Publications archive - Publications
Key departmental publications, e.g. annual reports, budget papers and program guidelines are available in our online archive.
Much of the material listed on these archived web pages has been superseded, or served a particular purpose at a particular time. It may contain references to activities or policies that have no current application. Many archived documents may link to web pages that have moved or no longer exist, or may refer to other documents that are no longer available.
Gregory J. Hollis
Environment Australia, October 1997
Note: This publication has been superseded by the National Recovery Plan for the Baw Baw Frog Philoria frosti
The conservation status of the Baw Baw Frog (Philoria frosti) up until recently was presumed to be secure (e.g. Tyler 1992). This assumption was based on population estimates made by Malone (1985a), who estimated the adult male population size to be in the vicinity of 10,000-15,000 in 1983 and 1984. In response to growing concern among many amphibian biologists about recent amphibian declines (e.g. Tyler and Davies 1985; Osborne 1990; Blaustein and Wake 1990; Watson et al. 1991; Richards et al. 1993) and extinctions (e.g. Barinaga 1990; Czechura and Ingram 1990; Phillips 1990), concerns were raised over the current conservation status of the Baw Baw Frog. Further investigations showed an even more alarming trend; that of those amphibians reported to have declined or disappeared, a significant number are restricted mountain-top environments in relatively pristine habitats, like the Baw Baw Frog (e.g. Osborne 1990; La Marca and Reinthaler 1991; Bradford 1991; Crump et al. 1992; Richards et al. 1993; Gillespie et al. 1995).
A subsequent review of historical information on the distribution and abundance of the Baw Baw Frog, and survey of the current distribution and abundance of the species, was initiated in 1992 by the then Victorian Department of Conservation and Environment. Although only anecdotal, records of frogs from the 1950's, 1960's, 1970's and early 1980's indicate that the species was relatively common. Similarly, the results of systematic surveys for the species in 1983 and 1984 also found it to be common, with an estimate of 10,000-15,000 adult males being present (Malone 1985a). No records exist for the species between 1985 and 1988. Records from 1989 to 1992 are scant and anecdotal, but all reports indicate only low numbers being present. The results of systematic surveys conducted from 1993 to 1996 are in accordance with the later observations. The relative abundance of the adult male population has declined by several orders of magnitude, being only 1-2% of the estimate reported by Malone (1985a) over a decade ago (Hollis 1995; G. Hollis unpubl. data). Along with this observed decline, is an apparent contraction in range of breeding habitat to topographically protected gullies (Hollis 1995; G. Hollis unpubl. data), compared with former known breeding sites distributed over a much broader area (see Malone 1985a). The factors responsible for the decline in the Baw Baw Frog are currently unknown, but if they continue to operate, may result in the extinction of the species. Population monitoring and research is urgently required to ascertain those factors contributing to the decline.
The Baw Baw Frog (Philoria frosti) is a medium-sized species in the family Myobatrachidae. Females are larger than males, measuring on average 51.6 and 44.5 mm snout-vent length (SVL) respectively (Malone 1985b), although males have longer legs and wider heads (Littlejohn 1963). The dorsal surface of adults is predominantly brown to dark brown with cream-yellow flecks or spots, and/or larger patches spreading from the top of the head and snout to the back. A characteristic brown-black lobe-shaped parotoid gland occurs behind each eye, extending back over each shoulder. The surface of the skin is scattered with small tubercles or warts, particularly on the sides and rump, giving a 'prickly' appearance. The ventral surface is smooth, ranging in colour from cream-dark yellow, and heavily pigmented with brown flecking. The throats of males are dark brown-black. The tympanum is hidden. Fingers and toes are unwebbed. Adult females are characterised by possessing a flanged inner first and second finger. (For other descriptions of adults see Spencer 1901; Littlejohn 1963; Cogger 1994; Barker et al. 1995).
Metamorphlings and juveniles look somewhat different from adults. Colour varies from cream to brown with dark brown irregular flecks and patches. A dark brown band occurs from the snout, through the eye, to the base of the forelimb. Young Baw Baw Frogs have been described as looking strikingly similar to their relatives (the genus Kyarranus) in northern New South Wales (Littlejohn pers. comm., G. Hollis pers. obs.).
Larvae are creamish-white and unpigmented when they hatch at Gosner stages 22 - 23 (Littlejohn 1963; Malone 1985b). During later development, the dorsal surface is light-brown (Littlejohn 1963) and the eyes are pigmented (G. Hollis pers. obs.). They are non-feeding, hatching with a residual yolk mass that maintain them through to metamorphosis (Malone 1985a, b). The mouth parts are rudimentary; containing partly keratinised beaks, approximately 20 oral papillae, with no labial tooth rows (Littlejohn 1963; Malone 1985b). At Gosner stage 38 (Gosner 1960), larvae measure 7.0 mm (body length), 12.9 mm (tail length) and 5.8 mm (maximum body width). The spiracle opens on the left side close to the ventral surface, and the vent is median (Littlejohn 1963).
The Baw Baw Frog is restricted entirely to the Baw Baw Plateau, located approximately 120 km east of Melbourne, Victoria (Malone 1985a; Hollis 1995) (Fig. 1). Despite searches in close proximity to the plateau, and in similar environments (e.g. Mt Useful, Mt Torongo), the species has not been found elsewhere (Malone 1985a). Encompassing an area of approximately 80 km2, the Baw Baw Plateau between approximately 1260-1560 m a.s.l. is contained within the Baw Baw National Park, except for approximately 3.5 km2 near Mt Baw Baw that is managed by the Victorian Alpine Resorts Commission. At lower elevations, between 1000 and 1300 m, records of the species occur from montane forest along the access road to the Mt Baw Baw Alpine Resort and in the vicinity of the track leading to The Mushroom Rocks (near Mt Erica) in the Baw Baw National Park.
The species was discovered at Mt Baw Baw in 1898 by C. Frost, where he retrieved five freshly regurgitated specimens from a Tiger Snake (Notechis scutatus) (Spencer 1901). Records relating to the distribution of Baw Baw Frog prior to 1982 are derived mostly from collecting trips made by the Museum of Victoria and from early biological studies on the species (e.g. Littlejohn 1963). These records pertain to accessible locations only, most of which are in the Mt Baw Baw area, with a smaller number from the Mt Whitelaw and Mt Erica areas (see appendix 1 for details). In 1983 and 1984, Malone (1985a) showed that the distribution stronghold of the species is primarily the western (Mt Baw Baw), central (Mt St Phillack) and north-western (Mt Whitelaw) regions of the plateau (Fig. 2). He recorded fewer records from the eastern region (Mt St Gwinear), and the species was conspicuously absent from the south-eastern region (Mt Erica).
Systematic surveys for calling males from 1993 to 1996 show that the distribution of the species is currently restricted to the western and central area of the Baw Baw plateau (Hollis 1995; G. Hollis unpubl. data) (Fig. 3; inside back cover). This distribution was also reported by Malone (1985a) as the stronghold area for the species above. Within this distribution, however, a number of breeding units that previously contained up to several hundred calling males (e.g. Moondara Flat and Freeman's Flat), have not had frogs recorded in them since (see Table 1). No calling males were recorded from the eastern and south-eastern regions of the Plateau during these recent surveys. These regions appear to be drier and rockier than other regions on the plateau, and seem to provide only small pockets of breeding habitat similar to that currently being used by the species elsewhere (G. Hollis pers. obs.). In 1996, a portion of the survey effort was directed towards unsurveyed areas in montane forest vegetation in State Forest along the southern escarpment of the Baw Baw plateau at elevations between 1000 - 1300 m. Calling males were recorded in 10 of the 27 gully systems surveyed, as low as 1080 m, and ranged in distribution from the Mt Whitelaw region in the north-west to the Mt Tyers region located towards to south-east end of the plateau (G. Hollis unpubl. data).
Until recently, most records of the Baw Baw Frog have been collected during the breeding season (October - December), and consequently most habitat descriptions relate to breeding localities only (e.g. M. Littlejohn unpubl. data; Malone 1985a; Hollis 1995). Details of the few records collected outside the breeding season (January - April) remain unpublished (M. Littlejohn unpubl. data; J. Coventry and P. Robertson pers. comm.). Research conducted recently in 1995 and 1996, involving pitfall trapping and radio-tracking of the Baw Baw Frog, now provides preliminary information on the utilisation of breeding and non-breeding habitats by the species (G. Hollis unpubl. data).
Fig. 1. Distribution of the Baw Baw Frog (Philoria frosti) within five minute grids. (Source: Atlas of Victorian Wildlife, Department of Natural Resources and Environment).
Fig. 2. Distribution of calling male Philoria frosti recorded in 1983 and 1984 population surveys on the Baw Baw Plateau. (Source: Malone (1985a)).
|*#Access Rd 1||43||21||2||3||1||2|
|*#Access Rd 2||30||26||0||0||0||0|
|*#Access Rd 3||0||0||0||0||0||0|
|Village Flat 2||24||-||-||-||2||1|
|*La Trobe Plain||206||-||3||2||0||4|
|#Currawong Flat (a)||174||231||2||3||0||0|
|Currawong Flat ds||-||-||-||-||3||1|
|Creek Corner (a)||9||12||-||-||5||0|
|Jeep Track 1||0||0||0||-||-||-|
|Jeep Track 2||6||0||0||-||-||-|
|Jeep Track 3||3||0||0||-||-||-|
|Jeep Track 4||0||0||0||-||-||-|
|*#Tyers River 1||14||9||0||0||0||0|
|*#Tyers River 2||7||3||0||0||0||0|
Note: Currawong Flat (a) refers to a small portion of Currawong Flat censused in all surveys; this statistic is also contained within the census statistic for Currawong Flat for years 1983, 1993, 1994, 1995 and 1996.
|Hope Creek (left branch)||0|
|Faith Creek (left branch)||2|
|Faith Creek (right branch)||2|
|Mt Tyers 1||3|
|Mt Tyers 2||5|
|Mt Tyers 3||0|
|Mt Tyers 4||0|
|Tyers River East Branch (tributary)||0|
|Tyers River 1 (right branch)||0|
|Tyers River 2 (left branch)||0|
|Tyers River 2 (right branch)||0|
|Charity Creek (right branch)||1|
|Charity Creek (left branch)||2|
|Tanjil River Tributary 1||9|
|Tanjil River Tributary (ridge)||12|
|Tanjil River Tributary 2||3|
|Tanjil River East Branch||0|
|Tanjil River West Branch 1||0|
|Tanjil River West Branch 2||4|
|Tanjil River West Branch 3||0|
|Tanjil River West Branch 4||0|
|Tanjil River West Branch 5||0|
|Tanjil River West Branch 6||1|
During the breeding season (October-December), calling males, gravid females and egg masses have been recorded, along seepage lines within or at the periphery of sub-alpine wet heathland and in gullies within montane vegetation communities (Malone 1985a; Hollis 1995). Using the vegetation classification of Walsh et al. (1984), Malone (1985a) recorded by far the greatest percentage of calling male sites in wet alpine heath - bog ecotones (75%), with remaining sites in wet alpine heath (22.5%), bogs (1.6%) and grassland (0.9% - modified areas within the Mt Baw Baw Alpine Resort). A small percentage of calling males (2.4%) was
recorded from montane riparian thicket; Leptospermum grandifolium and Nothofagus cunninghamii-dominated vegetation communities.
In 1993, Hollis (1995) recorded a shift in the proportion of calling males using the vegetation communities reported by Malone. In wet sub-alpine heathland (equivalent to Malone's wet alpine-heath - bog ecotone, wet alpine heath and bog area combined), 39.4% of calling males were recorded. Of these, only a small number were in the heath-bog ecotones as reported by Malone (1985a), with most being recorded further up-slope in wet alpine heath. A further 39.4% were recorded in ecotonal habitats at the periphery of wet sub-alpine heathland, adjacent to montane riparian thicket, dry sub-alpine shrubland and sub-alpine woodland. Malone (1985a) did not record aggregations of calling males in these ecotonal habitats. It is possible, however, that some frogs that were recorded by Malone as occurring in wet alpine heath may have actually occurred in these ecotonal habitats (B. Malone pers. comm.). Another apparent shift in habitat use was the greater proportion of calling males recorded in montane riparian thicket (21.2% compared to Malone's 2.4%). It is also possible that Malone may have overlooked a number of calling males in this habitat type, given the difficulty in accessing thickets, and having to contend with counting large numbers of calling males in wet alpine heath - bog ecotones where it appeared the majority were (B. Malone pers. comm.). No calling males were recorded in the modified areas within the Baw Baw Alpine Resort. During surveys conducted in 1994, 1995 and 1996, the number of calling males located in different vegetation communities was similar to the proportions recorded in 1993, except for montane riparian thicket, where a proportional increase in frog numbers has been recorded (G. Hollis unpubl. data). Overall, surveys from 1993-1996 indicate that breeding localities are restricted to topographically protected gully habitats compared with the broader breeding distribution of the species over a decade ago.
The breeding microhabitat of the Baw Baw Frog is predominantly natural cavities in or under vegetation, logs, peat and soil, and rock (or combinations of these), that act as catchments for water travelling down slope (Littlejohn 1963; Malone 1985a; Hollis 1995). These sites occur along seepages or rivulets that drain water from the slopes of sub-alpine wet heathland and in gullies that drain the plateau. Some breeding microhabitats have also been reported from beneath building materials in the Mt Baw Baw Alpine resort (Malone 1985a; Littlejohn unpubl. data; J. Coventry and P. Robertson pers. comm.). In 1983 and 1984, Malone (1985a) recorded 79% of oviposition sites in cavities of vegetation consisting of Sphagnum spp., Astelia alpina, Empodisma minus, Epacris paludosa, Richea continentis and Carex spp. The remaining 21% of egg masses were recorded from beneath logs, rocks and building materials. By comparison, in 1993 Hollis (1995) reported a similar proportion of breeding sites in the habitats reported by Malone, but the assemblage of plants was slightly different. Most calling sites (76%) were located amongst the roots of vegetation, dominated by the species R. continentis, E. paludosa, Baeckea utilis var. latifolia, Orites lancifolia, Callistemon pityoides, L. grandifolium, N. cunninghamii, Sphagnum cristatum, Wittsteinia vacciniacea, Polytrichum alpinum/commune, A. alpina, Carex gaudichaudiana and C. appressa The remaining 24% of calling sites were located in peat or soil cavities beneath large granite boulders or logs. The different assemblage of plants recorded by Hollis can largely be attributed to the proportionally greater number of calling males recorded in montane riparian thicket vegetation and ecotonal habitats at the periphery of sub-alpine wet heathland (as noted above).
Larvae usually remain at oviposition sites through to metamorphosis (Barker et al. 1995; G. Hollis pers. obs.). However, upon hatching, larvae have also been observed to move small distances in shallow water from oviposition sites, while remaining covered under vegetation and/or woody debris (G. Hollis pers. obs.) or to be washed into nearby pools (B. Malone pers. comm.).
During the breeding and post-breeding seasons (October-April), adult male and female Baw Baw Frogs also utilise areas other than the breeding habitats described above. Females have been pit-trapped in wet sub-alpine heathland, sub-alpine woodland and ecotonal habitats adjacent to breeding sites (G. Hollis unpubl. data). Similarly, both males and females have been radio-tracked in wet sub-alpine heathland, sub-alpine woodland, montane wet forest and ecotonal habitats adjacent to breeding sites and in gullies (G. Hollis unpubl. data).
The microhabitats of frogs in these vegetation communities includes both aquatic and non-aquatic situations. Aquatic sites are similar to those used for breeding above. In sub-alpine woodland and ecotonal habitats adjacent to wet sub-alpine heathland, non-aquatic sites have been located in moist positions amongst the roots of vegetation (including the species W. vacciniacea, Carex spp., Monotoca sp aff. elliptica (Alps), Pultenaea muelleri, Tasmannia aff. xerophylla (Baw Baws), Eucalyptus pauciflora, Olearia phlogopappa, N. cunninghamii, L. grandifolium, Stylidium graminifolium, Leucopogon gelidus, Prostanthera cuneata, Lycopodium maccraei, Trochocarpa clarkei, Poa spp., Astelia alpina, E. paludosa, B. utilis var. latifolia), and/or under logs, rocks, leaf litter and bark (G. Hollis unpubl. data). In montane habitats (1000-1300 m), sites have been located in moist positions amongst the roots of vegetation (most commonly Blechnum wattsii, Carex spp., Hydrocotyle spp., E. delegatensis, W. vacciniacea, N. cunninghamii, L. grandifolium, Polystichum proliferum, Dicksonia antarctica and P. cuneata) and/or under logs, rocks, leaf litter and bark (G. Hollis unpubl. data). Previous records of the species outside the breeding season are from beneath logs, rocks and building materials around the Mt Baw Baw Alpine Resort (M. Littlejohn, P. Robertson and J. Coventry pers. comm.). Over-wintering habitats are unknown, although it is likely that frogs remain sheltered in the above micro-habitats. The temperature of soils in the vegetation communities being utilised by the Baw Baw Frog are likely not to fall below zero when covered by snow (see Costin 1954).
The Baw Baw Frog is considered to have an unusual and specialised life history that is exhibited by very few amphibians (Malone 1985a). Calling activity by the species has been heard in early October, continuing as late as the last week in December (Malone 1985a; Hollis 1995; Grigg and Barker 1995), although some individuals have been heard in January (J. Coventry pers. comm.). Oviposition (egg laying) appears to be confined to a shorter interval of 2 - 3 weeks during this period, when a peak in calling activity occurs (Malone 1985a, b; Hollis 1995). The timing of this peak in calling activity may vary between breeding seasons (G. Hollis unpubl. data). The following peaks in calling activity have been recorded: 11 - 29 November 1983/84 (Malone 1985a, b); 17 - 29 November 1993 (Hollis 1995); 26 November - 5 December 1994, 8 - 15 November 1995 and 2 - 15 November 1996 (G. Hollis unpubl. data). Calling activity may occur at any time of the day or night, with the greatest levels of activity being recorded during relatively warm, humid conditions (G. Hollis unpubl. data). Very little calling activity occurs at temperatures near freezing (Malone 1985a; G. Hollis unpubl. data) or when temperatures are high with low humidity (G. Hollis unpubl. data).
The advertisement call structure of the male is highly variable, and may be described as a short 'clunk', repeated in sequences of up to 30 calls (Littlejohn 1963). More recent tape recordings of calling activity throughout the breeding season have recorded sequences of up to 200 consecutive calls during the peak in calling activity (G. Hollis unpubl. data). These large calling sequences usually occurred at sites where more than one male (usually two) was participating in calling. The timing of notes by frogs involved in these calling sequences occurred in an alternate pattern so as to not overlap. Males also broadcast another type of call during the breeding season that can be described as 'growling'. This type of calling also occurs most commonly during the peak in calling activity, although its purpose remains unknown. A possible explanation is intraspecific competition between males, or male - female interaction. An analysis of the breeding call structure of the Baw Baw frog, including call duration, frequency and repetition rate, is reported by Littlejohn (1963).
Egg masses are deposited in a transparent foam nest (8 cm diameter and 3-4 cm high; Littlejohn 1963) at the calling site, or nearby, during inguinal amplexus. It is not known how long pairs remain in amplexus. The foam nest is produced by the female beating air bubbles into the mucous and eggs with flanged fingers during egg laying (Littlejohn 1963). The egg mass may be deposited at varying depths in vegetation, or below the ground surface, depending on the structural attributes of the site. Depths of over a metre have been observed, whilst others are deposited in vegetation very close to the surface (G. Hollis per. obs.). Clutch sizes reported in the literature range from 50-185 (see Littlejohn 1963; Malone 1985a, b; Tyler 1992). Ova are white and unpigmented and are considered large, measuring on average 4 mm in diameter (Malone 1985b). Oviposition of more than one clutch may occur at a single site (Malone 1985b; G. Hollis pers. obs.). It is also possible that females deposit a portion of their eggs at more than one site (Malone 1985a.). Under natural conditions the embryonic period varies from 5-8 weeks, with individuals hatching at Gosner stages 22-23 (Malone 1985a, b).
Initially, larvae are also white and unpigmented. They are non-feeding, hatching with a large residual yolk mass. Under experimental conditions, Malone (1985a) found that the provision of food, and stocking of larvae at different densities, had no significant effect on development. Although larvae are non-feeding, and development usually takes place at the oviposition site with very little water (G. Hollis pers. obs.), the ability to swim has been retained (Malone 1985a, b). Under natural conditions, the larval period varies from 5-10 weeks, at which time individuals metamorphose at an average length of 6.72 mm SVL (Malone 1985a).
Very little is known about the dynamics and demographic aspects of the Baw Baw Frog population. The little work that has been conducted documents mortality during embryonic and larval stages of development (Malone 1985b). Mortality during these stages of development is relatively high, measuring 74.0% and 70.3% respectively, with recruitment to the terrestrial stage measuring only 8.1% (Malone 1985b). Overall survivorship, however, is extremely variable among breeding localities. Malone suggested that periodic drying of oviposition sites was responsible for the high mortality rates observed, and indicated that these were not different to mortality rates recorded for other anurans. Predation on embryos and larvae appears minimal, with fungal infestations rapidly covering the capsules of dead embryos, and live embryos remaining resistant to infection (Malone 1985b).
The structure of vegetation has also been shown to significantly affect mortality of embryos and larvae. Mean embryonic and larval mortality rates are significantly higher in disturbed habitats (80.1% and 68.3% respectively) when compared to undisturbed habitats (62.5% and 44.5%) (Malone 1985a).
During the 1994 and 1995 breeding and post-breeding seasons, a mark-recapture study using 170 pitfall traps distributed at 2-3 metre intervals around the perimeter of a breeding site in sub-alpine wet heathland, had limited success in capturing frogs. All captures occurred during the breeding season, and most were females (1 male, 6 females - 1994; 5 females - 1995) (G. Hollis unpubl. data). Only one individual (a female) was ever recaptured, this occurring only several days after the first capture. In both years, gravid females were captured early in the breeding season, and spent females later in the breeding season, suggesting movement to and from breeding sites to lay eggs. No metamorphlings, juveniles or sub-adults were captured. From these very low capture rates, it has not been possible to glean information on growth rate, age-specific mortality or survivorship, age at sexual maturity or population structure of the species. Although mortality rates of frogs through predation are unknown, the Tiger Snake (N. scutatus) is known to consume the species (Spencer 1901), and other predators might include: the Alpine Copperhead (Austrelaps ramsayi), White-lipped Snake (Drysdalia coronoides), Southern Water Skink (Eulamprus tympanum), Bush Rat (Rattus fuscipes), Broad-toothed Rat (Mastacomys fuscus), Grey Currawong, Kookaburra, spiders and the introduced Cat, Fox and Dog/Dingo.
The relative abundance of calling males along different survey transects on the Baw Baw Plateau has been shown to vary from year to year. Surveys conducted by Malone (1985a) in 1983 and 1984 showed that numbers of calling males within some breeding units remained relatively stable between the two years of survey whilst numbers within others did not (see Table 1). Those that varied between 1983 and 1984 were mainly on the eastern side of the plateau, where a contraction in range of calling males was recorded in 1984. Malone speculated that this observation may have been due to the combination of lower spring and summer rainfall received on the Baw Baw Plateau in 1984, and the eastern side being drier in nature than other regions on the plateau. However, surveys conducted from 1993-1996 indicate no relationship between rainfall and numbers of calling males (G. Hollis unpubl. data). If no relationship exists between prevailing climatic conditions, such as rainfall, and rates of calling activity in a breeding season, the number of calling males active in a breeding season may therefore be related to breeding success and subsequent recruitment into the population several years prior. If desiccation of oviposition sites is the major cause of embryonic and larvae mortality, the low numbers of calling males recorded during 1993-1996 surveys may be a result of poor recruitment into the population during previously dry spring and summer seasons. In the mid to late 1980's, for example, relatively low spring and summer rainfall was received compared with records for the rest of the decade (Hollis 1995). This could be an explanation for the low number of frogs in the population at present, however, the effect of fluctuating rainfall patterns on the frog population needs to be examined over a longer time period rather than in the short-term where interpretation is more difficult.
A radio-tracking study conducted on the Baw Baw Frog over 6 months (Nov. 1995 - Apr. 1996) showed that movement by adults is variable and restricted to particular weather patterns. During the breeding season, some males remain at calling sites for most of the season, whilst others move away after several weeks (G. Hollis unpubl. data). By contrast, Malone (1985a, b) found that in the 1983 and 1984 surveys of the Baw Baw Frog, when they were relatively common, calling males continually disappeared from calling sites throughout the breeding season, only to be replaced by others. Radio-tracking and trapping studies indicate that females move into breeding sites from adjacent habitats to lay their eggs, then return to the same adjacent habitats (see 1.3.2). Movement by males at the completion of the breeding season varies, with some individuals remaining in close proximity to the breeding sites (<6 m), whilst others moved considerable distances (up to 70 m linear distance) (G. Hollis unpubl. data). Similarly, radio-tracking showed that some females also move relatively short distances from breeding sites (<8 m), whilst others move considerable distances away (up to 40 m) (G. Hollis unpubl. data). Gravid females have been pit-trapped in sub-alpine woodland habitats up to 70 m from known male calling sites early in the breeding season, and spent females at similar distances in the same habitat at the end of the breeding season (G. Hollis unpubl. data). It is not known whether frogs move during the cold months of the year when snow lies on the ground (June - September), although it is presumed that they remain relatively inactive like other ectotherms.
Detection of movement in both adult males and females has been found to be restricted almost entirely to rainfall events following several days of dry, relatively warm to hot weather (G. Hollis unpubl. data). During warm to hot, dry weather, frogs that were radio-tracked retained their position until the arrival of a rain-bearing frontal system. Similarly, frogs also retained their position during periods of very cold weather. Dial patterns of movement are unknown, although on a number of occasions diurnal movements of radio-tracked frogs were observed (G. Hollis unpubl. data). Movement is probably dependent upon local temperature and humidity conditions, and may occur at any time of the day or night. Individuals appear to retain their position at a site until the external weather conditions become similar to microclimatic conditions of the sheltering site (G. Hollis unpubl. data). The restriction of movement to these weather patterns may also correspond to periods of feeding activity. These observations indicate that movement and dispersal in the Baw Baw Frog is likely to be highly influenced by rainfall frequency and temperature conditions.
Virtually nothing is known about the movements of juveniles and sub-adults. The few observations that have been made indicate that metamorphlings remain in close proximity to breeding sites (M. Littlejohn pers. comm.; G. Hollis pers. obs.). Sub-adults maybe be more mobile.
The diet of the Baw Baw Frog is currently unknown.
The Baw Baw Frog coexists with four other amphibians, including the Alpine Tree Frog (Litoria verreauxii alpina), Southern Brown Tree Frog (L. ewingii), Common Froglet (Crinia signifera) and Southern Smooth Froglet (Geocrinia victoriana). The ecological relationships between these four species and the Baw Baw Frog are largely unknown. The breeding habitat of the Baw Baw Frog appears different from that of the other species. It utilises seepages on slopes and in gullies, whilst the Alpine Tree Frog, Common Froglet and Southern Smooth Froglet utilise pools or pockets of accumulated water and saturated sphagnum moss, soil and litter occurring in the basins of sub-alpine wet heathland or montane creek lines. The Southern Brown Tree Frog occurs at intermediate altitudes (800 - 1300 m a.s.l.) on the Baw Baw Plateau, where it hybridises with the Alpine Tree Frog (Watson et al. 1985). The timing of breeding seasons of Alpine Tree Frog, Common Froglet and Southern Smooth Froglet has been described as sequential, with the Alpine Tree Frog commencing breeding at the onset of the snow thaw, followed by the Common Froglet, whilst the Victorian Smooth Froglet breeds in autumn and has over-wintering larvae (Malone 1985b).
During 1993-1996 Baw Baw Frog surveys, Common Froglet larvae have been recorded in very large numbers in ponds on the Baw Baw Plateau (G. Hollis unpubl. data). Likewise the presence of G. victoriana larvae in many ponds across the plateau at this time would suggest that this species is also breeding successfully. In contrast to this, only one record of the Alpine Tree Frog was obtained during the three survey years (during late spring 1994), this a calling male in the vicinity of the Mt Baw Baw Alpine Resort. By comparison, Alpine Tree Frog larvae and frogs were found to be very common during the 1983 and 1984 surveys for the Baw Baw Frog (B. Malone pers. comm.). According to Gillespie et al. (1995), prior to the mid 1980's the Alpine Tree Frog was considered abundant throughout its distribution, but now appears to have declined at many localities, particularly those at higher altitudes. The status of the Southern Brown Tree Frog above 1000 m is unknown.
Revised assessments of the conservation status of the Baw Baw Frog have listed the species as nationally endangered (Tyler 1997) and critically endangered in Victoria (NRE 1997a). Evidence from the recent surveys of a precipitous population decline has necessitated this change from a listing as vulnerable.
The discovery of the Baw Baw Frog at Mt Baw Baw in 1898 by C. Frost, where he retrieved five relatively fresh regurgitated specimens from a Tiger Snake (Notechis scutatus), and later collected two live specimens, might suggest that the species was then relatively easy to find (see Spencer 1901). Records collected from the plateau by University of Melbourne and Museum of Victoria personnel during the 1950's, 1960's and 1970's indicate that the species was then relatively common and easy to locate (M. Littlejohn unpubl. data and pers. comm., J. Coventry pers. comm.). Likewise, a brief search for the species in the vicinity of the Mt Baw Baw Alpine Resort in 1982 recorded approximately 12 individuals (AVW). These frogs were also relatively easy to locate (P. Robertson pers. comm.).
The first systematic survey for the Baw Baw Frog in 1983 and 1984 also found the species to be common (see Malone 1985a). Malone estimated the number of adult males in the Baw Baw Frog population to be 10,000 - 15,000. He actually recorded a total of 4248 calling males, but because he frequently locating silent males, and more than one male at a calling site, a more realistic estimate of the adult male population was taken to be two to three times the number recorded. Between 1985 and 1988 there are no records of the Baw Baw Frog (Atlas of Victorian Wildlife). Records of the Baw Baw Frog between 1989 and 1992, although scant and anecdotal, indicate that numbers of calling males were much lower than that reported by Malone in 1983 and 1984 (W. Osborne, J. Morey, P. Johnson pers. comm.).
Results of systematic surveys for calling males conducted from 1993-1996 are in accordance with the later observations of low numbers. Using the same census methodology as Malone (1985a), and from surveys of the same areas, the abundance of calling males in each of the three years was 1-2% of the number recorded by Malone (1985a) over a decade ago (Hollis 1995; G. Hollis unpubl data) (Table 1). Table 1 contains the census statistics for individual breeding units (transects) surveyed across all years. From 13 breeding units censused across all survey years (1983-1996), the total number of calling males recorded in each year was: 1983-867, 1984-847, 1993-18, 1994-10, 1995-11 and 1996-6 (Table 1) (Fig 5). Hollis (1995) showed also that measures of relative abundance obtained from individual breeding units during the 1983, 1984 and 1993 surveys were highly correlated, indicating that the pattern of decline is uniform across the plateau, with numbers within each breeding unit remaining proportionally the same. Also, most breeding unit statistics obtained from the 1994, 1995 and 1996 surveys were proportionally similar to 1993 (Table 1).
In 22 comparable breeding units censused from 1993 - 1996, the total number of calling males recorded in each year was : 1993-76, 1994-47, 1995-31 and 1996-42 (Table 1) (Fig. 6). Although there is an overall decline trend in the census results, it is not known if this decline is real, or if it can be explained by seasonal variability or variation in the timing of the surveys.
No calling males were recorded from the eastern region of the Baw Baw Plateau and its vicinity during the 1993-1996 surveys (Hollis 1995; G. Hollis unpubl. data). Although Malone (1985a) recorded calling males in this region in 1983, he also recorded an absence of frogs there in 1984. It appears as though this region may only provide sub-optimal conditions for the Baw Baw Frog, with recruitment only being successful during more favourable years of breeding. This region is drier and rockier then other regions on the plateau (Hollis 1995), and seems to provide only small pockets of the breeding habitat (G. Hollis pers. obs.). A rain shadow is known to occur on the eastern side of the Baw Baw Plateau (DCE 1991) which may explain the drier conditions that seem to be present. It is not known whether numbers of adult males have declined in lower altitude montane forest areas. Except for Long Creek where 35 calling males was recorded (Table 2), numbers of calling males recorded in other montane forest areas are not dissimilar from numbers recorded at higher elevation.
The proportion of calling males recorded in different habitats during the 1993-1996 surveys (Hollis 1995; G. Hollis unpubl. data) appears different to that reported by Malone (1985a) in his 1983 and 1984 surveys (see 1.3.1 above). Calling males recorded by Malone were distributed over a wider area within sub-alpine wet heathland than found in the later surveys, where males were restricted predominantly to topographically protected gullies in wet sub-alpine heath, montane riparian thicket and montane wet forest vegetation. This apparent contraction in range of breeding habitat corresponds with an altitudinal decrease in the distribution of the species (Hollis 1995; G. Hollis unpubl. data).
Due to the cryptic nature of the species, it has not been possible to obtain population estimates of the female, sub-adult and juvenile frog populations. By examining gonadal development at the early terrestrial stage (metamorphlings), Malone (1985a) attempted to determine whether sex ratios were equal in the population and thus provide an indication of the female population size. This failed due to the lack of development of gonads in the metamorphlings he examined.
The dearth of information available on the population dynamics and demography of the Baw Baw Frog makes it is very difficult to interpret the observed decline in the adult male population. This decline however, coincides with a recent global phenomenon of declining amphibians (e.g. Barinaga 1990; Blaustein and Wake 1990; Tyler 1991), and perhaps of more concern, with the reported decline of species restricted to mountain-top and alpine environments in Australia (e.g. Osborne 1990; Richards et al. 1993) and overseas (La Marca and Reinthaler 1991; Crump et al. 1992). Numerous hypotheses have been generated to explain these declines, but few as yet have been investigated thoroughly. Those that have, or are currently being investigated, include climate (Osborne 1990; Richards et al. 1993), ultraviolet radiation (Blaustein et al. 1994), atmospheric pollution (Bradford et al. 1994) and pathogens (Blaustein et al. 1994;
Figure 5. Relative abundance of calling male Philoria frosti recorded from 13 breeding units censused in each of years 1983, 1984 and 1993 - 1996 (see table 1). NB. Y scale is logged.
Figure 6. Relative abundance of calling male Philoria frosti recorded from 22 breeding units censused in each of years 1993 - 1996 (see table 1).
Trenerry et al. 1994). No thorough investigation of these hypotheses has been undertaken for the Baw Baw Frog, or for any other Australian 'mountain-top' species.
Obvious anthropogenic disturbances on the Baw Baw Plateau are few when compared to habitats outside the Baw Baw National Park. The main disturbance on the plateau is the presence of the Mt Baw Baw Alpine Resort in the vicinity of Mt Baw Baw. This area it is restricted to 3.5 km2 of the 80 km2 plateau. Other disturbances include a network of bushwalking and ski trails traversing the plateau, areas at montane elevations that have undergone clearfell timber harvesting, and the presence of introduced animals, including cattle, Sambar deer, rabbit, cat, fox and dog. Few introduced plants occur on the plateau (Walsh et al. 1984; J. Davies pers. comm.), with most restricted to the vicinity of areas subjected to clearance or ground disturbance (DCE 1992). Salix cinerea (Grey Sallow), however, is becoming increasingly abundant throughout the wetland systems on the plateau (G. Hollis pers. obs.).
The impact that vegetation clearance may have on biology and ecology of the Baw Baw Frog is only partially understood. Malone (1985b) showed that mortality rates of both embryos and larvae are significantly higher in breeding habitats that have been disturbed (see 1.4.2). These early stages of amphibian development are considered to be the most vulnerable phases of the anuran life cycle (e.g. Duellman and Trueb 1986; Williamson and Bull 1994). The recent decline in the number of calling males within the Mt Baw Baw Alpine Resort (where disturbance to vegetation has occurred), however, appears not to be significantly different to declines recorded elsewhere on the plateau, although the frogs recorded in the resort area occurred primarily in habitats that had not been cleared or modified (G. Hollis unpubl. data). Located on the western side of the plateau, the Mt Baw Baw Alpine Resort is positioned within the stronghold area of the species distribution. Remaining intact sub-alpine and montane vegetation in this area represents an important component of the species optimal habitat.
Until recently, the literature had described the habitat of the Baw Baw Frog as only in wetland systems on the Baw Baw Plateau (e.g. Malone 1985a; Cogger 1994; Barker et al. 1995; Hollis 1995) (see 1.3.1). The species is now known to also utilise sub-alpine woodland, montane wet forest and ecotonal habitats adjacent to breeding sites (G. Hollis unpubl. data) (see 1.3.2 and 1.4.3). These non-breeding habitats are likely to be extremely important for shelter, foraging and dispersal. Fragmentation or modification of these habitats may limit dispersal, increase predation, destroy sheltering sites, alter sheltering microclimates and affect local diversity and/or abundance of invertebrates that are important sources of prey for the species.
At lower elevations (1000-1300 m), timber harvesting activities may pose a threat to populations occurring in montane forest and montane riparian thicket communities. Harvesting activities have the potential to reduce local populations, destroy sheltering sites, affect prey abundance, alter micro-climates, fragment habitat and allow the ingress of exotic predators and weeds if appropriate forest management strategies are not put in place (see 1.6.4 below).
Bushfires on the Baw Baw Plateau are uncommon due to its wet climate (DCE 1991), but may pose a threat to the Baw Baw Frog, particularly when populations are low in number as they are now. Current fire policy excludes burning on the plateau (sub-alpine) section of the Baw Baw National Park (DCE 1992). Historical records, although incomplete, indicate that the Baw Baw Plateau, or parts of it, have been burnt on a number of occasions previously, including 1851, 1898, 1926, 1932 and 1939 (DCE 1991). The plateau has been free of serious fires for the past 55 years. Fires that burn during spring, summer and autumn periods are likely to have the greatest impact on the Baw Baw Frog as this is the time when adults are known to shelter amongst vegetation and litter in snow-gum woodland and montane wet forest. Fire protection responsibility, objectives and strategies put in place to prevent and control fire on State Forest, is contained in DCE (1990).
The impact of introduced animals on the Baw Baw Frog and its habitat is unknown. Cats, dogs and foxes are relatively common on the plateau and are potential predators (G. Hollis pers. obs.). Exotic species may also be potential vectors for pathogens that may be detrimental to amphibians (e.g. Blaustein et al. 1994; Speare et al. 1994; Laurance et al. 1996). During spring, summer and autumn, Sambar deer are quite common on the plateau, particularly in montane forest and montane riparian thicket (G. Hollis pers. obs.). Recent surveys for the Baw Baw Frog show montane riparian thicket to be increasingly important for breeding purposes (see 1.3.1). Despite attempts to eradicate cattle from the Baw Baw Plateau, a small, but increasing, number (approximately 12-24) are still present (DCE 1992; G. Hollis pers. obs.). Cattle occupy higher altitudes on the plateau than Sambar deer, grazing within sub-alpine wet heathland, grassy sub-alpine woodland and cleared areas. Damage caused by cattle to vegetation and soil is localised but significant (G. Hollis pers. obs.) and may affect both breeding and non-breeding habitats.
The impact that recreational activities, and the associated network of bushwalking and ski trails, has on the Baw Baw Frog is also unknown. The placement of some trails, particularly in wetland habitats, has resulted in damage to local vegetation communities through the removal of vegetation during construction, degradation through use and altered drainage patterns (G. Hollis pers. obs.). This damage may be detrimental to local breeding populations, and large areas of cleared trails and ski areas may impede dispersal and movement or result in increased predation on the species.
Climate change has also been suggested as a potential threat to the Baw Baw Frog, given that the species is geographically localised, restricted to alpine and montane environments, and has specialised life history characteristics (see Bennett et al. 1991). Although excluded from the BIOCLIM analysis conducted by Bennett et al. (1991) because of its very small distribution, the authors suspect that the bio-climate of the Baw Baw Frog would disappear with a 1-3 oC rise in temperature. Climate change has been implicated as being responsible, in part, for the decline in the Corroboree Frog (Pseudophryne corroboree) (Osborne 1990), also an alpine species with a restricted distribution. Climate change, either due to long-term changes through enhanced greenhouse effects, or other shorter-term natural environmental fluctuations, may be responsible for the recent decline observed in the Baw Baw Frog population.
The unexplained nature of the recent decline of the Baw Baw Frog, and the relatively pristine environment in which the species occurs, leads to consideration of atmospheric factors such as increased ultra-violet (UV) radiation and air pollution as potential threats. Increased levels of terrestrial UV radiation associated with depletion of the stratospheric ozone may be a potential threat to the Baw Baw Frog. Recent evidence shows that increases in UV-B have occurred in relatively undisturbed temperate latitudes (Blumthaler and Ambach 1990), and progressive expansion of affected areas to lower latitudes is expected (Worrest and Grant 1989). The Baw Baw Frog lives in a relatively high altitude environment (1000 - 1564 m a.s.l.) and is located at a temperate latitude (37-38o).
A study of the impact of UV-B radiation on a selection of amphibian species from the United States showed that species that had declined possessed lower levels of UV-damage-specific repair enzyme (photolyase) than species that had not declined (Blaustein et al. 1994). Both field and laboratory experiments have also shown that exposure of embryos to levels UV radiation resulted in increased mortality (Blaustein et al. 1994). Both ova and larval stages of the Baw Baw Frog are unpigmented, indicating a potential to be sensitive to UV radiation exposure. However, recent observations of oviposition sites shows that in most cases embryonic and larval development occurs beneath the substrate (G. Hollis unpubl. data). This would suggest that these developmental stages are subjected to very little UV radiation. However, B. Malone (pers. comm.) found that during his studies on embryonic and larval development, some of the egg masses he was monitoring were located in semi-exposed positions within the vegetation.
Alterations to air quality through industrial and agricultural emissions has the potential to be detrimental to the Baw Baw Frog. The Latrobe Valley, located in close proximity south of the Baw Baw Plateau, is largely an agricultural and industrial area where activities such as coal mining, electricity generation, timber milling, paper production and farming occur. Acidification (acid deposition) from industrial emissions has been implicated as a potential reason for the decline in some amphibian populations, particularly those restricted to montane, high elevation or mountain areas (e.g. Blaustein and Wake 1990; Wyman 1990; Pounds and Crump 1994). Fallout of contaminants from industry in the Latrobe valley may be a threat to the Baw Baw Frog. However, a recent assessment of the affects of emissions from power stations in the Latrobe Valley region over 10 years (late 1980's to early 90's) found that they are not significantly adding to acidification (Ayers 1995). The direction of wind associated with frontal systems that frequent southern Victoria are usually from the west or south-west. The location of the Baw Baw Plateau in respect to the Latrobe Valley would result in many emissions from the Latrobe valley being blown away from the Baw Baw Plateau.
Potential changes to water quality and altered flow regimes from the Mt Baw Baw Alpine Resort may also pose a threat to local populations of the Baw Baw Frog by affecting habitat quality, breeding microclimates, embryonic and larval development and subsequent recruitment.
Disease has been implicated as a possible explanation for the decline and disappearance in some amphibians (e.g. Hunter et al. 1989; Bradford 1991; Blaustein et al. 1994; Trenerry et al. 1994; Laurance et al. 1996 ). However, in very few studies has the demise of an amphibian population been attributed largely to a pathogen, with identification of the pathogenic species (e.g. Blaustein et al. 1994), with most remaining circumstantial or speculative. In many cases, the disease hypothesis is largely supported by the discovery of moribund frogs only, with no rigorous assessment and experimental design to directly link pathogenic species with the demise of amphibian taxa or groups of taxa.
The Baw Baw Frog is currently listed as a threatened species under Schedule 2 of the Victorian Flora and Fauna Guarantee Act 1998. An Action Statement was completed for the species by the then Department of Conservation and Natural Resources in 1993 (CNR 1993), outlining the actions to be taken to ensure its long term survival. This Action Statement is due for revision in 1998. The following intended management actions are listed in the Statement: (a) Monitoring and survey, (b) Critical habitat assessment, (c) Pest plants and animals, (d) Recreation management and (e) Visitor education.
Since the Action Statement was approved in 1993, the majority of the actions proposed have been implemented or undertaken. These include:
The vegetation community 'wet sub-alpine heathland' is also listed on Schedule 2 of the Victorian Flora and Fauna Guarantee Act 1988, and subsequently provides for its protection. This community is frequently used by the Baw Baw Frog for breeding and non-breeding purposes.
The Baw Baw Frog is protected under the Wildlife Act 1975 as an endangered species. The species is also protected under the National Parks Act 1979. Measures for the conservation and management of the Baw Baw Frog, and the Baw Baw Plateau, are outlined in the Baw Baw National Park Management Plan (DCE 1992).
The Baw Baw Frog is currently listed as vulnerable on Schedule 1 of the Endangered Species Protection Act 1992. This Act provides a framework within which States and Territories can seek assistance from the Commonwealth for the preparation and implementation of Recovery Plans (this document).
The Baw Baw Frog is listed on The Action Plan for Australian Frogs (Tyler 1997) as an endangered species. The objectives outlined in this document are:
Recent records of the Baw Baw Frog located on State Forest lands in 1996 have resulted in the establishment of management guidelines to protect the Baw Baw Frog on State Forest. These guidelines are contained within the draft Central Highlands Management Plan (NRE 1997b). The following text is taken directly from this plan.
'Threats to the Baw Baw Frog on State Forest may include physical disturbance, habitat modification and fragmentation and increased sedimentation of drainage lines in Montane Wet Forest and Montane Riparian Thicket.
To ensure that viable populations of the Baw Baw Frog persist in the long-term, a sufficient area of its habitat must be protected. Areas of habitat on State Forest will be protected once the frogs' use of breeding and non-breeding habitat is better understood. To achieve this, the following research and survey will be undertaken:
The Baw Baw Frog Recovery Plan includes research proposals which will increase knowledge of the frog's biology, ecology and habitat distribution.
The results of the research and survey listed above are expected to be available in three to five years. In the interim, NRE will adopt a precautionary approach and will minimise disturbance to Baw Baw Frog habitat by:
Where timber harvesting above the 1000 metre level on the Baw Baw Plateau is proposed in the next 3 - 5 years, this interim strategy provides for:
Long-term protection measures for the Baw Baw Frog will be guided by the results of the survey and research obtained during the next 3 - 5 years. It is expected that these measures will be a combination of zoning for conservation and guidelines for scheduling and spatial distribution of timber harvesting coupes'.
The use of computer programs to model the long-term viability of species (e.g. VORTEX) requires detailed information about population demography and dynamics. For the Baw Baw Frog, information on age at sexual maturity, longevity, juvenile, sub-adult and adult mortality are required. Clutch size and embryonic and larval mortality are already known. A Population Viability Analysis cannot be conducted on the Baw Baw Frog until this demographic information is obtained, as well as knowledge of the dynamics of the population, including movement and dispersal rates between breeding sites and the likelihood and effect of environmental stochastic events such as drought.