Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Review of the distribution, causes for the decline andrecommendations for management of the quokka,Setonix brachyurus (Macropodidae: Marsupialia), anendemic macropodid marsupial from south-westWestern Australia
PAUL J. DE TORES1*, MATT W. HAYWARD2,3,4, MICHAEL J. DILLON3 AND ROBERT I. BRAZELL5
1 Department of Environment and Conservation, Wildlife Research Centre,P.O. Box 51 Wanneroo, Western Australia 6946
2 School of Biological, Earth and Environmental Science, University of New South Wales,Sydney, New South Wales 2052
3 Department of Environment and Conservation, Dwellingup Research Centre,Banksiadale Road, Dwellingup, Western Australia 6213
4 Current address: Mammal Research Institute, Polish Academy of Science,17-230 Biglowieja, Poland
5 Department of Environment and Conservation, Wellington District, P.O. Box 809 Collie, Western Australia 6225* Contact person for correspondence: Paul de Tores, email: [email protected]
ABSTRACT
The former and current distribution of the quokka,Setonix brachyurus, was mapped from published and allavailable unpublished records. At the time of Europeansettlement the quokka was widespread and abundant andits distribution encompassed an area of approximately41 200 km2 of south-west Western Australia inclusive oftwo offshore islands, Bald Island and Rottnest Island.Historical reports indicated an extensive populationdecline occurred in the 1930s. The decline continued,with a previously undocumented decline apparent in theperiod from 1980 to 1992. However, this decline maybe an artefact of the time scales used for mapping andmay well equate with a previously reported decline for asuite of south-west mammals in the 1970s. By 1992 thequokka’s distribution had been reduced to an area ofapproximately 17 800 km2. An increased awareness ofthe presence of the quokka on the mainland has resultedin numerous reportings of quokka presence since 1992,has confirmed the existence of several populations at thenorthern extent of the quokka’s known geographic rangeand indicated the current, 2005, distribution to be similarto that in 1992. However, survey and populationestimates at six of these mainland locations from thenorthern jarrah forest indicated low abundance. Therehave been no population estimates elsewhere on themainland. Two populations have been reported fromthe Swan Coastal Plain, but neither has been confirmedextant.Predation by the introduced fox, Vulpes vulpes, isimplicated as a major cause of the quokka’s initial decline,while ongoing predation, habitat destruction andmodification through altered fire regimes havecontributed to the continued decline.
Specific conservation management actions arerecommended, namely: (i) Implementing an activeadaptive management program in the northern jarrahforest to determine quokka population response tohabitat manipulation through the use of fire, fox baitingand pig control; (ii) Surveying the Stirling Range andGreen Range populations with emphasis placed ondetermining population size and population geneticstructure; (iii) Surveying the reported occurrences fromthe Swan Coastal Plain, with emphasis on unambiguouslydetermining presence. If confirmed, priority should bedirected to assessing population size and determiningthe management requirements to ensure persistence ofthe population; (iv) Surveying southern forest and southcoast populations to assess quokka population size, theextent of movement between subpopulations andassessment of the range of habitat types used by quokkas.The latter should be combined with spatial analyses ofknown extant populations and suitable and potentiallysuitable habitat; (v) Determining the role of fire inestablishing and maintaining preferred habitat of southernforest and south coast populations; and (vi) Establishinga program to assess the potential effects frommanagement operations.
Keywords: quokka, Setonix brachyurus, distribution, foxpredation, fire, adaptive management
INTRODUCTION
The quokka, Setonix brachyurus (Macropodidae:Marsupialia) (Quoy & Gaimard 1830) is a small to mediumsized macropodid marsupial, endemic to the mainland ofsouth-west Western Australia and two offshore islands –
Conservation Science W. Aust. 6 (1) : 13–73 (2007)
14 P.J. de Torres et al.
Bald Island, east of Albany and Rottnest Island, 20kmwest of Perth (Fig. 1). Adult males range in weight from2.7 to 4.2 kg and adult females from 1.6 to 3.5kg(Hayward et al. 2003; Kitchener 1995). The quokka isknown to the Aboriginal, or Noongar, people of south-west Western Australia by a range of names including‘Ban-gup’, ‘Bungeup’, ‘Quak –a’ (Gould 1863;Shortridge 1909), ‘kwoka’ and ‘bangop’ (Abbott 2001a).
The quokka was the second Australian marsupialrecorded by Europeans, the first appears to have beenFrancis Pelsart’s description of the Tammar Wallaby fromthe Houtman Abrolhos on 1629. The first Europeanrecord of a quokka is attributed to Samuel Volckertzoon(or Volckersen) who, in 1658, when visiting the thenun-named Rottnest Island off the coast near Perth,described the quokka as ‘a wild cat resembling a civet-cat but with browner hair’ (Alexander 1914). In 1696when Willem de Vlamingh visited the island he describedthe quokka as ‘a kind of rat as big as a common cat’(Alexander 1914). He named the island Rottenest (nowRottnest), meaning rat’s nest. The quokka is oftenerroneously reported by local media as occurring onlyon Rottnest Island. Its presence on the mainland wasdescribed in 1957 as largely unreported (Barker et al.1957) and it is still widely thought of as only occurringon Rottnest Island.
At the time of European settlement of south-westWestern Australia, the mainland distribution of thequokka was thought to extend from the Moore Riverdistrict 80–100 km north of Perth (Baynes 1979), toareas south and south-east of Perth at locations withinthe Swan Coastal Plain, the jarrah, Eucalyptus marginata,and karri, E. diversicolor, forests south-east of Perth, theCape Leeuwin–Cape Naturaliste region, the south coastand east to the Hunter River east of Bremer Bay (Baynes1979; Baynes et al. 1975; Glauert 1933; Kitchener 1995;Ride 1970) (Fig. 1). In addition to the Rottnest Islandand Bald Island populations, quokkas were reported fromBreaksea Island, near Albany (Western AustralianMuseum records), ‘Twin Peak’ Island and other off shoreislands near Esperance (Shortridge 1909). Subsequentreferences to the quokka’s historic distribution on themainland, at or prior to the time of European settlement,have included the south coast of Western Australia as fareast as Esperance (Shortridge 1909; Troughton 1965).
This inferred former distribution was based on theabove reports and location records. All location recordswere within four biogeographical regions, or bioregions,as identified in the Interim Biogeographical Regionalisationfor Australia (IBRA) (Thackway & Cresswell 1995),namely the Swan Coastal Plain, the Jarrah Forest, theWarren and the Esperance Plains bioregions (Fig. 1).
Figure 1. The four biogeographical regions, as defined by the Interim Biogeographical Regionalisation for Australia (IBRA)(Thackway and Cresswell, 1995), within south-west Western Australia which encompass all records of occurrence of the quokka,Setonix brachyurus. Previously unpublished trapping locations are shown as + for sites trapped in the period 1992 to 2002 toconfirm presence and/or to derive an estimate of abundance
Distribution of the quokka 15
Various authors have expressed concern at the extentof the quokka’s decline from this inferred historicdistribution and attributed different levels of threat to itsconservation status. Clarke (1948) suggested the quokkawas almost confined to Rottnest Island. White (1952),although noting there was a misleading impression ofrarity on the mainland, conceded the quokka wasbecoming less abundant there. Loaring (in Serventy etal. 1954) believed quokkas had vanished from their ‘gullyhaunts’ in the Darling Plateau by the 1920s. Sharman(1954) noted the quokka appeared common only onRottnest and Bald Islands with isolated colonies at somemainland sites.
Barker et al. (1957) cited a 1957 newspaper articlewhich suggested the quokka was extinct on the mainland.That report generated interest in quokka populations,and in the same year, led to the confirmation of thepresence of a small colony of quokkas near Byford,immediately south of the Perth metropolitan area (Barkeret al. 1957). Hart et al. (1986) and similarly Perry (1973)also noted there was a widely held, yet incorrect, beliefthe quokka had become extinct on the mainland afterthe population crash of the 1930s. This collapse of quokkapopulations on the mainland in the 1930s is widelyreported and Ride (1970) recorded, prior to this collapse,that the quokka was common in the south-west andpopulations were sufficiently abundant for quokkashooting to be a ‘familiar sport’.
Serventy (in Serventy et al. 1954) reported anapparent increase in quokka numbers in the DarlingRange and Manjimup areas in the early 1950s. However,this seems to be contradicted by reports from Loaringand Glauert (also reporting in Serventy et al. 1954).Dell (1983) believed Serventy’s (1954) reported increasein numbers of the quokka and other species could notbe substantiated and was an artefact of an increasedawareness by observers. How et al. (1987), whenreporting on a vertebrate fauna survey for the areabetween Busselton and Albany, believed quokkapopulations had persisted on the mainland but wererapidly diminishing.
In an attempt to quantify the conservation status andcauses of decline for a suite of macropod species, Johnsonet al. (1989) concluded the quokka had experienced asubstantial (85–90%) decline in its geographic range onthe mainland. These authors identified the quokka aswarranting a high priority for conservation management.
The quokka is currently listed as a Threatened Species,in the sub-category Vulnerable, pursuant to the WorldConservation Union (IUCN) Red List of ThreatenedSpecies (Hilton-Taylor 2000). It is listed nationally asThreatened Fauna, in the sub-category Vulnerable,pursuant to the Commonwealth of Australia’sEnvironment Protection and Biodiversity Conservation Act1999 (EPBC Act) criteria for Vulnerable. In 1996 itwas placed on the Western Australian list of ‘fauna whichis rare or likely to become extinct’ pursuant to Section14(2)(ba) of the Western Australian WildlifeConservation Act 1950.
This paper reviews the distribution of the quokka,
assesses the probable causes of the species decline andproposes specific conservation managementrecommendations. The conservation status is reviewedin a forthcoming publication.
METHODS
Distribution maps and terminology
Location records for the quokka were obtained frompublished and unpublished accounts of fossil, subfossiland surface cave deposits, published and unpublishedrecords of distribution/presence, database records ofthe Western Australian Museum (WAM), databaserecords of the Australian Museum, database records ofthe Western Australian Department of Environment andConservation (DEC, formerly the Department ofConservation and Land Management, CALM) (CALMunpublished; Gilfillan unpublished). Additionalinformation on WAM records was supplied by NorahCooper (pers. comm. to PJdeT)1 and additional sub-fossil records were obtained from searches of theuncatalogued collection of the Palaeontology andAnthropology Sections of the Western AustralianMuseum (E. Jefferys pers. comm. to MWH)2 .
Records of presence were provided from ourpreviously unpublished field survey and trapping datafrom the early 1970s to 2002 with additional detaileddata from the period 1992 to 2002 from trapping at sixlocations within the northern jarrah forest – GervasseForest Block, Rosella Road, Albany Highway (Travellers’Arms Site), Holyoake Swamp, Kesners Swamp and Lewis(Wild Pig) Swamp (Fig. 1.). Field survey consideredquokkas to be present at a site if confirmed throughtrapping, recent roadkills or by the presence of scatswithin characteristic runways in densely vegetated areasconsidered typical of quokka habitat (see Barker et al.1957; Christensen et al. 1985; Christensen & Kimber1975; Dillon 1993; Kitchener 1995; Maxwell et al. 1996;Sinclair 1999). The presence of scats in open areas andaway from characteristic runways, was consideredinsufficient to infer presence (Alacs et al. 2003).
Locations from all sources were mapped using thegeographic information system (GIS) software ArcGIS(ESRI 1999-2004). Location records were excludedwhen they were unable to be validated and/or if notsupported by location co-ordinates, or if location co-ordinates could not be determined. The inferred pre-European distribution (inclusive of sub-fossil records)was mapped from these records, as was the distributionat six subsequent times, namely at the time of Europeansettlement, at 1920, 1950, 1980, 1992 and 2005. Theinferred distribution at the time of European settlementand at 1920 pre-dates arrival of the fox and the firstevidence of fox predation in south-west Western Australia.The 1950 distribution includes records from 1921 to
1 Norah Cooper: Curator of Mammals, Western Australian Museum, Perth2 Elizabeth Jefferys: University of New South Wales
16 P.J. de Torres et al.
1950. The 1980 distribution includes records from 1951to 1980. The 1992 distribution includes records from1981 to 1992. The 2005 distribution is inferred fromrecords post 1992, where populations were verified orwhere there was no evidence to indicate previously knownpopulations had not persisted. These time frames allowquantitative and visual assessment of the decline fromthe pre-European distribution. Our use of the term ‘pre-European distribution’ is synonymous with the term‘original distribution’ as used by Baynes (1987). Thecurrent (2005) distribution was also mapped to depictdensity of sighting and location records. Densitycategories were derived through kernel analysis inArcGIS. Densities reflect the density of sighting/locationrecords and do not necessarily equate with populationdensities.
Areas surveyed, population estimates andsurvivorship
Estimates of population size were obtained frompublished accounts and our previously unpublishedsurvey and trapping records at locations within thenorthern and southern jarrah forests, where the northernjarrah forest is broadly defined as the jarrah, marri andwandoo forests north of the Preston River (Fig. 1). Thesouthern forest is broadly defined as the jarrah, marri,
karri, tingle and wandoo forests south of the PrestonRiver, north and west of the Frankland River. The southcoast refers to the area east of the Frankland River (Fig.1).
The Gervasse site (Fig. 1) is subject to an ongoingmonitoring program (capture-mark-release-recapture)and capture data were analysed for eleven trappingsessions conducted in June/July 1992, March 1993,June/July 1993, January 1994, August 1994, April 1995,January 1996, January 1997, February 1998, March1999 and February 2000. To enable comparison withpopulation estimates from five other northern jarrahforest sites (Hayward et al. 2003), estimates of populationsize from the Gervasse site were derived from the ‘deathbut no immigration’ Jolly-Seber (JS) model using thesoftware interface modified to accommodate unequaltime periods between trapping sessions (Barker &McGlinchy 2001).
‘Known to be alive’ (KTBA) estimates were alsoderived for the Gervasse site, where an animal trappedat trapping session t(n-i), not trapped at t(n) andsubsequently trapped at t(n+i) is assumed to be alive att(n). The number of captures at the other five sitesreported from the northern jarrah forest was too low toderive an estimate of, or index to, population size. Wherepresence was confirmed, data from these sites are reportedas the total number of individuals trapped.
Figure 2. The southwest of Western Australia showing location records for the quokka, Setonix brachyurus. The full source for eachrecord is listed in tables 1 to 8.
Distribution of the quokka 17
RESULTS
Records of occurrenceFossil and subfossil records are shown in Table 1,Australian Museum records in Table 2, published recordsin Table 3, unpublished records (from various sources)in Table 4, unpublished records from the Departmentof Conservation and Land Management databases, nowDepartment of Environment and Conservation (DEC)(CALM unpublished; Gilfillan unpublished) in Table 5.Our previously unpublished location data for extantpopulations are shown in Tables 6 and 7. All recordswere mapped and shown in Fig. 2. Data from theWestern Australian Museum (Kitchener & Vicker 1981)and additional WAM records (WA Museum databaserecords; pers. comm. from E. Jefferys; pers. comm.from N. Cooper) are presented in mapped form only.Five of the records unable to be validated or unable tobe confirmed are shown in Table 8. These records arealso mapped in Fig. 2, but excluded when inferringdistribution. The population translocated to a predator-proof fenced sanctuary (Karakamia Sanctuary, east ofPerth) is also listed in Table 8 and mapped in Fig. 2,but also excluded from inferred distribution, as is thecaptive colony formerly housed at the University ofWestern Australia and the location of a failedtranslocation – the Harry Waring Marsupial Reserve,Jandakot (Short et al. 1992). Two unconfirmed records
(Muddy Lake, south of Bunbury and a WaterCorporation reserve near Dunsborough) are also listedin Table 8 and mapped in Fig. 2, but excluded frominferred distribution mapping.
Estimates of population size andsurvivorship
The ‘deaths but no immigration’ Jolly-Seber estimate ofpopulation size for Gervasse was 49 ±7.9 and is shownin Table 7, as are results from opportunistic trappingprograms initiated to determine presence at five othersites. Known to be alive (KTBA) estimates for the period1992 to 2000 for the Gervasse population are shown inTable 9.
DISTRIBUTION
Distribution pre European settlement
There is no evidence to suggest the quokka has extendedits distribution since European settlement and allpublished and unpublished accounts imply a progressivecontraction of this distribution. Therefore, distributionpre-European settlement is inferred from current andhistorically known location records and from subfossilrecords and is shown in Fig. 3. The age, i.e. sub-fossil,of the Hasting’s Cave deposits east of Jurien Bay (Table 1and Fig. 2) indicate this to be the northern extent of the
Figure 3. The inferred distribution of the quokka, Setonix brachyurus, pre European settlement and at five subsequent timeperiods, 1920, 1950, 1980, 1992 and 2005. The distribution at all time periods includes Rottnest Island and Bald Island. The brokenline indicates the 700mm rainfall isohyet, the solid line indicates the 1000mm rainfall isohyet.
18 P.J. de Torres et al.
quokka’s pre-European distribution, as concluded byBaynes (1979). Sub fossil deposits at Yanchep (Merrilees1965) and the series of deposits from Perth to JurienBay (E. Jefferys pers. comm. to MWH) indicate thisdistribution was continuous. The Hunter River record(Kitchener & Vicker 1981), east of Albany is the easternmost record. Therefore, the inferred pre-Europeanextent of occurrence on the mainland extends from JurienBay, southward through the Swan Coastal Plain, thejarrah and karri forest areas, the Leeuwin-Naturalisteregion, the south coast and east to the Hunter River,east of Albany. This distribution encompassed an areaof approximately 44 300 km2, inclusive of Rottnest Island(15.5 km2) and Bald Island (7.8 km2). This area isexclusive of Breaksea Island, North Twin Peak Islandand South Twin Peak Island and exclusive of the southcoast, east of the Hunter River, considered by Shortridge(1909) to be within the quokka’s original distribution.
The two records from Breaksea Island are from skullscollected by Ian Abbott3 in 1975 (Kitchener & Vicker1981). There have been no further records fromBreaksea Island and the true origin of these skulls isunclear. The skulls may be sub-fossil specimensrepresenting a natural population which subsequentlydied out after it became isolated on the Breaksea peninsulawhen it was separated from the mainland approximately7 000 years ago (Ian Abbott pers. comm. to PJdeT).Alternatively, explanations (as proposed by Ian Abbott,pers. comm. to PJdeT) include: (i) quokkas still occuron Breaksea Island but have been overlooked; (ii) sealersknown to have been operating in the area from the 1790s,and/or Noongar women held captive on the Island,caught quokkas on the mainland and left their remainson Breaksea Island. This is supported by historicalrecords which refer to sealers marooning Noongarwomen on islands; and (iii) residents on the Island,including a lighthouse keeper present there from 1858,or other visitors, attempted and failed to establish aquokka population on the Island.
Recent work examining the effect of weeds on nestingseabirds has involved considerable time traversing allvegetation types on Breaksea Island and has failed todetect any sign of quokka activity or presence and studiesexamining sub-fossil remains of seabirds have notidentified fossil or subfossil quokka remains (PeterCollins, pers. comm. to PJdeT)4 . Although quokkaremains may have been overlooked (Peter Collins, pers.comm. to PJdeT), there seems to be insufficient evidenceto include Breaksea Island within the original distributionof the quokka.
Shortridge (1909) provided the sole reference toquokka presence on ‘Twin Peak Island’, presumablyreferring to either North Twin Peak Island or SouthTwin Peak Island in the Recherche Archipelago, off the
coast at Esperance. Glauert (in Serventy et al. 1954)described the quokka as plentiful on Rottnest Island and‘some islands off the south coast’. However, there are noother records of occurrence of the quokka from theislands off Esperance and numerous authors (Abbott &Burbidge 1995; Burbidge 1977; Calaby 1971; Glauert1933; Maxwell et al. 1996; Sharman 1954; Waring 1956)list the quokka as present on only Rottnest Island andBald Island. Quokka presence was not recorded on anyof the islands in the Recherche group by Serventy(1953), however, the tammar wallaby, Macropus eugenii,was recorded as plentiful on Middle Island and NorthTwin Peak Island. Similarly, Kabay and Start (1976) wereunable to confirm presence of quokkas on either Northor South Twin Peak Island when they surveyed bothislands in 1976. The tammar and quokka do not occursympatrically on any island and various hypotheses havebeen forwarded to explain this (see Clarke 1948; Main1961; Serventy 1951; Serventy 1953). The mostparsimonious explanation for the lack of sympatricoccurrence is that the small to medium size insularmacropod fauna is determined by the climatic conditionsat the time of separation from the mainland, withpersistence determined by continued availability ofsuitable habitat. Where islands are small, inter-specificcompetition operates and one species only will persist.Main (1961) hypothesised an island area in excess of sixsquare miles was necessary to support a single speciesof small macropod. This hypothesis is consistent withthe occurrence of tammars only on Garden Island andquokkas only on Rottnest Island. It is also consistentwith the presence of quokkas only on Bald Island.
Given the size of North and South Twin Peak Islandand the presence of the tammar on North Twin PeakIsland, it is unlikely the quokka also occurred there. The‘Twin Peak Island’ record of Shortridge (1909) thereforeappears to be spurious and we believe the record is areference to the tammar wallaby, presumably from NorthTwin Peak Island. Thus, we have interpreted the pre-European distribution of the quokka to be exclusive ofthe islands of the Recherche and exclusive of the southcoast, east of the Hunter River. The absence of quokkasfrom the fossil and sub fossil records from this region isconsistent with this inferred distribution.
The inferred pre-historic distribution is not differentfrom the inferred distribution, pre-European settlement.Although abundant in the fossil record dating to thePleistocene (Balme et al. 1978) (Table 1), the quokkaappears to have been restricted to the south-west cornerof Australia. The mainland population was presumablysplit by rising sea levels, with the Rottnest Islandpopulation separated between 6 000 and 8 000 yearsbefore present (BP) (Churchill 1959). The Bald Islandpopulation was separated almost 10 000 years BP (Storr1965).
Quokka fossil deposits from Devil’s Lair were datedto 35 000 years BP (Balme et al. 1978), although morerecent techniques suggest an age in excess of 40 000years (Turney et al. 2001) (Table 1). Balme et al. (1978)concluded quokkas became increasingly abundant from
3 Ian Abbott: Science Advisor and Senior Principal Research Scientist,Western Australian Department of Environment and Conservation, ScienceDivision, Kensington
4 Peter Collins: Fauna Conservation Officer, Western Australian Departmentof Environment and Conservation, South Coast Region, Albany
Distribution of the quokka 19
32 000 years BP, then showed a slow decline, followedby an increase in abundance. The peak in abundancecoincided with a wetter but cooler climate and the periodof decline coincided with a period of intense aridity,relative to present conditions. Despite these variations,the quokka was still one of the most abundant speciesthroughout the Devil’s Lair deposits (Balme et al. 1978).Balme et al. (1978) concluded the persistence of thequokka throughout this period of climatic variationimplies a continuous presence of thickly vegetatedwatercourses and forest, although the forest speciescomposition may have changed. Importantly, the post-glacial period of aridity experienced in other parts ofsouthern Australia was not as severe in the south-westcorner of Australia, thus allowing forest dwellingmammals to persist (Balme et al. 1978).
Distribution at the time of Europeansettlement
The inferred distribution at the time of Europeansettlement is shown in Fig. 3. Published and unpublishedreports (Table 3 and 4) indicate the quokka was locallyabundant and the distribution encompassed an area ofapproximately 41 200 km2. This indicates a relativelyminor southern contraction from the pre-Europeandistribution.
With the exception of records north and north-eastof Albany, including the Stirling Range National Park,the distribution follows the pattern of rainfall, with recordsof occurrence confined to areas now receiving an averageannual rainfall of 700 mm or more (Fig. 3). Althoughthe distribution is shown as continuous, the Stirling Rangeand Green Range populations are isolated from all othersouth coast populations. The Stirling Range populationmay be in an area of higher rainfall (see climaticinfluences, below).
Distribution at 1920
The inferred distribution at 1920 shows no contractionfrom that at the time of European settlement (Fig. 3).
Distribution at 1950
The inferred distribution at 1950 is shown in Fig. 3encompasses an area of occurrence of approximately37 800 km2 and shows a contraction from that at 1920.
Reports of quokka abundance prior to the mid 1930scontrast to later reports (Tables 3 and 4). Prince (1984)included the quokka amongst the species of kangaroosand wallabies commercially exploited in the first 30 yearsof the 20th century. Records of the specific number ofquokkas harvested were not kept and harvests of thequokka and the tammar wallaby were combined withother small macropods. Potential commercial exploitationof the quokka ceased in 1952, with proclamation of theFauna Protection Act of 1950 (Prince 1984). As none
of these harvests was from Rottnest Island (R.I.T. Prince,pers. comm. to PJdeT)5 , the implication is the quokkawas sufficiently abundant on the mainland in the earlypart of the 20th century to enable it to be commerciallyexploited. These harvested animals were in addition tothe large number of quokkas shot in locally organized‘quokka shoots’ which occurred in the 1940s (Les Wilson,pers. comm. to PJdeT)6 and it seems the quokka waswidespread and abundant prior to the 1930s, albeit inrestricted habitat (Gould 1863; Perry 1971; Perry 1973;Shortridge 1909; White 1952). Long-term residents fromthe Northcliffe area (pers. comm. to PJdeT) reportedthat in the 1930s, and perhaps into the 1940s, quokkasregularly browsed in pasture well away from areas ofrestricted creekline habitat. Further evidence indicatingthe quokka was at high density prior to the decline inthe 1930s is provided by the 1933 proclamation of thequokka as ‘vermin’ pursuant to the Vermin Act, 1918(Government Gazette of Western Australia 1933).
The subsequent decline is reported anecdotally only.This decline is also only partially reflected by the declinein inferred distribution at 1950 (Fig. 3) which showsonly a moderate southward contraction from the inferreddistribution at 1920. Populations were still present, orinferred by subsequent ‘rediscoveries’ to have been present,as far north as the Perth metropolitan area. No westwardcontraction in range is apparent and similarly, populationspersisted on the Swan Coastal Plain. Although thedistribution at 1950 shows only a moderate contractionfrom 1920, the number of known locations clearlydeclined and abundance at each also appears to havedeclined. Quokkas may have persisted as far east as theHunter River area, east of Albany. However, supportiveevidence is provided by one WAM registered specimenrecord only, collected in 1970 (Kitchener & Vicker1981). The specimen is a part of a mandible (NorahCooper, pers. comm. to PJdeT), is undated and may beconsiderably older than the collection date implies. Thus,the westward contraction may have commenced priorto the collection date of the Hunter River specimen andthe eastern extent of the distribution at 1950 (and 1980,see below) may have extended no farther than theWaychinicup Bay/Waychinicup National Park and GreenRange area (Fig. 3).
Distribution at 1980
The inferred distribution at 1980 is shown in Fig. 3 andshows no contraction from that at 1950. However, asfor the pattern of decline described at 1950, recordsindicate the number of known populations had decreasedand most published accounts referred to the quokka asrare on the mainland (Table 3).
Distribution at 1992
The inferred distribution at 1992 is shown in Fig. 3encompasses an area of approximately 17 800 km2 and
5 R.I.T. (Bob) Prince: Senior Research Scientist, Western Australian Department of Environment and Conservation, Science Division, Woodvale6 Les Wilson: long-term resident of the Northcliffe area.
20 P.J. de Torres et al.
shows a marked contraction from the distribution at1980. With the exception of a recently reportedrediscovery at Muddy Lake, south of Bunbury (Dell &Hyder-Griffiths 2002) (Table 4) and an unconfirmedrecord from the Dunsborough area (Table 8), the lastrecords from the Swan Coastal Plain are from 1961 fromBibra Lake, at the southern outskirts of the Perthmetropolitan area (Kitchener & Vicker 1981) and from1975 and 1976 from the Bunbury/Muddy Lake area(Table 4) (Hart et al. 1986; Kitchener & Vicker 1981).The last confirmed record from the Cape Leeuwin–CapeNaturaliste area is from 1979 from Wardanup Hill, southof Dunsborough (Kitchener & Vicker 1981).
Populations were present, or inferred as still presentby subsequent ‘rediscoveries’, at Churchman Forest Block,within 10 km of the Perth metropolitan area. Exclusiveof Bald Island, the eastern most records are from theWaychinicup National Park area east of Albany and GreenRange, northeast of Albany. The Stirling Rangepopulation(s) is shown as isolated from other south coastpopulations and may have been isolated prior to 1992.
Current distribution
The inferred current distribution is shown in Figs 3 and4 and is unchanged from 1992.
Trapping in 1995–96 in the Dwellingup andManjimup areas at sites known to support quokkapopulations in the 1990s failed to confirm presence atnumerous sites (Dillon 1996) (Table 10). With theexception of the unconfirmed records from Muddy Lakeand Dunsborough, quokkas are now considered absentfrom the Swan Coastal Plain. The skull collected at MuddyLake (Dell & Hyder-Griffiths 2002) in September 2002has no associated soft tissue and is not aged (NorahCooper, pers. comm. to PJdeT). Its collection date doesnot necessarily infer presence of an extant population atthe time of collection. Quokka presence is otherwiseonly inferred at this location by the presence of runways.There is an unconfirmed report of quokka occurrencefrom a Water Corporation managed reserve in theDunsborough area and unsubstantiated anecdotal reportsof presence in the Cape Leeuwin–Cape Naturaliste area.
Figure 4. The inferred current (2005) distribution of the quokka, Setonix brachyurus, showing density of location / sightingrecords. The solid line indicates the 1000mm rainfall isohyet.
Distribution of the quokka 21
Presence of suitable habitat and anecdotal reports suggestquokkas may also still occur in the lower catchment ofthe Blackwood River, north-east of Augusta (RogerHearn, pers. comm. to PJdeT)7 . Data unavailable forthis review and referred to by Liddelow (2006) indicatedseveral populations may occur west of Nannup. Thecurrent inferred distribution (Fig. 3) includes theserecords, however the pattern of these records ofoccurrence combined with sites also examined byLiddelow (2006) where quokkas were previously knownto occur and are considered to no longer occur, suggestthis western forest margin may be experiencing the samedecline detected in the northern jarrah forest.
The most northerly records from the Darling Plateauare from Churchman Forest Block, approximately10 km south-east of the Perth metropolitan area (Table4). The confirmed sites from the northern jarrah forestextend in a narrow band along the Darling Plateau andare bounded by the 1 000 mm annual rainfall isohyet(Fig. 3). There are no confirmed records of extantpopulations between Collie (Davis Forest Block) andNannup (Boronia, Gregory and Helms forest blocks)(Table 6, Fig. 3). With the exception of the recordsfrom Boronia Forest Block and records east of Nannup(Liddelow 2006), the Nannup population(s) and forestpopulations farther south are also almost entirely withinthe 1 000 mm annual rainfall isohyet (Fig. 3). The mosteasterly mainland records are from the south coastnear Waychinicup National Park, east of Albany, frompresence inferred by collection of hair from GreenRange north-north-east of Albany (J.A. Friend, pers.comm. to PJdeT)8 and from the Stirling RangeNational Park, northeast of Albany. As depicted bythe inferred distribution at 1992, the Stirling Rangeand Green Range population(s) appear to be, and mayalways have been, isolated from other south coastpopulations.
Figure 4 shows the current inferred distribution basedon records of occurrence from 1992 to present. Thedensity gradient depicted (Fig. 4) reflects the density ofsighting/location records. It does not necessarily reflectpopulation densities and the nodes of inferred highpopulation density are, to some extent, likely to be anartefact of proximity to townsites and local interest inreporting sightings.
DISCUSSION
Distribution and population estimates
The contraction of distribution has been greatest fromthe northern extent of the geographic range and fromthe Swan Coastal Plain. The northern-most known extantpopulation is approximately 10 km south-east of Perth.With the exception of the unconfirmed records from
Muddy Lake and the Water Corporation reserve nearDunsborough, the quokka is now considered absent fromthe Swan Coastal Plain and the Cape Leeuwin–CapeNaturaliste area and occurs only as far east asWaychinicup National Park (Figs 2 and 3).
Surveys aimed at determining presence in forest areassince the early 1970s have confirmed quokka populationsfrom the northern (Dwellingup) and southern(Manjimup) forest areas are patchy and discontinuous.The known northern limit of the present distributionwas extended in 1995 by discovery of a population nearJarrahdale (Rosella Road) (Tables 6 and 7), in 1996 byre-discovery of a population north-east of Jarrahdale(Albany Highway/Travellers’ Arms) (Tables 6 and 7)and in 2001 and 2002 by discovery of populations withinChurchman Forest Block, 10 km south-east of the Perthmetropolitan area (John Liddington, pers. comm. toPJdeT, Table 4)9 . Although there are in excess of 40records of occurrence in the northern jarrah forest post1992 (Tables 3, 4, 5, 6 and 7), several of these recordsof occurrence may constitute a single population. Equallyimportantly, confirmation of presence does not equatewith persistence of a population. Figure 4 shouldtherefore be interpreted cautiously, as each sighting/location record does not necessarily reflect a separatepopulation and the density gradient depicted reflects thedensity of location/sighting records, not populationdensity. Although the nodes which depict highest densityin the southern forest, near Manjimup, and the southcoast, near Walpole-Nornalup National Park, concur withour observational and unquantified survey data whichsuggest these areas support the highest densitypopulations (exclusive of Rottenest Island), these nodesalso reflect proximity to townsites or areas where thereis local interest in reporting sightings. Therefore, thehigh density nodes are also likely to reflect the level oflocal survey effort and the level of local interest inreporting sighting records. They should not be seen asquantified estimates of abundance.
Estimates of abundance were available from six sitesonly, rarely reported more than twenty to thirty individualsand often comprised considerably fewer (Table 3 and7). The Gervasse population is the only knownpopulation shown to support in excess of forty individuals(Table 7). The northern-most sites appear to be atcritically low density. The Rosella Road population isthought to be at risk of local extinction (Hayward et al.2003). The Wild Pig Swamp (Lewis Swamp) sitesupported quokkas in the early 1990s (Dillon 1993),however despite the evidence of fresh activity, trappingfailed to confirm quokka presence when trapped in 1995(de Tores, Dillon, Tomkinson and Buehrig, Tables 6and 7) and in 1998–2000 (Hayward et al. 2003). TheHolyoake site is thought to be locally extinct (our dataand Hayward et al. 2003).
Results from published accounts (Christensen et al.
7 Roger Hearn: Regional Ecologist, Western Australian Department of Environment and Conservation, Manjimup8 J.A. Friend: Principal Research Scientist, Western Australian Department of Environment and Conservation, Science Division, Albany9 John Liddington: Water Authority of Western Australia, Armadale
22 P.J. de Torres et al.
1985) and our trapping surveys (Tables 6 and 7) haveconsistently returned low capture rates from these forestareas. Capture rates of 0.3 (Dillon 1993) and 0.07 to0.99 (Hayward et al. 2003) quokkas per 100 trap nightsfor sites in the Dwellingup area and 1.2 per 100 trapnights over three sites north of Dwellingup (Sinclair1999), are representative of trapping success at theselow density sites.
Hayward et al. (2003) believed the northern jarrahforest population formerly constituted a functionalmetapopulation which is now under threat of collapse.Collapse of this metapopulation, or collapse of theremaining catchment-confined populations, would resultin a considerable contraction of the quokkas’ distribution.Given the lag time required to detect a decline in apopulation, we believe there is an urgent need for acoordinated and site-prioritised monitoring program atknown quokka sites and appropriate quantitative analysesof monitoring results (see section on managementrecommendations).
An additional series of surveys has been conductedpost collation of the data reported here (GraemeLiddelow10 and Alan Wright11 , pers. com. to PJdeT)and indicate presence of quokkas at numerous additionallocations within the southern forest and northern jarrahforest. However, no data were collected on populationsize or movement between populations. The likelihoodof persistence of these populations is unknown.
Habitat use
Hayward (unpublished, reported in summary in de Toreset al. 2004) concluded all extant northern jarrah forestpopulations were associated with creeklines characterisedand dominated by the ti-tree, Taxandria linearifolia(formerly Agonis linearifolia), and further concludedquokka populations from the northern jarrah forest arerestricted to areas supporting a structural mosaic of burntand unburnt vegetation. These findings concur withrecords from the mid 1970s to 2002 (our records,various unpublished accounts and confirmed sightings).
Hayward (unpublished, reported in summary in deTores et al. 2004) also examined 66 sites in the northernjarrah forest to assess quokka presence/absence. GeneralLinear Modeling (GLM) and model selection throughuse of a mixed approach of stepwise removal of variablesand the Information-Theoretic approach and AkaikeInformation Criterion (AIC) (Burnham & Anderson2002) was used to identify the ‘best’ model to describethe preferred habitat of the quokka. We do not advocatethis mixed approach and recommend use of theInformation-Theoretic approach and selection from aset of a priori candidate models for any further analyses.
From the mixed approach, three models equally well
described the preferred habitat. The explanatory variablesof the preferred models were:(i) the number of 1080 meat baits delivered per hectare
(an increased number of baits was correlated withquokka presence);
(ii) the average age of the swamp in terms of yearssince last burn (there was a positive correlationbetween years since last burn and quokka presence);
(iii) a habitat factor score (NJF4) (characterised bypossessing large areas of Taxandria swamp burntfive to nine years previously – positively correlatedwith quokka presence);
(iv) a habitat factor score (NJF2) (characterized by ahigh proportion of jarrah – marri open forest andTaxandria swamp burnt 15–19 years previously –this variable was negatively correlated with quokkapresence); and
(v) increasing distance to anthropogenic disturbance.Hayward (unpublished, reported in summary in de
Tores et al. 2004) concluded these features highlightthe importance of a mosaic of age classes within theswamp. This mosaic needs to support early (< 10 yearspost-fire) and late (long unburnt) seral stages. Theintermediate seral stage (15–19 years post fire) is avoided.Therefore, this is a mosaic of specific age classes (youngand old) rather than simply a mosaic of mixed age classes.We recommend adopting an active adaptive managementprogram to determine whether conservationmanagement of the quokka can be improved, throughthe use of fire, to create and maintain this preferredmosaic (see management recommendations section).
The relative importance to the quokka of plant speciesadapted to different fire intervals also needs to beassessed. For example, species such as Taxandrialinearifolia will rapidly generate from root stocks andprovide a dense cover within one year post a spring fire(Kimber 1974), whereas many other species fromriparian zones generate from seed and are adapted tolonger intervals between fire (Abbott 1999; Burrows &Wardell-Johnson 2003). Potential conflict with otherconservation management priorities must also beaddressed. For example, riparian systems supportingquokka populations may also support fire sensitiverelictual plant taxa and communities (Burrows & Wardell-Johnson 2003). Therefore, we caution that any imposedfire regime and adaptive management approach shouldbe supported by an appropriate level of monitoring ofall species potentially affected by the regime.
Although present in the northern jarrah forest, thehabitat occupied there suggests the quokka is not aforest-dwelling species, or more specifically, is not aspecies of the more open northern jarrah forest. Historicaccounts of the appearance of the south-west forestsand the pattern of burning by Noongar people innorthern jarrah forest have been interpreted by Hallam(1975) as indicating the forests ‘comprised tall, straight,mature trees, all frequently scorched but clear of undergrowthand easy to move through’. If the forest structure was asdescribed by Hallam (1975), it would not have been
10 Graeme Liddelow: Senior Technical Officer, Western Australian Departmentof Environment and Conservation, Science Division, Manjimup
11 Alan Wright: Nature Conservation Officer, Western Australian Departmentof Environment and Conservation, Swan Region, Jarrahdale
Distribution of the quokka 23
conducive to a contiguous distribution of a speciesdependent upon swamp and creekline vegetation andmay have restricted the quokka to discrete catchments,with minimal dispersal and movement of individualsbetween catchments.
In southern forest areas, populations appear to beless discrete and the quokka inhabits a broader range ofhabitats including ti-tree thickets in the upper reaches ofcreek systems, dense streamside beds of rushes, karriregrowth, ridges supporting karri and tingle, Eucalyptusguilfoylei and E. jacksonii, forests (Christensen et al.1985) and occurs widely within forest areas supportingan understorey of spreading sword-sedge, Lepidospermaeffusum, and Anarthria scabra (Greg Freebury, pers.comm. to PJdeT)12 . Populations from the south coastalso appear to be less discrete than the northern jarrahforest populations and occur from Walpole-NornalupNational Park to the Mt Manypeaks area, northeast ofAlbany (Fig. 3). The extent of mixing (dispersal,immigration and emigration) between sub-populationsfrom the southern forest and the extent of mixingbetween sub-populations from the south coast areas isnot known.
Population genetics – the northern jarrahforest metapopulation
Hayward et al. (2003) hypothesised the northern jarrahforest formerly supported a metapopulation which is nowin a state of ‘terminal collapse’. Conversely, results froma study on the genetic structure (Alacs 2001) of thesame populations revealed these northern jarrah forestpopulations showed no evidence of mixing. The latterstudy suggests these populations may never haveconstituted a single metapopulation. However, the twofindings are not mutually exclusive. Alacs (2001) andHayward et al. (2003) examined the same geographicallyseparated populations (different catchments) from withinthe northern jarrah forest. We hypothesise there mayhave been mixing (gene flow) within catchments, butlittle or no mixing between catchments. Each catchmentmay have constituted a functional metapopulation.
Historical evidence from the Swan Coastal Plain isconsistent with this hypothesis and suggests thepopulations from the Coastal Plain were discrete andconfined to individual swamps and/or catchments orsub-catchments. We further hypothesise the populationsfrom the northern jarrah forest, although confined tocatchments or sub-catchments, mixed with populationsfrom the same catchment on the Swan Coastal Plain.Hence metapopulations may have existed, but under thisscenario, each catchment or subcatchment would havesupported a metapopulation. With progressivefragmentation of suitable habitat in the jarrah forest andon the coastal plain (draining of swamps and increaseddistances between suitable habitat patches), movement
between habitat fragments would have been less frequentand eventually ceased.
Factors influencing decline of the quokka
Numerous studies have attempted to determine the causesof the widespread decline in Australia’s mammalian faunasince European settlement (see for example Burbidge& McKenzie 1989; Calaby & Grigg 1989; Calver &Dell 1998a; Calver & Dell 1998b; Morton 1990; Recher& Lim 1990; Short et al. 2002; Short & Calaby 2001;Smith & Quinn 1996; Wilson & Friend 1999). Thesereviews have been selective and addressed some, butnot all of the plausible causes for decline. In acomprehensive analysis, Abbott (2001b) reviewed therole of fifteen possible causes for the decline of the bilby,Macrotis lagotis. His conclusion concurred with Watts(1969) i.e. that the fox was ‘the necessary and sufficientfactor associated with regional declines’ of the bilby.
Different studies have attributed the decline indistribution and abundance of the quokka to specific,but varying causes. A marked decline of the quokkawas recorded during the 1930s (White 1952). Christensen(1978; Christensen 1980b) presented evidence for adecline in a range of mammal species from south-westWestern Australia in the period from 1973 to 1978 andbelieved this decline was the result of an increased levelof predation by foxes. Christensen (1978) furthersuggested this increase in fox predation coincided with acessation of widespread 1080 baiting for rabbits, whichin turn led to a reduction of secondary poisoning offoxes. King et al. (1981) concurred with this belief. Thephenomenon of secondary poisoning of foxes wassubsequently demonstrated by Algar and Kinnear (1996).Our data suggest a period of marked decline in the extentof occurrence between 1980 and 1992 (Fig. 3).However, this period of decline may be an artefact ofthe time periods used for mapping. The mapped declinemay reflect the decline identified by Christensen (1978).
A widespread decline of many species, particularly inmore arid areas, was recorded in the 1880s (Shortridge1909). However, Shortridge (1909) also noted themammals of the south-west, to as far north as the MooreRiver, had not ‘disappeared in the same extraordinaryway’. Burbidge and McKenzie (1989) attributed themodern (post 1900) decline of Western Australia’s non-volant terrestrial, critical weight range (35 g–5.5 kg)mammalian fauna to environmental changes sinceEuropean settlement. They believed these changesreduced available productivity by diverting resources tohumans and introduced species and reduced vegetativecover through grazing and altered fire regimes. Criticalweight range mammals suffered the greatest declinesdue to their limited mobility and relatively high metabolicrequirements (Burbidge & McKenzie 1989). Theyfurther identified factors likely to ameliorate susceptibilityto decline and believed critical weight range mammalspecies with the ability to use, or with a requirement forrockpiles were afforded additional protection. Kinnearet al. (1988) had previously identified the predation
12 Greg Freebury: Operations Officer – Nature Conservation, WesternAustralian Department of Environment and Conservation, South CoastRegion, Albany
24 P.J. de Torres et al.
refuge value of rockpiles and subsequently (Kinnear etal. 1998) provided more empirical evidence for this.Burbidge and McKenzie (1989) believed species usingburrows, sheltering in hollow logs or to some extent,those with an arboreal habit, would also be affordedprotection from predation. We believe the quokka, witha requirement for a vegetation mosaic which includesareas difficult to penetrate, represents a further groupof critical weight range mammal fauna benefiting fromuse of a predation refuge, analogous to the rock wallabyuse of rockpiles. However, we caution that althoughrefuges may provide temporal respite from predation,long-term persistence of predation may still lead to localand regional extinctions, despite the presence of refuges.
Burbidge and McKenzie (1989) claimed the arid zonefauna were most at risk and fauna from the mesic zonewere somewhat buffered from the changes which resultedin reduced resource availability. The more mesic south-west Western Australian environment may have bufferedthe fauna from this initial decline. Recent researchsuggests fauna from mesic areas are no longer secure(Woinarski et al. 2001). Mesic areas may have moreavailable environmental productivity and may toleratelonger periods of the various disturbances responsiblefor the decline observed in arid areas. However,continued depletion of environmental resources and areduction in refugia may be contributing to a moremodern decline. The decline of the quokka is consistentwith this pattern, with the distribution contracting fromthe more arid northern and west coast areas, whilepopulations from the more mesic southern forest andsouth coast areas have persisted (Figs 3 and 4).
Predation by the dingo, European red fox andferal cat and interaction with other species
Three eutherian predators, the dingo, Canis lupus dingo,the European red fox, Vulpes vulpes, and the cat, Feliscatus, have been introduced to Australia since the arrivalof humans. These eutherian predators acted on apredator-naïve marsupial prey which had not interactedwith medium-sized cursorial predators for the preceding20 000 to 30 000 years (Johnson et al. 1989).
The Dingo
The dingo was introduced to Australia by Asian seafarersapproximately 3 500 to 4 000 years ago (Corbett 1995b).Where present, the primary prey of the dingo is usuallymedium sized mammals (Corbett 1995a) and largemacropods (Corbett & Newsome 1987). However,dingoes will prey upon a wide range of species (Corbett& Newsome 1987; Newsome et al. 1983; Robertshaw& Harden 1985; Thomson 1992; Whitehouse 1977)and the preferred prey in any particular area appears tobe determined not only by prey abundance, but by preyavailability (Corbett 1995a; Corbett & Newsome 1987;Vernes et al. 2001). In north-west Western Australia,Thomson (1992) found large prey items, mostlykangaroos, featured prominently, with smaller prey items
taken less often. Taking smaller prey was associated witha breakdown of dingo packs and an increase in thenumber of solitary dingoes (Thomson 1992).
In south-west Western Australia, the quokka wascommon prior to the 1930s and may have been anoccasional prey item for the dingo, although the quokka’spreference for densely-vegetated habitat may have offeredrefuge from predation. However, the established presenceof the dingo within the region prior to the modern declineof the quokka, implies dingo predation was not the majorcause of that decline. Our own observations indicatedingoes are now exceedingly rare in the jarrah forest andare thus unlikely to be generating predation pressure ofany consequence on quokka populations. Historically, thedingo is believed to have had little major impact on themajority of Australian fauna with the exception of theThylacine, Thylacinus cynocephalus, and the Tasmaniandevil, Sarcophilus harrisii, (see Corbett 1995b; Jones 1995;Rounsevell & Mooney 1995).
The impact of the dingo on Australian native faunamay have been minimised through competitive inhibitionby the sympatric mesopredator, humans. Similarly,dingoes may exhibit an indirect effect as a mesopredator(Newsome 2001). Dingoes are thought to prey on theEuropean red fox and the feral cat where they occursympatrically in the Nullarbor region of Western Australia(Marsack & Campbell 1990), however Marsack andCampbell (1990) implied the evidence for this wascircumstantial. Thomson (1992), as part of an extensivestudy in north-west Western Australia, found the feralcat was present as a prey item only once, and the foxwas not present at all in the stomach contents from asample of 95 dingoes. Newsome et al. (1983) recordedthe fox and the cat present in 1.6 and 0.3 percentrespectively from the stomach contents of 530 dingoesin a trapping and dietary analysis study in south-easternAustralia. The lack of any other comprehensive datasupporting the assertion of intraguild predation suggestsdingoes are more likely to competitively inhibit foxesand cats than prey directly on them.
The European red fox
The red fox was introduced to eastern Australia in thelate 1860s to 1870s (Rolls 1969; Troughton 1965) andhad spread to the south-west of Western Australia bythe early 1930s (Gooding 1955; Jarman 1986; King &Smith 1985; Long 1988). The arrival of the fox in south-west Western Australia coincides with the reporteddecline of the quokka (White 1952) and numerous smallto medium size terrestrial native mammals from mainlandAustralia (Burbidge & McKenzie 1989; Daubney et al.undated; Friend 1990; Jenkins 1974; Richards & Short1996). Numerous authors have suggested the fox wasresponsible for this fauna decline, see for example thereview by Abbott (2001b). Although it is only relativelyrecently that the fox has been empirically linked to thesuppression of native mammal populations (see Kinnearet al. 1988; Kinnear et al. 1998), it seems very likelyfoxes were responsible for the initial decline of the quokka
Distribution of the quokka 25
on the mainland and have contributed to its continueddecline. The abundance of quokkas on Rottnest Islandin the absence of foxes (Dunnet 1963; Holsworth 1964;Holsworth 1967; Kitchener 1970; Shield 1959) isconsistent with this hypothesis. Further evidence of theeffect of foxes on quokkas comes from a series ofreintroductions, and ultimate failure, of nearly 700Rottnest Island quokkas between 1972 and 1983 to afenced, but not fox-proof, enclosure at Jandakot, nearPerth (Short et al. 1992) (Table 8). By 1988 only ninequokkas remained in the 254 ha enclosure. The lowsurvivorship was attributed to predation by foxes andferal cats (Short et al. 1992).
The fox is now generally accepted as a significantpredator of medium size terrestrial native Australianmammals. Recognition has been at a national and statelevel, as demonstrated by the Commonwealth ofAustralia’s Threat Abatement Plan for Predation by theEuropean Red Fox (Biodiversity Group EnvironmentAustralia 1999) and two recently published state plansfor control of the red fox – the Victorian PestManagement. A framework for Action. Fox ManagementStrategy (DNRE 2002) and the New South Wales ThreatAbatement Plan for Predation by the Red Fox (Vulpesvulpes) (NSW NPWS 2001). Fauna managementprograms in Western Australia have recognised theimportance of fox control and the Western AustralianDepartment of Environment and Conservation (DEC)has implemented a broad scale fox control and faunarecovery program, Western Shield. The program isfocused on recovery of threatened fauna througheffective fox control and translocation of threatened faunaspecies in the presence of fox control.
The predecessor to Western Shield was OperationFoxglove, a large scale experimental 1080 baitingprogram within the northern jarrah forest of south-westWestern Australia (de Tores 1999). The areas baitedwithin the 550 000 ha study area of Operation Foxglove,combined with local baiting at the Gervasse site, includeor abutted all but one of the known extant occurrencesof the quokka in the northern jarrah forest. The Foxglovebaiting program commenced in 1994 and, despite thefact that quokka populations have not shown an increasein abundance since this baiting program commenced(Hayward et al. 2003), we hypothesise the quokka haspersisted in the northern jarrah forest as a result of thisbaiting and further believe effective long termconservation of the quokka requires implementing andmaintaining fox control programs (see managementrecommendations).
The feral cat
Aborigines from the desert regions of central Australiaand central Western Australia believe cats to have eitheralways been present or to have arrived ‘from the west’(Burbidge & McKenzie 1989). Although far fromclaiming the assertion as definitive, Gaynor (2000)believed this arrival from the west may have been a resultof cats escaping from Dutch shipwrecks off the west
coast of Australia. Dickman (1996) noted cats couldhave been introduced to north-western Australia as earlyas the 16th Century. However, Abbott (2002), from adetailed search of historical records, concluded cats werenot present on mainland Australia prior to Europeansettlement. Abbott (2002) presented a compellingargument and conceptual model to support his beliefthat cats spread from multiple points of introduction atcoastal locations in the period 1824 to 1886. Abbott(2002) and others before him (see Morton 1990) alsoargued there is insufficient evidence to support theassertion that cats were responsible for any mammalextinctions from arid Australia. However, there is generalconsensus in the literature (see Dickman 1996) tosuggest the cat has contributed to the decline of manyspecies. Although there is some empirical evidencecorrelating the presence of cats with extinctions of criticalweight range mammals from islands (Burbidge & Manly2002) and empirical evidence indicating predation byferal cats led to reduced population sizes of small nativevertebrates (Risbey et al. 2000), we caution againstinferring a causal relationship between cat predation andfauna extinctions on the mainland generally, and betweencat predation and quokka decline specifically.
Abbott (2002), citing Catling , Coman (1991) andDickman (1996) noted the optimal prey size for catswas approximately 200g. The weight of adult quokkasranges from 1.6 to 3.5kg for females and 2.7 to 4.2 kgfor males (Hayward et al. 2003; Kitchener 1995), whichsuggests if cats prey on quokkas, predation may berestricted to juveniles, or will favour predation ofjuveniles. Circumstantial evidence suggests feral cats preyon the young of bridled nailtail wallabies, Onychogaleafraenata, (Horsup & Evans 1993) and the brush-tailedrock-wallaby, Petrogale penicillata, (Short et al. 1992).More convincing evidence was presented implicating thecat as a significant predator of juvenile and adult alliedrock-wallabies, Petrogale assimilis, (Spencer 1991). Allthree species are larger than the quokka (Eldridge &Close 1995; Evans & Gordon 1995; Prince 1995),suggesting cats may well prey on juvenile and/or adultquokkas.
Cats have also been identified as predators of severalspecies of Australian mammals with adult body weightsgreater than 200 g. For example, cats were responsiblefor between 25 and 32% of predation events of adultwoylies, Bettongia penicillata, at translocation release sitesin the northern jarrah forest (de Tores 1999) andresponsible for predation of adult brushtail possums,Trichosurus vulpecula, in the same study (de Tores,Himbeck, MacArthur, Maxwell, White and Rosier,unpublished radio telemetry data). Cats have also beenreported as preying on adult rufous hare-wallabies,Lagorchestes hirsutus, (Gibson et al. 1994), the burrowingbettong, Bettongia lesueur, (Christensen & Burrows1994), the numbat, Myrmecobius fasciatus, (Friend &Thomas 1994) and the western ringtail possum (de Tores2005). Short et al. (1992) concluded cat predationcontributed to the failure of a translocation of the bandedhare-wallaby, Lagostrophus fasciatus. The above suggests
26 P.J. de Torres et al.
cats prey on a suite of native mammal fauna, inclusive ofspecies larger than 200g and have the potential to preyon quokkas. Short et al. (1992) further concluded directpredation by foxes and cats was a more plausibleexplanation, than overgrazing by rabbits and othermacropods, for the failure of the quokka translocationsto Jandakot in the 1970s.
However, the failure of the Jandakot translocationsand the suspicion that cat predation contributed to thosefailures are insufficient to conclude cat predation wasresponsible for the initial and continued decline of quokkapopulations on the mainland. Equally confounding is theco-existence of cats and quokkas on Rottnest Island withno discernable reduction in quokka abundance. Similarlyconfounding is the presence of cats in Tasmania with nodiscernable reduction in quokka-size prey species.
Molsher (1999) concluded, from an experimentalstudy at Lake Burrendong, New South Wales,interspecific competition between foxes and cats was themost likely mechanism limiting feral cats. Interspecificcompetition between cats and foxes was also thought tobe a possible explanation for the observed increase incat numbers when fox density was experimentally reducedat Heirisson Prong, north-west Western Australia (Risbeyet al. 2000). Such a response (mesopredator release) isa potential outcome of DEC’s large-scale fox controland fauna recovery program, Western Shield. However,the lack of records of cat predation from Rottnest Islandand the absence of evidence to support the assertionthat the cat was responsible for other mammal extinctions(Abbott 2002), suggests cats are unlikely to be solely orprimarily responsible for the widespread decline of thequokka.
Other predators and interactions
Other predators are unlikely to have been responsible forthe decline in quokka populations. Nocturnal birds of preyare reported to have been responsible for fossil depositsfrom several cave sites (see Table 1 – owl accumulatedfossil deposits). Two owl species from the south-west ofWestern Australia, the barking owl, Ninox connivens, andthe masked owl, Tyto novaehollandiae, have been recordedtaking prey as large as young rabbits (Schodde &Tidemann 1982), so could presumably prey on juvenileor sub-adult quokkas. Although Johnstone (cited as apers. comm. by Abbott 1999) regarded the barking owlas a species favouring swamps and edges of rivers, it isnot a forest species (Abbott 1999) and was not recordedat any of the seventy forest sites surveyed by Liddelow etal. (2002). The masked owl is infrequently recorded inforest (Abbott 1999) and is more common in woodlandor at the interface of agricultural land and forest (Liddelowet al. 2002). Therefore, neither species is likely to havebeen responsible for the decline of the quokka, particularlywithin forest areas. The presence of quokka bonesrecovered from a wedge-tail eagle’s eyrie on Bald Island(Storr 1965) indicates the quokka may be an occasionalprey item of the wedge-tail eagle. However, we cautionagainst inferring presence of a prey species in the diet of a
predator equates with a predation induced decline of thedetected prey species. The quokka’s diurnal use of denselyvegetated swamps within forest areas on the mainland(Christensen et al. 1985; Storr 1964b; White 1952), useof Spreading Sword-sedge, Lepidosperma effusum, andAnarthria scabra habitat within other forest areas anduse of heath habitat on the south coast further minimisesthe risk of predation by eagles.
Large predatory snakes, e.g. the carpet python,Morelia spilota, and to a lesser extent goannas, Varanusspp., are able to prey upon species such as the southernbrown bandicoot, Isoodon obesulus, the common brushtailpossum, Trichosurus vulpecula , the brush-tailed bettong,Bettongia penicillata, (de Tores, Himbeck, MacArthur,Maxwell, White and Rosier, unpublished radio telemetrydata), the western ringtail possum, Pseudocheirusoccidentalis, (de Tores 2005) and the tammar wallaby,Macropus eugenii, (David Pearson, pers. comm. toMWH)13 and may also pose a predation threat to sub-adult quokkas. However, these predators have co-existedwith the quokka in the south-west of Western Australiathroughout the quokkas existence and there is noevidence to suggest they initiated the decline in quokkaabundance. The co-existence of high density populationsof the carpet python and the tammar wallaby on GardenIsland is consistent with this hypothesis.
The feral pig, Sus scrofa, an introduced opportunisticomnivore, is present in forest areas of south-west WesternAustralia and most climatic regions in Australia (Pavlov1995). In addition to the commercial health risks posed,where investigated, pigs have been found to pose asignificant environmental and management problem(Pavlov et al. 1992). Ecological damage associated withpig activity (rooting) in wet tropical forests in Queenslandwas shown to vary with forest type, with wet sclerophyllforest sites showing the greatest disturbance (Laurance& Harrington 1997). Our observations (MJD inparticular) have confirmed the presence of pigs atnumerous quokka sites and, although not supported byany quantitative data, there appears to be a trend ofincreasing occurrence of pig activity and pig abundanceat quokka sites in the south-west forests (MJD, personalobservations; Graeme Liddelow, pers. comm. to PJdeT).Non-target captures of quokkas at sites trapped for pigcontrol purposes (see records by Liddington and Stainesin Table 4 for the years 2001 and 2002) may indicate anoverlap of quokka and pig preferred habitat within thesouth-west forests. The relatively sedentary nature ofpigs (Caley 1997; Saunders & Kay 1996) and thepotential for pigs to significantly disturb quokka habitatby creating large openings and easier access for foxes,may pose an additional threat to the conservation statusof the quokka.
Climatic influencesThe mainland quokka populations appear to have beenhistorically limited to areas of the south-west of Western
13 David Pearson: Principal Research Scientist, Western AustralianDepartment of Environment and Conservation, Science Division, Woodvale
Distribution of the quokka 27
Australia with an annual average rainfall in excess of700 mm. The current distribution of the quokka closelyfollows the limit of the 1 000 mm rainfall isohyet and,for all but the eastern extent of the distribution near theStirling Range National Park, appears to be limited toareas with an annual average rainfall of 700 mm. or more(Figs 3 and 4). This may reflect the quokka’s relativelyhigh water requirements (Main & Yadav 1971).
Populations in the vicinity of Stirling Range and GreenRange appear to be isolated from all other south coastpopulations and also appear to be within a lower rainfallzone (Figs 3 and 4). The locations of rainfall recordingstations provided by the Australian Bureau ofMeteorology (BoM) indicate stations are located to thenorth and south of the Stirling Range, with no stationsin the immediate vicinity of the known extant quokkapopulations. The Stirling Range may be in a pocket ofhigher, orographically generated rainfall, not detectedby the BoM rainfall recording stations and not reflectedin the pattern of rainfall isohyets. This hypothesis issupported by the presence of an isolated population ofthe red flowering gum, Corymbia ficifolia, (Brown et al.1998) and the occurrence several wet region bird species(Ian Abbott, pers. comm. to PJdeT) in the StirlingRange. The Corymbia ficifolia population is 100 kmnorth of its main distribution and the species is thoughtto be a relic from a past wetter climate (Brown et al.1998). Further evidence of this orographic effect wasprovided by Courtney (1993) who prepared a rainfallisohyet map depicting the highest peaks of the StirlingRange as receiving more than 700mm of rainfall annually.The Stirling Range quokka population(s) may have beentraditionally isolated or may represent a remnant of aonce contiguous population. In either case, the statusand security of these isolated populations and the isolatedoccurrence at Green Range warrant further investigation.
The Rottnest Island population suffers seasonalmortality over summer and this has been attributedindirectly to dehydration (Barker 1961; Holsworth 1964;Main et al. 1959; Packer 1968; Storr 1964b).Nonetheless, quokkas persisted in the fossil record fromsouth-west Western Australia throughout periods ofsignificant climatic variation from warm and wet to glacialaridity. Cook (1960) suggested climate change was thecause of the modern decline of the quokka. There seemsto be insufficient evidence to attribute climate changealone as the major cause of the decline of the quokkasince the 1930s. However, Balme et al. noted thepersistence of the quokka in the fossil record throughperiods of aridity may have been because the south-westcorner of Australia did not experience the extremes inaridity experienced by other parts of southern Australia.Therefore, the consequences of increasing aridityassociated with current patterns of climate change shouldbe seen as a potential threat to the continued persistenceof the quokka.
European colonisation and development
Reports from naturalists in the early 1900s indicated the
quokka was common prior to and until the 1930s (Gould1863; Shortridge 1909; White 1952) and, with very fewexceptions, most reports suggest quokkas were restrictedto specific habitats (see Holsworth 1967; Storr 1964b).Much of the coastal heath and shrubland habitats in thesouth-west of Western Australia where quokkas onceoccurred (Shortridge 1909; White 1952) have beencleared for urban development. The coastal plain, fromnorth of Perth to Busselton (Fig. 1) has few remainingpockets of undisturbed vegetation and the remainingfragments are small and easily invaded by introducedpredators. The areas known to have historicallysupported quokka populations from the Swan CoastalPlain were swamps or low lying seasonally waterloggedareas. These coastal plain swamps have been progressivelydrained, with very few retaining their original drainagepatterns. Similarly, the upper reaches of many creeksystems in the forested areas of the Darling Range havebeen cleared for agriculture, dammed to supply waterto Perth or split from connecting habitat by roads tosuch an extent that the remaining quokka habitat is highlyfragmented and in places may be too small to supportviable populations.
Other factors associated with European colonisationmay have similarly affected the quokka. Calver and Dell(1998b) believed all mammal species from the south-west forests of Western Australia historically had a widerdistribution and suggested the extant mammal fauna maybe resilient to changes in forest structure. However, theycautioned this resilience may not be the case andemphasised the importance of experimentallydemonstrating that there are no direct and indirect linksbetween forestry practices and declines in distributionand abundance on the suite of resident native fauna fromthe south-west of Western Australia. Increases inpredation rates (Wayne et al. 2000), increases in thearea of edge affected habitat (Wilson & Friend 1999),increases in interpatch distances (Hayward et al. 2003)and increases in roading disturbance (Calver & Dell1998a) may be indirect effects from management of thejarrah forest. The effect of these activities on quokkapopulations has not been quantified. Roading and otherdisturbances associated with mining and harvesting,resulting in removal of habitat and alteration of drainagepatterns, also have the potential to contribute to increasesin interpatch distances. Construction of logging accessroads in 2001–2002 in Nairn Forest Block in the southernforest region of south-west Western Australia has beenimplicated as an example of the detrimental effects fromharvesting and the associated roading activities. Postcommencement of roading there was an observedincrease in the number of reported quokka roadkills. Aminimum of 10 quokka roadkills was reported by a singleobserver within a six month period in close proximityto, and coinciding with roading activity at Nairn ForestBlock (John Austin, pers. comm. to PJdeT)14 . However,there are alternative and equally plausible explanations
14 John Austin: Long-term local resident, Northcliffe, Western Australia
28 P.J. de Torres et al.
for this increase in observed roadkills. Quokkas mayhave been attracted to areas of new growth resultingfrom recent burns of roadside verges. This explanationis also consistent with the quokkas’ preference for avegetation mosaic which includes recently burnt areas(de Tores et al. 2004). Nonetheless, this uncertaintyhighlights the need to identity the cause of the largenumber of roadkills and to quantify the direct and indirecteffects from anthropogenic disturbance.
Fire
The Noongar people have occupied the south-westcorner of Australia for at least the last 40 000 years(Balme et al. 1978; Merrilees et al. 1973) and possiblylonger (Turney et al. 2001). In the period prior toEuropean settlement, the Noongar people of the south-west are thought to have burnt the jarrah forest, withlow intensity fires, with a minimum fire interval of twoto three years in the moist to high rainfall areas, and twoto five years in the forest areas with lower annual rainfalland lower fuel accumulation rates (Burrows et al. 1995).Burrows et al. (1995) concluded this pattern of burningwould have led to a vegetation mosaic which would haveincluded patches of unburnt forest, with most patches‘being less than 6 years since last fire’ (see also Wilson &Friend 1999). Riparian environments may haveexperienced longer intervals between fires (Abbott 2000;Burrows & Friend 1998).
Ward et al. (2001) (see also Ward & Sneeuwjagt1999) believed examining fire scars on grass trees,Xanthorrhoea spp., provided a more sensitive techniquefor determining fire history. From examination of grasstrees, Ward et al. (2001) believed Noongar burning, orthe pre-European frequency of burning in the jarrahforest, was once every three to four years. Lamont et al.(2003) cautioned that the susceptibility of grass trees tofire, and therefore their ability to reflect fire history,depends on their location in the landscape. Burrows andWardell-Johnson (2003) further cautioned againstinterpreting historic patterns of fire on a regional scaleon the basis of patterns observed among individual grasstrees. The findings of Ward et al. (2001) suggested ashorter interval between fires than concluded by Burrowset al. (1995). However, both studies indicated thefrequency of burning in forest areas of south-westWestern Australia has changed from the frequency usedby the Noongar people. Ward and Sneeuwjagt (1999)further implied these changes also apply to coastal areas.Abbott (2003) suggested the interval between fires setby the Noongar people in coastal areas may have beenas short as two to four years.
Alterations to fire regimes following Europeansettlement have been implicated with declines in faunaabundance and range (see review by Wilson & Friend1999). Burrows et al. (1995) believed three distinct fireregime periods or eras could be defined for the forestsof south-west Western Australia post Europeansettlement, namely (i) the first European era (1855–1920) where there was a cessation of Noongar burning
practices which led to longer intervals between fires andmore intense fires; (ii) the second European era (1920–1965) when forestry practices adopted a policy of fireexclusion. Burning was restricted to strategic buffers andfire suppression purposes. Much of the forest remainedunburnt and large high intensity wildfires occurred,fuelled by logging debris and naturally occurring highfuel loads; (iii) the third European era (1965–1995)where fuel reduction burning was based on a six to tenyear rotation. Although fire frequency for this periodwas comparable to the pre-European regimes, fireintensity was higher, fuelled primarily by logging debris(Burrows et al. 1995). Towards the end of this thirdera, and from 1962 in particular, the area burnt byprescribed fire greatly increased (Lachie McCaw15 pers.comm. to PJdeT and unpublished data). This was areflection of the then Forest Department’s expandedprogram of broad-scale burning which commenced after1954 (McCaw et al. 2005; Wallace 1966).
Although this third ‘era’ encompasses the period 1980to 1992, i.e. the period within which our data indicate alarge contraction in the distribution of the quokka, nodirect causal relationship can be drawn, particularly asthe areas of greatest quokka decline are outside forestareas subject to this fire regime. Fire regimes in forestareas post 1995 have been similar to the third Europeanera described by Burrows et al. (1995). Discussions withDepartmental operational staff involved in prescriptionburning, combined with the reported reduction in theextent of the area burnt annually in the south-west ofWestern Australia, indicated a trend of progressivelydecreasing prescribed-burn fire frequency in the south-west. The 1998–99 prescribed burn program achievedless than 50% of the planned prescribed burning programand was the lowest achieved since 1961 (CALM 1999).The areas ‘prescription burnt’ in the south-west forestregions in 1999–2000, 2000–2001 and 2001–2002 were134 308, 87 866 and 74 739 ha. respectively (CALM2000; CALM 2001; CALM 2002). From 2002 on thistrend of decreasing area burnt annually has been reversed,with 144 835 ha burnt as part of prescribed burns in2002–2003 and 192 119 ha in 2003–2004 (CALM2003; CALM 2004). We caution that increasing the area‘prescription burnt’ alone does not equate with burningto provide the preferred habitat mosaic. Our personalobservations (PJdeT) suggest broad scale burning has atendency to use natural barriers as fire boundaries. TheTaxandria linearifolia creeklines often act as suchnatural barriers and the long term effect has been toencroach upon these barriers and progressively reducethe size and extent of the T. linearifolia creeklines. Thistype of encroachment does not equate with providingthe burn conducive to maintaining the preferred habitatof the quokka. However, effective use of prescribed burnsdoes provide the opportunity to use fire to create thequokka’s preferred habitat mosaic as described by
15 Lachie McCaw: Senior Research Scientist, Western Australian Departmentof Environment and Conservation, Science Division, Manjimup
Distribution of the quokka 29
Hayward (2002) and Hayward (unpublished, reportedin summary in de Tores et al. 2004).
The finding by Hayward (unpublished, reported insummary in de Tores et al. 2004) is only partly consistentwith findings from a small-scale study (Christensen &Kimber 1975) specifically examining the effect of fire onthe quokka in forest sites near Dwellingup in the 1970s.Christensen and Kimber (1975) concluded quokkasreturned to swamps to forage almost immediately aftera burn, showed an influx of new individuals (i.e.previously un-trapped animals) and became resident 18months after the fire. There appeared to be no residentquokka population where vegetative cover was entirelyremoved by the fire (Christensen & Kimber 1975). Anolder (unburnt for 15 years) site trapped in that studyled the authors to conclude quokkas desert sites unburntfor 15 years and have a preference for a spatial mosaicor patchy burn which provide areas of refuge and areasof foraging habitat (Christensen & Kimber 1975).Hayward (unpublished, reported in summary in de Toreset al. 2004) came to a different conclusion and suggestedthis 15 to 19 year post-fire component of the mosaic didnot represent the upper fuel age of the quokkas preferredmosaic. Hayward (unpublished, reported in summaryin de Tores et al. 2004) showed this component of themosaic was negatively correlated with quokka presence(avoided by quokkas) and the preferred mosaic includedan additional component which was long unburnt (seesection on habitat use). We therefore suggest a burningregime which will create and maintain the mosaicidentified by Hayward (unpublished, reported insummary in de Tores et al. 2004) is required in thenorthern jarrah forest. This would necessitate differentburning regimes from those currently used at quokkasites in the northern jarrah forest (see section onmanagement recommendations).
Small, scattered populations are also likely to besusceptible to stochastic events such as wildfires whichmay result in localised and more extensive extinctions.Kirke (1983) reported quokkas fleeing from wildfires atGreen Range, north of Albany. Similarly, in theNorthcliffe area, quokkas observed in large numbers inthe 1940s and earlier, and known to feed in paddocksaway from vegetated creeklines, were reported as lastseen in number at this location when fleeing from a firein the 1940s (Laurie Wilson, pers. comm. to PJdeT)16 .Numerous quokkas were also observed in an openpaddock after fleeing a wildfire in the Allen Road/HilltopRoad area in Walpole-Nornalup National Park, circa1987 (John Asher, pers. comm. to PJdeT)17 . Quokkasare now reported to have repopulated this area and arethought to be in large numbers (Greg Freebury, pers.comm. to PJdeT).
The only known record of occurrence from KarnetForest Block, west of Jarrahdale, is from quokkas
observed fleeing a fire in 1991 (unpublished records fromA.N. Start, Table 4). A high-intensity wildfire burnt alarge portion of the Stirling Range National Park inNovember 2000 (pers. obs. of MWH) and numerousquokka deaths were reported. A similar fire in the StirlingRange National Park in 1991 was thought to have beenequally damaging to quokkas. However, quokkas werereported to have repopulated burnt areas in subsequentyears (Sinclair 1999). A large number of quokka deathswas also recorded as a result of a wildfire in the NuytsWilderness area, near Walpole in 2001 (Middleton 2001).Quokkas have subsequently been detected in unburntpatches within the boundary of the Nuyts fire and quokkapresence has been inferred at three locations immediatelyoutside the burn boundary. Presence was inferred bydetection of scats within typical quokka runways in densepatches of Spreading Sword-sedge, Lepidospermaeffusum, and Anarthria scabra (Greg Freebury, pers.comm. to PJdeT). With the exception of the programestablished to monitor recovery from the Nuyts fire, theextent of documenting any recovery from these stochasticevents has been largely anecdotal.
The quokka appears to be capable of persisting in afire-prone environment and the absence of low intensityfire from many sites may be a contributing factor to thecollapse of the northern jarrah forest metapopulation asdescribed by Hayward et al. (2003). However, there issufficient evidence from the northern jarrah forest tosuggest quokka populations there are dependent on thepresence of a structural mosaic which incorporates areasburnt within the previous nine years, but also has a largeoverall fuel age (i.e. must also include areas longunburnt) (Hayward, unpublished, reported in summaryin de Tores et al. 2004).
Further support for the requirement of a mosaic,and not simply a requirement for the presence of firewithin the past nine years, is provided by the results fromsurveys in 1995–1996 (Dillon 1996). From 28 northernand southern forest locations previously known to supportquokka populations, nine no longer supported quokkasand seven of these nine sites had been burnt within theprevious 10 years (Table 10) (Dillon 1996). Althoughthese areas provide the component of the mosaic whichhas been burnt within the last 9 to 10 years, they may nolonger support areas of long unburnt habitat.
A fire regime of low intensity burns and with afrequency comparable to the third European era ofBurrows et al. (1995) may be appropriate to generatethe vegetation mosaic preferred by the quokka. However,no single fire regime will benefit all taxa. Of particularconcern is the threat to the Noisy Scrub-bird, Atrichornisclamosus. The noisy scrub-bird has a preferred habitatof densely vegetated creekline and gully vegetation(Abbott 1999; Burbidge 2003). A too frequent burningregime of creeklines where it and the quokka aresympatric may compromise noisy scrub-bird habitat and‘even a mild fire can render an area unsuitable for manyyears’ (Burbidge 2003). Although the only knownlocations where extant populations of the noisy scrub-bird and quokka occur sympatrically are at Two Peoples
16 Laurie Wilson: Long-term local resident, Northcliffe, Western Australia17 John Asher: Environmental Officer, Western Australian Department of
Environment and Conservation, Nature Conservation Division, Bunbury
30 P.J. de Torres et al.
Bay, on the south coast, the noisy scrub-bird has recentlybeen translocated to sites within the northern jarrah forestwhere quokkas may also occur. Therefore, there is thepotential for conflicting fire management requirementsat these translocation release sites.
There are also indirect detrimental effects from fire.Christensen (1980a) recorded increases in the predationrate on the brush-tailed bettong following fire. This typeof indirect effect may also result in localised increasedpredation risk for other mammal populations. Kinnearet al. (Kinnear et al. 1988) suggested this type of indirectthreat may increase the risk of localised extinctions.
Hunting by Aboriginal people
Gardner (1957), citing John Lort Stokes, referred toobservations of Aboriginal people burning the bush forthe purpose of catching snakes, lizards and wallabies.These prey were speared as they fled the burningvegetation. Gould (1863) and Evans (undated) weremore specific and noted quokkas were eaten byAboriginal people. Evans (undated) noted the Aboriginalpeople from the Northcliffe area supplemented their dietwith quokka, kangaroo and marron. Gould (1863) notedquokkas were killed in great numbers at the end of theseason by Aboriginal people. Gould’s (1863) descriptionof Aboriginal people burning the bush to flush out theirprey is consistent with Gardner’s (1957) description. Theend of the season referred to by Gould (1863) is nodoubt a reference to the end of summer, as Meagher(1974), citing numerous historical records, noted towardsthe end of summer, Aboriginal people set fire to thebush to drive the wallabies from their retreats. Green(1989) referred to use of fire at Bald Head in January(mid summer) to ‘burn off the land for wallaby’ and theJanuary reference of ‘brought home … three wallaby fromBald Head where the Aborigines had fired the land to hunt’indicates burning for this purpose was not restricted tothe end of the summer season. At other times of theyear dogs were used to drive out prey (Meagher 1974).
Although Aboriginal people may have beenresponsible for exterminating some island populationsof macropods (Abbott 1980), there is no evidence tosuggest Aboriginal hunting of the quokka contributedto the species’ decline.
Disease
Reports from studies of colonies of captive quokkas showthe susceptibility of the species to disease (see Bradshaw1991). Anecdotal accounts in 1967 suggested a fatalherpes epidemic transmitted by a handler affected acaptive quokka colony (Burnet 1968). Salmonellainfections are common in the Rottnest Island and BaldIsland quokka populations but far less so on the mainland(Hart 1977; Hart et al. 1986). Hart (1981) foundcaptured mainland quokkas died within 24 hours ofexposure to bags in which Rottnest Island quokkas hadbeen held. These deaths were attributed to infectionsfrom Salmonella meunchen. Hart (1981) hypothesisedthese deaths may have been the result of cross infection
from use of the Rottenest Island bags, but conceded S.meunchen may have been present at an undetectablylow level in the mainland quokka population and becamelethal only when the animals were stressed throughtrapping.
Barker et al. (1957) recorded the presence of theparasite Ixodes australiensis in mainland populations ofthe quokka and several other parasites and diseases havebeen isolated from quokkas.
Mass mortality of quokkas was reported from theNorthcliffe area where swamps were reportedly ‘full ofquokka bodies’ in the early 1920s (George Gardner citedas a pers. comm. by How et al. 1987). How et al. (1987)attributed this to disease. Similarly, there are records ofan epidemic within the quokka population in the WarrenRiver area, near Manjimup in 1921 (Aldrich 1921; Lane-Poole 1921; Weston 1921). Unexplained quokka deathswere also reported from Crown reserves in the vicinityof Yallingup in 1933 (Aldrich 1933). Other references(Cook 1960; Perry 1973; Waring 1956; White 1952)also indicated mass deaths occurred as a result of diseasein the 1930s. Mass deaths have recently been attributedto surplus killing by foxes (Short et al. 2002).
The decline in arid zone mammals of the late 1800snoted by Shortridge (1909) and reported from theNullarbor Plain (Richards & Short 1996), has beensuggested to be a result of a ‘strange virus’ whichRichards and Short (1996) suggest, albeit based onanecdotal accounts, may have been the protozoanparasitic disease toxoplasmosis, passed on from feral cats.Toxoplasmosis was first recorded in the Rottnest Islandquokkas in 1961 (Gibb et al. 1966) and has beenreported in other marsupials including the eastern barredbandicoot, Perameles gunnii, (Lenghaus et al. 1990),the western ringtail possum, Pseudocheirus occidentalis,(de Tores 2005) and the chuditch, Dasyurus geoffroii,(Haigh et al. 1994). Berdoy et al. (2000) notedtoxoplasmosis may alter the behaviour of its intermediatehosts and increase its susceptibility to predation.
White (1952) believed fox predation, competition withrabbits, destruction of habitat through clearing andbushfires were supplementary to disease as the causalfactor for the quokkas’ virtual disappearance on themainland. Despite the references to quokkas dying fromdisease in the 1920s (Aldrich 1921; Lane-Poole 1921;Weston 1921) and the 1930s (Cook 1960; Perry 1973;Waring 1956; White 1952), the persistence of quokkason Rottnest Island and Bald Island during these periodssuggests disease was not the major contributor to theirdecline, or alternatively, if disease was responsible forthe decline on the mainland, it did not have the sameeffect on, or did not reach, Rottnest Island or Bald Island.It was during this period of decline on the mainland inthe 1930s when the quokka was no doubt at high densityon Rottnest Island, as it was in this period when it wasfirst referred to as a pest on Rottnest Island (Storr 1963).
Although White (1952) believed a decline caused bypredation, competition and habitat destruction wassupplementary to that caused by disease and Cook (1960)linked the decline of the quokka with disease, Recher
Distribution of the quokka 31
and Lim (1990) considered disease to be a contributing,not causal, factor to the decline of the Australia’s mammalfauna. Johnson et al. (1989) discounted disease as a factorassociated with the decline of critical weight rangemacropods altogether. Dickman (1992) noted there wasno evidence to indicate widespread disease as the causeof the decline of mammals within Australasia. We concurwith these authors and suggest there is insufficientevidence to conclude disease alone was the cause of thedecline of the quokka. However, we caution againsttrivializing the potential effects from disease. The 1826King George’s Sound record of Quoy and Giamard notedthe quokka specimen described ‘was recently dead whenwe found it, probably from disease’ (Alexander 1916). Ifdisease was the cause of death, it is reasonable tohypothesise disease was brought to King George’s Soundby Europeans. Although the Rottnest Island populationappears to be secure, its possible exposure to disease asa result of human contact should be seen as a potentialthreat. Introduction of a disease, or a single catastrophicevent, could result in a chance extinction of this insularpopulation.
Management recommendations
The northern jarrah forest
We recommend an active adaptive managementapproach for conservation management of the quokkawithin the northern jarrah forest. The active adaptivemanagement approach requires implementing an agreedexperimental approach (i.e. agreed to by allstakeholders), whereby a set of models and managementactions have been formulated and appropriatemonitoring protocols established. We believe monitoringshould be focused on examining the response of thequokka to a variety of management practices.
Quantified data has shown the northern jarrah forestpopulations are at low density. Despite the presence offox baiting, these populations have not responded andpredation is still potentially limiting population response.Based on the rapid rate of fox re-invasion of the northernjarrah forest post aerial baiting events (de Tores 1999),the lower probability of survivorship of the woylie,Bettongia penicillata, in treatments baited four times peryear compared with six times per year, and thesignificantly lower levels of survivorship in areas abuttingagricultural land (de Tores 1999), we recommend theactive adaptive management program incorporates anassessment of the effectiveness of an increased frequencyof 1080 baiting.
The northern jarrah forest populations are highlyfragmented, the populations are not mixing and thepreferred habitat within the Taxandria swamps is acomplex mosaic of recently burnt and long unburntareas. An additional requirement of the preferred mosaicis to have a minimal area burnt 15 to 19 years previously.The preferred spatial configuration of these seral stagesis not known – specifically, the relative proportion ofeach required seral stage is not known and the upper
limit of ‘time since last fire’ , i.e. the maximum period oftime without fire within the long unburnt component ofthe mosaic, is not known. Fire management practiceswhich selectively burn long unburnt components of themosaic may be detrimental to the long-term conservationof quokkas in the northern jarrah forest. This practicehas the potential to increase fragmentation of quokkahabitat by removing a component of the preferredmosaic. This practice should not be seen as a methodfor creation of quokka habitat, and, at best, constitutesa program more akin to a trial and error approach thanan adaptive management program.
Therefore, we recommend the active adaptivemanagement program should assign high priority tospatial analyses of existing, historic and potential quokkasites in the northern jarrah forest, with the objective ofstratifying the existing Taxandria mosaic (with stratabased on the number of years post fire) and determiningwhether any of these swamps can be better managedthrough the use of fire. Under this scenario, fire wouldbe used to create the preferred mosaic. We alsorecommend monitoring should incorporate geneticanalyses to test the northern jarrah forest metapopulationhypothesis. We further advocate any monitoring programassociated with quokka conservation needs to incorporatea component to enable quantitative assessment of pigdamage to quokka habitat. Although we advocate useof conventional trapping techniques to determine quokkaabundance, we caution against the overuse of invasivetrapping and strongly recommend use of alternativemethods where applicable. These methods includemolecular techniques (Alacs et al. 2003). We alsorecommend development of techniques to quantifyabundance based on the extent of activity in quokka-likerunways. Currently assessment of activity in runways canmeasure activity levels only and should not be extrapolatedto infer abundance. Further development of thetechnique adopted by Hayward et al. (2005) isrecommended.
The active adaptive management program shouldspecify quantifiable conservation outcomes. Suchmeasurable long-term conservation outcomes includedetermining:
• the number of known extant quokka sites where thepreferred structural mosaic has been establishedthrough the use of fire;
• the number of new sites where the preferred structuralmosaic has been created through the use of fire;
• the number of sites where quokka populations remainstable or show an increase in abundance (asmeasured through survival analyses and populationestimates, respectively);
• the number of sites currently thought to be supportingpotentially suitable habitat, where quokkapresence has not been detected/confirmed, andwhere, post habitat manipulation, quokkapresence is confirmed and a populationestablished;
32 P.J. de Torres et al.
• the number of sites where population mixing has beenconfirmed (i.e. sites where animals have beenknown to disperse to and where the source site isstill viable [the source sites can be identifiedthrough the use of molecular techniques, andpopulation viability can be determined byconventional monitoring [trapping] andpopulation modeling]).
The Swan Coastal Plain
High priority is recommended to unambiguouslydetermine quokka presence at Muddy Lake and the waterauthority reserve near Dunsborough. If presence isconfirmed, we recommend a monitoring program beimplemented to ensure any trend of increase or decreasein population can be detected. The requirement forintroduced predator control and habitat manipulationshould also be assessed. Minimal data were available onquokka presence elsewhere on the Swan Coastal Plain.We recommend a review of the former known locationsfrom the Swan Coastal Plain (historically these wereswamps), assessment of quokka presence at these sitesand examination of the effects from draining theseswamps. If appropriate, and where possible, theconservation value of re-instating former drainagepatterns should be examined.
The southern forest and south coast
The dearth of information on the size of each population/sub-population from the southern forest and south coastareas should be addressed. We recommend an initialapproach of confirming presence, through trapping, atsites where quokkas are thought to occur. We furtherrecommend undertaking spatial analysis of these extantpopulations and locations of known and potentiallysuitable habitat in conjunction with a survey andmonitoring program to assess population size. The extentof dispersal/immigration/emigration between habitatpatches should be quantified to determine whether thesub-populations constitute a functional metapopulation,discrete sub-populations or a panmictic population. Thisprocess would enable populations of high conservationvalue to be identified, where conservation value is assessedin terms of the population’s strategic value locally,regionally and globally as determined by its geographiclocation, demographics, genetic structure and importanceas a source population for re-stocking other populations/subpopulations.
As a matter of urgency, we recommend a morerigorous and strategic approach be implemented to assessthe potential effect from timber harvesting and associatedoperational activities. Assessment should identify theextent of quokka habitat to be modified, destroyed orretained by each proposed operation, the size of thepopulation(s) affected by the proposed operation, theconservation significance of the population, the potentialfor dispersal, numbers likely to disperse and dispersalpatterns, availability of suitable habitat within dispersaldistances, population size within areas of suitable habitat
within dispersal distance and the potential effect on thesepopulations.
We recommend investigation of the population(s)from the Stirling Range to assess the security of thesepopulations. Survey is recommended to assesspopulation size, habitat used and the security of thishabitat from stochastic events, in particular from wildfire.We also recommend examination of the genetic structureto determine whether the Stirling Range and GreenRange population(s) was, or is still, contiguous with othersouth coast populations.
CONCLUSION
The distribution of the quokka appears to have beentraditionally limited by climate and by rainfall in particular.Reduction in availability of suitable habitat, in conjunctionwith predation and changed fire regimes, appear to havelimited this distribution, or more specifically has furtherconfined the quokka to specific habitats within the limitsof its geographic range.
For the quokka, like numerous other native mammalspecies, the arrival of Europeans to Australia coincidedwith a slow but continual decline in abundance and range(Fig. 3). From 1900 in particular, the increasing humanpopulation in the south-west of Western Australia resultedin anthropogenic disturbances including vegetationclearance, logging, mining, hunting and changed fireregimes. Introduced predators have been implicated in40% of historic extinctions (Caughley & Gunn 1996)and it seems likely that predation pressure from theintroduced red fox, in conjunction with continued habitatalteration through fire exclusion and colonisation, furthercompromised the conservation status of the quokka.
However, the quokka, like many Australian nativemammals (Wilson & Friend 1999) appears to be resilientto individual disturbance factors but also appears to beincreasingly susceptible to the cumulative effect ofmultiple factors. None of these factors has occurred inisolation. Many commenced more or less in synchronywith the arrival of the fox (Recher & Lim 1990) and allcontinue to operate.
Although there is considerable agreement thatmultiple factors combined to contribute to the decline,there is also considerable disagreement as to the ultimatecause of the decline of critical weight range mammalssince European arrival (see Burbidge & McKenzie 1989;Lunney et al. 2001; Morton 1990; Recher & Lim 1990;Short et al. 2002; Short & Calaby 2001; Smith & Quinn1996; Wilson & Friend 1999). Abbott (2001b) concludedthe fox was the primary agent responsible for the declineof the bilby. However, a single explanatory hypothesisto account for past and continuing declines is unlikely tobe applicable to all species and it would seem logical thateach species is affected to a differing degree by eachfactor. Similarly, the relative importance of any factorwill vary spatially for each species. Kinnear et al. (2002)proposed two hypotheses, niche loss/damage andpredation, to account for the initial and continued decline
Distribution of the quokka 33
of Australia’s critical weight range mammal fauna. Thehypotheses are not necessarily mutually exclusive (Kinnearet al. 2002). There is a wealth of evidence linking twofactors to the decline of the quokka; predation and aloss of the quokkas’ preferred habitat mosaic. The latterequates with the niche loss/damage hypothesis proposedby Kinnear et al. (2002). We believe the two hypotheses,niche loss/damage and predation, when acting inconcert, are sufficient to account for the pattern of declineof the quokka.
Islands, although not immune from disturbancefactors, often act as refuges for threatened species largelybecause the factors responsible for the decline on themainland are often absent from islands (Dickman 1992).The quokka population on Rottnest Island may provideclues to the significance of the various disturbance factorswhich led to its decline on the mainland. Compared tothe mainland, the Rottnest Island population has a highersusceptibility to disease (Salmonella) (Hart et al. 1986),encounters seasonal aridity leading to summer mortality(Hodgkin & Sheard 1959; Shield 1964) and a period ofanoestrus (Shield 1964). Rottnest Island and mainlandsouth-west Western Australia have experienced the effectsof development and have been subject to extensive habitatfragmentation. The success of the quokka populationon Rottnest Island, despite these disturbances, suggeststhe species is quite resilient. With respect to disturbancefactors, the most significant difference between RottnestIsland and the south-west mainland appears to be thepresence of the fox on the mainland. Recent interestingadvances in our understanding of the effect of the foxon native fauna reinforce this conclusion. The predationefficiency of foxes combined with the inadequate anti-predator defences of a predator-naïve fauna may lead to‘high and unsustainable’ levels of predation (Short et al.2002). Short et al. (2002) concluded this is the case forthe Australian mammal fauna which has evolved in relativeisolation since the break-up of Gondwana 50-60 millionyears BP (Heatwole 1987). Short et al. (2002) furtherconcluded surplus killing may be an outward sign of thismismatch and provided case studies to support theirassertion that the declines of many prey species, and inparticular those species with limited refugia, may be aresult of surplus killing by foxes.
The hypothesis that species with little or no refugewould have suffered most from this predation (Short etal. 2002) is consistent with the pattern of decline of criticalweight range mammals in the arid zone described byBurbidge and McKenzie (1989). The quokka is nowrestricted to areas of refuge (islands and dense vegetationon the mainland). The presence of this latter refuge seemslikely to have been the sole factor staving off extinctionresulting from fox predation and the combined effectsof altered fire regimes and habitat loss.
With the continued depletion of resources (nutrients,water, refuge), which have previously buffered the faunadecline in the more mesic areas of the mainland (Recher& Lim 1990; Woinarski et al. 2001), the bufferingcapability may be in the process of being compromised.Further extinctions may result (Recher & Lim 1990;
Woinarski et al. 2001) and the assumption that criticalweight range mammals in mesic areas are secure(Burbidge & McKenzie 1989) may no longer be valid.
ACKNOWLEDGEMENTS
Staf f from the Western Australia Department ofEnvironment and Conservation (DEC, formerly theDepartment of Conservation and Land Management,CALM) provided numerous records of the occurrenceof the quokka, and in particular we thank Sandra Gilfillan,Peter Orell and Peter Mawson for supplying databaserecords and Peter Collins, Greg Freebury, Ian Wilsonand Graeme Liddelow for additional information onsouthern forest and south coast populations. Thanks toRalph Staines for providing additional trapping andlocation records, Antoinette Tomkinson, Rob Buehrig,Kathy Himbeck, Cathy Lambert, Kaylene Parker andSuzanne Rosier for providing field assistance at varioustrapping sites and to John Liddington, from the WaterCorporation of Western Australia for providing additionaltrapping and location records. Thanks especially to LizJef ferys for collating quokka records from theuncatalogued records of the Palaeontology Departmentof the Museum of Western Australia, Norah Cooper forclarifying WAM records, Tony Start for providing manyunpublished records, Sandy Ingleby for providing accessto the database records of the Australian Museum, ClaireWright from the Rottnest Island Authority for providingadditional references and Tony Start and Jim Lane foralerting the authors to the issues relevant to quokkahabitat on the Swan Coastal Plain. Special thanks toMark Sheehan, Northcliffe, for his commitment toensuring all quokka records were forwarded to theauthors. We thank Robyn Wilson, Shane French andDavid Robertson from the Western AustralianDepartment of Environment and Conservation,Information Management Branch for GIS support. Inparticular we thank Lisa Wright, Wildlife ResearchLibrarian, DEC, for the tracking down requests forobscure references and Ian Abbott, Neil Burrows, R.I.T.(Bob) Prince, Lachie McCaw, Tony Start, AndrewBurbidge and Suzanne Rosier for advice on preliminarydrafts. We also thank Per Christensen for his constructivecriticism as a formal referee and for his valuable inputand support generally. We thank Don Bradshaw andDarrell Kitchener for comment as referees of a previousdraft and two anonymous referees for valuable commenton an earlier draft which encompassed a conservationstatus review. The work was funded by the WesternAustralian Department of Conservation and LandManagement and Alcoa World Alumina, Australia.
REFERENCES
Abbott I (1980). Aboriginal man as an exterminator ofkangaroo and wallaby populations on islands roundAustralia. Oecologia 44:347–354.
34 P.J. de Torres et al.
Abbott I (1998). Avifauna of the Irwin Inlet-Broke Inlet-Mt Frankland region of south-west Western Australia,1912-1913. CALMScience 2(4):373–386.
Abbott I (1999). The avifauna of the forests of south-west Western Australia: changes in speciescomposition, distribution, and abundance followinganthropogenic disturbance. CALMScienceSupplement 5:1–175.
Abbott I (2000). Impact of agricultural developmentand changed fire regimes on species composition ofthe avifauna in the Denmark region of south-westWestern Australia, 1889–1999. CALMScience3(3):279–308.
Abbott I (2001a). Aboriginal names of mammal speciesin south-west Western Australia. CALMScience3(4):433–486.
Abbott I (2001b). The Bilby Macrotis lagotis(Marsupialia: Peramelidae) in south-western Australia:original range limits, subsequent decline, andpresumed regional extinction. Records of the WesternAustralian Museum 20:271–305.
Abbott I (2002). Origin and spread of the cat, Feliscatus, on mainland Australia, with a discussion ofthe magnitude of its early impact on native fauna.Wildlife Research 29(1):51–74.
Abbott I. (2003). Aboriginal fire regimes in south-westWestern Australia: evidence from historicaldocuments. In: Fire in Ecosystems of south-west WesternAustralia: impacts and management. (I AbbottNBurrows, Eds). Backhuys Publishing, Leiden, theNetherlands. Perth, Western Australia, pp. 119–146.
Abbott I, Burbidge AA (1995). The occurrence ofmammal species on the islands of Australia: asummary of existing knowledge. CALMScience1(3):259–324.
Alacs E, Aplers D, de Tores PJ et al. (2003). Identifyingthe presence of quokkas (Setonix brachyurus) andother macropods using cytochrome b analyses fromfaeces. Wildlife Research 30(1):41–47.
Alacs EA (2001). Conservation genetics of the Quokka,Setonix brachyurus. Honours thesis. VeterinaryBiology. Murdoch University. Perth, Western Australia
Aldrich J (1921). File note dated 9 May 1921 in WesternAustralian Government State Records OfficeArchives: 56/21 AN 108 Acc 477. Game Act –Diseases among Quaggas (Wallibies) in South West.
Aldrich J (1933). File note dated 9 September 1933 inWestern Australian Government State Records OfficeArchives: 56/21 AN 108 Acc 477. Game Act –Diseases among Quaggas (Wallibies) in South West.
Alexander WB (1914). The history of zoology in WesternAustralia. Part I. – Discoveries in the 17th Century.The Journal of the Natural History and Science Societyof Western Australia 5:49–64.
Alexander WB (1916). History of zoology in WesternAustralia. Part II. – 1797–1829. Journal andProceedings of the Royal Society of Western Australia1:83–149.
Alexander WB (1918). History of zoology in WesternAustralia. Part 3. – 1829–1840. Journal and
Proceedings of the Royal Society of Western AustraliaIII:37–69.
Algar D, Kinnear JE (1996). Secondary poisoning offoxes following a routine 1080 rabbit-baitingcampaign in the Western Australian wheatbelt.CALMScience 2(2):149–152.
Archer M (1974a). New information about thequaternary distribution of the Thylacine (Marsupialia,Thylacinidae) in Australia. Journal of the Royal Societyof Western Australia 57(2):43–50.
Archer M (1974b). Excavations in the Orchestra ShellCave, Wanneroo, Western Australia. Archaeology andPhysical Anthropology in Oceania 9(2):156–162.
Archer M, Baynes A (1972). Prehistoric mammal faunasfrom two small caves in the extreme south-west ofWestern Australia. Journal of the Royal Society ofWestern Australia 55(3):80–90.
Austin C (1979). The Well-Being of the JandakotQuokka Population. Population Analysis. Thesissubmitted in partial fulfillment of Bachelor of Science(Honours). Zoology Department. University ofWestern Australia. Perth. Western Australia. Perth
Balme J, Merrilees D, Porter JK (1978). Late Quaternarymammal remains, spanning about 30 000 years, fromexcavations in Devil’s Lair, Western Australia. Journalof the Royal Society of Western Australia 61(2):33–65.
Bannister JL (1970). Report on Biological Survey, TwoPeoples Bay Nature Reserve (No. 27956) February3 to 17, 1970. Unpublished Report for the WesternAustralian Museum. Perth, Western Australia.
Barker R, McGlinchy A. (2001). M-R-I. MarkRecapture Interface for Programs Capture and Jolly.Version 2. Landcare Research. Manaaki Whenua.New Zealand. http://www.landcareresearch.co.nz
Barker S (1961). Studies on marsupial nutrition. III.The copper-molybdenum-inorganic sulphateinteraction in the Rottnest quokka, Setonix brachyurus(Quoy & Gaimard). Australian Journal of BiologicalScience 14:646–658.
Barker S, Main AR, Sadleir RM (1957). Recent captureof the quokka (Setonix brachyurus) on the mainland.The Western Australian Naturalist 6:53–55.
Barrett S (1996). A Biological Survey of Mountains inSouthern Western Australia. Unpublished reportprepared for the Western Australian Department ofConservation and Land Management and theAustralian Nature Conservation Agency.
Baynes A (1979). The Analysis of a Late QuarternaryMammal Fauna from Hastings Cave, Jurien. PhDThesis. University of Western Australia. Perth
Baynes A. (1987). The original mammal fauna of theNullarbor and southern peripheral regions: evidencefrom skeletal remains in the superficial cave deposits.In: A Biological Survey of the Nullarbor Region Southand Western Australia in 1984. (NL McKenzieACRobinson, Eds). South Australian Department ofEnvironment and Planning. Adelaide, pp. 139–152.
Baynes A, Merrilees D, Porter JK (1975). Mammalremains from the upper levels of a late Pleistocene
Distribution of the quokka 35
deposit in Devil’s Lair, Western Australia. Journal ofthe Royal Society of Western Australia 58(4):97–126.
Berdoy M, Webster JP, MacDonald DW (2000). Fatalattraction in rats infected with Toxoplasma gondii.Proceedings of the Royal Society of London (Series B)267:1591–1594.
Biodiversity Group Environment Australia (1999). ThreatAbatement Plan for Predation by the European RedFox. Environment Australia under the NaturalHeritage Trust. Canberra, ACT, Australia.
Bradshaw SD. (1991). Rottnest Quokka Workshop. Aninquiry into the current status of the quokka, held in theRottnest Research Station on 9–10 March, 1991.Unpublished report prepared for the Rottnest IslandResearch Committee.
Brown KL, Hickman EJ, Lally TR (1998). Discoveryof the red flowering Corymbia ficifolia in the StirlingRange. The Western Australian Naturalist 22(1):23–24.
Burbidge AA (Ed.).) (1977). The Status of Kangaroosand Wallabies in Australia. Report of the WorkingGroup on Macropod Habitat of the Standing Committeeof the Council of Nature Conservation Ministers.Department of Environment, Housing andCommunity Development. Australian GovernmentPublishing Service, Canberra.
Burbidge AA, McKenzie NL (1989). Patterns in themodern decline of Western Australia’s vertebratefauna: causes and conservation implications.Biological Conservation 50:143–198.
Burbidge AA, Manly BFJ (2002). Mammal extinctionson Australian islands: causes and conservationimplications. Journal of Biogeography 29:465–473.
Burbidge AH. (2003). Birds and fire in theMediterranean climate of south-west WesternAustralia. In: Fire in Ecosystems of south-west WesternAustralia: impacts and management. (I AbbottNBurrows, Eds). Backhuys Publishing, Leiden, theNetherlands. Perth, Western Australia, pp. 321–348.
Burnet SM (1968). Changing Patterns: an atypicalautobiography. Sun Books. Melbourne.
Burnham KP, Anderson DR (2002). Model Selection andMultimodel Inference. A Practical Information-Theoretic Approach. Second Edition. Springer-Verlag.New York.
Burrows N, Wardell-Johnson G. (2003). Fire and plantinteractions in forested ecosystems of south-westAustralia. In: Fire in Ecosystems of south-west WesternAustralia: impacts and management. (I AbbottNBurrows, Eds). Backhuys Publishing, Leiden, theNetherlands. Perth, Western Australia, pp. 225–268.
Burrows ND, Friend GR. (1998). Biological indicatorsof appropriate fire regimes in southwest Australianecosystems. In: Fire in Ecosystem Management:Shifting the Paradigm from Suppression to Prescription.(TL PrudenLA Brennan, Eds). Proceedings of theTall Timbers Fire Ecology Conference No. 20. TallTimbers Research Station, Tallahassee, Florida. 413–421.
Burrows ND, Ward B, Robinson AD (1995). Jarrahforest fire history from stem analysis andanthropological evidence. Australian Forestry58(1):7–16.
Butler WH (1969). Remains of Sarcophilus the‘Tasmanian’ Devil (Marsupialia, Dasyuridae) fromcoastal dunes south of Scott River, Western Australia.The Western Australian Naturalist 11(4):87–88.
Calaby JH (1971). The current status of the Australianmacropodidae. Australian Journal of Zoology16(1):17–28.
Calaby JH, Grigg GC. (1989). Changes in macropodcommunities and populations in the past 200 years,and the future. In: Kangaroos, Wallabies and Rat-Kangaroos. Vol. 2 (G Grigg, P JarmanI Hume, Eds).Surrey Beatty & Sons Pty Limited. Chipping Norton,New South Wales, pp. 813–820.
Caley P (1997). Movements, activity patterns and habitatuse of feral pigs (Sus scrofa) in a tropical habitat.Wildlife Research 24(1):77–88.
CALM (1999). Department of Conservation and LandManagement Annual Report 1998–1999. Perth,Western Australia.
CALM (2000). Department of Conservation and LandManagement Annual Report 1999–2000. Perth,Western Australia.
CALM (2001). Department of Conservation and LandManagement Annual Report 2000–2001. Perth,Western Australia.
CALM (2002). Department of Conservation and LandManagement Annual Report 2001–2002. Perth,Western Australia.
CALM (2003). Department of Conservation and LandManagement Annual Report 2002–2003. Perth,Western Australia.
CALM (2004). Department of Conservation and LandManagement Annual Report 2003–2004. Perth,Western Australia.
CALM. (unpublished). Unpublished Department ofConservation and Land Management databaserecords of fauna occurrence in Western Australia.
Calver MC, Dell J (1998a). Conservation status ofmammals and birds in the southwestern Australianforests. II. Are there unstudied, indirect or long-termlinks between forestry practices and species declineand extinction? Pacific Conservation Biology 4:315–325.
Calver MC, Dell J (1998b). Conservation status ofmammals and birds in the southwestern Australianforests. I. Is there evidence of direct links betweenforestry practices and species decline and extinction?Pacific Conservation Biology 4:296–314.
Catling PC (1988). Similarities and contrasts in the dietsof foxes, Vulpes vulpes, and Cats, Felis catus, relativeto fluctuating prey populations and drought.Australian Wildlife Research 15:307–317.
Caughley G, Gunn A (1996). Conservation biology intheory and practice. Blackwell Science. Carlton.
36 P.J. de Torres et al.
Christensen P (1978). The introduced european red fox(Vulpes vulpes): a serious threat to native mammalsof the south-west. Forests Department of WesternAustralia. Perth. November 1978.
Christensen P, Burrows N. (1994). Project desertdreaming: experimental reintroduction of mammalsto the Gibson Desert, Western Australia. In:Reintroduction Biology of Australian and NewZealand Fauna. (M Serena, Ed.). Surrey Beatty andSons. Sydney, pp. 199–207.
Christensen P, Annels A, Liddelow G et al. (1985).Vertebrate fauna in the southern forests of WesternAustralia: a survey. Bulletin 94. Forests Departmentof Western Australia. Perth. 1985.
Christensen PE (undated). Effect of Prescribed Burningon One Population of Quokka (Setonix brachyurus).Local Expt. 117. Unpublished report for the ForestsDepartment of Western Australia.
Christensen PE, Kimber PC. (1975). Effects ofprescribed burning on the flora and fauna of south-west Australian forests. In: Managing TerrestrialEcosystems. Brisbane. (J KikkawaHA Nix, Eds).Ecological Society of Australia. Proceedings of theEcological Society of Australia, Vol 9: 85–106.
Christensen PES (1980a). The biology of Bettongiapenicillata Gray, 1837, and Macropus eugenii(Desmarest, 1817) in relation to fire. Bulletin 91.Forests Department of Western Australia. Perth.1980.
Christensen PES (1980b). A sad day for native fauna.Forest Focus 23:3–12.
Churchill DM (1959). Late Quarternery eustaticchanges in the Swan River District. Journal of theRoyal Society of Western Australia 42:53–55.
Clarke JR (1948). Anatomy of the Quokka. Part 1. –External morphology and large intestine. Journal ofthe Royal Society of Western Australia 33:59–150.
Coman BJ. (1991). The ecology and control of feralcats in Australia. In: The Impact of Cats on NativeWildlife. (C Potter, Ed.). National EndangeredSpecies Program, Australian National Parks andWildlife Service, Canberra. 70–74.
Cook DL (1960). Some mammal remains found in cavesnear Margaret River. The Western AustralianNaturalist 7(4):107–108.
Cook DL (1963). The fossil vertebrate fauna of Strongs’Cave, Boranup, Western Australia. The WesternAustralian Naturalist 8(7):153–162.
Corbett LK (1995a). The Dingo in Australia and Asia.University of New South Wales Press. Sydney.
Corbett LK. (1995b). Dingo, Canis lupus dingo. In:The Mammals of Australia. (R Strahan, Ed.). ReedBooks. Sydney, pp. 696–698.
Corbett LK, Newsome AE (1987). The feeding ecologyof the dingo. III. Dietary relationships with widelyfluctuating prey populations in arid Australia: anhypothesis of alternation of predation. Oecologia74:215–227.
Courtney J. (1993). Climate. In: Mountains of Mystery.A Natural History Of The Striling Range. (CThomson, G HallG Friend, Eds). Western AustralianDepartment of Conservation and Land Management.Perth, Western Australia.
Crabb D (1973). Quokka on the Albany Highway. TheWestern Australian Naturalist 12(6):141.
Daubney ME, Gardner G, Moore EM et al. (Eds.)(undated). Northcliffe Remembers. NorthcliffePioneer Museum, Northcliffe, Western Australia.
de Tores P (1999). 1997–1998 Report to EnvironmentAustralia National Feral Animal Control ProgramISF06638. Control and Ecology of the Red Fox inWestern Australia – Prey Response to 1080 BaitingOver Large Areas. Unpublished report prepared forEnvironment Australia, National Feral AnimalControl Program. ISF06638. Department ofConservation and Land Management. Perth.
de Tores P (2005). A proposal for translocation of thewestern ringtail possum, Pseudocheirus occidentalis,an arboreal marsupial endemic to the south-west ofWestern Australia. Unpublished report prepared forthe Western Australian Department of Conservationand Land Management. Perth, Western Australia.
de Tores PJ, Hayward MW, Rosier SM (2004). Thewestern ringtail possum, Pseudocheirus occidentalis,and the quokka, Setonix brachyurus, case studies:Western Shield review – February 2003. ConservationScience Western Australia 5(2):235–257.
Dell J. (1983). The Importance of the Darling Scarp toFauna. In: Scarp Symposium. A one day symposium toreview land use and conflicts on the Darling Scarp.Perth. Western Australian Institute of Technology.WAIT Environmental Studies Group. Report No.10:
Dell J, Hyder-Griffiths B (2002). A PreliminaryAssessment of the Faunal Values of the Muddy LakesPortion of the South Bunbury to Capel CoastalCorridor. Unpublished Draft Report. Departmentof Environmental Protection.
Dickman CR. (1992). Conservation of Mammals in theAustralasian Region: the Importance of Islands. In:Australia and the Global Environmental Crisis. (JNColesJM Drew, Eds). Academy Press. Canberra, pp.175–214.
Dickman CR (1996). Overview of the impacts of feralcats on Australian native fauna. Australian NatureConservation Agency. Canberra.
Dillon M (1993). Summary of Studies Involving theMainland Quokka in the Dwellingup District Sincethe Early 1970s. Unpublished Report for theDepartment of Conservation and Land Management.Perth, Western Australia.
Dillon MJ (1996). Summary of all Sites Assessed forQuokka Abundance. Unpublished report preparedfor the Western Australian Department ofConservation and Land Management. Perth. WesternAustralia.
DNRE (2002). Victorian Pest Management. A
Distribution of the quokka 37
framework for Action. Fox Management Strategy. TheState of Victoria, Department of Natural Resourcesand Environment.
Dortch C, Merrilees D (1971). A Salvage Excavation inDevil’s Lair, Western Australia. Journal of the RoyalSociety of Western Australia 54(4):103–113.
Dortch CE, Merrilees D (1973). Human occupation ofDevil’s Lair, Western Australia during the Pleistocene.Archaeology & Physical Anthropology in OceaniaVIII(2):89–115.
Dunnet GM (1962). A population study of the quokka,Setonix brachyurus Quoy and Gaimard (Marsupialia).II. Habitat, movements, breeding and growth.CSIRO Wildlife Research 7(1):13–32.
Dunnet GM (1963). A population study of the quokka,Setonix brachyurus Quoy and Gaimard (Marsupialia).III. The estimation of population parameters bymeans of the recapture technique. CSIRO WildlifeResearch 8:78–117.
Eldridge MDB, Close RL. (1995). Allied rock-wallaby,Petrogale assimilis. In: The Mammals of Australia.(R Strahan, Ed.). Reed Books. Sydney, pp. 365–366.
ESRI. (1999–2004). ArcGIS 9. Environmental SystemsResearch Institute, Inc.
Evans HD (undated). Pioneering in the Karri Country.Pemberton-Northcliffe Tourist Bureau.
Evans M, Gordon G. (1995). Bridled Nailtail Wallaby,Onychogalea fraenata. In: The Mammals of Australia.(R Strahan, Ed.). Reed Books. Sydney, pp. 356–358.
Friend JA (1990). The Numbat Myrmecobius fasciatus(Myrmecobiidae): history of decline and potential forrecovery. Proceedings of the Ecological Society ofAustralia 16:369–377.
Friend JA, Thomas ND. (1994). Reintroduction andthe numbat recovery programme. In: ReintroductionBiology of Australian and New Zealand Fauna. (MSerena, Ed.). Surrey Beatty and Sons. Sydney, pp.189–198.
Friend JA, Butler T (nominally 1998). Gilbert’s PotorooRecovery Team Annual Report 1998. Unpublishedreport for the Western Australian Department ofConservation and Land Management. Albany,Western Australia.
Friend JA, Butler T (nominally 1999). Gilbert’s PotorooRecovery Team Annual Report 1999. Unpublishedreport for the Western Australian Department ofConservation and Land Management. Albany,Western Australia.
Gardner CA (1957). The fire factor in relation to thevegetation of Western Australia. The WesternAustralian Naturalist 5(7):166–173.
Gaynor A (2000). Report on the history of the arrival ofthe feral cat population in Western Australia.CALMScience 3(2):149–179.
Gibb DGA, Kakulas BA, Perrett DH et al. (1966).Toxoplasmosis in the Rottnest quokka (Setonixbrachyurus). Australian Journal of ExperimentalBiology and Medical Science 44:665–672.
Gibson DF, Lundie-Jenkins G, Langford DG et al.(1994). Predation by Feral Cats, Felis catus, on theRufous Hare-wallaby, Lagorchestes hirsutus, in theTanami Desert. Australian Mammalogy 17:103–108.
Gilfillan S. (unpublished). Unpublished database recordsof fauna occurrence from the south coast region ofWestern Australia.
Glauert L (1926). A List of Western Australian Fossils.Geological Survey of Western Australia Bulletin 88:36–71.
Glauert L (1933). The distribution of the marsupials inWestern Australia. Journal of the Royal Society ofWestern Australia 19:17–32.
Glauert L (1948). The cave fossils of the south-west.The Western Australian Naturalist 1:100–104.
Glauert L (1950). The development of our knowledgeof the marsupials of Western Australia. Journal of theRoyal Society of Western Australia:115–134.
Gooding CD (1955). The vermin bonus system inWestern Australia. Part 1 – The distribution ofpayments. The Journal of Agriculture of WesternAustralia 4(4):433–449.
Gould J (1863). The Mammals of Australia. Volume ll.The author. 26 Charlotte Street, BedfordSquare,London.
Government Gazette of Western Australia. (1933). TheVermin Act, 1918. Proclamation. Perth, WesternAustralia.
Green N. (1989). Aborigines of the Albany Region 1821–1898. The Bicentennial Dictionary of WesternAustralians. Volume VI. The University of WesternAustralia Press, Perth, Western Australia.
Haigh SA, Gaynor WT, Morris KD. (1994). A healthmonitoring program for captive, wild and translocatedchuditch (Dasyurus geoffroii). In: Proceedings of the1994 Conference of the Australian Association ofVeterinary Conservation Biologists. Canberra. (AWEnglishGR Phelps, Eds). Australian Association ofVeterinary Conservation Biologists. 52–66.
Hallam SJ (1975). Fire and Hearth a study of Aboriginalusage and European usurpation in south-westernAustralia. Australian Institue of Aboriginal Studies.Canberra.
Hart R (1977). Report on trapping program andassessment of health of quokkas at Holyoake Site,near Dwellingup. Unpublished report.
Hart RP (1981). The ecology of Salmonella in a wildmarsupial, the Quokka (Setonix brachyurus). PhDthesis. Department of Zoology. University of WesternAustralia. Perth
Hart RP, Bradshaw SD, Iveson JB (1986). Salmonellainfections and animal condition in the mainland andBald Island populations of the quokka (Setonixbrachyurus: Marsupialia). Journal of the Royal Societyof Western Australia 69(1):7–11.
Hayward MW (2002). The ecology of the quokka(Setonix brachyurus) (Macropodidae: Marsupialia) inthe northern jarrah forest of Australia. PhD thesis.
38 P.J. de Torres et al.
School of Biological, Earth and EnvironmentalScience. University of New South Wales. Sydney
Hayward MW, de Tores PJ, Dillon MJ et al. (2003).Local population structure of a naturally occurringmetapopulation of the quokka (Setonix brachyurusMacropodidae: Marsupialia). Biological Conservation110:343–355.
Hayward MW, de Tores PJ, Dillon MJ et al. (2005).Using faecal pellet counts along transects to estimatequokka (Setonix brachyurus) population density.Wildlife Research 32(6):503–507.
Heatwole H. (1987). Major components and distributionof the terrestrial fauna. In: Fauna of Australia.General Articles. Vol. 1A (GR DyneDW Walton,Eds). Australian Government Publishing Service.Canberra, pp. 101–135.
Herrick EH (1961). Some preliminary experiments onadrenal function during seasonal stresses in a wildmarsupial (Setonix brachyurus). Journal of the RoyalSociety of Western Australia 44(2):61–64.
Hilton-Taylor C. (2000). 2000 IUCN Red List ofThreatened Species. IUCN, Gland, Switzerland andCambridge, UK.
Hodgkin EP, Sheard K (1959). Rottnest Island: theRottnest Biological Station and recent scientificresearch. 42(3):65–95.
Holsworth WN (1964). Marsupial Behaviour with specialreference to Population Homeostasis in the Quokkason the West End of Rottnest Island. PhD Thesis.Department of Zoology. University of WesternAustralia. Perth, Western Australia
Holsworth WN (1967). Population dynamics of thequokka, Setonix brachyurus, on the west end ofRottnest I., Western Australia. I. Habitat anddistribution of the quokka. Australian Journal ofZoology 15:29–46.
Horsup A, Evans M (1993). Predation by feral cats,Felis catus, on an endangered marsupial, the bridlednailtail wallaby, Onychogalea franenata. AustralianMammalogy 16(1):83–84.
How RA, Dell J, Humphreys WF (1987). The groundvertebrate fauna of coastal areas between Busseltonand Albany, Western Australia. Records of the WesternAustralian Museum 13(4):553–574.
Jarman P. (1986). The red fox – an exotic, large predator.In: The ecology of exotic animals and plants: someAustralian case histories. (RL Kitching, Ed.). JohnWiley and Sons. Brisbane, pp. 45–61.
Jenkins CFH (1974). The decline of the Dalgite (Macrotislagotis) and other wild life in the Avon Valley. TheWestern Australian Naturalist 12(8):169–172.
Johnson KA, Burbidge AA, McKenzie NL. (1989).Australian Macropodoidea: Status, causes of declineand future research and management. In: Kangaroos,Wallabies and Rat-Kangaroos. Vol. 2 (G Grigg, PJarmanI Hume, Eds). Surrey Beatty & Sons PtyLimited. Chipping Norton, New South Wales, pp.641–657.
Jones M. (1995). Tasmanina Devil, Sarcophilus harrisii.
In: The Mammals of Australia. (R Strahan, Ed.).Reed Books. Sydney, pp. 82–84.
Kabay ED, Start AN (1976). Results of the Search forthe Potoroo in South West and South Coast ofWestern Australia 1975/76. Unpublished reportprepared for the Western Australian Department ofFisheries and Wildlife. Perth, Western Australia.
Kimber PC. (1974). Some effects of prescribed burningon jarrah forest birds. In: 3rd Fire Ecology Symposium.Monash University, Melbourne. 49–57.
King DR, Smith LA (1985). The distribution of theEuropean red fox (Vulpes vulpes) in Western Australia.Records of the Western Australian Museum 12(2):197–205.
King DR, Oliver AJ, Mead RJ (1981). Bettongia andfluoroacetate: a role for 1080 in fauna management.Australian Wildlife Research 8:529–536.
Kinnear JE, Onus ML, Bromilow RN (1988). Foxcontrol and rock-wallaby population dynamics.Australian Wildlife Research 15:435–450.
Kinnear JE, Onus ML, Sumner NR (1998). Fox Controland rock-wallaby population dynamics – II. Anupdate. Wildlife Research 25(1):81–88.
Kinnear JE, Sumner NR, Onus ML (2002). The redfox in Australia – an exotic predator turned biocontrolagent. Biological Conservation 108:335–359.
Kirke A (1983). Quokka, Setonix brachyurus, at GreenRange. The Western Australian Naturalist15(6):146.
Kitchener DJ (1970). Aspects of the Response of theQuokka to Environmental Stress. PhD Thesis.University of Western Australia. Perth, WesternAustralia
Kitchener DJ (1972). The importance of shelter to thequokka, Setonix brachyurus (Marsupialia), on RottnestIsland. Australian Journal of Zoology 20:281–299.
Kitchener DJ (1973). Notes on home range andmovement in two small macropods, the potoroo(Potorous apicalis) and the quokka (Setonixbrachyurus). Mammalia 37:231–240.
Kitchener DJ. (1995). Quokka, Setonix brachyurus. In:The Mammals of Australia. (R Strahan, Ed.). ReedBooks. Sydney, pp. 401–403.
Kitchener DJ, Vicker E (1981). Catalogue of modernmammals in the Western Australian Museum 1895to 1981. Western Australian Museum. Perth.
Kitchener DJ, Chapman A, Barron G. (1978). Mammalsof the Northern Swan Coastal Plain. In: FaunalStudies of the Northern Swan Coastal Plain. AConsideration of Past and Future Changes. TheWestern Australian Museum, for the Department ofConservation and Environment. Perth, WesternAustralia.
Krefft G (1867). List of specimens collected by MrGeorge Masters, Assistant Curator, during the year1866. Annual Report. Australian Museum. pp: 6–7.
Krefft G (1869). Specimens collected by Mr GeorgeMasters, Assistant Curator, in Western Australia.Annual Report. Australian Museum. pp: 7–8.
Distribution of the quokka 39
Lamont BB, Ward DJ, Eldridge J et al. (2003). Believingthe Balga: a new method for gauging the fire historyof vegetation using grass trees. In: Fire in Ecosystemsof south-west Western Australia: impacts andmanagement. (I AbbottN Burrows, Eds). BackhuysPublishing, Leiden, the Netherlands. Perth, WesternAustralia, pp. 147–169.
Lane-Poole CE (1921). File note dated 17 May 1921 inWestern Australian Government State Records OfficeArchives: 56/21 AN 108 Acc 477. Game Act –Diseases among Quaggas (Wallibies) in South West.
Laurance WF, Harrington GN (1997). Ecologicalassociations of feeding sites of feral pigs in theQueensland wet tropics. Wildlife Research 24(5):579–590.
Lenghaus C, Obendorf DL, Wright FH. (1990).Veterinary aspects of Perameles gunnii biology withspecial reference to species conservation. In:Management and Conservation of Small Populations.Proceedings of a conference held in Melbourne,Australia. September 26–27, 1989. (TW ClarkJHSeebeck, Eds). Chicago Zoological Society.Brookfield, Illinois.
Liddelow GL (2006). Rapid survey of quokka (Setonixbrachyurus) in the southern forests. In AnnualResearch Activity Report. July 2005 – June 2006.Science Division. Discovering the nature of WA.Department of Consrevation and Land Management.Perth, Western Australia. pp: 71–72.
Liddelow GL, Wheeler IB, Kavanagh RP. (2002). Owlsin the Southwest Forests of Western Australia. In:Ecology and Conservation of Owls. (I Newton, RKavanagh, J Olsen et al., Eds). CSIRO. Melbourne,pp. 233–241.
Lilley I (1993). Recent research in southwestern WesternAustralia: a summary of initial findings. AustralianArchaeology 39:34–41.
Long JL (1988). Introduced birds and mammals inWestern Australia. Technical Series 1 (2nd Edn).Agricultural Protection Board, Perth, WesternAustralia.
Lowry DC. (1967). Geological Survey of WesternAustralia. Bureau of Mineral Resources, Geologyand Geophysics, Perth, Western Australia.
Lundelius E (1957). Additions to knowledge of theranges of Western Australian mammals. The WesternAustralian Naturalist 5:173–182.
Lundelius EL (1960). Post Pleistocene Faunal Successionin Western Australia and its Climatic Interpretation.Report of the International Geological Congress. XXISession. Part IV Chronology and Climatology of theQuaternary. Norden. 1960. pp: 142–153.
Lundelius EL (1966). Marsupial carnivore dens inAustralian caves. Studies in Speleology 1(4):174–181.
Lunney D, Matthews A, Grigg J (2001). The diet ofAntechinus agilis and A. swainsonii in unlogged andregenerating sites in Mumbula State Forest, south-eastern New South Wales. Wildlife Research28(5):459–464.
Main AR (1961). The occurrence of macropodidae onislands and its climatic and ecological implications.Journal of the Royal Society of Western Australia44(3):84–89.
Main AR, Yadav M (1971). Conservation of macropodsin reserves in Western Australia. BiologicalConservation 3(2):123–133.
Main AR, Shield JW, Waring H. (1959). Recent studieson marsupial ecology. In: Biogeography and Ecologyin Australia. (RL Keast, RL Crocker, CS Christian,Eds). The Hague, Netherlands, pp. 315–331.
Marsack P, Campbell G (1990). Feeding behaviour anddiet of dingoes in the Nullarbor Region, WesternAustralia. Australian Wildlife Research 17:349–357.
Maxwell S, Burbidge AA, Morris K (Eds.) (1996). The1996 Action Plan For Australian Marsupials andMonotremes. Endangered Species Program. ProjectNumber 500. Wildlife Australia, Canberra, ACT.
McCaw L, Hamilton T, Rumley C. (2005). Applicationof fire hstory records to contemporary managementissues in south-west Australian forests. In: A forestconscienceness: proceedings 6th National Conferenceof the Australian Forest History Society Inc, 12–17September 2004. Augusta, Western Australia. (MCalver, H Bigler-Cole, G Bolton et al., Eds).Millpress, Rotterdam.
Mead RJ, Twigg LE, King DR et al. (1985). Thetolerance to fluoroacetate of geographically separatedpopulations of the quokka (Setonix brachyurus).Australian Journal of Zoology 21(6):503–511.
Meagher SJ (1974). The food resources of the aboriginesof the south-west of Western Australia. Records of theWestern Australian Museum 3:14–65.
Merrilees D (1965). Quokka at Yanchep in historic time.The Western Australian Naturalist 10(1):18.
Merrilees D (1968). Man the destroyer: late Quaternarychanges in the Australian marsupial fauna. Journalof the Royal Society of Western Australia 51(1):1–24.
Merrilees D, Dix WC, Hallam SJ et al. (1973).Aboriginal man in southwest Australia. Journal of theRoyal Society of Western Australia 56:44–55.
Middleton E (2001). Nuyts Wilderness Wildfire ImpactReport. 7–11 March 2001, Walpole, WesternAustralia. Unpublished report prepared for theDepartment of Conservation and Land Management,Perth, Western Australia.
Molsher RL (1999). The ecology of feral cats, Felis catus,in the open forest in New South Wales: interactionswith food resources and foxes. PhD Thesis. Schoolof Biological Sciences. University of Sydney. Sydney
Morton SR. (1990). The impact of European settlementon the vertebrate animals of arid Australia: aconceptual model. In: Australian Ecosystems: 200years of Utilization, Degradation and Reconstruction.Geraldton, Western Australia. (DA Saunders, AJMHopkinsRA How, Eds). Ecological Society ofAustralia. Proceedings of the Ecological Society ofAustralia, Vol 16: 201–213.
Newsome AE. (2001). The biology and ecology of thedingo. In: A Symposium on the Dingo. (CR Dickman,
40 P.J. de Torres et al.
D Lunney, Eds). Royal Zoological Society of New SouthWales. Mosman, New South Wales, pp. 20–33.
Newsome AE, Corbett LK, Catling PC et al. (1983).The feeding ecology of the dingo. I. Stomach contentsfrom trapping in southeastern Australia, and the non-target wildlife also caught in dingo traps. AustralianWildlife Research 10:477–486.
Nicholls DG (1971). Daily and seasonal movements ofthe quokka, Setonix brachyurus (Marsupialia), onRottnest Island. Australian Journal of Zoology 19:215–226.
NSW NPWS (2001). Threat Abatement Plan forPredation by the red fox (Vulpes vulpes). NSW NationalParks and Wildlife Service. Hurstville, Sydney.
Packer WC (1965). Environmental influences on dailyand seasonal activity in Setonix brachyurus (Quoy andGaimard) (Marsupialia). Animal Behaviour 13:270–283.
Packer WC (1968). Eosinophils and population densityin the marsupial Setonix. Journal of Mammalogy49(1):124–126.
Pavlov PM. (1995). Pig, Sus scrofa. In: The Mammals ofAustralia. (R Strahan, Ed.). Reed Books. Sydney,pp. 698–699.
Pavlov PM, Crome FHJ, Moore LA (1992). Feral pigs,rainforest conservation and exotic disease in northQueensland. Wildlife Research 19(2):179–194.
Pearce AT (1969). File note dated 24 February 1969 inWestern Australian Government State Records OfficeArchives: 65/50 Acc 477. Folio 120. F.C. Act –Idnetification of Fauna – General File.
Perry DH (1971). Observations of a young naturalistin the Lower Blackwood Valley in the year 1919. TheWestern Australian Naturalist 12(2):46–47.
Perry DH (1973). Some notes on the decline andsubsequent recovery of mammal populations in thesouth-west. The Western Australian Naturalist12(6):128–130.
Porter JK (1979). Vertebrate remains from a stratifiedHolocene deposit in Skull Cave, Western Australia,and a review of their significance. Journal of the RoyalSociety of Western Australia 61(4):109–117.
Prince RIT (1984). Exploitation of Kangaroos andWallabies in Western Australia I. A Review to 1970,with special reference on Red and Western GreyKangaroos. Wildlife Research Bulletin Number 13Western Australia. Department of Fisheries andWildlife, Perth, Western Australia.
Prince RIT. (1995). Banded hare-wallaby, Lagostrophusfasciatus. In: The Mammals of Australia. (R Strahan,Ed.). Reed Books. Sydney, pp. 406–408.
Recher HF, Lim L. (1990). A review of current ideas ofthe extinction, conservation and management ofAustralia’s terrestrial vertebrate fauna. In: AustralianEcosystems: 200 years of Utilization, Degradation andReconstruction. Geraldton, Western Australia. (DASaunders, AJM HopkinsRA How, Eds). EcologicalSociety of Australia. Proceedings of the EcologicalSociety of Australia, Vol.16: 287–301.
Richards JD, Short J (1996). History of the disappearanceof native fauna from the Nullarbor Plain through theeyes of long term resident Amy Crocker. The WesternAustralian Naturalist 21(2):89–96.
Ride WDL (1970). A Guide to the Native Mammals ofAustralia. Oxford University Press. Melbourne.
Risbey DA, Calver MC, Short JC et al. (2000). Theimpact of cats and foxes on the small vertebrate faunaof Heirisson Prong, Western Australia. WildlifeResearch 27(3):223–236.
Robertshaw JD, Harden RH (1985). The ecology ofthe dingo in north-eastern New South Wales III.Analysis of macropod bone fragments found in dingoscats. Australian Wildlife Research 12:163–171.
Roe R (1971). Trial excavation in a small cave, Gingin.The Western Australian Naturalist 11:183–184.
Rolls EC (1969). They All Ran Wild: The enthrallingstory of pests on the land in Australia. Angus andRobertson Ltd. Sydney.
Rounsevell DE, Mooney N. (1995). Thylacine,Thylacinus cynocephalus. In: The Mammals ofAustralia. (R Strahan, Ed.). Reed Books. Sydney,pp. 164–165.
Sadleir RMFS (1959). Comparative Aspects of theEcology and Physiology of Rottnest and ByfordPopulations of the Quokka (Setonix brachyurus Quoyand Gaimard). Honours thesis. University of WesternAustralia. Perth. Western Australia
Saunders G, Kay B (1996). Movements and home rangesof feral pigs (Sus scrofa) in Kosciusko National Park,New South Wales. Wildlife Research 23(6):711–720.
Schmidt W, Mason M (1973). Fire and fauna in thenorthern jarrah forest of Western Australia. TheWestern Australian Naturalist 12(6):162–164.
Schodde R, Tidemann SC (Eds.) (1982). The Reader’sDigest Complete Book of Australian Birds. Reader’sDigest Services Pty Ltd, Sydney.
Serventy DL (1951). Inter-specific competion on smallislands. The Western Australian Naturalist 3(1):59–60.
Serventy DL, Glauert L, Carnaby IC et al. (1954). Therecent increase of the rarer native mammals. TheWestern Australian Naturalist 4:128–141.
Serventy VN (1953). The archipelago of the Recherche.4. Mammals. Australian Geographical Society Report1:40–48.
Sharman GB (1954). The relationships of the quokka(Setonix brachyurus). The Western AustralianNaturalist 4:159–168.
Shield JW (1959). Rottnest Island: the Rottnest BiologicalStation and recent scientific research. 15. – Rottnestfield studies concerned with the Quokka. Journal ofthe Royal Society of Western Australia 42(3):76–82.
Shield JW (1964). A breeding season difference in twopopulations of the Australian macropod marsupial(Setonix brachyurus). Journal of Mammalogy45(4):616–625.
Short J, Kinnear JE, Robley A (2002). Surplus killing
Distribution of the quokka 41
by introduced predators in Australia – evidence forineffective anti-predator adaptions in native preyspecies? Biological Conservation 103:283–301.
Short J, Bradshaw SD, Giles J et al. (1992).Reintroduction of macropods (Marsupialia:Macropodoidea) in Australia – a review. BiologicalConservation 62:189–204.
Short JC, Calaby JH (2001). The status of Australianmammals in 1922 – collections and field notes ofmuseum collector Charles Hoy. Australian Zoologist31(4):533–562.
Shortridge GC (1909). An account of the geographicaldistribution of marsupials and monotremes of south-western Australia having special reference to thespecimens collected during the Balston Expeditionof 1904–1907. Proceedings of the Zoological Society(London) 1909:803–848.
Sinclair EA (1999). Genetic variation in two endangeredmarsupials, the quokka, Setonix brachyurus, andGilbert’s potoroo, Potorous gilbertii. PhD thesis.University of Western Australia. Perth, WesternAustralia
Smith AP, Quinn DG (1996). Patterns and causes ofextinction and decline in Australian conilurinerodents. Biological Conservation 77:243–267.
Smitz A (1998). Progress report on fate of quokkastranslocated to Karakamia Sanctuary, east of Perth.Unpublished report for Karakamia Sanctuary for theDepartment of Conservation and Land Management,Perth Western Australia.
Spencer PBS (1991). Evidence of predation by a feralcat Felis catus (Carnivora: Felidae) on an isolatedrock-wallaby colony in tropical Queensland.Australian Mammalogy 14:143–144.
Stewart DWR (1936). Notes on marsupial damage inpine plantations. Australian Forestry 1(2):41–42.
Storr GM (1963). Some factors inducing change in thevegetation of Rottnest Island. The Western AustralianNaturalist 9(1):15–22.
Storr GM (1964a). Studies on marsupial nutrition. IV.Diet of the quokka, Setonix brachyurus (Quoy &Gaimard), on Rottnest Island, Western Australia.Australian Journal of Biological Science 17(2):469–481.
Storr GM (1964b). The environment of the quokka(Setonix brachyurus) in the Darling Range, WesternAustralia. Journal of the Royal Society of WesternAustralia 47(1):1–2.
Storr GM (1965). Notes on Bald Island and the adjacentmainland. The Western Australian Naturalist9(8):187–196.
Thackway R, Cresswell ID (Eds.) (1995). An InterimBiogeographical Regionalisation for Australia: aFramework for Setting Priorities in the NationalReserves Cooperative Program. Version 4.0. AustralianNature Conservation Agency, Canberra.
Thomas O (1906). On mammals collected in South-west Australia for Mr. W. E. Balston. Proceedings ofthe Zoological Society of London 1906:468–478.
Thomson PC (1992). The behavioural ecology ofdingoes in north-western Australia. III. Hunting andfeeding behaviour, and diet. Wildlife Research19(5):531–542.
Troughton E (1965). Furred Animals of Australia.Angus and Robertson. Sydney, London, Melbourne.
Turney CSM, Bird MI, Fifield LK et al. (2001). Earlyhuman occupation at Devil’s Lair, southwesternAustralia 50,000 years ago. Quaternary Research55:3–13.
Vernes K, Dennis A, Winter J (2001). Mammalian dietand broad hunting strategy of the dingo (Canisfamiliaris dingo) in the wet tropical rain forests ofnortheastern Australia. Biotropica 33(2):339–345.
Wallace WR (1966). Fire in the jarrah forest. Journal ofthe Royal Society of Western Australia 49(2):33–44.
Walton DW (Ed.).) (1988). Zoological catalogue ofAustralia. 5 Mammalia. AGPS, Canberra.
Ward D, Sneeuwjagt R (1999). Believing the balga. Atime capsule of fire management. Landscope14(3):10–16.
Ward DJ, Lamont BB, Burrows CL (2001). Grasstreesreveal contrasting fire regimes in eucalypt forestsbefore and after European settlement in southwesternAustralia. Forest Ecology and Management 150:323–329.
Waring H (1956). Marsupial studies in Western Australia.The Australian Journal of Science 18:66–73.
Watts CHS (1969). Distribution and habits of the rabbitbandicoot. Transactions of the Royal Society of SouthAustralia 93:135–141.
Wayne A, Ward C, Rooney J et al. (2000). The immediateimpacts of timber harvesting and associated activitieson the Ngwayir (Pseudocheirus occidentalis) in theJarrah forest of Kingston State Forest Block. ProgressReport December 2000. Unpublished reportprepared for the Western Australian Department ofConservation and Land Management.
Weston LN (1921). File note dated 18 March 1921 inWestern Australian Government State Records OfficeArchives: 56/21 AN 108 Acc 477. Game Act –Diseases among Quaggas (Wallibies) in South West.
White SR (1952). The occurrence of the quokka in thesouth-west. The Western Australian Naturalist3(5):101–103.
Whitehouse SJO (1977). The diet of the Dingo inWestern Australia. Australian Wildlife Research4:145–150.
Wilson BA, Friend GR (1999). Responses of Australianmammals to disturbance factors. AustralianMammalogy 21(1):87–105.
Woinarski JCZ, Milne DJ, Wanganeen G (2001).Changes in mammal populations in relatively intactlandscapes of Kakadu National Park, NorthernTerritory, Australia. Austral Ecology 26:360–370.
42P.J. de Torres et al.
Table 1
Fossil and sub-fossil records of Setonix brachyurus. The site names are listed in a north to south latitudinal cline and are mapped in Figure 2. Bolding and an asterisk indicatethe deposits at that site are considered to be part of the original (modern) mammal fauna as noted by the source for the record or as inferred by the age of the deposit. Boldingonly indicates a more subjective assessment of age, where deposits were implied as part of the original (modern) mammal fauna. The abbreviation BP means before present.
Site name Location Description Comment Age Source Comment on Setonix record
Hastings Cave / 8 km north east of Jurien, Described by (Lundelius Hastings Cave / Drover’s Cave listed 5 900 ± 140 years BP Baynes (1979) Listed from 1 layer only,Drovers Cave * approx 210 km north 1957; Lundelius 1960) as by (Lundelius 1957) as representative (Lundelius 1960), however mammal occurrences were from
of Perth entrance formed by collapse of Jurien Bay caves. Bone deposits Lundelius (1957) reported top deposits considered to lessof roof. Cave formed from were concentrated near cave mouth most foot of surface deposits to than 1,000 years old andarches developing (Lundelius 1957) and thought to be represent fauna prior to effect considered to be part of the(coherence) over dissolved derived from owl pellets. Concurs with of introduced species. Also original faunaaeolian calcarenite Baynes (1979). Fragmentation of listed presence of rabbits and(Baynes 1979). bones of larger mammals considered house mouse and interpreted Lundelius (1960) Lower levels only
to be indicative of mammalian this as indicating youngerpredator accumulation (Baynes deposits. Baynes (1979) listed Merrilees (1968) Merrilees noted the only1979), with a minor contribution as a ages (Hastings Cave) as 400 ± records of Setonix from theresult of human occupation. 70 years BP at a depth of Moore River – Dongara region
10 mm to 11 400 ± 200 years are from lower parts of HastingsBP at 2.98m. Cave deposit. However,
Jefferys (pers comm. to MWH)subsequently listed Setonixoccurrence at Echidna Cave(Dongara), an un-named cavenear the northern boundary ofNambung National Park, HouseCave (Moore – Dongara riverarea), Greensand Cliffs (Gingin)and McIntyre Gully (Gingin)
Yanchep * North of Perth A complete quokka skeleton The skull was penetrated by a (Merrilees 1965) Skeleton with iron rod throughrecovered in 1965 registered small iron stake implanted skullwith the WAM as part of the during the animal’s life time,fossil collection. thereby inferring persistence of
quokkas in Perth area inmodern times
Orchestra Shell Wanneroo, North of 4 stratigraphic levels, Lair, deposits presumed to be not specified Archer (1974b) Lower two levelsCave Perth. Location noted as progressive faunal accumulated by Sarcophilus.
approx 16km south of impoverishment towards top.Yanchep cave describedby Lundelius (1957)
Murray Cave Approximately 40km Considered to represent a 3 090 ± 90 years BP. Contains (Archer 1974a)north of Perth carnivore’s accumulation, youngest dated occurrence
typical of Sarcophilus (from charcoal at 1–7cm depth)of thylacine
Rainbow Cave * 1 km inland, 1.5km south Collapsed limestone cave, Lilley (1993) aged at 340 ± 45 Lilley (1993) At depths of 10,15, 25, 35, 50of the mouth of the 30m deep, 15m wide years BP at depth of 5–10cm; and 75cm. Deposits aboveMargaret river, south- 790 ± 50 years BP at depth of 30cm are therefore presumablywest Western Australia 25–30cm; 8340 ± 45 years BP at from the modern fauna
depth of 35–40cm; 4150 ± 70years BP at depth of 70–75cm;and a second age of 1030 ± 50
Distribution of the quokka 43ye
ars
BP
was
det
erm
ined
at
dept
h of
25–
30cm
Mam
mot
h C
ave
Mar
gare
t R
iver
are
a,W
as t
he f
irst
cave
inD
epos
its c
onsi
dere
d by
Gla
uert
Lund
eliu
s (1
960)
dat
ed d
epos
itsG
laue
rt (
1948
)so
uth
wes
t W
este
rnW
este
rn A
ustr
alia
to
yiel
d(1
948)
to
be w
ater
lain
. M
erril
ees
from
the
upp
er s
and
laye
r as
Aus
tral
iafo
ssil
vert
ebra
te r
emai
ns &
(196
8) c
onsi
dere
d de
posi
ts w
ere
grea
ter
than
37
000
year
s B
P.C
ook
(196
3)de
scrib
ed b
y G
laue
rt (
1948
).m
ore
cons
iste
nt w
ith t
alus
and
not
Mer
rilee
s (1
968)
dat
ed u
pper
Orig
inal
ly r
ecog
nise
d as
wat
er t
rans
port
ed a
s su
gges
ted
part
s of
dep
osit
as 3
1,00
0M
erril
ees
(196
8)ha
ving
tw
o st
ratig
raph
icby
Gla
uert
year
s B
Pun
its:
a ye
llow
-red
san
dLu
ndel
ius
(196
0)la
yer
abov
e co
arse
red
san
d.B
oth
laye
rs c
appe
d w
ithtr
aver
tine
(Lun
deliu
s 19
60).
Rem
nant
s of
a p
revi
ousl
ylo
nger
tun
nel
in c
alca
reou
sae
olia
nite
s
Mar
gare
t R
iver
not
desc
ribed
Gla
uert
(19
26)
refe
rred
to
colle
ctio
nsP
leis
toce
neG
laue
rt (
1926
)ca
ves
from
‘cer
tain
cav
es n
ear
the
Mar
gare
tR
iver
in
the
extr
eme
Sou
th-W
est’.
Loca
tion
take
n no
min
ally
her
e as
Mam
mot
h C
ave
Har
ley’
s C
ave
Cap
e Le
euw
in –
Cap
eno
t de
scrib
edLi
sted
by
Mer
rilee
s (1
968)
as
List
ed a
s fa
una
char
acte
ristic
of
Mer
rilee
s (1
968)
Nat
ural
iste
reg
ion,
sou
thre
pres
enta
tive
of s
peci
men
s co
llect
edth
e re
gion
for
som
e hu
ndre
d to
wes
t W
este
rn A
ustr
alia
by L
owry
(19
67)
from
sur
face
litte
ra
few
tho
usan
d ye
ars
prio
r to
from
cav
es i
n C
ape
Leeu
win
–Cap
eE
urop
ean
settl
emen
tN
atur
alis
te r
egio
n. H
owev
er,
noin
dica
tion
of t
hese
col
lect
ions
bei
ngm
ade
by L
owry
(19
67)
who
dis
cuss
edge
omor
phol
ogy
Lake
Cav
eM
arga
ret
Riv
er a
rea,
not
desc
ribed
not
give
nG
laue
rt (
1948
)so
uth
wes
t W
este
rnA
ustr
alia
Dev
il’s L
air,
orS
outh
wes
t of
Witc
hclif
fe,
Sm
all
cave
in
aeol
ian
Dep
osits
con
side
red
(Lun
deliu
sS
urfic
ial
laye
rs d
escr
ibed
as
Dor
tch
and
Nan
nup
Cav
e of
sout
h w
est
Wes
tern
calc
aren
ite. T
rave
rtin
e flo
or19
60)
to b
e a
resu
lt of
Sar
coph
ilus,
rece
nt a
s 32
0 ye
ars
BP.
Oth
erM
erril
ees
(197
1)Lu
ndel
ius
(196
0;A
ustr
alia
over
red
cla
yey
sand
she
nce
the
nam
e D
evil’
s La
ir.de
posi
ts e
xten
ding
to
35,1
6019
66)
and
thou
ght
to b
e w
ashe
d in
.D
ortc
h an
d M
erril
ees
(197
1)ye
ars
BP
Bal
me
et a
l. (1
978)
.D
ortc
h an
dC
ook
(196
0).
*D
epos
its d
escr
ibed
fro
m t
wo
desc
ribed
as
arte
fact
s an
d fo
odLu
ndel
ius
(196
0) d
ated
the
Mer
rilee
s (1
973)
leve
ls:
imm
edia
tely
bel
owre
mai
ns p
resu
mab
ly le
ft as
a r
esul
t of
uppe
r le
vel
8,50
0 ±
160
year
str
aver
tine
floor
; an
d at
Abo
rigin
al u
se o
f th
e ca
ve.
Som
eB
P, w
ith d
epos
its a
t 4
feet
as
Lund
eliu
s (1
966)
dept
h of
4 f
eet.
rem
ains
may
hav
e be
en l
eft
by12
,175
± 2
75 y
ears
BP
Sar
coph
ilus.
Bay
nes
et a
l. (1
975)
Dor
tch
and
Mer
rilee
s (1
973)
Bal
me
et a
l.R
epre
sent
ed c
onsi
sten
tlyco
nsid
ered
upp
er l
ayer
s m
ay h
ave
liste
d up
perm
ost
stra
tigra
phic
(197
8)th
roug
hout
dep
osit
(all
ages
)be
en a
ccum
ulat
ed b
y hu
man
s an
dun
it to
be
less
tha
n 12
,000
belie
ved
Sar
coph
ilus
was
res
pons
ible
year
s ol
d, w
ith o
ther
maj
orB
ayne
s et
al.
Dep
osits
con
side
red
to b
e pa
rtfo
r su
bseq
uent
ly ‘w
orki
ng o
ver’
stra
tigra
phic
uni
ts a
s 19
,000
–(1
975)
of t
he m
oder
n fa
una
(occ
urrin
gth
e de
posi
ts.
25,0
00 y
ears
old
and
the
in t
he a
rea
imm
edia
tely
pre
botto
m d
epos
it m
ore
than
Eur
opea
n ar
rival
)31
,000
yea
rs o
ld.
Sub
sequ
ent
datin
g te
chni
ques
sup
port
edLu
ndel
ius
(196
0)B
oth
leve
lsth
ese
date
s fo
r up
per
laye
rs a
ndin
dica
ted
low
er l
evel
s m
ay b
e in
Coo
k (1
960)
exce
ss o
f 40
,000
yea
rs o
ld(T
urne
y et
al.
2001
).
44 P.J. de Torres et al.
Str
ong’
s C
ave,
or
7 m
iles
sout
h of
As
for
Mam
mot
h C
ave
has
aC
entr
al p
ortio
n of
tal
us s
lope
Coo
k (1
963)
Teet
h on
lyS
tron
gs’ C
ave
ofM
amm
oth
Cav
eco
mpl
ex d
evel
opm
ent
has
mor
e re
cent
, yo
ung
depo
sits
,C
ook
(196
3)hi
stor
y. M
ain
depo
sits
in
but
exca
vatio
ns f
rom
str
eam
bed
Mer
rilee
s (1
968)
entr
ance
cha
mbe
r. H
asco
nsid
ered
old
er,
but
youn
ger
talu
s sl
ope
unde
rmin
ed b
yth
an M
amm
oth
Cav
e. S
ome
ofst
ream
act
ion
& c
entr
alM
amm
oth
Cav
e de
posi
ts n
otpo
rtio
n of
tal
us c
olla
psin
g.pr
esen
t in
Str
ong’
s C
ave
also
Mat
eria
l w
ashe
d an
dim
ply
Str
ong’
s C
ave
depo
sits
tran
spor
ted
by s
trea
m a
ndm
ay b
e yo
unge
rth
en m
ixed
with
sed
imen
tary
depo
sits
Two
smal
lN
ear
Turn
er B
rook
,S
urfa
ce/s
ub-s
urfa
ce c
ave
Dep
osits
inc
lude
d un
wor
n43
0 ±
160
BP
bas
ed o
n ra
dio
Arc
her
and
Pre
sent
in
Cav
e 3
only
un-n
amed
cav
esA
ugus
tade
posi
ts w
ithin
8cm
of
Sar
coph
ilus
teet
h w
ith p
oorly
for
med
carb
on d
ated
hai
r fr
om C
ave
1.B
ayne
s (1
972)
(Cav
e 1
and
surf
ace
(Cav
e 1)
and
13c
mro
ots.
Int
erpr
eted
as
juve
nile
ani
mal
s.C
ave
3 m
ay b
e ol
der.
Nei
ther
Cav
e 3)
of s
urfa
ce (
Cav
e 3)
Thi
s, c
ombi
ned
with
kno
wn
cave
had
dep
osits
of
the
geog
raph
ic r
ange
of
med
ium
siz
ein
trod
uced
mam
mal
s (h
ouse
owl
spec
ies
and
the
inac
cess
ibili
ty o
fm
ouse
, bl
ack
rat
or r
abbi
t),
all o
fca
ves,
sug
gest
s de
posi
ts a
res
ult
ofw
hich
may
hav
e ar
rived
pos
tow
l pr
edat
ion.
accu
mul
atio
n of
dep
osits
Brid
e’s
Cav
eM
arga
ret
Riv
er a
rea,
not
desc
ribed
not
give
nG
laue
rt (
1948
)so
uth
wes
t W
este
rnA
ustr
alia
Sku
ll C
ave
Cap
e Le
euw
in r
egio
n of
Com
pris
ed o
f a
larg
e m
ain
May
hav
e fu
nctio
ned
as a
pit
trap
Upp
er l
ayer
(21
–28c
m)
aged
at
Por
ter
(197
9)P
rese
nt a
t de
pths
to
170c
mso
uth
wes
t W
este
rnch
ambe
r w
ith p
artia
llyw
ith a
dditi
onal
dep
osits
res
ultin
g2,
900
± 80
yea
rs B
PA
ustr
alia
colla
psed
roo
f an
d a
smal
ler
from
use
by
owls
Inte
rmed
iate
lay
er (
100–
115c
m)
cham
ber
lead
ing
from
7,87
5 ±
100
year
s B
P. D
epth
of
wes
tern
par
t of
mai
n19
0cm
con
side
red
late
cham
ber.
Bot
h ch
ambe
rsP
leis
toce
nepa
rtia
lly f
illed
with
san
dyse
dim
ents
, po
ssib
ly w
ashe
d in
Sco
tt R
iver
Coa
stal
dun
es s
outh
of
Exp
osed
fos
sil
soils
and
not
give
n, p
rese
nce
of f
ox a
ndB
utle
r (1
969)
the
Sco
tt R
iver
surf
ace
depo
sits
fro
mra
bbit
rem
ains
impl
y de
posi
t is
mob
ile-p
artly
sta
bilis
edm
ix o
f fo
ssil
and
mod
ern
anim
alsa
nd d
unes
rem
ains
Sco
tt R
iver
Bet
wee
n Le
dge
Poi
ntB
ones
col
lect
ed f
rom
colle
cted
in
perio
d 12
Mar
ch 1
976
not
date
d, p
resu
mab
ly s
ub f
ossi
lK
abay
and
Sta
rtP
resu
mab
ly s
ub f
ossi
l or
rec
ent
Are
a *
and
Bla
ck P
oint
sand
dun
esto
22
Mar
ch 1
976
or r
ecen
t. P
rese
nce
of r
abbi
t(1
976)
rem
ains
with
oth
er s
peci
men
sim
plie
s th
ese
are
rece
nt (
part
of
the
mod
ern
faun
a)
Tabl
e 1
(co
nt.)
Site
nam
eLo
catio
nD
escr
iptio
nC
omm
ent
Age
Sou
rce
Com
men
t on
S
eton
ix r
ecor
d
Distribution of the quokka 45
War
ren
Riv
er /
Des
crib
ed a
s bo
unde
dS
peci
men
s co
llect
ed f
rom
colle
cted
in
perio
d 27
to
31 M
ayno
t da
ted,
pre
sum
ably
sub
Kab
ay a
nd S
tart
Pre
sum
ably
sub
fos
sil
or r
ecen
tD
onne
lly R
iver
by t
he W
arre
n an
dsa
nd d
unes
19
76
foss
il or
rec
ent
(197
6)a
rea
Don
nelly
riv
ers,
the
sout
h co
ast
and
Vas
seH
ighw
ay.
Bon
e m
ater
ial
from
Yea
geru
p S
and
Du
ne
s
Nor
man
’sS
and
blow
out
sout
h of
desc
ribed
onl
y as
bon
e(s)
colle
cted
in
perio
d 15
to
20no
t da
ted,
pre
sum
ably
sub
Kab
ay a
nd S
tart
Pre
sum
ably
sub
fos
sil
or r
ecen
tB
each
/P
lant
agen
et l
ocat
ion
colle
cted
Aug
ust
1976
foss
il or
rec
ent
(197
6)P
lant
agen
et43
48 a
nd 1
km
eas
t of
Nor
man
’s B
each
Wal
pole
Des
crib
ed a
s co
asta
lS
peci
es l
iste
d as
obt
aine
dco
llect
ed i
n pe
riod
15 t
o 20
not
date
d, h
owev
er t
heK
abay
and
Sta
rtP
resu
mab
ly s
ub f
ossi
l or
rec
ent
sand
blo
w o
ut,
in s
and
dune
sA
ugus
t 19
76pr
esen
ce o
f ra
bbit
and
fox
(197
6)W
alpo
le a
rea
rem
ains
with
oth
er s
peci
men
sco
llect
ed i
mpl
ies
thes
e ar
ere
cent
dep
osits
(pa
rt o
f th
em
oder
n fa
una)
Two
Peo
ples
Bay
Eas
t of
Alb
any
Spe
cies
lis
ted
as o
btai
ned
colle
cted
in
perio
d 23
Apr
il 19
76 t
o 5
not
date
d, p
resu
mab
ly s
ubK
abay
and
Sta
rtP
resu
mab
ly s
ub f
ossi
l or
rec
ent
Nat
ure
Res
erve
in s
and
dune
sM
ay 1
976
and
15 t
o 20
Aug
ust
1976
foss
il or
rec
ent.
Pre
senc
e of
(197
6)ra
bbit
rem
ains
with
oth
ersp
ecim
ens
impl
ies
thes
e ar
ere
cent
(pa
rt o
f th
e m
oder
nfa
una)
Pla
ntag
enet
San
d du
nes,
sou
th o
fde
scrib
ed o
nly
as b
one(
s)co
llect
ed i
n pe
riod
15 t
o 20
not
date
d, p
resu
mab
ly s
ubK
abay
and
Sta
rtP
resu
mab
ly s
ub f
ossi
l or
rec
ent
Pla
ntag
enet
Loc
atio
nco
llect
edA
ugus
t 19
76fo
ssil
or r
ecen
t(1
976)
41
30
Will
iam
Bay
Spe
cies
lis
ted
as m
ater
ial
colle
cted
in
perio
d 4
to 1
0no
t da
ted,
pre
sum
ably
sub
Kab
ay a
nd S
tart
Pre
sum
ably
sub
fos
sil
or r
ecen
tN
atio
nal
Par
kpi
cked
up
in s
and
dune
sF
ebru
ary
1976
foss
il or
rec
ent
(197
6)
Boa
t Har
bour
San
d bl
owou
t at
Boa
tde
scrib
ed o
nly
as b
one(
s)co
llect
ed i
n pe
riod
15 t
o 20
not
date
d, p
resu
mab
ly s
ubK
abay
and
Sta
rtP
resu
mab
ly s
ub f
ossi
l or
rec
ent
Har
bour
eas
t of
colle
cted
Aug
ust
1976
foss
il or
rec
ent
(197
6)R
eser
ve 7
723
46P.J. de Torres et al.
Table 2
Australian Museum records of the quokka, Setonix brachyurus. Additional records by Masters lacking location records are not shown.
Year of Location Collector Form of Comments Australian Museumcollection specimens Registration number
Undated Rottnest Island brain in Formalin M 18429
Undated Rottnest Island G. P. Whitley-Staff skull and mandibles Shown as G. P. Whitney-Staff in Australian Museum database S 1936
Undated Rottnest Island G. P. Whitley-Staff skull only Shown as G. P. Whitney-Staff in Australian Museum database S 1937
Undated Rottnest Island G. P. Whitley-Staff skull, odd mandible Shown as G. P. Whitney-Staff in Australian Museum database S 1938
Undated Rottnest Island Taronga Park Trust skin M 7994
Undated Rottnest Island Taronga Park Zoo skin skull M 8177
1866 King George’s Sound Masters mount Year of collection nominally shown here as 1866, as Masters is known to have P 1048collected in King George’s Sound area from Jan to April 1866 (Abbott 1999; Glauert1950) and Sept 1868 to April 1869 (Glauert 1950), however this record not listed byKrefft (1867; 1869). No co-ordinates listed, co-ordinates for Albany used, howeverMasters known to have collected as far north as the Pallinup River (Salt River)(Glauert 1950) and the Stirling Range (Abbott 1999)
1866 King George’s Sound Masters mount See comment for record P 1048 P 1049
1866 King George’s Sound Masters spirit See comment for record P 1048 P 1060
1866 King George’s Sound Masters spirit See comment for record P 1048 P 1061
1866 King George’s Sound Masters spirit See comment for record P 1048 P 1062
1920 Nornalup near Denmark A. S. Le Souef skin skull Year of collection not listed. Nominally listed here 1920 as Le Souef known to have M 4212collected in Porongurup area in 1920 (Abbott 1999). Co-ordinates map to SouthernOcean, directly south of Nornalup Inlet and Walpole, corrected to map near Nornalup
1920 Nornalup, near Denmark A. S. Le Souef skull only No co-ordinates with this record, corrected co-ordinates for Le Souef record (M 4212) S 1799used. Other comments for record M 4212 apply.
1920 Nornalup, near Denmark A. S. Le Souef skull only No co-ordinates with this record, corrected co-ordinates for Le Souef record (M 4212) S 1800used. Other comments for record M 4212 apply.
1921 King River, 10 miles E. Le G. Troughton skin skull Shown as LE G. Troughton in Australian Museum database M 3083from Albany J. Wright
1921 King River, 10 miles E. Le G. Troughton skin skull Shown as LE G. Troughton in Australian Museum database M 3084from Albany J. Wright
1921 King River, 10 miles E. Le G. Troughton Shown as LE G. Troughton in Australian Museum database M 3085from Albany J. Wright
Distribution of the quokka 47
19
21
Kin
g R
iver
, 10
mile
sE
. Le
G.
Trou
ghto
nsk
in s
kull
Sho
wn
as L
E G
. Tr
ough
ton
in A
ustr
alia
n M
useu
m d
atab
ase
M 3
086
from
Alb
any
J. W
right
19
21
Prin
cess
Roy
al B
ay,
E.
Le G
. Tr
ough
ton
skin
sku
llS
how
n as
LE
G.
Trou
ghto
n in
Aus
tral
ian
Mus
eum
dat
abas
e. C
o-or
dina
tes
M 3
087
Alb
an
yJ.
Wrig
ht(3
5 de
g 3
min
Sou
th a
nd 1
17 d
eg 5
3min
Eas
t) m
ap t
o w
ithin
Prin
cess
Roy
alH
arbo
ur,
corr
ecte
d to
map
to
map
on
Prin
cess
Roy
al H
arbo
ur f
ores
hore
nea
rA
lban
y to
wns
ite
19
35
Rot
tnes
t Is
land
S.
Larn
ach
spiri
tM
182
14
19
35
Rot
tnes
t Is
land
S.
Larn
ach
spiri
tM
182
15
19
35
Rot
tnes
t Is
land
S.
Larn
ach
spiri
tM
182
16
19
35
Rot
tnes
t Is
land
S.
Larn
ach
spiri
tM
182
17
19
35
Rot
tnes
t Is
land
S.
Larn
ach
spiri
tM
182
18
19
35
Rot
tnes
t Is
land
S.
Larn
ach
spiri
tM
182
19
19
61
Rot
tnes
t Is
land
Taro
nga
Par
ksk
in s
kull
M 9
056
19
87
edge
of
Lake
Bag
dad,
Gav
in G
atte
nby
ske
leto
nM
186
26R
ottn
est
Isla
nd
19
97
Rot
tnes
t Is
land
E.
Dov
eysk
ull
M 3
3146
48P.J. de Torres et al.
Table 3
Published records of occurrence of the quokka, Setonix brachyurus. Records are listed chronologically.
Year of Location or Source Commentrecord Site name
1658 Rottnest Island Samuel Volckersen (Volckertzoon), Considered the first record of quokka sighting by a European and only the second record of a marsupial from Australiacited by (Alexander 1914) by a EuropeanGlauert (1950)
1696 Rottnest Island Willem de Vlamingh, cited by Mistook the quokka on Rottnest island for ‘ … A kind of rat as big as a common cat … ‘(Alexander 1914) Glauert (1950)
1801 Rottnest Island M. Freycinet,, cited by M. Freycinet, observed a quadruped ‘which the old Dutch navigators actually mistook for a rat’Alexander (1916)
1826 King George’s Sound Records of Quoy and Gaimard Reported the new species, the specimen described as recently dead and ... ‘probably from disease’as cited by Alexander (1916)
1829 King George’s Sound Scott Nind, cited by (Alexander Scott Nind, medical officer at the King George’s Sound penal settlement recorded the presence of the quokka1916) and Glauert (1950)
1829 Rottnest Island Dr T.B. Wilson, cited by Dr T.B. Wilson, visitor to the Swan River Settlement, when visiting Rottnest Island recorded ‘the dogs caught two wallabi’Alexander (1918)
1830 King George Sound Walton (1988) The type specimen of Quoy and Gaimard
1837 Swan River, Perth George Gray, cited by Glauert (1950) reported Gray’s list of species from the Swan River. The list included the quokka Kitchener et al. (1978)
Glauert (1950) noted Gray’s reference to ‘Swan River’ may have included the Darling Range & the York & Avon River valleys. Nominallymapped as Piesse Brook
1842 Coastal areas Gilbert, cited by Gould (1863) Gilbert found the quokka abundant in all the swampy tracts which ‘skirt nearly the whole of Western Australia at a shortdistance from the sea’. Presumably this was referring to the south west of WA only. Gilbert was known to have visited WAand collected from this area in 1839–1840 and 1842–1843 Abbott (1999). Date nominally listed here as 1842 andlocations nominally shown as Albany and Margaret River area
1905 Bald Island Shortridge (1909) see note 1, below
1905 Rottnest Island Shortridge (1909) see note 1, below
1905 Margaret River, Burnside Shortridge (1909) see note 1, below
1905 Busselton, Yallingup Shortridge (1909) see note 1, below
1905 Albany, King River Shortridge (1909) see note 1, below
1905 Albany, King River Thomas (1906) Descriptions of collections made by Shortridge. Listed as collected during end of 1904 and during 1905. Nominally listedhere as 1905 and mapped as same location as Shortridge (1909) for King River
Distribution of the quokka 49
19
05
Alb
any,
Big
Gro
ve,
Tho
mas
(19
06)
Des
crip
tions
of
colle
ctio
ns m
ade
by S
hort
ridge
. Li
sted
as
colle
cted
dur
ing
end
of 1
904
and
durin
g 19
05.
Nom
inal
ly l
iste
dK
ing
Geo
rge’
s S
ound
here
as
1905
and
nom
inal
ly m
appe
d as
the
bet
wee
n A
lban
y an
d K
ing
Riv
er
19
07
Mar
gare
t R
iver
are
aW
. H
. Loa
ring,
cite
d by
Quo
kkas
rec
orde
d as
ver
y nu
mer
ous
1907
–191
0. L
ocat
ion
nom
inal
ly m
appe
d as
cre
eklin
e in
for
este
d ar
ea,
imm
edia
tely
Whi
te (
1952
)w
est
of M
arga
ret
Riv
er
19
12
Irw
in I
nlet
– M
t F
rank
land
Dia
ries
of S
.W.
Jack
son,
Jack
son’
s di
arie
s fo
r 19
12–1
913
reco
rd ‘c
augh
t a
wal
laby
in
trap
’ and
sle
epin
g un
der
swor
d gr
ass
here
(D
eep
Riv
er)
are
aci
ted
by A
bbot
t (1
998)
inte
rpre
ted
by A
bbot
t as
ref
errin
g to
quo
kkas
. N
omin
ally
sho
wn
as 1
912
and
nom
inal
ly m
appe
d as
Dee
p R
iver
.
19
19
Low
er B
lack
woo
d V
alle
yP
erry
(19
71)
Rec
orde
d th
e qu
okka
as
num
erou
s an
d fr
eque
ntly
see
n in
the
Low
er B
lack
woo
d va
lley.
App
roxi
mat
e lo
catio
n on
ly
19
20
Mar
gare
t R
iver
Hoy
, ci
ted
by S
hort
and
From
col
lect
ions
by
Cha
rles
Hoy
, 19
19–1
922.
Hoy
col
lect
ed 9
spe
cim
ens
from
the
Mar
gare
t R
iver
are
a in
192
0. W
ith t
heC
alab
y (2
001)
exce
ptio
n of
Tric
hosu
rus
vulp
ecul
a, h
e de
scri
bed
none
of
the
colle
cted
mam
mal
s as
ple
ntifu
l. T
he q
uokk
a w
as d
escr
ibed
as s
eldo
m s
een
(due
to
noct
urna
l ha
bits
). L
ocat
ion
nom
inal
ly m
appe
d as
im
med
iate
ly n
orth
wes
t of
Mar
gare
t R
iver
19
20
Un-
nam
ed c
ave
near
Roe
(19
71)
Roe
(19
71)
reco
rded
thi
s lo
catio
n as
a f
ossi
l/sub
foss
il re
cord
and
not
ed l
ong
term
res
iden
ts w
ere
aw
are
quok
kas
(alo
ngG
ingi
nw
ith b
oodi
es,
Bet
tong
ia l
esue
ur,
and
bilb
ies,
Mac
rotis
lag
otis
) w
ere
pres
ent
in t
he d
istr
ict
until
arr
ival
of
the
fox
19
20
Bic
kley
, D
arlin
g S
carp
,W
. H
. Loa
ring,
cite
d by
Not
ed a
s pa
rtic
ular
ly p
lent
iful
in t
he e
arly
192
0s w
ith e
xten
sive
run
way
s in
thi
ck s
crub
bor
derin
g st
ream
sno
w c
onsi
dere
d a
subu
rbW
hite
(19
52)
of P
erth
19
20
Dar
ling
Ran
ge,
Pie
sse
W.H
. Loa
ring
inT
he q
uokk
a de
scrib
ed t
o ha
ve ‘v
anis
hed
from
its
gul
ly h
aunt
s …
30
year
s ag
o’. T
his
reco
rd i
s no
min
ally
sho
wn
asB
rook
- B
ickl
eyS
erve
nty
et a
l. (1
954)
pres
ent
here
30
year
s ea
rlier
(19
20)
and
loca
tion
map
ped
as P
iess
e B
rook
(P
iess
e G
ully
)
19
22
Yarl
oop,
Log
ue B
rook
,W
hite
(19
52)
Rec
orde
d as
num
erou
s in
low
tan
gled
scr
ub w
hen
auth
or w
as a
sch
oolb
oy a
nd s
till
plen
tiful
unt
il 19
26 w
hen
he l
eft
Dar
ling
Ran
ge b
etw
een
scho
ol.
App
roxi
mat
e lo
catio
n on
ly,
nom
inal
ly m
appe
d as
upp
er c
atch
men
t of
Log
ue B
rook
and
dat
e no
min
ally
sho
wn
1922
Per
th a
nd H
arve
y
19
29
Bu
sse
lton
Whi
te (
1952
)R
ecor
ded
as ‘
in g
reat
num
bers
’, lo
catio
n ap
prox
imat
e
19
29
Cap
e Le
euw
in –
Cap
eW
hite
(19
52)
Rec
orde
d as
‘in
gre
at n
umbe
rs’,
loca
tion
appr
oxim
ate
Nat
ural
iste
19
20
s/H
elen
a R
iver
, ne
arP
erry
(19
73)
Rec
orde
d as
abu
ndan
t in
den
se l
ow c
over
alo
ng c
reek
s an
d riv
er c
ours
es f
rom
val
ley
of H
elen
a R
iver
eas
twar
ds.
19
30
sM
unda
ring
Loca
tion
iden
tifie
d as
Gre
ysto
nes
pine
pla
ntat
ion
whi
ch w
as p
lant
ed b
y D
ick
Per
ry i
n th
e 19
20s/
1930
s an
d re
cord
assu
med
to
be f
rom
the
192
0s t
o 19
30s
(Ian
Abb
ott,
pers
. co
m.
to P
JdeT
).
19
30
Nor
thcl
iffe
area
Dau
bney
et
al.
(und
ated
)D
iscu
ssio
ns b
etw
een
one
of u
s (P
JdeT
) an
d lo
cal
land
hold
ers
plus
rep
orts
in
Dau
bney
et
al.
(und
ated
) in
dica
te t
hequ
okka
was
con
side
rabl
y m
ore
abun
dant
in
the
Nor
thcl
iffe
area
prio
r to
the
arr
ival
of
the
fox.
Gla
dys
Buc
king
ham
, sc
hool
teac
her
in N
orth
cliff
e, r
ecal
led
of t
he 1
930s
‘eve
ry s
wam
p w
as a
maz
e of
tra
cks
and
tunn
els
of w
alla
bies
, qu
okka
s,ba
ndic
oots
…’
(Dau
bney
et
al.
unda
ted)
. Lo
catio
n ap
prox
imat
e an
d m
appe
d as
im
med
iate
ly e
ast
of N
orth
cliff
e to
wns
ite,
date
her
e no
min
ally
lis
ted
as 1
930
19
31
Vas
se E
stua
ry,
near
Whi
te (
1952
)Q
uokk
as ‘i
n nu
mbe
rs’ i
n lo
w s
crub
bet
wee
n co
asta
l du
nes,
loc
atio
n ap
prox
imat
eB
uss
elto
n
19
32
Pin
e pl
anta
tions
,S
tew
art
(193
6)R
ecor
ded
as r
espo
nsib
le f
or d
amag
e to
pin
e pl
anta
tions
with
in k
arri
fore
st a
reas
and
rec
orde
d gr
azin
g up
to
2 km
s fr
omP
em
be
rto
nsw
amps
. Lo
catio
n m
appe
d he
re n
omin
ally
as
pine
pla
ntat
ions
, im
med
iate
ly e
ast
of P
embe
rton
19
32
Rot
tnes
t Is
land
Sto
rr (
1963
)S
torr
cite
d H
.T.P
ears
e, a
s th
e fir
st r
efer
ence
(19
32)
to t
he l
arge
num
ber
of q
uokk
as o
n R
ottn
est
Isla
nd b
eing
reg
arde
das
a p
est
50 P.J. de Torres et al.
19
33
Com
men
ts o
n ge
nera
lG
laue
rt (
1933
)P
ublis
hed
to i
ndic
ate
the
pres
ent
rang
e (in
193
3) o
f m
arsu
pial
fau
na i
n W
este
rn A
ustr
alia
and
as
an u
pdat
e to
the
dist
ribu
tion
dist
ribut
ions
des
crib
ed b
y S
hort
ridge
(19
09).
Des
crib
ed d
istr
ibut
ion
of t
he q
uokk
a as
ext
endi
ng f
rom
Moo
re R
iver
to
the
sout
h co
ast
and
incl
usiv
e of
Rot
tnes
t Is
land
and
Bal
d Is
land
. N
o re
fere
nce
to t
he p
rese
nce
on t
he i
slan
ds o
ff E
sper
ance
as s
ugge
sted
by
Sho
rtrid
ge.
Des
crib
ed a
s ab
unda
nt i
n su
itabl
e sw
ampy
loc
aliti
es.
Nom
inal
ly m
appe
d he
re a
s A
lban
yan
d D
wel
lingu
p, H
olyo
ake
19
33
Bic
kley
, D
arlin
g S
carp
,W
. H
. Loa
ring,
cite
d by
Quo
kkas
stil
l pr
esen
t in
gul
lies
in 1
933–
34no
w c
onsi
dere
d a
subu
rbW
hite
(19
52)
of P
erth
19
33
Mar
gare
t R
iver
are
aW
. H
. Loa
ring,
cite
d by
Quo
kkas
rep
orte
d as
far
les
s pl
entif
ul t
han
prev
ious
ly r
ecor
ded
in 1
907–
10.
Loca
tion
nom
inal
ly m
appe
d as
for
190
7–W
hite
(19
52)
1910
and
as
cree
klin
e in
for
este
d ar
ea,
imm
edia
tely
wes
t of
Mar
gare
t R
iver
19
33
Can
al R
ocks
, ne
ar Y
allin
gup
Whi
te (
1952
)R
ecor
ded
as p
artic
ular
ly n
umer
ous
in 1
933,
stil
l pr
esen
t 4
year
s la
ter,
les
s no
ticea
ble
with
in a
few
yea
rs
19
54
Man
jimup
, P
erup
are
aA
.D.
Jone
s in
Ser
vent
y et
al.
Rec
ogni
sed
the
area
bou
nded
by
the
Tone
and
Per
up r
iver
s, n
orth
war
d to
lat
itude
34
degr
ees
12 m
inut
es S
outh
was
(195
4)ric
h in
mar
supi
als.
Dat
e no
min
ally
lis
ted
here
as
1954
and
loc
atio
n m
appe
d as
with
in t
he P
erup
For
est,
and
betw
een
Tone
& P
erup
riv
ers,
nor
thea
st o
f M
anjim
up
19
54
Man
jimup
, M
oral
lup
A.D
. Jo
nes
in S
erve
nty
et a
l.R
ecor
d re
port
ed t
o A
.D.
Jone
s (in
Ser
vent
y et
al.
1954
) as
a s
ight
ing
10 m
iles
nort
h of
Mor
dallu
p. P
resu
mab
ly t
his
is a
(195
4)re
fere
nce
to M
oral
lup,
whi
ch p
lace
s th
is r
ecor
d ap
prox
imat
ely
23km
nor
thea
st o
f M
anjim
up.
Dat
e no
min
ally
lis
ted
here
as
1954
19
54
Com
men
ts o
n ge
nera
lL.
Gla
uert
in
Ser
vent
y et
al.
Des
crib
ed a
s pl
entif
ul o
n R
ottn
est
Isla
nd a
nd ‘s
ome
isla
nds
off
the
sout
h co
ast’
and
in s
wam
py c
ount
ry i
n th
e lo
wer
dist
ribu
tion
(195
4)so
uth
wes
t. N
omin
ally
map
ped
here
as
Alb
any
and
the
Bro
ke I
nlet
(D
’Ent
reca
stea
ux)
area
s, t
he l
atte
r as
map
ped
byC
hris
tens
en e
t al
. (1
985)
19
54
Com
men
ts o
n ge
nera
lL.
Gla
uert
in
Ser
vent
y et
al.
Des
crib
ed a
s no
lon
ger
pres
ent
in t
he v
alle
ys o
f th
e D
arlin
g R
ange
. P
resu
mab
ly t
he ‘i
slan
ds’ r
efer
ence
is
refe
rrin
g to
Bal
ddi
strib
utio
n (c
ont)
fro
m L
.(1
954)
Isla
nd,
but
this
may
als
o be
a r
efer
ence
to
the
isla
nds
of E
sper
ance
whi
ch w
ere
othe
rwis
e re
ferr
ed t
o on
ly b
yG
laue
rt i
n S
even
tyS
hort
ridge
(19
09).
et a
l. (1
954)
19
54
Yarl
oop,
Dar
ling
Ran
geS
erve
nty
et a
l. (1
954)
Ser
vent
y no
ted
quok
kas
seen
occ
asio
nally
, w
ith p
lent
y of
sig
ns i
n al
l sw
amps
. N
omin
ally
rec
orde
d as
195
4. E
xact
betw
een
Per
th a
nd H
arve
ylo
catio
n un
clea
r, n
omin
ally
map
ped
as L
ogue
Bro
ok a
rea
and
the
sam
e lo
catio
n as
Whi
te (
1952
). S
erve
nty’
s re
port
appe
ars
to c
ontr
adic
t th
e re
port
s of
Gla
uert
and
Loa
ring,
bot
h al
so i
n S
erve
nty
et a
l. (1
954)
, w
ho b
elie
ved
quok
kas
had
vani
shed
fro
m g
ully
hau
nts
of t
he D
arlin
g R
ange
30
year
s pr
evio
usly
. H
owev
er,
the
area
ref
erre
d to
by
Ser
vent
y et
al.
(195
4) i
s 10
0km
Sou
th o
f th
e se
ctio
n of
Dar
ling
Ran
ge L
oarin
g an
d G
laue
rt m
ay h
ave
been
ref
errin
g to
.
19
54
Man
jimup
, ne
arS
erve
nty
et a
l. (1
954)
Rec
orde
d as
firs
t re
cord
for
10
year
s. N
omin
ally
rec
orde
d as
195
4, e
xact
loc
atio
n un
clea
r. P
ossi
bly
refe
rrin
g to
the
‘B
roke
Bro
ke I
nlet
tur
noff
Inle
t tu
rn-o
ff’ f
rom
the
Sou
th W
este
rn H
ighw
ay (
Man
jimup
-Wal
pole
Hig
hway
), m
appe
d at
thi
s lo
catio
n
19
54
Tool
brun
up,
Stir
ling
Sha
rman
(19
54)
Exa
min
ed q
uokk
a ch
rom
osom
e nu
mbe
r an
d ur
ogen
ital
syst
em a
nd c
ompa
red
with
rel
ated
spe
cies
. N
oted
quo
kkas
Ra
ng
es
form
erly
wid
ely
dist
ribut
ed,
now
com
mon
onl
y on
Rot
tnes
t an
d B
ald
isla
nds.
Not
ed a
sku
ll co
llect
ed f
rom
Stir
ling
Ran
ges.
Loca
tion
nom
inal
ly s
how
n as
Blu
ff K
noll,
Stir
ling
Ran
ge.
Tabl
e 3
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 51
19
55
Wal
pole
J.A
. R
ate,
cite
d by
Cap
ture
of
an i
mm
atur
e qu
okka
. A
ppro
xim
ate
loca
tion
only
, no
min
ally
map
ped
as w
ithin
Wal
pole
-Nor
nalu
p N
atio
nal
Par
kB
arke
r et
al.
(195
7)
19
56
Rot
tnes
t Is
land
Dun
net
(196
3)E
xam
ined
quo
kka
sub
popu
latio
ns a
t B
agda
d an
d S
erpe
ntin
e so
aks
at e
aste
rn/c
entr
al a
rea
of R
ottn
est
Isla
nd i
n th
epe
riod
1954
–195
8. N
omin
ally
lis
ted
here
as
1956
, se
e al
so D
unne
t (1
962)
. E
stim
ates
of
popu
latio
n si
ze w
ere
varia
ble
and
diffe
red
betw
een
and
with
in s
ampl
ing
perio
ds.
Dun
net
caut
ione
d th
ese
estim
ates
wer
e al
so s
ubje
ct t
o bi
as
19
56
Alb
any
Hig
hway
, ne
arR
. A
itken
, ci
ted
by B
arke
rTw
o ro
adki
ll re
cord
s. L
ocat
ion
appr
oxim
ate
only
and
ide
ntifi
ed f
rom
map
in
Sad
leir
(195
9).
See
196
6 re
cord
by
Ken
tTr
avel
lers
’ Arm
set
al.
(195
7)W
illia
ms
(Tab
le 4
) an
d 19
96 r
ecor
d by
de
Tore
s, D
illon
, To
mki
nson
and
Bue
hrig
(Ta
ble
6)(s
outh
east
of
Per
th)
19
56
Rot
tnes
t Is
land
Shi
eld
(195
9)45
0 qu
okka
s sa
mpl
ed i
n 19
56–5
7 as
par
t of
stu
dy a
sses
sing
whe
ther
quo
kka
popu
latio
n w
as l
imite
d by
env
ironm
enta
l/ph
ysio
logi
cal
fact
ors.
No
estim
ate
of t
he s
ize
of t
he R
ottn
est
popu
latio
n
19
56
Rot
tnes
t Is
land
War
ing
(195
6)N
oted
quo
kkas
kno
wn
from
tw
o of
f-sh
ore
isla
nds
only
(R
ottn
est
and
Bal
d) a
nd c
onsi
dere
d th
e on
ly m
ainl
and
popu
latio
nto
be
‘a r
emna
nt p
opul
atio
n in
Kar
ri fo
rest
’ in
the
sout
h-w
est.
Ref
erre
d to
the
Rot
tnes
t po
pula
tion
as c
onsi
stin
g of
‘per
haps
5,0
00 i
ndiv
idua
ls’.
19
57
Byf
ord,
Man
jeda
l B
rook
Bar
ker
et a
l. (1
957)
Thr
ee q
uokk
as t
rapp
ed i
n th
e pe
riod
8 M
ay t
o 6
June
195
7. T
rapp
ing
surv
ey i
nitia
ted
in r
espo
nse
to a
new
spap
er a
rtic
leon
the
sou
thea
st e
dge
ofw
hich
sug
gest
ed q
uokk
as h
ad b
ecom
e ex
tinct
on
the
mai
nlan
d.th
e P
erth
met
ropo
litan
are
a
19
58
Rot
tnes
t Is
land
Her
rick
(196
1)E
xam
ined
adr
enal
fun
ctio
n of
the
quo
kka
and
incl
uded
ani
mal
s co
llect
ed f
rom
Rot
tnes
t Is
land
(S
umm
er 1
958)
and
obse
rvat
ions
fro
m t
wo
site
s on
Rot
tnes
t
19
58
Man
jeda
l B
rook
, no
rth
Sto
rr (
1964
b)S
tudy
of
quok
ka h
abita
t, no
est
imat
es o
f po
pula
tion
size
. P
resu
mab
ly t
he s
ite(s
) us
ed b
y S
adle
ir (1
959)
and
of J
arra
hdal
eB
arke
r et
al.
(195
7)
19
58
Rot
tnes
t Is
land
Sto
rr (
1964
a)N
utrit
iona
l st
udy,
511
quo
kkas
cau
ght
over
uns
peci
fied
num
ber
of m
onth
ly v
isits
in
1958
. N
o es
timat
e of
pop
ulat
ion
size
19
59
Rot
tnes
t Is
land
Hol
swor
th (
1967
)E
xam
ined
hom
e ra
nge
and
terr
itory
use
in
the
perio
d 19
54 t
o 19
64.
Nom
inal
ly s
how
n he
re a
s 19
59.
No
estim
ate
ofpo
pula
tion
size
, ho
wev
er s
how
ed v
aria
tion
in t
he n
umbe
r of
ind
ivid
ual
anim
als
caug
ht f
rom
38
in 1
960
to 8
13 i
n 19
63
19
59
Bal
d Is
land
Sto
rr (
1965
)R
ecor
ded
quok
kas
as a
bund
ant,
occu
rrin
g fr
om s
ea l
evel
to
the
peak
, bu
t de
nsity
var
ied
grea
tly
19
59
Way
chin
icup
Sto
rr (
1965
)N
one
trap
ped
or s
nare
d, a
bund
ant
evid
ence
in
‘sw
ampy
val
leys
dra
inin
g in
to t
he l
ower
Way
chin
icup
’. E
xact
loc
atio
n no
tgi
ven,
map
ped
as a
ppro
xim
ate
loca
tion
only
.
19
63
Rot
tnes
t Is
land
Pac
ker
(196
5)O
bser
vatio
nal
stud
y in
196
3, n
o es
timat
e of
pop
ulat
ion
size
19
67
Rot
tnes
t Is
land
Nic
holls
(19
71)
Cau
ght
and
mon
itore
d 27
quo
kkas
fro
m W
est
End
of
Rot
tnes
t Is
land
to
dete
rmin
e ho
me
rang
e an
d m
ovem
ents
,19
67–1
968.
No
estim
ate
of p
opul
atio
n/su
b-po
pula
tion
size
19
69
Rot
tnes
t Is
land
Kitc
hene
r (1
973)
Rec
orde
d an
ave
rage
of
53 r
esid
ent
quok
kas
at t
he B
arke
rs S
wam
p st
udy
site
, w
ith a
vera
ge d
ensi
ty o
f 1
quok
ka p
er0.
06ha
and
not
ed t
his
was
con
side
rabl
y gr
eate
r th
an t
he a
vera
ge o
f 1
quok
ka p
er 0
.4 –
1.2
ha r
ecor
ded
by M
ain
and
Yada
v (1
971)
. T
he l
ocal
ised
hig
h de
nsity
at
Bar
kers
Sw
amp
was
con
side
red
to r
efle
ct t
he l
ocal
ised
hig
h qu
ality
site
and
the
high
den
sity
was
mai
ntai
ned
by ‘
recr
uitm
ent
of c
onsi
dera
ble
num
ber
of q
uokk
as i
nto
the
loca
l po
pula
tion’
19
70
Rot
tnes
t Is
land
Kitc
hene
r (1
972)
Rep
orte
d on
obs
erve
d be
havi
ours
, in
tera
ctio
ns a
nd d
omin
ance
of
sub
popu
latio
n at
Bar
kers
Sw
amp.
Sub
pop
ulat
ion
estim
ated
by
dire
ct c
ount
dur
ing
noct
urna
l ob
serv
atio
ns.
How
ever
, su
bpop
ulat
ion
estim
ate
not
repo
rted
52 P.J. de Torres et al.
19
70
Dar
ling
Ran
ge,
clos
eR
ide
(197
0)R
ecor
ded
as r
are
on t
he m
ainl
and,
kno
wn
only
fro
m a
few
sw
ampy
val
leys
in
the
Dar
ling
Ran
ge.
Nom
inal
ly s
how
n he
reto
Per
thas
Dw
ellin
gup
(Hol
yoak
e) w
here
kno
wn
to o
ccur
in
the
1970
s
19
71
Com
men
ts o
n ge
nera
lC
alab
y (1
971)
Cal
aby
note
d th
e qu
okka
was
kno
wn
from
a f
ew s
wam
py a
reas
on
the
mai
nlan
d an
d w
as s
till
com
mon
on
Rot
tnes
t an
ddi
stri
butio
nB
ald
isla
nds.
Nom
inal
ly m
appe
d as
Dw
ellin
gup,
Hol
yoak
e
19
71
Rot
tnes
t Is
land
Mai
n an
d Ya
dav
(197
1)R
evie
wed
the
con
serv
atio
n in
form
atio
n fo
r m
acro
pods
fro
m B
arro
w I
slan
d an
d co
mpa
red
this
to
othe
r is
land
s. G
ave
anes
timat
e of
quo
kka
popu
latio
n de
nsity
for
Rot
tnes
t Is
land
as
rang
ing
from
1 q
uokk
a pe
r 1.
2ha
to 1
per
0.4
ha.
How
ever
,no
dat
a su
pplie
d to
sho
w h
ow t
hese
val
ues
wer
e de
rived
and
den
sity
may
hav
e be
en d
eter
min
ed o
n ba
sis
of t
he e
ntire
isla
nd b
eing
occ
upie
d. D
ate
nom
inal
ly s
how
n he
re a
s 19
71
19
71
Bal
d Is
land
Mai
n an
d Ya
dav
(197
1)G
ave
an e
stim
ate
of q
uokk
a po
pula
tion
size
at
Bal
d Is
land
as
rang
ing
from
600
–190
0 qu
okka
s (if
ent
ire i
slan
d w
asoc
cupi
ed).
Qua
lifie
d, i
n re
cogn
ition
not
all
of t
he i
slan
d w
as o
ccup
ied,
to
give
an
uppe
r es
timat
e of
600
. D
ate
nom
inal
lylis
ted
here
as
1971
19
71
Dw
ellin
gup
Sch
mid
t an
d M
ason
(19
73)
Rec
orde
d a
suite
of
mam
mal
spe
cies
in
fore
st n
ear
Dw
ellin
gup
in 1
971
and
conc
lude
d po
pula
tions
wer
e co
ncen
trat
ed i
nsw
amps
and
oth
er a
reas
of
dens
e ve
geta
tion.
Quo
kkas
inc
lude
d in
lis
t of
spe
cies
rec
orde
d, n
o si
te d
etai
ls,
nom
inal
lym
appe
d he
re a
s H
olyo
ake
19
72
Dw
ellin
gup,
Wre
ns R
oad
Chr
iste
nsen
and
Kim
ber
(197
5)P
opul
atio
n tr
appe
d (1
972–
74)
befo
re a
nd a
fter
a pa
rtia
l bu
rn.
Trap
suc
cess
rat
e in
crea
sed
afte
r th
e bu
rn.
Indi
ces
toS
wam
p (e
ast)
abun
danc
e im
ply
mod
erat
e si
ze p
opul
atio
n, h
owev
er n
o es
timat
e of
pop
ulat
ion
size
. D
ate
assu
med
to
be 1
972,
see
Chr
iste
nsen
(un
date
d) i
n ta
ble
of u
npub
lishe
d re
cord
s
19
72
Alb
any
Hig
hway
,C
rabb
(19
73)
Roa
dkill
, ne
ar t
he ‘2
7 m
ile p
eg o
n A
lban
y H
ighw
ay’
sout
heas
t of
Per
th
19
74
Dw
ellin
gup,
Dun
cans
Chr
iste
nsen
and
Kim
ber
(197
5)P
opul
atio
n tr
appe
d 19
72–7
4. T
his
site
unb
urnt
, cf
Wre
ns R
oad.
Tra
p su
cces
s co
mp
arab
le t
o po
st b
urn
capt
ure
succ
ess
Roa
d S
wam
pat
Wre
ns R
oad.
Ind
ices
to
abun
danc
e im
ply
a m
oder
ate
size
pop
ulat
ion,
no
estim
ate
of p
opul
atio
n si
ze.
Dat
e no
min
ally
show
n as
197
4
19
75
Kar
ri, n
orth
wes
t of
Chr
iste
nsen
et
al.
(198
5)Li
sted
and
map
ped
by C
hris
tens
en e
t al
. (1
985)
as
Sur
vey
Are
a L,
Kar
ri. S
ee n
ote
2, b
elow
Nor
thcl
iffe
19
75
Yea
ga
rup,
be
twe
en
th
eC
hris
tens
en e
t al
. (1
985)
List
ed a
nd m
appe
d by
Chr
iste
nsen
et
al.
(198
5) a
s S
urve
y A
rea
A,
Yeag
arup
. S
ee n
ote
2, b
elow
Don
nelly
and
War
ren
river
s
19
75
Mitc
hell,
nor
th n
orth
wes
tC
hris
tens
en e
t al
. (1
985)
List
ed a
nd m
appe
d by
Chr
iste
nsen
et
al.
(198
5) a
s S
urve
y A
rea
I, M
itche
ll. S
ee n
ote
2, b
elow
of W
alpo
le
19
75
Soh
o, n
orth
east
of
Wal
pole
Chr
iste
nsen
et
al.
(198
5)Li
sted
and
map
ped
by C
hris
tens
en e
t al
. (1
985)
as
Sur
vey
Are
a H
, S
oho.
See
not
e 2,
bel
ow
19
75
Bor
anup
, so
uth
ofC
hris
tens
en e
t al
. (1
985)
List
ed a
nd m
appe
d by
Chr
iste
nsen
et
al.
(198
5) a
s S
urve
y A
rea
D,
Bor
anup
. S
ee n
ote
2, b
elow
Mar
gare
t R
iver
Tabl
e 3
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 53
19
75
Gia
nts,
bet
wee
n N
orna
lup
Chr
iste
nsen
et
al.
(198
5)Li
sted
and
map
ped
by C
hris
tens
en e
t al
. (1
985)
as
Sur
vey
Are
a O
, G
iant
s. S
ee n
ote
2, b
elow
Inle
t an
d Ir
win
Inl
et,
east
of
Nor
nalu
p
19
75
Dom
baku
p, w
est
Chr
iste
nsen
et
al.
(198
5)O
ne o
f tw
o si
tes
liste
d an
d m
appe
d by
Chr
iste
nsen
et
al.
(198
5) a
s S
urve
y A
rea
C,
Dom
baku
p. S
ee n
ote
2, b
elow
nort
hwes
t of
Nor
thcl
iffe
19
75
Woo
lbal
es,
sout
heas
t of
Chr
iste
nsen
et
al.
(198
5)Li
sted
and
map
ped
by C
hris
tens
en e
t al
. (1
985)
as
Sur
vey
Are
a B
, W
oolb
ales
. S
ee n
ote
2, b
elow
Bro
ke I
nlet
19
75
Dom
baku
p, w
est
of B
roke
Chr
iste
nsen
et
al.
(198
5)O
ne o
f tw
o si
tes
liste
d an
d m
appe
d by
Chr
iste
nsen
et
al.
(198
5) a
s S
urve
y A
rea
C,
Dom
baku
p. S
ee n
ote
2, b
elow
Inle
t (D
’Ent
reca
stea
uxN
atio
nal
Par
k)
19
75
Sun
klan
ds,
sout
h-C
hris
tens
en e
t al
. (1
985)
List
ed a
nd m
appe
d by
Chr
iste
nsen
et
al.
(198
5) a
s S
urve
y A
rea
E,
Sun
klan
ds.
See
not
e 2,
bel
owso
uthe
ast
of B
usse
lton
19
76
Nan
nup,
Mow
en/S
toat
sH
art
et a
l. (1
986)
Trap
ped
as p
art
of c
ompa
rativ
e st
udy
of S
alm
onel
la i
nfec
tion.
Im
plie
d to
be
at l
ow d
ensi
tyR
oad
Cro
ssin
g
19
76
Mud
dy L
ake,
sou
th o
fH
art
et a
l. (1
986)
Trap
ped
as p
art
of c
ompa
rativ
e st
udy
of S
alm
onel
la i
nfec
tion.
Im
plie
d to
be
at l
ow d
ensi
tyB
un
bu
ry
19
76
Dw
ellin
gup,
Hol
yoak
eH
art
et a
l. (1
986)
Rep
orte
d hi
gh t
rap
yiel
d (1
0 an
imal
s fr
om 1
00 t
rap
nigh
ts)
usin
g el
abor
ate
trap
s co
mpr
ised
of
fenc
e lin
es a
nden
clos
ures
, w
here
mul
tiple
tra
p ni
ghts
ass
umed
for
eac
h tr
ap
19
76
Byf
ord
Har
t et
al.
(198
6)Tr
appe
d as
par
t of
com
para
tive
stud
y of
Sal
mon
ella
inf
ectio
n. I
mpl
ied
to b
e at
low
den
sity
. P
resu
mab
ly s
ame
site
as
trap
ped
by B
arke
r et
al.
(195
7) a
nd S
adle
ir (1
959)
19
77
Dw
ellin
gup,
Hol
yoak
eH
art
et a
l. (1
986)
Rep
orte
d hi
gh t
rap
yiel
d (2
7 an
imal
s fr
om 2
04 t
rap
nigh
ts)
usin
g el
abor
ate
trap
s co
mpr
ised
of
fenc
e lin
es a
nden
clos
ures
, w
here
mul
tiple
tra
p ni
ghts
ass
umed
for
eac
h tr
ap
19
80
Bal
d is
land
Har
t et
al.
(198
6)14
quo
kkas
cau
ght
from
a s
ingl
e on
e-ni
ght
trip
. N
o es
timat
e of
abu
ndan
ce r
epor
ted,
how
ever
pop
ulat
ion
leve
l re
port
edto
flu
ctua
te a
s a
resu
lt of
sum
mer
sta
rvat
ion
19
83
Dar
ling
Sca
rpD
ell
(198
3)T
his
publ
icat
ion
prod
uced
to
upda
te t
he W
este
rn A
ustr
alia
n M
useu
m m
amm
al r
ecor
ds w
ith r
ecen
t ob
serv
atio
ns a
ndre
view
the
lite
ratu
re.
The
quo
kka
was
lis
ted
as o
ne o
f se
vera
l sp
ecie
s to
hav
e de
clin
ed.
Nom
inal
ly m
appe
d he
re a
sD
wel
lingu
p, U
rbra
e F
ores
t B
lock
19
83
Gre
en R
ange
, 60
kmK
irke
(198
3)Q
uokk
as r
epor
ted
as f
leei
ng f
rom
bur
ning
bus
hlan
d. S
kele
ton
subs
eque
ntly
col
lect
ed.
The
onl
y re
cord
at
this
loc
atio
nno
rthe
ast
of A
lban
ypr
ior
to t
his
was
fro
m P
earc
e (1
969)
, se
e 19
69 e
ntry
in
Tabl
e 4,
Unp
ublis
hed
Rec
ords
.
19
85
Way
chin
icup
Inl
et,
east
Har
t et
al.
(198
6)Tr
appe
d as
par
t of
com
para
tive
stud
y of
Sal
mon
ella
inf
ectio
n. P
opul
atio
n im
plie
d to
be
at l
ow d
ensi
tyof
Alb
any
19
85
Pt
D’E
ntre
cast
eaux
How
et
al.
(198
7)S
urve
y co
nduc
ted
in t
he p
erio
d M
arch
198
5, O
ctob
er t
o N
ovem
ber
1985
and
Jan
uary
to
Feb
ruar
y 19
86.
Nom
inal
lysh
own
here
as
1985
. Q
uokk
a re
port
ed f
rom
onl
y 1
of 1
0 lo
catio
ns s
urve
yed.
Thi
s re
cord
of
pres
ence
im
plie
d by
How
et
al.
(198
7) t
o be
que
stio
nabl
e. L
ocat
ion
appr
oxim
ate
only
54 P.J. de Torres et al.
19
85
Pem
bert
on a
nd D
wel
lingu
pM
ead
et a
l. (1
985)
Ani
mal
s co
llect
ed f
rom
mai
nlan
d (D
wel
lingu
p an
d P
embe
rton
) an
d is
land
s fo
r la
bora
tory
ass
essm
ent
of t
oler
ance
to
1080
. S
ampl
e si
ze n
ot s
peci
fied
and
no e
stim
ate
of p
opul
atio
n si
ze a
t co
llect
ion
poin
t. D
ate
nom
inal
ly s
how
n he
re a
s19
85.
Loca
tion
map
ped
nom
inal
ly a
s st
ate
fore
st,
imm
edia
tely
eas
t of
Pem
bert
on a
nd D
wel
lingu
p (H
olyo
ake)
19
85
Bal
d Is
land
Mea
d et
al.
(198
5)A
nim
als
colle
cted
for
lab
orat
ory
asse
ssm
ent
of t
oler
ance
to
1080
. S
ampl
e si
ze n
ot s
peci
fied
and
no e
stim
ate
ofpo
pula
tion
size
at
colle
ctio
n po
int.
Dat
e no
min
ally
sho
wn
here
as
1985
19
85
Rot
tnes
t Is
land
Mea
d et
al.
(198
5)A
nim
als
colle
cted
for
lab
orat
ory
asse
ssm
ent
of t
oler
ance
to
1080
. S
ampl
e si
ze n
ot s
peci
fied
and
no e
stim
ate
ofpo
pula
tion
size
at
colle
ctio
n po
int.
Dat
e no
min
ally
sho
wn
here
as
1985
20
00
Col
lie,
Vic
tor
Roa
d,H
ayw
ard
et a
l.(20
03)
Trap
ped
over
the
per
iod
1998
–200
0. P
opul
atio
n es
timat
ed t
o be
9 +
/- 1
(Jo
lly-S
eber
mar
k-re
capt
ure
deriv
ed e
stim
ates
)H
amilt
on F
ores
t B
lock
20
00
Col
lie/H
arve
y, H
adfie
ldH
ayw
ard
et a
l.(20
03)
Trap
ped
over
the
per
iod
1998
–200
0. P
opul
atio
n es
timat
ed t
o be
29
+/-
5 (
Jolly
-Seb
er m
ark-
reca
ptur
e de
rived
est
imat
es)
For
est
Blo
ck
20
00
Dw
ellin
gup,
Kes
ners
Hay
war
d et
al.(
2003
)Tr
appe
d ov
er t
he p
erio
d 19
98–2
000.
Pop
ulat
ion
estim
ated
to
be 3
6 +
/- 6
(Jo
lly-S
eber
mar
k-re
capt
ure
deriv
ed e
stim
ates
)S
wam
p, T
urne
r F
ores
tB
lock
20
00
Jarr
ahda
le,
Cha
ndle
r R
oad
Hay
war
d et
al.(
2003
)Tr
appe
d ov
er t
he p
erio
d 19
98–2
000.
Pop
ulat
ion
estim
ated
to
be 1
0 (J
olly
-Seb
er m
ark-
reca
ptur
e de
rived
est
imat
es)
Site
, C
hand
ler
For
est
Blo
ck
20
00
Jarr
ahda
le,
Ros
ella
Roa
d,H
ayw
ard
et a
l.(20
03)
Trap
ped
over
the
per
iod
1998
–200
0. N
o po
pula
tion
estim
ate,
onl
y on
e an
imal
tra
pped
Gor
don
For
est
Blo
ck
Not
es:
1.S
hort
ridge
col
lect
ed 3
8 sp
ecim
ens
and
desc
ribed
the
quo
kka
as p
lent
iful
amon
g co
asta
l th
icke
ts a
nd s
wam
ps o
f th
e so
uth
wes
t, no
t ex
tend
ing
inla
nd.
The
dis
trib
utio
n w
as r
ecor
ded
as e
xten
ding
from
the
Moo
re R
iver
in
the
nort
h, a
nd m
appe
d to
inc
lude
the
sou
th c
oast
as
far
east
as
Esp
eran
ce. T
he e
aste
rn e
xten
t ap
pear
s to
be
base
d of
rep
orts
fro
m T
win
Pea
k &
oth
er i
slan
ds o
ffE
sper
ance
. T
his
is t
he o
nly
repo
rt o
f qu
okka
occ
urre
nce
from
the
isl
ands
off
Esp
eran
ce a
nd n
o co
llect
ions
hav
e be
en d
ocum
ente
d fr
om t
hese
isl
ands
. S
hort
ridge
’s r
ecor
d is
ass
umed
to
bere
ferr
ing
to t
amm
ars,
see
tex
t.2.
Chr
iste
nsen
et
al.
(198
5) f
rom
sur
veys
197
0–19
82 n
oted
the
quo
kka
as w
ides
prea
d an
d lo
cally
com
mon
. E
stim
ates
of
popu
latio
n si
ze w
ere
not
prov
ided
and
loc
atio
ns a
re a
ppro
xim
ate
and
rem
appe
d fr
om m
ap o
n p4
of
Chr
iste
nsen
et
al.
(198
5).
Dat
e no
min
ally
sho
wn
as 1
975
Tabl
e 3
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka
55
Table 4
Unpublished records of occurrence of the quokka, Setonix brachyurus. Records are listed chronologically.
Year of Location or Source Commentrecord Site name
1940 Northcliffe, Bashford Road L. Wilson (personal communication Quokkas observed in large numbers in 1940s and earlier and known to feed in paddocks away from vegetatedarea, 10km east to PJdeT) creeklines. Reported as ‘last seen in number when fleeing from fire’ in the 1940s. Location approximate and shownof Northcliffe nominally as 1940
1958 Rottnest Island Sadleir (1959) Conducted a trapping program at the Byford/Manjedal Swamp site on the mainland and Rottnest Island to examinephysiological differences between the two populations
1958 Byford, Manjedal Sadleir (1959) Trapped at Manjedal Swamp in April – May and September – October 1958. Six quokkas only caught, all in theSwamp/Brook April – May period
1962 Rottnest Island Holsworth (1964) Trapped in period May 1961 to May 1963 and incorporated data from 1954 to 1961 and from December 1963 to April1964. Nominally shown here as 1962. Population estimates were derived for 4 areas within the West End only. Eacharea was comprised of several group territories. Population estimates were described by Holsworth as stable for theperiod 1955 to 1963, however data indicates large variations in abundance.
1965 Travellers’ Arms, Albany Kent Williams (pers. comm. Quokkas successfully trapped in 1965 at swamp on the opposite side of Albany highway from the Travellers’ Arms Hotel.Highway, south-east of the to PJdeT) Subsequent trapping in the same year, post fire was unable to confirm continued presence. See record for 1996Perth metropolitan area (Table 6) when presence at this location was re-confirmed
1969 Two Peoples Bay Nature Bannister (1970) Reported carcass found in 1969 by N. Robinson, in thick scrub at bottom of a shallow gully, near CSIRO hut. OldReserve, near CSIRO hut evidence and no indication of recent activity when surveyed by Bannister (1970). Location mapped here is approximate
1969 Rottnest Island Kitchener (1970) Estimated a mean monthly population size from a study area comprised of 48 quadrats, each 12m x 12m at BarkersSwamp, near the centre/eastern end of the Island. Estimates derived from data collected in the period December 1967to December 1969. Listed here as 1969. Estimates were from direct observations and included permanent andtemporary residents, recruits and transients. Mean monthly estimate was approximately 45 adults and 13 juveniles withinthe Barkers Swamp study site.
1969 Green Range, 50 miles Pearce (1969) Local landholder recorded the presence of small kangaroos. Subsequent investigation by WA Department of Fisheries(30km) northeast of Albany Fauna Warden detected spoor consistent with quokka. Drawing of spoor provided in Departmental file – we have
presumed this to be a quokka. Presence recorded here in 1983, see 1983 entry in Table 3.
1970 Two Peoples Bay Nature Bannister (1970) Unable to confirm by trapping, presence inferred by runs and scats detected in thickly vegetated gullies, near weatherReserve, near weather station above CSIRO hut. Location mapped here is approximatestation above CSIRO hut
1970 Two Peoples Bay Nature Bannister (1970) Unable to confirm by trapping, presence inferred by runs and scats detected in thickly vegetated gullies, bottom ofReserve, south-west of valley, south-west of Pt GardnerPt Gardner
1970 Two Peoples Bay Nature Bannister (1970) Unable to confirm by trapping, presence inferred by runs and scats detected in thickly vegetated gullies, thick swampyReserve, Moates Lagoon areas at west end of Moates Lagoon
56 P.J. de Torres et al.
19
72
Dw
ellin
gup,
Wre
ns R
oad
Chr
iste
nsen
(un
date
d)Tr
appi
ng p
rogr
am b
efor
e (A
ugus
t to
Nov
embe
r 19
72)
and
afte
r (1
2 to
21
Dec
embe
r 19
72)
a lo
w i
nten
sity
bur
n of
60%
of
Sw
amp
(wes
t),
Urb
rae
quok
ka h
abita
t w
ithin
the
sw
amp.
No
estim
ates
of
dens
ity,
how
ever
a t
otal
of
22 i
ndiv
idua
ls t
rapp
ed.
For
est
Blo
ck
19
72
Dw
ellin
gup,
Wre
ns R
oad
Chr
iste
nsen
(un
date
d)Tr
appi
ng p
rogr
am b
efor
e (A
ugus
t to
Nov
embe
r 19
72)
and
afte
r (1
2 to
21
Dec
embe
r 19
72)
a lo
w i
nten
sity
bur
n of
60%
of
Sw
amp
(eas
t),
Whi
tequ
okka
hab
itat
with
in t
he s
wam
p. N
o es
timat
es o
f de
nsity
, ho
wev
er a
tot
al o
f 22
ind
ivid
uals
tra
pped
.F
ores
t B
lock
19
75
Den
bark
er,
5km
eas
t of
Kab
ay a
nd S
tart
(19
76),
150
cage
tra
p ni
ghts
and
180
sna
re t
rap
nigh
ts o
ver
9 ni
ghts
(28
–30
July
197
5 an
d 11
–20
Aug
ust
1975
). S
ix q
uokk
asD
enba
rker
Roa
d br
idge
note
1,
belo
wtr
appe
d or
sna
red
over
Mitc
hell
Riv
er
19
75
Mou
nt M
anyp
eaks
, N
orth
Kab
ay a
nd S
tart
(19
76),
130
cage
tra
p ni
ghts
ove
r th
ree
nigh
ts p
lus
60 s
nare
tra
p ni
ghts
ove
r th
ree
nigh
ts (
5–7
July
197
5).
One
quo
kka
only
of M
ount
Man
ypea
ksno
te 1
, be
low
seen
, ru
nway
s an
d sc
ats
foun
d
19
75
Two
Peo
ples
Bay
,K
abay
and
Sta
rt (
1976
),18
0 ca
ge t
rap
nigh
ts o
ver
4 ni
ghts
(1–
8 Ju
ly 1
975)
. N
o qu
okka
cap
ture
s, b
ut p
rese
nce
infe
rred
by
pres
ence
of
scat
sM
t G
ardn
erno
te 1
, be
low
and
runw
ays
in g
ullie
s an
d in
den
se h
eath
bor
derin
g gu
llies
19
75
Nan
nup,
Mow
en R
oad
Kab
ay a
nd S
tart
(19
76),
Sur
veye
d Ju
ne 1
975
with
50
cage
tra
p ni
ghts
and
10
snar
e tr
ap n
ight
s ov
er o
ne n
ight
(in
per
iod
6–17
Jun
e 19
75).
cree
k cr
ossi
ng,
east
of
itsno
te 1
, be
low
Thr
ee q
uokk
as t
rapp
ed,
scat
s an
d ru
nway
s pr
esen
tju
nctio
n w
ith S
toat
s R
oad
19
76
Bal
d Is
land
Kab
ay a
nd S
tart
(19
76),
360
cage
tra
p ni
ghts
ove
r th
ree
nigh
ts (
29 M
ay –
3 J
une
1976
). O
ne q
uokk
a tr
appe
d, l
arge
num
ber
seen
and
a n
umbe
rno
te 1
, be
low
of s
kulls
col
lect
ed
19
76
Two
Peo
ples
Bay
,K
abay
and
Sta
rt (
1976
),36
0 ca
ge t
rap
nigh
ts o
ver
4–9
nigh
ts a
t a
rang
e of
loc
atio
ns (
13 A
pril
– 2
May
197
6). T
wo
juve
nile
quo
kkas
tra
pped
at
‘Rob
inso
n’s
Gul
ly’
note
1,
belo
w‘R
obin
son’
s G
ully
’. A
ppro
xim
ate
loca
tion
only
map
ped
here
19
76
Bun
bury
, M
uddy
Lak
e,K
abay
and
Sta
rt (
1976
),15
0 ca
ge t
rap
nigh
ts o
ver
thre
e ni
ghts
Jun
e 19
75 (
10–1
2 Ju
ne 1
975)
. Q
uokk
a sk
elet
al m
ater
ial
reco
vere
d, r
unw
ays
and
11km
Sou
th o
f B
unbu
ryno
te 1
, be
low
scat
s co
mm
on.
No
live
anim
als
trap
ped
in J
une
1975
, bu
t w
ere
seen
sub
sequ
ently
(4
Feb
197
6) b
y th
ese
inve
stig
ator
s.C
o-or
dina
tes
corr
ecte
d to
plo
t to
loc
atio
n of
Mud
dy L
ake
19
77
Dw
ellin
gup,
Hol
yoak
eH
art
(197
7)Tr
appe
d 10
ind
ivid
ual
quok
kas
over
the
per
iod
6–11
Nov
embe
r 19
76 a
nd 2
7 ov
er t
he p
erio
d 17
–24
Mar
ch 1
977
19
79
Jan
da
kot
Aus
tin (
1979
)E
stim
ated
pop
ulat
ion
size
of
a tr
ansl
ocat
ed,
fenc
ed p
opul
atio
n, w
here
393
quo
kkas
ex
Rot
tnes
t Is
land
had
bee
nre
leas
ed i
n th
e pe
riod
197
2–19
78.
Pop
ulat
ion
estim
ated
to
be a
ppro
xim
atel
y 84
ani
mal
s in
197
9. T
rans
loca
ted
popu
latio
n su
bseq
uent
ly s
how
n to
hav
e fa
iled
to p
ersi
st
19
82
Jarr
ahda
le (
Mun
dilu
pD
. G
iles
(per
sona
l co
mm
unic
atio
nR
ecor
d of
roa
dkill
(s)
repo
rted
fro
m t
he e
arly
198
0s (
nom
inal
ly 1
982)
by
loca
l la
ndho
lder
. R
epor
ted
to D
oug
Gile
s,F
ores
t B
lock
)to
PJd
eT)
Dep
artm
ent
of C
onse
rvat
ion
and
Land
Man
agem
ent,
Jarr
ahda
le
19
86
Stir
ling
Ran
ge N
atio
nal
A.N
. S
tart
, no
te 2
bel
owC
onfir
med
rec
ords
of
pres
ence
(si
ghtin
gs)
by A
llan
Ros
e in
the
per
iod
from
198
6 to
199
1. N
omin
ally
lis
ted
and
map
ped
Par
k –
Blu
ff K
noll
here
as
1986
, se
e al
so r
ecor
d fo
r 19
91(1
98
6–
19
91
)
Tabl
e 4
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 57
19
88
Wal
pole
-Nor
nalu
p N
atio
nal
K.
Gill
en,
note
2 b
elow
Rec
ords
of
scat
s an
d ru
nway
s (e
ast
of M
t H
opki
ns,
Kar
ri H
ill G
ully
eas
t of
Cry
stal
Lak
e an
d so
uth
of D
eep
Riv
er)
over
the
Par
k, N
uyts
Wild
erne
sspe
riod
June
198
6 to
Jun
e 19
88.
Nom
inal
ly s
how
n he
re a
s 19
88 a
nd m
appe
d at
eas
t of
Cry
stal
Lak
e
19
88
Mou
nt M
anyp
eaks
are
aK
. G
illen
, no
te 2
, be
low
Sev
eral
rec
ords
of
quok
ka s
cats
and
run
way
s no
ted
in p
erio
d S
epte
mbe
r 19
87 t
o S
epte
mbe
r 19
89.
Map
ped
loca
tion
isap
prox
imat
e on
ly a
nd l
iste
d as
198
8
19
89
Mou
nt F
rank
land
Nat
iona
lA
.N.
Sta
rt,
note
2 b
elow
Inci
dent
al s
ight
ing
by A
ndre
w M
orto
n. S
ight
ing
of o
ne a
nim
al o
nly
Par
k, M
iddl
e R
oad
19
89
Two
Peo
ples
Bay
Nat
ure
K.
Gill
en,
note
2 b
elow
Mul
tiple
rec
ords
of
quok
ka s
cats
and
run
way
s no
ted
in p
erio
d Ju
ne 1
985
to J
uly
1991
. T
hree
con
firm
ed s
ight
ings
in
July
Res
erve
, M
t G
ardn
er a
rea
1988
(Lo
wer
Rob
inso
n’s
Gul
ly/C
offin
Gul
ly),
Jul
y 19
89 (
Cof
fin G
ully
) an
d Ju
ly 1
990
(Rob
inso
n’s
Gul
ly/C
offin
Gul
ly)
19
89
Bal
d Is
land
K.
Gill
en,
note
2 b
elow
Rec
ord
of s
cats
and
sub
sequ
ent
quok
ka s
ight
ing
on r
ocks
nea
r w
este
rn s
hore
19
91
Stir
ling
Ran
ge N
atio
nal
Sm
ith,
R.
Car
cass
rec
orde
d by
Bob
Sm
ith (
CA
LM M
anjim
up),
via
Lac
hie
McC
aw (
CA
LM M
anjim
up).
Quo
kka
thou
ght
to b
eP
ark,
Suc
cess
Rid
ge T
rack
disp
lace
d as
a r
esul
t of
fire
19
91
Na
nn
up
, L
ew
an
aA
.N.
Sta
rt,
note
2 b
elow
Car
cass
of
juve
nile
fox
-kill
ed q
uokk
a co
llect
ed i
n pi
ne p
lant
atio
n. A
ppro
xim
ate
map
ping
of
loca
tion
only
. N
o de
tails
of
Pla
nta
tion
sour
ce p
opul
atio
n
19
91
Stir
ling
Ran
ge N
atio
nal
A.N
. S
tart
, no
te 2
bel
owC
onfir
med
rec
ords
of
pres
ence
(si
ghtin
gs)
by A
llan
Ros
e in
the
per
iod
from
198
6 to
199
1. N
omin
ally
lis
ted
here
as
1991
,P
ark
- B
luff
Kno
llse
e al
so 1
986
reco
rd a
bove
. A
rea
burn
t in
Apr
il 19
91 a
nd o
nly
one
reco
rd a
t th
is l
ocat
ion
sinc
e (s
ee 1
991
reco
rd b
elow
).(1
98
6–
19
91
)P
rese
nce
othe
rwis
e su
gges
ted
by s
cats
, se
e 19
95 r
ecor
d of
Bar
rett
(199
6).
19
91
Stir
ling
Ran
ge N
atio
nal
A.N
. S
tart
, no
te 2
bel
owC
arca
ss (
bone
s on
ly)
colle
cted
by
Alla
n R
ose.
Pre
sum
ed k
illed
as
a re
sult
of f
ireP
ark
- B
luff
Kno
ll (1
991)
19
91
Jarr
ahda
le,
Ros
ella
A.N
. S
tart
, no
te 2
bel
owP
rese
nce
repo
rted
fro
m s
cat
only
. S
ubse
quen
tly t
rapp
ed t
o co
nfir
m p
rese
nce
in 1
995
(pre
viou
sly
unpu
blis
hed
resu
lts,
Roa
d, G
ordo
n F
ores
t B
lock
see
Tabl
e 9)
and
to
estim
ate
popu
latio
n si
ze 1
998–
2000
(H
ayw
ard
et a
l. 20
03)
19
91
Jarr
ahda
le,
Kar
net
A.N
. S
tart
, no
te 2
bel
owS
ight
ed b
y G
rant
Pro
nk w
hen
flush
ed f
rom
veg
etat
ion
as e
scap
ing
bush
fire
For
est
Blo
ck
19
92
Wal
pole
, R
oe F
ores
t B
lock
A.N
. S
tart
, no
te 2
bel
owK
now
n fo
r on
e ro
adki
ll ca
rcas
s re
cord
ed b
y P
hil
Dur
ell
19
94
Mt
Man
ypea
ksK
. G
illen
, no
te 2
bel
owR
epor
ted
as d
ead
sub-
adul
t m
ale
19
94
Two
Peo
ples
Bay
(Sin
clai
r 19
99)
Non
e ca
ught
dur
ing
Sin
clai
r’s s
urve
y pe
riod.
Sub
sequ
ently
cau
ght
(by
WA
Dep
t of
Con
serv
atio
n an
d La
nd M
anag
emen
tN
atur
e R
eser
vest
aff)
fro
m D
ecem
ber
1994
onw
ards
fro
m s
ever
al d
iffer
ent
loca
tions
. Lo
catio
n co
-ord
inat
es h
ere
are
ther
efor
eap
prox
imat
e
19
95
Stir
ling
Ran
ge N
atio
nal
Bar
rett
(199
6)P
rese
nce
indi
cate
d by
det
ectio
n of
hai
r (h
air
tube
). F
resh
sca
ts a
nd r
unw
ays
also
obs
erve
d by
Sin
clai
r (1
999)
.P
ark
- E
llen
Pea
k
19
95
Stir
ling
Ran
ge N
atio
nal
Bar
rett
(199
6)P
rese
nce
indi
cate
d by
det
ectio
n of
sca
tsP
ark
- To
olbr
unup
Pea
k
19
95
Stir
ling
Ran
ge N
atio
nal
Bar
rett
(199
6)P
rese
nce
indi
cate
d by
det
ectio
n of
sca
ts f
rom
sou
ther
n sl
opes
and
cas
cade
are
aP
ark
- B
luff
Kno
ll
58 P.J. de Torres et al.
19
95
Mou
nt M
anyp
eaks
Bar
rett
(199
6)P
rese
nce
indi
cate
d by
det
ectio
n of
hai
r (h
air
tube
)N
atur
e R
eser
ve
19
95
Stir
ling
Ran
ge N
atio
nal
Bar
rett
(199
6)P
rese
nce
indi
cate
d by
det
ectio
n of
sca
tsP
ark
- H
ume
Pea
k
19
96
Pem
bert
on,
Poo
le F
ores
tG
. Li
ddel
ow,
note
3 b
elow
App
roxi
mat
e da
te.
Abu
ndan
ce r
ated
as
1 (lo
w)
on s
cale
of
1–3
for
exte
nt o
f vi
sibl
e ac
tivity
(sc
ats
with
in r
unw
ays,
ext
ent
Blo
ck (
seco
nd s
ite)
of r
unw
ays
and
spoo
r)
19
96
Pem
bert
on,
Tin
kers
Bro
ok,
G.
Lidd
elow
, no
te 3
bel
owA
ppro
xim
ate
date
. A
bund
ance
rat
ed a
s 1
(low
) on
sca
le o
f 1–
3 fo
r ex
tent
of
visi
ble
activ
ity (
scat
s w
ithin
run
way
s, e
xten
tS
utto
n F
ores
t B
lock
of r
unw
ays
and
spoo
r)
19
96
Pem
bert
on,
Poo
leG
. Li
ddel
ow,
note
3 b
elow
App
roxi
mat
e da
te.
Abu
ndan
ce r
ated
as
1 (lo
w)
on s
cale
of
1–3
for
exte
nt o
f vi
sibl
e ac
tivity
(sc
ats
with
in r
unw
ays,
ext
ent
For
est
Blo
ckof
run
way
s an
d sp
oor)
19
96
Jarr
ahda
le,
Cha
ndle
rD
. G
iles
(per
sona
l co
mm
unic
atio
nK
now
n fr
om o
ne s
ight
ed i
ndiv
idua
l. P
resu
mab
le c
ontig
uous
and
/or
from
Cha
ndle
r R
oad
popu
latio
nR
oad
Orc
hard
to P
JdeT
)
19
96
Stir
ling
Ran
ge N
atio
nal
Sin
clai
r (1
999)
Not
ed a
s a
sing
le d
ead
anim
al f
ound
on
Che
ster
Pas
s R
oad
in 1
996.
Dat
e pr
ovid
ed a
nd l
ocat
ion
give
n to
Mar
k R
oddy
,P
ark,
Che
ster
Pas
s R
oad
Par
k R
ange
r, as
opp
osite
Tol
l P
eak
car
park
, C
hest
er P
ass
Roa
d
19
97
Jarr
ahda
le,
Gor
don
P. B
att
Gile
s (p
erso
nal
Iden
tifie
d fr
om e
xten
sive
run
way
s, m
ay b
e co
ntin
uos
with
the
Ros
ella
Roa
d po
pula
tion
For
est
Blo
ckco
mm
unic
atio
n to
PJd
eT)
19
98
Two
Peo
ples
Bay
Nat
ure
Frie
nd a
nd B
utle
r (n
omin
ally
199
8)In
cide
ntal
, no
n-ta
rget
cap
ture
of
quok
kas
at 3
site
s at
Tw
o P
eopl
es B
ay N
atur
e R
eser
ve w
hen
mon
itorin
g/su
rvey
ing
for
Res
erve
, M
t G
ardn
er a
rea
Gilb
ert’s
pot
oroo
. S
ites
wer
e E
ast
Fire
brea
k, L
ower
Fire
brea
k &
Wes
t 6.
Map
ped
here
as
1 lo
catio
n, n
ear
Mt
Gar
dner
19
99
Two
Peo
ples
Bay
Nat
ure
Frie
nd a
nd B
utle
r (n
omin
ally
199
9)In
cide
ntal
, no
n-ta
rget
cap
ture
of
quok
kas
at 4
site
s at
Tw
o P
eopl
es B
ay N
atur
e R
eser
ve w
hen
mon
itorin
g/su
rvey
ing
for
Res
erve
, M
t G
ardn
er a
rea
Gilb
ert’s
pot
oroo
. S
ites
wer
e E
ast
Fire
brea
k, L
ower
Fire
brea
k, N
orth
Fire
brea
k &
Hak
ea.
Map
ped
here
as
1 lo
catio
n, n
ear
Mt
Gar
dner
20
01
Per
th w
ater
sup
ply
J. L
iddi
ngto
n (p
erso
nal
One
ind
ivid
ual
quok
ka t
rapp
ed a
s a
non-
targ
et i
ncid
enta
l ca
ptur
e w
hen
trap
ping
for
pig
s. F
irst
repo
rted
cap
ture
catc
hmen
t, C
hurc
hman
com
mun
icat
ion
to P
JdeT
)fr
om t
his
site
For
est
Blo
ck
20
01
Per
th w
ater
sup
ply
J. L
iddi
ngto
n (p
erso
nal
Quo
kka
caug
ht a
s a
non-
targ
et i
ncid
enta
l ca
ptur
e w
hen
trap
ping
for
pig
s. S
econ
d ca
ptur
e fr
om t
his
catc
hmen
t, w
ater
com
mun
icat
ion
to P
JdeT
)se
ctio
n of
cat
chm
ent
catc
hmen
t ex
clus
ion
with
inC
hurc
hman
For
est
Blo
ck
20
02
Per
th w
ater
sup
ply
J. L
iddi
ngto
n (p
erso
nal
Quo
kka
caug
ht a
s a
non-
targ
et i
ncid
enta
l ca
ptur
e w
hen
trap
ping
for
pig
s. T
hird
cap
ture
fro
m t
his
sect
ion
of c
atch
men
tca
tchm
ent,
Chu
rchm
anco
mm
unic
atio
n to
PJd
eT)
For
est
Blo
ck (
seco
nd s
ite)
Tabl
e 4
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 59
20
02
Per
th w
ater
sup
ply
J. L
iddi
ngto
n (p
erso
nal
Con
firm
ed r
oadk
ill.
Fou
rth
reco
rd w
ithin
12–
15 m
onth
s fr
om t
his
sect
ion
of c
atch
men
tca
tchm
ent,
Chu
rchm
anco
mm
unic
atio
n to
PJd
eT)
For
est
Blo
ck (
third
site
)
20
02
Nor
thcl
iffe,
Nai
rn F
ores
tM
. S
heeh
an (
pers
onal
Roa
dsid
e si
ghtin
g of
liv
e qu
okka
s an
d se
vera
l co
nfirm
ed r
oadk
ill r
ecor
ds.
Roa
dkill
s ap
pear
to
be a
ssoc
iate
d w
ithB
lock
com
mun
icat
ion
to P
JdeT
)di
stur
banc
e re
sulti
ng f
rom
roa
ding
act
ivity
, pr
ior
to t
imbe
r ha
rves
ting
20
02
Dw
ellin
gup,
Urb
rae
R.
Sta
ines
(pe
rson
alT
hree
quo
kkas
tra
pped
and
one
car
cass
fou
nd i
n pe
riod
21
Febr
uary
to
11 M
arch
200
2. T
rapp
ed q
uokk
as w
ere
non-
For
est
Blo
ckco
mm
unic
atio
n to
PJd
eT)
targ
et i
ncid
enta
l ca
ptur
es w
hen
trap
ping
for
pig
s. W
ithin
3.5
km o
f su
b ca
tchm
ent
whe
re C
hris
tens
en (
unda
ted)
tra
pped
in 1
972
20
02
Dw
ellin
gup,
Kya
bram
R.
Sta
ines
(pe
rson
alW
ithin
Alc
oa’s
Will
owda
le m
ine
site
. Tra
pped
as
a no
n-ta
rget
inc
iden
tal
capt
ure
whe
n tr
appi
ng f
or p
igs.
Kno
wn
at t
his
site
Bro
ok,
Par
k F
ores
t B
lock
com
mun
icat
ion
to P
JdeT
)fr
om o
ne c
aptu
re o
nly
20
02
Qua
rrum
Nat
ure
Res
erve
,G
. F
reeb
ury
(per
sona
lQ
uokk
a pr
esen
ce c
onfid
ently
inf
erre
d by
det
ectio
n of
sca
ts i
n ch
arac
teris
tic q
uokk
a r
unw
ays
east
of
Wal
pole
com
mun
icat
ion
to P
JdeT
)
20
02
Pea
k F
ores
t B
lock
,G
. F
reeb
ury
(per
sona
lQ
uokk
a pr
esen
ce c
onfid
ently
inf
erre
d by
det
ectio
n of
sca
ts i
n ch
arac
teris
tic q
uokk
a ru
nway
sN
orth
of
Wal
pole
com
mun
icat
ion
to P
JdeT
)
20
02
Mud
dy L
ake,
sou
th(D
ell
& H
yder
-Grif
fiths
200
2)T
he a
utho
rs r
epor
ted
‘road
kill
spe
cim
ens
had
been
col
lect
ed r
ecen
tly’ a
nd ‘c
arca
sses
had
bee
n br
ough
t …
in
… u
p un
tilof
Bun
bury
abou
t tw
o ye
ars
ago’
. A
sku
ll co
llect
ed f
orm
the
site
on
25 S
epte
mbe
r 20
02 h
as b
een
conf
irmed
as
a qu
okka
(W
AM
reco
rd M
5413
2),
how
ever
the
sku
ll is
not
age
d (N
orah
Coo
per,
per
s. c
omm
. to
PJd
eT)
and
quok
ka p
rese
nce
is o
ther
wis
eon
ly i
nfer
red
from
pre
senc
e of
run
way
s
20
03
Dw
ellin
gup
(Ban
ksia
dale
M.
Bur
t (p
erso
nal
com
mun
icat
ion
Roa
dkill
. C
arca
ss r
epor
ted
to M
erv
Bur
t, D
wel
lingu
p. C
arca
ss c
onfir
med
at
swam
p si
te,
not
colle
cted
For
est
Blo
ck)
- w
ithin
to P
JdeT
)A
lcoa
’s H
untly
Min
e S
item
inin
g en
velo
pe
20
03
Del
Par
k R
oad,
Dw
ellin
gup
M.
Bur
t (p
erso
nal
com
mun
icat
ion
Roa
dkill
. C
arca
ss r
epor
ted
and
colle
cted
by
Mer
v B
urt,
Dw
ellin
gup.
Dec
ompo
sed
carc
ass
colle
cted
(Tur
ner/
Mar
rinu
p F
ores
tto
PJd
eT)
Blo
ck)
20
03
Big
hill
Bro
ok (
Nai
rnM
. S
heeh
an (
pers
onal
Roa
dkill
, W
heat
ley
Coa
st R
oad,
300
m N
orth
cliff
e si
de o
f B
ighi
ll B
rook
For
est
Blo
ck)
com
mun
icat
ion
to P
JdeT
)
20
03
Nan
nup
(Bor
onia
G.
Voi
gt (
pers
onal
com
mun
icat
ion
Cha
ract
eris
tic q
uokk
a ru
nway
s id
entif
ied
and
conf
irmed
as
quok
ka b
y G
. Li
ddel
ow,
Dep
artm
ent
of C
onse
rvat
ion
and
For
est
Blo
ck)
to P
JdeT
)La
nd M
anag
emen
t. Id
entif
ied
prio
r to
pro
pose
d si
lvic
ultu
ral
burn
20
03
Nan
nup
(Nel
son
For
est
M.
Max
wel
l (pe
rson
alD
epar
tmen
t of
Con
serv
atio
n an
d La
nd M
anag
emen
t B
lack
woo
d D
istr
ict
reco
rd n
o 70
, no
ctur
nal
sigh
ting
Blo
ck)
com
mun
icat
ion
to P
JdeT
)
20
03
Nan
nup
(Bea
ton
For
est
M.
Max
wel
l (pe
rson
alD
epar
tmen
t of
Con
serv
atio
n an
d La
nd M
anag
emen
t B
lack
woo
d D
istr
ict
reco
rd n
o 74
. D
ead,
pre
sum
ably
a r
oadk
illB
lock
)co
mm
unic
atio
n to
PJd
eT)
20
03
Nan
nup
(Gre
gory
For
est
M.
Max
wel
l (pe
rson
alD
epar
tmen
t of
Con
serv
atio
n an
d La
nd M
anag
emen
t B
lack
woo
d D
istr
ict
reco
rd n
o 76
. D
ead,
pre
sum
ably
a r
oadk
illB
lock
)co
mm
unic
atio
n to
PJd
eT)
60 P.J. de Torres et al.
20
03
Mt
Chu
dalu
p, W
indy
M.
She
ehan
(pe
rson
alTw
o qu
okka
s, n
ight
sig
htin
g, e
aste
rn s
ide
of r
oad
near
‘w
ater
poin
t’ at
Mt
Chu
dalu
p m
onad
nock
Har
bour
Roa
dco
mm
unic
atio
n to
PJd
eT)
20
03
Kar
ri G
ully
, D
alga
rup
K.
Red
man
(pe
rson
alT
hree
sep
arat
e ro
ad k
ill r
ecor
ds o
n B
rock
man
Hig
hway
, be
twee
n N
annu
p an
d B
ridge
tow
n, a
ll w
ithin
the
firs
t ha
lf of
200
3F
ores
t, B
rock
man
hig
hway
,co
mm
unic
atio
n to
PJd
eT)
betw
een
Nan
nup
and
Bri
dg
eto
wn
20
03
Sal
mon
Bea
ch R
oad,
off
M.
She
ehan
(pe
rson
alF
our
sepa
rate
sig
htin
gs o
ver
thre
e w
eeks
on
road
side
in
coas
tal
heat
h/st
unte
d pe
pper
min
t, ne
ar p
ump
shed
Win
dy H
arbo
ur R
oad
com
mun
icat
ion
to P
JdeT
)
20
03
Jarr
ahda
le,
Fro
llet
D.
Gile
s (p
erso
nal
com
mun
icat
ion
Fou
r qu
okka
s si
ghte
d fle
eing
a b
urn
Pla
ntat
ion
(bet
wee
nto
PJd
eT)
Mun
dlim
up a
nd C
obia
cfo
rest
blo
cks)
20
03
Nor
thcl
iffe,
Whe
atle
yP.
Sar
giso
n (p
erso
nal
One
sub
-adu
lt m
ale
road
kill,
200
–300
m s
outh
of
Orc
hid
Roa
d, o
n ea
ster
n si
de o
f W
heat
ley
Coa
st R
oad.
Car
cass
Coa
st R
oad
com
mun
icat
ion
to P
JdeT
)re
tain
ed f
or c
olle
ctio
n by
Dep
artm
ent
of C
onse
rvat
ion
and
Land
Man
agem
ent
staf
f
20
03
Na
nn
up
/Bri
dg
eto
wn
,K
. R
edm
an (
pers
onal
One
adu
lt m
ale
road
kill
. H
alf
way
bet
wee
n N
annu
p an
d B
ridge
tow
n –
400m
W o
f Ja
rrah
Par
kB
rock
man
Hig
hway
com
mun
icat
ion
to P
JdeT
)
20
03
Gre
en R
ange
J.A
. F
riend
(pe
rson
alH
air
sam
ple
colle
cted
whe
n su
rvey
ing
(hai
r tu
bes)
for
Gilb
ert’s
Pot
oroo
.co
mm
unic
atio
n to
PJd
eT)
Not
es:
1K
abay
and
Sta
rt (
1976
) S
ough
t in
form
atio
n fr
om t
he p
ublic
, fr
om f
ield
sta
ff fr
om r
elev
ant
gove
rnm
ent
agen
cies
and
car
ried
out
biol
ogic
al s
urve
y to
det
ect
the
pres
ence
of
Gilb
ert’s
pot
oroo
,P
otor
ous
trid
acty
lus
gilb
ertii
, (n
ow P
. gi
lber
tii)
and
the
broa
d fa
ced
poto
roo,
P.
plat
yops
, in
the
per
iod
Apr
il 19
75 t
o O
ctob
er 1
976.
Quo
kka
pres
ence
was
als
o re
cord
ed a
t nu
mer
ous
site
sin
vest
iga
ted
.2.
Unp
ublis
hed
data
base
rec
ords
of
A.N
. S
tart
, W
este
rn A
ustr
alia
n D
epar
tmen
t of
Con
serv
atio
n an
d La
nd M
anag
emen
t.3.
Unp
ublis
hed
reco
rds
from
inc
iden
tal
sigh
tings
and
opp
ortu
nist
ic t
rapp
ing
by G
. Li
ddel
ow,
Wes
tern
Aus
tral
ian
Dep
artm
ent
of C
onse
rvat
ion
and
Land
Man
agem
ent
Tabl
e 4
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 61
Table 5
Records of occurrence of the quokka, Setonix brachyurus, extracted from unpublished Department of Conservationand Land Management databases (CALM unpublished; Gilfillan unpublished). Original source for most records wasDepartmental operational district staff. Records are predominantly from incidental and opportunistic sightings. Recordsare listed chronologically.
Date of Location Commentrecord
1987 Pemberton, Brockman National Park, near Pemberton-Northcliffe Road carcass – 3 animals
1992 William Bay National Park Information forwarded by Greg Freebury, WADepartment of Conservation and LandManagement. Date is estimate. Diurnal sighting of1 individual
1993 Pemberton, Brockman National Park roadkill
1994 Manjimup, Lewin Forest Block, Davidson Road / Pine Creek gully system No other details
1994 Manjimup, Lewin Forest Block. Davidson Road / Pine Creek gully system
1994 Manjimup, 7 Day Road / Court Rd No other details
1994 Manjimup
1995 Manjimup, Lewin Forest Block, Junction Davidson & Easter Rd No other details
1995 Manjimup, Andrew Forest Block No other details
1995 Manjimup, Lewin Forest Block, junction of Davidson and Easter roads No other details
1995 Manjimup, Lewin Forest Block, junction of Davidson and Easter roads
1995 Manjimup, Lewin Forest Block, junction Davidson and Easter roads
1995 Manjimup, Andrew Forest Block
1996 Walpole-Nornalup National Park, Cemetery Road Reserve 31362
1996 Walpole-Nornalup National Park, Monastery Road, 1.5 km fromSouth Coast Highway
1996 Walpole-Nornalup National Park, South Coast Highway, 0.5 km westof Gully Road
1996 Walpole-Nornalup National Park, near Spike Road Reserve 31362
1996 Walpole-Nornalup National Park, Frankland River, nearMonastery Landing
1996 Walpole-Nornalup National Park, Cemetery Road Reserve 31362
1996 Walpole-Nornalup National Park, Allen Road, 0.5 km from Hilltop Road
1996 Walpole-Nornalup National Park, Pool Road
1996 Walpole-Nornalup National Park, eastern boundary of Reserve 31362
1996 Walpole, Valley of the Giants, Valley of the Giants Road
1996 Walpole, Loc. 10190, 1 km northwest of junction of Jones Road andHilltop Road
1996 Walpole, Valley of the Giants, Bohall Road, 2.5 km from South CoastHighway
1996 Walpole, Valley of the Giants, I km along track running southeast ofPedro firebreak
62 P.J. de Torres et al.
1996 Walpole, Valley of the Giants, northern end of Twin Creek Road
1996 Walpole, Valley of the Giants, Court Road, 1 km from South Coast Hwy
1996 Walpole, Valley of the Giants, South Coast Highway, 250 m west ofConspicuous Beach Road.
1996 Walpole, 1.5 km northwest of junction of Jones Road and Hilltop Road
1996 Walpole, Valley of the Giants, Valley of the Giants Road
1996 Walpole-Nornalup National Park, Monastery Road, 0.5 km west ofZig Zag Road
1996 Walpole, Valley of the Giants, Valley of the Giants Road
1996 Walpole, Valley of the Giants, Rate Road 3.3 km north ofSouth Coast Hwy
1996 Pemberton, Brockman NP, adjacent to farmland Diurnal sighting of 2 quokka disturbed by andescaping from machine (dozer) disturbance
1996 Walpole, Valley of the Giants, South Coast Hwy, 1.1 km east ofConspicuous Beach Road
1996 Frankland, Ford Road 3 km from Talbot Road
1996 Walpole Inlet, near Collier Creek, eastern edge
1996 Nornalup, Nornalup Bridge, South Coast Highway.
1996 Manjimup, Solai Forest Block, Solai Rd HC6395 No other details
1996 Walpole, Valley of the Giants, South Coast Highway, 1km southof 28 Mile Road
1996 Manjimup, Yardup Forest Block, 1km North along Edwards Roadfrom Perup Road Database records indicate two different sightings,
same day. The source is identified for one sightingonly and is possibly a duplicate record. The Yarduprecord(s), combined with the 1954 record by A.D.Jones in Serventy et al. (1954) represent the onlyquokka records from the Perup forest.
1996 Manjimup, Lewin Forest Block, Easter & Eastwin roads No other details
1996 Manjimup, Nelson Forest Block, Bibbulmun Track approx 500m south No other detailsof Willow Springs
1996 Walpole, Valley of the Giants, near South Coast Highway, 2 km westof Nut Road
1996 Manjimup, Solai Forest Block, Solai Road
1996 Manjimup, Yardup Forest Block, 1km north along Edwards Road from No other detailsPerup Road
1996 Roe, first creek system west of Claude Road, on Roe Road.(near Mt. Roe)
1996 D’Entrecasteaux National Park, Mandalay Beach Road, near beach
1996 Walpole-Nornalup National Park, Tinglewood Road, 1.5 km from SouthCoast Highway
1996 Manjimup, Nelson Forest Block, Bibbulmun Track approx 500m south ofWillow Springs
1996 Walpole, Keystone Rd. 1.5 km north of South Coast Highway
Table 5 (cont.)
Date of Location Commentrecord
Distribution of the quokka 63
1996 Manjimup, Easter Forest Block, Easter and Eastwin roads
1996 Walpole-Nornalup National Park, Shelleys Beach
1996 Walpole-Nornalup National Park, South Coast Highway, 0.75 km westof Tinglewood Road
1996 Walpole-Nornalup National Park, walking track between Sealers Coveand Circus Beach
1996 Walpole-Nornalup National Park, gully north-northwest of Circus Beach
1996 Walpole-Nornalup National Park, 1 km northwest of Circus Beach
1996 Walpole, Rest Point Road, 1 km from South Coast Highway
1996 Denbarker, Denmark- Mt. Barker Road, near Mitchell River Bridge
1997 Manjimup, Netic Forest Block, Pool Road
1997 Walpole, Approximately 200m east of the junction of Tree Top Walk RoadkillRoad and Valley of the Giants Road on Valley of the Giants Road
1997 Pemberton, D’Entrecasteaux National Park, near Landslide Road in diurnal sighting of 1 adult quokkaaerial burn area DC16
1997 Walpole, Valley of the Giants Road approximately 200m west of the Roadkill, carcass identifiedTree Top Walk turn-off
1997 Manjimup, Graphite Forest Block, Austin Road
1997 Manjimup, Andrew Forest Block, Top Road
1997 Manjimup, Netic Forest Block, Netic quokka exclusion (WHAT IS THIS)
1997 Manjimup, Netic Forest Block, Pool Road
1997 Manjimup, Netic Forest Block, junction of Sexton and Parky Roads
1997 Manjimup, Netic Forest Block, Sexton Road
1997 Manjimup, Netic Forest Block, Kanny Road
1997 Manjimup, Andrew Forest Block, Dalberg Road waterpoint – 2.5 kmwest of Austin Road
1997 Pemberton, D’Entrecasteaux National Park Roadkill on Salmon Beach Road 1km from WindyHarbour. Definite signs of quokka activity in 4yearold heath north of road
1997 Manjimup, Mack Forest Block, Well Road No other details
1997 Manjimup, Netic Forest Block, Kanny / Parky Roads
1997 Pemberton, Crowea Forest Block, McAlpine Road waterpoint Presence inferred by signs (scats, runways)
1997 Walpole, Walpole-Nornalup National Park, walk trail from town to Diurnal sighting of one individual, crossingCoalmine Beach walk trail
1997 Manjimup, Solai Forest Block, Monk / Solai Roads
1997 Pemberton, Crowea Forest Block, thick ti-tree between coupes Presence inferred by signs (scats, runways)on Crowea Road
1997 Pemberton, Crowea Forest Block Diurnal sighting
1997 Pemberton, Crowea Forest Block, track crossing Orchid Rd,1.1km from Crowea Road
1997 Pemberton, Crowea Forest Block, south-west corner 1984 regen Presence inferred by signs (scats, runways)on south side of road
Date of Location Commentrecord
64 P.J. de Torres et al.
1997 Pemberton, Crowea Forest Block, McAlpine Road, water point Presence inferred by signs (scats, runways)near Hawkins Road and Dan Road
1997 Pemberton, Crowea Forest Block, Rowney Road north of Crowea Rd Presence inferred by signs (scats, runways)
1997 Pemberton, Crowea Forest Block, Karri/Marri gully 1.9km fromWheatley Coast Road
1997 Pemberton, Crowea Forest Block, track crossing Cederman Road400m from Wheatley Coast Road. Karri/Marri regrowth
1997 Walpole, Wye Forest Block Presence inferred by signs (spoor)
1997 Manjimup, Mack Forest Block, Mt Mack Road / Tom Road intersection
1997 Manjimup, Netic Forest Block, Pool Road No other details
1997 Walpole, Rocky Forest Block No detail, presumably a sighting
1997 Walpole, Rocky Forest Block Old (not fresh) evidence in unburnt gully
1997 Walpole, Sharpe Forest Block Presumably determined by evidence of activity, atcreek in Sharpe Block (North of Sharpe 6), flowsinto the Deep River
1997 Manjimup, Solai Forest Block, Monk / Solai Roads No other details
1997 Manjimup, Andrew Forest Block, Top Road No other details
1997 Manjimup, Graphite Forest Block, Austin Road No other details
1997 Manjimup, Graphite Forest Block, Cow Brook – Davidson Road No other details
1997 Manjimup, Gordon Forest Block, Mobil Road No other details
1997 Manjimup, Andrew Forest Block, Dalberg Road waterpoint – 2.5 km No other details
west of Austin Road
1997 Manjimup, Mack Forest Block, McNab Well No other details
1997 Manjimup, Netic Forest Block, Penny Road No other details
1997 Manjimup, Netic Forest Block, Kanny Road No other details
1997 Manjimup, Netic Forest Block, Kanny / Parky Roads No other details
1997 Manjimup, Netic Forest Block, Junction of Sexton and Parky Roads No other details
1997 Manjimup, Netic Forest Block, Pool Road No other details
1997 Manjimup, Netic Forest Block No other details
1997 Manjimup, Mack Forest Block, Mt Mack Road / Tom Road intersection No other details
1997 Walpole, Rocky Forest Block Recorded as gully crossing road, presumablypresence inferred by evidence of activity
1997 Manjimup, Mack Forest Block, McNab Well
1997 Manjimup, Gordon Forest Block, Mobil Road
1997 Manjimup, Netic Forest Block, Penny Road
1997 Manjimup, Graphite Forest Block, Cow Brook – Davidson Road
1997 Manjimup, Netic Forest Block, Sexton Road No other details
Table 5 (cont.)
Date of Location Commentrecord
Distribution of the quokka 65
1997 Walpole, Ford Road approximately 2.0km south west from the corner diurnal sighting of 1 quokka crossing roadof Ford and Collis roads
1997 Walpole, approximately 200m west of the Conspicuous Beach Roadkillturn-off on the South Coast Highway (northern verge)
1997 Manjimup, Mack Forest Block, Well Road
1998 Walpole, Walpole-Nornalup National Park, Valley of the diurnal sighting of 1 quokka crossing carparkGiants old carpark
1998 Walpole, Walpole-Nornalup National Park, Valley of the Giants Road, night sighting of adult quokka crossing roadapproximately 500m south of Howe Road
1998 Walpole, 100m west on Beardmore and Thomson roads diurnal sighting of 1 quokka crossing road
1998 Walpole, approximately 400m west of the Jack Rate Lookout on nocturnal sighting of 1 quokka at roadsidethe South West Highway
1999 Walpole, South Coast Highway approximately 3.5km east of Walpole Roadkill 1 adult
1999 Walpole, Jones Road, 1km south of Clarke Road Nocturnal sighting of 3 quokkas, at least 1 adult.Runways present in vegetation
1999 Manjimup, Beavis Forest Block, Beavis 8 diurnal sighting of 2 quokkas
1999 Manjimup, Lewin Forest Block, Pine Creek Road, 1.3 km from CarcassDavidson Road
1999 Walpole, Walpole-Nornalup National Park, 100m west of Nornalup Bridge Roadkill, carcass identified
1999 Manjimup, 220 metres east of Boundary Road on Willow Spring Road identified from carcass
1999 Manjimup, Andrew Forest Block diurnal sighting of 2 quokkas
1999 Manjimup, near intersection Penny Road and Willow Spring Road diurnal sighting of 1 quokka
1999 Stirling Range National Park, Pyungoorup Peak Carcass forwarded to Western Australian Museum
1999 Two Peoples Bay Nature Reserve, on access track to research quarters Nocturnal sighting of 1 quokka
1999 Walpole, South West Highway, approximately 400m west of the Roadkill 1 adultJack Rate Lookout and approximately 700m east of Tinglewood Road
2000 Waychinicup National Park carcass
2000 Two Peoples Bay Nature Reserve, Little Beach Road, start of slashed Nocturnal sighting of 1 quokkafirebreak to south of road
2000 Two Peoples Bay Nature Reserve, approximately 2–300 m from Nocturnal sighting of 2 quokkasstart of Little Beach Road
2000 Waychinicup National Park, Waychinicup Road, just north carcassof creek crossing
2000 Manjimup, Nelsons Location 9466 Jones Rd, Yanmah / Glenoran area day sighting of 3 individuals, feeding onhousehold scraps
2000 Waychinicup National Park, campsite Nocturnal sighting of 2 quokkas
2000 Two Peoples Bay Nature Reserve, Little Beach Road, start of slashed carcassfirebreak to south of road
2000 Manjimup, Beavis Forest Block
2000 Manjimup, Donnelly Mill Road nocturnal sighting of 1 quokka
2000 Manjimup
2000 Manjimup, 100m South of Palings Road and Coronation Road Carcass
Date of Location Commentrecord
66 P.J. de Torres et al.
2000 Manjimup, 1km east along Telephone Road from Coronation Road Carcass
2000 Manjimup one individual sighted running from road to scrub
2001 Walpole, 1/2 way between Deep River and Crystal Springs on carcassSouth Western Highway
2001 Walpole, Crystal Springs, Coalmine Beach turnoff on South Coast CarcassHighway, approximately 3km east of Walpole
2001 Stirling Range National Park, Mt. James Road capture of 1 quokka during Western Shieldmonitoring
2001 Dwellingup, Del Park Road, south of Alcoa minesite Roadkill 1 adult
Table 5 (cont.)
Date of Location Commentrecord
Distribution of the quokka
67
Table 6
Previously unpublished records of occurrence of the quokka, Setonix brachyurus, held by the authors. Records are from opportunistic sightings and trapping programsimplemented to determine presence. Records are listed chronologically.
Year of Location or Source Commentrecord Site name
1971 Dwellingup, Lewis (Wild Pig) Swamp Dillon (1993) see note 1, below Reported on the 1971 survey of the wetter western creek systems in vicinity of Dwellingup. A total of 52quokkas caught at 6 sites. Population density estimates of 1 animal per 2 ha in swamps unburnt for 5–10years. Locations nominally mapped as Lewis and Holyoake swamps
1971 Dwellingup, Holyoake Dillon (1993) see note 1, below As above
1988 Dwellingup, Alcoa’s Huntly Mine Site Dillon (1993) see note 1, below Four of 5 sites inspected in 1988 showed evidence of quokka activity. Nominally mapped as Lewisenvelope Swamp, within mine site envelope. No estimates of population size
1992 Dwellingup, Holyoake Dillon (1993) see note 1, below 15 of 30 swamps surveyed in vicinity of Dwellingup in 1992 showed evidence of quokka activity. Three ofthese were then trapped unsuccessfully (fence funnel trap), subsequently one quokka only was trapped(wire cage trap) at the Holyoake site only.
1992 Collie/Mornington, Hadfield R. Brazell, note 2 below Population presumed to be contiguous with population trapped and population size reported byForest Block Hayward et al. (2003)
1992 Collie/Mornington, Gervasse R. Brazell, note 2 below This site known to support a relatively large population. Population estimate reported in this study (seeForest Block Table 7)
1992 Collie/Mornington, Hamilton R. Brazell, note 2 below 3 quokkas trapped when site trapped to determine presence in 1992. Population presumably contiguousForest Block with the population trapped by Hayward et al. (2003)
1992 Collie/Mornington, Victor Road Site, R. Brazell, note 2 below This site known to support a relatively large population. Population presumably contiguous with theHamilton Forest Block population trapped by Hayward et al. (2003)
1992 Collie/Mornington, Hadfield R. Brazell, note 2 below This site known to support a quokka population – population estimate reported by Hayward et al (2002).Forest Block Other incidental quokka reportings and opportunistic trapping from Hadfield Forest Block presumed to be
contiguous with this population
1995 Pemberton, Chudalup Forest Block Dillon (1996), note 3 below Presence indicated, abundance rated at 1 on scale of 0 (absence) to 3 (relatively abundant). Abundanceconsidered comparable with survey in 1978–79
1995 Manjimup, Andrew Forest Block Dillon (1996), note 3 below Presence indicated, abundance rated at 2 on scale of 0 (absence) to 3 (relatively abundant). Unknown if(second site) abundance has changed since survey in 1978–79
1995 Manjimup, Lindsay Forest Block Dillon (1996), note 3 below Presence indicated. Abundance rated at 1 on scale of 0 (absence) to 3 (relatively abundant) andconsidered to be reduced from previous survey in 1978–79. Logging operations close to swamp.
1995 Nannup, Mack Forest Block Dillon (1996), note 3 below Presumably contiguous with the Andrew Forest Block (third site) record. Presence indicated, abundancerated at 1 on scale of 0 (absence) to 3 (relatively abundant). Abundance considered comparable withsurvey in 1992–93
68 P.J. de Torres et al.
19
95
Wal
pole
, W
eld
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
cons
ider
ed c
ompa
rabl
e w
ith s
urve
y in
197
8–79
19
95
Wal
pole
-Nor
nalu
p N
atio
nal
Par
kD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 2
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
(Hill
top
Roa
d or
for
mer
lyco
nsid
ered
com
para
ble
with
sur
vey
in 1
978–
79.
Was
bur
nt i
n w
ildfir
e in
198
7, q
uokk
as s
een
disp
ersi
ng a
tH
illto
p F
ores
t B
lock
)tim
e of
fire
19
95
Wal
pole
, C
ross
ing
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
cons
ider
ed c
ompa
rabl
e w
ith s
urve
y in
197
8–79
19
95
Wal
pole
, G
iant
s F
ores
t B
lock
Dill
on (
1996
), n
ote
3 be
low
Pre
senc
e in
dica
ted,
abu
ndan
ce r
ated
at
1 on
sca
le o
f 0
(abs
ence
) to
3 (
rela
tivel
y ab
unda
nt).
Abu
ndan
ceco
nsid
ered
com
para
ble
with
sur
vey
in 1
978–
79
19
95
Nan
nup,
Gre
gory
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
cons
ider
ed c
ompa
rabl
e w
ith s
urve
y in
199
2–93
. S
ite i
s lo
ng u
nbur
nt
19
95
Nan
nup,
Bea
ton
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
cons
ider
ed c
ompa
rabl
e w
ith s
urve
y in
199
2–93
19
95
Nan
nup,
Bea
ton
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
(thi
rd s
ite)
cons
ider
ed c
ompa
rabl
e w
ith s
urve
y in
199
2–93
19
95
Man
jimup
, A
ndre
w F
ores
t B
lock
Dill
on (
1996
), n
ote
3 be
low
Pre
sum
ably
con
tiguo
us w
ith t
he M
ack
For
est
Blo
ck r
ecor
d. P
rese
nce
indi
cate
d. A
bund
ance
rat
ed a
t 2
on(t
hird
site
)sc
ale
of 0
(ab
senc
e) t
o 3
(rel
ativ
ely
abun
dant
) an
d co
nsid
ered
to
be c
ompa
rabl
e w
ith p
revi
ous
surv
ey i
n19
78–7
9. S
ite d
istu
rbed
by
cons
truc
tion
of n
ewly
dug
wat
er p
oint
.
19
95
Nan
nup,
Hel
ms
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). A
bund
ance
cons
ider
ed r
educ
ed f
rom
sur
vey
in 1
975.
sw
amp
patc
h bu
rnt
in 1
994
(1 y
ear
prio
r to
thi
s as
sess
men
t)
19
95
Nan
nup,
Bea
ton
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). U
nabl
e to
(sec
ond
site
)de
term
ine
if ch
ange
in
abun
danc
e fr
om p
revi
ous
surv
ey
19
95
Nan
nup,
Bea
ton
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). U
nabl
e to
(fift
h si
te)
dete
rmin
e if
abun
danc
e ha
s ch
ange
d si
nce
prev
ious
sur
vey
19
95
Man
jimup
, A
ndre
w F
ores
t B
lock
Dill
on (
1996
), n
ote
3 be
low
Pre
senc
e in
dica
ted,
abu
ndan
ce r
ated
at
2 on
sca
le o
f 0
(abs
ence
) to
3 (
rela
tivel
y ab
unda
nt).
Unk
now
n if
abun
danc
e ha
s ch
ange
d si
nce
surv
ey i
n 19
78–7
9
19
95
Nan
nup,
Bea
ton
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 2
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). U
nabl
e to
(fou
rth
site
)de
term
ine
if ab
unda
nce
chan
ged
from
pre
viou
s su
rvey
19
95
Nan
nup,
Gre
gory
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). u
nabl
e to
(sec
ond
site
)de
term
ine
if ab
unda
nce
has
chan
ged
sinc
e la
st s
urve
y. r
unw
ays
appe
ar u
nuse
d, o
ld s
cat
only
det
ecte
d
Tabl
e 6
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 69
19
95
Dw
ellin
gup,
Kes
ners
Sw
amp,
P. d
e To
res,
M.
Dill
on,
A. T
omki
nson
Kno
wn
to s
uppo
rt a
pop
ulat
ion
in t
he 1
970s
, po
pula
tion
pres
ence
con
firm
ed t
hrou
gh t
rapp
ing
inM
arrin
up F
ores
t B
lock
and
R.
Bue
hrig
, no
te 4
bel
owF
ebru
ary
to M
arch
199
5. P
opul
atio
n si
ze e
stim
ated
by
Hay
war
d (2
003)
19
95
Dw
ellin
gup,
Lew
is (
Wild
Pig
) S
wam
p,P.
de
Tore
s, M
. D
illon
, A
. Tom
kins
onS
how
ed s
igns
of
fres
h ac
tivity
in
1995
, pr
esen
ce u
nabl
e to
be
conf
irmed
by
trap
ping
. P
rese
nce
also
Wils
on F
ores
t B
lock
and
R.
Bue
hrig
, no
te 4
bel
owun
able
to
be c
onfir
med
in
2000
by
Hay
war
d et
al.
(200
3) a
nd t
his
loca
tion
may
no
long
er s
uppo
rts
ap
op
ula
tion
19
95
Dw
ellin
gup,
Hol
yoak
e F
ores
t B
lock
P. d
e To
res,
M.
Dill
on,
A. T
omki
nson
Sho
wed
evi
denc
e of
act
ivity
in
1995
, ho
wev
er,
by 2
000
cons
ider
ed t
o no
lon
ger
supp
ort
a po
pula
tion
and
R.
Bue
hrig
, no
te 4
bel
ow
19
95
Jarr
ahda
le,
Ros
ella
Roa
d,P.
de
Tore
s, M
. D
illon
, A
. Tom
kins
onTr
appe
d to
con
firm
pre
senc
e. S
even
cap
ture
eve
nts
of 4
quo
kkas
, A
pril
1995
Gor
don
For
est
Blo
ckan
d R
. B
uehr
ig,
note
4 b
elow
19
96
Man
jimup
, S
torr
y F
ores
t B
lock
Dill
on (
1996
), n
ote
3 be
low
Pre
senc
e in
dica
ted,
abu
ndan
ce r
ated
at
1 on
sca
le o
f 0
(abs
ence
) to
3 (
rela
tivel
y ab
unda
nt).
Una
ble
tode
term
ine
if ab
unda
nce
has
chan
ged
sinc
e la
st s
urve
y
19
96
Nan
nup,
Net
ic F
ores
t B
lock
Dill
on (
1996
), n
ote
3 be
low
Pre
senc
e in
dica
ted,
abu
ndan
ce r
ated
at
2 on
sca
le o
f 0
(abs
ence
) to
3 (
rela
tivel
y ab
unda
nt).
Una
ble
tode
term
ine
if ab
unda
nce
has
chan
ged
sinc
e pr
evio
us s
urve
y. f
resh
tra
cks
and
activ
ity b
eyon
d sw
amp
area
19
96
Nan
nup,
Nel
son
For
est
Blo
ckD
illon
(19
96),
not
e 3
belo
wP
rese
nce
indi
cate
d, a
bund
ance
rat
ed a
t 1
on s
cale
of
0 (a
bsen
ce)
to 3
(re
lativ
ely
abun
dant
). U
nabl
e to
dete
rmin
e if
abun
danc
e ha
s ch
ange
d si
nce
prev
ious
sur
vey.
app
ears
6–1
0 ye
ars
sinc
e la
st (
patc
hy)
burn
19
96
Jarr
ahda
le,
Cha
ndle
r R
oad,
Cha
ndle
rP.
de
Tore
s, M
. D
illon
, A
. Tom
kins
onK
now
n to
sup
port
a p
opul
atio
n -p
opul
atio
n es
timat
e re
port
ed b
y H
ayw
ard
et a
l. (2
003)
For
est
Blo
ckan
d R
. B
uehr
ig,
note
4 b
elow
19
96
Jarr
ahda
le,
Alb
any
Hig
hway
Site
.P.
de
Tore
s, M
. D
illon
, A
. Tom
kins
onP
opul
atio
n re
port
ed f
rom
a r
oadk
ill i
n 19
96 a
nd s
ubse
quen
t fo
llow
up
trap
ping
(in
the
sam
e ye
ar)
(Tra
velle
rs’ A
rms
site
).an
d R
. B
uehr
ig,
note
4 b
elow
conf
irm
ed p
rese
nce.
Thi
s si
te p
resu
med
to
be t
he T
rave
llers
’ Arm
s si
te r
efer
red
to i
n 19
56 b
y R
. A
itken
,ci
ted
by B
arke
r et
al.
(195
7) (
Tabl
e 3)
and
tra
pped
by
Ken
t W
illia
ms
in 1
966
(Tab
le 4
)
19
96
Col
lie/
Mor
ning
ton
Roa
d,R
. B
raze
ll, n
ote
2 be
low
Kno
wn
from
1 r
oadk
ill,
sour
ce p
opul
atio
n no
t co
nfirm
edH
amilt
on F
ores
t B
lock
20
00
Col
lie,
Dav
is F
ores
t B
lock
M.
Hay
war
d, n
ote
5 be
low
Pre
senc
e in
ferr
ed b
y de
tect
ion
of s
cats
with
in r
unw
ays
20
00
Dw
ellin
gup,
Mar
rinup
For
est
Blo
ckM
. H
ayw
ard,
not
e 5
belo
wP
rese
nce
infe
rred
by
dete
ctio
n of
sca
ts w
ithin
run
way
s
20
00
Dw
ellin
gup,
Fed
eral
For
est
Blo
ckM
. H
ayw
ard,
not
e 5
belo
wP
rese
nce
infe
rred
by
dete
ctio
n of
sca
ts w
ithin
run
way
s
20
00
Pem
bert
on,
Dic
kson
Roa
dM
. H
ayw
ard,
not
e 5
belo
wP
rese
nce
infe
rred
by
dete
ctio
n of
sca
ts w
ithin
run
way
s
20
00
Dw
ellin
gup,
Fin
lay
Bro
ok,
Wils
onM
. H
ayw
ard,
not
e 5
belo
wP
rese
nce
infe
rred
by
dete
ctio
n of
sca
ts w
ithin
run
way
s. S
poor
sub
sequ
ently
obs
erve
d by
For
est
Blo
ckde
Tor
es a
nd D
illon
20
00
Dw
ellin
gup,
Clin
ton
For
est
Blo
ckM
. H
ayw
ard,
not
e 5
belo
wP
rese
nce
infe
rred
by
dete
ctio
n of
sca
ts w
ithin
run
way
s.
20
00
Dw
ellin
gup,
Alc
oa C
onve
yor
Bel
tM
. H
ayw
ard,
not
e 5
belo
wP
rese
nce
infe
rred
by
dete
ctio
n of
sca
ts w
ithin
run
way
s.
20
00
Dw
ellin
gup,
Sou
th D
anda
lup
Dam
M.
Hay
war
d, n
ote
5 be
low
Pre
senc
e in
ferr
ed b
y de
tect
ion
of s
cats
with
in r
unw
ays.
Als
o kn
own
prev
ious
ly f
rom
inc
iden
tal
sigh
tings
.P
opul
atio
n m
ay b
e co
ntig
uous
with
pop
ulat
ion
at K
esne
rs S
wam
p an
d th
e C
onve
yor
Bel
t S
ite
70 P.J. de Torres et al.
20
00
Dw
ellin
gup,
Kea
ts F
ores
t B
lock
M.
Hay
war
d, n
ote
5 be
low
Pre
senc
e in
ferr
ed b
y de
tect
ion
of s
cats
with
in r
unw
ays.
20
00
Dw
ellin
gup,
Tur
ner
For
est
Blo
ckP.
de
Tore
s, n
ote
4 be
low
Inci
dent
al s
ight
ing
in M
ay 2
000,
sub
sequ
ent
addi
tiona
l in
cide
ntal
sig
htin
g re
port
ed i
n O
ctob
er 2
000
and
road
kill
(car
cass
col
lect
ed b
y M
. M
axw
ell
and
C. T
reth
owan
) in
Dec
embe
r 20
00.
Als
o su
bseq
uent
lyco
nsid
ered
by
Hay
war
d to
sup
port
a q
uokk
a po
pula
tion
Not
es:
1.D
illon
(19
93)
repo
rted
on
a 19
71 s
urve
y of
cre
ek s
yste
ms
in t
he w
ette
r w
este
rn c
reek
sys
tem
s in
vic
inity
of
Dw
ellin
gup,
a 1
988
surv
ey w
ithin
Alc
oa’s
Hun
tly m
ine
site
env
elop
e ne
ar D
wel
lingu
p an
d19
92 b
road
scal
e su
rvey
to
dete
rmin
e pr
esen
ce i
n sw
amps
nea
r D
wel
lingu
p2.
Unp
ublis
hed
reco
rds
from
inc
iden
tal
sigh
tings
and
opp
ortu
nist
ic t
rapp
ing
by R
.I. B
raze
ll, W
este
rn A
ustr
alia
n D
epar
tmen
t of
Con
serv
atio
n an
d La
nd M
anag
emen
t.3.
Dill
on (
1996
) re
port
on
surv
ey o
f si
tes
surv
eyed
fro
m 1
995
–199
6, w
here
pre
senc
e w
as c
onfir
med
thr
ough
tra
ppin
g, o
r in
dica
ted
by s
cats
obs
erve
d w
ithin
run
way
s an
d ot
her
fres
h si
gns
of a
ctiv
ity–
scat
s al
one
not
cons
ider
ed s
uffic
ient
to
conf
irm p
rese
nce.
4.U
npub
lishe
d re
cord
s fr
om i
ncid
enta
l si
ghtin
gs,
trap
ping
pro
gram
s in
itiat
ed t
o de
term
ine
pres
ence
and
opp
ortu
nist
ic t
rapp
ing
by P
. J.
de
Tore
s, M
. J.
Dill
on,
A.
Tom
kins
on a
nd R
. B
uehr
ig,
Wes
tern
Aus
tral
ian
Dep
artm
ent
of C
onse
rvat
ion
and
Land
Man
agem
ent.
5.R
esul
ts f
rom
unp
ublis
hed
field
sur
vey
cond
ucte
d by
M.
W.
Hay
war
d be
twee
n S
epte
mbe
r 19
98 a
nd S
epte
mbe
r 20
00 w
here
a r
ange
of
pote
ntia
l lo
catio
ns w
ere
exam
ined
for
evi
denc
e of
quo
kka
pres
ence
. W
ith t
he e
xcep
tion
of t
he S
outh
Dan
dalu
p D
am s
ite,
the
site
s lis
ted
wer
e pr
evio
us n
ot k
now
n to
sup
port
quo
kkas
.
Tabl
e 6
(con
t.)
Year
of
Loca
tion
orS
ourc
eC
omm
ent
reco
rdS
ite n
ame
Distribution of the quokka 71
Table 7
Previously unpublished trapping records for the quokka, Setonix brachyurus, from sites in the northern jarrah forest ofsouth-west Western Australia
Site name Y ear(s) Purpose of trapping Method of analysis Resulttrapped program
Gervasse Forest 1992 to Long-term monitoring Abundance estimates derived from the Population estimate ofBlock, near Collie 2000 of abundance and ‘deaths and no immigration’ Jolly-Seber 49.21 ±7.85 individuals
survivorship model Survivorship estimates derivedfrom the Program MARK Cormack-Jolly-Seber model
Rosella Road, April 1995 Opportunistic trapping to Listing of total number of captured 7 captures of 4 quokkasnear Jarrahdale determine presence only individuals
Subsequently trapped byHayward et al. (2003) toderive populationestimates
Albany Highway, 1996 Opportunistic trapping to Listing of total number of captured 1 capture of an adult malesoutheast of the determine presence only individualsPerth metropolitanarea. This location ispresumed to be theTravelers’ Arms sitereferred to by Barkeret al. (1957) (Table 3)and Williams in 1966(Table 4)
Holyoake, August Opportunistic trapping to N/A No captures, no evidenceDwellingup 1995 determine presence only of recent presence. Old
runways and no freshscats
Lewis (Wild Pig) August Opportunistic trapping to N/A No captures, someSwamp, Dwellingup 1995 determine presence only evidence of recent activity
in runways
Kesners Swamp, February to Opportunistic trapping to Listing of total number of captured 22 captures of 10 quokkasDwellingup June 1995 determine presence only individuals
Subsequently trapped byHayward et al. (2003) toderive population estimates
72 P.J. de Torres et al.
Table 8
Location records for the quokka, Setonix brachyurus, excluded from maps used to infer distribution and excluded fromestimates of extent of occurrence. Records listed and/or the location described were either unable to be confirmed, orwere from captive colonies/sanctuaries
Location Year of Source of Commentrecord record
North Twin 1905 Shortridge (1909) Shortridge included ‘Twin Peak’ island as a known site for the quokka. UnclearPeak Island if this was a reference to North Twin Peak Island or South Twin Peak Island.
Quokkas have not been recorded from either island by any other author andthe record is considered to be a tammar wallaby, see comments in text
South Twin 1905 Shortridge (1909) Comments for the North Twin Peak Island, above apply.Peak Island Western Australian Museum Identification confirmed as a quokka in 1982 by D.J. Kitchener (WA Museum)Point Culver records. WAM registration (Norah Cooper pers. comm. to PJdeT). The record is outside the known
number 024346 historic range and the range as determined by the fossil record and istherefore considered spurious. No additional location details available and nodate of collection or any other details. Accuracy of location thereforeconsidered insufficient to include within inferred distribution.
Coorow 1949 Western Australian Museum Validity of record questioned by Nora Cooper (WA Museum). Remains wererecords. WAM registration collected from a cave in 1949, subsequently destroyed and unable to benumber 002781 verified (Norah Cooper pers. comm. to PJdeT).
Breaksea 1975 Western Australian Museum See comments in text.Island records. WAM registration
number 002781
Department of 1958 Western Australian Museum Specimen presumed to be from the captive colony held at the University ofZoology, and records. WAM registration Western Australia, ex Rottnest IslandUniversity of 1959 numbers 003365 andWestern 003618Australia
Karakamia 1998 Smitz (1998) Four quokkas originally sourced from mainland locations and released withinSanctuary, a predator proof fenced sanctuary. Three quokkas released in 1996, two ofChidlow, east which were originally sourced from the Gervasse population, near Collie andof Perth held by the ‘Big Swamp Wildlife Park’, Bunbury. The third was a captive bred
young of the first two. The fourth animal sourced from the Rosella Road siteand released in 1997. One death has been recorded and at least 1 younghas been produced, population size unknown at May 2002
Harry Waring 1979 Austin (1979) Population sourced from Rottnest Island and released at the fenced HarryMarsupial Waring Marsupial Reserve. The colony failed to persist.Reserve,Jandakot
Muddy Lake, 2002 Dell and Hyder-Griffiths Unconfirmed record – See listing in Table 4south of (2002)Bunbury
Water 2004 Jim Lane (pers. comm. Unconfirmed recordCorporation to PJdeT)reserve,Dunsborough
Distribution of the quokka 73
Table 10
Locations previously known to support populations of the quokka, Setonix brachyurus, re-surveyed in the period1995–1996 by Dillon (1996) and found to show no evidence of presence or where presence was unable to beconfirmed.
Location Year of last Commentknown record
Netic Forest Block, 1978–79 0 to 2 years since last burnt. Quokkas subsequently reported from other locations within thisManjimup and neighbouring forest block (see Table 5)
Shannon, Walpole 1978–79 0 to 2 years since last burnt, swamp vegetation completely burnt. Quokkas subsequentlyreported from other locations within Shannon River area (see Table 5)
Giants Forest Block, 1978–79 Very heavy litter layer and 11 years or more since last burn. Previously known from a roadkillWalpole only at this site. Quokkas subsequently reported from other locations within this forest block
(see Tables 5 and 6)
Walpole-Nornalup 1978–79 Very heavy litter layer and 11 years or more since last burn. No recent evidence of activity.National Park (Hilltop Quokkas subsequently reported from other locations within this vicinity and may be at highRoad area), Walpole density (see text and Tables 5 and 6)
Sheepwash Forest 1978–79 6 to 10 years since last burn. No sign of recent activityBlock, Denmark
Farmland area, 1976 No evidence of activity, habitat now water logged. Location determined on basis of incorrectBunbury geographic co-ordinates for Muddy Lake listed by Kabay and Start (1976)
Urbrae Forest Block, 1988 11 years or more since last burn. Quokkas subsequently reported from other locations withinDwellingup this forest block (see Table 6)
Holyoake, 1992 3 to 5 years since last burn. Subsequently trapped more intensively (August 1995) (Table 7)Dwellingup with no evidence of quokka activity or presence
Inglehope Forest 1972–73 6 to 10 years since last burn. No sign of activity when surveyed by Dillon (1996) nor whenBlock, Dwellingup surveyed by Hayward (2002) in 1999–2000.
Table 9
Known To Be Alive (KTBA) estimates for the quokka,Setonix brachyurus, at the Gervasse Forest Block site, forthe period 1992 to 2002.
Month and Year KTBA estimateof trapping
June/July 1992 First trapping session, no KTBA estimateavailable
March 1993 14
June/July 1993 20
January 1994 22
August 1994 20
April 1995 19
January 1996 15
January 1997 12
February 1998 12
March 1999 11
February 2000 Last trapping session, no KTBA estimateavailable