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

13. Fire ecology of invertebrates - implications for nature conservation, fire management and future research

Gordon Friend
Science and Information Division, Department of Conservation and Land Management, W.A.

13.1 Abstract

Invertebrates are now recognised as critical elements in the maintenance of ecosystems, and many are seen to have potential as bio-indicators of environmental conditions. Information on their responses to environmental disturbance such as fire is therefore critical. There is a clear need to assess our current knowledge of invertebrate fire ecology, and examine the implications of the findings for fire management, nature conservation and future research.

An examination of data from two recent experimental studies, and a review of a broad cross-section of research carried out in temperate Australia over the past 40 years raises doubt that sound management decisions can be made from invertebrate fire ecology information collected to date. Invertebrate populations are so variable in time and space, being largely driven by environmental factors, that the impacts of fire may be largely over-ridden and almost impossible to predict. This is particularly the case where a broad level of taxonomic resolution is adopted. Furthermore, since post-fire succession may proceed to a new transient state also occupied by undisturbed communities, there is no 'ideal' stage or normal condition towards which management can aim.

In addition, considerable variability and shortcomings in experimental design and length of study exist. These deficiencies make it difficult to determine whether the outcomes observed are a true feature of invertebrate responses to fire, or are largely artefacts of the sampling procedure. Bearing this in mind, available data indicate a trend in invertebrate resilience to fire across broad habitat and climatic gradients. It is suggested that there is a need for conservatism in the application of high frequency and large scale fire regimes, particularly in the more mesic forested areas of Australia.

Given the variability and complexity of invertebrate assemblages and their responses to fire, a much more rigorous approach to methodology must be adopted if optimal fire regimes for invertebrate conservation in various habitat types are to be devised.

Key words: fire ecology, invertebrates, conservation implications, fire management, future research, bio-indications, temperate Australia.

13.2 Introduction

Although fire has been, and continues to be, one of the most important factors influencing the structure and distribution of plant and animal communities in Australia, our understanding of it as a management tool is incomplete. On the one hand we have quite sound knowledge of the principles of fire physics, and how weather, fuel and other site factors influence fire behaviour. Such knowledge forms the rationale for the prescribed burning of forested areas in the south-west and south-east of Australia to reduce fuel loads and wildfire hazard (Shea et al. 1981; Cheney 1985; McCaw & Burrows 1989). On the other hand, however, there is relatively little information available regarding the impacts of fire on the biota, particularly in the longer term, and of the precise role of fire in ecosysytem management. No doubt this partly reflects the enormous complexity of fire and its effects, but this does not negate the need to obtain reliable data on this topic to optimise the planned use of fire in habitat management.

Most fire effects studies to date have focussed on single fire events, and many have contemporaneously sampled adjacent areas with different fire histories, making the assumption that any differences in the abundance or composition of the biota are attributable to fire. Such an assumption is rather tenuous, given the inherent within-site variability of plant and animal populations (Campbell & Tanton 1981). Furthermore, there has been an emphasis on fire effects in forest ecosystems, particularly on plants and mammals. Relatively few studies have examined semi-arid woodlands and shrublands, or concentrated on herpetofauna or invertebrates (see reviews by Suckling & Macfarlane 1984; Christensen & Abbott 1989; Friend 1993).

With respect to invertebrates, this lack of focus is surprising since there is a considerable and growing body of evidence that such organisms play a crucial role in the structure and maintenance of ecosystems (Hill & Michaelis 1988; Kim 1993). Indeed, invertebrates constitute about 99 per cent of the world's biodiversity (Wilson 1987,1988), and in Australia outnumber vertebrate species by about 150:1 (CONCOM 1989; Yen & Butcher 1992). They are fundamental to processes such as decomposition, herbivory, parasitism and pollination, to name only a few, and form important dietary linkages throughout the food chain (Kim 1993; Samways 1993). Furthermore, many authors maintain that certain guilds of invertebrates, such as ants (Majer 1983; Andersen 1987), spiders (Clausen 1986; Main 1987), wasps and flies (Disney 1986) are proving to be reliable indicators of environmental trends.

There is a clear need to assess current knowledge of invertebrate fire ecology, and in particular to critically examine whether the information at hand is sufficiently robust to enable sound management decisions to be made. This paper briefly describes (a) the results from two recent studies of the impact of fire on invertebrate communities in semi-arid shrublands of the south-west of Western Australia, and (b) the outcome of a recent review of sampling methodologies and invertebrate post-fire response patterns, and assesses the implications of the findings for fire management, nature conservation and future research.

13.3 Recent studies

13.3.1 Stirling Range National Park

Friend and Williams (1993) documented the results from a three-year study (1989-1992) of invertebrate responses to fire in mallee-heath shrublands of the Stirling Range National Park, in the south-west of Western Australia. Two approaches were adopted in this work. Firstly, an experimental approach, whereby two 20ha areas and a 400ha area were subjected to fires of different intensities, scale and seasons of burn. One of the 20ha areas was burnt under low-moderate intensity in spring 1990, while the other two areas were burnt by moderate-high intensity autumn fires in 1991. Samples of invertebrates were obtained for 12-18 months before and after these burns, along with data on vegetation and habitat structure. In addition, some data were obtained between October 1991 and May 1992 from an area which was affected by a large-scale high intensity wildfire in April 1991.

The second approach adopted was a space-for-time substitution (Pickett 1989), whereby a variety of mallee-heath shrublands of different ages were contemporaneously sampled for invertebrates, vegetation and habitat structure. Ages of vegetation available for sampling included four, eight, 20, 40 and c. 50-year-old stands (ages as at 1990). The 20 and 40-year-old areas were burned in the experimental approach outlined above. The space-for-time approach enabled longer-term successional aspects to be examined in a relatively short-term project.

Invertebrates were sampled using pitfall traps, and the abundance of major orders (Coleoptera, Araneae, Hymenoptera, Diptera Hemiptera and Orthoptera) was assessed over time. Coleoptera were further identified to species (morphospecies) level, allowing a more detailed assessment of the impacts of fire and comparison with the broader Order-level data.

The results obtained showed the enormous complexity of factors which influence the post-fire response patterns of invertebrates. While some groups increased in abundance after fire, others declined. These trends were both inconsistent for a particular taxon between different experimentally burnt areas and quite ephemeral. Amongst the Diptera, for example, numbers increased significantly after the experimental spring fire and the autumn wildfire, but declined significantly after both experimental autumn fires. Some of these trends were apparent only at the acute stage, while others were more chronic and still apparent 12 months after fire. The greatest changes in Dipteran numbers, however, simply reflected year-to-year variation.

Within the Coleoptera large fluctuations in numbers occurred on a seasonal basis. Although primarily due to only a few species, these fluctuations were independent of fire. No significant changes in beetle abundance or species richness occurred following either spring or autumn control burns, and none were apparent within six months of the high intensity wildfire. Of the major beetle families examined in detail, only the Carabidae exhibited a consistent response, with numbers generally declining in the short term after fire. Although abundance as a whole was greatest in the 40-year-old area, year-to-year changes and locality effects tended to outweigh any fire-related trends.

Hymenoptera and Araneae showed significant fire-related trends only in one area, there being higher abundances of both groups in the eight-year-old area than in the adjacent long-unburnt area. Neither group showed any distinct and consistent relationship with time since fire. Similarly, although Hemiptera generally increased after fire, this trend was significant only in the wildfire area and may be related to the flush of new growth which had occurred by the time sampling commenced. Such young growth probably represents a significant food resource for this group. New vegetative growth also may explain the higher abundances of Orthoptera recorded in the chronic and longer-term phases following fire; lack of plant food resources in the acute phase was correlated with a significant short-term decline in abundance after one of the experimental autumn fires.

Ordinations of the Stirling Range data using Principal Components Analysis reflected changes in absolute and relative abundance of orders from one year to the next, and to a lesser extent seasonal changes, all probably tied to climate. Any effects of either prescribed burns or wildfire were not apparent at the order level in the context of these changes. These analyses also indicated a lack of correlation between changes in invertebrate abundance and composition, and changes in floristics or vegetation structure. Although direct relationships no doubt existed between individual species of invertebrates and plants, these patterns were at a much finer level of resolution than was addressed in this (or any previous) study. This lack of correlation between invertebrates, floristics and vegetation structure further emphasises Yen's (1987) point that categorising and protecting areas on the basis of high vertebrate species richness does not ensure that large numbers of invertebrate species are also conserved.

Thus, in Friend and Williams' (1993) study which were assessed at the order level, and others (eg. Bornemissa 1969; Leonard 1972; Whelan et al. 1980; Majer 1980, 1984, 1985; Abbott 1984; Tap & Whelan 1984 and Neumann & Tolhurst 1991), there were few clear-cut trends in the patterns of abundance and composition of invertebrates assessed at the ordinal level, which could be directly related to fire, floristics or vegetation structure. Patterns clearly related to fire only emerged when Coleoptera were examined at the morphospecies level. Then, three major groups were obvious: those associated with pre-fire grids, post-fire control (unburnt) grids and burnt grids. Further examination of the data was able to indicate which species and/or families of beetles showed relationships with fire, and thus signified which groups are worthy of more detailed study.

13.3.2 Durokoppin Nature Reserve

Addressing the need for more detailed species-level studies of invertebrate responses to fire, Strehlow (1993) analysed pitfall trap data on ground-dwelling spiders collected over five years (1987-1991) from Durokoppin Nature Reserve in the central wheatbelt of Western Australia. Four sites were studied, two of these being subjected to a high intensity fire in autumn 1989, and the other two remaining unburnt.

The fire had a significant short-term impact on spider abundance and species richness, with web building species being much more severely affected than ground-dwelling and burrowing forms. Families such as the Zodaridae, Salticidae and Gnaphosidae were greatly reduced in abundance after the fire, while others such as the Lycosidae, Zoridae and mygalomorphs were little affected. Recolonisation occurred from the immediate survivors of the fire and from immigration of spiders from the adjacent unburnt areas. The impact of the fire was mitigated by a relatively wet winter and mild, wet summer following the burn which allowed rapid recovery of the spider community.

As with the Stirling Range National Park study, environmental variables appeared to be the principal factors determining the rate and extent of recovery of the spider populations after fire. Ordination of the spider family data by years showed that although the 1989 fire caused a significant divergence in similarities between burnt and unburnt sites in the year following the fire (1990), by 1991 the ecological distance between control and impact sites was similar to that before fire. Interestingly however, the community composition of all sites in 1991 was significantly different from other years. Post-fire succession of the spider communities tended to move not towards the original pre-fire state, but to a new transient state of the undisturbed communities (Strehlow 1993). This outcome suggests that no steady state existed with respect to spider community composition, and that the effects from a single fire event were only temporary, and secondary to those caused by seasonal and year-to-year variability in environmental factors.

13.4 Methodology and post-fire response patterns

In examining results from these recent studies, and earlier studies of invertebrates and fire, two findings became apparent. These are: that a wide variety of post-fire response patterns may occur; and that reponses are often not consistent within taxonomic group or habitat type between different studies. In many instances invertebrate groups show marked locality, season and year-to-year effects which outweigh any changes attributable to fire. Furthermore, many of these inconsistencies could have arisen because of variations or shortcomings in experimental design, taxonomic treatment and length of study.

To place these factors in perspective, Friend (1995) conducted a more detailed examination of 24 invertebrate fire ecology papers which represented a broad cross-section of research carried out in temperate Australia over the past 40 years. This review aimed to: (a) quantitatively assess whether or not various groups (eg. Orders) have consistent post-fire response patterns; (b) highlight groups or taxa which are sensitive to fire and therefore could serve as indicator species of pyric status; and (c) examine trends in relation to habitat type across a broad climatic gradient. In addition, the opportunity was taken to gather some statistics on experimental design and sampling methodologies employed in studies of invertebrate responses to fire.

13.4.1 Post-fire response patterns

Variation across habitats

Response patterns of the major invertebrate groups were examined across all 24 studies, and with the studies separated according to three major habitat types; these were tall open-forest, open-forest and woodland/shrubland. For each group, for which post-fire response information was available the proportion of cases showing no post-fire change in abundance, a post-fire increase or a post-fire decrease were calculated and tabulated (Friend 1995). This provided information on both the strength and consistency of post-fire responses which could be compared either between different groups or between the same groups in different habitat types.

Examination of these data revealed very clear and significant differences in the response patterns for the three different habitat types (Friend 1995). In tall open-forest there were more cases than expected in the decrease category; in open-forest more cases than expected occurred in the no change and decrease categories. In woodlands and shrublands the response pattern was similar to that expected by chance.

This outcome strongly suggests that there is a gradient in invertebrate responses to fire related to habitat type, and ultimately, climate (Friend 1995). High levels of resilience in the drier habitats is probably a reflection of the invertebrate fauna's adaptations to survive seasonal aridity (Friend 1995). Low resilience in the more mesic habitats in turn reflects the relictual nature of many invertebrate communities which arose in and are adapted to much earlier, wetter climatic regimes.

Indicator groups

Across all 24 studies, six invertebrate groups proved to be common, sensitive to fire (ie. showed a decrease in abundance post-fire) and exhibited a consistent response to it; these were Araneae, Lepidoptera, Isopoda, Isoptera, Thysanura and Blattodea. A seventh group, the Diptera, also qualified because the two cases where a post-fire increase occurred both related to high-intensity wildfires (Neumann 1991; Friend & Williams 1993), indicating some differential but consistent responses associated with fire type.

In open-forest Araneae, Diptera, Lepidoptera, Isopoda and Blattodea were quite consistent, and in addition Coleoptera, Acarina and Collembola qualified. Interestingly, ants, which have been frequently used as indicator species, registered inconsistent responses in this habitat type. In the drier woodland and shrubland habitats, however, ants showed very consistent post-fire responses (increases), suggesting that their value as indicators of disturbance or pyric status may vary according to climate and habitat (Friend 1995).

In the woodland/shrubland habitats there was a paucity of data, but the Araneae group stood out as a potential indicator. This is not surprising considering that spiders are at the apex of the invertebrate food pyramid, and some representatives (especially the mygalomorphs or trapdoor spiders) are relictual in their distribution, long-lived and relatively sedentary with poor dispersal powers. They also have very specific microhabitat preferences (Main 1987). Araneae thus appeared to be the most promising indicator group to use in fire ecology studies across a broad range of habitats (Friend 1995). Lepidoptera, Isopoda, Blattodea and Thysanura also have potential as indicator groups, but more data are needed to confirm this. Friend (1995) considered that at this stage there are insufficient fire ecology data available at the species level (perhaps with the exception of ants) to categorise invertebrate species as indicators.

13.4.2 Methodology

The majority (71 per cent) of the 24 studies reviewed by Friend (1995) used pitfall traps, but the dimensions varied greatly and were often tailored for specific purposes. Small diameter pitfall traps (eg. 18mm test tubes) were often used by researchers with a primary interest in ants, whilst more generalised studies usually utilised larger traps (eg. plastic vials or cups up to 90mm diameter). A mixture of ethanol and glycerol was the most common preservative used in pitfall traps and most were left open for between seven to ten days. Heat extraction of soil cores and/or leaf litter samples using Tullgren or Berlese funnels was also a common method used, although two studies employed hand sorting, a method considered by Campbell and Tanton (1981) to be biased against small, cryptic animals.

Although eighteen of the 24 studies examined were based on sampling before and after fire, some did not specify the length of post-fire sampling,
and many studies had acquired only a minimal amount of pre-fire data (<12 months), but had adequate post-fire data. Indeed, those studies with the most post-fire data were also amongst those with the least pre-fire data. Friend (1995) considered such an unbalanced sampling schedule to be of questionable benefit.

Nine studies adopted a space-for-time approach either wholly (ie. with no pre-fire information) or as an adjunct to before/after experimental work. Friend (1994) considered the latter approach optimal since it provides both short-term and long-term data on the impact of several fires within a relatively short research time-frame. The duration of monitoring in such studies, however, was highly variable (CV = 145 per cent; Friend 1995).

With respect to identification levels, Friend (1995) found that the majority of studies (about 75 per cent) examined one or two groups in detail to the family or species level and identified the remainder to order level. The family or species level identifications generally concerned ants, and to a lesser extent spiders. Five studies concentrated only on specific groups (ants, spiders, mites or springtails) and did not examine other order level data. From this Friend (1995) concluded that studies have either been very general or highly specific, and that to date no fire ecology study has identified a broad range of invertebrates to family or species level. This situation is understandable given the huge diversity of invertebrates and the current (and specialised) taxonomic knowledge of this fauna. This is the taxonomic impediment situation referred to by New (1984) and makes the search for species or groups which can be used to indicate certain environmental regimes all the more pressing (New 1984).

13.5 Implications of findings

The information briefly reviewed above has serious implications for the conservation of invertebrates in relation to fire and fire management, the use of invertebrates in fire ecology studies, and the design and direction of future studies. Three important factors impinge on these issues.

Firstly, invertebrate populations are extremely variable in time and space at a very fine scale. This variability, which is primarily driven by environmental factors (eg. temperature, rainfall, insolation), has the capacity to override any changes in populations attributable to fire. As pointed out by Campbell and Tanton (1981), this frequently makes prediction of responses to fire almost impossible. This is certainly likely to be a problem if data are analysed at the Order level of identification; individual species may be markedly affected by fire (both increases and decreases), but at the broad level of identification these trends will tend to cancel out and thus not be apparent. If causal relationships are to be defined there is a need to examine abundance patterns of invertebrate communities at a finer level of taxonomic resolution, as exemplified in both studies outlined above (Friend & Williams 1993; Strehlow 1993).

Furthermore, because invertebrate post-fire succession does not necessarily return to the pre-fire state, there is no 'climax' community or baseline against which we can measure long-term change. This has profound implications for land managers because there is no 'ideal' stage or normal condition towards which managment can aim (Strehlow 1993). In essence, every stage in community development is important (Strehlow 1993), and managers need to be flexible enough in their operations to allow for such diversity.

A second major factor concerns the general finding that invertebrates appear to be less resilient (sensu Westman 1986) to fire in the more mesic environments (particularly tall open-forest), than in drier woodland or shrubland ecosystems. This means that management prescriptions developed for one habitat or ecosystem type do not necessarily apply to other ecosystems. But more importantly, it suggests that high frequency/large scale fire regimes may be inappropriate for the mesic forested areas of south-eastern and south-western Australia.

Given present knowledge, it is clearly better to err on the side of conservatism in developing fire regimes for any habitat types. For remnant shrubland and woodland vegetation, larger-scale block burning should not be carried out except in special circumstances (eg. for specific regeneration purposes, experimental research or where it contributes to a well considered strategic fire management objective). Protection of areas from large-scale, high-intensity wildfires through a system of internal and external low fuel zones should remain a high priority for managers of remnant vegetation. Such a strategy is also applicable to the broader forested habitats of temperate Australia.

The third factor relates to the design and direction of future research on fire and invertebrates. Given the high levels of variability inherent in invertebrate populations, the fact that effects may be masked if analyses are undertaken only at a general level of identification, and the enormous taxonomic impediment surrounding invertebrate studies (New 1984), a much more rigorous approach is necessary. More systematic methods relating to sampling, experimental design and analysis need to be adopted. Carefully designed experimental studies incorporating repeated samplings, to provide site-based information before and after the treatment, which can be compared with an untreated control (Green 1979; Underwood 1991; 1993) are required. Such studies need to be reasonably long-term and temporally balanced each side of the treatment, so as to provide sufficient measure of underlying variability and direction of population or community change over time. The studies also need to be focussed on species, or morphospecies or recognisable taxonomic units (Oliver & Beattie 1993) which have proven to be reliable indicators of pyric status and environmental trends. It is only by adopting such approaches that we can minimise experimental variability and determine whether observed outcomes are a true reflection of invertebrate responses to fire, or due to human-related factors.

Clearly, we have a considerable way to go before sufficiently thorough information will be available to enable us to confidently devise optimal fire regimes for invertebrates in various habitat types.

13.6 References

Abbott, I. 1984, 'Changes in the abundance and activity of certain soil and litter fauna in the jarrah forest of Western Australia after a moderate intensity fire', Australian Journal of Soil Research, vol. 22, pp. 463-469.

Andersen, A.N. 1987, 'Ant community organization and environmental assessment', in The Role of Invertebrates in Conservation and Biological Survey. ed. J.D. Majer, Western Australian Department of Conservation and Land Management Report. pp. 43-52.

Bornemissza, G. F. 1969, 'The reinvasion of burnt woodland areas by insects and mites', Proceedings of the Ecological Society of Australia, vol. 4, pp. 138.

Campbell, A.J., & Tanton, M.T. 1981, 'Effects of fire on the invertebrate fauna of soil and litter of a eucalypt forest', in Fire and the Australian Biota. Eds. A.M. Gill, R.H. Groves, & I.R. Noble, Australian Academy of Science, Canberra.

Cheney, N.P. 1985, 'Forest fire management in Australia', in Proceedings of the Intermountain Fire Council 1983 Fire Management Workshop. Report NOR-X-271, Canadian Forest Service, pp 75-83.

Christensen, P.E.S. & Abbott, I. 1989, 'Impact of fire in the forest ecosysytem of southern Western Australia: a critical review', Australian Forestry, vol. 52, pp. 103-121.

CONCOM 1989, 'Council of Nature Conservation Ministers – Australian statement on invertebrates', Myrmecia, vol. 25, pp. 123-125.

Clausen, I.H.S. 1986, 'The use of spiders (Araneae) as ecological indicators', Bulletin of the British Arachnology Society, vol. 7, pp. 83-86.

Disney, R.H.L. 1986, 'Assessments using invertebrates; posing the problem', in Wildlife Conservation Evaluation, Ed. M.B. Usher, Chapman and Hall, London.

Friend, G.R. 1993, 'Impact of fire on small vertebrates in mallee woodlands and shrublands of temperate Australia – a review', Biological Conservation, vol. 65, pp. 99-114.

Friend, G.R. 1995, 'Fire and invertebrates – a review of research methodology and the predictability of post-fire response patterns', in Landscale Fires '93: Proceedings of an Australian Bushfires Conference, Perth, Western Australia. Eds, W.L. McCaw, N.D. Burrows, G.R. Friend & A.M. Gill. CALMScience Suppl. no. 4, pp. 165-174.

Friend, G.R. & Williams M.R. 1993, Fire and invertebrate conservation in mallee-heath remnants. Final Report, Project P144, World Wide Fund for Nature Australia.

Green, R.H. 1979, Sampling Design and Statistical Methods for Envionmental Biologists. John Wiley & Sons, New York.

Hill, L. & Michaelis, F.B. 1988, 'Conservation of insects and related wildlife', Occasional Paper no. 13. Australian National Parks and Wildlife Service, Canberra.

Kim, K.C. 1993, 'Biodiversity, conservation and inventory: why insects matter', Biodiversity and Conservation, vol. 2, pp. 191-214.

Leonard, B.V. 1972, 'The effect of fire upon selected small mammals and leaf litter fauna in sclerophyll forest in southern Australia', M.Sc. thesis, Monash University, Melbourne.

Main, B.Y. 1987, 'Persistence of invertebrates in small areas : case studies of trapdoor spiders in Western Australia', in Nature Conservation: The Role of Remnants of Native Vegetation, eds. D.A. Saunders, G.W. Arnold, A.A. Burbidge & A.J.M. Hopkins, Surrey Beatty & Sons, Chipping Norton, N.S.W., pp 29-39.

Majer, J.D. 1980, 'Report on a study of invertebrates in relation to the Kojonup Fire Management Plan', Western Australian Institute of Technology, Department of Biology Bulletin no. 2., pp 1-22.

Majer, J.D. 1983, 'Ants : bio-indicators of minesite rehabilitation, land-use, and land conservation', Environmental Management, vol. 7, pp. 375-383.

Majer, J.D. 1984, 'Short-term response of soil and litter invertebrates to a cool autumn in Jarrah (Eucalyptus marginata) forest in Western Australia', Pedobiologia, vol. 26, pp. 229-247.

Majer, J.D. 1985, 'Fire effects on invertebrate fauna of forest and woodland', in Fire Ecology and Management in Western Australian Ecosystems, WAIT, Environmental Studies Group Report no. 14. ed. J. Ford, Western Australian Institute of Technology, Perth, pp. 103-106.

McCaw, W.L. & Burrows, N.D. 1989, 'Fire management', in The Jarrah Forest: A Complex Mediterranean Ecosysytem. eds. B. Dell, J.J. Havel & N. Malajczuk, Kluwer Academic Publishers, Dordrecht.

Neumann, F.G. 1991, 'Responses of litter arthropods to major natural or artificial ecological disturbances in mountain ash forest', Australian Journal of Ecology, vol. 16, pp. 19-32.

Neumann F.G. & Tolhurst, K. 1991, 'Effects of fuel reduction burning on epigeal arthropods and earthworms in dry sclerophyll eucalypt forest of west-central Victoria', Australian Journal of Ecology, vol. 16, pp. 315-330.

New, T.R. 1984, Insect Conservation: An Australian Perspective, Dr W. Junk Publishers, Netherlands.

Oliver, I. & Beattie, A.J. 1993, 'A possible method for the rapid assessment of biodiversity', Conservation Biology, vol. 7, pp. 562-568.

Pickett, S.T.A. 1989, 'Space-for-time substitution as an alternative to long-term studies', in Long-term Studies In Ecology: Approaches and Alternatives. ed. G.E. Likens, Springer-Verlag, New York.

Samways, M.J. 1993, 'Insects in biodiversity conservation: some perspectives and directives', Biodiversity and Conservation, vol. 2, pp. 258-282.

Shea, S.R. Peet, G.B. & Cheney, N.P. 1981, 'The role of fire in forest management', in Fire and the Australian Biota, eds. A.M. Gill, R.H. Groves, & I.R. Noble, Australian Academy of Science, Canberra.

Strehlow, K.H. 1993, 'Impact of fires on spider communities inhabiting semi-arid shrublands in Western Australia's wheatbelt', B.Sc. (Hons) thesis, Murdoch University.

Suckling, G.C. & Macfarlane, M.A. 1984, 'The effects of fire on fauna: a review', in Fighting Fire With Fire, ed. E.H.M. Ealey, Graduate School of Environmental Science, Monash University, Melbourne.

Tap, P. & Whelan, R.J. 1984, 'The effect of fire on populations of heathland invertebrates', in Medecos IV. Proceedings of the Fourth International Conference on Mediterranean Ecosysytems, ed. B. Dell, University of Western Australia, pp. 147-148.

Underwood, A.J. 1991, 'Beyond BACI: experimental designs for detecting human environmental impacts on temporal variations in natural populations', Australian Journal of Marine and Freshwater Research, vol. 42, pp. 569-587.

Underwood, 1993, 'The mechanics of spatially replicated sampling programmes to detect environmental impacts in a variable world', Australian Journal of Ecology, vol. 18, pp. 99-116.

Westman, W.E. 1986, 'Resilience: concepts and measures', in Resilience in Mediterranean-Type Ecosysytems, eds. B. Dell, A.J.M. Hopkins & B.B. Lamont, Dr W. Junk Publishers, Dordrecht.

Whelan, R.J., Langdyk, W. & Pashby, A.S. 1980, 'The effects of wildfire on arthropod populations in jarrah-Banksia woodland', Western Australian Naturalist, vol. 14, pp. 214-220.

Wilson, E.O. 1987, 'The little things that run the world (the importance and conservation of invertebrates)', Conservation Biology, vol. 1, pp. 344-346.

Wilson, E.O. 1988, 'The current state of biological diversity', in Biodiversity, Ed. E.O. Wilson, National Academy Press, Washington D.C.

Yen, A.L. 1987, 'A preliminary assessment of the correlation between plant, vertebrate and Coleoptera communities in the Victorian mallee', in The Role of Invertebrates in Conservation and Biological Survey, ed. J.D. Majer, Department of Conservation and Land Management, Perth.

Yen, A.L. & Butcher, R.J. 1992, 'Practical conservation of non-marine invertebrates', Search, vol. 23, pp. 103-105.