, recolonisation of areas totally burnt by fire would only occur when their habitat had recovered sufficiently to provide adequate food, shelter and breeding sites. If a greater proportion of these animals' habitat was burnt, the recovery time could be expected to be longer. The results suggest that fire management prescriptions should aim to minimise the burning of gully vegetation and burn no more than 70-80 per cent of an area overall if rapid recovery of small mammals is considered desirable.

Unburnt patches of litter and fallen logs act as refuges for small skinks and provide food, shelter and breeding sites after a fire. A fire regime that allows the periodic build-up of litter and other sheltering and basking sites (e.g. logs), and thereby an adequate food supply (litter invertebrates), should enable most reptiles to persist.

The spring burn caused short-term reductions in activity among both common invertebrate taxa Collembola (springtails) and Diptera (flies), and rarely trapped taxa Opilionida (harvestmen), Lepidoptera (moths) and Apocrita (parasitic wasps) for up to one year. These reductions were associated with low fine fuel loads in the 12 months after the burn. Populations of earthworms declined substantially, but recovered within three years. The autumn burn suppressed the Collembola and the minor taxa Blattodea, Polydesmida, Thysanura and Tettigoniidae for up to ten months, but earthworms were not affected. Very dry soil conditions were associated with depressed collembolan activity at all study sites irrespective of burning. Autumn burning is favoured to minimise effects on invertebrates.

10.6 Acknowledgments

I would like to acknowledge that the data presented here were collected by various researchers including: Richard Loyn, Robert Humphries, Fred Neumann and students from Ballarat University.

10.7 References

Braithwaite, R.W. 1987, Effects of fire regime on lizards in the wet-dry tropics of Australia. Journal of Tropical Ecology, vol. 3, pp. 265-275.

Catling, P.C. and Newsome, A.E. 1981, 'Responses of the Australian vertebrate fauna to fire: an evolutionary approach', in Fire and the Australian Biota, eds. Gill, A.M., Groves, R.H. & Noble, I.R., Australian Academy of Science, Canberra.

Caughley, J. 1985, 'Effect of fire on the reptile fauna of the mallee', in Biology of Australian frogs and reptiles, eds. Grigg, G., Shine, R. & Erman, H., Surrey Beatty, Chipping Norton, NSW.

Christensen, P. 1974, 'The concept of fauna priority areas', in Proceedings Fire Ecology Symposium, Monash Univ., Clayton.

Christensen, P., Recher, H. and Hoare, J. 1981, 'Responses of open forest to fire regimes', in Fire and the Australian Biota, eds. Gill, A.M. Groves, R.H. & Noble, I.R., Australian Academy Science, Canberra.

Christensen, P.E. and Kimber, P.C. 1975, 'Effect of prescribed burning on the flora and fauna of south-west Australian forests', Proceedings of Ecological Society Australia, vol. 9, pp. 85-106.

Cogger, H.G. 1969, 'A study of the ecology and biology of the Mallee Dragon (Amphibolurus fordii) and its adaptations to survival in an arid environment. Ph. D. thesis, Macquarie University.

Cowley, R.D. 1974, 'Effects of prescribed burning on birds of the mixed species forests of west central Victoria', in Proceedings Third Fire Ecology Symposium, Monash Univ., Clayton.

Cromack, K., Sollins, P., Todd, R.L., Crossley, D.A., Fender, W.M., Fogel, R. and Todd, A.W. 1977, 'Soil micro-organisms – Arthropod interactions: fungi as major calcium and sodium sources', in The role of arthropods in forest ecosystems, ed. Mattson, W.J. Springer-Verlag, New York, Heidelberg, Berlin.

Crossley, D.A. 1970, 'Roles of micorflora and fauna in soil systems', in Pesticides in the soil: ecology, degradation and movement, Michigan State University, East Lansing.

Crossley, D.A. 1977, 'The role of terrestrial saprophagous arthropods in forest soil; current status and concepts, in The role of arthropods in forest ecosystems, ed. Mattson, W.J., Springer-Verlag, New York, Heidelberg, Berlin.

Earl, G. & Lunt, I. 1989, 'Bibliography of ecological survey reports, Department of Conservation Forests and Lands, 1983 to August 1989', Victorian Naturalist, vol. 106, pp. 205-207.

Fyfe, G. 1980, 'The effect of fire on lizard communities in central Australia', Herpetofauna, vol. 12, pp. 1-9.

Gilmore, A.M., Emison, W.B. and Wheeler, J.R. 1979, 'Vertebrate fauna of the Ballarat area, Victoria', Memoirs of the National Museum of Victoria, vol. 40.

Greenslade, P.J.M. & Greenslade, P. 1983, 'Ecology of soil invertebrates', in Soils, An Australian Viewpoint. CSIRO, Melbourne.

Hall, I.S. 1981, 'The biogeography of Australian bats', in The Ecological Biogeography of Australia, ed. A. Keast, Dr. W. Junk Publishers.

Heatwole, H.F. & Taylor, J. 1987, Ecology of Reptiles, Surrey Beatty & Sons, Chipping Norton, NSW, 325pp.

Heislers, A. 1974, 'Effects of wildfire on mammals in a Victorian eucalypt forest', in Proceedings Third Fire Ecology Symposium, Monash University, Clayton.

Heislers, A. 1980, The impact of fire in foothil forest on small mammals and vegetation at Daylesford, Victoria. M.Sc. thesis, Monash University, Clayton, Victoria.

Hewish, M.J. & Loyn, R.H. 1989, Popularity and effectiveness of four survey methods for monitoring populations of Australian land birds. RAOU (Royal Australasian Ornithologists Union) Report no. 55.

Hodgson, A. & Heislers, A. 1972, 'Some aspests of the role of forest fire in south-eastern Australia', Forests Commission, Victoria Bulletin, no. 21, 1-40pp.

Humphries, R.K. 1992, 'The effects of single spring and autumn prescribed fires on reptile populations in Wombat State Forest', Dept. Conservation & Environment, Victoria, Forest Research Report no. 349.

Humphries, R.K. & Tolhurst, K.G. 1992, 'The effects of singele autumn and spring prescribed fires on Antechinus stuartii and Rattus fuscipes in Wombat State Forest', Dept. Conservation & Environment, Victoria, Forest Research Report no. 349.

Kemp, J., Westbrooke, M., van Gaans, G. and Fowler, A. 1992, 'An assessment of bat populations in the fire effects study areas, Wombat State Forest', Dept. Conservation & Environment, Victoria, Forest Research Report no. 349.

Kimber, P. 1974, 'Some effects of prescribed burning on Jarrah forest birds', in Proceedings Third Fire Ecology Symposium., Monash University, Clayton.

Lee, K. E. 1983, 'Soil animals and pedological processes', in Soils An Australian Viewpoint. CSIRO, Melbourne.

Leonard, B.V. 1972, The effects of fire upon selected small mammals and leaf litter fauna in sclerophyll forest in southern Australia. M.Sc. thesis, Unpub, Monash Univ.

Loyn, R.H. 1986, 'The 20 minute search – a simple method for counting forest birds', Corella, vol. 10, pp. 58-60.

Loyn, R.H., Hewish, M.J. & Considine, M. 1992, Short-term effects of fuel reduction burning on bird populations in Wombat State Forest', Dept. Conservation & Environment, Victoria, Forest Research Report no. 349.

Lunney, D., Eby, P. & O'Connell, M. 1991, 'Effects of logging, fire and drought on three species of lizards in Mumbulla State Foresst on the south coast of New South Wales', Australian Journal of Ecology, vol. 16, pp. 33-46.

Manton, S. M. 1977, The Arthropoda, Oxford University Press, Oxford.

Mather, P.B. 1978, An assessment of the ecological requirements of Sphenomorphus tympanum (c.t.f.) (Scindidae) – a forest dependent resident. M.Sc. thesis, La Trobe University, Bundoora. Victoria.

McKenzie, N.L. 1983, 'Bats – a part of our heritage', Swans, vol. 13, no. 3.

Menkhorst, P.W. 1989, 'Working list of Victorian mammals', Arthur Rylah Institute Tech. Rep. no. 50, Dept. Conservation Forests and Lands, Victoria.

Neumann, F.G. 1992, Effects of low intensity prescribed fire on invertebrates in mixed eucalypt forest – Wombat State Forest', Dept. Conservation & Environment, Victoria, Forest Research Report no. 349.

Newsome, A., McIlroy, J. & Catling, P. 1975, 'The effects of an extensive wildfire on populations of twenty ground vertebrates in south-east Australia', Proceedings Ecological Society Australia, vol. 9, pp. 107-23.

Parnaby, H. 1977, 'Bat survey of the Daylesford area, Victoria', Victoria Naturalist, vol. 94, pp. 191-97.

Rawlinson, P.A. 1971, Reptiles of Victoria', in Victorian Year Book No.85, ed. Arnold, V.H., Govt. Printer, Melbourne.

Recher, H.F., Lunney, D. & Posamentier, H. 1974, 'Effects of high intensity wildfire on small mammals at Nadgee Nature Reserve, N.S.W.', in Proceedings Third Fire Ecology Symposium, Monash University, Clayton.

Reilly, P. 1991, Emu 91: 100-106.

Reynolds, J. W. 1973, Earthworm (Annelida: Oligocheta) ecology and systematics, Proceedings of 1st Soil Microcommunities Conference, Syracuse, N.Y. 1971, ed. D. Dindal, USAEC, Office of Information Services.

Squire, R.O. 1990, Report on the progress of the Silvicultural Systems Project, Dept. Conservation & Environment, Victoria.

Stokes, T. 1975, 'The effect of a bushfire on the banding of flame robins in the Brindabella Ranges', Australian Bird Bander, vol. 13, pp. 75-76.

Suckling, G.A. & Macfarlane, M.A. 1984, 'The effects of fire on fauna – a review, in Proceedings of Symposium on Fuel Reduction Burning, ed. Ealey, E.H.M., Grad. School. Env. Sci., Monash University, Melbourne.

Woinarski, J.C.Z. 1990, 'Effects of fire on the bird communities of tropical woodlands and open forests in northern Australia', Australian Journal of Ecology, vol. 15, pp. 1-22.