Fungal consumption by marsupials in southern Tasmania

Introduction

Globally, over 600 vertebrates contribute to fungal spore dispersal (Elliott et al. 2019a, 2019b, 2022a; Nest et al. 2023) by eating fungi that form mycorrhizal associations key to plant nutrient uptake (Claridge and May 1994; Johnson 1996). Many fungal species ingested by animals form sequestrate (enclosed) fruiting bodies that are hypogeous (fruit underground). These habits have likely developed in response to animal feeding behaviours, as strong aromas released by the fungi are attractive to animals (Donaldson and Stoddart 1994; Stephens et al. 2020). In the absence of the aid of animal dispersal, most fungi disperse spores by wind (Geml et al. 2008) or move vegetatively via mycelial spread (Jonsson et al. 1999). However, only a small percentage of spores disperse beyond a few metres of the parent fruiting body (Li 2005; Galante et al. 2011) and a variety of factors can impede mycelial spread. The excavation, consumption and subsequent dispersal of fungi provided by vertebrate mycophagists is therefore a key factor in the health of mycorrhizal forest systems.

There has been significant research focus on these animal–fungal interactions within Australia (Claridge and May 1994; Nuske et al. 2017) but these associations also are widespread beyond this region. Rodents have been shown to be the most diverse group of mycophagists globally, including many Australian taxa (Vernes et al. 2015; Elliott et al. 2022a). In Australia, potoroid marsupials are the most thoroughly studied obligate mycophagists. All eight extant members of this Australian endemic family eat a diversity of fungi (Taylor 1992; Claridge and Trappe 2005; Nuske et al. 2017; Elliott et al. 2022a).

The fungal foraging activities of potoroids are important to soil disturbance and organic matter decomposition (Garkaklis et al. 2004; Newell 2009; Fleming et al. 2014; Davies et al. 2019; Palmer et al. 2020). This ecosystem service is vital to soil hydration and organic matter decomposition (Fleming et al. 2014; Davies et al. 2019; Palmer et al. 2020). While estimating soil disturbance was beyond the scope of this study, Tasmanian potoroids and other mycophagists would be expected to play a similar role in bioturbation as has been shown for related species on mainland Australia. Since colonial invasion, Australia’s potoroid mammals have suffered serious population declines and extinctions (Short and Smith 1994; Short 1998; Vernes et al. 2021). This loss has led to a reduction in spore dispersal that in turn has impacted plant and fungal communities. Given the ecological significance of services provided by members of the Potoroidae, our aim was to explore the fungal diets of these mammals in an ecosystem in which the animals remain in healthy numbers.

Materials and methods

We established two trapping sites in the South East Tasmanian bioregion: The Peter Murrell Conservation Area and State Reserve near Huntingfield, Tasmania (43°00′S, 147°18′E) and on private property near The Lea, Tasmania (42°56′08.2″S, 147°18′55.9″E). These sites are distinct and ~7 km apart. The Peter Murrell Reserves are dominated by Eucalyptus amygdalina coastal woodland with a heathy understory on sandy soils and include stands of Acacia spp. and Allocasuarina sp. The Reserves also include heathlands and some wetland areas. The site near The Lea is dominated by Eucalyptus pulchella forest and woodland, pasture (~9 ha) and some Eucalyptus globulus dry forest and woodland.

We used standard cage trapping protocols approved by the University of New England Animal Ethics Committee (Approval Number AEC18-065) and the Department of Primary Industries, Parks, Water and Environment’s Standard Operating Procedures for Live-trapping and Handling of Wild Tasmanian Mammals 2013. Traps were baited with peanut butter and vanilla on bread, and set from dusk to dawn. Captured animals were transferred to clean cloth bags to be identified, weighed and sexed. To ensure identification of recaptures, we clipped a small patch of fur from the flank of each animal for temporary identification or used subcutaneous microchips for more permanent identification. Scats were collected from each trap and placed in 70% alcohol. Traps were cleaned after each capture to prevent contamination between captures. We had 120 trap nights in the Peter Murrell Conservation Area between 26 and 29 August 2019 and 160 trap nights at The Lea site between 29 August 2019 and 1 September 2019.

Scat samples were collected from each capture and multiple pellets (from the same individual) were collected when possible to ensure accurate assessment of which fungal taxa had been eaten. Samples were wet-mounted on glass slides and analysed at 400× magnification to determine presence and species richness of fungal spores (Fig. 1). To avoid reporting incidental consumption of fungi from spores that happened to be present in the environment, only spores found to occur frequently (a minimum of five times) in each sample were considered. Spores were identified based on standard microscopic characters. Ambiguous identities were confirmed by comparison with spores of identified fungal fruiting body collections made by the authors in the same or similar habitats. We generally did not attempt to refine fungal identification beyond family or genus level. We focused on the diversity of fungi consumed and therefore apply some older nomenclatural concepts that help to outline the diversity of taxa. For example, we mention the genera Thaxterogaster and Quadrispora and our use refers to sequestrate taxa allied with Cortinarius, a speciose genus that contains both epigeous and sequestrate hypogeous species. Many recent taxonomic studies of sequestrate fungi have shown that members of historically sequestrate genera are often synonymous with epigeous genera (Gasparini 2014; Elliott and Trappe 2018; Pastor et al. 2019; Elliott et al. 2020a; Nouhra et al. 2021). We agree with these taxonomic revisions but in the context of this study we mention Thaxterogaster and Quadrispora in reference to spores of the Cortinariaceae but have characters indicating that these likely belong to sequestrate taxa. The taxa we list as Cortinarius could also be sequestrate but we have no evidence to indicate the fruiting habits.

Fig. 1.

A micrograph showing a typical view of a diversity of fungal spores in the scat of a long-nosed potoroo.

To compare fungal consumption between mammal species and sites, we conducted analysis of similarities (ANOSIM) in Primer 7 using presence/absence of fungal species in scat samples to generate Bray–Curtis similarity measures. ANOSIM returns an R-statistic that yields a measure of similarity between categories. R-values most commonly range from zero to one; the closer the R-value is to one, the more the categories differ; a value close to zero indicates that assemblages can barely be separated (Clarke and Warwick 2001). ANOSIM tests are robust to differences in the number of samples in factor groups (Clarke and Warwick 2001). To visualise multivariate patterns in vertebrate assemblage structure, multidimensional scaling was performed. Scats without fungi were excluded from the analysis, as was the single eastern barred bandicoot sample (see Table 1).

Results

Fungi were a major component of the diets of the five mammal species trapped during this study. A total of 24 fungal spore types was recognised in the scats (Table 1 and Fig. 1). We had eight captures of brush-tailed possums (Trichosurus vulpecula), one of eastern barred bandicoot (Perameles gunnii) and three of southern brown bandicoots (Isoodon obesulus) (Table 1). We had six captures of eastern bettongs (Bettongia gaimardi) and 31 captures of long-nosed potoroos (Potorous tridactylus), and every scat from these two species contained fungal spores (Table 1 and Fig. 2).

Fig. 2.

Ordination of individual mammal diets based on presence/absence of fungal taxa. Brown triangles = southern brown bandicoot, light blue circles = long-nosed potoroo, dark blue squares = eastern bettong and red inverted triangles = brush-tailed possum.

We found significant differences in fungal diet among species (ANOSIM R = 0.698, P = 0.001, Fig. 2). Composition of fungi in the long-nosed potoroo samples was well separated from that in southern brown bandicoots (R = 1.00) and brush-tailed possums (R = 0.97) but not well separated from eastern bettongs (R = 0.29). Fungi composition in eastern bettongs was well separated from that in southern brown bandicoots (R = 1.00) and moderately separated from that in brush-tailed possums (R = 0.59). Fungi composition in brush-tailed possums was moderately separated from that in southern brown bandicoots (R = 0.47).

The single brush-tailed possum sample from the Peter Murrell Reserves is clustered with that of the southern brown bandicoots from the same site (Fig. 2). The other brush-tailed possum samples are from The Lea. Eastern bettong samples are all from The Lea. The long-nosed potoroos are the only mammal for which we had sufficient data to compare between the two study sites. There was no significant difference in fungal consumption between The Lea and the Peter Murrell Reserves (R = 0.08, P = 0.23, Fig. 3).

Fig. 3.

Ordination of individual long-nosed potoroo diets based on presence/absence of fungal taxa. Red circles = Peter Murrell Reserves, blue squares = The Lea.

Discussion

This study builds on existing knowledge about mycophagy in Australia and highlights the ecological importance of conservation of potoroids and other critical weight range mammals. Mycophagy levels among potoroids have been shown to vary between seasons and with the availability of fungi (Taylor 1992; Johnson 1994; Johnson and McIlwee 1997; Tory et al. 1997; Vernes 1999; Robley et al. 2001; Vernes et al. 2015). Our study is a temporal snapshot and does not represent consumption throughout the entire year but many studies of this group have shown these to be regular fungal consumers (Nuske et al. 2017; Elliott et al. 2022a). Our study supports the idea that obligate mycophagists such as our two potoroids share similar fungal dietary components. Our observation of mycophagy among brush-tailed possums, eastern barred bandicoots and southern brown bandicoots supports more robust mycophagy data sets about the diets of these marsupials (see: Nuske et al. 2019; Elliott et al. 2022a, 2023).

Testing viability of spores after transit through the digestive systems was beyond the scope of our study. However, among the more than 40 mammal species that have been studied – including members of Peramelidae, Phalangeridae and Potoroidae – all have been shown to pass viable fungal spores and some have been shown to increase spore germination rates (Elliott et al. 2022a). Although we did not test for evidence of detrimental degradation of spores, no evidence of this was observed.

Accurately estimating spore dispersal potential requires data on digestive transit time and telemetry data on movement. These data were lacking, therefore we could not model dispersal potential in this study. However, for comparison, modeling of swamp wallabies (Wallabia bicolor) has shown that these animals frequently disperse mycorrhizal fungal spores hundreds of metres and possibly as far as 1265 m from the point of consumption (Danks et al. 2020). Dispersal estimates based on home range size indicate that some murid rodents could also move spores hundreds and possibly over 1000 m (Elliott et al. 2020b, 2022b), and bandicoots likely hundreds of metres (Elliott et al. 2022c).

Spores often linger longer than other dietary items in digestive systems and therefore have a longer period of potential dispersal (Danks 2012). Ultimately, spore dispersal potential is dependent upon movement patterns, home range size and gut passage rate. No studies have assessed passage rate of fungal spores through potoroid digestive systems but based on what is known of passage rates of other animals discussed in the aforementioned studies, maximum retention times are likely to be 2 days. Mean home range sizes vary significantly between the two potoroids in this study. Mean home range size is 61 ha for the eastern bettong (Taylor 1993), and 4 ha among males and 1.9 ha among females for long-nosed potoroos (Long 2001). The differences in home range size suggest that bettongs may contribute to longer distance dispersal than potoroos. Another variable not considered in this study is secondary dispersal that occurs when these prey species are ingested by a predator or scavenger. In mainland Australia, dingoes (Canis familiaris) contribute to spore dispersal by eating primary mycophagists (including long-nosed potoroos), with modelling suggesting that dingoes potentially disperse fungal spores over more than 10 km (Elliott et al. 2024). Avian raptors, Tasmanian devils (Sarcophilus harrisii) and quolls (Dasyurus spp.) likely provide – and at one point the Tasmanian tiger (Thylacinus cynocephalus) provided – a similar function of long-distance spore dispersal.

Our data show that possums, bandicoots, bettongs and potoroos consume a diversity of mycorrhizal fungi in Tasmania. We hope this study will stimulate further research into viability, passage rates and dispersal estimation of fungal spores, and provide an additional ecological rationale for the value of conserving Australian mammal communities.