The parasites of free-ranging terrestrial wildlife from Australia’s south-west

Up to 1 mL of blood was collected from the lateral caudal (tail) vein in B. penicillata, and either the jugular or tail vein in other marsupials, into EDTA MiniCollect tubes (Greiner Bio-One, Germany), and frozen at −20°C before processing. In T. rugosa, up to 0.5 mL of blood was collected from the ventral coccygeal vein. Peripheral blood smears were prepared in the field using the ‘wedge’ method (see Clark et al. 2004) and stained with Diff-Quik prior to mounting with dibutylphthalate polystyrene xylene. Blood smears from a subset of Trypanosoma-positive hosts (B. penicillata and T. v. hypoleucus only) were inspected by light microscopy for the presence of haemoprotozoans. Trypanosoma spp. were the main haemoparasites monitored during this study, though the seroprevalence of T. gondii was also determined in a subset of B. penicillata. Trypanosome detection methods (PCR and Sanger sequencing) have been described by Northover et al. (2019a). While clade specific primers were used to identify T. vegrandis, T. copemani and T. noyesi, Sanger sequencing was used to identify T. sp. ANU2 and T. gilletti, as species specific primers for these trypanosomes were not available. Targeted amplicon next generation sequencing of the Trypanosoma 18S rDNA loci (see Cooper et al. 2018) was performed on samples in which species identification could not be achieved with Sanger sequencing.

A commercial modified agglutination test (Toxo-Screen DA, bioMerieux, France) was used as per the manufacturer’s instructions to determine the seroprevalence of T. gondii. In this test, formalin-treated T. gondii antigen was added to the sera and agglutination was expected to occur in the presence of specific IgG antibodies (Couzineau and Baufine-Ducrocq 1970). A denaturing agent (2-mercaptoethanol) was used to avoid agglutination by non-specific IgM antibodies (Couzineau and Baufine-Ducrocq 1970). Each sample used two dilutions (1/40 and 1/4000) and every plate contained a positive, negative, and antigen control. Results were interpreted as per the manufacturer’s guidelines, whereby a seropositive result presented as an agglutination mat covering more than half of the well, and a seronegative result presented as a red button. Borderline reactions were retested.

Newspaper was placed under each trap to collect faeces. Faecal samples were preserved in 10% buffered formalin and refrigerated at 4°C until processing. Faeces were examined for the presence of gastrointestinal parasite eggs, oocysts, and larvae, using the modified sodium nitrate (NaNO₃) faecal flotation protocol described by Northover et al. (2017); see Appendices (Figs 4–7) for criteria used to differentiate gastrointestinal parasite taxa.

Each individual was examined in a systematic manner for the presence of ectoparasites (fleas, lice, mites and ticks). A subset of the different ectoparasites observed at the time of sampling were collected from each animal and stored in 70% ethanol. Ectoparasites were morphologically identified by Murdoch University parasitology staff using keys developed by Roberts (1970), von Kéler (1971), Dunnet and Mardon (1974), and Domrow (1987).

The Trypanosoma spp. (and piroplasms) infecting B. penicillata have previously been documented, based on the analysis of 1211 blood samples from 596 individuals (Northover et al. 2019a). Here we analysed an additional 194 blood samples from 151 (119 not previously sampled) individuals for the presence of trypanosomes. With both datasets combined (1405 blood samples from 715 individuals) 54.0% of samples were Trypanosoma-positive (Table 1); 80 samples could not be identified to species. Prevalence of infection among individuals (i.e. a positive result at any time point during the monitoring period) was 65.3%. Of the remaining 1325 samples, five Trypanosoma species (T. vegrandis, T. copemani, T. noyesi, T. gilletti and T. sp. ANU2) were detected (Table 1). Trypanosoma vegrandis was most prevalent, followed by T. copemani, T. noyesi, T. sp. ANU2 and T. gilletti (Table 1).

Of the Trypanosoma-positive samples that could be identified to species (n = 679), the prevalence of Trypanosoma spp. coinfection was 23.1% (95% CI: 20.1–26.4%), with up to three Trypanosoma species detected in a single host (Appendix Fig. 1). Trypanosoma prevalence and diversity varied between sites, with the highest overall prevalence detected within Walcott and the lowest species diversity within Dryandra (Appendix Table 1); T. sp. ANU2 and T. gilletti were not detected within Dryandra. Bettongia penicillata translocated into Dryandra (UWR origin) were also negative for T. gilletti. Both trypanosomes and piroplasms (Theileria and Babesia spp.) were observed in peripheral blood smears from B. penicillata (Fig. 2).

Fig. 2.

Haemoparasites identified in peripheral blood smears from woylies (Bettongia penicillata): (a) Trypanosoma copemani; (b) Theileria sp. (black arrow); (c) Babesia sp. (black arrows).

A subset of B. penicillata blood samples (n = 617) collected from 349 individuals was also screened for T. gondii, of which 11 samples (1.8%) were positive. Of these, five were from Warrup East (0.8%), four were from Walcott (0.6%; includes two positive results from the same host at different time points) and two were from Dryandra (0.3%).

In total, 317 blood samples collected from 226 individuals were screened for trypanosomes, of which 262 samples (82.6%) were positive (Table 1). Four samples that were positive on general PCR screening, could not be identified to species. Prevalence of infection among individuals was 86.7%. Four Trypanosoma species were detected in T. v. hypoleucus (T. copemani, T. vegrandis, T. noyesi and T. sp. ANU2) (Table 1), including up to three Trypanosoma species in a single host (all T. copemani, T. vegrandis, T. noyesi coinfection) (Appendix Fig. 2). Overall, T. copemani was most prevalent, followed by T. vegrandis and T. noyesi (Table 1). Trypanosoma sp. ANU2 was detected in only three hosts from Warrup East (Appendix Table 2). Coinfection with two or more trypanosomes was detected in 26.0% (95% CI: 21.0–31.7%) of positive samples. Warrup East had the greatest Trypanosoma spp. prevalence (100%) and diversity (Appendix Table 2). An unknown Hepatozoon species was identified in three blood smears from T. v. hypoleucus (Fig. 3); all three hosts originated from Dryandra and Trypanosoma spp. infection (T. copemani and T. vegrandis coinfection in two cases; T. copemani in the other) was also apparent. Trypanosomes and unidentified piroplasms were also detected in blood smears (Fig. 3). Four (UWR) T. v. hypoleucus that were screened for T. gondii were seronegative.

Fig. 3.

Unidentified Hepatozoon spp. (a–c), a piroplasm (c; black arrow) and a Trypanosoma sp. (d) identified in blood smears from brush-tailed possums (Trichosurus vulpecula hypoleucus).

We screened 119 blood samples from 97 individuals for the presence of trypanosomes. Of these, 14 samples (11.8%) were positive (Table 1); one sample could not be identified to species. Prevalence of infection among individuals was 14.4%. Four Trypanosoma species were identified in D. geoffroii (Table 1), including up to three Trypanosoma species in a single host (Appendix Fig. 3). Trypanosoma copemani and T. noyesi were equally most prevalent, followed by T. vegrandis and T. sp. ANU2 (Table 1). Coinfection was detected in 46.2% (95% CI: 20.4–73.9%) of positive samples. Despite the high number of D. geoffroii sampled within Dryandra (n = 54), only one host was Trypanosoma-positive (T. noyesi). Except for T. sp. ANU2, which was detected in a single host from Warrup East, Walcott had the highest prevalence of Trypanosoma spp. infection (Appendix Table 3). Trypanosoma vegrandis was not detected in D. geoffroii from Warrup East or Dryandra, despite Warrup East (and Walcott) having high Trypanosoma spp. diversity.

Seven T. rugosa were screened for the presence of trypanosomes, three of which (42.9%) were positive; Amblyomma albolimbatum ticks were parasitising two of the Trypanosoma-positive hosts, the third had no visible ectoparasites. Trypanosoma noyesi was sequenced from two of the positive samples (Dryandra origin). The third positive sample (UWR origin) could not be interpreted with confidence due to difficulties encountered during the extraction process, thus species-level identification was not pursued. A single feral cat (Felis catus) captured during trapping in Warrup East tested negative for Trypanosoma spp. on general PCR screening. Trypanosoma vegrandis was sequenced in a quenda or south-western brown bandicoot (Isoodon fusciventer) from Perup Sanctuary; this was the only I. fusciventer screened for trypanosomes.

We examined 946 faecal samples collected from 377 individuals for the presence of gastrointestinal parasites. Five visually distinctive nematode eggs (typical strongyle, Strongyloides-like, Potoroxyuris sp., Trichuris sp. and Linstowinema sp.), a single type of cestode egg (undescribed species), coccidian oocysts (including Eimeria woyliei), and first stage (metastrongyloid) lungworm larvae were identified (Table 2, Appendix Fig. 4). Eimeria woyliei was identified in cases where sporulated oocysts were present and distinguishing morphological characteristics were observed (see Northover et al. 2019c); coccidian oocyst counts (Table 2) include E. woyliei. Unidentified nematode eggs/larvae were also found and may represent either infective nematode larvae or free-living nematodes.

Overall, 90.8% of samples contained at least one type of gastrointestinal parasite (including unidentified nematode eggs/larvae). Prevalence of infection among individuals was 95.2%. Excluding unidentified nematode eggs/larvae, strongyle eggs were most prevalent, with other parasite taxa detected less frequently (Table 2). Of the samples that were negative, the majority (92.1%) were from Dryandra. Except for cestode eggs, which were only detected in hosts from Dryandra, the prevalence and diversity of gastrointestinal parasite taxa was lowest within this site (Appendix Table 4). Linstowinema sp. eggs were found in just two hosts, while strongyle eggs, Strongyloides-like eggs and coccidian oocysts were ubiquitous across all sites. Potoroxyuris sp., Trichuris sp. eggs and lungworm larvae were only identified in hosts within or originating from Perup Sanctuary.

We examined 234 faecal samples from 112 hosts and detected four distinct types of nematode eggs (typical strongyle, Parastrongyloides sp., Protospirura sp. and oxyurid), a single type of anoplocephalid egg (Bertiella sp.), coccidian oocysts and unidentified nematode eggs/larvae (Table 2, Appendix Fig. 5). Distinctive, large gravid nematodes (approximately 200 μm in length) were also frequently observed (Appendix Fig. 5); we suspect these are free-living spp. based on their reproductive anatomy (Chitwood and Chitwood 1937). Overall, gastrointestinal parasites were identified in 74.4% of samples, and strongyle eggs were the most prevalent taxon detected (Table 2). Prevalence of infection among individuals was 92.0%. Species diversity was similar across all sites, with prevalences generally lower in Dryandra, except for coccidian oocysts and Bertiella sp. (Appendix Table 5). Parastrongyloides eggs were only found in hosts from the UWR, whereas strongyle eggs and coccidian oocysts were detected in all sites (Appendix Table 5).

We examined 112 faecal samples collected from 80 individuals and identified typical strongyle eggs, oxyurid eggs, spiruroid eggs (including Physaloptera sp.), two distinct anoplocephalid eggs [i.e. an unidentified sp. (Anoplocephalid egg) and Bertiella sp.], a Capillaria sp. egg, coccidian oocysts and unidentified nematode eggs/larvae (Table 2, Appendix Fig. 6). We also found a distinct morphotype of strongylid nematode egg (unidentified species referred to as ‘megastrongyle’ egg), which was much larger (approximately 150 μm × 75 μm) than typical strongyle eggs (Appendix Fig. 6). Overall, gastrointestinal parasites were present in 88.4% of samples, with strongyle eggs detected in 82.1% of samples (Table 2). Prevalence of infection among individuals was 90.0%. Species diversity was lowest in Walcott, with prevalences generally lower in Dryandra, except for Bertiella sp. (Appendix Table 6). Strongyle eggs, ‘megastrongyle’ eggs and coccidian oocysts were ubiquitous across all sites (Appendix Table 6).

Twelve faecal samples from I. fusciventer were screened for gastrointestinal parasites. Typical strongyle eggs, coccidian oocysts, Strongyloides-like eggs, lungworm (L1) larvae, Trichuris sp. eggs, Linstowinema sp. eggs, Labiobulura sp. eggs, Potorolepis sp. eggs, an Australiformis sp. egg and unidentified nematode eggs/larvae were observed (Table 2, Appendix Fig. 7). All 12 samples were positive for parasites, with strongyle eggs detected in all but two samples; coccidia were detected in 50% of samples. Two faecal samples from P. t. wambenger (Walcott origin) were also examined; both were positive for strongyle eggs; a Linstowinema sp. egg was detected in the faeces from one host and unidentified nematode larvae were present in the other (Appendix Fig. 8). Faecal samples from T. rugosa were not examined.

In total, 923 samples collected from 404 individuals were morphologically identified. Five species of tick (Amblyomma triguttatum, Ixodes australiensis, Ixodes myrmecobii, Ixodes tasmani and Ixodes woyliei), seven species of flea (Acidesta chera, Choristopsylla ochi, Echidnophaga myrmecobii, Echidnophaga perilis, Pygiopsylla hilli, Pygiopsylla tunneyi and Stephanocircus dasyuri), two species of louse (Boopia uncinata and Paraheterodoxus calcaratus), and four species of mite (Androlaelaps fahrenholzi, Haemolaelaps hattanae, Haemolaelaps marsupialis and Haemolaelaps quartus) were identified (Table 3). Larval/nymph stages of ticks and mites were also frequently encountered, but in most instances could not be identified to species. Trombiculid mites (Acari: Trombiculidae) were detected (based on their distinctive macroscopic appearance; Fig. 4a) on five hosts [Dryandra (origin UWR) n = 1; Perup Sanctuary n = 4] but were not collected for morphological identification.

Fig. 4.

Trombiculid mites on a woylie (Bettongia penicillata) (a; black arrow), ‘stick-fast’ fleas (Echidnophaga myrmecobii) on a brush-tailed possum (Trichosurus vulpecula hypoleucus) (b), and trombiculid mites on a chuditch (Dasyurus geoffroii) (c).

Of the 1314 B. penicillata (641 individuals) that were examined for the presence of ectoparasites in the field, 95.2% (95% CI: 93.9–96.2%) were positive for at least one ectoparasite taxon. Prevalence of infection among individuals was 98.0%. Fleas were recorded on only six hosts from Dryandra (one translocated, five resident); however, specimens were not collected for identification. Of the four ectoparasite taxonomic groups (ticks, fleas, lice and mites), ticks were most common (Appendix Table 7). From these groups, I. australiensis, P. tunneyi, P. calcaratus and H. hattanae were most prevalent, respectively; however, site differences were apparent (Appendix Table 8). For example, I. australiensis was not detected in Dryandra, despite this tick being the most prevalent species overall. In contrast, the prevalence of I. woyliei and Amblyomma spp. ticks was highest within Dryandra (Appendix Table 8), despite this site having the lowest overall prevalence of ticks (Appendix Table 7).

We morphologically identified 194 ectoparasite samples from 138 individuals. Seven species of tick (Amblyomma sp., Haemaphysalis bancrofti, Haemaphysalis bremneri, Haemaphysalis humerosa, I. australiensis, I. myrmecobii and I. tasmani), four species of flea (C. ochi, E. myrmecobii, P. tunneyi and S. dasyuri), one species of louse (P. calcaratus), and six mite taxa (H. hattanae, H. marsupialis, Liponyssoides lukoschusi, Ornithonyssus praedo, trombiculid mites and Ulyxes penelope) were identified (Table 3). Unidentified larval/nymph stages of ticks and mites were also common. ‘Stick-fast’ fleas were identified in the field based on their characteristic appearance embedded in the skin on seven hosts from Perup Sanctuary (Fig. 4b; confirmed as E. myrmecobii in five hosts).

Of the 323 T. v. hypoleucus (230 individuals) that were assessed for the presence of ectoparasites in the field, 81.1% (95% CI: 76.5–85.0%) were host to at least one ectoparasite taxon. Prevalence of infection among individuals was 88.7%. Of the three major ectoparasite taxonomic groups found on T. v. hypoleucus (ticks, fleas and mites), ticks were most common (Appendix Table 9). From these groups, I. tasmani, C. ochi, and trombiculid mites were most prevalent, respectively, though site differences were evident (Appendix Table 10).

Eighty-four ectoparasite samples collected from 80 individuals were morphologically identified. All four ectoparasite taxa were present on D. geoffroii, including three species of tick (I. australiensis, Ixodes fecialis and I. tasmani), six species of flea (A. chera, C. ochi, E. myrmecobii, E. perilis, P. tunneyi and S. dasyuri), two species of louse (B. uncinata and P. calcaratus), and four mite taxa (H. hattanae, H. marsupialis, Ornithonyssus dasyuri and trombiculid mites) (Table 3); unidentified larval/nymph stages of ticks and mites were also present.

Of the 120 D. geoffroii (99 individuals) that were examined for the presence of ectoparasites, at least one ectoparasite taxon was detected in 74.2% (95% CI: 65.6–81.2%) of cases. Prevalence of infection among individuals was 84.8%. Of the four major ectoparasite taxonomic groups, mites were the most common (Appendix Table 11). From these groups, I. australiensis, P. tunneyi, B. uncinata and trombiculid mites (Fig. 4c) were most prevalent, respectively (Appendix Table 12). Overall, the prevalence and diversity of ectoparasites was lowest in Dryandra (Appendix Tables 11 and 12).

Ticks were collected from seven T. rugosa and morphologically identified as Amblyomma albolimbatum in six hosts, and Amblyomma sp. in the other host. Fleas, lice and mites were not observed on T. rugosa. The fleas S. dasyuri and P. tunneyi were present on three I. fusciventer (origin Perup, Walcott and Warrup East); mites and ticks were also recorded as present on one of these hosts (Walcott) but were not collected for morphological identification. Amblyomma sp., I. australiensis and Ixodes sp. ticks were isolated from a P. t. wambenger.

Host–ectoparasite associations not previously reported in B. penicillata (von Kéler 1971; Dunnet and Mardon 1974; Domrow 1987; Kaewmongkol et al. 2011; Ash et al. 2017; Thompson et al. 2018; Krige et al. 2021a) include A. chera, C. ochi, B. uncinata, H. marsupialis, A. fahrenholzi and trombiculid mites. Pygiopsylla tunneyi was the most common flea species identified on B. penicillata during this study. In contrast, B. penicillata may be an accidental host for A. chera, C. ochi and H. marsupialis as the nominal host [I. obesulus for A. chera (Dunnet and Mardon 1974) and H. marsupialis (Domrow 1987); and T. vulpecula for C. ochi (Dunnet and Mardon 1974)] was captured beforehand, making bag contamination a possibility. The presence of A. fahrenholzi on a single host from Warrup East is intriguing, as birds or their nests are the only known hosts within (south-eastern) Australia (Domrow 1987). Outside of Australia A. fahrenholzi is known to parasitise a wide range of hosts, in particular rodents (Silva-de la Fuente et al. 2020), but there is no record of this parasite in WA until now. Dasyurus species are the predominant host of B. uncinata (von Kéler 1971); however, we did not detect B. uncinata on D. geoffroii in Dryandra, and only B. penicillata were captured on the day B. uncinata was detected. While this is the first published record of trombiculid (larval stage) mites on B. penicillata, specimens were not obtained to confirm the species.

Stark site variation in parasite type and prevalence were also apparent. For example, I. australiensis and P. tunneyi were not detected on B. penicillata from Dryandra, despite these species being the most prevalent overall within their respective taxon groups. Ixodes australiensis was not found on sympatric species within Dryandra either, and P. tunneyi was only detected on a single host (T. v. hypoleucus) from Dryandra. In contrast, the host-specific tick I. woyliei was most prevalent within Dryandra, which is unexpected given the lack of I. australiensis and low prevalence of I. tasmani (n = 1), as these three species are considered sympatric, sharing the same geographical region, habitat and host (Ash et al. 2017). Fleas were also extremely rare (excluding C. ochi on T. v. hypoleucus and E. myrmecobii on D. geoffroii) within Dryandra. Encouragingly, the threatened host-specific flea, P. hilli (Kwak 2018), is persisting in all UWR sites. The critically endangered status and current reliance of B. penicillata upon periodic translocations, however, places this density dependent and site-specific coendangered parasite at risk of extinction.

Host–ectoparasite associations not previously reported in T. vulpecula (Roberts 1970; Dunnet and Mardon 1974; Viggers and Spratt 1995; Burmej et al. 2008; Clark 2011; Hillman 2016; Krige et al. 2021a) include P. calcaratus, H. hattanae, H. marsupialis, O. praedo and U. penelope. Additionally, this is the first report of P. tunneyi, H. bancrofti, H. bremneri, H. humerosa, L. lukoschusi and trombiculid mites on T. v. hypoleucus from south-western WA.) While H. bremneri (and another Haemaphysalis spp. tick) have been isolated from a T. vulpecula in the greater Perth region (Hillman 2016), H. humerosa and H. bancrofti have not been identified on T. vulpecula from WA (Roberts 1970). We identified H. humerosa on two hosts from Warrup East, one of which had two specimens. Previous studies (Roberts 1963) suggest that bandicoots are the primary host species for H. humerosa. Although the host with two H. humerosa ticks was captured after an I. fusciventer, the other host was not. We also found H. bancrofti (and H. bremneri) on T. v. hypoleucus from Dryandra. Haemaphysalis bancrofti has been reported from T. vulpecula in eastern Australia (Roberts 1963, 1970; Viggers and Spratt 1995; Laan et al. 2011), but this is the first report in WA.

The presence of P. calcaratus on T. v. hypoleucus (n = 2) from Walcott is unusual, as no lice have been recorded on T. vulpecula (Johnson and Hemsley 2008). While it is likely that T. v. hypoleucus is an accidental host (i.e. P. calcaratus-positive B. penicillata captured beforehand in one instance), we cannot rule out T. v. hypoleucus being an alternative host; particularly when lice (unidentified species) were recorded on a further six hosts in the field (Appendix Table 9).

The presence of six mite taxa is noteworthy. Haemolaelaps hattanae was found on four hosts (from all sites), though its presence could be attributed to bag contamination in two cases. The presence of two Haemolaelaps mite species and U. penelope (previously Haemolaelaps penelope; Shaw 2014), however, may indicate that T. v. hypoleucus is an alternative host. Haemolaelaps sisyphus has been recorded from T. vulpecula in NSW and Qld (Viggers and Spratt 1995). Ornithonyssus praedo has only been documented on other marsupials (Domrow 1987), and birds are the recognised host of L. lukoschusi (Domrow 1987). As we identified L. lukoschusi on six hosts from Dryandra, and Hillman et al. (2018) found a Liponyssoides sp. mite on T. vulpecula (n = 1) from the greater Perth region, T. v. hypoleucus may be an alternative host, though its arboreal nature may expose it to such parasites. The presence of O. praedo on a single host is more difficult to interpret, though no known hosts (i.e. P. tapoatafa: Domrow 1987) were captured during this trapping session.

As for B. penicillata, site differences in ectoparasite prevalence and diversity were evident. The prevalence of Ixodes spp. ticks was lowest within Dryandra; neither I. australiensis nor I. myrmecobii were detected on T. v. hypoleucus within this site. In contrast, L. lukoschusi, H. bancrofti and H. bremneri were detected only on T. v. hypoleucus within Dryandra, and the prevalence of C. ochi, trombiculid mites and Amblyomma spp. ticks was highest within this site. The majority of Haemolaelaps spp. mites (5/7) were isolated from T. v. hypoleucus in Walcott (as was S. dasyuri), whereas O. praedo was found only in Warrup East. Amblyomma spp. ticks, I. tasmani, C. ochi, P. tunneyi and H. hattanae were ubiquitous across sites.

Host–ectoparasite associations not previously reported in D. geoffroii (von Kéler 1971; Haigh 1994; Burmej et al. 2008; Thomasz 2014) include I. australiensis, I. tasmani, C. ochi, P. calcaratus, H. hattanae, H. marsupialis and O. dasyuri. Ixodes australiensis was detected on 17 animals and had the highest prevalence of any tick species, indicating that D. geoffroii is a true host. Likewise, the presence of I. tasmani on D. geoffroii across all sites suggests a legitimate host–parasite association. In contrast, the presence of C. ochi in a single host from Warrup East is questionable. Choristopsylla ochi is commonly found in arboreal species such as T. vulpecula (Dunnet and Mardon 1974) and although D. geoffroii may be an accidental host, its semiarboreal nature may expose it to such parasites. Bettongia penicillata was the only other animal captured before this host (using the same handling bag) on that day. Paraheterodoxus calcaratus has not been recorded in D. geoffroii. We believe D. geoffroii is an accidental host of P. calcaratus as two of the three hosts detected with P. calcaratus were captured immediately after B. penicillata infected with P. calcaratus using the same handling bag. The presence of four mite species was surprising. Although we detected a low prevalence of H. hattanae, H. marsupialis and O. dasyuri, O. dasyuri has been found on the eastern quoll (Dasyurus viverrinus) (Domrow 1987) and both Haemolaelaps spp. were found on D. geoffroii at multiple (UWR) sites. Trombiculid mites had the highest prevalence of all mite taxa and were observed on D. geoffroii at all sites during this study.

As for other host species, site variation in parasite prevalence and diversity was identified. For instance, nine parasite taxa that were present on D. geoffroii from Warrup East and Walcott were not detected on D. geoffroii within Dryandra. This includes I. australiensis, P. tunneyi and B. uncinata, which were the most prevalent species identified within their respective ectoparasite taxa groups. The opposite trend was observed for trombiculid mites and E. myrmecobii, which had the highest prevalence within Dryandra; O. dasyuri (n = 1) was detected only within this site. In comparison to B. penicillata and T. v. hypoleucus, in which ticks were the most prevalent ectoparasite found, mites were most prevalent on D. geoffroii.

Ticks collected from T. rugosa were identified as A. albolimbatum on six hosts, and Amblyomma sp. in the other host. Tiliqua rugosa is the primary host of A. albolimbatum (Sharrad and King 1981; Barker and Walker 2014). The two flea species found on I. fusciventer (S. dasyuri and P. tunneyi) have been previously documented on I. obesulus from WA (Dunnet and Mardon 1974; Hillman 2016). In contrast, this is the first report of I. australiensis and Amblyomma sp. ticks from P. t. wambenger. Phascogale tapoatafa wambenger is a known host of other Ixodes spp. ticks (Roberts 1970) and Amblyomma spp. ticks have a wide host range, including domestic animals and humans (Roberts 1970; Waudby et al. 2007). As we only collected a larval Amblyomma specimen from this host, we were unable to identify the species.

For all host species, many of the ectoparasites collected were larval stages of ticks and mites, which could not be identified to species using morphological keys. Unidentified ticks had the highest prevalence within their taxonomic group. Molecular techniques could be used in future to determine the species of ectoparasite and provide more accurate prevalence data. Introducing measures to reduce the risk of cross contamination will also help to reduce the likelihood of spurious host–ectoparasite associations. For example, single use handling bags or providing handling bags and equipment for each species.