(Egernia slateri): An Endangered Australian Desert ...

4 downloads 0 Views 307KB Size Report
Acacia victoriae and Senna artemisioides (Table 3). Mounded shrubs occupied by E. slateri had a mean of 3.7 burrows per shrub, with a range of. 1 to 11.
Foraging Ecology and Habitat Use of Slater's Skink (Egernia slateri): An Endangered Australian Desert Lizard Author(s): Chris R. Pavey, Chris J. Burwell, and Catherine E. M. Nano Source: Journal of Herpetology, 44(4):563-571. 2010. Published By: The Society for the Study of Amphibians and Reptiles DOI: 10.1670/09-102.1 URL: http://www.bioone.org/doi/full/10.1670/09-102.1

BioOne (www.bioone.org) is an electronic aggregator of bioscience research content, and the online home to over 160 journals and books published by not-for-profit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

Journal of Herpetology, Vol. 44, No. 4, pp. 563–571, 2010 Copyright 2010 Society for the Study of Amphibians and Reptiles

Foraging Ecology and Habitat Use of Slater’s Skink (Egernia slateri): An Endangered Australian Desert Lizard CHRIS R. PAVEY,1,2,3,4 CHRIS J. BURWELL,3,5

AND

CATHERINE E. M. NANO1

1

Biodiversity Unit, Department of Natural Resources, Environment, the Arts and Sport, PO Box 1120, Alice Springs, Northern Territory 0871, Australia 2 Centre for Behavioural and Physiological Ecology, Zoology, University of New England, Armidale, New South Wales 2351, Australia 3 Queensland Museum, PO Box 3300, South Brisbane, Queensland 4101, Australia 5 Environmental Futures Centre and Griffith School of Environment, Griffith University, Nathan, Queensland 4111, Australia

ABSTRACT.—The nominate subspecies of Slater’s Skink, Egernia slateri, is nationally endangered with a severely restricted range on desert river floodplains in central Australia. Here, we provide the first details of foraging ecology and habitat use of one of the remaining wild populations. Ants (35.0% volume, 95.5% occurrence) and termites (Isoptera, 23.3% volume, 63.6% occurrence) were the major prey of adult E. slateri. Other important prey included Coleoptera, Orthoptera, and spiders. Egernia slateri depended on soil mounds formed at the base of shrubs as habitat in which to dig burrows. Ninety-five percent of shrubs in the genus Eremophila sampled in two subpopulations were mounded and 58% of the mounded shrubs had lizard burrows. Behavioral observations indicate that E. slateri is largely solitary with a group size of one (94% of bouts), except for three observations of adult–juvenile associations. Animals showed diurnal activity with basking individuals present up to 327 min postsunrise. The study reveals E. slateri as a species occurring in small populations in limited habitat. It faces imminent threats to its persistence as a result of high disturbance regimes on floodplains in central Australia and is considered one of Australia’s most threatened reptiles.

Egernia is an endemic Australian genus that currently consists of 30 species of medium-tolarge lizards (Chapple et al., 2004). Within the genus, the Egernia whitii species-group forms a distinct clade of 11 species (Chapple et al., 2004) that some authorities regard as a distinct genus, Liopholis (Gardener et al., 2008). The E. whitii species-group includes five species that are restricted to arid and semiarid Australia. Of these species, Slater’s Skink (Egernia slateri) has a severely restricted range, is poorly known, and until May 2004 was considered likely to be extinct (Pavey, 2004). Although the phylogeny, based on molecular analysis, and aspects of the ecology of the other 10 species in the whitiigroup are relatively well known (Chapple, 2003; Chapple and Keogh, 2004; Chapple et al., 2004), very little information is available on the ecology of E. slateri. Egernia s. slateri is listed as endangered nationally under the Environment Protection and Biodiversity Conservation Act and is restricted to the southern arid regions of the Northern Territory (Storr, 1968; Horner, 1992). It has undergone a significant decline over the 4 Corresponding Author. E-mail: chris.pavey@nt. gov.au

past 40 yr, largely in response to changes in land use (Pavey, 2004). The other subspecies, E. s. virgata, is known from only a handful of specimens collected from far northern South Australia, all before 1915 (Storr, 1968). The latter subspecies has not been assessed nationally but is listed as endangered under South Australian legislation (Chapple, 2003). In contrast to three of the other arid-restricted species in the E. whitii species-group (E. kintorei, E. inornata, and E. striata), which occur on sandy soils (Wilson and Swan, 2008), E. s. slateri occupies shrubland and open shrubland on alluvial soils close to drainage lines (Pavey, 2007) and has been negatively impacted by disturbance to floodplain environments (Pavey, 2004). It is a burrowing species that constructs multientrance burrows in the pedestal of soil that builds up around some shrubs and small trees especially native fuschia (Eremophila) and corkwood (Hakea) (Pavey, 2004). At all recorded localities of E. slateri, wind-carried soil accumulates at the bases of shrubs and trees, resulting in the formation of these soil pedestals to a height of $10 cm above the ground (Henzell, 1972; CRP, unpubl. data). The lack of information on the ecology of E. slateri largely results from an absence of records from the mid-1970s onward, including its

564

C. R. PAVEY ET AL.

disappearance from sites where it was formerly abundant (Pavey, 2007). However, E. s. slateri is currently the focus of a conservation program implementing the national recovery plan adopted in 2004 (Pavey, 2004). During field surveys undertaken between 2004 and 2008, seven small and highly fragmented populations of E. s. slateri were located within the MacDonnell Ranges and Finke bioregions of central Australia. These findings provided the opportunity to assess the ecology of the species for the first time. We studied the diet, habitat use and behavior of the species as a basis for understanding reasons for its decline and implementing management programs. Our research also furthered our understanding of critical aspects of the biology and ecology of the genus Egernia as a whole, including life-history characteristics, reproduction, degree of sociality, and diet (Chapple, 2003). MATERIALS AND METHODS Study Site.—The study site was located in the footslopes of the Waterhouse Range at Owen Springs Reserve within the MacDonnell Ranges bioregion, west of Alice Springs. Despite extensive searches, only four small subpopulations were located at the site within an area of ,10 ha. Field observations were carried out at two of these subpopulations, which were separated from each other by 142 m at the closest point. Sexing and Ageing.—Reliable identification of the sexes in Egernia requires inspection for the presence or absence of hemipenes (e.g., Duffield and Bull, 2002). Male and female E. slateri could not be reliably separated by external characters in the field. Egernia live to $10 yr in the wild and typically reach maturity at 2 to 3 yr of age (Chapple, 2003). As a consequence of this life history, we recognized three age classes: juvenile (,6 months of age), subadult (.6 months of age but not yet adult size), and adult (snout–vent length [SVL] of 85–97 mm). Fecal Analysis.—We assessed diet by analyzing prey material in 44 scats of adult or subadult E. slateri. This species deposits scats close to burrow entrances, and these scats accumulate into small scat piles. This enabled the collection of moderate numbers of scats for a noninvasive assessment of the diet of the species. We identified scats as belonging to E. slateri based on their location close to occupied burrows, size, and occurrence in scat piles. No other lizard in the study area piles scats of this size. Scats of adults were separated from those of juveniles by their larger size; however, scats of subadults could not be reliably separated from

those of adults. Only a small number of juvenile scats were collected (N 5 10); therefore, we present data from analysis only of adult and subadult scats. Material was stored in an airtight container and dried at 200uC for 24 h before analysis. Because of the rarity and endangered status of the species, care was taken in searching for scats in the vicinity of active burrows. Scat collection took place in late summer (March) and early autumn (May) 2006. We placed each scat in a petri dish, soaked it with four to five drops of 10% KOH, teased it apart with jeweller’s forceps, and then covered it in 70% ethanol. We systematically searched each scat for identifiable material under a low power (36.4–40) binocular microscope. Prey fragments were identified to the lowest taxonomic level possible. Material was identified by reference to collections in the Queensland Museum, Brisbane. We estimated the percentage volume that each prey category contributed to identifiable material within a scat. We spread all the identifiable fragments in a petri dish with graph paper underneath and estimated for each prey item the space (area) occupied by its fragments (including insect exoskeleton, wing fragments, scales, and plant material). Percentage volume was estimated to the nearest 5%. Taxa that contributed ,2.5% were not included in percentage volume estimates. We then calculated the mean percentage volume of each prey category in the diet of E. slateri by summing the scores and dividing by the number of scats analyzed (44). The number of fragments of each prey category was counted or estimated for large numbers of fragments. We also recorded the presence–absence of each prey category in each scat. Percentage occurrence (also known as percentage frequency) was measured as the number of scats containing the specific prey taxon divided by 44 (the total number of scats analyzed). Behavioral Observations.—Observations of the behavior of individual E. slateri were made at both subpopulations in February and March 2006 and March 2007 by a single observer. Fourteen observation sessions (10 in 2006, four in 2007) were carried out over 12 days, with a total of 2,325 min (38.75 h) spent searching for and observing lizards. Most search effort was concentrated within the 4 h postsunrise and 3 h presunset. A standardized searching method was developed to detect lizards. The observer initially looked for active or basking animals by scanning each shrub with a soil pedestal (hereafter referred to as ‘‘mounded shrubs’’) that contained burrows from a distance of .10 m by using a pair of 10 3 40 binoculars

ECOLOGY AND HABITAT OF AN ENDANGERED EGERNIA (Carl Zeiss, Jena, Germany). The observer then moved to within 5–6 m and repeated the search. Finally, if no animal was seen from 5–6 m, the observer moved to within 1 m of the mounded shrub and thoroughly checked to determine whether a skink was at or just below the burrow entrance. All mounded shrubs with burrows were searched using this method on each visit to either subpopulation. The behavior of each visible lizard was recorded and observed approximately every 30 min. Each mounded shrub in each of the subpopulations was scanned at least once per 60 min. An individual lizard was observed for at least 60 sec before moving on to the next individual. We defined a bout of behavior as being continuous until a lizard entered a burrow and remained out of view for at least 30 min. Lizards were not artificially marked, but the majority of individuals were recognized by a combination of size, shape, and distinctive markings (such as an ablated tail). This method enabled us to estimate the minimum number of individuals present, but it did not allow separation of similar sized and marked animals (if any were present). In April 2005, we attempted to capture individuals by setting a variety of traps, including aluminium box traps (Elliott traps), baited with commercial cat food, and funnel traps, in clusters around burrows over 16 nights. We did not trial pitfall traps because of our concerns that this would damage the burrow systems of the lizards; however, box traps are effective in capturing other Egernia species (Douch, 1994; Duffield and Bull, 2002; Arena and Wooller, 2003). We captured only two individuals by using these methods (and a further two by hand capture); therefore, we decided that it was not possible to capture and individually mark animals for the behavioral study. The behavior of each lizard was classified into the categories described below. Basking. This behavior was recorded when an individual was resting either at the mouth of a burrow entrance or, if outside, within 50 cm of a burrow entrance. Animals either had all limbs on the soil surface or occasionally had the forelimbs raised on a low branch. Basking individuals often had their eyes shut. Foraging. Prey capture and consumption were both included in this category. A foraging lizard typically moved rapidly from a stationary position close to a burrow entrance to capture a ground-dwelling invertebrate. Digging. This behavior was observed at the entrance to burrows. Animals spent 10–120 sec clearing sand from the burrow by using either the hindlimbs or both sets of limbs.

565

Other activity. This category included entering and leaving burrows and commuting from one location to another. Vegetation Assessment and Burrow Availability.— Within each of the two subpopulations studied, environmental attributes were measured in situ within a 20- 3 20-m quadrat located within the area occupied by E. slateri. The areas occupied by E. slateri at each site were small (0.14 and 0.07 ha) such that a larger quadrat size would have included areas unsuitable for the species. In each quadrat, we recorded every plant species and estimated its projected foliage cover by using a scale of 1 (,10% cover) to 6 (.70% cover). We measured several abiotic variables including slope, aspect, rockiness, and landform element. Disturbance variables assessed were as follows: introduced large herbivore (including horse, cattle, camel, and donkey) impact (based on presence of tracks and scats), rabbit impact (based on presence of scats, diggings, and burrow systems), and weed cover. A representative soil sample (to a depth of 40 cm) was taken at both sites and analyzed for texture. At each subpopulation, we counted the number of mounded individuals of each plant species and the number of lizard burrows in each mound. Burrows of E. slateri do not have any diagnostic characteristics, so we could not be 100% certain that these mounds were dug by the study species; however, we rarely detected other burrowing lizard species around the mounds; therefore, the majority of burrows were likely to have been dug by E. slateri. This assessment of mounded shrubs was completed in March 2006. After our behavioral study, we determined the number of mounded shrubs of each species used by E. slateri. Statistical Analysis.—We estimated the total number of prey taxa from our dietary samples by using the nonparametric Chao 2, first-order jackknife, and second-order jackknife estimators. We chose these three richness estimators because they provide the least biased estimates of total richness (Cowell and Coddington, 1994; Toti et al., 2000) and all are incidence-based, thus they use presence–absence data to quantify rarity (Colwell and Coddington, 1994; Toti et al., 2000; Seaby and Henderson, 2006). We also constructed a prey accumulation curve to compare the number of prey taxa recorded against the number of scats sampled. Calculations were run in the program Species Diversity and Richness IV (Seaby and Henderson, 2006) by using 10 random selections of the sample order. We used a chi-square test of independence to compare use and availability by E. slateri of different species of mounded shrubs. Yates’

566

C. R. PAVEY ET AL. RESULTS

FIG. 1. Prey accumulation curve and curves of three prey richness estimators based on dietary data from 44 scats of Egernia slateri from Owen Springs, central Australia.

correction for continuity was used for this calculation because it involved a 2 3 2 contingency table (Siegel and Castellan, 1988). We performed a linear regression to examine the relationship between percentage volume of ants in the diet and SVL within the genus Egernia. N denotes the number of observations, whereas n denotes the number of individuals.

Adult Diet.—We analyzed the prey contents of scats collected from six separate mounded shrubs (four from subpopulation 1, the remainder from subpopulation 2). Our approach to identify individuals by unique markings showed that the six shrubs were occupied by a minimum of four adults, two subadults, and one juvenile during the study period. The number of adult and subadult scats collected per mound was 12, 2, 5, 10, 10, and 5. The prey accumulation curve and the curves of the three prey richness estimators did not reach an asymptote with prey richness continuing to increase as sample size increased (Fig. 1); however, the prey accumulation curve was beginning to plateau. The Chao 2 estimator appeared to be erratic, whereas the shape of the curve of the first-order jackknife was very similar to that of the prey taxa accumulation curve. The diet of E. slateri included both invertebrate and plant material. Invertebrates consisted of eight orders of insects, centipedes, spiders, and scorpions (Table 1). Plant material consisted of leaves, flowers, and seeds. Ants (Formicidae) and termites (Isoptera) were the major prey of adult and subadult E. slateri. Ants were

TABLE 1. Summary of dietary items in scats (N 5 44) of adult Egernia slateri at Owen Springs, central Australia. Higher classification

Arthropoda Hexapoda

Class

Insecta

Order

Isoptera Orthoptera Hemiptera Coleoptera

Diptera Lepidoptera Blattodea Hymenoptera

Myriapoda Chelicerata Planta

Chilopoda Arachnida

Suborder/ superfamily

Araneae Scorpionida

Family

Unidentified Termitidae Unidentified Caelifera Ensifera Heteroptera

Unidentified Acrididae Unidentified Pentatomidae Auchenorrhyncha Unidentified Unidentified Curculionidae Scarabaeidae Tenebrionidae Trogidae Bombyliidae Unidentified Unidentified Unidentified Chalcidoidea Pteromalidae Ichneumonoidea Apoidea Scolioidea Formicidae Tiphiidae

% % % occurrence fragments volume

59.1 29.6 11.4 2.3 9.1 15.9 6.8 6.8 20.5 4.5 11.4 2.3 2.3 2.3 2.3 2.3 2.3 2.3 2.3 6.8 95.5 2.3 2.3 27.3 6.8 36.4

27.3 10.2 2.6 0 1.6 0.6 0.9 0.2 1.8 0.2 2.6 0.7 0.6 0.5 0 0.1 0 0 0.2 0.8 46.5 0 ,0.1 1.9 0.5

23.3 9.1 1.8 0.3 16.5

0 0 0.1 1.3

35.0 0 2.7 1.5 8.4

ECOLOGY AND HABITAT OF AN ENDANGERED EGERNIA TABLE 2. Summary of ant genera present in the scats of adult Egernia slateri at Owen Springs, central Australia.

Subfamily

Genus

Cerapachyinae Dolichoderinae

Cerapachys Iridomyrmex purpureus-group Iridomyrmex other Camponotus Calomyrmex Melophorus Meranoplus Tetramorium Monomorium Pheidole Rhytidoponera Leptogenys Odontomachus Pachycondyla

Formicinae Myrmicinae

Ectatomminae Ponerinae

Proportion of scats (N 5 44)

0.02 0.23 0.27 0.23 0.02 0.11 0.05 0.05 0.05 0.39 0.14 0.05 0.02 0.07

present in almost all scats examined (95.5% occurrence, Table 1) and termites (including Termitidae) were present in 63.6% of scats. Other taxa present in at least 20% of scats included beetles (Coleoptera), spiders (Araneae), and plants. Ants were also the main prey based on the number of fragments in the scats (46.5% of fragments, Table 1) followed by termites (37.5% of fragments). Despite the comparatively small size of ants and termites, both taxa dominated the diet of E. slateri when measured by volume. Ants contributed 35% by volume and termites 23.3%. Other taxa contributing .5% by volume to E. slateri diet were Coleoptera (16.5%), Orthoptera (9.1%), and plants (8.4%). Egernia slateri captured ants from six subfamilies and 13 genera (Table 2). All ant material observed in scats was from adults including workers (both minor and major castes) and alates. The dominant ant genera in the diet were Pheidole, Iridomyrmex, and Camponotus (Table 2).

567

Vegetation Assessment and Burrow Availability.—One subpopulation of E. slateri (site 1) was situated at the footslope of a small rise, adjacent to a floodplain. The site was gently inclined (,2u slope) with a westerly aspect. Vegetation at the site was a sparse shrubland (,10% canopy cover) of mainly mature Eremophila (also Hakea and Acacia). The ground layer was extremely sparse (,5% cover), comprising predominantly chenopod herbs and subshrubs. Likewise, litter cover and surface stone cover (mainly pebbles) were minimal (5% and 1%, respectively), giving rise to a ground stratum of mainly bare ground. Soil was a duplex of sandy-loam (surface to 40 cm in depth) over medium clay. The other site (site 2) was on the lower gentle slope (2u) of a northeast facing low stony rise. This site also supported a sparse shrubland (,10% canopy cover) of mature Eremophila, Senna, and Acacia. Ground vegetation cover (chenopod herbs and short grasses) was marginally higher (6%) than that at site 1, as was rocky surface cover (3%). Again, most of the site consisted of bare ground. Soil was clayey sand over sandy clay. All vegetation at both sites was ,5 m in height. Disturbance at both sites was low and consisted of a low level of impact by introduced herbivores and invasive grasses (namely, Cynodon dactylon and Cenchrus ciliaris). Neither site had been burnt recently. Six species of shrubs had mounded individuals in one or both E. slateri subpopulations (Table 3). The dominant shrubs at both sites were species of Eremophila: E. maculata at site 1, E. duttonii at site 2, and E. sturtii at both sites (Table 3). Ninety-five percent of eremophilas within the two vegetation quadrats were mounded, and 58% of these mounded shrubs contained burrows in the soil pedestal. The highest number of burrows per mound was under E. sturtii then E. maculata. Mounded E. maculata shrubs were preferred as burrow sites by E. slateri; skinks occupied four of 11 shrubs of this species but only four of

TABLE 3. Number of mounded individuals of each shrub species present at each site (site 1/site 2) and details of presence of burrows and use by Egernia slateri. A dash indicates that either mounded shrubs were absent or, if present, contained no burrows. Species

Eremophila maculata Eremophila sturtii Eremophila duttonii Acacia victoriae Acacia tetragonophylla Senna artemisioides subsp. filifolia Dead

No. mounded shrubs

Proportion mounded shrubs with burrows

Avg. no. burrows/ mounded shrub

No. mounded shrubs used by E. slateri

11/0 15/21 0/13 1/1 0/1

0.55/— 0.60/0.71 —/0.54 1.00/0 —/1.00

3.50/— 3.33/2.80 —/1.29 1.00/— —/1.00

4/— 2/2 —/0 1/— —/0

—/1.00 —/0

—/4.00

0/1 0/2

—/1

568

C. R. PAVEY ET AL.

49 shrubs of the other two Eremophila species combined (Table 3), a significant difference (x2 corrected for continuity 5 3.98, P 5 0.046). Egernia slateri also used burrows in mounded Acacia victoriae and Senna artemisioides (Table 3). Mounded shrubs occupied by E. slateri had a mean of 3.7 burrows per shrub, with a range of 1 to 11. Behavior, Sociality, and Activity Patterns.—We observed 52 behavioral bouts in total, involving a minimum of eight individuals: six (three adult, one subadult, and two juvenile) in subpopulation 1 and two (one adult and one subadult) in subpopulation 2. Adults and subadults were observed by themselves during 94% of behavioral bouts (N 5 47, n 5 6). The exception to this pattern was three bouts when an adult was accompanied by a juvenile. During these bouts, the adult and juvenile were in proximity, typically within 5 cm, often at the entrance to a burrow. However, juveniles also were seen at the same burrow entrances without adults. An individual adult or subadult seemed to be the sole occupant of all burrows in a particular mounded shrub. During the course of daily observations, one adult was observed to move between three mounded shrubs and a second adult moved between two mounded shrubs. We observed only one instance of two animals (apart from adult–juvenile combinations) using the same mounded shrub. This case was of an adult and subadult; however, the adult was observed at this shrub only once and at the time the subadult was not present. Apart from this situation, adults seemed to be solitary. The focal animal basked during 71% (N 5 37, n 5 6) of behavioral bouts. Basking occurred either at a burrow entrance (with up to 70% of the animal still in the burrow) or on the mound of soil near a burrow entrance. Digging was observed during six bouts (n 5 4). In addition to basking, animals searched for food while at the burrow entrance. A ‘‘sit and wait’’ foraging method was observed. After detecting prey, the lizard moved rapidly to capture and consume it. The distance moved from the basking site to capture prey ranged from 10 to 120 cm (N 5 4, n 5 4). Prey capture attempts involved ants (N 5 3) and a locust (Orthoptera: Acrididae). Ants were picked off the ground, whereas a subadult E. slateri captured the locust after climbing along a branch, 2–3 cm above the ground, to reach it. We observed animals at the burrow entrance postsunrise during 29 bouts. The latest bout was observed 327 min postsunrise; however, the majority of bouts were during the second and third hours postsunrise (Fig. 2). A minimum of five individuals was observed during this period (Fig. 2). We did not observe any activity

FIG. 2. Activity of Egernia slateri during the 6 h immediately postsunrise at Owen Springs, central Australia, showing the number of animals observed and the rate at which bouts occurred for each 60-min interval. The total search effort (minutes) is given above each 60-min interval.

by E. slateri during the 60 min immediately postsunrise. We observed activity by E. slateri presunset in the evening during 23 bouts. Activity did not commence until the final 2 h presunset (Fig. 3), with most activity during the 60 min immediately before sunset. A minimum of seven individuals was observed during this period (Fig. 3). DISCUSSION This study provides the first detailed information on the foraging ecology and habitat use of the nominate subspecies of E. slateri, a highly threatened arid zone lizard. The dietary data suggest a degree of specialization on ants and termites. The ranking of prey taxa was consistent across the three methods used to assess dietary importance; percentage occurrence in scats, percentage identifiable fragments, and percentage volume. Ants were the major prey of adult E. slateri, contributing 35.0% by volume despite their small overall size (Table 1). Isoptera were the next major prey taxon followed by Coleoptera and Orthoptera (Table 1).

FIG. 3. Activity of Egernia slateri during the 3 h immediately presunset and the first hour postsunset at Owen Springs, central Australia, showing the number of animals observed and the rate at which bouts occurred for each 60-min interval. The total search effort (minutes) is given above each 60min interval.

ECOLOGY AND HABITAT OF AN ENDANGERED EGERNIA

FIG. 4. Incorporation of ants in the diet by Egernia species, using data from studies that estimated percentage volume of prey. Body size is based on the median snout–vent length given in Chapple (2003). Source data are as follows: E. cunninghami (Brown, 1983), E. hosmeri (Shea, 1995), E. striolata (Brown, 1983), E. saxatilis (Brown, 1983), E. whitii (Brown, 1983), E. inornata (Pianka and Giles, 1982), E. striata (Pianka and Giles, 1982), E. major (Shea, 1999), E. stokesii (Duffield and Bull, 1998), E. coventryi (Douch, 1994), and E. slateri (this study).

Egernia slateri consumed 13 ant genera in total from six subfamilies, representing almost half the genera of ground-nesting ants expected to occur in central Australia (Shattuck, 1999; Hertog, 2008). The prey richness estimators suggest that increasing the sample of scats analyzed will increase the number of prey taxa in the diet; therefore, it seems highly likely that further dietary sampling will increase the number of ant genera taken by E. slateri. Both ants and termites represent small prey for a skink the size of E. slateri (SVL 85–97 mm). A comparison of the degree of incorporation of ants in the diet, measured as percentage volume, within the genus Egernia reveals that ants are generally a minor prey item and that once SVL exceeds 110 mm, ants are almost never captured (Fig. 4). Egernia slateri and E. inornata are the only Egernia species known to have a diet dominated by ants. As with E. slateri, ants were the major prey of E. inornata, irrespective of the method used to assess dietary importance (Pianka and Giles, 1982). The only Egernia species that had a volume of ants in the diet of .0.1% were those with an SVL , 110 mm (Fig. 4). Within this group of five species, there was a significant inverse relationship between SVL and incorporation of ants in the diet (linear regression, R2 5 0.87, P , 0.05). The relationship is not as clear-cut as indicated because some studies, that do not use percentage volume, indicate that some of the larger species do take small numbers of ants. Furthermore, some of

569

the larger species do take more insects during the juvenile stage (e.g., Duffield and Bull, 1998). Our data confirm the positive relationship between body size and incorporation of plant material in the diet within the genus Egernia (Chapple, 2003). Based on the regression equation given in figure 3 of Chapple (2003), the percentage plant material in the diet of E. slateri (based on a SVL of 85 mm) should be 8.65%. We recorded a volume of 8.4% (Table 1), similar to that found for other small Egernia species. Our habitat assessment demonstrates the reliance of E. slateri on pedestals formed by accumulation of wind-carried soil as habitat in which to dig burrows. Henzell (1972) noted that such mounded shrubs were present at all known localities of E. slateri, and they also were at the seven sites where we recorded E. slateri between 2004 and 2008. Furthermore, our data from Owen Springs show that E. slateri occupies areas where the majority of shrubs are mounded (Table 3) and is absent from adjacent sites where shrubs do not support mounds. Egernia slateri is one of four arid zone species in this genus that are obligate burrowers (Chapple et al., 2004); however, it is the only one of these species that occupies alluvial plains close to the boundary with stony hills and rises. The other three obligate burrowers in the arid zone (E. kintorei, E. striata, and E. inornata) shelter in burrows in sandy substrates, whereas other arid congeners (E. margaretae, E. stokesii, and E. depressa) use existing structures (rock slabs and hollow logs) for shelter (Wilson and Swan, 2008). Clay-based soils dominate the alluvial plains to which E. slateri is largely restricted, and these soils provide limited burrowing opportunities. Therefore, mounded shrubs probably play an important role in enabling the persistence of populations of E. slateri on alluvial plains by providing a suitable substrate in which to burrow. Henzell (1972) developed this relationship further by proposing that burrowing into mounded shrubs provided physiological advantages for E. slateri because the humidity in the air spaces in the soil around the roots of the shrubs increased burrow humidity and resulted in a reduced rate of evaporative water loss. Behavioral observations from this study indicate that E. slateri is a largely solitary species. Despite complex sociality being widespread in the genus Egernia (Duffield and Bull, 2002; Gardner et al., 2008) and anecdotal claims of social behavior in E. slateri (Ehmann cited in Chapple, 2003), we obtained no evidence of sociality in our study population. Group size was one except when an adult was observed with a juvenile. Our observations were limited to late summer (February–March); however, in

570

C. R. PAVEY ET AL.

other Egernia species, stable social groups are present throughout the active season, which is typically late August to April, and they include readily observable aggregations of group members basking together, which involves animals either physically touching or in piles with some on top of others (e.g., Duffield and Bull, 2002). As a consequence, if E. slateri is a social species our observation intensity was more than sufficient to obtain evidence of social groups. Our findings support Henzell’s conclusion that the species is largely solitary (Henzell, 1972). Our observations indicate that each adult E. slateri is likely to maintain exclusive use of at least a single mounded shrub, except during the mating period, when males and females will overlap. Information on spatial ecology of congeners shows that home range size and number of occupied crevices and burrows are typically small, even in group living species (Bustard, 1970; Duffield and Bull, 2002). For example, groups of two to 17 individuals of the arid, rock-inhabiting E. stokesii occupied between two and 11 crevices per group and had a mean home range size of 0.0864 ha (Duffield and Bull, 2002). Such a small range size is possible because of the energetically conservative ‘‘sit and wait’’ foraging behaviour used by E. slateri and its congeners. Individuals in both subpopulations were active during the day; between 2 and 6 h postsunrise and the final 2 h presunset. Although no activity was noted during the first hour postsunrise, we did observe activity after sunset. Diurnal foraging also was noted. This diurnal activity took place despite February and March being two of the four hottest months in central Australia (Australian Bureau of Meteorology, unpubl. data). However, our study did not assess the level of nocturnal activity in these animals. Although the majority of Egernia species are considered diurnal, some arid and semiarid sand plain species (E. kintorei, E. inornata, and E. striata) are crepuscular and/or nocturnal (Chapple, 2003). Our study covered two of four subpopulations at one of seven sites where E. slateri is known to occur. We observed a minimum of eight individuals (based on body size and distinctive markings) with an estimated maximum count of 12 individuals in these two subpopulations. In comparison, a mark–recapture study at the same two subpopulations between February and April 2005 resulted in the capture of four individuals (two by hand) and recorded only three unmarked animals at the conclusion of the study, giving a count of seven individuals (Derez and CRP, unpubl. data). We recorded these low population counts despite a significant portion of mounded shrubs being unoccupied (i.e., not containing any burrows) in both subpopulations

(Table 3). For example, five of 11 E. maculata and six of 15 E. sturtii within the area of subpopulation 1 had no burrows. The other two subpopulations at this site have not been studied intensively, but neither is larger in area nor contains more suitable habitat than the two subpopulations we assessed. Therefore, we estimate that the maximum size of the entire metapopulation between 2005 and 2007 was approximately 20–25 individuals. Such low counts, together with the limited amount of suitable habitat available, highlight the vulnerability of this lizard to extinction. This situation is exacerbated by the high disturbance regimes on the floodplains occupied by E. slateri in central Australia (Pavey and Nano, 2009). To emphasise the threats faced by this metapopulation, no individuals were observed at subpopulation 1 during 7 wk of searching between 5 January and 10 December 2008 (Ford, unpubl. data). At the time of our study, the Owen Springs population experienced only low levels of impact by weeds and introduced herbivores; however, weeds have infested other potentially suitable areas throughout the local area. Research assessing the size and extent of other known populations is currently being undertaken together with on-ground management of threatening processes such as weed invasion. Evidence indicates that E. slateri is one of the most highly threatened reptiles in Australia. Acknowledgments.—We acknowledge the efforts of S. McAlpin in raising awareness of the plight of E. slateri and preparing the submission for listing of E. s. slateri under Australia’s Environment Protection and Biodiversity Conservation Act. We are grateful to G. Fyfe for undertaking field surveys and for locating the populations assessed in this study. We thank C. Derez for her role in studying the species on Owen Springs during 2005. We thank P. Couper, S. Wright, and O. Seeman at the Queensland Museum for her assistance with identifying prey material. CRP thanks M. Bull for support and encouragement of this study and M. Ford for insight into the ecology of E. slateri and the study area.

LITERATURE CITED ARENA, P. C., AND R. D. WOOLLER. 2003. The reproduction and diet of Egernia kingii (Reptilia: Scincidae) on Penguin Island, Western Australia. Australian Journal of Zoology 51:495–504. BROWN, G. W. 1983. Comparative Feeding Ecology of South-eastern Australian Scincids. Unpubl. Ph.D. diss., La Trobe University, Melbourne, Victoria, Australia.

ECOLOGY AND HABITAT OF AN ENDANGERED EGERNIA BUSTARD, H. R. 1970. A population study of the scincid lizard Egernia striolata in northern New South Wales. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen C 73:186– 213. CHAPPLE, D. G. 2003. Ecology, life-history, and behaviour in the Australian scincid genus Egernia, with comments on the evolution of complex sociality in lizards. Herpetological Monographs 17:145–180. CHAPPLE, D. G., AND J. S. KEOGH. 2004. Parallel adaptive radiations in arid and temperate Australia: molecular phylogeography and systematics of the Egernia whitii (Lacertilia: Scincidae) species group. Biological Journal of the Linnean Society 83:157– 173. CHAPPLE, D. G., J. S. KEOGH, AND M. N. HUTCHINSON. 2004. Molecular phylogeography and systematics of the arid-zone members of the Egernia whitii (Lacertilia: Scincidae) species group. Molecular Phylogenetics and Evolution 33:549–561. COLWELL, R. K., AND J. A. CODDINGTON. 1994. Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society of London B 345:101–118. DOUCH, P. M. 1994. Comparative ecophysiology of two species of scincid lizard, Egernia coventryi and Egernia whitii. Unpubl. Honours thesis, University of Melbourne, Melbourne, Victoria, Australia. DUFFIELD, G. A., AND M. C. BULL. 1998. Seasonal and ontogenetic change in the diet of the Australian skink Egernia stokesii. Herpetologica 54:414–419. ———. 2002. Stable social aggregations in an Australian lizard, Egernia stokesii. Naturwissenschaften 89:424–427. GARDNER, M. G., A. F. HUGGALL, S. C. DONNELLAN, M. N. HUTCHINSON, AND R. FOSTER. 2008. Molecular systematics of social skinks: phylogeny and taxonomy of the Egernia group (Reptilia: Scincidae). Zoological Journal of the Linnean Society 154:781–794. HENZELL, R. P. 1972. Adaptation to Aridity in Lizards of the Egernia whitei Species-group. Unpubl. Ph.D. diss., University of Adelaide, Adelaide, South Australia, Australia. HERTOG, T. 2008. Ant species of the western region of the Simpson Desert, NT. In Colson Pinnacle Project, pp. 35–40. Desert Discovery Inc., Smithfield, New South Wales, Australia. HORNER, P. 1992. The Skinks of the Northern Territory. Northern Territory Government Printing Office, Darwin, Northern Territory, Australia.

571

PAVEY, C. 2007. Slater’s Skink Egernia slateri. In J. Woinarski, C. Pavey, R. Kerrigan, I. Cowie, and S. Ward (eds.), Lost from Our Landscape: Threatened Species of the Northern Territory, pp. 176–177. Northern Territory Government, Darwin, Northern Territory, Australia. PAVEY, C. R. 2004. Recovery Plan for Slater’s Skink, Egernia slateri, 2005–2010. Northern Territory Department of Infrastructure, Planning and Environment, Northern Territory Government, Alice Springs, Northern Territory, Australia. PAVEY , C. R., AND C. E. M. N ANO . 2009. Bird assemblages of arid Australia: vegetation patterns have a greater effect than disturbance and resource pulses. Journal of Arid Environments 73:634–642. PIANKA, E. R., AND W. F. GILES. 1982. Notes on the biology of two species of nocturnal skinks, Egernia inornata and Egernia striata, in the Great Victoria Desert. Western Australian Naturalist 15:44–49. SEABY, R. M., AND P. A. HENDERSON. 2006. Species Diversity and Richness, version 4. Pisces Conservation International, Lymington, England. SHATTUCK, S. O. 1999. Australian Ants: Their Biology and Identification. Commonwealth Scientific and Industrial Research Organization, Melbourne, Victoria, Australia. SHEA, G. M. 1995. Herbivory in Egernia hosmeri (Squamata: Scincidae). Herpetofauna 25:8–11. ———. 1999. Morphology and natural history of the Land Mullet Egernia major (Squamata: Scincidae). Australian Zoologist 31:351–364. SIEGEL, S., AND N. J. CASTELLAN JR. 1988. Nonparametric Statistics for the Behavioral Sciences. McGrawHill, New York. STORR, G. M. 1968. Revision of the Egernia whitei species-group (Lacertilia, Scincidae). Journal of the Royal Society of Western Australia 51:51–62. TOTI, D. S., F. A. COYLE, AND J. A. MILLER. 2000. A structured inventory of Appalachian grass bald and heath bald spider assemblages and a test of species richness estimator performance. Journal of Arachnology 28:329–345. WILSON, S., AND G. SWANN. 2008. A Complete Guide to Reptiles of Australia, 2nd ed. New Holland Publishers, Sydney, New South Wales, Australia.

Accepted: 30 March 2010.