Threats to the critically endangered Arid Bronze Azure (Ogyris subterrestris petrina) by proposed vegetation clearing

Matthew R. Williams, Senior Research Scientist, DEC Science Division. Andrew A. E. Williams, Senior Technical Officer, DEC Science Division.

4 November 2008

Summary

The critically endangered Arid Bronze Azure butterfly (Ogyris subterrestris petrina Field 1999) has been recorded from only two localities, both in south-west Western . No have been seen at one locality since 1993 and this population is considered extinct. The habitat of the remaining population, in remnant vegetation adjacent to Barbalin Nature Reserve, is subject to a clearing application. This report (i) records the significance of this site to the long-term survival of the butterfly; (ii) examines the likely effect of clearing and associated disturbance on the butterfly population; (iii) documents the surveys that have been conducted for this butterfly; (iv) discusses any potential for other populations to occur in the region; and (v) examines the likelihood that the butterfly population, if displaced, could disperse and colonise other sites nearby.

1 Introduction

The butterfly. The Arid bronze Azure butterfly (Ogyris subterrestris petrina) is a medium-sized, robust butterfly capable of rapid flight (Figures 1, 2). It is predominantly dark brown to black, with bronze to purple-bronze colouration on the uppersides of the wings. The underside is grey-brown, with iridescent bluish-white bars in the forewing cell and a series of brown-black markings on the hindwing. It has pronounced sexual dimorphism: males have dull dark-purplish bronze uppersides whereas females have purplish areas on the uppersides of both wings and a cream patch on the forewing.

This butterfly is readily identifiable because it has several distinctive features: (i) a very rapid and direct flight pattern, characteristic of the ; (ii) uniquely for butterflies in the region, a very dark colouration (adults, especially males, appear almost black in flight); (iii) a habit of perching on the ground in open areas (including on roads and roadsides); and (iv) are among the largest lycaenid butterflies in Australia. These characteristics make Ogyris butterflies easily identifiable in the field, to the extent that populations have been detected from vehicles. The Barbalin population of O. s. petrina was discovered in this way when an amateur entomologist drove through the site and recognised the butterflies as Ogyris sp. as they flew across the road.

More than 400 species of butterflies are recorded for Australia and few more are likely to be discovered, making them the best-known group (Braby 2000). Members of the genus Ogyris are large and spectacular lycaenid butterflies that are restricted to the Australian region. Of the 13 species in Australia, 10 were discovered before 1900 and remainder between 1900 and 1914. The last species to be described, O. subterrestris, was discovered in 1912 but was so rare that it was not described as a distinct species until extant colonies were discovered in the 1980s. No butterfly species of comparable size has been discovered in Western Australia since 1913. Because of their relatively large size and often spectacular colouration, Ogyris species are prized by butterfly collectors who actively seek out populations. Occurrence of most species is highly localised. The larvae are attended by ants and feed mainly on mistletoes and

2 related parasitic plants but two species (O. idmo and O. subterrestris) are myrmecophagous, feeding exclusively on the larvae and pupae of the attendant ants.

The Arid bronze Azure was described from a series of 72 specimens collected between 1982 and 1991 at Lake Douglas, 12 km SW of Kalgoorlie, Western Australia. The holotype female was collected at this site by Ross Field on 12 Nov 1989 (Museum of Victoria accession number NMV T-17268). The new species was distinguished from the related species Ogyris idmo, O. otanes and O. waterhouseri by a series of morphological traits. When described, O. subterrestris was recognised as comprising two subspecies: O. s. subterrestris (the Mallee Bronze Azure) from three extant localities in South Australia, New South Wales and Victoria, and O. s. petrina from the single locality in Western Australia. A subsequent study of mitochondrial DNA confirmed the taxonomic distinction between O. subterrestris and the related species, and also between the two subspecies of O. subterrestris (Schmidt et al. 2005). Based on molecular clock methods, Schmidt et al. (2005) concluded that the two subspecies had diverged between 3 and 4 million years ago.

In order to reproduce, the butterfly has an obligate association with a single form of one species of ant, the pale-coloured or ‘Goldfields’ form of Camponotus terebrans. The larvae (Figure 3) are fed by or predate the ants and live entirely within the ant’s nest during their development (Field 1999). Studies of other ant-dependent butterfly species have shown that a certain number and density of ant colonies is required in an area before the site can support a breeding colony of butterflies: an approximate ratio of ants to butterflies is 500 to 1 (Griebeler & Seitz 2002). Where the butterfly occurs, the ant colonies are enormous (R. Grund, Butterflies of South Australia website: http://users.sa.chariot.net.au/~rbg/subterrestris_ds.htm). Male butterflies typically form small flight areas or territories (leks), approximately 5-20m in extent, adjacent to the breeding area. They patrol these leks and defend them against other males. Males also establish territories on hilltops if these are near the breeding area, a common behaviour amongst butterflies. Unmated females visit the leks or hilltops to mate, but thereafter practice male avoidance and remain within the breeding area to lay eggs. Most of the eggs are probably laid in the original breeding area, but some females probably disperse to search for new breeding areas. This is the usual life history pattern amongst ant-tended

3 lycaenid butterflies (Pierce et al. 2002). Typically, each female lays more than half of her eggs in the original colony and dispersal commences thereafter. If successful in locating another ant population these dispersals may result in the establishment of new butterfly colonies. Typical dispersal distances are not known for females of O. s. subterrestris, but a study of a similar lycaenid butterfly in south-west Australia, Hypochrysops halyaetus, found that of over 1000 marked individuals the longest dispersal distance was 810 m for one female, but averaged less than 200 m (M. R. Williams, unpublished data).

The butterflies are active and breed continuously between October and April. All life stages (eggs, larvae and adults) are present during this period. Between May and September, the entire population consists of larvae only, all of which are within ant colonies. There is a peak in butterfly numbers in October, when the over-wintering larvae complete the life cycle and emerge as adults.

The Threatened Species Scientific Committee (TSSC) assessed the taxon as Critically Endangered (A3c, B1 + 2ab(i, ii, iii, iv, v)) and the butterfly was gazetted as Schedule 1 (rare or likely to become extinct) in August 2008 (Government Gazette 5 August 2008, p 3483).

Known populations. A single specimen of O. s. petrina was collected in October 1912 from the ‘Kalgoorlie District’ and lodged in the British Museum. No further specimens were known until Prof. Alan Graham of the WA School of Mines in 1982 discovered a population at Lake Douglas, 12 km SW of Kalgoorlie. Prof. Graham conducted regular (typically weekly) visits to the site and collected several specimens. These were provided to the Australian National Insect Collection, which confirmed that the butterfly represented a known but undescribed species of Ogyris. Several professional and amateur entomologists also visited the site during 1982 – 1992 and collected specimens which are held in private and public collections. At Lake Douglas the butterfly occurred intermittently over a large area, approximately 3 x 1.5 km.

Butterflies were locally abundant at the site between 1982 and 1991, but in 1992 underwent a severe decline. Prof. Graham observed only two individuals in 1993 and no

4 butterflies have been seen at the site since then, despite numerous visits by both amateur and professional entomologists (Table 1).

Table 1. Known searches for Ogyris subterrestris petrina at the Lake Douglas site between 1993 and 2007. Date(s) Observer(s) Time spent at Survey method Results the site 1993-2002 A Graham Weekly visits Walk surveys No butterflies seen Oct 1992 M Williams & 1 day Walk surveys No butterflies seen R Hay Oct 1994 M Williams & 1 day Walk surveys No butterflies seen R Hay Oct 1995 M Williams, 2 days Walk surveys No butterflies seen R Hay & T Lundstrom Oct 1997 S Johnson & 2 days Walk surveys No butterflies seen P Valentine Oct 2000 M Williams, 2 days Walk surveys No butterflies seen A Williams, D Lohman & A Graham Oct 2001 S Brown & 1 day Walk surveys No butterflies seen R Weir Oct 2002 J Landy 1 day Walk surveys No butterflies seen Oct 2003 J Landy 1 day Walk surveys No butterflies seen Oct 2003 P Samson 2 days Walk surveys No butterflies seen Oct 2003 A Williams 2 days 2 x 3km structured No butterflies seen walk transect surveys Dec 2003 A Williams 2 days 2 x 3km structured No butterflies seen walk transect surveys Apr 2004 A Williams 2 days 2 x 3km structured No butterflies seen walk transect surveys Nov 2005 J Landy 1 day Walk surveys No butterflies seen Oct 2007 S Brown, 1 day Walk surveys No butterflies seen G Miller

The surrounding areas have also been searched extensively but no butterflies have been seen. In particular, during the period that he resided in Kalgoorlie between 1982 and 2002, Prof. Graham undertook numerous field trips into the surrounding areas. He also

5 conducted a significant amount of field work in his role as Professor of Geology at the Curtin University WA School of Mines. During this time he did not locate any further populations of O. s. petrina. In 2005 an article on the butterfly was published in Landscope magazine, requesting that any sightings of the butterfly be reported to DEC (Williams & Williams 2005). No sightings of the Arid Bronze Azure butterfly were reported. The Lake Douglas population is therefore considered extinct.

The cause of the extinction of the Lake Douglas population is believed to be due to increased disturbance in the area from off-road vehicles and nearby mining activity (A Graham, pers. comm.). Detailed surveys of the known breeding area at Lake Douglas were conducted and these determined that the host ant was now absent. However, numerous colonies of the Meat Ant purpureus were found, a common disturbance-opportunist ant species. An experimental study has shown that other ant species are strongly and negatively impacted by I. purpureus, particularly Camponotus species (Andersen & Patel 1994). The replacement of the host ant by Meat Ant colonies at Lake Douglas supports the hypothesis that disturbance had enabled this ant to colonise the area. As a result, the Camponotus colonies were negatively impacted through direct predation or competition, being extirpated or reduced in density to a level where they were no longer able to support a viable population of the butterfly.

A second population of O. s. petrina was located 11 km west of Mukinbudin (S35°51.792′, E140° 27.532′), adjacent to Barbalin Nature Reserve, in October 2006 by Geoff Walker, a visiting amateur entomologist. Specimens collected at that time were subject to morphological and mitochondrial DNA analysis which confirmed them to be O. s. petrina (D. Schmidt, Griffith University Qld, pers. comm).

Subsequent visits to the site by DEC staff confirmed the presence of a breeding population and delineated the extent of the flight and breeding areas. The flight area was found to extend approximately 700 m east to west and 30 to 50 m north to south, on the southern side of the Wyalkatchem-Southern Cross road, east of the intersection with Barbalin North and South roads. Thus this colony is considerably smaller than the extinct Lake Douglas population. Surveys of ant colonies within this area, which includes the breeding centre of the O. s. petrina colony, and also in the adjacent Nature Reserve, have

6 been undertaken. There are particularly large colonies of ants both within and adjacent to the flight area. These large ant colonies directly correspond with the remnant vegetation within the area under application for clearing (A. Williams 2008, Brown 2008). The high density of ants within the area under application is believed to be the reason that the butterfly colony occurs and is concentrated within this area.

Searches of areas surrounding the Barbalin population. DEC technical officer Andrew Williams conducted surveys within Barbalin Nature Reserve to the north and south of the known flight area but did not sight any butterflies or find any additional populations. He also examined the larger portion of Barbalin NR to the east of the known site, but found no butterflies, additional populations, or areas of similar habitat. A detailed search of all the fire breaks surrounding the east and western portions of the Barbalin Nature Reserve did not find any evidence of butterflies. Remnant vegetation to the north of the Barbalin Nature Reserve has been searched and no evidence of the butterfly and no Camponotus ant colonies were observed. There have not yet been any detailed surveys of vegetation remnants to the south of Barbalin Nature Reserve by DEC staff.

At the request of DEC all of the visitors to the site have examined remnant vegetation in nearby areas in search of additional populations of the O. s. petrina. A survey was conducted over a six-day period in October 2007 by Dr. S. S. Brown and Dr. C. G. Miller. Both are experienced amateur entomologists with extensive knowledge of Australian butterflies and have made numerous contributions to the scientific literature. Their observations and subsequent report (Brown 2008) confirmed the findings of DEC staff. They found numerous colonies of Camponotus terebrans ants within the main flight area, but their surveys of nearby remnant vegetation and around the Mukinbudin area did not locate any other occurrences of the butterfly. Overall, searches of bushland remnants within a 25 km radius to the north and south of the known site failed to locate any butterflies or areas of apparently similar habitat (Table 2).

7 Table 2. Searches and surveys for Ogyris subterrestris petrina at sites around Barbalin NR between 2006 and 2008.

Date(s) Observer(s) Time spent at Survey method Results the site Oct-06 G Walker 1 day Walk surveys Numerous adults seen flying along the 700 metre stretch of the main road. Oct-06 A Williams 1 day Walk surveys More than 20 adults seen flying along the 700 metre stretch of the main road, none seen in Barbalin NR or elsewhere. Feb-07 D King 1 day Walk surveys No butterflies seen. Mar-07 A Williams 1 day Walk surveys Four male butterflies seen, all along the main road within the original flight area. None seen in Barbalin NR or elsewhere. Apr-07 A Williams 1 day 4 km transect No butterflies seen in Barbalin established NR or elsewhere. Oct-07 A Williams 2 days 3km structured 13 adults seen along the 700 walk transect metre stretch of the main road, surveys only one outside the main flight area. None seen in Barbalin NR or elsewhere. Oct-07 S Brown, 3 days Walk surveys ~ 25 butterflies seen, all within G Miller the main flight area along main road. None seen in Barbalin NR or elsewhere. Nov-07 D King 1 day Walk surveys No butterflies seen. Oct-08 D King 1 day Walk surveys Adult butterflies seen along the 700 metre stretch of the main road, none seen in Barbalin NR or elsewhere.

Regional searches for additional populations. Over 30 surveys have been conducted in the region by DEC staff and experienced volunteers between 1992 and 2008. The details of sites visited are documented in references 9-38, and include extensive surveys between Payne’s Find and Kalgoorlie, including most of the major conservation reserves. Hilltops have been targetted in most of these surveys. The surveys have covered extensive parts of the region in which O. s. petrina might occur, but have not detected any individuals or additional populations of the butterfly. The Western Australian Museum conducted a

8 major survey of invertebrates (including butterflies) around Comet Vale, 100 km N of Kalgoorlie, in 1979 but no specimens of the Arid Bronze Azure butterfly were recorded. A number of experienced amateur entomologists from other states of Australia have also conducted field work within this region over the past 30 years. The fact that further populations have not been located, despite the species being conspicuous, demonstrates the rarity of this butterfly and the significance of the Barbalin site.

Impacts of clearing

The area currently under application for clearing (Figure 4) is within the only known breeding area of O. s. petrina. The Camponotus ant colonies within this area form the centre of this breeding colony. Clearing the current area under application would have two major effects on the Camponotus ant population:

1) It would directly impact the number of ants through destruction of their colonies; and

2) The associated disturbance would likely favour the establishment of opportunist ants, such as the Meat Ant, leading to further declines and probable extirpation of any remaining ant colonies not impacted directly by clearing.

These impacts would disrupt the breeding cycle of the butterfly and cause extinction of the colony. Brown (2008) concluded that the destruction of trees would almost certainly decimate the butterfly population, as eggs of O. s. petrina have all been found at the base of the trees which have ant colonies beneath them (Field 1999). Similarly, Braby (2000) concluded that the species is particularly sensitive to habitat disturbance. A population of the closely related O. s. subterrestris in Mildura, NSW became extinct following earthworks and soil disturbance (Douglas 1995, cited in Braby 2000).

However, if clearing was to proceed the timing would be important to allow the maximum numbers of butterflies, particularly gravid females, to escape immediate destruction. If clearing of vegetation occurred between May and September, when the

9 population consists entirely of larvae, destruction of ant colonies would also destroy all of their resident O. s. petrina larvae. If clearing took place between October and April, when adults are actively flying and all life stages are present, some individuals would be able to disperse into adjacent areas. The time when the highest proportion of mated adult females are present is likely to be mid to late October. Clearing at this time would allow the maximum number of females to lay their eggs in any Camponotus ant nests that have been unaffected by the clearing. However, there are apparently no areas nearby containing suitable habitat for the butterfly. Translocation of gravid females to the Lake Douglas site to attempt to re-establish that population is possible, although surveys indicate that no host ants are now present at that site and the likelihood of success is negligible.

We conclude that if the proposed clearing takes place the butterfly population will be severely impacted through the destruction of all or the majority of its breeding area, and destruction of most or all of the colonies of the host ant. Because the butterfly requires large numbers of ant colonies to sustain a population, the proposed clearing would almost certainly result in destruction of the only known population of Ogyris subterrestris petrina. Although clearing during late October would provide an opportunity for dispersal to nearby areas there is virtually no likelihood that the butterflies would successfully colonise these areas, which lack the necessary large colonies of the host ant. Given the direct and indirect impacts of the disturbance and very low likelihood of successful dispersal it is considered that the clearing as proposed will almost certainly result in the loss of this butterfly population and therefore cause the extinction of Ogyris subterrestris petrina.

10 References

1. Andersen, A.N. & Patel, A.D. (1994) Meat ants as dominant members of Australian ant communities: an experimental test of their influence on the foraging success and forager abundance of other species. Oecologia 98: 15–24. 2. Braby M. F. (2000) The butterflies of Australia: their identification, biology and distribution. CSIRO publishing, Collingwood, VIC. 3. Brown, S.S. (2008). Report relating to Licence To Take Fauna For Scientific Purposes No. SF005970. Unpublished report to DEC, 2pp. 4. Douglas, F. (1995). Recovery plan for threatened diurnal in Western Victoria. Part 2: Family . Report to the Dept. of Conservation and Natural Resources, 100 pp. 5. Field, R.P. (1999) A new species of Ogyris Angas (Lepidoptera: Lycaenidae) from southern arid Australia. Memoirs of Museum Victoria 57: 251-259. 6. Griebeler E.M. & Seitz, A. (2002) An individual based model for the conservation of the endangered Large Blue Butterfly, Maculinea arion (Lepidoptera : Lycaenidae). Ecological Modelling 156, 43-60. 7. Pierce N. E., F B. M., Heath A., Lohman D. J., Mathew J., Rand D. B. & Travassos M. A. (2002) The ecology and evolution of ant association in the Lycaenidae (Lepidoptera). Annual review of Entomology 47, 733-771. 8. Schmidt, D., Grund, R. & Hughes, J. (2005). East-west relationships in Australia revisited: genealogical concordance among three threatened Ogyris butterflies restricted to southeastern and southwestern Australia. Paper presented at the 2005 Annual conference of the Society for Molecular Biology and Evolution, Auckland, NZ June 2005. 9. Atkins A. F., Williams A. A. E. & Williams M. R. (2002) Exometoeca nycteris Meyrick (Lepidoptera: Hesperiidae: Pyrginae): life history and morphological studies. Australian Entomologist 29, 1-10. 10. Graham A. J., Bollam H. H. & Williams M. R. (1996) An unusual temporally isolated population of Neolucia agricola Waterhouse & Turner in Western Australia (Lepidoptera: Lycaenidae). Australian Entomologist 23, 111-4. 11. Valentine P. S. & Johnson S. J. (2003) Notes on the distribution and conservation status of atkinsi Williams, Williams & Hay (Lepidoptera: Hesperiidae). Australian Entomologist 30, 87-91. 12. Williams A. A. E. (1991) New southern records of the Yellow Palmdart trichopepla (Lower) (Lepidoptera: Hesperiidae) in Western Australia. Australian entomological Magazine 18, 43-44.

11 13. Williams A. A. E. (1993) New distribution records for the yellow palmdart Cephrenes trichopepla and the orange palmdart Cephrenes augiades sperthias from Western Australia. Western Australian Naturalist 19, 255-256. 14. Williams A. A. E. (1993) A new larval foodplant for donnysa albina Waterhouse (Lepidoptera: Hesperiidae) in Western Australia. Australian Entomologist 20, 72. 15. Williams A. A. E. (1997) The butterflies (Lepidoptera) of Garden and Rottnest Islands, Western Australia. Australian Entomologist 24, 27-34. 16. Williams, A. (2008) Status of the population of Ogyris subterrestris petrina Field 1999 at Barbalin. Unpublished DEC internal report, 1pp. 17. Williams A. A. E. & Atkins A. F. (1997) Notes on the life history of the Western Australian hayi Edwards and Graham (Lepidoptera: Hesperiidae). Australian Entomologist 24, 81-85. 18. Williams A. A. E. & Hall G. P. (1993) New records of butterflies (Lepidoptera: Hesperioidae and Papilionoidea) from Bernier Island, Western Australia. Australian Entomologist 20, 45-46. 19. Williams A. A. E., Hay R. W. & Bollam H. H. (1992) New records for six lycaenid butterflies in Western Australia (Lepidoptera: Lycaenidae). Australian entomological Magazine 19, 25-27. 20. Williams A. A. E. & Powell R. J. (2006) The butterflies of Middle, Mondrain, Sandy Hook, Woody and Goose islands in the Recherche Archipelago, Western Australia. Australian Entomologist 33, 39-48. 21. Williams A. A. E. & Tomlinson A. G. (1994) Further distributional records and natural history notes on butterflies from Western Australia. Victorian Entomologist 24, 122-124. 22. Williams A. A. E., Tomlinson A. G. & Lundstrom T. D. (1996) Records of butterflies from the central desert region and semi-arid areas of Western Australia. Victorian Entomologist 26, 29-34. 23. Williams A. A. E. & Williams M. R. (2003) In search of the elusive western flat. Landscope 18, 48-53. 24. Williams, A. & Williams, M. (2005). Endangered or extinct?: Kalgoorlie's arid bronze azure. Landscope 21(2): 19-23 25. Williams A. A. E. & Williams M. R. (2006) A new locality record and notes on the distribution of Trapezites waterhousei Mayo & Atkins (Lepidoptera: Hesperiidae). Australian Entomologist 33, 139-141. 26. Williams A. A. E., Williams M. R. & Atkins A. F. (1997) Notes on some Western Australian butterflies. Victorian Entomologist 27, 44-49. 27. Williams A. A. E., Williams M. R. & Hay R. W. (1998) A new species of Trapezites Hübner (Lepidoptera: Hesperiidae) from Western Australia. Australian Entomologist 25, 7-12.

12 28. Williams A. A. E., Williams M. R., Hay R. W. & Tomlinson A. G. (1993) Some distributional records and natural history notes on butterflies from Western Australia. Victorian Entomologist 23, 126-131. 29. Williams M. R. (1990) New Gahnia Forst.& Forst.F. foodplant records for three Western Australian Skippers (Lepidoptera: Hesperiidae). Australian Entomological Magazine 17, 113-114. 30. Williams M. R. (1996) An extraordinary extralimital record of Acraea andromacha andromacha (Fabricius) 1775, (Lepidoptera: Nymphalidae) from Western Australia. Victorian Entomologist 26, 23-24. 31. Williams M. R. & Atkins A. F. (1997) The life history of Trapezites waterhousei Mayo and Atkins (Lepidoptera: Hesperiidae: ). Australian entomologist 24, 1-4. 32. Williams M. R., Atkins A. F., Hay R. W. & Bollam H. H. (1992) The life history of Ogyris otanes C. & R. Felder in the Stirling Range, Western Australia (Lepidoptera: Lycaenidae). Australian Entomological Magazine 19, 55-60. 33. Williams M. R. & Bollam H. H. (2001) A new subspecies of Candalides hyacinthinus (Semper) (Lepidoptera: Lycaenidae) from Western Australia. Australian entomologist 28, 49-54. 34. Williams M. R. & Hay R. W. (2001) Two new subspecies of Ogyris otanes C. & R. Felder (Lepidoptera: Lycaenidae) from Western Australia. Australian entomologist 28, 55-63. 35. Williams M. R., Williams A. A. E. & Atkins A. F. (1992) The life history of the Sciron Skipper Trapezites sciron sciron Waterhouse and Lyell (Lepidoptera: Hesperiidae). Australian Entomological Magazine 19, 29-32. 36. Williams M. R., Williams A. A. E. & Lundstrom T. D. (1998) Jewels of the west. Landscope 13, 49-53. 37. Williams M. R., Williams A. A. E., Lundstrom T. D., Hay R. W., Bollam H. H. & Graham A. J. (1995) Range extensions and natural history notes for some Western Australian butterflies. Victorian entomologist 25, 94-96. 38. Williams M. R., Williams A. A. E., Lundstrom T. E. & Hay R. W. (1996) The distribution of Waterhouse’s skipper Trapezites waterhousei Mayo and Atkins (Lepidoptera: Hesperiidae) in Western Australia. Australian entomologist 23, 83- 85.

13 Figure 1. Female (top) and male specimens of the Arid Bronze Azure butterfly (Ogyris subterrestris petrina), showing the upperside wing surfaces. Wingspan = 36 mm. (Photo S. Brown).

14 Figure 2. Female (top) and male specimens of the Arid Bronze Azure butterfly (Ogyris subterrestris petrina), showing the underside wing surfaces. Wingspan = 36 mm. (Photo S. Brown).

15 Figure 3. Mature larva of The Mallee Bronze Azure butterfly (Ogyris subterrestris subterrestris), attended by Camponotus terebrans worker ants. The larva of the Arid Bronze Azure is likely to be very similar. (Photo L. Hunt).

16 Figure 4. Area of occurrence of the Arid Bronze Azure butterfly and the area under application for clearing. The butterfly flight area corresponds to a rectangle delimited by the green GPS co-ordinates: it extends 700 m westwards from the Barbalin North and South roads, and is bounded in the north by the existing major highway and in the south by the railway line. The breeding colony of the butterfly occurs in the remnant vegetation between the mallee fowl sign in the west and the Barbalin North and South roads in the east.

17