Ann. Zool. Fennici 35: 163–171 ISSN 0003-455X Helsinki 11 December 1998 © Finnish Zoological and Botanical Publishing Board 1998

The role of ground (Coleoptera: Carabidae) in monitoring programmes in Australia

Timothy R. New

New, T. R., Department of Zoology, La Trobe University, Bundoora, Victoria 3083, Australia Received 4 May 1998, accepted 6 August 1998

Use of the diverse Australian carabid fauna in environmental assessment is at present restricted, because of lack of ecological and faunistic information and the difficulties of identifying most taxa. They are not yet a core focal group in monitoring studies. In contrast to groups such as ants, in particular, many carabids are usually captured in only small numbers, and functional groups of Australian Carabidae are poorly defined. Al- though a few studies have demonstrated the predominance of particular taxa in local assemblages and their responses to changes in land management, the presence of par- ticular taxa and the richness of assemblages is unpredictable because of lack of knowl- edge of habitat requirements. The general usefulness of carabids as indicators in Aus- tralia is not yet proven. Carabidae are one of several groups of epigaeic invertebrates with potential for incorporation into a portfolio of focal groups for environmental as- sessment, but are still far from being a ‘stand alone’ group in Australia.

1. Introduction vation assessment in Australia, especially of their values in helping to evaluate habitat quality. Most The use of Carabidae (ground beetles) for envi- of the problems of employing carabids result sim- ronmental assessment in Australia has received ply from our ignorance of the group, and of the little attention, and is by no means as common as responses of individual species to habitat change in much of the northern hemisphere, where con- at even very broad levels. As elsewhere, they are siderable ecological interpretation at species and among the widespread groups of epigaeic assemblage levels has been undertaken. There, ca- amenable to trapping by relatively simple tech- rabids are one of the most important groups niques, appear to show strong relationships with used as indicators in many terrestrial ecosystems particular habitats, and which also occur reason- (Stork 1990, Desender et al. 1994). Despite their ably strongly in heavily altered landscapes. Cara- undoubted values and potential as monitors in bidae, simply, merit increased study to determine some environments, their use in Australia is not their worth as ecological tools in Australia. Such likely to increase substantially during the next few studies must encompass the biology of the more years, at the least. In this paper I explore why this abundant taxa and seek to define the characteris- is so, in the context of a broader summary of the tic carabid assemblages of a variety of vegetation use of terrestrial epigaeic invertebrates in conser- types and microhabitats. 164 New • ANN. ZOOL. FENNICI Vol. 35

2. The Australian Carabidae series of some 50 papers by T. G. Sloane is diffi- cult for a non-specialist to digest. Sequences of The Australian Carabidae are a remarkable ele- papers more recently by B. P. Moore and M. Baehr, ment of the diverse fauna of the isolated in particular, have made notable advances to our island continent, with many phylogenetically im- knowledge of some groups, and their distributional portant taxa. Carabidae are by no means the larg- relationships. Cicindelinae were appraised by Frei- est family of beetles in Australia, but still comprise tag (1979), so that the 29 species of Cicindela L. an impressive, and very approximate, 2500 spe- can be recognised clearly, and the other genera of cies. Many of these are undescribed, although the this small subfamily diagnosed. However, by far listing in the Zoological Catalogue of Australia the majority of Australian Carabidae belong to (Moore et al. 1987) enumerates 266 genera and the Carabinae (with 32 tribes recognised: Law- about 1800 described species, many with addi- rence & Britton 1994). tional synonyms. More recent revisionary studies As with many insect groups, endemism in Aus- have extended the described fauna to slightly over tralian Carabidae is very high. A high proportion 2000 species. Much of the basic descriptive tax- of species is geophilous, although recent studies onomy does not incorporate biological informa- in northern Queensland have greatly increased the tion, and there have been few complementary eco- known diversity of arboreal species (such as the logical studies and surveys of ground beetles. Ba- species of Philipis Erwin: Baehr 1995). Many taxa sic inferences on regional diversity, local species appear to have very restricted distributions. Noto- richness, and distribution patterns of most groups nomus Chaudoir, for example, contains around are fragmentary or non-existent. The detailed tem- 100 described species, all flightless and many lim- plates of species incidence and seasonal patterns ited to forest environments in eastern Australia. taken for granted by entomologists working on In these habitats, there is considerable species turn- carabids in parts of Europe simply are unavailable. over along latitudinal and other gradients, as dem- Likewise, there is relatively little non-special- onstrated by Darlington (1961). ist literature on the family. Matthews (1980) pro- The broad geographical pattern of Australian vided keys to the genera that occur in South Aus- carabids is one of older elements in the south of tralia, and Moore (1980) included a general ap- the country, reflecting persistence of Gondwanan praisal of the family in his book. A rather more elements in the Bassian region and various de- technical account, with keys to tribes, is provided grees of northern incursives (taxa with strong re- by Lawrence and Britton (1994). However, almost lationships to the faunas of Indonesia, Papua New universally for species level studies, the informa- Guinea and, more broadly, the Oriental region) tion needed to identify species is in the primary extending from north to south. However, many literature, or does not exist in any synthesised species appear to be uncommon, and are known form. Some groups, such as Harpalinae, are par- from few specimens and, in some cases, from only ticularly difficult to identify. Even separation to their type locality. Their distributions are essen- generic level is difficult without access to a large tially unknown except in the most general terms, institutional collection and, as for many other and usually lack strong ecological correlates at groups of insects in Australia, the overall taxo- smaller scales. nomic impediment and dearth of secondary guides Ecological specialisations include a substan- largely precludes reliable non-specialist identifi- tial troglobiont fauna, with local taxa in caves in cations of Carabidae. Few larvae are recognisable many different parts of the country — including to species level, and life history data are fragmen- local endemics in Tasmania, New South Wales tary even for many of the most common species. and the Nullarbor Plain. Speotarus princeps, for As Moore et al. (1987) noted, many of the larger example, has been recorded from a single cave in genera have not been subjects of any recent revi- New South Wales, in which recent searches by sion, and contain varying amounts of undetected Eberhard and Spate (1995) failed to locate it. Sev- or undocumented synonymy, in addition to unde- eral obligate cave dwellers are listed as ‘Vulner- scribed taxa. The solid descriptive foundation on able’ on Tasmania’s interim list of threatened in- Australian Carabidae laid earlier this century in a vertebrates (IAC 1994). Many of the intriguing ANN. ZOOL. FENNICI Vol. 35 • The role of ground beetles in monitoring programmes 165 cave-dwelling carabids have been described by uation of any major habitat. The particular groups Moore over the last 35 years. involved will clearly differ in different contexts, Likewise, the flightless forest taxa, both geo- but should preferably incorporate taxa which show philous and arboreal, tend to be highly localised. some ecological complementarity and can be sam- They were incorporated by Monteith (1995) in pled straightforwardly by simple, standardised developing a conservation index for evaluating techniques (New 1993, 1995, 1998). This context and ranking the highly sensitive tropical wet for- is currently receiving considerable attention in est areas of Queensland. However, in those areas, evaluations involving invertebrates in Australia. arboreal taxa (collected by insecticide fogging of In altered landscapes, such as those heavily modi- tree trunks: method summarised by Baehr 1995) fied for intensive agriculture or softwood planta- were the predominant fauna collected. tion forestry (mainly of the exotic Monterey pine, There is little definitive information on carabid Pinus radiata) in the southeast, dramatic changes phenology, simply because surveys and detailed in the epigaeic invertebrate fauna can result. Con- life history studies have only rarely been done, siderable effort has been made to evaluate and even for many of the most abundant species. Near document such changes, and to implement moni- Melbourne, Notonomus gravis (Chaudoir) has been toring systems involving the responses of differ- captured in every month of the year (P. A. Horne, ent invertebrates and invertebrate groups. 1992), but in very small numbers during winter com- However, there is one important restriction on pared with mid-summer. As P. A. Horne (1990, the interpretation of these. Virtually all studies 1992) noted, the population has overlapping co- have been undertaken in those parts of the conti- horts of current year and previous year adults and nent already subject to massive alteration by this feature, together with many females being in- changes in land use patterns since European set- active for parts of their life by caring for their tlement over the last 200 years (Greenslade & New brood (so that they are then not accessible to pit- 1991, Graetz et al. 1995). These areas have been fall trapping) renders interpretation of population subjected to massive despoliation and invasion size and sex ratios difficult. For such brood-tend- by exotic biota (not including carabids), and much ing taxa, which are probably numerous among of it bears little resemblance to its condition of Australian pterostichines (Horne 1990), differen- only a century ago. For some critical ecosystems tial activity of the sexes may be the main cause of now represented by only small disjunct remnants, variations in pitfall catches. it is difficult to obtain reliable baseline data against which to evaluate the extent of faunal change. Methods for interpretation involving invertebrates 3. Conservation evaluation are still being developed, drawing in part from lessons learned from other parts of the world. Considerable attention has been paid in Australia As one example of habitat change, native to facilitating the use of terrestrial invertebrates grasslands in Australia’s southeast are regarded in evaluating and characterising habitats and land- widely as one of the country’s most endangered scape change (Yen & Butcher 1997, for overview). terrestrial ecosystems (Kirkpatrick et al. 1995), In evaluating the importance of both cave faunas and estimates of 93%–97% of the grasslands be- and forest assemblages, as above, carabids are ing lost, predominantly to pastoralisation and ex- simply one group of many invertebrate taxa em- otic grass species, indicate the likely extent of ployed, and their use as sole tools in these con- parallel disruption to native invertebrate commu- texts seems comparatively limited, notwithstand- nities. Data for changes in Carabidae are rather ing the values of particular notable taxa which sparse but, for example, using the Golden Sun could themselves be targeted (for example by Moth (Synemon plana Walker, Castniidae) as a some form of listing on protection schedules) for flagship species for high quality native grassland, individual conservation. The lack of detailed bio- it is clear that this species has disappeared from logical knowledge of any given taxonomic group many of the sites it formerly inhabited, and is now can in part be offset by using a combination or present almost entirely on a small number of rem- suite of different, complementary groups for eval- nant patches. Another flagship species in the same 166 New • ANN. ZOOL. FENNICI Vol. 35 habitat is the flightless morabine grasshopper, 4. Monitoring Keyacris scurra (Rehn). Extensive pitfall trapping in such areas gener- The twin needs for monitoring in terrestrial habi- ally reveals rather few species of Carabidae, many tats using invertebrates in Australia both draw on of them very sporadic in incidence. The same situ- the common presumption that invertebrate re- ation prevails in open woodlands and agricultural sponses may be among the most sensitive indica- lands. Horne and Edward (1997) found 28 spe- tors of environmental change and, hence, of ‘eco- cies of Carabidae, some of them commonly, in a system health’. These needs are: year of pitfall trapping in agricultural land in west- ern Victoria and believed some of these to be im- 1. Determining the influences of changes in land portant predators of pest . Indeed, con- use, including determining the levels of deg- servation tillage practices may even enhance the radation in relation to pristine areas, monitor- abundance of some species such as Rhytisternus ing attempts at restoration, estimating the con- liopleurus (Chaudoir) over conventional tillage servation value of private land, providing in- treatments (Horne & Edward 1998). Near Mel- ventory information and basic documentation bourne, in an even longer period of pitfall trap- to assess the value of protected areas and rem- ping, Horne (1992) recovered only 15 carabid spe- nant habitat patches, detecting spread of exotic cies, and over one season more than 80% of the taxa and their impacts, and broad evaluation 1939 individuals were Notonomus gravis (Chau- of conservation management. doir). Likewise, in the Australian Capital Terri- 2. Determining change along environmental gra- tory, B. A. Melbourne (unpubl., cited by Melbourne dients, including altitudinal gradients as possi- et al. 1997) found 24 species, representing 22 gen- ble harbingers of climatic change, and latitu- era. Ordination analyses indicated that two types dinal gradients, as well as more local scenarios of introduced grassland were characterised by es- such as studying the effects of habitat edges sentially the same carabid assemblage, although and barriers resulting from development. assemblages were somewhat more distinctive on some different native grasslands. However, each In attempting to determine whether a particu- of Melbourne’s grassland sites supported rather few lar taxonomic group merits attention as an ‘indi- species (range 2–9) and (as in Horne’s studies noted cator’, the suggestion by Kremen et al. (1993) of above) a very high proportion of total carabids (735/ examining responses across a steep environmen- 972) was a single species, Notiobia edwardsii tal gradient is an expedient way to proceed. ‘Grad- (Castelnau). sect analysis’ (Gillison & Brewer 1985) is the de- Similar inferences may be drawn from stud- liberate selection of transects which contain the ies in forests. Thus, only 18 of Tasmania’s ap- steepest environmental gradients present in an proximately 220 species of Carabidae were trapp- area, to ensure sampling of the greatest range of ed in a survey of eucalypt forests in the Picton variables; in their initial example, of vegetation, Valley in a survey involving 14 sites over 10 months Gillison and Brewer showed that gradsects cap- (Michaels & McQuillan 1985), and only two of tured more information than randomly placed tran- these were widespread and abundant. The assem- sects of similar lengths. C. Helman (unpubl., as blage in forest patches in New South Wales com- cited in Austin & Heyligers 1989, 1991) concluded prised 45 species (Davies & Margules 1998). that this approach is sound for surveying rainfor- Such results emphasise that many species pres- est patches, based on altitudinal gradients from ent may be scarce, so that — if we neglect such the coast to the tablelands of New South Wales, rare taxa of uncertain incidence and consequent and using the initial assumption that altitude was low monitoring value — few species of Carabidae may be useful as reliable monitors of habitat the major gradient. Within a gradsect other fac- change. Partly because of this, Carabidae have at- tors (such as substrate, size of habitat patch and tracted rather little attention compared to several degree of isolation) may determine the precise more diverse and abundant groups, such as Col- sites for replicated sampling, because many such lembola (Greenslade 1997) and ants (Andersen factors may not be included in the initial selec- 1997b), or to a variety of other groups reported in tion criteria for the gradient. A selection proce- some surveys. dure is discussed in detail by Austin and Heyligers ANN. ZOOL. FENNICI Vol. 35 • The role of ground beetles in monitoring programmes 167

(1991) and could form the basis for appropriate to using ants in monitoring restoration, for designs for arthropods. For Carabidae, vegetation example following mining activities when type and gross level of disturbance might be valid clear patterns of faunal turnover can sometimes for gradsect selection, but the multitude of local occur, and indices of biological integrity de- factors which influence the incidence and abun- vised (Majer & Beeston 1996), reflecting the dance of particular species must eventually be lessening of numbers of species from those in superimposed on this. pristine habitats under different regimes of There is urgent need to determine a series of land use. Parallel functional groups occur in ‘focal groups’ which can be used validly in such North America, and Andersen (1997a) sug- studies, and to establish adequate sampling pro- gested the likelihood of similar patterns else- tocols for their effective use. As noted earlier, good where in the world. In some ways, ants in Aus- baseline data involving carabids are sparse. Even tralia seem equivalent to carabids in the north- for highly sensitive and vulnerable areas, such as ern temperate region, where series of papers Australia’s restricted alpine zone (about which (such as those in Stork 1990, Desender et al. there is considerable concern over intensifying de- 1994) have demonstrated a wide variety of velopment for winter sports and its likely decline ecological groupings of considerable indicator with global warming), most groups of inverte- value. brates have not been enumerated comprehensively Characteristic levels of ant species richness despite many species being limited to those envi- are gradually being documented, and differ be- ronments. Some unusual flightless Carabidae are tween different regions and habitats. A series of narrowly endemic to alpine habitats. surveys of ants in grasslands at Mount Piper, Vic- Ants are by far the most widely documented toria yielded around 30 ant morphospecies on each group of insects involved in broader appraisal of (Miller & New 1997), whereas woodland surveys epigaeic faunas in Australia, and influences of hab- in the same region revealed around twice this num- itat change, predominantly through studies involv- ber on each plot (Hinkley & New 1997). These ing pitfall trapping (Andersen 1997a, Majer 1997). trends are borne out by other surveys in Victoria. Much of their interest stems from two main char- Change in the balance of functional groups has acteristics: both seasonal and habitat quality components, but 1. Australian ants are diverse and, although many it is also clear that differing levels of sampling of the 4 000 or so species are unnamed, the intensity may be needed at similar times of the genera are mostly recognisable unambiguous- year in different habitats to obtain reasonably com- ly. As a group, ants are trophically diverse and, plete inventory samples. Species accumulation therefore, ecologically informative. Local di- curves, for example, asymptote after different versity can be high: it is not unusual to exceed sampling periods in grassland and woodland, re- 100 species of ants/hectare in parts of the coun- flecting different levels of complexity in the ant try. faunas in these habitats (New et al. 1996). For 2. This ecological complexity, including associa- most groups of epigaeic invertebrates, including tions with a variety of other taxa, has led to carabids, such patterns have not yet been investi- designation of a series of ‘functional groups’ gated or established; phenological patterns are un- (Greenslade 1978, Andersen 1995), whereby known, but similar levels of continuous or inter- particular ant genera can be allocated (albeit val trapping may be needed to establish the basic at times tentatively) to a particular category, parameters for assessment. Any ‘spot’ samples and richness (based on ‘morphospecies’) with- can be very misleading in the absence of longer in genera and functional groups related to fea- term survey data to establish the template against tures and disturbance regimes of the habitats, which these can be better appraised. and ranked to help in assessment of broader For most pitfall trapped invertebrates, patchi- diversity. Thus, some genera (such as Rhytido- ness of incidence is high — and, therefore, pre- ponera Mayr) are ‘opportunists’ and can be- dictability and definable pattern is correspond- come abundant following disturbance in open ingly low. A similar inference was drawn by habitats. Considerable attention has been paid Greenslade (1997) who, writing on Collembola 168 New • ANN. ZOOL. FENNICI Vol. 35 in native grasslands, noted ‘intrinsic variation in rank abundance of beetle families in these sites, these ecological systems is too great for a linear carabids rank 3 (pasture), 4 (Grey Box) and 5 response to disturbance to be shown by the inver- (River Red Gum). tebrates studied here, which makes it unlikely that As another example, a temperate region for- a single, reliable indicator can be found’. No con- est context, a trapping programme to evaluate the gruence was found between species richness of effects of forest fragmentation on epigaeic inver- ants, carabids and Collembola, so that use of any tebrates at Wog Wog, New South Wales (Margu- of these as a surrogate group alone might be mis- les 1992) yielded approximately 554 species of leading. In relation to ants and Collembola, beetles, of which about 432 could not be named ground-dwelling Carabidae appear genuinely to species level. One species of Notonomus was scarce. This may reflect low ground humidity in appraised in more detail, but ‘patchiness’ may pre- many habitats. In discussing the abundance of sub- clude effective evaluation of trends: Margules cortical carabids in Australia, Baehr (1990) noted noted that two of his plots had very low numbers that much of the continent has few geophilous of the beetle, for no apparent reason. Such differ- ground beetles, but it is intriguing that this pau- ences in abundance may reflect microclimate or city seems to extend to the more mesic parts of other features not yet quantified, but apparently Australia, in which most surveys have been un- similar plots in a variety of vegetation types com- dertaken. monly yield very different arrays of beetles. The Another, broader example also indicates the carabid fauna of Wog Wog included representa- possible place of Carabidae in monitoring and eval- tives of 13 tribes (Davies & Margules 1998), but uation exercises. In an attempt to determine the only 8/45 species accounted for 92% of all indi- indicator values of epigaeic invertebrates in rem- viduals. Only five species were represented by nant woodland systems in northern Victoria, Yen 100 or more individuals in that extensive survey et al. (1996) compared River Red Gum (Eucalyptus of 144 monitoring sites. camaldulensis)-dominated sites, Grey Box (E. mi- Similar heterogeneity was found in Tasmanian crocarpa)- dominated sites and cleared land by a eucalypt forests (Michaels & McQuillan 1995), combination of pitfall trapping, sweeping and where species richness at particular sites ranged timed direct searches, the programme incorporat- from 4 or 5 (20 year old regeneration sites) to 11 ing 32 sites and extending at intervals for over a (at one old growth site). Only two common spe- year. Thirty five families of Coleoptera were col- cies (Sloaneana tasmaniae (Sloan), Rhabdotus re- lected, with several of these (Anthicidae, Curculio- flexus (Chaudoir)) occurred at all sites, and 8 spe- nidae, Elateridae, Mordellidae, Staphylinidae) cies occurred at three or fewer sites. Additional more abundant than Carabidae in pitfalls. The in- hand-collecting showed that some species (such ference may be that carabids are only one of sev- as Scopodes tasmanicus Bates) were probably eral beetle families which merit further attention under-represented in pitfall traps and were com- to determine their possible indicator values. In mon under bark of fallen trees. Michaels and

Table 1. Percentage occurrence of the five most abundant carabid species trapped in different regeneration age classes in Eucalyptus forest in southern Tasmania (from Michaels & McQuillan 1995). ————————————————————————————————————————————————— Regeneration age (years) ——————————————————————— Species 1–3 7–10 20+ old growth ————————————————————————————————————————————————— Sloaneana tasmaniae 10 70 10 10 Rhabdotus reflexus 10** 60** 20** 20* Promecoderus spp. 2 80 7 12 Mecylothorax ambiguus*90 10 0 0 Notonomus politus 26 7 1 65 ————————————————————————————————————————————————— *Winged, all other species flightless. ** As in published table. ANN. ZOOL. FENNICI Vol. 35 • The role of ground beetles in monitoring programmes 169

McQuillan suggested that Trechini (a diverse with groundbeetles here! In temperate southern group of small taxa frequenting moss and litter) Australia, in particular, several large and complex might also have been undersampled. Two points longterm trapping programmes for epigaeic in- of more general relevance emerge from this study. vertebrates have been undertaken recently: that First, that strong seasonal activity necessitates by Yen et al. noted above is only one example of sampling over most of the year in order to get those which have yielded standard samples from adequate representation of the fauna and its re- a variety of sites reflecting various forms of land sponses. Second, because many of the species in use or ecological gradients, as bases for ordina- a survey are local endemics, the degree of valid tion analyses. The difficulties of obtaining spe- extrapolation even to other notionally similar for- cies-level identifications and, therefore, of apprais- est environments is very limited. Nevertheless, ing the samples at the most meaningful level, have there were some clear differences in the carabid retarded analyses of most of these studies. How- assemblages of different stages of forest regen- ever, morphospecies can usually be delimited con- eration after felling, and the study reported by Mi- sistently within a genus. During the next few years, chaels and McQuillan (1995) was undertaken in it is likely that the Carabidae from several of these part to investigate parallels of Carabidae in Tas- surveys will be appraised in more detail, follow- mania to the more extensive studies in North ing the example set by B. A. Melbourne (unpubl.). American forests (for example by Holling 1992, At that time, we should be able to more effec- Niemela et al. 1993) Notwithstanding the low tively assess their potential as indicators of dis- abundance of many species, each of the five most turbance or habitat type, not least because the ‘typ- abundant species showed proportional peaks at ical’ habitat of some species of open environments particular stages of regeneration, and ranged from will be defined far more clearly than at present. early successional species to those representative At present, our data are fragmentary, our ex- of old growth forest (Table 1). As in other stud- pectations of Carabidae too vague and inconsist- ies, flightless species are assumed to be resident. ent, and there is insufficient detailed knowledge of carabid biology. We are unable to evaluate assem- blages of carabids effectively, because of the un- 5. The future known significance of the numerous rare species, but it is necessary to employ continuous sampling Development of monitoring techniques using in- regimes rather than interval trapping to address this vertebrates is an active field of endeavour in Aus- effectively, and to gain even reasonable inventory tralia, together with critical evaluation of the tech- data on regional faunas. Even this more compre- niques and sampling protocols needed to provide hensive approach may reveal many species to be reliable information. Debate on the optimal focal highly unpredictable and patchy in incidence, both groups to appraise is continuing. The shortcom- in space and time, and direct our attention to more ings of pitfall traps are recognised widely, to the detailed studies of the factors influencing more extent that more than a dozen contributors to a re- abundant taxa in the assemblage. At present these cent symposium on invertebrate biodiversity (Yen are largely unknown and undocumented. Carabid & New 1997) addressed these in some detail. diversity in many Australian temperate regions ap- Despite uncertainty over the most informative pears to be generally less than in comparable north- invertebrate groups to study in such programmes, ern hemisphere systems, with patterns of species ants, spiders (for which a series of functional dominance also unusual. However, the relatively groups are now definable: Churchill 1997) and few abundant species provide clear focal points for Collembola are at present particularly attractive investigating responses to environmental change to investigate further. In contrast to the widespread and correlations with land use. use of carabids in some other parts of the world, In selecting a suitable portfolio of taxa to use there is currently little effective advocacy for the in monitoring land use, condition and change in group in Australia, other than by expatriates fa- southern Australia, it is premature to confirm that miliar with their values in the northern hemisphere Carabidae would be included. Nevertheless, in rec- and who feel that we should be able to do better ognising their worth elsewhere, it is equally pre- 170 New • ANN. ZOOL. FENNICI Vol. 35 mature to dismiss them before more critical evalu- Eberhard, S. & Spate, A. 1995: Cave Invertebrate Survey: ation of their responses to change. toward an atlas of NSW cave fauna. — NSW Heritage Association Program Report, NSWNPWS, Queanbeyan, NSW. Acknowledgements: I greatly appreciate comments on Freitag, R. 1979: Reclassification, phylogeny and zoogeo- a draft of this paper from Dr. Paul Horne, Prof. Nigel Stork graphy of the Australian species of Cicindela (Coleopte- and an anonymous referee. This paper is based on a presen- ra: Cicindelidae). — Aust. J. Zool., Supplement 66: 1– tation at a symposium associated with the Helsinki Globenet 99. workshop and my participation was made possible by fund- Gillison, A. N. & Brewer, K. R. W. 1985: The use of gradi- ing organised through Prof. Jari Niemela and Dr. Johan Kot- ent directed transects or gradsects in nature resource ze, whom I thank also for their initiative in organising such surveys. — J. Environ. Manag. 20: 103–127. a stimulating enterprise. Graetz, R. D., Wilson, M. A. & Campbell, S. K. 1995: Land- cover disturbance over the Australian continent. A con- temporary assessment. — Paper no. 7. Biodiversity References series, Department of Environment, Sports and Terri- tories, Canberra. Andersen, A. N. 1995: Measuring more of biodiversity: ge- Greenslade, P. 1997: Are Collembola useful as indicators nus richness as a surrogate for species richness in Aus- of the conservation value of native grasslands? — Pedo- tralian ant faunas. — Biol. Conserv. 73: 39–43. biologia 41: 215–220. Andersen, A. N. 1997a: Functional groups and patterns of Greenslade, P. & New, T. R. 1991: Australia: conservation organisation in North American ant communities: a of a continental insect fauna. — In: Collins, N. M. & comparison with Australia. — J. Biogeogr. 24: 433– Thomas, J. A. (eds.), The conservation of insects and 460. their habitats: 33–70. Academic Press, London. Andersen, A. N. 1997b: Ants as indicators of ecosystem Greenslade, P. J. M. 1978: A guide to the ants of South restoration following mining: a functional group ap- Australia. South Australian Museum, Adelaide. proach. — In: Hale, P. & Lamb, D. (eds.), Conserva- Hinkley, S. & New, T. R. 1997: Pitfall trapping for survey- tion outside nature reserves: 319–323. Centre for Con- ing ant assemblages: lessons from a study at Mount servation Biology, University of Queensland. Piper, Victoria. — Mem. Mus. Vic. 56: 369–376. Austin, M. P. & Heyligers, P. C. 1989: Vegetation survey Holling, N. J. 1992: The carabid fauna (Coleoptera: Carabi- design for conservation: gradsect sampling of forests dae) during postfire regeneration of boreal forest: prop- in north-eastern New South Wales. — Biol. Conserv. erties and dynamics of species assemblages. — Can. J. 50: 13–32. Zool. 70: 440–452. Austin, M. P. & Heyligers, P. C. 1991: New approaches to Horne, P. A. 1990: Parental care in Notonomus Chaudoir vegetation survey: gradsect sampling. — In: Margules, (Coleoptera: Carabidae: ). — Aust. Ento- C. R. & Austin, M. P. (eds.), Nature conservation: cost mol. Mag. 17: 65–69. effective biological surveys and data analysis: 31–36. Horne, P. A. 1992: Comparative life histories of two spe- CSIRO, Melbourne. cies of Notonomus (Coleoptera: Carabidae) in Victo- Baehr, M. 1990: The carabid community living under the ria. — Aust. J. Zool. 40: 163–171. bark of Australian eucalypts. — In: Stork, N. E. (ed.), Horne, P. A. & Edward, C. L. 1997: Preliminary observa- The role of ground beetles in ecological and environ- tions on awareness, management and impact of biodi- mental studies: 3–11. Intercept, Andover. versity in agricultural ecosystems. — Mem. Mus. Vic. Baehr, M. 1995: Revision of Philipis (Coleoptera: Carabi- 56: 281–285. dae: Bembididiinae), a genus of arboreal tachyine bee- Horne, P. A. & Edward, C. L. 1998: Effects of tillage on tles from the rainforests of eastern Australia: taxonomy, pest and beneficial beetles in the Wimmera region of phylogeny and biogeography. — Mem. Qld. Mus. 38: Victoria, Australia. — Aust. J. Entomol. 37: 60–63. 315–381. IAC (Invertebrate Advisory Committee) 1994: Interim list Churchill, T. B. 1997: Spiders as ecological indicators: an of native invertebrates which are rare or threatened in overview for Australia. — Mem. Mus. Vict. 56: 331– Tasmania. Parks and Wildlife Service, Tasmania. 337. Kirkpatrick, J. B., McDougall, K. & Hyde, M. 1995: Aus- Darlington, P. J. 1961: Australian carabid beetles. 1. Tran- tralia’s most threatened ecosystems. The southeastern sition of wet forest faunas from New Guinea to Tasma- lowland native grasslands. — Surrey Beatty, Chipping nia. — Psyche 68: 1–24. Norton. Davies, K. F. & Margules, C. R. 1998: Effects of habitat Kremen, C., Colwell, R. K., Erwin, T. L., Murphy, D. D., fragmentation on carabid beetles; experimental evi- Noss, R. F. & Sanjayan, M. A. 1993: Terrestrial arthro- dence. — J. Anim. Ecol. 67: 460–471. pod assemblages: their use in conservation planning. Desender, K., Dufrene, M., Loreau, M., Luff, M. L. & Mael- — Conserv. Biol. 7: 796–808. fait, J. P. (eds.) 1994: Carabid beetles: ecology and evo- Lawrence, J. L. & Britton, E. B. 1994: Australian beetles. lution. Kluwer, Dordrecht. — Melbourne University Press, Melbourne. ANN. ZOOL. FENNICI Vol. 35 • The role of ground beetles in monitoring programmes 171

Majer, J. D. 1997: The use of pitfall traps for sampling ants Government Publishing Service, Canberra. — a critique. — Mem. Mus. Vic. 56: 323–329. New, T. R. 1993: Angels on a pin: dimensions of the crisis Majer, J. D. & Beeston, G. 1996: The biodiversity integrity in invertebrate conservation. — Amer. Zool. 33: 623– index: an illustration using ants in Western Australia. 630. — Conserv. Biol. 10: 65–73. New, T. R. 1995: Introduction to invertebrate conservation Margules, C. R. 1992: The Wog Wog habitat fragmenta- biology. — Oxford University Press, Oxford. tion experiment. — Environmental Conservation 19: New, T. R. 1998: Invertebrate surveys for conservation. — 316–325. Oxford University Press, Oxford. Matthews, E. G. 1980: A guide to the genera of beetles of New, T. R., Britton, D. R., Hinkley, S. D. & Miller, L. J. South Australia. Part 1. — South Australian Museum, 1996: The ant fauna of Mount Piper and its relevance Adelaide. to environmental assessment and the conservation of a Melbourne, B. A., Gullan, P. J. & Su, Y. N. 1997: Interpret- threatened invertebrate community. Flora and Fauna ing data from pitfall-trap surveys: crickets and slugs in Technical Report no. 143. — Department of Natural exotic and native grasslands of the Australian Capital Resources and Environment, Victoria. Territory. — Mem. Mus. Vic. 56: 361–367. Niemela, J., Langor, D. & Spence, J. R. 1993: Effects of Michaels, K. F. & McQuillan, P. B. 1985: Impact of com- clear-cut harvesting on boreal assem- mercial forest management on geophilous carabid bee- blages (Coleoptera: Carabidae) in western Canada. — tles (Coleoptera: Carabidae) in tall, wet Eucalyptus obli- Conserv. Biol. 7: 551–561. qua forest in southern Tasmania. — Aust. J. Ecol. 20: Stork, N. E. (ed.) 1990: The role of ground beetles in eco- 316–323. logical and environmental studies. — Intercept, An- Miller, L. J. & New, T. R. 1997: Mount Piper grasslands: dover. pitfall trapping of ants and interpretation of habitat vari- Yen, A. L. & Butcher, R. J. 1997: An overview of the con- ability. — Mem. Mus. Vic. 56: 377–381. servation of non-marine invertebrates in Australia. — Monteith, G. B. 1995: Distribution and altitudinal zonation Environment Australia, Canberra. of low vagility insects of the Queensland wet tropics. Yen, A. L. & New, T. R. (eds.) 1997: Proceedings of the Report to the Wet Tropics Management Authority, conference ‘Invertebrate Biodiversity and Conserva- Queensland. tion’. — Mem. Mus. Vic. 56: 261–675. Moore, B. P. 1980: A guide to the beetles of south-eastern Yen, A. L., Hinkley, S. D., Horne, P. A., Milledge, G. A. & Australia. Fasc. 1–2. — Australian Entomological Press, New, T. R. 1996: Development of invertebrate indica- Greenwich. tors of remnant grassy-woodland ecosystems. — Re- Moore, B. P., Weir, T. A. & Pyke, J. 1987: Zoological Cata- port for Australian Nature Conservation Agency Save logue of Australia. Vol. 4. Coleoptera. — Australian the Bush Program.